sfc: Move the Solarflare drivers

Moves the Solarflare drivers into drivers/net/ethernet/sfc/ and
make the necessary Kconfig and Makefile changes.

CC: Steve Hodgson <shodgson@solarflare.com>
CC: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
diff --git a/drivers/net/ethernet/Kconfig b/drivers/net/ethernet/Kconfig
index 5edd237..0bc6635 100644
--- a/drivers/net/ethernet/Kconfig
+++ b/drivers/net/ethernet/Kconfig
@@ -20,6 +20,7 @@
 source "drivers/net/ethernet/i825xx/Kconfig"
 source "drivers/net/ethernet/qlogic/Kconfig"
 source "drivers/net/ethernet/racal/Kconfig"
+source "drivers/net/ethernet/sfc/Kconfig"
 source "drivers/net/ethernet/smsc/Kconfig"
 source "drivers/net/ethernet/sun/Kconfig"
 
diff --git a/drivers/net/ethernet/Makefile b/drivers/net/ethernet/Makefile
index 18d8a89..50faab5 100644
--- a/drivers/net/ethernet/Makefile
+++ b/drivers/net/ethernet/Makefile
@@ -11,5 +11,6 @@
 obj-$(CONFIG_NET_VENDOR_I825XX) += i825xx/
 obj-$(CONFIG_NET_VENDOR_QLOGIC) += qlogic/
 obj-$(CONFIG_NET_VENDOR_RACAL) += racal/
+obj-$(CONFIG_SFC) += sfc/
 obj-$(CONFIG_NET_VENDOR_SMSC) += smsc/
 obj-$(CONFIG_NET_VENDOR_SUN) += sun/
diff --git a/drivers/net/ethernet/sfc/Kconfig b/drivers/net/ethernet/sfc/Kconfig
new file mode 100644
index 0000000..5d18841
--- /dev/null
+++ b/drivers/net/ethernet/sfc/Kconfig
@@ -0,0 +1,21 @@
+config SFC
+	tristate "Solarflare SFC4000/SFC9000-family support"
+	depends on PCI && INET
+	select MDIO
+	select CRC32
+	select I2C
+	select I2C_ALGOBIT
+	---help---
+	  This driver supports 10-gigabit Ethernet cards based on
+	  the Solarflare SFC4000 and SFC9000-family controllers.
+
+	  To compile this driver as a module, choose M here.  The module
+	  will be called sfc.
+config SFC_MTD
+	bool "Solarflare SFC4000/SFC9000-family MTD support"
+	depends on SFC && MTD && !(SFC=y && MTD=m)
+	default y
+	---help---
+	  This exposes the on-board flash memory as MTD devices (e.g.
+	  /dev/mtd1).  This makes it possible to upload new firmware
+	  to the NIC.
diff --git a/drivers/net/ethernet/sfc/Makefile b/drivers/net/ethernet/sfc/Makefile
new file mode 100644
index 0000000..ab31c71
--- /dev/null
+++ b/drivers/net/ethernet/sfc/Makefile
@@ -0,0 +1,8 @@
+sfc-y			+= efx.o nic.o falcon.o siena.o tx.o rx.o filter.o \
+			   falcon_xmac.o mcdi_mac.o \
+			   selftest.o ethtool.o qt202x_phy.o mdio_10g.o \
+			   tenxpress.o txc43128_phy.o falcon_boards.o \
+			   mcdi.o mcdi_phy.o
+sfc-$(CONFIG_SFC_MTD)	+= mtd.o
+
+obj-$(CONFIG_SFC)	+= sfc.o
diff --git a/drivers/net/ethernet/sfc/bitfield.h b/drivers/net/ethernet/sfc/bitfield.h
new file mode 100644
index 0000000..098ac2a
--- /dev/null
+++ b/drivers/net/ethernet/sfc/bitfield.h
@@ -0,0 +1,538 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2005-2006 Fen Systems Ltd.
+ * Copyright 2006-2009 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#ifndef EFX_BITFIELD_H
+#define EFX_BITFIELD_H
+
+/*
+ * Efx bitfield access
+ *
+ * Efx NICs make extensive use of bitfields up to 128 bits
+ * wide.  Since there is no native 128-bit datatype on most systems,
+ * and since 64-bit datatypes are inefficient on 32-bit systems and
+ * vice versa, we wrap accesses in a way that uses the most efficient
+ * datatype.
+ *
+ * The NICs are PCI devices and therefore little-endian.  Since most
+ * of the quantities that we deal with are DMAed to/from host memory,
+ * we define our datatypes (efx_oword_t, efx_qword_t and
+ * efx_dword_t) to be little-endian.
+ */
+
+/* Lowest bit numbers and widths */
+#define EFX_DUMMY_FIELD_LBN 0
+#define EFX_DUMMY_FIELD_WIDTH 0
+#define EFX_DWORD_0_LBN 0
+#define EFX_DWORD_0_WIDTH 32
+#define EFX_DWORD_1_LBN 32
+#define EFX_DWORD_1_WIDTH 32
+#define EFX_DWORD_2_LBN 64
+#define EFX_DWORD_2_WIDTH 32
+#define EFX_DWORD_3_LBN 96
+#define EFX_DWORD_3_WIDTH 32
+#define EFX_QWORD_0_LBN 0
+#define EFX_QWORD_0_WIDTH 64
+
+/* Specified attribute (e.g. LBN) of the specified field */
+#define EFX_VAL(field, attribute) field ## _ ## attribute
+/* Low bit number of the specified field */
+#define EFX_LOW_BIT(field) EFX_VAL(field, LBN)
+/* Bit width of the specified field */
+#define EFX_WIDTH(field) EFX_VAL(field, WIDTH)
+/* High bit number of the specified field */
+#define EFX_HIGH_BIT(field) (EFX_LOW_BIT(field) + EFX_WIDTH(field) - 1)
+/* Mask equal in width to the specified field.
+ *
+ * For example, a field with width 5 would have a mask of 0x1f.
+ *
+ * The maximum width mask that can be generated is 64 bits.
+ */
+#define EFX_MASK64(width)			\
+	((width) == 64 ? ~((u64) 0) :		\
+	 (((((u64) 1) << (width))) - 1))
+
+/* Mask equal in width to the specified field.
+ *
+ * For example, a field with width 5 would have a mask of 0x1f.
+ *
+ * The maximum width mask that can be generated is 32 bits.  Use
+ * EFX_MASK64 for higher width fields.
+ */
+#define EFX_MASK32(width)			\
+	((width) == 32 ? ~((u32) 0) :		\
+	 (((((u32) 1) << (width))) - 1))
+
+/* A doubleword (i.e. 4 byte) datatype - little-endian in HW */
+typedef union efx_dword {
+	__le32 u32[1];
+} efx_dword_t;
+
+/* A quadword (i.e. 8 byte) datatype - little-endian in HW */
+typedef union efx_qword {
+	__le64 u64[1];
+	__le32 u32[2];
+	efx_dword_t dword[2];
+} efx_qword_t;
+
+/* An octword (eight-word, i.e. 16 byte) datatype - little-endian in HW */
+typedef union efx_oword {
+	__le64 u64[2];
+	efx_qword_t qword[2];
+	__le32 u32[4];
+	efx_dword_t dword[4];
+} efx_oword_t;
+
+/* Format string and value expanders for printk */
+#define EFX_DWORD_FMT "%08x"
+#define EFX_QWORD_FMT "%08x:%08x"
+#define EFX_OWORD_FMT "%08x:%08x:%08x:%08x"
+#define EFX_DWORD_VAL(dword)				\
+	((unsigned int) le32_to_cpu((dword).u32[0]))
+#define EFX_QWORD_VAL(qword)				\
+	((unsigned int) le32_to_cpu((qword).u32[1])),	\
+	((unsigned int) le32_to_cpu((qword).u32[0]))
+#define EFX_OWORD_VAL(oword)				\
+	((unsigned int) le32_to_cpu((oword).u32[3])),	\
+	((unsigned int) le32_to_cpu((oword).u32[2])),	\
+	((unsigned int) le32_to_cpu((oword).u32[1])),	\
+	((unsigned int) le32_to_cpu((oword).u32[0]))
+
+/*
+ * Extract bit field portion [low,high) from the native-endian element
+ * which contains bits [min,max).
+ *
+ * For example, suppose "element" represents the high 32 bits of a
+ * 64-bit value, and we wish to extract the bits belonging to the bit
+ * field occupying bits 28-45 of this 64-bit value.
+ *
+ * Then EFX_EXTRACT ( element, 32, 63, 28, 45 ) would give
+ *
+ *   ( element ) << 4
+ *
+ * The result will contain the relevant bits filled in in the range
+ * [0,high-low), with garbage in bits [high-low+1,...).
+ */
+#define EFX_EXTRACT_NATIVE(native_element, min, max, low, high)		\
+	(((low > max) || (high < min)) ? 0 :				\
+	 ((low > min) ?							\
+	  ((native_element) >> (low - min)) :				\
+	  ((native_element) << (min - low))))
+
+/*
+ * Extract bit field portion [low,high) from the 64-bit little-endian
+ * element which contains bits [min,max)
+ */
+#define EFX_EXTRACT64(element, min, max, low, high)			\
+	EFX_EXTRACT_NATIVE(le64_to_cpu(element), min, max, low, high)
+
+/*
+ * Extract bit field portion [low,high) from the 32-bit little-endian
+ * element which contains bits [min,max)
+ */
+#define EFX_EXTRACT32(element, min, max, low, high)			\
+	EFX_EXTRACT_NATIVE(le32_to_cpu(element), min, max, low, high)
+
+#define EFX_EXTRACT_OWORD64(oword, low, high)				\
+	((EFX_EXTRACT64((oword).u64[0], 0, 63, low, high) |		\
+	  EFX_EXTRACT64((oword).u64[1], 64, 127, low, high)) &		\
+	 EFX_MASK64(high + 1 - low))
+
+#define EFX_EXTRACT_QWORD64(qword, low, high)				\
+	(EFX_EXTRACT64((qword).u64[0], 0, 63, low, high) &		\
+	 EFX_MASK64(high + 1 - low))
+
+#define EFX_EXTRACT_OWORD32(oword, low, high)				\
+	((EFX_EXTRACT32((oword).u32[0], 0, 31, low, high) |		\
+	  EFX_EXTRACT32((oword).u32[1], 32, 63, low, high) |		\
+	  EFX_EXTRACT32((oword).u32[2], 64, 95, low, high) |		\
+	  EFX_EXTRACT32((oword).u32[3], 96, 127, low, high)) &		\
+	 EFX_MASK32(high + 1 - low))
+
+#define EFX_EXTRACT_QWORD32(qword, low, high)				\
+	((EFX_EXTRACT32((qword).u32[0], 0, 31, low, high) |		\
+	  EFX_EXTRACT32((qword).u32[1], 32, 63, low, high)) &		\
+	 EFX_MASK32(high + 1 - low))
+
+#define EFX_EXTRACT_DWORD(dword, low, high)			\
+	(EFX_EXTRACT32((dword).u32[0], 0, 31, low, high) &	\
+	 EFX_MASK32(high + 1 - low))
+
+#define EFX_OWORD_FIELD64(oword, field)				\
+	EFX_EXTRACT_OWORD64(oword, EFX_LOW_BIT(field),		\
+			    EFX_HIGH_BIT(field))
+
+#define EFX_QWORD_FIELD64(qword, field)				\
+	EFX_EXTRACT_QWORD64(qword, EFX_LOW_BIT(field),		\
+			    EFX_HIGH_BIT(field))
+
+#define EFX_OWORD_FIELD32(oword, field)				\
+	EFX_EXTRACT_OWORD32(oword, EFX_LOW_BIT(field),		\
+			    EFX_HIGH_BIT(field))
+
+#define EFX_QWORD_FIELD32(qword, field)				\
+	EFX_EXTRACT_QWORD32(qword, EFX_LOW_BIT(field),		\
+			    EFX_HIGH_BIT(field))
+
+#define EFX_DWORD_FIELD(dword, field)				\
+	EFX_EXTRACT_DWORD(dword, EFX_LOW_BIT(field),		\
+			  EFX_HIGH_BIT(field))
+
+#define EFX_OWORD_IS_ZERO64(oword)					\
+	(((oword).u64[0] | (oword).u64[1]) == (__force __le64) 0)
+
+#define EFX_QWORD_IS_ZERO64(qword)					\
+	(((qword).u64[0]) == (__force __le64) 0)
+
+#define EFX_OWORD_IS_ZERO32(oword)					     \
+	(((oword).u32[0] | (oword).u32[1] | (oword).u32[2] | (oword).u32[3]) \
+	 == (__force __le32) 0)
+
+#define EFX_QWORD_IS_ZERO32(qword)					\
+	(((qword).u32[0] | (qword).u32[1]) == (__force __le32) 0)
+
+#define EFX_DWORD_IS_ZERO(dword)					\
+	(((dword).u32[0]) == (__force __le32) 0)
+
+#define EFX_OWORD_IS_ALL_ONES64(oword)					\
+	(((oword).u64[0] & (oword).u64[1]) == ~((__force __le64) 0))
+
+#define EFX_QWORD_IS_ALL_ONES64(qword)					\
+	((qword).u64[0] == ~((__force __le64) 0))
+
+#define EFX_OWORD_IS_ALL_ONES32(oword)					\
+	(((oword).u32[0] & (oword).u32[1] & (oword).u32[2] & (oword).u32[3]) \
+	 == ~((__force __le32) 0))
+
+#define EFX_QWORD_IS_ALL_ONES32(qword)					\
+	(((qword).u32[0] & (qword).u32[1]) == ~((__force __le32) 0))
+
+#define EFX_DWORD_IS_ALL_ONES(dword)					\
+	((dword).u32[0] == ~((__force __le32) 0))
+
+#if BITS_PER_LONG == 64
+#define EFX_OWORD_FIELD		EFX_OWORD_FIELD64
+#define EFX_QWORD_FIELD		EFX_QWORD_FIELD64
+#define EFX_OWORD_IS_ZERO	EFX_OWORD_IS_ZERO64
+#define EFX_QWORD_IS_ZERO	EFX_QWORD_IS_ZERO64
+#define EFX_OWORD_IS_ALL_ONES	EFX_OWORD_IS_ALL_ONES64
+#define EFX_QWORD_IS_ALL_ONES	EFX_QWORD_IS_ALL_ONES64
+#else
+#define EFX_OWORD_FIELD		EFX_OWORD_FIELD32
+#define EFX_QWORD_FIELD		EFX_QWORD_FIELD32
+#define EFX_OWORD_IS_ZERO	EFX_OWORD_IS_ZERO32
+#define EFX_QWORD_IS_ZERO	EFX_QWORD_IS_ZERO32
+#define EFX_OWORD_IS_ALL_ONES	EFX_OWORD_IS_ALL_ONES32
+#define EFX_QWORD_IS_ALL_ONES	EFX_QWORD_IS_ALL_ONES32
+#endif
+
+/*
+ * Construct bit field portion
+ *
+ * Creates the portion of the bit field [low,high) that lies within
+ * the range [min,max).
+ */
+#define EFX_INSERT_NATIVE64(min, max, low, high, value)		\
+	(((low > max) || (high < min)) ? 0 :			\
+	 ((low > min) ?						\
+	  (((u64) (value)) << (low - min)) :		\
+	  (((u64) (value)) >> (min - low))))
+
+#define EFX_INSERT_NATIVE32(min, max, low, high, value)		\
+	(((low > max) || (high < min)) ? 0 :			\
+	 ((low > min) ?						\
+	  (((u32) (value)) << (low - min)) :		\
+	  (((u32) (value)) >> (min - low))))
+
+#define EFX_INSERT_NATIVE(min, max, low, high, value)		\
+	((((max - min) >= 32) || ((high - low) >= 32)) ?	\
+	 EFX_INSERT_NATIVE64(min, max, low, high, value) :	\
+	 EFX_INSERT_NATIVE32(min, max, low, high, value))
+
+/*
+ * Construct bit field portion
+ *
+ * Creates the portion of the named bit field that lies within the
+ * range [min,max).
+ */
+#define EFX_INSERT_FIELD_NATIVE(min, max, field, value)		\
+	EFX_INSERT_NATIVE(min, max, EFX_LOW_BIT(field),		\
+			  EFX_HIGH_BIT(field), value)
+
+/*
+ * Construct bit field
+ *
+ * Creates the portion of the named bit fields that lie within the
+ * range [min,max).
+ */
+#define EFX_INSERT_FIELDS_NATIVE(min, max,				\
+				 field1, value1,			\
+				 field2, value2,			\
+				 field3, value3,			\
+				 field4, value4,			\
+				 field5, value5,			\
+				 field6, value6,			\
+				 field7, value7,			\
+				 field8, value8,			\
+				 field9, value9,			\
+				 field10, value10)			\
+	(EFX_INSERT_FIELD_NATIVE((min), (max), field1, (value1)) |	\
+	 EFX_INSERT_FIELD_NATIVE((min), (max), field2, (value2)) |	\
+	 EFX_INSERT_FIELD_NATIVE((min), (max), field3, (value3)) |	\
+	 EFX_INSERT_FIELD_NATIVE((min), (max), field4, (value4)) |	\
+	 EFX_INSERT_FIELD_NATIVE((min), (max), field5, (value5)) |	\
+	 EFX_INSERT_FIELD_NATIVE((min), (max), field6, (value6)) |	\
+	 EFX_INSERT_FIELD_NATIVE((min), (max), field7, (value7)) |	\
+	 EFX_INSERT_FIELD_NATIVE((min), (max), field8, (value8)) |	\
+	 EFX_INSERT_FIELD_NATIVE((min), (max), field9, (value9)) |	\
+	 EFX_INSERT_FIELD_NATIVE((min), (max), field10, (value10)))
+
+#define EFX_INSERT_FIELDS64(...)				\
+	cpu_to_le64(EFX_INSERT_FIELDS_NATIVE(__VA_ARGS__))
+
+#define EFX_INSERT_FIELDS32(...)				\
+	cpu_to_le32(EFX_INSERT_FIELDS_NATIVE(__VA_ARGS__))
+
+#define EFX_POPULATE_OWORD64(oword, ...) do {				\
+	(oword).u64[0] = EFX_INSERT_FIELDS64(0, 63, __VA_ARGS__);	\
+	(oword).u64[1] = EFX_INSERT_FIELDS64(64, 127, __VA_ARGS__);	\
+	} while (0)
+
+#define EFX_POPULATE_QWORD64(qword, ...) do {				\
+	(qword).u64[0] = EFX_INSERT_FIELDS64(0, 63, __VA_ARGS__);	\
+	} while (0)
+
+#define EFX_POPULATE_OWORD32(oword, ...) do {				\
+	(oword).u32[0] = EFX_INSERT_FIELDS32(0, 31, __VA_ARGS__);	\
+	(oword).u32[1] = EFX_INSERT_FIELDS32(32, 63, __VA_ARGS__);	\
+	(oword).u32[2] = EFX_INSERT_FIELDS32(64, 95, __VA_ARGS__);	\
+	(oword).u32[3] = EFX_INSERT_FIELDS32(96, 127, __VA_ARGS__);	\
+	} while (0)
+
+#define EFX_POPULATE_QWORD32(qword, ...) do {				\
+	(qword).u32[0] = EFX_INSERT_FIELDS32(0, 31, __VA_ARGS__);	\
+	(qword).u32[1] = EFX_INSERT_FIELDS32(32, 63, __VA_ARGS__);	\
+	} while (0)
+
+#define EFX_POPULATE_DWORD(dword, ...) do {				\
+	(dword).u32[0] = EFX_INSERT_FIELDS32(0, 31, __VA_ARGS__);	\
+	} while (0)
+
+#if BITS_PER_LONG == 64
+#define EFX_POPULATE_OWORD EFX_POPULATE_OWORD64
+#define EFX_POPULATE_QWORD EFX_POPULATE_QWORD64
+#else
+#define EFX_POPULATE_OWORD EFX_POPULATE_OWORD32
+#define EFX_POPULATE_QWORD EFX_POPULATE_QWORD32
+#endif
+
+/* Populate an octword field with various numbers of arguments */
+#define EFX_POPULATE_OWORD_10 EFX_POPULATE_OWORD
+#define EFX_POPULATE_OWORD_9(oword, ...) \
+	EFX_POPULATE_OWORD_10(oword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_OWORD_8(oword, ...) \
+	EFX_POPULATE_OWORD_9(oword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_OWORD_7(oword, ...) \
+	EFX_POPULATE_OWORD_8(oword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_OWORD_6(oword, ...) \
+	EFX_POPULATE_OWORD_7(oword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_OWORD_5(oword, ...) \
+	EFX_POPULATE_OWORD_6(oword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_OWORD_4(oword, ...) \
+	EFX_POPULATE_OWORD_5(oword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_OWORD_3(oword, ...) \
+	EFX_POPULATE_OWORD_4(oword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_OWORD_2(oword, ...) \
+	EFX_POPULATE_OWORD_3(oword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_OWORD_1(oword, ...) \
+	EFX_POPULATE_OWORD_2(oword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_ZERO_OWORD(oword) \
+	EFX_POPULATE_OWORD_1(oword, EFX_DUMMY_FIELD, 0)
+#define EFX_SET_OWORD(oword) \
+	EFX_POPULATE_OWORD_4(oword, \
+			     EFX_DWORD_0, 0xffffffff, \
+			     EFX_DWORD_1, 0xffffffff, \
+			     EFX_DWORD_2, 0xffffffff, \
+			     EFX_DWORD_3, 0xffffffff)
+
+/* Populate a quadword field with various numbers of arguments */
+#define EFX_POPULATE_QWORD_10 EFX_POPULATE_QWORD
+#define EFX_POPULATE_QWORD_9(qword, ...) \
+	EFX_POPULATE_QWORD_10(qword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_QWORD_8(qword, ...) \
+	EFX_POPULATE_QWORD_9(qword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_QWORD_7(qword, ...) \
+	EFX_POPULATE_QWORD_8(qword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_QWORD_6(qword, ...) \
+	EFX_POPULATE_QWORD_7(qword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_QWORD_5(qword, ...) \
+	EFX_POPULATE_QWORD_6(qword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_QWORD_4(qword, ...) \
+	EFX_POPULATE_QWORD_5(qword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_QWORD_3(qword, ...) \
+	EFX_POPULATE_QWORD_4(qword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_QWORD_2(qword, ...) \
+	EFX_POPULATE_QWORD_3(qword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_QWORD_1(qword, ...) \
+	EFX_POPULATE_QWORD_2(qword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_ZERO_QWORD(qword) \
+	EFX_POPULATE_QWORD_1(qword, EFX_DUMMY_FIELD, 0)
+#define EFX_SET_QWORD(qword) \
+	EFX_POPULATE_QWORD_2(qword, \
+			     EFX_DWORD_0, 0xffffffff, \
+			     EFX_DWORD_1, 0xffffffff)
+
+/* Populate a dword field with various numbers of arguments */
+#define EFX_POPULATE_DWORD_10 EFX_POPULATE_DWORD
+#define EFX_POPULATE_DWORD_9(dword, ...) \
+	EFX_POPULATE_DWORD_10(dword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_DWORD_8(dword, ...) \
+	EFX_POPULATE_DWORD_9(dword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_DWORD_7(dword, ...) \
+	EFX_POPULATE_DWORD_8(dword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_DWORD_6(dword, ...) \
+	EFX_POPULATE_DWORD_7(dword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_DWORD_5(dword, ...) \
+	EFX_POPULATE_DWORD_6(dword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_DWORD_4(dword, ...) \
+	EFX_POPULATE_DWORD_5(dword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_DWORD_3(dword, ...) \
+	EFX_POPULATE_DWORD_4(dword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_DWORD_2(dword, ...) \
+	EFX_POPULATE_DWORD_3(dword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_POPULATE_DWORD_1(dword, ...) \
+	EFX_POPULATE_DWORD_2(dword, EFX_DUMMY_FIELD, 0, __VA_ARGS__)
+#define EFX_ZERO_DWORD(dword) \
+	EFX_POPULATE_DWORD_1(dword, EFX_DUMMY_FIELD, 0)
+#define EFX_SET_DWORD(dword) \
+	EFX_POPULATE_DWORD_1(dword, EFX_DWORD_0, 0xffffffff)
+
+/*
+ * Modify a named field within an already-populated structure.  Used
+ * for read-modify-write operations.
+ *
+ */
+#define EFX_INVERT_OWORD(oword) do {		\
+	(oword).u64[0] = ~((oword).u64[0]);	\
+	(oword).u64[1] = ~((oword).u64[1]);	\
+	} while (0)
+
+#define EFX_AND_OWORD(oword, from, mask)			\
+	do {							\
+		(oword).u64[0] = (from).u64[0] & (mask).u64[0];	\
+		(oword).u64[1] = (from).u64[1] & (mask).u64[1];	\
+	} while (0)
+
+#define EFX_OR_OWORD(oword, from, mask)				\
+	do {							\
+		(oword).u64[0] = (from).u64[0] | (mask).u64[0];	\
+		(oword).u64[1] = (from).u64[1] | (mask).u64[1];	\
+	} while (0)
+
+#define EFX_INSERT64(min, max, low, high, value)			\
+	cpu_to_le64(EFX_INSERT_NATIVE(min, max, low, high, value))
+
+#define EFX_INSERT32(min, max, low, high, value)			\
+	cpu_to_le32(EFX_INSERT_NATIVE(min, max, low, high, value))
+
+#define EFX_INPLACE_MASK64(min, max, low, high)				\
+	EFX_INSERT64(min, max, low, high, EFX_MASK64(high + 1 - low))
+
+#define EFX_INPLACE_MASK32(min, max, low, high)				\
+	EFX_INSERT32(min, max, low, high, EFX_MASK32(high + 1 - low))
+
+#define EFX_SET_OWORD64(oword, low, high, value) do {			\
+	(oword).u64[0] = (((oword).u64[0] 				\
+			   & ~EFX_INPLACE_MASK64(0,  63, low, high))	\
+			  | EFX_INSERT64(0,  63, low, high, value));	\
+	(oword).u64[1] = (((oword).u64[1] 				\
+			   & ~EFX_INPLACE_MASK64(64, 127, low, high))	\
+			  | EFX_INSERT64(64, 127, low, high, value));	\
+	} while (0)
+
+#define EFX_SET_QWORD64(qword, low, high, value) do {			\
+	(qword).u64[0] = (((qword).u64[0] 				\
+			   & ~EFX_INPLACE_MASK64(0, 63, low, high))	\
+			  | EFX_INSERT64(0, 63, low, high, value));	\
+	} while (0)
+
+#define EFX_SET_OWORD32(oword, low, high, value) do {			\
+	(oword).u32[0] = (((oword).u32[0] 				\
+			   & ~EFX_INPLACE_MASK32(0, 31, low, high))	\
+			  | EFX_INSERT32(0, 31, low, high, value));	\
+	(oword).u32[1] = (((oword).u32[1] 				\
+			   & ~EFX_INPLACE_MASK32(32, 63, low, high))	\
+			  | EFX_INSERT32(32, 63, low, high, value));	\
+	(oword).u32[2] = (((oword).u32[2] 				\
+			   & ~EFX_INPLACE_MASK32(64, 95, low, high))	\
+			  | EFX_INSERT32(64, 95, low, high, value));	\
+	(oword).u32[3] = (((oword).u32[3] 				\
+			   & ~EFX_INPLACE_MASK32(96, 127, low, high))	\
+			  | EFX_INSERT32(96, 127, low, high, value));	\
+	} while (0)
+
+#define EFX_SET_QWORD32(qword, low, high, value) do {			\
+	(qword).u32[0] = (((qword).u32[0] 				\
+			   & ~EFX_INPLACE_MASK32(0, 31, low, high))	\
+			  | EFX_INSERT32(0, 31, low, high, value));	\
+	(qword).u32[1] = (((qword).u32[1] 				\
+			   & ~EFX_INPLACE_MASK32(32, 63, low, high))	\
+			  | EFX_INSERT32(32, 63, low, high, value));	\
+	} while (0)
+
+#define EFX_SET_DWORD32(dword, low, high, value) do {			\
+	(dword).u32[0] = (((dword).u32[0]				\
+			   & ~EFX_INPLACE_MASK32(0, 31, low, high))	\
+			  | EFX_INSERT32(0, 31, low, high, value));	\
+	} while (0)
+
+#define EFX_SET_OWORD_FIELD64(oword, field, value)			\
+	EFX_SET_OWORD64(oword, EFX_LOW_BIT(field),			\
+			 EFX_HIGH_BIT(field), value)
+
+#define EFX_SET_QWORD_FIELD64(qword, field, value)			\
+	EFX_SET_QWORD64(qword, EFX_LOW_BIT(field),			\
+			 EFX_HIGH_BIT(field), value)
+
+#define EFX_SET_OWORD_FIELD32(oword, field, value)			\
+	EFX_SET_OWORD32(oword, EFX_LOW_BIT(field),			\
+			 EFX_HIGH_BIT(field), value)
+
+#define EFX_SET_QWORD_FIELD32(qword, field, value)			\
+	EFX_SET_QWORD32(qword, EFX_LOW_BIT(field),			\
+			 EFX_HIGH_BIT(field), value)
+
+#define EFX_SET_DWORD_FIELD(dword, field, value)			\
+	EFX_SET_DWORD32(dword, EFX_LOW_BIT(field),			\
+			 EFX_HIGH_BIT(field), value)
+
+
+
+#if BITS_PER_LONG == 64
+#define EFX_SET_OWORD_FIELD EFX_SET_OWORD_FIELD64
+#define EFX_SET_QWORD_FIELD EFX_SET_QWORD_FIELD64
+#else
+#define EFX_SET_OWORD_FIELD EFX_SET_OWORD_FIELD32
+#define EFX_SET_QWORD_FIELD EFX_SET_QWORD_FIELD32
+#endif
+
+/* Used to avoid compiler warnings about shift range exceeding width
+ * of the data types when dma_addr_t is only 32 bits wide.
+ */
+#define DMA_ADDR_T_WIDTH	(8 * sizeof(dma_addr_t))
+#define EFX_DMA_TYPE_WIDTH(width) \
+	(((width) < DMA_ADDR_T_WIDTH) ? (width) : DMA_ADDR_T_WIDTH)
+
+
+/* Static initialiser */
+#define EFX_OWORD32(a, b, c, d)						\
+	{ .u32 = { cpu_to_le32(a), cpu_to_le32(b), \
+		   cpu_to_le32(c), cpu_to_le32(d) } }
+
+#endif /* EFX_BITFIELD_H */
diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
new file mode 100644
index 0000000..faca764
--- /dev/null
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -0,0 +1,2714 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2005-2006 Fen Systems Ltd.
+ * Copyright 2005-2011 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/delay.h>
+#include <linux/notifier.h>
+#include <linux/ip.h>
+#include <linux/tcp.h>
+#include <linux/in.h>
+#include <linux/crc32.h>
+#include <linux/ethtool.h>
+#include <linux/topology.h>
+#include <linux/gfp.h>
+#include <linux/cpu_rmap.h>
+#include "net_driver.h"
+#include "efx.h"
+#include "nic.h"
+
+#include "mcdi.h"
+#include "workarounds.h"
+
+/**************************************************************************
+ *
+ * Type name strings
+ *
+ **************************************************************************
+ */
+
+/* Loopback mode names (see LOOPBACK_MODE()) */
+const unsigned int efx_loopback_mode_max = LOOPBACK_MAX;
+const char *efx_loopback_mode_names[] = {
+	[LOOPBACK_NONE]		= "NONE",
+	[LOOPBACK_DATA]		= "DATAPATH",
+	[LOOPBACK_GMAC]		= "GMAC",
+	[LOOPBACK_XGMII]	= "XGMII",
+	[LOOPBACK_XGXS]		= "XGXS",
+	[LOOPBACK_XAUI]  	= "XAUI",
+	[LOOPBACK_GMII] 	= "GMII",
+	[LOOPBACK_SGMII] 	= "SGMII",
+	[LOOPBACK_XGBR]		= "XGBR",
+	[LOOPBACK_XFI]		= "XFI",
+	[LOOPBACK_XAUI_FAR]	= "XAUI_FAR",
+	[LOOPBACK_GMII_FAR]	= "GMII_FAR",
+	[LOOPBACK_SGMII_FAR]	= "SGMII_FAR",
+	[LOOPBACK_XFI_FAR]	= "XFI_FAR",
+	[LOOPBACK_GPHY]		= "GPHY",
+	[LOOPBACK_PHYXS]	= "PHYXS",
+	[LOOPBACK_PCS]	 	= "PCS",
+	[LOOPBACK_PMAPMD] 	= "PMA/PMD",
+	[LOOPBACK_XPORT]	= "XPORT",
+	[LOOPBACK_XGMII_WS]	= "XGMII_WS",
+	[LOOPBACK_XAUI_WS]  	= "XAUI_WS",
+	[LOOPBACK_XAUI_WS_FAR]  = "XAUI_WS_FAR",
+	[LOOPBACK_XAUI_WS_NEAR] = "XAUI_WS_NEAR",
+	[LOOPBACK_GMII_WS] 	= "GMII_WS",
+	[LOOPBACK_XFI_WS]	= "XFI_WS",
+	[LOOPBACK_XFI_WS_FAR]	= "XFI_WS_FAR",
+	[LOOPBACK_PHYXS_WS]  	= "PHYXS_WS",
+};
+
+const unsigned int efx_reset_type_max = RESET_TYPE_MAX;
+const char *efx_reset_type_names[] = {
+	[RESET_TYPE_INVISIBLE]     = "INVISIBLE",
+	[RESET_TYPE_ALL]           = "ALL",
+	[RESET_TYPE_WORLD]         = "WORLD",
+	[RESET_TYPE_DISABLE]       = "DISABLE",
+	[RESET_TYPE_TX_WATCHDOG]   = "TX_WATCHDOG",
+	[RESET_TYPE_INT_ERROR]     = "INT_ERROR",
+	[RESET_TYPE_RX_RECOVERY]   = "RX_RECOVERY",
+	[RESET_TYPE_RX_DESC_FETCH] = "RX_DESC_FETCH",
+	[RESET_TYPE_TX_DESC_FETCH] = "TX_DESC_FETCH",
+	[RESET_TYPE_TX_SKIP]       = "TX_SKIP",
+	[RESET_TYPE_MC_FAILURE]    = "MC_FAILURE",
+};
+
+#define EFX_MAX_MTU (9 * 1024)
+
+/* Reset workqueue. If any NIC has a hardware failure then a reset will be
+ * queued onto this work queue. This is not a per-nic work queue, because
+ * efx_reset_work() acquires the rtnl lock, so resets are naturally serialised.
+ */
+static struct workqueue_struct *reset_workqueue;
+
+/**************************************************************************
+ *
+ * Configurable values
+ *
+ *************************************************************************/
+
+/*
+ * Use separate channels for TX and RX events
+ *
+ * Set this to 1 to use separate channels for TX and RX. It allows us
+ * to control interrupt affinity separately for TX and RX.
+ *
+ * This is only used in MSI-X interrupt mode
+ */
+static unsigned int separate_tx_channels;
+module_param(separate_tx_channels, uint, 0444);
+MODULE_PARM_DESC(separate_tx_channels,
+		 "Use separate channels for TX and RX");
+
+/* This is the weight assigned to each of the (per-channel) virtual
+ * NAPI devices.
+ */
+static int napi_weight = 64;
+
+/* This is the time (in jiffies) between invocations of the hardware
+ * monitor.  On Falcon-based NICs, this will:
+ * - Check the on-board hardware monitor;
+ * - Poll the link state and reconfigure the hardware as necessary.
+ */
+static unsigned int efx_monitor_interval = 1 * HZ;
+
+/* This controls whether or not the driver will initialise devices
+ * with invalid MAC addresses stored in the EEPROM or flash.  If true,
+ * such devices will be initialised with a random locally-generated
+ * MAC address.  This allows for loading the sfc_mtd driver to
+ * reprogram the flash, even if the flash contents (including the MAC
+ * address) have previously been erased.
+ */
+static unsigned int allow_bad_hwaddr;
+
+/* Initial interrupt moderation settings.  They can be modified after
+ * module load with ethtool.
+ *
+ * The default for RX should strike a balance between increasing the
+ * round-trip latency and reducing overhead.
+ */
+static unsigned int rx_irq_mod_usec = 60;
+
+/* Initial interrupt moderation settings.  They can be modified after
+ * module load with ethtool.
+ *
+ * This default is chosen to ensure that a 10G link does not go idle
+ * while a TX queue is stopped after it has become full.  A queue is
+ * restarted when it drops below half full.  The time this takes (assuming
+ * worst case 3 descriptors per packet and 1024 descriptors) is
+ *   512 / 3 * 1.2 = 205 usec.
+ */
+static unsigned int tx_irq_mod_usec = 150;
+
+/* This is the first interrupt mode to try out of:
+ * 0 => MSI-X
+ * 1 => MSI
+ * 2 => legacy
+ */
+static unsigned int interrupt_mode;
+
+/* This is the requested number of CPUs to use for Receive-Side Scaling (RSS),
+ * i.e. the number of CPUs among which we may distribute simultaneous
+ * interrupt handling.
+ *
+ * Cards without MSI-X will only target one CPU via legacy or MSI interrupt.
+ * The default (0) means to assign an interrupt to each package (level II cache)
+ */
+static unsigned int rss_cpus;
+module_param(rss_cpus, uint, 0444);
+MODULE_PARM_DESC(rss_cpus, "Number of CPUs to use for Receive-Side Scaling");
+
+static int phy_flash_cfg;
+module_param(phy_flash_cfg, int, 0644);
+MODULE_PARM_DESC(phy_flash_cfg, "Set PHYs into reflash mode initially");
+
+static unsigned irq_adapt_low_thresh = 10000;
+module_param(irq_adapt_low_thresh, uint, 0644);
+MODULE_PARM_DESC(irq_adapt_low_thresh,
+		 "Threshold score for reducing IRQ moderation");
+
+static unsigned irq_adapt_high_thresh = 20000;
+module_param(irq_adapt_high_thresh, uint, 0644);
+MODULE_PARM_DESC(irq_adapt_high_thresh,
+		 "Threshold score for increasing IRQ moderation");
+
+static unsigned debug = (NETIF_MSG_DRV | NETIF_MSG_PROBE |
+			 NETIF_MSG_LINK | NETIF_MSG_IFDOWN |
+			 NETIF_MSG_IFUP | NETIF_MSG_RX_ERR |
+			 NETIF_MSG_TX_ERR | NETIF_MSG_HW);
+module_param(debug, uint, 0);
+MODULE_PARM_DESC(debug, "Bitmapped debugging message enable value");
+
+/**************************************************************************
+ *
+ * Utility functions and prototypes
+ *
+ *************************************************************************/
+
+static void efx_remove_channels(struct efx_nic *efx);
+static void efx_remove_port(struct efx_nic *efx);
+static void efx_init_napi(struct efx_nic *efx);
+static void efx_fini_napi(struct efx_nic *efx);
+static void efx_fini_napi_channel(struct efx_channel *channel);
+static void efx_fini_struct(struct efx_nic *efx);
+static void efx_start_all(struct efx_nic *efx);
+static void efx_stop_all(struct efx_nic *efx);
+
+#define EFX_ASSERT_RESET_SERIALISED(efx)		\
+	do {						\
+		if ((efx->state == STATE_RUNNING) ||	\
+		    (efx->state == STATE_DISABLED))	\
+			ASSERT_RTNL();			\
+	} while (0)
+
+/**************************************************************************
+ *
+ * Event queue processing
+ *
+ *************************************************************************/
+
+/* Process channel's event queue
+ *
+ * This function is responsible for processing the event queue of a
+ * single channel.  The caller must guarantee that this function will
+ * never be concurrently called more than once on the same channel,
+ * though different channels may be being processed concurrently.
+ */
+static int efx_process_channel(struct efx_channel *channel, int budget)
+{
+	struct efx_nic *efx = channel->efx;
+	int spent;
+
+	if (unlikely(efx->reset_pending || !channel->enabled))
+		return 0;
+
+	spent = efx_nic_process_eventq(channel, budget);
+	if (spent == 0)
+		return 0;
+
+	/* Deliver last RX packet. */
+	if (channel->rx_pkt) {
+		__efx_rx_packet(channel, channel->rx_pkt,
+				channel->rx_pkt_csummed);
+		channel->rx_pkt = NULL;
+	}
+
+	efx_rx_strategy(channel);
+
+	efx_fast_push_rx_descriptors(efx_channel_get_rx_queue(channel));
+
+	return spent;
+}
+
+/* Mark channel as finished processing
+ *
+ * Note that since we will not receive further interrupts for this
+ * channel before we finish processing and call the eventq_read_ack()
+ * method, there is no need to use the interrupt hold-off timers.
+ */
+static inline void efx_channel_processed(struct efx_channel *channel)
+{
+	/* The interrupt handler for this channel may set work_pending
+	 * as soon as we acknowledge the events we've seen.  Make sure
+	 * it's cleared before then. */
+	channel->work_pending = false;
+	smp_wmb();
+
+	efx_nic_eventq_read_ack(channel);
+}
+
+/* NAPI poll handler
+ *
+ * NAPI guarantees serialisation of polls of the same device, which
+ * provides the guarantee required by efx_process_channel().
+ */
+static int efx_poll(struct napi_struct *napi, int budget)
+{
+	struct efx_channel *channel =
+		container_of(napi, struct efx_channel, napi_str);
+	struct efx_nic *efx = channel->efx;
+	int spent;
+
+	netif_vdbg(efx, intr, efx->net_dev,
+		   "channel %d NAPI poll executing on CPU %d\n",
+		   channel->channel, raw_smp_processor_id());
+
+	spent = efx_process_channel(channel, budget);
+
+	if (spent < budget) {
+		if (channel->channel < efx->n_rx_channels &&
+		    efx->irq_rx_adaptive &&
+		    unlikely(++channel->irq_count == 1000)) {
+			if (unlikely(channel->irq_mod_score <
+				     irq_adapt_low_thresh)) {
+				if (channel->irq_moderation > 1) {
+					channel->irq_moderation -= 1;
+					efx->type->push_irq_moderation(channel);
+				}
+			} else if (unlikely(channel->irq_mod_score >
+					    irq_adapt_high_thresh)) {
+				if (channel->irq_moderation <
+				    efx->irq_rx_moderation) {
+					channel->irq_moderation += 1;
+					efx->type->push_irq_moderation(channel);
+				}
+			}
+			channel->irq_count = 0;
+			channel->irq_mod_score = 0;
+		}
+
+		efx_filter_rfs_expire(channel);
+
+		/* There is no race here; although napi_disable() will
+		 * only wait for napi_complete(), this isn't a problem
+		 * since efx_channel_processed() will have no effect if
+		 * interrupts have already been disabled.
+		 */
+		napi_complete(napi);
+		efx_channel_processed(channel);
+	}
+
+	return spent;
+}
+
+/* Process the eventq of the specified channel immediately on this CPU
+ *
+ * Disable hardware generated interrupts, wait for any existing
+ * processing to finish, then directly poll (and ack ) the eventq.
+ * Finally reenable NAPI and interrupts.
+ *
+ * This is for use only during a loopback self-test.  It must not
+ * deliver any packets up the stack as this can result in deadlock.
+ */
+void efx_process_channel_now(struct efx_channel *channel)
+{
+	struct efx_nic *efx = channel->efx;
+
+	BUG_ON(channel->channel >= efx->n_channels);
+	BUG_ON(!channel->enabled);
+	BUG_ON(!efx->loopback_selftest);
+
+	/* Disable interrupts and wait for ISRs to complete */
+	efx_nic_disable_interrupts(efx);
+	if (efx->legacy_irq) {
+		synchronize_irq(efx->legacy_irq);
+		efx->legacy_irq_enabled = false;
+	}
+	if (channel->irq)
+		synchronize_irq(channel->irq);
+
+	/* Wait for any NAPI processing to complete */
+	napi_disable(&channel->napi_str);
+
+	/* Poll the channel */
+	efx_process_channel(channel, channel->eventq_mask + 1);
+
+	/* Ack the eventq. This may cause an interrupt to be generated
+	 * when they are reenabled */
+	efx_channel_processed(channel);
+
+	napi_enable(&channel->napi_str);
+	if (efx->legacy_irq)
+		efx->legacy_irq_enabled = true;
+	efx_nic_enable_interrupts(efx);
+}
+
+/* Create event queue
+ * Event queue memory allocations are done only once.  If the channel
+ * is reset, the memory buffer will be reused; this guards against
+ * errors during channel reset and also simplifies interrupt handling.
+ */
+static int efx_probe_eventq(struct efx_channel *channel)
+{
+	struct efx_nic *efx = channel->efx;
+	unsigned long entries;
+
+	netif_dbg(channel->efx, probe, channel->efx->net_dev,
+		  "chan %d create event queue\n", channel->channel);
+
+	/* Build an event queue with room for one event per tx and rx buffer,
+	 * plus some extra for link state events and MCDI completions. */
+	entries = roundup_pow_of_two(efx->rxq_entries + efx->txq_entries + 128);
+	EFX_BUG_ON_PARANOID(entries > EFX_MAX_EVQ_SIZE);
+	channel->eventq_mask = max(entries, EFX_MIN_EVQ_SIZE) - 1;
+
+	return efx_nic_probe_eventq(channel);
+}
+
+/* Prepare channel's event queue */
+static void efx_init_eventq(struct efx_channel *channel)
+{
+	netif_dbg(channel->efx, drv, channel->efx->net_dev,
+		  "chan %d init event queue\n", channel->channel);
+
+	channel->eventq_read_ptr = 0;
+
+	efx_nic_init_eventq(channel);
+}
+
+static void efx_fini_eventq(struct efx_channel *channel)
+{
+	netif_dbg(channel->efx, drv, channel->efx->net_dev,
+		  "chan %d fini event queue\n", channel->channel);
+
+	efx_nic_fini_eventq(channel);
+}
+
+static void efx_remove_eventq(struct efx_channel *channel)
+{
+	netif_dbg(channel->efx, drv, channel->efx->net_dev,
+		  "chan %d remove event queue\n", channel->channel);
+
+	efx_nic_remove_eventq(channel);
+}
+
+/**************************************************************************
+ *
+ * Channel handling
+ *
+ *************************************************************************/
+
+/* Allocate and initialise a channel structure, optionally copying
+ * parameters (but not resources) from an old channel structure. */
+static struct efx_channel *
+efx_alloc_channel(struct efx_nic *efx, int i, struct efx_channel *old_channel)
+{
+	struct efx_channel *channel;
+	struct efx_rx_queue *rx_queue;
+	struct efx_tx_queue *tx_queue;
+	int j;
+
+	if (old_channel) {
+		channel = kmalloc(sizeof(*channel), GFP_KERNEL);
+		if (!channel)
+			return NULL;
+
+		*channel = *old_channel;
+
+		channel->napi_dev = NULL;
+		memset(&channel->eventq, 0, sizeof(channel->eventq));
+
+		rx_queue = &channel->rx_queue;
+		rx_queue->buffer = NULL;
+		memset(&rx_queue->rxd, 0, sizeof(rx_queue->rxd));
+
+		for (j = 0; j < EFX_TXQ_TYPES; j++) {
+			tx_queue = &channel->tx_queue[j];
+			if (tx_queue->channel)
+				tx_queue->channel = channel;
+			tx_queue->buffer = NULL;
+			memset(&tx_queue->txd, 0, sizeof(tx_queue->txd));
+		}
+	} else {
+		channel = kzalloc(sizeof(*channel), GFP_KERNEL);
+		if (!channel)
+			return NULL;
+
+		channel->efx = efx;
+		channel->channel = i;
+
+		for (j = 0; j < EFX_TXQ_TYPES; j++) {
+			tx_queue = &channel->tx_queue[j];
+			tx_queue->efx = efx;
+			tx_queue->queue = i * EFX_TXQ_TYPES + j;
+			tx_queue->channel = channel;
+		}
+	}
+
+	rx_queue = &channel->rx_queue;
+	rx_queue->efx = efx;
+	setup_timer(&rx_queue->slow_fill, efx_rx_slow_fill,
+		    (unsigned long)rx_queue);
+
+	return channel;
+}
+
+static int efx_probe_channel(struct efx_channel *channel)
+{
+	struct efx_tx_queue *tx_queue;
+	struct efx_rx_queue *rx_queue;
+	int rc;
+
+	netif_dbg(channel->efx, probe, channel->efx->net_dev,
+		  "creating channel %d\n", channel->channel);
+
+	rc = efx_probe_eventq(channel);
+	if (rc)
+		goto fail1;
+
+	efx_for_each_channel_tx_queue(tx_queue, channel) {
+		rc = efx_probe_tx_queue(tx_queue);
+		if (rc)
+			goto fail2;
+	}
+
+	efx_for_each_channel_rx_queue(rx_queue, channel) {
+		rc = efx_probe_rx_queue(rx_queue);
+		if (rc)
+			goto fail3;
+	}
+
+	channel->n_rx_frm_trunc = 0;
+
+	return 0;
+
+ fail3:
+	efx_for_each_channel_rx_queue(rx_queue, channel)
+		efx_remove_rx_queue(rx_queue);
+ fail2:
+	efx_for_each_channel_tx_queue(tx_queue, channel)
+		efx_remove_tx_queue(tx_queue);
+ fail1:
+	return rc;
+}
+
+
+static void efx_set_channel_names(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+	const char *type = "";
+	int number;
+
+	efx_for_each_channel(channel, efx) {
+		number = channel->channel;
+		if (efx->n_channels > efx->n_rx_channels) {
+			if (channel->channel < efx->n_rx_channels) {
+				type = "-rx";
+			} else {
+				type = "-tx";
+				number -= efx->n_rx_channels;
+			}
+		}
+		snprintf(efx->channel_name[channel->channel],
+			 sizeof(efx->channel_name[0]),
+			 "%s%s-%d", efx->name, type, number);
+	}
+}
+
+static int efx_probe_channels(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+	int rc;
+
+	/* Restart special buffer allocation */
+	efx->next_buffer_table = 0;
+
+	efx_for_each_channel(channel, efx) {
+		rc = efx_probe_channel(channel);
+		if (rc) {
+			netif_err(efx, probe, efx->net_dev,
+				  "failed to create channel %d\n",
+				  channel->channel);
+			goto fail;
+		}
+	}
+	efx_set_channel_names(efx);
+
+	return 0;
+
+fail:
+	efx_remove_channels(efx);
+	return rc;
+}
+
+/* Channels are shutdown and reinitialised whilst the NIC is running
+ * to propagate configuration changes (mtu, checksum offload), or
+ * to clear hardware error conditions
+ */
+static void efx_init_channels(struct efx_nic *efx)
+{
+	struct efx_tx_queue *tx_queue;
+	struct efx_rx_queue *rx_queue;
+	struct efx_channel *channel;
+
+	/* Calculate the rx buffer allocation parameters required to
+	 * support the current MTU, including padding for header
+	 * alignment and overruns.
+	 */
+	efx->rx_buffer_len = (max(EFX_PAGE_IP_ALIGN, NET_IP_ALIGN) +
+			      EFX_MAX_FRAME_LEN(efx->net_dev->mtu) +
+			      efx->type->rx_buffer_hash_size +
+			      efx->type->rx_buffer_padding);
+	efx->rx_buffer_order = get_order(efx->rx_buffer_len +
+					 sizeof(struct efx_rx_page_state));
+
+	/* Initialise the channels */
+	efx_for_each_channel(channel, efx) {
+		netif_dbg(channel->efx, drv, channel->efx->net_dev,
+			  "init chan %d\n", channel->channel);
+
+		efx_init_eventq(channel);
+
+		efx_for_each_channel_tx_queue(tx_queue, channel)
+			efx_init_tx_queue(tx_queue);
+
+		/* The rx buffer allocation strategy is MTU dependent */
+		efx_rx_strategy(channel);
+
+		efx_for_each_channel_rx_queue(rx_queue, channel)
+			efx_init_rx_queue(rx_queue);
+
+		WARN_ON(channel->rx_pkt != NULL);
+		efx_rx_strategy(channel);
+	}
+}
+
+/* This enables event queue processing and packet transmission.
+ *
+ * Note that this function is not allowed to fail, since that would
+ * introduce too much complexity into the suspend/resume path.
+ */
+static void efx_start_channel(struct efx_channel *channel)
+{
+	struct efx_rx_queue *rx_queue;
+
+	netif_dbg(channel->efx, ifup, channel->efx->net_dev,
+		  "starting chan %d\n", channel->channel);
+
+	/* The interrupt handler for this channel may set work_pending
+	 * as soon as we enable it.  Make sure it's cleared before
+	 * then.  Similarly, make sure it sees the enabled flag set. */
+	channel->work_pending = false;
+	channel->enabled = true;
+	smp_wmb();
+
+	/* Fill the queues before enabling NAPI */
+	efx_for_each_channel_rx_queue(rx_queue, channel)
+		efx_fast_push_rx_descriptors(rx_queue);
+
+	napi_enable(&channel->napi_str);
+}
+
+/* This disables event queue processing and packet transmission.
+ * This function does not guarantee that all queue processing
+ * (e.g. RX refill) is complete.
+ */
+static void efx_stop_channel(struct efx_channel *channel)
+{
+	if (!channel->enabled)
+		return;
+
+	netif_dbg(channel->efx, ifdown, channel->efx->net_dev,
+		  "stop chan %d\n", channel->channel);
+
+	channel->enabled = false;
+	napi_disable(&channel->napi_str);
+}
+
+static void efx_fini_channels(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+	struct efx_tx_queue *tx_queue;
+	struct efx_rx_queue *rx_queue;
+	int rc;
+
+	EFX_ASSERT_RESET_SERIALISED(efx);
+	BUG_ON(efx->port_enabled);
+
+	rc = efx_nic_flush_queues(efx);
+	if (rc && EFX_WORKAROUND_7803(efx)) {
+		/* Schedule a reset to recover from the flush failure. The
+		 * descriptor caches reference memory we're about to free,
+		 * but falcon_reconfigure_mac_wrapper() won't reconnect
+		 * the MACs because of the pending reset. */
+		netif_err(efx, drv, efx->net_dev,
+			  "Resetting to recover from flush failure\n");
+		efx_schedule_reset(efx, RESET_TYPE_ALL);
+	} else if (rc) {
+		netif_err(efx, drv, efx->net_dev, "failed to flush queues\n");
+	} else {
+		netif_dbg(efx, drv, efx->net_dev,
+			  "successfully flushed all queues\n");
+	}
+
+	efx_for_each_channel(channel, efx) {
+		netif_dbg(channel->efx, drv, channel->efx->net_dev,
+			  "shut down chan %d\n", channel->channel);
+
+		efx_for_each_channel_rx_queue(rx_queue, channel)
+			efx_fini_rx_queue(rx_queue);
+		efx_for_each_possible_channel_tx_queue(tx_queue, channel)
+			efx_fini_tx_queue(tx_queue);
+		efx_fini_eventq(channel);
+	}
+}
+
+static void efx_remove_channel(struct efx_channel *channel)
+{
+	struct efx_tx_queue *tx_queue;
+	struct efx_rx_queue *rx_queue;
+
+	netif_dbg(channel->efx, drv, channel->efx->net_dev,
+		  "destroy chan %d\n", channel->channel);
+
+	efx_for_each_channel_rx_queue(rx_queue, channel)
+		efx_remove_rx_queue(rx_queue);
+	efx_for_each_possible_channel_tx_queue(tx_queue, channel)
+		efx_remove_tx_queue(tx_queue);
+	efx_remove_eventq(channel);
+}
+
+static void efx_remove_channels(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+
+	efx_for_each_channel(channel, efx)
+		efx_remove_channel(channel);
+}
+
+int
+efx_realloc_channels(struct efx_nic *efx, u32 rxq_entries, u32 txq_entries)
+{
+	struct efx_channel *other_channel[EFX_MAX_CHANNELS], *channel;
+	u32 old_rxq_entries, old_txq_entries;
+	unsigned i;
+	int rc;
+
+	efx_stop_all(efx);
+	efx_fini_channels(efx);
+
+	/* Clone channels */
+	memset(other_channel, 0, sizeof(other_channel));
+	for (i = 0; i < efx->n_channels; i++) {
+		channel = efx_alloc_channel(efx, i, efx->channel[i]);
+		if (!channel) {
+			rc = -ENOMEM;
+			goto out;
+		}
+		other_channel[i] = channel;
+	}
+
+	/* Swap entry counts and channel pointers */
+	old_rxq_entries = efx->rxq_entries;
+	old_txq_entries = efx->txq_entries;
+	efx->rxq_entries = rxq_entries;
+	efx->txq_entries = txq_entries;
+	for (i = 0; i < efx->n_channels; i++) {
+		channel = efx->channel[i];
+		efx->channel[i] = other_channel[i];
+		other_channel[i] = channel;
+	}
+
+	rc = efx_probe_channels(efx);
+	if (rc)
+		goto rollback;
+
+	efx_init_napi(efx);
+
+	/* Destroy old channels */
+	for (i = 0; i < efx->n_channels; i++) {
+		efx_fini_napi_channel(other_channel[i]);
+		efx_remove_channel(other_channel[i]);
+	}
+out:
+	/* Free unused channel structures */
+	for (i = 0; i < efx->n_channels; i++)
+		kfree(other_channel[i]);
+
+	efx_init_channels(efx);
+	efx_start_all(efx);
+	return rc;
+
+rollback:
+	/* Swap back */
+	efx->rxq_entries = old_rxq_entries;
+	efx->txq_entries = old_txq_entries;
+	for (i = 0; i < efx->n_channels; i++) {
+		channel = efx->channel[i];
+		efx->channel[i] = other_channel[i];
+		other_channel[i] = channel;
+	}
+	goto out;
+}
+
+void efx_schedule_slow_fill(struct efx_rx_queue *rx_queue)
+{
+	mod_timer(&rx_queue->slow_fill, jiffies + msecs_to_jiffies(100));
+}
+
+/**************************************************************************
+ *
+ * Port handling
+ *
+ **************************************************************************/
+
+/* This ensures that the kernel is kept informed (via
+ * netif_carrier_on/off) of the link status, and also maintains the
+ * link status's stop on the port's TX queue.
+ */
+void efx_link_status_changed(struct efx_nic *efx)
+{
+	struct efx_link_state *link_state = &efx->link_state;
+
+	/* SFC Bug 5356: A net_dev notifier is registered, so we must ensure
+	 * that no events are triggered between unregister_netdev() and the
+	 * driver unloading. A more general condition is that NETDEV_CHANGE
+	 * can only be generated between NETDEV_UP and NETDEV_DOWN */
+	if (!netif_running(efx->net_dev))
+		return;
+
+	if (link_state->up != netif_carrier_ok(efx->net_dev)) {
+		efx->n_link_state_changes++;
+
+		if (link_state->up)
+			netif_carrier_on(efx->net_dev);
+		else
+			netif_carrier_off(efx->net_dev);
+	}
+
+	/* Status message for kernel log */
+	if (link_state->up) {
+		netif_info(efx, link, efx->net_dev,
+			   "link up at %uMbps %s-duplex (MTU %d)%s\n",
+			   link_state->speed, link_state->fd ? "full" : "half",
+			   efx->net_dev->mtu,
+			   (efx->promiscuous ? " [PROMISC]" : ""));
+	} else {
+		netif_info(efx, link, efx->net_dev, "link down\n");
+	}
+
+}
+
+void efx_link_set_advertising(struct efx_nic *efx, u32 advertising)
+{
+	efx->link_advertising = advertising;
+	if (advertising) {
+		if (advertising & ADVERTISED_Pause)
+			efx->wanted_fc |= (EFX_FC_TX | EFX_FC_RX);
+		else
+			efx->wanted_fc &= ~(EFX_FC_TX | EFX_FC_RX);
+		if (advertising & ADVERTISED_Asym_Pause)
+			efx->wanted_fc ^= EFX_FC_TX;
+	}
+}
+
+void efx_link_set_wanted_fc(struct efx_nic *efx, u8 wanted_fc)
+{
+	efx->wanted_fc = wanted_fc;
+	if (efx->link_advertising) {
+		if (wanted_fc & EFX_FC_RX)
+			efx->link_advertising |= (ADVERTISED_Pause |
+						  ADVERTISED_Asym_Pause);
+		else
+			efx->link_advertising &= ~(ADVERTISED_Pause |
+						   ADVERTISED_Asym_Pause);
+		if (wanted_fc & EFX_FC_TX)
+			efx->link_advertising ^= ADVERTISED_Asym_Pause;
+	}
+}
+
+static void efx_fini_port(struct efx_nic *efx);
+
+/* Push loopback/power/transmit disable settings to the PHY, and reconfigure
+ * the MAC appropriately. All other PHY configuration changes are pushed
+ * through phy_op->set_settings(), and pushed asynchronously to the MAC
+ * through efx_monitor().
+ *
+ * Callers must hold the mac_lock
+ */
+int __efx_reconfigure_port(struct efx_nic *efx)
+{
+	enum efx_phy_mode phy_mode;
+	int rc;
+
+	WARN_ON(!mutex_is_locked(&efx->mac_lock));
+
+	/* Serialise the promiscuous flag with efx_set_multicast_list. */
+	if (efx_dev_registered(efx)) {
+		netif_addr_lock_bh(efx->net_dev);
+		netif_addr_unlock_bh(efx->net_dev);
+	}
+
+	/* Disable PHY transmit in mac level loopbacks */
+	phy_mode = efx->phy_mode;
+	if (LOOPBACK_INTERNAL(efx))
+		efx->phy_mode |= PHY_MODE_TX_DISABLED;
+	else
+		efx->phy_mode &= ~PHY_MODE_TX_DISABLED;
+
+	rc = efx->type->reconfigure_port(efx);
+
+	if (rc)
+		efx->phy_mode = phy_mode;
+
+	return rc;
+}
+
+/* Reinitialise the MAC to pick up new PHY settings, even if the port is
+ * disabled. */
+int efx_reconfigure_port(struct efx_nic *efx)
+{
+	int rc;
+
+	EFX_ASSERT_RESET_SERIALISED(efx);
+
+	mutex_lock(&efx->mac_lock);
+	rc = __efx_reconfigure_port(efx);
+	mutex_unlock(&efx->mac_lock);
+
+	return rc;
+}
+
+/* Asynchronous work item for changing MAC promiscuity and multicast
+ * hash.  Avoid a drain/rx_ingress enable by reconfiguring the current
+ * MAC directly. */
+static void efx_mac_work(struct work_struct *data)
+{
+	struct efx_nic *efx = container_of(data, struct efx_nic, mac_work);
+
+	mutex_lock(&efx->mac_lock);
+	if (efx->port_enabled) {
+		efx->type->push_multicast_hash(efx);
+		efx->mac_op->reconfigure(efx);
+	}
+	mutex_unlock(&efx->mac_lock);
+}
+
+static int efx_probe_port(struct efx_nic *efx)
+{
+	unsigned char *perm_addr;
+	int rc;
+
+	netif_dbg(efx, probe, efx->net_dev, "create port\n");
+
+	if (phy_flash_cfg)
+		efx->phy_mode = PHY_MODE_SPECIAL;
+
+	/* Connect up MAC/PHY operations table */
+	rc = efx->type->probe_port(efx);
+	if (rc)
+		return rc;
+
+	/* Sanity check MAC address */
+	perm_addr = efx->net_dev->perm_addr;
+	if (is_valid_ether_addr(perm_addr)) {
+		memcpy(efx->net_dev->dev_addr, perm_addr, ETH_ALEN);
+	} else {
+		netif_err(efx, probe, efx->net_dev, "invalid MAC address %pM\n",
+			  perm_addr);
+		if (!allow_bad_hwaddr) {
+			rc = -EINVAL;
+			goto err;
+		}
+		random_ether_addr(efx->net_dev->dev_addr);
+		netif_info(efx, probe, efx->net_dev,
+			   "using locally-generated MAC %pM\n",
+			   efx->net_dev->dev_addr);
+	}
+
+	return 0;
+
+ err:
+	efx->type->remove_port(efx);
+	return rc;
+}
+
+static int efx_init_port(struct efx_nic *efx)
+{
+	int rc;
+
+	netif_dbg(efx, drv, efx->net_dev, "init port\n");
+
+	mutex_lock(&efx->mac_lock);
+
+	rc = efx->phy_op->init(efx);
+	if (rc)
+		goto fail1;
+
+	efx->port_initialized = true;
+
+	/* Reconfigure the MAC before creating dma queues (required for
+	 * Falcon/A1 where RX_INGR_EN/TX_DRAIN_EN isn't supported) */
+	efx->mac_op->reconfigure(efx);
+
+	/* Ensure the PHY advertises the correct flow control settings */
+	rc = efx->phy_op->reconfigure(efx);
+	if (rc)
+		goto fail2;
+
+	mutex_unlock(&efx->mac_lock);
+	return 0;
+
+fail2:
+	efx->phy_op->fini(efx);
+fail1:
+	mutex_unlock(&efx->mac_lock);
+	return rc;
+}
+
+static void efx_start_port(struct efx_nic *efx)
+{
+	netif_dbg(efx, ifup, efx->net_dev, "start port\n");
+	BUG_ON(efx->port_enabled);
+
+	mutex_lock(&efx->mac_lock);
+	efx->port_enabled = true;
+
+	/* efx_mac_work() might have been scheduled after efx_stop_port(),
+	 * and then cancelled by efx_flush_all() */
+	efx->type->push_multicast_hash(efx);
+	efx->mac_op->reconfigure(efx);
+
+	mutex_unlock(&efx->mac_lock);
+}
+
+/* Prevent efx_mac_work() and efx_monitor() from working */
+static void efx_stop_port(struct efx_nic *efx)
+{
+	netif_dbg(efx, ifdown, efx->net_dev, "stop port\n");
+
+	mutex_lock(&efx->mac_lock);
+	efx->port_enabled = false;
+	mutex_unlock(&efx->mac_lock);
+
+	/* Serialise against efx_set_multicast_list() */
+	if (efx_dev_registered(efx)) {
+		netif_addr_lock_bh(efx->net_dev);
+		netif_addr_unlock_bh(efx->net_dev);
+	}
+}
+
+static void efx_fini_port(struct efx_nic *efx)
+{
+	netif_dbg(efx, drv, efx->net_dev, "shut down port\n");
+
+	if (!efx->port_initialized)
+		return;
+
+	efx->phy_op->fini(efx);
+	efx->port_initialized = false;
+
+	efx->link_state.up = false;
+	efx_link_status_changed(efx);
+}
+
+static void efx_remove_port(struct efx_nic *efx)
+{
+	netif_dbg(efx, drv, efx->net_dev, "destroying port\n");
+
+	efx->type->remove_port(efx);
+}
+
+/**************************************************************************
+ *
+ * NIC handling
+ *
+ **************************************************************************/
+
+/* This configures the PCI device to enable I/O and DMA. */
+static int efx_init_io(struct efx_nic *efx)
+{
+	struct pci_dev *pci_dev = efx->pci_dev;
+	dma_addr_t dma_mask = efx->type->max_dma_mask;
+	bool use_wc;
+	int rc;
+
+	netif_dbg(efx, probe, efx->net_dev, "initialising I/O\n");
+
+	rc = pci_enable_device(pci_dev);
+	if (rc) {
+		netif_err(efx, probe, efx->net_dev,
+			  "failed to enable PCI device\n");
+		goto fail1;
+	}
+
+	pci_set_master(pci_dev);
+
+	/* Set the PCI DMA mask.  Try all possibilities from our
+	 * genuine mask down to 32 bits, because some architectures
+	 * (e.g. x86_64 with iommu_sac_force set) will allow 40 bit
+	 * masks event though they reject 46 bit masks.
+	 */
+	while (dma_mask > 0x7fffffffUL) {
+		if (pci_dma_supported(pci_dev, dma_mask) &&
+		    ((rc = pci_set_dma_mask(pci_dev, dma_mask)) == 0))
+			break;
+		dma_mask >>= 1;
+	}
+	if (rc) {
+		netif_err(efx, probe, efx->net_dev,
+			  "could not find a suitable DMA mask\n");
+		goto fail2;
+	}
+	netif_dbg(efx, probe, efx->net_dev,
+		  "using DMA mask %llx\n", (unsigned long long) dma_mask);
+	rc = pci_set_consistent_dma_mask(pci_dev, dma_mask);
+	if (rc) {
+		/* pci_set_consistent_dma_mask() is not *allowed* to
+		 * fail with a mask that pci_set_dma_mask() accepted,
+		 * but just in case...
+		 */
+		netif_err(efx, probe, efx->net_dev,
+			  "failed to set consistent DMA mask\n");
+		goto fail2;
+	}
+
+	efx->membase_phys = pci_resource_start(efx->pci_dev, EFX_MEM_BAR);
+	rc = pci_request_region(pci_dev, EFX_MEM_BAR, "sfc");
+	if (rc) {
+		netif_err(efx, probe, efx->net_dev,
+			  "request for memory BAR failed\n");
+		rc = -EIO;
+		goto fail3;
+	}
+
+	/* bug22643: If SR-IOV is enabled then tx push over a write combined
+	 * mapping is unsafe. We need to disable write combining in this case.
+	 * MSI is unsupported when SR-IOV is enabled, and the firmware will
+	 * have removed the MSI capability. So write combining is safe if
+	 * there is an MSI capability.
+	 */
+	use_wc = (!EFX_WORKAROUND_22643(efx) ||
+		  pci_find_capability(pci_dev, PCI_CAP_ID_MSI));
+	if (use_wc)
+		efx->membase = ioremap_wc(efx->membase_phys,
+					  efx->type->mem_map_size);
+	else
+		efx->membase = ioremap_nocache(efx->membase_phys,
+					       efx->type->mem_map_size);
+	if (!efx->membase) {
+		netif_err(efx, probe, efx->net_dev,
+			  "could not map memory BAR at %llx+%x\n",
+			  (unsigned long long)efx->membase_phys,
+			  efx->type->mem_map_size);
+		rc = -ENOMEM;
+		goto fail4;
+	}
+	netif_dbg(efx, probe, efx->net_dev,
+		  "memory BAR at %llx+%x (virtual %p)\n",
+		  (unsigned long long)efx->membase_phys,
+		  efx->type->mem_map_size, efx->membase);
+
+	return 0;
+
+ fail4:
+	pci_release_region(efx->pci_dev, EFX_MEM_BAR);
+ fail3:
+	efx->membase_phys = 0;
+ fail2:
+	pci_disable_device(efx->pci_dev);
+ fail1:
+	return rc;
+}
+
+static void efx_fini_io(struct efx_nic *efx)
+{
+	netif_dbg(efx, drv, efx->net_dev, "shutting down I/O\n");
+
+	if (efx->membase) {
+		iounmap(efx->membase);
+		efx->membase = NULL;
+	}
+
+	if (efx->membase_phys) {
+		pci_release_region(efx->pci_dev, EFX_MEM_BAR);
+		efx->membase_phys = 0;
+	}
+
+	pci_disable_device(efx->pci_dev);
+}
+
+/* Get number of channels wanted.  Each channel will have its own IRQ,
+ * 1 RX queue and/or 2 TX queues. */
+static int efx_wanted_channels(void)
+{
+	cpumask_var_t core_mask;
+	int count;
+	int cpu;
+
+	if (rss_cpus)
+		return rss_cpus;
+
+	if (unlikely(!zalloc_cpumask_var(&core_mask, GFP_KERNEL))) {
+		printk(KERN_WARNING
+		       "sfc: RSS disabled due to allocation failure\n");
+		return 1;
+	}
+
+	count = 0;
+	for_each_online_cpu(cpu) {
+		if (!cpumask_test_cpu(cpu, core_mask)) {
+			++count;
+			cpumask_or(core_mask, core_mask,
+				   topology_core_cpumask(cpu));
+		}
+	}
+
+	free_cpumask_var(core_mask);
+	return count;
+}
+
+static int
+efx_init_rx_cpu_rmap(struct efx_nic *efx, struct msix_entry *xentries)
+{
+#ifdef CONFIG_RFS_ACCEL
+	int i, rc;
+
+	efx->net_dev->rx_cpu_rmap = alloc_irq_cpu_rmap(efx->n_rx_channels);
+	if (!efx->net_dev->rx_cpu_rmap)
+		return -ENOMEM;
+	for (i = 0; i < efx->n_rx_channels; i++) {
+		rc = irq_cpu_rmap_add(efx->net_dev->rx_cpu_rmap,
+				      xentries[i].vector);
+		if (rc) {
+			free_irq_cpu_rmap(efx->net_dev->rx_cpu_rmap);
+			efx->net_dev->rx_cpu_rmap = NULL;
+			return rc;
+		}
+	}
+#endif
+	return 0;
+}
+
+/* Probe the number and type of interrupts we are able to obtain, and
+ * the resulting numbers of channels and RX queues.
+ */
+static int efx_probe_interrupts(struct efx_nic *efx)
+{
+	int max_channels =
+		min_t(int, efx->type->phys_addr_channels, EFX_MAX_CHANNELS);
+	int rc, i;
+
+	if (efx->interrupt_mode == EFX_INT_MODE_MSIX) {
+		struct msix_entry xentries[EFX_MAX_CHANNELS];
+		int n_channels;
+
+		n_channels = efx_wanted_channels();
+		if (separate_tx_channels)
+			n_channels *= 2;
+		n_channels = min(n_channels, max_channels);
+
+		for (i = 0; i < n_channels; i++)
+			xentries[i].entry = i;
+		rc = pci_enable_msix(efx->pci_dev, xentries, n_channels);
+		if (rc > 0) {
+			netif_err(efx, drv, efx->net_dev,
+				  "WARNING: Insufficient MSI-X vectors"
+				  " available (%d < %d).\n", rc, n_channels);
+			netif_err(efx, drv, efx->net_dev,
+				  "WARNING: Performance may be reduced.\n");
+			EFX_BUG_ON_PARANOID(rc >= n_channels);
+			n_channels = rc;
+			rc = pci_enable_msix(efx->pci_dev, xentries,
+					     n_channels);
+		}
+
+		if (rc == 0) {
+			efx->n_channels = n_channels;
+			if (separate_tx_channels) {
+				efx->n_tx_channels =
+					max(efx->n_channels / 2, 1U);
+				efx->n_rx_channels =
+					max(efx->n_channels -
+					    efx->n_tx_channels, 1U);
+			} else {
+				efx->n_tx_channels = efx->n_channels;
+				efx->n_rx_channels = efx->n_channels;
+			}
+			rc = efx_init_rx_cpu_rmap(efx, xentries);
+			if (rc) {
+				pci_disable_msix(efx->pci_dev);
+				return rc;
+			}
+			for (i = 0; i < n_channels; i++)
+				efx_get_channel(efx, i)->irq =
+					xentries[i].vector;
+		} else {
+			/* Fall back to single channel MSI */
+			efx->interrupt_mode = EFX_INT_MODE_MSI;
+			netif_err(efx, drv, efx->net_dev,
+				  "could not enable MSI-X\n");
+		}
+	}
+
+	/* Try single interrupt MSI */
+	if (efx->interrupt_mode == EFX_INT_MODE_MSI) {
+		efx->n_channels = 1;
+		efx->n_rx_channels = 1;
+		efx->n_tx_channels = 1;
+		rc = pci_enable_msi(efx->pci_dev);
+		if (rc == 0) {
+			efx_get_channel(efx, 0)->irq = efx->pci_dev->irq;
+		} else {
+			netif_err(efx, drv, efx->net_dev,
+				  "could not enable MSI\n");
+			efx->interrupt_mode = EFX_INT_MODE_LEGACY;
+		}
+	}
+
+	/* Assume legacy interrupts */
+	if (efx->interrupt_mode == EFX_INT_MODE_LEGACY) {
+		efx->n_channels = 1 + (separate_tx_channels ? 1 : 0);
+		efx->n_rx_channels = 1;
+		efx->n_tx_channels = 1;
+		efx->legacy_irq = efx->pci_dev->irq;
+	}
+
+	return 0;
+}
+
+static void efx_remove_interrupts(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+
+	/* Remove MSI/MSI-X interrupts */
+	efx_for_each_channel(channel, efx)
+		channel->irq = 0;
+	pci_disable_msi(efx->pci_dev);
+	pci_disable_msix(efx->pci_dev);
+
+	/* Remove legacy interrupt */
+	efx->legacy_irq = 0;
+}
+
+static void efx_set_channels(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+	struct efx_tx_queue *tx_queue;
+
+	efx->tx_channel_offset =
+		separate_tx_channels ? efx->n_channels - efx->n_tx_channels : 0;
+
+	/* We need to adjust the TX queue numbers if we have separate
+	 * RX-only and TX-only channels.
+	 */
+	efx_for_each_channel(channel, efx) {
+		efx_for_each_channel_tx_queue(tx_queue, channel)
+			tx_queue->queue -= (efx->tx_channel_offset *
+					    EFX_TXQ_TYPES);
+	}
+}
+
+static int efx_probe_nic(struct efx_nic *efx)
+{
+	size_t i;
+	int rc;
+
+	netif_dbg(efx, probe, efx->net_dev, "creating NIC\n");
+
+	/* Carry out hardware-type specific initialisation */
+	rc = efx->type->probe(efx);
+	if (rc)
+		return rc;
+
+	/* Determine the number of channels and queues by trying to hook
+	 * in MSI-X interrupts. */
+	rc = efx_probe_interrupts(efx);
+	if (rc)
+		goto fail;
+
+	if (efx->n_channels > 1)
+		get_random_bytes(&efx->rx_hash_key, sizeof(efx->rx_hash_key));
+	for (i = 0; i < ARRAY_SIZE(efx->rx_indir_table); i++)
+		efx->rx_indir_table[i] = i % efx->n_rx_channels;
+
+	efx_set_channels(efx);
+	netif_set_real_num_tx_queues(efx->net_dev, efx->n_tx_channels);
+	netif_set_real_num_rx_queues(efx->net_dev, efx->n_rx_channels);
+
+	/* Initialise the interrupt moderation settings */
+	efx_init_irq_moderation(efx, tx_irq_mod_usec, rx_irq_mod_usec, true);
+
+	return 0;
+
+fail:
+	efx->type->remove(efx);
+	return rc;
+}
+
+static void efx_remove_nic(struct efx_nic *efx)
+{
+	netif_dbg(efx, drv, efx->net_dev, "destroying NIC\n");
+
+	efx_remove_interrupts(efx);
+	efx->type->remove(efx);
+}
+
+/**************************************************************************
+ *
+ * NIC startup/shutdown
+ *
+ *************************************************************************/
+
+static int efx_probe_all(struct efx_nic *efx)
+{
+	int rc;
+
+	rc = efx_probe_nic(efx);
+	if (rc) {
+		netif_err(efx, probe, efx->net_dev, "failed to create NIC\n");
+		goto fail1;
+	}
+
+	rc = efx_probe_port(efx);
+	if (rc) {
+		netif_err(efx, probe, efx->net_dev, "failed to create port\n");
+		goto fail2;
+	}
+
+	efx->rxq_entries = efx->txq_entries = EFX_DEFAULT_DMAQ_SIZE;
+	rc = efx_probe_channels(efx);
+	if (rc)
+		goto fail3;
+
+	rc = efx_probe_filters(efx);
+	if (rc) {
+		netif_err(efx, probe, efx->net_dev,
+			  "failed to create filter tables\n");
+		goto fail4;
+	}
+
+	return 0;
+
+ fail4:
+	efx_remove_channels(efx);
+ fail3:
+	efx_remove_port(efx);
+ fail2:
+	efx_remove_nic(efx);
+ fail1:
+	return rc;
+}
+
+/* Called after previous invocation(s) of efx_stop_all, restarts the
+ * port, kernel transmit queue, NAPI processing and hardware interrupts,
+ * and ensures that the port is scheduled to be reconfigured.
+ * This function is safe to call multiple times when the NIC is in any
+ * state. */
+static void efx_start_all(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+
+	EFX_ASSERT_RESET_SERIALISED(efx);
+
+	/* Check that it is appropriate to restart the interface. All
+	 * of these flags are safe to read under just the rtnl lock */
+	if (efx->port_enabled)
+		return;
+	if ((efx->state != STATE_RUNNING) && (efx->state != STATE_INIT))
+		return;
+	if (efx_dev_registered(efx) && !netif_running(efx->net_dev))
+		return;
+
+	/* Mark the port as enabled so port reconfigurations can start, then
+	 * restart the transmit interface early so the watchdog timer stops */
+	efx_start_port(efx);
+
+	if (efx_dev_registered(efx) && netif_device_present(efx->net_dev))
+		netif_tx_wake_all_queues(efx->net_dev);
+
+	efx_for_each_channel(channel, efx)
+		efx_start_channel(channel);
+
+	if (efx->legacy_irq)
+		efx->legacy_irq_enabled = true;
+	efx_nic_enable_interrupts(efx);
+
+	/* Switch to event based MCDI completions after enabling interrupts.
+	 * If a reset has been scheduled, then we need to stay in polled mode.
+	 * Rather than serialising efx_mcdi_mode_event() [which sleeps] and
+	 * reset_pending [modified from an atomic context], we instead guarantee
+	 * that efx_mcdi_mode_poll() isn't reverted erroneously */
+	efx_mcdi_mode_event(efx);
+	if (efx->reset_pending)
+		efx_mcdi_mode_poll(efx);
+
+	/* Start the hardware monitor if there is one. Otherwise (we're link
+	 * event driven), we have to poll the PHY because after an event queue
+	 * flush, we could have a missed a link state change */
+	if (efx->type->monitor != NULL) {
+		queue_delayed_work(efx->workqueue, &efx->monitor_work,
+				   efx_monitor_interval);
+	} else {
+		mutex_lock(&efx->mac_lock);
+		if (efx->phy_op->poll(efx))
+			efx_link_status_changed(efx);
+		mutex_unlock(&efx->mac_lock);
+	}
+
+	efx->type->start_stats(efx);
+}
+
+/* Flush all delayed work. Should only be called when no more delayed work
+ * will be scheduled. This doesn't flush pending online resets (efx_reset),
+ * since we're holding the rtnl_lock at this point. */
+static void efx_flush_all(struct efx_nic *efx)
+{
+	/* Make sure the hardware monitor is stopped */
+	cancel_delayed_work_sync(&efx->monitor_work);
+	/* Stop scheduled port reconfigurations */
+	cancel_work_sync(&efx->mac_work);
+}
+
+/* Quiesce hardware and software without bringing the link down.
+ * Safe to call multiple times, when the nic and interface is in any
+ * state. The caller is guaranteed to subsequently be in a position
+ * to modify any hardware and software state they see fit without
+ * taking locks. */
+static void efx_stop_all(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+
+	EFX_ASSERT_RESET_SERIALISED(efx);
+
+	/* port_enabled can be read safely under the rtnl lock */
+	if (!efx->port_enabled)
+		return;
+
+	efx->type->stop_stats(efx);
+
+	/* Switch to MCDI polling on Siena before disabling interrupts */
+	efx_mcdi_mode_poll(efx);
+
+	/* Disable interrupts and wait for ISR to complete */
+	efx_nic_disable_interrupts(efx);
+	if (efx->legacy_irq) {
+		synchronize_irq(efx->legacy_irq);
+		efx->legacy_irq_enabled = false;
+	}
+	efx_for_each_channel(channel, efx) {
+		if (channel->irq)
+			synchronize_irq(channel->irq);
+	}
+
+	/* Stop all NAPI processing and synchronous rx refills */
+	efx_for_each_channel(channel, efx)
+		efx_stop_channel(channel);
+
+	/* Stop all asynchronous port reconfigurations. Since all
+	 * event processing has already been stopped, there is no
+	 * window to loose phy events */
+	efx_stop_port(efx);
+
+	/* Flush efx_mac_work(), refill_workqueue, monitor_work */
+	efx_flush_all(efx);
+
+	/* Stop the kernel transmit interface late, so the watchdog
+	 * timer isn't ticking over the flush */
+	if (efx_dev_registered(efx)) {
+		netif_tx_stop_all_queues(efx->net_dev);
+		netif_tx_lock_bh(efx->net_dev);
+		netif_tx_unlock_bh(efx->net_dev);
+	}
+}
+
+static void efx_remove_all(struct efx_nic *efx)
+{
+	efx_remove_filters(efx);
+	efx_remove_channels(efx);
+	efx_remove_port(efx);
+	efx_remove_nic(efx);
+}
+
+/**************************************************************************
+ *
+ * Interrupt moderation
+ *
+ **************************************************************************/
+
+static unsigned irq_mod_ticks(int usecs, int resolution)
+{
+	if (usecs <= 0)
+		return 0; /* cannot receive interrupts ahead of time :-) */
+	if (usecs < resolution)
+		return 1; /* never round down to 0 */
+	return usecs / resolution;
+}
+
+/* Set interrupt moderation parameters */
+void efx_init_irq_moderation(struct efx_nic *efx, int tx_usecs, int rx_usecs,
+			     bool rx_adaptive)
+{
+	struct efx_channel *channel;
+	unsigned tx_ticks = irq_mod_ticks(tx_usecs, EFX_IRQ_MOD_RESOLUTION);
+	unsigned rx_ticks = irq_mod_ticks(rx_usecs, EFX_IRQ_MOD_RESOLUTION);
+
+	EFX_ASSERT_RESET_SERIALISED(efx);
+
+	efx->irq_rx_adaptive = rx_adaptive;
+	efx->irq_rx_moderation = rx_ticks;
+	efx_for_each_channel(channel, efx) {
+		if (efx_channel_has_rx_queue(channel))
+			channel->irq_moderation = rx_ticks;
+		else if (efx_channel_has_tx_queues(channel))
+			channel->irq_moderation = tx_ticks;
+	}
+}
+
+/**************************************************************************
+ *
+ * Hardware monitor
+ *
+ **************************************************************************/
+
+/* Run periodically off the general workqueue */
+static void efx_monitor(struct work_struct *data)
+{
+	struct efx_nic *efx = container_of(data, struct efx_nic,
+					   monitor_work.work);
+
+	netif_vdbg(efx, timer, efx->net_dev,
+		   "hardware monitor executing on CPU %d\n",
+		   raw_smp_processor_id());
+	BUG_ON(efx->type->monitor == NULL);
+
+	/* If the mac_lock is already held then it is likely a port
+	 * reconfiguration is already in place, which will likely do
+	 * most of the work of monitor() anyway. */
+	if (mutex_trylock(&efx->mac_lock)) {
+		if (efx->port_enabled)
+			efx->type->monitor(efx);
+		mutex_unlock(&efx->mac_lock);
+	}
+
+	queue_delayed_work(efx->workqueue, &efx->monitor_work,
+			   efx_monitor_interval);
+}
+
+/**************************************************************************
+ *
+ * ioctls
+ *
+ *************************************************************************/
+
+/* Net device ioctl
+ * Context: process, rtnl_lock() held.
+ */
+static int efx_ioctl(struct net_device *net_dev, struct ifreq *ifr, int cmd)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	struct mii_ioctl_data *data = if_mii(ifr);
+
+	EFX_ASSERT_RESET_SERIALISED(efx);
+
+	/* Convert phy_id from older PRTAD/DEVAD format */
+	if ((cmd == SIOCGMIIREG || cmd == SIOCSMIIREG) &&
+	    (data->phy_id & 0xfc00) == 0x0400)
+		data->phy_id ^= MDIO_PHY_ID_C45 | 0x0400;
+
+	return mdio_mii_ioctl(&efx->mdio, data, cmd);
+}
+
+/**************************************************************************
+ *
+ * NAPI interface
+ *
+ **************************************************************************/
+
+static void efx_init_napi(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+
+	efx_for_each_channel(channel, efx) {
+		channel->napi_dev = efx->net_dev;
+		netif_napi_add(channel->napi_dev, &channel->napi_str,
+			       efx_poll, napi_weight);
+	}
+}
+
+static void efx_fini_napi_channel(struct efx_channel *channel)
+{
+	if (channel->napi_dev)
+		netif_napi_del(&channel->napi_str);
+	channel->napi_dev = NULL;
+}
+
+static void efx_fini_napi(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+
+	efx_for_each_channel(channel, efx)
+		efx_fini_napi_channel(channel);
+}
+
+/**************************************************************************
+ *
+ * Kernel netpoll interface
+ *
+ *************************************************************************/
+
+#ifdef CONFIG_NET_POLL_CONTROLLER
+
+/* Although in the common case interrupts will be disabled, this is not
+ * guaranteed. However, all our work happens inside the NAPI callback,
+ * so no locking is required.
+ */
+static void efx_netpoll(struct net_device *net_dev)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	struct efx_channel *channel;
+
+	efx_for_each_channel(channel, efx)
+		efx_schedule_channel(channel);
+}
+
+#endif
+
+/**************************************************************************
+ *
+ * Kernel net device interface
+ *
+ *************************************************************************/
+
+/* Context: process, rtnl_lock() held. */
+static int efx_net_open(struct net_device *net_dev)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	EFX_ASSERT_RESET_SERIALISED(efx);
+
+	netif_dbg(efx, ifup, efx->net_dev, "opening device on CPU %d\n",
+		  raw_smp_processor_id());
+
+	if (efx->state == STATE_DISABLED)
+		return -EIO;
+	if (efx->phy_mode & PHY_MODE_SPECIAL)
+		return -EBUSY;
+	if (efx_mcdi_poll_reboot(efx) && efx_reset(efx, RESET_TYPE_ALL))
+		return -EIO;
+
+	/* Notify the kernel of the link state polled during driver load,
+	 * before the monitor starts running */
+	efx_link_status_changed(efx);
+
+	efx_start_all(efx);
+	return 0;
+}
+
+/* Context: process, rtnl_lock() held.
+ * Note that the kernel will ignore our return code; this method
+ * should really be a void.
+ */
+static int efx_net_stop(struct net_device *net_dev)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+
+	netif_dbg(efx, ifdown, efx->net_dev, "closing on CPU %d\n",
+		  raw_smp_processor_id());
+
+	if (efx->state != STATE_DISABLED) {
+		/* Stop the device and flush all the channels */
+		efx_stop_all(efx);
+		efx_fini_channels(efx);
+		efx_init_channels(efx);
+	}
+
+	return 0;
+}
+
+/* Context: process, dev_base_lock or RTNL held, non-blocking. */
+static struct rtnl_link_stats64 *efx_net_stats(struct net_device *net_dev, struct rtnl_link_stats64 *stats)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	struct efx_mac_stats *mac_stats = &efx->mac_stats;
+
+	spin_lock_bh(&efx->stats_lock);
+	efx->type->update_stats(efx);
+	spin_unlock_bh(&efx->stats_lock);
+
+	stats->rx_packets = mac_stats->rx_packets;
+	stats->tx_packets = mac_stats->tx_packets;
+	stats->rx_bytes = mac_stats->rx_bytes;
+	stats->tx_bytes = mac_stats->tx_bytes;
+	stats->rx_dropped = efx->n_rx_nodesc_drop_cnt;
+	stats->multicast = mac_stats->rx_multicast;
+	stats->collisions = mac_stats->tx_collision;
+	stats->rx_length_errors = (mac_stats->rx_gtjumbo +
+				   mac_stats->rx_length_error);
+	stats->rx_crc_errors = mac_stats->rx_bad;
+	stats->rx_frame_errors = mac_stats->rx_align_error;
+	stats->rx_fifo_errors = mac_stats->rx_overflow;
+	stats->rx_missed_errors = mac_stats->rx_missed;
+	stats->tx_window_errors = mac_stats->tx_late_collision;
+
+	stats->rx_errors = (stats->rx_length_errors +
+			    stats->rx_crc_errors +
+			    stats->rx_frame_errors +
+			    mac_stats->rx_symbol_error);
+	stats->tx_errors = (stats->tx_window_errors +
+			    mac_stats->tx_bad);
+
+	return stats;
+}
+
+/* Context: netif_tx_lock held, BHs disabled. */
+static void efx_watchdog(struct net_device *net_dev)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+
+	netif_err(efx, tx_err, efx->net_dev,
+		  "TX stuck with port_enabled=%d: resetting channels\n",
+		  efx->port_enabled);
+
+	efx_schedule_reset(efx, RESET_TYPE_TX_WATCHDOG);
+}
+
+
+/* Context: process, rtnl_lock() held. */
+static int efx_change_mtu(struct net_device *net_dev, int new_mtu)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	int rc = 0;
+
+	EFX_ASSERT_RESET_SERIALISED(efx);
+
+	if (new_mtu > EFX_MAX_MTU)
+		return -EINVAL;
+
+	efx_stop_all(efx);
+
+	netif_dbg(efx, drv, efx->net_dev, "changing MTU to %d\n", new_mtu);
+
+	efx_fini_channels(efx);
+
+	mutex_lock(&efx->mac_lock);
+	/* Reconfigure the MAC before enabling the dma queues so that
+	 * the RX buffers don't overflow */
+	net_dev->mtu = new_mtu;
+	efx->mac_op->reconfigure(efx);
+	mutex_unlock(&efx->mac_lock);
+
+	efx_init_channels(efx);
+
+	efx_start_all(efx);
+	return rc;
+}
+
+static int efx_set_mac_address(struct net_device *net_dev, void *data)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	struct sockaddr *addr = data;
+	char *new_addr = addr->sa_data;
+
+	EFX_ASSERT_RESET_SERIALISED(efx);
+
+	if (!is_valid_ether_addr(new_addr)) {
+		netif_err(efx, drv, efx->net_dev,
+			  "invalid ethernet MAC address requested: %pM\n",
+			  new_addr);
+		return -EINVAL;
+	}
+
+	memcpy(net_dev->dev_addr, new_addr, net_dev->addr_len);
+
+	/* Reconfigure the MAC */
+	mutex_lock(&efx->mac_lock);
+	efx->mac_op->reconfigure(efx);
+	mutex_unlock(&efx->mac_lock);
+
+	return 0;
+}
+
+/* Context: netif_addr_lock held, BHs disabled. */
+static void efx_set_multicast_list(struct net_device *net_dev)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	struct netdev_hw_addr *ha;
+	union efx_multicast_hash *mc_hash = &efx->multicast_hash;
+	u32 crc;
+	int bit;
+
+	efx->promiscuous = !!(net_dev->flags & IFF_PROMISC);
+
+	/* Build multicast hash table */
+	if (efx->promiscuous || (net_dev->flags & IFF_ALLMULTI)) {
+		memset(mc_hash, 0xff, sizeof(*mc_hash));
+	} else {
+		memset(mc_hash, 0x00, sizeof(*mc_hash));
+		netdev_for_each_mc_addr(ha, net_dev) {
+			crc = ether_crc_le(ETH_ALEN, ha->addr);
+			bit = crc & (EFX_MCAST_HASH_ENTRIES - 1);
+			set_bit_le(bit, mc_hash->byte);
+		}
+
+		/* Broadcast packets go through the multicast hash filter.
+		 * ether_crc_le() of the broadcast address is 0xbe2612ff
+		 * so we always add bit 0xff to the mask.
+		 */
+		set_bit_le(0xff, mc_hash->byte);
+	}
+
+	if (efx->port_enabled)
+		queue_work(efx->workqueue, &efx->mac_work);
+	/* Otherwise efx_start_port() will do this */
+}
+
+static int efx_set_features(struct net_device *net_dev, u32 data)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+
+	/* If disabling RX n-tuple filtering, clear existing filters */
+	if (net_dev->features & ~data & NETIF_F_NTUPLE)
+		efx_filter_clear_rx(efx, EFX_FILTER_PRI_MANUAL);
+
+	return 0;
+}
+
+static const struct net_device_ops efx_netdev_ops = {
+	.ndo_open		= efx_net_open,
+	.ndo_stop		= efx_net_stop,
+	.ndo_get_stats64	= efx_net_stats,
+	.ndo_tx_timeout		= efx_watchdog,
+	.ndo_start_xmit		= efx_hard_start_xmit,
+	.ndo_validate_addr	= eth_validate_addr,
+	.ndo_do_ioctl		= efx_ioctl,
+	.ndo_change_mtu		= efx_change_mtu,
+	.ndo_set_mac_address	= efx_set_mac_address,
+	.ndo_set_multicast_list = efx_set_multicast_list,
+	.ndo_set_features	= efx_set_features,
+#ifdef CONFIG_NET_POLL_CONTROLLER
+	.ndo_poll_controller = efx_netpoll,
+#endif
+	.ndo_setup_tc		= efx_setup_tc,
+#ifdef CONFIG_RFS_ACCEL
+	.ndo_rx_flow_steer	= efx_filter_rfs,
+#endif
+};
+
+static void efx_update_name(struct efx_nic *efx)
+{
+	strcpy(efx->name, efx->net_dev->name);
+	efx_mtd_rename(efx);
+	efx_set_channel_names(efx);
+}
+
+static int efx_netdev_event(struct notifier_block *this,
+			    unsigned long event, void *ptr)
+{
+	struct net_device *net_dev = ptr;
+
+	if (net_dev->netdev_ops == &efx_netdev_ops &&
+	    event == NETDEV_CHANGENAME)
+		efx_update_name(netdev_priv(net_dev));
+
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block efx_netdev_notifier = {
+	.notifier_call = efx_netdev_event,
+};
+
+static ssize_t
+show_phy_type(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct efx_nic *efx = pci_get_drvdata(to_pci_dev(dev));
+	return sprintf(buf, "%d\n", efx->phy_type);
+}
+static DEVICE_ATTR(phy_type, 0644, show_phy_type, NULL);
+
+static int efx_register_netdev(struct efx_nic *efx)
+{
+	struct net_device *net_dev = efx->net_dev;
+	struct efx_channel *channel;
+	int rc;
+
+	net_dev->watchdog_timeo = 5 * HZ;
+	net_dev->irq = efx->pci_dev->irq;
+	net_dev->netdev_ops = &efx_netdev_ops;
+	SET_ETHTOOL_OPS(net_dev, &efx_ethtool_ops);
+
+	/* Clear MAC statistics */
+	efx->mac_op->update_stats(efx);
+	memset(&efx->mac_stats, 0, sizeof(efx->mac_stats));
+
+	rtnl_lock();
+
+	rc = dev_alloc_name(net_dev, net_dev->name);
+	if (rc < 0)
+		goto fail_locked;
+	efx_update_name(efx);
+
+	rc = register_netdevice(net_dev);
+	if (rc)
+		goto fail_locked;
+
+	efx_for_each_channel(channel, efx) {
+		struct efx_tx_queue *tx_queue;
+		efx_for_each_channel_tx_queue(tx_queue, channel)
+			efx_init_tx_queue_core_txq(tx_queue);
+	}
+
+	/* Always start with carrier off; PHY events will detect the link */
+	netif_carrier_off(efx->net_dev);
+
+	rtnl_unlock();
+
+	rc = device_create_file(&efx->pci_dev->dev, &dev_attr_phy_type);
+	if (rc) {
+		netif_err(efx, drv, efx->net_dev,
+			  "failed to init net dev attributes\n");
+		goto fail_registered;
+	}
+
+	return 0;
+
+fail_locked:
+	rtnl_unlock();
+	netif_err(efx, drv, efx->net_dev, "could not register net dev\n");
+	return rc;
+
+fail_registered:
+	unregister_netdev(net_dev);
+	return rc;
+}
+
+static void efx_unregister_netdev(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+	struct efx_tx_queue *tx_queue;
+
+	if (!efx->net_dev)
+		return;
+
+	BUG_ON(netdev_priv(efx->net_dev) != efx);
+
+	/* Free up any skbs still remaining. This has to happen before
+	 * we try to unregister the netdev as running their destructors
+	 * may be needed to get the device ref. count to 0. */
+	efx_for_each_channel(channel, efx) {
+		efx_for_each_channel_tx_queue(tx_queue, channel)
+			efx_release_tx_buffers(tx_queue);
+	}
+
+	if (efx_dev_registered(efx)) {
+		strlcpy(efx->name, pci_name(efx->pci_dev), sizeof(efx->name));
+		device_remove_file(&efx->pci_dev->dev, &dev_attr_phy_type);
+		unregister_netdev(efx->net_dev);
+	}
+}
+
+/**************************************************************************
+ *
+ * Device reset and suspend
+ *
+ **************************************************************************/
+
+/* Tears down the entire software state and most of the hardware state
+ * before reset.  */
+void efx_reset_down(struct efx_nic *efx, enum reset_type method)
+{
+	EFX_ASSERT_RESET_SERIALISED(efx);
+
+	efx_stop_all(efx);
+	mutex_lock(&efx->mac_lock);
+
+	efx_fini_channels(efx);
+	if (efx->port_initialized && method != RESET_TYPE_INVISIBLE)
+		efx->phy_op->fini(efx);
+	efx->type->fini(efx);
+}
+
+/* This function will always ensure that the locks acquired in
+ * efx_reset_down() are released. A failure return code indicates
+ * that we were unable to reinitialise the hardware, and the
+ * driver should be disabled. If ok is false, then the rx and tx
+ * engines are not restarted, pending a RESET_DISABLE. */
+int efx_reset_up(struct efx_nic *efx, enum reset_type method, bool ok)
+{
+	int rc;
+
+	EFX_ASSERT_RESET_SERIALISED(efx);
+
+	rc = efx->type->init(efx);
+	if (rc) {
+		netif_err(efx, drv, efx->net_dev, "failed to initialise NIC\n");
+		goto fail;
+	}
+
+	if (!ok)
+		goto fail;
+
+	if (efx->port_initialized && method != RESET_TYPE_INVISIBLE) {
+		rc = efx->phy_op->init(efx);
+		if (rc)
+			goto fail;
+		if (efx->phy_op->reconfigure(efx))
+			netif_err(efx, drv, efx->net_dev,
+				  "could not restore PHY settings\n");
+	}
+
+	efx->mac_op->reconfigure(efx);
+
+	efx_init_channels(efx);
+	efx_restore_filters(efx);
+
+	mutex_unlock(&efx->mac_lock);
+
+	efx_start_all(efx);
+
+	return 0;
+
+fail:
+	efx->port_initialized = false;
+
+	mutex_unlock(&efx->mac_lock);
+
+	return rc;
+}
+
+/* Reset the NIC using the specified method.  Note that the reset may
+ * fail, in which case the card will be left in an unusable state.
+ *
+ * Caller must hold the rtnl_lock.
+ */
+int efx_reset(struct efx_nic *efx, enum reset_type method)
+{
+	int rc, rc2;
+	bool disabled;
+
+	netif_info(efx, drv, efx->net_dev, "resetting (%s)\n",
+		   RESET_TYPE(method));
+
+	netif_device_detach(efx->net_dev);
+	efx_reset_down(efx, method);
+
+	rc = efx->type->reset(efx, method);
+	if (rc) {
+		netif_err(efx, drv, efx->net_dev, "failed to reset hardware\n");
+		goto out;
+	}
+
+	/* Clear flags for the scopes we covered.  We assume the NIC and
+	 * driver are now quiescent so that there is no race here.
+	 */
+	efx->reset_pending &= -(1 << (method + 1));
+
+	/* Reinitialise bus-mastering, which may have been turned off before
+	 * the reset was scheduled. This is still appropriate, even in the
+	 * RESET_TYPE_DISABLE since this driver generally assumes the hardware
+	 * can respond to requests. */
+	pci_set_master(efx->pci_dev);
+
+out:
+	/* Leave device stopped if necessary */
+	disabled = rc || method == RESET_TYPE_DISABLE;
+	rc2 = efx_reset_up(efx, method, !disabled);
+	if (rc2) {
+		disabled = true;
+		if (!rc)
+			rc = rc2;
+	}
+
+	if (disabled) {
+		dev_close(efx->net_dev);
+		netif_err(efx, drv, efx->net_dev, "has been disabled\n");
+		efx->state = STATE_DISABLED;
+	} else {
+		netif_dbg(efx, drv, efx->net_dev, "reset complete\n");
+		netif_device_attach(efx->net_dev);
+	}
+	return rc;
+}
+
+/* The worker thread exists so that code that cannot sleep can
+ * schedule a reset for later.
+ */
+static void efx_reset_work(struct work_struct *data)
+{
+	struct efx_nic *efx = container_of(data, struct efx_nic, reset_work);
+	unsigned long pending = ACCESS_ONCE(efx->reset_pending);
+
+	if (!pending)
+		return;
+
+	/* If we're not RUNNING then don't reset. Leave the reset_pending
+	 * flags set so that efx_pci_probe_main will be retried */
+	if (efx->state != STATE_RUNNING) {
+		netif_info(efx, drv, efx->net_dev,
+			   "scheduled reset quenched. NIC not RUNNING\n");
+		return;
+	}
+
+	rtnl_lock();
+	(void)efx_reset(efx, fls(pending) - 1);
+	rtnl_unlock();
+}
+
+void efx_schedule_reset(struct efx_nic *efx, enum reset_type type)
+{
+	enum reset_type method;
+
+	switch (type) {
+	case RESET_TYPE_INVISIBLE:
+	case RESET_TYPE_ALL:
+	case RESET_TYPE_WORLD:
+	case RESET_TYPE_DISABLE:
+		method = type;
+		netif_dbg(efx, drv, efx->net_dev, "scheduling %s reset\n",
+			  RESET_TYPE(method));
+		break;
+	default:
+		method = efx->type->map_reset_reason(type);
+		netif_dbg(efx, drv, efx->net_dev,
+			  "scheduling %s reset for %s\n",
+			  RESET_TYPE(method), RESET_TYPE(type));
+		break;
+	}
+
+	set_bit(method, &efx->reset_pending);
+
+	/* efx_process_channel() will no longer read events once a
+	 * reset is scheduled. So switch back to poll'd MCDI completions. */
+	efx_mcdi_mode_poll(efx);
+
+	queue_work(reset_workqueue, &efx->reset_work);
+}
+
+/**************************************************************************
+ *
+ * List of NICs we support
+ *
+ **************************************************************************/
+
+/* PCI device ID table */
+static DEFINE_PCI_DEVICE_TABLE(efx_pci_table) = {
+	{PCI_DEVICE(EFX_VENDID_SFC, FALCON_A_P_DEVID),
+	 .driver_data = (unsigned long) &falcon_a1_nic_type},
+	{PCI_DEVICE(EFX_VENDID_SFC, FALCON_B_P_DEVID),
+	 .driver_data = (unsigned long) &falcon_b0_nic_type},
+	{PCI_DEVICE(EFX_VENDID_SFC, BETHPAGE_A_P_DEVID),
+	 .driver_data = (unsigned long) &siena_a0_nic_type},
+	{PCI_DEVICE(EFX_VENDID_SFC, SIENA_A_P_DEVID),
+	 .driver_data = (unsigned long) &siena_a0_nic_type},
+	{0}			/* end of list */
+};
+
+/**************************************************************************
+ *
+ * Dummy PHY/MAC operations
+ *
+ * Can be used for some unimplemented operations
+ * Needed so all function pointers are valid and do not have to be tested
+ * before use
+ *
+ **************************************************************************/
+int efx_port_dummy_op_int(struct efx_nic *efx)
+{
+	return 0;
+}
+void efx_port_dummy_op_void(struct efx_nic *efx) {}
+
+static bool efx_port_dummy_op_poll(struct efx_nic *efx)
+{
+	return false;
+}
+
+static const struct efx_phy_operations efx_dummy_phy_operations = {
+	.init		 = efx_port_dummy_op_int,
+	.reconfigure	 = efx_port_dummy_op_int,
+	.poll		 = efx_port_dummy_op_poll,
+	.fini		 = efx_port_dummy_op_void,
+};
+
+/**************************************************************************
+ *
+ * Data housekeeping
+ *
+ **************************************************************************/
+
+/* This zeroes out and then fills in the invariants in a struct
+ * efx_nic (including all sub-structures).
+ */
+static int efx_init_struct(struct efx_nic *efx, const struct efx_nic_type *type,
+			   struct pci_dev *pci_dev, struct net_device *net_dev)
+{
+	int i;
+
+	/* Initialise common structures */
+	memset(efx, 0, sizeof(*efx));
+	spin_lock_init(&efx->biu_lock);
+#ifdef CONFIG_SFC_MTD
+	INIT_LIST_HEAD(&efx->mtd_list);
+#endif
+	INIT_WORK(&efx->reset_work, efx_reset_work);
+	INIT_DELAYED_WORK(&efx->monitor_work, efx_monitor);
+	efx->pci_dev = pci_dev;
+	efx->msg_enable = debug;
+	efx->state = STATE_INIT;
+	strlcpy(efx->name, pci_name(pci_dev), sizeof(efx->name));
+
+	efx->net_dev = net_dev;
+	spin_lock_init(&efx->stats_lock);
+	mutex_init(&efx->mac_lock);
+	efx->mac_op = type->default_mac_ops;
+	efx->phy_op = &efx_dummy_phy_operations;
+	efx->mdio.dev = net_dev;
+	INIT_WORK(&efx->mac_work, efx_mac_work);
+
+	for (i = 0; i < EFX_MAX_CHANNELS; i++) {
+		efx->channel[i] = efx_alloc_channel(efx, i, NULL);
+		if (!efx->channel[i])
+			goto fail;
+	}
+
+	efx->type = type;
+
+	EFX_BUG_ON_PARANOID(efx->type->phys_addr_channels > EFX_MAX_CHANNELS);
+
+	/* Higher numbered interrupt modes are less capable! */
+	efx->interrupt_mode = max(efx->type->max_interrupt_mode,
+				  interrupt_mode);
+
+	/* Would be good to use the net_dev name, but we're too early */
+	snprintf(efx->workqueue_name, sizeof(efx->workqueue_name), "sfc%s",
+		 pci_name(pci_dev));
+	efx->workqueue = create_singlethread_workqueue(efx->workqueue_name);
+	if (!efx->workqueue)
+		goto fail;
+
+	return 0;
+
+fail:
+	efx_fini_struct(efx);
+	return -ENOMEM;
+}
+
+static void efx_fini_struct(struct efx_nic *efx)
+{
+	int i;
+
+	for (i = 0; i < EFX_MAX_CHANNELS; i++)
+		kfree(efx->channel[i]);
+
+	if (efx->workqueue) {
+		destroy_workqueue(efx->workqueue);
+		efx->workqueue = NULL;
+	}
+}
+
+/**************************************************************************
+ *
+ * PCI interface
+ *
+ **************************************************************************/
+
+/* Main body of final NIC shutdown code
+ * This is called only at module unload (or hotplug removal).
+ */
+static void efx_pci_remove_main(struct efx_nic *efx)
+{
+#ifdef CONFIG_RFS_ACCEL
+	free_irq_cpu_rmap(efx->net_dev->rx_cpu_rmap);
+	efx->net_dev->rx_cpu_rmap = NULL;
+#endif
+	efx_nic_fini_interrupt(efx);
+	efx_fini_channels(efx);
+	efx_fini_port(efx);
+	efx->type->fini(efx);
+	efx_fini_napi(efx);
+	efx_remove_all(efx);
+}
+
+/* Final NIC shutdown
+ * This is called only at module unload (or hotplug removal).
+ */
+static void efx_pci_remove(struct pci_dev *pci_dev)
+{
+	struct efx_nic *efx;
+
+	efx = pci_get_drvdata(pci_dev);
+	if (!efx)
+		return;
+
+	/* Mark the NIC as fini, then stop the interface */
+	rtnl_lock();
+	efx->state = STATE_FINI;
+	dev_close(efx->net_dev);
+
+	/* Allow any queued efx_resets() to complete */
+	rtnl_unlock();
+
+	efx_unregister_netdev(efx);
+
+	efx_mtd_remove(efx);
+
+	/* Wait for any scheduled resets to complete. No more will be
+	 * scheduled from this point because efx_stop_all() has been
+	 * called, we are no longer registered with driverlink, and
+	 * the net_device's have been removed. */
+	cancel_work_sync(&efx->reset_work);
+
+	efx_pci_remove_main(efx);
+
+	efx_fini_io(efx);
+	netif_dbg(efx, drv, efx->net_dev, "shutdown successful\n");
+
+	pci_set_drvdata(pci_dev, NULL);
+	efx_fini_struct(efx);
+	free_netdev(efx->net_dev);
+};
+
+/* Main body of NIC initialisation
+ * This is called at module load (or hotplug insertion, theoretically).
+ */
+static int efx_pci_probe_main(struct efx_nic *efx)
+{
+	int rc;
+
+	/* Do start-of-day initialisation */
+	rc = efx_probe_all(efx);
+	if (rc)
+		goto fail1;
+
+	efx_init_napi(efx);
+
+	rc = efx->type->init(efx);
+	if (rc) {
+		netif_err(efx, probe, efx->net_dev,
+			  "failed to initialise NIC\n");
+		goto fail3;
+	}
+
+	rc = efx_init_port(efx);
+	if (rc) {
+		netif_err(efx, probe, efx->net_dev,
+			  "failed to initialise port\n");
+		goto fail4;
+	}
+
+	efx_init_channels(efx);
+
+	rc = efx_nic_init_interrupt(efx);
+	if (rc)
+		goto fail5;
+
+	return 0;
+
+ fail5:
+	efx_fini_channels(efx);
+	efx_fini_port(efx);
+ fail4:
+	efx->type->fini(efx);
+ fail3:
+	efx_fini_napi(efx);
+	efx_remove_all(efx);
+ fail1:
+	return rc;
+}
+
+/* NIC initialisation
+ *
+ * This is called at module load (or hotplug insertion,
+ * theoretically).  It sets up PCI mappings, tests and resets the NIC,
+ * sets up and registers the network devices with the kernel and hooks
+ * the interrupt service routine.  It does not prepare the device for
+ * transmission; this is left to the first time one of the network
+ * interfaces is brought up (i.e. efx_net_open).
+ */
+static int __devinit efx_pci_probe(struct pci_dev *pci_dev,
+				   const struct pci_device_id *entry)
+{
+	const struct efx_nic_type *type = (const struct efx_nic_type *) entry->driver_data;
+	struct net_device *net_dev;
+	struct efx_nic *efx;
+	int i, rc;
+
+	/* Allocate and initialise a struct net_device and struct efx_nic */
+	net_dev = alloc_etherdev_mqs(sizeof(*efx), EFX_MAX_CORE_TX_QUEUES,
+				     EFX_MAX_RX_QUEUES);
+	if (!net_dev)
+		return -ENOMEM;
+	net_dev->features |= (type->offload_features | NETIF_F_SG |
+			      NETIF_F_HIGHDMA | NETIF_F_TSO |
+			      NETIF_F_RXCSUM);
+	if (type->offload_features & NETIF_F_V6_CSUM)
+		net_dev->features |= NETIF_F_TSO6;
+	/* Mask for features that also apply to VLAN devices */
+	net_dev->vlan_features |= (NETIF_F_ALL_CSUM | NETIF_F_SG |
+				   NETIF_F_HIGHDMA | NETIF_F_ALL_TSO |
+				   NETIF_F_RXCSUM);
+	/* All offloads can be toggled */
+	net_dev->hw_features = net_dev->features & ~NETIF_F_HIGHDMA;
+	efx = netdev_priv(net_dev);
+	pci_set_drvdata(pci_dev, efx);
+	SET_NETDEV_DEV(net_dev, &pci_dev->dev);
+	rc = efx_init_struct(efx, type, pci_dev, net_dev);
+	if (rc)
+		goto fail1;
+
+	netif_info(efx, probe, efx->net_dev,
+		   "Solarflare NIC detected\n");
+
+	/* Set up basic I/O (BAR mappings etc) */
+	rc = efx_init_io(efx);
+	if (rc)
+		goto fail2;
+
+	/* No serialisation is required with the reset path because
+	 * we're in STATE_INIT. */
+	for (i = 0; i < 5; i++) {
+		rc = efx_pci_probe_main(efx);
+
+		/* Serialise against efx_reset(). No more resets will be
+		 * scheduled since efx_stop_all() has been called, and we
+		 * have not and never have been registered with either
+		 * the rtnetlink or driverlink layers. */
+		cancel_work_sync(&efx->reset_work);
+
+		if (rc == 0) {
+			if (efx->reset_pending) {
+				/* If there was a scheduled reset during
+				 * probe, the NIC is probably hosed anyway */
+				efx_pci_remove_main(efx);
+				rc = -EIO;
+			} else {
+				break;
+			}
+		}
+
+		/* Retry if a recoverably reset event has been scheduled */
+		if (efx->reset_pending &
+		    ~(1 << RESET_TYPE_INVISIBLE | 1 << RESET_TYPE_ALL) ||
+		    !efx->reset_pending)
+			goto fail3;
+
+		efx->reset_pending = 0;
+	}
+
+	if (rc) {
+		netif_err(efx, probe, efx->net_dev, "Could not reset NIC\n");
+		goto fail4;
+	}
+
+	/* Switch to the running state before we expose the device to the OS,
+	 * so that dev_open()|efx_start_all() will actually start the device */
+	efx->state = STATE_RUNNING;
+
+	rc = efx_register_netdev(efx);
+	if (rc)
+		goto fail5;
+
+	netif_dbg(efx, probe, efx->net_dev, "initialisation successful\n");
+
+	rtnl_lock();
+	efx_mtd_probe(efx); /* allowed to fail */
+	rtnl_unlock();
+	return 0;
+
+ fail5:
+	efx_pci_remove_main(efx);
+ fail4:
+ fail3:
+	efx_fini_io(efx);
+ fail2:
+	efx_fini_struct(efx);
+ fail1:
+	WARN_ON(rc > 0);
+	netif_dbg(efx, drv, efx->net_dev, "initialisation failed. rc=%d\n", rc);
+	free_netdev(net_dev);
+	return rc;
+}
+
+static int efx_pm_freeze(struct device *dev)
+{
+	struct efx_nic *efx = pci_get_drvdata(to_pci_dev(dev));
+
+	efx->state = STATE_FINI;
+
+	netif_device_detach(efx->net_dev);
+
+	efx_stop_all(efx);
+	efx_fini_channels(efx);
+
+	return 0;
+}
+
+static int efx_pm_thaw(struct device *dev)
+{
+	struct efx_nic *efx = pci_get_drvdata(to_pci_dev(dev));
+
+	efx->state = STATE_INIT;
+
+	efx_init_channels(efx);
+
+	mutex_lock(&efx->mac_lock);
+	efx->phy_op->reconfigure(efx);
+	mutex_unlock(&efx->mac_lock);
+
+	efx_start_all(efx);
+
+	netif_device_attach(efx->net_dev);
+
+	efx->state = STATE_RUNNING;
+
+	efx->type->resume_wol(efx);
+
+	/* Reschedule any quenched resets scheduled during efx_pm_freeze() */
+	queue_work(reset_workqueue, &efx->reset_work);
+
+	return 0;
+}
+
+static int efx_pm_poweroff(struct device *dev)
+{
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+	struct efx_nic *efx = pci_get_drvdata(pci_dev);
+
+	efx->type->fini(efx);
+
+	efx->reset_pending = 0;
+
+	pci_save_state(pci_dev);
+	return pci_set_power_state(pci_dev, PCI_D3hot);
+}
+
+/* Used for both resume and restore */
+static int efx_pm_resume(struct device *dev)
+{
+	struct pci_dev *pci_dev = to_pci_dev(dev);
+	struct efx_nic *efx = pci_get_drvdata(pci_dev);
+	int rc;
+
+	rc = pci_set_power_state(pci_dev, PCI_D0);
+	if (rc)
+		return rc;
+	pci_restore_state(pci_dev);
+	rc = pci_enable_device(pci_dev);
+	if (rc)
+		return rc;
+	pci_set_master(efx->pci_dev);
+	rc = efx->type->reset(efx, RESET_TYPE_ALL);
+	if (rc)
+		return rc;
+	rc = efx->type->init(efx);
+	if (rc)
+		return rc;
+	efx_pm_thaw(dev);
+	return 0;
+}
+
+static int efx_pm_suspend(struct device *dev)
+{
+	int rc;
+
+	efx_pm_freeze(dev);
+	rc = efx_pm_poweroff(dev);
+	if (rc)
+		efx_pm_resume(dev);
+	return rc;
+}
+
+static struct dev_pm_ops efx_pm_ops = {
+	.suspend	= efx_pm_suspend,
+	.resume		= efx_pm_resume,
+	.freeze		= efx_pm_freeze,
+	.thaw		= efx_pm_thaw,
+	.poweroff	= efx_pm_poweroff,
+	.restore	= efx_pm_resume,
+};
+
+static struct pci_driver efx_pci_driver = {
+	.name		= KBUILD_MODNAME,
+	.id_table	= efx_pci_table,
+	.probe		= efx_pci_probe,
+	.remove		= efx_pci_remove,
+	.driver.pm	= &efx_pm_ops,
+};
+
+/**************************************************************************
+ *
+ * Kernel module interface
+ *
+ *************************************************************************/
+
+module_param(interrupt_mode, uint, 0444);
+MODULE_PARM_DESC(interrupt_mode,
+		 "Interrupt mode (0=>MSIX 1=>MSI 2=>legacy)");
+
+static int __init efx_init_module(void)
+{
+	int rc;
+
+	printk(KERN_INFO "Solarflare NET driver v" EFX_DRIVER_VERSION "\n");
+
+	rc = register_netdevice_notifier(&efx_netdev_notifier);
+	if (rc)
+		goto err_notifier;
+
+	reset_workqueue = create_singlethread_workqueue("sfc_reset");
+	if (!reset_workqueue) {
+		rc = -ENOMEM;
+		goto err_reset;
+	}
+
+	rc = pci_register_driver(&efx_pci_driver);
+	if (rc < 0)
+		goto err_pci;
+
+	return 0;
+
+ err_pci:
+	destroy_workqueue(reset_workqueue);
+ err_reset:
+	unregister_netdevice_notifier(&efx_netdev_notifier);
+ err_notifier:
+	return rc;
+}
+
+static void __exit efx_exit_module(void)
+{
+	printk(KERN_INFO "Solarflare NET driver unloading\n");
+
+	pci_unregister_driver(&efx_pci_driver);
+	destroy_workqueue(reset_workqueue);
+	unregister_netdevice_notifier(&efx_netdev_notifier);
+
+}
+
+module_init(efx_init_module);
+module_exit(efx_exit_module);
+
+MODULE_AUTHOR("Solarflare Communications and "
+	      "Michael Brown <mbrown@fensystems.co.uk>");
+MODULE_DESCRIPTION("Solarflare Communications network driver");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(pci, efx_pci_table);
diff --git a/drivers/net/ethernet/sfc/efx.h b/drivers/net/ethernet/sfc/efx.h
new file mode 100644
index 0000000..b0d1209
--- /dev/null
+++ b/drivers/net/ethernet/sfc/efx.h
@@ -0,0 +1,147 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2005-2006 Fen Systems Ltd.
+ * Copyright 2006-2010 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#ifndef EFX_EFX_H
+#define EFX_EFX_H
+
+#include "net_driver.h"
+#include "filter.h"
+
+/* PCI IDs */
+#define EFX_VENDID_SFC	        0x1924
+#define FALCON_A_P_DEVID	0x0703
+#define FALCON_A_S_DEVID        0x6703
+#define FALCON_B_P_DEVID        0x0710
+#define BETHPAGE_A_P_DEVID      0x0803
+#define SIENA_A_P_DEVID         0x0813
+
+/* Solarstorm controllers use BAR 0 for I/O space and BAR 2(&3) for memory */
+#define EFX_MEM_BAR 2
+
+/* TX */
+extern int efx_probe_tx_queue(struct efx_tx_queue *tx_queue);
+extern void efx_remove_tx_queue(struct efx_tx_queue *tx_queue);
+extern void efx_init_tx_queue(struct efx_tx_queue *tx_queue);
+extern void efx_init_tx_queue_core_txq(struct efx_tx_queue *tx_queue);
+extern void efx_fini_tx_queue(struct efx_tx_queue *tx_queue);
+extern void efx_release_tx_buffers(struct efx_tx_queue *tx_queue);
+extern netdev_tx_t
+efx_hard_start_xmit(struct sk_buff *skb, struct net_device *net_dev);
+extern netdev_tx_t
+efx_enqueue_skb(struct efx_tx_queue *tx_queue, struct sk_buff *skb);
+extern void efx_xmit_done(struct efx_tx_queue *tx_queue, unsigned int index);
+extern int efx_setup_tc(struct net_device *net_dev, u8 num_tc);
+
+/* RX */
+extern int efx_probe_rx_queue(struct efx_rx_queue *rx_queue);
+extern void efx_remove_rx_queue(struct efx_rx_queue *rx_queue);
+extern void efx_init_rx_queue(struct efx_rx_queue *rx_queue);
+extern void efx_fini_rx_queue(struct efx_rx_queue *rx_queue);
+extern void efx_rx_strategy(struct efx_channel *channel);
+extern void efx_fast_push_rx_descriptors(struct efx_rx_queue *rx_queue);
+extern void efx_rx_slow_fill(unsigned long context);
+extern void __efx_rx_packet(struct efx_channel *channel,
+			    struct efx_rx_buffer *rx_buf, bool checksummed);
+extern void efx_rx_packet(struct efx_rx_queue *rx_queue, unsigned int index,
+			  unsigned int len, bool checksummed, bool discard);
+extern void efx_schedule_slow_fill(struct efx_rx_queue *rx_queue);
+
+#define EFX_MAX_DMAQ_SIZE 4096UL
+#define EFX_DEFAULT_DMAQ_SIZE 1024UL
+#define EFX_MIN_DMAQ_SIZE 512UL
+
+#define EFX_MAX_EVQ_SIZE 16384UL
+#define EFX_MIN_EVQ_SIZE 512UL
+
+/* The smallest [rt]xq_entries that the driver supports. Callers of
+ * efx_wake_queue() assume that they can subsequently send at least one
+ * skb. Falcon/A1 may require up to three descriptors per skb_frag. */
+#define EFX_MIN_RING_SIZE (roundup_pow_of_two(2 * 3 * MAX_SKB_FRAGS))
+
+/* Filters */
+extern int efx_probe_filters(struct efx_nic *efx);
+extern void efx_restore_filters(struct efx_nic *efx);
+extern void efx_remove_filters(struct efx_nic *efx);
+extern int efx_filter_insert_filter(struct efx_nic *efx,
+				    struct efx_filter_spec *spec,
+				    bool replace);
+extern int efx_filter_remove_filter(struct efx_nic *efx,
+				    struct efx_filter_spec *spec);
+extern void efx_filter_clear_rx(struct efx_nic *efx,
+				enum efx_filter_priority priority);
+#ifdef CONFIG_RFS_ACCEL
+extern int efx_filter_rfs(struct net_device *net_dev, const struct sk_buff *skb,
+			  u16 rxq_index, u32 flow_id);
+extern bool __efx_filter_rfs_expire(struct efx_nic *efx, unsigned quota);
+static inline void efx_filter_rfs_expire(struct efx_channel *channel)
+{
+	if (channel->rfs_filters_added >= 60 &&
+	    __efx_filter_rfs_expire(channel->efx, 100))
+		channel->rfs_filters_added -= 60;
+}
+#define efx_filter_rfs_enabled() 1
+#else
+static inline void efx_filter_rfs_expire(struct efx_channel *channel) {}
+#define efx_filter_rfs_enabled() 0
+#endif
+
+/* Channels */
+extern void efx_process_channel_now(struct efx_channel *channel);
+extern int
+efx_realloc_channels(struct efx_nic *efx, u32 rxq_entries, u32 txq_entries);
+
+/* Ports */
+extern int efx_reconfigure_port(struct efx_nic *efx);
+extern int __efx_reconfigure_port(struct efx_nic *efx);
+
+/* Ethtool support */
+extern const struct ethtool_ops efx_ethtool_ops;
+
+/* Reset handling */
+extern int efx_reset(struct efx_nic *efx, enum reset_type method);
+extern void efx_reset_down(struct efx_nic *efx, enum reset_type method);
+extern int efx_reset_up(struct efx_nic *efx, enum reset_type method, bool ok);
+
+/* Global */
+extern void efx_schedule_reset(struct efx_nic *efx, enum reset_type type);
+extern void efx_init_irq_moderation(struct efx_nic *efx, int tx_usecs,
+				    int rx_usecs, bool rx_adaptive);
+
+/* Dummy PHY ops for PHY drivers */
+extern int efx_port_dummy_op_int(struct efx_nic *efx);
+extern void efx_port_dummy_op_void(struct efx_nic *efx);
+
+
+/* MTD */
+#ifdef CONFIG_SFC_MTD
+extern int efx_mtd_probe(struct efx_nic *efx);
+extern void efx_mtd_rename(struct efx_nic *efx);
+extern void efx_mtd_remove(struct efx_nic *efx);
+#else
+static inline int efx_mtd_probe(struct efx_nic *efx) { return 0; }
+static inline void efx_mtd_rename(struct efx_nic *efx) {}
+static inline void efx_mtd_remove(struct efx_nic *efx) {}
+#endif
+
+static inline void efx_schedule_channel(struct efx_channel *channel)
+{
+	netif_vdbg(channel->efx, intr, channel->efx->net_dev,
+		   "channel %d scheduling NAPI poll on CPU%d\n",
+		   channel->channel, raw_smp_processor_id());
+	channel->work_pending = true;
+
+	napi_schedule(&channel->napi_str);
+}
+
+extern void efx_link_status_changed(struct efx_nic *efx);
+extern void efx_link_set_advertising(struct efx_nic *efx, u32);
+extern void efx_link_set_wanted_fc(struct efx_nic *efx, u8);
+
+#endif /* EFX_EFX_H */
diff --git a/drivers/net/ethernet/sfc/enum.h b/drivers/net/ethernet/sfc/enum.h
new file mode 100644
index 0000000..d725a8f
--- /dev/null
+++ b/drivers/net/ethernet/sfc/enum.h
@@ -0,0 +1,167 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2007-2009 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#ifndef EFX_ENUM_H
+#define EFX_ENUM_H
+
+/**
+ * enum efx_loopback_mode - loopback modes
+ * @LOOPBACK_NONE: no loopback
+ * @LOOPBACK_DATA: data path loopback
+ * @LOOPBACK_GMAC: loopback within GMAC
+ * @LOOPBACK_XGMII: loopback after XMAC
+ * @LOOPBACK_XGXS: loopback within BPX after XGXS
+ * @LOOPBACK_XAUI: loopback within BPX before XAUI serdes
+ * @LOOPBACK_GMII: loopback within BPX after GMAC
+ * @LOOPBACK_SGMII: loopback within BPX within SGMII
+ * @LOOPBACK_XGBR: loopback within BPX within XGBR
+ * @LOOPBACK_XFI: loopback within BPX before XFI serdes
+ * @LOOPBACK_XAUI_FAR: loopback within BPX after XAUI serdes
+ * @LOOPBACK_GMII_FAR: loopback within BPX before SGMII
+ * @LOOPBACK_SGMII_FAR: loopback within BPX after SGMII
+ * @LOOPBACK_XFI_FAR: loopback after XFI serdes
+ * @LOOPBACK_GPHY: loopback within 1G PHY at unspecified level
+ * @LOOPBACK_PHYXS: loopback within 10G PHY at PHYXS level
+ * @LOOPBACK_PCS: loopback within 10G PHY at PCS level
+ * @LOOPBACK_PMAPMD: loopback within 10G PHY at PMAPMD level
+ * @LOOPBACK_XPORT: cross port loopback
+ * @LOOPBACK_XGMII_WS: wireside loopback excluding XMAC
+ * @LOOPBACK_XAUI_WS: wireside loopback within BPX within XAUI serdes
+ * @LOOPBACK_XAUI_WS_FAR: wireside loopback within BPX including XAUI serdes
+ * @LOOPBACK_XAUI_WS_NEAR: wireside loopback within BPX excluding XAUI serdes
+ * @LOOPBACK_GMII_WS: wireside loopback excluding GMAC
+ * @LOOPBACK_XFI_WS: wireside loopback excluding XFI serdes
+ * @LOOPBACK_XFI_WS_FAR: wireside loopback including XFI serdes
+ * @LOOPBACK_PHYXS_WS: wireside loopback within 10G PHY at PHYXS level
+ */
+/* Please keep up-to-date w.r.t the following two #defines */
+enum efx_loopback_mode {
+	LOOPBACK_NONE = 0,
+	LOOPBACK_DATA = 1,
+	LOOPBACK_GMAC = 2,
+	LOOPBACK_XGMII = 3,
+	LOOPBACK_XGXS = 4,
+	LOOPBACK_XAUI = 5,
+	LOOPBACK_GMII = 6,
+	LOOPBACK_SGMII = 7,
+	LOOPBACK_XGBR = 8,
+	LOOPBACK_XFI = 9,
+	LOOPBACK_XAUI_FAR = 10,
+	LOOPBACK_GMII_FAR = 11,
+	LOOPBACK_SGMII_FAR = 12,
+	LOOPBACK_XFI_FAR = 13,
+	LOOPBACK_GPHY = 14,
+	LOOPBACK_PHYXS = 15,
+	LOOPBACK_PCS = 16,
+	LOOPBACK_PMAPMD = 17,
+	LOOPBACK_XPORT = 18,
+	LOOPBACK_XGMII_WS = 19,
+	LOOPBACK_XAUI_WS = 20,
+	LOOPBACK_XAUI_WS_FAR = 21,
+	LOOPBACK_XAUI_WS_NEAR = 22,
+	LOOPBACK_GMII_WS = 23,
+	LOOPBACK_XFI_WS = 24,
+	LOOPBACK_XFI_WS_FAR = 25,
+	LOOPBACK_PHYXS_WS = 26,
+	LOOPBACK_MAX
+};
+#define LOOPBACK_TEST_MAX LOOPBACK_PMAPMD
+
+/* These loopbacks occur within the controller */
+#define LOOPBACKS_INTERNAL ((1 << LOOPBACK_DATA) |		\
+			    (1 << LOOPBACK_GMAC) |		\
+			    (1 << LOOPBACK_XGMII)|		\
+			    (1 << LOOPBACK_XGXS) |		\
+			    (1 << LOOPBACK_XAUI) |		\
+			    (1 << LOOPBACK_GMII) |		\
+			    (1 << LOOPBACK_SGMII) |		\
+			    (1 << LOOPBACK_SGMII) |		\
+			    (1 << LOOPBACK_XGBR) |		\
+			    (1 << LOOPBACK_XFI) |		\
+			    (1 << LOOPBACK_XAUI_FAR) |		\
+			    (1 << LOOPBACK_GMII_FAR) |		\
+			    (1 << LOOPBACK_SGMII_FAR) |		\
+			    (1 << LOOPBACK_XFI_FAR) |		\
+			    (1 << LOOPBACK_XGMII_WS) |		\
+			    (1 << LOOPBACK_XAUI_WS) |		\
+			    (1 << LOOPBACK_XAUI_WS_FAR) |	\
+			    (1 << LOOPBACK_XAUI_WS_NEAR) |	\
+			    (1 << LOOPBACK_GMII_WS) |		\
+			    (1 << LOOPBACK_XFI_WS) |		\
+			    (1 << LOOPBACK_XFI_WS_FAR))
+
+#define LOOPBACKS_WS ((1 << LOOPBACK_XGMII_WS) |		\
+		      (1 << LOOPBACK_XAUI_WS) |			\
+		      (1 << LOOPBACK_XAUI_WS_FAR) |		\
+		      (1 << LOOPBACK_XAUI_WS_NEAR) |		\
+		      (1 << LOOPBACK_GMII_WS) |			\
+		      (1 << LOOPBACK_XFI_WS) |			\
+		      (1 << LOOPBACK_XFI_WS_FAR) |		\
+		      (1 << LOOPBACK_PHYXS_WS))
+
+#define LOOPBACKS_EXTERNAL(_efx)					\
+	((_efx)->loopback_modes & ~LOOPBACKS_INTERNAL &			\
+	 ~(1 << LOOPBACK_NONE))
+
+#define LOOPBACK_MASK(_efx)			\
+	(1 << (_efx)->loopback_mode)
+
+#define LOOPBACK_INTERNAL(_efx)				\
+	(!!(LOOPBACKS_INTERNAL & LOOPBACK_MASK(_efx)))
+
+#define LOOPBACK_EXTERNAL(_efx)				\
+	(!!(LOOPBACK_MASK(_efx) & LOOPBACKS_EXTERNAL(_efx)))
+
+#define LOOPBACK_CHANGED(_from, _to, _mask)				\
+	(!!((LOOPBACK_MASK(_from) ^ LOOPBACK_MASK(_to)) & (_mask)))
+
+#define LOOPBACK_OUT_OF(_from, _to, _mask)				\
+	((LOOPBACK_MASK(_from) & (_mask)) && !(LOOPBACK_MASK(_to) & (_mask)))
+
+/*****************************************************************************/
+
+/**
+ * enum reset_type - reset types
+ *
+ * %RESET_TYPE_INVSIBLE, %RESET_TYPE_ALL, %RESET_TYPE_WORLD and
+ * %RESET_TYPE_DISABLE specify the method/scope of the reset.  The
+ * other valuesspecify reasons, which efx_schedule_reset() will choose
+ * a method for.
+ *
+ * Reset methods are numbered in order of increasing scope.
+ *
+ * @RESET_TYPE_INVISIBLE: don't reset the PHYs or interrupts
+ * @RESET_TYPE_ALL: reset everything but PCI core blocks
+ * @RESET_TYPE_WORLD: reset everything, save & restore PCI config
+ * @RESET_TYPE_DISABLE: disable NIC
+ * @RESET_TYPE_TX_WATCHDOG: reset due to TX watchdog
+ * @RESET_TYPE_INT_ERROR: reset due to internal error
+ * @RESET_TYPE_RX_RECOVERY: reset to recover from RX datapath errors
+ * @RESET_TYPE_RX_DESC_FETCH: pcie error during rx descriptor fetch
+ * @RESET_TYPE_TX_DESC_FETCH: pcie error during tx descriptor fetch
+ * @RESET_TYPE_TX_SKIP: hardware completed empty tx descriptors
+ * @RESET_TYPE_MC_FAILURE: MC reboot/assertion
+ */
+enum reset_type {
+	RESET_TYPE_INVISIBLE = 0,
+	RESET_TYPE_ALL = 1,
+	RESET_TYPE_WORLD = 2,
+	RESET_TYPE_DISABLE = 3,
+	RESET_TYPE_MAX_METHOD,
+	RESET_TYPE_TX_WATCHDOG,
+	RESET_TYPE_INT_ERROR,
+	RESET_TYPE_RX_RECOVERY,
+	RESET_TYPE_RX_DESC_FETCH,
+	RESET_TYPE_TX_DESC_FETCH,
+	RESET_TYPE_TX_SKIP,
+	RESET_TYPE_MC_FAILURE,
+	RESET_TYPE_MAX,
+};
+
+#endif /* EFX_ENUM_H */
diff --git a/drivers/net/ethernet/sfc/ethtool.c b/drivers/net/ethernet/sfc/ethtool.c
new file mode 100644
index 0000000..bc4643a
--- /dev/null
+++ b/drivers/net/ethernet/sfc/ethtool.c
@@ -0,0 +1,1012 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2005-2006 Fen Systems Ltd.
+ * Copyright 2006-2010 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#include <linux/netdevice.h>
+#include <linux/ethtool.h>
+#include <linux/rtnetlink.h>
+#include <linux/in.h>
+#include "net_driver.h"
+#include "workarounds.h"
+#include "selftest.h"
+#include "efx.h"
+#include "filter.h"
+#include "nic.h"
+
+struct ethtool_string {
+	char name[ETH_GSTRING_LEN];
+};
+
+struct efx_ethtool_stat {
+	const char *name;
+	enum {
+		EFX_ETHTOOL_STAT_SOURCE_mac_stats,
+		EFX_ETHTOOL_STAT_SOURCE_nic,
+		EFX_ETHTOOL_STAT_SOURCE_channel,
+		EFX_ETHTOOL_STAT_SOURCE_tx_queue
+	} source;
+	unsigned offset;
+	u64(*get_stat) (void *field); /* Reader function */
+};
+
+/* Initialiser for a struct #efx_ethtool_stat with type-checking */
+#define EFX_ETHTOOL_STAT(stat_name, source_name, field, field_type, \
+				get_stat_function) {			\
+	.name = #stat_name,						\
+	.source = EFX_ETHTOOL_STAT_SOURCE_##source_name,		\
+	.offset = ((((field_type *) 0) ==				\
+		      &((struct efx_##source_name *)0)->field) ?	\
+		    offsetof(struct efx_##source_name, field) :		\
+		    offsetof(struct efx_##source_name, field)),		\
+	.get_stat = get_stat_function,					\
+}
+
+static u64 efx_get_uint_stat(void *field)
+{
+	return *(unsigned int *)field;
+}
+
+static u64 efx_get_ulong_stat(void *field)
+{
+	return *(unsigned long *)field;
+}
+
+static u64 efx_get_u64_stat(void *field)
+{
+	return *(u64 *) field;
+}
+
+static u64 efx_get_atomic_stat(void *field)
+{
+	return atomic_read((atomic_t *) field);
+}
+
+#define EFX_ETHTOOL_ULONG_MAC_STAT(field)			\
+	EFX_ETHTOOL_STAT(field, mac_stats, field, 		\
+			  unsigned long, efx_get_ulong_stat)
+
+#define EFX_ETHTOOL_U64_MAC_STAT(field)				\
+	EFX_ETHTOOL_STAT(field, mac_stats, field, 		\
+			  u64, efx_get_u64_stat)
+
+#define EFX_ETHTOOL_UINT_NIC_STAT(name)				\
+	EFX_ETHTOOL_STAT(name, nic, n_##name,			\
+			 unsigned int, efx_get_uint_stat)
+
+#define EFX_ETHTOOL_ATOMIC_NIC_ERROR_STAT(field)		\
+	EFX_ETHTOOL_STAT(field, nic, field,			\
+			 atomic_t, efx_get_atomic_stat)
+
+#define EFX_ETHTOOL_UINT_CHANNEL_STAT(field)			\
+	EFX_ETHTOOL_STAT(field, channel, n_##field,		\
+			 unsigned int, efx_get_uint_stat)
+
+#define EFX_ETHTOOL_UINT_TXQ_STAT(field)			\
+	EFX_ETHTOOL_STAT(tx_##field, tx_queue, field,		\
+			 unsigned int, efx_get_uint_stat)
+
+static struct efx_ethtool_stat efx_ethtool_stats[] = {
+	EFX_ETHTOOL_U64_MAC_STAT(tx_bytes),
+	EFX_ETHTOOL_U64_MAC_STAT(tx_good_bytes),
+	EFX_ETHTOOL_U64_MAC_STAT(tx_bad_bytes),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_packets),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_bad),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_pause),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_control),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_unicast),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_multicast),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_broadcast),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_lt64),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_64),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_65_to_127),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_128_to_255),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_256_to_511),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_512_to_1023),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_1024_to_15xx),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_15xx_to_jumbo),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_gtjumbo),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_collision),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_single_collision),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_multiple_collision),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_excessive_collision),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_deferred),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_late_collision),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_excessive_deferred),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_non_tcpudp),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_mac_src_error),
+	EFX_ETHTOOL_ULONG_MAC_STAT(tx_ip_src_error),
+	EFX_ETHTOOL_UINT_TXQ_STAT(tso_bursts),
+	EFX_ETHTOOL_UINT_TXQ_STAT(tso_long_headers),
+	EFX_ETHTOOL_UINT_TXQ_STAT(tso_packets),
+	EFX_ETHTOOL_UINT_TXQ_STAT(pushes),
+	EFX_ETHTOOL_U64_MAC_STAT(rx_bytes),
+	EFX_ETHTOOL_U64_MAC_STAT(rx_good_bytes),
+	EFX_ETHTOOL_U64_MAC_STAT(rx_bad_bytes),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_packets),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_good),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_bad),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_pause),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_control),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_unicast),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_multicast),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_broadcast),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_lt64),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_64),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_65_to_127),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_128_to_255),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_256_to_511),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_512_to_1023),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_1024_to_15xx),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_15xx_to_jumbo),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_gtjumbo),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_bad_lt64),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_bad_64_to_15xx),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_bad_15xx_to_jumbo),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_bad_gtjumbo),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_overflow),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_missed),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_false_carrier),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_symbol_error),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_align_error),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_length_error),
+	EFX_ETHTOOL_ULONG_MAC_STAT(rx_internal_error),
+	EFX_ETHTOOL_UINT_NIC_STAT(rx_nodesc_drop_cnt),
+	EFX_ETHTOOL_ATOMIC_NIC_ERROR_STAT(rx_reset),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_tobe_disc),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_ip_hdr_chksum_err),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_tcp_udp_chksum_err),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_mcast_mismatch),
+	EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_frm_trunc),
+};
+
+/* Number of ethtool statistics */
+#define EFX_ETHTOOL_NUM_STATS ARRAY_SIZE(efx_ethtool_stats)
+
+#define EFX_ETHTOOL_EEPROM_MAGIC 0xEFAB
+
+/**************************************************************************
+ *
+ * Ethtool operations
+ *
+ **************************************************************************
+ */
+
+/* Identify device by flashing LEDs */
+static int efx_ethtool_phys_id(struct net_device *net_dev,
+			       enum ethtool_phys_id_state state)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	enum efx_led_mode mode = EFX_LED_DEFAULT;
+
+	switch (state) {
+	case ETHTOOL_ID_ON:
+		mode = EFX_LED_ON;
+		break;
+	case ETHTOOL_ID_OFF:
+		mode = EFX_LED_OFF;
+		break;
+	case ETHTOOL_ID_INACTIVE:
+		mode = EFX_LED_DEFAULT;
+		break;
+	case ETHTOOL_ID_ACTIVE:
+		return 1;	/* cycle on/off once per second */
+	}
+
+	efx->type->set_id_led(efx, mode);
+	return 0;
+}
+
+/* This must be called with rtnl_lock held. */
+static int efx_ethtool_get_settings(struct net_device *net_dev,
+				    struct ethtool_cmd *ecmd)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	struct efx_link_state *link_state = &efx->link_state;
+
+	mutex_lock(&efx->mac_lock);
+	efx->phy_op->get_settings(efx, ecmd);
+	mutex_unlock(&efx->mac_lock);
+
+	/* GMAC does not support 1000Mbps HD */
+	ecmd->supported &= ~SUPPORTED_1000baseT_Half;
+	/* Both MACs support pause frames (bidirectional and respond-only) */
+	ecmd->supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause;
+
+	if (LOOPBACK_INTERNAL(efx)) {
+		ethtool_cmd_speed_set(ecmd, link_state->speed);
+		ecmd->duplex = link_state->fd ? DUPLEX_FULL : DUPLEX_HALF;
+	}
+
+	return 0;
+}
+
+/* This must be called with rtnl_lock held. */
+static int efx_ethtool_set_settings(struct net_device *net_dev,
+				    struct ethtool_cmd *ecmd)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	int rc;
+
+	/* GMAC does not support 1000Mbps HD */
+	if ((ethtool_cmd_speed(ecmd) == SPEED_1000) &&
+	    (ecmd->duplex != DUPLEX_FULL)) {
+		netif_dbg(efx, drv, efx->net_dev,
+			  "rejecting unsupported 1000Mbps HD setting\n");
+		return -EINVAL;
+	}
+
+	mutex_lock(&efx->mac_lock);
+	rc = efx->phy_op->set_settings(efx, ecmd);
+	mutex_unlock(&efx->mac_lock);
+	return rc;
+}
+
+static void efx_ethtool_get_drvinfo(struct net_device *net_dev,
+				    struct ethtool_drvinfo *info)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+
+	strlcpy(info->driver, KBUILD_MODNAME, sizeof(info->driver));
+	strlcpy(info->version, EFX_DRIVER_VERSION, sizeof(info->version));
+	if (efx_nic_rev(efx) >= EFX_REV_SIENA_A0)
+		efx_mcdi_print_fwver(efx, info->fw_version,
+				     sizeof(info->fw_version));
+	strlcpy(info->bus_info, pci_name(efx->pci_dev), sizeof(info->bus_info));
+}
+
+static int efx_ethtool_get_regs_len(struct net_device *net_dev)
+{
+	return efx_nic_get_regs_len(netdev_priv(net_dev));
+}
+
+static void efx_ethtool_get_regs(struct net_device *net_dev,
+				 struct ethtool_regs *regs, void *buf)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+
+	regs->version = efx->type->revision;
+	efx_nic_get_regs(efx, buf);
+}
+
+static u32 efx_ethtool_get_msglevel(struct net_device *net_dev)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	return efx->msg_enable;
+}
+
+static void efx_ethtool_set_msglevel(struct net_device *net_dev, u32 msg_enable)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	efx->msg_enable = msg_enable;
+}
+
+/**
+ * efx_fill_test - fill in an individual self-test entry
+ * @test_index:		Index of the test
+ * @strings:		Ethtool strings, or %NULL
+ * @data:		Ethtool test results, or %NULL
+ * @test:		Pointer to test result (used only if data != %NULL)
+ * @unit_format:	Unit name format (e.g. "chan\%d")
+ * @unit_id:		Unit id (e.g. 0 for "chan0")
+ * @test_format:	Test name format (e.g. "loopback.\%s.tx.sent")
+ * @test_id:		Test id (e.g. "PHYXS" for "loopback.PHYXS.tx_sent")
+ *
+ * Fill in an individual self-test entry.
+ */
+static void efx_fill_test(unsigned int test_index,
+			  struct ethtool_string *strings, u64 *data,
+			  int *test, const char *unit_format, int unit_id,
+			  const char *test_format, const char *test_id)
+{
+	struct ethtool_string unit_str, test_str;
+
+	/* Fill data value, if applicable */
+	if (data)
+		data[test_index] = *test;
+
+	/* Fill string, if applicable */
+	if (strings) {
+		if (strchr(unit_format, '%'))
+			snprintf(unit_str.name, sizeof(unit_str.name),
+				 unit_format, unit_id);
+		else
+			strcpy(unit_str.name, unit_format);
+		snprintf(test_str.name, sizeof(test_str.name),
+			 test_format, test_id);
+		snprintf(strings[test_index].name,
+			 sizeof(strings[test_index].name),
+			 "%-6s %-24s", unit_str.name, test_str.name);
+	}
+}
+
+#define EFX_CHANNEL_NAME(_channel) "chan%d", _channel->channel
+#define EFX_TX_QUEUE_NAME(_tx_queue) "txq%d", _tx_queue->queue
+#define EFX_RX_QUEUE_NAME(_rx_queue) "rxq%d", _rx_queue->queue
+#define EFX_LOOPBACK_NAME(_mode, _counter)			\
+	"loopback.%s." _counter, STRING_TABLE_LOOKUP(_mode, efx_loopback_mode)
+
+/**
+ * efx_fill_loopback_test - fill in a block of loopback self-test entries
+ * @efx:		Efx NIC
+ * @lb_tests:		Efx loopback self-test results structure
+ * @mode:		Loopback test mode
+ * @test_index:		Starting index of the test
+ * @strings:		Ethtool strings, or %NULL
+ * @data:		Ethtool test results, or %NULL
+ */
+static int efx_fill_loopback_test(struct efx_nic *efx,
+				  struct efx_loopback_self_tests *lb_tests,
+				  enum efx_loopback_mode mode,
+				  unsigned int test_index,
+				  struct ethtool_string *strings, u64 *data)
+{
+	struct efx_channel *channel = efx_get_channel(efx, 0);
+	struct efx_tx_queue *tx_queue;
+
+	efx_for_each_channel_tx_queue(tx_queue, channel) {
+		efx_fill_test(test_index++, strings, data,
+			      &lb_tests->tx_sent[tx_queue->queue],
+			      EFX_TX_QUEUE_NAME(tx_queue),
+			      EFX_LOOPBACK_NAME(mode, "tx_sent"));
+		efx_fill_test(test_index++, strings, data,
+			      &lb_tests->tx_done[tx_queue->queue],
+			      EFX_TX_QUEUE_NAME(tx_queue),
+			      EFX_LOOPBACK_NAME(mode, "tx_done"));
+	}
+	efx_fill_test(test_index++, strings, data,
+		      &lb_tests->rx_good,
+		      "rx", 0,
+		      EFX_LOOPBACK_NAME(mode, "rx_good"));
+	efx_fill_test(test_index++, strings, data,
+		      &lb_tests->rx_bad,
+		      "rx", 0,
+		      EFX_LOOPBACK_NAME(mode, "rx_bad"));
+
+	return test_index;
+}
+
+/**
+ * efx_ethtool_fill_self_tests - get self-test details
+ * @efx:		Efx NIC
+ * @tests:		Efx self-test results structure, or %NULL
+ * @strings:		Ethtool strings, or %NULL
+ * @data:		Ethtool test results, or %NULL
+ */
+static int efx_ethtool_fill_self_tests(struct efx_nic *efx,
+				       struct efx_self_tests *tests,
+				       struct ethtool_string *strings,
+				       u64 *data)
+{
+	struct efx_channel *channel;
+	unsigned int n = 0, i;
+	enum efx_loopback_mode mode;
+
+	efx_fill_test(n++, strings, data, &tests->phy_alive,
+		      "phy", 0, "alive", NULL);
+	efx_fill_test(n++, strings, data, &tests->nvram,
+		      "core", 0, "nvram", NULL);
+	efx_fill_test(n++, strings, data, &tests->interrupt,
+		      "core", 0, "interrupt", NULL);
+
+	/* Event queues */
+	efx_for_each_channel(channel, efx) {
+		efx_fill_test(n++, strings, data,
+			      &tests->eventq_dma[channel->channel],
+			      EFX_CHANNEL_NAME(channel),
+			      "eventq.dma", NULL);
+		efx_fill_test(n++, strings, data,
+			      &tests->eventq_int[channel->channel],
+			      EFX_CHANNEL_NAME(channel),
+			      "eventq.int", NULL);
+		efx_fill_test(n++, strings, data,
+			      &tests->eventq_poll[channel->channel],
+			      EFX_CHANNEL_NAME(channel),
+			      "eventq.poll", NULL);
+	}
+
+	efx_fill_test(n++, strings, data, &tests->registers,
+		      "core", 0, "registers", NULL);
+
+	if (efx->phy_op->run_tests != NULL) {
+		EFX_BUG_ON_PARANOID(efx->phy_op->test_name == NULL);
+
+		for (i = 0; true; ++i) {
+			const char *name;
+
+			EFX_BUG_ON_PARANOID(i >= EFX_MAX_PHY_TESTS);
+			name = efx->phy_op->test_name(efx, i);
+			if (name == NULL)
+				break;
+
+			efx_fill_test(n++, strings, data, &tests->phy_ext[i],
+				      "phy", 0, name, NULL);
+		}
+	}
+
+	/* Loopback tests */
+	for (mode = LOOPBACK_NONE; mode <= LOOPBACK_TEST_MAX; mode++) {
+		if (!(efx->loopback_modes & (1 << mode)))
+			continue;
+		n = efx_fill_loopback_test(efx,
+					   &tests->loopback[mode], mode, n,
+					   strings, data);
+	}
+
+	return n;
+}
+
+static int efx_ethtool_get_sset_count(struct net_device *net_dev,
+				      int string_set)
+{
+	switch (string_set) {
+	case ETH_SS_STATS:
+		return EFX_ETHTOOL_NUM_STATS;
+	case ETH_SS_TEST:
+		return efx_ethtool_fill_self_tests(netdev_priv(net_dev),
+						   NULL, NULL, NULL);
+	default:
+		return -EINVAL;
+	}
+}
+
+static void efx_ethtool_get_strings(struct net_device *net_dev,
+				    u32 string_set, u8 *strings)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	struct ethtool_string *ethtool_strings =
+		(struct ethtool_string *)strings;
+	int i;
+
+	switch (string_set) {
+	case ETH_SS_STATS:
+		for (i = 0; i < EFX_ETHTOOL_NUM_STATS; i++)
+			strncpy(ethtool_strings[i].name,
+				efx_ethtool_stats[i].name,
+				sizeof(ethtool_strings[i].name));
+		break;
+	case ETH_SS_TEST:
+		efx_ethtool_fill_self_tests(efx, NULL,
+					    ethtool_strings, NULL);
+		break;
+	default:
+		/* No other string sets */
+		break;
+	}
+}
+
+static void efx_ethtool_get_stats(struct net_device *net_dev,
+				  struct ethtool_stats *stats,
+				  u64 *data)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	struct efx_mac_stats *mac_stats = &efx->mac_stats;
+	struct efx_ethtool_stat *stat;
+	struct efx_channel *channel;
+	struct efx_tx_queue *tx_queue;
+	struct rtnl_link_stats64 temp;
+	int i;
+
+	EFX_BUG_ON_PARANOID(stats->n_stats != EFX_ETHTOOL_NUM_STATS);
+
+	/* Update MAC and NIC statistics */
+	dev_get_stats(net_dev, &temp);
+
+	/* Fill detailed statistics buffer */
+	for (i = 0; i < EFX_ETHTOOL_NUM_STATS; i++) {
+		stat = &efx_ethtool_stats[i];
+		switch (stat->source) {
+		case EFX_ETHTOOL_STAT_SOURCE_mac_stats:
+			data[i] = stat->get_stat((void *)mac_stats +
+						 stat->offset);
+			break;
+		case EFX_ETHTOOL_STAT_SOURCE_nic:
+			data[i] = stat->get_stat((void *)efx + stat->offset);
+			break;
+		case EFX_ETHTOOL_STAT_SOURCE_channel:
+			data[i] = 0;
+			efx_for_each_channel(channel, efx)
+				data[i] += stat->get_stat((void *)channel +
+							  stat->offset);
+			break;
+		case EFX_ETHTOOL_STAT_SOURCE_tx_queue:
+			data[i] = 0;
+			efx_for_each_channel(channel, efx) {
+				efx_for_each_channel_tx_queue(tx_queue, channel)
+					data[i] +=
+						stat->get_stat((void *)tx_queue
+							       + stat->offset);
+			}
+			break;
+		}
+	}
+}
+
+static void efx_ethtool_self_test(struct net_device *net_dev,
+				  struct ethtool_test *test, u64 *data)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	struct efx_self_tests *efx_tests;
+	int already_up;
+	int rc = -ENOMEM;
+
+	efx_tests = kzalloc(sizeof(*efx_tests), GFP_KERNEL);
+	if (!efx_tests)
+		goto fail;
+
+
+	ASSERT_RTNL();
+	if (efx->state != STATE_RUNNING) {
+		rc = -EIO;
+		goto fail1;
+	}
+
+	netif_info(efx, drv, efx->net_dev, "starting %sline testing\n",
+		   (test->flags & ETH_TEST_FL_OFFLINE) ? "off" : "on");
+
+	/* We need rx buffers and interrupts. */
+	already_up = (efx->net_dev->flags & IFF_UP);
+	if (!already_up) {
+		rc = dev_open(efx->net_dev);
+		if (rc) {
+			netif_err(efx, drv, efx->net_dev,
+				  "failed opening device.\n");
+			goto fail1;
+		}
+	}
+
+	rc = efx_selftest(efx, efx_tests, test->flags);
+
+	if (!already_up)
+		dev_close(efx->net_dev);
+
+	netif_info(efx, drv, efx->net_dev, "%s %sline self-tests\n",
+		   rc == 0 ? "passed" : "failed",
+		   (test->flags & ETH_TEST_FL_OFFLINE) ? "off" : "on");
+
+fail1:
+	/* Fill ethtool results structures */
+	efx_ethtool_fill_self_tests(efx, efx_tests, NULL, data);
+	kfree(efx_tests);
+fail:
+	if (rc)
+		test->flags |= ETH_TEST_FL_FAILED;
+}
+
+/* Restart autonegotiation */
+static int efx_ethtool_nway_reset(struct net_device *net_dev)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+
+	return mdio45_nway_restart(&efx->mdio);
+}
+
+static int efx_ethtool_get_coalesce(struct net_device *net_dev,
+				    struct ethtool_coalesce *coalesce)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	struct efx_channel *channel;
+
+	memset(coalesce, 0, sizeof(*coalesce));
+
+	/* Find lowest IRQ moderation across all used TX queues */
+	coalesce->tx_coalesce_usecs_irq = ~((u32) 0);
+	efx_for_each_channel(channel, efx) {
+		if (!efx_channel_has_tx_queues(channel))
+			continue;
+		if (channel->irq_moderation < coalesce->tx_coalesce_usecs_irq) {
+			if (channel->channel < efx->n_rx_channels)
+				coalesce->tx_coalesce_usecs_irq =
+					channel->irq_moderation;
+			else
+				coalesce->tx_coalesce_usecs_irq = 0;
+		}
+	}
+
+	coalesce->use_adaptive_rx_coalesce = efx->irq_rx_adaptive;
+	coalesce->rx_coalesce_usecs_irq = efx->irq_rx_moderation;
+
+	coalesce->tx_coalesce_usecs_irq *= EFX_IRQ_MOD_RESOLUTION;
+	coalesce->rx_coalesce_usecs_irq *= EFX_IRQ_MOD_RESOLUTION;
+
+	return 0;
+}
+
+/* Set coalescing parameters
+ * The difficulties occur for shared channels
+ */
+static int efx_ethtool_set_coalesce(struct net_device *net_dev,
+				    struct ethtool_coalesce *coalesce)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	struct efx_channel *channel;
+	unsigned tx_usecs, rx_usecs, adaptive;
+
+	if (coalesce->use_adaptive_tx_coalesce)
+		return -EOPNOTSUPP;
+
+	if (coalesce->rx_coalesce_usecs || coalesce->tx_coalesce_usecs) {
+		netif_err(efx, drv, efx->net_dev, "invalid coalescing setting. "
+			  "Only rx/tx_coalesce_usecs_irq are supported\n");
+		return -EOPNOTSUPP;
+	}
+
+	rx_usecs = coalesce->rx_coalesce_usecs_irq;
+	tx_usecs = coalesce->tx_coalesce_usecs_irq;
+	adaptive = coalesce->use_adaptive_rx_coalesce;
+
+	/* If the channel is shared only allow RX parameters to be set */
+	efx_for_each_channel(channel, efx) {
+		if (efx_channel_has_rx_queue(channel) &&
+		    efx_channel_has_tx_queues(channel) &&
+		    tx_usecs) {
+			netif_err(efx, drv, efx->net_dev, "Channel is shared. "
+				  "Only RX coalescing may be set\n");
+			return -EOPNOTSUPP;
+		}
+	}
+
+	efx_init_irq_moderation(efx, tx_usecs, rx_usecs, adaptive);
+	efx_for_each_channel(channel, efx)
+		efx->type->push_irq_moderation(channel);
+
+	return 0;
+}
+
+static void efx_ethtool_get_ringparam(struct net_device *net_dev,
+				      struct ethtool_ringparam *ring)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+
+	ring->rx_max_pending = EFX_MAX_DMAQ_SIZE;
+	ring->tx_max_pending = EFX_MAX_DMAQ_SIZE;
+	ring->rx_mini_max_pending = 0;
+	ring->rx_jumbo_max_pending = 0;
+	ring->rx_pending = efx->rxq_entries;
+	ring->tx_pending = efx->txq_entries;
+	ring->rx_mini_pending = 0;
+	ring->rx_jumbo_pending = 0;
+}
+
+static int efx_ethtool_set_ringparam(struct net_device *net_dev,
+				     struct ethtool_ringparam *ring)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+
+	if (ring->rx_mini_pending || ring->rx_jumbo_pending ||
+	    ring->rx_pending > EFX_MAX_DMAQ_SIZE ||
+	    ring->tx_pending > EFX_MAX_DMAQ_SIZE)
+		return -EINVAL;
+
+	if (ring->rx_pending < EFX_MIN_RING_SIZE ||
+	    ring->tx_pending < EFX_MIN_RING_SIZE) {
+		netif_err(efx, drv, efx->net_dev,
+			  "TX and RX queues cannot be smaller than %ld\n",
+			  EFX_MIN_RING_SIZE);
+		return -EINVAL;
+	}
+
+	return efx_realloc_channels(efx, ring->rx_pending, ring->tx_pending);
+}
+
+static int efx_ethtool_set_pauseparam(struct net_device *net_dev,
+				      struct ethtool_pauseparam *pause)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	u8 wanted_fc, old_fc;
+	u32 old_adv;
+	bool reset;
+	int rc = 0;
+
+	mutex_lock(&efx->mac_lock);
+
+	wanted_fc = ((pause->rx_pause ? EFX_FC_RX : 0) |
+		     (pause->tx_pause ? EFX_FC_TX : 0) |
+		     (pause->autoneg ? EFX_FC_AUTO : 0));
+
+	if ((wanted_fc & EFX_FC_TX) && !(wanted_fc & EFX_FC_RX)) {
+		netif_dbg(efx, drv, efx->net_dev,
+			  "Flow control unsupported: tx ON rx OFF\n");
+		rc = -EINVAL;
+		goto out;
+	}
+
+	if ((wanted_fc & EFX_FC_AUTO) && !efx->link_advertising) {
+		netif_dbg(efx, drv, efx->net_dev,
+			  "Autonegotiation is disabled\n");
+		rc = -EINVAL;
+		goto out;
+	}
+
+	/* TX flow control may automatically turn itself off if the
+	 * link partner (intermittently) stops responding to pause
+	 * frames. There isn't any indication that this has happened,
+	 * so the best we do is leave it up to the user to spot this
+	 * and fix it be cycling transmit flow control on this end. */
+	reset = (wanted_fc & EFX_FC_TX) && !(efx->wanted_fc & EFX_FC_TX);
+	if (EFX_WORKAROUND_11482(efx) && reset) {
+		if (efx_nic_rev(efx) == EFX_REV_FALCON_B0) {
+			/* Recover by resetting the EM block */
+			falcon_stop_nic_stats(efx);
+			falcon_drain_tx_fifo(efx);
+			efx->mac_op->reconfigure(efx);
+			falcon_start_nic_stats(efx);
+		} else {
+			/* Schedule a reset to recover */
+			efx_schedule_reset(efx, RESET_TYPE_INVISIBLE);
+		}
+	}
+
+	old_adv = efx->link_advertising;
+	old_fc = efx->wanted_fc;
+	efx_link_set_wanted_fc(efx, wanted_fc);
+	if (efx->link_advertising != old_adv ||
+	    (efx->wanted_fc ^ old_fc) & EFX_FC_AUTO) {
+		rc = efx->phy_op->reconfigure(efx);
+		if (rc) {
+			netif_err(efx, drv, efx->net_dev,
+				  "Unable to advertise requested flow "
+				  "control setting\n");
+			goto out;
+		}
+	}
+
+	/* Reconfigure the MAC. The PHY *may* generate a link state change event
+	 * if the user just changed the advertised capabilities, but there's no
+	 * harm doing this twice */
+	efx->mac_op->reconfigure(efx);
+
+out:
+	mutex_unlock(&efx->mac_lock);
+
+	return rc;
+}
+
+static void efx_ethtool_get_pauseparam(struct net_device *net_dev,
+				       struct ethtool_pauseparam *pause)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+
+	pause->rx_pause = !!(efx->wanted_fc & EFX_FC_RX);
+	pause->tx_pause = !!(efx->wanted_fc & EFX_FC_TX);
+	pause->autoneg = !!(efx->wanted_fc & EFX_FC_AUTO);
+}
+
+
+static void efx_ethtool_get_wol(struct net_device *net_dev,
+				struct ethtool_wolinfo *wol)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	return efx->type->get_wol(efx, wol);
+}
+
+
+static int efx_ethtool_set_wol(struct net_device *net_dev,
+			       struct ethtool_wolinfo *wol)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	return efx->type->set_wol(efx, wol->wolopts);
+}
+
+static int efx_ethtool_reset(struct net_device *net_dev, u32 *flags)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	int rc;
+
+	rc = efx->type->map_reset_flags(flags);
+	if (rc < 0)
+		return rc;
+
+	return efx_reset(efx, rc);
+}
+
+static int
+efx_ethtool_get_rxnfc(struct net_device *net_dev,
+		      struct ethtool_rxnfc *info, void *rules __always_unused)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+
+	switch (info->cmd) {
+	case ETHTOOL_GRXRINGS:
+		info->data = efx->n_rx_channels;
+		return 0;
+
+	case ETHTOOL_GRXFH: {
+		unsigned min_revision = 0;
+
+		info->data = 0;
+		switch (info->flow_type) {
+		case TCP_V4_FLOW:
+			info->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+			/* fall through */
+		case UDP_V4_FLOW:
+		case SCTP_V4_FLOW:
+		case AH_ESP_V4_FLOW:
+		case IPV4_FLOW:
+			info->data |= RXH_IP_SRC | RXH_IP_DST;
+			min_revision = EFX_REV_FALCON_B0;
+			break;
+		case TCP_V6_FLOW:
+			info->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+			/* fall through */
+		case UDP_V6_FLOW:
+		case SCTP_V6_FLOW:
+		case AH_ESP_V6_FLOW:
+		case IPV6_FLOW:
+			info->data |= RXH_IP_SRC | RXH_IP_DST;
+			min_revision = EFX_REV_SIENA_A0;
+			break;
+		default:
+			break;
+		}
+		if (efx_nic_rev(efx) < min_revision)
+			info->data = 0;
+		return 0;
+	}
+
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int efx_ethtool_set_rx_ntuple(struct net_device *net_dev,
+				     struct ethtool_rx_ntuple *ntuple)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	struct ethtool_tcpip4_spec *ip_entry = &ntuple->fs.h_u.tcp_ip4_spec;
+	struct ethtool_tcpip4_spec *ip_mask = &ntuple->fs.m_u.tcp_ip4_spec;
+	struct ethhdr *mac_entry = &ntuple->fs.h_u.ether_spec;
+	struct ethhdr *mac_mask = &ntuple->fs.m_u.ether_spec;
+	struct efx_filter_spec filter;
+	int rc;
+
+	/* Range-check action */
+	if (ntuple->fs.action < ETHTOOL_RXNTUPLE_ACTION_CLEAR ||
+	    ntuple->fs.action >= (s32)efx->n_rx_channels)
+		return -EINVAL;
+
+	if (~ntuple->fs.data_mask)
+		return -EINVAL;
+
+	efx_filter_init_rx(&filter, EFX_FILTER_PRI_MANUAL, 0,
+			   (ntuple->fs.action == ETHTOOL_RXNTUPLE_ACTION_DROP) ?
+			   0xfff : ntuple->fs.action);
+
+	switch (ntuple->fs.flow_type) {
+	case TCP_V4_FLOW:
+	case UDP_V4_FLOW: {
+		u8 proto = (ntuple->fs.flow_type == TCP_V4_FLOW ?
+			    IPPROTO_TCP : IPPROTO_UDP);
+
+		/* Must match all of destination, */
+		if (ip_mask->ip4dst | ip_mask->pdst)
+			return -EINVAL;
+		/* all or none of source, */
+		if ((ip_mask->ip4src | ip_mask->psrc) &&
+		    ((__force u32)~ip_mask->ip4src |
+		     (__force u16)~ip_mask->psrc))
+			return -EINVAL;
+		/* and nothing else */
+		if ((u8)~ip_mask->tos | (u16)~ntuple->fs.vlan_tag_mask)
+			return -EINVAL;
+
+		if (!ip_mask->ip4src)
+			rc = efx_filter_set_ipv4_full(&filter, proto,
+						      ip_entry->ip4dst,
+						      ip_entry->pdst,
+						      ip_entry->ip4src,
+						      ip_entry->psrc);
+		else
+			rc = efx_filter_set_ipv4_local(&filter, proto,
+						       ip_entry->ip4dst,
+						       ip_entry->pdst);
+		if (rc)
+			return rc;
+		break;
+	}
+
+	case ETHER_FLOW:
+		/* Must match all of destination, */
+		if (!is_zero_ether_addr(mac_mask->h_dest))
+			return -EINVAL;
+		/* all or none of VID, */
+		if (ntuple->fs.vlan_tag_mask != 0xf000 &&
+		    ntuple->fs.vlan_tag_mask != 0xffff)
+			return -EINVAL;
+		/* and nothing else */
+		if (!is_broadcast_ether_addr(mac_mask->h_source) ||
+		    mac_mask->h_proto != htons(0xffff))
+			return -EINVAL;
+
+		rc = efx_filter_set_eth_local(
+			&filter,
+			(ntuple->fs.vlan_tag_mask == 0xf000) ?
+			ntuple->fs.vlan_tag : EFX_FILTER_VID_UNSPEC,
+			mac_entry->h_dest);
+		if (rc)
+			return rc;
+		break;
+
+	default:
+		return -EINVAL;
+	}
+
+	if (ntuple->fs.action == ETHTOOL_RXNTUPLE_ACTION_CLEAR)
+		return efx_filter_remove_filter(efx, &filter);
+
+	rc = efx_filter_insert_filter(efx, &filter, true);
+	return rc < 0 ? rc : 0;
+}
+
+static int efx_ethtool_get_rxfh_indir(struct net_device *net_dev,
+				      struct ethtool_rxfh_indir *indir)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	size_t copy_size =
+		min_t(size_t, indir->size, ARRAY_SIZE(efx->rx_indir_table));
+
+	if (efx_nic_rev(efx) < EFX_REV_FALCON_B0)
+		return -EOPNOTSUPP;
+
+	indir->size = ARRAY_SIZE(efx->rx_indir_table);
+	memcpy(indir->ring_index, efx->rx_indir_table,
+	       copy_size * sizeof(indir->ring_index[0]));
+	return 0;
+}
+
+static int efx_ethtool_set_rxfh_indir(struct net_device *net_dev,
+				      const struct ethtool_rxfh_indir *indir)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	size_t i;
+
+	if (efx_nic_rev(efx) < EFX_REV_FALCON_B0)
+		return -EOPNOTSUPP;
+
+	/* Validate size and indices */
+	if (indir->size != ARRAY_SIZE(efx->rx_indir_table))
+		return -EINVAL;
+	for (i = 0; i < ARRAY_SIZE(efx->rx_indir_table); i++)
+		if (indir->ring_index[i] >= efx->n_rx_channels)
+			return -EINVAL;
+
+	memcpy(efx->rx_indir_table, indir->ring_index,
+	       sizeof(efx->rx_indir_table));
+	efx_nic_push_rx_indir_table(efx);
+	return 0;
+}
+
+const struct ethtool_ops efx_ethtool_ops = {
+	.get_settings		= efx_ethtool_get_settings,
+	.set_settings		= efx_ethtool_set_settings,
+	.get_drvinfo		= efx_ethtool_get_drvinfo,
+	.get_regs_len		= efx_ethtool_get_regs_len,
+	.get_regs		= efx_ethtool_get_regs,
+	.get_msglevel		= efx_ethtool_get_msglevel,
+	.set_msglevel		= efx_ethtool_set_msglevel,
+	.nway_reset		= efx_ethtool_nway_reset,
+	.get_link		= ethtool_op_get_link,
+	.get_coalesce		= efx_ethtool_get_coalesce,
+	.set_coalesce		= efx_ethtool_set_coalesce,
+	.get_ringparam		= efx_ethtool_get_ringparam,
+	.set_ringparam		= efx_ethtool_set_ringparam,
+	.get_pauseparam         = efx_ethtool_get_pauseparam,
+	.set_pauseparam         = efx_ethtool_set_pauseparam,
+	.get_sset_count		= efx_ethtool_get_sset_count,
+	.self_test		= efx_ethtool_self_test,
+	.get_strings		= efx_ethtool_get_strings,
+	.set_phys_id		= efx_ethtool_phys_id,
+	.get_ethtool_stats	= efx_ethtool_get_stats,
+	.get_wol                = efx_ethtool_get_wol,
+	.set_wol                = efx_ethtool_set_wol,
+	.reset			= efx_ethtool_reset,
+	.get_rxnfc		= efx_ethtool_get_rxnfc,
+	.set_rx_ntuple		= efx_ethtool_set_rx_ntuple,
+	.get_rxfh_indir		= efx_ethtool_get_rxfh_indir,
+	.set_rxfh_indir		= efx_ethtool_set_rxfh_indir,
+};
diff --git a/drivers/net/ethernet/sfc/falcon.c b/drivers/net/ethernet/sfc/falcon.c
new file mode 100644
index 0000000..94bf4aa
--- /dev/null
+++ b/drivers/net/ethernet/sfc/falcon.c
@@ -0,0 +1,1841 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2005-2006 Fen Systems Ltd.
+ * Copyright 2006-2010 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#include <linux/bitops.h>
+#include <linux/delay.h>
+#include <linux/pci.h>
+#include <linux/module.h>
+#include <linux/seq_file.h>
+#include <linux/i2c.h>
+#include <linux/mii.h>
+#include <linux/slab.h>
+#include "net_driver.h"
+#include "bitfield.h"
+#include "efx.h"
+#include "mac.h"
+#include "spi.h"
+#include "nic.h"
+#include "regs.h"
+#include "io.h"
+#include "phy.h"
+#include "workarounds.h"
+
+/* Hardware control for SFC4000 (aka Falcon). */
+
+static const unsigned int
+/* "Large" EEPROM device: Atmel AT25640 or similar
+ * 8 KB, 16-bit address, 32 B write block */
+large_eeprom_type = ((13 << SPI_DEV_TYPE_SIZE_LBN)
+		     | (2 << SPI_DEV_TYPE_ADDR_LEN_LBN)
+		     | (5 << SPI_DEV_TYPE_BLOCK_SIZE_LBN)),
+/* Default flash device: Atmel AT25F1024
+ * 128 KB, 24-bit address, 32 KB erase block, 256 B write block */
+default_flash_type = ((17 << SPI_DEV_TYPE_SIZE_LBN)
+		      | (3 << SPI_DEV_TYPE_ADDR_LEN_LBN)
+		      | (0x52 << SPI_DEV_TYPE_ERASE_CMD_LBN)
+		      | (15 << SPI_DEV_TYPE_ERASE_SIZE_LBN)
+		      | (8 << SPI_DEV_TYPE_BLOCK_SIZE_LBN));
+
+/**************************************************************************
+ *
+ * I2C bus - this is a bit-bashing interface using GPIO pins
+ * Note that it uses the output enables to tristate the outputs
+ * SDA is the data pin and SCL is the clock
+ *
+ **************************************************************************
+ */
+static void falcon_setsda(void *data, int state)
+{
+	struct efx_nic *efx = (struct efx_nic *)data;
+	efx_oword_t reg;
+
+	efx_reado(efx, &reg, FR_AB_GPIO_CTL);
+	EFX_SET_OWORD_FIELD(reg, FRF_AB_GPIO3_OEN, !state);
+	efx_writeo(efx, &reg, FR_AB_GPIO_CTL);
+}
+
+static void falcon_setscl(void *data, int state)
+{
+	struct efx_nic *efx = (struct efx_nic *)data;
+	efx_oword_t reg;
+
+	efx_reado(efx, &reg, FR_AB_GPIO_CTL);
+	EFX_SET_OWORD_FIELD(reg, FRF_AB_GPIO0_OEN, !state);
+	efx_writeo(efx, &reg, FR_AB_GPIO_CTL);
+}
+
+static int falcon_getsda(void *data)
+{
+	struct efx_nic *efx = (struct efx_nic *)data;
+	efx_oword_t reg;
+
+	efx_reado(efx, &reg, FR_AB_GPIO_CTL);
+	return EFX_OWORD_FIELD(reg, FRF_AB_GPIO3_IN);
+}
+
+static int falcon_getscl(void *data)
+{
+	struct efx_nic *efx = (struct efx_nic *)data;
+	efx_oword_t reg;
+
+	efx_reado(efx, &reg, FR_AB_GPIO_CTL);
+	return EFX_OWORD_FIELD(reg, FRF_AB_GPIO0_IN);
+}
+
+static struct i2c_algo_bit_data falcon_i2c_bit_operations = {
+	.setsda		= falcon_setsda,
+	.setscl		= falcon_setscl,
+	.getsda		= falcon_getsda,
+	.getscl		= falcon_getscl,
+	.udelay		= 5,
+	/* Wait up to 50 ms for slave to let us pull SCL high */
+	.timeout	= DIV_ROUND_UP(HZ, 20),
+};
+
+static void falcon_push_irq_moderation(struct efx_channel *channel)
+{
+	efx_dword_t timer_cmd;
+	struct efx_nic *efx = channel->efx;
+
+	/* Set timer register */
+	if (channel->irq_moderation) {
+		EFX_POPULATE_DWORD_2(timer_cmd,
+				     FRF_AB_TC_TIMER_MODE,
+				     FFE_BB_TIMER_MODE_INT_HLDOFF,
+				     FRF_AB_TC_TIMER_VAL,
+				     channel->irq_moderation - 1);
+	} else {
+		EFX_POPULATE_DWORD_2(timer_cmd,
+				     FRF_AB_TC_TIMER_MODE,
+				     FFE_BB_TIMER_MODE_DIS,
+				     FRF_AB_TC_TIMER_VAL, 0);
+	}
+	BUILD_BUG_ON(FR_AA_TIMER_COMMAND_KER != FR_BZ_TIMER_COMMAND_P0);
+	efx_writed_page_locked(efx, &timer_cmd, FR_BZ_TIMER_COMMAND_P0,
+			       channel->channel);
+}
+
+static void falcon_deconfigure_mac_wrapper(struct efx_nic *efx);
+
+static void falcon_prepare_flush(struct efx_nic *efx)
+{
+	falcon_deconfigure_mac_wrapper(efx);
+
+	/* Wait for the tx and rx fifo's to get to the next packet boundary
+	 * (~1ms without back-pressure), then to drain the remainder of the
+	 * fifo's at data path speeds (negligible), with a healthy margin. */
+	msleep(10);
+}
+
+/* Acknowledge a legacy interrupt from Falcon
+ *
+ * This acknowledges a legacy (not MSI) interrupt via INT_ACK_KER_REG.
+ *
+ * Due to SFC bug 3706 (silicon revision <=A1) reads can be duplicated in the
+ * BIU. Interrupt acknowledge is read sensitive so must write instead
+ * (then read to ensure the BIU collector is flushed)
+ *
+ * NB most hardware supports MSI interrupts
+ */
+inline void falcon_irq_ack_a1(struct efx_nic *efx)
+{
+	efx_dword_t reg;
+
+	EFX_POPULATE_DWORD_1(reg, FRF_AA_INT_ACK_KER_FIELD, 0xb7eb7e);
+	efx_writed(efx, &reg, FR_AA_INT_ACK_KER);
+	efx_readd(efx, &reg, FR_AA_WORK_AROUND_BROKEN_PCI_READS);
+}
+
+
+irqreturn_t falcon_legacy_interrupt_a1(int irq, void *dev_id)
+{
+	struct efx_nic *efx = dev_id;
+	efx_oword_t *int_ker = efx->irq_status.addr;
+	int syserr;
+	int queues;
+
+	/* Check to see if this is our interrupt.  If it isn't, we
+	 * exit without having touched the hardware.
+	 */
+	if (unlikely(EFX_OWORD_IS_ZERO(*int_ker))) {
+		netif_vdbg(efx, intr, efx->net_dev,
+			   "IRQ %d on CPU %d not for me\n", irq,
+			   raw_smp_processor_id());
+		return IRQ_NONE;
+	}
+	efx->last_irq_cpu = raw_smp_processor_id();
+	netif_vdbg(efx, intr, efx->net_dev,
+		   "IRQ %d on CPU %d status " EFX_OWORD_FMT "\n",
+		   irq, raw_smp_processor_id(), EFX_OWORD_VAL(*int_ker));
+
+	/* Determine interrupting queues, clear interrupt status
+	 * register and acknowledge the device interrupt.
+	 */
+	BUILD_BUG_ON(FSF_AZ_NET_IVEC_INT_Q_WIDTH > EFX_MAX_CHANNELS);
+	queues = EFX_OWORD_FIELD(*int_ker, FSF_AZ_NET_IVEC_INT_Q);
+
+	/* Check to see if we have a serious error condition */
+	if (queues & (1U << efx->fatal_irq_level)) {
+		syserr = EFX_OWORD_FIELD(*int_ker, FSF_AZ_NET_IVEC_FATAL_INT);
+		if (unlikely(syserr))
+			return efx_nic_fatal_interrupt(efx);
+	}
+
+	EFX_ZERO_OWORD(*int_ker);
+	wmb(); /* Ensure the vector is cleared before interrupt ack */
+	falcon_irq_ack_a1(efx);
+
+	if (queues & 1)
+		efx_schedule_channel(efx_get_channel(efx, 0));
+	if (queues & 2)
+		efx_schedule_channel(efx_get_channel(efx, 1));
+	return IRQ_HANDLED;
+}
+/**************************************************************************
+ *
+ * EEPROM/flash
+ *
+ **************************************************************************
+ */
+
+#define FALCON_SPI_MAX_LEN sizeof(efx_oword_t)
+
+static int falcon_spi_poll(struct efx_nic *efx)
+{
+	efx_oword_t reg;
+	efx_reado(efx, &reg, FR_AB_EE_SPI_HCMD);
+	return EFX_OWORD_FIELD(reg, FRF_AB_EE_SPI_HCMD_CMD_EN) ? -EBUSY : 0;
+}
+
+/* Wait for SPI command completion */
+static int falcon_spi_wait(struct efx_nic *efx)
+{
+	/* Most commands will finish quickly, so we start polling at
+	 * very short intervals.  Sometimes the command may have to
+	 * wait for VPD or expansion ROM access outside of our
+	 * control, so we allow up to 100 ms. */
+	unsigned long timeout = jiffies + 1 + DIV_ROUND_UP(HZ, 10);
+	int i;
+
+	for (i = 0; i < 10; i++) {
+		if (!falcon_spi_poll(efx))
+			return 0;
+		udelay(10);
+	}
+
+	for (;;) {
+		if (!falcon_spi_poll(efx))
+			return 0;
+		if (time_after_eq(jiffies, timeout)) {
+			netif_err(efx, hw, efx->net_dev,
+				  "timed out waiting for SPI\n");
+			return -ETIMEDOUT;
+		}
+		schedule_timeout_uninterruptible(1);
+	}
+}
+
+int falcon_spi_cmd(struct efx_nic *efx, const struct efx_spi_device *spi,
+		   unsigned int command, int address,
+		   const void *in, void *out, size_t len)
+{
+	bool addressed = (address >= 0);
+	bool reading = (out != NULL);
+	efx_oword_t reg;
+	int rc;
+
+	/* Input validation */
+	if (len > FALCON_SPI_MAX_LEN)
+		return -EINVAL;
+
+	/* Check that previous command is not still running */
+	rc = falcon_spi_poll(efx);
+	if (rc)
+		return rc;
+
+	/* Program address register, if we have an address */
+	if (addressed) {
+		EFX_POPULATE_OWORD_1(reg, FRF_AB_EE_SPI_HADR_ADR, address);
+		efx_writeo(efx, &reg, FR_AB_EE_SPI_HADR);
+	}
+
+	/* Program data register, if we have data */
+	if (in != NULL) {
+		memcpy(&reg, in, len);
+		efx_writeo(efx, &reg, FR_AB_EE_SPI_HDATA);
+	}
+
+	/* Issue read/write command */
+	EFX_POPULATE_OWORD_7(reg,
+			     FRF_AB_EE_SPI_HCMD_CMD_EN, 1,
+			     FRF_AB_EE_SPI_HCMD_SF_SEL, spi->device_id,
+			     FRF_AB_EE_SPI_HCMD_DABCNT, len,
+			     FRF_AB_EE_SPI_HCMD_READ, reading,
+			     FRF_AB_EE_SPI_HCMD_DUBCNT, 0,
+			     FRF_AB_EE_SPI_HCMD_ADBCNT,
+			     (addressed ? spi->addr_len : 0),
+			     FRF_AB_EE_SPI_HCMD_ENC, command);
+	efx_writeo(efx, &reg, FR_AB_EE_SPI_HCMD);
+
+	/* Wait for read/write to complete */
+	rc = falcon_spi_wait(efx);
+	if (rc)
+		return rc;
+
+	/* Read data */
+	if (out != NULL) {
+		efx_reado(efx, &reg, FR_AB_EE_SPI_HDATA);
+		memcpy(out, &reg, len);
+	}
+
+	return 0;
+}
+
+static size_t
+falcon_spi_write_limit(const struct efx_spi_device *spi, size_t start)
+{
+	return min(FALCON_SPI_MAX_LEN,
+		   (spi->block_size - (start & (spi->block_size - 1))));
+}
+
+static inline u8
+efx_spi_munge_command(const struct efx_spi_device *spi,
+		      const u8 command, const unsigned int address)
+{
+	return command | (((address >> 8) & spi->munge_address) << 3);
+}
+
+/* Wait up to 10 ms for buffered write completion */
+int
+falcon_spi_wait_write(struct efx_nic *efx, const struct efx_spi_device *spi)
+{
+	unsigned long timeout = jiffies + 1 + DIV_ROUND_UP(HZ, 100);
+	u8 status;
+	int rc;
+
+	for (;;) {
+		rc = falcon_spi_cmd(efx, spi, SPI_RDSR, -1, NULL,
+				    &status, sizeof(status));
+		if (rc)
+			return rc;
+		if (!(status & SPI_STATUS_NRDY))
+			return 0;
+		if (time_after_eq(jiffies, timeout)) {
+			netif_err(efx, hw, efx->net_dev,
+				  "SPI write timeout on device %d"
+				  " last status=0x%02x\n",
+				  spi->device_id, status);
+			return -ETIMEDOUT;
+		}
+		schedule_timeout_uninterruptible(1);
+	}
+}
+
+int falcon_spi_read(struct efx_nic *efx, const struct efx_spi_device *spi,
+		    loff_t start, size_t len, size_t *retlen, u8 *buffer)
+{
+	size_t block_len, pos = 0;
+	unsigned int command;
+	int rc = 0;
+
+	while (pos < len) {
+		block_len = min(len - pos, FALCON_SPI_MAX_LEN);
+
+		command = efx_spi_munge_command(spi, SPI_READ, start + pos);
+		rc = falcon_spi_cmd(efx, spi, command, start + pos, NULL,
+				    buffer + pos, block_len);
+		if (rc)
+			break;
+		pos += block_len;
+
+		/* Avoid locking up the system */
+		cond_resched();
+		if (signal_pending(current)) {
+			rc = -EINTR;
+			break;
+		}
+	}
+
+	if (retlen)
+		*retlen = pos;
+	return rc;
+}
+
+int
+falcon_spi_write(struct efx_nic *efx, const struct efx_spi_device *spi,
+		 loff_t start, size_t len, size_t *retlen, const u8 *buffer)
+{
+	u8 verify_buffer[FALCON_SPI_MAX_LEN];
+	size_t block_len, pos = 0;
+	unsigned int command;
+	int rc = 0;
+
+	while (pos < len) {
+		rc = falcon_spi_cmd(efx, spi, SPI_WREN, -1, NULL, NULL, 0);
+		if (rc)
+			break;
+
+		block_len = min(len - pos,
+				falcon_spi_write_limit(spi, start + pos));
+		command = efx_spi_munge_command(spi, SPI_WRITE, start + pos);
+		rc = falcon_spi_cmd(efx, spi, command, start + pos,
+				    buffer + pos, NULL, block_len);
+		if (rc)
+			break;
+
+		rc = falcon_spi_wait_write(efx, spi);
+		if (rc)
+			break;
+
+		command = efx_spi_munge_command(spi, SPI_READ, start + pos);
+		rc = falcon_spi_cmd(efx, spi, command, start + pos,
+				    NULL, verify_buffer, block_len);
+		if (memcmp(verify_buffer, buffer + pos, block_len)) {
+			rc = -EIO;
+			break;
+		}
+
+		pos += block_len;
+
+		/* Avoid locking up the system */
+		cond_resched();
+		if (signal_pending(current)) {
+			rc = -EINTR;
+			break;
+		}
+	}
+
+	if (retlen)
+		*retlen = pos;
+	return rc;
+}
+
+/**************************************************************************
+ *
+ * MAC wrapper
+ *
+ **************************************************************************
+ */
+
+static void falcon_push_multicast_hash(struct efx_nic *efx)
+{
+	union efx_multicast_hash *mc_hash = &efx->multicast_hash;
+
+	WARN_ON(!mutex_is_locked(&efx->mac_lock));
+
+	efx_writeo(efx, &mc_hash->oword[0], FR_AB_MAC_MC_HASH_REG0);
+	efx_writeo(efx, &mc_hash->oword[1], FR_AB_MAC_MC_HASH_REG1);
+}
+
+static void falcon_reset_macs(struct efx_nic *efx)
+{
+	struct falcon_nic_data *nic_data = efx->nic_data;
+	efx_oword_t reg, mac_ctrl;
+	int count;
+
+	if (efx_nic_rev(efx) < EFX_REV_FALCON_B0) {
+		/* It's not safe to use GLB_CTL_REG to reset the
+		 * macs, so instead use the internal MAC resets
+		 */
+		EFX_POPULATE_OWORD_1(reg, FRF_AB_XM_CORE_RST, 1);
+		efx_writeo(efx, &reg, FR_AB_XM_GLB_CFG);
+
+		for (count = 0; count < 10000; count++) {
+			efx_reado(efx, &reg, FR_AB_XM_GLB_CFG);
+			if (EFX_OWORD_FIELD(reg, FRF_AB_XM_CORE_RST) ==
+			    0)
+				return;
+			udelay(10);
+		}
+
+		netif_err(efx, hw, efx->net_dev,
+			  "timed out waiting for XMAC core reset\n");
+	}
+
+	/* Mac stats will fail whist the TX fifo is draining */
+	WARN_ON(nic_data->stats_disable_count == 0);
+
+	efx_reado(efx, &mac_ctrl, FR_AB_MAC_CTRL);
+	EFX_SET_OWORD_FIELD(mac_ctrl, FRF_BB_TXFIFO_DRAIN_EN, 1);
+	efx_writeo(efx, &mac_ctrl, FR_AB_MAC_CTRL);
+
+	efx_reado(efx, &reg, FR_AB_GLB_CTL);
+	EFX_SET_OWORD_FIELD(reg, FRF_AB_RST_XGTX, 1);
+	EFX_SET_OWORD_FIELD(reg, FRF_AB_RST_XGRX, 1);
+	EFX_SET_OWORD_FIELD(reg, FRF_AB_RST_EM, 1);
+	efx_writeo(efx, &reg, FR_AB_GLB_CTL);
+
+	count = 0;
+	while (1) {
+		efx_reado(efx, &reg, FR_AB_GLB_CTL);
+		if (!EFX_OWORD_FIELD(reg, FRF_AB_RST_XGTX) &&
+		    !EFX_OWORD_FIELD(reg, FRF_AB_RST_XGRX) &&
+		    !EFX_OWORD_FIELD(reg, FRF_AB_RST_EM)) {
+			netif_dbg(efx, hw, efx->net_dev,
+				  "Completed MAC reset after %d loops\n",
+				  count);
+			break;
+		}
+		if (count > 20) {
+			netif_err(efx, hw, efx->net_dev, "MAC reset failed\n");
+			break;
+		}
+		count++;
+		udelay(10);
+	}
+
+	/* Ensure the correct MAC is selected before statistics
+	 * are re-enabled by the caller */
+	efx_writeo(efx, &mac_ctrl, FR_AB_MAC_CTRL);
+
+	falcon_setup_xaui(efx);
+}
+
+void falcon_drain_tx_fifo(struct efx_nic *efx)
+{
+	efx_oword_t reg;
+
+	if ((efx_nic_rev(efx) < EFX_REV_FALCON_B0) ||
+	    (efx->loopback_mode != LOOPBACK_NONE))
+		return;
+
+	efx_reado(efx, &reg, FR_AB_MAC_CTRL);
+	/* There is no point in draining more than once */
+	if (EFX_OWORD_FIELD(reg, FRF_BB_TXFIFO_DRAIN_EN))
+		return;
+
+	falcon_reset_macs(efx);
+}
+
+static void falcon_deconfigure_mac_wrapper(struct efx_nic *efx)
+{
+	efx_oword_t reg;
+
+	if (efx_nic_rev(efx) < EFX_REV_FALCON_B0)
+		return;
+
+	/* Isolate the MAC -> RX */
+	efx_reado(efx, &reg, FR_AZ_RX_CFG);
+	EFX_SET_OWORD_FIELD(reg, FRF_BZ_RX_INGR_EN, 0);
+	efx_writeo(efx, &reg, FR_AZ_RX_CFG);
+
+	/* Isolate TX -> MAC */
+	falcon_drain_tx_fifo(efx);
+}
+
+void falcon_reconfigure_mac_wrapper(struct efx_nic *efx)
+{
+	struct efx_link_state *link_state = &efx->link_state;
+	efx_oword_t reg;
+	int link_speed, isolate;
+
+	isolate = !!ACCESS_ONCE(efx->reset_pending);
+
+	switch (link_state->speed) {
+	case 10000: link_speed = 3; break;
+	case 1000:  link_speed = 2; break;
+	case 100:   link_speed = 1; break;
+	default:    link_speed = 0; break;
+	}
+	/* MAC_LINK_STATUS controls MAC backpressure but doesn't work
+	 * as advertised.  Disable to ensure packets are not
+	 * indefinitely held and TX queue can be flushed at any point
+	 * while the link is down. */
+	EFX_POPULATE_OWORD_5(reg,
+			     FRF_AB_MAC_XOFF_VAL, 0xffff /* max pause time */,
+			     FRF_AB_MAC_BCAD_ACPT, 1,
+			     FRF_AB_MAC_UC_PROM, efx->promiscuous,
+			     FRF_AB_MAC_LINK_STATUS, 1, /* always set */
+			     FRF_AB_MAC_SPEED, link_speed);
+	/* On B0, MAC backpressure can be disabled and packets get
+	 * discarded. */
+	if (efx_nic_rev(efx) >= EFX_REV_FALCON_B0) {
+		EFX_SET_OWORD_FIELD(reg, FRF_BB_TXFIFO_DRAIN_EN,
+				    !link_state->up || isolate);
+	}
+
+	efx_writeo(efx, &reg, FR_AB_MAC_CTRL);
+
+	/* Restore the multicast hash registers. */
+	falcon_push_multicast_hash(efx);
+
+	efx_reado(efx, &reg, FR_AZ_RX_CFG);
+	/* Enable XOFF signal from RX FIFO (we enabled it during NIC
+	 * initialisation but it may read back as 0) */
+	EFX_SET_OWORD_FIELD(reg, FRF_AZ_RX_XOFF_MAC_EN, 1);
+	/* Unisolate the MAC -> RX */
+	if (efx_nic_rev(efx) >= EFX_REV_FALCON_B0)
+		EFX_SET_OWORD_FIELD(reg, FRF_BZ_RX_INGR_EN, !isolate);
+	efx_writeo(efx, &reg, FR_AZ_RX_CFG);
+}
+
+static void falcon_stats_request(struct efx_nic *efx)
+{
+	struct falcon_nic_data *nic_data = efx->nic_data;
+	efx_oword_t reg;
+
+	WARN_ON(nic_data->stats_pending);
+	WARN_ON(nic_data->stats_disable_count);
+
+	if (nic_data->stats_dma_done == NULL)
+		return;	/* no mac selected */
+
+	*nic_data->stats_dma_done = FALCON_STATS_NOT_DONE;
+	nic_data->stats_pending = true;
+	wmb(); /* ensure done flag is clear */
+
+	/* Initiate DMA transfer of stats */
+	EFX_POPULATE_OWORD_2(reg,
+			     FRF_AB_MAC_STAT_DMA_CMD, 1,
+			     FRF_AB_MAC_STAT_DMA_ADR,
+			     efx->stats_buffer.dma_addr);
+	efx_writeo(efx, &reg, FR_AB_MAC_STAT_DMA);
+
+	mod_timer(&nic_data->stats_timer, round_jiffies_up(jiffies + HZ / 2));
+}
+
+static void falcon_stats_complete(struct efx_nic *efx)
+{
+	struct falcon_nic_data *nic_data = efx->nic_data;
+
+	if (!nic_data->stats_pending)
+		return;
+
+	nic_data->stats_pending = 0;
+	if (*nic_data->stats_dma_done == FALCON_STATS_DONE) {
+		rmb(); /* read the done flag before the stats */
+		efx->mac_op->update_stats(efx);
+	} else {
+		netif_err(efx, hw, efx->net_dev,
+			  "timed out waiting for statistics\n");
+	}
+}
+
+static void falcon_stats_timer_func(unsigned long context)
+{
+	struct efx_nic *efx = (struct efx_nic *)context;
+	struct falcon_nic_data *nic_data = efx->nic_data;
+
+	spin_lock(&efx->stats_lock);
+
+	falcon_stats_complete(efx);
+	if (nic_data->stats_disable_count == 0)
+		falcon_stats_request(efx);
+
+	spin_unlock(&efx->stats_lock);
+}
+
+static bool falcon_loopback_link_poll(struct efx_nic *efx)
+{
+	struct efx_link_state old_state = efx->link_state;
+
+	WARN_ON(!mutex_is_locked(&efx->mac_lock));
+	WARN_ON(!LOOPBACK_INTERNAL(efx));
+
+	efx->link_state.fd = true;
+	efx->link_state.fc = efx->wanted_fc;
+	efx->link_state.up = true;
+	efx->link_state.speed = 10000;
+
+	return !efx_link_state_equal(&efx->link_state, &old_state);
+}
+
+static int falcon_reconfigure_port(struct efx_nic *efx)
+{
+	int rc;
+
+	WARN_ON(efx_nic_rev(efx) > EFX_REV_FALCON_B0);
+
+	/* Poll the PHY link state *before* reconfiguring it. This means we
+	 * will pick up the correct speed (in loopback) to select the correct
+	 * MAC.
+	 */
+	if (LOOPBACK_INTERNAL(efx))
+		falcon_loopback_link_poll(efx);
+	else
+		efx->phy_op->poll(efx);
+
+	falcon_stop_nic_stats(efx);
+	falcon_deconfigure_mac_wrapper(efx);
+
+	falcon_reset_macs(efx);
+
+	efx->phy_op->reconfigure(efx);
+	rc = efx->mac_op->reconfigure(efx);
+	BUG_ON(rc);
+
+	falcon_start_nic_stats(efx);
+
+	/* Synchronise efx->link_state with the kernel */
+	efx_link_status_changed(efx);
+
+	return 0;
+}
+
+/**************************************************************************
+ *
+ * PHY access via GMII
+ *
+ **************************************************************************
+ */
+
+/* Wait for GMII access to complete */
+static int falcon_gmii_wait(struct efx_nic *efx)
+{
+	efx_oword_t md_stat;
+	int count;
+
+	/* wait up to 50ms - taken max from datasheet */
+	for (count = 0; count < 5000; count++) {
+		efx_reado(efx, &md_stat, FR_AB_MD_STAT);
+		if (EFX_OWORD_FIELD(md_stat, FRF_AB_MD_BSY) == 0) {
+			if (EFX_OWORD_FIELD(md_stat, FRF_AB_MD_LNFL) != 0 ||
+			    EFX_OWORD_FIELD(md_stat, FRF_AB_MD_BSERR) != 0) {
+				netif_err(efx, hw, efx->net_dev,
+					  "error from GMII access "
+					  EFX_OWORD_FMT"\n",
+					  EFX_OWORD_VAL(md_stat));
+				return -EIO;
+			}
+			return 0;
+		}
+		udelay(10);
+	}
+	netif_err(efx, hw, efx->net_dev, "timed out waiting for GMII\n");
+	return -ETIMEDOUT;
+}
+
+/* Write an MDIO register of a PHY connected to Falcon. */
+static int falcon_mdio_write(struct net_device *net_dev,
+			     int prtad, int devad, u16 addr, u16 value)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	struct falcon_nic_data *nic_data = efx->nic_data;
+	efx_oword_t reg;
+	int rc;
+
+	netif_vdbg(efx, hw, efx->net_dev,
+		   "writing MDIO %d register %d.%d with 0x%04x\n",
+		    prtad, devad, addr, value);
+
+	mutex_lock(&nic_data->mdio_lock);
+
+	/* Check MDIO not currently being accessed */
+	rc = falcon_gmii_wait(efx);
+	if (rc)
+		goto out;
+
+	/* Write the address/ID register */
+	EFX_POPULATE_OWORD_1(reg, FRF_AB_MD_PHY_ADR, addr);
+	efx_writeo(efx, &reg, FR_AB_MD_PHY_ADR);
+
+	EFX_POPULATE_OWORD_2(reg, FRF_AB_MD_PRT_ADR, prtad,
+			     FRF_AB_MD_DEV_ADR, devad);
+	efx_writeo(efx, &reg, FR_AB_MD_ID);
+
+	/* Write data */
+	EFX_POPULATE_OWORD_1(reg, FRF_AB_MD_TXD, value);
+	efx_writeo(efx, &reg, FR_AB_MD_TXD);
+
+	EFX_POPULATE_OWORD_2(reg,
+			     FRF_AB_MD_WRC, 1,
+			     FRF_AB_MD_GC, 0);
+	efx_writeo(efx, &reg, FR_AB_MD_CS);
+
+	/* Wait for data to be written */
+	rc = falcon_gmii_wait(efx);
+	if (rc) {
+		/* Abort the write operation */
+		EFX_POPULATE_OWORD_2(reg,
+				     FRF_AB_MD_WRC, 0,
+				     FRF_AB_MD_GC, 1);
+		efx_writeo(efx, &reg, FR_AB_MD_CS);
+		udelay(10);
+	}
+
+out:
+	mutex_unlock(&nic_data->mdio_lock);
+	return rc;
+}
+
+/* Read an MDIO register of a PHY connected to Falcon. */
+static int falcon_mdio_read(struct net_device *net_dev,
+			    int prtad, int devad, u16 addr)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	struct falcon_nic_data *nic_data = efx->nic_data;
+	efx_oword_t reg;
+	int rc;
+
+	mutex_lock(&nic_data->mdio_lock);
+
+	/* Check MDIO not currently being accessed */
+	rc = falcon_gmii_wait(efx);
+	if (rc)
+		goto out;
+
+	EFX_POPULATE_OWORD_1(reg, FRF_AB_MD_PHY_ADR, addr);
+	efx_writeo(efx, &reg, FR_AB_MD_PHY_ADR);
+
+	EFX_POPULATE_OWORD_2(reg, FRF_AB_MD_PRT_ADR, prtad,
+			     FRF_AB_MD_DEV_ADR, devad);
+	efx_writeo(efx, &reg, FR_AB_MD_ID);
+
+	/* Request data to be read */
+	EFX_POPULATE_OWORD_2(reg, FRF_AB_MD_RDC, 1, FRF_AB_MD_GC, 0);
+	efx_writeo(efx, &reg, FR_AB_MD_CS);
+
+	/* Wait for data to become available */
+	rc = falcon_gmii_wait(efx);
+	if (rc == 0) {
+		efx_reado(efx, &reg, FR_AB_MD_RXD);
+		rc = EFX_OWORD_FIELD(reg, FRF_AB_MD_RXD);
+		netif_vdbg(efx, hw, efx->net_dev,
+			   "read from MDIO %d register %d.%d, got %04x\n",
+			   prtad, devad, addr, rc);
+	} else {
+		/* Abort the read operation */
+		EFX_POPULATE_OWORD_2(reg,
+				     FRF_AB_MD_RIC, 0,
+				     FRF_AB_MD_GC, 1);
+		efx_writeo(efx, &reg, FR_AB_MD_CS);
+
+		netif_dbg(efx, hw, efx->net_dev,
+			  "read from MDIO %d register %d.%d, got error %d\n",
+			  prtad, devad, addr, rc);
+	}
+
+out:
+	mutex_unlock(&nic_data->mdio_lock);
+	return rc;
+}
+
+/* This call is responsible for hooking in the MAC and PHY operations */
+static int falcon_probe_port(struct efx_nic *efx)
+{
+	struct falcon_nic_data *nic_data = efx->nic_data;
+	int rc;
+
+	switch (efx->phy_type) {
+	case PHY_TYPE_SFX7101:
+		efx->phy_op = &falcon_sfx7101_phy_ops;
+		break;
+	case PHY_TYPE_QT2022C2:
+	case PHY_TYPE_QT2025C:
+		efx->phy_op = &falcon_qt202x_phy_ops;
+		break;
+	case PHY_TYPE_TXC43128:
+		efx->phy_op = &falcon_txc_phy_ops;
+		break;
+	default:
+		netif_err(efx, probe, efx->net_dev, "Unknown PHY type %d\n",
+			  efx->phy_type);
+		return -ENODEV;
+	}
+
+	/* Fill out MDIO structure and loopback modes */
+	mutex_init(&nic_data->mdio_lock);
+	efx->mdio.mdio_read = falcon_mdio_read;
+	efx->mdio.mdio_write = falcon_mdio_write;
+	rc = efx->phy_op->probe(efx);
+	if (rc != 0)
+		return rc;
+
+	/* Initial assumption */
+	efx->link_state.speed = 10000;
+	efx->link_state.fd = true;
+
+	/* Hardware flow ctrl. FalconA RX FIFO too small for pause generation */
+	if (efx_nic_rev(efx) >= EFX_REV_FALCON_B0)
+		efx->wanted_fc = EFX_FC_RX | EFX_FC_TX;
+	else
+		efx->wanted_fc = EFX_FC_RX;
+	if (efx->mdio.mmds & MDIO_DEVS_AN)
+		efx->wanted_fc |= EFX_FC_AUTO;
+
+	/* Allocate buffer for stats */
+	rc = efx_nic_alloc_buffer(efx, &efx->stats_buffer,
+				  FALCON_MAC_STATS_SIZE);
+	if (rc)
+		return rc;
+	netif_dbg(efx, probe, efx->net_dev,
+		  "stats buffer at %llx (virt %p phys %llx)\n",
+		  (u64)efx->stats_buffer.dma_addr,
+		  efx->stats_buffer.addr,
+		  (u64)virt_to_phys(efx->stats_buffer.addr));
+	nic_data->stats_dma_done = efx->stats_buffer.addr + XgDmaDone_offset;
+
+	return 0;
+}
+
+static void falcon_remove_port(struct efx_nic *efx)
+{
+	efx->phy_op->remove(efx);
+	efx_nic_free_buffer(efx, &efx->stats_buffer);
+}
+
+/* Global events are basically PHY events */
+static bool
+falcon_handle_global_event(struct efx_channel *channel, efx_qword_t *event)
+{
+	struct efx_nic *efx = channel->efx;
+	struct falcon_nic_data *nic_data = efx->nic_data;
+
+	if (EFX_QWORD_FIELD(*event, FSF_AB_GLB_EV_G_PHY0_INTR) ||
+	    EFX_QWORD_FIELD(*event, FSF_AB_GLB_EV_XG_PHY0_INTR) ||
+	    EFX_QWORD_FIELD(*event, FSF_AB_GLB_EV_XFP_PHY0_INTR))
+		/* Ignored */
+		return true;
+
+	if ((efx_nic_rev(efx) == EFX_REV_FALCON_B0) &&
+	    EFX_QWORD_FIELD(*event, FSF_BB_GLB_EV_XG_MGT_INTR)) {
+		nic_data->xmac_poll_required = true;
+		return true;
+	}
+
+	if (efx_nic_rev(efx) <= EFX_REV_FALCON_A1 ?
+	    EFX_QWORD_FIELD(*event, FSF_AA_GLB_EV_RX_RECOVERY) :
+	    EFX_QWORD_FIELD(*event, FSF_BB_GLB_EV_RX_RECOVERY)) {
+		netif_err(efx, rx_err, efx->net_dev,
+			  "channel %d seen global RX_RESET event. Resetting.\n",
+			  channel->channel);
+
+		atomic_inc(&efx->rx_reset);
+		efx_schedule_reset(efx, EFX_WORKAROUND_6555(efx) ?
+				   RESET_TYPE_RX_RECOVERY : RESET_TYPE_DISABLE);
+		return true;
+	}
+
+	return false;
+}
+
+/**************************************************************************
+ *
+ * Falcon test code
+ *
+ **************************************************************************/
+
+static int
+falcon_read_nvram(struct efx_nic *efx, struct falcon_nvconfig *nvconfig_out)
+{
+	struct falcon_nic_data *nic_data = efx->nic_data;
+	struct falcon_nvconfig *nvconfig;
+	struct efx_spi_device *spi;
+	void *region;
+	int rc, magic_num, struct_ver;
+	__le16 *word, *limit;
+	u32 csum;
+
+	if (efx_spi_present(&nic_data->spi_flash))
+		spi = &nic_data->spi_flash;
+	else if (efx_spi_present(&nic_data->spi_eeprom))
+		spi = &nic_data->spi_eeprom;
+	else
+		return -EINVAL;
+
+	region = kmalloc(FALCON_NVCONFIG_END, GFP_KERNEL);
+	if (!region)
+		return -ENOMEM;
+	nvconfig = region + FALCON_NVCONFIG_OFFSET;
+
+	mutex_lock(&nic_data->spi_lock);
+	rc = falcon_spi_read(efx, spi, 0, FALCON_NVCONFIG_END, NULL, region);
+	mutex_unlock(&nic_data->spi_lock);
+	if (rc) {
+		netif_err(efx, hw, efx->net_dev, "Failed to read %s\n",
+			  efx_spi_present(&nic_data->spi_flash) ?
+			  "flash" : "EEPROM");
+		rc = -EIO;
+		goto out;
+	}
+
+	magic_num = le16_to_cpu(nvconfig->board_magic_num);
+	struct_ver = le16_to_cpu(nvconfig->board_struct_ver);
+
+	rc = -EINVAL;
+	if (magic_num != FALCON_NVCONFIG_BOARD_MAGIC_NUM) {
+		netif_err(efx, hw, efx->net_dev,
+			  "NVRAM bad magic 0x%x\n", magic_num);
+		goto out;
+	}
+	if (struct_ver < 2) {
+		netif_err(efx, hw, efx->net_dev,
+			  "NVRAM has ancient version 0x%x\n", struct_ver);
+		goto out;
+	} else if (struct_ver < 4) {
+		word = &nvconfig->board_magic_num;
+		limit = (__le16 *) (nvconfig + 1);
+	} else {
+		word = region;
+		limit = region + FALCON_NVCONFIG_END;
+	}
+	for (csum = 0; word < limit; ++word)
+		csum += le16_to_cpu(*word);
+
+	if (~csum & 0xffff) {
+		netif_err(efx, hw, efx->net_dev,
+			  "NVRAM has incorrect checksum\n");
+		goto out;
+	}
+
+	rc = 0;
+	if (nvconfig_out)
+		memcpy(nvconfig_out, nvconfig, sizeof(*nvconfig));
+
+ out:
+	kfree(region);
+	return rc;
+}
+
+static int falcon_test_nvram(struct efx_nic *efx)
+{
+	return falcon_read_nvram(efx, NULL);
+}
+
+static const struct efx_nic_register_test falcon_b0_register_tests[] = {
+	{ FR_AZ_ADR_REGION,
+	  EFX_OWORD32(0x0003FFFF, 0x0003FFFF, 0x0003FFFF, 0x0003FFFF) },
+	{ FR_AZ_RX_CFG,
+	  EFX_OWORD32(0xFFFFFFFE, 0x00017FFF, 0x00000000, 0x00000000) },
+	{ FR_AZ_TX_CFG,
+	  EFX_OWORD32(0x7FFF0037, 0x00000000, 0x00000000, 0x00000000) },
+	{ FR_AZ_TX_RESERVED,
+	  EFX_OWORD32(0xFFFEFE80, 0x1FFFFFFF, 0x020000FE, 0x007FFFFF) },
+	{ FR_AB_MAC_CTRL,
+	  EFX_OWORD32(0xFFFF0000, 0x00000000, 0x00000000, 0x00000000) },
+	{ FR_AZ_SRM_TX_DC_CFG,
+	  EFX_OWORD32(0x001FFFFF, 0x00000000, 0x00000000, 0x00000000) },
+	{ FR_AZ_RX_DC_CFG,
+	  EFX_OWORD32(0x0000000F, 0x00000000, 0x00000000, 0x00000000) },
+	{ FR_AZ_RX_DC_PF_WM,
+	  EFX_OWORD32(0x000003FF, 0x00000000, 0x00000000, 0x00000000) },
+	{ FR_BZ_DP_CTRL,
+	  EFX_OWORD32(0x00000FFF, 0x00000000, 0x00000000, 0x00000000) },
+	{ FR_AB_GM_CFG2,
+	  EFX_OWORD32(0x00007337, 0x00000000, 0x00000000, 0x00000000) },
+	{ FR_AB_GMF_CFG0,
+	  EFX_OWORD32(0x00001F1F, 0x00000000, 0x00000000, 0x00000000) },
+	{ FR_AB_XM_GLB_CFG,
+	  EFX_OWORD32(0x00000C68, 0x00000000, 0x00000000, 0x00000000) },
+	{ FR_AB_XM_TX_CFG,
+	  EFX_OWORD32(0x00080164, 0x00000000, 0x00000000, 0x00000000) },
+	{ FR_AB_XM_RX_CFG,
+	  EFX_OWORD32(0x07100A0C, 0x00000000, 0x00000000, 0x00000000) },
+	{ FR_AB_XM_RX_PARAM,
+	  EFX_OWORD32(0x00001FF8, 0x00000000, 0x00000000, 0x00000000) },
+	{ FR_AB_XM_FC,
+	  EFX_OWORD32(0xFFFF0001, 0x00000000, 0x00000000, 0x00000000) },
+	{ FR_AB_XM_ADR_LO,
+	  EFX_OWORD32(0xFFFFFFFF, 0x00000000, 0x00000000, 0x00000000) },
+	{ FR_AB_XX_SD_CTL,
+	  EFX_OWORD32(0x0003FF0F, 0x00000000, 0x00000000, 0x00000000) },
+};
+
+static int falcon_b0_test_registers(struct efx_nic *efx)
+{
+	return efx_nic_test_registers(efx, falcon_b0_register_tests,
+				      ARRAY_SIZE(falcon_b0_register_tests));
+}
+
+/**************************************************************************
+ *
+ * Device reset
+ *
+ **************************************************************************
+ */
+
+static enum reset_type falcon_map_reset_reason(enum reset_type reason)
+{
+	switch (reason) {
+	case RESET_TYPE_RX_RECOVERY:
+	case RESET_TYPE_RX_DESC_FETCH:
+	case RESET_TYPE_TX_DESC_FETCH:
+	case RESET_TYPE_TX_SKIP:
+		/* These can occasionally occur due to hardware bugs.
+		 * We try to reset without disrupting the link.
+		 */
+		return RESET_TYPE_INVISIBLE;
+	default:
+		return RESET_TYPE_ALL;
+	}
+}
+
+static int falcon_map_reset_flags(u32 *flags)
+{
+	enum {
+		FALCON_RESET_INVISIBLE = (ETH_RESET_DMA | ETH_RESET_FILTER |
+					  ETH_RESET_OFFLOAD | ETH_RESET_MAC),
+		FALCON_RESET_ALL = FALCON_RESET_INVISIBLE | ETH_RESET_PHY,
+		FALCON_RESET_WORLD = FALCON_RESET_ALL | ETH_RESET_IRQ,
+	};
+
+	if ((*flags & FALCON_RESET_WORLD) == FALCON_RESET_WORLD) {
+		*flags &= ~FALCON_RESET_WORLD;
+		return RESET_TYPE_WORLD;
+	}
+
+	if ((*flags & FALCON_RESET_ALL) == FALCON_RESET_ALL) {
+		*flags &= ~FALCON_RESET_ALL;
+		return RESET_TYPE_ALL;
+	}
+
+	if ((*flags & FALCON_RESET_INVISIBLE) == FALCON_RESET_INVISIBLE) {
+		*flags &= ~FALCON_RESET_INVISIBLE;
+		return RESET_TYPE_INVISIBLE;
+	}
+
+	return -EINVAL;
+}
+
+/* Resets NIC to known state.  This routine must be called in process
+ * context and is allowed to sleep. */
+static int __falcon_reset_hw(struct efx_nic *efx, enum reset_type method)
+{
+	struct falcon_nic_data *nic_data = efx->nic_data;
+	efx_oword_t glb_ctl_reg_ker;
+	int rc;
+
+	netif_dbg(efx, hw, efx->net_dev, "performing %s hardware reset\n",
+		  RESET_TYPE(method));
+
+	/* Initiate device reset */
+	if (method == RESET_TYPE_WORLD) {
+		rc = pci_save_state(efx->pci_dev);
+		if (rc) {
+			netif_err(efx, drv, efx->net_dev,
+				  "failed to backup PCI state of primary "
+				  "function prior to hardware reset\n");
+			goto fail1;
+		}
+		if (efx_nic_is_dual_func(efx)) {
+			rc = pci_save_state(nic_data->pci_dev2);
+			if (rc) {
+				netif_err(efx, drv, efx->net_dev,
+					  "failed to backup PCI state of "
+					  "secondary function prior to "
+					  "hardware reset\n");
+				goto fail2;
+			}
+		}
+
+		EFX_POPULATE_OWORD_2(glb_ctl_reg_ker,
+				     FRF_AB_EXT_PHY_RST_DUR,
+				     FFE_AB_EXT_PHY_RST_DUR_10240US,
+				     FRF_AB_SWRST, 1);
+	} else {
+		EFX_POPULATE_OWORD_7(glb_ctl_reg_ker,
+				     /* exclude PHY from "invisible" reset */
+				     FRF_AB_EXT_PHY_RST_CTL,
+				     method == RESET_TYPE_INVISIBLE,
+				     /* exclude EEPROM/flash and PCIe */
+				     FRF_AB_PCIE_CORE_RST_CTL, 1,
+				     FRF_AB_PCIE_NSTKY_RST_CTL, 1,
+				     FRF_AB_PCIE_SD_RST_CTL, 1,
+				     FRF_AB_EE_RST_CTL, 1,
+				     FRF_AB_EXT_PHY_RST_DUR,
+				     FFE_AB_EXT_PHY_RST_DUR_10240US,
+				     FRF_AB_SWRST, 1);
+	}
+	efx_writeo(efx, &glb_ctl_reg_ker, FR_AB_GLB_CTL);
+
+	netif_dbg(efx, hw, efx->net_dev, "waiting for hardware reset\n");
+	schedule_timeout_uninterruptible(HZ / 20);
+
+	/* Restore PCI configuration if needed */
+	if (method == RESET_TYPE_WORLD) {
+		if (efx_nic_is_dual_func(efx))
+			pci_restore_state(nic_data->pci_dev2);
+		pci_restore_state(efx->pci_dev);
+		netif_dbg(efx, drv, efx->net_dev,
+			  "successfully restored PCI config\n");
+	}
+
+	/* Assert that reset complete */
+	efx_reado(efx, &glb_ctl_reg_ker, FR_AB_GLB_CTL);
+	if (EFX_OWORD_FIELD(glb_ctl_reg_ker, FRF_AB_SWRST) != 0) {
+		rc = -ETIMEDOUT;
+		netif_err(efx, hw, efx->net_dev,
+			  "timed out waiting for hardware reset\n");
+		goto fail3;
+	}
+	netif_dbg(efx, hw, efx->net_dev, "hardware reset complete\n");
+
+	return 0;
+
+	/* pci_save_state() and pci_restore_state() MUST be called in pairs */
+fail2:
+	pci_restore_state(efx->pci_dev);
+fail1:
+fail3:
+	return rc;
+}
+
+static int falcon_reset_hw(struct efx_nic *efx, enum reset_type method)
+{
+	struct falcon_nic_data *nic_data = efx->nic_data;
+	int rc;
+
+	mutex_lock(&nic_data->spi_lock);
+	rc = __falcon_reset_hw(efx, method);
+	mutex_unlock(&nic_data->spi_lock);
+
+	return rc;
+}
+
+static void falcon_monitor(struct efx_nic *efx)
+{
+	bool link_changed;
+	int rc;
+
+	BUG_ON(!mutex_is_locked(&efx->mac_lock));
+
+	rc = falcon_board(efx)->type->monitor(efx);
+	if (rc) {
+		netif_err(efx, hw, efx->net_dev,
+			  "Board sensor %s; shutting down PHY\n",
+			  (rc == -ERANGE) ? "reported fault" : "failed");
+		efx->phy_mode |= PHY_MODE_LOW_POWER;
+		rc = __efx_reconfigure_port(efx);
+		WARN_ON(rc);
+	}
+
+	if (LOOPBACK_INTERNAL(efx))
+		link_changed = falcon_loopback_link_poll(efx);
+	else
+		link_changed = efx->phy_op->poll(efx);
+
+	if (link_changed) {
+		falcon_stop_nic_stats(efx);
+		falcon_deconfigure_mac_wrapper(efx);
+
+		falcon_reset_macs(efx);
+		rc = efx->mac_op->reconfigure(efx);
+		BUG_ON(rc);
+
+		falcon_start_nic_stats(efx);
+
+		efx_link_status_changed(efx);
+	}
+
+	falcon_poll_xmac(efx);
+}
+
+/* Zeroes out the SRAM contents.  This routine must be called in
+ * process context and is allowed to sleep.
+ */
+static int falcon_reset_sram(struct efx_nic *efx)
+{
+	efx_oword_t srm_cfg_reg_ker, gpio_cfg_reg_ker;
+	int count;
+
+	/* Set the SRAM wake/sleep GPIO appropriately. */
+	efx_reado(efx, &gpio_cfg_reg_ker, FR_AB_GPIO_CTL);
+	EFX_SET_OWORD_FIELD(gpio_cfg_reg_ker, FRF_AB_GPIO1_OEN, 1);
+	EFX_SET_OWORD_FIELD(gpio_cfg_reg_ker, FRF_AB_GPIO1_OUT, 1);
+	efx_writeo(efx, &gpio_cfg_reg_ker, FR_AB_GPIO_CTL);
+
+	/* Initiate SRAM reset */
+	EFX_POPULATE_OWORD_2(srm_cfg_reg_ker,
+			     FRF_AZ_SRM_INIT_EN, 1,
+			     FRF_AZ_SRM_NB_SZ, 0);
+	efx_writeo(efx, &srm_cfg_reg_ker, FR_AZ_SRM_CFG);
+
+	/* Wait for SRAM reset to complete */
+	count = 0;
+	do {
+		netif_dbg(efx, hw, efx->net_dev,
+			  "waiting for SRAM reset (attempt %d)...\n", count);
+
+		/* SRAM reset is slow; expect around 16ms */
+		schedule_timeout_uninterruptible(HZ / 50);
+
+		/* Check for reset complete */
+		efx_reado(efx, &srm_cfg_reg_ker, FR_AZ_SRM_CFG);
+		if (!EFX_OWORD_FIELD(srm_cfg_reg_ker, FRF_AZ_SRM_INIT_EN)) {
+			netif_dbg(efx, hw, efx->net_dev,
+				  "SRAM reset complete\n");
+
+			return 0;
+		}
+	} while (++count < 20);	/* wait up to 0.4 sec */
+
+	netif_err(efx, hw, efx->net_dev, "timed out waiting for SRAM reset\n");
+	return -ETIMEDOUT;
+}
+
+static void falcon_spi_device_init(struct efx_nic *efx,
+				  struct efx_spi_device *spi_device,
+				  unsigned int device_id, u32 device_type)
+{
+	if (device_type != 0) {
+		spi_device->device_id = device_id;
+		spi_device->size =
+			1 << SPI_DEV_TYPE_FIELD(device_type, SPI_DEV_TYPE_SIZE);
+		spi_device->addr_len =
+			SPI_DEV_TYPE_FIELD(device_type, SPI_DEV_TYPE_ADDR_LEN);
+		spi_device->munge_address = (spi_device->size == 1 << 9 &&
+					     spi_device->addr_len == 1);
+		spi_device->erase_command =
+			SPI_DEV_TYPE_FIELD(device_type, SPI_DEV_TYPE_ERASE_CMD);
+		spi_device->erase_size =
+			1 << SPI_DEV_TYPE_FIELD(device_type,
+						SPI_DEV_TYPE_ERASE_SIZE);
+		spi_device->block_size =
+			1 << SPI_DEV_TYPE_FIELD(device_type,
+						SPI_DEV_TYPE_BLOCK_SIZE);
+	} else {
+		spi_device->size = 0;
+	}
+}
+
+/* Extract non-volatile configuration */
+static int falcon_probe_nvconfig(struct efx_nic *efx)
+{
+	struct falcon_nic_data *nic_data = efx->nic_data;
+	struct falcon_nvconfig *nvconfig;
+	int rc;
+
+	nvconfig = kmalloc(sizeof(*nvconfig), GFP_KERNEL);
+	if (!nvconfig)
+		return -ENOMEM;
+
+	rc = falcon_read_nvram(efx, nvconfig);
+	if (rc)
+		goto out;
+
+	efx->phy_type = nvconfig->board_v2.port0_phy_type;
+	efx->mdio.prtad = nvconfig->board_v2.port0_phy_addr;
+
+	if (le16_to_cpu(nvconfig->board_struct_ver) >= 3) {
+		falcon_spi_device_init(
+			efx, &nic_data->spi_flash, FFE_AB_SPI_DEVICE_FLASH,
+			le32_to_cpu(nvconfig->board_v3
+				    .spi_device_type[FFE_AB_SPI_DEVICE_FLASH]));
+		falcon_spi_device_init(
+			efx, &nic_data->spi_eeprom, FFE_AB_SPI_DEVICE_EEPROM,
+			le32_to_cpu(nvconfig->board_v3
+				    .spi_device_type[FFE_AB_SPI_DEVICE_EEPROM]));
+	}
+
+	/* Read the MAC addresses */
+	memcpy(efx->net_dev->perm_addr, nvconfig->mac_address[0], ETH_ALEN);
+
+	netif_dbg(efx, probe, efx->net_dev, "PHY is %d phy_id %d\n",
+		  efx->phy_type, efx->mdio.prtad);
+
+	rc = falcon_probe_board(efx,
+				le16_to_cpu(nvconfig->board_v2.board_revision));
+out:
+	kfree(nvconfig);
+	return rc;
+}
+
+/* Probe all SPI devices on the NIC */
+static void falcon_probe_spi_devices(struct efx_nic *efx)
+{
+	struct falcon_nic_data *nic_data = efx->nic_data;
+	efx_oword_t nic_stat, gpio_ctl, ee_vpd_cfg;
+	int boot_dev;
+
+	efx_reado(efx, &gpio_ctl, FR_AB_GPIO_CTL);
+	efx_reado(efx, &nic_stat, FR_AB_NIC_STAT);
+	efx_reado(efx, &ee_vpd_cfg, FR_AB_EE_VPD_CFG0);
+
+	if (EFX_OWORD_FIELD(gpio_ctl, FRF_AB_GPIO3_PWRUP_VALUE)) {
+		boot_dev = (EFX_OWORD_FIELD(nic_stat, FRF_AB_SF_PRST) ?
+			    FFE_AB_SPI_DEVICE_FLASH : FFE_AB_SPI_DEVICE_EEPROM);
+		netif_dbg(efx, probe, efx->net_dev, "Booted from %s\n",
+			  boot_dev == FFE_AB_SPI_DEVICE_FLASH ?
+			  "flash" : "EEPROM");
+	} else {
+		/* Disable VPD and set clock dividers to safe
+		 * values for initial programming. */
+		boot_dev = -1;
+		netif_dbg(efx, probe, efx->net_dev,
+			  "Booted from internal ASIC settings;"
+			  " setting SPI config\n");
+		EFX_POPULATE_OWORD_3(ee_vpd_cfg, FRF_AB_EE_VPD_EN, 0,
+				     /* 125 MHz / 7 ~= 20 MHz */
+				     FRF_AB_EE_SF_CLOCK_DIV, 7,
+				     /* 125 MHz / 63 ~= 2 MHz */
+				     FRF_AB_EE_EE_CLOCK_DIV, 63);
+		efx_writeo(efx, &ee_vpd_cfg, FR_AB_EE_VPD_CFG0);
+	}
+
+	mutex_init(&nic_data->spi_lock);
+
+	if (boot_dev == FFE_AB_SPI_DEVICE_FLASH)
+		falcon_spi_device_init(efx, &nic_data->spi_flash,
+				       FFE_AB_SPI_DEVICE_FLASH,
+				       default_flash_type);
+	if (boot_dev == FFE_AB_SPI_DEVICE_EEPROM)
+		falcon_spi_device_init(efx, &nic_data->spi_eeprom,
+				       FFE_AB_SPI_DEVICE_EEPROM,
+				       large_eeprom_type);
+}
+
+static int falcon_probe_nic(struct efx_nic *efx)
+{
+	struct falcon_nic_data *nic_data;
+	struct falcon_board *board;
+	int rc;
+
+	/* Allocate storage for hardware specific data */
+	nic_data = kzalloc(sizeof(*nic_data), GFP_KERNEL);
+	if (!nic_data)
+		return -ENOMEM;
+	efx->nic_data = nic_data;
+
+	rc = -ENODEV;
+
+	if (efx_nic_fpga_ver(efx) != 0) {
+		netif_err(efx, probe, efx->net_dev,
+			  "Falcon FPGA not supported\n");
+		goto fail1;
+	}
+
+	if (efx_nic_rev(efx) <= EFX_REV_FALCON_A1) {
+		efx_oword_t nic_stat;
+		struct pci_dev *dev;
+		u8 pci_rev = efx->pci_dev->revision;
+
+		if ((pci_rev == 0xff) || (pci_rev == 0)) {
+			netif_err(efx, probe, efx->net_dev,
+				  "Falcon rev A0 not supported\n");
+			goto fail1;
+		}
+		efx_reado(efx, &nic_stat, FR_AB_NIC_STAT);
+		if (EFX_OWORD_FIELD(nic_stat, FRF_AB_STRAP_10G) == 0) {
+			netif_err(efx, probe, efx->net_dev,
+				  "Falcon rev A1 1G not supported\n");
+			goto fail1;
+		}
+		if (EFX_OWORD_FIELD(nic_stat, FRF_AA_STRAP_PCIE) == 0) {
+			netif_err(efx, probe, efx->net_dev,
+				  "Falcon rev A1 PCI-X not supported\n");
+			goto fail1;
+		}
+
+		dev = pci_dev_get(efx->pci_dev);
+		while ((dev = pci_get_device(EFX_VENDID_SFC, FALCON_A_S_DEVID,
+					     dev))) {
+			if (dev->bus == efx->pci_dev->bus &&
+			    dev->devfn == efx->pci_dev->devfn + 1) {
+				nic_data->pci_dev2 = dev;
+				break;
+			}
+		}
+		if (!nic_data->pci_dev2) {
+			netif_err(efx, probe, efx->net_dev,
+				  "failed to find secondary function\n");
+			rc = -ENODEV;
+			goto fail2;
+		}
+	}
+
+	/* Now we can reset the NIC */
+	rc = __falcon_reset_hw(efx, RESET_TYPE_ALL);
+	if (rc) {
+		netif_err(efx, probe, efx->net_dev, "failed to reset NIC\n");
+		goto fail3;
+	}
+
+	/* Allocate memory for INT_KER */
+	rc = efx_nic_alloc_buffer(efx, &efx->irq_status, sizeof(efx_oword_t));
+	if (rc)
+		goto fail4;
+	BUG_ON(efx->irq_status.dma_addr & 0x0f);
+
+	netif_dbg(efx, probe, efx->net_dev,
+		  "INT_KER at %llx (virt %p phys %llx)\n",
+		  (u64)efx->irq_status.dma_addr,
+		  efx->irq_status.addr,
+		  (u64)virt_to_phys(efx->irq_status.addr));
+
+	falcon_probe_spi_devices(efx);
+
+	/* Read in the non-volatile configuration */
+	rc = falcon_probe_nvconfig(efx);
+	if (rc) {
+		if (rc == -EINVAL)
+			netif_err(efx, probe, efx->net_dev, "NVRAM is invalid\n");
+		goto fail5;
+	}
+
+	/* Initialise I2C adapter */
+	board = falcon_board(efx);
+	board->i2c_adap.owner = THIS_MODULE;
+	board->i2c_data = falcon_i2c_bit_operations;
+	board->i2c_data.data = efx;
+	board->i2c_adap.algo_data = &board->i2c_data;
+	board->i2c_adap.dev.parent = &efx->pci_dev->dev;
+	strlcpy(board->i2c_adap.name, "SFC4000 GPIO",
+		sizeof(board->i2c_adap.name));
+	rc = i2c_bit_add_bus(&board->i2c_adap);
+	if (rc)
+		goto fail5;
+
+	rc = falcon_board(efx)->type->init(efx);
+	if (rc) {
+		netif_err(efx, probe, efx->net_dev,
+			  "failed to initialise board\n");
+		goto fail6;
+	}
+
+	nic_data->stats_disable_count = 1;
+	setup_timer(&nic_data->stats_timer, &falcon_stats_timer_func,
+		    (unsigned long)efx);
+
+	return 0;
+
+ fail6:
+	BUG_ON(i2c_del_adapter(&board->i2c_adap));
+	memset(&board->i2c_adap, 0, sizeof(board->i2c_adap));
+ fail5:
+	efx_nic_free_buffer(efx, &efx->irq_status);
+ fail4:
+ fail3:
+	if (nic_data->pci_dev2) {
+		pci_dev_put(nic_data->pci_dev2);
+		nic_data->pci_dev2 = NULL;
+	}
+ fail2:
+ fail1:
+	kfree(efx->nic_data);
+	return rc;
+}
+
+static void falcon_init_rx_cfg(struct efx_nic *efx)
+{
+	/* Prior to Siena the RX DMA engine will split each frame at
+	 * intervals of RX_USR_BUF_SIZE (32-byte units). We set it to
+	 * be so large that that never happens. */
+	const unsigned huge_buf_size = (3 * 4096) >> 5;
+	/* RX control FIFO thresholds (32 entries) */
+	const unsigned ctrl_xon_thr = 20;
+	const unsigned ctrl_xoff_thr = 25;
+	efx_oword_t reg;
+
+	efx_reado(efx, &reg, FR_AZ_RX_CFG);
+	if (efx_nic_rev(efx) <= EFX_REV_FALCON_A1) {
+		/* Data FIFO size is 5.5K */
+		EFX_SET_OWORD_FIELD(reg, FRF_AA_RX_DESC_PUSH_EN, 0);
+		EFX_SET_OWORD_FIELD(reg, FRF_AA_RX_USR_BUF_SIZE,
+				    huge_buf_size);
+		EFX_SET_OWORD_FIELD(reg, FRF_AA_RX_XON_MAC_TH, 512 >> 8);
+		EFX_SET_OWORD_FIELD(reg, FRF_AA_RX_XOFF_MAC_TH, 2048 >> 8);
+		EFX_SET_OWORD_FIELD(reg, FRF_AA_RX_XON_TX_TH, ctrl_xon_thr);
+		EFX_SET_OWORD_FIELD(reg, FRF_AA_RX_XOFF_TX_TH, ctrl_xoff_thr);
+	} else {
+		/* Data FIFO size is 80K; register fields moved */
+		EFX_SET_OWORD_FIELD(reg, FRF_BZ_RX_DESC_PUSH_EN, 0);
+		EFX_SET_OWORD_FIELD(reg, FRF_BZ_RX_USR_BUF_SIZE,
+				    huge_buf_size);
+		/* Send XON and XOFF at ~3 * max MTU away from empty/full */
+		EFX_SET_OWORD_FIELD(reg, FRF_BZ_RX_XON_MAC_TH, 27648 >> 8);
+		EFX_SET_OWORD_FIELD(reg, FRF_BZ_RX_XOFF_MAC_TH, 54272 >> 8);
+		EFX_SET_OWORD_FIELD(reg, FRF_BZ_RX_XON_TX_TH, ctrl_xon_thr);
+		EFX_SET_OWORD_FIELD(reg, FRF_BZ_RX_XOFF_TX_TH, ctrl_xoff_thr);
+		EFX_SET_OWORD_FIELD(reg, FRF_BZ_RX_INGR_EN, 1);
+
+		/* Enable hash insertion. This is broken for the
+		 * 'Falcon' hash so also select Toeplitz TCP/IPv4 and
+		 * IPv4 hashes. */
+		EFX_SET_OWORD_FIELD(reg, FRF_BZ_RX_HASH_INSRT_HDR, 1);
+		EFX_SET_OWORD_FIELD(reg, FRF_BZ_RX_HASH_ALG, 1);
+		EFX_SET_OWORD_FIELD(reg, FRF_BZ_RX_IP_HASH, 1);
+	}
+	/* Always enable XOFF signal from RX FIFO.  We enable
+	 * or disable transmission of pause frames at the MAC. */
+	EFX_SET_OWORD_FIELD(reg, FRF_AZ_RX_XOFF_MAC_EN, 1);
+	efx_writeo(efx, &reg, FR_AZ_RX_CFG);
+}
+
+/* This call performs hardware-specific global initialisation, such as
+ * defining the descriptor cache sizes and number of RSS channels.
+ * It does not set up any buffers, descriptor rings or event queues.
+ */
+static int falcon_init_nic(struct efx_nic *efx)
+{
+	efx_oword_t temp;
+	int rc;
+
+	/* Use on-chip SRAM */
+	efx_reado(efx, &temp, FR_AB_NIC_STAT);
+	EFX_SET_OWORD_FIELD(temp, FRF_AB_ONCHIP_SRAM, 1);
+	efx_writeo(efx, &temp, FR_AB_NIC_STAT);
+
+	rc = falcon_reset_sram(efx);
+	if (rc)
+		return rc;
+
+	/* Clear the parity enables on the TX data fifos as
+	 * they produce false parity errors because of timing issues
+	 */
+	if (EFX_WORKAROUND_5129(efx)) {
+		efx_reado(efx, &temp, FR_AZ_CSR_SPARE);
+		EFX_SET_OWORD_FIELD(temp, FRF_AB_MEM_PERR_EN_TX_DATA, 0);
+		efx_writeo(efx, &temp, FR_AZ_CSR_SPARE);
+	}
+
+	if (EFX_WORKAROUND_7244(efx)) {
+		efx_reado(efx, &temp, FR_BZ_RX_FILTER_CTL);
+		EFX_SET_OWORD_FIELD(temp, FRF_BZ_UDP_FULL_SRCH_LIMIT, 8);
+		EFX_SET_OWORD_FIELD(temp, FRF_BZ_UDP_WILD_SRCH_LIMIT, 8);
+		EFX_SET_OWORD_FIELD(temp, FRF_BZ_TCP_FULL_SRCH_LIMIT, 8);
+		EFX_SET_OWORD_FIELD(temp, FRF_BZ_TCP_WILD_SRCH_LIMIT, 8);
+		efx_writeo(efx, &temp, FR_BZ_RX_FILTER_CTL);
+	}
+
+	/* XXX This is documented only for Falcon A0/A1 */
+	/* Setup RX.  Wait for descriptor is broken and must
+	 * be disabled.  RXDP recovery shouldn't be needed, but is.
+	 */
+	efx_reado(efx, &temp, FR_AA_RX_SELF_RST);
+	EFX_SET_OWORD_FIELD(temp, FRF_AA_RX_NODESC_WAIT_DIS, 1);
+	EFX_SET_OWORD_FIELD(temp, FRF_AA_RX_SELF_RST_EN, 1);
+	if (EFX_WORKAROUND_5583(efx))
+		EFX_SET_OWORD_FIELD(temp, FRF_AA_RX_ISCSI_DIS, 1);
+	efx_writeo(efx, &temp, FR_AA_RX_SELF_RST);
+
+	/* Do not enable TX_NO_EOP_DISC_EN, since it limits packets to 16
+	 * descriptors (which is bad).
+	 */
+	efx_reado(efx, &temp, FR_AZ_TX_CFG);
+	EFX_SET_OWORD_FIELD(temp, FRF_AZ_TX_NO_EOP_DISC_EN, 0);
+	efx_writeo(efx, &temp, FR_AZ_TX_CFG);
+
+	falcon_init_rx_cfg(efx);
+
+	if (efx_nic_rev(efx) >= EFX_REV_FALCON_B0) {
+		/* Set hash key for IPv4 */
+		memcpy(&temp, efx->rx_hash_key, sizeof(temp));
+		efx_writeo(efx, &temp, FR_BZ_RX_RSS_TKEY);
+
+		/* Set destination of both TX and RX Flush events */
+		EFX_POPULATE_OWORD_1(temp, FRF_BZ_FLS_EVQ_ID, 0);
+		efx_writeo(efx, &temp, FR_BZ_DP_CTRL);
+	}
+
+	efx_nic_init_common(efx);
+
+	return 0;
+}
+
+static void falcon_remove_nic(struct efx_nic *efx)
+{
+	struct falcon_nic_data *nic_data = efx->nic_data;
+	struct falcon_board *board = falcon_board(efx);
+	int rc;
+
+	board->type->fini(efx);
+
+	/* Remove I2C adapter and clear it in preparation for a retry */
+	rc = i2c_del_adapter(&board->i2c_adap);
+	BUG_ON(rc);
+	memset(&board->i2c_adap, 0, sizeof(board->i2c_adap));
+
+	efx_nic_free_buffer(efx, &efx->irq_status);
+
+	__falcon_reset_hw(efx, RESET_TYPE_ALL);
+
+	/* Release the second function after the reset */
+	if (nic_data->pci_dev2) {
+		pci_dev_put(nic_data->pci_dev2);
+		nic_data->pci_dev2 = NULL;
+	}
+
+	/* Tear down the private nic state */
+	kfree(efx->nic_data);
+	efx->nic_data = NULL;
+}
+
+static void falcon_update_nic_stats(struct efx_nic *efx)
+{
+	struct falcon_nic_data *nic_data = efx->nic_data;
+	efx_oword_t cnt;
+
+	if (nic_data->stats_disable_count)
+		return;
+
+	efx_reado(efx, &cnt, FR_AZ_RX_NODESC_DROP);
+	efx->n_rx_nodesc_drop_cnt +=
+		EFX_OWORD_FIELD(cnt, FRF_AB_RX_NODESC_DROP_CNT);
+
+	if (nic_data->stats_pending &&
+	    *nic_data->stats_dma_done == FALCON_STATS_DONE) {
+		nic_data->stats_pending = false;
+		rmb(); /* read the done flag before the stats */
+		efx->mac_op->update_stats(efx);
+	}
+}
+
+void falcon_start_nic_stats(struct efx_nic *efx)
+{
+	struct falcon_nic_data *nic_data = efx->nic_data;
+
+	spin_lock_bh(&efx->stats_lock);
+	if (--nic_data->stats_disable_count == 0)
+		falcon_stats_request(efx);
+	spin_unlock_bh(&efx->stats_lock);
+}
+
+void falcon_stop_nic_stats(struct efx_nic *efx)
+{
+	struct falcon_nic_data *nic_data = efx->nic_data;
+	int i;
+
+	might_sleep();
+
+	spin_lock_bh(&efx->stats_lock);
+	++nic_data->stats_disable_count;
+	spin_unlock_bh(&efx->stats_lock);
+
+	del_timer_sync(&nic_data->stats_timer);
+
+	/* Wait enough time for the most recent transfer to
+	 * complete. */
+	for (i = 0; i < 4 && nic_data->stats_pending; i++) {
+		if (*nic_data->stats_dma_done == FALCON_STATS_DONE)
+			break;
+		msleep(1);
+	}
+
+	spin_lock_bh(&efx->stats_lock);
+	falcon_stats_complete(efx);
+	spin_unlock_bh(&efx->stats_lock);
+}
+
+static void falcon_set_id_led(struct efx_nic *efx, enum efx_led_mode mode)
+{
+	falcon_board(efx)->type->set_id_led(efx, mode);
+}
+
+/**************************************************************************
+ *
+ * Wake on LAN
+ *
+ **************************************************************************
+ */
+
+static void falcon_get_wol(struct efx_nic *efx, struct ethtool_wolinfo *wol)
+{
+	wol->supported = 0;
+	wol->wolopts = 0;
+	memset(&wol->sopass, 0, sizeof(wol->sopass));
+}
+
+static int falcon_set_wol(struct efx_nic *efx, u32 type)
+{
+	if (type != 0)
+		return -EINVAL;
+	return 0;
+}
+
+/**************************************************************************
+ *
+ * Revision-dependent attributes used by efx.c and nic.c
+ *
+ **************************************************************************
+ */
+
+const struct efx_nic_type falcon_a1_nic_type = {
+	.probe = falcon_probe_nic,
+	.remove = falcon_remove_nic,
+	.init = falcon_init_nic,
+	.fini = efx_port_dummy_op_void,
+	.monitor = falcon_monitor,
+	.map_reset_reason = falcon_map_reset_reason,
+	.map_reset_flags = falcon_map_reset_flags,
+	.reset = falcon_reset_hw,
+	.probe_port = falcon_probe_port,
+	.remove_port = falcon_remove_port,
+	.handle_global_event = falcon_handle_global_event,
+	.prepare_flush = falcon_prepare_flush,
+	.update_stats = falcon_update_nic_stats,
+	.start_stats = falcon_start_nic_stats,
+	.stop_stats = falcon_stop_nic_stats,
+	.set_id_led = falcon_set_id_led,
+	.push_irq_moderation = falcon_push_irq_moderation,
+	.push_multicast_hash = falcon_push_multicast_hash,
+	.reconfigure_port = falcon_reconfigure_port,
+	.get_wol = falcon_get_wol,
+	.set_wol = falcon_set_wol,
+	.resume_wol = efx_port_dummy_op_void,
+	.test_nvram = falcon_test_nvram,
+	.default_mac_ops = &falcon_xmac_operations,
+
+	.revision = EFX_REV_FALCON_A1,
+	.mem_map_size = 0x20000,
+	.txd_ptr_tbl_base = FR_AA_TX_DESC_PTR_TBL_KER,
+	.rxd_ptr_tbl_base = FR_AA_RX_DESC_PTR_TBL_KER,
+	.buf_tbl_base = FR_AA_BUF_FULL_TBL_KER,
+	.evq_ptr_tbl_base = FR_AA_EVQ_PTR_TBL_KER,
+	.evq_rptr_tbl_base = FR_AA_EVQ_RPTR_KER,
+	.max_dma_mask = DMA_BIT_MASK(FSF_AZ_TX_KER_BUF_ADDR_WIDTH),
+	.rx_buffer_padding = 0x24,
+	.max_interrupt_mode = EFX_INT_MODE_MSI,
+	.phys_addr_channels = 4,
+	.tx_dc_base = 0x130000,
+	.rx_dc_base = 0x100000,
+	.offload_features = NETIF_F_IP_CSUM,
+};
+
+const struct efx_nic_type falcon_b0_nic_type = {
+	.probe = falcon_probe_nic,
+	.remove = falcon_remove_nic,
+	.init = falcon_init_nic,
+	.fini = efx_port_dummy_op_void,
+	.monitor = falcon_monitor,
+	.map_reset_reason = falcon_map_reset_reason,
+	.map_reset_flags = falcon_map_reset_flags,
+	.reset = falcon_reset_hw,
+	.probe_port = falcon_probe_port,
+	.remove_port = falcon_remove_port,
+	.handle_global_event = falcon_handle_global_event,
+	.prepare_flush = falcon_prepare_flush,
+	.update_stats = falcon_update_nic_stats,
+	.start_stats = falcon_start_nic_stats,
+	.stop_stats = falcon_stop_nic_stats,
+	.set_id_led = falcon_set_id_led,
+	.push_irq_moderation = falcon_push_irq_moderation,
+	.push_multicast_hash = falcon_push_multicast_hash,
+	.reconfigure_port = falcon_reconfigure_port,
+	.get_wol = falcon_get_wol,
+	.set_wol = falcon_set_wol,
+	.resume_wol = efx_port_dummy_op_void,
+	.test_registers = falcon_b0_test_registers,
+	.test_nvram = falcon_test_nvram,
+	.default_mac_ops = &falcon_xmac_operations,
+
+	.revision = EFX_REV_FALCON_B0,
+	/* Map everything up to and including the RSS indirection
+	 * table.  Don't map MSI-X table, MSI-X PBA since Linux
+	 * requires that they not be mapped.  */
+	.mem_map_size = (FR_BZ_RX_INDIRECTION_TBL +
+			 FR_BZ_RX_INDIRECTION_TBL_STEP *
+			 FR_BZ_RX_INDIRECTION_TBL_ROWS),
+	.txd_ptr_tbl_base = FR_BZ_TX_DESC_PTR_TBL,
+	.rxd_ptr_tbl_base = FR_BZ_RX_DESC_PTR_TBL,
+	.buf_tbl_base = FR_BZ_BUF_FULL_TBL,
+	.evq_ptr_tbl_base = FR_BZ_EVQ_PTR_TBL,
+	.evq_rptr_tbl_base = FR_BZ_EVQ_RPTR,
+	.max_dma_mask = DMA_BIT_MASK(FSF_AZ_TX_KER_BUF_ADDR_WIDTH),
+	.rx_buffer_hash_size = 0x10,
+	.rx_buffer_padding = 0,
+	.max_interrupt_mode = EFX_INT_MODE_MSIX,
+	.phys_addr_channels = 32, /* Hardware limit is 64, but the legacy
+				   * interrupt handler only supports 32
+				   * channels */
+	.tx_dc_base = 0x130000,
+	.rx_dc_base = 0x100000,
+	.offload_features = NETIF_F_IP_CSUM | NETIF_F_RXHASH | NETIF_F_NTUPLE,
+};
+
diff --git a/drivers/net/ethernet/sfc/falcon_boards.c b/drivers/net/ethernet/sfc/falcon_boards.c
new file mode 100644
index 0000000..b9cc846
--- /dev/null
+++ b/drivers/net/ethernet/sfc/falcon_boards.c
@@ -0,0 +1,776 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2007-2010 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#include <linux/rtnetlink.h>
+
+#include "net_driver.h"
+#include "phy.h"
+#include "efx.h"
+#include "nic.h"
+#include "workarounds.h"
+
+/* Macros for unpacking the board revision */
+/* The revision info is in host byte order. */
+#define FALCON_BOARD_TYPE(_rev) (_rev >> 8)
+#define FALCON_BOARD_MAJOR(_rev) ((_rev >> 4) & 0xf)
+#define FALCON_BOARD_MINOR(_rev) (_rev & 0xf)
+
+/* Board types */
+#define FALCON_BOARD_SFE4001 0x01
+#define FALCON_BOARD_SFE4002 0x02
+#define FALCON_BOARD_SFE4003 0x03
+#define FALCON_BOARD_SFN4112F 0x52
+
+/* Board temperature is about 15°C above ambient when air flow is
+ * limited.  The maximum acceptable ambient temperature varies
+ * depending on the PHY specifications but the critical temperature
+ * above which we should shut down to avoid damage is 80°C. */
+#define FALCON_BOARD_TEMP_BIAS	15
+#define FALCON_BOARD_TEMP_CRIT	(80 + FALCON_BOARD_TEMP_BIAS)
+
+/* SFC4000 datasheet says: 'The maximum permitted junction temperature
+ * is 125°C; the thermal design of the environment for the SFC4000
+ * should aim to keep this well below 100°C.' */
+#define FALCON_JUNC_TEMP_MIN	0
+#define FALCON_JUNC_TEMP_MAX	90
+#define FALCON_JUNC_TEMP_CRIT	125
+
+/*****************************************************************************
+ * Support for LM87 sensor chip used on several boards
+ */
+#define LM87_REG_TEMP_HW_INT_LOCK	0x13
+#define LM87_REG_TEMP_HW_EXT_LOCK	0x14
+#define LM87_REG_TEMP_HW_INT		0x17
+#define LM87_REG_TEMP_HW_EXT		0x18
+#define LM87_REG_TEMP_EXT1		0x26
+#define LM87_REG_TEMP_INT		0x27
+#define LM87_REG_ALARMS1		0x41
+#define LM87_REG_ALARMS2		0x42
+#define LM87_IN_LIMITS(nr, _min, _max)			\
+	0x2B + (nr) * 2, _max, 0x2C + (nr) * 2, _min
+#define LM87_AIN_LIMITS(nr, _min, _max)			\
+	0x3B + (nr), _max, 0x1A + (nr), _min
+#define LM87_TEMP_INT_LIMITS(_min, _max)		\
+	0x39, _max, 0x3A, _min
+#define LM87_TEMP_EXT1_LIMITS(_min, _max)		\
+	0x37, _max, 0x38, _min
+
+#define LM87_ALARM_TEMP_INT		0x10
+#define LM87_ALARM_TEMP_EXT1		0x20
+
+#if defined(CONFIG_SENSORS_LM87) || defined(CONFIG_SENSORS_LM87_MODULE)
+
+static int efx_poke_lm87(struct i2c_client *client, const u8 *reg_values)
+{
+	while (*reg_values) {
+		u8 reg = *reg_values++;
+		u8 value = *reg_values++;
+		int rc = i2c_smbus_write_byte_data(client, reg, value);
+		if (rc)
+			return rc;
+	}
+	return 0;
+}
+
+static const u8 falcon_lm87_common_regs[] = {
+	LM87_REG_TEMP_HW_INT_LOCK, FALCON_BOARD_TEMP_CRIT,
+	LM87_REG_TEMP_HW_INT, FALCON_BOARD_TEMP_CRIT,
+	LM87_TEMP_EXT1_LIMITS(FALCON_JUNC_TEMP_MIN, FALCON_JUNC_TEMP_MAX),
+	LM87_REG_TEMP_HW_EXT_LOCK, FALCON_JUNC_TEMP_CRIT,
+	LM87_REG_TEMP_HW_EXT, FALCON_JUNC_TEMP_CRIT,
+	0
+};
+
+static int efx_init_lm87(struct efx_nic *efx, struct i2c_board_info *info,
+			 const u8 *reg_values)
+{
+	struct falcon_board *board = falcon_board(efx);
+	struct i2c_client *client = i2c_new_device(&board->i2c_adap, info);
+	int rc;
+
+	if (!client)
+		return -EIO;
+
+	/* Read-to-clear alarm/interrupt status */
+	i2c_smbus_read_byte_data(client, LM87_REG_ALARMS1);
+	i2c_smbus_read_byte_data(client, LM87_REG_ALARMS2);
+
+	rc = efx_poke_lm87(client, reg_values);
+	if (rc)
+		goto err;
+	rc = efx_poke_lm87(client, falcon_lm87_common_regs);
+	if (rc)
+		goto err;
+
+	board->hwmon_client = client;
+	return 0;
+
+err:
+	i2c_unregister_device(client);
+	return rc;
+}
+
+static void efx_fini_lm87(struct efx_nic *efx)
+{
+	i2c_unregister_device(falcon_board(efx)->hwmon_client);
+}
+
+static int efx_check_lm87(struct efx_nic *efx, unsigned mask)
+{
+	struct i2c_client *client = falcon_board(efx)->hwmon_client;
+	bool temp_crit, elec_fault, is_failure;
+	u16 alarms;
+	s32 reg;
+
+	/* If link is up then do not monitor temperature */
+	if (EFX_WORKAROUND_7884(efx) && efx->link_state.up)
+		return 0;
+
+	reg = i2c_smbus_read_byte_data(client, LM87_REG_ALARMS1);
+	if (reg < 0)
+		return reg;
+	alarms = reg;
+	reg = i2c_smbus_read_byte_data(client, LM87_REG_ALARMS2);
+	if (reg < 0)
+		return reg;
+	alarms |= reg << 8;
+	alarms &= mask;
+
+	temp_crit = false;
+	if (alarms & LM87_ALARM_TEMP_INT) {
+		reg = i2c_smbus_read_byte_data(client, LM87_REG_TEMP_INT);
+		if (reg < 0)
+			return reg;
+		if (reg > FALCON_BOARD_TEMP_CRIT)
+			temp_crit = true;
+	}
+	if (alarms & LM87_ALARM_TEMP_EXT1) {
+		reg = i2c_smbus_read_byte_data(client, LM87_REG_TEMP_EXT1);
+		if (reg < 0)
+			return reg;
+		if (reg > FALCON_JUNC_TEMP_CRIT)
+			temp_crit = true;
+	}
+	elec_fault = alarms & ~(LM87_ALARM_TEMP_INT | LM87_ALARM_TEMP_EXT1);
+	is_failure = temp_crit || elec_fault;
+
+	if (alarms)
+		netif_err(efx, hw, efx->net_dev,
+			  "LM87 detected a hardware %s (status %02x:%02x)"
+			  "%s%s%s%s\n",
+			  is_failure ? "failure" : "problem",
+			  alarms & 0xff, alarms >> 8,
+			  (alarms & LM87_ALARM_TEMP_INT) ?
+			  "; board is overheating" : "",
+			  (alarms & LM87_ALARM_TEMP_EXT1) ?
+			  "; controller is overheating" : "",
+			  temp_crit ? "; reached critical temperature" : "",
+			  elec_fault ? "; electrical fault" : "");
+
+	return is_failure ? -ERANGE : 0;
+}
+
+#else /* !CONFIG_SENSORS_LM87 */
+
+static inline int
+efx_init_lm87(struct efx_nic *efx, struct i2c_board_info *info,
+	      const u8 *reg_values)
+{
+	return 0;
+}
+static inline void efx_fini_lm87(struct efx_nic *efx)
+{
+}
+static inline int efx_check_lm87(struct efx_nic *efx, unsigned mask)
+{
+	return 0;
+}
+
+#endif /* CONFIG_SENSORS_LM87 */
+
+/*****************************************************************************
+ * Support for the SFE4001 NIC.
+ *
+ * The SFE4001 does not power-up fully at reset due to its high power
+ * consumption.  We control its power via a PCA9539 I/O expander.
+ * It also has a MAX6647 temperature monitor which we expose to
+ * the lm90 driver.
+ *
+ * This also provides minimal support for reflashing the PHY, which is
+ * initiated by resetting it with the FLASH_CFG_1 pin pulled down.
+ * On SFE4001 rev A2 and later this is connected to the 3V3X output of
+ * the IO-expander.
+ * We represent reflash mode as PHY_MODE_SPECIAL and make it mutually
+ * exclusive with the network device being open.
+ */
+
+/**************************************************************************
+ * Support for I2C IO Expander device on SFE4001
+ */
+#define	PCA9539 0x74
+
+#define	P0_IN 0x00
+#define	P0_OUT 0x02
+#define	P0_INVERT 0x04
+#define	P0_CONFIG 0x06
+
+#define	P0_EN_1V0X_LBN 0
+#define	P0_EN_1V0X_WIDTH 1
+#define	P0_EN_1V2_LBN 1
+#define	P0_EN_1V2_WIDTH 1
+#define	P0_EN_2V5_LBN 2
+#define	P0_EN_2V5_WIDTH 1
+#define	P0_EN_3V3X_LBN 3
+#define	P0_EN_3V3X_WIDTH 1
+#define	P0_EN_5V_LBN 4
+#define	P0_EN_5V_WIDTH 1
+#define	P0_SHORTEN_JTAG_LBN 5
+#define	P0_SHORTEN_JTAG_WIDTH 1
+#define	P0_X_TRST_LBN 6
+#define	P0_X_TRST_WIDTH 1
+#define	P0_DSP_RESET_LBN 7
+#define	P0_DSP_RESET_WIDTH 1
+
+#define	P1_IN 0x01
+#define	P1_OUT 0x03
+#define	P1_INVERT 0x05
+#define	P1_CONFIG 0x07
+
+#define	P1_AFE_PWD_LBN 0
+#define	P1_AFE_PWD_WIDTH 1
+#define	P1_DSP_PWD25_LBN 1
+#define	P1_DSP_PWD25_WIDTH 1
+#define	P1_RESERVED_LBN 2
+#define	P1_RESERVED_WIDTH 2
+#define	P1_SPARE_LBN 4
+#define	P1_SPARE_WIDTH 4
+
+/* Temperature Sensor */
+#define MAX664X_REG_RSL		0x02
+#define MAX664X_REG_WLHO	0x0B
+
+static void sfe4001_poweroff(struct efx_nic *efx)
+{
+	struct i2c_client *ioexp_client = falcon_board(efx)->ioexp_client;
+	struct i2c_client *hwmon_client = falcon_board(efx)->hwmon_client;
+
+	/* Turn off all power rails and disable outputs */
+	i2c_smbus_write_byte_data(ioexp_client, P0_OUT, 0xff);
+	i2c_smbus_write_byte_data(ioexp_client, P1_CONFIG, 0xff);
+	i2c_smbus_write_byte_data(ioexp_client, P0_CONFIG, 0xff);
+
+	/* Clear any over-temperature alert */
+	i2c_smbus_read_byte_data(hwmon_client, MAX664X_REG_RSL);
+}
+
+static int sfe4001_poweron(struct efx_nic *efx)
+{
+	struct i2c_client *ioexp_client = falcon_board(efx)->ioexp_client;
+	struct i2c_client *hwmon_client = falcon_board(efx)->hwmon_client;
+	unsigned int i, j;
+	int rc;
+	u8 out;
+
+	/* Clear any previous over-temperature alert */
+	rc = i2c_smbus_read_byte_data(hwmon_client, MAX664X_REG_RSL);
+	if (rc < 0)
+		return rc;
+
+	/* Enable port 0 and port 1 outputs on IO expander */
+	rc = i2c_smbus_write_byte_data(ioexp_client, P0_CONFIG, 0x00);
+	if (rc)
+		return rc;
+	rc = i2c_smbus_write_byte_data(ioexp_client, P1_CONFIG,
+				       0xff & ~(1 << P1_SPARE_LBN));
+	if (rc)
+		goto fail_on;
+
+	/* If PHY power is on, turn it all off and wait 1 second to
+	 * ensure a full reset.
+	 */
+	rc = i2c_smbus_read_byte_data(ioexp_client, P0_OUT);
+	if (rc < 0)
+		goto fail_on;
+	out = 0xff & ~((0 << P0_EN_1V2_LBN) | (0 << P0_EN_2V5_LBN) |
+		       (0 << P0_EN_3V3X_LBN) | (0 << P0_EN_5V_LBN) |
+		       (0 << P0_EN_1V0X_LBN));
+	if (rc != out) {
+		netif_info(efx, hw, efx->net_dev, "power-cycling PHY\n");
+		rc = i2c_smbus_write_byte_data(ioexp_client, P0_OUT, out);
+		if (rc)
+			goto fail_on;
+		schedule_timeout_uninterruptible(HZ);
+	}
+
+	for (i = 0; i < 20; ++i) {
+		/* Turn on 1.2V, 2.5V, 3.3V and 5V power rails */
+		out = 0xff & ~((1 << P0_EN_1V2_LBN) | (1 << P0_EN_2V5_LBN) |
+			       (1 << P0_EN_3V3X_LBN) | (1 << P0_EN_5V_LBN) |
+			       (1 << P0_X_TRST_LBN));
+		if (efx->phy_mode & PHY_MODE_SPECIAL)
+			out |= 1 << P0_EN_3V3X_LBN;
+
+		rc = i2c_smbus_write_byte_data(ioexp_client, P0_OUT, out);
+		if (rc)
+			goto fail_on;
+		msleep(10);
+
+		/* Turn on 1V power rail */
+		out &= ~(1 << P0_EN_1V0X_LBN);
+		rc = i2c_smbus_write_byte_data(ioexp_client, P0_OUT, out);
+		if (rc)
+			goto fail_on;
+
+		netif_info(efx, hw, efx->net_dev,
+			   "waiting for DSP boot (attempt %d)...\n", i);
+
+		/* In flash config mode, DSP does not turn on AFE, so
+		 * just wait 1 second.
+		 */
+		if (efx->phy_mode & PHY_MODE_SPECIAL) {
+			schedule_timeout_uninterruptible(HZ);
+			return 0;
+		}
+
+		for (j = 0; j < 10; ++j) {
+			msleep(100);
+
+			/* Check DSP has asserted AFE power line */
+			rc = i2c_smbus_read_byte_data(ioexp_client, P1_IN);
+			if (rc < 0)
+				goto fail_on;
+			if (rc & (1 << P1_AFE_PWD_LBN))
+				return 0;
+		}
+	}
+
+	netif_info(efx, hw, efx->net_dev, "timed out waiting for DSP boot\n");
+	rc = -ETIMEDOUT;
+fail_on:
+	sfe4001_poweroff(efx);
+	return rc;
+}
+
+static ssize_t show_phy_flash_cfg(struct device *dev,
+				  struct device_attribute *attr, char *buf)
+{
+	struct efx_nic *efx = pci_get_drvdata(to_pci_dev(dev));
+	return sprintf(buf, "%d\n", !!(efx->phy_mode & PHY_MODE_SPECIAL));
+}
+
+static ssize_t set_phy_flash_cfg(struct device *dev,
+				 struct device_attribute *attr,
+				 const char *buf, size_t count)
+{
+	struct efx_nic *efx = pci_get_drvdata(to_pci_dev(dev));
+	enum efx_phy_mode old_mode, new_mode;
+	int err;
+
+	rtnl_lock();
+	old_mode = efx->phy_mode;
+	if (count == 0 || *buf == '0')
+		new_mode = old_mode & ~PHY_MODE_SPECIAL;
+	else
+		new_mode = PHY_MODE_SPECIAL;
+	if (!((old_mode ^ new_mode) & PHY_MODE_SPECIAL)) {
+		err = 0;
+	} else if (efx->state != STATE_RUNNING || netif_running(efx->net_dev)) {
+		err = -EBUSY;
+	} else {
+		/* Reset the PHY, reconfigure the MAC and enable/disable
+		 * MAC stats accordingly. */
+		efx->phy_mode = new_mode;
+		if (new_mode & PHY_MODE_SPECIAL)
+			falcon_stop_nic_stats(efx);
+		err = sfe4001_poweron(efx);
+		if (!err)
+			err = efx_reconfigure_port(efx);
+		if (!(new_mode & PHY_MODE_SPECIAL))
+			falcon_start_nic_stats(efx);
+	}
+	rtnl_unlock();
+
+	return err ? err : count;
+}
+
+static DEVICE_ATTR(phy_flash_cfg, 0644, show_phy_flash_cfg, set_phy_flash_cfg);
+
+static void sfe4001_fini(struct efx_nic *efx)
+{
+	struct falcon_board *board = falcon_board(efx);
+
+	netif_info(efx, drv, efx->net_dev, "%s\n", __func__);
+
+	device_remove_file(&efx->pci_dev->dev, &dev_attr_phy_flash_cfg);
+	sfe4001_poweroff(efx);
+	i2c_unregister_device(board->ioexp_client);
+	i2c_unregister_device(board->hwmon_client);
+}
+
+static int sfe4001_check_hw(struct efx_nic *efx)
+{
+	struct falcon_nic_data *nic_data = efx->nic_data;
+	s32 status;
+
+	/* If XAUI link is up then do not monitor */
+	if (EFX_WORKAROUND_7884(efx) && !nic_data->xmac_poll_required)
+		return 0;
+
+	/* Check the powered status of the PHY. Lack of power implies that
+	 * the MAX6647 has shut down power to it, probably due to a temp.
+	 * alarm. Reading the power status rather than the MAX6647 status
+	 * directly because the later is read-to-clear and would thus
+	 * start to power up the PHY again when polled, causing us to blip
+	 * the power undesirably.
+	 * We know we can read from the IO expander because we did
+	 * it during power-on. Assume failure now is bad news. */
+	status = i2c_smbus_read_byte_data(falcon_board(efx)->ioexp_client, P1_IN);
+	if (status >= 0 &&
+	    (status & ((1 << P1_AFE_PWD_LBN) | (1 << P1_DSP_PWD25_LBN))) != 0)
+		return 0;
+
+	/* Use board power control, not PHY power control */
+	sfe4001_poweroff(efx);
+	efx->phy_mode = PHY_MODE_OFF;
+
+	return (status < 0) ? -EIO : -ERANGE;
+}
+
+static struct i2c_board_info sfe4001_hwmon_info = {
+	I2C_BOARD_INFO("max6647", 0x4e),
+};
+
+/* This board uses an I2C expander to provider power to the PHY, which needs to
+ * be turned on before the PHY can be used.
+ * Context: Process context, rtnl lock held
+ */
+static int sfe4001_init(struct efx_nic *efx)
+{
+	struct falcon_board *board = falcon_board(efx);
+	int rc;
+
+#if defined(CONFIG_SENSORS_LM90) || defined(CONFIG_SENSORS_LM90_MODULE)
+	board->hwmon_client =
+		i2c_new_device(&board->i2c_adap, &sfe4001_hwmon_info);
+#else
+	board->hwmon_client =
+		i2c_new_dummy(&board->i2c_adap, sfe4001_hwmon_info.addr);
+#endif
+	if (!board->hwmon_client)
+		return -EIO;
+
+	/* Raise board/PHY high limit from 85 to 90 degrees Celsius */
+	rc = i2c_smbus_write_byte_data(board->hwmon_client,
+				       MAX664X_REG_WLHO, 90);
+	if (rc)
+		goto fail_hwmon;
+
+	board->ioexp_client = i2c_new_dummy(&board->i2c_adap, PCA9539);
+	if (!board->ioexp_client) {
+		rc = -EIO;
+		goto fail_hwmon;
+	}
+
+	if (efx->phy_mode & PHY_MODE_SPECIAL) {
+		/* PHY won't generate a 156.25 MHz clock and MAC stats fetch
+		 * will fail. */
+		falcon_stop_nic_stats(efx);
+	}
+	rc = sfe4001_poweron(efx);
+	if (rc)
+		goto fail_ioexp;
+
+	rc = device_create_file(&efx->pci_dev->dev, &dev_attr_phy_flash_cfg);
+	if (rc)
+		goto fail_on;
+
+	netif_info(efx, hw, efx->net_dev, "PHY is powered on\n");
+	return 0;
+
+fail_on:
+	sfe4001_poweroff(efx);
+fail_ioexp:
+	i2c_unregister_device(board->ioexp_client);
+fail_hwmon:
+	i2c_unregister_device(board->hwmon_client);
+	return rc;
+}
+
+/*****************************************************************************
+ * Support for the SFE4002
+ *
+ */
+static u8 sfe4002_lm87_channel = 0x03; /* use AIN not FAN inputs */
+
+static const u8 sfe4002_lm87_regs[] = {
+	LM87_IN_LIMITS(0, 0x7c, 0x99),		/* 2.5V:  1.8V +/- 10% */
+	LM87_IN_LIMITS(1, 0x4c, 0x5e),		/* Vccp1: 1.2V +/- 10% */
+	LM87_IN_LIMITS(2, 0xac, 0xd4),		/* 3.3V:  3.3V +/- 10% */
+	LM87_IN_LIMITS(3, 0xac, 0xd4),		/* 5V:    5.0V +/- 10% */
+	LM87_IN_LIMITS(4, 0xac, 0xe0),		/* 12V:   10.8-14V */
+	LM87_IN_LIMITS(5, 0x3f, 0x4f),		/* Vccp2: 1.0V +/- 10% */
+	LM87_AIN_LIMITS(0, 0x98, 0xbb),		/* AIN1:  1.66V +/- 10% */
+	LM87_AIN_LIMITS(1, 0x8a, 0xa9),		/* AIN2:  1.5V +/- 10% */
+	LM87_TEMP_INT_LIMITS(0, 80 + FALCON_BOARD_TEMP_BIAS),
+	LM87_TEMP_EXT1_LIMITS(0, FALCON_JUNC_TEMP_MAX),
+	0
+};
+
+static struct i2c_board_info sfe4002_hwmon_info = {
+	I2C_BOARD_INFO("lm87", 0x2e),
+	.platform_data	= &sfe4002_lm87_channel,
+};
+
+/****************************************************************************/
+/* LED allocations. Note that on rev A0 boards the schematic and the reality
+ * differ: red and green are swapped. Below is the fixed (A1) layout (there
+ * are only 3 A0 boards in existence, so no real reason to make this
+ * conditional).
+ */
+#define SFE4002_FAULT_LED (2)	/* Red */
+#define SFE4002_RX_LED    (0)	/* Green */
+#define SFE4002_TX_LED    (1)	/* Amber */
+
+static void sfe4002_init_phy(struct efx_nic *efx)
+{
+	/* Set the TX and RX LEDs to reflect status and activity, and the
+	 * fault LED off */
+	falcon_qt202x_set_led(efx, SFE4002_TX_LED,
+			      QUAKE_LED_TXLINK | QUAKE_LED_LINK_ACTSTAT);
+	falcon_qt202x_set_led(efx, SFE4002_RX_LED,
+			      QUAKE_LED_RXLINK | QUAKE_LED_LINK_ACTSTAT);
+	falcon_qt202x_set_led(efx, SFE4002_FAULT_LED, QUAKE_LED_OFF);
+}
+
+static void sfe4002_set_id_led(struct efx_nic *efx, enum efx_led_mode mode)
+{
+	falcon_qt202x_set_led(
+		efx, SFE4002_FAULT_LED,
+		(mode == EFX_LED_ON) ? QUAKE_LED_ON : QUAKE_LED_OFF);
+}
+
+static int sfe4002_check_hw(struct efx_nic *efx)
+{
+	struct falcon_board *board = falcon_board(efx);
+
+	/* A0 board rev. 4002s report a temperature fault the whole time
+	 * (bad sensor) so we mask it out. */
+	unsigned alarm_mask =
+		(board->major == 0 && board->minor == 0) ?
+		~LM87_ALARM_TEMP_EXT1 : ~0;
+
+	return efx_check_lm87(efx, alarm_mask);
+}
+
+static int sfe4002_init(struct efx_nic *efx)
+{
+	return efx_init_lm87(efx, &sfe4002_hwmon_info, sfe4002_lm87_regs);
+}
+
+/*****************************************************************************
+ * Support for the SFN4112F
+ *
+ */
+static u8 sfn4112f_lm87_channel = 0x03; /* use AIN not FAN inputs */
+
+static const u8 sfn4112f_lm87_regs[] = {
+	LM87_IN_LIMITS(0, 0x7c, 0x99),		/* 2.5V:  1.8V +/- 10% */
+	LM87_IN_LIMITS(1, 0x4c, 0x5e),		/* Vccp1: 1.2V +/- 10% */
+	LM87_IN_LIMITS(2, 0xac, 0xd4),		/* 3.3V:  3.3V +/- 10% */
+	LM87_IN_LIMITS(4, 0xac, 0xe0),		/* 12V:   10.8-14V */
+	LM87_IN_LIMITS(5, 0x3f, 0x4f),		/* Vccp2: 1.0V +/- 10% */
+	LM87_AIN_LIMITS(1, 0x8a, 0xa9),		/* AIN2:  1.5V +/- 10% */
+	LM87_TEMP_INT_LIMITS(0, 60 + FALCON_BOARD_TEMP_BIAS),
+	LM87_TEMP_EXT1_LIMITS(0, FALCON_JUNC_TEMP_MAX),
+	0
+};
+
+static struct i2c_board_info sfn4112f_hwmon_info = {
+	I2C_BOARD_INFO("lm87", 0x2e),
+	.platform_data	= &sfn4112f_lm87_channel,
+};
+
+#define SFN4112F_ACT_LED	0
+#define SFN4112F_LINK_LED	1
+
+static void sfn4112f_init_phy(struct efx_nic *efx)
+{
+	falcon_qt202x_set_led(efx, SFN4112F_ACT_LED,
+			      QUAKE_LED_RXLINK | QUAKE_LED_LINK_ACT);
+	falcon_qt202x_set_led(efx, SFN4112F_LINK_LED,
+			      QUAKE_LED_RXLINK | QUAKE_LED_LINK_STAT);
+}
+
+static void sfn4112f_set_id_led(struct efx_nic *efx, enum efx_led_mode mode)
+{
+	int reg;
+
+	switch (mode) {
+	case EFX_LED_OFF:
+		reg = QUAKE_LED_OFF;
+		break;
+	case EFX_LED_ON:
+		reg = QUAKE_LED_ON;
+		break;
+	default:
+		reg = QUAKE_LED_RXLINK | QUAKE_LED_LINK_STAT;
+		break;
+	}
+
+	falcon_qt202x_set_led(efx, SFN4112F_LINK_LED, reg);
+}
+
+static int sfn4112f_check_hw(struct efx_nic *efx)
+{
+	/* Mask out unused sensors */
+	return efx_check_lm87(efx, ~0x48);
+}
+
+static int sfn4112f_init(struct efx_nic *efx)
+{
+	return efx_init_lm87(efx, &sfn4112f_hwmon_info, sfn4112f_lm87_regs);
+}
+
+/*****************************************************************************
+ * Support for the SFE4003
+ *
+ */
+static u8 sfe4003_lm87_channel = 0x03; /* use AIN not FAN inputs */
+
+static const u8 sfe4003_lm87_regs[] = {
+	LM87_IN_LIMITS(0, 0x67, 0x7f),		/* 2.5V:  1.5V +/- 10% */
+	LM87_IN_LIMITS(1, 0x4c, 0x5e),		/* Vccp1: 1.2V +/- 10% */
+	LM87_IN_LIMITS(2, 0xac, 0xd4),		/* 3.3V:  3.3V +/- 10% */
+	LM87_IN_LIMITS(4, 0xac, 0xe0),		/* 12V:   10.8-14V */
+	LM87_IN_LIMITS(5, 0x3f, 0x4f),		/* Vccp2: 1.0V +/- 10% */
+	LM87_TEMP_INT_LIMITS(0, 70 + FALCON_BOARD_TEMP_BIAS),
+	0
+};
+
+static struct i2c_board_info sfe4003_hwmon_info = {
+	I2C_BOARD_INFO("lm87", 0x2e),
+	.platform_data	= &sfe4003_lm87_channel,
+};
+
+/* Board-specific LED info. */
+#define SFE4003_RED_LED_GPIO	11
+#define SFE4003_LED_ON		1
+#define SFE4003_LED_OFF		0
+
+static void sfe4003_set_id_led(struct efx_nic *efx, enum efx_led_mode mode)
+{
+	struct falcon_board *board = falcon_board(efx);
+
+	/* The LEDs were not wired to GPIOs before A3 */
+	if (board->minor < 3 && board->major == 0)
+		return;
+
+	falcon_txc_set_gpio_val(
+		efx, SFE4003_RED_LED_GPIO,
+		(mode == EFX_LED_ON) ? SFE4003_LED_ON : SFE4003_LED_OFF);
+}
+
+static void sfe4003_init_phy(struct efx_nic *efx)
+{
+	struct falcon_board *board = falcon_board(efx);
+
+	/* The LEDs were not wired to GPIOs before A3 */
+	if (board->minor < 3 && board->major == 0)
+		return;
+
+	falcon_txc_set_gpio_dir(efx, SFE4003_RED_LED_GPIO, TXC_GPIO_DIR_OUTPUT);
+	falcon_txc_set_gpio_val(efx, SFE4003_RED_LED_GPIO, SFE4003_LED_OFF);
+}
+
+static int sfe4003_check_hw(struct efx_nic *efx)
+{
+	struct falcon_board *board = falcon_board(efx);
+
+	/* A0/A1/A2 board rev. 4003s  report a temperature fault the whole time
+	 * (bad sensor) so we mask it out. */
+	unsigned alarm_mask =
+		(board->major == 0 && board->minor <= 2) ?
+		~LM87_ALARM_TEMP_EXT1 : ~0;
+
+	return efx_check_lm87(efx, alarm_mask);
+}
+
+static int sfe4003_init(struct efx_nic *efx)
+{
+	return efx_init_lm87(efx, &sfe4003_hwmon_info, sfe4003_lm87_regs);
+}
+
+static const struct falcon_board_type board_types[] = {
+	{
+		.id		= FALCON_BOARD_SFE4001,
+		.ref_model	= "SFE4001",
+		.gen_type	= "10GBASE-T adapter",
+		.init		= sfe4001_init,
+		.init_phy	= efx_port_dummy_op_void,
+		.fini		= sfe4001_fini,
+		.set_id_led	= tenxpress_set_id_led,
+		.monitor	= sfe4001_check_hw,
+	},
+	{
+		.id		= FALCON_BOARD_SFE4002,
+		.ref_model	= "SFE4002",
+		.gen_type	= "XFP adapter",
+		.init		= sfe4002_init,
+		.init_phy	= sfe4002_init_phy,
+		.fini		= efx_fini_lm87,
+		.set_id_led	= sfe4002_set_id_led,
+		.monitor	= sfe4002_check_hw,
+	},
+	{
+		.id		= FALCON_BOARD_SFE4003,
+		.ref_model	= "SFE4003",
+		.gen_type	= "10GBASE-CX4 adapter",
+		.init		= sfe4003_init,
+		.init_phy	= sfe4003_init_phy,
+		.fini		= efx_fini_lm87,
+		.set_id_led	= sfe4003_set_id_led,
+		.monitor	= sfe4003_check_hw,
+	},
+	{
+		.id		= FALCON_BOARD_SFN4112F,
+		.ref_model	= "SFN4112F",
+		.gen_type	= "SFP+ adapter",
+		.init		= sfn4112f_init,
+		.init_phy	= sfn4112f_init_phy,
+		.fini		= efx_fini_lm87,
+		.set_id_led	= sfn4112f_set_id_led,
+		.monitor	= sfn4112f_check_hw,
+	},
+};
+
+int falcon_probe_board(struct efx_nic *efx, u16 revision_info)
+{
+	struct falcon_board *board = falcon_board(efx);
+	u8 type_id = FALCON_BOARD_TYPE(revision_info);
+	int i;
+
+	board->major = FALCON_BOARD_MAJOR(revision_info);
+	board->minor = FALCON_BOARD_MINOR(revision_info);
+
+	for (i = 0; i < ARRAY_SIZE(board_types); i++)
+		if (board_types[i].id == type_id)
+			board->type = &board_types[i];
+
+	if (board->type) {
+		netif_info(efx, probe, efx->net_dev, "board is %s rev %c%d\n",
+			 (efx->pci_dev->subsystem_vendor == EFX_VENDID_SFC)
+			 ? board->type->ref_model : board->type->gen_type,
+			 'A' + board->major, board->minor);
+		return 0;
+	} else {
+		netif_err(efx, probe, efx->net_dev, "unknown board type %d\n",
+			  type_id);
+		return -ENODEV;
+	}
+}
diff --git a/drivers/net/ethernet/sfc/falcon_xmac.c b/drivers/net/ethernet/sfc/falcon_xmac.c
new file mode 100644
index 0000000..9516452
--- /dev/null
+++ b/drivers/net/ethernet/sfc/falcon_xmac.c
@@ -0,0 +1,369 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2005-2006 Fen Systems Ltd.
+ * Copyright 2006-2010 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#include <linux/delay.h>
+#include "net_driver.h"
+#include "efx.h"
+#include "nic.h"
+#include "regs.h"
+#include "io.h"
+#include "mac.h"
+#include "mdio_10g.h"
+#include "workarounds.h"
+
+/**************************************************************************
+ *
+ * MAC operations
+ *
+ *************************************************************************/
+
+/* Configure the XAUI driver that is an output from Falcon */
+void falcon_setup_xaui(struct efx_nic *efx)
+{
+	efx_oword_t sdctl, txdrv;
+
+	/* Move the XAUI into low power, unless there is no PHY, in
+	 * which case the XAUI will have to drive a cable. */
+	if (efx->phy_type == PHY_TYPE_NONE)
+		return;
+
+	efx_reado(efx, &sdctl, FR_AB_XX_SD_CTL);
+	EFX_SET_OWORD_FIELD(sdctl, FRF_AB_XX_HIDRVD, FFE_AB_XX_SD_CTL_DRV_DEF);
+	EFX_SET_OWORD_FIELD(sdctl, FRF_AB_XX_LODRVD, FFE_AB_XX_SD_CTL_DRV_DEF);
+	EFX_SET_OWORD_FIELD(sdctl, FRF_AB_XX_HIDRVC, FFE_AB_XX_SD_CTL_DRV_DEF);
+	EFX_SET_OWORD_FIELD(sdctl, FRF_AB_XX_LODRVC, FFE_AB_XX_SD_CTL_DRV_DEF);
+	EFX_SET_OWORD_FIELD(sdctl, FRF_AB_XX_HIDRVB, FFE_AB_XX_SD_CTL_DRV_DEF);
+	EFX_SET_OWORD_FIELD(sdctl, FRF_AB_XX_LODRVB, FFE_AB_XX_SD_CTL_DRV_DEF);
+	EFX_SET_OWORD_FIELD(sdctl, FRF_AB_XX_HIDRVA, FFE_AB_XX_SD_CTL_DRV_DEF);
+	EFX_SET_OWORD_FIELD(sdctl, FRF_AB_XX_LODRVA, FFE_AB_XX_SD_CTL_DRV_DEF);
+	efx_writeo(efx, &sdctl, FR_AB_XX_SD_CTL);
+
+	EFX_POPULATE_OWORD_8(txdrv,
+			     FRF_AB_XX_DEQD, FFE_AB_XX_TXDRV_DEQ_DEF,
+			     FRF_AB_XX_DEQC, FFE_AB_XX_TXDRV_DEQ_DEF,
+			     FRF_AB_XX_DEQB, FFE_AB_XX_TXDRV_DEQ_DEF,
+			     FRF_AB_XX_DEQA, FFE_AB_XX_TXDRV_DEQ_DEF,
+			     FRF_AB_XX_DTXD, FFE_AB_XX_TXDRV_DTX_DEF,
+			     FRF_AB_XX_DTXC, FFE_AB_XX_TXDRV_DTX_DEF,
+			     FRF_AB_XX_DTXB, FFE_AB_XX_TXDRV_DTX_DEF,
+			     FRF_AB_XX_DTXA, FFE_AB_XX_TXDRV_DTX_DEF);
+	efx_writeo(efx, &txdrv, FR_AB_XX_TXDRV_CTL);
+}
+
+int falcon_reset_xaui(struct efx_nic *efx)
+{
+	struct falcon_nic_data *nic_data = efx->nic_data;
+	efx_oword_t reg;
+	int count;
+
+	/* Don't fetch MAC statistics over an XMAC reset */
+	WARN_ON(nic_data->stats_disable_count == 0);
+
+	/* Start reset sequence */
+	EFX_POPULATE_OWORD_1(reg, FRF_AB_XX_RST_XX_EN, 1);
+	efx_writeo(efx, &reg, FR_AB_XX_PWR_RST);
+
+	/* Wait up to 10 ms for completion, then reinitialise */
+	for (count = 0; count < 1000; count++) {
+		efx_reado(efx, &reg, FR_AB_XX_PWR_RST);
+		if (EFX_OWORD_FIELD(reg, FRF_AB_XX_RST_XX_EN) == 0 &&
+		    EFX_OWORD_FIELD(reg, FRF_AB_XX_SD_RST_ACT) == 0) {
+			falcon_setup_xaui(efx);
+			return 0;
+		}
+		udelay(10);
+	}
+	netif_err(efx, hw, efx->net_dev,
+		  "timed out waiting for XAUI/XGXS reset\n");
+	return -ETIMEDOUT;
+}
+
+static void falcon_ack_status_intr(struct efx_nic *efx)
+{
+	struct falcon_nic_data *nic_data = efx->nic_data;
+	efx_oword_t reg;
+
+	if ((efx_nic_rev(efx) != EFX_REV_FALCON_B0) || LOOPBACK_INTERNAL(efx))
+		return;
+
+	/* We expect xgmii faults if the wireside link is down */
+	if (!EFX_WORKAROUND_5147(efx) || !efx->link_state.up)
+		return;
+
+	/* We can only use this interrupt to signal the negative edge of
+	 * xaui_align [we have to poll the positive edge]. */
+	if (nic_data->xmac_poll_required)
+		return;
+
+	efx_reado(efx, &reg, FR_AB_XM_MGT_INT_MSK);
+}
+
+static bool falcon_xgxs_link_ok(struct efx_nic *efx)
+{
+	efx_oword_t reg;
+	bool align_done, link_ok = false;
+	int sync_status;
+
+	/* Read link status */
+	efx_reado(efx, &reg, FR_AB_XX_CORE_STAT);
+
+	align_done = EFX_OWORD_FIELD(reg, FRF_AB_XX_ALIGN_DONE);
+	sync_status = EFX_OWORD_FIELD(reg, FRF_AB_XX_SYNC_STAT);
+	if (align_done && (sync_status == FFE_AB_XX_STAT_ALL_LANES))
+		link_ok = true;
+
+	/* Clear link status ready for next read */
+	EFX_SET_OWORD_FIELD(reg, FRF_AB_XX_COMMA_DET, FFE_AB_XX_STAT_ALL_LANES);
+	EFX_SET_OWORD_FIELD(reg, FRF_AB_XX_CHAR_ERR, FFE_AB_XX_STAT_ALL_LANES);
+	EFX_SET_OWORD_FIELD(reg, FRF_AB_XX_DISPERR, FFE_AB_XX_STAT_ALL_LANES);
+	efx_writeo(efx, &reg, FR_AB_XX_CORE_STAT);
+
+	return link_ok;
+}
+
+static bool falcon_xmac_link_ok(struct efx_nic *efx)
+{
+	/*
+	 * Check MAC's XGXS link status except when using XGMII loopback
+	 * which bypasses the XGXS block.
+	 * If possible, check PHY's XGXS link status except when using
+	 * MAC loopback.
+	 */
+	return (efx->loopback_mode == LOOPBACK_XGMII ||
+		falcon_xgxs_link_ok(efx)) &&
+		(!(efx->mdio.mmds & (1 << MDIO_MMD_PHYXS)) ||
+		 LOOPBACK_INTERNAL(efx) || 
+		 efx_mdio_phyxgxs_lane_sync(efx));
+}
+
+static void falcon_reconfigure_xmac_core(struct efx_nic *efx)
+{
+	unsigned int max_frame_len;
+	efx_oword_t reg;
+	bool rx_fc = !!(efx->link_state.fc & EFX_FC_RX);
+	bool tx_fc = !!(efx->link_state.fc & EFX_FC_TX);
+
+	/* Configure MAC  - cut-thru mode is hard wired on */
+	EFX_POPULATE_OWORD_3(reg,
+			     FRF_AB_XM_RX_JUMBO_MODE, 1,
+			     FRF_AB_XM_TX_STAT_EN, 1,
+			     FRF_AB_XM_RX_STAT_EN, 1);
+	efx_writeo(efx, &reg, FR_AB_XM_GLB_CFG);
+
+	/* Configure TX */
+	EFX_POPULATE_OWORD_6(reg,
+			     FRF_AB_XM_TXEN, 1,
+			     FRF_AB_XM_TX_PRMBL, 1,
+			     FRF_AB_XM_AUTO_PAD, 1,
+			     FRF_AB_XM_TXCRC, 1,
+			     FRF_AB_XM_FCNTL, tx_fc,
+			     FRF_AB_XM_IPG, 0x3);
+	efx_writeo(efx, &reg, FR_AB_XM_TX_CFG);
+
+	/* Configure RX */
+	EFX_POPULATE_OWORD_5(reg,
+			     FRF_AB_XM_RXEN, 1,
+			     FRF_AB_XM_AUTO_DEPAD, 0,
+			     FRF_AB_XM_ACPT_ALL_MCAST, 1,
+			     FRF_AB_XM_ACPT_ALL_UCAST, efx->promiscuous,
+			     FRF_AB_XM_PASS_CRC_ERR, 1);
+	efx_writeo(efx, &reg, FR_AB_XM_RX_CFG);
+
+	/* Set frame length */
+	max_frame_len = EFX_MAX_FRAME_LEN(efx->net_dev->mtu);
+	EFX_POPULATE_OWORD_1(reg, FRF_AB_XM_MAX_RX_FRM_SIZE, max_frame_len);
+	efx_writeo(efx, &reg, FR_AB_XM_RX_PARAM);
+	EFX_POPULATE_OWORD_2(reg,
+			     FRF_AB_XM_MAX_TX_FRM_SIZE, max_frame_len,
+			     FRF_AB_XM_TX_JUMBO_MODE, 1);
+	efx_writeo(efx, &reg, FR_AB_XM_TX_PARAM);
+
+	EFX_POPULATE_OWORD_2(reg,
+			     FRF_AB_XM_PAUSE_TIME, 0xfffe, /* MAX PAUSE TIME */
+			     FRF_AB_XM_DIS_FCNTL, !rx_fc);
+	efx_writeo(efx, &reg, FR_AB_XM_FC);
+
+	/* Set MAC address */
+	memcpy(&reg, &efx->net_dev->dev_addr[0], 4);
+	efx_writeo(efx, &reg, FR_AB_XM_ADR_LO);
+	memcpy(&reg, &efx->net_dev->dev_addr[4], 2);
+	efx_writeo(efx, &reg, FR_AB_XM_ADR_HI);
+}
+
+static void falcon_reconfigure_xgxs_core(struct efx_nic *efx)
+{
+	efx_oword_t reg;
+	bool xgxs_loopback = (efx->loopback_mode == LOOPBACK_XGXS);
+	bool xaui_loopback = (efx->loopback_mode == LOOPBACK_XAUI);
+	bool xgmii_loopback = (efx->loopback_mode == LOOPBACK_XGMII);
+
+	/* XGXS block is flaky and will need to be reset if moving
+	 * into our out of XGMII, XGXS or XAUI loopbacks. */
+	if (EFX_WORKAROUND_5147(efx)) {
+		bool old_xgmii_loopback, old_xgxs_loopback, old_xaui_loopback;
+		bool reset_xgxs;
+
+		efx_reado(efx, &reg, FR_AB_XX_CORE_STAT);
+		old_xgxs_loopback = EFX_OWORD_FIELD(reg, FRF_AB_XX_XGXS_LB_EN);
+		old_xgmii_loopback =
+			EFX_OWORD_FIELD(reg, FRF_AB_XX_XGMII_LB_EN);
+
+		efx_reado(efx, &reg, FR_AB_XX_SD_CTL);
+		old_xaui_loopback = EFX_OWORD_FIELD(reg, FRF_AB_XX_LPBKA);
+
+		/* The PHY driver may have turned XAUI off */
+		reset_xgxs = ((xgxs_loopback != old_xgxs_loopback) ||
+			      (xaui_loopback != old_xaui_loopback) ||
+			      (xgmii_loopback != old_xgmii_loopback));
+
+		if (reset_xgxs)
+			falcon_reset_xaui(efx);
+	}
+
+	efx_reado(efx, &reg, FR_AB_XX_CORE_STAT);
+	EFX_SET_OWORD_FIELD(reg, FRF_AB_XX_FORCE_SIG,
+			    (xgxs_loopback || xaui_loopback) ?
+			    FFE_AB_XX_FORCE_SIG_ALL_LANES : 0);
+	EFX_SET_OWORD_FIELD(reg, FRF_AB_XX_XGXS_LB_EN, xgxs_loopback);
+	EFX_SET_OWORD_FIELD(reg, FRF_AB_XX_XGMII_LB_EN, xgmii_loopback);
+	efx_writeo(efx, &reg, FR_AB_XX_CORE_STAT);
+
+	efx_reado(efx, &reg, FR_AB_XX_SD_CTL);
+	EFX_SET_OWORD_FIELD(reg, FRF_AB_XX_LPBKD, xaui_loopback);
+	EFX_SET_OWORD_FIELD(reg, FRF_AB_XX_LPBKC, xaui_loopback);
+	EFX_SET_OWORD_FIELD(reg, FRF_AB_XX_LPBKB, xaui_loopback);
+	EFX_SET_OWORD_FIELD(reg, FRF_AB_XX_LPBKA, xaui_loopback);
+	efx_writeo(efx, &reg, FR_AB_XX_SD_CTL);
+}
+
+
+/* Try to bring up the Falcon side of the Falcon-Phy XAUI link */
+static bool falcon_xmac_link_ok_retry(struct efx_nic *efx, int tries)
+{
+	bool mac_up = falcon_xmac_link_ok(efx);
+
+	if (LOOPBACK_MASK(efx) & LOOPBACKS_EXTERNAL(efx) & LOOPBACKS_WS ||
+	    efx_phy_mode_disabled(efx->phy_mode))
+		/* XAUI link is expected to be down */
+		return mac_up;
+
+	falcon_stop_nic_stats(efx);
+
+	while (!mac_up && tries) {
+		netif_dbg(efx, hw, efx->net_dev, "bashing xaui\n");
+		falcon_reset_xaui(efx);
+		udelay(200);
+
+		mac_up = falcon_xmac_link_ok(efx);
+		--tries;
+	}
+
+	falcon_start_nic_stats(efx);
+
+	return mac_up;
+}
+
+static bool falcon_xmac_check_fault(struct efx_nic *efx)
+{
+	return !falcon_xmac_link_ok_retry(efx, 5);
+}
+
+static int falcon_reconfigure_xmac(struct efx_nic *efx)
+{
+	struct falcon_nic_data *nic_data = efx->nic_data;
+
+	falcon_reconfigure_xgxs_core(efx);
+	falcon_reconfigure_xmac_core(efx);
+
+	falcon_reconfigure_mac_wrapper(efx);
+
+	nic_data->xmac_poll_required = !falcon_xmac_link_ok_retry(efx, 5);
+	falcon_ack_status_intr(efx);
+
+	return 0;
+}
+
+static void falcon_update_stats_xmac(struct efx_nic *efx)
+{
+	struct efx_mac_stats *mac_stats = &efx->mac_stats;
+
+	/* Update MAC stats from DMAed values */
+	FALCON_STAT(efx, XgRxOctets, rx_bytes);
+	FALCON_STAT(efx, XgRxOctetsOK, rx_good_bytes);
+	FALCON_STAT(efx, XgRxPkts, rx_packets);
+	FALCON_STAT(efx, XgRxPktsOK, rx_good);
+	FALCON_STAT(efx, XgRxBroadcastPkts, rx_broadcast);
+	FALCON_STAT(efx, XgRxMulticastPkts, rx_multicast);
+	FALCON_STAT(efx, XgRxUnicastPkts, rx_unicast);
+	FALCON_STAT(efx, XgRxUndersizePkts, rx_lt64);
+	FALCON_STAT(efx, XgRxOversizePkts, rx_gtjumbo);
+	FALCON_STAT(efx, XgRxJabberPkts, rx_bad_gtjumbo);
+	FALCON_STAT(efx, XgRxUndersizeFCSerrorPkts, rx_bad_lt64);
+	FALCON_STAT(efx, XgRxDropEvents, rx_overflow);
+	FALCON_STAT(efx, XgRxFCSerrorPkts, rx_bad);
+	FALCON_STAT(efx, XgRxAlignError, rx_align_error);
+	FALCON_STAT(efx, XgRxSymbolError, rx_symbol_error);
+	FALCON_STAT(efx, XgRxInternalMACError, rx_internal_error);
+	FALCON_STAT(efx, XgRxControlPkts, rx_control);
+	FALCON_STAT(efx, XgRxPausePkts, rx_pause);
+	FALCON_STAT(efx, XgRxPkts64Octets, rx_64);
+	FALCON_STAT(efx, XgRxPkts65to127Octets, rx_65_to_127);
+	FALCON_STAT(efx, XgRxPkts128to255Octets, rx_128_to_255);
+	FALCON_STAT(efx, XgRxPkts256to511Octets, rx_256_to_511);
+	FALCON_STAT(efx, XgRxPkts512to1023Octets, rx_512_to_1023);
+	FALCON_STAT(efx, XgRxPkts1024to15xxOctets, rx_1024_to_15xx);
+	FALCON_STAT(efx, XgRxPkts15xxtoMaxOctets, rx_15xx_to_jumbo);
+	FALCON_STAT(efx, XgRxLengthError, rx_length_error);
+	FALCON_STAT(efx, XgTxPkts, tx_packets);
+	FALCON_STAT(efx, XgTxOctets, tx_bytes);
+	FALCON_STAT(efx, XgTxMulticastPkts, tx_multicast);
+	FALCON_STAT(efx, XgTxBroadcastPkts, tx_broadcast);
+	FALCON_STAT(efx, XgTxUnicastPkts, tx_unicast);
+	FALCON_STAT(efx, XgTxControlPkts, tx_control);
+	FALCON_STAT(efx, XgTxPausePkts, tx_pause);
+	FALCON_STAT(efx, XgTxPkts64Octets, tx_64);
+	FALCON_STAT(efx, XgTxPkts65to127Octets, tx_65_to_127);
+	FALCON_STAT(efx, XgTxPkts128to255Octets, tx_128_to_255);
+	FALCON_STAT(efx, XgTxPkts256to511Octets, tx_256_to_511);
+	FALCON_STAT(efx, XgTxPkts512to1023Octets, tx_512_to_1023);
+	FALCON_STAT(efx, XgTxPkts1024to15xxOctets, tx_1024_to_15xx);
+	FALCON_STAT(efx, XgTxPkts1519toMaxOctets, tx_15xx_to_jumbo);
+	FALCON_STAT(efx, XgTxUndersizePkts, tx_lt64);
+	FALCON_STAT(efx, XgTxOversizePkts, tx_gtjumbo);
+	FALCON_STAT(efx, XgTxNonTcpUdpPkt, tx_non_tcpudp);
+	FALCON_STAT(efx, XgTxMacSrcErrPkt, tx_mac_src_error);
+	FALCON_STAT(efx, XgTxIpSrcErrPkt, tx_ip_src_error);
+
+	/* Update derived statistics */
+	mac_stats->tx_good_bytes =
+		(mac_stats->tx_bytes - mac_stats->tx_bad_bytes -
+		 mac_stats->tx_control * 64);
+	mac_stats->rx_bad_bytes =
+		(mac_stats->rx_bytes - mac_stats->rx_good_bytes -
+		 mac_stats->rx_control * 64);
+}
+
+void falcon_poll_xmac(struct efx_nic *efx)
+{
+	struct falcon_nic_data *nic_data = efx->nic_data;
+
+	if (!EFX_WORKAROUND_5147(efx) || !efx->link_state.up ||
+	    !nic_data->xmac_poll_required)
+		return;
+
+	nic_data->xmac_poll_required = !falcon_xmac_link_ok_retry(efx, 1);
+	falcon_ack_status_intr(efx);
+}
+
+const struct efx_mac_operations falcon_xmac_operations = {
+	.reconfigure	= falcon_reconfigure_xmac,
+	.update_stats	= falcon_update_stats_xmac,
+	.check_fault	= falcon_xmac_check_fault,
+};
diff --git a/drivers/net/ethernet/sfc/filter.c b/drivers/net/ethernet/sfc/filter.c
new file mode 100644
index 0000000..2b9636f
--- /dev/null
+++ b/drivers/net/ethernet/sfc/filter.c
@@ -0,0 +1,727 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2005-2010 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#include <linux/in.h>
+#include <net/ip.h>
+#include "efx.h"
+#include "filter.h"
+#include "io.h"
+#include "nic.h"
+#include "regs.h"
+
+/* "Fudge factors" - difference between programmed value and actual depth.
+ * Due to pipelined implementation we need to program H/W with a value that
+ * is larger than the hop limit we want.
+ */
+#define FILTER_CTL_SRCH_FUDGE_WILD 3
+#define FILTER_CTL_SRCH_FUDGE_FULL 1
+
+/* Hard maximum hop limit.  Hardware will time-out beyond 200-something.
+ * We also need to avoid infinite loops in efx_filter_search() when the
+ * table is full.
+ */
+#define FILTER_CTL_SRCH_MAX 200
+
+/* Don't try very hard to find space for performance hints, as this is
+ * counter-productive. */
+#define FILTER_CTL_SRCH_HINT_MAX 5
+
+enum efx_filter_table_id {
+	EFX_FILTER_TABLE_RX_IP = 0,
+	EFX_FILTER_TABLE_RX_MAC,
+	EFX_FILTER_TABLE_COUNT,
+};
+
+struct efx_filter_table {
+	enum efx_filter_table_id id;
+	u32		offset;		/* address of table relative to BAR */
+	unsigned	size;		/* number of entries */
+	unsigned	step;		/* step between entries */
+	unsigned	used;		/* number currently used */
+	unsigned long	*used_bitmap;
+	struct efx_filter_spec *spec;
+	unsigned	search_depth[EFX_FILTER_TYPE_COUNT];
+};
+
+struct efx_filter_state {
+	spinlock_t	lock;
+	struct efx_filter_table table[EFX_FILTER_TABLE_COUNT];
+#ifdef CONFIG_RFS_ACCEL
+	u32		*rps_flow_id;
+	unsigned	rps_expire_index;
+#endif
+};
+
+/* The filter hash function is LFSR polynomial x^16 + x^3 + 1 of a 32-bit
+ * key derived from the n-tuple.  The initial LFSR state is 0xffff. */
+static u16 efx_filter_hash(u32 key)
+{
+	u16 tmp;
+
+	/* First 16 rounds */
+	tmp = 0x1fff ^ key >> 16;
+	tmp = tmp ^ tmp >> 3 ^ tmp >> 6;
+	tmp = tmp ^ tmp >> 9;
+	/* Last 16 rounds */
+	tmp = tmp ^ tmp << 13 ^ key;
+	tmp = tmp ^ tmp >> 3 ^ tmp >> 6;
+	return tmp ^ tmp >> 9;
+}
+
+/* To allow for hash collisions, filter search continues at these
+ * increments from the first possible entry selected by the hash. */
+static u16 efx_filter_increment(u32 key)
+{
+	return key * 2 - 1;
+}
+
+static enum efx_filter_table_id
+efx_filter_spec_table_id(const struct efx_filter_spec *spec)
+{
+	BUILD_BUG_ON(EFX_FILTER_TABLE_RX_IP != (EFX_FILTER_TCP_FULL >> 2));
+	BUILD_BUG_ON(EFX_FILTER_TABLE_RX_IP != (EFX_FILTER_TCP_WILD >> 2));
+	BUILD_BUG_ON(EFX_FILTER_TABLE_RX_IP != (EFX_FILTER_UDP_FULL >> 2));
+	BUILD_BUG_ON(EFX_FILTER_TABLE_RX_IP != (EFX_FILTER_UDP_WILD >> 2));
+	BUILD_BUG_ON(EFX_FILTER_TABLE_RX_MAC != (EFX_FILTER_MAC_FULL >> 2));
+	BUILD_BUG_ON(EFX_FILTER_TABLE_RX_MAC != (EFX_FILTER_MAC_WILD >> 2));
+	EFX_BUG_ON_PARANOID(spec->type == EFX_FILTER_UNSPEC);
+	return spec->type >> 2;
+}
+
+static struct efx_filter_table *
+efx_filter_spec_table(struct efx_filter_state *state,
+		      const struct efx_filter_spec *spec)
+{
+	if (spec->type == EFX_FILTER_UNSPEC)
+		return NULL;
+	else
+		return &state->table[efx_filter_spec_table_id(spec)];
+}
+
+static void efx_filter_table_reset_search_depth(struct efx_filter_table *table)
+{
+	memset(table->search_depth, 0, sizeof(table->search_depth));
+}
+
+static void efx_filter_push_rx_limits(struct efx_nic *efx)
+{
+	struct efx_filter_state *state = efx->filter_state;
+	struct efx_filter_table *table;
+	efx_oword_t filter_ctl;
+
+	efx_reado(efx, &filter_ctl, FR_BZ_RX_FILTER_CTL);
+
+	table = &state->table[EFX_FILTER_TABLE_RX_IP];
+	EFX_SET_OWORD_FIELD(filter_ctl, FRF_BZ_TCP_FULL_SRCH_LIMIT,
+			    table->search_depth[EFX_FILTER_TCP_FULL] +
+			    FILTER_CTL_SRCH_FUDGE_FULL);
+	EFX_SET_OWORD_FIELD(filter_ctl, FRF_BZ_TCP_WILD_SRCH_LIMIT,
+			    table->search_depth[EFX_FILTER_TCP_WILD] +
+			    FILTER_CTL_SRCH_FUDGE_WILD);
+	EFX_SET_OWORD_FIELD(filter_ctl, FRF_BZ_UDP_FULL_SRCH_LIMIT,
+			    table->search_depth[EFX_FILTER_UDP_FULL] +
+			    FILTER_CTL_SRCH_FUDGE_FULL);
+	EFX_SET_OWORD_FIELD(filter_ctl, FRF_BZ_UDP_WILD_SRCH_LIMIT,
+			    table->search_depth[EFX_FILTER_UDP_WILD] +
+			    FILTER_CTL_SRCH_FUDGE_WILD);
+
+	table = &state->table[EFX_FILTER_TABLE_RX_MAC];
+	if (table->size) {
+		EFX_SET_OWORD_FIELD(
+			filter_ctl, FRF_CZ_ETHERNET_FULL_SEARCH_LIMIT,
+			table->search_depth[EFX_FILTER_MAC_FULL] +
+			FILTER_CTL_SRCH_FUDGE_FULL);
+		EFX_SET_OWORD_FIELD(
+			filter_ctl, FRF_CZ_ETHERNET_WILDCARD_SEARCH_LIMIT,
+			table->search_depth[EFX_FILTER_MAC_WILD] +
+			FILTER_CTL_SRCH_FUDGE_WILD);
+	}
+
+	efx_writeo(efx, &filter_ctl, FR_BZ_RX_FILTER_CTL);
+}
+
+static inline void __efx_filter_set_ipv4(struct efx_filter_spec *spec,
+					 __be32 host1, __be16 port1,
+					 __be32 host2, __be16 port2)
+{
+	spec->data[0] = ntohl(host1) << 16 | ntohs(port1);
+	spec->data[1] = ntohs(port2) << 16 | ntohl(host1) >> 16;
+	spec->data[2] = ntohl(host2);
+}
+
+/**
+ * efx_filter_set_ipv4_local - specify IPv4 host, transport protocol and port
+ * @spec: Specification to initialise
+ * @proto: Transport layer protocol number
+ * @host: Local host address (network byte order)
+ * @port: Local port (network byte order)
+ */
+int efx_filter_set_ipv4_local(struct efx_filter_spec *spec, u8 proto,
+			      __be32 host, __be16 port)
+{
+	__be32 host1;
+	__be16 port1;
+
+	EFX_BUG_ON_PARANOID(!(spec->flags & EFX_FILTER_FLAG_RX));
+
+	/* This cannot currently be combined with other filtering */
+	if (spec->type != EFX_FILTER_UNSPEC)
+		return -EPROTONOSUPPORT;
+
+	if (port == 0)
+		return -EINVAL;
+
+	switch (proto) {
+	case IPPROTO_TCP:
+		spec->type = EFX_FILTER_TCP_WILD;
+		break;
+	case IPPROTO_UDP:
+		spec->type = EFX_FILTER_UDP_WILD;
+		break;
+	default:
+		return -EPROTONOSUPPORT;
+	}
+
+	/* Filter is constructed in terms of source and destination,
+	 * with the odd wrinkle that the ports are swapped in a UDP
+	 * wildcard filter.  We need to convert from local and remote
+	 * (= zero for wildcard) addresses.
+	 */
+	host1 = 0;
+	if (proto != IPPROTO_UDP) {
+		port1 = 0;
+	} else {
+		port1 = port;
+		port = 0;
+	}
+
+	__efx_filter_set_ipv4(spec, host1, port1, host, port);
+	return 0;
+}
+
+/**
+ * efx_filter_set_ipv4_full - specify IPv4 hosts, transport protocol and ports
+ * @spec: Specification to initialise
+ * @proto: Transport layer protocol number
+ * @host: Local host address (network byte order)
+ * @port: Local port (network byte order)
+ * @rhost: Remote host address (network byte order)
+ * @rport: Remote port (network byte order)
+ */
+int efx_filter_set_ipv4_full(struct efx_filter_spec *spec, u8 proto,
+			     __be32 host, __be16 port,
+			     __be32 rhost, __be16 rport)
+{
+	EFX_BUG_ON_PARANOID(!(spec->flags & EFX_FILTER_FLAG_RX));
+
+	/* This cannot currently be combined with other filtering */
+	if (spec->type != EFX_FILTER_UNSPEC)
+		return -EPROTONOSUPPORT;
+
+	if (port == 0 || rport == 0)
+		return -EINVAL;
+
+	switch (proto) {
+	case IPPROTO_TCP:
+		spec->type = EFX_FILTER_TCP_FULL;
+		break;
+	case IPPROTO_UDP:
+		spec->type = EFX_FILTER_UDP_FULL;
+		break;
+	default:
+		return -EPROTONOSUPPORT;
+	}
+
+	__efx_filter_set_ipv4(spec, rhost, rport, host, port);
+	return 0;
+}
+
+/**
+ * efx_filter_set_eth_local - specify local Ethernet address and optional VID
+ * @spec: Specification to initialise
+ * @vid: VLAN ID to match, or %EFX_FILTER_VID_UNSPEC
+ * @addr: Local Ethernet MAC address
+ */
+int efx_filter_set_eth_local(struct efx_filter_spec *spec,
+			     u16 vid, const u8 *addr)
+{
+	EFX_BUG_ON_PARANOID(!(spec->flags & EFX_FILTER_FLAG_RX));
+
+	/* This cannot currently be combined with other filtering */
+	if (spec->type != EFX_FILTER_UNSPEC)
+		return -EPROTONOSUPPORT;
+
+	if (vid == EFX_FILTER_VID_UNSPEC) {
+		spec->type = EFX_FILTER_MAC_WILD;
+		spec->data[0] = 0;
+	} else {
+		spec->type = EFX_FILTER_MAC_FULL;
+		spec->data[0] = vid;
+	}
+
+	spec->data[1] = addr[2] << 24 | addr[3] << 16 | addr[4] << 8 | addr[5];
+	spec->data[2] = addr[0] << 8 | addr[1];
+	return 0;
+}
+
+/* Build a filter entry and return its n-tuple key. */
+static u32 efx_filter_build(efx_oword_t *filter, struct efx_filter_spec *spec)
+{
+	u32 data3;
+
+	switch (efx_filter_spec_table_id(spec)) {
+	case EFX_FILTER_TABLE_RX_IP: {
+		bool is_udp = (spec->type == EFX_FILTER_UDP_FULL ||
+			       spec->type == EFX_FILTER_UDP_WILD);
+		EFX_POPULATE_OWORD_7(
+			*filter,
+			FRF_BZ_RSS_EN,
+			!!(spec->flags & EFX_FILTER_FLAG_RX_RSS),
+			FRF_BZ_SCATTER_EN,
+			!!(spec->flags & EFX_FILTER_FLAG_RX_SCATTER),
+			FRF_BZ_TCP_UDP, is_udp,
+			FRF_BZ_RXQ_ID, spec->dmaq_id,
+			EFX_DWORD_2, spec->data[2],
+			EFX_DWORD_1, spec->data[1],
+			EFX_DWORD_0, spec->data[0]);
+		data3 = is_udp;
+		break;
+	}
+
+	case EFX_FILTER_TABLE_RX_MAC: {
+		bool is_wild = spec->type == EFX_FILTER_MAC_WILD;
+		EFX_POPULATE_OWORD_8(
+			*filter,
+			FRF_CZ_RMFT_RSS_EN,
+			!!(spec->flags & EFX_FILTER_FLAG_RX_RSS),
+			FRF_CZ_RMFT_SCATTER_EN,
+			!!(spec->flags & EFX_FILTER_FLAG_RX_SCATTER),
+			FRF_CZ_RMFT_IP_OVERRIDE,
+			!!(spec->flags & EFX_FILTER_FLAG_RX_OVERRIDE_IP),
+			FRF_CZ_RMFT_RXQ_ID, spec->dmaq_id,
+			FRF_CZ_RMFT_WILDCARD_MATCH, is_wild,
+			FRF_CZ_RMFT_DEST_MAC_HI, spec->data[2],
+			FRF_CZ_RMFT_DEST_MAC_LO, spec->data[1],
+			FRF_CZ_RMFT_VLAN_ID, spec->data[0]);
+		data3 = is_wild;
+		break;
+	}
+
+	default:
+		BUG();
+	}
+
+	return spec->data[0] ^ spec->data[1] ^ spec->data[2] ^ data3;
+}
+
+static bool efx_filter_equal(const struct efx_filter_spec *left,
+			     const struct efx_filter_spec *right)
+{
+	if (left->type != right->type ||
+	    memcmp(left->data, right->data, sizeof(left->data)))
+		return false;
+
+	return true;
+}
+
+static int efx_filter_search(struct efx_filter_table *table,
+			     struct efx_filter_spec *spec, u32 key,
+			     bool for_insert, int *depth_required)
+{
+	unsigned hash, incr, filter_idx, depth, depth_max;
+
+	hash = efx_filter_hash(key);
+	incr = efx_filter_increment(key);
+
+	filter_idx = hash & (table->size - 1);
+	depth = 1;
+	depth_max = (for_insert ?
+		     (spec->priority <= EFX_FILTER_PRI_HINT ?
+		      FILTER_CTL_SRCH_HINT_MAX : FILTER_CTL_SRCH_MAX) :
+		     table->search_depth[spec->type]);
+
+	for (;;) {
+		/* Return success if entry is used and matches this spec
+		 * or entry is unused and we are trying to insert.
+		 */
+		if (test_bit(filter_idx, table->used_bitmap) ?
+		    efx_filter_equal(spec, &table->spec[filter_idx]) :
+		    for_insert) {
+			*depth_required = depth;
+			return filter_idx;
+		}
+
+		/* Return failure if we reached the maximum search depth */
+		if (depth == depth_max)
+			return for_insert ? -EBUSY : -ENOENT;
+
+		filter_idx = (filter_idx + incr) & (table->size - 1);
+		++depth;
+	}
+}
+
+/* Construct/deconstruct external filter IDs */
+
+static inline int
+efx_filter_make_id(enum efx_filter_table_id table_id, unsigned index)
+{
+	return table_id << 16 | index;
+}
+
+/**
+ * efx_filter_insert_filter - add or replace a filter
+ * @efx: NIC in which to insert the filter
+ * @spec: Specification for the filter
+ * @replace: Flag for whether the specified filter may replace a filter
+ *	with an identical match expression and equal or lower priority
+ *
+ * On success, return the filter ID.
+ * On failure, return a negative error code.
+ */
+int efx_filter_insert_filter(struct efx_nic *efx, struct efx_filter_spec *spec,
+			     bool replace)
+{
+	struct efx_filter_state *state = efx->filter_state;
+	struct efx_filter_table *table = efx_filter_spec_table(state, spec);
+	struct efx_filter_spec *saved_spec;
+	efx_oword_t filter;
+	int filter_idx, depth;
+	u32 key;
+	int rc;
+
+	if (!table || table->size == 0)
+		return -EINVAL;
+
+	key = efx_filter_build(&filter, spec);
+
+	netif_vdbg(efx, hw, efx->net_dev,
+		   "%s: type %d search_depth=%d", __func__, spec->type,
+		   table->search_depth[spec->type]);
+
+	spin_lock_bh(&state->lock);
+
+	rc = efx_filter_search(table, spec, key, true, &depth);
+	if (rc < 0)
+		goto out;
+	filter_idx = rc;
+	BUG_ON(filter_idx >= table->size);
+	saved_spec = &table->spec[filter_idx];
+
+	if (test_bit(filter_idx, table->used_bitmap)) {
+		/* Should we replace the existing filter? */
+		if (!replace) {
+			rc = -EEXIST;
+			goto out;
+		}
+		if (spec->priority < saved_spec->priority) {
+			rc = -EPERM;
+			goto out;
+		}
+	} else {
+		__set_bit(filter_idx, table->used_bitmap);
+		++table->used;
+	}
+	*saved_spec = *spec;
+
+	if (table->search_depth[spec->type] < depth) {
+		table->search_depth[spec->type] = depth;
+		efx_filter_push_rx_limits(efx);
+	}
+
+	efx_writeo(efx, &filter, table->offset + table->step * filter_idx);
+
+	netif_vdbg(efx, hw, efx->net_dev,
+		   "%s: filter type %d index %d rxq %u set",
+		   __func__, spec->type, filter_idx, spec->dmaq_id);
+	rc = efx_filter_make_id(table->id, filter_idx);
+
+out:
+	spin_unlock_bh(&state->lock);
+	return rc;
+}
+
+static void efx_filter_table_clear_entry(struct efx_nic *efx,
+					 struct efx_filter_table *table,
+					 int filter_idx)
+{
+	static efx_oword_t filter;
+
+	if (test_bit(filter_idx, table->used_bitmap)) {
+		__clear_bit(filter_idx, table->used_bitmap);
+		--table->used;
+		memset(&table->spec[filter_idx], 0, sizeof(table->spec[0]));
+
+		efx_writeo(efx, &filter,
+			   table->offset + table->step * filter_idx);
+	}
+}
+
+/**
+ * efx_filter_remove_filter - remove a filter by specification
+ * @efx: NIC from which to remove the filter
+ * @spec: Specification for the filter
+ *
+ * On success, return zero.
+ * On failure, return a negative error code.
+ */
+int efx_filter_remove_filter(struct efx_nic *efx, struct efx_filter_spec *spec)
+{
+	struct efx_filter_state *state = efx->filter_state;
+	struct efx_filter_table *table = efx_filter_spec_table(state, spec);
+	struct efx_filter_spec *saved_spec;
+	efx_oword_t filter;
+	int filter_idx, depth;
+	u32 key;
+	int rc;
+
+	if (!table)
+		return -EINVAL;
+
+	key = efx_filter_build(&filter, spec);
+
+	spin_lock_bh(&state->lock);
+
+	rc = efx_filter_search(table, spec, key, false, &depth);
+	if (rc < 0)
+		goto out;
+	filter_idx = rc;
+	saved_spec = &table->spec[filter_idx];
+
+	if (spec->priority < saved_spec->priority) {
+		rc = -EPERM;
+		goto out;
+	}
+
+	efx_filter_table_clear_entry(efx, table, filter_idx);
+	if (table->used == 0)
+		efx_filter_table_reset_search_depth(table);
+	rc = 0;
+
+out:
+	spin_unlock_bh(&state->lock);
+	return rc;
+}
+
+static void efx_filter_table_clear(struct efx_nic *efx,
+				   enum efx_filter_table_id table_id,
+				   enum efx_filter_priority priority)
+{
+	struct efx_filter_state *state = efx->filter_state;
+	struct efx_filter_table *table = &state->table[table_id];
+	int filter_idx;
+
+	spin_lock_bh(&state->lock);
+
+	for (filter_idx = 0; filter_idx < table->size; ++filter_idx)
+		if (table->spec[filter_idx].priority <= priority)
+			efx_filter_table_clear_entry(efx, table, filter_idx);
+	if (table->used == 0)
+		efx_filter_table_reset_search_depth(table);
+
+	spin_unlock_bh(&state->lock);
+}
+
+/**
+ * efx_filter_clear_rx - remove RX filters by priority
+ * @efx: NIC from which to remove the filters
+ * @priority: Maximum priority to remove
+ */
+void efx_filter_clear_rx(struct efx_nic *efx, enum efx_filter_priority priority)
+{
+	efx_filter_table_clear(efx, EFX_FILTER_TABLE_RX_IP, priority);
+	efx_filter_table_clear(efx, EFX_FILTER_TABLE_RX_MAC, priority);
+}
+
+/* Restore filter stater after reset */
+void efx_restore_filters(struct efx_nic *efx)
+{
+	struct efx_filter_state *state = efx->filter_state;
+	enum efx_filter_table_id table_id;
+	struct efx_filter_table *table;
+	efx_oword_t filter;
+	int filter_idx;
+
+	spin_lock_bh(&state->lock);
+
+	for (table_id = 0; table_id < EFX_FILTER_TABLE_COUNT; table_id++) {
+		table = &state->table[table_id];
+		for (filter_idx = 0; filter_idx < table->size; filter_idx++) {
+			if (!test_bit(filter_idx, table->used_bitmap))
+				continue;
+			efx_filter_build(&filter, &table->spec[filter_idx]);
+			efx_writeo(efx, &filter,
+				   table->offset + table->step * filter_idx);
+		}
+	}
+
+	efx_filter_push_rx_limits(efx);
+
+	spin_unlock_bh(&state->lock);
+}
+
+int efx_probe_filters(struct efx_nic *efx)
+{
+	struct efx_filter_state *state;
+	struct efx_filter_table *table;
+	unsigned table_id;
+
+	state = kzalloc(sizeof(*efx->filter_state), GFP_KERNEL);
+	if (!state)
+		return -ENOMEM;
+	efx->filter_state = state;
+
+	spin_lock_init(&state->lock);
+
+	if (efx_nic_rev(efx) >= EFX_REV_FALCON_B0) {
+#ifdef CONFIG_RFS_ACCEL
+		state->rps_flow_id = kcalloc(FR_BZ_RX_FILTER_TBL0_ROWS,
+					     sizeof(*state->rps_flow_id),
+					     GFP_KERNEL);
+		if (!state->rps_flow_id)
+			goto fail;
+#endif
+		table = &state->table[EFX_FILTER_TABLE_RX_IP];
+		table->id = EFX_FILTER_TABLE_RX_IP;
+		table->offset = FR_BZ_RX_FILTER_TBL0;
+		table->size = FR_BZ_RX_FILTER_TBL0_ROWS;
+		table->step = FR_BZ_RX_FILTER_TBL0_STEP;
+	}
+
+	if (efx_nic_rev(efx) >= EFX_REV_SIENA_A0) {
+		table = &state->table[EFX_FILTER_TABLE_RX_MAC];
+		table->id = EFX_FILTER_TABLE_RX_MAC;
+		table->offset = FR_CZ_RX_MAC_FILTER_TBL0;
+		table->size = FR_CZ_RX_MAC_FILTER_TBL0_ROWS;
+		table->step = FR_CZ_RX_MAC_FILTER_TBL0_STEP;
+	}
+
+	for (table_id = 0; table_id < EFX_FILTER_TABLE_COUNT; table_id++) {
+		table = &state->table[table_id];
+		if (table->size == 0)
+			continue;
+		table->used_bitmap = kcalloc(BITS_TO_LONGS(table->size),
+					     sizeof(unsigned long),
+					     GFP_KERNEL);
+		if (!table->used_bitmap)
+			goto fail;
+		table->spec = vzalloc(table->size * sizeof(*table->spec));
+		if (!table->spec)
+			goto fail;
+	}
+
+	return 0;
+
+fail:
+	efx_remove_filters(efx);
+	return -ENOMEM;
+}
+
+void efx_remove_filters(struct efx_nic *efx)
+{
+	struct efx_filter_state *state = efx->filter_state;
+	enum efx_filter_table_id table_id;
+
+	for (table_id = 0; table_id < EFX_FILTER_TABLE_COUNT; table_id++) {
+		kfree(state->table[table_id].used_bitmap);
+		vfree(state->table[table_id].spec);
+	}
+#ifdef CONFIG_RFS_ACCEL
+	kfree(state->rps_flow_id);
+#endif
+	kfree(state);
+}
+
+#ifdef CONFIG_RFS_ACCEL
+
+int efx_filter_rfs(struct net_device *net_dev, const struct sk_buff *skb,
+		   u16 rxq_index, u32 flow_id)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	struct efx_channel *channel;
+	struct efx_filter_state *state = efx->filter_state;
+	struct efx_filter_spec spec;
+	const struct iphdr *ip;
+	const __be16 *ports;
+	int nhoff;
+	int rc;
+
+	nhoff = skb_network_offset(skb);
+
+	if (skb->protocol != htons(ETH_P_IP))
+		return -EPROTONOSUPPORT;
+
+	/* RFS must validate the IP header length before calling us */
+	EFX_BUG_ON_PARANOID(skb_headlen(skb) < nhoff + sizeof(*ip));
+	ip = (const struct iphdr *)(skb->data + nhoff);
+	if (ip_is_fragment(ip))
+		return -EPROTONOSUPPORT;
+	EFX_BUG_ON_PARANOID(skb_headlen(skb) < nhoff + 4 * ip->ihl + 4);
+	ports = (const __be16 *)(skb->data + nhoff + 4 * ip->ihl);
+
+	efx_filter_init_rx(&spec, EFX_FILTER_PRI_HINT, 0, rxq_index);
+	rc = efx_filter_set_ipv4_full(&spec, ip->protocol,
+				      ip->daddr, ports[1], ip->saddr, ports[0]);
+	if (rc)
+		return rc;
+
+	rc = efx_filter_insert_filter(efx, &spec, true);
+	if (rc < 0)
+		return rc;
+
+	/* Remember this so we can check whether to expire the filter later */
+	state->rps_flow_id[rc] = flow_id;
+	channel = efx_get_channel(efx, skb_get_rx_queue(skb));
+	++channel->rfs_filters_added;
+
+	netif_info(efx, rx_status, efx->net_dev,
+		   "steering %s %pI4:%u:%pI4:%u to queue %u [flow %u filter %d]\n",
+		   (ip->protocol == IPPROTO_TCP) ? "TCP" : "UDP",
+		   &ip->saddr, ntohs(ports[0]), &ip->daddr, ntohs(ports[1]),
+		   rxq_index, flow_id, rc);
+
+	return rc;
+}
+
+bool __efx_filter_rfs_expire(struct efx_nic *efx, unsigned quota)
+{
+	struct efx_filter_state *state = efx->filter_state;
+	struct efx_filter_table *table = &state->table[EFX_FILTER_TABLE_RX_IP];
+	unsigned mask = table->size - 1;
+	unsigned index;
+	unsigned stop;
+
+	if (!spin_trylock_bh(&state->lock))
+		return false;
+
+	index = state->rps_expire_index;
+	stop = (index + quota) & mask;
+
+	while (index != stop) {
+		if (test_bit(index, table->used_bitmap) &&
+		    table->spec[index].priority == EFX_FILTER_PRI_HINT &&
+		    rps_may_expire_flow(efx->net_dev,
+					table->spec[index].dmaq_id,
+					state->rps_flow_id[index], index)) {
+			netif_info(efx, rx_status, efx->net_dev,
+				   "expiring filter %d [flow %u]\n",
+				   index, state->rps_flow_id[index]);
+			efx_filter_table_clear_entry(efx, table, index);
+		}
+		index = (index + 1) & mask;
+	}
+
+	state->rps_expire_index = stop;
+	if (table->used == 0)
+		efx_filter_table_reset_search_depth(table);
+
+	spin_unlock_bh(&state->lock);
+	return true;
+}
+
+#endif /* CONFIG_RFS_ACCEL */
diff --git a/drivers/net/ethernet/sfc/filter.h b/drivers/net/ethernet/sfc/filter.h
new file mode 100644
index 0000000..872f213
--- /dev/null
+++ b/drivers/net/ethernet/sfc/filter.h
@@ -0,0 +1,112 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2005-2010 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#ifndef EFX_FILTER_H
+#define EFX_FILTER_H
+
+#include <linux/types.h>
+
+/**
+ * enum efx_filter_type - type of hardware filter
+ * @EFX_FILTER_TCP_FULL: Matching TCP/IPv4 4-tuple
+ * @EFX_FILTER_TCP_WILD: Matching TCP/IPv4 destination (host, port)
+ * @EFX_FILTER_UDP_FULL: Matching UDP/IPv4 4-tuple
+ * @EFX_FILTER_UDP_WILD: Matching UDP/IPv4 destination (host, port)
+ * @EFX_FILTER_MAC_FULL: Matching Ethernet destination MAC address, VID
+ * @EFX_FILTER_MAC_WILD: Matching Ethernet destination MAC address
+ * @EFX_FILTER_UNSPEC: Match type is unspecified
+ *
+ * Falcon NICs only support the TCP/IPv4 and UDP/IPv4 filter types.
+ */
+enum efx_filter_type {
+	EFX_FILTER_TCP_FULL = 0,
+	EFX_FILTER_TCP_WILD,
+	EFX_FILTER_UDP_FULL,
+	EFX_FILTER_UDP_WILD,
+	EFX_FILTER_MAC_FULL = 4,
+	EFX_FILTER_MAC_WILD,
+	EFX_FILTER_TYPE_COUNT,		/* number of specific types */
+	EFX_FILTER_UNSPEC = 0xf,
+};
+
+/**
+ * enum efx_filter_priority - priority of a hardware filter specification
+ * @EFX_FILTER_PRI_HINT: Performance hint
+ * @EFX_FILTER_PRI_MANUAL: Manually configured filter
+ * @EFX_FILTER_PRI_REQUIRED: Required for correct behaviour
+ */
+enum efx_filter_priority {
+	EFX_FILTER_PRI_HINT = 0,
+	EFX_FILTER_PRI_MANUAL,
+	EFX_FILTER_PRI_REQUIRED,
+};
+
+/**
+ * enum efx_filter_flags - flags for hardware filter specifications
+ * @EFX_FILTER_FLAG_RX_RSS: Use RSS to spread across multiple queues.
+ *	By default, matching packets will be delivered only to the
+ *	specified queue. If this flag is set, they will be delivered
+ *	to a range of queues offset from the specified queue number
+ *	according to the indirection table.
+ * @EFX_FILTER_FLAG_RX_SCATTER: Enable DMA scatter on the receiving
+ *	queue.
+ * @EFX_FILTER_FLAG_RX_OVERRIDE_IP: Enables a MAC filter to override
+ *	any IP filter that matches the same packet.  By default, IP
+ *	filters take precedence.
+ * @EFX_FILTER_FLAG_RX: Filter is for RX
+ */
+enum efx_filter_flags {
+	EFX_FILTER_FLAG_RX_RSS = 0x01,
+	EFX_FILTER_FLAG_RX_SCATTER = 0x02,
+	EFX_FILTER_FLAG_RX_OVERRIDE_IP = 0x04,
+	EFX_FILTER_FLAG_RX = 0x08,
+};
+
+/**
+ * struct efx_filter_spec - specification for a hardware filter
+ * @type: Type of match to be performed, from &enum efx_filter_type
+ * @priority: Priority of the filter, from &enum efx_filter_priority
+ * @flags: Miscellaneous flags, from &enum efx_filter_flags
+ * @dmaq_id: Source/target queue index
+ * @data: Match data (type-dependent)
+ *
+ * Use the efx_filter_set_*() functions to initialise the @type and
+ * @data fields.
+ */
+struct efx_filter_spec {
+	u8	type:4;
+	u8	priority:4;
+	u8	flags;
+	u16	dmaq_id;
+	u32	data[3];
+};
+
+static inline void efx_filter_init_rx(struct efx_filter_spec *spec,
+				      enum efx_filter_priority priority,
+				      enum efx_filter_flags flags,
+				      unsigned rxq_id)
+{
+	spec->type = EFX_FILTER_UNSPEC;
+	spec->priority = priority;
+	spec->flags = EFX_FILTER_FLAG_RX | flags;
+	spec->dmaq_id = rxq_id;
+}
+
+extern int efx_filter_set_ipv4_local(struct efx_filter_spec *spec, u8 proto,
+				     __be32 host, __be16 port);
+extern int efx_filter_set_ipv4_full(struct efx_filter_spec *spec, u8 proto,
+				    __be32 host, __be16 port,
+				    __be32 rhost, __be16 rport);
+extern int efx_filter_set_eth_local(struct efx_filter_spec *spec,
+				    u16 vid, const u8 *addr);
+enum {
+	EFX_FILTER_VID_UNSPEC = 0xffff,
+};
+
+#endif /* EFX_FILTER_H */
diff --git a/drivers/net/ethernet/sfc/io.h b/drivers/net/ethernet/sfc/io.h
new file mode 100644
index 0000000..cc97880
--- /dev/null
+++ b/drivers/net/ethernet/sfc/io.h
@@ -0,0 +1,299 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2005-2006 Fen Systems Ltd.
+ * Copyright 2006-2010 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#ifndef EFX_IO_H
+#define EFX_IO_H
+
+#include <linux/io.h>
+#include <linux/spinlock.h>
+
+/**************************************************************************
+ *
+ * NIC register I/O
+ *
+ **************************************************************************
+ *
+ * Notes on locking strategy:
+ *
+ * Most CSRs are 128-bit (oword) and therefore cannot be read or
+ * written atomically.  Access from the host is buffered by the Bus
+ * Interface Unit (BIU).  Whenever the host reads from the lowest
+ * address of such a register, or from the address of a different such
+ * register, the BIU latches the register's value.  Subsequent reads
+ * from higher addresses of the same register will read the latched
+ * value.  Whenever the host writes part of such a register, the BIU
+ * collects the written value and does not write to the underlying
+ * register until all 4 dwords have been written.  A similar buffering
+ * scheme applies to host access to the NIC's 64-bit SRAM.
+ *
+ * Access to different CSRs and 64-bit SRAM words must be serialised,
+ * since interleaved access can result in lost writes or lost
+ * information from read-to-clear fields.  We use efx_nic::biu_lock
+ * for this.  (We could use separate locks for read and write, but
+ * this is not normally a performance bottleneck.)
+ *
+ * The DMA descriptor pointers (RX_DESC_UPD and TX_DESC_UPD) are
+ * 128-bit but are special-cased in the BIU to avoid the need for
+ * locking in the host:
+ *
+ * - They are write-only.
+ * - The semantics of writing to these registers are such that
+ *   replacing the low 96 bits with zero does not affect functionality.
+ * - If the host writes to the last dword address of such a register
+ *   (i.e. the high 32 bits) the underlying register will always be
+ *   written.  If the collector and the current write together do not
+ *   provide values for all 128 bits of the register, the low 96 bits
+ *   will be written as zero.
+ * - If the host writes to the address of any other part of such a
+ *   register while the collector already holds values for some other
+ *   register, the write is discarded and the collector maintains its
+ *   current state.
+ */
+
+#if BITS_PER_LONG == 64
+#define EFX_USE_QWORD_IO 1
+#endif
+
+#ifdef EFX_USE_QWORD_IO
+static inline void _efx_writeq(struct efx_nic *efx, __le64 value,
+				  unsigned int reg)
+{
+	__raw_writeq((__force u64)value, efx->membase + reg);
+}
+static inline __le64 _efx_readq(struct efx_nic *efx, unsigned int reg)
+{
+	return (__force __le64)__raw_readq(efx->membase + reg);
+}
+#endif
+
+static inline void _efx_writed(struct efx_nic *efx, __le32 value,
+				  unsigned int reg)
+{
+	__raw_writel((__force u32)value, efx->membase + reg);
+}
+static inline __le32 _efx_readd(struct efx_nic *efx, unsigned int reg)
+{
+	return (__force __le32)__raw_readl(efx->membase + reg);
+}
+
+/* Write a normal 128-bit CSR, locking as appropriate. */
+static inline void efx_writeo(struct efx_nic *efx, efx_oword_t *value,
+			      unsigned int reg)
+{
+	unsigned long flags __attribute__ ((unused));
+
+	netif_vdbg(efx, hw, efx->net_dev,
+		   "writing register %x with " EFX_OWORD_FMT "\n", reg,
+		   EFX_OWORD_VAL(*value));
+
+	spin_lock_irqsave(&efx->biu_lock, flags);
+#ifdef EFX_USE_QWORD_IO
+	_efx_writeq(efx, value->u64[0], reg + 0);
+	_efx_writeq(efx, value->u64[1], reg + 8);
+#else
+	_efx_writed(efx, value->u32[0], reg + 0);
+	_efx_writed(efx, value->u32[1], reg + 4);
+	_efx_writed(efx, value->u32[2], reg + 8);
+	_efx_writed(efx, value->u32[3], reg + 12);
+#endif
+	wmb();
+	mmiowb();
+	spin_unlock_irqrestore(&efx->biu_lock, flags);
+}
+
+/* Write 64-bit SRAM through the supplied mapping, locking as appropriate. */
+static inline void efx_sram_writeq(struct efx_nic *efx, void __iomem *membase,
+				   efx_qword_t *value, unsigned int index)
+{
+	unsigned int addr = index * sizeof(*value);
+	unsigned long flags __attribute__ ((unused));
+
+	netif_vdbg(efx, hw, efx->net_dev,
+		   "writing SRAM address %x with " EFX_QWORD_FMT "\n",
+		   addr, EFX_QWORD_VAL(*value));
+
+	spin_lock_irqsave(&efx->biu_lock, flags);
+#ifdef EFX_USE_QWORD_IO
+	__raw_writeq((__force u64)value->u64[0], membase + addr);
+#else
+	__raw_writel((__force u32)value->u32[0], membase + addr);
+	__raw_writel((__force u32)value->u32[1], membase + addr + 4);
+#endif
+	wmb();
+	mmiowb();
+	spin_unlock_irqrestore(&efx->biu_lock, flags);
+}
+
+/* Write a 32-bit CSR or the last dword of a special 128-bit CSR */
+static inline void efx_writed(struct efx_nic *efx, efx_dword_t *value,
+			      unsigned int reg)
+{
+	netif_vdbg(efx, hw, efx->net_dev,
+		   "writing register %x with "EFX_DWORD_FMT"\n",
+		   reg, EFX_DWORD_VAL(*value));
+
+	/* No lock required */
+	_efx_writed(efx, value->u32[0], reg);
+	wmb();
+}
+
+/* Read a 128-bit CSR, locking as appropriate. */
+static inline void efx_reado(struct efx_nic *efx, efx_oword_t *value,
+			     unsigned int reg)
+{
+	unsigned long flags __attribute__ ((unused));
+
+	spin_lock_irqsave(&efx->biu_lock, flags);
+	value->u32[0] = _efx_readd(efx, reg + 0);
+	rmb();
+	value->u32[1] = _efx_readd(efx, reg + 4);
+	value->u32[2] = _efx_readd(efx, reg + 8);
+	value->u32[3] = _efx_readd(efx, reg + 12);
+	spin_unlock_irqrestore(&efx->biu_lock, flags);
+
+	netif_vdbg(efx, hw, efx->net_dev,
+		   "read from register %x, got " EFX_OWORD_FMT "\n", reg,
+		   EFX_OWORD_VAL(*value));
+}
+
+/* Read 64-bit SRAM through the supplied mapping, locking as appropriate. */
+static inline void efx_sram_readq(struct efx_nic *efx, void __iomem *membase,
+				  efx_qword_t *value, unsigned int index)
+{
+	unsigned int addr = index * sizeof(*value);
+	unsigned long flags __attribute__ ((unused));
+
+	spin_lock_irqsave(&efx->biu_lock, flags);
+#ifdef EFX_USE_QWORD_IO
+	value->u64[0] = (__force __le64)__raw_readq(membase + addr);
+#else
+	value->u32[0] = (__force __le32)__raw_readl(membase + addr);
+	rmb();
+	value->u32[1] = (__force __le32)__raw_readl(membase + addr + 4);
+#endif
+	spin_unlock_irqrestore(&efx->biu_lock, flags);
+
+	netif_vdbg(efx, hw, efx->net_dev,
+		   "read from SRAM address %x, got "EFX_QWORD_FMT"\n",
+		   addr, EFX_QWORD_VAL(*value));
+}
+
+/* Read a 32-bit CSR or SRAM */
+static inline void efx_readd(struct efx_nic *efx, efx_dword_t *value,
+				unsigned int reg)
+{
+	value->u32[0] = _efx_readd(efx, reg);
+	netif_vdbg(efx, hw, efx->net_dev,
+		   "read from register %x, got "EFX_DWORD_FMT"\n",
+		   reg, EFX_DWORD_VAL(*value));
+}
+
+/* Write a 128-bit CSR forming part of a table */
+static inline void efx_writeo_table(struct efx_nic *efx, efx_oword_t *value,
+				      unsigned int reg, unsigned int index)
+{
+	efx_writeo(efx, value, reg + index * sizeof(efx_oword_t));
+}
+
+/* Read a 128-bit CSR forming part of a table */
+static inline void efx_reado_table(struct efx_nic *efx, efx_oword_t *value,
+				     unsigned int reg, unsigned int index)
+{
+	efx_reado(efx, value, reg + index * sizeof(efx_oword_t));
+}
+
+/* Write a 32-bit CSR forming part of a table, or 32-bit SRAM */
+static inline void efx_writed_table(struct efx_nic *efx, efx_dword_t *value,
+				       unsigned int reg, unsigned int index)
+{
+	efx_writed(efx, value, reg + index * sizeof(efx_oword_t));
+}
+
+/* Read a 32-bit CSR forming part of a table, or 32-bit SRAM */
+static inline void efx_readd_table(struct efx_nic *efx, efx_dword_t *value,
+				   unsigned int reg, unsigned int index)
+{
+	efx_readd(efx, value, reg + index * sizeof(efx_dword_t));
+}
+
+/* Page-mapped register block size */
+#define EFX_PAGE_BLOCK_SIZE 0x2000
+
+/* Calculate offset to page-mapped register block */
+#define EFX_PAGED_REG(page, reg) \
+	((page) * EFX_PAGE_BLOCK_SIZE + (reg))
+
+/* Write the whole of RX_DESC_UPD or TX_DESC_UPD */
+static inline void _efx_writeo_page(struct efx_nic *efx, efx_oword_t *value,
+				    unsigned int reg, unsigned int page)
+{
+	reg = EFX_PAGED_REG(page, reg);
+
+	netif_vdbg(efx, hw, efx->net_dev,
+		   "writing register %x with " EFX_OWORD_FMT "\n", reg,
+		   EFX_OWORD_VAL(*value));
+
+#ifdef EFX_USE_QWORD_IO
+	_efx_writeq(efx, value->u64[0], reg + 0);
+	_efx_writeq(efx, value->u64[1], reg + 8);
+#else
+	_efx_writed(efx, value->u32[0], reg + 0);
+	_efx_writed(efx, value->u32[1], reg + 4);
+	_efx_writed(efx, value->u32[2], reg + 8);
+	_efx_writed(efx, value->u32[3], reg + 12);
+#endif
+	wmb();
+}
+#define efx_writeo_page(efx, value, reg, page)				\
+	_efx_writeo_page(efx, value,					\
+			 reg +						\
+			 BUILD_BUG_ON_ZERO((reg) != 0x830 && (reg) != 0xa10), \
+			 page)
+
+/* Write a page-mapped 32-bit CSR (EVQ_RPTR or the high bits of
+ * RX_DESC_UPD or TX_DESC_UPD)
+ */
+static inline void _efx_writed_page(struct efx_nic *efx, efx_dword_t *value,
+				    unsigned int reg, unsigned int page)
+{
+	efx_writed(efx, value, EFX_PAGED_REG(page, reg));
+}
+#define efx_writed_page(efx, value, reg, page)				\
+	_efx_writed_page(efx, value,					\
+			 reg +						\
+			 BUILD_BUG_ON_ZERO((reg) != 0x400 && (reg) != 0x83c \
+					   && (reg) != 0xa1c),		\
+			 page)
+
+/* Write TIMER_COMMAND.  This is a page-mapped 32-bit CSR, but a bug
+ * in the BIU means that writes to TIMER_COMMAND[0] invalidate the
+ * collector register.
+ */
+static inline void _efx_writed_page_locked(struct efx_nic *efx,
+					   efx_dword_t *value,
+					   unsigned int reg,
+					   unsigned int page)
+{
+	unsigned long flags __attribute__ ((unused));
+
+	if (page == 0) {
+		spin_lock_irqsave(&efx->biu_lock, flags);
+		efx_writed(efx, value, EFX_PAGED_REG(page, reg));
+		spin_unlock_irqrestore(&efx->biu_lock, flags);
+	} else {
+		efx_writed(efx, value, EFX_PAGED_REG(page, reg));
+	}
+}
+#define efx_writed_page_locked(efx, value, reg, page)			\
+	_efx_writed_page_locked(efx, value,				\
+				reg + BUILD_BUG_ON_ZERO((reg) != 0x420), \
+				page)
+
+#endif /* EFX_IO_H */
diff --git a/drivers/net/ethernet/sfc/mac.h b/drivers/net/ethernet/sfc/mac.h
new file mode 100644
index 0000000..d6a255d
--- /dev/null
+++ b/drivers/net/ethernet/sfc/mac.h
@@ -0,0 +1,21 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2005-2006 Fen Systems Ltd.
+ * Copyright 2006-2009 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#ifndef EFX_MAC_H
+#define EFX_MAC_H
+
+#include "net_driver.h"
+
+extern const struct efx_mac_operations falcon_xmac_operations;
+extern const struct efx_mac_operations efx_mcdi_mac_operations;
+extern int efx_mcdi_mac_stats(struct efx_nic *efx, dma_addr_t dma_addr,
+			      u32 dma_len, int enable, int clear);
+
+#endif
diff --git a/drivers/net/ethernet/sfc/mcdi.c b/drivers/net/ethernet/sfc/mcdi.c
new file mode 100644
index 0000000..3dd45ed
--- /dev/null
+++ b/drivers/net/ethernet/sfc/mcdi.c
@@ -0,0 +1,1203 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2008-2011 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#include <linux/delay.h>
+#include "net_driver.h"
+#include "nic.h"
+#include "io.h"
+#include "regs.h"
+#include "mcdi_pcol.h"
+#include "phy.h"
+
+/**************************************************************************
+ *
+ * Management-Controller-to-Driver Interface
+ *
+ **************************************************************************
+ */
+
+/* Software-defined structure to the shared-memory */
+#define CMD_NOTIFY_PORT0 0
+#define CMD_NOTIFY_PORT1 4
+#define CMD_PDU_PORT0    0x008
+#define CMD_PDU_PORT1    0x108
+#define REBOOT_FLAG_PORT0 0x3f8
+#define REBOOT_FLAG_PORT1 0x3fc
+
+#define MCDI_RPC_TIMEOUT       10 /*seconds */
+
+#define MCDI_PDU(efx)							\
+	(efx_port_num(efx) ? CMD_PDU_PORT1 : CMD_PDU_PORT0)
+#define MCDI_DOORBELL(efx)						\
+	(efx_port_num(efx) ? CMD_NOTIFY_PORT1 : CMD_NOTIFY_PORT0)
+#define MCDI_REBOOT_FLAG(efx)						\
+	(efx_port_num(efx) ? REBOOT_FLAG_PORT1 : REBOOT_FLAG_PORT0)
+
+#define SEQ_MASK							\
+	EFX_MASK32(EFX_WIDTH(MCDI_HEADER_SEQ))
+
+static inline struct efx_mcdi_iface *efx_mcdi(struct efx_nic *efx)
+{
+	struct siena_nic_data *nic_data;
+	EFX_BUG_ON_PARANOID(efx_nic_rev(efx) < EFX_REV_SIENA_A0);
+	nic_data = efx->nic_data;
+	return &nic_data->mcdi;
+}
+
+static inline void
+efx_mcdi_readd(struct efx_nic *efx, efx_dword_t *value, unsigned reg)
+{
+	struct siena_nic_data *nic_data = efx->nic_data;
+	value->u32[0] = (__force __le32)__raw_readl(nic_data->mcdi_smem + reg);
+}
+
+static inline void
+efx_mcdi_writed(struct efx_nic *efx, const efx_dword_t *value, unsigned reg)
+{
+	struct siena_nic_data *nic_data = efx->nic_data;
+	__raw_writel((__force u32)value->u32[0], nic_data->mcdi_smem + reg);
+}
+
+void efx_mcdi_init(struct efx_nic *efx)
+{
+	struct efx_mcdi_iface *mcdi;
+
+	if (efx_nic_rev(efx) < EFX_REV_SIENA_A0)
+		return;
+
+	mcdi = efx_mcdi(efx);
+	init_waitqueue_head(&mcdi->wq);
+	spin_lock_init(&mcdi->iface_lock);
+	atomic_set(&mcdi->state, MCDI_STATE_QUIESCENT);
+	mcdi->mode = MCDI_MODE_POLL;
+
+	(void) efx_mcdi_poll_reboot(efx);
+}
+
+static void efx_mcdi_copyin(struct efx_nic *efx, unsigned cmd,
+			    const u8 *inbuf, size_t inlen)
+{
+	struct efx_mcdi_iface *mcdi = efx_mcdi(efx);
+	unsigned pdu = MCDI_PDU(efx);
+	unsigned doorbell = MCDI_DOORBELL(efx);
+	unsigned int i;
+	efx_dword_t hdr;
+	u32 xflags, seqno;
+
+	BUG_ON(atomic_read(&mcdi->state) == MCDI_STATE_QUIESCENT);
+	BUG_ON(inlen & 3 || inlen >= 0x100);
+
+	seqno = mcdi->seqno & SEQ_MASK;
+	xflags = 0;
+	if (mcdi->mode == MCDI_MODE_EVENTS)
+		xflags |= MCDI_HEADER_XFLAGS_EVREQ;
+
+	EFX_POPULATE_DWORD_6(hdr,
+			     MCDI_HEADER_RESPONSE, 0,
+			     MCDI_HEADER_RESYNC, 1,
+			     MCDI_HEADER_CODE, cmd,
+			     MCDI_HEADER_DATALEN, inlen,
+			     MCDI_HEADER_SEQ, seqno,
+			     MCDI_HEADER_XFLAGS, xflags);
+
+	efx_mcdi_writed(efx, &hdr, pdu);
+
+	for (i = 0; i < inlen; i += 4)
+		efx_mcdi_writed(efx, (const efx_dword_t *)(inbuf + i),
+				pdu + 4 + i);
+
+	/* ring the doorbell with a distinctive value */
+	EFX_POPULATE_DWORD_1(hdr, EFX_DWORD_0, 0x45789abc);
+	efx_mcdi_writed(efx, &hdr, doorbell);
+}
+
+static void efx_mcdi_copyout(struct efx_nic *efx, u8 *outbuf, size_t outlen)
+{
+	struct efx_mcdi_iface *mcdi = efx_mcdi(efx);
+	unsigned int pdu = MCDI_PDU(efx);
+	int i;
+
+	BUG_ON(atomic_read(&mcdi->state) == MCDI_STATE_QUIESCENT);
+	BUG_ON(outlen & 3 || outlen >= 0x100);
+
+	for (i = 0; i < outlen; i += 4)
+		efx_mcdi_readd(efx, (efx_dword_t *)(outbuf + i), pdu + 4 + i);
+}
+
+static int efx_mcdi_poll(struct efx_nic *efx)
+{
+	struct efx_mcdi_iface *mcdi = efx_mcdi(efx);
+	unsigned int time, finish;
+	unsigned int respseq, respcmd, error;
+	unsigned int pdu = MCDI_PDU(efx);
+	unsigned int rc, spins;
+	efx_dword_t reg;
+
+	/* Check for a reboot atomically with respect to efx_mcdi_copyout() */
+	rc = -efx_mcdi_poll_reboot(efx);
+	if (rc)
+		goto out;
+
+	/* Poll for completion. Poll quickly (once a us) for the 1st jiffy,
+	 * because generally mcdi responses are fast. After that, back off
+	 * and poll once a jiffy (approximately)
+	 */
+	spins = TICK_USEC;
+	finish = get_seconds() + MCDI_RPC_TIMEOUT;
+
+	while (1) {
+		if (spins != 0) {
+			--spins;
+			udelay(1);
+		} else {
+			schedule_timeout_uninterruptible(1);
+		}
+
+		time = get_seconds();
+
+		efx_mcdi_readd(efx, &reg, pdu);
+
+		/* All 1's indicates that shared memory is in reset (and is
+		 * not a valid header). Wait for it to come out reset before
+		 * completing the command */
+		if (EFX_DWORD_FIELD(reg, EFX_DWORD_0) != 0xffffffff &&
+		    EFX_DWORD_FIELD(reg, MCDI_HEADER_RESPONSE))
+			break;
+
+		if (time >= finish)
+			return -ETIMEDOUT;
+	}
+
+	mcdi->resplen = EFX_DWORD_FIELD(reg, MCDI_HEADER_DATALEN);
+	respseq = EFX_DWORD_FIELD(reg, MCDI_HEADER_SEQ);
+	respcmd = EFX_DWORD_FIELD(reg, MCDI_HEADER_CODE);
+	error = EFX_DWORD_FIELD(reg, MCDI_HEADER_ERROR);
+
+	if (error && mcdi->resplen == 0) {
+		netif_err(efx, hw, efx->net_dev, "MC rebooted\n");
+		rc = EIO;
+	} else if ((respseq ^ mcdi->seqno) & SEQ_MASK) {
+		netif_err(efx, hw, efx->net_dev,
+			  "MC response mismatch tx seq 0x%x rx seq 0x%x\n",
+			  respseq, mcdi->seqno);
+		rc = EIO;
+	} else if (error) {
+		efx_mcdi_readd(efx, &reg, pdu + 4);
+		switch (EFX_DWORD_FIELD(reg, EFX_DWORD_0)) {
+#define TRANSLATE_ERROR(name)					\
+		case MC_CMD_ERR_ ## name:			\
+			rc = name;				\
+			break
+			TRANSLATE_ERROR(ENOENT);
+			TRANSLATE_ERROR(EINTR);
+			TRANSLATE_ERROR(EACCES);
+			TRANSLATE_ERROR(EBUSY);
+			TRANSLATE_ERROR(EINVAL);
+			TRANSLATE_ERROR(EDEADLK);
+			TRANSLATE_ERROR(ENOSYS);
+			TRANSLATE_ERROR(ETIME);
+#undef TRANSLATE_ERROR
+		default:
+			rc = EIO;
+			break;
+		}
+	} else
+		rc = 0;
+
+out:
+	mcdi->resprc = rc;
+	if (rc)
+		mcdi->resplen = 0;
+
+	/* Return rc=0 like wait_event_timeout() */
+	return 0;
+}
+
+/* Test and clear MC-rebooted flag for this port/function */
+int efx_mcdi_poll_reboot(struct efx_nic *efx)
+{
+	unsigned int addr = MCDI_REBOOT_FLAG(efx);
+	efx_dword_t reg;
+	uint32_t value;
+
+	if (efx_nic_rev(efx) < EFX_REV_SIENA_A0)
+		return false;
+
+	efx_mcdi_readd(efx, &reg, addr);
+	value = EFX_DWORD_FIELD(reg, EFX_DWORD_0);
+
+	if (value == 0)
+		return 0;
+
+	EFX_ZERO_DWORD(reg);
+	efx_mcdi_writed(efx, &reg, addr);
+
+	if (value == MC_STATUS_DWORD_ASSERT)
+		return -EINTR;
+	else
+		return -EIO;
+}
+
+static void efx_mcdi_acquire(struct efx_mcdi_iface *mcdi)
+{
+	/* Wait until the interface becomes QUIESCENT and we win the race
+	 * to mark it RUNNING. */
+	wait_event(mcdi->wq,
+		   atomic_cmpxchg(&mcdi->state,
+				  MCDI_STATE_QUIESCENT,
+				  MCDI_STATE_RUNNING)
+		   == MCDI_STATE_QUIESCENT);
+}
+
+static int efx_mcdi_await_completion(struct efx_nic *efx)
+{
+	struct efx_mcdi_iface *mcdi = efx_mcdi(efx);
+
+	if (wait_event_timeout(
+		    mcdi->wq,
+		    atomic_read(&mcdi->state) == MCDI_STATE_COMPLETED,
+		    msecs_to_jiffies(MCDI_RPC_TIMEOUT * 1000)) == 0)
+		return -ETIMEDOUT;
+
+	/* Check if efx_mcdi_set_mode() switched us back to polled completions.
+	 * In which case, poll for completions directly. If efx_mcdi_ev_cpl()
+	 * completed the request first, then we'll just end up completing the
+	 * request again, which is safe.
+	 *
+	 * We need an smp_rmb() to synchronise with efx_mcdi_mode_poll(), which
+	 * wait_event_timeout() implicitly provides.
+	 */
+	if (mcdi->mode == MCDI_MODE_POLL)
+		return efx_mcdi_poll(efx);
+
+	return 0;
+}
+
+static bool efx_mcdi_complete(struct efx_mcdi_iface *mcdi)
+{
+	/* If the interface is RUNNING, then move to COMPLETED and wake any
+	 * waiters. If the interface isn't in RUNNING then we've received a
+	 * duplicate completion after we've already transitioned back to
+	 * QUIESCENT. [A subsequent invocation would increment seqno, so would
+	 * have failed the seqno check].
+	 */
+	if (atomic_cmpxchg(&mcdi->state,
+			   MCDI_STATE_RUNNING,
+			   MCDI_STATE_COMPLETED) == MCDI_STATE_RUNNING) {
+		wake_up(&mcdi->wq);
+		return true;
+	}
+
+	return false;
+}
+
+static void efx_mcdi_release(struct efx_mcdi_iface *mcdi)
+{
+	atomic_set(&mcdi->state, MCDI_STATE_QUIESCENT);
+	wake_up(&mcdi->wq);
+}
+
+static void efx_mcdi_ev_cpl(struct efx_nic *efx, unsigned int seqno,
+			    unsigned int datalen, unsigned int errno)
+{
+	struct efx_mcdi_iface *mcdi = efx_mcdi(efx);
+	bool wake = false;
+
+	spin_lock(&mcdi->iface_lock);
+
+	if ((seqno ^ mcdi->seqno) & SEQ_MASK) {
+		if (mcdi->credits)
+			/* The request has been cancelled */
+			--mcdi->credits;
+		else
+			netif_err(efx, hw, efx->net_dev,
+				  "MC response mismatch tx seq 0x%x rx "
+				  "seq 0x%x\n", seqno, mcdi->seqno);
+	} else {
+		mcdi->resprc = errno;
+		mcdi->resplen = datalen;
+
+		wake = true;
+	}
+
+	spin_unlock(&mcdi->iface_lock);
+
+	if (wake)
+		efx_mcdi_complete(mcdi);
+}
+
+/* Issue the given command by writing the data into the shared memory PDU,
+ * ring the doorbell and wait for completion. Copyout the result. */
+int efx_mcdi_rpc(struct efx_nic *efx, unsigned cmd,
+		 const u8 *inbuf, size_t inlen, u8 *outbuf, size_t outlen,
+		 size_t *outlen_actual)
+{
+	struct efx_mcdi_iface *mcdi = efx_mcdi(efx);
+	int rc;
+	BUG_ON(efx_nic_rev(efx) < EFX_REV_SIENA_A0);
+
+	efx_mcdi_acquire(mcdi);
+
+	/* Serialise with efx_mcdi_ev_cpl() and efx_mcdi_ev_death() */
+	spin_lock_bh(&mcdi->iface_lock);
+	++mcdi->seqno;
+	spin_unlock_bh(&mcdi->iface_lock);
+
+	efx_mcdi_copyin(efx, cmd, inbuf, inlen);
+
+	if (mcdi->mode == MCDI_MODE_POLL)
+		rc = efx_mcdi_poll(efx);
+	else
+		rc = efx_mcdi_await_completion(efx);
+
+	if (rc != 0) {
+		/* Close the race with efx_mcdi_ev_cpl() executing just too late
+		 * and completing a request we've just cancelled, by ensuring
+		 * that the seqno check therein fails.
+		 */
+		spin_lock_bh(&mcdi->iface_lock);
+		++mcdi->seqno;
+		++mcdi->credits;
+		spin_unlock_bh(&mcdi->iface_lock);
+
+		netif_err(efx, hw, efx->net_dev,
+			  "MC command 0x%x inlen %d mode %d timed out\n",
+			  cmd, (int)inlen, mcdi->mode);
+	} else {
+		size_t resplen;
+
+		/* At the very least we need a memory barrier here to ensure
+		 * we pick up changes from efx_mcdi_ev_cpl(). Protect against
+		 * a spurious efx_mcdi_ev_cpl() running concurrently by
+		 * acquiring the iface_lock. */
+		spin_lock_bh(&mcdi->iface_lock);
+		rc = -mcdi->resprc;
+		resplen = mcdi->resplen;
+		spin_unlock_bh(&mcdi->iface_lock);
+
+		if (rc == 0) {
+			efx_mcdi_copyout(efx, outbuf,
+					 min(outlen, mcdi->resplen + 3) & ~0x3);
+			if (outlen_actual != NULL)
+				*outlen_actual = resplen;
+		} else if (cmd == MC_CMD_REBOOT && rc == -EIO)
+			; /* Don't reset if MC_CMD_REBOOT returns EIO */
+		else if (rc == -EIO || rc == -EINTR) {
+			netif_err(efx, hw, efx->net_dev, "MC fatal error %d\n",
+				  -rc);
+			efx_schedule_reset(efx, RESET_TYPE_MC_FAILURE);
+		} else
+			netif_dbg(efx, hw, efx->net_dev,
+				  "MC command 0x%x inlen %d failed rc=%d\n",
+				  cmd, (int)inlen, -rc);
+	}
+
+	efx_mcdi_release(mcdi);
+	return rc;
+}
+
+void efx_mcdi_mode_poll(struct efx_nic *efx)
+{
+	struct efx_mcdi_iface *mcdi;
+
+	if (efx_nic_rev(efx) < EFX_REV_SIENA_A0)
+		return;
+
+	mcdi = efx_mcdi(efx);
+	if (mcdi->mode == MCDI_MODE_POLL)
+		return;
+
+	/* We can switch from event completion to polled completion, because
+	 * mcdi requests are always completed in shared memory. We do this by
+	 * switching the mode to POLL'd then completing the request.
+	 * efx_mcdi_await_completion() will then call efx_mcdi_poll().
+	 *
+	 * We need an smp_wmb() to synchronise with efx_mcdi_await_completion(),
+	 * which efx_mcdi_complete() provides for us.
+	 */
+	mcdi->mode = MCDI_MODE_POLL;
+
+	efx_mcdi_complete(mcdi);
+}
+
+void efx_mcdi_mode_event(struct efx_nic *efx)
+{
+	struct efx_mcdi_iface *mcdi;
+
+	if (efx_nic_rev(efx) < EFX_REV_SIENA_A0)
+		return;
+
+	mcdi = efx_mcdi(efx);
+
+	if (mcdi->mode == MCDI_MODE_EVENTS)
+		return;
+
+	/* We can't switch from polled to event completion in the middle of a
+	 * request, because the completion method is specified in the request.
+	 * So acquire the interface to serialise the requestors. We don't need
+	 * to acquire the iface_lock to change the mode here, but we do need a
+	 * write memory barrier ensure that efx_mcdi_rpc() sees it, which
+	 * efx_mcdi_acquire() provides.
+	 */
+	efx_mcdi_acquire(mcdi);
+	mcdi->mode = MCDI_MODE_EVENTS;
+	efx_mcdi_release(mcdi);
+}
+
+static void efx_mcdi_ev_death(struct efx_nic *efx, int rc)
+{
+	struct efx_mcdi_iface *mcdi = efx_mcdi(efx);
+
+	/* If there is an outstanding MCDI request, it has been terminated
+	 * either by a BADASSERT or REBOOT event. If the mcdi interface is
+	 * in polled mode, then do nothing because the MC reboot handler will
+	 * set the header correctly. However, if the mcdi interface is waiting
+	 * for a CMDDONE event it won't receive it [and since all MCDI events
+	 * are sent to the same queue, we can't be racing with
+	 * efx_mcdi_ev_cpl()]
+	 *
+	 * There's a race here with efx_mcdi_rpc(), because we might receive
+	 * a REBOOT event *before* the request has been copied out. In polled
+	 * mode (during startup) this is irrelevant, because efx_mcdi_complete()
+	 * is ignored. In event mode, this condition is just an edge-case of
+	 * receiving a REBOOT event after posting the MCDI request. Did the mc
+	 * reboot before or after the copyout? The best we can do always is
+	 * just return failure.
+	 */
+	spin_lock(&mcdi->iface_lock);
+	if (efx_mcdi_complete(mcdi)) {
+		if (mcdi->mode == MCDI_MODE_EVENTS) {
+			mcdi->resprc = rc;
+			mcdi->resplen = 0;
+			++mcdi->credits;
+		}
+	} else
+		/* Nobody was waiting for an MCDI request, so trigger a reset */
+		efx_schedule_reset(efx, RESET_TYPE_MC_FAILURE);
+
+	spin_unlock(&mcdi->iface_lock);
+}
+
+static unsigned int efx_mcdi_event_link_speed[] = {
+	[MCDI_EVENT_LINKCHANGE_SPEED_100M] = 100,
+	[MCDI_EVENT_LINKCHANGE_SPEED_1G] = 1000,
+	[MCDI_EVENT_LINKCHANGE_SPEED_10G] = 10000,
+};
+
+
+static void efx_mcdi_process_link_change(struct efx_nic *efx, efx_qword_t *ev)
+{
+	u32 flags, fcntl, speed, lpa;
+
+	speed = EFX_QWORD_FIELD(*ev, MCDI_EVENT_LINKCHANGE_SPEED);
+	EFX_BUG_ON_PARANOID(speed >= ARRAY_SIZE(efx_mcdi_event_link_speed));
+	speed = efx_mcdi_event_link_speed[speed];
+
+	flags = EFX_QWORD_FIELD(*ev, MCDI_EVENT_LINKCHANGE_LINK_FLAGS);
+	fcntl = EFX_QWORD_FIELD(*ev, MCDI_EVENT_LINKCHANGE_FCNTL);
+	lpa = EFX_QWORD_FIELD(*ev, MCDI_EVENT_LINKCHANGE_LP_CAP);
+
+	/* efx->link_state is only modified by efx_mcdi_phy_get_link(),
+	 * which is only run after flushing the event queues. Therefore, it
+	 * is safe to modify the link state outside of the mac_lock here.
+	 */
+	efx_mcdi_phy_decode_link(efx, &efx->link_state, speed, flags, fcntl);
+
+	efx_mcdi_phy_check_fcntl(efx, lpa);
+
+	efx_link_status_changed(efx);
+}
+
+static const char *sensor_names[] = {
+	[MC_CMD_SENSOR_CONTROLLER_TEMP] = "Controller temp. sensor",
+	[MC_CMD_SENSOR_PHY_COMMON_TEMP] = "PHY shared temp. sensor",
+	[MC_CMD_SENSOR_CONTROLLER_COOLING] = "Controller cooling",
+	[MC_CMD_SENSOR_PHY0_TEMP] = "PHY 0 temp. sensor",
+	[MC_CMD_SENSOR_PHY0_COOLING] = "PHY 0 cooling",
+	[MC_CMD_SENSOR_PHY1_TEMP] = "PHY 1 temp. sensor",
+	[MC_CMD_SENSOR_PHY1_COOLING] = "PHY 1 cooling",
+	[MC_CMD_SENSOR_IN_1V0] = "1.0V supply sensor",
+	[MC_CMD_SENSOR_IN_1V2] = "1.2V supply sensor",
+	[MC_CMD_SENSOR_IN_1V8] = "1.8V supply sensor",
+	[MC_CMD_SENSOR_IN_2V5] = "2.5V supply sensor",
+	[MC_CMD_SENSOR_IN_3V3] = "3.3V supply sensor",
+	[MC_CMD_SENSOR_IN_12V0] = "12V supply sensor"
+};
+
+static const char *sensor_status_names[] = {
+	[MC_CMD_SENSOR_STATE_OK] = "OK",
+	[MC_CMD_SENSOR_STATE_WARNING] = "Warning",
+	[MC_CMD_SENSOR_STATE_FATAL] = "Fatal",
+	[MC_CMD_SENSOR_STATE_BROKEN] = "Device failure",
+};
+
+static void efx_mcdi_sensor_event(struct efx_nic *efx, efx_qword_t *ev)
+{
+	unsigned int monitor, state, value;
+	const char *name, *state_txt;
+	monitor = EFX_QWORD_FIELD(*ev, MCDI_EVENT_SENSOREVT_MONITOR);
+	state = EFX_QWORD_FIELD(*ev, MCDI_EVENT_SENSOREVT_STATE);
+	value = EFX_QWORD_FIELD(*ev, MCDI_EVENT_SENSOREVT_VALUE);
+	/* Deal gracefully with the board having more drivers than we
+	 * know about, but do not expect new sensor states. */
+	name = (monitor >= ARRAY_SIZE(sensor_names))
+				    ? "No sensor name available" :
+				    sensor_names[monitor];
+	EFX_BUG_ON_PARANOID(state >= ARRAY_SIZE(sensor_status_names));
+	state_txt = sensor_status_names[state];
+
+	netif_err(efx, hw, efx->net_dev,
+		  "Sensor %d (%s) reports condition '%s' for raw value %d\n",
+		  monitor, name, state_txt, value);
+}
+
+/* Called from  falcon_process_eventq for MCDI events */
+void efx_mcdi_process_event(struct efx_channel *channel,
+			    efx_qword_t *event)
+{
+	struct efx_nic *efx = channel->efx;
+	int code = EFX_QWORD_FIELD(*event, MCDI_EVENT_CODE);
+	u32 data = EFX_QWORD_FIELD(*event, MCDI_EVENT_DATA);
+
+	switch (code) {
+	case MCDI_EVENT_CODE_BADSSERT:
+		netif_err(efx, hw, efx->net_dev,
+			  "MC watchdog or assertion failure at 0x%x\n", data);
+		efx_mcdi_ev_death(efx, EINTR);
+		break;
+
+	case MCDI_EVENT_CODE_PMNOTICE:
+		netif_info(efx, wol, efx->net_dev, "MCDI PM event.\n");
+		break;
+
+	case MCDI_EVENT_CODE_CMDDONE:
+		efx_mcdi_ev_cpl(efx,
+				MCDI_EVENT_FIELD(*event, CMDDONE_SEQ),
+				MCDI_EVENT_FIELD(*event, CMDDONE_DATALEN),
+				MCDI_EVENT_FIELD(*event, CMDDONE_ERRNO));
+		break;
+
+	case MCDI_EVENT_CODE_LINKCHANGE:
+		efx_mcdi_process_link_change(efx, event);
+		break;
+	case MCDI_EVENT_CODE_SENSOREVT:
+		efx_mcdi_sensor_event(efx, event);
+		break;
+	case MCDI_EVENT_CODE_SCHEDERR:
+		netif_info(efx, hw, efx->net_dev,
+			   "MC Scheduler error address=0x%x\n", data);
+		break;
+	case MCDI_EVENT_CODE_REBOOT:
+		netif_info(efx, hw, efx->net_dev, "MC Reboot\n");
+		efx_mcdi_ev_death(efx, EIO);
+		break;
+	case MCDI_EVENT_CODE_MAC_STATS_DMA:
+		/* MAC stats are gather lazily.  We can ignore this. */
+		break;
+
+	default:
+		netif_err(efx, hw, efx->net_dev, "Unknown MCDI event 0x%x\n",
+			  code);
+	}
+}
+
+/**************************************************************************
+ *
+ * Specific request functions
+ *
+ **************************************************************************
+ */
+
+void efx_mcdi_print_fwver(struct efx_nic *efx, char *buf, size_t len)
+{
+	u8 outbuf[ALIGN(MC_CMD_GET_VERSION_V1_OUT_LEN, 4)];
+	size_t outlength;
+	const __le16 *ver_words;
+	int rc;
+
+	BUILD_BUG_ON(MC_CMD_GET_VERSION_IN_LEN != 0);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_GET_VERSION, NULL, 0,
+			  outbuf, sizeof(outbuf), &outlength);
+	if (rc)
+		goto fail;
+
+	if (outlength < MC_CMD_GET_VERSION_V1_OUT_LEN) {
+		rc = -EIO;
+		goto fail;
+	}
+
+	ver_words = (__le16 *)MCDI_PTR(outbuf, GET_VERSION_OUT_VERSION);
+	snprintf(buf, len, "%u.%u.%u.%u",
+		 le16_to_cpu(ver_words[0]), le16_to_cpu(ver_words[1]),
+		 le16_to_cpu(ver_words[2]), le16_to_cpu(ver_words[3]));
+	return;
+
+fail:
+	netif_err(efx, probe, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	buf[0] = 0;
+}
+
+int efx_mcdi_drv_attach(struct efx_nic *efx, bool driver_operating,
+			bool *was_attached)
+{
+	u8 inbuf[MC_CMD_DRV_ATTACH_IN_LEN];
+	u8 outbuf[MC_CMD_DRV_ATTACH_OUT_LEN];
+	size_t outlen;
+	int rc;
+
+	MCDI_SET_DWORD(inbuf, DRV_ATTACH_IN_NEW_STATE,
+		       driver_operating ? 1 : 0);
+	MCDI_SET_DWORD(inbuf, DRV_ATTACH_IN_UPDATE, 1);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_DRV_ATTACH, inbuf, sizeof(inbuf),
+			  outbuf, sizeof(outbuf), &outlen);
+	if (rc)
+		goto fail;
+	if (outlen < MC_CMD_DRV_ATTACH_OUT_LEN) {
+		rc = -EIO;
+		goto fail;
+	}
+
+	if (was_attached != NULL)
+		*was_attached = MCDI_DWORD(outbuf, DRV_ATTACH_OUT_OLD_STATE);
+	return 0;
+
+fail:
+	netif_err(efx, probe, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	return rc;
+}
+
+int efx_mcdi_get_board_cfg(struct efx_nic *efx, u8 *mac_address,
+			   u16 *fw_subtype_list)
+{
+	uint8_t outbuf[MC_CMD_GET_BOARD_CFG_OUT_LEN];
+	size_t outlen;
+	int port_num = efx_port_num(efx);
+	int offset;
+	int rc;
+
+	BUILD_BUG_ON(MC_CMD_GET_BOARD_CFG_IN_LEN != 0);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_GET_BOARD_CFG, NULL, 0,
+			  outbuf, sizeof(outbuf), &outlen);
+	if (rc)
+		goto fail;
+
+	if (outlen < MC_CMD_GET_BOARD_CFG_OUT_LEN) {
+		rc = -EIO;
+		goto fail;
+	}
+
+	offset = (port_num)
+		? MC_CMD_GET_BOARD_CFG_OUT_MAC_ADDR_BASE_PORT1_OFST
+		: MC_CMD_GET_BOARD_CFG_OUT_MAC_ADDR_BASE_PORT0_OFST;
+	if (mac_address)
+		memcpy(mac_address, outbuf + offset, ETH_ALEN);
+	if (fw_subtype_list)
+		memcpy(fw_subtype_list,
+		       outbuf + MC_CMD_GET_BOARD_CFG_OUT_FW_SUBTYPE_LIST_OFST,
+		       MC_CMD_GET_BOARD_CFG_OUT_FW_SUBTYPE_LIST_LEN);
+
+	return 0;
+
+fail:
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d len=%d\n",
+		  __func__, rc, (int)outlen);
+
+	return rc;
+}
+
+int efx_mcdi_log_ctrl(struct efx_nic *efx, bool evq, bool uart, u32 dest_evq)
+{
+	u8 inbuf[MC_CMD_LOG_CTRL_IN_LEN];
+	u32 dest = 0;
+	int rc;
+
+	if (uart)
+		dest |= MC_CMD_LOG_CTRL_IN_LOG_DEST_UART;
+	if (evq)
+		dest |= MC_CMD_LOG_CTRL_IN_LOG_DEST_EVQ;
+
+	MCDI_SET_DWORD(inbuf, LOG_CTRL_IN_LOG_DEST, dest);
+	MCDI_SET_DWORD(inbuf, LOG_CTRL_IN_LOG_DEST_EVQ, dest_evq);
+
+	BUILD_BUG_ON(MC_CMD_LOG_CTRL_OUT_LEN != 0);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_LOG_CTRL, inbuf, sizeof(inbuf),
+			  NULL, 0, NULL);
+	if (rc)
+		goto fail;
+
+	return 0;
+
+fail:
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	return rc;
+}
+
+int efx_mcdi_nvram_types(struct efx_nic *efx, u32 *nvram_types_out)
+{
+	u8 outbuf[MC_CMD_NVRAM_TYPES_OUT_LEN];
+	size_t outlen;
+	int rc;
+
+	BUILD_BUG_ON(MC_CMD_NVRAM_TYPES_IN_LEN != 0);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_NVRAM_TYPES, NULL, 0,
+			  outbuf, sizeof(outbuf), &outlen);
+	if (rc)
+		goto fail;
+	if (outlen < MC_CMD_NVRAM_TYPES_OUT_LEN) {
+		rc = -EIO;
+		goto fail;
+	}
+
+	*nvram_types_out = MCDI_DWORD(outbuf, NVRAM_TYPES_OUT_TYPES);
+	return 0;
+
+fail:
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n",
+		  __func__, rc);
+	return rc;
+}
+
+int efx_mcdi_nvram_info(struct efx_nic *efx, unsigned int type,
+			size_t *size_out, size_t *erase_size_out,
+			bool *protected_out)
+{
+	u8 inbuf[MC_CMD_NVRAM_INFO_IN_LEN];
+	u8 outbuf[MC_CMD_NVRAM_INFO_OUT_LEN];
+	size_t outlen;
+	int rc;
+
+	MCDI_SET_DWORD(inbuf, NVRAM_INFO_IN_TYPE, type);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_NVRAM_INFO, inbuf, sizeof(inbuf),
+			  outbuf, sizeof(outbuf), &outlen);
+	if (rc)
+		goto fail;
+	if (outlen < MC_CMD_NVRAM_INFO_OUT_LEN) {
+		rc = -EIO;
+		goto fail;
+	}
+
+	*size_out = MCDI_DWORD(outbuf, NVRAM_INFO_OUT_SIZE);
+	*erase_size_out = MCDI_DWORD(outbuf, NVRAM_INFO_OUT_ERASESIZE);
+	*protected_out = !!(MCDI_DWORD(outbuf, NVRAM_INFO_OUT_FLAGS) &
+				(1 << MC_CMD_NVRAM_PROTECTED_LBN));
+	return 0;
+
+fail:
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	return rc;
+}
+
+int efx_mcdi_nvram_update_start(struct efx_nic *efx, unsigned int type)
+{
+	u8 inbuf[MC_CMD_NVRAM_UPDATE_START_IN_LEN];
+	int rc;
+
+	MCDI_SET_DWORD(inbuf, NVRAM_UPDATE_START_IN_TYPE, type);
+
+	BUILD_BUG_ON(MC_CMD_NVRAM_UPDATE_START_OUT_LEN != 0);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_NVRAM_UPDATE_START, inbuf, sizeof(inbuf),
+			  NULL, 0, NULL);
+	if (rc)
+		goto fail;
+
+	return 0;
+
+fail:
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	return rc;
+}
+
+int efx_mcdi_nvram_read(struct efx_nic *efx, unsigned int type,
+			loff_t offset, u8 *buffer, size_t length)
+{
+	u8 inbuf[MC_CMD_NVRAM_READ_IN_LEN];
+	u8 outbuf[MC_CMD_NVRAM_READ_OUT_LEN(EFX_MCDI_NVRAM_LEN_MAX)];
+	size_t outlen;
+	int rc;
+
+	MCDI_SET_DWORD(inbuf, NVRAM_READ_IN_TYPE, type);
+	MCDI_SET_DWORD(inbuf, NVRAM_READ_IN_OFFSET, offset);
+	MCDI_SET_DWORD(inbuf, NVRAM_READ_IN_LENGTH, length);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_NVRAM_READ, inbuf, sizeof(inbuf),
+			  outbuf, sizeof(outbuf), &outlen);
+	if (rc)
+		goto fail;
+
+	memcpy(buffer, MCDI_PTR(outbuf, NVRAM_READ_OUT_READ_BUFFER), length);
+	return 0;
+
+fail:
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	return rc;
+}
+
+int efx_mcdi_nvram_write(struct efx_nic *efx, unsigned int type,
+			   loff_t offset, const u8 *buffer, size_t length)
+{
+	u8 inbuf[MC_CMD_NVRAM_WRITE_IN_LEN(EFX_MCDI_NVRAM_LEN_MAX)];
+	int rc;
+
+	MCDI_SET_DWORD(inbuf, NVRAM_WRITE_IN_TYPE, type);
+	MCDI_SET_DWORD(inbuf, NVRAM_WRITE_IN_OFFSET, offset);
+	MCDI_SET_DWORD(inbuf, NVRAM_WRITE_IN_LENGTH, length);
+	memcpy(MCDI_PTR(inbuf, NVRAM_WRITE_IN_WRITE_BUFFER), buffer, length);
+
+	BUILD_BUG_ON(MC_CMD_NVRAM_WRITE_OUT_LEN != 0);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_NVRAM_WRITE, inbuf,
+			  ALIGN(MC_CMD_NVRAM_WRITE_IN_LEN(length), 4),
+			  NULL, 0, NULL);
+	if (rc)
+		goto fail;
+
+	return 0;
+
+fail:
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	return rc;
+}
+
+int efx_mcdi_nvram_erase(struct efx_nic *efx, unsigned int type,
+			 loff_t offset, size_t length)
+{
+	u8 inbuf[MC_CMD_NVRAM_ERASE_IN_LEN];
+	int rc;
+
+	MCDI_SET_DWORD(inbuf, NVRAM_ERASE_IN_TYPE, type);
+	MCDI_SET_DWORD(inbuf, NVRAM_ERASE_IN_OFFSET, offset);
+	MCDI_SET_DWORD(inbuf, NVRAM_ERASE_IN_LENGTH, length);
+
+	BUILD_BUG_ON(MC_CMD_NVRAM_ERASE_OUT_LEN != 0);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_NVRAM_ERASE, inbuf, sizeof(inbuf),
+			  NULL, 0, NULL);
+	if (rc)
+		goto fail;
+
+	return 0;
+
+fail:
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	return rc;
+}
+
+int efx_mcdi_nvram_update_finish(struct efx_nic *efx, unsigned int type)
+{
+	u8 inbuf[MC_CMD_NVRAM_UPDATE_FINISH_IN_LEN];
+	int rc;
+
+	MCDI_SET_DWORD(inbuf, NVRAM_UPDATE_FINISH_IN_TYPE, type);
+
+	BUILD_BUG_ON(MC_CMD_NVRAM_UPDATE_FINISH_OUT_LEN != 0);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_NVRAM_UPDATE_FINISH, inbuf, sizeof(inbuf),
+			  NULL, 0, NULL);
+	if (rc)
+		goto fail;
+
+	return 0;
+
+fail:
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	return rc;
+}
+
+static int efx_mcdi_nvram_test(struct efx_nic *efx, unsigned int type)
+{
+	u8 inbuf[MC_CMD_NVRAM_TEST_IN_LEN];
+	u8 outbuf[MC_CMD_NVRAM_TEST_OUT_LEN];
+	int rc;
+
+	MCDI_SET_DWORD(inbuf, NVRAM_TEST_IN_TYPE, type);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_NVRAM_TEST, inbuf, sizeof(inbuf),
+			  outbuf, sizeof(outbuf), NULL);
+	if (rc)
+		return rc;
+
+	switch (MCDI_DWORD(outbuf, NVRAM_TEST_OUT_RESULT)) {
+	case MC_CMD_NVRAM_TEST_PASS:
+	case MC_CMD_NVRAM_TEST_NOTSUPP:
+		return 0;
+	default:
+		return -EIO;
+	}
+}
+
+int efx_mcdi_nvram_test_all(struct efx_nic *efx)
+{
+	u32 nvram_types;
+	unsigned int type;
+	int rc;
+
+	rc = efx_mcdi_nvram_types(efx, &nvram_types);
+	if (rc)
+		goto fail1;
+
+	type = 0;
+	while (nvram_types != 0) {
+		if (nvram_types & 1) {
+			rc = efx_mcdi_nvram_test(efx, type);
+			if (rc)
+				goto fail2;
+		}
+		type++;
+		nvram_types >>= 1;
+	}
+
+	return 0;
+
+fail2:
+	netif_err(efx, hw, efx->net_dev, "%s: failed type=%u\n",
+		  __func__, type);
+fail1:
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	return rc;
+}
+
+static int efx_mcdi_read_assertion(struct efx_nic *efx)
+{
+	u8 inbuf[MC_CMD_GET_ASSERTS_IN_LEN];
+	u8 outbuf[MC_CMD_GET_ASSERTS_OUT_LEN];
+	unsigned int flags, index, ofst;
+	const char *reason;
+	size_t outlen;
+	int retry;
+	int rc;
+
+	/* Attempt to read any stored assertion state before we reboot
+	 * the mcfw out of the assertion handler. Retry twice, once
+	 * because a boot-time assertion might cause this command to fail
+	 * with EINTR. And once again because GET_ASSERTS can race with
+	 * MC_CMD_REBOOT running on the other port. */
+	retry = 2;
+	do {
+		MCDI_SET_DWORD(inbuf, GET_ASSERTS_IN_CLEAR, 1);
+		rc = efx_mcdi_rpc(efx, MC_CMD_GET_ASSERTS,
+				  inbuf, MC_CMD_GET_ASSERTS_IN_LEN,
+				  outbuf, sizeof(outbuf), &outlen);
+	} while ((rc == -EINTR || rc == -EIO) && retry-- > 0);
+
+	if (rc)
+		return rc;
+	if (outlen < MC_CMD_GET_ASSERTS_OUT_LEN)
+		return -EIO;
+
+	/* Print out any recorded assertion state */
+	flags = MCDI_DWORD(outbuf, GET_ASSERTS_OUT_GLOBAL_FLAGS);
+	if (flags == MC_CMD_GET_ASSERTS_FLAGS_NO_FAILS)
+		return 0;
+
+	reason = (flags == MC_CMD_GET_ASSERTS_FLAGS_SYS_FAIL)
+		? "system-level assertion"
+		: (flags == MC_CMD_GET_ASSERTS_FLAGS_THR_FAIL)
+		? "thread-level assertion"
+		: (flags == MC_CMD_GET_ASSERTS_FLAGS_WDOG_FIRED)
+		? "watchdog reset"
+		: "unknown assertion";
+	netif_err(efx, hw, efx->net_dev,
+		  "MCPU %s at PC = 0x%.8x in thread 0x%.8x\n", reason,
+		  MCDI_DWORD(outbuf, GET_ASSERTS_OUT_SAVED_PC_OFFS),
+		  MCDI_DWORD(outbuf, GET_ASSERTS_OUT_THREAD_OFFS));
+
+	/* Print out the registers */
+	ofst = MC_CMD_GET_ASSERTS_OUT_GP_REGS_OFFS_OFST;
+	for (index = 1; index < 32; index++) {
+		netif_err(efx, hw, efx->net_dev, "R%.2d (?): 0x%.8x\n", index,
+			MCDI_DWORD2(outbuf, ofst));
+		ofst += sizeof(efx_dword_t);
+	}
+
+	return 0;
+}
+
+static void efx_mcdi_exit_assertion(struct efx_nic *efx)
+{
+	u8 inbuf[MC_CMD_REBOOT_IN_LEN];
+
+	/* Atomically reboot the mcfw out of the assertion handler */
+	BUILD_BUG_ON(MC_CMD_REBOOT_OUT_LEN != 0);
+	MCDI_SET_DWORD(inbuf, REBOOT_IN_FLAGS,
+		       MC_CMD_REBOOT_FLAGS_AFTER_ASSERTION);
+	efx_mcdi_rpc(efx, MC_CMD_REBOOT, inbuf, MC_CMD_REBOOT_IN_LEN,
+		     NULL, 0, NULL);
+}
+
+int efx_mcdi_handle_assertion(struct efx_nic *efx)
+{
+	int rc;
+
+	rc = efx_mcdi_read_assertion(efx);
+	if (rc)
+		return rc;
+
+	efx_mcdi_exit_assertion(efx);
+
+	return 0;
+}
+
+void efx_mcdi_set_id_led(struct efx_nic *efx, enum efx_led_mode mode)
+{
+	u8 inbuf[MC_CMD_SET_ID_LED_IN_LEN];
+	int rc;
+
+	BUILD_BUG_ON(EFX_LED_OFF != MC_CMD_LED_OFF);
+	BUILD_BUG_ON(EFX_LED_ON != MC_CMD_LED_ON);
+	BUILD_BUG_ON(EFX_LED_DEFAULT != MC_CMD_LED_DEFAULT);
+
+	BUILD_BUG_ON(MC_CMD_SET_ID_LED_OUT_LEN != 0);
+
+	MCDI_SET_DWORD(inbuf, SET_ID_LED_IN_STATE, mode);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_SET_ID_LED, inbuf, sizeof(inbuf),
+			  NULL, 0, NULL);
+	if (rc)
+		netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n",
+			  __func__, rc);
+}
+
+int efx_mcdi_reset_port(struct efx_nic *efx)
+{
+	int rc = efx_mcdi_rpc(efx, MC_CMD_PORT_RESET, NULL, 0, NULL, 0, NULL);
+	if (rc)
+		netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n",
+			  __func__, rc);
+	return rc;
+}
+
+int efx_mcdi_reset_mc(struct efx_nic *efx)
+{
+	u8 inbuf[MC_CMD_REBOOT_IN_LEN];
+	int rc;
+
+	BUILD_BUG_ON(MC_CMD_REBOOT_OUT_LEN != 0);
+	MCDI_SET_DWORD(inbuf, REBOOT_IN_FLAGS, 0);
+	rc = efx_mcdi_rpc(efx, MC_CMD_REBOOT, inbuf, sizeof(inbuf),
+			  NULL, 0, NULL);
+	/* White is black, and up is down */
+	if (rc == -EIO)
+		return 0;
+	if (rc == 0)
+		rc = -EIO;
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	return rc;
+}
+
+static int efx_mcdi_wol_filter_set(struct efx_nic *efx, u32 type,
+				   const u8 *mac, int *id_out)
+{
+	u8 inbuf[MC_CMD_WOL_FILTER_SET_IN_LEN];
+	u8 outbuf[MC_CMD_WOL_FILTER_SET_OUT_LEN];
+	size_t outlen;
+	int rc;
+
+	MCDI_SET_DWORD(inbuf, WOL_FILTER_SET_IN_WOL_TYPE, type);
+	MCDI_SET_DWORD(inbuf, WOL_FILTER_SET_IN_FILTER_MODE,
+		       MC_CMD_FILTER_MODE_SIMPLE);
+	memcpy(MCDI_PTR(inbuf, WOL_FILTER_SET_IN_MAGIC_MAC), mac, ETH_ALEN);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_WOL_FILTER_SET, inbuf, sizeof(inbuf),
+			  outbuf, sizeof(outbuf), &outlen);
+	if (rc)
+		goto fail;
+
+	if (outlen < MC_CMD_WOL_FILTER_SET_OUT_LEN) {
+		rc = -EIO;
+		goto fail;
+	}
+
+	*id_out = (int)MCDI_DWORD(outbuf, WOL_FILTER_SET_OUT_FILTER_ID);
+
+	return 0;
+
+fail:
+	*id_out = -1;
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	return rc;
+
+}
+
+
+int
+efx_mcdi_wol_filter_set_magic(struct efx_nic *efx,  const u8 *mac, int *id_out)
+{
+	return efx_mcdi_wol_filter_set(efx, MC_CMD_WOL_TYPE_MAGIC, mac, id_out);
+}
+
+
+int efx_mcdi_wol_filter_get_magic(struct efx_nic *efx, int *id_out)
+{
+	u8 outbuf[MC_CMD_WOL_FILTER_GET_OUT_LEN];
+	size_t outlen;
+	int rc;
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_WOL_FILTER_GET, NULL, 0,
+			  outbuf, sizeof(outbuf), &outlen);
+	if (rc)
+		goto fail;
+
+	if (outlen < MC_CMD_WOL_FILTER_GET_OUT_LEN) {
+		rc = -EIO;
+		goto fail;
+	}
+
+	*id_out = (int)MCDI_DWORD(outbuf, WOL_FILTER_GET_OUT_FILTER_ID);
+
+	return 0;
+
+fail:
+	*id_out = -1;
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	return rc;
+}
+
+
+int efx_mcdi_wol_filter_remove(struct efx_nic *efx, int id)
+{
+	u8 inbuf[MC_CMD_WOL_FILTER_REMOVE_IN_LEN];
+	int rc;
+
+	MCDI_SET_DWORD(inbuf, WOL_FILTER_REMOVE_IN_FILTER_ID, (u32)id);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_WOL_FILTER_REMOVE, inbuf, sizeof(inbuf),
+			  NULL, 0, NULL);
+	if (rc)
+		goto fail;
+
+	return 0;
+
+fail:
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	return rc;
+}
+
+
+int efx_mcdi_wol_filter_reset(struct efx_nic *efx)
+{
+	int rc;
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_WOL_FILTER_RESET, NULL, 0, NULL, 0, NULL);
+	if (rc)
+		goto fail;
+
+	return 0;
+
+fail:
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	return rc;
+}
+
diff --git a/drivers/net/ethernet/sfc/mcdi.h b/drivers/net/ethernet/sfc/mcdi.h
new file mode 100644
index 0000000..aced2a7
--- /dev/null
+++ b/drivers/net/ethernet/sfc/mcdi.h
@@ -0,0 +1,130 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2008-2010 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#ifndef EFX_MCDI_H
+#define EFX_MCDI_H
+
+/**
+ * enum efx_mcdi_state
+ * @MCDI_STATE_QUIESCENT: No pending MCDI requests. If the caller holds the
+ *	mcdi_lock then they are able to move to MCDI_STATE_RUNNING
+ * @MCDI_STATE_RUNNING: There is an MCDI request pending. Only the thread that
+ *	moved into this state is allowed to move out of it.
+ * @MCDI_STATE_COMPLETED: An MCDI request has completed, but the owning thread
+ *	has not yet consumed the result. For all other threads, equivalent to
+ *	MCDI_STATE_RUNNING.
+ */
+enum efx_mcdi_state {
+	MCDI_STATE_QUIESCENT,
+	MCDI_STATE_RUNNING,
+	MCDI_STATE_COMPLETED,
+};
+
+enum efx_mcdi_mode {
+	MCDI_MODE_POLL,
+	MCDI_MODE_EVENTS,
+};
+
+/**
+ * struct efx_mcdi_iface
+ * @state: Interface state. Waited for by mcdi_wq.
+ * @wq: Wait queue for threads waiting for state != STATE_RUNNING
+ * @iface_lock: Protects @credits, @seqno, @resprc, @resplen
+ * @mode: Poll for mcdi completion, or wait for an mcdi_event.
+ *	Serialised by @lock
+ * @seqno: The next sequence number to use for mcdi requests.
+ *	Serialised by @lock
+ * @credits: Number of spurious MCDI completion events allowed before we
+ *	trigger a fatal error. Protected by @lock
+ * @resprc: Returned MCDI completion
+ * @resplen: Returned payload length
+ */
+struct efx_mcdi_iface {
+	atomic_t state;
+	wait_queue_head_t wq;
+	spinlock_t iface_lock;
+	enum efx_mcdi_mode mode;
+	unsigned int credits;
+	unsigned int seqno;
+	unsigned int resprc;
+	size_t resplen;
+};
+
+extern void efx_mcdi_init(struct efx_nic *efx);
+
+extern int efx_mcdi_rpc(struct efx_nic *efx, unsigned cmd, const u8 *inbuf,
+			size_t inlen, u8 *outbuf, size_t outlen,
+			size_t *outlen_actual);
+
+extern int efx_mcdi_poll_reboot(struct efx_nic *efx);
+extern void efx_mcdi_mode_poll(struct efx_nic *efx);
+extern void efx_mcdi_mode_event(struct efx_nic *efx);
+
+extern void efx_mcdi_process_event(struct efx_channel *channel,
+				   efx_qword_t *event);
+
+#define MCDI_PTR2(_buf, _ofst)						\
+	(((u8 *)_buf) + _ofst)
+#define MCDI_SET_DWORD2(_buf, _ofst, _value)				\
+	EFX_POPULATE_DWORD_1(*((efx_dword_t *)MCDI_PTR2(_buf, _ofst)),	\
+			     EFX_DWORD_0, _value)
+#define MCDI_DWORD2(_buf, _ofst)					\
+	EFX_DWORD_FIELD(*((efx_dword_t *)MCDI_PTR2(_buf, _ofst)),	\
+			EFX_DWORD_0)
+#define MCDI_QWORD2(_buf, _ofst)					\
+	EFX_QWORD_FIELD64(*((efx_qword_t *)MCDI_PTR2(_buf, _ofst)),	\
+			  EFX_QWORD_0)
+
+#define MCDI_PTR(_buf, _ofst)						\
+	MCDI_PTR2(_buf, MC_CMD_ ## _ofst ## _OFST)
+#define MCDI_SET_DWORD(_buf, _ofst, _value)				\
+	MCDI_SET_DWORD2(_buf, MC_CMD_ ## _ofst ## _OFST, _value)
+#define MCDI_DWORD(_buf, _ofst)						\
+	MCDI_DWORD2(_buf, MC_CMD_ ## _ofst ## _OFST)
+#define MCDI_QWORD(_buf, _ofst)						\
+	MCDI_QWORD2(_buf, MC_CMD_ ## _ofst ## _OFST)
+
+#define MCDI_EVENT_FIELD(_ev, _field)			\
+	EFX_QWORD_FIELD(_ev, MCDI_EVENT_ ## _field)
+
+extern void efx_mcdi_print_fwver(struct efx_nic *efx, char *buf, size_t len);
+extern int efx_mcdi_drv_attach(struct efx_nic *efx, bool driver_operating,
+			       bool *was_attached_out);
+extern int efx_mcdi_get_board_cfg(struct efx_nic *efx, u8 *mac_address,
+				  u16 *fw_subtype_list);
+extern int efx_mcdi_log_ctrl(struct efx_nic *efx, bool evq, bool uart,
+			     u32 dest_evq);
+extern int efx_mcdi_nvram_types(struct efx_nic *efx, u32 *nvram_types_out);
+extern int efx_mcdi_nvram_info(struct efx_nic *efx, unsigned int type,
+			       size_t *size_out, size_t *erase_size_out,
+			       bool *protected_out);
+extern int efx_mcdi_nvram_update_start(struct efx_nic *efx,
+				       unsigned int type);
+extern int efx_mcdi_nvram_read(struct efx_nic *efx, unsigned int type,
+			       loff_t offset, u8 *buffer, size_t length);
+extern int efx_mcdi_nvram_write(struct efx_nic *efx, unsigned int type,
+				loff_t offset, const u8 *buffer,
+				size_t length);
+#define EFX_MCDI_NVRAM_LEN_MAX 128
+extern int efx_mcdi_nvram_erase(struct efx_nic *efx, unsigned int type,
+				loff_t offset, size_t length);
+extern int efx_mcdi_nvram_update_finish(struct efx_nic *efx,
+					unsigned int type);
+extern int efx_mcdi_nvram_test_all(struct efx_nic *efx);
+extern int efx_mcdi_handle_assertion(struct efx_nic *efx);
+extern void efx_mcdi_set_id_led(struct efx_nic *efx, enum efx_led_mode mode);
+extern int efx_mcdi_reset_port(struct efx_nic *efx);
+extern int efx_mcdi_reset_mc(struct efx_nic *efx);
+extern int efx_mcdi_wol_filter_set_magic(struct efx_nic *efx,
+					 const u8 *mac, int *id_out);
+extern int efx_mcdi_wol_filter_get_magic(struct efx_nic *efx, int *id_out);
+extern int efx_mcdi_wol_filter_remove(struct efx_nic *efx, int id);
+extern int efx_mcdi_wol_filter_reset(struct efx_nic *efx);
+
+#endif /* EFX_MCDI_H */
diff --git a/drivers/net/ethernet/sfc/mcdi_mac.c b/drivers/net/ethernet/sfc/mcdi_mac.c
new file mode 100644
index 0000000..50c2077
--- /dev/null
+++ b/drivers/net/ethernet/sfc/mcdi_mac.c
@@ -0,0 +1,145 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2009-2010 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#include "net_driver.h"
+#include "efx.h"
+#include "mac.h"
+#include "mcdi.h"
+#include "mcdi_pcol.h"
+
+static int efx_mcdi_set_mac(struct efx_nic *efx)
+{
+	u32 reject, fcntl;
+	u8 cmdbytes[MC_CMD_SET_MAC_IN_LEN];
+
+	memcpy(cmdbytes + MC_CMD_SET_MAC_IN_ADDR_OFST,
+	       efx->net_dev->dev_addr, ETH_ALEN);
+
+	MCDI_SET_DWORD(cmdbytes, SET_MAC_IN_MTU,
+			EFX_MAX_FRAME_LEN(efx->net_dev->mtu));
+	MCDI_SET_DWORD(cmdbytes, SET_MAC_IN_DRAIN, 0);
+
+	/* The MCDI command provides for controlling accept/reject
+	 * of broadcast packets too, but the driver doesn't currently
+	 * expose this. */
+	reject = (efx->promiscuous) ? 0 :
+		(1 << MC_CMD_SET_MAC_IN_REJECT_UNCST_LBN);
+	MCDI_SET_DWORD(cmdbytes, SET_MAC_IN_REJECT, reject);
+
+	switch (efx->wanted_fc) {
+	case EFX_FC_RX | EFX_FC_TX:
+		fcntl = MC_CMD_FCNTL_BIDIR;
+		break;
+	case EFX_FC_RX:
+		fcntl = MC_CMD_FCNTL_RESPOND;
+		break;
+	default:
+		fcntl = MC_CMD_FCNTL_OFF;
+		break;
+	}
+	if (efx->wanted_fc & EFX_FC_AUTO)
+		fcntl = MC_CMD_FCNTL_AUTO;
+
+	MCDI_SET_DWORD(cmdbytes, SET_MAC_IN_FCNTL, fcntl);
+
+	return efx_mcdi_rpc(efx, MC_CMD_SET_MAC, cmdbytes, sizeof(cmdbytes),
+			    NULL, 0, NULL);
+}
+
+static int efx_mcdi_get_mac_faults(struct efx_nic *efx, u32 *faults)
+{
+	u8 outbuf[MC_CMD_GET_LINK_OUT_LEN];
+	size_t outlength;
+	int rc;
+
+	BUILD_BUG_ON(MC_CMD_GET_LINK_IN_LEN != 0);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_GET_LINK, NULL, 0,
+			  outbuf, sizeof(outbuf), &outlength);
+	if (rc)
+		goto fail;
+
+	*faults = MCDI_DWORD(outbuf, GET_LINK_OUT_MAC_FAULT);
+	return 0;
+
+fail:
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n",
+		  __func__, rc);
+	return rc;
+}
+
+int efx_mcdi_mac_stats(struct efx_nic *efx, dma_addr_t dma_addr,
+		       u32 dma_len, int enable, int clear)
+{
+	u8 inbuf[MC_CMD_MAC_STATS_IN_LEN];
+	int rc;
+	efx_dword_t *cmd_ptr;
+	int period = enable ? 1000 : 0;
+	u32 addr_hi;
+	u32 addr_lo;
+
+	BUILD_BUG_ON(MC_CMD_MAC_STATS_OUT_LEN != 0);
+
+	addr_lo = ((u64)dma_addr) >> 0;
+	addr_hi = ((u64)dma_addr) >> 32;
+
+	MCDI_SET_DWORD(inbuf, MAC_STATS_IN_DMA_ADDR_LO, addr_lo);
+	MCDI_SET_DWORD(inbuf, MAC_STATS_IN_DMA_ADDR_HI, addr_hi);
+	cmd_ptr = (efx_dword_t *)MCDI_PTR(inbuf, MAC_STATS_IN_CMD);
+	EFX_POPULATE_DWORD_7(*cmd_ptr,
+			     MC_CMD_MAC_STATS_CMD_DMA, !!enable,
+			     MC_CMD_MAC_STATS_CMD_CLEAR, clear,
+			     MC_CMD_MAC_STATS_CMD_PERIODIC_CHANGE, 1,
+			     MC_CMD_MAC_STATS_CMD_PERIODIC_ENABLE, !!enable,
+			     MC_CMD_MAC_STATS_CMD_PERIODIC_CLEAR, 0,
+			     MC_CMD_MAC_STATS_CMD_PERIODIC_NOEVENT, 1,
+			     MC_CMD_MAC_STATS_CMD_PERIOD_MS, period);
+	MCDI_SET_DWORD(inbuf, MAC_STATS_IN_DMA_LEN, dma_len);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_MAC_STATS, inbuf, sizeof(inbuf),
+			  NULL, 0, NULL);
+	if (rc)
+		goto fail;
+
+	return 0;
+
+fail:
+	netif_err(efx, hw, efx->net_dev, "%s: %s failed rc=%d\n",
+		  __func__, enable ? "enable" : "disable", rc);
+	return rc;
+}
+
+static int efx_mcdi_mac_reconfigure(struct efx_nic *efx)
+{
+	int rc;
+
+	rc = efx_mcdi_set_mac(efx);
+	if (rc != 0)
+		return rc;
+
+	/* Restore the multicast hash registers. */
+	efx->type->push_multicast_hash(efx);
+
+	return 0;
+}
+
+
+static bool efx_mcdi_mac_check_fault(struct efx_nic *efx)
+{
+	u32 faults;
+	int rc = efx_mcdi_get_mac_faults(efx, &faults);
+	return (rc != 0) || (faults != 0);
+}
+
+
+const struct efx_mac_operations efx_mcdi_mac_operations = {
+	.reconfigure	= efx_mcdi_mac_reconfigure,
+	.update_stats	= efx_port_dummy_op_void,
+	.check_fault 	= efx_mcdi_mac_check_fault,
+};
diff --git a/drivers/net/ethernet/sfc/mcdi_pcol.h b/drivers/net/ethernet/sfc/mcdi_pcol.h
new file mode 100644
index 0000000..41fe06f
--- /dev/null
+++ b/drivers/net/ethernet/sfc/mcdi_pcol.h
@@ -0,0 +1,1775 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2009-2011 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+
+#ifndef MCDI_PCOL_H
+#define MCDI_PCOL_H
+
+/* Values to be written into FMCR_CZ_RESET_STATE_REG to control boot. */
+/* Power-on reset state */
+#define MC_FW_STATE_POR (1)
+/* If this is set in MC_RESET_STATE_REG then it should be
+ * possible to jump into IMEM without loading code from flash. */
+#define MC_FW_WARM_BOOT_OK (2)
+/* The MC main image has started to boot. */
+#define MC_FW_STATE_BOOTING (4)
+/* The Scheduler has started. */
+#define MC_FW_STATE_SCHED (8)
+
+/* Values to be written to the per-port status dword in shared
+ * memory on reboot and assert */
+#define MC_STATUS_DWORD_REBOOT (0xb007b007)
+#define MC_STATUS_DWORD_ASSERT (0xdeaddead)
+
+/* The current version of the MCDI protocol.
+ *
+ * Note that the ROM burnt into the card only talks V0, so at the very
+ * least every driver must support version 0 and MCDI_PCOL_VERSION
+ */
+#define MCDI_PCOL_VERSION 1
+
+/**
+ * MCDI version 1
+ *
+ * Each MCDI request starts with an MCDI_HEADER, which is a 32byte
+ * structure, filled in by the client.
+ *
+ *       0       7  8     16    20     22  23  24    31
+ *      | CODE | R | LEN | SEQ | Rsvd | E | R | XFLAGS |
+ *               |                      |   |
+ *               |                      |   \--- Response
+ *               |                      \------- Error
+ *               \------------------------------ Resync (always set)
+ *
+ * The client writes it's request into MC shared memory, and rings the
+ * doorbell. Each request is completed by either by the MC writting
+ * back into shared memory, or by writting out an event.
+ *
+ * All MCDI commands support completion by shared memory response. Each
+ * request may also contain additional data (accounted for by HEADER.LEN),
+ * and some response's may also contain additional data (again, accounted
+ * for by HEADER.LEN).
+ *
+ * Some MCDI commands support completion by event, in which any associated
+ * response data is included in the event.
+ *
+ * The protocol requires one response to be delivered for every request, a
+ * request should not be sent unless the response for the previous request
+ * has been received (either by polling shared memory, or by receiving
+ * an event).
+ */
+
+/** Request/Response structure */
+#define MCDI_HEADER_OFST 0
+#define MCDI_HEADER_CODE_LBN 0
+#define MCDI_HEADER_CODE_WIDTH 7
+#define MCDI_HEADER_RESYNC_LBN 7
+#define MCDI_HEADER_RESYNC_WIDTH 1
+#define MCDI_HEADER_DATALEN_LBN 8
+#define MCDI_HEADER_DATALEN_WIDTH 8
+#define MCDI_HEADER_SEQ_LBN 16
+#define MCDI_HEADER_RSVD_LBN 20
+#define MCDI_HEADER_RSVD_WIDTH 2
+#define MCDI_HEADER_SEQ_WIDTH 4
+#define MCDI_HEADER_ERROR_LBN 22
+#define MCDI_HEADER_ERROR_WIDTH 1
+#define MCDI_HEADER_RESPONSE_LBN 23
+#define MCDI_HEADER_RESPONSE_WIDTH 1
+#define MCDI_HEADER_XFLAGS_LBN 24
+#define MCDI_HEADER_XFLAGS_WIDTH 8
+/* Request response using event */
+#define MCDI_HEADER_XFLAGS_EVREQ 0x01
+
+/* Maximum number of payload bytes */
+#define MCDI_CTL_SDU_LEN_MAX 0xfc
+
+/* The MC can generate events for two reasons:
+ *   - To complete a shared memory request if XFLAGS_EVREQ was set
+ *   - As a notification (link state, i2c event), controlled
+ *     via MC_CMD_LOG_CTRL
+ *
+ * Both events share a common structure:
+ *
+ *  0      32     33      36    44     52     60
+ * | Data | Cont | Level | Src | Code | Rsvd |
+ *           |
+ *           \ There is another event pending in this notification
+ *
+ * If Code==CMDDONE, then the fields are further interpreted as:
+ *
+ *   - LEVEL==INFO    Command succeeded
+ *   - LEVEL==ERR     Command failed
+ *
+ *    0     8         16      24     32
+ *   | Seq | Datalen | Errno | Rsvd |
+ *
+ *   These fields are taken directly out of the standard MCDI header, i.e.,
+ *   LEVEL==ERR, Datalen == 0 => Reboot
+ *
+ * Events can be squirted out of the UART (using LOG_CTRL) without a
+ * MCDI header.  An event can be distinguished from a MCDI response by
+ * examining the first byte which is 0xc0.  This corresponds to the
+ * non-existent MCDI command MC_CMD_DEBUG_LOG.
+ *
+ *      0         7        8
+ *     | command | Resync |     = 0xc0
+ *
+ * Since the event is written in big-endian byte order, this works
+ * providing bits 56-63 of the event are 0xc0.
+ *
+ *      56     60  63
+ *     | Rsvd | Code |    = 0xc0
+ *
+ * Which means for convenience the event code is 0xc for all MC
+ * generated events.
+ */
+#define FSE_AZ_EV_CODE_MCDI_EVRESPONSE 0xc
+
+#define MCDI_EVENT_DATA_LBN 0
+#define MCDI_EVENT_DATA_WIDTH 32
+#define MCDI_EVENT_CONT_LBN 32
+#define MCDI_EVENT_CONT_WIDTH 1
+#define MCDI_EVENT_LEVEL_LBN 33
+#define MCDI_EVENT_LEVEL_WIDTH 3
+#define MCDI_EVENT_LEVEL_INFO (0)
+#define MCDI_EVENT_LEVEL_WARN (1)
+#define MCDI_EVENT_LEVEL_ERR (2)
+#define MCDI_EVENT_LEVEL_FATAL (3)
+#define MCDI_EVENT_SRC_LBN 36
+#define MCDI_EVENT_SRC_WIDTH 8
+#define MCDI_EVENT_CODE_LBN 44
+#define MCDI_EVENT_CODE_WIDTH 8
+#define MCDI_EVENT_CODE_BADSSERT (1)
+#define MCDI_EVENT_CODE_PMNOTICE (2)
+#define MCDI_EVENT_CODE_CMDDONE (3)
+#define  MCDI_EVENT_CMDDONE_SEQ_LBN 0
+#define  MCDI_EVENT_CMDDONE_SEQ_WIDTH 8
+#define  MCDI_EVENT_CMDDONE_DATALEN_LBN 8
+#define  MCDI_EVENT_CMDDONE_DATALEN_WIDTH 8
+#define  MCDI_EVENT_CMDDONE_ERRNO_LBN 16
+#define  MCDI_EVENT_CMDDONE_ERRNO_WIDTH 8
+#define MCDI_EVENT_CODE_LINKCHANGE (4)
+#define  MCDI_EVENT_LINKCHANGE_LP_CAP_LBN 0
+#define  MCDI_EVENT_LINKCHANGE_LP_CAP_WIDTH 16
+#define  MCDI_EVENT_LINKCHANGE_SPEED_LBN 16
+#define  MCDI_EVENT_LINKCHANGE_SPEED_WIDTH 4
+#define  MCDI_EVENT_LINKCHANGE_SPEED_100M 1
+#define  MCDI_EVENT_LINKCHANGE_SPEED_1G 2
+#define  MCDI_EVENT_LINKCHANGE_SPEED_10G 3
+#define  MCDI_EVENT_LINKCHANGE_FCNTL_LBN 20
+#define  MCDI_EVENT_LINKCHANGE_FCNTL_WIDTH 4
+#define  MCDI_EVENT_LINKCHANGE_LINK_FLAGS_LBN 24
+#define  MCDI_EVENT_LINKCHANGE_LINK_FLAGS_WIDTH 8
+#define MCDI_EVENT_CODE_SENSOREVT (5)
+#define  MCDI_EVENT_SENSOREVT_MONITOR_LBN 0
+#define  MCDI_EVENT_SENSOREVT_MONITOR_WIDTH 8
+#define  MCDI_EVENT_SENSOREVT_STATE_LBN 8
+#define  MCDI_EVENT_SENSOREVT_STATE_WIDTH 8
+#define  MCDI_EVENT_SENSOREVT_VALUE_LBN 16
+#define  MCDI_EVENT_SENSOREVT_VALUE_WIDTH 16
+#define MCDI_EVENT_CODE_SCHEDERR (6)
+#define MCDI_EVENT_CODE_REBOOT (7)
+#define MCDI_EVENT_CODE_MAC_STATS_DMA (8)
+#define  MCDI_EVENT_MAC_STATS_DMA_GENERATION_LBN 0
+#define  MCDI_EVENT_MAC_STATS_DMA_GENERATION_WIDTH 32
+
+/* Non-existent command target */
+#define MC_CMD_ERR_ENOENT 2
+/* assert() has killed the MC */
+#define MC_CMD_ERR_EINTR 4
+/* Caller does not hold required locks */
+#define MC_CMD_ERR_EACCES 13
+/* Resource is currently unavailable (e.g. lock contention) */
+#define MC_CMD_ERR_EBUSY 16
+/* Invalid argument to target */
+#define MC_CMD_ERR_EINVAL 22
+/* Non-recursive resource is already acquired */
+#define MC_CMD_ERR_EDEADLK 35
+/* Operation not implemented */
+#define MC_CMD_ERR_ENOSYS 38
+/* Operation timed out */
+#define MC_CMD_ERR_ETIME 62
+
+#define MC_CMD_ERR_CODE_OFST 0
+
+
+/* MC_CMD_READ32: (debug, variadic out)
+ * Read multiple 32byte words from MC memory
+ */
+#define MC_CMD_READ32 0x01
+#define MC_CMD_READ32_IN_LEN 8
+#define MC_CMD_READ32_IN_ADDR_OFST 0
+#define MC_CMD_READ32_IN_NUMWORDS_OFST 4
+#define MC_CMD_READ32_OUT_LEN(_numwords) \
+	(4 * (_numwords))
+#define MC_CMD_READ32_OUT_BUFFER_OFST 0
+
+/* MC_CMD_WRITE32: (debug, variadic in)
+ * Write multiple 32byte words to MC memory
+ */
+#define MC_CMD_WRITE32 0x02
+#define MC_CMD_WRITE32_IN_LEN(_numwords) (((_numwords) * 4) + 4)
+#define MC_CMD_WRITE32_IN_ADDR_OFST 0
+#define MC_CMD_WRITE32_IN_BUFFER_OFST 4
+#define MC_CMD_WRITE32_OUT_LEN 0
+
+/* MC_CMD_COPYCODE: (debug)
+ * Copy MC code between two locations and jump
+ */
+#define MC_CMD_COPYCODE 0x03
+#define MC_CMD_COPYCODE_IN_LEN 16
+#define MC_CMD_COPYCODE_IN_SRC_ADDR_OFST 0
+#define MC_CMD_COPYCODE_IN_DEST_ADDR_OFST 4
+#define MC_CMD_COPYCODE_IN_NUMWORDS_OFST 8
+#define MC_CMD_COPYCODE_IN_JUMP_OFST 12
+/* Control should return to the caller rather than jumping */
+#define MC_CMD_COPYCODE_JUMP_NONE 1
+#define MC_CMD_COPYCODE_OUT_LEN 0
+
+/* MC_CMD_SET_FUNC: (debug)
+ * Select function for function-specific commands.
+ */
+#define MC_CMD_SET_FUNC 0x04
+#define MC_CMD_SET_FUNC_IN_LEN 4
+#define MC_CMD_SET_FUNC_IN_FUNC_OFST 0
+#define MC_CMD_SET_FUNC_OUT_LEN 0
+
+/* MC_CMD_GET_BOOT_STATUS:
+ * Get the instruction address from which the MC booted.
+ */
+#define MC_CMD_GET_BOOT_STATUS 0x05
+#define MC_CMD_GET_BOOT_STATUS_IN_LEN 0
+#define MC_CMD_GET_BOOT_STATUS_OUT_LEN 8
+#define MC_CMD_GET_BOOT_STATUS_OUT_BOOT_OFFSET_OFST 0
+#define MC_CMD_GET_BOOT_STATUS_OUT_FLAGS_OFST 4
+/* Reboot caused by watchdog */
+#define MC_CMD_GET_BOOT_STATUS_FLAGS_WATCHDOG_LBN   (0)
+#define MC_CMD_GET_BOOT_STATUS_FLAGS_WATCHDOG_WIDTH (1)
+/* MC booted from primary flash partition */
+#define MC_CMD_GET_BOOT_STATUS_FLAGS_PRIMARY_LBN    (1)
+#define MC_CMD_GET_BOOT_STATUS_FLAGS_PRIMARY_WIDTH  (1)
+/* MC booted from backup flash partition */
+#define MC_CMD_GET_BOOT_STATUS_FLAGS_BACKUP_LBN     (2)
+#define MC_CMD_GET_BOOT_STATUS_FLAGS_BACKUP_WIDTH   (1)
+
+/* MC_CMD_GET_ASSERTS: (debug, variadic out)
+ * Get (and optionally clear) the current assertion status.
+ *
+ * Only OUT.GLOBAL_FLAGS is guaranteed to exist in the completion
+ * payload. The other fields will only be present if
+ * OUT.GLOBAL_FLAGS != NO_FAILS
+ */
+#define MC_CMD_GET_ASSERTS 0x06
+#define MC_CMD_GET_ASSERTS_IN_LEN 4
+#define MC_CMD_GET_ASSERTS_IN_CLEAR_OFST 0
+#define MC_CMD_GET_ASSERTS_OUT_LEN 140
+/* Assertion status flag */
+#define MC_CMD_GET_ASSERTS_OUT_GLOBAL_FLAGS_OFST 0
+/*! No assertions have failed. */
+#define MC_CMD_GET_ASSERTS_FLAGS_NO_FAILS 1
+/*! A system-level assertion has failed. */
+#define MC_CMD_GET_ASSERTS_FLAGS_SYS_FAIL 2
+/*! A thread-level assertion has failed. */
+#define MC_CMD_GET_ASSERTS_FLAGS_THR_FAIL 3
+/*! The system was reset by the watchdog. */
+#define MC_CMD_GET_ASSERTS_FLAGS_WDOG_FIRED 4
+/* Failing PC value */
+#define MC_CMD_GET_ASSERTS_OUT_SAVED_PC_OFFS_OFST 4
+/* Saved GP regs */
+#define MC_CMD_GET_ASSERTS_OUT_GP_REGS_OFFS_OFST 8
+#define MC_CMD_GET_ASSERTS_OUT_GP_REGS_LEN 124
+/* Failing thread address */
+#define MC_CMD_GET_ASSERTS_OUT_THREAD_OFFS_OFST 132
+
+/* MC_CMD_LOG_CTRL:
+ * Determine the output stream for various events and messages
+ */
+#define MC_CMD_LOG_CTRL 0x07
+#define MC_CMD_LOG_CTRL_IN_LEN 8
+#define MC_CMD_LOG_CTRL_IN_LOG_DEST_OFST 0
+#define MC_CMD_LOG_CTRL_IN_LOG_DEST_UART (1)
+#define MC_CMD_LOG_CTRL_IN_LOG_DEST_EVQ (2)
+#define MC_CMD_LOG_CTRL_IN_LOG_DEST_EVQ_OFST 4
+#define MC_CMD_LOG_CTRL_OUT_LEN 0
+
+/* MC_CMD_GET_VERSION:
+ * Get version information about the MC firmware
+ */
+#define MC_CMD_GET_VERSION 0x08
+#define MC_CMD_GET_VERSION_IN_LEN 0
+#define MC_CMD_GET_VERSION_V0_OUT_LEN 4
+#define MC_CMD_GET_VERSION_V1_OUT_LEN 32
+#define MC_CMD_GET_VERSION_OUT_FIRMWARE_OFST 0
+/* Reserved version number to indicate "any" version. */
+#define MC_CMD_GET_VERSION_OUT_FIRMWARE_ANY 0xffffffff
+/* The version response of a boot ROM awaiting rescue */
+#define MC_CMD_GET_VERSION_OUT_FIRMWARE_BOOTROM 0xb0070000
+#define MC_CMD_GET_VERSION_V1_OUT_PCOL_OFST 4
+/* 128bit mask of functions supported by the current firmware */
+#define MC_CMD_GET_VERSION_V1_OUT_SUPPORTED_FUNCS_OFST 8
+/* The command set exported by the boot ROM (MCDI v0) */
+#define MC_CMD_GET_VERSION_V0_SUPPORTED_FUNCS {		\
+	(1 << MC_CMD_READ32)	|			\
+	(1 << MC_CMD_WRITE32)	|			\
+	(1 << MC_CMD_COPYCODE)	|			\
+	(1 << MC_CMD_GET_VERSION),			\
+	0, 0, 0 }
+#define MC_CMD_GET_VERSION_OUT_VERSION_OFST 24
+
+/* Vectors in the boot ROM */
+/* Point to the copycode entry point. */
+#define MC_BOOTROM_COPYCODE_VEC (0x7f4)
+/* Points to the recovery mode entry point. */
+#define MC_BOOTROM_NOFLASH_VEC (0x7f8)
+
+/* Test execution limits */
+#define MC_TESTEXEC_VARIANT_COUNT 16
+#define MC_TESTEXEC_RESULT_COUNT 7
+
+/* MC_CMD_SET_TESTVARS: (debug, variadic in)
+ * Write variant words for test.
+ *
+ * The user supplies a bitmap of the variants they wish to set.
+ * They must ensure that IN.LEN >= 4 + 4 * ffs(BITMAP)
+ */
+#define MC_CMD_SET_TESTVARS 0x09
+#define MC_CMD_SET_TESTVARS_IN_LEN(_numwords)	\
+  (4 + 4*(_numwords))
+#define MC_CMD_SET_TESTVARS_IN_ARGS_BITMAP_OFST 0
+/* Up to MC_TESTEXEC_VARIANT_COUNT of 32byte words start here */
+#define MC_CMD_SET_TESTVARS_IN_ARGS_BUFFER_OFST 4
+#define MC_CMD_SET_TESTVARS_OUT_LEN 0
+
+/* MC_CMD_GET_TESTRCS: (debug, variadic out)
+ * Return result words from test.
+ */
+#define MC_CMD_GET_TESTRCS 0x0a
+#define MC_CMD_GET_TESTRCS_IN_LEN 4
+#define MC_CMD_GET_TESTRCS_IN_NUMWORDS_OFST 0
+#define MC_CMD_GET_TESTRCS_OUT_LEN(_numwords) \
+	(4 * (_numwords))
+#define MC_CMD_GET_TESTRCS_OUT_BUFFER_OFST 0
+
+/* MC_CMD_RUN_TEST: (debug)
+ * Run the test exported by this firmware image
+ */
+#define MC_CMD_RUN_TEST 0x0b
+#define MC_CMD_RUN_TEST_IN_LEN 0
+#define MC_CMD_RUN_TEST_OUT_LEN 0
+
+/* MC_CMD_CSR_READ32: (debug, variadic out)
+ * Read 32bit words from the indirect memory map
+ */
+#define MC_CMD_CSR_READ32 0x0c
+#define MC_CMD_CSR_READ32_IN_LEN 12
+#define MC_CMD_CSR_READ32_IN_ADDR_OFST 0
+#define MC_CMD_CSR_READ32_IN_STEP_OFST 4
+#define MC_CMD_CSR_READ32_IN_NUMWORDS_OFST 8
+#define MC_CMD_CSR_READ32_OUT_LEN(_numwords)	\
+	(((_numwords) * 4) + 4)
+/* IN.NUMWORDS of 32bit words start here */
+#define MC_CMD_CSR_READ32_OUT_BUFFER_OFST 0
+#define MC_CMD_CSR_READ32_OUT_IREG_STATUS_OFST(_numwords)	\
+	((_numwords) * 4)
+
+/* MC_CMD_CSR_WRITE32: (debug, variadic in)
+ * Write 32bit dwords to the indirect memory map
+ */
+#define MC_CMD_CSR_WRITE32 0x0d
+#define MC_CMD_CSR_WRITE32_IN_LEN(_numwords)	\
+	(((_numwords) * 4) + 8)
+#define MC_CMD_CSR_WRITE32_IN_ADDR_OFST 0
+#define MC_CMD_CSR_WRITE32_IN_STEP_OFST 4
+/* Multiple 32bit words of data to write start here */
+#define MC_CMD_CSR_WRITE32_IN_BUFFER_OFST 8
+#define MC_CMD_CSR_WRITE32_OUT_LEN 4
+#define MC_CMD_CSR_WRITE32_OUT_STATUS_OFST 0
+
+/* MC_CMD_JTAG_WORK: (debug, fpga only)
+ * Process JTAG work buffer for RBF acceleration.
+ *
+ *  Host: bit count, (up to) 32 words of data to clock out to JTAG
+ *   (bits 1,0=TMS,TDO for first bit; bits 3,2=TMS,TDO for second bit, etc.)
+ *  MC: bit count, (up to) 32 words of data clocked in from JTAG
+ *   (bit 0=TDI for first bit, bit 1=TDI for second bit, etc.; [31:16] unused)
+ */
+#define MC_CMD_JTAG_WORK 0x0e
+
+/* MC_CMD_STACKINFO: (debug, variadic out)
+ * Get stack information
+ *
+ * Host: nothing
+ * MC: (thread ptr, stack size, free space) for each thread in system
+ */
+#define MC_CMD_STACKINFO 0x0f
+
+/* MC_CMD_MDIO_READ:
+ * MDIO register read
+ */
+#define MC_CMD_MDIO_READ 0x10
+#define MC_CMD_MDIO_READ_IN_LEN 16
+#define MC_CMD_MDIO_READ_IN_BUS_OFST 0
+#define MC_CMD_MDIO_READ_IN_PRTAD_OFST 4
+#define MC_CMD_MDIO_READ_IN_DEVAD_OFST 8
+#define MC_CMD_MDIO_READ_IN_ADDR_OFST 12
+#define MC_CMD_MDIO_READ_OUT_LEN 8
+#define MC_CMD_MDIO_READ_OUT_VALUE_OFST 0
+#define MC_CMD_MDIO_READ_OUT_STATUS_OFST 4
+
+/* MC_CMD_MDIO_WRITE:
+ * MDIO register write
+ */
+#define MC_CMD_MDIO_WRITE 0x11
+#define MC_CMD_MDIO_WRITE_IN_LEN 20
+#define MC_CMD_MDIO_WRITE_IN_BUS_OFST 0
+#define MC_CMD_MDIO_WRITE_IN_PRTAD_OFST 4
+#define MC_CMD_MDIO_WRITE_IN_DEVAD_OFST 8
+#define MC_CMD_MDIO_WRITE_IN_ADDR_OFST 12
+#define MC_CMD_MDIO_WRITE_IN_VALUE_OFST 16
+#define MC_CMD_MDIO_WRITE_OUT_LEN 4
+#define MC_CMD_MDIO_WRITE_OUT_STATUS_OFST 0
+
+/* By default all the MCDI MDIO operations perform clause45 mode.
+ * If you want to use clause22 then set DEVAD = MC_CMD_MDIO_CLAUSE22.
+ */
+#define MC_CMD_MDIO_CLAUSE22 32
+
+/* There are two MDIO buses: one for the internal PHY, and one for external
+ * devices.
+ */
+#define MC_CMD_MDIO_BUS_INTERNAL 0
+#define MC_CMD_MDIO_BUS_EXTERNAL 1
+
+/* The MDIO commands return the raw status bits from the MDIO block.  A "good"
+ * transaction should have the DONE bit set and all other bits clear.
+ */
+#define MC_CMD_MDIO_STATUS_GOOD 0x08
+
+
+/* MC_CMD_DBI_WRITE: (debug)
+ * Write DBI register(s)
+ *
+ * Host: address, byte-enables (and VF selection, and cs2 flag),
+ *       value [,address ...]
+ * MC: nothing
+ */
+#define MC_CMD_DBI_WRITE 0x12
+#define MC_CMD_DBI_WRITE_IN_LEN(_numwords)		\
+	(12 * (_numwords))
+#define MC_CMD_DBI_WRITE_IN_ADDRESS_OFST(_word)		\
+	(((_word) * 12) + 0)
+#define MC_CMD_DBI_WRITE_IN_BYTE_MASK_OFST(_word)	\
+	(((_word) * 12) + 4)
+#define MC_CMD_DBI_WRITE_IN_VALUE_OFST(_word)		\
+	(((_word) * 12) + 8)
+#define MC_CMD_DBI_WRITE_OUT_LEN 0
+
+/* MC_CMD_DBI_READ: (debug)
+ * Read DBI register(s)
+ *
+ * Host: address, [,address ...]
+ * MC: value [,value ...]
+ * (note: this does not support reading from VFs, but is retained for backwards
+ * compatibility; see MC_CMD_DBI_READX below)
+ */
+#define MC_CMD_DBI_READ 0x13
+#define MC_CMD_DBI_READ_IN_LEN(_numwords)		\
+	(4 * (_numwords))
+#define MC_CMD_DBI_READ_OUT_LEN(_numwords)		\
+	(4 * (_numwords))
+
+/* MC_CMD_PORT_READ32: (debug)
+ * Read a 32-bit register from the indirect port register map.
+ *
+ * The port to access is implied by the Shared memory channel used.
+ */
+#define MC_CMD_PORT_READ32 0x14
+#define MC_CMD_PORT_READ32_IN_LEN 4
+#define MC_CMD_PORT_READ32_IN_ADDR_OFST 0
+#define MC_CMD_PORT_READ32_OUT_LEN 8
+#define MC_CMD_PORT_READ32_OUT_VALUE_OFST 0
+#define MC_CMD_PORT_READ32_OUT_STATUS_OFST 4
+
+/* MC_CMD_PORT_WRITE32: (debug)
+ * Write a 32-bit register to the indirect port register map.
+ *
+ * The port to access is implied by the Shared memory channel used.
+ */
+#define MC_CMD_PORT_WRITE32 0x15
+#define MC_CMD_PORT_WRITE32_IN_LEN 8
+#define MC_CMD_PORT_WRITE32_IN_ADDR_OFST 0
+#define MC_CMD_PORT_WRITE32_IN_VALUE_OFST 4
+#define MC_CMD_PORT_WRITE32_OUT_LEN 4
+#define MC_CMD_PORT_WRITE32_OUT_STATUS_OFST 0
+
+/* MC_CMD_PORT_READ128: (debug)
+ * Read a 128-bit register from indirect port register map
+ *
+ * The port to access is implied by the Shared memory channel used.
+ */
+#define MC_CMD_PORT_READ128 0x16
+#define MC_CMD_PORT_READ128_IN_LEN 4
+#define MC_CMD_PORT_READ128_IN_ADDR_OFST 0
+#define MC_CMD_PORT_READ128_OUT_LEN 20
+#define MC_CMD_PORT_READ128_OUT_VALUE_OFST 0
+#define MC_CMD_PORT_READ128_OUT_STATUS_OFST 16
+
+/* MC_CMD_PORT_WRITE128: (debug)
+ * Write a 128-bit register to indirect port register map.
+ *
+ * The port to access is implied by the Shared memory channel used.
+ */
+#define MC_CMD_PORT_WRITE128 0x17
+#define MC_CMD_PORT_WRITE128_IN_LEN 20
+#define MC_CMD_PORT_WRITE128_IN_ADDR_OFST 0
+#define MC_CMD_PORT_WRITE128_IN_VALUE_OFST 4
+#define MC_CMD_PORT_WRITE128_OUT_LEN 4
+#define MC_CMD_PORT_WRITE128_OUT_STATUS_OFST 0
+
+/* MC_CMD_GET_BOARD_CFG:
+ * Returns the MC firmware configuration structure
+ *
+ * The FW_SUBTYPE_LIST contains a 16-bit value for each of the 12 types of
+ * NVRAM area.  The values are defined in the firmware/mc/platform/<xxx>.c file
+ * for a specific board type, but otherwise have no meaning to the MC; they
+ * are used by the driver to manage selection of appropriate firmware updates.
+ */
+#define MC_CMD_GET_BOARD_CFG 0x18
+#define MC_CMD_GET_BOARD_CFG_IN_LEN 0
+#define MC_CMD_GET_BOARD_CFG_OUT_LEN 96
+#define MC_CMD_GET_BOARD_CFG_OUT_BOARD_TYPE_OFST 0
+#define MC_CMD_GET_BOARD_CFG_OUT_BOARD_NAME_OFST 4
+#define MC_CMD_GET_BOARD_CFG_OUT_BOARD_NAME_LEN 32
+#define MC_CMD_GET_BOARD_CFG_OUT_CAPABILITIES_PORT0_OFST 36
+#define MC_CMD_GET_BOARD_CFG_OUT_CAPABILITIES_PORT1_OFST 40
+#define MC_CMD_GET_BOARD_CFG_OUT_MAC_ADDR_BASE_PORT0_OFST 44
+#define MC_CMD_GET_BOARD_CFG_OUT_MAC_ADDR_BASE_PORT0_LEN 6
+#define MC_CMD_GET_BOARD_CFG_OUT_MAC_ADDR_BASE_PORT1_OFST 50
+#define MC_CMD_GET_BOARD_CFG_OUT_MAC_ADDR_BASE_PORT1_LEN 6
+#define MC_CMD_GET_BOARD_CFG_OUT_MAC_COUNT_PORT0_OFST 56
+#define MC_CMD_GET_BOARD_CFG_OUT_MAC_COUNT_PORT1_OFST 60
+#define MC_CMD_GET_BOARD_CFG_OUT_MAC_STRIDE_PORT0_OFST 64
+#define MC_CMD_GET_BOARD_CFG_OUT_MAC_STRIDE_PORT1_OFST 68
+#define MC_CMD_GET_BOARD_CFG_OUT_FW_SUBTYPE_LIST_OFST 72
+#define MC_CMD_GET_BOARD_CFG_OUT_FW_SUBTYPE_LIST_LEN 24
+
+/* MC_CMD_DBI_READX: (debug)
+ * Read DBI register(s) -- extended functionality
+ *
+ * Host: vf selection, address, [,vf selection ...]
+ * MC: value [,value ...]
+ */
+#define MC_CMD_DBI_READX 0x19
+#define MC_CMD_DBI_READX_IN_LEN(_numwords)	\
+  (8*(_numwords))
+#define MC_CMD_DBI_READX_OUT_LEN(_numwords)	\
+  (4*(_numwords))
+
+/* MC_CMD_SET_RAND_SEED:
+ * Set the 16byte seed for the MC pseudo-random generator
+ */
+#define MC_CMD_SET_RAND_SEED 0x1a
+#define MC_CMD_SET_RAND_SEED_IN_LEN 16
+#define MC_CMD_SET_RAND_SEED_IN_SEED_OFST 0
+#define MC_CMD_SET_RAND_SEED_OUT_LEN 0
+
+/* MC_CMD_LTSSM_HIST: (debug)
+ * Retrieve the history of the LTSSM, if the build supports it.
+ *
+ * Host: nothing
+ * MC: variable number of LTSSM values, as bytes
+ * The history is read-to-clear.
+ */
+#define MC_CMD_LTSSM_HIST 0x1b
+
+/* MC_CMD_DRV_ATTACH:
+ * Inform MCPU that this port is managed on the host (i.e. driver active)
+ */
+#define MC_CMD_DRV_ATTACH 0x1c
+#define MC_CMD_DRV_ATTACH_IN_LEN 8
+#define MC_CMD_DRV_ATTACH_IN_NEW_STATE_OFST 0
+#define MC_CMD_DRV_ATTACH_IN_UPDATE_OFST 4
+#define MC_CMD_DRV_ATTACH_OUT_LEN 4
+#define MC_CMD_DRV_ATTACH_OUT_OLD_STATE_OFST 0
+
+/* MC_CMD_NCSI_PROD: (debug)
+ * Trigger an NC-SI event (and possibly an AEN in response)
+ */
+#define MC_CMD_NCSI_PROD 0x1d
+#define MC_CMD_NCSI_PROD_IN_LEN 4
+#define MC_CMD_NCSI_PROD_IN_EVENTS_OFST 0
+#define MC_CMD_NCSI_PROD_LINKCHANGE_LBN 0
+#define MC_CMD_NCSI_PROD_LINKCHANGE_WIDTH 1
+#define MC_CMD_NCSI_PROD_RESET_LBN 1
+#define MC_CMD_NCSI_PROD_RESET_WIDTH 1
+#define MC_CMD_NCSI_PROD_DRVATTACH_LBN 2
+#define MC_CMD_NCSI_PROD_DRVATTACH_WIDTH 1
+#define MC_CMD_NCSI_PROD_OUT_LEN 0
+
+/* Enumeration */
+#define MC_CMD_NCSI_PROD_LINKCHANGE 0
+#define MC_CMD_NCSI_PROD_RESET 1
+#define MC_CMD_NCSI_PROD_DRVATTACH 2
+
+/* MC_CMD_DEVEL: (debug)
+ * Reserved for development
+ */
+#define MC_CMD_DEVEL 0x1e
+
+/* MC_CMD_SHMUART: (debug)
+ * Route UART output to circular buffer in shared memory instead.
+ */
+#define MC_CMD_SHMUART 0x1f
+#define MC_CMD_SHMUART_IN_FLAG_OFST 0
+#define MC_CMD_SHMUART_IN_LEN 4
+#define MC_CMD_SHMUART_OUT_LEN 0
+
+/* MC_CMD_PORT_RESET:
+ * Generic per-port reset. There is no equivalent for per-board reset.
+ *
+ * Locks required: None
+ * Return code: 0, ETIME
+ */
+#define MC_CMD_PORT_RESET 0x20
+#define MC_CMD_PORT_RESET_IN_LEN 0
+#define MC_CMD_PORT_RESET_OUT_LEN 0
+
+/* MC_CMD_RESOURCE_LOCK:
+ * Generic resource lock/unlock interface.
+ *
+ * Locks required: None
+ * Return code: 0,
+ *              EBUSY (if trylock is contended by other port),
+ *              EDEADLK (if trylock is already acquired by this port)
+ *              EINVAL (if unlock doesn't own the lock)
+ */
+#define MC_CMD_RESOURCE_LOCK 0x21
+#define MC_CMD_RESOURCE_LOCK_IN_LEN 8
+#define MC_CMD_RESOURCE_LOCK_IN_ACTION_OFST 0
+#define MC_CMD_RESOURCE_LOCK_ACTION_TRYLOCK 1
+#define MC_CMD_RESOURCE_LOCK_ACTION_UNLOCK 0
+#define MC_CMD_RESOURCE_LOCK_IN_RESOURCE_OFST 4
+#define MC_CMD_RESOURCE_LOCK_I2C 2
+#define MC_CMD_RESOURCE_LOCK_PHY 3
+#define MC_CMD_RESOURCE_LOCK_OUT_LEN 0
+
+/* MC_CMD_SPI_COMMAND: (variadic in, variadic out)
+ * Read/Write to/from the SPI device.
+ *
+ * Locks required: SPI_LOCK
+ * Return code: 0, ETIME, EINVAL, EACCES (if SPI_LOCK is not held)
+ */
+#define MC_CMD_SPI_COMMAND 0x22
+#define MC_CMD_SPI_COMMAND_IN_LEN(_write_bytes)	(12 + (_write_bytes))
+#define MC_CMD_SPI_COMMAND_IN_ARGS_OFST 0
+#define MC_CMD_SPI_COMMAND_IN_ARGS_ADDRESS_OFST 0
+#define MC_CMD_SPI_COMMAND_IN_ARGS_READ_BYTES_OFST 4
+#define MC_CMD_SPI_COMMAND_IN_ARGS_CHIP_SELECT_OFST 8
+/* Data to write here */
+#define MC_CMD_SPI_COMMAND_IN_WRITE_BUFFER_OFST 12
+#define MC_CMD_SPI_COMMAND_OUT_LEN(_read_bytes) (_read_bytes)
+/* Data read here */
+#define MC_CMD_SPI_COMMAND_OUT_READ_BUFFER_OFST 0
+
+/* MC_CMD_I2C_READ_WRITE: (variadic in, variadic out)
+ * Read/Write to/from the I2C bus.
+ *
+ * Locks required: I2C_LOCK
+ * Return code: 0, ETIME, EINVAL, EACCES (if I2C_LOCK is not held)
+ */
+#define MC_CMD_I2C_RW 0x23
+#define MC_CMD_I2C_RW_IN_LEN(_write_bytes) (8 + (_write_bytes))
+#define MC_CMD_I2C_RW_IN_ARGS_OFST 0
+#define MC_CMD_I2C_RW_IN_ARGS_ADDR_OFST 0
+#define MC_CMD_I2C_RW_IN_ARGS_READ_BYTES_OFST 4
+/* Data to write here */
+#define MC_CMD_I2C_RW_IN_WRITE_BUFFER_OFSET 8
+#define MC_CMD_I2C_RW_OUT_LEN(_read_bytes) (_read_bytes)
+/* Data read here */
+#define MC_CMD_I2C_RW_OUT_READ_BUFFER_OFST 0
+
+/* Generic phy capability bitmask */
+#define MC_CMD_PHY_CAP_10HDX_LBN 1
+#define MC_CMD_PHY_CAP_10HDX_WIDTH 1
+#define MC_CMD_PHY_CAP_10FDX_LBN 2
+#define MC_CMD_PHY_CAP_10FDX_WIDTH 1
+#define MC_CMD_PHY_CAP_100HDX_LBN 3
+#define MC_CMD_PHY_CAP_100HDX_WIDTH 1
+#define MC_CMD_PHY_CAP_100FDX_LBN 4
+#define MC_CMD_PHY_CAP_100FDX_WIDTH 1
+#define MC_CMD_PHY_CAP_1000HDX_LBN 5
+#define MC_CMD_PHY_CAP_1000HDX_WIDTH 1
+#define MC_CMD_PHY_CAP_1000FDX_LBN 6
+#define MC_CMD_PHY_CAP_1000FDX_WIDTH 1
+#define MC_CMD_PHY_CAP_10000FDX_LBN 7
+#define MC_CMD_PHY_CAP_10000FDX_WIDTH 1
+#define MC_CMD_PHY_CAP_PAUSE_LBN 8
+#define MC_CMD_PHY_CAP_PAUSE_WIDTH 1
+#define MC_CMD_PHY_CAP_ASYM_LBN 9
+#define MC_CMD_PHY_CAP_ASYM_WIDTH 1
+#define MC_CMD_PHY_CAP_AN_LBN 10
+#define MC_CMD_PHY_CAP_AN_WIDTH 1
+
+/* Generic loopback enumeration */
+#define MC_CMD_LOOPBACK_NONE 0
+#define MC_CMD_LOOPBACK_DATA 1
+#define MC_CMD_LOOPBACK_GMAC 2
+#define MC_CMD_LOOPBACK_XGMII 3
+#define MC_CMD_LOOPBACK_XGXS 4
+#define MC_CMD_LOOPBACK_XAUI 5
+#define MC_CMD_LOOPBACK_GMII 6
+#define MC_CMD_LOOPBACK_SGMII 7
+#define MC_CMD_LOOPBACK_XGBR 8
+#define MC_CMD_LOOPBACK_XFI 9
+#define MC_CMD_LOOPBACK_XAUI_FAR 10
+#define MC_CMD_LOOPBACK_GMII_FAR 11
+#define MC_CMD_LOOPBACK_SGMII_FAR 12
+#define MC_CMD_LOOPBACK_XFI_FAR 13
+#define MC_CMD_LOOPBACK_GPHY 14
+#define MC_CMD_LOOPBACK_PHYXS 15
+#define MC_CMD_LOOPBACK_PCS 16
+#define MC_CMD_LOOPBACK_PMAPMD 17
+#define MC_CMD_LOOPBACK_XPORT 18
+#define MC_CMD_LOOPBACK_XGMII_WS 19
+#define MC_CMD_LOOPBACK_XAUI_WS 20
+#define MC_CMD_LOOPBACK_XAUI_WS_FAR 21
+#define MC_CMD_LOOPBACK_XAUI_WS_NEAR 22
+#define MC_CMD_LOOPBACK_GMII_WS 23
+#define MC_CMD_LOOPBACK_XFI_WS 24
+#define MC_CMD_LOOPBACK_XFI_WS_FAR 25
+#define MC_CMD_LOOPBACK_PHYXS_WS 26
+
+/* Generic PHY statistics enumeration */
+#define MC_CMD_OUI 0
+#define MC_CMD_PMA_PMD_LINK_UP 1
+#define MC_CMD_PMA_PMD_RX_FAULT 2
+#define MC_CMD_PMA_PMD_TX_FAULT 3
+#define MC_CMD_PMA_PMD_SIGNAL 4
+#define MC_CMD_PMA_PMD_SNR_A 5
+#define MC_CMD_PMA_PMD_SNR_B 6
+#define MC_CMD_PMA_PMD_SNR_C 7
+#define MC_CMD_PMA_PMD_SNR_D 8
+#define MC_CMD_PCS_LINK_UP 9
+#define MC_CMD_PCS_RX_FAULT 10
+#define MC_CMD_PCS_TX_FAULT 11
+#define MC_CMD_PCS_BER 12
+#define MC_CMD_PCS_BLOCK_ERRORS 13
+#define MC_CMD_PHYXS_LINK_UP 14
+#define MC_CMD_PHYXS_RX_FAULT 15
+#define MC_CMD_PHYXS_TX_FAULT 16
+#define MC_CMD_PHYXS_ALIGN 17
+#define MC_CMD_PHYXS_SYNC 18
+#define MC_CMD_AN_LINK_UP 19
+#define MC_CMD_AN_COMPLETE 20
+#define MC_CMD_AN_10GBT_STATUS 21
+#define MC_CMD_CL22_LINK_UP 22
+#define MC_CMD_PHY_NSTATS 23
+
+/* MC_CMD_GET_PHY_CFG:
+ * Report PHY configuration.  This guarantees to succeed even if the PHY is in
+ * a "zombie" state.
+ *
+ * Locks required: None
+ * Return code: 0
+ */
+#define MC_CMD_GET_PHY_CFG 0x24
+
+#define MC_CMD_GET_PHY_CFG_IN_LEN 0
+#define MC_CMD_GET_PHY_CFG_OUT_LEN 72
+
+#define MC_CMD_GET_PHY_CFG_OUT_FLAGS_OFST 0
+#define MC_CMD_GET_PHY_CFG_PRESENT_LBN 0
+#define MC_CMD_GET_PHY_CFG_PRESENT_WIDTH 1
+#define MC_CMD_GET_PHY_CFG_BIST_CABLE_SHORT_LBN 1
+#define MC_CMD_GET_PHY_CFG_BIST_CABLE_SHORT_WIDTH 1
+#define MC_CMD_GET_PHY_CFG_BIST_CABLE_LONG_LBN 2
+#define MC_CMD_GET_PHY_CFG_BIST_CABLE_LONG_WIDTH 1
+#define MC_CMD_GET_PHY_CFG_LOWPOWER_LBN 3
+#define MC_CMD_GET_PHY_CFG_LOWPOWER_WIDTH 1
+#define MC_CMD_GET_PHY_CFG_POWEROFF_LBN 4
+#define MC_CMD_GET_PHY_CFG_POWEROFF_WIDTH 1
+#define MC_CMD_GET_PHY_CFG_TXDIS_LBN 5
+#define MC_CMD_GET_PHY_CFG_TXDIS_WIDTH 1
+#define MC_CMD_GET_PHY_CFG_BIST_LBN 6
+#define MC_CMD_GET_PHY_CFG_BIST_WIDTH 1
+#define MC_CMD_GET_PHY_CFG_OUT_TYPE_OFST 4
+/* Bitmask of supported capabilities */
+#define MC_CMD_GET_PHY_CFG_OUT_SUPPORTED_CAP_OFST 8
+#define MC_CMD_GET_PHY_CFG_OUT_CHANNEL_OFST 12
+#define MC_CMD_GET_PHY_CFG_OUT_PRT_OFST 16
+/* PHY statistics bitmap */
+#define MC_CMD_GET_PHY_CFG_OUT_STATS_MASK_OFST 20
+/* PHY type/name string */
+#define MC_CMD_GET_PHY_CFG_OUT_NAME_OFST 24
+#define MC_CMD_GET_PHY_CFG_OUT_NAME_LEN 20
+#define MC_CMD_GET_PHY_CFG_OUT_MEDIA_TYPE_OFST 44
+#define MC_CMD_MEDIA_XAUI 1
+#define MC_CMD_MEDIA_CX4 2
+#define MC_CMD_MEDIA_KX4 3
+#define MC_CMD_MEDIA_XFP 4
+#define MC_CMD_MEDIA_SFP_PLUS 5
+#define MC_CMD_MEDIA_BASE_T 6
+/* MDIO "MMDS" supported */
+#define MC_CMD_GET_PHY_CFG_OUT_MMD_MASK_OFST 48
+/* Native clause 22 */
+#define MC_CMD_MMD_CLAUSE22  0
+#define MC_CMD_MMD_CLAUSE45_PMAPMD 1
+#define MC_CMD_MMD_CLAUSE45_WIS 2
+#define MC_CMD_MMD_CLAUSE45_PCS 3
+#define MC_CMD_MMD_CLAUSE45_PHYXS 4
+#define MC_CMD_MMD_CLAUSE45_DTEXS 5
+#define MC_CMD_MMD_CLAUSE45_TC 6
+#define MC_CMD_MMD_CLAUSE45_AN 7
+/* Clause22 proxied over clause45 by PHY */
+#define MC_CMD_MMD_CLAUSE45_C22EXT 29
+#define MC_CMD_MMD_CLAUSE45_VEND1 30
+#define MC_CMD_MMD_CLAUSE45_VEND2 31
+/* PHY stepping version */
+#define MC_CMD_GET_PHY_CFG_OUT_REVISION_OFST 52
+#define MC_CMD_GET_PHY_CFG_OUT_REVISION_LEN 20
+
+/* MC_CMD_START_BIST:
+ * Start a BIST test on the PHY.
+ *
+ * Locks required: PHY_LOCK if doing a  PHY BIST
+ * Return code: 0, EINVAL, EACCES (if PHY_LOCK is not held)
+ */
+#define MC_CMD_START_BIST 0x25
+#define MC_CMD_START_BIST_IN_LEN 4
+#define MC_CMD_START_BIST_IN_TYPE_OFST 0
+#define MC_CMD_START_BIST_OUT_LEN 0
+
+/* Run the PHY's short cable BIST */
+#define MC_CMD_PHY_BIST_CABLE_SHORT  1
+/* Run the PHY's long cable BIST */
+#define MC_CMD_PHY_BIST_CABLE_LONG   2
+/* Run BIST on the currently selected BPX Serdes (XAUI or XFI) */
+#define MC_CMD_BPX_SERDES_BIST 3
+/* Run the MC loopback tests */
+#define MC_CMD_MC_LOOPBACK_BIST 4
+/* Run the PHY's standard BIST */
+#define MC_CMD_PHY_BIST 5
+
+/* MC_CMD_POLL_PHY_BIST: (variadic output)
+ * Poll for BIST completion
+ *
+ * Returns a single status code, and optionally some PHY specific
+ * bist output. The driver should only consume the BIST output
+ * after validating OUTLEN and PHY_CFG.PHY_TYPE.
+ *
+ * If a driver can't successfully parse the BIST output, it should
+ * still respect the pass/Fail in OUT.RESULT
+ *
+ * Locks required: PHY_LOCK if doing a  PHY BIST
+ * Return code: 0, EACCES (if PHY_LOCK is not held)
+ */
+#define MC_CMD_POLL_BIST 0x26
+#define MC_CMD_POLL_BIST_IN_LEN 0
+#define MC_CMD_POLL_BIST_OUT_LEN UNKNOWN
+#define MC_CMD_POLL_BIST_OUT_SFT9001_LEN 36
+#define MC_CMD_POLL_BIST_OUT_MRSFP_LEN 8
+#define MC_CMD_POLL_BIST_OUT_RESULT_OFST 0
+#define MC_CMD_POLL_BIST_RUNNING 1
+#define MC_CMD_POLL_BIST_PASSED 2
+#define MC_CMD_POLL_BIST_FAILED 3
+#define MC_CMD_POLL_BIST_TIMEOUT 4
+/* Generic: */
+#define MC_CMD_POLL_BIST_OUT_PRIVATE_OFST 4
+/* SFT9001-specific: */
+#define MC_CMD_POLL_BIST_OUT_SFT9001_CABLE_LENGTH_A_OFST 4
+#define MC_CMD_POLL_BIST_OUT_SFT9001_CABLE_LENGTH_B_OFST 8
+#define MC_CMD_POLL_BIST_OUT_SFT9001_CABLE_LENGTH_C_OFST 12
+#define MC_CMD_POLL_BIST_OUT_SFT9001_CABLE_LENGTH_D_OFST 16
+#define MC_CMD_POLL_BIST_OUT_SFT9001_CABLE_STATUS_A_OFST 20
+#define MC_CMD_POLL_BIST_OUT_SFT9001_CABLE_STATUS_B_OFST 24
+#define MC_CMD_POLL_BIST_OUT_SFT9001_CABLE_STATUS_C_OFST 28
+#define MC_CMD_POLL_BIST_OUT_SFT9001_CABLE_STATUS_D_OFST 32
+#define MC_CMD_POLL_BIST_SFT9001_PAIR_OK 1
+#define MC_CMD_POLL_BIST_SFT9001_PAIR_OPEN 2
+#define MC_CMD_POLL_BIST_SFT9001_INTRA_PAIR_SHORT 3
+#define MC_CMD_POLL_BIST_SFT9001_INTER_PAIR_SHORT 4
+#define MC_CMD_POLL_BIST_SFT9001_PAIR_BUSY 9
+/* mrsfp "PHY" driver: */
+#define MC_CMD_POLL_BIST_OUT_MRSFP_TEST_OFST 4
+#define MC_CMD_POLL_BIST_MRSFP_TEST_COMPLETE 0
+#define MC_CMD_POLL_BIST_MRSFP_TEST_BUS_SWITCH_OFF_I2C_WRITE 1
+#define MC_CMD_POLL_BIST_MRSFP_TEST_BUS_SWITCH_OFF_I2C_NO_ACCESS_IO_EXP 2
+#define MC_CMD_POLL_BIST_MRSFP_TEST_BUS_SWITCH_OFF_I2C_NO_ACCESS_MODULE 3
+#define MC_CMD_POLL_BIST_MRSFP_TEST_IO_EXP_I2C_CONFIGURE 4
+#define MC_CMD_POLL_BIST_MRSFP_TEST_BUS_SWITCH_I2C_NO_CROSSTALK 5
+#define MC_CMD_POLL_BIST_MRSFP_TEST_MODULE_PRESENCE 6
+#define MC_CMD_POLL_BIST_MRSFP_TEST_MODULE_ID_I2C_ACCESS 7
+#define MC_CMD_POLL_BIST_MRSFP_TEST_MODULE_ID_SANE_VALUE 8
+
+/* MC_CMD_PHY_SPI: (variadic in, variadic out)
+ * Read/Write/Erase the PHY SPI device
+ *
+ * Locks required: PHY_LOCK
+ * Return code: 0, ETIME, EINVAL, EACCES (if PHY_LOCK is not held)
+ */
+#define MC_CMD_PHY_SPI 0x27
+#define MC_CMD_PHY_SPI_IN_LEN(_write_bytes) (12 + (_write_bytes))
+#define MC_CMD_PHY_SPI_IN_ARGS_OFST 0
+#define MC_CMD_PHY_SPI_IN_ARGS_ADDR_OFST 0
+#define MC_CMD_PHY_SPI_IN_ARGS_READ_BYTES_OFST 4
+#define MC_CMD_PHY_SPI_IN_ARGS_ERASE_ALL_OFST 8
+/* Data to write here */
+#define MC_CMD_PHY_SPI_IN_WRITE_BUFFER_OFSET 12
+#define MC_CMD_PHY_SPI_OUT_LEN(_read_bytes) (_read_bytes)
+/* Data read here */
+#define MC_CMD_PHY_SPI_OUT_READ_BUFFER_OFST 0
+
+
+/* MC_CMD_GET_LOOPBACK_MODES:
+ * Returns a bitmask of loopback modes evailable at each speed.
+ *
+ * Locks required: None
+ * Return code: 0
+ */
+#define MC_CMD_GET_LOOPBACK_MODES 0x28
+#define MC_CMD_GET_LOOPBACK_MODES_IN_LEN 0
+#define MC_CMD_GET_LOOPBACK_MODES_OUT_LEN 32
+#define MC_CMD_GET_LOOPBACK_MODES_100M_OFST 0
+#define MC_CMD_GET_LOOPBACK_MODES_1G_OFST 8
+#define MC_CMD_GET_LOOPBACK_MODES_10G_OFST 16
+#define MC_CMD_GET_LOOPBACK_MODES_SUGGESTED_OFST 24
+
+/* Flow control enumeration */
+#define MC_CMD_FCNTL_OFF 0
+#define MC_CMD_FCNTL_RESPOND 1
+#define MC_CMD_FCNTL_BIDIR 2
+/* Auto - Use what the link has autonegotiated
+ *      - The driver should modify the advertised capabilities via SET_LINK.CAP
+ *        to control the negotiated flow control mode.
+ *      - Can only be set if the PHY supports PAUSE+ASYM capabilities
+ *      - Never returned by GET_LINK as the value programmed into the MAC
+ */
+#define MC_CMD_FCNTL_AUTO 3
+
+/* Generic mac fault bitmask */
+#define MC_CMD_MAC_FAULT_XGMII_LOCAL_LBN 0
+#define MC_CMD_MAC_FAULT_XGMII_LOCAL_WIDTH 1
+#define MC_CMD_MAC_FAULT_XGMII_REMOTE_LBN 1
+#define MC_CMD_MAC_FAULT_XGMII_REMOTE_WIDTH 1
+#define MC_CMD_MAC_FAULT_SGMII_REMOTE_LBN 2
+#define MC_CMD_MAC_FAULT_SGMII_REMOTE_WIDTH 1
+
+/* MC_CMD_GET_LINK:
+ * Read the unified MAC/PHY link state
+ *
+ * Locks required: None
+ * Return code: 0, ETIME
+ */
+#define MC_CMD_GET_LINK 0x29
+#define MC_CMD_GET_LINK_IN_LEN 0
+#define MC_CMD_GET_LINK_OUT_LEN 28
+/* near-side and link-partner advertised capabilities */
+#define MC_CMD_GET_LINK_OUT_CAP_OFST 0
+#define MC_CMD_GET_LINK_OUT_LP_CAP_OFST 4
+/* Autonegotiated speed in mbit/s. The link may still be down
+ * even if this reads non-zero */
+#define MC_CMD_GET_LINK_OUT_LINK_SPEED_OFST 8
+#define MC_CMD_GET_LINK_OUT_LOOPBACK_MODE_OFST 12
+#define MC_CMD_GET_LINK_OUT_FLAGS_OFST 16
+/* Whether we have overall link up */
+#define MC_CMD_GET_LINK_LINK_UP_LBN 0
+#define MC_CMD_GET_LINK_LINK_UP_WIDTH 1
+#define MC_CMD_GET_LINK_FULL_DUPLEX_LBN 1
+#define MC_CMD_GET_LINK_FULL_DUPLEX_WIDTH 1
+/* Whether we have link at the layers provided by the BPX */
+#define MC_CMD_GET_LINK_BPX_LINK_LBN 2
+#define MC_CMD_GET_LINK_BPX_LINK_WIDTH 1
+/* Whether the PHY has external link */
+#define MC_CMD_GET_LINK_PHY_LINK_LBN 3
+#define MC_CMD_GET_LINK_PHY_LINK_WIDTH 1
+#define MC_CMD_GET_LINK_OUT_FCNTL_OFST 20
+#define MC_CMD_GET_LINK_OUT_MAC_FAULT_OFST 24
+
+/* MC_CMD_SET_LINK:
+ * Write the unified MAC/PHY link configuration
+ *
+ * A loopback speed of "0" is supported, and means
+ * (choose any available speed)
+ *
+ * Locks required: None
+ * Return code: 0, EINVAL, ETIME
+ */
+#define MC_CMD_SET_LINK 0x2a
+#define MC_CMD_SET_LINK_IN_LEN 16
+#define MC_CMD_SET_LINK_IN_CAP_OFST 0
+#define MC_CMD_SET_LINK_IN_FLAGS_OFST 4
+#define MC_CMD_SET_LINK_LOWPOWER_LBN 0
+#define MC_CMD_SET_LINK_LOWPOWER_WIDTH 1
+#define MC_CMD_SET_LINK_POWEROFF_LBN 1
+#define MC_CMD_SET_LINK_POWEROFF_WIDTH 1
+#define MC_CMD_SET_LINK_TXDIS_LBN 2
+#define MC_CMD_SET_LINK_TXDIS_WIDTH 1
+#define MC_CMD_SET_LINK_IN_LOOPBACK_MODE_OFST 8
+#define MC_CMD_SET_LINK_IN_LOOPBACK_SPEED_OFST 12
+#define MC_CMD_SET_LINK_OUT_LEN 0
+
+/* MC_CMD_SET_ID_LED:
+ * Set indentification LED state
+ *
+ * Locks required: None
+ * Return code: 0, EINVAL
+ */
+#define MC_CMD_SET_ID_LED 0x2b
+#define MC_CMD_SET_ID_LED_IN_LEN 4
+#define MC_CMD_SET_ID_LED_IN_STATE_OFST 0
+#define  MC_CMD_LED_OFF 0
+#define  MC_CMD_LED_ON 1
+#define  MC_CMD_LED_DEFAULT 2
+#define MC_CMD_SET_ID_LED_OUT_LEN 0
+
+/* MC_CMD_SET_MAC:
+ * Set MAC configuration
+ *
+ * The MTU is the MTU programmed directly into the XMAC/GMAC
+ * (inclusive of EtherII, VLAN, bug16011 padding)
+ *
+ * Locks required: None
+ * Return code: 0, EINVAL
+ */
+#define MC_CMD_SET_MAC 0x2c
+#define MC_CMD_SET_MAC_IN_LEN 24
+#define MC_CMD_SET_MAC_IN_MTU_OFST 0
+#define MC_CMD_SET_MAC_IN_DRAIN_OFST 4
+#define MC_CMD_SET_MAC_IN_ADDR_OFST 8
+#define MC_CMD_SET_MAC_IN_REJECT_OFST 16
+#define MC_CMD_SET_MAC_IN_REJECT_UNCST_LBN 0
+#define MC_CMD_SET_MAC_IN_REJECT_UNCST_WIDTH 1
+#define MC_CMD_SET_MAC_IN_REJECT_BRDCST_LBN 1
+#define MC_CMD_SET_MAC_IN_REJECT_BRDCST_WIDTH 1
+#define MC_CMD_SET_MAC_IN_FCNTL_OFST 20
+#define MC_CMD_SET_MAC_OUT_LEN 0
+
+/* MC_CMD_PHY_STATS:
+ * Get generic PHY statistics
+ *
+ * This call returns the statistics for a generic PHY in a sparse
+ * array (indexed by the enumerate). Each value is represented by
+ * a 32bit number.
+ *
+ * If the DMA_ADDR is 0, then no DMA is performed, and the statistics
+ * may be read directly out of shared memory. If DMA_ADDR != 0, then
+ * the statistics are dmad to that (page-aligned location)
+ *
+ * Locks required: None
+ * Returns: 0, ETIME
+ * Response methods: shared memory, event
+ */
+#define MC_CMD_PHY_STATS 0x2d
+#define MC_CMD_PHY_STATS_IN_LEN 8
+#define MC_CMD_PHY_STATS_IN_DMA_ADDR_LO_OFST 0
+#define MC_CMD_PHY_STATS_IN_DMA_ADDR_HI_OFST 4
+#define MC_CMD_PHY_STATS_OUT_DMA_LEN 0
+#define MC_CMD_PHY_STATS_OUT_NO_DMA_LEN (MC_CMD_PHY_NSTATS * 4)
+
+/* Unified MAC statistics enumeration */
+#define MC_CMD_MAC_GENERATION_START 0
+#define MC_CMD_MAC_TX_PKTS 1
+#define MC_CMD_MAC_TX_PAUSE_PKTS 2
+#define MC_CMD_MAC_TX_CONTROL_PKTS 3
+#define MC_CMD_MAC_TX_UNICAST_PKTS 4
+#define MC_CMD_MAC_TX_MULTICAST_PKTS 5
+#define MC_CMD_MAC_TX_BROADCAST_PKTS 6
+#define MC_CMD_MAC_TX_BYTES 7
+#define MC_CMD_MAC_TX_BAD_BYTES 8
+#define MC_CMD_MAC_TX_LT64_PKTS 9
+#define MC_CMD_MAC_TX_64_PKTS 10
+#define MC_CMD_MAC_TX_65_TO_127_PKTS 11
+#define MC_CMD_MAC_TX_128_TO_255_PKTS 12
+#define MC_CMD_MAC_TX_256_TO_511_PKTS 13
+#define MC_CMD_MAC_TX_512_TO_1023_PKTS 14
+#define MC_CMD_MAC_TX_1024_TO_15XX_PKTS 15
+#define MC_CMD_MAC_TX_15XX_TO_JUMBO_PKTS 16
+#define MC_CMD_MAC_TX_GTJUMBO_PKTS 17
+#define MC_CMD_MAC_TX_BAD_FCS_PKTS 18
+#define MC_CMD_MAC_TX_SINGLE_COLLISION_PKTS 19
+#define MC_CMD_MAC_TX_MULTIPLE_COLLISION_PKTS 20
+#define MC_CMD_MAC_TX_EXCESSIVE_COLLISION_PKTS 21
+#define MC_CMD_MAC_TX_LATE_COLLISION_PKTS 22
+#define MC_CMD_MAC_TX_DEFERRED_PKTS 23
+#define MC_CMD_MAC_TX_EXCESSIVE_DEFERRED_PKTS 24
+#define MC_CMD_MAC_TX_NON_TCPUDP_PKTS 25
+#define MC_CMD_MAC_TX_MAC_SRC_ERR_PKTS 26
+#define MC_CMD_MAC_TX_IP_SRC_ERR_PKTS 27
+#define MC_CMD_MAC_RX_PKTS 28
+#define MC_CMD_MAC_RX_PAUSE_PKTS 29
+#define MC_CMD_MAC_RX_GOOD_PKTS 30
+#define MC_CMD_MAC_RX_CONTROL_PKTS 31
+#define MC_CMD_MAC_RX_UNICAST_PKTS 32
+#define MC_CMD_MAC_RX_MULTICAST_PKTS 33
+#define MC_CMD_MAC_RX_BROADCAST_PKTS 34
+#define MC_CMD_MAC_RX_BYTES 35
+#define MC_CMD_MAC_RX_BAD_BYTES 36
+#define MC_CMD_MAC_RX_64_PKTS 37
+#define MC_CMD_MAC_RX_65_TO_127_PKTS 38
+#define MC_CMD_MAC_RX_128_TO_255_PKTS 39
+#define MC_CMD_MAC_RX_256_TO_511_PKTS 40
+#define MC_CMD_MAC_RX_512_TO_1023_PKTS 41
+#define MC_CMD_MAC_RX_1024_TO_15XX_PKTS 42
+#define MC_CMD_MAC_RX_15XX_TO_JUMBO_PKTS 43
+#define MC_CMD_MAC_RX_GTJUMBO_PKTS 44
+#define MC_CMD_MAC_RX_UNDERSIZE_PKTS 45
+#define MC_CMD_MAC_RX_BAD_FCS_PKTS 46
+#define MC_CMD_MAC_RX_OVERFLOW_PKTS 47
+#define MC_CMD_MAC_RX_FALSE_CARRIER_PKTS 48
+#define MC_CMD_MAC_RX_SYMBOL_ERROR_PKTS 49
+#define MC_CMD_MAC_RX_ALIGN_ERROR_PKTS 50
+#define MC_CMD_MAC_RX_LENGTH_ERROR_PKTS 51
+#define MC_CMD_MAC_RX_INTERNAL_ERROR_PKTS 52
+#define MC_CMD_MAC_RX_JABBER_PKTS 53
+#define MC_CMD_MAC_RX_NODESC_DROPS 54
+#define MC_CMD_MAC_RX_LANES01_CHAR_ERR 55
+#define MC_CMD_MAC_RX_LANES23_CHAR_ERR 56
+#define MC_CMD_MAC_RX_LANES01_DISP_ERR 57
+#define MC_CMD_MAC_RX_LANES23_DISP_ERR 58
+#define MC_CMD_MAC_RX_MATCH_FAULT 59
+#define MC_CMD_GMAC_DMABUF_START 64
+#define MC_CMD_GMAC_DMABUF_END   95
+/* Insert new members here. */
+#define MC_CMD_MAC_GENERATION_END 96
+#define MC_CMD_MAC_NSTATS (MC_CMD_MAC_GENERATION_END+1)
+
+/* MC_CMD_MAC_STATS:
+ * Get unified GMAC/XMAC statistics
+ *
+ * This call returns unified statistics maintained by the MC as it
+ * switches between the GMAC and XMAC. The MC will write out all
+ * supported stats.  The driver should zero initialise the buffer to
+ * guarantee consistent results.
+ *
+ * Locks required: None
+ * Returns: 0
+ * Response methods: shared memory, event
+ */
+#define MC_CMD_MAC_STATS 0x2e
+#define MC_CMD_MAC_STATS_IN_LEN 16
+#define MC_CMD_MAC_STATS_IN_DMA_ADDR_LO_OFST 0
+#define MC_CMD_MAC_STATS_IN_DMA_ADDR_HI_OFST 4
+#define MC_CMD_MAC_STATS_IN_CMD_OFST 8
+#define MC_CMD_MAC_STATS_CMD_DMA_LBN 0
+#define MC_CMD_MAC_STATS_CMD_DMA_WIDTH 1
+#define MC_CMD_MAC_STATS_CMD_CLEAR_LBN 1
+#define MC_CMD_MAC_STATS_CMD_CLEAR_WIDTH 1
+#define MC_CMD_MAC_STATS_CMD_PERIODIC_CHANGE_LBN 2
+#define MC_CMD_MAC_STATS_CMD_PERIODIC_CHANGE_WIDTH 1
+/* Remaining PERIOD* fields only relevant when PERIODIC_CHANGE is set */
+#define MC_CMD_MAC_STATS_CMD_PERIODIC_ENABLE_LBN 3
+#define MC_CMD_MAC_STATS_CMD_PERIODIC_ENABLE_WIDTH 1
+#define MC_CMD_MAC_STATS_CMD_PERIODIC_CLEAR_LBN 4
+#define MC_CMD_MAC_STATS_CMD_PERIODIC_CLEAR_WIDTH 1
+#define MC_CMD_MAC_STATS_CMD_PERIODIC_NOEVENT_LBN 5
+#define MC_CMD_MAC_STATS_CMD_PERIODIC_NOEVENT_WIDTH 1
+#define MC_CMD_MAC_STATS_CMD_PERIOD_MS_LBN 16
+#define MC_CMD_MAC_STATS_CMD_PERIOD_MS_WIDTH 16
+#define MC_CMD_MAC_STATS_IN_DMA_LEN_OFST 12
+
+#define MC_CMD_MAC_STATS_OUT_LEN 0
+
+/* Callisto flags */
+#define MC_CMD_SFT9001_ROBUST_LBN 0
+#define MC_CMD_SFT9001_ROBUST_WIDTH 1
+#define MC_CMD_SFT9001_SHORT_REACH_LBN 1
+#define MC_CMD_SFT9001_SHORT_REACH_WIDTH 1
+
+/* MC_CMD_SFT9001_GET:
+ * Read current callisto specific setting
+ *
+ * Locks required: None
+ * Returns: 0, ETIME
+ */
+#define MC_CMD_SFT9001_GET 0x30
+#define MC_CMD_SFT9001_GET_IN_LEN 0
+#define MC_CMD_SFT9001_GET_OUT_LEN 4
+#define MC_CMD_SFT9001_GET_OUT_FLAGS_OFST 0
+
+/* MC_CMD_SFT9001_SET:
+ * Write current callisto specific setting
+ *
+ * Locks required: None
+ * Returns: 0, ETIME, EINVAL
+ */
+#define MC_CMD_SFT9001_SET 0x31
+#define MC_CMD_SFT9001_SET_IN_LEN 4
+#define MC_CMD_SFT9001_SET_IN_FLAGS_OFST 0
+#define MC_CMD_SFT9001_SET_OUT_LEN 0
+
+
+/* MC_CMD_WOL_FILTER_SET:
+ * Set a WoL filter
+ *
+ * Locks required: None
+ * Returns: 0, EBUSY, EINVAL, ENOSYS
+ */
+#define MC_CMD_WOL_FILTER_SET 0x32
+#define MC_CMD_WOL_FILTER_SET_IN_LEN 192 /* 190 rounded up to a word */
+#define MC_CMD_WOL_FILTER_SET_IN_FILTER_MODE_OFST 0
+#define MC_CMD_WOL_FILTER_SET_IN_WOL_TYPE_OFST 4
+
+/* There is a union at offset 8, following defines overlap due to
+ * this */
+#define MC_CMD_WOL_FILTER_SET_IN_DATA_OFST 8
+
+#define MC_CMD_WOL_FILTER_SET_IN_MAGIC_MAC_OFST		\
+	MC_CMD_WOL_FILTER_SET_IN_DATA_OFST
+
+#define MC_CMD_WOL_FILTER_SET_IN_IPV4_SYN_SRC_IP_OFST   \
+	MC_CMD_WOL_FILTER_SET_IN_DATA_OFST
+#define MC_CMD_WOL_FILTER_SET_IN_IPV4_SYN_DST_IP_OFST   \
+	(MC_CMD_WOL_FILTER_SET_IN_DATA_OFST + 4)
+#define MC_CMD_WOL_FILTER_SET_IN_IPV4_SYN_SRC_PORT_OFST \
+	(MC_CMD_WOL_FILTER_SET_IN_DATA_OFST + 8)
+#define MC_CMD_WOL_FILTER_SET_IN_IPV4_SYN_DST_PORT_OFST \
+	(MC_CMD_WOL_FILTER_SET_IN_DATA_OFST + 10)
+
+#define MC_CMD_WOL_FILTER_SET_IN_IPV6_SYN_SRC_IP_OFST   \
+	MC_CMD_WOL_FILTER_SET_IN_DATA_OFST
+#define MC_CMD_WOL_FILTER_SET_IN_IPV6_SYN_DST_IP_OFST   \
+	(MC_CMD_WOL_FILTER_SET_IN_DATA_OFST + 16)
+#define MC_CMD_WOL_FILTER_SET_IN_IPV6_SYN_SRC_PORT_OFST \
+	(MC_CMD_WOL_FILTER_SET_IN_DATA_OFST + 32)
+#define MC_CMD_WOL_FILTER_SET_IN_IPV6_SYN_DST_PORT_OFST \
+	(MC_CMD_WOL_FILTER_SET_IN_DATA_OFST + 34)
+
+#define MC_CMD_WOL_FILTER_SET_IN_BITMAP_MASK_OFST	\
+	MC_CMD_WOL_FILTER_SET_IN_DATA_OFST
+#define MC_CMD_WOL_FILTER_SET_IN_BITMAP_OFST		\
+	(MC_CMD_WOL_FILTER_SET_IN_DATA_OFST + 48)
+#define MC_CMD_WOL_FILTER_SET_IN_BITMAP_LEN_OFST	\
+	(MC_CMD_WOL_FILTER_SET_IN_DATA_OFST + 176)
+#define MC_CMD_WOL_FILTER_SET_IN_BITMAP_LAYER3_OFST	\
+	(MC_CMD_WOL_FILTER_SET_IN_DATA_OFST + 177)
+#define MC_CMD_WOL_FILTER_SET_IN_BITMAP_LAYER4_OFST	\
+	(MC_CMD_WOL_FILTER_SET_IN_DATA_OFST + 178)
+
+#define MC_CMD_WOL_FILTER_SET_IN_LINK_MASK_OFST	\
+	MC_CMD_WOL_FILTER_SET_IN_DATA_OFST
+#define MC_CMD_WOL_FILTER_SET_IN_LINK_UP_LBN	0
+#define MC_CMD_WOL_FILTER_SET_IN_LINK_UP_WIDTH	1
+#define MC_CMD_WOL_FILTER_SET_IN_LINK_DOWN_LBN	1
+#define MC_CMD_WOL_FILTER_SET_IN_LINK_DOWN_WIDTH 1
+
+#define MC_CMD_WOL_FILTER_SET_OUT_LEN 4
+#define MC_CMD_WOL_FILTER_SET_OUT_FILTER_ID_OFST 0
+
+/* WOL Filter types enumeration */
+#define MC_CMD_WOL_TYPE_MAGIC      0x0
+			 /* unused 0x1 */
+#define MC_CMD_WOL_TYPE_WIN_MAGIC  0x2
+#define MC_CMD_WOL_TYPE_IPV4_SYN   0x3
+#define MC_CMD_WOL_TYPE_IPV6_SYN   0x4
+#define MC_CMD_WOL_TYPE_BITMAP     0x5
+#define MC_CMD_WOL_TYPE_LINK       0x6
+#define MC_CMD_WOL_TYPE_MAX        0x7
+
+#define MC_CMD_FILTER_MODE_SIMPLE     0x0
+#define MC_CMD_FILTER_MODE_STRUCTURED 0xffffffff
+
+/* MC_CMD_WOL_FILTER_REMOVE:
+ * Remove a WoL filter
+ *
+ * Locks required: None
+ * Returns: 0, EINVAL, ENOSYS
+ */
+#define MC_CMD_WOL_FILTER_REMOVE 0x33
+#define MC_CMD_WOL_FILTER_REMOVE_IN_LEN 4
+#define MC_CMD_WOL_FILTER_REMOVE_IN_FILTER_ID_OFST 0
+#define MC_CMD_WOL_FILTER_REMOVE_OUT_LEN 0
+
+
+/* MC_CMD_WOL_FILTER_RESET:
+ * Reset (i.e. remove all) WoL filters
+ *
+ * Locks required: None
+ * Returns: 0, ENOSYS
+ */
+#define MC_CMD_WOL_FILTER_RESET 0x34
+#define MC_CMD_WOL_FILTER_RESET_IN_LEN 0
+#define MC_CMD_WOL_FILTER_RESET_OUT_LEN 0
+
+/* MC_CMD_SET_MCAST_HASH:
+ * Set the MCASH hash value without otherwise
+ * reconfiguring the MAC
+ */
+#define MC_CMD_SET_MCAST_HASH 0x35
+#define MC_CMD_SET_MCAST_HASH_IN_LEN 32
+#define MC_CMD_SET_MCAST_HASH_IN_HASH0_OFST 0
+#define MC_CMD_SET_MCAST_HASH_IN_HASH1_OFST 16
+#define MC_CMD_SET_MCAST_HASH_OUT_LEN 0
+
+/* MC_CMD_NVRAM_TYPES:
+ * Return bitfield indicating available types of virtual NVRAM partitions
+ *
+ * Locks required: none
+ * Returns: 0
+ */
+#define MC_CMD_NVRAM_TYPES 0x36
+#define MC_CMD_NVRAM_TYPES_IN_LEN 0
+#define MC_CMD_NVRAM_TYPES_OUT_LEN 4
+#define MC_CMD_NVRAM_TYPES_OUT_TYPES_OFST 0
+
+/* Supported NVRAM types */
+#define MC_CMD_NVRAM_TYPE_DISABLED_CALLISTO 0
+#define MC_CMD_NVRAM_TYPE_MC_FW 1
+#define MC_CMD_NVRAM_TYPE_MC_FW_BACKUP 2
+#define MC_CMD_NVRAM_TYPE_STATIC_CFG_PORT0 3
+#define MC_CMD_NVRAM_TYPE_STATIC_CFG_PORT1 4
+#define MC_CMD_NVRAM_TYPE_DYNAMIC_CFG_PORT0 5
+#define MC_CMD_NVRAM_TYPE_DYNAMIC_CFG_PORT1 6
+#define MC_CMD_NVRAM_TYPE_EXP_ROM 7
+#define MC_CMD_NVRAM_TYPE_EXP_ROM_CFG_PORT0 8
+#define MC_CMD_NVRAM_TYPE_EXP_ROM_CFG_PORT1 9
+#define MC_CMD_NVRAM_TYPE_PHY_PORT0 10
+#define MC_CMD_NVRAM_TYPE_PHY_PORT1 11
+#define MC_CMD_NVRAM_TYPE_LOG 12
+
+/* MC_CMD_NVRAM_INFO:
+ * Read info about a virtual NVRAM partition
+ *
+ * Locks required: none
+ * Returns: 0, EINVAL (bad type)
+ */
+#define MC_CMD_NVRAM_INFO 0x37
+#define MC_CMD_NVRAM_INFO_IN_LEN 4
+#define MC_CMD_NVRAM_INFO_IN_TYPE_OFST 0
+#define MC_CMD_NVRAM_INFO_OUT_LEN 24
+#define MC_CMD_NVRAM_INFO_OUT_TYPE_OFST 0
+#define MC_CMD_NVRAM_INFO_OUT_SIZE_OFST 4
+#define MC_CMD_NVRAM_INFO_OUT_ERASESIZE_OFST 8
+#define MC_CMD_NVRAM_INFO_OUT_FLAGS_OFST 12
+#define   MC_CMD_NVRAM_PROTECTED_LBN 0
+#define   MC_CMD_NVRAM_PROTECTED_WIDTH 1
+#define MC_CMD_NVRAM_INFO_OUT_PHYSDEV_OFST 16
+#define MC_CMD_NVRAM_INFO_OUT_PHYSADDR_OFST 20
+
+/* MC_CMD_NVRAM_UPDATE_START:
+ * Start a group of update operations on a virtual NVRAM partition
+ *
+ * Locks required: PHY_LOCK if type==*PHY*
+ * Returns: 0, EINVAL (bad type), EACCES (if PHY_LOCK required and not held)
+ */
+#define MC_CMD_NVRAM_UPDATE_START 0x38
+#define MC_CMD_NVRAM_UPDATE_START_IN_LEN 4
+#define MC_CMD_NVRAM_UPDATE_START_IN_TYPE_OFST 0
+#define MC_CMD_NVRAM_UPDATE_START_OUT_LEN 0
+
+/* MC_CMD_NVRAM_READ:
+ * Read data from a virtual NVRAM partition
+ *
+ * Locks required: PHY_LOCK if type==*PHY*
+ * Returns: 0, EINVAL (bad type/offset/length), EACCES (if PHY_LOCK required and not held)
+ */
+#define MC_CMD_NVRAM_READ 0x39
+#define MC_CMD_NVRAM_READ_IN_LEN 12
+#define MC_CMD_NVRAM_READ_IN_TYPE_OFST 0
+#define MC_CMD_NVRAM_READ_IN_OFFSET_OFST 4
+#define MC_CMD_NVRAM_READ_IN_LENGTH_OFST 8
+#define MC_CMD_NVRAM_READ_OUT_LEN(_read_bytes) (_read_bytes)
+#define MC_CMD_NVRAM_READ_OUT_READ_BUFFER_OFST 0
+
+/* MC_CMD_NVRAM_WRITE:
+ * Write data to a virtual NVRAM partition
+ *
+ * Locks required: PHY_LOCK if type==*PHY*
+ * Returns: 0, EINVAL (bad type/offset/length), EACCES (if PHY_LOCK required and not held)
+ */
+#define MC_CMD_NVRAM_WRITE 0x3a
+#define MC_CMD_NVRAM_WRITE_IN_TYPE_OFST 0
+#define MC_CMD_NVRAM_WRITE_IN_OFFSET_OFST 4
+#define MC_CMD_NVRAM_WRITE_IN_LENGTH_OFST 8
+#define MC_CMD_NVRAM_WRITE_IN_WRITE_BUFFER_OFST 12
+#define MC_CMD_NVRAM_WRITE_IN_LEN(_write_bytes) (12 + _write_bytes)
+#define MC_CMD_NVRAM_WRITE_OUT_LEN 0
+
+/* MC_CMD_NVRAM_ERASE:
+ * Erase sector(s) from a virtual NVRAM partition
+ *
+ * Locks required: PHY_LOCK if type==*PHY*
+ * Returns: 0, EINVAL (bad type/offset/length), EACCES (if PHY_LOCK required and not held)
+ */
+#define MC_CMD_NVRAM_ERASE 0x3b
+#define MC_CMD_NVRAM_ERASE_IN_LEN 12
+#define MC_CMD_NVRAM_ERASE_IN_TYPE_OFST 0
+#define MC_CMD_NVRAM_ERASE_IN_OFFSET_OFST 4
+#define MC_CMD_NVRAM_ERASE_IN_LENGTH_OFST 8
+#define MC_CMD_NVRAM_ERASE_OUT_LEN 0
+
+/* MC_CMD_NVRAM_UPDATE_FINISH:
+ * Finish a group of update operations on a virtual NVRAM partition
+ *
+ * Locks required: PHY_LOCK if type==*PHY*
+ * Returns: 0, EINVAL (bad type/offset/length), EACCES (if PHY_LOCK required and not held)
+ */
+#define MC_CMD_NVRAM_UPDATE_FINISH 0x3c
+#define MC_CMD_NVRAM_UPDATE_FINISH_IN_LEN 8
+#define MC_CMD_NVRAM_UPDATE_FINISH_IN_TYPE_OFST 0
+#define MC_CMD_NVRAM_UPDATE_FINISH_IN_REBOOT_OFST 4
+#define MC_CMD_NVRAM_UPDATE_FINISH_OUT_LEN 0
+
+/* MC_CMD_REBOOT:
+ * Reboot the MC.
+ *
+ * The AFTER_ASSERTION flag is intended to be used when the driver notices
+ * an assertion failure (at which point it is expected to perform a complete
+ * tear down and reinitialise), to allow both ports to reset the MC once
+ * in an atomic fashion.
+ *
+ * Production mc firmwares are generally compiled with REBOOT_ON_ASSERT=1,
+ * which means that they will automatically reboot out of the assertion
+ * handler, so this is in practise an optional operation. It is still
+ * recommended that drivers execute this to support custom firmwares
+ * with REBOOT_ON_ASSERT=0.
+ *
+ * Locks required: NONE
+ * Returns: Nothing. You get back a response with ERR=1, DATALEN=0
+ */
+#define MC_CMD_REBOOT 0x3d
+#define MC_CMD_REBOOT_IN_LEN 4
+#define MC_CMD_REBOOT_IN_FLAGS_OFST 0
+#define MC_CMD_REBOOT_FLAGS_AFTER_ASSERTION 1
+#define MC_CMD_REBOOT_OUT_LEN 0
+
+/* MC_CMD_SCHEDINFO:
+ * Request scheduler info. from the MC.
+ *
+ * Locks required: NONE
+ * Returns: An array of (timeslice,maximum overrun), one for each thread,
+ * in ascending order of thread address.s
+ */
+#define MC_CMD_SCHEDINFO 0x3e
+#define MC_CMD_SCHEDINFO_IN_LEN 0
+
+
+/* MC_CMD_SET_REBOOT_MODE: (debug)
+ * Set the mode for the next MC reboot.
+ *
+ * Locks required: NONE
+ *
+ * Sets the reboot mode to the specified value.  Returns the old mode.
+ */
+#define MC_CMD_REBOOT_MODE 0x3f
+#define MC_CMD_REBOOT_MODE_IN_LEN 4
+#define MC_CMD_REBOOT_MODE_IN_VALUE_OFST 0
+#define MC_CMD_REBOOT_MODE_OUT_LEN 4
+#define MC_CMD_REBOOT_MODE_OUT_VALUE_OFST 0
+#define   MC_CMD_REBOOT_MODE_NORMAL 0
+#define   MC_CMD_REBOOT_MODE_SNAPPER 3
+
+/* MC_CMD_DEBUG_LOG:
+ * Null request/response command (debug)
+ * - sequence number is always zero
+ * - only supported on the UART interface
+ * (the same set of bytes is delivered as an
+ * event over PCI)
+ */
+#define MC_CMD_DEBUG_LOG 0x40
+#define MC_CMD_DEBUG_LOG_IN_LEN 0
+#define MC_CMD_DEBUG_LOG_OUT_LEN 0
+
+/* Generic sensor enumeration. Note that a dual port NIC
+ * will EITHER expose PHY_COMMON_TEMP OR PHY0_TEMP and
+ * PHY1_TEMP depending on whether there is a single sensor
+ * in the vicinity of the two port, or one per port.
+ */
+#define MC_CMD_SENSOR_CONTROLLER_TEMP 0		/* degC */
+#define MC_CMD_SENSOR_PHY_COMMON_TEMP 1		/* degC */
+#define MC_CMD_SENSOR_CONTROLLER_COOLING 2	/* bool */
+#define MC_CMD_SENSOR_PHY0_TEMP 3		/* degC */
+#define MC_CMD_SENSOR_PHY0_COOLING 4		/* bool */
+#define MC_CMD_SENSOR_PHY1_TEMP 5		/* degC */
+#define MC_CMD_SENSOR_PHY1_COOLING 6		/* bool */
+#define MC_CMD_SENSOR_IN_1V0 7			/* mV */
+#define MC_CMD_SENSOR_IN_1V2 8			/* mV */
+#define MC_CMD_SENSOR_IN_1V8 9			/* mV */
+#define MC_CMD_SENSOR_IN_2V5 10			/* mV */
+#define MC_CMD_SENSOR_IN_3V3 11			/* mV */
+#define MC_CMD_SENSOR_IN_12V0 12		/* mV */
+
+
+/* Sensor state */
+#define MC_CMD_SENSOR_STATE_OK 0
+#define MC_CMD_SENSOR_STATE_WARNING 1
+#define MC_CMD_SENSOR_STATE_FATAL 2
+#define MC_CMD_SENSOR_STATE_BROKEN 3
+
+/* MC_CMD_SENSOR_INFO:
+ * Returns information about every available sensor.
+ *
+ * Each sensor has a single (16bit) value, and a corresponding state.
+ * The mapping between value and sensor is nominally determined by the
+ * MC, but in practise is implemented as zero (BROKEN), one (TEMPERATURE),
+ * or two (VOLTAGE) ranges per sensor per state.
+ *
+ * This call returns a mask (32bit) of the sensors that are supported
+ * by this platform, then an array (indexed by MC_CMD_SENSOR) of byte
+ * offsets to the per-sensor arrays. Each sensor array has four 16bit
+ * numbers, min1, max1, min2, max2.
+ *
+ * Locks required: None
+ * Returns: 0
+ */
+#define MC_CMD_SENSOR_INFO 0x41
+#define MC_CMD_SENSOR_INFO_IN_LEN 0
+#define MC_CMD_SENSOR_INFO_OUT_MASK_OFST 0
+#define MC_CMD_SENSOR_INFO_OUT_OFFSET_OFST(_x) \
+	(4 + (_x))
+#define MC_CMD_SENSOR_INFO_OUT_MIN1_OFST(_ofst) \
+	((_ofst) + 0)
+#define MC_CMD_SENSOR_INFO_OUT_MAX1_OFST(_ofst) \
+	((_ofst) + 2)
+#define MC_CMD_SENSOR_INFO_OUT_MIN2_OFST(_ofst) \
+	((_ofst) + 4)
+#define MC_CMD_SENSOR_INFO_OUT_MAX2_OFST(_ofst) \
+	((_ofst) + 6)
+
+/* MC_CMD_READ_SENSORS
+ * Returns the current reading from each sensor
+ *
+ * Returns a sparse array of sensor readings (indexed by the sensor
+ * type) into host memory.  Each array element is a dword.
+ *
+ * The MC will send a SENSOREVT event every time any sensor changes state. The
+ * driver is responsible for ensuring that it doesn't miss any events. The board
+ * will function normally if all sensors are in STATE_OK or state_WARNING.
+ * Otherwise the board should not be expected to function.
+ */
+#define MC_CMD_READ_SENSORS 0x42
+#define MC_CMD_READ_SENSORS_IN_LEN 8
+#define MC_CMD_READ_SENSORS_IN_DMA_ADDR_LO_OFST 0
+#define MC_CMD_READ_SENSORS_IN_DMA_ADDR_HI_OFST 4
+#define MC_CMD_READ_SENSORS_OUT_LEN 0
+
+/* Sensor reading fields */
+#define MC_CMD_READ_SENSOR_VALUE_LBN 0
+#define MC_CMD_READ_SENSOR_VALUE_WIDTH 16
+#define MC_CMD_READ_SENSOR_STATE_LBN 16
+#define MC_CMD_READ_SENSOR_STATE_WIDTH 8
+
+
+/* MC_CMD_GET_PHY_STATE:
+ * Report current state of PHY.  A "zombie" PHY is a PHY that has failed to
+ * boot (e.g. due to missing or corrupted firmware).
+ *
+ * Locks required: None
+ * Return code: 0
+ */
+#define MC_CMD_GET_PHY_STATE 0x43
+
+#define MC_CMD_GET_PHY_STATE_IN_LEN 0
+#define MC_CMD_GET_PHY_STATE_OUT_LEN 4
+#define MC_CMD_GET_PHY_STATE_STATE_OFST 0
+/* PHY state enumeration: */
+#define MC_CMD_PHY_STATE_OK 1
+#define MC_CMD_PHY_STATE_ZOMBIE 2
+
+
+/* 802.1Qbb control. 8 Tx queues that map to priorities 0 - 7. Use all 1s to
+ * disable 802.Qbb for a given priority. */
+#define MC_CMD_SETUP_8021QBB 0x44
+#define MC_CMD_SETUP_8021QBB_IN_LEN 32
+#define MC_CMD_SETUP_8021QBB_OUT_LEN 0
+#define MC_CMD_SETUP_8021QBB_IN_TXQS_OFFST 0
+
+
+/* MC_CMD_WOL_FILTER_GET:
+ * Retrieve ID of any WoL filters
+ *
+ * Locks required: None
+ * Returns: 0, ENOSYS
+ */
+#define MC_CMD_WOL_FILTER_GET 0x45
+#define MC_CMD_WOL_FILTER_GET_IN_LEN 0
+#define MC_CMD_WOL_FILTER_GET_OUT_LEN 4
+#define MC_CMD_WOL_FILTER_GET_OUT_FILTER_ID_OFST 0
+
+
+/* MC_CMD_ADD_LIGHTSOUT_OFFLOAD:
+ * Offload a protocol to NIC for lights-out state
+ *
+ * Locks required: None
+ * Returns: 0, ENOSYS
+ */
+#define MC_CMD_ADD_LIGHTSOUT_OFFLOAD 0x46
+
+#define MC_CMD_ADD_LIGHTSOUT_OFFLOAD_IN_LEN 16
+#define MC_CMD_ADD_LIGHTSOUT_OFFLOAD_IN_PROTOCOL_OFST 0
+
+/* There is a union at offset 4, following defines overlap due to
+ * this */
+#define MC_CMD_ADD_LIGHTSOUT_OFFLOAD_IN_DATA_OFST 4
+#define MC_CMD_ADD_LIGHTSOUT_OFFLOAD_IN_ARPMAC_OFST 4
+#define MC_CMD_ADD_LIGHTSOUT_OFFLOAD_IN_ARPIP_OFST 10
+#define MC_CMD_ADD_LIGHTSOUT_OFFLOAD_IN_NSMAC_OFST 4
+#define MC_CMD_ADD_LIGHTSOUT_OFFLOAD_IN_NSSNIPV6_OFST 10
+#define MC_CMD_ADD_LIGHTSOUT_OFFLOAD_IN_NSIPV6_OFST 26
+
+#define MC_CMD_ADD_LIGHTSOUT_OFFLOAD_OUT_LEN 4
+#define MC_CMD_ADD_LIGHTSOUT_OFFLOAD_OUT_FILTER_ID_OFST 0
+
+
+/* MC_CMD_REMOVE_LIGHTSOUT_PROTOCOL_OFFLOAD:
+ * Offload a protocol to NIC for lights-out state
+ *
+ * Locks required: None
+ * Returns: 0, ENOSYS
+ */
+#define MC_CMD_REMOVE_LIGHTSOUT_OFFLOAD 0x47
+#define MC_CMD_REMOVE_LIGHTSOUT_OFFLOAD_IN_LEN 8
+#define MC_CMD_REMOVE_LIGHTSOUT_OFFLOAD_OUT_LEN 0
+
+#define MC_CMD_REMOVE_LIGHTSOUT_OFFLOAD_IN_PROTOCOL_OFST 0
+#define MC_CMD_REMOVE_LIGHTSOUT_OFFLOAD_IN_FILTER_ID_OFST 4
+
+/* Lights-out offload protocols enumeration */
+#define MC_CMD_LIGHTSOUT_OFFLOAD_PROTOCOL_ARP 0x1
+#define MC_CMD_LIGHTSOUT_OFFLOAD_PROTOCOL_NS  0x2
+
+
+/* MC_CMD_MAC_RESET_RESTORE:
+ * Restore MAC after block reset
+ *
+ * Locks required: None
+ * Returns: 0
+ */
+
+#define MC_CMD_MAC_RESET_RESTORE 0x48
+#define MC_CMD_MAC_RESET_RESTORE_IN_LEN 0
+#define MC_CMD_MAC_RESET_RESTORE_OUT_LEN 0
+
+
+/* MC_CMD_TEST_ASSERT:
+ * Deliberately trigger an assert-detonation in the firmware for testing
+ * purposes (i.e. to allow tests that the driver copes gracefully).
+ *
+ * Locks required: None
+ * Returns: 0
+ */
+
+#define MC_CMD_TESTASSERT 0x49
+#define MC_CMD_TESTASSERT_IN_LEN 0
+#define MC_CMD_TESTASSERT_OUT_LEN 0
+
+/* MC_CMD_WORKAROUND 0x4a
+ *
+ * Enable/Disable a given workaround. The mcfw will return EINVAL if it
+ * doesn't understand the given workaround number - which should not
+ * be treated as a hard error by client code.
+ *
+ * This op does not imply any semantics about each workaround, that's between
+ * the driver and the mcfw on a per-workaround basis.
+ *
+ * Locks required: None
+ * Returns: 0, EINVAL
+ */
+#define MC_CMD_WORKAROUND 0x4a
+#define MC_CMD_WORKAROUND_IN_LEN 8
+#define MC_CMD_WORKAROUND_IN_TYPE_OFST 0
+#define MC_CMD_WORKAROUND_BUG17230 1
+#define MC_CMD_WORKAROUND_IN_ENABLED_OFST 4
+#define MC_CMD_WORKAROUND_OUT_LEN 0
+
+/* MC_CMD_GET_PHY_MEDIA_INFO:
+ * Read media-specific data from PHY (e.g. SFP/SFP+ module ID information for
+ * SFP+ PHYs).
+ *
+ * The "media type" can be found via GET_PHY_CFG (GET_PHY_CFG_OUT_MEDIA_TYPE);
+ * the valid "page number" input values, and the output data, are interpreted
+ * on a per-type basis.
+ *
+ * For SFP+: PAGE=0 or 1 returns a 128-byte block read from module I2C address
+ *           0xA0 offset 0 or 0x80.
+ * Anything else: currently undefined.
+ *
+ * Locks required: None
+ * Return code: 0
+ */
+#define MC_CMD_GET_PHY_MEDIA_INFO 0x4b
+#define MC_CMD_GET_PHY_MEDIA_INFO_IN_LEN 4
+#define MC_CMD_GET_PHY_MEDIA_INFO_IN_PAGE_OFST 0
+#define MC_CMD_GET_PHY_MEDIA_INFO_OUT_LEN(_num_bytes) (4 + (_num_bytes))
+#define MC_CMD_GET_PHY_MEDIA_INFO_OUT_DATALEN_OFST 0
+#define MC_CMD_GET_PHY_MEDIA_INFO_OUT_DATA_OFST 4
+
+/* MC_CMD_NVRAM_TEST:
+ * Test a particular NVRAM partition for valid contents (where "valid"
+ * depends on the type of partition).
+ *
+ * Locks required: None
+ * Return code: 0
+ */
+#define MC_CMD_NVRAM_TEST 0x4c
+#define MC_CMD_NVRAM_TEST_IN_LEN 4
+#define MC_CMD_NVRAM_TEST_IN_TYPE_OFST 0
+#define MC_CMD_NVRAM_TEST_OUT_LEN 4
+#define MC_CMD_NVRAM_TEST_OUT_RESULT_OFST 0
+#define MC_CMD_NVRAM_TEST_PASS 0
+#define MC_CMD_NVRAM_TEST_FAIL 1
+#define MC_CMD_NVRAM_TEST_NOTSUPP 2
+
+/* MC_CMD_MRSFP_TWEAK: (debug)
+ * Read status and/or set parameters for the "mrsfp" driver in mr_rusty builds.
+ * I2C I/O expander bits are always read; if equaliser parameters are supplied,
+ * they are configured first.
+ *
+ * Locks required: None
+ * Return code: 0, EINVAL
+ */
+#define MC_CMD_MRSFP_TWEAK 0x4d
+#define MC_CMD_MRSFP_TWEAK_IN_LEN_READ_ONLY 0
+#define MC_CMD_MRSFP_TWEAK_IN_LEN_EQ_CONFIG 16
+#define MC_CMD_MRSFP_TWEAK_IN_TXEQ_LEVEL_OFST 0    /* 0-6 low->high de-emph. */
+#define MC_CMD_MRSFP_TWEAK_IN_TXEQ_DT_CFG_OFST 4   /* 0-8 low->high ref.V */
+#define MC_CMD_MRSFP_TWEAK_IN_RXEQ_BOOST_OFST 8    /* 0-8 low->high boost */
+#define MC_CMD_MRSFP_TWEAK_IN_RXEQ_DT_CFG_OFST 12  /* 0-8 low->high ref.V */
+#define MC_CMD_MRSFP_TWEAK_OUT_LEN 12
+#define MC_CMD_MRSFP_TWEAK_OUT_IOEXP_INPUTS_OFST 0     /* input bits */
+#define MC_CMD_MRSFP_TWEAK_OUT_IOEXP_OUTPUTS_OFST 4    /* output bits */
+#define MC_CMD_MRSFP_TWEAK_OUT_IOEXP_DIRECTION_OFST 8  /* dirs: 0=out, 1=in */
+
+/* MC_CMD_TEST_HACK: (debug (unsurprisingly))
+  * Change bits of network port state for test purposes in ways that would never be
+  * useful in normal operation and so need a special command to change. */
+#define MC_CMD_TEST_HACK 0x2f
+#define MC_CMD_TEST_HACK_IN_LEN 8
+#define MC_CMD_TEST_HACK_IN_TXPAD_OFST 0
+#define   MC_CMD_TEST_HACK_IN_TXPAD_AUTO  0 /* Let the MC manage things */
+#define   MC_CMD_TEST_HACK_IN_TXPAD_ON    1 /* Force on */
+#define   MC_CMD_TEST_HACK_IN_TXPAD_OFF   2 /* Force on */
+#define MC_CMD_TEST_HACK_IN_IPG_OFST   4 /* Takes a value in bits */
+#define   MC_CMD_TEST_HACK_IN_IPG_AUTO    0 /* The MC picks the value */
+#define MC_CMD_TEST_HACK_OUT_LEN 0
+
+/* MC_CMD_SENSOR_SET_LIMS: (debug) (mostly) adjust the sensor limits. This
+ * is a warranty-voiding operation.
+  *
+ * IN: sensor identifier (one of the enumeration starting with MC_CMD_SENSOR_CONTROLLER_TEMP
+ * followed by 4 32-bit values: min(warning) max(warning), min(fatal), max(fatal). Which
+ * of these limits are meaningful and what their interpretation is is sensor-specific.
+ *
+ * OUT: nothing
+ *
+ * Returns: ENOENT if the sensor specified does not exist, EINVAL if the limits are
+  * out of range.
+ */
+#define MC_CMD_SENSOR_SET_LIMS 0x4e
+#define MC_CMD_SENSOR_SET_LIMS_IN_LEN 20
+#define MC_CMD_SENSOR_SET_LIMS_IN_SENSOR_OFST 0
+#define MC_CMD_SENSOR_SET_LIMS_IN_LOW0_OFST 4
+#define MC_CMD_SENSOR_SET_LIMS_IN_HI0_OFST  8
+#define MC_CMD_SENSOR_SET_LIMS_IN_LOW1_OFST 12
+#define MC_CMD_SENSOR_SET_LIMS_IN_HI1_OFST  16
+
+/* Do NOT add new commands beyond 0x4f as part of 3.0 : 0x50 - 0x7f will be
+ * used for post-3.0 extensions. If you run out of space, look for gaps or
+ * commands that are unused in the existing range. */
+
+#endif /* MCDI_PCOL_H */
diff --git a/drivers/net/ethernet/sfc/mcdi_phy.c b/drivers/net/ethernet/sfc/mcdi_phy.c
new file mode 100644
index 0000000..6c63ab0
--- /dev/null
+++ b/drivers/net/ethernet/sfc/mcdi_phy.c
@@ -0,0 +1,754 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2009-2010 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+/*
+ * Driver for PHY related operations via MCDI.
+ */
+
+#include <linux/slab.h>
+#include "efx.h"
+#include "phy.h"
+#include "mcdi.h"
+#include "mcdi_pcol.h"
+#include "nic.h"
+#include "selftest.h"
+
+struct efx_mcdi_phy_data {
+	u32 flags;
+	u32 type;
+	u32 supported_cap;
+	u32 channel;
+	u32 port;
+	u32 stats_mask;
+	u8 name[20];
+	u32 media;
+	u32 mmd_mask;
+	u8 revision[20];
+	u32 forced_cap;
+};
+
+static int
+efx_mcdi_get_phy_cfg(struct efx_nic *efx, struct efx_mcdi_phy_data *cfg)
+{
+	u8 outbuf[MC_CMD_GET_PHY_CFG_OUT_LEN];
+	size_t outlen;
+	int rc;
+
+	BUILD_BUG_ON(MC_CMD_GET_PHY_CFG_IN_LEN != 0);
+	BUILD_BUG_ON(MC_CMD_GET_PHY_CFG_OUT_NAME_LEN != sizeof(cfg->name));
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_GET_PHY_CFG, NULL, 0,
+			  outbuf, sizeof(outbuf), &outlen);
+	if (rc)
+		goto fail;
+
+	if (outlen < MC_CMD_GET_PHY_CFG_OUT_LEN) {
+		rc = -EIO;
+		goto fail;
+	}
+
+	cfg->flags = MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_FLAGS);
+	cfg->type = MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_TYPE);
+	cfg->supported_cap =
+		MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_SUPPORTED_CAP);
+	cfg->channel = MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_CHANNEL);
+	cfg->port = MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_PRT);
+	cfg->stats_mask = MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_STATS_MASK);
+	memcpy(cfg->name, MCDI_PTR(outbuf, GET_PHY_CFG_OUT_NAME),
+	       sizeof(cfg->name));
+	cfg->media = MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_MEDIA_TYPE);
+	cfg->mmd_mask = MCDI_DWORD(outbuf, GET_PHY_CFG_OUT_MMD_MASK);
+	memcpy(cfg->revision, MCDI_PTR(outbuf, GET_PHY_CFG_OUT_REVISION),
+	       sizeof(cfg->revision));
+
+	return 0;
+
+fail:
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	return rc;
+}
+
+static int efx_mcdi_set_link(struct efx_nic *efx, u32 capabilities,
+			     u32 flags, u32 loopback_mode,
+			     u32 loopback_speed)
+{
+	u8 inbuf[MC_CMD_SET_LINK_IN_LEN];
+	int rc;
+
+	BUILD_BUG_ON(MC_CMD_SET_LINK_OUT_LEN != 0);
+
+	MCDI_SET_DWORD(inbuf, SET_LINK_IN_CAP, capabilities);
+	MCDI_SET_DWORD(inbuf, SET_LINK_IN_FLAGS, flags);
+	MCDI_SET_DWORD(inbuf, SET_LINK_IN_LOOPBACK_MODE, loopback_mode);
+	MCDI_SET_DWORD(inbuf, SET_LINK_IN_LOOPBACK_SPEED, loopback_speed);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_SET_LINK, inbuf, sizeof(inbuf),
+			  NULL, 0, NULL);
+	if (rc)
+		goto fail;
+
+	return 0;
+
+fail:
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	return rc;
+}
+
+static int efx_mcdi_loopback_modes(struct efx_nic *efx, u64 *loopback_modes)
+{
+	u8 outbuf[MC_CMD_GET_LOOPBACK_MODES_OUT_LEN];
+	size_t outlen;
+	int rc;
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_GET_LOOPBACK_MODES, NULL, 0,
+			  outbuf, sizeof(outbuf), &outlen);
+	if (rc)
+		goto fail;
+
+	if (outlen < MC_CMD_GET_LOOPBACK_MODES_OUT_LEN) {
+		rc = -EIO;
+		goto fail;
+	}
+
+	*loopback_modes = MCDI_QWORD(outbuf, GET_LOOPBACK_MODES_SUGGESTED);
+
+	return 0;
+
+fail:
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	return rc;
+}
+
+int efx_mcdi_mdio_read(struct efx_nic *efx, unsigned int bus,
+			 unsigned int prtad, unsigned int devad, u16 addr,
+			 u16 *value_out, u32 *status_out)
+{
+	u8 inbuf[MC_CMD_MDIO_READ_IN_LEN];
+	u8 outbuf[MC_CMD_MDIO_READ_OUT_LEN];
+	size_t outlen;
+	int rc;
+
+	MCDI_SET_DWORD(inbuf, MDIO_READ_IN_BUS, bus);
+	MCDI_SET_DWORD(inbuf, MDIO_READ_IN_PRTAD, prtad);
+	MCDI_SET_DWORD(inbuf, MDIO_READ_IN_DEVAD, devad);
+	MCDI_SET_DWORD(inbuf, MDIO_READ_IN_ADDR, addr);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_MDIO_READ, inbuf, sizeof(inbuf),
+			  outbuf, sizeof(outbuf), &outlen);
+	if (rc)
+		goto fail;
+
+	*value_out = (u16)MCDI_DWORD(outbuf, MDIO_READ_OUT_VALUE);
+	*status_out = MCDI_DWORD(outbuf, MDIO_READ_OUT_STATUS);
+	return 0;
+
+fail:
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	return rc;
+}
+
+int efx_mcdi_mdio_write(struct efx_nic *efx, unsigned int bus,
+			  unsigned int prtad, unsigned int devad, u16 addr,
+			  u16 value, u32 *status_out)
+{
+	u8 inbuf[MC_CMD_MDIO_WRITE_IN_LEN];
+	u8 outbuf[MC_CMD_MDIO_WRITE_OUT_LEN];
+	size_t outlen;
+	int rc;
+
+	MCDI_SET_DWORD(inbuf, MDIO_WRITE_IN_BUS, bus);
+	MCDI_SET_DWORD(inbuf, MDIO_WRITE_IN_PRTAD, prtad);
+	MCDI_SET_DWORD(inbuf, MDIO_WRITE_IN_DEVAD, devad);
+	MCDI_SET_DWORD(inbuf, MDIO_WRITE_IN_ADDR, addr);
+	MCDI_SET_DWORD(inbuf, MDIO_WRITE_IN_VALUE, value);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_MDIO_WRITE, inbuf, sizeof(inbuf),
+			  outbuf, sizeof(outbuf), &outlen);
+	if (rc)
+		goto fail;
+
+	*status_out = MCDI_DWORD(outbuf, MDIO_WRITE_OUT_STATUS);
+	return 0;
+
+fail:
+	netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+	return rc;
+}
+
+static u32 mcdi_to_ethtool_cap(u32 media, u32 cap)
+{
+	u32 result = 0;
+
+	switch (media) {
+	case MC_CMD_MEDIA_KX4:
+		result |= SUPPORTED_Backplane;
+		if (cap & (1 << MC_CMD_PHY_CAP_1000FDX_LBN))
+			result |= SUPPORTED_1000baseKX_Full;
+		if (cap & (1 << MC_CMD_PHY_CAP_10000FDX_LBN))
+			result |= SUPPORTED_10000baseKX4_Full;
+		break;
+
+	case MC_CMD_MEDIA_XFP:
+	case MC_CMD_MEDIA_SFP_PLUS:
+		result |= SUPPORTED_FIBRE;
+		break;
+
+	case MC_CMD_MEDIA_BASE_T:
+		result |= SUPPORTED_TP;
+		if (cap & (1 << MC_CMD_PHY_CAP_10HDX_LBN))
+			result |= SUPPORTED_10baseT_Half;
+		if (cap & (1 << MC_CMD_PHY_CAP_10FDX_LBN))
+			result |= SUPPORTED_10baseT_Full;
+		if (cap & (1 << MC_CMD_PHY_CAP_100HDX_LBN))
+			result |= SUPPORTED_100baseT_Half;
+		if (cap & (1 << MC_CMD_PHY_CAP_100FDX_LBN))
+			result |= SUPPORTED_100baseT_Full;
+		if (cap & (1 << MC_CMD_PHY_CAP_1000HDX_LBN))
+			result |= SUPPORTED_1000baseT_Half;
+		if (cap & (1 << MC_CMD_PHY_CAP_1000FDX_LBN))
+			result |= SUPPORTED_1000baseT_Full;
+		if (cap & (1 << MC_CMD_PHY_CAP_10000FDX_LBN))
+			result |= SUPPORTED_10000baseT_Full;
+		break;
+	}
+
+	if (cap & (1 << MC_CMD_PHY_CAP_PAUSE_LBN))
+		result |= SUPPORTED_Pause;
+	if (cap & (1 << MC_CMD_PHY_CAP_ASYM_LBN))
+		result |= SUPPORTED_Asym_Pause;
+	if (cap & (1 << MC_CMD_PHY_CAP_AN_LBN))
+		result |= SUPPORTED_Autoneg;
+
+	return result;
+}
+
+static u32 ethtool_to_mcdi_cap(u32 cap)
+{
+	u32 result = 0;
+
+	if (cap & SUPPORTED_10baseT_Half)
+		result |= (1 << MC_CMD_PHY_CAP_10HDX_LBN);
+	if (cap & SUPPORTED_10baseT_Full)
+		result |= (1 << MC_CMD_PHY_CAP_10FDX_LBN);
+	if (cap & SUPPORTED_100baseT_Half)
+		result |= (1 << MC_CMD_PHY_CAP_100HDX_LBN);
+	if (cap & SUPPORTED_100baseT_Full)
+		result |= (1 << MC_CMD_PHY_CAP_100FDX_LBN);
+	if (cap & SUPPORTED_1000baseT_Half)
+		result |= (1 << MC_CMD_PHY_CAP_1000HDX_LBN);
+	if (cap & (SUPPORTED_1000baseT_Full | SUPPORTED_1000baseKX_Full))
+		result |= (1 << MC_CMD_PHY_CAP_1000FDX_LBN);
+	if (cap & (SUPPORTED_10000baseT_Full | SUPPORTED_10000baseKX4_Full))
+		result |= (1 << MC_CMD_PHY_CAP_10000FDX_LBN);
+	if (cap & SUPPORTED_Pause)
+		result |= (1 << MC_CMD_PHY_CAP_PAUSE_LBN);
+	if (cap & SUPPORTED_Asym_Pause)
+		result |= (1 << MC_CMD_PHY_CAP_ASYM_LBN);
+	if (cap & SUPPORTED_Autoneg)
+		result |= (1 << MC_CMD_PHY_CAP_AN_LBN);
+
+	return result;
+}
+
+static u32 efx_get_mcdi_phy_flags(struct efx_nic *efx)
+{
+	struct efx_mcdi_phy_data *phy_cfg = efx->phy_data;
+	enum efx_phy_mode mode, supported;
+	u32 flags;
+
+	/* TODO: Advertise the capabilities supported by this PHY */
+	supported = 0;
+	if (phy_cfg->flags & (1 << MC_CMD_GET_PHY_CFG_TXDIS_LBN))
+		supported |= PHY_MODE_TX_DISABLED;
+	if (phy_cfg->flags & (1 << MC_CMD_GET_PHY_CFG_LOWPOWER_LBN))
+		supported |= PHY_MODE_LOW_POWER;
+	if (phy_cfg->flags & (1 << MC_CMD_GET_PHY_CFG_POWEROFF_LBN))
+		supported |= PHY_MODE_OFF;
+
+	mode = efx->phy_mode & supported;
+
+	flags = 0;
+	if (mode & PHY_MODE_TX_DISABLED)
+		flags |= (1 << MC_CMD_SET_LINK_TXDIS_LBN);
+	if (mode & PHY_MODE_LOW_POWER)
+		flags |= (1 << MC_CMD_SET_LINK_LOWPOWER_LBN);
+	if (mode & PHY_MODE_OFF)
+		flags |= (1 << MC_CMD_SET_LINK_POWEROFF_LBN);
+
+	return flags;
+}
+
+static u32 mcdi_to_ethtool_media(u32 media)
+{
+	switch (media) {
+	case MC_CMD_MEDIA_XAUI:
+	case MC_CMD_MEDIA_CX4:
+	case MC_CMD_MEDIA_KX4:
+		return PORT_OTHER;
+
+	case MC_CMD_MEDIA_XFP:
+	case MC_CMD_MEDIA_SFP_PLUS:
+		return PORT_FIBRE;
+
+	case MC_CMD_MEDIA_BASE_T:
+		return PORT_TP;
+
+	default:
+		return PORT_OTHER;
+	}
+}
+
+static int efx_mcdi_phy_probe(struct efx_nic *efx)
+{
+	struct efx_mcdi_phy_data *phy_data;
+	u8 outbuf[MC_CMD_GET_LINK_OUT_LEN];
+	u32 caps;
+	int rc;
+
+	/* Initialise and populate phy_data */
+	phy_data = kzalloc(sizeof(*phy_data), GFP_KERNEL);
+	if (phy_data == NULL)
+		return -ENOMEM;
+
+	rc = efx_mcdi_get_phy_cfg(efx, phy_data);
+	if (rc != 0)
+		goto fail;
+
+	/* Read initial link advertisement */
+	BUILD_BUG_ON(MC_CMD_GET_LINK_IN_LEN != 0);
+	rc = efx_mcdi_rpc(efx, MC_CMD_GET_LINK, NULL, 0,
+			  outbuf, sizeof(outbuf), NULL);
+	if (rc)
+		goto fail;
+
+	/* Fill out nic state */
+	efx->phy_data = phy_data;
+	efx->phy_type = phy_data->type;
+
+	efx->mdio_bus = phy_data->channel;
+	efx->mdio.prtad = phy_data->port;
+	efx->mdio.mmds = phy_data->mmd_mask & ~(1 << MC_CMD_MMD_CLAUSE22);
+	efx->mdio.mode_support = 0;
+	if (phy_data->mmd_mask & (1 << MC_CMD_MMD_CLAUSE22))
+		efx->mdio.mode_support |= MDIO_SUPPORTS_C22;
+	if (phy_data->mmd_mask & ~(1 << MC_CMD_MMD_CLAUSE22))
+		efx->mdio.mode_support |= MDIO_SUPPORTS_C45 | MDIO_EMULATE_C22;
+
+	caps = MCDI_DWORD(outbuf, GET_LINK_OUT_CAP);
+	if (caps & (1 << MC_CMD_PHY_CAP_AN_LBN))
+		efx->link_advertising =
+			mcdi_to_ethtool_cap(phy_data->media, caps);
+	else
+		phy_data->forced_cap = caps;
+
+	/* Assert that we can map efx -> mcdi loopback modes */
+	BUILD_BUG_ON(LOOPBACK_NONE != MC_CMD_LOOPBACK_NONE);
+	BUILD_BUG_ON(LOOPBACK_DATA != MC_CMD_LOOPBACK_DATA);
+	BUILD_BUG_ON(LOOPBACK_GMAC != MC_CMD_LOOPBACK_GMAC);
+	BUILD_BUG_ON(LOOPBACK_XGMII != MC_CMD_LOOPBACK_XGMII);
+	BUILD_BUG_ON(LOOPBACK_XGXS != MC_CMD_LOOPBACK_XGXS);
+	BUILD_BUG_ON(LOOPBACK_XAUI != MC_CMD_LOOPBACK_XAUI);
+	BUILD_BUG_ON(LOOPBACK_GMII != MC_CMD_LOOPBACK_GMII);
+	BUILD_BUG_ON(LOOPBACK_SGMII != MC_CMD_LOOPBACK_SGMII);
+	BUILD_BUG_ON(LOOPBACK_XGBR != MC_CMD_LOOPBACK_XGBR);
+	BUILD_BUG_ON(LOOPBACK_XFI != MC_CMD_LOOPBACK_XFI);
+	BUILD_BUG_ON(LOOPBACK_XAUI_FAR != MC_CMD_LOOPBACK_XAUI_FAR);
+	BUILD_BUG_ON(LOOPBACK_GMII_FAR != MC_CMD_LOOPBACK_GMII_FAR);
+	BUILD_BUG_ON(LOOPBACK_SGMII_FAR != MC_CMD_LOOPBACK_SGMII_FAR);
+	BUILD_BUG_ON(LOOPBACK_XFI_FAR != MC_CMD_LOOPBACK_XFI_FAR);
+	BUILD_BUG_ON(LOOPBACK_GPHY != MC_CMD_LOOPBACK_GPHY);
+	BUILD_BUG_ON(LOOPBACK_PHYXS != MC_CMD_LOOPBACK_PHYXS);
+	BUILD_BUG_ON(LOOPBACK_PCS != MC_CMD_LOOPBACK_PCS);
+	BUILD_BUG_ON(LOOPBACK_PMAPMD != MC_CMD_LOOPBACK_PMAPMD);
+	BUILD_BUG_ON(LOOPBACK_XPORT != MC_CMD_LOOPBACK_XPORT);
+	BUILD_BUG_ON(LOOPBACK_XGMII_WS != MC_CMD_LOOPBACK_XGMII_WS);
+	BUILD_BUG_ON(LOOPBACK_XAUI_WS != MC_CMD_LOOPBACK_XAUI_WS);
+	BUILD_BUG_ON(LOOPBACK_XAUI_WS_FAR != MC_CMD_LOOPBACK_XAUI_WS_FAR);
+	BUILD_BUG_ON(LOOPBACK_XAUI_WS_NEAR != MC_CMD_LOOPBACK_XAUI_WS_NEAR);
+	BUILD_BUG_ON(LOOPBACK_GMII_WS != MC_CMD_LOOPBACK_GMII_WS);
+	BUILD_BUG_ON(LOOPBACK_XFI_WS != MC_CMD_LOOPBACK_XFI_WS);
+	BUILD_BUG_ON(LOOPBACK_XFI_WS_FAR != MC_CMD_LOOPBACK_XFI_WS_FAR);
+	BUILD_BUG_ON(LOOPBACK_PHYXS_WS != MC_CMD_LOOPBACK_PHYXS_WS);
+
+	rc = efx_mcdi_loopback_modes(efx, &efx->loopback_modes);
+	if (rc != 0)
+		goto fail;
+	/* The MC indicates that LOOPBACK_NONE is a valid loopback mode,
+	 * but by convention we don't */
+	efx->loopback_modes &= ~(1 << LOOPBACK_NONE);
+
+	/* Set the initial link mode */
+	efx_mcdi_phy_decode_link(
+		efx, &efx->link_state,
+		MCDI_DWORD(outbuf, GET_LINK_OUT_LINK_SPEED),
+		MCDI_DWORD(outbuf, GET_LINK_OUT_FLAGS),
+		MCDI_DWORD(outbuf, GET_LINK_OUT_FCNTL));
+
+	/* Default to Autonegotiated flow control if the PHY supports it */
+	efx->wanted_fc = EFX_FC_RX | EFX_FC_TX;
+	if (phy_data->supported_cap & (1 << MC_CMD_PHY_CAP_AN_LBN))
+		efx->wanted_fc |= EFX_FC_AUTO;
+	efx_link_set_wanted_fc(efx, efx->wanted_fc);
+
+	return 0;
+
+fail:
+	kfree(phy_data);
+	return rc;
+}
+
+int efx_mcdi_phy_reconfigure(struct efx_nic *efx)
+{
+	struct efx_mcdi_phy_data *phy_cfg = efx->phy_data;
+	u32 caps = (efx->link_advertising ?
+		    ethtool_to_mcdi_cap(efx->link_advertising) :
+		    phy_cfg->forced_cap);
+
+	return efx_mcdi_set_link(efx, caps, efx_get_mcdi_phy_flags(efx),
+				 efx->loopback_mode, 0);
+}
+
+void efx_mcdi_phy_decode_link(struct efx_nic *efx,
+			      struct efx_link_state *link_state,
+			      u32 speed, u32 flags, u32 fcntl)
+{
+	switch (fcntl) {
+	case MC_CMD_FCNTL_AUTO:
+		WARN_ON(1);	/* This is not a link mode */
+		link_state->fc = EFX_FC_AUTO | EFX_FC_TX | EFX_FC_RX;
+		break;
+	case MC_CMD_FCNTL_BIDIR:
+		link_state->fc = EFX_FC_TX | EFX_FC_RX;
+		break;
+	case MC_CMD_FCNTL_RESPOND:
+		link_state->fc = EFX_FC_RX;
+		break;
+	default:
+		WARN_ON(1);
+	case MC_CMD_FCNTL_OFF:
+		link_state->fc = 0;
+		break;
+	}
+
+	link_state->up = !!(flags & (1 << MC_CMD_GET_LINK_LINK_UP_LBN));
+	link_state->fd = !!(flags & (1 << MC_CMD_GET_LINK_FULL_DUPLEX_LBN));
+	link_state->speed = speed;
+}
+
+/* Verify that the forced flow control settings (!EFX_FC_AUTO) are
+ * supported by the link partner. Warn the user if this isn't the case
+ */
+void efx_mcdi_phy_check_fcntl(struct efx_nic *efx, u32 lpa)
+{
+	struct efx_mcdi_phy_data *phy_cfg = efx->phy_data;
+	u32 rmtadv;
+
+	/* The link partner capabilities are only relevant if the
+	 * link supports flow control autonegotiation */
+	if (~phy_cfg->supported_cap & (1 << MC_CMD_PHY_CAP_AN_LBN))
+		return;
+
+	/* If flow control autoneg is supported and enabled, then fine */
+	if (efx->wanted_fc & EFX_FC_AUTO)
+		return;
+
+	rmtadv = 0;
+	if (lpa & (1 << MC_CMD_PHY_CAP_PAUSE_LBN))
+		rmtadv |= ADVERTISED_Pause;
+	if (lpa & (1 << MC_CMD_PHY_CAP_ASYM_LBN))
+		rmtadv |=  ADVERTISED_Asym_Pause;
+
+	if ((efx->wanted_fc & EFX_FC_TX) && rmtadv == ADVERTISED_Asym_Pause)
+		netif_err(efx, link, efx->net_dev,
+			  "warning: link partner doesn't support pause frames");
+}
+
+static bool efx_mcdi_phy_poll(struct efx_nic *efx)
+{
+	struct efx_link_state old_state = efx->link_state;
+	u8 outbuf[MC_CMD_GET_LINK_OUT_LEN];
+	int rc;
+
+	WARN_ON(!mutex_is_locked(&efx->mac_lock));
+
+	BUILD_BUG_ON(MC_CMD_GET_LINK_IN_LEN != 0);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_GET_LINK, NULL, 0,
+			  outbuf, sizeof(outbuf), NULL);
+	if (rc) {
+		netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n",
+			  __func__, rc);
+		efx->link_state.up = false;
+	} else {
+		efx_mcdi_phy_decode_link(
+			efx, &efx->link_state,
+			MCDI_DWORD(outbuf, GET_LINK_OUT_LINK_SPEED),
+			MCDI_DWORD(outbuf, GET_LINK_OUT_FLAGS),
+			MCDI_DWORD(outbuf, GET_LINK_OUT_FCNTL));
+	}
+
+	return !efx_link_state_equal(&efx->link_state, &old_state);
+}
+
+static void efx_mcdi_phy_remove(struct efx_nic *efx)
+{
+	struct efx_mcdi_phy_data *phy_data = efx->phy_data;
+
+	efx->phy_data = NULL;
+	kfree(phy_data);
+}
+
+static void efx_mcdi_phy_get_settings(struct efx_nic *efx, struct ethtool_cmd *ecmd)
+{
+	struct efx_mcdi_phy_data *phy_cfg = efx->phy_data;
+	u8 outbuf[MC_CMD_GET_LINK_OUT_LEN];
+	int rc;
+
+	ecmd->supported =
+		mcdi_to_ethtool_cap(phy_cfg->media, phy_cfg->supported_cap);
+	ecmd->advertising = efx->link_advertising;
+	ethtool_cmd_speed_set(ecmd, efx->link_state.speed);
+	ecmd->duplex = efx->link_state.fd;
+	ecmd->port = mcdi_to_ethtool_media(phy_cfg->media);
+	ecmd->phy_address = phy_cfg->port;
+	ecmd->transceiver = XCVR_INTERNAL;
+	ecmd->autoneg = !!(efx->link_advertising & ADVERTISED_Autoneg);
+	ecmd->mdio_support = (efx->mdio.mode_support &
+			      (MDIO_SUPPORTS_C45 | MDIO_SUPPORTS_C22));
+
+	BUILD_BUG_ON(MC_CMD_GET_LINK_IN_LEN != 0);
+	rc = efx_mcdi_rpc(efx, MC_CMD_GET_LINK, NULL, 0,
+			  outbuf, sizeof(outbuf), NULL);
+	if (rc) {
+		netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n",
+			  __func__, rc);
+		return;
+	}
+	ecmd->lp_advertising =
+		mcdi_to_ethtool_cap(phy_cfg->media,
+				    MCDI_DWORD(outbuf, GET_LINK_OUT_LP_CAP));
+}
+
+static int efx_mcdi_phy_set_settings(struct efx_nic *efx, struct ethtool_cmd *ecmd)
+{
+	struct efx_mcdi_phy_data *phy_cfg = efx->phy_data;
+	u32 caps;
+	int rc;
+
+	if (ecmd->autoneg) {
+		caps = (ethtool_to_mcdi_cap(ecmd->advertising) |
+			 1 << MC_CMD_PHY_CAP_AN_LBN);
+	} else if (ecmd->duplex) {
+		switch (ethtool_cmd_speed(ecmd)) {
+		case 10:    caps = 1 << MC_CMD_PHY_CAP_10FDX_LBN;    break;
+		case 100:   caps = 1 << MC_CMD_PHY_CAP_100FDX_LBN;   break;
+		case 1000:  caps = 1 << MC_CMD_PHY_CAP_1000FDX_LBN;  break;
+		case 10000: caps = 1 << MC_CMD_PHY_CAP_10000FDX_LBN; break;
+		default:    return -EINVAL;
+		}
+	} else {
+		switch (ethtool_cmd_speed(ecmd)) {
+		case 10:    caps = 1 << MC_CMD_PHY_CAP_10HDX_LBN;    break;
+		case 100:   caps = 1 << MC_CMD_PHY_CAP_100HDX_LBN;   break;
+		case 1000:  caps = 1 << MC_CMD_PHY_CAP_1000HDX_LBN;  break;
+		default:    return -EINVAL;
+		}
+	}
+
+	rc = efx_mcdi_set_link(efx, caps, efx_get_mcdi_phy_flags(efx),
+			       efx->loopback_mode, 0);
+	if (rc)
+		return rc;
+
+	if (ecmd->autoneg) {
+		efx_link_set_advertising(
+			efx, ecmd->advertising | ADVERTISED_Autoneg);
+		phy_cfg->forced_cap = 0;
+	} else {
+		efx_link_set_advertising(efx, 0);
+		phy_cfg->forced_cap = caps;
+	}
+	return 0;
+}
+
+static int efx_mcdi_phy_test_alive(struct efx_nic *efx)
+{
+	u8 outbuf[MC_CMD_GET_PHY_STATE_OUT_LEN];
+	size_t outlen;
+	int rc;
+
+	BUILD_BUG_ON(MC_CMD_GET_PHY_STATE_IN_LEN != 0);
+
+	rc = efx_mcdi_rpc(efx, MC_CMD_GET_PHY_STATE, NULL, 0,
+			  outbuf, sizeof(outbuf), &outlen);
+	if (rc)
+		return rc;
+
+	if (outlen < MC_CMD_GET_PHY_STATE_OUT_LEN)
+		return -EIO;
+	if (MCDI_DWORD(outbuf, GET_PHY_STATE_STATE) != MC_CMD_PHY_STATE_OK)
+		return -EINVAL;
+
+	return 0;
+}
+
+static const char *const mcdi_sft9001_cable_diag_names[] = {
+	"cable.pairA.length",
+	"cable.pairB.length",
+	"cable.pairC.length",
+	"cable.pairD.length",
+	"cable.pairA.status",
+	"cable.pairB.status",
+	"cable.pairC.status",
+	"cable.pairD.status",
+};
+
+static int efx_mcdi_bist(struct efx_nic *efx, unsigned int bist_mode,
+			 int *results)
+{
+	unsigned int retry, i, count = 0;
+	size_t outlen;
+	u32 status;
+	u8 *buf, *ptr;
+	int rc;
+
+	buf = kzalloc(0x100, GFP_KERNEL);
+	if (buf == NULL)
+		return -ENOMEM;
+
+	BUILD_BUG_ON(MC_CMD_START_BIST_OUT_LEN != 0);
+	MCDI_SET_DWORD(buf, START_BIST_IN_TYPE, bist_mode);
+	rc = efx_mcdi_rpc(efx, MC_CMD_START_BIST, buf, MC_CMD_START_BIST_IN_LEN,
+			  NULL, 0, NULL);
+	if (rc)
+		goto out;
+
+	/* Wait up to 10s for BIST to finish */
+	for (retry = 0; retry < 100; ++retry) {
+		BUILD_BUG_ON(MC_CMD_POLL_BIST_IN_LEN != 0);
+		rc = efx_mcdi_rpc(efx, MC_CMD_POLL_BIST, NULL, 0,
+				  buf, 0x100, &outlen);
+		if (rc)
+			goto out;
+
+		status = MCDI_DWORD(buf, POLL_BIST_OUT_RESULT);
+		if (status != MC_CMD_POLL_BIST_RUNNING)
+			goto finished;
+
+		msleep(100);
+	}
+
+	rc = -ETIMEDOUT;
+	goto out;
+
+finished:
+	results[count++] = (status == MC_CMD_POLL_BIST_PASSED) ? 1 : -1;
+
+	/* SFT9001 specific cable diagnostics output */
+	if (efx->phy_type == PHY_TYPE_SFT9001B &&
+	    (bist_mode == MC_CMD_PHY_BIST_CABLE_SHORT ||
+	     bist_mode == MC_CMD_PHY_BIST_CABLE_LONG)) {
+		ptr = MCDI_PTR(buf, POLL_BIST_OUT_SFT9001_CABLE_LENGTH_A);
+		if (status == MC_CMD_POLL_BIST_PASSED &&
+		    outlen >= MC_CMD_POLL_BIST_OUT_SFT9001_LEN) {
+			for (i = 0; i < 8; i++) {
+				results[count + i] =
+					EFX_DWORD_FIELD(((efx_dword_t *)ptr)[i],
+							EFX_DWORD_0);
+			}
+		}
+		count += 8;
+	}
+	rc = count;
+
+out:
+	kfree(buf);
+
+	return rc;
+}
+
+static int efx_mcdi_phy_run_tests(struct efx_nic *efx, int *results,
+				  unsigned flags)
+{
+	struct efx_mcdi_phy_data *phy_cfg = efx->phy_data;
+	u32 mode;
+	int rc;
+
+	if (phy_cfg->flags & (1 << MC_CMD_GET_PHY_CFG_BIST_LBN)) {
+		rc = efx_mcdi_bist(efx, MC_CMD_PHY_BIST, results);
+		if (rc < 0)
+			return rc;
+
+		results += rc;
+	}
+
+	/* If we support both LONG and SHORT, then run each in response to
+	 * break or not. Otherwise, run the one we support */
+	mode = 0;
+	if (phy_cfg->flags & (1 << MC_CMD_GET_PHY_CFG_BIST_CABLE_SHORT_LBN)) {
+		if ((flags & ETH_TEST_FL_OFFLINE) &&
+		    (phy_cfg->flags &
+		     (1 << MC_CMD_GET_PHY_CFG_BIST_CABLE_LONG_LBN)))
+			mode = MC_CMD_PHY_BIST_CABLE_LONG;
+		else
+			mode = MC_CMD_PHY_BIST_CABLE_SHORT;
+	} else if (phy_cfg->flags &
+		   (1 << MC_CMD_GET_PHY_CFG_BIST_CABLE_LONG_LBN))
+		mode = MC_CMD_PHY_BIST_CABLE_LONG;
+
+	if (mode != 0) {
+		rc = efx_mcdi_bist(efx, mode, results);
+		if (rc < 0)
+			return rc;
+		results += rc;
+	}
+
+	return 0;
+}
+
+static const char *efx_mcdi_phy_test_name(struct efx_nic *efx,
+					  unsigned int index)
+{
+	struct efx_mcdi_phy_data *phy_cfg = efx->phy_data;
+
+	if (phy_cfg->flags & (1 << MC_CMD_GET_PHY_CFG_BIST_LBN)) {
+		if (index == 0)
+			return "bist";
+		--index;
+	}
+
+	if (phy_cfg->flags & ((1 << MC_CMD_GET_PHY_CFG_BIST_CABLE_SHORT_LBN) |
+			      (1 << MC_CMD_GET_PHY_CFG_BIST_CABLE_LONG_LBN))) {
+		if (index == 0)
+			return "cable";
+		--index;
+
+		if (efx->phy_type == PHY_TYPE_SFT9001B) {
+			if (index < ARRAY_SIZE(mcdi_sft9001_cable_diag_names))
+				return mcdi_sft9001_cable_diag_names[index];
+			index -= ARRAY_SIZE(mcdi_sft9001_cable_diag_names);
+		}
+	}
+
+	return NULL;
+}
+
+const struct efx_phy_operations efx_mcdi_phy_ops = {
+	.probe		= efx_mcdi_phy_probe,
+	.init 	 	= efx_port_dummy_op_int,
+	.reconfigure	= efx_mcdi_phy_reconfigure,
+	.poll		= efx_mcdi_phy_poll,
+	.fini		= efx_port_dummy_op_void,
+	.remove		= efx_mcdi_phy_remove,
+	.get_settings	= efx_mcdi_phy_get_settings,
+	.set_settings	= efx_mcdi_phy_set_settings,
+	.test_alive	= efx_mcdi_phy_test_alive,
+	.run_tests	= efx_mcdi_phy_run_tests,
+	.test_name	= efx_mcdi_phy_test_name,
+};
diff --git a/drivers/net/ethernet/sfc/mdio_10g.c b/drivers/net/ethernet/sfc/mdio_10g.c
new file mode 100644
index 0000000..7ab385c
--- /dev/null
+++ b/drivers/net/ethernet/sfc/mdio_10g.c
@@ -0,0 +1,323 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2006-2011 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+/*
+ * Useful functions for working with MDIO clause 45 PHYs
+ */
+#include <linux/types.h>
+#include <linux/ethtool.h>
+#include <linux/delay.h>
+#include "net_driver.h"
+#include "mdio_10g.h"
+#include "workarounds.h"
+
+unsigned efx_mdio_id_oui(u32 id)
+{
+	unsigned oui = 0;
+	int i;
+
+	/* The bits of the OUI are designated a..x, with a=0 and b variable.
+	 * In the id register c is the MSB but the OUI is conventionally
+	 * written as bytes h..a, p..i, x..q.  Reorder the bits accordingly. */
+	for (i = 0; i < 22; ++i)
+		if (id & (1 << (i + 10)))
+			oui |= 1 << (i ^ 7);
+
+	return oui;
+}
+
+int efx_mdio_reset_mmd(struct efx_nic *port, int mmd,
+			    int spins, int spintime)
+{
+	u32 ctrl;
+
+	/* Catch callers passing values in the wrong units (or just silly) */
+	EFX_BUG_ON_PARANOID(spins * spintime >= 5000);
+
+	efx_mdio_write(port, mmd, MDIO_CTRL1, MDIO_CTRL1_RESET);
+	/* Wait for the reset bit to clear. */
+	do {
+		msleep(spintime);
+		ctrl = efx_mdio_read(port, mmd, MDIO_CTRL1);
+		spins--;
+
+	} while (spins && (ctrl & MDIO_CTRL1_RESET));
+
+	return spins ? spins : -ETIMEDOUT;
+}
+
+static int efx_mdio_check_mmd(struct efx_nic *efx, int mmd)
+{
+	int status;
+
+	if (mmd != MDIO_MMD_AN) {
+		/* Read MMD STATUS2 to check it is responding. */
+		status = efx_mdio_read(efx, mmd, MDIO_STAT2);
+		if ((status & MDIO_STAT2_DEVPRST) != MDIO_STAT2_DEVPRST_VAL) {
+			netif_err(efx, hw, efx->net_dev,
+				  "PHY MMD %d not responding.\n", mmd);
+			return -EIO;
+		}
+	}
+
+	return 0;
+}
+
+/* This ought to be ridiculous overkill. We expect it to fail rarely */
+#define MDIO45_RESET_TIME	1000 /* ms */
+#define MDIO45_RESET_ITERS	100
+
+int efx_mdio_wait_reset_mmds(struct efx_nic *efx, unsigned int mmd_mask)
+{
+	const int spintime = MDIO45_RESET_TIME / MDIO45_RESET_ITERS;
+	int tries = MDIO45_RESET_ITERS;
+	int rc = 0;
+	int in_reset;
+
+	while (tries) {
+		int mask = mmd_mask;
+		int mmd = 0;
+		int stat;
+		in_reset = 0;
+		while (mask) {
+			if (mask & 1) {
+				stat = efx_mdio_read(efx, mmd, MDIO_CTRL1);
+				if (stat < 0) {
+					netif_err(efx, hw, efx->net_dev,
+						  "failed to read status of"
+						  " MMD %d\n", mmd);
+					return -EIO;
+				}
+				if (stat & MDIO_CTRL1_RESET)
+					in_reset |= (1 << mmd);
+			}
+			mask = mask >> 1;
+			mmd++;
+		}
+		if (!in_reset)
+			break;
+		tries--;
+		msleep(spintime);
+	}
+	if (in_reset != 0) {
+		netif_err(efx, hw, efx->net_dev,
+			  "not all MMDs came out of reset in time."
+			  " MMDs still in reset: %x\n", in_reset);
+		rc = -ETIMEDOUT;
+	}
+	return rc;
+}
+
+int efx_mdio_check_mmds(struct efx_nic *efx, unsigned int mmd_mask)
+{
+	int mmd = 0, probe_mmd, devs1, devs2;
+	u32 devices;
+
+	/* Historically we have probed the PHYXS to find out what devices are
+	 * present,but that doesn't work so well if the PHYXS isn't expected
+	 * to exist, if so just find the first item in the list supplied. */
+	probe_mmd = (mmd_mask & MDIO_DEVS_PHYXS) ? MDIO_MMD_PHYXS :
+	    __ffs(mmd_mask);
+
+	/* Check all the expected MMDs are present */
+	devs1 = efx_mdio_read(efx, probe_mmd, MDIO_DEVS1);
+	devs2 = efx_mdio_read(efx, probe_mmd, MDIO_DEVS2);
+	if (devs1 < 0 || devs2 < 0) {
+		netif_err(efx, hw, efx->net_dev,
+			  "failed to read devices present\n");
+		return -EIO;
+	}
+	devices = devs1 | (devs2 << 16);
+	if ((devices & mmd_mask) != mmd_mask) {
+		netif_err(efx, hw, efx->net_dev,
+			  "required MMDs not present: got %x, wanted %x\n",
+			  devices, mmd_mask);
+		return -ENODEV;
+	}
+	netif_vdbg(efx, hw, efx->net_dev, "Devices present: %x\n", devices);
+
+	/* Check all required MMDs are responding and happy. */
+	while (mmd_mask) {
+		if ((mmd_mask & 1) && efx_mdio_check_mmd(efx, mmd))
+			return -EIO;
+		mmd_mask = mmd_mask >> 1;
+		mmd++;
+	}
+
+	return 0;
+}
+
+bool efx_mdio_links_ok(struct efx_nic *efx, unsigned int mmd_mask)
+{
+	/* If the port is in loopback, then we should only consider a subset
+	 * of mmd's */
+	if (LOOPBACK_INTERNAL(efx))
+		return true;
+	else if (LOOPBACK_MASK(efx) & LOOPBACKS_WS)
+		return false;
+	else if (efx_phy_mode_disabled(efx->phy_mode))
+		return false;
+	else if (efx->loopback_mode == LOOPBACK_PHYXS)
+		mmd_mask &= ~(MDIO_DEVS_PHYXS |
+			      MDIO_DEVS_PCS |
+			      MDIO_DEVS_PMAPMD |
+			      MDIO_DEVS_AN);
+	else if (efx->loopback_mode == LOOPBACK_PCS)
+		mmd_mask &= ~(MDIO_DEVS_PCS |
+			      MDIO_DEVS_PMAPMD |
+			      MDIO_DEVS_AN);
+	else if (efx->loopback_mode == LOOPBACK_PMAPMD)
+		mmd_mask &= ~(MDIO_DEVS_PMAPMD |
+			      MDIO_DEVS_AN);
+
+	return mdio45_links_ok(&efx->mdio, mmd_mask);
+}
+
+void efx_mdio_transmit_disable(struct efx_nic *efx)
+{
+	efx_mdio_set_flag(efx, MDIO_MMD_PMAPMD,
+			  MDIO_PMA_TXDIS, MDIO_PMD_TXDIS_GLOBAL,
+			  efx->phy_mode & PHY_MODE_TX_DISABLED);
+}
+
+void efx_mdio_phy_reconfigure(struct efx_nic *efx)
+{
+	efx_mdio_set_flag(efx, MDIO_MMD_PMAPMD,
+			  MDIO_CTRL1, MDIO_PMA_CTRL1_LOOPBACK,
+			  efx->loopback_mode == LOOPBACK_PMAPMD);
+	efx_mdio_set_flag(efx, MDIO_MMD_PCS,
+			  MDIO_CTRL1, MDIO_PCS_CTRL1_LOOPBACK,
+			  efx->loopback_mode == LOOPBACK_PCS);
+	efx_mdio_set_flag(efx, MDIO_MMD_PHYXS,
+			  MDIO_CTRL1, MDIO_PHYXS_CTRL1_LOOPBACK,
+			  efx->loopback_mode == LOOPBACK_PHYXS_WS);
+}
+
+static void efx_mdio_set_mmd_lpower(struct efx_nic *efx,
+				    int lpower, int mmd)
+{
+	int stat = efx_mdio_read(efx, mmd, MDIO_STAT1);
+
+	netif_vdbg(efx, drv, efx->net_dev, "Setting low power mode for MMD %d to %d\n",
+		  mmd, lpower);
+
+	if (stat & MDIO_STAT1_LPOWERABLE) {
+		efx_mdio_set_flag(efx, mmd, MDIO_CTRL1,
+				  MDIO_CTRL1_LPOWER, lpower);
+	}
+}
+
+void efx_mdio_set_mmds_lpower(struct efx_nic *efx,
+			      int low_power, unsigned int mmd_mask)
+{
+	int mmd = 0;
+	mmd_mask &= ~MDIO_DEVS_AN;
+	while (mmd_mask) {
+		if (mmd_mask & 1)
+			efx_mdio_set_mmd_lpower(efx, low_power, mmd);
+		mmd_mask = (mmd_mask >> 1);
+		mmd++;
+	}
+}
+
+/**
+ * efx_mdio_set_settings - Set (some of) the PHY settings over MDIO.
+ * @efx:		Efx NIC
+ * @ecmd: 		New settings
+ */
+int efx_mdio_set_settings(struct efx_nic *efx, struct ethtool_cmd *ecmd)
+{
+	struct ethtool_cmd prev = { .cmd = ETHTOOL_GSET };
+
+	efx->phy_op->get_settings(efx, &prev);
+
+	if (ecmd->advertising == prev.advertising &&
+	    ethtool_cmd_speed(ecmd) == ethtool_cmd_speed(&prev) &&
+	    ecmd->duplex == prev.duplex &&
+	    ecmd->port == prev.port &&
+	    ecmd->autoneg == prev.autoneg)
+		return 0;
+
+	/* We can only change these settings for -T PHYs */
+	if (prev.port != PORT_TP || ecmd->port != PORT_TP)
+		return -EINVAL;
+
+	/* Check that PHY supports these settings */
+	if (!ecmd->autoneg ||
+	    (ecmd->advertising | SUPPORTED_Autoneg) & ~prev.supported)
+		return -EINVAL;
+
+	efx_link_set_advertising(efx, ecmd->advertising | ADVERTISED_Autoneg);
+	efx_mdio_an_reconfigure(efx);
+	return 0;
+}
+
+/**
+ * efx_mdio_an_reconfigure - Push advertising flags and restart autonegotiation
+ * @efx:		Efx NIC
+ */
+void efx_mdio_an_reconfigure(struct efx_nic *efx)
+{
+	int reg;
+
+	WARN_ON(!(efx->mdio.mmds & MDIO_DEVS_AN));
+
+	/* Set up the base page */
+	reg = ADVERTISE_CSMA | ADVERTISE_RESV;
+	if (efx->link_advertising & ADVERTISED_Pause)
+		reg |= ADVERTISE_PAUSE_CAP;
+	if (efx->link_advertising & ADVERTISED_Asym_Pause)
+		reg |= ADVERTISE_PAUSE_ASYM;
+	efx_mdio_write(efx, MDIO_MMD_AN, MDIO_AN_ADVERTISE, reg);
+
+	/* Set up the (extended) next page */
+	efx->phy_op->set_npage_adv(efx, efx->link_advertising);
+
+	/* Enable and restart AN */
+	reg = efx_mdio_read(efx, MDIO_MMD_AN, MDIO_CTRL1);
+	reg |= MDIO_AN_CTRL1_ENABLE | MDIO_AN_CTRL1_RESTART | MDIO_AN_CTRL1_XNP;
+	efx_mdio_write(efx, MDIO_MMD_AN, MDIO_CTRL1, reg);
+}
+
+u8 efx_mdio_get_pause(struct efx_nic *efx)
+{
+	BUILD_BUG_ON(EFX_FC_AUTO & (EFX_FC_RX | EFX_FC_TX));
+
+	if (!(efx->wanted_fc & EFX_FC_AUTO))
+		return efx->wanted_fc;
+
+	WARN_ON(!(efx->mdio.mmds & MDIO_DEVS_AN));
+
+	return mii_resolve_flowctrl_fdx(
+		mii_advertise_flowctrl(efx->wanted_fc),
+		efx_mdio_read(efx, MDIO_MMD_AN, MDIO_AN_LPA));
+}
+
+int efx_mdio_test_alive(struct efx_nic *efx)
+{
+	int rc;
+	int devad = __ffs(efx->mdio.mmds);
+	u16 physid1, physid2;
+
+	mutex_lock(&efx->mac_lock);
+
+	physid1 = efx_mdio_read(efx, devad, MDIO_DEVID1);
+	physid2 = efx_mdio_read(efx, devad, MDIO_DEVID2);
+
+	if ((physid1 == 0x0000) || (physid1 == 0xffff) ||
+	    (physid2 == 0x0000) || (physid2 == 0xffff)) {
+		netif_err(efx, hw, efx->net_dev,
+			  "no MDIO PHY present with ID %d\n", efx->mdio.prtad);
+		rc = -EINVAL;
+	} else {
+		rc = efx_mdio_check_mmds(efx, efx->mdio.mmds);
+	}
+
+	mutex_unlock(&efx->mac_lock);
+	return rc;
+}
diff --git a/drivers/net/ethernet/sfc/mdio_10g.h b/drivers/net/ethernet/sfc/mdio_10g.h
new file mode 100644
index 0000000..a97dbbd
--- /dev/null
+++ b/drivers/net/ethernet/sfc/mdio_10g.h
@@ -0,0 +1,112 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2006-2011 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#ifndef EFX_MDIO_10G_H
+#define EFX_MDIO_10G_H
+
+#include <linux/mdio.h>
+
+/*
+ * Helper functions for doing 10G MDIO as specified in IEEE 802.3 clause 45.
+ */
+
+#include "efx.h"
+
+static inline unsigned efx_mdio_id_rev(u32 id) { return id & 0xf; }
+static inline unsigned efx_mdio_id_model(u32 id) { return (id >> 4) & 0x3f; }
+extern unsigned efx_mdio_id_oui(u32 id);
+
+static inline int efx_mdio_read(struct efx_nic *efx, int devad, int addr)
+{
+	return efx->mdio.mdio_read(efx->net_dev, efx->mdio.prtad, devad, addr);
+}
+
+static inline void
+efx_mdio_write(struct efx_nic *efx, int devad, int addr, int value)
+{
+	efx->mdio.mdio_write(efx->net_dev, efx->mdio.prtad, devad, addr, value);
+}
+
+static inline u32 efx_mdio_read_id(struct efx_nic *efx, int mmd)
+{
+	u16 id_low = efx_mdio_read(efx, mmd, MDIO_DEVID2);
+	u16 id_hi = efx_mdio_read(efx, mmd, MDIO_DEVID1);
+	return (id_hi << 16) | (id_low);
+}
+
+static inline bool efx_mdio_phyxgxs_lane_sync(struct efx_nic *efx)
+{
+	int i, lane_status;
+	bool sync;
+
+	for (i = 0; i < 2; ++i)
+		lane_status = efx_mdio_read(efx, MDIO_MMD_PHYXS,
+					    MDIO_PHYXS_LNSTAT);
+
+	sync = !!(lane_status & MDIO_PHYXS_LNSTAT_ALIGN);
+	if (!sync)
+		netif_dbg(efx, hw, efx->net_dev, "XGXS lane status: %x\n",
+			  lane_status);
+	return sync;
+}
+
+extern const char *efx_mdio_mmd_name(int mmd);
+
+/*
+ * Reset a specific MMD and wait for reset to clear.
+ * Return number of spins left (>0) on success, -%ETIMEDOUT on failure.
+ *
+ * This function will sleep
+ */
+extern int efx_mdio_reset_mmd(struct efx_nic *efx, int mmd,
+			      int spins, int spintime);
+
+/* As efx_mdio_check_mmd but for multiple MMDs */
+int efx_mdio_check_mmds(struct efx_nic *efx, unsigned int mmd_mask);
+
+/* Check the link status of specified mmds in bit mask */
+extern bool efx_mdio_links_ok(struct efx_nic *efx, unsigned int mmd_mask);
+
+/* Generic transmit disable support though PMAPMD */
+extern void efx_mdio_transmit_disable(struct efx_nic *efx);
+
+/* Generic part of reconfigure: set/clear loopback bits */
+extern void efx_mdio_phy_reconfigure(struct efx_nic *efx);
+
+/* Set the power state of the specified MMDs */
+extern void efx_mdio_set_mmds_lpower(struct efx_nic *efx,
+				     int low_power, unsigned int mmd_mask);
+
+/* Set (some of) the PHY settings over MDIO */
+extern int efx_mdio_set_settings(struct efx_nic *efx, struct ethtool_cmd *ecmd);
+
+/* Push advertising flags and restart autonegotiation */
+extern void efx_mdio_an_reconfigure(struct efx_nic *efx);
+
+/* Get pause parameters from AN if available (otherwise return
+ * requested pause parameters)
+ */
+u8 efx_mdio_get_pause(struct efx_nic *efx);
+
+/* Wait for specified MMDs to exit reset within a timeout */
+extern int efx_mdio_wait_reset_mmds(struct efx_nic *efx,
+				    unsigned int mmd_mask);
+
+/* Set or clear flag, debouncing */
+static inline void
+efx_mdio_set_flag(struct efx_nic *efx, int devad, int addr,
+		  int mask, bool state)
+{
+	mdio_set_flag(&efx->mdio, efx->mdio.prtad, devad, addr, mask, state);
+}
+
+/* Liveness self-test for MDIO PHYs */
+extern int efx_mdio_test_alive(struct efx_nic *efx);
+
+#endif /* EFX_MDIO_10G_H */
diff --git a/drivers/net/ethernet/sfc/mtd.c b/drivers/net/ethernet/sfc/mtd.c
new file mode 100644
index 0000000..b630448
--- /dev/null
+++ b/drivers/net/ethernet/sfc/mtd.c
@@ -0,0 +1,693 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2005-2006 Fen Systems Ltd.
+ * Copyright 2006-2010 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#include <linux/bitops.h>
+#include <linux/module.h>
+#include <linux/mtd/mtd.h>
+#include <linux/delay.h>
+#include <linux/slab.h>
+#include <linux/rtnetlink.h>
+
+#include "net_driver.h"
+#include "spi.h"
+#include "efx.h"
+#include "nic.h"
+#include "mcdi.h"
+#include "mcdi_pcol.h"
+
+#define EFX_SPI_VERIFY_BUF_LEN 16
+
+struct efx_mtd_partition {
+	struct mtd_info mtd;
+	union {
+		struct {
+			bool updating;
+			u8 nvram_type;
+			u16 fw_subtype;
+		} mcdi;
+		size_t offset;
+	};
+	const char *type_name;
+	char name[IFNAMSIZ + 20];
+};
+
+struct efx_mtd_ops {
+	int (*read)(struct mtd_info *mtd, loff_t start, size_t len,
+		    size_t *retlen, u8 *buffer);
+	int (*erase)(struct mtd_info *mtd, loff_t start, size_t len);
+	int (*write)(struct mtd_info *mtd, loff_t start, size_t len,
+		     size_t *retlen, const u8 *buffer);
+	int (*sync)(struct mtd_info *mtd);
+};
+
+struct efx_mtd {
+	struct list_head node;
+	struct efx_nic *efx;
+	const struct efx_spi_device *spi;
+	const char *name;
+	const struct efx_mtd_ops *ops;
+	size_t n_parts;
+	struct efx_mtd_partition part[0];
+};
+
+#define efx_for_each_partition(part, efx_mtd)			\
+	for ((part) = &(efx_mtd)->part[0];			\
+	     (part) != &(efx_mtd)->part[(efx_mtd)->n_parts];	\
+	     (part)++)
+
+#define to_efx_mtd_partition(mtd)				\
+	container_of(mtd, struct efx_mtd_partition, mtd)
+
+static int falcon_mtd_probe(struct efx_nic *efx);
+static int siena_mtd_probe(struct efx_nic *efx);
+
+/* SPI utilities */
+
+static int
+efx_spi_slow_wait(struct efx_mtd_partition *part, bool uninterruptible)
+{
+	struct efx_mtd *efx_mtd = part->mtd.priv;
+	const struct efx_spi_device *spi = efx_mtd->spi;
+	struct efx_nic *efx = efx_mtd->efx;
+	u8 status;
+	int rc, i;
+
+	/* Wait up to 4s for flash/EEPROM to finish a slow operation. */
+	for (i = 0; i < 40; i++) {
+		__set_current_state(uninterruptible ?
+				    TASK_UNINTERRUPTIBLE : TASK_INTERRUPTIBLE);
+		schedule_timeout(HZ / 10);
+		rc = falcon_spi_cmd(efx, spi, SPI_RDSR, -1, NULL,
+				    &status, sizeof(status));
+		if (rc)
+			return rc;
+		if (!(status & SPI_STATUS_NRDY))
+			return 0;
+		if (signal_pending(current))
+			return -EINTR;
+	}
+	pr_err("%s: timed out waiting for %s\n", part->name, efx_mtd->name);
+	return -ETIMEDOUT;
+}
+
+static int
+efx_spi_unlock(struct efx_nic *efx, const struct efx_spi_device *spi)
+{
+	const u8 unlock_mask = (SPI_STATUS_BP2 | SPI_STATUS_BP1 |
+				SPI_STATUS_BP0);
+	u8 status;
+	int rc;
+
+	rc = falcon_spi_cmd(efx, spi, SPI_RDSR, -1, NULL,
+			    &status, sizeof(status));
+	if (rc)
+		return rc;
+
+	if (!(status & unlock_mask))
+		return 0; /* already unlocked */
+
+	rc = falcon_spi_cmd(efx, spi, SPI_WREN, -1, NULL, NULL, 0);
+	if (rc)
+		return rc;
+	rc = falcon_spi_cmd(efx, spi, SPI_SST_EWSR, -1, NULL, NULL, 0);
+	if (rc)
+		return rc;
+
+	status &= ~unlock_mask;
+	rc = falcon_spi_cmd(efx, spi, SPI_WRSR, -1, &status,
+			    NULL, sizeof(status));
+	if (rc)
+		return rc;
+	rc = falcon_spi_wait_write(efx, spi);
+	if (rc)
+		return rc;
+
+	return 0;
+}
+
+static int
+efx_spi_erase(struct efx_mtd_partition *part, loff_t start, size_t len)
+{
+	struct efx_mtd *efx_mtd = part->mtd.priv;
+	const struct efx_spi_device *spi = efx_mtd->spi;
+	struct efx_nic *efx = efx_mtd->efx;
+	unsigned pos, block_len;
+	u8 empty[EFX_SPI_VERIFY_BUF_LEN];
+	u8 buffer[EFX_SPI_VERIFY_BUF_LEN];
+	int rc;
+
+	if (len != spi->erase_size)
+		return -EINVAL;
+
+	if (spi->erase_command == 0)
+		return -EOPNOTSUPP;
+
+	rc = efx_spi_unlock(efx, spi);
+	if (rc)
+		return rc;
+	rc = falcon_spi_cmd(efx, spi, SPI_WREN, -1, NULL, NULL, 0);
+	if (rc)
+		return rc;
+	rc = falcon_spi_cmd(efx, spi, spi->erase_command, start, NULL,
+			    NULL, 0);
+	if (rc)
+		return rc;
+	rc = efx_spi_slow_wait(part, false);
+
+	/* Verify the entire region has been wiped */
+	memset(empty, 0xff, sizeof(empty));
+	for (pos = 0; pos < len; pos += block_len) {
+		block_len = min(len - pos, sizeof(buffer));
+		rc = falcon_spi_read(efx, spi, start + pos, block_len,
+				     NULL, buffer);
+		if (rc)
+			return rc;
+		if (memcmp(empty, buffer, block_len))
+			return -EIO;
+
+		/* Avoid locking up the system */
+		cond_resched();
+		if (signal_pending(current))
+			return -EINTR;
+	}
+
+	return rc;
+}
+
+/* MTD interface */
+
+static int efx_mtd_erase(struct mtd_info *mtd, struct erase_info *erase)
+{
+	struct efx_mtd *efx_mtd = mtd->priv;
+	int rc;
+
+	rc = efx_mtd->ops->erase(mtd, erase->addr, erase->len);
+	if (rc == 0) {
+		erase->state = MTD_ERASE_DONE;
+	} else {
+		erase->state = MTD_ERASE_FAILED;
+		erase->fail_addr = 0xffffffff;
+	}
+	mtd_erase_callback(erase);
+	return rc;
+}
+
+static void efx_mtd_sync(struct mtd_info *mtd)
+{
+	struct efx_mtd_partition *part = to_efx_mtd_partition(mtd);
+	struct efx_mtd *efx_mtd = mtd->priv;
+	int rc;
+
+	rc = efx_mtd->ops->sync(mtd);
+	if (rc)
+		pr_err("%s: %s sync failed (%d)\n",
+		       part->name, efx_mtd->name, rc);
+}
+
+static void efx_mtd_remove_partition(struct efx_mtd_partition *part)
+{
+	int rc;
+
+	for (;;) {
+		rc = mtd_device_unregister(&part->mtd);
+		if (rc != -EBUSY)
+			break;
+		ssleep(1);
+	}
+	WARN_ON(rc);
+}
+
+static void efx_mtd_remove_device(struct efx_mtd *efx_mtd)
+{
+	struct efx_mtd_partition *part;
+
+	efx_for_each_partition(part, efx_mtd)
+		efx_mtd_remove_partition(part);
+	list_del(&efx_mtd->node);
+	kfree(efx_mtd);
+}
+
+static void efx_mtd_rename_device(struct efx_mtd *efx_mtd)
+{
+	struct efx_mtd_partition *part;
+
+	efx_for_each_partition(part, efx_mtd)
+		if (efx_nic_rev(efx_mtd->efx) >= EFX_REV_SIENA_A0)
+			snprintf(part->name, sizeof(part->name),
+				 "%s %s:%02x", efx_mtd->efx->name,
+				 part->type_name, part->mcdi.fw_subtype);
+		else
+			snprintf(part->name, sizeof(part->name),
+				 "%s %s", efx_mtd->efx->name,
+				 part->type_name);
+}
+
+static int efx_mtd_probe_device(struct efx_nic *efx, struct efx_mtd *efx_mtd)
+{
+	struct efx_mtd_partition *part;
+
+	efx_mtd->efx = efx;
+
+	efx_mtd_rename_device(efx_mtd);
+
+	efx_for_each_partition(part, efx_mtd) {
+		part->mtd.writesize = 1;
+
+		part->mtd.owner = THIS_MODULE;
+		part->mtd.priv = efx_mtd;
+		part->mtd.name = part->name;
+		part->mtd.erase = efx_mtd_erase;
+		part->mtd.read = efx_mtd->ops->read;
+		part->mtd.write = efx_mtd->ops->write;
+		part->mtd.sync = efx_mtd_sync;
+
+		if (mtd_device_register(&part->mtd, NULL, 0))
+			goto fail;
+	}
+
+	list_add(&efx_mtd->node, &efx->mtd_list);
+	return 0;
+
+fail:
+	while (part != &efx_mtd->part[0]) {
+		--part;
+		efx_mtd_remove_partition(part);
+	}
+	/* mtd_device_register() returns 1 if the MTD table is full */
+	return -ENOMEM;
+}
+
+void efx_mtd_remove(struct efx_nic *efx)
+{
+	struct efx_mtd *efx_mtd, *next;
+
+	WARN_ON(efx_dev_registered(efx));
+
+	list_for_each_entry_safe(efx_mtd, next, &efx->mtd_list, node)
+		efx_mtd_remove_device(efx_mtd);
+}
+
+void efx_mtd_rename(struct efx_nic *efx)
+{
+	struct efx_mtd *efx_mtd;
+
+	ASSERT_RTNL();
+
+	list_for_each_entry(efx_mtd, &efx->mtd_list, node)
+		efx_mtd_rename_device(efx_mtd);
+}
+
+int efx_mtd_probe(struct efx_nic *efx)
+{
+	if (efx_nic_rev(efx) >= EFX_REV_SIENA_A0)
+		return siena_mtd_probe(efx);
+	else
+		return falcon_mtd_probe(efx);
+}
+
+/* Implementation of MTD operations for Falcon */
+
+static int falcon_mtd_read(struct mtd_info *mtd, loff_t start,
+			   size_t len, size_t *retlen, u8 *buffer)
+{
+	struct efx_mtd_partition *part = to_efx_mtd_partition(mtd);
+	struct efx_mtd *efx_mtd = mtd->priv;
+	const struct efx_spi_device *spi = efx_mtd->spi;
+	struct efx_nic *efx = efx_mtd->efx;
+	struct falcon_nic_data *nic_data = efx->nic_data;
+	int rc;
+
+	rc = mutex_lock_interruptible(&nic_data->spi_lock);
+	if (rc)
+		return rc;
+	rc = falcon_spi_read(efx, spi, part->offset + start, len,
+			     retlen, buffer);
+	mutex_unlock(&nic_data->spi_lock);
+	return rc;
+}
+
+static int falcon_mtd_erase(struct mtd_info *mtd, loff_t start, size_t len)
+{
+	struct efx_mtd_partition *part = to_efx_mtd_partition(mtd);
+	struct efx_mtd *efx_mtd = mtd->priv;
+	struct efx_nic *efx = efx_mtd->efx;
+	struct falcon_nic_data *nic_data = efx->nic_data;
+	int rc;
+
+	rc = mutex_lock_interruptible(&nic_data->spi_lock);
+	if (rc)
+		return rc;
+	rc = efx_spi_erase(part, part->offset + start, len);
+	mutex_unlock(&nic_data->spi_lock);
+	return rc;
+}
+
+static int falcon_mtd_write(struct mtd_info *mtd, loff_t start,
+			    size_t len, size_t *retlen, const u8 *buffer)
+{
+	struct efx_mtd_partition *part = to_efx_mtd_partition(mtd);
+	struct efx_mtd *efx_mtd = mtd->priv;
+	const struct efx_spi_device *spi = efx_mtd->spi;
+	struct efx_nic *efx = efx_mtd->efx;
+	struct falcon_nic_data *nic_data = efx->nic_data;
+	int rc;
+
+	rc = mutex_lock_interruptible(&nic_data->spi_lock);
+	if (rc)
+		return rc;
+	rc = falcon_spi_write(efx, spi, part->offset + start, len,
+			      retlen, buffer);
+	mutex_unlock(&nic_data->spi_lock);
+	return rc;
+}
+
+static int falcon_mtd_sync(struct mtd_info *mtd)
+{
+	struct efx_mtd_partition *part = to_efx_mtd_partition(mtd);
+	struct efx_mtd *efx_mtd = mtd->priv;
+	struct efx_nic *efx = efx_mtd->efx;
+	struct falcon_nic_data *nic_data = efx->nic_data;
+	int rc;
+
+	mutex_lock(&nic_data->spi_lock);
+	rc = efx_spi_slow_wait(part, true);
+	mutex_unlock(&nic_data->spi_lock);
+	return rc;
+}
+
+static struct efx_mtd_ops falcon_mtd_ops = {
+	.read	= falcon_mtd_read,
+	.erase	= falcon_mtd_erase,
+	.write	= falcon_mtd_write,
+	.sync	= falcon_mtd_sync,
+};
+
+static int falcon_mtd_probe(struct efx_nic *efx)
+{
+	struct falcon_nic_data *nic_data = efx->nic_data;
+	struct efx_spi_device *spi;
+	struct efx_mtd *efx_mtd;
+	int rc = -ENODEV;
+
+	ASSERT_RTNL();
+
+	spi = &nic_data->spi_flash;
+	if (efx_spi_present(spi) && spi->size > FALCON_FLASH_BOOTCODE_START) {
+		efx_mtd = kzalloc(sizeof(*efx_mtd) + sizeof(efx_mtd->part[0]),
+				  GFP_KERNEL);
+		if (!efx_mtd)
+			return -ENOMEM;
+
+		efx_mtd->spi = spi;
+		efx_mtd->name = "flash";
+		efx_mtd->ops = &falcon_mtd_ops;
+
+		efx_mtd->n_parts = 1;
+		efx_mtd->part[0].mtd.type = MTD_NORFLASH;
+		efx_mtd->part[0].mtd.flags = MTD_CAP_NORFLASH;
+		efx_mtd->part[0].mtd.size = spi->size - FALCON_FLASH_BOOTCODE_START;
+		efx_mtd->part[0].mtd.erasesize = spi->erase_size;
+		efx_mtd->part[0].offset = FALCON_FLASH_BOOTCODE_START;
+		efx_mtd->part[0].type_name = "sfc_flash_bootrom";
+
+		rc = efx_mtd_probe_device(efx, efx_mtd);
+		if (rc) {
+			kfree(efx_mtd);
+			return rc;
+		}
+	}
+
+	spi = &nic_data->spi_eeprom;
+	if (efx_spi_present(spi) && spi->size > EFX_EEPROM_BOOTCONFIG_START) {
+		efx_mtd = kzalloc(sizeof(*efx_mtd) + sizeof(efx_mtd->part[0]),
+				  GFP_KERNEL);
+		if (!efx_mtd)
+			return -ENOMEM;
+
+		efx_mtd->spi = spi;
+		efx_mtd->name = "EEPROM";
+		efx_mtd->ops = &falcon_mtd_ops;
+
+		efx_mtd->n_parts = 1;
+		efx_mtd->part[0].mtd.type = MTD_RAM;
+		efx_mtd->part[0].mtd.flags = MTD_CAP_RAM;
+		efx_mtd->part[0].mtd.size =
+			min(spi->size, EFX_EEPROM_BOOTCONFIG_END) -
+			EFX_EEPROM_BOOTCONFIG_START;
+		efx_mtd->part[0].mtd.erasesize = spi->erase_size;
+		efx_mtd->part[0].offset = EFX_EEPROM_BOOTCONFIG_START;
+		efx_mtd->part[0].type_name = "sfc_bootconfig";
+
+		rc = efx_mtd_probe_device(efx, efx_mtd);
+		if (rc) {
+			kfree(efx_mtd);
+			return rc;
+		}
+	}
+
+	return rc;
+}
+
+/* Implementation of MTD operations for Siena */
+
+static int siena_mtd_read(struct mtd_info *mtd, loff_t start,
+			  size_t len, size_t *retlen, u8 *buffer)
+{
+	struct efx_mtd_partition *part = to_efx_mtd_partition(mtd);
+	struct efx_mtd *efx_mtd = mtd->priv;
+	struct efx_nic *efx = efx_mtd->efx;
+	loff_t offset = start;
+	loff_t end = min_t(loff_t, start + len, mtd->size);
+	size_t chunk;
+	int rc = 0;
+
+	while (offset < end) {
+		chunk = min_t(size_t, end - offset, EFX_MCDI_NVRAM_LEN_MAX);
+		rc = efx_mcdi_nvram_read(efx, part->mcdi.nvram_type, offset,
+					 buffer, chunk);
+		if (rc)
+			goto out;
+		offset += chunk;
+		buffer += chunk;
+	}
+out:
+	*retlen = offset - start;
+	return rc;
+}
+
+static int siena_mtd_erase(struct mtd_info *mtd, loff_t start, size_t len)
+{
+	struct efx_mtd_partition *part = to_efx_mtd_partition(mtd);
+	struct efx_mtd *efx_mtd = mtd->priv;
+	struct efx_nic *efx = efx_mtd->efx;
+	loff_t offset = start & ~((loff_t)(mtd->erasesize - 1));
+	loff_t end = min_t(loff_t, start + len, mtd->size);
+	size_t chunk = part->mtd.erasesize;
+	int rc = 0;
+
+	if (!part->mcdi.updating) {
+		rc = efx_mcdi_nvram_update_start(efx, part->mcdi.nvram_type);
+		if (rc)
+			goto out;
+		part->mcdi.updating = 1;
+	}
+
+	/* The MCDI interface can in fact do multiple erase blocks at once;
+	 * but erasing may be slow, so we make multiple calls here to avoid
+	 * tripping the MCDI RPC timeout. */
+	while (offset < end) {
+		rc = efx_mcdi_nvram_erase(efx, part->mcdi.nvram_type, offset,
+					  chunk);
+		if (rc)
+			goto out;
+		offset += chunk;
+	}
+out:
+	return rc;
+}
+
+static int siena_mtd_write(struct mtd_info *mtd, loff_t start,
+			   size_t len, size_t *retlen, const u8 *buffer)
+{
+	struct efx_mtd_partition *part = to_efx_mtd_partition(mtd);
+	struct efx_mtd *efx_mtd = mtd->priv;
+	struct efx_nic *efx = efx_mtd->efx;
+	loff_t offset = start;
+	loff_t end = min_t(loff_t, start + len, mtd->size);
+	size_t chunk;
+	int rc = 0;
+
+	if (!part->mcdi.updating) {
+		rc = efx_mcdi_nvram_update_start(efx, part->mcdi.nvram_type);
+		if (rc)
+			goto out;
+		part->mcdi.updating = 1;
+	}
+
+	while (offset < end) {
+		chunk = min_t(size_t, end - offset, EFX_MCDI_NVRAM_LEN_MAX);
+		rc = efx_mcdi_nvram_write(efx, part->mcdi.nvram_type, offset,
+					  buffer, chunk);
+		if (rc)
+			goto out;
+		offset += chunk;
+		buffer += chunk;
+	}
+out:
+	*retlen = offset - start;
+	return rc;
+}
+
+static int siena_mtd_sync(struct mtd_info *mtd)
+{
+	struct efx_mtd_partition *part = to_efx_mtd_partition(mtd);
+	struct efx_mtd *efx_mtd = mtd->priv;
+	struct efx_nic *efx = efx_mtd->efx;
+	int rc = 0;
+
+	if (part->mcdi.updating) {
+		part->mcdi.updating = 0;
+		rc = efx_mcdi_nvram_update_finish(efx, part->mcdi.nvram_type);
+	}
+
+	return rc;
+}
+
+static struct efx_mtd_ops siena_mtd_ops = {
+	.read	= siena_mtd_read,
+	.erase	= siena_mtd_erase,
+	.write	= siena_mtd_write,
+	.sync	= siena_mtd_sync,
+};
+
+struct siena_nvram_type_info {
+	int port;
+	const char *name;
+};
+
+static struct siena_nvram_type_info siena_nvram_types[] = {
+	[MC_CMD_NVRAM_TYPE_DISABLED_CALLISTO]	= { 0, "sfc_dummy_phy" },
+	[MC_CMD_NVRAM_TYPE_MC_FW]		= { 0, "sfc_mcfw" },
+	[MC_CMD_NVRAM_TYPE_MC_FW_BACKUP]	= { 0, "sfc_mcfw_backup" },
+	[MC_CMD_NVRAM_TYPE_STATIC_CFG_PORT0]	= { 0, "sfc_static_cfg" },
+	[MC_CMD_NVRAM_TYPE_STATIC_CFG_PORT1]	= { 1, "sfc_static_cfg" },
+	[MC_CMD_NVRAM_TYPE_DYNAMIC_CFG_PORT0]	= { 0, "sfc_dynamic_cfg" },
+	[MC_CMD_NVRAM_TYPE_DYNAMIC_CFG_PORT1]	= { 1, "sfc_dynamic_cfg" },
+	[MC_CMD_NVRAM_TYPE_EXP_ROM]		= { 0, "sfc_exp_rom" },
+	[MC_CMD_NVRAM_TYPE_EXP_ROM_CFG_PORT0]	= { 0, "sfc_exp_rom_cfg" },
+	[MC_CMD_NVRAM_TYPE_EXP_ROM_CFG_PORT1]	= { 1, "sfc_exp_rom_cfg" },
+	[MC_CMD_NVRAM_TYPE_PHY_PORT0]		= { 0, "sfc_phy_fw" },
+	[MC_CMD_NVRAM_TYPE_PHY_PORT1]		= { 1, "sfc_phy_fw" },
+};
+
+static int siena_mtd_probe_partition(struct efx_nic *efx,
+				     struct efx_mtd *efx_mtd,
+				     unsigned int part_id,
+				     unsigned int type)
+{
+	struct efx_mtd_partition *part = &efx_mtd->part[part_id];
+	struct siena_nvram_type_info *info;
+	size_t size, erase_size;
+	bool protected;
+	int rc;
+
+	if (type >= ARRAY_SIZE(siena_nvram_types))
+		return -ENODEV;
+
+	info = &siena_nvram_types[type];
+
+	if (info->port != efx_port_num(efx))
+		return -ENODEV;
+
+	rc = efx_mcdi_nvram_info(efx, type, &size, &erase_size, &protected);
+	if (rc)
+		return rc;
+	if (protected)
+		return -ENODEV; /* hide it */
+
+	part->mcdi.nvram_type = type;
+	part->type_name = info->name;
+
+	part->mtd.type = MTD_NORFLASH;
+	part->mtd.flags = MTD_CAP_NORFLASH;
+	part->mtd.size = size;
+	part->mtd.erasesize = erase_size;
+
+	return 0;
+}
+
+static int siena_mtd_get_fw_subtypes(struct efx_nic *efx,
+				     struct efx_mtd *efx_mtd)
+{
+	struct efx_mtd_partition *part;
+	uint16_t fw_subtype_list[MC_CMD_GET_BOARD_CFG_OUT_FW_SUBTYPE_LIST_LEN /
+				 sizeof(uint16_t)];
+	int rc;
+
+	rc = efx_mcdi_get_board_cfg(efx, NULL, fw_subtype_list);
+	if (rc)
+		return rc;
+
+	efx_for_each_partition(part, efx_mtd)
+		part->mcdi.fw_subtype = fw_subtype_list[part->mcdi.nvram_type];
+
+	return 0;
+}
+
+static int siena_mtd_probe(struct efx_nic *efx)
+{
+	struct efx_mtd *efx_mtd;
+	int rc = -ENODEV;
+	u32 nvram_types;
+	unsigned int type;
+
+	ASSERT_RTNL();
+
+	rc = efx_mcdi_nvram_types(efx, &nvram_types);
+	if (rc)
+		return rc;
+
+	efx_mtd = kzalloc(sizeof(*efx_mtd) +
+			  hweight32(nvram_types) * sizeof(efx_mtd->part[0]),
+			  GFP_KERNEL);
+	if (!efx_mtd)
+		return -ENOMEM;
+
+	efx_mtd->name = "Siena NVRAM manager";
+
+	efx_mtd->ops = &siena_mtd_ops;
+
+	type = 0;
+	efx_mtd->n_parts = 0;
+
+	while (nvram_types != 0) {
+		if (nvram_types & 1) {
+			rc = siena_mtd_probe_partition(efx, efx_mtd,
+						       efx_mtd->n_parts, type);
+			if (rc == 0)
+				efx_mtd->n_parts++;
+			else if (rc != -ENODEV)
+				goto fail;
+		}
+		type++;
+		nvram_types >>= 1;
+	}
+
+	rc = siena_mtd_get_fw_subtypes(efx, efx_mtd);
+	if (rc)
+		goto fail;
+
+	rc = efx_mtd_probe_device(efx, efx_mtd);
+fail:
+	if (rc)
+		kfree(efx_mtd);
+	return rc;
+}
+
diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h
new file mode 100644
index 0000000..b8e251a
--- /dev/null
+++ b/drivers/net/ethernet/sfc/net_driver.h
@@ -0,0 +1,1060 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2005-2006 Fen Systems Ltd.
+ * Copyright 2005-2011 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+/* Common definitions for all Efx net driver code */
+
+#ifndef EFX_NET_DRIVER_H
+#define EFX_NET_DRIVER_H
+
+#if defined(EFX_ENABLE_DEBUG) && !defined(DEBUG)
+#define DEBUG
+#endif
+
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/ethtool.h>
+#include <linux/if_vlan.h>
+#include <linux/timer.h>
+#include <linux/mdio.h>
+#include <linux/list.h>
+#include <linux/pci.h>
+#include <linux/device.h>
+#include <linux/highmem.h>
+#include <linux/workqueue.h>
+#include <linux/vmalloc.h>
+#include <linux/i2c.h>
+
+#include "enum.h"
+#include "bitfield.h"
+
+/**************************************************************************
+ *
+ * Build definitions
+ *
+ **************************************************************************/
+
+#define EFX_DRIVER_VERSION	"3.1"
+
+#ifdef EFX_ENABLE_DEBUG
+#define EFX_BUG_ON_PARANOID(x) BUG_ON(x)
+#define EFX_WARN_ON_PARANOID(x) WARN_ON(x)
+#else
+#define EFX_BUG_ON_PARANOID(x) do {} while (0)
+#define EFX_WARN_ON_PARANOID(x) do {} while (0)
+#endif
+
+/**************************************************************************
+ *
+ * Efx data structures
+ *
+ **************************************************************************/
+
+#define EFX_MAX_CHANNELS 32
+#define EFX_MAX_RX_QUEUES EFX_MAX_CHANNELS
+
+/* Checksum generation is a per-queue option in hardware, so each
+ * queue visible to the networking core is backed by two hardware TX
+ * queues. */
+#define EFX_MAX_TX_TC		2
+#define EFX_MAX_CORE_TX_QUEUES	(EFX_MAX_TX_TC * EFX_MAX_CHANNELS)
+#define EFX_TXQ_TYPE_OFFLOAD	1	/* flag */
+#define EFX_TXQ_TYPE_HIGHPRI	2	/* flag */
+#define EFX_TXQ_TYPES		4
+#define EFX_MAX_TX_QUEUES	(EFX_TXQ_TYPES * EFX_MAX_CHANNELS)
+
+/**
+ * struct efx_special_buffer - An Efx special buffer
+ * @addr: CPU base address of the buffer
+ * @dma_addr: DMA base address of the buffer
+ * @len: Buffer length, in bytes
+ * @index: Buffer index within controller;s buffer table
+ * @entries: Number of buffer table entries
+ *
+ * Special buffers are used for the event queues and the TX and RX
+ * descriptor queues for each channel.  They are *not* used for the
+ * actual transmit and receive buffers.
+ */
+struct efx_special_buffer {
+	void *addr;
+	dma_addr_t dma_addr;
+	unsigned int len;
+	int index;
+	int entries;
+};
+
+enum efx_flush_state {
+	FLUSH_NONE,
+	FLUSH_PENDING,
+	FLUSH_FAILED,
+	FLUSH_DONE,
+};
+
+/**
+ * struct efx_tx_buffer - An Efx TX buffer
+ * @skb: The associated socket buffer.
+ *	Set only on the final fragment of a packet; %NULL for all other
+ *	fragments.  When this fragment completes, then we can free this
+ *	skb.
+ * @tsoh: The associated TSO header structure, or %NULL if this
+ *	buffer is not a TSO header.
+ * @dma_addr: DMA address of the fragment.
+ * @len: Length of this fragment.
+ *	This field is zero when the queue slot is empty.
+ * @continuation: True if this fragment is not the end of a packet.
+ * @unmap_single: True if pci_unmap_single should be used.
+ * @unmap_len: Length of this fragment to unmap
+ */
+struct efx_tx_buffer {
+	const struct sk_buff *skb;
+	struct efx_tso_header *tsoh;
+	dma_addr_t dma_addr;
+	unsigned short len;
+	bool continuation;
+	bool unmap_single;
+	unsigned short unmap_len;
+};
+
+/**
+ * struct efx_tx_queue - An Efx TX queue
+ *
+ * This is a ring buffer of TX fragments.
+ * Since the TX completion path always executes on the same
+ * CPU and the xmit path can operate on different CPUs,
+ * performance is increased by ensuring that the completion
+ * path and the xmit path operate on different cache lines.
+ * This is particularly important if the xmit path is always
+ * executing on one CPU which is different from the completion
+ * path.  There is also a cache line for members which are
+ * read but not written on the fast path.
+ *
+ * @efx: The associated Efx NIC
+ * @queue: DMA queue number
+ * @channel: The associated channel
+ * @core_txq: The networking core TX queue structure
+ * @buffer: The software buffer ring
+ * @txd: The hardware descriptor ring
+ * @ptr_mask: The size of the ring minus 1.
+ * @initialised: Has hardware queue been initialised?
+ * @flushed: Used when handling queue flushing
+ * @read_count: Current read pointer.
+ *	This is the number of buffers that have been removed from both rings.
+ * @old_write_count: The value of @write_count when last checked.
+ *	This is here for performance reasons.  The xmit path will
+ *	only get the up-to-date value of @write_count if this
+ *	variable indicates that the queue is empty.  This is to
+ *	avoid cache-line ping-pong between the xmit path and the
+ *	completion path.
+ * @insert_count: Current insert pointer
+ *	This is the number of buffers that have been added to the
+ *	software ring.
+ * @write_count: Current write pointer
+ *	This is the number of buffers that have been added to the
+ *	hardware ring.
+ * @old_read_count: The value of read_count when last checked.
+ *	This is here for performance reasons.  The xmit path will
+ *	only get the up-to-date value of read_count if this
+ *	variable indicates that the queue is full.  This is to
+ *	avoid cache-line ping-pong between the xmit path and the
+ *	completion path.
+ * @tso_headers_free: A list of TSO headers allocated for this TX queue
+ *	that are not in use, and so available for new TSO sends. The list
+ *	is protected by the TX queue lock.
+ * @tso_bursts: Number of times TSO xmit invoked by kernel
+ * @tso_long_headers: Number of packets with headers too long for standard
+ *	blocks
+ * @tso_packets: Number of packets via the TSO xmit path
+ * @pushes: Number of times the TX push feature has been used
+ * @empty_read_count: If the completion path has seen the queue as empty
+ *	and the transmission path has not yet checked this, the value of
+ *	@read_count bitwise-added to %EFX_EMPTY_COUNT_VALID; otherwise 0.
+ */
+struct efx_tx_queue {
+	/* Members which don't change on the fast path */
+	struct efx_nic *efx ____cacheline_aligned_in_smp;
+	unsigned queue;
+	struct efx_channel *channel;
+	struct netdev_queue *core_txq;
+	struct efx_tx_buffer *buffer;
+	struct efx_special_buffer txd;
+	unsigned int ptr_mask;
+	bool initialised;
+	enum efx_flush_state flushed;
+
+	/* Members used mainly on the completion path */
+	unsigned int read_count ____cacheline_aligned_in_smp;
+	unsigned int old_write_count;
+
+	/* Members used only on the xmit path */
+	unsigned int insert_count ____cacheline_aligned_in_smp;
+	unsigned int write_count;
+	unsigned int old_read_count;
+	struct efx_tso_header *tso_headers_free;
+	unsigned int tso_bursts;
+	unsigned int tso_long_headers;
+	unsigned int tso_packets;
+	unsigned int pushes;
+
+	/* Members shared between paths and sometimes updated */
+	unsigned int empty_read_count ____cacheline_aligned_in_smp;
+#define EFX_EMPTY_COUNT_VALID 0x80000000
+};
+
+/**
+ * struct efx_rx_buffer - An Efx RX data buffer
+ * @dma_addr: DMA base address of the buffer
+ * @skb: The associated socket buffer, if any.
+ *	If both this and page are %NULL, the buffer slot is currently free.
+ * @page: The associated page buffer, if any.
+ *	If both this and skb are %NULL, the buffer slot is currently free.
+ * @len: Buffer length, in bytes.
+ * @is_page: Indicates if @page is valid. If false, @skb is valid.
+ */
+struct efx_rx_buffer {
+	dma_addr_t dma_addr;
+	union {
+		struct sk_buff *skb;
+		struct page *page;
+	} u;
+	unsigned int len;
+	bool is_page;
+};
+
+/**
+ * struct efx_rx_page_state - Page-based rx buffer state
+ *
+ * Inserted at the start of every page allocated for receive buffers.
+ * Used to facilitate sharing dma mappings between recycled rx buffers
+ * and those passed up to the kernel.
+ *
+ * @refcnt: Number of struct efx_rx_buffer's referencing this page.
+ *	When refcnt falls to zero, the page is unmapped for dma
+ * @dma_addr: The dma address of this page.
+ */
+struct efx_rx_page_state {
+	unsigned refcnt;
+	dma_addr_t dma_addr;
+
+	unsigned int __pad[0] ____cacheline_aligned;
+};
+
+/**
+ * struct efx_rx_queue - An Efx RX queue
+ * @efx: The associated Efx NIC
+ * @buffer: The software buffer ring
+ * @rxd: The hardware descriptor ring
+ * @ptr_mask: The size of the ring minus 1.
+ * @added_count: Number of buffers added to the receive queue.
+ * @notified_count: Number of buffers given to NIC (<= @added_count).
+ * @removed_count: Number of buffers removed from the receive queue.
+ * @max_fill: RX descriptor maximum fill level (<= ring size)
+ * @fast_fill_trigger: RX descriptor fill level that will trigger a fast fill
+ *	(<= @max_fill)
+ * @fast_fill_limit: The level to which a fast fill will fill
+ *	(@fast_fill_trigger <= @fast_fill_limit <= @max_fill)
+ * @min_fill: RX descriptor minimum non-zero fill level.
+ *	This records the minimum fill level observed when a ring
+ *	refill was triggered.
+ * @alloc_page_count: RX allocation strategy counter.
+ * @alloc_skb_count: RX allocation strategy counter.
+ * @slow_fill: Timer used to defer efx_nic_generate_fill_event().
+ * @flushed: Use when handling queue flushing
+ */
+struct efx_rx_queue {
+	struct efx_nic *efx;
+	struct efx_rx_buffer *buffer;
+	struct efx_special_buffer rxd;
+	unsigned int ptr_mask;
+
+	int added_count;
+	int notified_count;
+	int removed_count;
+	unsigned int max_fill;
+	unsigned int fast_fill_trigger;
+	unsigned int fast_fill_limit;
+	unsigned int min_fill;
+	unsigned int min_overfill;
+	unsigned int alloc_page_count;
+	unsigned int alloc_skb_count;
+	struct timer_list slow_fill;
+	unsigned int slow_fill_count;
+
+	enum efx_flush_state flushed;
+};
+
+/**
+ * struct efx_buffer - An Efx general-purpose buffer
+ * @addr: host base address of the buffer
+ * @dma_addr: DMA base address of the buffer
+ * @len: Buffer length, in bytes
+ *
+ * The NIC uses these buffers for its interrupt status registers and
+ * MAC stats dumps.
+ */
+struct efx_buffer {
+	void *addr;
+	dma_addr_t dma_addr;
+	unsigned int len;
+};
+
+
+enum efx_rx_alloc_method {
+	RX_ALLOC_METHOD_AUTO = 0,
+	RX_ALLOC_METHOD_SKB = 1,
+	RX_ALLOC_METHOD_PAGE = 2,
+};
+
+/**
+ * struct efx_channel - An Efx channel
+ *
+ * A channel comprises an event queue, at least one TX queue, at least
+ * one RX queue, and an associated tasklet for processing the event
+ * queue.
+ *
+ * @efx: Associated Efx NIC
+ * @channel: Channel instance number
+ * @enabled: Channel enabled indicator
+ * @irq: IRQ number (MSI and MSI-X only)
+ * @irq_moderation: IRQ moderation value (in hardware ticks)
+ * @napi_dev: Net device used with NAPI
+ * @napi_str: NAPI control structure
+ * @work_pending: Is work pending via NAPI?
+ * @eventq: Event queue buffer
+ * @eventq_mask: Event queue pointer mask
+ * @eventq_read_ptr: Event queue read pointer
+ * @last_eventq_read_ptr: Last event queue read pointer value.
+ * @irq_count: Number of IRQs since last adaptive moderation decision
+ * @irq_mod_score: IRQ moderation score
+ * @rx_alloc_level: Watermark based heuristic counter for pushing descriptors
+ *	and diagnostic counters
+ * @rx_alloc_push_pages: RX allocation method currently in use for pushing
+ *	descriptors
+ * @n_rx_tobe_disc: Count of RX_TOBE_DISC errors
+ * @n_rx_ip_hdr_chksum_err: Count of RX IP header checksum errors
+ * @n_rx_tcp_udp_chksum_err: Count of RX TCP and UDP checksum errors
+ * @n_rx_mcast_mismatch: Count of unmatched multicast frames
+ * @n_rx_frm_trunc: Count of RX_FRM_TRUNC errors
+ * @n_rx_overlength: Count of RX_OVERLENGTH errors
+ * @n_skbuff_leaks: Count of skbuffs leaked due to RX overrun
+ * @rx_queue: RX queue for this channel
+ * @tx_queue: TX queues for this channel
+ */
+struct efx_channel {
+	struct efx_nic *efx;
+	int channel;
+	bool enabled;
+	int irq;
+	unsigned int irq_moderation;
+	struct net_device *napi_dev;
+	struct napi_struct napi_str;
+	bool work_pending;
+	struct efx_special_buffer eventq;
+	unsigned int eventq_mask;
+	unsigned int eventq_read_ptr;
+	unsigned int last_eventq_read_ptr;
+
+	unsigned int irq_count;
+	unsigned int irq_mod_score;
+#ifdef CONFIG_RFS_ACCEL
+	unsigned int rfs_filters_added;
+#endif
+
+	int rx_alloc_level;
+	int rx_alloc_push_pages;
+
+	unsigned n_rx_tobe_disc;
+	unsigned n_rx_ip_hdr_chksum_err;
+	unsigned n_rx_tcp_udp_chksum_err;
+	unsigned n_rx_mcast_mismatch;
+	unsigned n_rx_frm_trunc;
+	unsigned n_rx_overlength;
+	unsigned n_skbuff_leaks;
+
+	/* Used to pipeline received packets in order to optimise memory
+	 * access with prefetches.
+	 */
+	struct efx_rx_buffer *rx_pkt;
+	bool rx_pkt_csummed;
+
+	struct efx_rx_queue rx_queue;
+	struct efx_tx_queue tx_queue[EFX_TXQ_TYPES];
+};
+
+enum efx_led_mode {
+	EFX_LED_OFF	= 0,
+	EFX_LED_ON	= 1,
+	EFX_LED_DEFAULT	= 2
+};
+
+#define STRING_TABLE_LOOKUP(val, member) \
+	((val) < member ## _max) ? member ## _names[val] : "(invalid)"
+
+extern const char *efx_loopback_mode_names[];
+extern const unsigned int efx_loopback_mode_max;
+#define LOOPBACK_MODE(efx) \
+	STRING_TABLE_LOOKUP((efx)->loopback_mode, efx_loopback_mode)
+
+extern const char *efx_reset_type_names[];
+extern const unsigned int efx_reset_type_max;
+#define RESET_TYPE(type) \
+	STRING_TABLE_LOOKUP(type, efx_reset_type)
+
+enum efx_int_mode {
+	/* Be careful if altering to correct macro below */
+	EFX_INT_MODE_MSIX = 0,
+	EFX_INT_MODE_MSI = 1,
+	EFX_INT_MODE_LEGACY = 2,
+	EFX_INT_MODE_MAX	/* Insert any new items before this */
+};
+#define EFX_INT_MODE_USE_MSI(x) (((x)->interrupt_mode) <= EFX_INT_MODE_MSI)
+
+enum nic_state {
+	STATE_INIT = 0,
+	STATE_RUNNING = 1,
+	STATE_FINI = 2,
+	STATE_DISABLED = 3,
+	STATE_MAX,
+};
+
+/*
+ * Alignment of page-allocated RX buffers
+ *
+ * Controls the number of bytes inserted at the start of an RX buffer.
+ * This is the equivalent of NET_IP_ALIGN [which controls the alignment
+ * of the skb->head for hardware DMA].
+ */
+#ifdef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
+#define EFX_PAGE_IP_ALIGN 0
+#else
+#define EFX_PAGE_IP_ALIGN NET_IP_ALIGN
+#endif
+
+/*
+ * Alignment of the skb->head which wraps a page-allocated RX buffer
+ *
+ * The skb allocated to wrap an rx_buffer can have this alignment. Since
+ * the data is memcpy'd from the rx_buf, it does not need to be equal to
+ * EFX_PAGE_IP_ALIGN.
+ */
+#define EFX_PAGE_SKB_ALIGN 2
+
+/* Forward declaration */
+struct efx_nic;
+
+/* Pseudo bit-mask flow control field */
+#define EFX_FC_RX	FLOW_CTRL_RX
+#define EFX_FC_TX	FLOW_CTRL_TX
+#define EFX_FC_AUTO	4
+
+/**
+ * struct efx_link_state - Current state of the link
+ * @up: Link is up
+ * @fd: Link is full-duplex
+ * @fc: Actual flow control flags
+ * @speed: Link speed (Mbps)
+ */
+struct efx_link_state {
+	bool up;
+	bool fd;
+	u8 fc;
+	unsigned int speed;
+};
+
+static inline bool efx_link_state_equal(const struct efx_link_state *left,
+					const struct efx_link_state *right)
+{
+	return left->up == right->up && left->fd == right->fd &&
+		left->fc == right->fc && left->speed == right->speed;
+}
+
+/**
+ * struct efx_mac_operations - Efx MAC operations table
+ * @reconfigure: Reconfigure MAC. Serialised by the mac_lock
+ * @update_stats: Update statistics
+ * @check_fault: Check fault state. True if fault present.
+ */
+struct efx_mac_operations {
+	int (*reconfigure) (struct efx_nic *efx);
+	void (*update_stats) (struct efx_nic *efx);
+	bool (*check_fault)(struct efx_nic *efx);
+};
+
+/**
+ * struct efx_phy_operations - Efx PHY operations table
+ * @probe: Probe PHY and initialise efx->mdio.mode_support, efx->mdio.mmds,
+ *	efx->loopback_modes.
+ * @init: Initialise PHY
+ * @fini: Shut down PHY
+ * @reconfigure: Reconfigure PHY (e.g. for new link parameters)
+ * @poll: Update @link_state and report whether it changed.
+ *	Serialised by the mac_lock.
+ * @get_settings: Get ethtool settings. Serialised by the mac_lock.
+ * @set_settings: Set ethtool settings. Serialised by the mac_lock.
+ * @set_npage_adv: Set abilities advertised in (Extended) Next Page
+ *	(only needed where AN bit is set in mmds)
+ * @test_alive: Test that PHY is 'alive' (online)
+ * @test_name: Get the name of a PHY-specific test/result
+ * @run_tests: Run tests and record results as appropriate (offline).
+ *	Flags are the ethtool tests flags.
+ */
+struct efx_phy_operations {
+	int (*probe) (struct efx_nic *efx);
+	int (*init) (struct efx_nic *efx);
+	void (*fini) (struct efx_nic *efx);
+	void (*remove) (struct efx_nic *efx);
+	int (*reconfigure) (struct efx_nic *efx);
+	bool (*poll) (struct efx_nic *efx);
+	void (*get_settings) (struct efx_nic *efx,
+			      struct ethtool_cmd *ecmd);
+	int (*set_settings) (struct efx_nic *efx,
+			     struct ethtool_cmd *ecmd);
+	void (*set_npage_adv) (struct efx_nic *efx, u32);
+	int (*test_alive) (struct efx_nic *efx);
+	const char *(*test_name) (struct efx_nic *efx, unsigned int index);
+	int (*run_tests) (struct efx_nic *efx, int *results, unsigned flags);
+};
+
+/**
+ * @enum efx_phy_mode - PHY operating mode flags
+ * @PHY_MODE_NORMAL: on and should pass traffic
+ * @PHY_MODE_TX_DISABLED: on with TX disabled
+ * @PHY_MODE_LOW_POWER: set to low power through MDIO
+ * @PHY_MODE_OFF: switched off through external control
+ * @PHY_MODE_SPECIAL: on but will not pass traffic
+ */
+enum efx_phy_mode {
+	PHY_MODE_NORMAL		= 0,
+	PHY_MODE_TX_DISABLED	= 1,
+	PHY_MODE_LOW_POWER	= 2,
+	PHY_MODE_OFF		= 4,
+	PHY_MODE_SPECIAL	= 8,
+};
+
+static inline bool efx_phy_mode_disabled(enum efx_phy_mode mode)
+{
+	return !!(mode & ~PHY_MODE_TX_DISABLED);
+}
+
+/*
+ * Efx extended statistics
+ *
+ * Not all statistics are provided by all supported MACs.  The purpose
+ * is this structure is to contain the raw statistics provided by each
+ * MAC.
+ */
+struct efx_mac_stats {
+	u64 tx_bytes;
+	u64 tx_good_bytes;
+	u64 tx_bad_bytes;
+	unsigned long tx_packets;
+	unsigned long tx_bad;
+	unsigned long tx_pause;
+	unsigned long tx_control;
+	unsigned long tx_unicast;
+	unsigned long tx_multicast;
+	unsigned long tx_broadcast;
+	unsigned long tx_lt64;
+	unsigned long tx_64;
+	unsigned long tx_65_to_127;
+	unsigned long tx_128_to_255;
+	unsigned long tx_256_to_511;
+	unsigned long tx_512_to_1023;
+	unsigned long tx_1024_to_15xx;
+	unsigned long tx_15xx_to_jumbo;
+	unsigned long tx_gtjumbo;
+	unsigned long tx_collision;
+	unsigned long tx_single_collision;
+	unsigned long tx_multiple_collision;
+	unsigned long tx_excessive_collision;
+	unsigned long tx_deferred;
+	unsigned long tx_late_collision;
+	unsigned long tx_excessive_deferred;
+	unsigned long tx_non_tcpudp;
+	unsigned long tx_mac_src_error;
+	unsigned long tx_ip_src_error;
+	u64 rx_bytes;
+	u64 rx_good_bytes;
+	u64 rx_bad_bytes;
+	unsigned long rx_packets;
+	unsigned long rx_good;
+	unsigned long rx_bad;
+	unsigned long rx_pause;
+	unsigned long rx_control;
+	unsigned long rx_unicast;
+	unsigned long rx_multicast;
+	unsigned long rx_broadcast;
+	unsigned long rx_lt64;
+	unsigned long rx_64;
+	unsigned long rx_65_to_127;
+	unsigned long rx_128_to_255;
+	unsigned long rx_256_to_511;
+	unsigned long rx_512_to_1023;
+	unsigned long rx_1024_to_15xx;
+	unsigned long rx_15xx_to_jumbo;
+	unsigned long rx_gtjumbo;
+	unsigned long rx_bad_lt64;
+	unsigned long rx_bad_64_to_15xx;
+	unsigned long rx_bad_15xx_to_jumbo;
+	unsigned long rx_bad_gtjumbo;
+	unsigned long rx_overflow;
+	unsigned long rx_missed;
+	unsigned long rx_false_carrier;
+	unsigned long rx_symbol_error;
+	unsigned long rx_align_error;
+	unsigned long rx_length_error;
+	unsigned long rx_internal_error;
+	unsigned long rx_good_lt64;
+};
+
+/* Number of bits used in a multicast filter hash address */
+#define EFX_MCAST_HASH_BITS 8
+
+/* Number of (single-bit) entries in a multicast filter hash */
+#define EFX_MCAST_HASH_ENTRIES (1 << EFX_MCAST_HASH_BITS)
+
+/* An Efx multicast filter hash */
+union efx_multicast_hash {
+	u8 byte[EFX_MCAST_HASH_ENTRIES / 8];
+	efx_oword_t oword[EFX_MCAST_HASH_ENTRIES / sizeof(efx_oword_t) / 8];
+};
+
+struct efx_filter_state;
+
+/**
+ * struct efx_nic - an Efx NIC
+ * @name: Device name (net device name or bus id before net device registered)
+ * @pci_dev: The PCI device
+ * @type: Controller type attributes
+ * @legacy_irq: IRQ number
+ * @legacy_irq_enabled: Are IRQs enabled on NIC (INT_EN_KER register)?
+ * @workqueue: Workqueue for port reconfigures and the HW monitor.
+ *	Work items do not hold and must not acquire RTNL.
+ * @workqueue_name: Name of workqueue
+ * @reset_work: Scheduled reset workitem
+ * @membase_phys: Memory BAR value as physical address
+ * @membase: Memory BAR value
+ * @interrupt_mode: Interrupt mode
+ * @irq_rx_adaptive: Adaptive IRQ moderation enabled for RX event queues
+ * @irq_rx_moderation: IRQ moderation time for RX event queues
+ * @msg_enable: Log message enable flags
+ * @state: Device state flag. Serialised by the rtnl_lock.
+ * @reset_pending: Bitmask for pending resets
+ * @tx_queue: TX DMA queues
+ * @rx_queue: RX DMA queues
+ * @channel: Channels
+ * @channel_name: Names for channels and their IRQs
+ * @rxq_entries: Size of receive queues requested by user.
+ * @txq_entries: Size of transmit queues requested by user.
+ * @next_buffer_table: First available buffer table id
+ * @n_channels: Number of channels in use
+ * @n_rx_channels: Number of channels used for RX (= number of RX queues)
+ * @n_tx_channels: Number of channels used for TX
+ * @rx_buffer_len: RX buffer length
+ * @rx_buffer_order: Order (log2) of number of pages for each RX buffer
+ * @rx_hash_key: Toeplitz hash key for RSS
+ * @rx_indir_table: Indirection table for RSS
+ * @int_error_count: Number of internal errors seen recently
+ * @int_error_expire: Time at which error count will be expired
+ * @irq_status: Interrupt status buffer
+ * @irq_zero_count: Number of legacy IRQs seen with queue flags == 0
+ * @fatal_irq_level: IRQ level (bit number) used for serious errors
+ * @mtd_list: List of MTDs attached to the NIC
+ * @nic_data: Hardware dependent state
+ * @mac_lock: MAC access lock. Protects @port_enabled, @phy_mode,
+ *	efx_monitor() and efx_reconfigure_port()
+ * @port_enabled: Port enabled indicator.
+ *	Serialises efx_stop_all(), efx_start_all(), efx_monitor() and
+ *	efx_mac_work() with kernel interfaces. Safe to read under any
+ *	one of the rtnl_lock, mac_lock, or netif_tx_lock, but all three must
+ *	be held to modify it.
+ * @port_initialized: Port initialized?
+ * @net_dev: Operating system network device. Consider holding the rtnl lock
+ * @stats_buffer: DMA buffer for statistics
+ * @mac_op: MAC interface
+ * @phy_type: PHY type
+ * @phy_op: PHY interface
+ * @phy_data: PHY private data (including PHY-specific stats)
+ * @mdio: PHY MDIO interface
+ * @mdio_bus: PHY MDIO bus ID (only used by Siena)
+ * @phy_mode: PHY operating mode. Serialised by @mac_lock.
+ * @link_advertising: Autonegotiation advertising flags
+ * @link_state: Current state of the link
+ * @n_link_state_changes: Number of times the link has changed state
+ * @promiscuous: Promiscuous flag. Protected by netif_tx_lock.
+ * @multicast_hash: Multicast hash table
+ * @wanted_fc: Wanted flow control flags
+ * @mac_work: Work item for changing MAC promiscuity and multicast hash
+ * @loopback_mode: Loopback status
+ * @loopback_modes: Supported loopback mode bitmask
+ * @loopback_selftest: Offline self-test private state
+ * @monitor_work: Hardware monitor workitem
+ * @biu_lock: BIU (bus interface unit) lock
+ * @last_irq_cpu: Last CPU to handle interrupt.
+ *	This register is written with the SMP processor ID whenever an
+ *	interrupt is handled.  It is used by efx_nic_test_interrupt()
+ *	to verify that an interrupt has occurred.
+ * @n_rx_nodesc_drop_cnt: RX no descriptor drop count
+ * @mac_stats: MAC statistics. These include all statistics the MACs
+ *	can provide.  Generic code converts these into a standard
+ *	&struct net_device_stats.
+ * @stats_lock: Statistics update lock. Serialises statistics fetches
+ *
+ * This is stored in the private area of the &struct net_device.
+ */
+struct efx_nic {
+	/* The following fields should be written very rarely */
+
+	char name[IFNAMSIZ];
+	struct pci_dev *pci_dev;
+	const struct efx_nic_type *type;
+	int legacy_irq;
+	bool legacy_irq_enabled;
+	struct workqueue_struct *workqueue;
+	char workqueue_name[16];
+	struct work_struct reset_work;
+	resource_size_t membase_phys;
+	void __iomem *membase;
+
+	enum efx_int_mode interrupt_mode;
+	bool irq_rx_adaptive;
+	unsigned int irq_rx_moderation;
+	u32 msg_enable;
+
+	enum nic_state state;
+	unsigned long reset_pending;
+
+	struct efx_channel *channel[EFX_MAX_CHANNELS];
+	char channel_name[EFX_MAX_CHANNELS][IFNAMSIZ + 6];
+
+	unsigned rxq_entries;
+	unsigned txq_entries;
+	unsigned next_buffer_table;
+	unsigned n_channels;
+	unsigned n_rx_channels;
+	unsigned tx_channel_offset;
+	unsigned n_tx_channels;
+	unsigned int rx_buffer_len;
+	unsigned int rx_buffer_order;
+	u8 rx_hash_key[40];
+	u32 rx_indir_table[128];
+
+	unsigned int_error_count;
+	unsigned long int_error_expire;
+
+	struct efx_buffer irq_status;
+	unsigned irq_zero_count;
+	unsigned fatal_irq_level;
+
+#ifdef CONFIG_SFC_MTD
+	struct list_head mtd_list;
+#endif
+
+	void *nic_data;
+
+	struct mutex mac_lock;
+	struct work_struct mac_work;
+	bool port_enabled;
+
+	bool port_initialized;
+	struct net_device *net_dev;
+
+	struct efx_buffer stats_buffer;
+
+	const struct efx_mac_operations *mac_op;
+
+	unsigned int phy_type;
+	const struct efx_phy_operations *phy_op;
+	void *phy_data;
+	struct mdio_if_info mdio;
+	unsigned int mdio_bus;
+	enum efx_phy_mode phy_mode;
+
+	u32 link_advertising;
+	struct efx_link_state link_state;
+	unsigned int n_link_state_changes;
+
+	bool promiscuous;
+	union efx_multicast_hash multicast_hash;
+	u8 wanted_fc;
+
+	atomic_t rx_reset;
+	enum efx_loopback_mode loopback_mode;
+	u64 loopback_modes;
+
+	void *loopback_selftest;
+
+	struct efx_filter_state *filter_state;
+
+	/* The following fields may be written more often */
+
+	struct delayed_work monitor_work ____cacheline_aligned_in_smp;
+	spinlock_t biu_lock;
+	volatile signed int last_irq_cpu;
+	unsigned n_rx_nodesc_drop_cnt;
+	struct efx_mac_stats mac_stats;
+	spinlock_t stats_lock;
+};
+
+static inline int efx_dev_registered(struct efx_nic *efx)
+{
+	return efx->net_dev->reg_state == NETREG_REGISTERED;
+}
+
+/* Net device name, for inclusion in log messages if it has been registered.
+ * Use efx->name not efx->net_dev->name so that races with (un)registration
+ * are harmless.
+ */
+static inline const char *efx_dev_name(struct efx_nic *efx)
+{
+	return efx_dev_registered(efx) ? efx->name : "";
+}
+
+static inline unsigned int efx_port_num(struct efx_nic *efx)
+{
+	return efx->net_dev->dev_id;
+}
+
+/**
+ * struct efx_nic_type - Efx device type definition
+ * @probe: Probe the controller
+ * @remove: Free resources allocated by probe()
+ * @init: Initialise the controller
+ * @fini: Shut down the controller
+ * @monitor: Periodic function for polling link state and hardware monitor
+ * @map_reset_reason: Map ethtool reset reason to a reset method
+ * @map_reset_flags: Map ethtool reset flags to a reset method, if possible
+ * @reset: Reset the controller hardware and possibly the PHY.  This will
+ *	be called while the controller is uninitialised.
+ * @probe_port: Probe the MAC and PHY
+ * @remove_port: Free resources allocated by probe_port()
+ * @handle_global_event: Handle a "global" event (may be %NULL)
+ * @prepare_flush: Prepare the hardware for flushing the DMA queues
+ * @update_stats: Update statistics not provided by event handling
+ * @start_stats: Start the regular fetching of statistics
+ * @stop_stats: Stop the regular fetching of statistics
+ * @set_id_led: Set state of identifying LED or revert to automatic function
+ * @push_irq_moderation: Apply interrupt moderation value
+ * @push_multicast_hash: Apply multicast hash table
+ * @reconfigure_port: Push loopback/power/txdis changes to the MAC and PHY
+ * @get_wol: Get WoL configuration from driver state
+ * @set_wol: Push WoL configuration to the NIC
+ * @resume_wol: Synchronise WoL state between driver and MC (e.g. after resume)
+ * @test_registers: Test read/write functionality of control registers
+ * @test_nvram: Test validity of NVRAM contents
+ * @default_mac_ops: efx_mac_operations to set at startup
+ * @revision: Hardware architecture revision
+ * @mem_map_size: Memory BAR mapped size
+ * @txd_ptr_tbl_base: TX descriptor ring base address
+ * @rxd_ptr_tbl_base: RX descriptor ring base address
+ * @buf_tbl_base: Buffer table base address
+ * @evq_ptr_tbl_base: Event queue pointer table base address
+ * @evq_rptr_tbl_base: Event queue read-pointer table base address
+ * @max_dma_mask: Maximum possible DMA mask
+ * @rx_buffer_hash_size: Size of hash at start of RX buffer
+ * @rx_buffer_padding: Size of padding at end of RX buffer
+ * @max_interrupt_mode: Highest capability interrupt mode supported
+ *	from &enum efx_init_mode.
+ * @phys_addr_channels: Number of channels with physically addressed
+ *	descriptors
+ * @tx_dc_base: Base address in SRAM of TX queue descriptor caches
+ * @rx_dc_base: Base address in SRAM of RX queue descriptor caches
+ * @offload_features: net_device feature flags for protocol offload
+ *	features implemented in hardware
+ */
+struct efx_nic_type {
+	int (*probe)(struct efx_nic *efx);
+	void (*remove)(struct efx_nic *efx);
+	int (*init)(struct efx_nic *efx);
+	void (*fini)(struct efx_nic *efx);
+	void (*monitor)(struct efx_nic *efx);
+	enum reset_type (*map_reset_reason)(enum reset_type reason);
+	int (*map_reset_flags)(u32 *flags);
+	int (*reset)(struct efx_nic *efx, enum reset_type method);
+	int (*probe_port)(struct efx_nic *efx);
+	void (*remove_port)(struct efx_nic *efx);
+	bool (*handle_global_event)(struct efx_channel *channel, efx_qword_t *);
+	void (*prepare_flush)(struct efx_nic *efx);
+	void (*update_stats)(struct efx_nic *efx);
+	void (*start_stats)(struct efx_nic *efx);
+	void (*stop_stats)(struct efx_nic *efx);
+	void (*set_id_led)(struct efx_nic *efx, enum efx_led_mode mode);
+	void (*push_irq_moderation)(struct efx_channel *channel);
+	void (*push_multicast_hash)(struct efx_nic *efx);
+	int (*reconfigure_port)(struct efx_nic *efx);
+	void (*get_wol)(struct efx_nic *efx, struct ethtool_wolinfo *wol);
+	int (*set_wol)(struct efx_nic *efx, u32 type);
+	void (*resume_wol)(struct efx_nic *efx);
+	int (*test_registers)(struct efx_nic *efx);
+	int (*test_nvram)(struct efx_nic *efx);
+	const struct efx_mac_operations *default_mac_ops;
+
+	int revision;
+	unsigned int mem_map_size;
+	unsigned int txd_ptr_tbl_base;
+	unsigned int rxd_ptr_tbl_base;
+	unsigned int buf_tbl_base;
+	unsigned int evq_ptr_tbl_base;
+	unsigned int evq_rptr_tbl_base;
+	u64 max_dma_mask;
+	unsigned int rx_buffer_hash_size;
+	unsigned int rx_buffer_padding;
+	unsigned int max_interrupt_mode;
+	unsigned int phys_addr_channels;
+	unsigned int tx_dc_base;
+	unsigned int rx_dc_base;
+	u32 offload_features;
+};
+
+/**************************************************************************
+ *
+ * Prototypes and inline functions
+ *
+ *************************************************************************/
+
+static inline struct efx_channel *
+efx_get_channel(struct efx_nic *efx, unsigned index)
+{
+	EFX_BUG_ON_PARANOID(index >= efx->n_channels);
+	return efx->channel[index];
+}
+
+/* Iterate over all used channels */
+#define efx_for_each_channel(_channel, _efx)				\
+	for (_channel = (_efx)->channel[0];				\
+	     _channel;							\
+	     _channel = (_channel->channel + 1 < (_efx)->n_channels) ?	\
+		     (_efx)->channel[_channel->channel + 1] : NULL)
+
+static inline struct efx_tx_queue *
+efx_get_tx_queue(struct efx_nic *efx, unsigned index, unsigned type)
+{
+	EFX_BUG_ON_PARANOID(index >= efx->n_tx_channels ||
+			    type >= EFX_TXQ_TYPES);
+	return &efx->channel[efx->tx_channel_offset + index]->tx_queue[type];
+}
+
+static inline bool efx_channel_has_tx_queues(struct efx_channel *channel)
+{
+	return channel->channel - channel->efx->tx_channel_offset <
+		channel->efx->n_tx_channels;
+}
+
+static inline struct efx_tx_queue *
+efx_channel_get_tx_queue(struct efx_channel *channel, unsigned type)
+{
+	EFX_BUG_ON_PARANOID(!efx_channel_has_tx_queues(channel) ||
+			    type >= EFX_TXQ_TYPES);
+	return &channel->tx_queue[type];
+}
+
+static inline bool efx_tx_queue_used(struct efx_tx_queue *tx_queue)
+{
+	return !(tx_queue->efx->net_dev->num_tc < 2 &&
+		 tx_queue->queue & EFX_TXQ_TYPE_HIGHPRI);
+}
+
+/* Iterate over all TX queues belonging to a channel */
+#define efx_for_each_channel_tx_queue(_tx_queue, _channel)		\
+	if (!efx_channel_has_tx_queues(_channel))			\
+		;							\
+	else								\
+		for (_tx_queue = (_channel)->tx_queue;			\
+		     _tx_queue < (_channel)->tx_queue + EFX_TXQ_TYPES && \
+			     efx_tx_queue_used(_tx_queue);		\
+		     _tx_queue++)
+
+/* Iterate over all possible TX queues belonging to a channel */
+#define efx_for_each_possible_channel_tx_queue(_tx_queue, _channel)	\
+	for (_tx_queue = (_channel)->tx_queue;				\
+	     _tx_queue < (_channel)->tx_queue + EFX_TXQ_TYPES;		\
+	     _tx_queue++)
+
+static inline struct efx_rx_queue *
+efx_get_rx_queue(struct efx_nic *efx, unsigned index)
+{
+	EFX_BUG_ON_PARANOID(index >= efx->n_rx_channels);
+	return &efx->channel[index]->rx_queue;
+}
+
+static inline bool efx_channel_has_rx_queue(struct efx_channel *channel)
+{
+	return channel->channel < channel->efx->n_rx_channels;
+}
+
+static inline struct efx_rx_queue *
+efx_channel_get_rx_queue(struct efx_channel *channel)
+{
+	EFX_BUG_ON_PARANOID(!efx_channel_has_rx_queue(channel));
+	return &channel->rx_queue;
+}
+
+/* Iterate over all RX queues belonging to a channel */
+#define efx_for_each_channel_rx_queue(_rx_queue, _channel)		\
+	if (!efx_channel_has_rx_queue(_channel))			\
+		;							\
+	else								\
+		for (_rx_queue = &(_channel)->rx_queue;			\
+		     _rx_queue;						\
+		     _rx_queue = NULL)
+
+static inline struct efx_channel *
+efx_rx_queue_channel(struct efx_rx_queue *rx_queue)
+{
+	return container_of(rx_queue, struct efx_channel, rx_queue);
+}
+
+static inline int efx_rx_queue_index(struct efx_rx_queue *rx_queue)
+{
+	return efx_rx_queue_channel(rx_queue)->channel;
+}
+
+/* Returns a pointer to the specified receive buffer in the RX
+ * descriptor queue.
+ */
+static inline struct efx_rx_buffer *efx_rx_buffer(struct efx_rx_queue *rx_queue,
+						  unsigned int index)
+{
+	return &rx_queue->buffer[index];
+}
+
+/* Set bit in a little-endian bitfield */
+static inline void set_bit_le(unsigned nr, unsigned char *addr)
+{
+	addr[nr / 8] |= (1 << (nr % 8));
+}
+
+/* Clear bit in a little-endian bitfield */
+static inline void clear_bit_le(unsigned nr, unsigned char *addr)
+{
+	addr[nr / 8] &= ~(1 << (nr % 8));
+}
+
+
+/**
+ * EFX_MAX_FRAME_LEN - calculate maximum frame length
+ *
+ * This calculates the maximum frame length that will be used for a
+ * given MTU.  The frame length will be equal to the MTU plus a
+ * constant amount of header space and padding.  This is the quantity
+ * that the net driver will program into the MAC as the maximum frame
+ * length.
+ *
+ * The 10G MAC requires 8-byte alignment on the frame
+ * length, so we round up to the nearest 8.
+ *
+ * Re-clocking by the XGXS on RX can reduce an IPG to 32 bits (half an
+ * XGMII cycle).  If the frame length reaches the maximum value in the
+ * same cycle, the XMAC can miss the IPG altogether.  We work around
+ * this by adding a further 16 bytes.
+ */
+#define EFX_MAX_FRAME_LEN(mtu) \
+	((((mtu) + ETH_HLEN + VLAN_HLEN + 4/* FCS */ + 7) & ~7) + 16)
+
+
+#endif /* EFX_NET_DRIVER_H */
diff --git a/drivers/net/ethernet/sfc/nic.c b/drivers/net/ethernet/sfc/nic.c
new file mode 100644
index 0000000..bafa23a
--- /dev/null
+++ b/drivers/net/ethernet/sfc/nic.c
@@ -0,0 +1,1969 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2005-2006 Fen Systems Ltd.
+ * Copyright 2006-2011 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#include <linux/bitops.h>
+#include <linux/delay.h>
+#include <linux/interrupt.h>
+#include <linux/pci.h>
+#include <linux/module.h>
+#include <linux/seq_file.h>
+#include "net_driver.h"
+#include "bitfield.h"
+#include "efx.h"
+#include "nic.h"
+#include "regs.h"
+#include "io.h"
+#include "workarounds.h"
+
+/**************************************************************************
+ *
+ * Configurable values
+ *
+ **************************************************************************
+ */
+
+/* This is set to 16 for a good reason.  In summary, if larger than
+ * 16, the descriptor cache holds more than a default socket
+ * buffer's worth of packets (for UDP we can only have at most one
+ * socket buffer's worth outstanding).  This combined with the fact
+ * that we only get 1 TX event per descriptor cache means the NIC
+ * goes idle.
+ */
+#define TX_DC_ENTRIES 16
+#define TX_DC_ENTRIES_ORDER 1
+
+#define RX_DC_ENTRIES 64
+#define RX_DC_ENTRIES_ORDER 3
+
+/* If EFX_MAX_INT_ERRORS internal errors occur within
+ * EFX_INT_ERROR_EXPIRE seconds, we consider the NIC broken and
+ * disable it.
+ */
+#define EFX_INT_ERROR_EXPIRE 3600
+#define EFX_MAX_INT_ERRORS 5
+
+/* We poll for events every FLUSH_INTERVAL ms, and check FLUSH_POLL_COUNT times
+ */
+#define EFX_FLUSH_INTERVAL 10
+#define EFX_FLUSH_POLL_COUNT 100
+
+/* Size and alignment of special buffers (4KB) */
+#define EFX_BUF_SIZE 4096
+
+/* Depth of RX flush request fifo */
+#define EFX_RX_FLUSH_COUNT 4
+
+/* Generated event code for efx_generate_test_event() */
+#define EFX_CHANNEL_MAGIC_TEST(_channel)	\
+	(0x00010100 + (_channel)->channel)
+
+/* Generated event code for efx_generate_fill_event() */
+#define EFX_CHANNEL_MAGIC_FILL(_channel)	\
+	(0x00010200 + (_channel)->channel)
+
+/**************************************************************************
+ *
+ * Solarstorm hardware access
+ *
+ **************************************************************************/
+
+static inline void efx_write_buf_tbl(struct efx_nic *efx, efx_qword_t *value,
+				     unsigned int index)
+{
+	efx_sram_writeq(efx, efx->membase + efx->type->buf_tbl_base,
+			value, index);
+}
+
+/* Read the current event from the event queue */
+static inline efx_qword_t *efx_event(struct efx_channel *channel,
+				     unsigned int index)
+{
+	return ((efx_qword_t *) (channel->eventq.addr)) +
+		(index & channel->eventq_mask);
+}
+
+/* See if an event is present
+ *
+ * We check both the high and low dword of the event for all ones.  We
+ * wrote all ones when we cleared the event, and no valid event can
+ * have all ones in either its high or low dwords.  This approach is
+ * robust against reordering.
+ *
+ * Note that using a single 64-bit comparison is incorrect; even
+ * though the CPU read will be atomic, the DMA write may not be.
+ */
+static inline int efx_event_present(efx_qword_t *event)
+{
+	return !(EFX_DWORD_IS_ALL_ONES(event->dword[0]) |
+		  EFX_DWORD_IS_ALL_ONES(event->dword[1]));
+}
+
+static bool efx_masked_compare_oword(const efx_oword_t *a, const efx_oword_t *b,
+				     const efx_oword_t *mask)
+{
+	return ((a->u64[0] ^ b->u64[0]) & mask->u64[0]) ||
+		((a->u64[1] ^ b->u64[1]) & mask->u64[1]);
+}
+
+int efx_nic_test_registers(struct efx_nic *efx,
+			   const struct efx_nic_register_test *regs,
+			   size_t n_regs)
+{
+	unsigned address = 0, i, j;
+	efx_oword_t mask, imask, original, reg, buf;
+
+	/* Falcon should be in loopback to isolate the XMAC from the PHY */
+	WARN_ON(!LOOPBACK_INTERNAL(efx));
+
+	for (i = 0; i < n_regs; ++i) {
+		address = regs[i].address;
+		mask = imask = regs[i].mask;
+		EFX_INVERT_OWORD(imask);
+
+		efx_reado(efx, &original, address);
+
+		/* bit sweep on and off */
+		for (j = 0; j < 128; j++) {
+			if (!EFX_EXTRACT_OWORD32(mask, j, j))
+				continue;
+
+			/* Test this testable bit can be set in isolation */
+			EFX_AND_OWORD(reg, original, mask);
+			EFX_SET_OWORD32(reg, j, j, 1);
+
+			efx_writeo(efx, &reg, address);
+			efx_reado(efx, &buf, address);
+
+			if (efx_masked_compare_oword(&reg, &buf, &mask))
+				goto fail;
+
+			/* Test this testable bit can be cleared in isolation */
+			EFX_OR_OWORD(reg, original, mask);
+			EFX_SET_OWORD32(reg, j, j, 0);
+
+			efx_writeo(efx, &reg, address);
+			efx_reado(efx, &buf, address);
+
+			if (efx_masked_compare_oword(&reg, &buf, &mask))
+				goto fail;
+		}
+
+		efx_writeo(efx, &original, address);
+	}
+
+	return 0;
+
+fail:
+	netif_err(efx, hw, efx->net_dev,
+		  "wrote "EFX_OWORD_FMT" read "EFX_OWORD_FMT
+		  " at address 0x%x mask "EFX_OWORD_FMT"\n", EFX_OWORD_VAL(reg),
+		  EFX_OWORD_VAL(buf), address, EFX_OWORD_VAL(mask));
+	return -EIO;
+}
+
+/**************************************************************************
+ *
+ * Special buffer handling
+ * Special buffers are used for event queues and the TX and RX
+ * descriptor rings.
+ *
+ *************************************************************************/
+
+/*
+ * Initialise a special buffer
+ *
+ * This will define a buffer (previously allocated via
+ * efx_alloc_special_buffer()) in the buffer table, allowing
+ * it to be used for event queues, descriptor rings etc.
+ */
+static void
+efx_init_special_buffer(struct efx_nic *efx, struct efx_special_buffer *buffer)
+{
+	efx_qword_t buf_desc;
+	int index;
+	dma_addr_t dma_addr;
+	int i;
+
+	EFX_BUG_ON_PARANOID(!buffer->addr);
+
+	/* Write buffer descriptors to NIC */
+	for (i = 0; i < buffer->entries; i++) {
+		index = buffer->index + i;
+		dma_addr = buffer->dma_addr + (i * 4096);
+		netif_dbg(efx, probe, efx->net_dev,
+			  "mapping special buffer %d at %llx\n",
+			  index, (unsigned long long)dma_addr);
+		EFX_POPULATE_QWORD_3(buf_desc,
+				     FRF_AZ_BUF_ADR_REGION, 0,
+				     FRF_AZ_BUF_ADR_FBUF, dma_addr >> 12,
+				     FRF_AZ_BUF_OWNER_ID_FBUF, 0);
+		efx_write_buf_tbl(efx, &buf_desc, index);
+	}
+}
+
+/* Unmaps a buffer and clears the buffer table entries */
+static void
+efx_fini_special_buffer(struct efx_nic *efx, struct efx_special_buffer *buffer)
+{
+	efx_oword_t buf_tbl_upd;
+	unsigned int start = buffer->index;
+	unsigned int end = (buffer->index + buffer->entries - 1);
+
+	if (!buffer->entries)
+		return;
+
+	netif_dbg(efx, hw, efx->net_dev, "unmapping special buffers %d-%d\n",
+		  buffer->index, buffer->index + buffer->entries - 1);
+
+	EFX_POPULATE_OWORD_4(buf_tbl_upd,
+			     FRF_AZ_BUF_UPD_CMD, 0,
+			     FRF_AZ_BUF_CLR_CMD, 1,
+			     FRF_AZ_BUF_CLR_END_ID, end,
+			     FRF_AZ_BUF_CLR_START_ID, start);
+	efx_writeo(efx, &buf_tbl_upd, FR_AZ_BUF_TBL_UPD);
+}
+
+/*
+ * Allocate a new special buffer
+ *
+ * This allocates memory for a new buffer, clears it and allocates a
+ * new buffer ID range.  It does not write into the buffer table.
+ *
+ * This call will allocate 4KB buffers, since 8KB buffers can't be
+ * used for event queues and descriptor rings.
+ */
+static int efx_alloc_special_buffer(struct efx_nic *efx,
+				    struct efx_special_buffer *buffer,
+				    unsigned int len)
+{
+	len = ALIGN(len, EFX_BUF_SIZE);
+
+	buffer->addr = dma_alloc_coherent(&efx->pci_dev->dev, len,
+					  &buffer->dma_addr, GFP_KERNEL);
+	if (!buffer->addr)
+		return -ENOMEM;
+	buffer->len = len;
+	buffer->entries = len / EFX_BUF_SIZE;
+	BUG_ON(buffer->dma_addr & (EFX_BUF_SIZE - 1));
+
+	/* All zeros is a potentially valid event so memset to 0xff */
+	memset(buffer->addr, 0xff, len);
+
+	/* Select new buffer ID */
+	buffer->index = efx->next_buffer_table;
+	efx->next_buffer_table += buffer->entries;
+
+	netif_dbg(efx, probe, efx->net_dev,
+		  "allocating special buffers %d-%d at %llx+%x "
+		  "(virt %p phys %llx)\n", buffer->index,
+		  buffer->index + buffer->entries - 1,
+		  (u64)buffer->dma_addr, len,
+		  buffer->addr, (u64)virt_to_phys(buffer->addr));
+
+	return 0;
+}
+
+static void
+efx_free_special_buffer(struct efx_nic *efx, struct efx_special_buffer *buffer)
+{
+	if (!buffer->addr)
+		return;
+
+	netif_dbg(efx, hw, efx->net_dev,
+		  "deallocating special buffers %d-%d at %llx+%x "
+		  "(virt %p phys %llx)\n", buffer->index,
+		  buffer->index + buffer->entries - 1,
+		  (u64)buffer->dma_addr, buffer->len,
+		  buffer->addr, (u64)virt_to_phys(buffer->addr));
+
+	dma_free_coherent(&efx->pci_dev->dev, buffer->len, buffer->addr,
+			  buffer->dma_addr);
+	buffer->addr = NULL;
+	buffer->entries = 0;
+}
+
+/**************************************************************************
+ *
+ * Generic buffer handling
+ * These buffers are used for interrupt status and MAC stats
+ *
+ **************************************************************************/
+
+int efx_nic_alloc_buffer(struct efx_nic *efx, struct efx_buffer *buffer,
+			 unsigned int len)
+{
+	buffer->addr = pci_alloc_consistent(efx->pci_dev, len,
+					    &buffer->dma_addr);
+	if (!buffer->addr)
+		return -ENOMEM;
+	buffer->len = len;
+	memset(buffer->addr, 0, len);
+	return 0;
+}
+
+void efx_nic_free_buffer(struct efx_nic *efx, struct efx_buffer *buffer)
+{
+	if (buffer->addr) {
+		pci_free_consistent(efx->pci_dev, buffer->len,
+				    buffer->addr, buffer->dma_addr);
+		buffer->addr = NULL;
+	}
+}
+
+/**************************************************************************
+ *
+ * TX path
+ *
+ **************************************************************************/
+
+/* Returns a pointer to the specified transmit descriptor in the TX
+ * descriptor queue belonging to the specified channel.
+ */
+static inline efx_qword_t *
+efx_tx_desc(struct efx_tx_queue *tx_queue, unsigned int index)
+{
+	return ((efx_qword_t *) (tx_queue->txd.addr)) + index;
+}
+
+/* This writes to the TX_DESC_WPTR; write pointer for TX descriptor ring */
+static inline void efx_notify_tx_desc(struct efx_tx_queue *tx_queue)
+{
+	unsigned write_ptr;
+	efx_dword_t reg;
+
+	write_ptr = tx_queue->write_count & tx_queue->ptr_mask;
+	EFX_POPULATE_DWORD_1(reg, FRF_AZ_TX_DESC_WPTR_DWORD, write_ptr);
+	efx_writed_page(tx_queue->efx, &reg,
+			FR_AZ_TX_DESC_UPD_DWORD_P0, tx_queue->queue);
+}
+
+/* Write pointer and first descriptor for TX descriptor ring */
+static inline void efx_push_tx_desc(struct efx_tx_queue *tx_queue,
+				    const efx_qword_t *txd)
+{
+	unsigned write_ptr;
+	efx_oword_t reg;
+
+	BUILD_BUG_ON(FRF_AZ_TX_DESC_LBN != 0);
+	BUILD_BUG_ON(FR_AA_TX_DESC_UPD_KER != FR_BZ_TX_DESC_UPD_P0);
+
+	write_ptr = tx_queue->write_count & tx_queue->ptr_mask;
+	EFX_POPULATE_OWORD_2(reg, FRF_AZ_TX_DESC_PUSH_CMD, true,
+			     FRF_AZ_TX_DESC_WPTR, write_ptr);
+	reg.qword[0] = *txd;
+	efx_writeo_page(tx_queue->efx, &reg,
+			FR_BZ_TX_DESC_UPD_P0, tx_queue->queue);
+}
+
+static inline bool
+efx_may_push_tx_desc(struct efx_tx_queue *tx_queue, unsigned int write_count)
+{
+	unsigned empty_read_count = ACCESS_ONCE(tx_queue->empty_read_count);
+
+	if (empty_read_count == 0)
+		return false;
+
+	tx_queue->empty_read_count = 0;
+	return ((empty_read_count ^ write_count) & ~EFX_EMPTY_COUNT_VALID) == 0;
+}
+
+/* For each entry inserted into the software descriptor ring, create a
+ * descriptor in the hardware TX descriptor ring (in host memory), and
+ * write a doorbell.
+ */
+void efx_nic_push_buffers(struct efx_tx_queue *tx_queue)
+{
+
+	struct efx_tx_buffer *buffer;
+	efx_qword_t *txd;
+	unsigned write_ptr;
+	unsigned old_write_count = tx_queue->write_count;
+
+	BUG_ON(tx_queue->write_count == tx_queue->insert_count);
+
+	do {
+		write_ptr = tx_queue->write_count & tx_queue->ptr_mask;
+		buffer = &tx_queue->buffer[write_ptr];
+		txd = efx_tx_desc(tx_queue, write_ptr);
+		++tx_queue->write_count;
+
+		/* Create TX descriptor ring entry */
+		EFX_POPULATE_QWORD_4(*txd,
+				     FSF_AZ_TX_KER_CONT, buffer->continuation,
+				     FSF_AZ_TX_KER_BYTE_COUNT, buffer->len,
+				     FSF_AZ_TX_KER_BUF_REGION, 0,
+				     FSF_AZ_TX_KER_BUF_ADDR, buffer->dma_addr);
+	} while (tx_queue->write_count != tx_queue->insert_count);
+
+	wmb(); /* Ensure descriptors are written before they are fetched */
+
+	if (efx_may_push_tx_desc(tx_queue, old_write_count)) {
+		txd = efx_tx_desc(tx_queue,
+				  old_write_count & tx_queue->ptr_mask);
+		efx_push_tx_desc(tx_queue, txd);
+		++tx_queue->pushes;
+	} else {
+		efx_notify_tx_desc(tx_queue);
+	}
+}
+
+/* Allocate hardware resources for a TX queue */
+int efx_nic_probe_tx(struct efx_tx_queue *tx_queue)
+{
+	struct efx_nic *efx = tx_queue->efx;
+	unsigned entries;
+
+	entries = tx_queue->ptr_mask + 1;
+	return efx_alloc_special_buffer(efx, &tx_queue->txd,
+					entries * sizeof(efx_qword_t));
+}
+
+void efx_nic_init_tx(struct efx_tx_queue *tx_queue)
+{
+	struct efx_nic *efx = tx_queue->efx;
+	efx_oword_t reg;
+
+	tx_queue->flushed = FLUSH_NONE;
+
+	/* Pin TX descriptor ring */
+	efx_init_special_buffer(efx, &tx_queue->txd);
+
+	/* Push TX descriptor ring to card */
+	EFX_POPULATE_OWORD_10(reg,
+			      FRF_AZ_TX_DESCQ_EN, 1,
+			      FRF_AZ_TX_ISCSI_DDIG_EN, 0,
+			      FRF_AZ_TX_ISCSI_HDIG_EN, 0,
+			      FRF_AZ_TX_DESCQ_BUF_BASE_ID, tx_queue->txd.index,
+			      FRF_AZ_TX_DESCQ_EVQ_ID,
+			      tx_queue->channel->channel,
+			      FRF_AZ_TX_DESCQ_OWNER_ID, 0,
+			      FRF_AZ_TX_DESCQ_LABEL, tx_queue->queue,
+			      FRF_AZ_TX_DESCQ_SIZE,
+			      __ffs(tx_queue->txd.entries),
+			      FRF_AZ_TX_DESCQ_TYPE, 0,
+			      FRF_BZ_TX_NON_IP_DROP_DIS, 1);
+
+	if (efx_nic_rev(efx) >= EFX_REV_FALCON_B0) {
+		int csum = tx_queue->queue & EFX_TXQ_TYPE_OFFLOAD;
+		EFX_SET_OWORD_FIELD(reg, FRF_BZ_TX_IP_CHKSM_DIS, !csum);
+		EFX_SET_OWORD_FIELD(reg, FRF_BZ_TX_TCP_CHKSM_DIS,
+				    !csum);
+	}
+
+	efx_writeo_table(efx, &reg, efx->type->txd_ptr_tbl_base,
+			 tx_queue->queue);
+
+	if (efx_nic_rev(efx) < EFX_REV_FALCON_B0) {
+		/* Only 128 bits in this register */
+		BUILD_BUG_ON(EFX_MAX_TX_QUEUES > 128);
+
+		efx_reado(efx, &reg, FR_AA_TX_CHKSM_CFG);
+		if (tx_queue->queue & EFX_TXQ_TYPE_OFFLOAD)
+			clear_bit_le(tx_queue->queue, (void *)&reg);
+		else
+			set_bit_le(tx_queue->queue, (void *)&reg);
+		efx_writeo(efx, &reg, FR_AA_TX_CHKSM_CFG);
+	}
+
+	if (efx_nic_rev(efx) >= EFX_REV_FALCON_B0) {
+		EFX_POPULATE_OWORD_1(reg,
+				     FRF_BZ_TX_PACE,
+				     (tx_queue->queue & EFX_TXQ_TYPE_HIGHPRI) ?
+				     FFE_BZ_TX_PACE_OFF :
+				     FFE_BZ_TX_PACE_RESERVED);
+		efx_writeo_table(efx, &reg, FR_BZ_TX_PACE_TBL,
+				 tx_queue->queue);
+	}
+}
+
+static void efx_flush_tx_queue(struct efx_tx_queue *tx_queue)
+{
+	struct efx_nic *efx = tx_queue->efx;
+	efx_oword_t tx_flush_descq;
+
+	tx_queue->flushed = FLUSH_PENDING;
+
+	/* Post a flush command */
+	EFX_POPULATE_OWORD_2(tx_flush_descq,
+			     FRF_AZ_TX_FLUSH_DESCQ_CMD, 1,
+			     FRF_AZ_TX_FLUSH_DESCQ, tx_queue->queue);
+	efx_writeo(efx, &tx_flush_descq, FR_AZ_TX_FLUSH_DESCQ);
+}
+
+void efx_nic_fini_tx(struct efx_tx_queue *tx_queue)
+{
+	struct efx_nic *efx = tx_queue->efx;
+	efx_oword_t tx_desc_ptr;
+
+	/* The queue should have been flushed */
+	WARN_ON(tx_queue->flushed != FLUSH_DONE);
+
+	/* Remove TX descriptor ring from card */
+	EFX_ZERO_OWORD(tx_desc_ptr);
+	efx_writeo_table(efx, &tx_desc_ptr, efx->type->txd_ptr_tbl_base,
+			 tx_queue->queue);
+
+	/* Unpin TX descriptor ring */
+	efx_fini_special_buffer(efx, &tx_queue->txd);
+}
+
+/* Free buffers backing TX queue */
+void efx_nic_remove_tx(struct efx_tx_queue *tx_queue)
+{
+	efx_free_special_buffer(tx_queue->efx, &tx_queue->txd);
+}
+
+/**************************************************************************
+ *
+ * RX path
+ *
+ **************************************************************************/
+
+/* Returns a pointer to the specified descriptor in the RX descriptor queue */
+static inline efx_qword_t *
+efx_rx_desc(struct efx_rx_queue *rx_queue, unsigned int index)
+{
+	return ((efx_qword_t *) (rx_queue->rxd.addr)) + index;
+}
+
+/* This creates an entry in the RX descriptor queue */
+static inline void
+efx_build_rx_desc(struct efx_rx_queue *rx_queue, unsigned index)
+{
+	struct efx_rx_buffer *rx_buf;
+	efx_qword_t *rxd;
+
+	rxd = efx_rx_desc(rx_queue, index);
+	rx_buf = efx_rx_buffer(rx_queue, index);
+	EFX_POPULATE_QWORD_3(*rxd,
+			     FSF_AZ_RX_KER_BUF_SIZE,
+			     rx_buf->len -
+			     rx_queue->efx->type->rx_buffer_padding,
+			     FSF_AZ_RX_KER_BUF_REGION, 0,
+			     FSF_AZ_RX_KER_BUF_ADDR, rx_buf->dma_addr);
+}
+
+/* This writes to the RX_DESC_WPTR register for the specified receive
+ * descriptor ring.
+ */
+void efx_nic_notify_rx_desc(struct efx_rx_queue *rx_queue)
+{
+	struct efx_nic *efx = rx_queue->efx;
+	efx_dword_t reg;
+	unsigned write_ptr;
+
+	while (rx_queue->notified_count != rx_queue->added_count) {
+		efx_build_rx_desc(
+			rx_queue,
+			rx_queue->notified_count & rx_queue->ptr_mask);
+		++rx_queue->notified_count;
+	}
+
+	wmb();
+	write_ptr = rx_queue->added_count & rx_queue->ptr_mask;
+	EFX_POPULATE_DWORD_1(reg, FRF_AZ_RX_DESC_WPTR_DWORD, write_ptr);
+	efx_writed_page(efx, &reg, FR_AZ_RX_DESC_UPD_DWORD_P0,
+			efx_rx_queue_index(rx_queue));
+}
+
+int efx_nic_probe_rx(struct efx_rx_queue *rx_queue)
+{
+	struct efx_nic *efx = rx_queue->efx;
+	unsigned entries;
+
+	entries = rx_queue->ptr_mask + 1;
+	return efx_alloc_special_buffer(efx, &rx_queue->rxd,
+					entries * sizeof(efx_qword_t));
+}
+
+void efx_nic_init_rx(struct efx_rx_queue *rx_queue)
+{
+	efx_oword_t rx_desc_ptr;
+	struct efx_nic *efx = rx_queue->efx;
+	bool is_b0 = efx_nic_rev(efx) >= EFX_REV_FALCON_B0;
+	bool iscsi_digest_en = is_b0;
+
+	netif_dbg(efx, hw, efx->net_dev,
+		  "RX queue %d ring in special buffers %d-%d\n",
+		  efx_rx_queue_index(rx_queue), rx_queue->rxd.index,
+		  rx_queue->rxd.index + rx_queue->rxd.entries - 1);
+
+	rx_queue->flushed = FLUSH_NONE;
+
+	/* Pin RX descriptor ring */
+	efx_init_special_buffer(efx, &rx_queue->rxd);
+
+	/* Push RX descriptor ring to card */
+	EFX_POPULATE_OWORD_10(rx_desc_ptr,
+			      FRF_AZ_RX_ISCSI_DDIG_EN, iscsi_digest_en,
+			      FRF_AZ_RX_ISCSI_HDIG_EN, iscsi_digest_en,
+			      FRF_AZ_RX_DESCQ_BUF_BASE_ID, rx_queue->rxd.index,
+			      FRF_AZ_RX_DESCQ_EVQ_ID,
+			      efx_rx_queue_channel(rx_queue)->channel,
+			      FRF_AZ_RX_DESCQ_OWNER_ID, 0,
+			      FRF_AZ_RX_DESCQ_LABEL,
+			      efx_rx_queue_index(rx_queue),
+			      FRF_AZ_RX_DESCQ_SIZE,
+			      __ffs(rx_queue->rxd.entries),
+			      FRF_AZ_RX_DESCQ_TYPE, 0 /* kernel queue */ ,
+			      /* For >=B0 this is scatter so disable */
+			      FRF_AZ_RX_DESCQ_JUMBO, !is_b0,
+			      FRF_AZ_RX_DESCQ_EN, 1);
+	efx_writeo_table(efx, &rx_desc_ptr, efx->type->rxd_ptr_tbl_base,
+			 efx_rx_queue_index(rx_queue));
+}
+
+static void efx_flush_rx_queue(struct efx_rx_queue *rx_queue)
+{
+	struct efx_nic *efx = rx_queue->efx;
+	efx_oword_t rx_flush_descq;
+
+	rx_queue->flushed = FLUSH_PENDING;
+
+	/* Post a flush command */
+	EFX_POPULATE_OWORD_2(rx_flush_descq,
+			     FRF_AZ_RX_FLUSH_DESCQ_CMD, 1,
+			     FRF_AZ_RX_FLUSH_DESCQ,
+			     efx_rx_queue_index(rx_queue));
+	efx_writeo(efx, &rx_flush_descq, FR_AZ_RX_FLUSH_DESCQ);
+}
+
+void efx_nic_fini_rx(struct efx_rx_queue *rx_queue)
+{
+	efx_oword_t rx_desc_ptr;
+	struct efx_nic *efx = rx_queue->efx;
+
+	/* The queue should already have been flushed */
+	WARN_ON(rx_queue->flushed != FLUSH_DONE);
+
+	/* Remove RX descriptor ring from card */
+	EFX_ZERO_OWORD(rx_desc_ptr);
+	efx_writeo_table(efx, &rx_desc_ptr, efx->type->rxd_ptr_tbl_base,
+			 efx_rx_queue_index(rx_queue));
+
+	/* Unpin RX descriptor ring */
+	efx_fini_special_buffer(efx, &rx_queue->rxd);
+}
+
+/* Free buffers backing RX queue */
+void efx_nic_remove_rx(struct efx_rx_queue *rx_queue)
+{
+	efx_free_special_buffer(rx_queue->efx, &rx_queue->rxd);
+}
+
+/**************************************************************************
+ *
+ * Event queue processing
+ * Event queues are processed by per-channel tasklets.
+ *
+ **************************************************************************/
+
+/* Update a channel's event queue's read pointer (RPTR) register
+ *
+ * This writes the EVQ_RPTR_REG register for the specified channel's
+ * event queue.
+ */
+void efx_nic_eventq_read_ack(struct efx_channel *channel)
+{
+	efx_dword_t reg;
+	struct efx_nic *efx = channel->efx;
+
+	EFX_POPULATE_DWORD_1(reg, FRF_AZ_EVQ_RPTR,
+			     channel->eventq_read_ptr & channel->eventq_mask);
+	efx_writed_table(efx, &reg, efx->type->evq_rptr_tbl_base,
+			 channel->channel);
+}
+
+/* Use HW to insert a SW defined event */
+static void efx_generate_event(struct efx_channel *channel, efx_qword_t *event)
+{
+	efx_oword_t drv_ev_reg;
+
+	BUILD_BUG_ON(FRF_AZ_DRV_EV_DATA_LBN != 0 ||
+		     FRF_AZ_DRV_EV_DATA_WIDTH != 64);
+	drv_ev_reg.u32[0] = event->u32[0];
+	drv_ev_reg.u32[1] = event->u32[1];
+	drv_ev_reg.u32[2] = 0;
+	drv_ev_reg.u32[3] = 0;
+	EFX_SET_OWORD_FIELD(drv_ev_reg, FRF_AZ_DRV_EV_QID, channel->channel);
+	efx_writeo(channel->efx, &drv_ev_reg, FR_AZ_DRV_EV);
+}
+
+/* Handle a transmit completion event
+ *
+ * The NIC batches TX completion events; the message we receive is of
+ * the form "complete all TX events up to this index".
+ */
+static int
+efx_handle_tx_event(struct efx_channel *channel, efx_qword_t *event)
+{
+	unsigned int tx_ev_desc_ptr;
+	unsigned int tx_ev_q_label;
+	struct efx_tx_queue *tx_queue;
+	struct efx_nic *efx = channel->efx;
+	int tx_packets = 0;
+
+	if (likely(EFX_QWORD_FIELD(*event, FSF_AZ_TX_EV_COMP))) {
+		/* Transmit completion */
+		tx_ev_desc_ptr = EFX_QWORD_FIELD(*event, FSF_AZ_TX_EV_DESC_PTR);
+		tx_ev_q_label = EFX_QWORD_FIELD(*event, FSF_AZ_TX_EV_Q_LABEL);
+		tx_queue = efx_channel_get_tx_queue(
+			channel, tx_ev_q_label % EFX_TXQ_TYPES);
+		tx_packets = ((tx_ev_desc_ptr - tx_queue->read_count) &
+			      tx_queue->ptr_mask);
+		channel->irq_mod_score += tx_packets;
+		efx_xmit_done(tx_queue, tx_ev_desc_ptr);
+	} else if (EFX_QWORD_FIELD(*event, FSF_AZ_TX_EV_WQ_FF_FULL)) {
+		/* Rewrite the FIFO write pointer */
+		tx_ev_q_label = EFX_QWORD_FIELD(*event, FSF_AZ_TX_EV_Q_LABEL);
+		tx_queue = efx_channel_get_tx_queue(
+			channel, tx_ev_q_label % EFX_TXQ_TYPES);
+
+		if (efx_dev_registered(efx))
+			netif_tx_lock(efx->net_dev);
+		efx_notify_tx_desc(tx_queue);
+		if (efx_dev_registered(efx))
+			netif_tx_unlock(efx->net_dev);
+	} else if (EFX_QWORD_FIELD(*event, FSF_AZ_TX_EV_PKT_ERR) &&
+		   EFX_WORKAROUND_10727(efx)) {
+		efx_schedule_reset(efx, RESET_TYPE_TX_DESC_FETCH);
+	} else {
+		netif_err(efx, tx_err, efx->net_dev,
+			  "channel %d unexpected TX event "
+			  EFX_QWORD_FMT"\n", channel->channel,
+			  EFX_QWORD_VAL(*event));
+	}
+
+	return tx_packets;
+}
+
+/* Detect errors included in the rx_evt_pkt_ok bit. */
+static void efx_handle_rx_not_ok(struct efx_rx_queue *rx_queue,
+				 const efx_qword_t *event,
+				 bool *rx_ev_pkt_ok,
+				 bool *discard)
+{
+	struct efx_channel *channel = efx_rx_queue_channel(rx_queue);
+	struct efx_nic *efx = rx_queue->efx;
+	bool rx_ev_buf_owner_id_err, rx_ev_ip_hdr_chksum_err;
+	bool rx_ev_tcp_udp_chksum_err, rx_ev_eth_crc_err;
+	bool rx_ev_frm_trunc, rx_ev_drib_nib, rx_ev_tobe_disc;
+	bool rx_ev_other_err, rx_ev_pause_frm;
+	bool rx_ev_hdr_type, rx_ev_mcast_pkt;
+	unsigned rx_ev_pkt_type;
+
+	rx_ev_hdr_type = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_HDR_TYPE);
+	rx_ev_mcast_pkt = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_MCAST_PKT);
+	rx_ev_tobe_disc = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_TOBE_DISC);
+	rx_ev_pkt_type = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_PKT_TYPE);
+	rx_ev_buf_owner_id_err = EFX_QWORD_FIELD(*event,
+						 FSF_AZ_RX_EV_BUF_OWNER_ID_ERR);
+	rx_ev_ip_hdr_chksum_err = EFX_QWORD_FIELD(*event,
+						  FSF_AZ_RX_EV_IP_HDR_CHKSUM_ERR);
+	rx_ev_tcp_udp_chksum_err = EFX_QWORD_FIELD(*event,
+						   FSF_AZ_RX_EV_TCP_UDP_CHKSUM_ERR);
+	rx_ev_eth_crc_err = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_ETH_CRC_ERR);
+	rx_ev_frm_trunc = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_FRM_TRUNC);
+	rx_ev_drib_nib = ((efx_nic_rev(efx) >= EFX_REV_FALCON_B0) ?
+			  0 : EFX_QWORD_FIELD(*event, FSF_AA_RX_EV_DRIB_NIB));
+	rx_ev_pause_frm = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_PAUSE_FRM_ERR);
+
+	/* Every error apart from tobe_disc and pause_frm */
+	rx_ev_other_err = (rx_ev_drib_nib | rx_ev_tcp_udp_chksum_err |
+			   rx_ev_buf_owner_id_err | rx_ev_eth_crc_err |
+			   rx_ev_frm_trunc | rx_ev_ip_hdr_chksum_err);
+
+	/* Count errors that are not in MAC stats.  Ignore expected
+	 * checksum errors during self-test. */
+	if (rx_ev_frm_trunc)
+		++channel->n_rx_frm_trunc;
+	else if (rx_ev_tobe_disc)
+		++channel->n_rx_tobe_disc;
+	else if (!efx->loopback_selftest) {
+		if (rx_ev_ip_hdr_chksum_err)
+			++channel->n_rx_ip_hdr_chksum_err;
+		else if (rx_ev_tcp_udp_chksum_err)
+			++channel->n_rx_tcp_udp_chksum_err;
+	}
+
+	/* The frame must be discarded if any of these are true. */
+	*discard = (rx_ev_eth_crc_err | rx_ev_frm_trunc | rx_ev_drib_nib |
+		    rx_ev_tobe_disc | rx_ev_pause_frm);
+
+	/* TOBE_DISC is expected on unicast mismatches; don't print out an
+	 * error message.  FRM_TRUNC indicates RXDP dropped the packet due
+	 * to a FIFO overflow.
+	 */
+#ifdef EFX_ENABLE_DEBUG
+	if (rx_ev_other_err && net_ratelimit()) {
+		netif_dbg(efx, rx_err, efx->net_dev,
+			  " RX queue %d unexpected RX event "
+			  EFX_QWORD_FMT "%s%s%s%s%s%s%s%s\n",
+			  efx_rx_queue_index(rx_queue), EFX_QWORD_VAL(*event),
+			  rx_ev_buf_owner_id_err ? " [OWNER_ID_ERR]" : "",
+			  rx_ev_ip_hdr_chksum_err ?
+			  " [IP_HDR_CHKSUM_ERR]" : "",
+			  rx_ev_tcp_udp_chksum_err ?
+			  " [TCP_UDP_CHKSUM_ERR]" : "",
+			  rx_ev_eth_crc_err ? " [ETH_CRC_ERR]" : "",
+			  rx_ev_frm_trunc ? " [FRM_TRUNC]" : "",
+			  rx_ev_drib_nib ? " [DRIB_NIB]" : "",
+			  rx_ev_tobe_disc ? " [TOBE_DISC]" : "",
+			  rx_ev_pause_frm ? " [PAUSE]" : "");
+	}
+#endif
+}
+
+/* Handle receive events that are not in-order. */
+static void
+efx_handle_rx_bad_index(struct efx_rx_queue *rx_queue, unsigned index)
+{
+	struct efx_nic *efx = rx_queue->efx;
+	unsigned expected, dropped;
+
+	expected = rx_queue->removed_count & rx_queue->ptr_mask;
+	dropped = (index - expected) & rx_queue->ptr_mask;
+	netif_info(efx, rx_err, efx->net_dev,
+		   "dropped %d events (index=%d expected=%d)\n",
+		   dropped, index, expected);
+
+	efx_schedule_reset(efx, EFX_WORKAROUND_5676(efx) ?
+			   RESET_TYPE_RX_RECOVERY : RESET_TYPE_DISABLE);
+}
+
+/* Handle a packet received event
+ *
+ * The NIC gives a "discard" flag if it's a unicast packet with the
+ * wrong destination address
+ * Also "is multicast" and "matches multicast filter" flags can be used to
+ * discard non-matching multicast packets.
+ */
+static void
+efx_handle_rx_event(struct efx_channel *channel, const efx_qword_t *event)
+{
+	unsigned int rx_ev_desc_ptr, rx_ev_byte_cnt;
+	unsigned int rx_ev_hdr_type, rx_ev_mcast_pkt;
+	unsigned expected_ptr;
+	bool rx_ev_pkt_ok, discard = false, checksummed;
+	struct efx_rx_queue *rx_queue;
+
+	/* Basic packet information */
+	rx_ev_byte_cnt = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_BYTE_CNT);
+	rx_ev_pkt_ok = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_PKT_OK);
+	rx_ev_hdr_type = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_HDR_TYPE);
+	WARN_ON(EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_JUMBO_CONT));
+	WARN_ON(EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_SOP) != 1);
+	WARN_ON(EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_Q_LABEL) !=
+		channel->channel);
+
+	rx_queue = efx_channel_get_rx_queue(channel);
+
+	rx_ev_desc_ptr = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_DESC_PTR);
+	expected_ptr = rx_queue->removed_count & rx_queue->ptr_mask;
+	if (unlikely(rx_ev_desc_ptr != expected_ptr))
+		efx_handle_rx_bad_index(rx_queue, rx_ev_desc_ptr);
+
+	if (likely(rx_ev_pkt_ok)) {
+		/* If packet is marked as OK and packet type is TCP/IP or
+		 * UDP/IP, then we can rely on the hardware checksum.
+		 */
+		checksummed =
+			rx_ev_hdr_type == FSE_CZ_RX_EV_HDR_TYPE_IPV4V6_TCP ||
+			rx_ev_hdr_type == FSE_CZ_RX_EV_HDR_TYPE_IPV4V6_UDP;
+	} else {
+		efx_handle_rx_not_ok(rx_queue, event, &rx_ev_pkt_ok, &discard);
+		checksummed = false;
+	}
+
+	/* Detect multicast packets that didn't match the filter */
+	rx_ev_mcast_pkt = EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_MCAST_PKT);
+	if (rx_ev_mcast_pkt) {
+		unsigned int rx_ev_mcast_hash_match =
+			EFX_QWORD_FIELD(*event, FSF_AZ_RX_EV_MCAST_HASH_MATCH);
+
+		if (unlikely(!rx_ev_mcast_hash_match)) {
+			++channel->n_rx_mcast_mismatch;
+			discard = true;
+		}
+	}
+
+	channel->irq_mod_score += 2;
+
+	/* Handle received packet */
+	efx_rx_packet(rx_queue, rx_ev_desc_ptr, rx_ev_byte_cnt,
+		      checksummed, discard);
+}
+
+static void
+efx_handle_generated_event(struct efx_channel *channel, efx_qword_t *event)
+{
+	struct efx_nic *efx = channel->efx;
+	unsigned code;
+
+	code = EFX_QWORD_FIELD(*event, FSF_AZ_DRV_GEN_EV_MAGIC);
+	if (code == EFX_CHANNEL_MAGIC_TEST(channel))
+		; /* ignore */
+	else if (code == EFX_CHANNEL_MAGIC_FILL(channel))
+		/* The queue must be empty, so we won't receive any rx
+		 * events, so efx_process_channel() won't refill the
+		 * queue. Refill it here */
+		efx_fast_push_rx_descriptors(efx_channel_get_rx_queue(channel));
+	else
+		netif_dbg(efx, hw, efx->net_dev, "channel %d received "
+			  "generated event "EFX_QWORD_FMT"\n",
+			  channel->channel, EFX_QWORD_VAL(*event));
+}
+
+static void
+efx_handle_driver_event(struct efx_channel *channel, efx_qword_t *event)
+{
+	struct efx_nic *efx = channel->efx;
+	unsigned int ev_sub_code;
+	unsigned int ev_sub_data;
+
+	ev_sub_code = EFX_QWORD_FIELD(*event, FSF_AZ_DRIVER_EV_SUBCODE);
+	ev_sub_data = EFX_QWORD_FIELD(*event, FSF_AZ_DRIVER_EV_SUBDATA);
+
+	switch (ev_sub_code) {
+	case FSE_AZ_TX_DESCQ_FLS_DONE_EV:
+		netif_vdbg(efx, hw, efx->net_dev, "channel %d TXQ %d flushed\n",
+			   channel->channel, ev_sub_data);
+		break;
+	case FSE_AZ_RX_DESCQ_FLS_DONE_EV:
+		netif_vdbg(efx, hw, efx->net_dev, "channel %d RXQ %d flushed\n",
+			   channel->channel, ev_sub_data);
+		break;
+	case FSE_AZ_EVQ_INIT_DONE_EV:
+		netif_dbg(efx, hw, efx->net_dev,
+			  "channel %d EVQ %d initialised\n",
+			  channel->channel, ev_sub_data);
+		break;
+	case FSE_AZ_SRM_UPD_DONE_EV:
+		netif_vdbg(efx, hw, efx->net_dev,
+			   "channel %d SRAM update done\n", channel->channel);
+		break;
+	case FSE_AZ_WAKE_UP_EV:
+		netif_vdbg(efx, hw, efx->net_dev,
+			   "channel %d RXQ %d wakeup event\n",
+			   channel->channel, ev_sub_data);
+		break;
+	case FSE_AZ_TIMER_EV:
+		netif_vdbg(efx, hw, efx->net_dev,
+			   "channel %d RX queue %d timer expired\n",
+			   channel->channel, ev_sub_data);
+		break;
+	case FSE_AA_RX_RECOVER_EV:
+		netif_err(efx, rx_err, efx->net_dev,
+			  "channel %d seen DRIVER RX_RESET event. "
+			"Resetting.\n", channel->channel);
+		atomic_inc(&efx->rx_reset);
+		efx_schedule_reset(efx,
+				   EFX_WORKAROUND_6555(efx) ?
+				   RESET_TYPE_RX_RECOVERY :
+				   RESET_TYPE_DISABLE);
+		break;
+	case FSE_BZ_RX_DSC_ERROR_EV:
+		netif_err(efx, rx_err, efx->net_dev,
+			  "RX DMA Q %d reports descriptor fetch error."
+			  " RX Q %d is disabled.\n", ev_sub_data, ev_sub_data);
+		efx_schedule_reset(efx, RESET_TYPE_RX_DESC_FETCH);
+		break;
+	case FSE_BZ_TX_DSC_ERROR_EV:
+		netif_err(efx, tx_err, efx->net_dev,
+			  "TX DMA Q %d reports descriptor fetch error."
+			  " TX Q %d is disabled.\n", ev_sub_data, ev_sub_data);
+		efx_schedule_reset(efx, RESET_TYPE_TX_DESC_FETCH);
+		break;
+	default:
+		netif_vdbg(efx, hw, efx->net_dev,
+			   "channel %d unknown driver event code %d "
+			   "data %04x\n", channel->channel, ev_sub_code,
+			   ev_sub_data);
+		break;
+	}
+}
+
+int efx_nic_process_eventq(struct efx_channel *channel, int budget)
+{
+	struct efx_nic *efx = channel->efx;
+	unsigned int read_ptr;
+	efx_qword_t event, *p_event;
+	int ev_code;
+	int tx_packets = 0;
+	int spent = 0;
+
+	read_ptr = channel->eventq_read_ptr;
+
+	for (;;) {
+		p_event = efx_event(channel, read_ptr);
+		event = *p_event;
+
+		if (!efx_event_present(&event))
+			/* End of events */
+			break;
+
+		netif_vdbg(channel->efx, intr, channel->efx->net_dev,
+			   "channel %d event is "EFX_QWORD_FMT"\n",
+			   channel->channel, EFX_QWORD_VAL(event));
+
+		/* Clear this event by marking it all ones */
+		EFX_SET_QWORD(*p_event);
+
+		++read_ptr;
+
+		ev_code = EFX_QWORD_FIELD(event, FSF_AZ_EV_CODE);
+
+		switch (ev_code) {
+		case FSE_AZ_EV_CODE_RX_EV:
+			efx_handle_rx_event(channel, &event);
+			if (++spent == budget)
+				goto out;
+			break;
+		case FSE_AZ_EV_CODE_TX_EV:
+			tx_packets += efx_handle_tx_event(channel, &event);
+			if (tx_packets > efx->txq_entries) {
+				spent = budget;
+				goto out;
+			}
+			break;
+		case FSE_AZ_EV_CODE_DRV_GEN_EV:
+			efx_handle_generated_event(channel, &event);
+			break;
+		case FSE_AZ_EV_CODE_DRIVER_EV:
+			efx_handle_driver_event(channel, &event);
+			break;
+		case FSE_CZ_EV_CODE_MCDI_EV:
+			efx_mcdi_process_event(channel, &event);
+			break;
+		case FSE_AZ_EV_CODE_GLOBAL_EV:
+			if (efx->type->handle_global_event &&
+			    efx->type->handle_global_event(channel, &event))
+				break;
+			/* else fall through */
+		default:
+			netif_err(channel->efx, hw, channel->efx->net_dev,
+				  "channel %d unknown event type %d (data "
+				  EFX_QWORD_FMT ")\n", channel->channel,
+				  ev_code, EFX_QWORD_VAL(event));
+		}
+	}
+
+out:
+	channel->eventq_read_ptr = read_ptr;
+	return spent;
+}
+
+/* Check whether an event is present in the eventq at the current
+ * read pointer.  Only useful for self-test.
+ */
+bool efx_nic_event_present(struct efx_channel *channel)
+{
+	return efx_event_present(efx_event(channel, channel->eventq_read_ptr));
+}
+
+/* Allocate buffer table entries for event queue */
+int efx_nic_probe_eventq(struct efx_channel *channel)
+{
+	struct efx_nic *efx = channel->efx;
+	unsigned entries;
+
+	entries = channel->eventq_mask + 1;
+	return efx_alloc_special_buffer(efx, &channel->eventq,
+					entries * sizeof(efx_qword_t));
+}
+
+void efx_nic_init_eventq(struct efx_channel *channel)
+{
+	efx_oword_t reg;
+	struct efx_nic *efx = channel->efx;
+
+	netif_dbg(efx, hw, efx->net_dev,
+		  "channel %d event queue in special buffers %d-%d\n",
+		  channel->channel, channel->eventq.index,
+		  channel->eventq.index + channel->eventq.entries - 1);
+
+	if (efx_nic_rev(efx) >= EFX_REV_SIENA_A0) {
+		EFX_POPULATE_OWORD_3(reg,
+				     FRF_CZ_TIMER_Q_EN, 1,
+				     FRF_CZ_HOST_NOTIFY_MODE, 0,
+				     FRF_CZ_TIMER_MODE, FFE_CZ_TIMER_MODE_DIS);
+		efx_writeo_table(efx, &reg, FR_BZ_TIMER_TBL, channel->channel);
+	}
+
+	/* Pin event queue buffer */
+	efx_init_special_buffer(efx, &channel->eventq);
+
+	/* Fill event queue with all ones (i.e. empty events) */
+	memset(channel->eventq.addr, 0xff, channel->eventq.len);
+
+	/* Push event queue to card */
+	EFX_POPULATE_OWORD_3(reg,
+			     FRF_AZ_EVQ_EN, 1,
+			     FRF_AZ_EVQ_SIZE, __ffs(channel->eventq.entries),
+			     FRF_AZ_EVQ_BUF_BASE_ID, channel->eventq.index);
+	efx_writeo_table(efx, &reg, efx->type->evq_ptr_tbl_base,
+			 channel->channel);
+
+	efx->type->push_irq_moderation(channel);
+}
+
+void efx_nic_fini_eventq(struct efx_channel *channel)
+{
+	efx_oword_t reg;
+	struct efx_nic *efx = channel->efx;
+
+	/* Remove event queue from card */
+	EFX_ZERO_OWORD(reg);
+	efx_writeo_table(efx, &reg, efx->type->evq_ptr_tbl_base,
+			 channel->channel);
+	if (efx_nic_rev(efx) >= EFX_REV_SIENA_A0)
+		efx_writeo_table(efx, &reg, FR_BZ_TIMER_TBL, channel->channel);
+
+	/* Unpin event queue */
+	efx_fini_special_buffer(efx, &channel->eventq);
+}
+
+/* Free buffers backing event queue */
+void efx_nic_remove_eventq(struct efx_channel *channel)
+{
+	efx_free_special_buffer(channel->efx, &channel->eventq);
+}
+
+
+void efx_nic_generate_test_event(struct efx_channel *channel)
+{
+	unsigned int magic = EFX_CHANNEL_MAGIC_TEST(channel);
+	efx_qword_t test_event;
+
+	EFX_POPULATE_QWORD_2(test_event, FSF_AZ_EV_CODE,
+			     FSE_AZ_EV_CODE_DRV_GEN_EV,
+			     FSF_AZ_DRV_GEN_EV_MAGIC, magic);
+	efx_generate_event(channel, &test_event);
+}
+
+void efx_nic_generate_fill_event(struct efx_channel *channel)
+{
+	unsigned int magic = EFX_CHANNEL_MAGIC_FILL(channel);
+	efx_qword_t test_event;
+
+	EFX_POPULATE_QWORD_2(test_event, FSF_AZ_EV_CODE,
+			     FSE_AZ_EV_CODE_DRV_GEN_EV,
+			     FSF_AZ_DRV_GEN_EV_MAGIC, magic);
+	efx_generate_event(channel, &test_event);
+}
+
+/**************************************************************************
+ *
+ * Flush handling
+ *
+ **************************************************************************/
+
+
+static void efx_poll_flush_events(struct efx_nic *efx)
+{
+	struct efx_channel *channel = efx_get_channel(efx, 0);
+	struct efx_tx_queue *tx_queue;
+	struct efx_rx_queue *rx_queue;
+	unsigned int read_ptr = channel->eventq_read_ptr;
+	unsigned int end_ptr = read_ptr + channel->eventq_mask - 1;
+
+	do {
+		efx_qword_t *event = efx_event(channel, read_ptr);
+		int ev_code, ev_sub_code, ev_queue;
+		bool ev_failed;
+
+		if (!efx_event_present(event))
+			break;
+
+		ev_code = EFX_QWORD_FIELD(*event, FSF_AZ_EV_CODE);
+		ev_sub_code = EFX_QWORD_FIELD(*event,
+					      FSF_AZ_DRIVER_EV_SUBCODE);
+		if (ev_code == FSE_AZ_EV_CODE_DRIVER_EV &&
+		    ev_sub_code == FSE_AZ_TX_DESCQ_FLS_DONE_EV) {
+			ev_queue = EFX_QWORD_FIELD(*event,
+						   FSF_AZ_DRIVER_EV_SUBDATA);
+			if (ev_queue < EFX_TXQ_TYPES * efx->n_tx_channels) {
+				tx_queue = efx_get_tx_queue(
+					efx, ev_queue / EFX_TXQ_TYPES,
+					ev_queue % EFX_TXQ_TYPES);
+				tx_queue->flushed = FLUSH_DONE;
+			}
+		} else if (ev_code == FSE_AZ_EV_CODE_DRIVER_EV &&
+			   ev_sub_code == FSE_AZ_RX_DESCQ_FLS_DONE_EV) {
+			ev_queue = EFX_QWORD_FIELD(
+				*event, FSF_AZ_DRIVER_EV_RX_DESCQ_ID);
+			ev_failed = EFX_QWORD_FIELD(
+				*event, FSF_AZ_DRIVER_EV_RX_FLUSH_FAIL);
+			if (ev_queue < efx->n_rx_channels) {
+				rx_queue = efx_get_rx_queue(efx, ev_queue);
+				rx_queue->flushed =
+					ev_failed ? FLUSH_FAILED : FLUSH_DONE;
+			}
+		}
+
+		/* We're about to destroy the queue anyway, so
+		 * it's ok to throw away every non-flush event */
+		EFX_SET_QWORD(*event);
+
+		++read_ptr;
+	} while (read_ptr != end_ptr);
+
+	channel->eventq_read_ptr = read_ptr;
+}
+
+/* Handle tx and rx flushes at the same time, since they run in
+ * parallel in the hardware and there's no reason for us to
+ * serialise them */
+int efx_nic_flush_queues(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+	struct efx_rx_queue *rx_queue;
+	struct efx_tx_queue *tx_queue;
+	int i, tx_pending, rx_pending;
+
+	/* If necessary prepare the hardware for flushing */
+	efx->type->prepare_flush(efx);
+
+	/* Flush all tx queues in parallel */
+	efx_for_each_channel(channel, efx) {
+		efx_for_each_possible_channel_tx_queue(tx_queue, channel) {
+			if (tx_queue->initialised)
+				efx_flush_tx_queue(tx_queue);
+		}
+	}
+
+	/* The hardware supports four concurrent rx flushes, each of which may
+	 * need to be retried if there is an outstanding descriptor fetch */
+	for (i = 0; i < EFX_FLUSH_POLL_COUNT; ++i) {
+		rx_pending = tx_pending = 0;
+		efx_for_each_channel(channel, efx) {
+			efx_for_each_channel_rx_queue(rx_queue, channel) {
+				if (rx_queue->flushed == FLUSH_PENDING)
+					++rx_pending;
+			}
+		}
+		efx_for_each_channel(channel, efx) {
+			efx_for_each_channel_rx_queue(rx_queue, channel) {
+				if (rx_pending == EFX_RX_FLUSH_COUNT)
+					break;
+				if (rx_queue->flushed == FLUSH_FAILED ||
+				    rx_queue->flushed == FLUSH_NONE) {
+					efx_flush_rx_queue(rx_queue);
+					++rx_pending;
+				}
+			}
+			efx_for_each_possible_channel_tx_queue(tx_queue, channel) {
+				if (tx_queue->initialised &&
+				    tx_queue->flushed != FLUSH_DONE)
+					++tx_pending;
+			}
+		}
+
+		if (rx_pending == 0 && tx_pending == 0)
+			return 0;
+
+		msleep(EFX_FLUSH_INTERVAL);
+		efx_poll_flush_events(efx);
+	}
+
+	/* Mark the queues as all flushed. We're going to return failure
+	 * leading to a reset, or fake up success anyway */
+	efx_for_each_channel(channel, efx) {
+		efx_for_each_possible_channel_tx_queue(tx_queue, channel) {
+			if (tx_queue->initialised &&
+			    tx_queue->flushed != FLUSH_DONE)
+				netif_err(efx, hw, efx->net_dev,
+					  "tx queue %d flush command timed out\n",
+					  tx_queue->queue);
+			tx_queue->flushed = FLUSH_DONE;
+		}
+		efx_for_each_channel_rx_queue(rx_queue, channel) {
+			if (rx_queue->flushed != FLUSH_DONE)
+				netif_err(efx, hw, efx->net_dev,
+					  "rx queue %d flush command timed out\n",
+					  efx_rx_queue_index(rx_queue));
+			rx_queue->flushed = FLUSH_DONE;
+		}
+	}
+
+	return -ETIMEDOUT;
+}
+
+/**************************************************************************
+ *
+ * Hardware interrupts
+ * The hardware interrupt handler does very little work; all the event
+ * queue processing is carried out by per-channel tasklets.
+ *
+ **************************************************************************/
+
+/* Enable/disable/generate interrupts */
+static inline void efx_nic_interrupts(struct efx_nic *efx,
+				      bool enabled, bool force)
+{
+	efx_oword_t int_en_reg_ker;
+
+	EFX_POPULATE_OWORD_3(int_en_reg_ker,
+			     FRF_AZ_KER_INT_LEVE_SEL, efx->fatal_irq_level,
+			     FRF_AZ_KER_INT_KER, force,
+			     FRF_AZ_DRV_INT_EN_KER, enabled);
+	efx_writeo(efx, &int_en_reg_ker, FR_AZ_INT_EN_KER);
+}
+
+void efx_nic_enable_interrupts(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+
+	EFX_ZERO_OWORD(*((efx_oword_t *) efx->irq_status.addr));
+	wmb(); /* Ensure interrupt vector is clear before interrupts enabled */
+
+	/* Enable interrupts */
+	efx_nic_interrupts(efx, true, false);
+
+	/* Force processing of all the channels to get the EVQ RPTRs up to
+	   date */
+	efx_for_each_channel(channel, efx)
+		efx_schedule_channel(channel);
+}
+
+void efx_nic_disable_interrupts(struct efx_nic *efx)
+{
+	/* Disable interrupts */
+	efx_nic_interrupts(efx, false, false);
+}
+
+/* Generate a test interrupt
+ * Interrupt must already have been enabled, otherwise nasty things
+ * may happen.
+ */
+void efx_nic_generate_interrupt(struct efx_nic *efx)
+{
+	efx_nic_interrupts(efx, true, true);
+}
+
+/* Process a fatal interrupt
+ * Disable bus mastering ASAP and schedule a reset
+ */
+irqreturn_t efx_nic_fatal_interrupt(struct efx_nic *efx)
+{
+	struct falcon_nic_data *nic_data = efx->nic_data;
+	efx_oword_t *int_ker = efx->irq_status.addr;
+	efx_oword_t fatal_intr;
+	int error, mem_perr;
+
+	efx_reado(efx, &fatal_intr, FR_AZ_FATAL_INTR_KER);
+	error = EFX_OWORD_FIELD(fatal_intr, FRF_AZ_FATAL_INTR);
+
+	netif_err(efx, hw, efx->net_dev, "SYSTEM ERROR "EFX_OWORD_FMT" status "
+		  EFX_OWORD_FMT ": %s\n", EFX_OWORD_VAL(*int_ker),
+		  EFX_OWORD_VAL(fatal_intr),
+		  error ? "disabling bus mastering" : "no recognised error");
+
+	/* If this is a memory parity error dump which blocks are offending */
+	mem_perr = (EFX_OWORD_FIELD(fatal_intr, FRF_AZ_MEM_PERR_INT_KER) ||
+		    EFX_OWORD_FIELD(fatal_intr, FRF_AZ_SRM_PERR_INT_KER));
+	if (mem_perr) {
+		efx_oword_t reg;
+		efx_reado(efx, &reg, FR_AZ_MEM_STAT);
+		netif_err(efx, hw, efx->net_dev,
+			  "SYSTEM ERROR: memory parity error "EFX_OWORD_FMT"\n",
+			  EFX_OWORD_VAL(reg));
+	}
+
+	/* Disable both devices */
+	pci_clear_master(efx->pci_dev);
+	if (efx_nic_is_dual_func(efx))
+		pci_clear_master(nic_data->pci_dev2);
+	efx_nic_disable_interrupts(efx);
+
+	/* Count errors and reset or disable the NIC accordingly */
+	if (efx->int_error_count == 0 ||
+	    time_after(jiffies, efx->int_error_expire)) {
+		efx->int_error_count = 0;
+		efx->int_error_expire =
+			jiffies + EFX_INT_ERROR_EXPIRE * HZ;
+	}
+	if (++efx->int_error_count < EFX_MAX_INT_ERRORS) {
+		netif_err(efx, hw, efx->net_dev,
+			  "SYSTEM ERROR - reset scheduled\n");
+		efx_schedule_reset(efx, RESET_TYPE_INT_ERROR);
+	} else {
+		netif_err(efx, hw, efx->net_dev,
+			  "SYSTEM ERROR - max number of errors seen."
+			  "NIC will be disabled\n");
+		efx_schedule_reset(efx, RESET_TYPE_DISABLE);
+	}
+
+	return IRQ_HANDLED;
+}
+
+/* Handle a legacy interrupt
+ * Acknowledges the interrupt and schedule event queue processing.
+ */
+static irqreturn_t efx_legacy_interrupt(int irq, void *dev_id)
+{
+	struct efx_nic *efx = dev_id;
+	efx_oword_t *int_ker = efx->irq_status.addr;
+	irqreturn_t result = IRQ_NONE;
+	struct efx_channel *channel;
+	efx_dword_t reg;
+	u32 queues;
+	int syserr;
+
+	/* Could this be ours?  If interrupts are disabled then the
+	 * channel state may not be valid.
+	 */
+	if (!efx->legacy_irq_enabled)
+		return result;
+
+	/* Read the ISR which also ACKs the interrupts */
+	efx_readd(efx, &reg, FR_BZ_INT_ISR0);
+	queues = EFX_EXTRACT_DWORD(reg, 0, 31);
+
+	/* Check to see if we have a serious error condition */
+	if (queues & (1U << efx->fatal_irq_level)) {
+		syserr = EFX_OWORD_FIELD(*int_ker, FSF_AZ_NET_IVEC_FATAL_INT);
+		if (unlikely(syserr))
+			return efx_nic_fatal_interrupt(efx);
+	}
+
+	if (queues != 0) {
+		if (EFX_WORKAROUND_15783(efx))
+			efx->irq_zero_count = 0;
+
+		/* Schedule processing of any interrupting queues */
+		efx_for_each_channel(channel, efx) {
+			if (queues & 1)
+				efx_schedule_channel(channel);
+			queues >>= 1;
+		}
+		result = IRQ_HANDLED;
+
+	} else if (EFX_WORKAROUND_15783(efx)) {
+		efx_qword_t *event;
+
+		/* We can't return IRQ_HANDLED more than once on seeing ISR=0
+		 * because this might be a shared interrupt. */
+		if (efx->irq_zero_count++ == 0)
+			result = IRQ_HANDLED;
+
+		/* Ensure we schedule or rearm all event queues */
+		efx_for_each_channel(channel, efx) {
+			event = efx_event(channel, channel->eventq_read_ptr);
+			if (efx_event_present(event))
+				efx_schedule_channel(channel);
+			else
+				efx_nic_eventq_read_ack(channel);
+		}
+	}
+
+	if (result == IRQ_HANDLED) {
+		efx->last_irq_cpu = raw_smp_processor_id();
+		netif_vdbg(efx, intr, efx->net_dev,
+			   "IRQ %d on CPU %d status " EFX_DWORD_FMT "\n",
+			   irq, raw_smp_processor_id(), EFX_DWORD_VAL(reg));
+	}
+
+	return result;
+}
+
+/* Handle an MSI interrupt
+ *
+ * Handle an MSI hardware interrupt.  This routine schedules event
+ * queue processing.  No interrupt acknowledgement cycle is necessary.
+ * Also, we never need to check that the interrupt is for us, since
+ * MSI interrupts cannot be shared.
+ */
+static irqreturn_t efx_msi_interrupt(int irq, void *dev_id)
+{
+	struct efx_channel *channel = *(struct efx_channel **)dev_id;
+	struct efx_nic *efx = channel->efx;
+	efx_oword_t *int_ker = efx->irq_status.addr;
+	int syserr;
+
+	efx->last_irq_cpu = raw_smp_processor_id();
+	netif_vdbg(efx, intr, efx->net_dev,
+		   "IRQ %d on CPU %d status " EFX_OWORD_FMT "\n",
+		   irq, raw_smp_processor_id(), EFX_OWORD_VAL(*int_ker));
+
+	/* Check to see if we have a serious error condition */
+	if (channel->channel == efx->fatal_irq_level) {
+		syserr = EFX_OWORD_FIELD(*int_ker, FSF_AZ_NET_IVEC_FATAL_INT);
+		if (unlikely(syserr))
+			return efx_nic_fatal_interrupt(efx);
+	}
+
+	/* Schedule processing of the channel */
+	efx_schedule_channel(channel);
+
+	return IRQ_HANDLED;
+}
+
+
+/* Setup RSS indirection table.
+ * This maps from the hash value of the packet to RXQ
+ */
+void efx_nic_push_rx_indir_table(struct efx_nic *efx)
+{
+	size_t i = 0;
+	efx_dword_t dword;
+
+	if (efx_nic_rev(efx) < EFX_REV_FALCON_B0)
+		return;
+
+	BUILD_BUG_ON(ARRAY_SIZE(efx->rx_indir_table) !=
+		     FR_BZ_RX_INDIRECTION_TBL_ROWS);
+
+	for (i = 0; i < FR_BZ_RX_INDIRECTION_TBL_ROWS; i++) {
+		EFX_POPULATE_DWORD_1(dword, FRF_BZ_IT_QUEUE,
+				     efx->rx_indir_table[i]);
+		efx_writed_table(efx, &dword, FR_BZ_RX_INDIRECTION_TBL, i);
+	}
+}
+
+/* Hook interrupt handler(s)
+ * Try MSI and then legacy interrupts.
+ */
+int efx_nic_init_interrupt(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+	int rc;
+
+	if (!EFX_INT_MODE_USE_MSI(efx)) {
+		irq_handler_t handler;
+		if (efx_nic_rev(efx) >= EFX_REV_FALCON_B0)
+			handler = efx_legacy_interrupt;
+		else
+			handler = falcon_legacy_interrupt_a1;
+
+		rc = request_irq(efx->legacy_irq, handler, IRQF_SHARED,
+				 efx->name, efx);
+		if (rc) {
+			netif_err(efx, drv, efx->net_dev,
+				  "failed to hook legacy IRQ %d\n",
+				  efx->pci_dev->irq);
+			goto fail1;
+		}
+		return 0;
+	}
+
+	/* Hook MSI or MSI-X interrupt */
+	efx_for_each_channel(channel, efx) {
+		rc = request_irq(channel->irq, efx_msi_interrupt,
+				 IRQF_PROBE_SHARED, /* Not shared */
+				 efx->channel_name[channel->channel],
+				 &efx->channel[channel->channel]);
+		if (rc) {
+			netif_err(efx, drv, efx->net_dev,
+				  "failed to hook IRQ %d\n", channel->irq);
+			goto fail2;
+		}
+	}
+
+	return 0;
+
+ fail2:
+	efx_for_each_channel(channel, efx)
+		free_irq(channel->irq, &efx->channel[channel->channel]);
+ fail1:
+	return rc;
+}
+
+void efx_nic_fini_interrupt(struct efx_nic *efx)
+{
+	struct efx_channel *channel;
+	efx_oword_t reg;
+
+	/* Disable MSI/MSI-X interrupts */
+	efx_for_each_channel(channel, efx) {
+		if (channel->irq)
+			free_irq(channel->irq, &efx->channel[channel->channel]);
+	}
+
+	/* ACK legacy interrupt */
+	if (efx_nic_rev(efx) >= EFX_REV_FALCON_B0)
+		efx_reado(efx, &reg, FR_BZ_INT_ISR0);
+	else
+		falcon_irq_ack_a1(efx);
+
+	/* Disable legacy interrupt */
+	if (efx->legacy_irq)
+		free_irq(efx->legacy_irq, efx);
+}
+
+u32 efx_nic_fpga_ver(struct efx_nic *efx)
+{
+	efx_oword_t altera_build;
+	efx_reado(efx, &altera_build, FR_AZ_ALTERA_BUILD);
+	return EFX_OWORD_FIELD(altera_build, FRF_AZ_ALTERA_BUILD_VER);
+}
+
+void efx_nic_init_common(struct efx_nic *efx)
+{
+	efx_oword_t temp;
+
+	/* Set positions of descriptor caches in SRAM. */
+	EFX_POPULATE_OWORD_1(temp, FRF_AZ_SRM_TX_DC_BASE_ADR,
+			     efx->type->tx_dc_base / 8);
+	efx_writeo(efx, &temp, FR_AZ_SRM_TX_DC_CFG);
+	EFX_POPULATE_OWORD_1(temp, FRF_AZ_SRM_RX_DC_BASE_ADR,
+			     efx->type->rx_dc_base / 8);
+	efx_writeo(efx, &temp, FR_AZ_SRM_RX_DC_CFG);
+
+	/* Set TX descriptor cache size. */
+	BUILD_BUG_ON(TX_DC_ENTRIES != (8 << TX_DC_ENTRIES_ORDER));
+	EFX_POPULATE_OWORD_1(temp, FRF_AZ_TX_DC_SIZE, TX_DC_ENTRIES_ORDER);
+	efx_writeo(efx, &temp, FR_AZ_TX_DC_CFG);
+
+	/* Set RX descriptor cache size.  Set low watermark to size-8, as
+	 * this allows most efficient prefetching.
+	 */
+	BUILD_BUG_ON(RX_DC_ENTRIES != (8 << RX_DC_ENTRIES_ORDER));
+	EFX_POPULATE_OWORD_1(temp, FRF_AZ_RX_DC_SIZE, RX_DC_ENTRIES_ORDER);
+	efx_writeo(efx, &temp, FR_AZ_RX_DC_CFG);
+	EFX_POPULATE_OWORD_1(temp, FRF_AZ_RX_DC_PF_LWM, RX_DC_ENTRIES - 8);
+	efx_writeo(efx, &temp, FR_AZ_RX_DC_PF_WM);
+
+	/* Program INT_KER address */
+	EFX_POPULATE_OWORD_2(temp,
+			     FRF_AZ_NORM_INT_VEC_DIS_KER,
+			     EFX_INT_MODE_USE_MSI(efx),
+			     FRF_AZ_INT_ADR_KER, efx->irq_status.dma_addr);
+	efx_writeo(efx, &temp, FR_AZ_INT_ADR_KER);
+
+	if (EFX_WORKAROUND_17213(efx) && !EFX_INT_MODE_USE_MSI(efx))
+		/* Use an interrupt level unused by event queues */
+		efx->fatal_irq_level = 0x1f;
+	else
+		/* Use a valid MSI-X vector */
+		efx->fatal_irq_level = 0;
+
+	/* Enable all the genuinely fatal interrupts.  (They are still
+	 * masked by the overall interrupt mask, controlled by
+	 * falcon_interrupts()).
+	 *
+	 * Note: All other fatal interrupts are enabled
+	 */
+	EFX_POPULATE_OWORD_3(temp,
+			     FRF_AZ_ILL_ADR_INT_KER_EN, 1,
+			     FRF_AZ_RBUF_OWN_INT_KER_EN, 1,
+			     FRF_AZ_TBUF_OWN_INT_KER_EN, 1);
+	if (efx_nic_rev(efx) >= EFX_REV_SIENA_A0)
+		EFX_SET_OWORD_FIELD(temp, FRF_CZ_SRAM_PERR_INT_P_KER_EN, 1);
+	EFX_INVERT_OWORD(temp);
+	efx_writeo(efx, &temp, FR_AZ_FATAL_INTR_KER);
+
+	efx_nic_push_rx_indir_table(efx);
+
+	/* Disable the ugly timer-based TX DMA backoff and allow TX DMA to be
+	 * controlled by the RX FIFO fill level. Set arbitration to one pkt/Q.
+	 */
+	efx_reado(efx, &temp, FR_AZ_TX_RESERVED);
+	EFX_SET_OWORD_FIELD(temp, FRF_AZ_TX_RX_SPACER, 0xfe);
+	EFX_SET_OWORD_FIELD(temp, FRF_AZ_TX_RX_SPACER_EN, 1);
+	EFX_SET_OWORD_FIELD(temp, FRF_AZ_TX_ONE_PKT_PER_Q, 1);
+	EFX_SET_OWORD_FIELD(temp, FRF_AZ_TX_PUSH_EN, 1);
+	EFX_SET_OWORD_FIELD(temp, FRF_AZ_TX_DIS_NON_IP_EV, 1);
+	/* Enable SW_EV to inherit in char driver - assume harmless here */
+	EFX_SET_OWORD_FIELD(temp, FRF_AZ_TX_SOFT_EVT_EN, 1);
+	/* Prefetch threshold 2 => fetch when descriptor cache half empty */
+	EFX_SET_OWORD_FIELD(temp, FRF_AZ_TX_PREF_THRESHOLD, 2);
+	/* Disable hardware watchdog which can misfire */
+	EFX_SET_OWORD_FIELD(temp, FRF_AZ_TX_PREF_WD_TMR, 0x3fffff);
+	/* Squash TX of packets of 16 bytes or less */
+	if (efx_nic_rev(efx) >= EFX_REV_FALCON_B0)
+		EFX_SET_OWORD_FIELD(temp, FRF_BZ_TX_FLUSH_MIN_LEN_EN, 1);
+	efx_writeo(efx, &temp, FR_AZ_TX_RESERVED);
+
+	if (efx_nic_rev(efx) >= EFX_REV_FALCON_B0) {
+		EFX_POPULATE_OWORD_4(temp,
+				     /* Default values */
+				     FRF_BZ_TX_PACE_SB_NOT_AF, 0x15,
+				     FRF_BZ_TX_PACE_SB_AF, 0xb,
+				     FRF_BZ_TX_PACE_FB_BASE, 0,
+				     /* Allow large pace values in the
+				      * fast bin. */
+				     FRF_BZ_TX_PACE_BIN_TH,
+				     FFE_BZ_TX_PACE_RESERVED);
+		efx_writeo(efx, &temp, FR_BZ_TX_PACE);
+	}
+}
+
+/* Register dump */
+
+#define REGISTER_REVISION_A	1
+#define REGISTER_REVISION_B	2
+#define REGISTER_REVISION_C	3
+#define REGISTER_REVISION_Z	3	/* latest revision */
+
+struct efx_nic_reg {
+	u32 offset:24;
+	u32 min_revision:2, max_revision:2;
+};
+
+#define REGISTER(name, min_rev, max_rev) {				\
+	FR_ ## min_rev ## max_rev ## _ ## name,				\
+	REGISTER_REVISION_ ## min_rev, REGISTER_REVISION_ ## max_rev	\
+}
+#define REGISTER_AA(name) REGISTER(name, A, A)
+#define REGISTER_AB(name) REGISTER(name, A, B)
+#define REGISTER_AZ(name) REGISTER(name, A, Z)
+#define REGISTER_BB(name) REGISTER(name, B, B)
+#define REGISTER_BZ(name) REGISTER(name, B, Z)
+#define REGISTER_CZ(name) REGISTER(name, C, Z)
+
+static const struct efx_nic_reg efx_nic_regs[] = {
+	REGISTER_AZ(ADR_REGION),
+	REGISTER_AZ(INT_EN_KER),
+	REGISTER_BZ(INT_EN_CHAR),
+	REGISTER_AZ(INT_ADR_KER),
+	REGISTER_BZ(INT_ADR_CHAR),
+	/* INT_ACK_KER is WO */
+	/* INT_ISR0 is RC */
+	REGISTER_AZ(HW_INIT),
+	REGISTER_CZ(USR_EV_CFG),
+	REGISTER_AB(EE_SPI_HCMD),
+	REGISTER_AB(EE_SPI_HADR),
+	REGISTER_AB(EE_SPI_HDATA),
+	REGISTER_AB(EE_BASE_PAGE),
+	REGISTER_AB(EE_VPD_CFG0),
+	/* EE_VPD_SW_CNTL and EE_VPD_SW_DATA are not used */
+	/* PMBX_DBG_IADDR and PBMX_DBG_IDATA are indirect */
+	/* PCIE_CORE_INDIRECT is indirect */
+	REGISTER_AB(NIC_STAT),
+	REGISTER_AB(GPIO_CTL),
+	REGISTER_AB(GLB_CTL),
+	/* FATAL_INTR_KER and FATAL_INTR_CHAR are partly RC */
+	REGISTER_BZ(DP_CTRL),
+	REGISTER_AZ(MEM_STAT),
+	REGISTER_AZ(CS_DEBUG),
+	REGISTER_AZ(ALTERA_BUILD),
+	REGISTER_AZ(CSR_SPARE),
+	REGISTER_AB(PCIE_SD_CTL0123),
+	REGISTER_AB(PCIE_SD_CTL45),
+	REGISTER_AB(PCIE_PCS_CTL_STAT),
+	/* DEBUG_DATA_OUT is not used */
+	/* DRV_EV is WO */
+	REGISTER_AZ(EVQ_CTL),
+	REGISTER_AZ(EVQ_CNT1),
+	REGISTER_AZ(EVQ_CNT2),
+	REGISTER_AZ(BUF_TBL_CFG),
+	REGISTER_AZ(SRM_RX_DC_CFG),
+	REGISTER_AZ(SRM_TX_DC_CFG),
+	REGISTER_AZ(SRM_CFG),
+	/* BUF_TBL_UPD is WO */
+	REGISTER_AZ(SRM_UPD_EVQ),
+	REGISTER_AZ(SRAM_PARITY),
+	REGISTER_AZ(RX_CFG),
+	REGISTER_BZ(RX_FILTER_CTL),
+	/* RX_FLUSH_DESCQ is WO */
+	REGISTER_AZ(RX_DC_CFG),
+	REGISTER_AZ(RX_DC_PF_WM),
+	REGISTER_BZ(RX_RSS_TKEY),
+	/* RX_NODESC_DROP is RC */
+	REGISTER_AA(RX_SELF_RST),
+	/* RX_DEBUG, RX_PUSH_DROP are not used */
+	REGISTER_CZ(RX_RSS_IPV6_REG1),
+	REGISTER_CZ(RX_RSS_IPV6_REG2),
+	REGISTER_CZ(RX_RSS_IPV6_REG3),
+	/* TX_FLUSH_DESCQ is WO */
+	REGISTER_AZ(TX_DC_CFG),
+	REGISTER_AA(TX_CHKSM_CFG),
+	REGISTER_AZ(TX_CFG),
+	/* TX_PUSH_DROP is not used */
+	REGISTER_AZ(TX_RESERVED),
+	REGISTER_BZ(TX_PACE),
+	/* TX_PACE_DROP_QID is RC */
+	REGISTER_BB(TX_VLAN),
+	REGISTER_BZ(TX_IPFIL_PORTEN),
+	REGISTER_AB(MD_TXD),
+	REGISTER_AB(MD_RXD),
+	REGISTER_AB(MD_CS),
+	REGISTER_AB(MD_PHY_ADR),
+	REGISTER_AB(MD_ID),
+	/* MD_STAT is RC */
+	REGISTER_AB(MAC_STAT_DMA),
+	REGISTER_AB(MAC_CTRL),
+	REGISTER_BB(GEN_MODE),
+	REGISTER_AB(MAC_MC_HASH_REG0),
+	REGISTER_AB(MAC_MC_HASH_REG1),
+	REGISTER_AB(GM_CFG1),
+	REGISTER_AB(GM_CFG2),
+	/* GM_IPG and GM_HD are not used */
+	REGISTER_AB(GM_MAX_FLEN),
+	/* GM_TEST is not used */
+	REGISTER_AB(GM_ADR1),
+	REGISTER_AB(GM_ADR2),
+	REGISTER_AB(GMF_CFG0),
+	REGISTER_AB(GMF_CFG1),
+	REGISTER_AB(GMF_CFG2),
+	REGISTER_AB(GMF_CFG3),
+	REGISTER_AB(GMF_CFG4),
+	REGISTER_AB(GMF_CFG5),
+	REGISTER_BB(TX_SRC_MAC_CTL),
+	REGISTER_AB(XM_ADR_LO),
+	REGISTER_AB(XM_ADR_HI),
+	REGISTER_AB(XM_GLB_CFG),
+	REGISTER_AB(XM_TX_CFG),
+	REGISTER_AB(XM_RX_CFG),
+	REGISTER_AB(XM_MGT_INT_MASK),
+	REGISTER_AB(XM_FC),
+	REGISTER_AB(XM_PAUSE_TIME),
+	REGISTER_AB(XM_TX_PARAM),
+	REGISTER_AB(XM_RX_PARAM),
+	/* XM_MGT_INT_MSK (note no 'A') is RC */
+	REGISTER_AB(XX_PWR_RST),
+	REGISTER_AB(XX_SD_CTL),
+	REGISTER_AB(XX_TXDRV_CTL),
+	/* XX_PRBS_CTL, XX_PRBS_CHK and XX_PRBS_ERR are not used */
+	/* XX_CORE_STAT is partly RC */
+};
+
+struct efx_nic_reg_table {
+	u32 offset:24;
+	u32 min_revision:2, max_revision:2;
+	u32 step:6, rows:21;
+};
+
+#define REGISTER_TABLE_DIMENSIONS(_, offset, min_rev, max_rev, step, rows) { \
+	offset,								\
+	REGISTER_REVISION_ ## min_rev, REGISTER_REVISION_ ## max_rev,	\
+	step, rows							\
+}
+#define REGISTER_TABLE(name, min_rev, max_rev) 				\
+	REGISTER_TABLE_DIMENSIONS(					\
+		name, FR_ ## min_rev ## max_rev ## _ ## name,		\
+		min_rev, max_rev,					\
+		FR_ ## min_rev ## max_rev ## _ ## name ## _STEP,	\
+		FR_ ## min_rev ## max_rev ## _ ## name ## _ROWS)
+#define REGISTER_TABLE_AA(name) REGISTER_TABLE(name, A, A)
+#define REGISTER_TABLE_AZ(name) REGISTER_TABLE(name, A, Z)
+#define REGISTER_TABLE_BB(name) REGISTER_TABLE(name, B, B)
+#define REGISTER_TABLE_BZ(name) REGISTER_TABLE(name, B, Z)
+#define REGISTER_TABLE_BB_CZ(name)					\
+	REGISTER_TABLE_DIMENSIONS(name, FR_BZ_ ## name, B, B,		\
+				  FR_BZ_ ## name ## _STEP,		\
+				  FR_BB_ ## name ## _ROWS),		\
+	REGISTER_TABLE_DIMENSIONS(name, FR_BZ_ ## name, C, Z,		\
+				  FR_BZ_ ## name ## _STEP,		\
+				  FR_CZ_ ## name ## _ROWS)
+#define REGISTER_TABLE_CZ(name) REGISTER_TABLE(name, C, Z)
+
+static const struct efx_nic_reg_table efx_nic_reg_tables[] = {
+	/* DRIVER is not used */
+	/* EVQ_RPTR, TIMER_COMMAND, USR_EV and {RX,TX}_DESC_UPD are WO */
+	REGISTER_TABLE_BB(TX_IPFIL_TBL),
+	REGISTER_TABLE_BB(TX_SRC_MAC_TBL),
+	REGISTER_TABLE_AA(RX_DESC_PTR_TBL_KER),
+	REGISTER_TABLE_BB_CZ(RX_DESC_PTR_TBL),
+	REGISTER_TABLE_AA(TX_DESC_PTR_TBL_KER),
+	REGISTER_TABLE_BB_CZ(TX_DESC_PTR_TBL),
+	REGISTER_TABLE_AA(EVQ_PTR_TBL_KER),
+	REGISTER_TABLE_BB_CZ(EVQ_PTR_TBL),
+	/* We can't reasonably read all of the buffer table (up to 8MB!).
+	 * However this driver will only use a few entries.  Reading
+	 * 1K entries allows for some expansion of queue count and
+	 * size before we need to change the version. */
+	REGISTER_TABLE_DIMENSIONS(BUF_FULL_TBL_KER, FR_AA_BUF_FULL_TBL_KER,
+				  A, A, 8, 1024),
+	REGISTER_TABLE_DIMENSIONS(BUF_FULL_TBL, FR_BZ_BUF_FULL_TBL,
+				  B, Z, 8, 1024),
+	REGISTER_TABLE_CZ(RX_MAC_FILTER_TBL0),
+	REGISTER_TABLE_BB_CZ(TIMER_TBL),
+	REGISTER_TABLE_BB_CZ(TX_PACE_TBL),
+	REGISTER_TABLE_BZ(RX_INDIRECTION_TBL),
+	/* TX_FILTER_TBL0 is huge and not used by this driver */
+	REGISTER_TABLE_CZ(TX_MAC_FILTER_TBL0),
+	REGISTER_TABLE_CZ(MC_TREG_SMEM),
+	/* MSIX_PBA_TABLE is not mapped */
+	/* SRM_DBG is not mapped (and is redundant with BUF_FLL_TBL) */
+	REGISTER_TABLE_BZ(RX_FILTER_TBL0),
+};
+
+size_t efx_nic_get_regs_len(struct efx_nic *efx)
+{
+	const struct efx_nic_reg *reg;
+	const struct efx_nic_reg_table *table;
+	size_t len = 0;
+
+	for (reg = efx_nic_regs;
+	     reg < efx_nic_regs + ARRAY_SIZE(efx_nic_regs);
+	     reg++)
+		if (efx->type->revision >= reg->min_revision &&
+		    efx->type->revision <= reg->max_revision)
+			len += sizeof(efx_oword_t);
+
+	for (table = efx_nic_reg_tables;
+	     table < efx_nic_reg_tables + ARRAY_SIZE(efx_nic_reg_tables);
+	     table++)
+		if (efx->type->revision >= table->min_revision &&
+		    efx->type->revision <= table->max_revision)
+			len += table->rows * min_t(size_t, table->step, 16);
+
+	return len;
+}
+
+void efx_nic_get_regs(struct efx_nic *efx, void *buf)
+{
+	const struct efx_nic_reg *reg;
+	const struct efx_nic_reg_table *table;
+
+	for (reg = efx_nic_regs;
+	     reg < efx_nic_regs + ARRAY_SIZE(efx_nic_regs);
+	     reg++) {
+		if (efx->type->revision >= reg->min_revision &&
+		    efx->type->revision <= reg->max_revision) {
+			efx_reado(efx, (efx_oword_t *)buf, reg->offset);
+			buf += sizeof(efx_oword_t);
+		}
+	}
+
+	for (table = efx_nic_reg_tables;
+	     table < efx_nic_reg_tables + ARRAY_SIZE(efx_nic_reg_tables);
+	     table++) {
+		size_t size, i;
+
+		if (!(efx->type->revision >= table->min_revision &&
+		      efx->type->revision <= table->max_revision))
+			continue;
+
+		size = min_t(size_t, table->step, 16);
+
+		if (table->offset >= efx->type->mem_map_size) {
+			/* No longer mapped; return dummy data */
+			memcpy(buf, "\xde\xc0\xad\xde", 4);
+			buf += table->rows * size;
+			continue;
+		}
+
+		for (i = 0; i < table->rows; i++) {
+			switch (table->step) {
+			case 4: /* 32-bit register or SRAM */
+				efx_readd_table(efx, buf, table->offset, i);
+				break;
+			case 8: /* 64-bit SRAM */
+				efx_sram_readq(efx,
+					       efx->membase + table->offset,
+					       buf, i);
+				break;
+			case 16: /* 128-bit register */
+				efx_reado_table(efx, buf, table->offset, i);
+				break;
+			case 32: /* 128-bit register, interleaved */
+				efx_reado_table(efx, buf, table->offset, 2 * i);
+				break;
+			default:
+				WARN_ON(1);
+				return;
+			}
+			buf += size;
+		}
+	}
+}
diff --git a/drivers/net/ethernet/sfc/nic.h b/drivers/net/ethernet/sfc/nic.h
new file mode 100644
index 0000000..4bd1f28
--- /dev/null
+++ b/drivers/net/ethernet/sfc/nic.h
@@ -0,0 +1,273 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2005-2006 Fen Systems Ltd.
+ * Copyright 2006-2011 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#ifndef EFX_NIC_H
+#define EFX_NIC_H
+
+#include <linux/i2c-algo-bit.h>
+#include "net_driver.h"
+#include "efx.h"
+#include "mcdi.h"
+#include "spi.h"
+
+/*
+ * Falcon hardware control
+ */
+
+enum {
+	EFX_REV_FALCON_A0 = 0,
+	EFX_REV_FALCON_A1 = 1,
+	EFX_REV_FALCON_B0 = 2,
+	EFX_REV_SIENA_A0 = 3,
+};
+
+static inline int efx_nic_rev(struct efx_nic *efx)
+{
+	return efx->type->revision;
+}
+
+extern u32 efx_nic_fpga_ver(struct efx_nic *efx);
+
+static inline bool efx_nic_has_mc(struct efx_nic *efx)
+{
+	return efx_nic_rev(efx) >= EFX_REV_SIENA_A0;
+}
+/* NIC has two interlinked PCI functions for the same port. */
+static inline bool efx_nic_is_dual_func(struct efx_nic *efx)
+{
+	return efx_nic_rev(efx) < EFX_REV_FALCON_B0;
+}
+
+enum {
+	PHY_TYPE_NONE = 0,
+	PHY_TYPE_TXC43128 = 1,
+	PHY_TYPE_88E1111 = 2,
+	PHY_TYPE_SFX7101 = 3,
+	PHY_TYPE_QT2022C2 = 4,
+	PHY_TYPE_PM8358 = 6,
+	PHY_TYPE_SFT9001A = 8,
+	PHY_TYPE_QT2025C = 9,
+	PHY_TYPE_SFT9001B = 10,
+};
+
+#define FALCON_XMAC_LOOPBACKS			\
+	((1 << LOOPBACK_XGMII) |		\
+	 (1 << LOOPBACK_XGXS) |			\
+	 (1 << LOOPBACK_XAUI))
+
+#define FALCON_GMAC_LOOPBACKS			\
+	(1 << LOOPBACK_GMAC)
+
+/**
+ * struct falcon_board_type - board operations and type information
+ * @id: Board type id, as found in NVRAM
+ * @ref_model: Model number of Solarflare reference design
+ * @gen_type: Generic board type description
+ * @init: Allocate resources and initialise peripheral hardware
+ * @init_phy: Do board-specific PHY initialisation
+ * @fini: Shut down hardware and free resources
+ * @set_id_led: Set state of identifying LED or revert to automatic function
+ * @monitor: Board-specific health check function
+ */
+struct falcon_board_type {
+	u8 id;
+	const char *ref_model;
+	const char *gen_type;
+	int (*init) (struct efx_nic *nic);
+	void (*init_phy) (struct efx_nic *efx);
+	void (*fini) (struct efx_nic *nic);
+	void (*set_id_led) (struct efx_nic *efx, enum efx_led_mode mode);
+	int (*monitor) (struct efx_nic *nic);
+};
+
+/**
+ * struct falcon_board - board information
+ * @type: Type of board
+ * @major: Major rev. ('A', 'B' ...)
+ * @minor: Minor rev. (0, 1, ...)
+ * @i2c_adap: I2C adapter for on-board peripherals
+ * @i2c_data: Data for bit-banging algorithm
+ * @hwmon_client: I2C client for hardware monitor
+ * @ioexp_client: I2C client for power/port control
+ */
+struct falcon_board {
+	const struct falcon_board_type *type;
+	int major;
+	int minor;
+	struct i2c_adapter i2c_adap;
+	struct i2c_algo_bit_data i2c_data;
+	struct i2c_client *hwmon_client, *ioexp_client;
+};
+
+/**
+ * struct falcon_nic_data - Falcon NIC state
+ * @pci_dev2: Secondary function of Falcon A
+ * @board: Board state and functions
+ * @stats_disable_count: Nest count for disabling statistics fetches
+ * @stats_pending: Is there a pending DMA of MAC statistics.
+ * @stats_timer: A timer for regularly fetching MAC statistics.
+ * @stats_dma_done: Pointer to the flag which indicates DMA completion.
+ * @spi_flash: SPI flash device
+ * @spi_eeprom: SPI EEPROM device
+ * @spi_lock: SPI bus lock
+ * @mdio_lock: MDIO bus lock
+ * @xmac_poll_required: XMAC link state needs polling
+ */
+struct falcon_nic_data {
+	struct pci_dev *pci_dev2;
+	struct falcon_board board;
+	unsigned int stats_disable_count;
+	bool stats_pending;
+	struct timer_list stats_timer;
+	u32 *stats_dma_done;
+	struct efx_spi_device spi_flash;
+	struct efx_spi_device spi_eeprom;
+	struct mutex spi_lock;
+	struct mutex mdio_lock;
+	bool xmac_poll_required;
+};
+
+static inline struct falcon_board *falcon_board(struct efx_nic *efx)
+{
+	struct falcon_nic_data *data = efx->nic_data;
+	return &data->board;
+}
+
+/**
+ * struct siena_nic_data - Siena NIC state
+ * @mcdi: Management-Controller-to-Driver Interface
+ * @mcdi_smem: MCDI shared memory mapping. The mapping is always uncacheable.
+ * @wol_filter_id: Wake-on-LAN packet filter id
+ */
+struct siena_nic_data {
+	struct efx_mcdi_iface mcdi;
+	void __iomem *mcdi_smem;
+	int wol_filter_id;
+};
+
+extern const struct efx_nic_type falcon_a1_nic_type;
+extern const struct efx_nic_type falcon_b0_nic_type;
+extern const struct efx_nic_type siena_a0_nic_type;
+
+/**************************************************************************
+ *
+ * Externs
+ *
+ **************************************************************************
+ */
+
+extern int falcon_probe_board(struct efx_nic *efx, u16 revision_info);
+
+/* TX data path */
+extern int efx_nic_probe_tx(struct efx_tx_queue *tx_queue);
+extern void efx_nic_init_tx(struct efx_tx_queue *tx_queue);
+extern void efx_nic_fini_tx(struct efx_tx_queue *tx_queue);
+extern void efx_nic_remove_tx(struct efx_tx_queue *tx_queue);
+extern void efx_nic_push_buffers(struct efx_tx_queue *tx_queue);
+
+/* RX data path */
+extern int efx_nic_probe_rx(struct efx_rx_queue *rx_queue);
+extern void efx_nic_init_rx(struct efx_rx_queue *rx_queue);
+extern void efx_nic_fini_rx(struct efx_rx_queue *rx_queue);
+extern void efx_nic_remove_rx(struct efx_rx_queue *rx_queue);
+extern void efx_nic_notify_rx_desc(struct efx_rx_queue *rx_queue);
+
+/* Event data path */
+extern int efx_nic_probe_eventq(struct efx_channel *channel);
+extern void efx_nic_init_eventq(struct efx_channel *channel);
+extern void efx_nic_fini_eventq(struct efx_channel *channel);
+extern void efx_nic_remove_eventq(struct efx_channel *channel);
+extern int efx_nic_process_eventq(struct efx_channel *channel, int rx_quota);
+extern void efx_nic_eventq_read_ack(struct efx_channel *channel);
+extern bool efx_nic_event_present(struct efx_channel *channel);
+
+/* MAC/PHY */
+extern void falcon_drain_tx_fifo(struct efx_nic *efx);
+extern void falcon_reconfigure_mac_wrapper(struct efx_nic *efx);
+
+/* Interrupts and test events */
+extern int efx_nic_init_interrupt(struct efx_nic *efx);
+extern void efx_nic_enable_interrupts(struct efx_nic *efx);
+extern void efx_nic_generate_test_event(struct efx_channel *channel);
+extern void efx_nic_generate_fill_event(struct efx_channel *channel);
+extern void efx_nic_generate_interrupt(struct efx_nic *efx);
+extern void efx_nic_disable_interrupts(struct efx_nic *efx);
+extern void efx_nic_fini_interrupt(struct efx_nic *efx);
+extern irqreturn_t efx_nic_fatal_interrupt(struct efx_nic *efx);
+extern irqreturn_t falcon_legacy_interrupt_a1(int irq, void *dev_id);
+extern void falcon_irq_ack_a1(struct efx_nic *efx);
+
+#define EFX_IRQ_MOD_RESOLUTION 5
+
+/* Global Resources */
+extern int efx_nic_flush_queues(struct efx_nic *efx);
+extern void falcon_start_nic_stats(struct efx_nic *efx);
+extern void falcon_stop_nic_stats(struct efx_nic *efx);
+extern void falcon_setup_xaui(struct efx_nic *efx);
+extern int falcon_reset_xaui(struct efx_nic *efx);
+extern void efx_nic_init_common(struct efx_nic *efx);
+extern void efx_nic_push_rx_indir_table(struct efx_nic *efx);
+
+int efx_nic_alloc_buffer(struct efx_nic *efx, struct efx_buffer *buffer,
+			 unsigned int len);
+void efx_nic_free_buffer(struct efx_nic *efx, struct efx_buffer *buffer);
+
+/* Tests */
+struct efx_nic_register_test {
+	unsigned address;
+	efx_oword_t mask;
+};
+extern int efx_nic_test_registers(struct efx_nic *efx,
+				  const struct efx_nic_register_test *regs,
+				  size_t n_regs);
+
+extern size_t efx_nic_get_regs_len(struct efx_nic *efx);
+extern void efx_nic_get_regs(struct efx_nic *efx, void *buf);
+
+/**************************************************************************
+ *
+ * Falcon MAC stats
+ *
+ **************************************************************************
+ */
+
+#define FALCON_STAT_OFFSET(falcon_stat) EFX_VAL(falcon_stat, offset)
+#define FALCON_STAT_WIDTH(falcon_stat) EFX_VAL(falcon_stat, WIDTH)
+
+/* Retrieve statistic from statistics block */
+#define FALCON_STAT(efx, falcon_stat, efx_stat) do {		\
+	if (FALCON_STAT_WIDTH(falcon_stat) == 16)		\
+		(efx)->mac_stats.efx_stat += le16_to_cpu(	\
+			*((__force __le16 *)				\
+			  (efx->stats_buffer.addr +		\
+			   FALCON_STAT_OFFSET(falcon_stat))));	\
+	else if (FALCON_STAT_WIDTH(falcon_stat) == 32)		\
+		(efx)->mac_stats.efx_stat += le32_to_cpu(	\
+			*((__force __le32 *)				\
+			  (efx->stats_buffer.addr +		\
+			   FALCON_STAT_OFFSET(falcon_stat))));	\
+	else							\
+		(efx)->mac_stats.efx_stat += le64_to_cpu(	\
+			*((__force __le64 *)				\
+			  (efx->stats_buffer.addr +		\
+			   FALCON_STAT_OFFSET(falcon_stat))));	\
+	} while (0)
+
+#define FALCON_MAC_STATS_SIZE 0x100
+
+#define MAC_DATA_LBN 0
+#define MAC_DATA_WIDTH 32
+
+extern void efx_nic_generate_event(struct efx_channel *channel,
+				   efx_qword_t *event);
+
+extern void falcon_poll_xmac(struct efx_nic *efx);
+
+#endif /* EFX_NIC_H */
diff --git a/drivers/net/ethernet/sfc/phy.h b/drivers/net/ethernet/sfc/phy.h
new file mode 100644
index 0000000..11d148c
--- /dev/null
+++ b/drivers/net/ethernet/sfc/phy.h
@@ -0,0 +1,67 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2007-2010 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#ifndef EFX_PHY_H
+#define EFX_PHY_H
+
+/****************************************************************************
+ * 10Xpress (SFX7101) PHY
+ */
+extern const struct efx_phy_operations falcon_sfx7101_phy_ops;
+
+extern void tenxpress_set_id_led(struct efx_nic *efx, enum efx_led_mode mode);
+
+/****************************************************************************
+ * AMCC/Quake QT202x PHYs
+ */
+extern const struct efx_phy_operations falcon_qt202x_phy_ops;
+
+/* These PHYs provide various H/W control states for LEDs */
+#define QUAKE_LED_LINK_INVAL	(0)
+#define QUAKE_LED_LINK_STAT	(1)
+#define QUAKE_LED_LINK_ACT	(2)
+#define QUAKE_LED_LINK_ACTSTAT	(3)
+#define QUAKE_LED_OFF		(4)
+#define QUAKE_LED_ON		(5)
+#define QUAKE_LED_LINK_INPUT	(6)	/* Pin is an input. */
+/* What link the LED tracks */
+#define QUAKE_LED_TXLINK	(0)
+#define QUAKE_LED_RXLINK	(8)
+
+extern void falcon_qt202x_set_led(struct efx_nic *p, int led, int state);
+
+/****************************************************************************
+* Transwitch CX4 retimer
+*/
+extern const struct efx_phy_operations falcon_txc_phy_ops;
+
+#define TXC_GPIO_DIR_INPUT	0
+#define TXC_GPIO_DIR_OUTPUT	1
+
+extern void falcon_txc_set_gpio_dir(struct efx_nic *efx, int pin, int dir);
+extern void falcon_txc_set_gpio_val(struct efx_nic *efx, int pin, int val);
+
+/****************************************************************************
+ * Siena managed PHYs
+ */
+extern const struct efx_phy_operations efx_mcdi_phy_ops;
+
+extern int efx_mcdi_mdio_read(struct efx_nic *efx, unsigned int bus,
+			      unsigned int prtad, unsigned int devad,
+			      u16 addr, u16 *value_out, u32 *status_out);
+extern int efx_mcdi_mdio_write(struct efx_nic *efx, unsigned int bus,
+			       unsigned int prtad, unsigned int devad,
+			       u16 addr, u16 value, u32 *status_out);
+extern void efx_mcdi_phy_decode_link(struct efx_nic *efx,
+				     struct efx_link_state *link_state,
+				     u32 speed, u32 flags, u32 fcntl);
+extern int efx_mcdi_phy_reconfigure(struct efx_nic *efx);
+extern void efx_mcdi_phy_check_fcntl(struct efx_nic *efx, u32 lpa);
+
+#endif
diff --git a/drivers/net/ethernet/sfc/qt202x_phy.c b/drivers/net/ethernet/sfc/qt202x_phy.c
new file mode 100644
index 0000000..7ad97e3
--- /dev/null
+++ b/drivers/net/ethernet/sfc/qt202x_phy.c
@@ -0,0 +1,462 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2006-2010 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+/*
+ * Driver for AMCC QT202x SFP+ and XFP adapters; see www.amcc.com for details
+ */
+
+#include <linux/slab.h>
+#include <linux/timer.h>
+#include <linux/delay.h>
+#include "efx.h"
+#include "mdio_10g.h"
+#include "phy.h"
+#include "nic.h"
+
+#define QT202X_REQUIRED_DEVS (MDIO_DEVS_PCS |		\
+			      MDIO_DEVS_PMAPMD |	\
+			      MDIO_DEVS_PHYXS)
+
+#define QT202X_LOOPBACKS ((1 << LOOPBACK_PCS) |		\
+			  (1 << LOOPBACK_PMAPMD) |	\
+			  (1 << LOOPBACK_PHYXS_WS))
+
+/****************************************************************************/
+/* Quake-specific MDIO registers */
+#define MDIO_QUAKE_LED0_REG	(0xD006)
+
+/* QT2025C only */
+#define PCS_FW_HEARTBEAT_REG	0xd7ee
+#define PCS_FW_HEARTB_LBN	0
+#define PCS_FW_HEARTB_WIDTH	8
+#define PCS_FW_PRODUCT_CODE_1	0xd7f0
+#define PCS_FW_VERSION_1	0xd7f3
+#define PCS_FW_BUILD_1		0xd7f6
+#define PCS_UC8051_STATUS_REG	0xd7fd
+#define PCS_UC_STATUS_LBN	0
+#define PCS_UC_STATUS_WIDTH	8
+#define PCS_UC_STATUS_FW_SAVE	0x20
+#define PMA_PMD_MODE_REG	0xc301
+#define PMA_PMD_RXIN_SEL_LBN	6
+#define PMA_PMD_FTX_CTRL2_REG	0xc309
+#define PMA_PMD_FTX_STATIC_LBN	13
+#define PMA_PMD_VEND1_REG	0xc001
+#define PMA_PMD_VEND1_LBTXD_LBN	15
+#define PCS_VEND1_REG	   	0xc000
+#define PCS_VEND1_LBTXD_LBN	5
+
+void falcon_qt202x_set_led(struct efx_nic *p, int led, int mode)
+{
+	int addr = MDIO_QUAKE_LED0_REG + led;
+	efx_mdio_write(p, MDIO_MMD_PMAPMD, addr, mode);
+}
+
+struct qt202x_phy_data {
+	enum efx_phy_mode phy_mode;
+	bool bug17190_in_bad_state;
+	unsigned long bug17190_timer;
+	u32 firmware_ver;
+};
+
+#define QT2022C2_MAX_RESET_TIME 500
+#define QT2022C2_RESET_WAIT 10
+
+#define QT2025C_MAX_HEARTB_TIME (5 * HZ)
+#define QT2025C_HEARTB_WAIT 100
+#define QT2025C_MAX_FWSTART_TIME (25 * HZ / 10)
+#define QT2025C_FWSTART_WAIT 100
+
+#define BUG17190_INTERVAL (2 * HZ)
+
+static int qt2025c_wait_heartbeat(struct efx_nic *efx)
+{
+	unsigned long timeout = jiffies + QT2025C_MAX_HEARTB_TIME;
+	int reg, old_counter = 0;
+
+	/* Wait for firmware heartbeat to start */
+	for (;;) {
+		int counter;
+		reg = efx_mdio_read(efx, MDIO_MMD_PCS, PCS_FW_HEARTBEAT_REG);
+		if (reg < 0)
+			return reg;
+		counter = ((reg >> PCS_FW_HEARTB_LBN) &
+			    ((1 << PCS_FW_HEARTB_WIDTH) - 1));
+		if (old_counter == 0)
+			old_counter = counter;
+		else if (counter != old_counter)
+			break;
+		if (time_after(jiffies, timeout)) {
+			/* Some cables have EEPROMs that conflict with the
+			 * PHY's on-board EEPROM so it cannot load firmware */
+			netif_err(efx, hw, efx->net_dev,
+				  "If an SFP+ direct attach cable is"
+				  " connected, please check that it complies"
+				  " with the SFP+ specification\n");
+			return -ETIMEDOUT;
+		}
+		msleep(QT2025C_HEARTB_WAIT);
+	}
+
+	return 0;
+}
+
+static int qt2025c_wait_fw_status_good(struct efx_nic *efx)
+{
+	unsigned long timeout = jiffies + QT2025C_MAX_FWSTART_TIME;
+	int reg;
+
+	/* Wait for firmware status to look good */
+	for (;;) {
+		reg = efx_mdio_read(efx, MDIO_MMD_PCS, PCS_UC8051_STATUS_REG);
+		if (reg < 0)
+			return reg;
+		if ((reg &
+		     ((1 << PCS_UC_STATUS_WIDTH) - 1) << PCS_UC_STATUS_LBN) >=
+		    PCS_UC_STATUS_FW_SAVE)
+			break;
+		if (time_after(jiffies, timeout))
+			return -ETIMEDOUT;
+		msleep(QT2025C_FWSTART_WAIT);
+	}
+
+	return 0;
+}
+
+static void qt2025c_restart_firmware(struct efx_nic *efx)
+{
+	/* Restart microcontroller execution of firmware from RAM */
+	efx_mdio_write(efx, 3, 0xe854, 0x00c0);
+	efx_mdio_write(efx, 3, 0xe854, 0x0040);
+	msleep(50);
+}
+
+static int qt2025c_wait_reset(struct efx_nic *efx)
+{
+	int rc;
+
+	rc = qt2025c_wait_heartbeat(efx);
+	if (rc != 0)
+		return rc;
+
+	rc = qt2025c_wait_fw_status_good(efx);
+	if (rc == -ETIMEDOUT) {
+		/* Bug 17689: occasionally heartbeat starts but firmware status
+		 * code never progresses beyond 0x00.  Try again, once, after
+		 * restarting execution of the firmware image. */
+		netif_dbg(efx, hw, efx->net_dev,
+			  "bashing QT2025C microcontroller\n");
+		qt2025c_restart_firmware(efx);
+		rc = qt2025c_wait_heartbeat(efx);
+		if (rc != 0)
+			return rc;
+		rc = qt2025c_wait_fw_status_good(efx);
+	}
+
+	return rc;
+}
+
+static void qt2025c_firmware_id(struct efx_nic *efx)
+{
+	struct qt202x_phy_data *phy_data = efx->phy_data;
+	u8 firmware_id[9];
+	size_t i;
+
+	for (i = 0; i < sizeof(firmware_id); i++)
+		firmware_id[i] = efx_mdio_read(efx, MDIO_MMD_PCS,
+					       PCS_FW_PRODUCT_CODE_1 + i);
+	netif_info(efx, probe, efx->net_dev,
+		   "QT2025C firmware %xr%d v%d.%d.%d.%d [20%02d-%02d-%02d]\n",
+		   (firmware_id[0] << 8) | firmware_id[1], firmware_id[2],
+		   firmware_id[3] >> 4, firmware_id[3] & 0xf,
+		   firmware_id[4], firmware_id[5],
+		   firmware_id[6], firmware_id[7], firmware_id[8]);
+	phy_data->firmware_ver = ((firmware_id[3] & 0xf0) << 20) |
+				 ((firmware_id[3] & 0x0f) << 16) |
+				 (firmware_id[4] << 8) | firmware_id[5];
+}
+
+static void qt2025c_bug17190_workaround(struct efx_nic *efx)
+{
+	struct qt202x_phy_data *phy_data = efx->phy_data;
+
+	/* The PHY can get stuck in a state where it reports PHY_XS and PMA/PMD
+	 * layers up, but PCS down (no block_lock).  If we notice this state
+	 * persisting for a couple of seconds, we switch PMA/PMD loopback
+	 * briefly on and then off again, which is normally sufficient to
+	 * recover it.
+	 */
+	if (efx->link_state.up ||
+	    !efx_mdio_links_ok(efx, MDIO_DEVS_PMAPMD | MDIO_DEVS_PHYXS)) {
+		phy_data->bug17190_in_bad_state = false;
+		return;
+	}
+
+	if (!phy_data->bug17190_in_bad_state) {
+		phy_data->bug17190_in_bad_state = true;
+		phy_data->bug17190_timer = jiffies + BUG17190_INTERVAL;
+		return;
+	}
+
+	if (time_after_eq(jiffies, phy_data->bug17190_timer)) {
+		netif_dbg(efx, hw, efx->net_dev, "bashing QT2025C PMA/PMD\n");
+		efx_mdio_set_flag(efx, MDIO_MMD_PMAPMD, MDIO_CTRL1,
+				  MDIO_PMA_CTRL1_LOOPBACK, true);
+		msleep(100);
+		efx_mdio_set_flag(efx, MDIO_MMD_PMAPMD, MDIO_CTRL1,
+				  MDIO_PMA_CTRL1_LOOPBACK, false);
+		phy_data->bug17190_timer = jiffies + BUG17190_INTERVAL;
+	}
+}
+
+static int qt2025c_select_phy_mode(struct efx_nic *efx)
+{
+	struct qt202x_phy_data *phy_data = efx->phy_data;
+	struct falcon_board *board = falcon_board(efx);
+	int reg, rc, i;
+	uint16_t phy_op_mode;
+
+	/* Only 2.0.1.0+ PHY firmware supports the more optimal SFP+
+	 * Self-Configure mode.  Don't attempt any switching if we encounter
+	 * older firmware. */
+	if (phy_data->firmware_ver < 0x02000100)
+		return 0;
+
+	/* In general we will get optimal behaviour in "SFP+ Self-Configure"
+	 * mode; however, that powers down most of the PHY when no module is
+	 * present, so we must use a different mode (any fixed mode will do)
+	 * to be sure that loopbacks will work. */
+	phy_op_mode = (efx->loopback_mode == LOOPBACK_NONE) ? 0x0038 : 0x0020;
+
+	/* Only change mode if really necessary */
+	reg = efx_mdio_read(efx, 1, 0xc319);
+	if ((reg & 0x0038) == phy_op_mode)
+		return 0;
+	netif_dbg(efx, hw, efx->net_dev, "Switching PHY to mode 0x%04x\n",
+		  phy_op_mode);
+
+	/* This sequence replicates the register writes configured in the boot
+	 * EEPROM (including the differences between board revisions), except
+	 * that the operating mode is changed, and the PHY is prevented from
+	 * unnecessarily reloading the main firmware image again. */
+	efx_mdio_write(efx, 1, 0xc300, 0x0000);
+	/* (Note: this portion of the boot EEPROM sequence, which bit-bashes 9
+	 * STOPs onto the firmware/module I2C bus to reset it, varies across
+	 * board revisions, as the bus is connected to different GPIO/LED
+	 * outputs on the PHY.) */
+	if (board->major == 0 && board->minor < 2) {
+		efx_mdio_write(efx, 1, 0xc303, 0x4498);
+		for (i = 0; i < 9; i++) {
+			efx_mdio_write(efx, 1, 0xc303, 0x4488);
+			efx_mdio_write(efx, 1, 0xc303, 0x4480);
+			efx_mdio_write(efx, 1, 0xc303, 0x4490);
+			efx_mdio_write(efx, 1, 0xc303, 0x4498);
+		}
+	} else {
+		efx_mdio_write(efx, 1, 0xc303, 0x0920);
+		efx_mdio_write(efx, 1, 0xd008, 0x0004);
+		for (i = 0; i < 9; i++) {
+			efx_mdio_write(efx, 1, 0xc303, 0x0900);
+			efx_mdio_write(efx, 1, 0xd008, 0x0005);
+			efx_mdio_write(efx, 1, 0xc303, 0x0920);
+			efx_mdio_write(efx, 1, 0xd008, 0x0004);
+		}
+		efx_mdio_write(efx, 1, 0xc303, 0x4900);
+	}
+	efx_mdio_write(efx, 1, 0xc303, 0x4900);
+	efx_mdio_write(efx, 1, 0xc302, 0x0004);
+	efx_mdio_write(efx, 1, 0xc316, 0x0013);
+	efx_mdio_write(efx, 1, 0xc318, 0x0054);
+	efx_mdio_write(efx, 1, 0xc319, phy_op_mode);
+	efx_mdio_write(efx, 1, 0xc31a, 0x0098);
+	efx_mdio_write(efx, 3, 0x0026, 0x0e00);
+	efx_mdio_write(efx, 3, 0x0027, 0x0013);
+	efx_mdio_write(efx, 3, 0x0028, 0xa528);
+	efx_mdio_write(efx, 1, 0xd006, 0x000a);
+	efx_mdio_write(efx, 1, 0xd007, 0x0009);
+	efx_mdio_write(efx, 1, 0xd008, 0x0004);
+	/* This additional write is not present in the boot EEPROM.  It
+	 * prevents the PHY's internal boot ROM doing another pointless (and
+	 * slow) reload of the firmware image (the microcontroller's code
+	 * memory is not affected by the microcontroller reset). */
+	efx_mdio_write(efx, 1, 0xc317, 0x00ff);
+	/* PMA/PMD loopback sets RXIN to inverse polarity and the firmware
+	 * restart doesn't reset it. We need to do that ourselves. */
+	efx_mdio_set_flag(efx, 1, PMA_PMD_MODE_REG,
+			  1 << PMA_PMD_RXIN_SEL_LBN, false);
+	efx_mdio_write(efx, 1, 0xc300, 0x0002);
+	msleep(20);
+
+	/* Restart microcontroller execution of firmware from RAM */
+	qt2025c_restart_firmware(efx);
+
+	/* Wait for the microcontroller to be ready again */
+	rc = qt2025c_wait_reset(efx);
+	if (rc < 0) {
+		netif_err(efx, hw, efx->net_dev,
+			  "PHY microcontroller reset during mode switch "
+			  "timed out\n");
+		return rc;
+	}
+
+	return 0;
+}
+
+static int qt202x_reset_phy(struct efx_nic *efx)
+{
+	int rc;
+
+	if (efx->phy_type == PHY_TYPE_QT2025C) {
+		/* Wait for the reset triggered by falcon_reset_hw()
+		 * to complete */
+		rc = qt2025c_wait_reset(efx);
+		if (rc < 0)
+			goto fail;
+	} else {
+		/* Reset the PHYXS MMD. This is documented as doing
+		 * a complete soft reset. */
+		rc = efx_mdio_reset_mmd(efx, MDIO_MMD_PHYXS,
+					QT2022C2_MAX_RESET_TIME /
+					QT2022C2_RESET_WAIT,
+					QT2022C2_RESET_WAIT);
+		if (rc < 0)
+			goto fail;
+	}
+
+	/* Wait 250ms for the PHY to complete bootup */
+	msleep(250);
+
+	falcon_board(efx)->type->init_phy(efx);
+
+	return 0;
+
+ fail:
+	netif_err(efx, hw, efx->net_dev, "PHY reset timed out\n");
+	return rc;
+}
+
+static int qt202x_phy_probe(struct efx_nic *efx)
+{
+	struct qt202x_phy_data *phy_data;
+
+	phy_data = kzalloc(sizeof(struct qt202x_phy_data), GFP_KERNEL);
+	if (!phy_data)
+		return -ENOMEM;
+	efx->phy_data = phy_data;
+	phy_data->phy_mode = efx->phy_mode;
+	phy_data->bug17190_in_bad_state = false;
+	phy_data->bug17190_timer = 0;
+
+	efx->mdio.mmds = QT202X_REQUIRED_DEVS;
+	efx->mdio.mode_support = MDIO_SUPPORTS_C45 | MDIO_EMULATE_C22;
+	efx->loopback_modes = QT202X_LOOPBACKS | FALCON_XMAC_LOOPBACKS;
+	return 0;
+}
+
+static int qt202x_phy_init(struct efx_nic *efx)
+{
+	u32 devid;
+	int rc;
+
+	rc = qt202x_reset_phy(efx);
+	if (rc) {
+		netif_err(efx, probe, efx->net_dev, "PHY init failed\n");
+		return rc;
+	}
+
+	devid = efx_mdio_read_id(efx, MDIO_MMD_PHYXS);
+	netif_info(efx, probe, efx->net_dev,
+		   "PHY ID reg %x (OUI %06x model %02x revision %x)\n",
+		   devid, efx_mdio_id_oui(devid), efx_mdio_id_model(devid),
+		   efx_mdio_id_rev(devid));
+
+	if (efx->phy_type == PHY_TYPE_QT2025C)
+		qt2025c_firmware_id(efx);
+
+	return 0;
+}
+
+static int qt202x_link_ok(struct efx_nic *efx)
+{
+	return efx_mdio_links_ok(efx, QT202X_REQUIRED_DEVS);
+}
+
+static bool qt202x_phy_poll(struct efx_nic *efx)
+{
+	bool was_up = efx->link_state.up;
+
+	efx->link_state.up = qt202x_link_ok(efx);
+	efx->link_state.speed = 10000;
+	efx->link_state.fd = true;
+	efx->link_state.fc = efx->wanted_fc;
+
+	if (efx->phy_type == PHY_TYPE_QT2025C)
+		qt2025c_bug17190_workaround(efx);
+
+	return efx->link_state.up != was_up;
+}
+
+static int qt202x_phy_reconfigure(struct efx_nic *efx)
+{
+	struct qt202x_phy_data *phy_data = efx->phy_data;
+
+	if (efx->phy_type == PHY_TYPE_QT2025C) {
+		int rc = qt2025c_select_phy_mode(efx);
+		if (rc)
+			return rc;
+
+		/* There are several different register bits which can
+		 * disable TX (and save power) on direct-attach cables
+		 * or optical transceivers, varying somewhat between
+		 * firmware versions.  Only 'static mode' appears to
+		 * cover everything. */
+		mdio_set_flag(
+			&efx->mdio, efx->mdio.prtad, MDIO_MMD_PMAPMD,
+			PMA_PMD_FTX_CTRL2_REG, 1 << PMA_PMD_FTX_STATIC_LBN,
+			efx->phy_mode & PHY_MODE_TX_DISABLED ||
+			efx->phy_mode & PHY_MODE_LOW_POWER ||
+			efx->loopback_mode == LOOPBACK_PCS ||
+			efx->loopback_mode == LOOPBACK_PMAPMD);
+	} else {
+		/* Reset the PHY when moving from tx off to tx on */
+		if (!(efx->phy_mode & PHY_MODE_TX_DISABLED) &&
+		    (phy_data->phy_mode & PHY_MODE_TX_DISABLED))
+			qt202x_reset_phy(efx);
+
+		efx_mdio_transmit_disable(efx);
+	}
+
+	efx_mdio_phy_reconfigure(efx);
+
+	phy_data->phy_mode = efx->phy_mode;
+
+	return 0;
+}
+
+static void qt202x_phy_get_settings(struct efx_nic *efx, struct ethtool_cmd *ecmd)
+{
+	mdio45_ethtool_gset(&efx->mdio, ecmd);
+}
+
+static void qt202x_phy_remove(struct efx_nic *efx)
+{
+	/* Free the context block */
+	kfree(efx->phy_data);
+	efx->phy_data = NULL;
+}
+
+const struct efx_phy_operations falcon_qt202x_phy_ops = {
+	.probe		 = qt202x_phy_probe,
+	.init		 = qt202x_phy_init,
+	.reconfigure	 = qt202x_phy_reconfigure,
+	.poll	     	 = qt202x_phy_poll,
+	.fini		 = efx_port_dummy_op_void,
+	.remove	  	 = qt202x_phy_remove,
+	.get_settings	 = qt202x_phy_get_settings,
+	.set_settings	 = efx_mdio_set_settings,
+	.test_alive	 = efx_mdio_test_alive,
+};
diff --git a/drivers/net/ethernet/sfc/regs.h b/drivers/net/ethernet/sfc/regs.h
new file mode 100644
index 0000000..cc2c86b
--- /dev/null
+++ b/drivers/net/ethernet/sfc/regs.h
@@ -0,0 +1,3188 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2005-2006 Fen Systems Ltd.
+ * Copyright 2006-2010 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#ifndef EFX_REGS_H
+#define EFX_REGS_H
+
+/*
+ * Falcon hardware architecture definitions have a name prefix following
+ * the format:
+ *
+ *     F<type>_<min-rev><max-rev>_
+ *
+ * The following <type> strings are used:
+ *
+ *             MMIO register  MC register  Host memory structure
+ * -------------------------------------------------------------
+ * Address     R              MCR
+ * Bitfield    RF             MCRF         SF
+ * Enumerator  FE             MCFE         SE
+ *
+ * <min-rev> is the first revision to which the definition applies:
+ *
+ *     A: Falcon A1 (SFC4000AB)
+ *     B: Falcon B0 (SFC4000BA)
+ *     C: Siena A0 (SFL9021AA)
+ *
+ * If the definition has been changed or removed in later revisions
+ * then <max-rev> is the last revision to which the definition applies;
+ * otherwise it is "Z".
+ */
+
+/**************************************************************************
+ *
+ * Falcon/Siena registers and descriptors
+ *
+ **************************************************************************
+ */
+
+/* ADR_REGION_REG: Address region register */
+#define	FR_AZ_ADR_REGION 0x00000000
+#define	FRF_AZ_ADR_REGION3_LBN 96
+#define	FRF_AZ_ADR_REGION3_WIDTH 18
+#define	FRF_AZ_ADR_REGION2_LBN 64
+#define	FRF_AZ_ADR_REGION2_WIDTH 18
+#define	FRF_AZ_ADR_REGION1_LBN 32
+#define	FRF_AZ_ADR_REGION1_WIDTH 18
+#define	FRF_AZ_ADR_REGION0_LBN 0
+#define	FRF_AZ_ADR_REGION0_WIDTH 18
+
+/* INT_EN_REG_KER: Kernel driver Interrupt enable register */
+#define	FR_AZ_INT_EN_KER 0x00000010
+#define	FRF_AZ_KER_INT_LEVE_SEL_LBN 8
+#define	FRF_AZ_KER_INT_LEVE_SEL_WIDTH 6
+#define	FRF_AZ_KER_INT_CHAR_LBN 4
+#define	FRF_AZ_KER_INT_CHAR_WIDTH 1
+#define	FRF_AZ_KER_INT_KER_LBN 3
+#define	FRF_AZ_KER_INT_KER_WIDTH 1
+#define	FRF_AZ_DRV_INT_EN_KER_LBN 0
+#define	FRF_AZ_DRV_INT_EN_KER_WIDTH 1
+
+/* INT_EN_REG_CHAR: Char Driver interrupt enable register */
+#define	FR_BZ_INT_EN_CHAR 0x00000020
+#define	FRF_BZ_CHAR_INT_LEVE_SEL_LBN 8
+#define	FRF_BZ_CHAR_INT_LEVE_SEL_WIDTH 6
+#define	FRF_BZ_CHAR_INT_CHAR_LBN 4
+#define	FRF_BZ_CHAR_INT_CHAR_WIDTH 1
+#define	FRF_BZ_CHAR_INT_KER_LBN 3
+#define	FRF_BZ_CHAR_INT_KER_WIDTH 1
+#define	FRF_BZ_DRV_INT_EN_CHAR_LBN 0
+#define	FRF_BZ_DRV_INT_EN_CHAR_WIDTH 1
+
+/* INT_ADR_REG_KER: Interrupt host address for Kernel driver */
+#define	FR_AZ_INT_ADR_KER 0x00000030
+#define	FRF_AZ_NORM_INT_VEC_DIS_KER_LBN 64
+#define	FRF_AZ_NORM_INT_VEC_DIS_KER_WIDTH 1
+#define	FRF_AZ_INT_ADR_KER_LBN 0
+#define	FRF_AZ_INT_ADR_KER_WIDTH 64
+
+/* INT_ADR_REG_CHAR: Interrupt host address for Char driver */
+#define	FR_BZ_INT_ADR_CHAR 0x00000040
+#define	FRF_BZ_NORM_INT_VEC_DIS_CHAR_LBN 64
+#define	FRF_BZ_NORM_INT_VEC_DIS_CHAR_WIDTH 1
+#define	FRF_BZ_INT_ADR_CHAR_LBN 0
+#define	FRF_BZ_INT_ADR_CHAR_WIDTH 64
+
+/* INT_ACK_KER: Kernel interrupt acknowledge register */
+#define	FR_AA_INT_ACK_KER 0x00000050
+#define	FRF_AA_INT_ACK_KER_FIELD_LBN 0
+#define	FRF_AA_INT_ACK_KER_FIELD_WIDTH 32
+
+/* INT_ISR0_REG: Function 0 Interrupt Acknowledge Status register */
+#define	FR_BZ_INT_ISR0 0x00000090
+#define	FRF_BZ_INT_ISR_REG_LBN 0
+#define	FRF_BZ_INT_ISR_REG_WIDTH 64
+
+/* HW_INIT_REG: Hardware initialization register */
+#define	FR_AZ_HW_INIT 0x000000c0
+#define	FRF_BB_BDMRD_CPLF_FULL_LBN 124
+#define	FRF_BB_BDMRD_CPLF_FULL_WIDTH 1
+#define	FRF_BB_PCIE_CPL_TIMEOUT_CTRL_LBN 121
+#define	FRF_BB_PCIE_CPL_TIMEOUT_CTRL_WIDTH 3
+#define	FRF_CZ_TX_MRG_TAGS_LBN 120
+#define	FRF_CZ_TX_MRG_TAGS_WIDTH 1
+#define	FRF_AB_TRGT_MASK_ALL_LBN 100
+#define	FRF_AB_TRGT_MASK_ALL_WIDTH 1
+#define	FRF_AZ_DOORBELL_DROP_LBN 92
+#define	FRF_AZ_DOORBELL_DROP_WIDTH 8
+#define	FRF_AB_TX_RREQ_MASK_EN_LBN 76
+#define	FRF_AB_TX_RREQ_MASK_EN_WIDTH 1
+#define	FRF_AB_PE_EIDLE_DIS_LBN 75
+#define	FRF_AB_PE_EIDLE_DIS_WIDTH 1
+#define	FRF_AA_FC_BLOCKING_EN_LBN 45
+#define	FRF_AA_FC_BLOCKING_EN_WIDTH 1
+#define	FRF_BZ_B2B_REQ_EN_LBN 45
+#define	FRF_BZ_B2B_REQ_EN_WIDTH 1
+#define	FRF_AA_B2B_REQ_EN_LBN 44
+#define	FRF_AA_B2B_REQ_EN_WIDTH 1
+#define	FRF_BB_FC_BLOCKING_EN_LBN 44
+#define	FRF_BB_FC_BLOCKING_EN_WIDTH 1
+#define	FRF_AZ_POST_WR_MASK_LBN 40
+#define	FRF_AZ_POST_WR_MASK_WIDTH 4
+#define	FRF_AZ_TLP_TC_LBN 34
+#define	FRF_AZ_TLP_TC_WIDTH 3
+#define	FRF_AZ_TLP_ATTR_LBN 32
+#define	FRF_AZ_TLP_ATTR_WIDTH 2
+#define	FRF_AB_INTB_VEC_LBN 24
+#define	FRF_AB_INTB_VEC_WIDTH 5
+#define	FRF_AB_INTA_VEC_LBN 16
+#define	FRF_AB_INTA_VEC_WIDTH 5
+#define	FRF_AZ_WD_TIMER_LBN 8
+#define	FRF_AZ_WD_TIMER_WIDTH 8
+#define	FRF_AZ_US_DISABLE_LBN 5
+#define	FRF_AZ_US_DISABLE_WIDTH 1
+#define	FRF_AZ_TLP_EP_LBN 4
+#define	FRF_AZ_TLP_EP_WIDTH 1
+#define	FRF_AZ_ATTR_SEL_LBN 3
+#define	FRF_AZ_ATTR_SEL_WIDTH 1
+#define	FRF_AZ_TD_SEL_LBN 1
+#define	FRF_AZ_TD_SEL_WIDTH 1
+#define	FRF_AZ_TLP_TD_LBN 0
+#define	FRF_AZ_TLP_TD_WIDTH 1
+
+/* EE_SPI_HCMD_REG: SPI host command register */
+#define	FR_AB_EE_SPI_HCMD 0x00000100
+#define	FRF_AB_EE_SPI_HCMD_CMD_EN_LBN 31
+#define	FRF_AB_EE_SPI_HCMD_CMD_EN_WIDTH 1
+#define	FRF_AB_EE_WR_TIMER_ACTIVE_LBN 28
+#define	FRF_AB_EE_WR_TIMER_ACTIVE_WIDTH 1
+#define	FRF_AB_EE_SPI_HCMD_SF_SEL_LBN 24
+#define	FRF_AB_EE_SPI_HCMD_SF_SEL_WIDTH 1
+#define	FRF_AB_EE_SPI_HCMD_DABCNT_LBN 16
+#define	FRF_AB_EE_SPI_HCMD_DABCNT_WIDTH 5
+#define	FRF_AB_EE_SPI_HCMD_READ_LBN 15
+#define	FRF_AB_EE_SPI_HCMD_READ_WIDTH 1
+#define	FRF_AB_EE_SPI_HCMD_DUBCNT_LBN 12
+#define	FRF_AB_EE_SPI_HCMD_DUBCNT_WIDTH 2
+#define	FRF_AB_EE_SPI_HCMD_ADBCNT_LBN 8
+#define	FRF_AB_EE_SPI_HCMD_ADBCNT_WIDTH 2
+#define	FRF_AB_EE_SPI_HCMD_ENC_LBN 0
+#define	FRF_AB_EE_SPI_HCMD_ENC_WIDTH 8
+
+/* USR_EV_CFG: User Level Event Configuration register */
+#define	FR_CZ_USR_EV_CFG 0x00000100
+#define	FRF_CZ_USREV_DIS_LBN 16
+#define	FRF_CZ_USREV_DIS_WIDTH 1
+#define	FRF_CZ_DFLT_EVQ_LBN 0
+#define	FRF_CZ_DFLT_EVQ_WIDTH 10
+
+/* EE_SPI_HADR_REG: SPI host address register */
+#define	FR_AB_EE_SPI_HADR 0x00000110
+#define	FRF_AB_EE_SPI_HADR_DUBYTE_LBN 24
+#define	FRF_AB_EE_SPI_HADR_DUBYTE_WIDTH 8
+#define	FRF_AB_EE_SPI_HADR_ADR_LBN 0
+#define	FRF_AB_EE_SPI_HADR_ADR_WIDTH 24
+
+/* EE_SPI_HDATA_REG: SPI host data register */
+#define	FR_AB_EE_SPI_HDATA 0x00000120
+#define	FRF_AB_EE_SPI_HDATA3_LBN 96
+#define	FRF_AB_EE_SPI_HDATA3_WIDTH 32
+#define	FRF_AB_EE_SPI_HDATA2_LBN 64
+#define	FRF_AB_EE_SPI_HDATA2_WIDTH 32
+#define	FRF_AB_EE_SPI_HDATA1_LBN 32
+#define	FRF_AB_EE_SPI_HDATA1_WIDTH 32
+#define	FRF_AB_EE_SPI_HDATA0_LBN 0
+#define	FRF_AB_EE_SPI_HDATA0_WIDTH 32
+
+/* EE_BASE_PAGE_REG: Expansion ROM base mirror register */
+#define	FR_AB_EE_BASE_PAGE 0x00000130
+#define	FRF_AB_EE_EXPROM_MASK_LBN 16
+#define	FRF_AB_EE_EXPROM_MASK_WIDTH 13
+#define	FRF_AB_EE_EXP_ROM_WINDOW_BASE_LBN 0
+#define	FRF_AB_EE_EXP_ROM_WINDOW_BASE_WIDTH 13
+
+/* EE_VPD_CFG0_REG: SPI/VPD configuration register 0 */
+#define	FR_AB_EE_VPD_CFG0 0x00000140
+#define	FRF_AB_EE_SF_FASTRD_EN_LBN 127
+#define	FRF_AB_EE_SF_FASTRD_EN_WIDTH 1
+#define	FRF_AB_EE_SF_CLOCK_DIV_LBN 120
+#define	FRF_AB_EE_SF_CLOCK_DIV_WIDTH 7
+#define	FRF_AB_EE_VPD_WIP_POLL_LBN 119
+#define	FRF_AB_EE_VPD_WIP_POLL_WIDTH 1
+#define	FRF_AB_EE_EE_CLOCK_DIV_LBN 112
+#define	FRF_AB_EE_EE_CLOCK_DIV_WIDTH 7
+#define	FRF_AB_EE_EE_WR_TMR_VALUE_LBN 96
+#define	FRF_AB_EE_EE_WR_TMR_VALUE_WIDTH 16
+#define	FRF_AB_EE_VPDW_LENGTH_LBN 80
+#define	FRF_AB_EE_VPDW_LENGTH_WIDTH 15
+#define	FRF_AB_EE_VPDW_BASE_LBN 64
+#define	FRF_AB_EE_VPDW_BASE_WIDTH 15
+#define	FRF_AB_EE_VPD_WR_CMD_EN_LBN 56
+#define	FRF_AB_EE_VPD_WR_CMD_EN_WIDTH 8
+#define	FRF_AB_EE_VPD_BASE_LBN 32
+#define	FRF_AB_EE_VPD_BASE_WIDTH 24
+#define	FRF_AB_EE_VPD_LENGTH_LBN 16
+#define	FRF_AB_EE_VPD_LENGTH_WIDTH 15
+#define	FRF_AB_EE_VPD_AD_SIZE_LBN 8
+#define	FRF_AB_EE_VPD_AD_SIZE_WIDTH 5
+#define	FRF_AB_EE_VPD_ACCESS_ON_LBN 5
+#define	FRF_AB_EE_VPD_ACCESS_ON_WIDTH 1
+#define	FRF_AB_EE_VPD_ACCESS_BLOCK_LBN 4
+#define	FRF_AB_EE_VPD_ACCESS_BLOCK_WIDTH 1
+#define	FRF_AB_EE_VPD_DEV_SF_SEL_LBN 2
+#define	FRF_AB_EE_VPD_DEV_SF_SEL_WIDTH 1
+#define	FRF_AB_EE_VPD_EN_AD9_MODE_LBN 1
+#define	FRF_AB_EE_VPD_EN_AD9_MODE_WIDTH 1
+#define	FRF_AB_EE_VPD_EN_LBN 0
+#define	FRF_AB_EE_VPD_EN_WIDTH 1
+
+/* EE_VPD_SW_CNTL_REG: VPD access SW control register */
+#define	FR_AB_EE_VPD_SW_CNTL 0x00000150
+#define	FRF_AB_EE_VPD_CYCLE_PENDING_LBN 31
+#define	FRF_AB_EE_VPD_CYCLE_PENDING_WIDTH 1
+#define	FRF_AB_EE_VPD_CYC_WRITE_LBN 28
+#define	FRF_AB_EE_VPD_CYC_WRITE_WIDTH 1
+#define	FRF_AB_EE_VPD_CYC_ADR_LBN 0
+#define	FRF_AB_EE_VPD_CYC_ADR_WIDTH 15
+
+/* EE_VPD_SW_DATA_REG: VPD access SW data register */
+#define	FR_AB_EE_VPD_SW_DATA 0x00000160
+#define	FRF_AB_EE_VPD_CYC_DAT_LBN 0
+#define	FRF_AB_EE_VPD_CYC_DAT_WIDTH 32
+
+/* PBMX_DBG_IADDR_REG: Capture Module address register */
+#define	FR_CZ_PBMX_DBG_IADDR 0x000001f0
+#define	FRF_CZ_PBMX_DBG_IADDR_LBN 0
+#define	FRF_CZ_PBMX_DBG_IADDR_WIDTH 32
+
+/* PCIE_CORE_INDIRECT_REG: Indirect Access to PCIE Core registers */
+#define	FR_BB_PCIE_CORE_INDIRECT 0x000001f0
+#define	FRF_BB_PCIE_CORE_TARGET_DATA_LBN 32
+#define	FRF_BB_PCIE_CORE_TARGET_DATA_WIDTH 32
+#define	FRF_BB_PCIE_CORE_INDIRECT_ACCESS_DIR_LBN 15
+#define	FRF_BB_PCIE_CORE_INDIRECT_ACCESS_DIR_WIDTH 1
+#define	FRF_BB_PCIE_CORE_TARGET_REG_ADRS_LBN 0
+#define	FRF_BB_PCIE_CORE_TARGET_REG_ADRS_WIDTH 12
+
+/* PBMX_DBG_IDATA_REG: Capture Module data register */
+#define	FR_CZ_PBMX_DBG_IDATA 0x000001f8
+#define	FRF_CZ_PBMX_DBG_IDATA_LBN 0
+#define	FRF_CZ_PBMX_DBG_IDATA_WIDTH 64
+
+/* NIC_STAT_REG: NIC status register */
+#define	FR_AB_NIC_STAT 0x00000200
+#define	FRF_BB_AER_DIS_LBN 34
+#define	FRF_BB_AER_DIS_WIDTH 1
+#define	FRF_BB_EE_STRAP_EN_LBN 31
+#define	FRF_BB_EE_STRAP_EN_WIDTH 1
+#define	FRF_BB_EE_STRAP_LBN 24
+#define	FRF_BB_EE_STRAP_WIDTH 4
+#define	FRF_BB_REVISION_ID_LBN 17
+#define	FRF_BB_REVISION_ID_WIDTH 7
+#define	FRF_AB_ONCHIP_SRAM_LBN 16
+#define	FRF_AB_ONCHIP_SRAM_WIDTH 1
+#define	FRF_AB_SF_PRST_LBN 9
+#define	FRF_AB_SF_PRST_WIDTH 1
+#define	FRF_AB_EE_PRST_LBN 8
+#define	FRF_AB_EE_PRST_WIDTH 1
+#define	FRF_AB_ATE_MODE_LBN 3
+#define	FRF_AB_ATE_MODE_WIDTH 1
+#define	FRF_AB_STRAP_PINS_LBN 0
+#define	FRF_AB_STRAP_PINS_WIDTH 3
+
+/* GPIO_CTL_REG: GPIO control register */
+#define	FR_AB_GPIO_CTL 0x00000210
+#define	FRF_AB_GPIO_OUT3_LBN 112
+#define	FRF_AB_GPIO_OUT3_WIDTH 16
+#define	FRF_AB_GPIO_IN3_LBN 104
+#define	FRF_AB_GPIO_IN3_WIDTH 8
+#define	FRF_AB_GPIO_PWRUP_VALUE3_LBN 96
+#define	FRF_AB_GPIO_PWRUP_VALUE3_WIDTH 8
+#define	FRF_AB_GPIO_OUT2_LBN 80
+#define	FRF_AB_GPIO_OUT2_WIDTH 16
+#define	FRF_AB_GPIO_IN2_LBN 72
+#define	FRF_AB_GPIO_IN2_WIDTH 8
+#define	FRF_AB_GPIO_PWRUP_VALUE2_LBN 64
+#define	FRF_AB_GPIO_PWRUP_VALUE2_WIDTH 8
+#define	FRF_AB_GPIO15_OEN_LBN 63
+#define	FRF_AB_GPIO15_OEN_WIDTH 1
+#define	FRF_AB_GPIO14_OEN_LBN 62
+#define	FRF_AB_GPIO14_OEN_WIDTH 1
+#define	FRF_AB_GPIO13_OEN_LBN 61
+#define	FRF_AB_GPIO13_OEN_WIDTH 1
+#define	FRF_AB_GPIO12_OEN_LBN 60
+#define	FRF_AB_GPIO12_OEN_WIDTH 1
+#define	FRF_AB_GPIO11_OEN_LBN 59
+#define	FRF_AB_GPIO11_OEN_WIDTH 1
+#define	FRF_AB_GPIO10_OEN_LBN 58
+#define	FRF_AB_GPIO10_OEN_WIDTH 1
+#define	FRF_AB_GPIO9_OEN_LBN 57
+#define	FRF_AB_GPIO9_OEN_WIDTH 1
+#define	FRF_AB_GPIO8_OEN_LBN 56
+#define	FRF_AB_GPIO8_OEN_WIDTH 1
+#define	FRF_AB_GPIO15_OUT_LBN 55
+#define	FRF_AB_GPIO15_OUT_WIDTH 1
+#define	FRF_AB_GPIO14_OUT_LBN 54
+#define	FRF_AB_GPIO14_OUT_WIDTH 1
+#define	FRF_AB_GPIO13_OUT_LBN 53
+#define	FRF_AB_GPIO13_OUT_WIDTH 1
+#define	FRF_AB_GPIO12_OUT_LBN 52
+#define	FRF_AB_GPIO12_OUT_WIDTH 1
+#define	FRF_AB_GPIO11_OUT_LBN 51
+#define	FRF_AB_GPIO11_OUT_WIDTH 1
+#define	FRF_AB_GPIO10_OUT_LBN 50
+#define	FRF_AB_GPIO10_OUT_WIDTH 1
+#define	FRF_AB_GPIO9_OUT_LBN 49
+#define	FRF_AB_GPIO9_OUT_WIDTH 1
+#define	FRF_AB_GPIO8_OUT_LBN 48
+#define	FRF_AB_GPIO8_OUT_WIDTH 1
+#define	FRF_AB_GPIO15_IN_LBN 47
+#define	FRF_AB_GPIO15_IN_WIDTH 1
+#define	FRF_AB_GPIO14_IN_LBN 46
+#define	FRF_AB_GPIO14_IN_WIDTH 1
+#define	FRF_AB_GPIO13_IN_LBN 45
+#define	FRF_AB_GPIO13_IN_WIDTH 1
+#define	FRF_AB_GPIO12_IN_LBN 44
+#define	FRF_AB_GPIO12_IN_WIDTH 1
+#define	FRF_AB_GPIO11_IN_LBN 43
+#define	FRF_AB_GPIO11_IN_WIDTH 1
+#define	FRF_AB_GPIO10_IN_LBN 42
+#define	FRF_AB_GPIO10_IN_WIDTH 1
+#define	FRF_AB_GPIO9_IN_LBN 41
+#define	FRF_AB_GPIO9_IN_WIDTH 1
+#define	FRF_AB_GPIO8_IN_LBN 40
+#define	FRF_AB_GPIO8_IN_WIDTH 1
+#define	FRF_AB_GPIO15_PWRUP_VALUE_LBN 39
+#define	FRF_AB_GPIO15_PWRUP_VALUE_WIDTH 1
+#define	FRF_AB_GPIO14_PWRUP_VALUE_LBN 38
+#define	FRF_AB_GPIO14_PWRUP_VALUE_WIDTH 1
+#define	FRF_AB_GPIO13_PWRUP_VALUE_LBN 37
+#define	FRF_AB_GPIO13_PWRUP_VALUE_WIDTH 1
+#define	FRF_AB_GPIO12_PWRUP_VALUE_LBN 36
+#define	FRF_AB_GPIO12_PWRUP_VALUE_WIDTH 1
+#define	FRF_AB_GPIO11_PWRUP_VALUE_LBN 35
+#define	FRF_AB_GPIO11_PWRUP_VALUE_WIDTH 1
+#define	FRF_AB_GPIO10_PWRUP_VALUE_LBN 34
+#define	FRF_AB_GPIO10_PWRUP_VALUE_WIDTH 1
+#define	FRF_AB_GPIO9_PWRUP_VALUE_LBN 33
+#define	FRF_AB_GPIO9_PWRUP_VALUE_WIDTH 1
+#define	FRF_AB_GPIO8_PWRUP_VALUE_LBN 32
+#define	FRF_AB_GPIO8_PWRUP_VALUE_WIDTH 1
+#define	FRF_AB_CLK156_OUT_EN_LBN 31
+#define	FRF_AB_CLK156_OUT_EN_WIDTH 1
+#define	FRF_AB_USE_NIC_CLK_LBN 30
+#define	FRF_AB_USE_NIC_CLK_WIDTH 1
+#define	FRF_AB_GPIO5_OEN_LBN 29
+#define	FRF_AB_GPIO5_OEN_WIDTH 1
+#define	FRF_AB_GPIO4_OEN_LBN 28
+#define	FRF_AB_GPIO4_OEN_WIDTH 1
+#define	FRF_AB_GPIO3_OEN_LBN 27
+#define	FRF_AB_GPIO3_OEN_WIDTH 1
+#define	FRF_AB_GPIO2_OEN_LBN 26
+#define	FRF_AB_GPIO2_OEN_WIDTH 1
+#define	FRF_AB_GPIO1_OEN_LBN 25
+#define	FRF_AB_GPIO1_OEN_WIDTH 1
+#define	FRF_AB_GPIO0_OEN_LBN 24
+#define	FRF_AB_GPIO0_OEN_WIDTH 1
+#define	FRF_AB_GPIO7_OUT_LBN 23
+#define	FRF_AB_GPIO7_OUT_WIDTH 1
+#define	FRF_AB_GPIO6_OUT_LBN 22
+#define	FRF_AB_GPIO6_OUT_WIDTH 1
+#define	FRF_AB_GPIO5_OUT_LBN 21
+#define	FRF_AB_GPIO5_OUT_WIDTH 1
+#define	FRF_AB_GPIO4_OUT_LBN 20
+#define	FRF_AB_GPIO4_OUT_WIDTH 1
+#define	FRF_AB_GPIO3_OUT_LBN 19
+#define	FRF_AB_GPIO3_OUT_WIDTH 1
+#define	FRF_AB_GPIO2_OUT_LBN 18
+#define	FRF_AB_GPIO2_OUT_WIDTH 1
+#define	FRF_AB_GPIO1_OUT_LBN 17
+#define	FRF_AB_GPIO1_OUT_WIDTH 1
+#define	FRF_AB_GPIO0_OUT_LBN 16
+#define	FRF_AB_GPIO0_OUT_WIDTH 1
+#define	FRF_AB_GPIO7_IN_LBN 15
+#define	FRF_AB_GPIO7_IN_WIDTH 1
+#define	FRF_AB_GPIO6_IN_LBN 14
+#define	FRF_AB_GPIO6_IN_WIDTH 1
+#define	FRF_AB_GPIO5_IN_LBN 13
+#define	FRF_AB_GPIO5_IN_WIDTH 1
+#define	FRF_AB_GPIO4_IN_LBN 12
+#define	FRF_AB_GPIO4_IN_WIDTH 1
+#define	FRF_AB_GPIO3_IN_LBN 11
+#define	FRF_AB_GPIO3_IN_WIDTH 1
+#define	FRF_AB_GPIO2_IN_LBN 10
+#define	FRF_AB_GPIO2_IN_WIDTH 1
+#define	FRF_AB_GPIO1_IN_LBN 9
+#define	FRF_AB_GPIO1_IN_WIDTH 1
+#define	FRF_AB_GPIO0_IN_LBN 8
+#define	FRF_AB_GPIO0_IN_WIDTH 1
+#define	FRF_AB_GPIO7_PWRUP_VALUE_LBN 7
+#define	FRF_AB_GPIO7_PWRUP_VALUE_WIDTH 1
+#define	FRF_AB_GPIO6_PWRUP_VALUE_LBN 6
+#define	FRF_AB_GPIO6_PWRUP_VALUE_WIDTH 1
+#define	FRF_AB_GPIO5_PWRUP_VALUE_LBN 5
+#define	FRF_AB_GPIO5_PWRUP_VALUE_WIDTH 1
+#define	FRF_AB_GPIO4_PWRUP_VALUE_LBN 4
+#define	FRF_AB_GPIO4_PWRUP_VALUE_WIDTH 1
+#define	FRF_AB_GPIO3_PWRUP_VALUE_LBN 3
+#define	FRF_AB_GPIO3_PWRUP_VALUE_WIDTH 1
+#define	FRF_AB_GPIO2_PWRUP_VALUE_LBN 2
+#define	FRF_AB_GPIO2_PWRUP_VALUE_WIDTH 1
+#define	FRF_AB_GPIO1_PWRUP_VALUE_LBN 1
+#define	FRF_AB_GPIO1_PWRUP_VALUE_WIDTH 1
+#define	FRF_AB_GPIO0_PWRUP_VALUE_LBN 0
+#define	FRF_AB_GPIO0_PWRUP_VALUE_WIDTH 1
+
+/* GLB_CTL_REG: Global control register */
+#define	FR_AB_GLB_CTL 0x00000220
+#define	FRF_AB_EXT_PHY_RST_CTL_LBN 63
+#define	FRF_AB_EXT_PHY_RST_CTL_WIDTH 1
+#define	FRF_AB_XAUI_SD_RST_CTL_LBN 62
+#define	FRF_AB_XAUI_SD_RST_CTL_WIDTH 1
+#define	FRF_AB_PCIE_SD_RST_CTL_LBN 61
+#define	FRF_AB_PCIE_SD_RST_CTL_WIDTH 1
+#define	FRF_AA_PCIX_RST_CTL_LBN 60
+#define	FRF_AA_PCIX_RST_CTL_WIDTH 1
+#define	FRF_BB_BIU_RST_CTL_LBN 60
+#define	FRF_BB_BIU_RST_CTL_WIDTH 1
+#define	FRF_AB_PCIE_STKY_RST_CTL_LBN 59
+#define	FRF_AB_PCIE_STKY_RST_CTL_WIDTH 1
+#define	FRF_AB_PCIE_NSTKY_RST_CTL_LBN 58
+#define	FRF_AB_PCIE_NSTKY_RST_CTL_WIDTH 1
+#define	FRF_AB_PCIE_CORE_RST_CTL_LBN 57
+#define	FRF_AB_PCIE_CORE_RST_CTL_WIDTH 1
+#define	FRF_AB_XGRX_RST_CTL_LBN 56
+#define	FRF_AB_XGRX_RST_CTL_WIDTH 1
+#define	FRF_AB_XGTX_RST_CTL_LBN 55
+#define	FRF_AB_XGTX_RST_CTL_WIDTH 1
+#define	FRF_AB_EM_RST_CTL_LBN 54
+#define	FRF_AB_EM_RST_CTL_WIDTH 1
+#define	FRF_AB_EV_RST_CTL_LBN 53
+#define	FRF_AB_EV_RST_CTL_WIDTH 1
+#define	FRF_AB_SR_RST_CTL_LBN 52
+#define	FRF_AB_SR_RST_CTL_WIDTH 1
+#define	FRF_AB_RX_RST_CTL_LBN 51
+#define	FRF_AB_RX_RST_CTL_WIDTH 1
+#define	FRF_AB_TX_RST_CTL_LBN 50
+#define	FRF_AB_TX_RST_CTL_WIDTH 1
+#define	FRF_AB_EE_RST_CTL_LBN 49
+#define	FRF_AB_EE_RST_CTL_WIDTH 1
+#define	FRF_AB_CS_RST_CTL_LBN 48
+#define	FRF_AB_CS_RST_CTL_WIDTH 1
+#define	FRF_AB_HOT_RST_CTL_LBN 40
+#define	FRF_AB_HOT_RST_CTL_WIDTH 2
+#define	FRF_AB_RST_EXT_PHY_LBN 31
+#define	FRF_AB_RST_EXT_PHY_WIDTH 1
+#define	FRF_AB_RST_XAUI_SD_LBN 30
+#define	FRF_AB_RST_XAUI_SD_WIDTH 1
+#define	FRF_AB_RST_PCIE_SD_LBN 29
+#define	FRF_AB_RST_PCIE_SD_WIDTH 1
+#define	FRF_AA_RST_PCIX_LBN 28
+#define	FRF_AA_RST_PCIX_WIDTH 1
+#define	FRF_BB_RST_BIU_LBN 28
+#define	FRF_BB_RST_BIU_WIDTH 1
+#define	FRF_AB_RST_PCIE_STKY_LBN 27
+#define	FRF_AB_RST_PCIE_STKY_WIDTH 1
+#define	FRF_AB_RST_PCIE_NSTKY_LBN 26
+#define	FRF_AB_RST_PCIE_NSTKY_WIDTH 1
+#define	FRF_AB_RST_PCIE_CORE_LBN 25
+#define	FRF_AB_RST_PCIE_CORE_WIDTH 1
+#define	FRF_AB_RST_XGRX_LBN 24
+#define	FRF_AB_RST_XGRX_WIDTH 1
+#define	FRF_AB_RST_XGTX_LBN 23
+#define	FRF_AB_RST_XGTX_WIDTH 1
+#define	FRF_AB_RST_EM_LBN 22
+#define	FRF_AB_RST_EM_WIDTH 1
+#define	FRF_AB_RST_EV_LBN 21
+#define	FRF_AB_RST_EV_WIDTH 1
+#define	FRF_AB_RST_SR_LBN 20
+#define	FRF_AB_RST_SR_WIDTH 1
+#define	FRF_AB_RST_RX_LBN 19
+#define	FRF_AB_RST_RX_WIDTH 1
+#define	FRF_AB_RST_TX_LBN 18
+#define	FRF_AB_RST_TX_WIDTH 1
+#define	FRF_AB_RST_SF_LBN 17
+#define	FRF_AB_RST_SF_WIDTH 1
+#define	FRF_AB_RST_CS_LBN 16
+#define	FRF_AB_RST_CS_WIDTH 1
+#define	FRF_AB_INT_RST_DUR_LBN 4
+#define	FRF_AB_INT_RST_DUR_WIDTH 3
+#define	FRF_AB_EXT_PHY_RST_DUR_LBN 1
+#define	FRF_AB_EXT_PHY_RST_DUR_WIDTH 3
+#define	FFE_AB_EXT_PHY_RST_DUR_10240US 7
+#define	FFE_AB_EXT_PHY_RST_DUR_5120US 6
+#define	FFE_AB_EXT_PHY_RST_DUR_2560US 5
+#define	FFE_AB_EXT_PHY_RST_DUR_1280US 4
+#define	FFE_AB_EXT_PHY_RST_DUR_640US 3
+#define	FFE_AB_EXT_PHY_RST_DUR_320US 2
+#define	FFE_AB_EXT_PHY_RST_DUR_160US 1
+#define	FFE_AB_EXT_PHY_RST_DUR_80US 0
+#define	FRF_AB_SWRST_LBN 0
+#define	FRF_AB_SWRST_WIDTH 1
+
+/* FATAL_INTR_REG_KER: Fatal interrupt register for Kernel */
+#define	FR_AZ_FATAL_INTR_KER 0x00000230
+#define	FRF_CZ_SRAM_PERR_INT_P_KER_EN_LBN 44
+#define	FRF_CZ_SRAM_PERR_INT_P_KER_EN_WIDTH 1
+#define	FRF_AB_PCI_BUSERR_INT_KER_EN_LBN 43
+#define	FRF_AB_PCI_BUSERR_INT_KER_EN_WIDTH 1
+#define	FRF_CZ_MBU_PERR_INT_KER_EN_LBN 43
+#define	FRF_CZ_MBU_PERR_INT_KER_EN_WIDTH 1
+#define	FRF_AZ_SRAM_OOB_INT_KER_EN_LBN 42
+#define	FRF_AZ_SRAM_OOB_INT_KER_EN_WIDTH 1
+#define	FRF_AZ_BUFID_OOB_INT_KER_EN_LBN 41
+#define	FRF_AZ_BUFID_OOB_INT_KER_EN_WIDTH 1
+#define	FRF_AZ_MEM_PERR_INT_KER_EN_LBN 40
+#define	FRF_AZ_MEM_PERR_INT_KER_EN_WIDTH 1
+#define	FRF_AZ_RBUF_OWN_INT_KER_EN_LBN 39
+#define	FRF_AZ_RBUF_OWN_INT_KER_EN_WIDTH 1
+#define	FRF_AZ_TBUF_OWN_INT_KER_EN_LBN 38
+#define	FRF_AZ_TBUF_OWN_INT_KER_EN_WIDTH 1
+#define	FRF_AZ_RDESCQ_OWN_INT_KER_EN_LBN 37
+#define	FRF_AZ_RDESCQ_OWN_INT_KER_EN_WIDTH 1
+#define	FRF_AZ_TDESCQ_OWN_INT_KER_EN_LBN 36
+#define	FRF_AZ_TDESCQ_OWN_INT_KER_EN_WIDTH 1
+#define	FRF_AZ_EVQ_OWN_INT_KER_EN_LBN 35
+#define	FRF_AZ_EVQ_OWN_INT_KER_EN_WIDTH 1
+#define	FRF_AZ_EVF_OFLO_INT_KER_EN_LBN 34
+#define	FRF_AZ_EVF_OFLO_INT_KER_EN_WIDTH 1
+#define	FRF_AZ_ILL_ADR_INT_KER_EN_LBN 33
+#define	FRF_AZ_ILL_ADR_INT_KER_EN_WIDTH 1
+#define	FRF_AZ_SRM_PERR_INT_KER_EN_LBN 32
+#define	FRF_AZ_SRM_PERR_INT_KER_EN_WIDTH 1
+#define	FRF_CZ_SRAM_PERR_INT_P_KER_LBN 12
+#define	FRF_CZ_SRAM_PERR_INT_P_KER_WIDTH 1
+#define	FRF_AB_PCI_BUSERR_INT_KER_LBN 11
+#define	FRF_AB_PCI_BUSERR_INT_KER_WIDTH 1
+#define	FRF_CZ_MBU_PERR_INT_KER_LBN 11
+#define	FRF_CZ_MBU_PERR_INT_KER_WIDTH 1
+#define	FRF_AZ_SRAM_OOB_INT_KER_LBN 10
+#define	FRF_AZ_SRAM_OOB_INT_KER_WIDTH 1
+#define	FRF_AZ_BUFID_DC_OOB_INT_KER_LBN 9
+#define	FRF_AZ_BUFID_DC_OOB_INT_KER_WIDTH 1
+#define	FRF_AZ_MEM_PERR_INT_KER_LBN 8
+#define	FRF_AZ_MEM_PERR_INT_KER_WIDTH 1
+#define	FRF_AZ_RBUF_OWN_INT_KER_LBN 7
+#define	FRF_AZ_RBUF_OWN_INT_KER_WIDTH 1
+#define	FRF_AZ_TBUF_OWN_INT_KER_LBN 6
+#define	FRF_AZ_TBUF_OWN_INT_KER_WIDTH 1
+#define	FRF_AZ_RDESCQ_OWN_INT_KER_LBN 5
+#define	FRF_AZ_RDESCQ_OWN_INT_KER_WIDTH 1
+#define	FRF_AZ_TDESCQ_OWN_INT_KER_LBN 4
+#define	FRF_AZ_TDESCQ_OWN_INT_KER_WIDTH 1
+#define	FRF_AZ_EVQ_OWN_INT_KER_LBN 3
+#define	FRF_AZ_EVQ_OWN_INT_KER_WIDTH 1
+#define	FRF_AZ_EVF_OFLO_INT_KER_LBN 2
+#define	FRF_AZ_EVF_OFLO_INT_KER_WIDTH 1
+#define	FRF_AZ_ILL_ADR_INT_KER_LBN 1
+#define	FRF_AZ_ILL_ADR_INT_KER_WIDTH 1
+#define	FRF_AZ_SRM_PERR_INT_KER_LBN 0
+#define	FRF_AZ_SRM_PERR_INT_KER_WIDTH 1
+
+/* FATAL_INTR_REG_CHAR: Fatal interrupt register for Char */
+#define	FR_BZ_FATAL_INTR_CHAR 0x00000240
+#define	FRF_CZ_SRAM_PERR_INT_P_CHAR_EN_LBN 44
+#define	FRF_CZ_SRAM_PERR_INT_P_CHAR_EN_WIDTH 1
+#define	FRF_BB_PCI_BUSERR_INT_CHAR_EN_LBN 43
+#define	FRF_BB_PCI_BUSERR_INT_CHAR_EN_WIDTH 1
+#define	FRF_CZ_MBU_PERR_INT_CHAR_EN_LBN 43
+#define	FRF_CZ_MBU_PERR_INT_CHAR_EN_WIDTH 1
+#define	FRF_BZ_SRAM_OOB_INT_CHAR_EN_LBN 42
+#define	FRF_BZ_SRAM_OOB_INT_CHAR_EN_WIDTH 1
+#define	FRF_BZ_BUFID_OOB_INT_CHAR_EN_LBN 41
+#define	FRF_BZ_BUFID_OOB_INT_CHAR_EN_WIDTH 1
+#define	FRF_BZ_MEM_PERR_INT_CHAR_EN_LBN 40
+#define	FRF_BZ_MEM_PERR_INT_CHAR_EN_WIDTH 1
+#define	FRF_BZ_RBUF_OWN_INT_CHAR_EN_LBN 39
+#define	FRF_BZ_RBUF_OWN_INT_CHAR_EN_WIDTH 1
+#define	FRF_BZ_TBUF_OWN_INT_CHAR_EN_LBN 38
+#define	FRF_BZ_TBUF_OWN_INT_CHAR_EN_WIDTH 1
+#define	FRF_BZ_RDESCQ_OWN_INT_CHAR_EN_LBN 37
+#define	FRF_BZ_RDESCQ_OWN_INT_CHAR_EN_WIDTH 1
+#define	FRF_BZ_TDESCQ_OWN_INT_CHAR_EN_LBN 36
+#define	FRF_BZ_TDESCQ_OWN_INT_CHAR_EN_WIDTH 1
+#define	FRF_BZ_EVQ_OWN_INT_CHAR_EN_LBN 35
+#define	FRF_BZ_EVQ_OWN_INT_CHAR_EN_WIDTH 1
+#define	FRF_BZ_EVF_OFLO_INT_CHAR_EN_LBN 34
+#define	FRF_BZ_EVF_OFLO_INT_CHAR_EN_WIDTH 1
+#define	FRF_BZ_ILL_ADR_INT_CHAR_EN_LBN 33
+#define	FRF_BZ_ILL_ADR_INT_CHAR_EN_WIDTH 1
+#define	FRF_BZ_SRM_PERR_INT_CHAR_EN_LBN 32
+#define	FRF_BZ_SRM_PERR_INT_CHAR_EN_WIDTH 1
+#define	FRF_CZ_SRAM_PERR_INT_P_CHAR_LBN 12
+#define	FRF_CZ_SRAM_PERR_INT_P_CHAR_WIDTH 1
+#define	FRF_BB_PCI_BUSERR_INT_CHAR_LBN 11
+#define	FRF_BB_PCI_BUSERR_INT_CHAR_WIDTH 1
+#define	FRF_CZ_MBU_PERR_INT_CHAR_LBN 11
+#define	FRF_CZ_MBU_PERR_INT_CHAR_WIDTH 1
+#define	FRF_BZ_SRAM_OOB_INT_CHAR_LBN 10
+#define	FRF_BZ_SRAM_OOB_INT_CHAR_WIDTH 1
+#define	FRF_BZ_BUFID_DC_OOB_INT_CHAR_LBN 9
+#define	FRF_BZ_BUFID_DC_OOB_INT_CHAR_WIDTH 1
+#define	FRF_BZ_MEM_PERR_INT_CHAR_LBN 8
+#define	FRF_BZ_MEM_PERR_INT_CHAR_WIDTH 1
+#define	FRF_BZ_RBUF_OWN_INT_CHAR_LBN 7
+#define	FRF_BZ_RBUF_OWN_INT_CHAR_WIDTH 1
+#define	FRF_BZ_TBUF_OWN_INT_CHAR_LBN 6
+#define	FRF_BZ_TBUF_OWN_INT_CHAR_WIDTH 1
+#define	FRF_BZ_RDESCQ_OWN_INT_CHAR_LBN 5
+#define	FRF_BZ_RDESCQ_OWN_INT_CHAR_WIDTH 1
+#define	FRF_BZ_TDESCQ_OWN_INT_CHAR_LBN 4
+#define	FRF_BZ_TDESCQ_OWN_INT_CHAR_WIDTH 1
+#define	FRF_BZ_EVQ_OWN_INT_CHAR_LBN 3
+#define	FRF_BZ_EVQ_OWN_INT_CHAR_WIDTH 1
+#define	FRF_BZ_EVF_OFLO_INT_CHAR_LBN 2
+#define	FRF_BZ_EVF_OFLO_INT_CHAR_WIDTH 1
+#define	FRF_BZ_ILL_ADR_INT_CHAR_LBN 1
+#define	FRF_BZ_ILL_ADR_INT_CHAR_WIDTH 1
+#define	FRF_BZ_SRM_PERR_INT_CHAR_LBN 0
+#define	FRF_BZ_SRM_PERR_INT_CHAR_WIDTH 1
+
+/* DP_CTRL_REG: Datapath control register */
+#define	FR_BZ_DP_CTRL 0x00000250
+#define	FRF_BZ_FLS_EVQ_ID_LBN 0
+#define	FRF_BZ_FLS_EVQ_ID_WIDTH 12
+
+/* MEM_STAT_REG: Memory status register */
+#define	FR_AZ_MEM_STAT 0x00000260
+#define	FRF_AB_MEM_PERR_VEC_LBN 53
+#define	FRF_AB_MEM_PERR_VEC_WIDTH 38
+#define	FRF_AB_MBIST_CORR_LBN 38
+#define	FRF_AB_MBIST_CORR_WIDTH 15
+#define	FRF_AB_MBIST_ERR_LBN 0
+#define	FRF_AB_MBIST_ERR_WIDTH 40
+#define	FRF_CZ_MEM_PERR_VEC_LBN 0
+#define	FRF_CZ_MEM_PERR_VEC_WIDTH 35
+
+/* CS_DEBUG_REG: Debug register */
+#define	FR_AZ_CS_DEBUG 0x00000270
+#define	FRF_AB_GLB_DEBUG2_SEL_LBN 50
+#define	FRF_AB_GLB_DEBUG2_SEL_WIDTH 3
+#define	FRF_AB_DEBUG_BLK_SEL2_LBN 47
+#define	FRF_AB_DEBUG_BLK_SEL2_WIDTH 3
+#define	FRF_AB_DEBUG_BLK_SEL1_LBN 44
+#define	FRF_AB_DEBUG_BLK_SEL1_WIDTH 3
+#define	FRF_AB_DEBUG_BLK_SEL0_LBN 41
+#define	FRF_AB_DEBUG_BLK_SEL0_WIDTH 3
+#define	FRF_CZ_CS_PORT_NUM_LBN 40
+#define	FRF_CZ_CS_PORT_NUM_WIDTH 2
+#define	FRF_AB_MISC_DEBUG_ADDR_LBN 36
+#define	FRF_AB_MISC_DEBUG_ADDR_WIDTH 5
+#define	FRF_AB_SERDES_DEBUG_ADDR_LBN 31
+#define	FRF_AB_SERDES_DEBUG_ADDR_WIDTH 5
+#define	FRF_CZ_CS_PORT_FPE_LBN 1
+#define	FRF_CZ_CS_PORT_FPE_WIDTH 35
+#define	FRF_AB_EM_DEBUG_ADDR_LBN 26
+#define	FRF_AB_EM_DEBUG_ADDR_WIDTH 5
+#define	FRF_AB_SR_DEBUG_ADDR_LBN 21
+#define	FRF_AB_SR_DEBUG_ADDR_WIDTH 5
+#define	FRF_AB_EV_DEBUG_ADDR_LBN 16
+#define	FRF_AB_EV_DEBUG_ADDR_WIDTH 5
+#define	FRF_AB_RX_DEBUG_ADDR_LBN 11
+#define	FRF_AB_RX_DEBUG_ADDR_WIDTH 5
+#define	FRF_AB_TX_DEBUG_ADDR_LBN 6
+#define	FRF_AB_TX_DEBUG_ADDR_WIDTH 5
+#define	FRF_AB_CS_BIU_DEBUG_ADDR_LBN 1
+#define	FRF_AB_CS_BIU_DEBUG_ADDR_WIDTH 5
+#define	FRF_AZ_CS_DEBUG_EN_LBN 0
+#define	FRF_AZ_CS_DEBUG_EN_WIDTH 1
+
+/* DRIVER_REG: Driver scratch register [0-7] */
+#define	FR_AZ_DRIVER 0x00000280
+#define	FR_AZ_DRIVER_STEP 16
+#define	FR_AZ_DRIVER_ROWS 8
+#define	FRF_AZ_DRIVER_DW0_LBN 0
+#define	FRF_AZ_DRIVER_DW0_WIDTH 32
+
+/* ALTERA_BUILD_REG: Altera build register */
+#define	FR_AZ_ALTERA_BUILD 0x00000300
+#define	FRF_AZ_ALTERA_BUILD_VER_LBN 0
+#define	FRF_AZ_ALTERA_BUILD_VER_WIDTH 32
+
+/* CSR_SPARE_REG: Spare register */
+#define	FR_AZ_CSR_SPARE 0x00000310
+#define	FRF_AB_MEM_PERR_EN_LBN 64
+#define	FRF_AB_MEM_PERR_EN_WIDTH 38
+#define	FRF_CZ_MEM_PERR_EN_LBN 64
+#define	FRF_CZ_MEM_PERR_EN_WIDTH 35
+#define	FRF_AB_MEM_PERR_EN_TX_DATA_LBN 72
+#define	FRF_AB_MEM_PERR_EN_TX_DATA_WIDTH 2
+#define	FRF_AZ_CSR_SPARE_BITS_LBN 0
+#define	FRF_AZ_CSR_SPARE_BITS_WIDTH 32
+
+/* PCIE_SD_CTL0123_REG: PCIE SerDes control register 0 to 3 */
+#define	FR_AB_PCIE_SD_CTL0123 0x00000320
+#define	FRF_AB_PCIE_TESTSIG_H_LBN 96
+#define	FRF_AB_PCIE_TESTSIG_H_WIDTH 19
+#define	FRF_AB_PCIE_TESTSIG_L_LBN 64
+#define	FRF_AB_PCIE_TESTSIG_L_WIDTH 19
+#define	FRF_AB_PCIE_OFFSET_LBN 56
+#define	FRF_AB_PCIE_OFFSET_WIDTH 8
+#define	FRF_AB_PCIE_OFFSETEN_H_LBN 55
+#define	FRF_AB_PCIE_OFFSETEN_H_WIDTH 1
+#define	FRF_AB_PCIE_OFFSETEN_L_LBN 54
+#define	FRF_AB_PCIE_OFFSETEN_L_WIDTH 1
+#define	FRF_AB_PCIE_HIVMODE_H_LBN 53
+#define	FRF_AB_PCIE_HIVMODE_H_WIDTH 1
+#define	FRF_AB_PCIE_HIVMODE_L_LBN 52
+#define	FRF_AB_PCIE_HIVMODE_L_WIDTH 1
+#define	FRF_AB_PCIE_PARRESET_H_LBN 51
+#define	FRF_AB_PCIE_PARRESET_H_WIDTH 1
+#define	FRF_AB_PCIE_PARRESET_L_LBN 50
+#define	FRF_AB_PCIE_PARRESET_L_WIDTH 1
+#define	FRF_AB_PCIE_LPBKWDRV_H_LBN 49
+#define	FRF_AB_PCIE_LPBKWDRV_H_WIDTH 1
+#define	FRF_AB_PCIE_LPBKWDRV_L_LBN 48
+#define	FRF_AB_PCIE_LPBKWDRV_L_WIDTH 1
+#define	FRF_AB_PCIE_LPBK_LBN 40
+#define	FRF_AB_PCIE_LPBK_WIDTH 8
+#define	FRF_AB_PCIE_PARLPBK_LBN 32
+#define	FRF_AB_PCIE_PARLPBK_WIDTH 8
+#define	FRF_AB_PCIE_RXTERMADJ_H_LBN 30
+#define	FRF_AB_PCIE_RXTERMADJ_H_WIDTH 2
+#define	FRF_AB_PCIE_RXTERMADJ_L_LBN 28
+#define	FRF_AB_PCIE_RXTERMADJ_L_WIDTH 2
+#define	FFE_AB_PCIE_RXTERMADJ_MIN15PCNT 3
+#define	FFE_AB_PCIE_RXTERMADJ_PL10PCNT 2
+#define	FFE_AB_PCIE_RXTERMADJ_MIN17PCNT 1
+#define	FFE_AB_PCIE_RXTERMADJ_NOMNL 0
+#define	FRF_AB_PCIE_TXTERMADJ_H_LBN 26
+#define	FRF_AB_PCIE_TXTERMADJ_H_WIDTH 2
+#define	FRF_AB_PCIE_TXTERMADJ_L_LBN 24
+#define	FRF_AB_PCIE_TXTERMADJ_L_WIDTH 2
+#define	FFE_AB_PCIE_TXTERMADJ_MIN15PCNT 3
+#define	FFE_AB_PCIE_TXTERMADJ_PL10PCNT 2
+#define	FFE_AB_PCIE_TXTERMADJ_MIN17PCNT 1
+#define	FFE_AB_PCIE_TXTERMADJ_NOMNL 0
+#define	FRF_AB_PCIE_RXEQCTL_H_LBN 18
+#define	FRF_AB_PCIE_RXEQCTL_H_WIDTH 2
+#define	FRF_AB_PCIE_RXEQCTL_L_LBN 16
+#define	FRF_AB_PCIE_RXEQCTL_L_WIDTH 2
+#define	FFE_AB_PCIE_RXEQCTL_OFF_ALT 3
+#define	FFE_AB_PCIE_RXEQCTL_OFF 2
+#define	FFE_AB_PCIE_RXEQCTL_MIN 1
+#define	FFE_AB_PCIE_RXEQCTL_MAX 0
+#define	FRF_AB_PCIE_HIDRV_LBN 8
+#define	FRF_AB_PCIE_HIDRV_WIDTH 8
+#define	FRF_AB_PCIE_LODRV_LBN 0
+#define	FRF_AB_PCIE_LODRV_WIDTH 8
+
+/* PCIE_SD_CTL45_REG: PCIE SerDes control register 4 and 5 */
+#define	FR_AB_PCIE_SD_CTL45 0x00000330
+#define	FRF_AB_PCIE_DTX7_LBN 60
+#define	FRF_AB_PCIE_DTX7_WIDTH 4
+#define	FRF_AB_PCIE_DTX6_LBN 56
+#define	FRF_AB_PCIE_DTX6_WIDTH 4
+#define	FRF_AB_PCIE_DTX5_LBN 52
+#define	FRF_AB_PCIE_DTX5_WIDTH 4
+#define	FRF_AB_PCIE_DTX4_LBN 48
+#define	FRF_AB_PCIE_DTX4_WIDTH 4
+#define	FRF_AB_PCIE_DTX3_LBN 44
+#define	FRF_AB_PCIE_DTX3_WIDTH 4
+#define	FRF_AB_PCIE_DTX2_LBN 40
+#define	FRF_AB_PCIE_DTX2_WIDTH 4
+#define	FRF_AB_PCIE_DTX1_LBN 36
+#define	FRF_AB_PCIE_DTX1_WIDTH 4
+#define	FRF_AB_PCIE_DTX0_LBN 32
+#define	FRF_AB_PCIE_DTX0_WIDTH 4
+#define	FRF_AB_PCIE_DEQ7_LBN 28
+#define	FRF_AB_PCIE_DEQ7_WIDTH 4
+#define	FRF_AB_PCIE_DEQ6_LBN 24
+#define	FRF_AB_PCIE_DEQ6_WIDTH 4
+#define	FRF_AB_PCIE_DEQ5_LBN 20
+#define	FRF_AB_PCIE_DEQ5_WIDTH 4
+#define	FRF_AB_PCIE_DEQ4_LBN 16
+#define	FRF_AB_PCIE_DEQ4_WIDTH 4
+#define	FRF_AB_PCIE_DEQ3_LBN 12
+#define	FRF_AB_PCIE_DEQ3_WIDTH 4
+#define	FRF_AB_PCIE_DEQ2_LBN 8
+#define	FRF_AB_PCIE_DEQ2_WIDTH 4
+#define	FRF_AB_PCIE_DEQ1_LBN 4
+#define	FRF_AB_PCIE_DEQ1_WIDTH 4
+#define	FRF_AB_PCIE_DEQ0_LBN 0
+#define	FRF_AB_PCIE_DEQ0_WIDTH 4
+
+/* PCIE_PCS_CTL_STAT_REG: PCIE PCS control and status register */
+#define	FR_AB_PCIE_PCS_CTL_STAT 0x00000340
+#define	FRF_AB_PCIE_PRBSERRCOUNT0_H_LBN 52
+#define	FRF_AB_PCIE_PRBSERRCOUNT0_H_WIDTH 4
+#define	FRF_AB_PCIE_PRBSERRCOUNT0_L_LBN 48
+#define	FRF_AB_PCIE_PRBSERRCOUNT0_L_WIDTH 4
+#define	FRF_AB_PCIE_PRBSERR_LBN 40
+#define	FRF_AB_PCIE_PRBSERR_WIDTH 8
+#define	FRF_AB_PCIE_PRBSERRH0_LBN 32
+#define	FRF_AB_PCIE_PRBSERRH0_WIDTH 8
+#define	FRF_AB_PCIE_FASTINIT_H_LBN 15
+#define	FRF_AB_PCIE_FASTINIT_H_WIDTH 1
+#define	FRF_AB_PCIE_FASTINIT_L_LBN 14
+#define	FRF_AB_PCIE_FASTINIT_L_WIDTH 1
+#define	FRF_AB_PCIE_CTCDISABLE_H_LBN 13
+#define	FRF_AB_PCIE_CTCDISABLE_H_WIDTH 1
+#define	FRF_AB_PCIE_CTCDISABLE_L_LBN 12
+#define	FRF_AB_PCIE_CTCDISABLE_L_WIDTH 1
+#define	FRF_AB_PCIE_PRBSSYNC_H_LBN 11
+#define	FRF_AB_PCIE_PRBSSYNC_H_WIDTH 1
+#define	FRF_AB_PCIE_PRBSSYNC_L_LBN 10
+#define	FRF_AB_PCIE_PRBSSYNC_L_WIDTH 1
+#define	FRF_AB_PCIE_PRBSERRACK_H_LBN 9
+#define	FRF_AB_PCIE_PRBSERRACK_H_WIDTH 1
+#define	FRF_AB_PCIE_PRBSERRACK_L_LBN 8
+#define	FRF_AB_PCIE_PRBSERRACK_L_WIDTH 1
+#define	FRF_AB_PCIE_PRBSSEL_LBN 0
+#define	FRF_AB_PCIE_PRBSSEL_WIDTH 8
+
+/* DEBUG_DATA_OUT_REG: Live Debug and Debug 2 out ports */
+#define	FR_BB_DEBUG_DATA_OUT 0x00000350
+#define	FRF_BB_DEBUG2_PORT_LBN 25
+#define	FRF_BB_DEBUG2_PORT_WIDTH 15
+#define	FRF_BB_DEBUG1_PORT_LBN 0
+#define	FRF_BB_DEBUG1_PORT_WIDTH 25
+
+/* EVQ_RPTR_REGP0: Event queue read pointer register */
+#define	FR_BZ_EVQ_RPTR_P0 0x00000400
+#define	FR_BZ_EVQ_RPTR_P0_STEP 8192
+#define	FR_BZ_EVQ_RPTR_P0_ROWS 1024
+/* EVQ_RPTR_REG_KER: Event queue read pointer register */
+#define	FR_AA_EVQ_RPTR_KER 0x00011b00
+#define	FR_AA_EVQ_RPTR_KER_STEP 4
+#define	FR_AA_EVQ_RPTR_KER_ROWS 4
+/* EVQ_RPTR_REG: Event queue read pointer register */
+#define	FR_BZ_EVQ_RPTR 0x00fa0000
+#define	FR_BZ_EVQ_RPTR_STEP 16
+#define	FR_BB_EVQ_RPTR_ROWS 4096
+#define	FR_CZ_EVQ_RPTR_ROWS 1024
+/* EVQ_RPTR_REGP123: Event queue read pointer register */
+#define	FR_BB_EVQ_RPTR_P123 0x01000400
+#define	FR_BB_EVQ_RPTR_P123_STEP 8192
+#define	FR_BB_EVQ_RPTR_P123_ROWS 3072
+#define	FRF_AZ_EVQ_RPTR_VLD_LBN 15
+#define	FRF_AZ_EVQ_RPTR_VLD_WIDTH 1
+#define	FRF_AZ_EVQ_RPTR_LBN 0
+#define	FRF_AZ_EVQ_RPTR_WIDTH 15
+
+/* TIMER_COMMAND_REGP0: Timer Command Registers */
+#define	FR_BZ_TIMER_COMMAND_P0 0x00000420
+#define	FR_BZ_TIMER_COMMAND_P0_STEP 8192
+#define	FR_BZ_TIMER_COMMAND_P0_ROWS 1024
+/* TIMER_COMMAND_REG_KER: Timer Command Registers */
+#define	FR_AA_TIMER_COMMAND_KER 0x00000420
+#define	FR_AA_TIMER_COMMAND_KER_STEP 8192
+#define	FR_AA_TIMER_COMMAND_KER_ROWS 4
+/* TIMER_COMMAND_REGP123: Timer Command Registers */
+#define	FR_BB_TIMER_COMMAND_P123 0x01000420
+#define	FR_BB_TIMER_COMMAND_P123_STEP 8192
+#define	FR_BB_TIMER_COMMAND_P123_ROWS 3072
+#define	FRF_CZ_TC_TIMER_MODE_LBN 14
+#define	FRF_CZ_TC_TIMER_MODE_WIDTH 2
+#define	FRF_AB_TC_TIMER_MODE_LBN 12
+#define	FRF_AB_TC_TIMER_MODE_WIDTH 2
+#define	FRF_CZ_TC_TIMER_VAL_LBN 0
+#define	FRF_CZ_TC_TIMER_VAL_WIDTH 14
+#define	FRF_AB_TC_TIMER_VAL_LBN 0
+#define	FRF_AB_TC_TIMER_VAL_WIDTH 12
+
+/* DRV_EV_REG: Driver generated event register */
+#define	FR_AZ_DRV_EV 0x00000440
+#define	FRF_AZ_DRV_EV_QID_LBN 64
+#define	FRF_AZ_DRV_EV_QID_WIDTH 12
+#define	FRF_AZ_DRV_EV_DATA_LBN 0
+#define	FRF_AZ_DRV_EV_DATA_WIDTH 64
+
+/* EVQ_CTL_REG: Event queue control register */
+#define	FR_AZ_EVQ_CTL 0x00000450
+#define	FRF_CZ_RX_EVQ_WAKEUP_MASK_LBN 15
+#define	FRF_CZ_RX_EVQ_WAKEUP_MASK_WIDTH 10
+#define	FRF_BB_RX_EVQ_WAKEUP_MASK_LBN 15
+#define	FRF_BB_RX_EVQ_WAKEUP_MASK_WIDTH 6
+#define	FRF_AZ_EVQ_OWNERR_CTL_LBN 14
+#define	FRF_AZ_EVQ_OWNERR_CTL_WIDTH 1
+#define	FRF_AZ_EVQ_FIFO_AF_TH_LBN 7
+#define	FRF_AZ_EVQ_FIFO_AF_TH_WIDTH 7
+#define	FRF_AZ_EVQ_FIFO_NOTAF_TH_LBN 0
+#define	FRF_AZ_EVQ_FIFO_NOTAF_TH_WIDTH 7
+
+/* EVQ_CNT1_REG: Event counter 1 register */
+#define	FR_AZ_EVQ_CNT1 0x00000460
+#define	FRF_AZ_EVQ_CNT_PRE_FIFO_LBN 120
+#define	FRF_AZ_EVQ_CNT_PRE_FIFO_WIDTH 7
+#define	FRF_AZ_EVQ_CNT_TOBIU_LBN 100
+#define	FRF_AZ_EVQ_CNT_TOBIU_WIDTH 20
+#define	FRF_AZ_EVQ_TX_REQ_CNT_LBN 80
+#define	FRF_AZ_EVQ_TX_REQ_CNT_WIDTH 20
+#define	FRF_AZ_EVQ_RX_REQ_CNT_LBN 60
+#define	FRF_AZ_EVQ_RX_REQ_CNT_WIDTH 20
+#define	FRF_AZ_EVQ_EM_REQ_CNT_LBN 40
+#define	FRF_AZ_EVQ_EM_REQ_CNT_WIDTH 20
+#define	FRF_AZ_EVQ_CSR_REQ_CNT_LBN 20
+#define	FRF_AZ_EVQ_CSR_REQ_CNT_WIDTH 20
+#define	FRF_AZ_EVQ_ERR_REQ_CNT_LBN 0
+#define	FRF_AZ_EVQ_ERR_REQ_CNT_WIDTH 20
+
+/* EVQ_CNT2_REG: Event counter 2 register */
+#define	FR_AZ_EVQ_CNT2 0x00000470
+#define	FRF_AZ_EVQ_UPD_REQ_CNT_LBN 104
+#define	FRF_AZ_EVQ_UPD_REQ_CNT_WIDTH 20
+#define	FRF_AZ_EVQ_CLR_REQ_CNT_LBN 84
+#define	FRF_AZ_EVQ_CLR_REQ_CNT_WIDTH 20
+#define	FRF_AZ_EVQ_RDY_CNT_LBN 80
+#define	FRF_AZ_EVQ_RDY_CNT_WIDTH 4
+#define	FRF_AZ_EVQ_WU_REQ_CNT_LBN 60
+#define	FRF_AZ_EVQ_WU_REQ_CNT_WIDTH 20
+#define	FRF_AZ_EVQ_WET_REQ_CNT_LBN 40
+#define	FRF_AZ_EVQ_WET_REQ_CNT_WIDTH 20
+#define	FRF_AZ_EVQ_INIT_REQ_CNT_LBN 20
+#define	FRF_AZ_EVQ_INIT_REQ_CNT_WIDTH 20
+#define	FRF_AZ_EVQ_TM_REQ_CNT_LBN 0
+#define	FRF_AZ_EVQ_TM_REQ_CNT_WIDTH 20
+
+/* USR_EV_REG: Event mailbox register */
+#define	FR_CZ_USR_EV 0x00000540
+#define	FR_CZ_USR_EV_STEP 8192
+#define	FR_CZ_USR_EV_ROWS 1024
+#define	FRF_CZ_USR_EV_DATA_LBN 0
+#define	FRF_CZ_USR_EV_DATA_WIDTH 32
+
+/* BUF_TBL_CFG_REG: Buffer table configuration register */
+#define	FR_AZ_BUF_TBL_CFG 0x00000600
+#define	FRF_AZ_BUF_TBL_MODE_LBN 3
+#define	FRF_AZ_BUF_TBL_MODE_WIDTH 1
+
+/* SRM_RX_DC_CFG_REG: SRAM receive descriptor cache configuration register */
+#define	FR_AZ_SRM_RX_DC_CFG 0x00000610
+#define	FRF_AZ_SRM_CLK_TMP_EN_LBN 21
+#define	FRF_AZ_SRM_CLK_TMP_EN_WIDTH 1
+#define	FRF_AZ_SRM_RX_DC_BASE_ADR_LBN 0
+#define	FRF_AZ_SRM_RX_DC_BASE_ADR_WIDTH 21
+
+/* SRM_TX_DC_CFG_REG: SRAM transmit descriptor cache configuration register */
+#define	FR_AZ_SRM_TX_DC_CFG 0x00000620
+#define	FRF_AZ_SRM_TX_DC_BASE_ADR_LBN 0
+#define	FRF_AZ_SRM_TX_DC_BASE_ADR_WIDTH 21
+
+/* SRM_CFG_REG: SRAM configuration register */
+#define	FR_AZ_SRM_CFG 0x00000630
+#define	FRF_AZ_SRM_OOB_ADR_INTEN_LBN 5
+#define	FRF_AZ_SRM_OOB_ADR_INTEN_WIDTH 1
+#define	FRF_AZ_SRM_OOB_BUF_INTEN_LBN 4
+#define	FRF_AZ_SRM_OOB_BUF_INTEN_WIDTH 1
+#define	FRF_AZ_SRM_INIT_EN_LBN 3
+#define	FRF_AZ_SRM_INIT_EN_WIDTH 1
+#define	FRF_AZ_SRM_NUM_BANK_LBN 2
+#define	FRF_AZ_SRM_NUM_BANK_WIDTH 1
+#define	FRF_AZ_SRM_BANK_SIZE_LBN 0
+#define	FRF_AZ_SRM_BANK_SIZE_WIDTH 2
+
+/* BUF_TBL_UPD_REG: Buffer table update register */
+#define	FR_AZ_BUF_TBL_UPD 0x00000650
+#define	FRF_AZ_BUF_UPD_CMD_LBN 63
+#define	FRF_AZ_BUF_UPD_CMD_WIDTH 1
+#define	FRF_AZ_BUF_CLR_CMD_LBN 62
+#define	FRF_AZ_BUF_CLR_CMD_WIDTH 1
+#define	FRF_AZ_BUF_CLR_END_ID_LBN 32
+#define	FRF_AZ_BUF_CLR_END_ID_WIDTH 20
+#define	FRF_AZ_BUF_CLR_START_ID_LBN 0
+#define	FRF_AZ_BUF_CLR_START_ID_WIDTH 20
+
+/* SRM_UPD_EVQ_REG: Buffer table update register */
+#define	FR_AZ_SRM_UPD_EVQ 0x00000660
+#define	FRF_AZ_SRM_UPD_EVQ_ID_LBN 0
+#define	FRF_AZ_SRM_UPD_EVQ_ID_WIDTH 12
+
+/* SRAM_PARITY_REG: SRAM parity register. */
+#define	FR_AZ_SRAM_PARITY 0x00000670
+#define	FRF_CZ_BYPASS_ECC_LBN 3
+#define	FRF_CZ_BYPASS_ECC_WIDTH 1
+#define	FRF_CZ_SEC_INT_LBN 2
+#define	FRF_CZ_SEC_INT_WIDTH 1
+#define	FRF_CZ_FORCE_SRAM_DOUBLE_ERR_LBN 1
+#define	FRF_CZ_FORCE_SRAM_DOUBLE_ERR_WIDTH 1
+#define	FRF_AB_FORCE_SRAM_PERR_LBN 0
+#define	FRF_AB_FORCE_SRAM_PERR_WIDTH 1
+#define	FRF_CZ_FORCE_SRAM_SINGLE_ERR_LBN 0
+#define	FRF_CZ_FORCE_SRAM_SINGLE_ERR_WIDTH 1
+
+/* RX_CFG_REG: Receive configuration register */
+#define	FR_AZ_RX_CFG 0x00000800
+#define	FRF_CZ_RX_MIN_KBUF_SIZE_LBN 72
+#define	FRF_CZ_RX_MIN_KBUF_SIZE_WIDTH 14
+#define	FRF_CZ_RX_HDR_SPLIT_EN_LBN 71
+#define	FRF_CZ_RX_HDR_SPLIT_EN_WIDTH 1
+#define	FRF_CZ_RX_HDR_SPLIT_PLD_BUF_SIZE_LBN 62
+#define	FRF_CZ_RX_HDR_SPLIT_PLD_BUF_SIZE_WIDTH 9
+#define	FRF_CZ_RX_HDR_SPLIT_HDR_BUF_SIZE_LBN 53
+#define	FRF_CZ_RX_HDR_SPLIT_HDR_BUF_SIZE_WIDTH 9
+#define	FRF_CZ_RX_PRE_RFF_IPG_LBN 49
+#define	FRF_CZ_RX_PRE_RFF_IPG_WIDTH 4
+#define	FRF_BZ_RX_TCP_SUP_LBN 48
+#define	FRF_BZ_RX_TCP_SUP_WIDTH 1
+#define	FRF_BZ_RX_INGR_EN_LBN 47
+#define	FRF_BZ_RX_INGR_EN_WIDTH 1
+#define	FRF_BZ_RX_IP_HASH_LBN 46
+#define	FRF_BZ_RX_IP_HASH_WIDTH 1
+#define	FRF_BZ_RX_HASH_ALG_LBN 45
+#define	FRF_BZ_RX_HASH_ALG_WIDTH 1
+#define	FRF_BZ_RX_HASH_INSRT_HDR_LBN 44
+#define	FRF_BZ_RX_HASH_INSRT_HDR_WIDTH 1
+#define	FRF_BZ_RX_DESC_PUSH_EN_LBN 43
+#define	FRF_BZ_RX_DESC_PUSH_EN_WIDTH 1
+#define	FRF_BZ_RX_RDW_PATCH_EN_LBN 42
+#define	FRF_BZ_RX_RDW_PATCH_EN_WIDTH 1
+#define	FRF_BB_RX_PCI_BURST_SIZE_LBN 39
+#define	FRF_BB_RX_PCI_BURST_SIZE_WIDTH 3
+#define	FRF_BZ_RX_OWNERR_CTL_LBN 38
+#define	FRF_BZ_RX_OWNERR_CTL_WIDTH 1
+#define	FRF_BZ_RX_XON_TX_TH_LBN 33
+#define	FRF_BZ_RX_XON_TX_TH_WIDTH 5
+#define	FRF_AA_RX_DESC_PUSH_EN_LBN 35
+#define	FRF_AA_RX_DESC_PUSH_EN_WIDTH 1
+#define	FRF_AA_RX_RDW_PATCH_EN_LBN 34
+#define	FRF_AA_RX_RDW_PATCH_EN_WIDTH 1
+#define	FRF_AA_RX_PCI_BURST_SIZE_LBN 31
+#define	FRF_AA_RX_PCI_BURST_SIZE_WIDTH 3
+#define	FRF_BZ_RX_XOFF_TX_TH_LBN 28
+#define	FRF_BZ_RX_XOFF_TX_TH_WIDTH 5
+#define	FRF_AA_RX_OWNERR_CTL_LBN 30
+#define	FRF_AA_RX_OWNERR_CTL_WIDTH 1
+#define	FRF_AA_RX_XON_TX_TH_LBN 25
+#define	FRF_AA_RX_XON_TX_TH_WIDTH 5
+#define	FRF_BZ_RX_USR_BUF_SIZE_LBN 19
+#define	FRF_BZ_RX_USR_BUF_SIZE_WIDTH 9
+#define	FRF_AA_RX_XOFF_TX_TH_LBN 20
+#define	FRF_AA_RX_XOFF_TX_TH_WIDTH 5
+#define	FRF_AA_RX_USR_BUF_SIZE_LBN 11
+#define	FRF_AA_RX_USR_BUF_SIZE_WIDTH 9
+#define	FRF_BZ_RX_XON_MAC_TH_LBN 10
+#define	FRF_BZ_RX_XON_MAC_TH_WIDTH 9
+#define	FRF_AA_RX_XON_MAC_TH_LBN 6
+#define	FRF_AA_RX_XON_MAC_TH_WIDTH 5
+#define	FRF_BZ_RX_XOFF_MAC_TH_LBN 1
+#define	FRF_BZ_RX_XOFF_MAC_TH_WIDTH 9
+#define	FRF_AA_RX_XOFF_MAC_TH_LBN 1
+#define	FRF_AA_RX_XOFF_MAC_TH_WIDTH 5
+#define	FRF_AZ_RX_XOFF_MAC_EN_LBN 0
+#define	FRF_AZ_RX_XOFF_MAC_EN_WIDTH 1
+
+/* RX_FILTER_CTL_REG: Receive filter control registers */
+#define	FR_BZ_RX_FILTER_CTL 0x00000810
+#define	FRF_CZ_ETHERNET_WILDCARD_SEARCH_LIMIT_LBN 94
+#define	FRF_CZ_ETHERNET_WILDCARD_SEARCH_LIMIT_WIDTH 8
+#define	FRF_CZ_ETHERNET_FULL_SEARCH_LIMIT_LBN 86
+#define	FRF_CZ_ETHERNET_FULL_SEARCH_LIMIT_WIDTH 8
+#define	FRF_CZ_RX_FILTER_ALL_VLAN_ETHERTYPES_LBN 85
+#define	FRF_CZ_RX_FILTER_ALL_VLAN_ETHERTYPES_WIDTH 1
+#define	FRF_CZ_RX_VLAN_MATCH_ETHERTYPE_LBN 69
+#define	FRF_CZ_RX_VLAN_MATCH_ETHERTYPE_WIDTH 16
+#define	FRF_CZ_MULTICAST_NOMATCH_Q_ID_LBN 57
+#define	FRF_CZ_MULTICAST_NOMATCH_Q_ID_WIDTH 12
+#define	FRF_CZ_MULTICAST_NOMATCH_RSS_ENABLED_LBN 56
+#define	FRF_CZ_MULTICAST_NOMATCH_RSS_ENABLED_WIDTH 1
+#define	FRF_CZ_MULTICAST_NOMATCH_IP_OVERRIDE_LBN 55
+#define	FRF_CZ_MULTICAST_NOMATCH_IP_OVERRIDE_WIDTH 1
+#define	FRF_CZ_UNICAST_NOMATCH_Q_ID_LBN 43
+#define	FRF_CZ_UNICAST_NOMATCH_Q_ID_WIDTH 12
+#define	FRF_CZ_UNICAST_NOMATCH_RSS_ENABLED_LBN 42
+#define	FRF_CZ_UNICAST_NOMATCH_RSS_ENABLED_WIDTH 1
+#define	FRF_CZ_UNICAST_NOMATCH_IP_OVERRIDE_LBN 41
+#define	FRF_CZ_UNICAST_NOMATCH_IP_OVERRIDE_WIDTH 1
+#define	FRF_BZ_SCATTER_ENBL_NO_MATCH_Q_LBN 40
+#define	FRF_BZ_SCATTER_ENBL_NO_MATCH_Q_WIDTH 1
+#define	FRF_BZ_UDP_FULL_SRCH_LIMIT_LBN 32
+#define	FRF_BZ_UDP_FULL_SRCH_LIMIT_WIDTH 8
+#define	FRF_BZ_NUM_KER_LBN 24
+#define	FRF_BZ_NUM_KER_WIDTH 2
+#define	FRF_BZ_UDP_WILD_SRCH_LIMIT_LBN 16
+#define	FRF_BZ_UDP_WILD_SRCH_LIMIT_WIDTH 8
+#define	FRF_BZ_TCP_WILD_SRCH_LIMIT_LBN 8
+#define	FRF_BZ_TCP_WILD_SRCH_LIMIT_WIDTH 8
+#define	FRF_BZ_TCP_FULL_SRCH_LIMIT_LBN 0
+#define	FRF_BZ_TCP_FULL_SRCH_LIMIT_WIDTH 8
+
+/* RX_FLUSH_DESCQ_REG: Receive flush descriptor queue register */
+#define	FR_AZ_RX_FLUSH_DESCQ 0x00000820
+#define	FRF_AZ_RX_FLUSH_DESCQ_CMD_LBN 24
+#define	FRF_AZ_RX_FLUSH_DESCQ_CMD_WIDTH 1
+#define	FRF_AZ_RX_FLUSH_DESCQ_LBN 0
+#define	FRF_AZ_RX_FLUSH_DESCQ_WIDTH 12
+
+/* RX_DESC_UPD_REGP0: Receive descriptor update register. */
+#define	FR_BZ_RX_DESC_UPD_P0 0x00000830
+#define	FR_BZ_RX_DESC_UPD_P0_STEP 8192
+#define	FR_BZ_RX_DESC_UPD_P0_ROWS 1024
+/* RX_DESC_UPD_REG_KER: Receive descriptor update register. */
+#define	FR_AA_RX_DESC_UPD_KER 0x00000830
+#define	FR_AA_RX_DESC_UPD_KER_STEP 8192
+#define	FR_AA_RX_DESC_UPD_KER_ROWS 4
+/* RX_DESC_UPD_REGP123: Receive descriptor update register. */
+#define	FR_BB_RX_DESC_UPD_P123 0x01000830
+#define	FR_BB_RX_DESC_UPD_P123_STEP 8192
+#define	FR_BB_RX_DESC_UPD_P123_ROWS 3072
+#define	FRF_AZ_RX_DESC_WPTR_LBN 96
+#define	FRF_AZ_RX_DESC_WPTR_WIDTH 12
+#define	FRF_AZ_RX_DESC_PUSH_CMD_LBN 95
+#define	FRF_AZ_RX_DESC_PUSH_CMD_WIDTH 1
+#define	FRF_AZ_RX_DESC_LBN 0
+#define	FRF_AZ_RX_DESC_WIDTH 64
+
+/* RX_DC_CFG_REG: Receive descriptor cache configuration register */
+#define	FR_AZ_RX_DC_CFG 0x00000840
+#define	FRF_AB_RX_MAX_PF_LBN 2
+#define	FRF_AB_RX_MAX_PF_WIDTH 2
+#define	FRF_AZ_RX_DC_SIZE_LBN 0
+#define	FRF_AZ_RX_DC_SIZE_WIDTH 2
+#define	FFE_AZ_RX_DC_SIZE_64 3
+#define	FFE_AZ_RX_DC_SIZE_32 2
+#define	FFE_AZ_RX_DC_SIZE_16 1
+#define	FFE_AZ_RX_DC_SIZE_8 0
+
+/* RX_DC_PF_WM_REG: Receive descriptor cache pre-fetch watermark register */
+#define	FR_AZ_RX_DC_PF_WM 0x00000850
+#define	FRF_AZ_RX_DC_PF_HWM_LBN 6
+#define	FRF_AZ_RX_DC_PF_HWM_WIDTH 6
+#define	FRF_AZ_RX_DC_PF_LWM_LBN 0
+#define	FRF_AZ_RX_DC_PF_LWM_WIDTH 6
+
+/* RX_RSS_TKEY_REG: RSS Toeplitz hash key */
+#define	FR_BZ_RX_RSS_TKEY 0x00000860
+#define	FRF_BZ_RX_RSS_TKEY_HI_LBN 64
+#define	FRF_BZ_RX_RSS_TKEY_HI_WIDTH 64
+#define	FRF_BZ_RX_RSS_TKEY_LO_LBN 0
+#define	FRF_BZ_RX_RSS_TKEY_LO_WIDTH 64
+
+/* RX_NODESC_DROP_REG: Receive dropped packet counter register */
+#define	FR_AZ_RX_NODESC_DROP 0x00000880
+#define	FRF_CZ_RX_NODESC_DROP_CNT_LBN 0
+#define	FRF_CZ_RX_NODESC_DROP_CNT_WIDTH 32
+#define	FRF_AB_RX_NODESC_DROP_CNT_LBN 0
+#define	FRF_AB_RX_NODESC_DROP_CNT_WIDTH 16
+
+/* RX_SELF_RST_REG: Receive self reset register */
+#define	FR_AA_RX_SELF_RST 0x00000890
+#define	FRF_AA_RX_ISCSI_DIS_LBN 17
+#define	FRF_AA_RX_ISCSI_DIS_WIDTH 1
+#define	FRF_AA_RX_SW_RST_REG_LBN 16
+#define	FRF_AA_RX_SW_RST_REG_WIDTH 1
+#define FRF_AA_RX_NODESC_WAIT_DIS_LBN 9
+#define FRF_AA_RX_NODESC_WAIT_DIS_WIDTH 1
+#define	FRF_AA_RX_SELF_RST_EN_LBN 8
+#define	FRF_AA_RX_SELF_RST_EN_WIDTH 1
+#define	FRF_AA_RX_MAX_PF_LAT_LBN 4
+#define	FRF_AA_RX_MAX_PF_LAT_WIDTH 4
+#define	FRF_AA_RX_MAX_LU_LAT_LBN 0
+#define	FRF_AA_RX_MAX_LU_LAT_WIDTH 4
+
+/* RX_DEBUG_REG: undocumented register */
+#define	FR_AZ_RX_DEBUG 0x000008a0
+#define	FRF_AZ_RX_DEBUG_LBN 0
+#define	FRF_AZ_RX_DEBUG_WIDTH 64
+
+/* RX_PUSH_DROP_REG: Receive descriptor push dropped counter register */
+#define	FR_AZ_RX_PUSH_DROP 0x000008b0
+#define	FRF_AZ_RX_PUSH_DROP_CNT_LBN 0
+#define	FRF_AZ_RX_PUSH_DROP_CNT_WIDTH 32
+
+/* RX_RSS_IPV6_REG1: IPv6 RSS Toeplitz hash key low bytes */
+#define	FR_CZ_RX_RSS_IPV6_REG1 0x000008d0
+#define	FRF_CZ_RX_RSS_IPV6_TKEY_LO_LBN 0
+#define	FRF_CZ_RX_RSS_IPV6_TKEY_LO_WIDTH 128
+
+/* RX_RSS_IPV6_REG2: IPv6 RSS Toeplitz hash key middle bytes */
+#define	FR_CZ_RX_RSS_IPV6_REG2 0x000008e0
+#define	FRF_CZ_RX_RSS_IPV6_TKEY_MID_LBN 0
+#define	FRF_CZ_RX_RSS_IPV6_TKEY_MID_WIDTH 128
+
+/* RX_RSS_IPV6_REG3: IPv6 RSS Toeplitz hash key upper bytes and IPv6 RSS settings */
+#define	FR_CZ_RX_RSS_IPV6_REG3 0x000008f0
+#define	FRF_CZ_RX_RSS_IPV6_THASH_ENABLE_LBN 66
+#define	FRF_CZ_RX_RSS_IPV6_THASH_ENABLE_WIDTH 1
+#define	FRF_CZ_RX_RSS_IPV6_IP_THASH_ENABLE_LBN 65
+#define	FRF_CZ_RX_RSS_IPV6_IP_THASH_ENABLE_WIDTH 1
+#define	FRF_CZ_RX_RSS_IPV6_TCP_SUPPRESS_LBN 64
+#define	FRF_CZ_RX_RSS_IPV6_TCP_SUPPRESS_WIDTH 1
+#define	FRF_CZ_RX_RSS_IPV6_TKEY_HI_LBN 0
+#define	FRF_CZ_RX_RSS_IPV6_TKEY_HI_WIDTH 64
+
+/* TX_FLUSH_DESCQ_REG: Transmit flush descriptor queue register */
+#define	FR_AZ_TX_FLUSH_DESCQ 0x00000a00
+#define	FRF_AZ_TX_FLUSH_DESCQ_CMD_LBN 12
+#define	FRF_AZ_TX_FLUSH_DESCQ_CMD_WIDTH 1
+#define	FRF_AZ_TX_FLUSH_DESCQ_LBN 0
+#define	FRF_AZ_TX_FLUSH_DESCQ_WIDTH 12
+
+/* TX_DESC_UPD_REGP0: Transmit descriptor update register. */
+#define	FR_BZ_TX_DESC_UPD_P0 0x00000a10
+#define	FR_BZ_TX_DESC_UPD_P0_STEP 8192
+#define	FR_BZ_TX_DESC_UPD_P0_ROWS 1024
+/* TX_DESC_UPD_REG_KER: Transmit descriptor update register. */
+#define	FR_AA_TX_DESC_UPD_KER 0x00000a10
+#define	FR_AA_TX_DESC_UPD_KER_STEP 8192
+#define	FR_AA_TX_DESC_UPD_KER_ROWS 8
+/* TX_DESC_UPD_REGP123: Transmit descriptor update register. */
+#define	FR_BB_TX_DESC_UPD_P123 0x01000a10
+#define	FR_BB_TX_DESC_UPD_P123_STEP 8192
+#define	FR_BB_TX_DESC_UPD_P123_ROWS 3072
+#define	FRF_AZ_TX_DESC_WPTR_LBN 96
+#define	FRF_AZ_TX_DESC_WPTR_WIDTH 12
+#define	FRF_AZ_TX_DESC_PUSH_CMD_LBN 95
+#define	FRF_AZ_TX_DESC_PUSH_CMD_WIDTH 1
+#define	FRF_AZ_TX_DESC_LBN 0
+#define	FRF_AZ_TX_DESC_WIDTH 95
+
+/* TX_DC_CFG_REG: Transmit descriptor cache configuration register */
+#define	FR_AZ_TX_DC_CFG 0x00000a20
+#define	FRF_AZ_TX_DC_SIZE_LBN 0
+#define	FRF_AZ_TX_DC_SIZE_WIDTH 2
+#define	FFE_AZ_TX_DC_SIZE_32 2
+#define	FFE_AZ_TX_DC_SIZE_16 1
+#define	FFE_AZ_TX_DC_SIZE_8 0
+
+/* TX_CHKSM_CFG_REG: Transmit checksum configuration register */
+#define	FR_AA_TX_CHKSM_CFG 0x00000a30
+#define	FRF_AA_TX_Q_CHKSM_DIS_96_127_LBN 96
+#define	FRF_AA_TX_Q_CHKSM_DIS_96_127_WIDTH 32
+#define	FRF_AA_TX_Q_CHKSM_DIS_64_95_LBN 64
+#define	FRF_AA_TX_Q_CHKSM_DIS_64_95_WIDTH 32
+#define	FRF_AA_TX_Q_CHKSM_DIS_32_63_LBN 32
+#define	FRF_AA_TX_Q_CHKSM_DIS_32_63_WIDTH 32
+#define	FRF_AA_TX_Q_CHKSM_DIS_0_31_LBN 0
+#define	FRF_AA_TX_Q_CHKSM_DIS_0_31_WIDTH 32
+
+/* TX_CFG_REG: Transmit configuration register */
+#define	FR_AZ_TX_CFG 0x00000a50
+#define	FRF_CZ_TX_CONT_LOOKUP_THRESH_RANGE_LBN 114
+#define	FRF_CZ_TX_CONT_LOOKUP_THRESH_RANGE_WIDTH 8
+#define	FRF_CZ_TX_FILTER_TEST_MODE_BIT_LBN 113
+#define	FRF_CZ_TX_FILTER_TEST_MODE_BIT_WIDTH 1
+#define	FRF_CZ_TX_ETH_FILTER_WILD_SEARCH_RANGE_LBN 105
+#define	FRF_CZ_TX_ETH_FILTER_WILD_SEARCH_RANGE_WIDTH 8
+#define	FRF_CZ_TX_ETH_FILTER_FULL_SEARCH_RANGE_LBN 97
+#define	FRF_CZ_TX_ETH_FILTER_FULL_SEARCH_RANGE_WIDTH 8
+#define	FRF_CZ_TX_UDPIP_FILTER_WILD_SEARCH_RANGE_LBN 89
+#define	FRF_CZ_TX_UDPIP_FILTER_WILD_SEARCH_RANGE_WIDTH 8
+#define	FRF_CZ_TX_UDPIP_FILTER_FULL_SEARCH_RANGE_LBN 81
+#define	FRF_CZ_TX_UDPIP_FILTER_FULL_SEARCH_RANGE_WIDTH 8
+#define	FRF_CZ_TX_TCPIP_FILTER_WILD_SEARCH_RANGE_LBN 73
+#define	FRF_CZ_TX_TCPIP_FILTER_WILD_SEARCH_RANGE_WIDTH 8
+#define	FRF_CZ_TX_TCPIP_FILTER_FULL_SEARCH_RANGE_LBN 65
+#define	FRF_CZ_TX_TCPIP_FILTER_FULL_SEARCH_RANGE_WIDTH 8
+#define	FRF_CZ_TX_FILTER_ALL_VLAN_ETHERTYPES_BIT_LBN 64
+#define	FRF_CZ_TX_FILTER_ALL_VLAN_ETHERTYPES_BIT_WIDTH 1
+#define	FRF_CZ_TX_VLAN_MATCH_ETHERTYPE_RANGE_LBN 48
+#define	FRF_CZ_TX_VLAN_MATCH_ETHERTYPE_RANGE_WIDTH 16
+#define	FRF_CZ_TX_FILTER_EN_BIT_LBN 47
+#define	FRF_CZ_TX_FILTER_EN_BIT_WIDTH 1
+#define	FRF_AZ_TX_IP_ID_P0_OFS_LBN 16
+#define	FRF_AZ_TX_IP_ID_P0_OFS_WIDTH 15
+#define	FRF_AZ_TX_NO_EOP_DISC_EN_LBN 5
+#define	FRF_AZ_TX_NO_EOP_DISC_EN_WIDTH 1
+#define	FRF_AZ_TX_P1_PRI_EN_LBN 4
+#define	FRF_AZ_TX_P1_PRI_EN_WIDTH 1
+#define	FRF_AZ_TX_OWNERR_CTL_LBN 2
+#define	FRF_AZ_TX_OWNERR_CTL_WIDTH 1
+#define	FRF_AA_TX_NON_IP_DROP_DIS_LBN 1
+#define	FRF_AA_TX_NON_IP_DROP_DIS_WIDTH 1
+#define	FRF_AZ_TX_IP_ID_REP_EN_LBN 0
+#define	FRF_AZ_TX_IP_ID_REP_EN_WIDTH 1
+
+/* TX_PUSH_DROP_REG: Transmit push dropped register */
+#define	FR_AZ_TX_PUSH_DROP 0x00000a60
+#define	FRF_AZ_TX_PUSH_DROP_CNT_LBN 0
+#define	FRF_AZ_TX_PUSH_DROP_CNT_WIDTH 32
+
+/* TX_RESERVED_REG: Transmit configuration register */
+#define	FR_AZ_TX_RESERVED 0x00000a80
+#define	FRF_AZ_TX_EVT_CNT_LBN 121
+#define	FRF_AZ_TX_EVT_CNT_WIDTH 7
+#define	FRF_AZ_TX_PREF_AGE_CNT_LBN 119
+#define	FRF_AZ_TX_PREF_AGE_CNT_WIDTH 2
+#define	FRF_AZ_TX_RD_COMP_TMR_LBN 96
+#define	FRF_AZ_TX_RD_COMP_TMR_WIDTH 23
+#define	FRF_AZ_TX_PUSH_EN_LBN 89
+#define	FRF_AZ_TX_PUSH_EN_WIDTH 1
+#define	FRF_AZ_TX_PUSH_CHK_DIS_LBN 88
+#define	FRF_AZ_TX_PUSH_CHK_DIS_WIDTH 1
+#define	FRF_AZ_TX_D_FF_FULL_P0_LBN 85
+#define	FRF_AZ_TX_D_FF_FULL_P0_WIDTH 1
+#define	FRF_AZ_TX_DMAR_ST_P0_LBN 81
+#define	FRF_AZ_TX_DMAR_ST_P0_WIDTH 1
+#define	FRF_AZ_TX_DMAQ_ST_LBN 78
+#define	FRF_AZ_TX_DMAQ_ST_WIDTH 1
+#define	FRF_AZ_TX_RX_SPACER_LBN 64
+#define	FRF_AZ_TX_RX_SPACER_WIDTH 8
+#define	FRF_AZ_TX_DROP_ABORT_EN_LBN 60
+#define	FRF_AZ_TX_DROP_ABORT_EN_WIDTH 1
+#define	FRF_AZ_TX_SOFT_EVT_EN_LBN 59
+#define	FRF_AZ_TX_SOFT_EVT_EN_WIDTH 1
+#define	FRF_AZ_TX_PS_EVT_DIS_LBN 58
+#define	FRF_AZ_TX_PS_EVT_DIS_WIDTH 1
+#define	FRF_AZ_TX_RX_SPACER_EN_LBN 57
+#define	FRF_AZ_TX_RX_SPACER_EN_WIDTH 1
+#define	FRF_AZ_TX_XP_TIMER_LBN 52
+#define	FRF_AZ_TX_XP_TIMER_WIDTH 5
+#define	FRF_AZ_TX_PREF_SPACER_LBN 44
+#define	FRF_AZ_TX_PREF_SPACER_WIDTH 8
+#define	FRF_AZ_TX_PREF_WD_TMR_LBN 22
+#define	FRF_AZ_TX_PREF_WD_TMR_WIDTH 22
+#define	FRF_AZ_TX_ONLY1TAG_LBN 21
+#define	FRF_AZ_TX_ONLY1TAG_WIDTH 1
+#define	FRF_AZ_TX_PREF_THRESHOLD_LBN 19
+#define	FRF_AZ_TX_PREF_THRESHOLD_WIDTH 2
+#define	FRF_AZ_TX_ONE_PKT_PER_Q_LBN 18
+#define	FRF_AZ_TX_ONE_PKT_PER_Q_WIDTH 1
+#define	FRF_AZ_TX_DIS_NON_IP_EV_LBN 17
+#define	FRF_AZ_TX_DIS_NON_IP_EV_WIDTH 1
+#define	FRF_AA_TX_DMA_FF_THR_LBN 16
+#define	FRF_AA_TX_DMA_FF_THR_WIDTH 1
+#define	FRF_AZ_TX_DMA_SPACER_LBN 8
+#define	FRF_AZ_TX_DMA_SPACER_WIDTH 8
+#define	FRF_AA_TX_TCP_DIS_LBN 7
+#define	FRF_AA_TX_TCP_DIS_WIDTH 1
+#define	FRF_BZ_TX_FLUSH_MIN_LEN_EN_LBN 7
+#define	FRF_BZ_TX_FLUSH_MIN_LEN_EN_WIDTH 1
+#define	FRF_AA_TX_IP_DIS_LBN 6
+#define	FRF_AA_TX_IP_DIS_WIDTH 1
+#define	FRF_AZ_TX_MAX_CPL_LBN 2
+#define	FRF_AZ_TX_MAX_CPL_WIDTH 2
+#define	FFE_AZ_TX_MAX_CPL_16 3
+#define	FFE_AZ_TX_MAX_CPL_8 2
+#define	FFE_AZ_TX_MAX_CPL_4 1
+#define	FFE_AZ_TX_MAX_CPL_NOLIMIT 0
+#define	FRF_AZ_TX_MAX_PREF_LBN 0
+#define	FRF_AZ_TX_MAX_PREF_WIDTH 2
+#define	FFE_AZ_TX_MAX_PREF_32 3
+#define	FFE_AZ_TX_MAX_PREF_16 2
+#define	FFE_AZ_TX_MAX_PREF_8 1
+#define	FFE_AZ_TX_MAX_PREF_OFF 0
+
+/* TX_PACE_REG: Transmit pace control register */
+#define	FR_BZ_TX_PACE 0x00000a90
+#define	FRF_BZ_TX_PACE_SB_NOT_AF_LBN 19
+#define	FRF_BZ_TX_PACE_SB_NOT_AF_WIDTH 10
+#define	FRF_BZ_TX_PACE_SB_AF_LBN 9
+#define	FRF_BZ_TX_PACE_SB_AF_WIDTH 10
+#define	FRF_BZ_TX_PACE_FB_BASE_LBN 5
+#define	FRF_BZ_TX_PACE_FB_BASE_WIDTH 4
+#define	FRF_BZ_TX_PACE_BIN_TH_LBN 0
+#define	FRF_BZ_TX_PACE_BIN_TH_WIDTH 5
+
+/* TX_PACE_DROP_QID_REG: PACE Drop QID Counter */
+#define	FR_BZ_TX_PACE_DROP_QID 0x00000aa0
+#define	FRF_BZ_TX_PACE_QID_DRP_CNT_LBN 0
+#define	FRF_BZ_TX_PACE_QID_DRP_CNT_WIDTH 16
+
+/* TX_VLAN_REG: Transmit VLAN tag register */
+#define	FR_BB_TX_VLAN 0x00000ae0
+#define	FRF_BB_TX_VLAN_EN_LBN 127
+#define	FRF_BB_TX_VLAN_EN_WIDTH 1
+#define	FRF_BB_TX_VLAN7_PORT1_EN_LBN 125
+#define	FRF_BB_TX_VLAN7_PORT1_EN_WIDTH 1
+#define	FRF_BB_TX_VLAN7_PORT0_EN_LBN 124
+#define	FRF_BB_TX_VLAN7_PORT0_EN_WIDTH 1
+#define	FRF_BB_TX_VLAN7_LBN 112
+#define	FRF_BB_TX_VLAN7_WIDTH 12
+#define	FRF_BB_TX_VLAN6_PORT1_EN_LBN 109
+#define	FRF_BB_TX_VLAN6_PORT1_EN_WIDTH 1
+#define	FRF_BB_TX_VLAN6_PORT0_EN_LBN 108
+#define	FRF_BB_TX_VLAN6_PORT0_EN_WIDTH 1
+#define	FRF_BB_TX_VLAN6_LBN 96
+#define	FRF_BB_TX_VLAN6_WIDTH 12
+#define	FRF_BB_TX_VLAN5_PORT1_EN_LBN 93
+#define	FRF_BB_TX_VLAN5_PORT1_EN_WIDTH 1
+#define	FRF_BB_TX_VLAN5_PORT0_EN_LBN 92
+#define	FRF_BB_TX_VLAN5_PORT0_EN_WIDTH 1
+#define	FRF_BB_TX_VLAN5_LBN 80
+#define	FRF_BB_TX_VLAN5_WIDTH 12
+#define	FRF_BB_TX_VLAN4_PORT1_EN_LBN 77
+#define	FRF_BB_TX_VLAN4_PORT1_EN_WIDTH 1
+#define	FRF_BB_TX_VLAN4_PORT0_EN_LBN 76
+#define	FRF_BB_TX_VLAN4_PORT0_EN_WIDTH 1
+#define	FRF_BB_TX_VLAN4_LBN 64
+#define	FRF_BB_TX_VLAN4_WIDTH 12
+#define	FRF_BB_TX_VLAN3_PORT1_EN_LBN 61
+#define	FRF_BB_TX_VLAN3_PORT1_EN_WIDTH 1
+#define	FRF_BB_TX_VLAN3_PORT0_EN_LBN 60
+#define	FRF_BB_TX_VLAN3_PORT0_EN_WIDTH 1
+#define	FRF_BB_TX_VLAN3_LBN 48
+#define	FRF_BB_TX_VLAN3_WIDTH 12
+#define	FRF_BB_TX_VLAN2_PORT1_EN_LBN 45
+#define	FRF_BB_TX_VLAN2_PORT1_EN_WIDTH 1
+#define	FRF_BB_TX_VLAN2_PORT0_EN_LBN 44
+#define	FRF_BB_TX_VLAN2_PORT0_EN_WIDTH 1
+#define	FRF_BB_TX_VLAN2_LBN 32
+#define	FRF_BB_TX_VLAN2_WIDTH 12
+#define	FRF_BB_TX_VLAN1_PORT1_EN_LBN 29
+#define	FRF_BB_TX_VLAN1_PORT1_EN_WIDTH 1
+#define	FRF_BB_TX_VLAN1_PORT0_EN_LBN 28
+#define	FRF_BB_TX_VLAN1_PORT0_EN_WIDTH 1
+#define	FRF_BB_TX_VLAN1_LBN 16
+#define	FRF_BB_TX_VLAN1_WIDTH 12
+#define	FRF_BB_TX_VLAN0_PORT1_EN_LBN 13
+#define	FRF_BB_TX_VLAN0_PORT1_EN_WIDTH 1
+#define	FRF_BB_TX_VLAN0_PORT0_EN_LBN 12
+#define	FRF_BB_TX_VLAN0_PORT0_EN_WIDTH 1
+#define	FRF_BB_TX_VLAN0_LBN 0
+#define	FRF_BB_TX_VLAN0_WIDTH 12
+
+/* TX_IPFIL_PORTEN_REG: Transmit filter control register */
+#define	FR_BZ_TX_IPFIL_PORTEN 0x00000af0
+#define	FRF_BZ_TX_MADR0_FIL_EN_LBN 64
+#define	FRF_BZ_TX_MADR0_FIL_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL31_PORT_EN_LBN 62
+#define	FRF_BB_TX_IPFIL31_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL30_PORT_EN_LBN 60
+#define	FRF_BB_TX_IPFIL30_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL29_PORT_EN_LBN 58
+#define	FRF_BB_TX_IPFIL29_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL28_PORT_EN_LBN 56
+#define	FRF_BB_TX_IPFIL28_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL27_PORT_EN_LBN 54
+#define	FRF_BB_TX_IPFIL27_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL26_PORT_EN_LBN 52
+#define	FRF_BB_TX_IPFIL26_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL25_PORT_EN_LBN 50
+#define	FRF_BB_TX_IPFIL25_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL24_PORT_EN_LBN 48
+#define	FRF_BB_TX_IPFIL24_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL23_PORT_EN_LBN 46
+#define	FRF_BB_TX_IPFIL23_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL22_PORT_EN_LBN 44
+#define	FRF_BB_TX_IPFIL22_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL21_PORT_EN_LBN 42
+#define	FRF_BB_TX_IPFIL21_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL20_PORT_EN_LBN 40
+#define	FRF_BB_TX_IPFIL20_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL19_PORT_EN_LBN 38
+#define	FRF_BB_TX_IPFIL19_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL18_PORT_EN_LBN 36
+#define	FRF_BB_TX_IPFIL18_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL17_PORT_EN_LBN 34
+#define	FRF_BB_TX_IPFIL17_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL16_PORT_EN_LBN 32
+#define	FRF_BB_TX_IPFIL16_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL15_PORT_EN_LBN 30
+#define	FRF_BB_TX_IPFIL15_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL14_PORT_EN_LBN 28
+#define	FRF_BB_TX_IPFIL14_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL13_PORT_EN_LBN 26
+#define	FRF_BB_TX_IPFIL13_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL12_PORT_EN_LBN 24
+#define	FRF_BB_TX_IPFIL12_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL11_PORT_EN_LBN 22
+#define	FRF_BB_TX_IPFIL11_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL10_PORT_EN_LBN 20
+#define	FRF_BB_TX_IPFIL10_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL9_PORT_EN_LBN 18
+#define	FRF_BB_TX_IPFIL9_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL8_PORT_EN_LBN 16
+#define	FRF_BB_TX_IPFIL8_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL7_PORT_EN_LBN 14
+#define	FRF_BB_TX_IPFIL7_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL6_PORT_EN_LBN 12
+#define	FRF_BB_TX_IPFIL6_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL5_PORT_EN_LBN 10
+#define	FRF_BB_TX_IPFIL5_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL4_PORT_EN_LBN 8
+#define	FRF_BB_TX_IPFIL4_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL3_PORT_EN_LBN 6
+#define	FRF_BB_TX_IPFIL3_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL2_PORT_EN_LBN 4
+#define	FRF_BB_TX_IPFIL2_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL1_PORT_EN_LBN 2
+#define	FRF_BB_TX_IPFIL1_PORT_EN_WIDTH 1
+#define	FRF_BB_TX_IPFIL0_PORT_EN_LBN 0
+#define	FRF_BB_TX_IPFIL0_PORT_EN_WIDTH 1
+
+/* TX_IPFIL_TBL: Transmit IP source address filter table */
+#define	FR_BB_TX_IPFIL_TBL 0x00000b00
+#define	FR_BB_TX_IPFIL_TBL_STEP 16
+#define	FR_BB_TX_IPFIL_TBL_ROWS 16
+#define	FRF_BB_TX_IPFIL_MASK_1_LBN 96
+#define	FRF_BB_TX_IPFIL_MASK_1_WIDTH 32
+#define	FRF_BB_TX_IP_SRC_ADR_1_LBN 64
+#define	FRF_BB_TX_IP_SRC_ADR_1_WIDTH 32
+#define	FRF_BB_TX_IPFIL_MASK_0_LBN 32
+#define	FRF_BB_TX_IPFIL_MASK_0_WIDTH 32
+#define	FRF_BB_TX_IP_SRC_ADR_0_LBN 0
+#define	FRF_BB_TX_IP_SRC_ADR_0_WIDTH 32
+
+/* MD_TXD_REG: PHY management transmit data register */
+#define	FR_AB_MD_TXD 0x00000c00
+#define	FRF_AB_MD_TXD_LBN 0
+#define	FRF_AB_MD_TXD_WIDTH 16
+
+/* MD_RXD_REG: PHY management receive data register */
+#define	FR_AB_MD_RXD 0x00000c10
+#define	FRF_AB_MD_RXD_LBN 0
+#define	FRF_AB_MD_RXD_WIDTH 16
+
+/* MD_CS_REG: PHY management configuration & status register */
+#define	FR_AB_MD_CS 0x00000c20
+#define	FRF_AB_MD_RD_EN_CMD_LBN 15
+#define	FRF_AB_MD_RD_EN_CMD_WIDTH 1
+#define	FRF_AB_MD_WR_EN_CMD_LBN 14
+#define	FRF_AB_MD_WR_EN_CMD_WIDTH 1
+#define	FRF_AB_MD_ADDR_CMD_LBN 13
+#define	FRF_AB_MD_ADDR_CMD_WIDTH 1
+#define	FRF_AB_MD_PT_LBN 7
+#define	FRF_AB_MD_PT_WIDTH 3
+#define	FRF_AB_MD_PL_LBN 6
+#define	FRF_AB_MD_PL_WIDTH 1
+#define	FRF_AB_MD_INT_CLR_LBN 5
+#define	FRF_AB_MD_INT_CLR_WIDTH 1
+#define	FRF_AB_MD_GC_LBN 4
+#define	FRF_AB_MD_GC_WIDTH 1
+#define	FRF_AB_MD_PRSP_LBN 3
+#define	FRF_AB_MD_PRSP_WIDTH 1
+#define	FRF_AB_MD_RIC_LBN 2
+#define	FRF_AB_MD_RIC_WIDTH 1
+#define	FRF_AB_MD_RDC_LBN 1
+#define	FRF_AB_MD_RDC_WIDTH 1
+#define	FRF_AB_MD_WRC_LBN 0
+#define	FRF_AB_MD_WRC_WIDTH 1
+
+/* MD_PHY_ADR_REG: PHY management PHY address register */
+#define	FR_AB_MD_PHY_ADR 0x00000c30
+#define	FRF_AB_MD_PHY_ADR_LBN 0
+#define	FRF_AB_MD_PHY_ADR_WIDTH 16
+
+/* MD_ID_REG: PHY management ID register */
+#define	FR_AB_MD_ID 0x00000c40
+#define	FRF_AB_MD_PRT_ADR_LBN 11
+#define	FRF_AB_MD_PRT_ADR_WIDTH 5
+#define	FRF_AB_MD_DEV_ADR_LBN 6
+#define	FRF_AB_MD_DEV_ADR_WIDTH 5
+
+/* MD_STAT_REG: PHY management status & mask register */
+#define	FR_AB_MD_STAT 0x00000c50
+#define	FRF_AB_MD_PINT_LBN 4
+#define	FRF_AB_MD_PINT_WIDTH 1
+#define	FRF_AB_MD_DONE_LBN 3
+#define	FRF_AB_MD_DONE_WIDTH 1
+#define	FRF_AB_MD_BSERR_LBN 2
+#define	FRF_AB_MD_BSERR_WIDTH 1
+#define	FRF_AB_MD_LNFL_LBN 1
+#define	FRF_AB_MD_LNFL_WIDTH 1
+#define	FRF_AB_MD_BSY_LBN 0
+#define	FRF_AB_MD_BSY_WIDTH 1
+
+/* MAC_STAT_DMA_REG: Port MAC statistical counter DMA register */
+#define	FR_AB_MAC_STAT_DMA 0x00000c60
+#define	FRF_AB_MAC_STAT_DMA_CMD_LBN 48
+#define	FRF_AB_MAC_STAT_DMA_CMD_WIDTH 1
+#define	FRF_AB_MAC_STAT_DMA_ADR_LBN 0
+#define	FRF_AB_MAC_STAT_DMA_ADR_WIDTH 48
+
+/* MAC_CTRL_REG: Port MAC control register */
+#define	FR_AB_MAC_CTRL 0x00000c80
+#define	FRF_AB_MAC_XOFF_VAL_LBN 16
+#define	FRF_AB_MAC_XOFF_VAL_WIDTH 16
+#define	FRF_BB_TXFIFO_DRAIN_EN_LBN 7
+#define	FRF_BB_TXFIFO_DRAIN_EN_WIDTH 1
+#define	FRF_AB_MAC_XG_DISTXCRC_LBN 5
+#define	FRF_AB_MAC_XG_DISTXCRC_WIDTH 1
+#define	FRF_AB_MAC_BCAD_ACPT_LBN 4
+#define	FRF_AB_MAC_BCAD_ACPT_WIDTH 1
+#define	FRF_AB_MAC_UC_PROM_LBN 3
+#define	FRF_AB_MAC_UC_PROM_WIDTH 1
+#define	FRF_AB_MAC_LINK_STATUS_LBN 2
+#define	FRF_AB_MAC_LINK_STATUS_WIDTH 1
+#define	FRF_AB_MAC_SPEED_LBN 0
+#define	FRF_AB_MAC_SPEED_WIDTH 2
+#define	FFE_AB_MAC_SPEED_10G 3
+#define	FFE_AB_MAC_SPEED_1G 2
+#define	FFE_AB_MAC_SPEED_100M 1
+#define	FFE_AB_MAC_SPEED_10M 0
+
+/* GEN_MODE_REG: General Purpose mode register (external interrupt mask) */
+#define	FR_BB_GEN_MODE 0x00000c90
+#define	FRF_BB_XFP_PHY_INT_POL_SEL_LBN 3
+#define	FRF_BB_XFP_PHY_INT_POL_SEL_WIDTH 1
+#define	FRF_BB_XG_PHY_INT_POL_SEL_LBN 2
+#define	FRF_BB_XG_PHY_INT_POL_SEL_WIDTH 1
+#define	FRF_BB_XFP_PHY_INT_MASK_LBN 1
+#define	FRF_BB_XFP_PHY_INT_MASK_WIDTH 1
+#define	FRF_BB_XG_PHY_INT_MASK_LBN 0
+#define	FRF_BB_XG_PHY_INT_MASK_WIDTH 1
+
+/* MAC_MC_HASH_REG0: Multicast address hash table */
+#define	FR_AB_MAC_MC_HASH_REG0 0x00000ca0
+#define	FRF_AB_MAC_MCAST_HASH0_LBN 0
+#define	FRF_AB_MAC_MCAST_HASH0_WIDTH 128
+
+/* MAC_MC_HASH_REG1: Multicast address hash table */
+#define	FR_AB_MAC_MC_HASH_REG1 0x00000cb0
+#define	FRF_AB_MAC_MCAST_HASH1_LBN 0
+#define	FRF_AB_MAC_MCAST_HASH1_WIDTH 128
+
+/* GM_CFG1_REG: GMAC configuration register 1 */
+#define	FR_AB_GM_CFG1 0x00000e00
+#define	FRF_AB_GM_SW_RST_LBN 31
+#define	FRF_AB_GM_SW_RST_WIDTH 1
+#define	FRF_AB_GM_SIM_RST_LBN 30
+#define	FRF_AB_GM_SIM_RST_WIDTH 1
+#define	FRF_AB_GM_RST_RX_MAC_CTL_LBN 19
+#define	FRF_AB_GM_RST_RX_MAC_CTL_WIDTH 1
+#define	FRF_AB_GM_RST_TX_MAC_CTL_LBN 18
+#define	FRF_AB_GM_RST_TX_MAC_CTL_WIDTH 1
+#define	FRF_AB_GM_RST_RX_FUNC_LBN 17
+#define	FRF_AB_GM_RST_RX_FUNC_WIDTH 1
+#define	FRF_AB_GM_RST_TX_FUNC_LBN 16
+#define	FRF_AB_GM_RST_TX_FUNC_WIDTH 1
+#define	FRF_AB_GM_LOOP_LBN 8
+#define	FRF_AB_GM_LOOP_WIDTH 1
+#define	FRF_AB_GM_RX_FC_EN_LBN 5
+#define	FRF_AB_GM_RX_FC_EN_WIDTH 1
+#define	FRF_AB_GM_TX_FC_EN_LBN 4
+#define	FRF_AB_GM_TX_FC_EN_WIDTH 1
+#define	FRF_AB_GM_SYNC_RXEN_LBN 3
+#define	FRF_AB_GM_SYNC_RXEN_WIDTH 1
+#define	FRF_AB_GM_RX_EN_LBN 2
+#define	FRF_AB_GM_RX_EN_WIDTH 1
+#define	FRF_AB_GM_SYNC_TXEN_LBN 1
+#define	FRF_AB_GM_SYNC_TXEN_WIDTH 1
+#define	FRF_AB_GM_TX_EN_LBN 0
+#define	FRF_AB_GM_TX_EN_WIDTH 1
+
+/* GM_CFG2_REG: GMAC configuration register 2 */
+#define	FR_AB_GM_CFG2 0x00000e10
+#define	FRF_AB_GM_PAMBL_LEN_LBN 12
+#define	FRF_AB_GM_PAMBL_LEN_WIDTH 4
+#define	FRF_AB_GM_IF_MODE_LBN 8
+#define	FRF_AB_GM_IF_MODE_WIDTH 2
+#define	FFE_AB_IF_MODE_BYTE_MODE 2
+#define	FFE_AB_IF_MODE_NIBBLE_MODE 1
+#define	FRF_AB_GM_HUGE_FRM_EN_LBN 5
+#define	FRF_AB_GM_HUGE_FRM_EN_WIDTH 1
+#define	FRF_AB_GM_LEN_CHK_LBN 4
+#define	FRF_AB_GM_LEN_CHK_WIDTH 1
+#define	FRF_AB_GM_PAD_CRC_EN_LBN 2
+#define	FRF_AB_GM_PAD_CRC_EN_WIDTH 1
+#define	FRF_AB_GM_CRC_EN_LBN 1
+#define	FRF_AB_GM_CRC_EN_WIDTH 1
+#define	FRF_AB_GM_FD_LBN 0
+#define	FRF_AB_GM_FD_WIDTH 1
+
+/* GM_IPG_REG: GMAC IPG register */
+#define	FR_AB_GM_IPG 0x00000e20
+#define	FRF_AB_GM_NONB2B_IPG1_LBN 24
+#define	FRF_AB_GM_NONB2B_IPG1_WIDTH 7
+#define	FRF_AB_GM_NONB2B_IPG2_LBN 16
+#define	FRF_AB_GM_NONB2B_IPG2_WIDTH 7
+#define	FRF_AB_GM_MIN_IPG_ENF_LBN 8
+#define	FRF_AB_GM_MIN_IPG_ENF_WIDTH 8
+#define	FRF_AB_GM_B2B_IPG_LBN 0
+#define	FRF_AB_GM_B2B_IPG_WIDTH 7
+
+/* GM_HD_REG: GMAC half duplex register */
+#define	FR_AB_GM_HD 0x00000e30
+#define	FRF_AB_GM_ALT_BOFF_VAL_LBN 20
+#define	FRF_AB_GM_ALT_BOFF_VAL_WIDTH 4
+#define	FRF_AB_GM_ALT_BOFF_EN_LBN 19
+#define	FRF_AB_GM_ALT_BOFF_EN_WIDTH 1
+#define	FRF_AB_GM_BP_NO_BOFF_LBN 18
+#define	FRF_AB_GM_BP_NO_BOFF_WIDTH 1
+#define	FRF_AB_GM_DIS_BOFF_LBN 17
+#define	FRF_AB_GM_DIS_BOFF_WIDTH 1
+#define	FRF_AB_GM_EXDEF_TX_EN_LBN 16
+#define	FRF_AB_GM_EXDEF_TX_EN_WIDTH 1
+#define	FRF_AB_GM_RTRY_LIMIT_LBN 12
+#define	FRF_AB_GM_RTRY_LIMIT_WIDTH 4
+#define	FRF_AB_GM_COL_WIN_LBN 0
+#define	FRF_AB_GM_COL_WIN_WIDTH 10
+
+/* GM_MAX_FLEN_REG: GMAC maximum frame length register */
+#define	FR_AB_GM_MAX_FLEN 0x00000e40
+#define	FRF_AB_GM_MAX_FLEN_LBN 0
+#define	FRF_AB_GM_MAX_FLEN_WIDTH 16
+
+/* GM_TEST_REG: GMAC test register */
+#define	FR_AB_GM_TEST 0x00000e70
+#define	FRF_AB_GM_MAX_BOFF_LBN 3
+#define	FRF_AB_GM_MAX_BOFF_WIDTH 1
+#define	FRF_AB_GM_REG_TX_FLOW_EN_LBN 2
+#define	FRF_AB_GM_REG_TX_FLOW_EN_WIDTH 1
+#define	FRF_AB_GM_TEST_PAUSE_LBN 1
+#define	FRF_AB_GM_TEST_PAUSE_WIDTH 1
+#define	FRF_AB_GM_SHORT_SLOT_LBN 0
+#define	FRF_AB_GM_SHORT_SLOT_WIDTH 1
+
+/* GM_ADR1_REG: GMAC station address register 1 */
+#define	FR_AB_GM_ADR1 0x00000f00
+#define	FRF_AB_GM_ADR_B0_LBN 24
+#define	FRF_AB_GM_ADR_B0_WIDTH 8
+#define	FRF_AB_GM_ADR_B1_LBN 16
+#define	FRF_AB_GM_ADR_B1_WIDTH 8
+#define	FRF_AB_GM_ADR_B2_LBN 8
+#define	FRF_AB_GM_ADR_B2_WIDTH 8
+#define	FRF_AB_GM_ADR_B3_LBN 0
+#define	FRF_AB_GM_ADR_B3_WIDTH 8
+
+/* GM_ADR2_REG: GMAC station address register 2 */
+#define	FR_AB_GM_ADR2 0x00000f10
+#define	FRF_AB_GM_ADR_B4_LBN 24
+#define	FRF_AB_GM_ADR_B4_WIDTH 8
+#define	FRF_AB_GM_ADR_B5_LBN 16
+#define	FRF_AB_GM_ADR_B5_WIDTH 8
+
+/* GMF_CFG0_REG: GMAC FIFO configuration register 0 */
+#define	FR_AB_GMF_CFG0 0x00000f20
+#define	FRF_AB_GMF_FTFENRPLY_LBN 20
+#define	FRF_AB_GMF_FTFENRPLY_WIDTH 1
+#define	FRF_AB_GMF_STFENRPLY_LBN 19
+#define	FRF_AB_GMF_STFENRPLY_WIDTH 1
+#define	FRF_AB_GMF_FRFENRPLY_LBN 18
+#define	FRF_AB_GMF_FRFENRPLY_WIDTH 1
+#define	FRF_AB_GMF_SRFENRPLY_LBN 17
+#define	FRF_AB_GMF_SRFENRPLY_WIDTH 1
+#define	FRF_AB_GMF_WTMENRPLY_LBN 16
+#define	FRF_AB_GMF_WTMENRPLY_WIDTH 1
+#define	FRF_AB_GMF_FTFENREQ_LBN 12
+#define	FRF_AB_GMF_FTFENREQ_WIDTH 1
+#define	FRF_AB_GMF_STFENREQ_LBN 11
+#define	FRF_AB_GMF_STFENREQ_WIDTH 1
+#define	FRF_AB_GMF_FRFENREQ_LBN 10
+#define	FRF_AB_GMF_FRFENREQ_WIDTH 1
+#define	FRF_AB_GMF_SRFENREQ_LBN 9
+#define	FRF_AB_GMF_SRFENREQ_WIDTH 1
+#define	FRF_AB_GMF_WTMENREQ_LBN 8
+#define	FRF_AB_GMF_WTMENREQ_WIDTH 1
+#define	FRF_AB_GMF_HSTRSTFT_LBN 4
+#define	FRF_AB_GMF_HSTRSTFT_WIDTH 1
+#define	FRF_AB_GMF_HSTRSTST_LBN 3
+#define	FRF_AB_GMF_HSTRSTST_WIDTH 1
+#define	FRF_AB_GMF_HSTRSTFR_LBN 2
+#define	FRF_AB_GMF_HSTRSTFR_WIDTH 1
+#define	FRF_AB_GMF_HSTRSTSR_LBN 1
+#define	FRF_AB_GMF_HSTRSTSR_WIDTH 1
+#define	FRF_AB_GMF_HSTRSTWT_LBN 0
+#define	FRF_AB_GMF_HSTRSTWT_WIDTH 1
+
+/* GMF_CFG1_REG: GMAC FIFO configuration register 1 */
+#define	FR_AB_GMF_CFG1 0x00000f30
+#define	FRF_AB_GMF_CFGFRTH_LBN 16
+#define	FRF_AB_GMF_CFGFRTH_WIDTH 5
+#define	FRF_AB_GMF_CFGXOFFRTX_LBN 0
+#define	FRF_AB_GMF_CFGXOFFRTX_WIDTH 16
+
+/* GMF_CFG2_REG: GMAC FIFO configuration register 2 */
+#define	FR_AB_GMF_CFG2 0x00000f40
+#define	FRF_AB_GMF_CFGHWM_LBN 16
+#define	FRF_AB_GMF_CFGHWM_WIDTH 6
+#define	FRF_AB_GMF_CFGLWM_LBN 0
+#define	FRF_AB_GMF_CFGLWM_WIDTH 6
+
+/* GMF_CFG3_REG: GMAC FIFO configuration register 3 */
+#define	FR_AB_GMF_CFG3 0x00000f50
+#define	FRF_AB_GMF_CFGHWMFT_LBN 16
+#define	FRF_AB_GMF_CFGHWMFT_WIDTH 6
+#define	FRF_AB_GMF_CFGFTTH_LBN 0
+#define	FRF_AB_GMF_CFGFTTH_WIDTH 6
+
+/* GMF_CFG4_REG: GMAC FIFO configuration register 4 */
+#define	FR_AB_GMF_CFG4 0x00000f60
+#define	FRF_AB_GMF_HSTFLTRFRM_LBN 0
+#define	FRF_AB_GMF_HSTFLTRFRM_WIDTH 18
+
+/* GMF_CFG5_REG: GMAC FIFO configuration register 5 */
+#define	FR_AB_GMF_CFG5 0x00000f70
+#define	FRF_AB_GMF_CFGHDPLX_LBN 22
+#define	FRF_AB_GMF_CFGHDPLX_WIDTH 1
+#define	FRF_AB_GMF_SRFULL_LBN 21
+#define	FRF_AB_GMF_SRFULL_WIDTH 1
+#define	FRF_AB_GMF_HSTSRFULLCLR_LBN 20
+#define	FRF_AB_GMF_HSTSRFULLCLR_WIDTH 1
+#define	FRF_AB_GMF_CFGBYTMODE_LBN 19
+#define	FRF_AB_GMF_CFGBYTMODE_WIDTH 1
+#define	FRF_AB_GMF_HSTDRPLT64_LBN 18
+#define	FRF_AB_GMF_HSTDRPLT64_WIDTH 1
+#define	FRF_AB_GMF_HSTFLTRFRMDC_LBN 0
+#define	FRF_AB_GMF_HSTFLTRFRMDC_WIDTH 18
+
+/* TX_SRC_MAC_TBL: Transmit IP source address filter table */
+#define	FR_BB_TX_SRC_MAC_TBL 0x00001000
+#define	FR_BB_TX_SRC_MAC_TBL_STEP 16
+#define	FR_BB_TX_SRC_MAC_TBL_ROWS 16
+#define	FRF_BB_TX_SRC_MAC_ADR_1_LBN 64
+#define	FRF_BB_TX_SRC_MAC_ADR_1_WIDTH 48
+#define	FRF_BB_TX_SRC_MAC_ADR_0_LBN 0
+#define	FRF_BB_TX_SRC_MAC_ADR_0_WIDTH 48
+
+/* TX_SRC_MAC_CTL_REG: Transmit MAC source address filter control */
+#define	FR_BB_TX_SRC_MAC_CTL 0x00001100
+#define	FRF_BB_TX_SRC_DROP_CTR_LBN 16
+#define	FRF_BB_TX_SRC_DROP_CTR_WIDTH 16
+#define	FRF_BB_TX_SRC_FLTR_EN_LBN 15
+#define	FRF_BB_TX_SRC_FLTR_EN_WIDTH 1
+#define	FRF_BB_TX_DROP_CTR_CLR_LBN 12
+#define	FRF_BB_TX_DROP_CTR_CLR_WIDTH 1
+#define	FRF_BB_TX_MAC_QID_SEL_LBN 0
+#define	FRF_BB_TX_MAC_QID_SEL_WIDTH 3
+
+/* XM_ADR_LO_REG: XGMAC address register low */
+#define	FR_AB_XM_ADR_LO 0x00001200
+#define	FRF_AB_XM_ADR_LO_LBN 0
+#define	FRF_AB_XM_ADR_LO_WIDTH 32
+
+/* XM_ADR_HI_REG: XGMAC address register high */
+#define	FR_AB_XM_ADR_HI 0x00001210
+#define	FRF_AB_XM_ADR_HI_LBN 0
+#define	FRF_AB_XM_ADR_HI_WIDTH 16
+
+/* XM_GLB_CFG_REG: XGMAC global configuration */
+#define	FR_AB_XM_GLB_CFG 0x00001220
+#define	FRF_AB_XM_RMTFLT_GEN_LBN 17
+#define	FRF_AB_XM_RMTFLT_GEN_WIDTH 1
+#define	FRF_AB_XM_DEBUG_MODE_LBN 16
+#define	FRF_AB_XM_DEBUG_MODE_WIDTH 1
+#define	FRF_AB_XM_RX_STAT_EN_LBN 11
+#define	FRF_AB_XM_RX_STAT_EN_WIDTH 1
+#define	FRF_AB_XM_TX_STAT_EN_LBN 10
+#define	FRF_AB_XM_TX_STAT_EN_WIDTH 1
+#define	FRF_AB_XM_RX_JUMBO_MODE_LBN 6
+#define	FRF_AB_XM_RX_JUMBO_MODE_WIDTH 1
+#define	FRF_AB_XM_WAN_MODE_LBN 5
+#define	FRF_AB_XM_WAN_MODE_WIDTH 1
+#define	FRF_AB_XM_INTCLR_MODE_LBN 3
+#define	FRF_AB_XM_INTCLR_MODE_WIDTH 1
+#define	FRF_AB_XM_CORE_RST_LBN 0
+#define	FRF_AB_XM_CORE_RST_WIDTH 1
+
+/* XM_TX_CFG_REG: XGMAC transmit configuration */
+#define	FR_AB_XM_TX_CFG 0x00001230
+#define	FRF_AB_XM_TX_PROG_LBN 24
+#define	FRF_AB_XM_TX_PROG_WIDTH 1
+#define	FRF_AB_XM_IPG_LBN 16
+#define	FRF_AB_XM_IPG_WIDTH 4
+#define	FRF_AB_XM_FCNTL_LBN 10
+#define	FRF_AB_XM_FCNTL_WIDTH 1
+#define	FRF_AB_XM_TXCRC_LBN 8
+#define	FRF_AB_XM_TXCRC_WIDTH 1
+#define	FRF_AB_XM_EDRC_LBN 6
+#define	FRF_AB_XM_EDRC_WIDTH 1
+#define	FRF_AB_XM_AUTO_PAD_LBN 5
+#define	FRF_AB_XM_AUTO_PAD_WIDTH 1
+#define	FRF_AB_XM_TX_PRMBL_LBN 2
+#define	FRF_AB_XM_TX_PRMBL_WIDTH 1
+#define	FRF_AB_XM_TXEN_LBN 1
+#define	FRF_AB_XM_TXEN_WIDTH 1
+#define	FRF_AB_XM_TX_RST_LBN 0
+#define	FRF_AB_XM_TX_RST_WIDTH 1
+
+/* XM_RX_CFG_REG: XGMAC receive configuration */
+#define	FR_AB_XM_RX_CFG 0x00001240
+#define	FRF_AB_XM_PASS_LENERR_LBN 26
+#define	FRF_AB_XM_PASS_LENERR_WIDTH 1
+#define	FRF_AB_XM_PASS_CRC_ERR_LBN 25
+#define	FRF_AB_XM_PASS_CRC_ERR_WIDTH 1
+#define	FRF_AB_XM_PASS_PRMBLE_ERR_LBN 24
+#define	FRF_AB_XM_PASS_PRMBLE_ERR_WIDTH 1
+#define	FRF_AB_XM_REJ_BCAST_LBN 20
+#define	FRF_AB_XM_REJ_BCAST_WIDTH 1
+#define	FRF_AB_XM_ACPT_ALL_MCAST_LBN 11
+#define	FRF_AB_XM_ACPT_ALL_MCAST_WIDTH 1
+#define	FRF_AB_XM_ACPT_ALL_UCAST_LBN 9
+#define	FRF_AB_XM_ACPT_ALL_UCAST_WIDTH 1
+#define	FRF_AB_XM_AUTO_DEPAD_LBN 8
+#define	FRF_AB_XM_AUTO_DEPAD_WIDTH 1
+#define	FRF_AB_XM_RXCRC_LBN 3
+#define	FRF_AB_XM_RXCRC_WIDTH 1
+#define	FRF_AB_XM_RX_PRMBL_LBN 2
+#define	FRF_AB_XM_RX_PRMBL_WIDTH 1
+#define	FRF_AB_XM_RXEN_LBN 1
+#define	FRF_AB_XM_RXEN_WIDTH 1
+#define	FRF_AB_XM_RX_RST_LBN 0
+#define	FRF_AB_XM_RX_RST_WIDTH 1
+
+/* XM_MGT_INT_MASK: documentation to be written for sum_XM_MGT_INT_MASK */
+#define	FR_AB_XM_MGT_INT_MASK 0x00001250
+#define	FRF_AB_XM_MSK_STA_INTR_LBN 16
+#define	FRF_AB_XM_MSK_STA_INTR_WIDTH 1
+#define	FRF_AB_XM_MSK_STAT_CNTR_HF_LBN 9
+#define	FRF_AB_XM_MSK_STAT_CNTR_HF_WIDTH 1
+#define	FRF_AB_XM_MSK_STAT_CNTR_OF_LBN 8
+#define	FRF_AB_XM_MSK_STAT_CNTR_OF_WIDTH 1
+#define	FRF_AB_XM_MSK_PRMBLE_ERR_LBN 2
+#define	FRF_AB_XM_MSK_PRMBLE_ERR_WIDTH 1
+#define	FRF_AB_XM_MSK_RMTFLT_LBN 1
+#define	FRF_AB_XM_MSK_RMTFLT_WIDTH 1
+#define	FRF_AB_XM_MSK_LCLFLT_LBN 0
+#define	FRF_AB_XM_MSK_LCLFLT_WIDTH 1
+
+/* XM_FC_REG: XGMAC flow control register */
+#define	FR_AB_XM_FC 0x00001270
+#define	FRF_AB_XM_PAUSE_TIME_LBN 16
+#define	FRF_AB_XM_PAUSE_TIME_WIDTH 16
+#define	FRF_AB_XM_RX_MAC_STAT_LBN 11
+#define	FRF_AB_XM_RX_MAC_STAT_WIDTH 1
+#define	FRF_AB_XM_TX_MAC_STAT_LBN 10
+#define	FRF_AB_XM_TX_MAC_STAT_WIDTH 1
+#define	FRF_AB_XM_MCNTL_PASS_LBN 8
+#define	FRF_AB_XM_MCNTL_PASS_WIDTH 2
+#define	FRF_AB_XM_REJ_CNTL_UCAST_LBN 6
+#define	FRF_AB_XM_REJ_CNTL_UCAST_WIDTH 1
+#define	FRF_AB_XM_REJ_CNTL_MCAST_LBN 5
+#define	FRF_AB_XM_REJ_CNTL_MCAST_WIDTH 1
+#define	FRF_AB_XM_ZPAUSE_LBN 2
+#define	FRF_AB_XM_ZPAUSE_WIDTH 1
+#define	FRF_AB_XM_XMIT_PAUSE_LBN 1
+#define	FRF_AB_XM_XMIT_PAUSE_WIDTH 1
+#define	FRF_AB_XM_DIS_FCNTL_LBN 0
+#define	FRF_AB_XM_DIS_FCNTL_WIDTH 1
+
+/* XM_PAUSE_TIME_REG: XGMAC pause time register */
+#define	FR_AB_XM_PAUSE_TIME 0x00001290
+#define	FRF_AB_XM_TX_PAUSE_CNT_LBN 16
+#define	FRF_AB_XM_TX_PAUSE_CNT_WIDTH 16
+#define	FRF_AB_XM_RX_PAUSE_CNT_LBN 0
+#define	FRF_AB_XM_RX_PAUSE_CNT_WIDTH 16
+
+/* XM_TX_PARAM_REG: XGMAC transmit parameter register */
+#define	FR_AB_XM_TX_PARAM 0x000012d0
+#define	FRF_AB_XM_TX_JUMBO_MODE_LBN 31
+#define	FRF_AB_XM_TX_JUMBO_MODE_WIDTH 1
+#define	FRF_AB_XM_MAX_TX_FRM_SIZE_HI_LBN 19
+#define	FRF_AB_XM_MAX_TX_FRM_SIZE_HI_WIDTH 11
+#define	FRF_AB_XM_MAX_TX_FRM_SIZE_LO_LBN 16
+#define	FRF_AB_XM_MAX_TX_FRM_SIZE_LO_WIDTH 3
+#define	FRF_AB_XM_PAD_CHAR_LBN 0
+#define	FRF_AB_XM_PAD_CHAR_WIDTH 8
+
+/* XM_RX_PARAM_REG: XGMAC receive parameter register */
+#define	FR_AB_XM_RX_PARAM 0x000012e0
+#define	FRF_AB_XM_MAX_RX_FRM_SIZE_HI_LBN 3
+#define	FRF_AB_XM_MAX_RX_FRM_SIZE_HI_WIDTH 11
+#define	FRF_AB_XM_MAX_RX_FRM_SIZE_LO_LBN 0
+#define	FRF_AB_XM_MAX_RX_FRM_SIZE_LO_WIDTH 3
+
+/* XM_MGT_INT_MSK_REG: XGMAC management interrupt mask register */
+#define	FR_AB_XM_MGT_INT_MSK 0x000012f0
+#define	FRF_AB_XM_STAT_CNTR_OF_LBN 9
+#define	FRF_AB_XM_STAT_CNTR_OF_WIDTH 1
+#define	FRF_AB_XM_STAT_CNTR_HF_LBN 8
+#define	FRF_AB_XM_STAT_CNTR_HF_WIDTH 1
+#define	FRF_AB_XM_PRMBLE_ERR_LBN 2
+#define	FRF_AB_XM_PRMBLE_ERR_WIDTH 1
+#define	FRF_AB_XM_RMTFLT_LBN 1
+#define	FRF_AB_XM_RMTFLT_WIDTH 1
+#define	FRF_AB_XM_LCLFLT_LBN 0
+#define	FRF_AB_XM_LCLFLT_WIDTH 1
+
+/* XX_PWR_RST_REG: XGXS/XAUI powerdown/reset register */
+#define	FR_AB_XX_PWR_RST 0x00001300
+#define	FRF_AB_XX_PWRDND_SIG_LBN 31
+#define	FRF_AB_XX_PWRDND_SIG_WIDTH 1
+#define	FRF_AB_XX_PWRDNC_SIG_LBN 30
+#define	FRF_AB_XX_PWRDNC_SIG_WIDTH 1
+#define	FRF_AB_XX_PWRDNB_SIG_LBN 29
+#define	FRF_AB_XX_PWRDNB_SIG_WIDTH 1
+#define	FRF_AB_XX_PWRDNA_SIG_LBN 28
+#define	FRF_AB_XX_PWRDNA_SIG_WIDTH 1
+#define	FRF_AB_XX_SIM_MODE_LBN 27
+#define	FRF_AB_XX_SIM_MODE_WIDTH 1
+#define	FRF_AB_XX_RSTPLLCD_SIG_LBN 25
+#define	FRF_AB_XX_RSTPLLCD_SIG_WIDTH 1
+#define	FRF_AB_XX_RSTPLLAB_SIG_LBN 24
+#define	FRF_AB_XX_RSTPLLAB_SIG_WIDTH 1
+#define	FRF_AB_XX_RESETD_SIG_LBN 23
+#define	FRF_AB_XX_RESETD_SIG_WIDTH 1
+#define	FRF_AB_XX_RESETC_SIG_LBN 22
+#define	FRF_AB_XX_RESETC_SIG_WIDTH 1
+#define	FRF_AB_XX_RESETB_SIG_LBN 21
+#define	FRF_AB_XX_RESETB_SIG_WIDTH 1
+#define	FRF_AB_XX_RESETA_SIG_LBN 20
+#define	FRF_AB_XX_RESETA_SIG_WIDTH 1
+#define	FRF_AB_XX_RSTXGXSRX_SIG_LBN 18
+#define	FRF_AB_XX_RSTXGXSRX_SIG_WIDTH 1
+#define	FRF_AB_XX_RSTXGXSTX_SIG_LBN 17
+#define	FRF_AB_XX_RSTXGXSTX_SIG_WIDTH 1
+#define	FRF_AB_XX_SD_RST_ACT_LBN 16
+#define	FRF_AB_XX_SD_RST_ACT_WIDTH 1
+#define	FRF_AB_XX_PWRDND_EN_LBN 15
+#define	FRF_AB_XX_PWRDND_EN_WIDTH 1
+#define	FRF_AB_XX_PWRDNC_EN_LBN 14
+#define	FRF_AB_XX_PWRDNC_EN_WIDTH 1
+#define	FRF_AB_XX_PWRDNB_EN_LBN 13
+#define	FRF_AB_XX_PWRDNB_EN_WIDTH 1
+#define	FRF_AB_XX_PWRDNA_EN_LBN 12
+#define	FRF_AB_XX_PWRDNA_EN_WIDTH 1
+#define	FRF_AB_XX_RSTPLLCD_EN_LBN 9
+#define	FRF_AB_XX_RSTPLLCD_EN_WIDTH 1
+#define	FRF_AB_XX_RSTPLLAB_EN_LBN 8
+#define	FRF_AB_XX_RSTPLLAB_EN_WIDTH 1
+#define	FRF_AB_XX_RESETD_EN_LBN 7
+#define	FRF_AB_XX_RESETD_EN_WIDTH 1
+#define	FRF_AB_XX_RESETC_EN_LBN 6
+#define	FRF_AB_XX_RESETC_EN_WIDTH 1
+#define	FRF_AB_XX_RESETB_EN_LBN 5
+#define	FRF_AB_XX_RESETB_EN_WIDTH 1
+#define	FRF_AB_XX_RESETA_EN_LBN 4
+#define	FRF_AB_XX_RESETA_EN_WIDTH 1
+#define	FRF_AB_XX_RSTXGXSRX_EN_LBN 2
+#define	FRF_AB_XX_RSTXGXSRX_EN_WIDTH 1
+#define	FRF_AB_XX_RSTXGXSTX_EN_LBN 1
+#define	FRF_AB_XX_RSTXGXSTX_EN_WIDTH 1
+#define	FRF_AB_XX_RST_XX_EN_LBN 0
+#define	FRF_AB_XX_RST_XX_EN_WIDTH 1
+
+/* XX_SD_CTL_REG: XGXS/XAUI powerdown/reset control register */
+#define	FR_AB_XX_SD_CTL 0x00001310
+#define	FRF_AB_XX_TERMADJ1_LBN 17
+#define	FRF_AB_XX_TERMADJ1_WIDTH 1
+#define	FRF_AB_XX_TERMADJ0_LBN 16
+#define	FRF_AB_XX_TERMADJ0_WIDTH 1
+#define	FRF_AB_XX_HIDRVD_LBN 15
+#define	FRF_AB_XX_HIDRVD_WIDTH 1
+#define	FRF_AB_XX_LODRVD_LBN 14
+#define	FRF_AB_XX_LODRVD_WIDTH 1
+#define	FRF_AB_XX_HIDRVC_LBN 13
+#define	FRF_AB_XX_HIDRVC_WIDTH 1
+#define	FRF_AB_XX_LODRVC_LBN 12
+#define	FRF_AB_XX_LODRVC_WIDTH 1
+#define	FRF_AB_XX_HIDRVB_LBN 11
+#define	FRF_AB_XX_HIDRVB_WIDTH 1
+#define	FRF_AB_XX_LODRVB_LBN 10
+#define	FRF_AB_XX_LODRVB_WIDTH 1
+#define	FRF_AB_XX_HIDRVA_LBN 9
+#define	FRF_AB_XX_HIDRVA_WIDTH 1
+#define	FRF_AB_XX_LODRVA_LBN 8
+#define	FRF_AB_XX_LODRVA_WIDTH 1
+#define	FRF_AB_XX_LPBKD_LBN 3
+#define	FRF_AB_XX_LPBKD_WIDTH 1
+#define	FRF_AB_XX_LPBKC_LBN 2
+#define	FRF_AB_XX_LPBKC_WIDTH 1
+#define	FRF_AB_XX_LPBKB_LBN 1
+#define	FRF_AB_XX_LPBKB_WIDTH 1
+#define	FRF_AB_XX_LPBKA_LBN 0
+#define	FRF_AB_XX_LPBKA_WIDTH 1
+
+/* XX_TXDRV_CTL_REG: XAUI SerDes transmit drive control register */
+#define	FR_AB_XX_TXDRV_CTL 0x00001320
+#define	FRF_AB_XX_DEQD_LBN 28
+#define	FRF_AB_XX_DEQD_WIDTH 4
+#define	FRF_AB_XX_DEQC_LBN 24
+#define	FRF_AB_XX_DEQC_WIDTH 4
+#define	FRF_AB_XX_DEQB_LBN 20
+#define	FRF_AB_XX_DEQB_WIDTH 4
+#define	FRF_AB_XX_DEQA_LBN 16
+#define	FRF_AB_XX_DEQA_WIDTH 4
+#define	FRF_AB_XX_DTXD_LBN 12
+#define	FRF_AB_XX_DTXD_WIDTH 4
+#define	FRF_AB_XX_DTXC_LBN 8
+#define	FRF_AB_XX_DTXC_WIDTH 4
+#define	FRF_AB_XX_DTXB_LBN 4
+#define	FRF_AB_XX_DTXB_WIDTH 4
+#define	FRF_AB_XX_DTXA_LBN 0
+#define	FRF_AB_XX_DTXA_WIDTH 4
+
+/* XX_PRBS_CTL_REG: documentation to be written for sum_XX_PRBS_CTL_REG */
+#define	FR_AB_XX_PRBS_CTL 0x00001330
+#define	FRF_AB_XX_CH3_RX_PRBS_SEL_LBN 30
+#define	FRF_AB_XX_CH3_RX_PRBS_SEL_WIDTH 2
+#define	FRF_AB_XX_CH3_RX_PRBS_INV_LBN 29
+#define	FRF_AB_XX_CH3_RX_PRBS_INV_WIDTH 1
+#define	FRF_AB_XX_CH3_RX_PRBS_CHKEN_LBN 28
+#define	FRF_AB_XX_CH3_RX_PRBS_CHKEN_WIDTH 1
+#define	FRF_AB_XX_CH2_RX_PRBS_SEL_LBN 26
+#define	FRF_AB_XX_CH2_RX_PRBS_SEL_WIDTH 2
+#define	FRF_AB_XX_CH2_RX_PRBS_INV_LBN 25
+#define	FRF_AB_XX_CH2_RX_PRBS_INV_WIDTH 1
+#define	FRF_AB_XX_CH2_RX_PRBS_CHKEN_LBN 24
+#define	FRF_AB_XX_CH2_RX_PRBS_CHKEN_WIDTH 1
+#define	FRF_AB_XX_CH1_RX_PRBS_SEL_LBN 22
+#define	FRF_AB_XX_CH1_RX_PRBS_SEL_WIDTH 2
+#define	FRF_AB_XX_CH1_RX_PRBS_INV_LBN 21
+#define	FRF_AB_XX_CH1_RX_PRBS_INV_WIDTH 1
+#define	FRF_AB_XX_CH1_RX_PRBS_CHKEN_LBN 20
+#define	FRF_AB_XX_CH1_RX_PRBS_CHKEN_WIDTH 1
+#define	FRF_AB_XX_CH0_RX_PRBS_SEL_LBN 18
+#define	FRF_AB_XX_CH0_RX_PRBS_SEL_WIDTH 2
+#define	FRF_AB_XX_CH0_RX_PRBS_INV_LBN 17
+#define	FRF_AB_XX_CH0_RX_PRBS_INV_WIDTH 1
+#define	FRF_AB_XX_CH0_RX_PRBS_CHKEN_LBN 16
+#define	FRF_AB_XX_CH0_RX_PRBS_CHKEN_WIDTH 1
+#define	FRF_AB_XX_CH3_TX_PRBS_SEL_LBN 14
+#define	FRF_AB_XX_CH3_TX_PRBS_SEL_WIDTH 2
+#define	FRF_AB_XX_CH3_TX_PRBS_INV_LBN 13
+#define	FRF_AB_XX_CH3_TX_PRBS_INV_WIDTH 1
+#define	FRF_AB_XX_CH3_TX_PRBS_CHKEN_LBN 12
+#define	FRF_AB_XX_CH3_TX_PRBS_CHKEN_WIDTH 1
+#define	FRF_AB_XX_CH2_TX_PRBS_SEL_LBN 10
+#define	FRF_AB_XX_CH2_TX_PRBS_SEL_WIDTH 2
+#define	FRF_AB_XX_CH2_TX_PRBS_INV_LBN 9
+#define	FRF_AB_XX_CH2_TX_PRBS_INV_WIDTH 1
+#define	FRF_AB_XX_CH2_TX_PRBS_CHKEN_LBN 8
+#define	FRF_AB_XX_CH2_TX_PRBS_CHKEN_WIDTH 1
+#define	FRF_AB_XX_CH1_TX_PRBS_SEL_LBN 6
+#define	FRF_AB_XX_CH1_TX_PRBS_SEL_WIDTH 2
+#define	FRF_AB_XX_CH1_TX_PRBS_INV_LBN 5
+#define	FRF_AB_XX_CH1_TX_PRBS_INV_WIDTH 1
+#define	FRF_AB_XX_CH1_TX_PRBS_CHKEN_LBN 4
+#define	FRF_AB_XX_CH1_TX_PRBS_CHKEN_WIDTH 1
+#define	FRF_AB_XX_CH0_TX_PRBS_SEL_LBN 2
+#define	FRF_AB_XX_CH0_TX_PRBS_SEL_WIDTH 2
+#define	FRF_AB_XX_CH0_TX_PRBS_INV_LBN 1
+#define	FRF_AB_XX_CH0_TX_PRBS_INV_WIDTH 1
+#define	FRF_AB_XX_CH0_TX_PRBS_CHKEN_LBN 0
+#define	FRF_AB_XX_CH0_TX_PRBS_CHKEN_WIDTH 1
+
+/* XX_PRBS_CHK_REG: documentation to be written for sum_XX_PRBS_CHK_REG */
+#define	FR_AB_XX_PRBS_CHK 0x00001340
+#define	FRF_AB_XX_REV_LB_EN_LBN 16
+#define	FRF_AB_XX_REV_LB_EN_WIDTH 1
+#define	FRF_AB_XX_CH3_DEG_DET_LBN 15
+#define	FRF_AB_XX_CH3_DEG_DET_WIDTH 1
+#define	FRF_AB_XX_CH3_LFSR_LOCK_IND_LBN 14
+#define	FRF_AB_XX_CH3_LFSR_LOCK_IND_WIDTH 1
+#define	FRF_AB_XX_CH3_PRBS_FRUN_LBN 13
+#define	FRF_AB_XX_CH3_PRBS_FRUN_WIDTH 1
+#define	FRF_AB_XX_CH3_ERR_CHK_LBN 12
+#define	FRF_AB_XX_CH3_ERR_CHK_WIDTH 1
+#define	FRF_AB_XX_CH2_DEG_DET_LBN 11
+#define	FRF_AB_XX_CH2_DEG_DET_WIDTH 1
+#define	FRF_AB_XX_CH2_LFSR_LOCK_IND_LBN 10
+#define	FRF_AB_XX_CH2_LFSR_LOCK_IND_WIDTH 1
+#define	FRF_AB_XX_CH2_PRBS_FRUN_LBN 9
+#define	FRF_AB_XX_CH2_PRBS_FRUN_WIDTH 1
+#define	FRF_AB_XX_CH2_ERR_CHK_LBN 8
+#define	FRF_AB_XX_CH2_ERR_CHK_WIDTH 1
+#define	FRF_AB_XX_CH1_DEG_DET_LBN 7
+#define	FRF_AB_XX_CH1_DEG_DET_WIDTH 1
+#define	FRF_AB_XX_CH1_LFSR_LOCK_IND_LBN 6
+#define	FRF_AB_XX_CH1_LFSR_LOCK_IND_WIDTH 1
+#define	FRF_AB_XX_CH1_PRBS_FRUN_LBN 5
+#define	FRF_AB_XX_CH1_PRBS_FRUN_WIDTH 1
+#define	FRF_AB_XX_CH1_ERR_CHK_LBN 4
+#define	FRF_AB_XX_CH1_ERR_CHK_WIDTH 1
+#define	FRF_AB_XX_CH0_DEG_DET_LBN 3
+#define	FRF_AB_XX_CH0_DEG_DET_WIDTH 1
+#define	FRF_AB_XX_CH0_LFSR_LOCK_IND_LBN 2
+#define	FRF_AB_XX_CH0_LFSR_LOCK_IND_WIDTH 1
+#define	FRF_AB_XX_CH0_PRBS_FRUN_LBN 1
+#define	FRF_AB_XX_CH0_PRBS_FRUN_WIDTH 1
+#define	FRF_AB_XX_CH0_ERR_CHK_LBN 0
+#define	FRF_AB_XX_CH0_ERR_CHK_WIDTH 1
+
+/* XX_PRBS_ERR_REG: documentation to be written for sum_XX_PRBS_ERR_REG */
+#define	FR_AB_XX_PRBS_ERR 0x00001350
+#define	FRF_AB_XX_CH3_PRBS_ERR_CNT_LBN 24
+#define	FRF_AB_XX_CH3_PRBS_ERR_CNT_WIDTH 8
+#define	FRF_AB_XX_CH2_PRBS_ERR_CNT_LBN 16
+#define	FRF_AB_XX_CH2_PRBS_ERR_CNT_WIDTH 8
+#define	FRF_AB_XX_CH1_PRBS_ERR_CNT_LBN 8
+#define	FRF_AB_XX_CH1_PRBS_ERR_CNT_WIDTH 8
+#define	FRF_AB_XX_CH0_PRBS_ERR_CNT_LBN 0
+#define	FRF_AB_XX_CH0_PRBS_ERR_CNT_WIDTH 8
+
+/* XX_CORE_STAT_REG: XAUI XGXS core status register */
+#define	FR_AB_XX_CORE_STAT 0x00001360
+#define	FRF_AB_XX_FORCE_SIG3_LBN 31
+#define	FRF_AB_XX_FORCE_SIG3_WIDTH 1
+#define	FRF_AB_XX_FORCE_SIG3_VAL_LBN 30
+#define	FRF_AB_XX_FORCE_SIG3_VAL_WIDTH 1
+#define	FRF_AB_XX_FORCE_SIG2_LBN 29
+#define	FRF_AB_XX_FORCE_SIG2_WIDTH 1
+#define	FRF_AB_XX_FORCE_SIG2_VAL_LBN 28
+#define	FRF_AB_XX_FORCE_SIG2_VAL_WIDTH 1
+#define	FRF_AB_XX_FORCE_SIG1_LBN 27
+#define	FRF_AB_XX_FORCE_SIG1_WIDTH 1
+#define	FRF_AB_XX_FORCE_SIG1_VAL_LBN 26
+#define	FRF_AB_XX_FORCE_SIG1_VAL_WIDTH 1
+#define	FRF_AB_XX_FORCE_SIG0_LBN 25
+#define	FRF_AB_XX_FORCE_SIG0_WIDTH 1
+#define	FRF_AB_XX_FORCE_SIG0_VAL_LBN 24
+#define	FRF_AB_XX_FORCE_SIG0_VAL_WIDTH 1
+#define	FRF_AB_XX_XGXS_LB_EN_LBN 23
+#define	FRF_AB_XX_XGXS_LB_EN_WIDTH 1
+#define	FRF_AB_XX_XGMII_LB_EN_LBN 22
+#define	FRF_AB_XX_XGMII_LB_EN_WIDTH 1
+#define	FRF_AB_XX_MATCH_FAULT_LBN 21
+#define	FRF_AB_XX_MATCH_FAULT_WIDTH 1
+#define	FRF_AB_XX_ALIGN_DONE_LBN 20
+#define	FRF_AB_XX_ALIGN_DONE_WIDTH 1
+#define	FRF_AB_XX_SYNC_STAT3_LBN 19
+#define	FRF_AB_XX_SYNC_STAT3_WIDTH 1
+#define	FRF_AB_XX_SYNC_STAT2_LBN 18
+#define	FRF_AB_XX_SYNC_STAT2_WIDTH 1
+#define	FRF_AB_XX_SYNC_STAT1_LBN 17
+#define	FRF_AB_XX_SYNC_STAT1_WIDTH 1
+#define	FRF_AB_XX_SYNC_STAT0_LBN 16
+#define	FRF_AB_XX_SYNC_STAT0_WIDTH 1
+#define	FRF_AB_XX_COMMA_DET_CH3_LBN 15
+#define	FRF_AB_XX_COMMA_DET_CH3_WIDTH 1
+#define	FRF_AB_XX_COMMA_DET_CH2_LBN 14
+#define	FRF_AB_XX_COMMA_DET_CH2_WIDTH 1
+#define	FRF_AB_XX_COMMA_DET_CH1_LBN 13
+#define	FRF_AB_XX_COMMA_DET_CH1_WIDTH 1
+#define	FRF_AB_XX_COMMA_DET_CH0_LBN 12
+#define	FRF_AB_XX_COMMA_DET_CH0_WIDTH 1
+#define	FRF_AB_XX_CGRP_ALIGN_CH3_LBN 11
+#define	FRF_AB_XX_CGRP_ALIGN_CH3_WIDTH 1
+#define	FRF_AB_XX_CGRP_ALIGN_CH2_LBN 10
+#define	FRF_AB_XX_CGRP_ALIGN_CH2_WIDTH 1
+#define	FRF_AB_XX_CGRP_ALIGN_CH1_LBN 9
+#define	FRF_AB_XX_CGRP_ALIGN_CH1_WIDTH 1
+#define	FRF_AB_XX_CGRP_ALIGN_CH0_LBN 8
+#define	FRF_AB_XX_CGRP_ALIGN_CH0_WIDTH 1
+#define	FRF_AB_XX_CHAR_ERR_CH3_LBN 7
+#define	FRF_AB_XX_CHAR_ERR_CH3_WIDTH 1
+#define	FRF_AB_XX_CHAR_ERR_CH2_LBN 6
+#define	FRF_AB_XX_CHAR_ERR_CH2_WIDTH 1
+#define	FRF_AB_XX_CHAR_ERR_CH1_LBN 5
+#define	FRF_AB_XX_CHAR_ERR_CH1_WIDTH 1
+#define	FRF_AB_XX_CHAR_ERR_CH0_LBN 4
+#define	FRF_AB_XX_CHAR_ERR_CH0_WIDTH 1
+#define	FRF_AB_XX_DISPERR_CH3_LBN 3
+#define	FRF_AB_XX_DISPERR_CH3_WIDTH 1
+#define	FRF_AB_XX_DISPERR_CH2_LBN 2
+#define	FRF_AB_XX_DISPERR_CH2_WIDTH 1
+#define	FRF_AB_XX_DISPERR_CH1_LBN 1
+#define	FRF_AB_XX_DISPERR_CH1_WIDTH 1
+#define	FRF_AB_XX_DISPERR_CH0_LBN 0
+#define	FRF_AB_XX_DISPERR_CH0_WIDTH 1
+
+/* RX_DESC_PTR_TBL_KER: Receive descriptor pointer table */
+#define	FR_AA_RX_DESC_PTR_TBL_KER 0x00011800
+#define	FR_AA_RX_DESC_PTR_TBL_KER_STEP 16
+#define	FR_AA_RX_DESC_PTR_TBL_KER_ROWS 4
+/* RX_DESC_PTR_TBL: Receive descriptor pointer table */
+#define	FR_BZ_RX_DESC_PTR_TBL 0x00f40000
+#define	FR_BZ_RX_DESC_PTR_TBL_STEP 16
+#define	FR_BB_RX_DESC_PTR_TBL_ROWS 4096
+#define	FR_CZ_RX_DESC_PTR_TBL_ROWS 1024
+#define	FRF_CZ_RX_HDR_SPLIT_LBN 90
+#define	FRF_CZ_RX_HDR_SPLIT_WIDTH 1
+#define	FRF_AA_RX_RESET_LBN 89
+#define	FRF_AA_RX_RESET_WIDTH 1
+#define	FRF_AZ_RX_ISCSI_DDIG_EN_LBN 88
+#define	FRF_AZ_RX_ISCSI_DDIG_EN_WIDTH 1
+#define	FRF_AZ_RX_ISCSI_HDIG_EN_LBN 87
+#define	FRF_AZ_RX_ISCSI_HDIG_EN_WIDTH 1
+#define	FRF_AZ_RX_DESC_PREF_ACT_LBN 86
+#define	FRF_AZ_RX_DESC_PREF_ACT_WIDTH 1
+#define	FRF_AZ_RX_DC_HW_RPTR_LBN 80
+#define	FRF_AZ_RX_DC_HW_RPTR_WIDTH 6
+#define	FRF_AZ_RX_DESCQ_HW_RPTR_LBN 68
+#define	FRF_AZ_RX_DESCQ_HW_RPTR_WIDTH 12
+#define	FRF_AZ_RX_DESCQ_SW_WPTR_LBN 56
+#define	FRF_AZ_RX_DESCQ_SW_WPTR_WIDTH 12
+#define	FRF_AZ_RX_DESCQ_BUF_BASE_ID_LBN 36
+#define	FRF_AZ_RX_DESCQ_BUF_BASE_ID_WIDTH 20
+#define	FRF_AZ_RX_DESCQ_EVQ_ID_LBN 24
+#define	FRF_AZ_RX_DESCQ_EVQ_ID_WIDTH 12
+#define	FRF_AZ_RX_DESCQ_OWNER_ID_LBN 10
+#define	FRF_AZ_RX_DESCQ_OWNER_ID_WIDTH 14
+#define	FRF_AZ_RX_DESCQ_LABEL_LBN 5
+#define	FRF_AZ_RX_DESCQ_LABEL_WIDTH 5
+#define	FRF_AZ_RX_DESCQ_SIZE_LBN 3
+#define	FRF_AZ_RX_DESCQ_SIZE_WIDTH 2
+#define	FFE_AZ_RX_DESCQ_SIZE_4K 3
+#define	FFE_AZ_RX_DESCQ_SIZE_2K 2
+#define	FFE_AZ_RX_DESCQ_SIZE_1K 1
+#define	FFE_AZ_RX_DESCQ_SIZE_512 0
+#define	FRF_AZ_RX_DESCQ_TYPE_LBN 2
+#define	FRF_AZ_RX_DESCQ_TYPE_WIDTH 1
+#define	FRF_AZ_RX_DESCQ_JUMBO_LBN 1
+#define	FRF_AZ_RX_DESCQ_JUMBO_WIDTH 1
+#define	FRF_AZ_RX_DESCQ_EN_LBN 0
+#define	FRF_AZ_RX_DESCQ_EN_WIDTH 1
+
+/* TX_DESC_PTR_TBL_KER: Transmit descriptor pointer */
+#define	FR_AA_TX_DESC_PTR_TBL_KER 0x00011900
+#define	FR_AA_TX_DESC_PTR_TBL_KER_STEP 16
+#define	FR_AA_TX_DESC_PTR_TBL_KER_ROWS 8
+/* TX_DESC_PTR_TBL: Transmit descriptor pointer */
+#define	FR_BZ_TX_DESC_PTR_TBL 0x00f50000
+#define	FR_BZ_TX_DESC_PTR_TBL_STEP 16
+#define	FR_BB_TX_DESC_PTR_TBL_ROWS 4096
+#define	FR_CZ_TX_DESC_PTR_TBL_ROWS 1024
+#define	FRF_CZ_TX_DPT_Q_MASK_WIDTH_LBN 94
+#define	FRF_CZ_TX_DPT_Q_MASK_WIDTH_WIDTH 2
+#define	FRF_CZ_TX_DPT_ETH_FILT_EN_LBN 93
+#define	FRF_CZ_TX_DPT_ETH_FILT_EN_WIDTH 1
+#define	FRF_CZ_TX_DPT_IP_FILT_EN_LBN 92
+#define	FRF_CZ_TX_DPT_IP_FILT_EN_WIDTH 1
+#define	FRF_BZ_TX_NON_IP_DROP_DIS_LBN 91
+#define	FRF_BZ_TX_NON_IP_DROP_DIS_WIDTH 1
+#define	FRF_BZ_TX_IP_CHKSM_DIS_LBN 90
+#define	FRF_BZ_TX_IP_CHKSM_DIS_WIDTH 1
+#define	FRF_BZ_TX_TCP_CHKSM_DIS_LBN 89
+#define	FRF_BZ_TX_TCP_CHKSM_DIS_WIDTH 1
+#define	FRF_AZ_TX_DESCQ_EN_LBN 88
+#define	FRF_AZ_TX_DESCQ_EN_WIDTH 1
+#define	FRF_AZ_TX_ISCSI_DDIG_EN_LBN 87
+#define	FRF_AZ_TX_ISCSI_DDIG_EN_WIDTH 1
+#define	FRF_AZ_TX_ISCSI_HDIG_EN_LBN 86
+#define	FRF_AZ_TX_ISCSI_HDIG_EN_WIDTH 1
+#define	FRF_AZ_TX_DC_HW_RPTR_LBN 80
+#define	FRF_AZ_TX_DC_HW_RPTR_WIDTH 6
+#define	FRF_AZ_TX_DESCQ_HW_RPTR_LBN 68
+#define	FRF_AZ_TX_DESCQ_HW_RPTR_WIDTH 12
+#define	FRF_AZ_TX_DESCQ_SW_WPTR_LBN 56
+#define	FRF_AZ_TX_DESCQ_SW_WPTR_WIDTH 12
+#define	FRF_AZ_TX_DESCQ_BUF_BASE_ID_LBN 36
+#define	FRF_AZ_TX_DESCQ_BUF_BASE_ID_WIDTH 20
+#define	FRF_AZ_TX_DESCQ_EVQ_ID_LBN 24
+#define	FRF_AZ_TX_DESCQ_EVQ_ID_WIDTH 12
+#define	FRF_AZ_TX_DESCQ_OWNER_ID_LBN 10
+#define	FRF_AZ_TX_DESCQ_OWNER_ID_WIDTH 14
+#define	FRF_AZ_TX_DESCQ_LABEL_LBN 5
+#define	FRF_AZ_TX_DESCQ_LABEL_WIDTH 5
+#define	FRF_AZ_TX_DESCQ_SIZE_LBN 3
+#define	FRF_AZ_TX_DESCQ_SIZE_WIDTH 2
+#define	FFE_AZ_TX_DESCQ_SIZE_4K 3
+#define	FFE_AZ_TX_DESCQ_SIZE_2K 2
+#define	FFE_AZ_TX_DESCQ_SIZE_1K 1
+#define	FFE_AZ_TX_DESCQ_SIZE_512 0
+#define	FRF_AZ_TX_DESCQ_TYPE_LBN 1
+#define	FRF_AZ_TX_DESCQ_TYPE_WIDTH 2
+#define	FRF_AZ_TX_DESCQ_FLUSH_LBN 0
+#define	FRF_AZ_TX_DESCQ_FLUSH_WIDTH 1
+
+/* EVQ_PTR_TBL_KER: Event queue pointer table */
+#define	FR_AA_EVQ_PTR_TBL_KER 0x00011a00
+#define	FR_AA_EVQ_PTR_TBL_KER_STEP 16
+#define	FR_AA_EVQ_PTR_TBL_KER_ROWS 4
+/* EVQ_PTR_TBL: Event queue pointer table */
+#define	FR_BZ_EVQ_PTR_TBL 0x00f60000
+#define	FR_BZ_EVQ_PTR_TBL_STEP 16
+#define	FR_CZ_EVQ_PTR_TBL_ROWS 1024
+#define	FR_BB_EVQ_PTR_TBL_ROWS 4096
+#define	FRF_BZ_EVQ_RPTR_IGN_LBN 40
+#define	FRF_BZ_EVQ_RPTR_IGN_WIDTH 1
+#define	FRF_AB_EVQ_WKUP_OR_INT_EN_LBN 39
+#define	FRF_AB_EVQ_WKUP_OR_INT_EN_WIDTH 1
+#define	FRF_CZ_EVQ_DOS_PROTECT_EN_LBN 39
+#define	FRF_CZ_EVQ_DOS_PROTECT_EN_WIDTH 1
+#define	FRF_AZ_EVQ_NXT_WPTR_LBN 24
+#define	FRF_AZ_EVQ_NXT_WPTR_WIDTH 15
+#define	FRF_AZ_EVQ_EN_LBN 23
+#define	FRF_AZ_EVQ_EN_WIDTH 1
+#define	FRF_AZ_EVQ_SIZE_LBN 20
+#define	FRF_AZ_EVQ_SIZE_WIDTH 3
+#define	FFE_AZ_EVQ_SIZE_32K 6
+#define	FFE_AZ_EVQ_SIZE_16K 5
+#define	FFE_AZ_EVQ_SIZE_8K 4
+#define	FFE_AZ_EVQ_SIZE_4K 3
+#define	FFE_AZ_EVQ_SIZE_2K 2
+#define	FFE_AZ_EVQ_SIZE_1K 1
+#define	FFE_AZ_EVQ_SIZE_512 0
+#define	FRF_AZ_EVQ_BUF_BASE_ID_LBN 0
+#define	FRF_AZ_EVQ_BUF_BASE_ID_WIDTH 20
+
+/* BUF_HALF_TBL_KER: Buffer table in half buffer table mode direct access by driver */
+#define	FR_AA_BUF_HALF_TBL_KER 0x00018000
+#define	FR_AA_BUF_HALF_TBL_KER_STEP 8
+#define	FR_AA_BUF_HALF_TBL_KER_ROWS 4096
+/* BUF_HALF_TBL: Buffer table in half buffer table mode direct access by driver */
+#define	FR_BZ_BUF_HALF_TBL 0x00800000
+#define	FR_BZ_BUF_HALF_TBL_STEP 8
+#define	FR_CZ_BUF_HALF_TBL_ROWS 147456
+#define	FR_BB_BUF_HALF_TBL_ROWS 524288
+#define	FRF_AZ_BUF_ADR_HBUF_ODD_LBN 44
+#define	FRF_AZ_BUF_ADR_HBUF_ODD_WIDTH 20
+#define	FRF_AZ_BUF_OWNER_ID_HBUF_ODD_LBN 32
+#define	FRF_AZ_BUF_OWNER_ID_HBUF_ODD_WIDTH 12
+#define	FRF_AZ_BUF_ADR_HBUF_EVEN_LBN 12
+#define	FRF_AZ_BUF_ADR_HBUF_EVEN_WIDTH 20
+#define	FRF_AZ_BUF_OWNER_ID_HBUF_EVEN_LBN 0
+#define	FRF_AZ_BUF_OWNER_ID_HBUF_EVEN_WIDTH 12
+
+/* BUF_FULL_TBL_KER: Buffer table in full buffer table mode direct access by driver */
+#define	FR_AA_BUF_FULL_TBL_KER 0x00018000
+#define	FR_AA_BUF_FULL_TBL_KER_STEP 8
+#define	FR_AA_BUF_FULL_TBL_KER_ROWS 4096
+/* BUF_FULL_TBL: Buffer table in full buffer table mode direct access by driver */
+#define	FR_BZ_BUF_FULL_TBL 0x00800000
+#define	FR_BZ_BUF_FULL_TBL_STEP 8
+#define	FR_CZ_BUF_FULL_TBL_ROWS 147456
+#define	FR_BB_BUF_FULL_TBL_ROWS 917504
+#define	FRF_AZ_BUF_FULL_UNUSED_LBN 51
+#define	FRF_AZ_BUF_FULL_UNUSED_WIDTH 13
+#define	FRF_AZ_IP_DAT_BUF_SIZE_LBN 50
+#define	FRF_AZ_IP_DAT_BUF_SIZE_WIDTH 1
+#define	FRF_AZ_BUF_ADR_REGION_LBN 48
+#define	FRF_AZ_BUF_ADR_REGION_WIDTH 2
+#define	FFE_AZ_BUF_ADR_REGN3 3
+#define	FFE_AZ_BUF_ADR_REGN2 2
+#define	FFE_AZ_BUF_ADR_REGN1 1
+#define	FFE_AZ_BUF_ADR_REGN0 0
+#define	FRF_AZ_BUF_ADR_FBUF_LBN 14
+#define	FRF_AZ_BUF_ADR_FBUF_WIDTH 34
+#define	FRF_AZ_BUF_OWNER_ID_FBUF_LBN 0
+#define	FRF_AZ_BUF_OWNER_ID_FBUF_WIDTH 14
+
+/* RX_FILTER_TBL0: TCP/IPv4 Receive filter table */
+#define	FR_BZ_RX_FILTER_TBL0 0x00f00000
+#define	FR_BZ_RX_FILTER_TBL0_STEP 32
+#define	FR_BZ_RX_FILTER_TBL0_ROWS 8192
+/* RX_FILTER_TBL1: TCP/IPv4 Receive filter table */
+#define	FR_BB_RX_FILTER_TBL1 0x00f00010
+#define	FR_BB_RX_FILTER_TBL1_STEP 32
+#define	FR_BB_RX_FILTER_TBL1_ROWS 8192
+#define	FRF_BZ_RSS_EN_LBN 110
+#define	FRF_BZ_RSS_EN_WIDTH 1
+#define	FRF_BZ_SCATTER_EN_LBN 109
+#define	FRF_BZ_SCATTER_EN_WIDTH 1
+#define	FRF_BZ_TCP_UDP_LBN 108
+#define	FRF_BZ_TCP_UDP_WIDTH 1
+#define	FRF_BZ_RXQ_ID_LBN 96
+#define	FRF_BZ_RXQ_ID_WIDTH 12
+#define	FRF_BZ_DEST_IP_LBN 64
+#define	FRF_BZ_DEST_IP_WIDTH 32
+#define	FRF_BZ_DEST_PORT_TCP_LBN 48
+#define	FRF_BZ_DEST_PORT_TCP_WIDTH 16
+#define	FRF_BZ_SRC_IP_LBN 16
+#define	FRF_BZ_SRC_IP_WIDTH 32
+#define	FRF_BZ_SRC_TCP_DEST_UDP_LBN 0
+#define	FRF_BZ_SRC_TCP_DEST_UDP_WIDTH 16
+
+/* RX_MAC_FILTER_TBL0: Receive Ethernet filter table */
+#define	FR_CZ_RX_MAC_FILTER_TBL0 0x00f00010
+#define	FR_CZ_RX_MAC_FILTER_TBL0_STEP 32
+#define	FR_CZ_RX_MAC_FILTER_TBL0_ROWS 512
+#define	FRF_CZ_RMFT_RSS_EN_LBN 75
+#define	FRF_CZ_RMFT_RSS_EN_WIDTH 1
+#define	FRF_CZ_RMFT_SCATTER_EN_LBN 74
+#define	FRF_CZ_RMFT_SCATTER_EN_WIDTH 1
+#define	FRF_CZ_RMFT_IP_OVERRIDE_LBN 73
+#define	FRF_CZ_RMFT_IP_OVERRIDE_WIDTH 1
+#define	FRF_CZ_RMFT_RXQ_ID_LBN 61
+#define	FRF_CZ_RMFT_RXQ_ID_WIDTH 12
+#define	FRF_CZ_RMFT_WILDCARD_MATCH_LBN 60
+#define	FRF_CZ_RMFT_WILDCARD_MATCH_WIDTH 1
+#define	FRF_CZ_RMFT_DEST_MAC_LBN 16
+#define	FRF_CZ_RMFT_DEST_MAC_WIDTH 44
+#define	FRF_CZ_RMFT_VLAN_ID_LBN 0
+#define	FRF_CZ_RMFT_VLAN_ID_WIDTH 12
+
+/* TIMER_TBL: Timer table */
+#define	FR_BZ_TIMER_TBL 0x00f70000
+#define	FR_BZ_TIMER_TBL_STEP 16
+#define	FR_CZ_TIMER_TBL_ROWS 1024
+#define	FR_BB_TIMER_TBL_ROWS 4096
+#define	FRF_CZ_TIMER_Q_EN_LBN 33
+#define	FRF_CZ_TIMER_Q_EN_WIDTH 1
+#define	FRF_CZ_INT_ARMD_LBN 32
+#define	FRF_CZ_INT_ARMD_WIDTH 1
+#define	FRF_CZ_INT_PEND_LBN 31
+#define	FRF_CZ_INT_PEND_WIDTH 1
+#define	FRF_CZ_HOST_NOTIFY_MODE_LBN 30
+#define	FRF_CZ_HOST_NOTIFY_MODE_WIDTH 1
+#define	FRF_CZ_RELOAD_TIMER_VAL_LBN 16
+#define	FRF_CZ_RELOAD_TIMER_VAL_WIDTH 14
+#define	FRF_CZ_TIMER_MODE_LBN 14
+#define	FRF_CZ_TIMER_MODE_WIDTH 2
+#define	FFE_CZ_TIMER_MODE_INT_HLDOFF 3
+#define	FFE_CZ_TIMER_MODE_TRIG_START 2
+#define	FFE_CZ_TIMER_MODE_IMMED_START 1
+#define	FFE_CZ_TIMER_MODE_DIS 0
+#define	FRF_BB_TIMER_MODE_LBN 12
+#define	FRF_BB_TIMER_MODE_WIDTH 2
+#define	FFE_BB_TIMER_MODE_INT_HLDOFF 2
+#define	FFE_BB_TIMER_MODE_TRIG_START 2
+#define	FFE_BB_TIMER_MODE_IMMED_START 1
+#define	FFE_BB_TIMER_MODE_DIS 0
+#define	FRF_CZ_TIMER_VAL_LBN 0
+#define	FRF_CZ_TIMER_VAL_WIDTH 14
+#define	FRF_BB_TIMER_VAL_LBN 0
+#define	FRF_BB_TIMER_VAL_WIDTH 12
+
+/* TX_PACE_TBL: Transmit pacing table */
+#define	FR_BZ_TX_PACE_TBL 0x00f80000
+#define	FR_BZ_TX_PACE_TBL_STEP 16
+#define	FR_CZ_TX_PACE_TBL_ROWS 1024
+#define	FR_BB_TX_PACE_TBL_ROWS 4096
+#define	FRF_BZ_TX_PACE_LBN 0
+#define	FRF_BZ_TX_PACE_WIDTH 5
+
+/* RX_INDIRECTION_TBL: RX Indirection Table */
+#define	FR_BZ_RX_INDIRECTION_TBL 0x00fb0000
+#define	FR_BZ_RX_INDIRECTION_TBL_STEP 16
+#define	FR_BZ_RX_INDIRECTION_TBL_ROWS 128
+#define	FRF_BZ_IT_QUEUE_LBN 0
+#define	FRF_BZ_IT_QUEUE_WIDTH 6
+
+/* TX_FILTER_TBL0: TCP/IPv4 Transmit filter table */
+#define	FR_CZ_TX_FILTER_TBL0 0x00fc0000
+#define	FR_CZ_TX_FILTER_TBL0_STEP 16
+#define	FR_CZ_TX_FILTER_TBL0_ROWS 8192
+#define	FRF_CZ_TIFT_TCP_UDP_LBN 108
+#define	FRF_CZ_TIFT_TCP_UDP_WIDTH 1
+#define	FRF_CZ_TIFT_TXQ_ID_LBN 96
+#define	FRF_CZ_TIFT_TXQ_ID_WIDTH 12
+#define	FRF_CZ_TIFT_DEST_IP_LBN 64
+#define	FRF_CZ_TIFT_DEST_IP_WIDTH 32
+#define	FRF_CZ_TIFT_DEST_PORT_TCP_LBN 48
+#define	FRF_CZ_TIFT_DEST_PORT_TCP_WIDTH 16
+#define	FRF_CZ_TIFT_SRC_IP_LBN 16
+#define	FRF_CZ_TIFT_SRC_IP_WIDTH 32
+#define	FRF_CZ_TIFT_SRC_TCP_DEST_UDP_LBN 0
+#define	FRF_CZ_TIFT_SRC_TCP_DEST_UDP_WIDTH 16
+
+/* TX_MAC_FILTER_TBL0: Transmit Ethernet filter table */
+#define	FR_CZ_TX_MAC_FILTER_TBL0 0x00fe0000
+#define	FR_CZ_TX_MAC_FILTER_TBL0_STEP 16
+#define	FR_CZ_TX_MAC_FILTER_TBL0_ROWS 512
+#define	FRF_CZ_TMFT_TXQ_ID_LBN 61
+#define	FRF_CZ_TMFT_TXQ_ID_WIDTH 12
+#define	FRF_CZ_TMFT_WILDCARD_MATCH_LBN 60
+#define	FRF_CZ_TMFT_WILDCARD_MATCH_WIDTH 1
+#define	FRF_CZ_TMFT_SRC_MAC_LBN 16
+#define	FRF_CZ_TMFT_SRC_MAC_WIDTH 44
+#define	FRF_CZ_TMFT_VLAN_ID_LBN 0
+#define	FRF_CZ_TMFT_VLAN_ID_WIDTH 12
+
+/* MC_TREG_SMEM: MC Shared Memory */
+#define	FR_CZ_MC_TREG_SMEM 0x00ff0000
+#define	FR_CZ_MC_TREG_SMEM_STEP 4
+#define	FR_CZ_MC_TREG_SMEM_ROWS 512
+#define	FRF_CZ_MC_TREG_SMEM_ROW_LBN 0
+#define	FRF_CZ_MC_TREG_SMEM_ROW_WIDTH 32
+
+/* MSIX_VECTOR_TABLE: MSIX Vector Table */
+#define	FR_BB_MSIX_VECTOR_TABLE 0x00ff0000
+#define	FR_BZ_MSIX_VECTOR_TABLE_STEP 16
+#define	FR_BB_MSIX_VECTOR_TABLE_ROWS 64
+/* MSIX_VECTOR_TABLE: MSIX Vector Table */
+#define	FR_CZ_MSIX_VECTOR_TABLE 0x00000000
+/* FR_BZ_MSIX_VECTOR_TABLE_STEP 16 */
+#define	FR_CZ_MSIX_VECTOR_TABLE_ROWS 1024
+#define	FRF_BZ_MSIX_VECTOR_RESERVED_LBN 97
+#define	FRF_BZ_MSIX_VECTOR_RESERVED_WIDTH 31
+#define	FRF_BZ_MSIX_VECTOR_MASK_LBN 96
+#define	FRF_BZ_MSIX_VECTOR_MASK_WIDTH 1
+#define	FRF_BZ_MSIX_MESSAGE_DATA_LBN 64
+#define	FRF_BZ_MSIX_MESSAGE_DATA_WIDTH 32
+#define	FRF_BZ_MSIX_MESSAGE_ADDRESS_HI_LBN 32
+#define	FRF_BZ_MSIX_MESSAGE_ADDRESS_HI_WIDTH 32
+#define	FRF_BZ_MSIX_MESSAGE_ADDRESS_LO_LBN 0
+#define	FRF_BZ_MSIX_MESSAGE_ADDRESS_LO_WIDTH 32
+
+/* MSIX_PBA_TABLE: MSIX Pending Bit Array */
+#define	FR_BB_MSIX_PBA_TABLE 0x00ff2000
+#define	FR_BZ_MSIX_PBA_TABLE_STEP 4
+#define	FR_BB_MSIX_PBA_TABLE_ROWS 2
+/* MSIX_PBA_TABLE: MSIX Pending Bit Array */
+#define	FR_CZ_MSIX_PBA_TABLE 0x00008000
+/* FR_BZ_MSIX_PBA_TABLE_STEP 4 */
+#define	FR_CZ_MSIX_PBA_TABLE_ROWS 32
+#define	FRF_BZ_MSIX_PBA_PEND_DWORD_LBN 0
+#define	FRF_BZ_MSIX_PBA_PEND_DWORD_WIDTH 32
+
+/* SRM_DBG_REG: SRAM debug access */
+#define	FR_BZ_SRM_DBG 0x03000000
+#define	FR_BZ_SRM_DBG_STEP 8
+#define	FR_CZ_SRM_DBG_ROWS 262144
+#define	FR_BB_SRM_DBG_ROWS 2097152
+#define	FRF_BZ_SRM_DBG_LBN 0
+#define	FRF_BZ_SRM_DBG_WIDTH 64
+
+/* TB_MSIX_PBA_TABLE: MSIX Pending Bit Array */
+#define	FR_CZ_TB_MSIX_PBA_TABLE 0x00008000
+#define	FR_CZ_TB_MSIX_PBA_TABLE_STEP 4
+#define	FR_CZ_TB_MSIX_PBA_TABLE_ROWS 1024
+#define	FRF_CZ_TB_MSIX_PBA_PEND_DWORD_LBN 0
+#define	FRF_CZ_TB_MSIX_PBA_PEND_DWORD_WIDTH 32
+
+/* DRIVER_EV */
+#define	FSF_AZ_DRIVER_EV_SUBCODE_LBN 56
+#define	FSF_AZ_DRIVER_EV_SUBCODE_WIDTH 4
+#define	FSE_BZ_TX_DSC_ERROR_EV 15
+#define	FSE_BZ_RX_DSC_ERROR_EV 14
+#define	FSE_AA_RX_RECOVER_EV 11
+#define	FSE_AZ_TIMER_EV 10
+#define	FSE_AZ_TX_PKT_NON_TCP_UDP 9
+#define	FSE_AZ_WAKE_UP_EV 6
+#define	FSE_AZ_SRM_UPD_DONE_EV 5
+#define	FSE_AB_EVQ_NOT_EN_EV 3
+#define	FSE_AZ_EVQ_INIT_DONE_EV 2
+#define	FSE_AZ_RX_DESCQ_FLS_DONE_EV 1
+#define	FSE_AZ_TX_DESCQ_FLS_DONE_EV 0
+#define	FSF_AZ_DRIVER_EV_SUBDATA_LBN 0
+#define	FSF_AZ_DRIVER_EV_SUBDATA_WIDTH 14
+
+/* EVENT_ENTRY */
+#define	FSF_AZ_EV_CODE_LBN 60
+#define	FSF_AZ_EV_CODE_WIDTH 4
+#define	FSE_CZ_EV_CODE_MCDI_EV 12
+#define	FSE_CZ_EV_CODE_USER_EV 8
+#define	FSE_AZ_EV_CODE_DRV_GEN_EV 7
+#define	FSE_AZ_EV_CODE_GLOBAL_EV 6
+#define	FSE_AZ_EV_CODE_DRIVER_EV 5
+#define	FSE_AZ_EV_CODE_TX_EV 2
+#define	FSE_AZ_EV_CODE_RX_EV 0
+#define	FSF_AZ_EV_DATA_LBN 0
+#define	FSF_AZ_EV_DATA_WIDTH 60
+
+/* GLOBAL_EV */
+#define	FSF_BB_GLB_EV_RX_RECOVERY_LBN 12
+#define	FSF_BB_GLB_EV_RX_RECOVERY_WIDTH 1
+#define	FSF_AA_GLB_EV_RX_RECOVERY_LBN 11
+#define	FSF_AA_GLB_EV_RX_RECOVERY_WIDTH 1
+#define	FSF_BB_GLB_EV_XG_MGT_INTR_LBN 11
+#define	FSF_BB_GLB_EV_XG_MGT_INTR_WIDTH 1
+#define	FSF_AB_GLB_EV_XFP_PHY0_INTR_LBN 10
+#define	FSF_AB_GLB_EV_XFP_PHY0_INTR_WIDTH 1
+#define	FSF_AB_GLB_EV_XG_PHY0_INTR_LBN 9
+#define	FSF_AB_GLB_EV_XG_PHY0_INTR_WIDTH 1
+#define	FSF_AB_GLB_EV_G_PHY0_INTR_LBN 7
+#define	FSF_AB_GLB_EV_G_PHY0_INTR_WIDTH 1
+
+/* LEGACY_INT_VEC */
+#define	FSF_AZ_NET_IVEC_FATAL_INT_LBN 64
+#define	FSF_AZ_NET_IVEC_FATAL_INT_WIDTH 1
+#define	FSF_AZ_NET_IVEC_INT_Q_LBN 40
+#define	FSF_AZ_NET_IVEC_INT_Q_WIDTH 4
+#define	FSF_AZ_NET_IVEC_INT_FLAG_LBN 32
+#define	FSF_AZ_NET_IVEC_INT_FLAG_WIDTH 1
+#define	FSF_AZ_NET_IVEC_EVQ_FIFO_HF_LBN 1
+#define	FSF_AZ_NET_IVEC_EVQ_FIFO_HF_WIDTH 1
+#define	FSF_AZ_NET_IVEC_EVQ_FIFO_AF_LBN 0
+#define	FSF_AZ_NET_IVEC_EVQ_FIFO_AF_WIDTH 1
+
+/* MC_XGMAC_FLTR_RULE_DEF */
+#define	FSF_CZ_MC_XFRC_MODE_LBN 416
+#define	FSF_CZ_MC_XFRC_MODE_WIDTH 1
+#define	FSE_CZ_MC_XFRC_MODE_LAYERED 1
+#define	FSE_CZ_MC_XFRC_MODE_SIMPLE 0
+#define	FSF_CZ_MC_XFRC_HASH_LBN 384
+#define	FSF_CZ_MC_XFRC_HASH_WIDTH 32
+#define	FSF_CZ_MC_XFRC_LAYER4_BYTE_MASK_LBN 256
+#define	FSF_CZ_MC_XFRC_LAYER4_BYTE_MASK_WIDTH 128
+#define	FSF_CZ_MC_XFRC_LAYER3_BYTE_MASK_LBN 128
+#define	FSF_CZ_MC_XFRC_LAYER3_BYTE_MASK_WIDTH 128
+#define	FSF_CZ_MC_XFRC_LAYER2_OR_SIMPLE_BYTE_MASK_LBN 0
+#define	FSF_CZ_MC_XFRC_LAYER2_OR_SIMPLE_BYTE_MASK_WIDTH 128
+
+/* RX_EV */
+#define	FSF_CZ_RX_EV_PKT_NOT_PARSED_LBN 58
+#define	FSF_CZ_RX_EV_PKT_NOT_PARSED_WIDTH 1
+#define	FSF_CZ_RX_EV_IPV6_PKT_LBN 57
+#define	FSF_CZ_RX_EV_IPV6_PKT_WIDTH 1
+#define	FSF_AZ_RX_EV_PKT_OK_LBN 56
+#define	FSF_AZ_RX_EV_PKT_OK_WIDTH 1
+#define	FSF_AZ_RX_EV_PAUSE_FRM_ERR_LBN 55
+#define	FSF_AZ_RX_EV_PAUSE_FRM_ERR_WIDTH 1
+#define	FSF_AZ_RX_EV_BUF_OWNER_ID_ERR_LBN 54
+#define	FSF_AZ_RX_EV_BUF_OWNER_ID_ERR_WIDTH 1
+#define	FSF_AZ_RX_EV_IP_FRAG_ERR_LBN 53
+#define	FSF_AZ_RX_EV_IP_FRAG_ERR_WIDTH 1
+#define	FSF_AZ_RX_EV_IP_HDR_CHKSUM_ERR_LBN 52
+#define	FSF_AZ_RX_EV_IP_HDR_CHKSUM_ERR_WIDTH 1
+#define	FSF_AZ_RX_EV_TCP_UDP_CHKSUM_ERR_LBN 51
+#define	FSF_AZ_RX_EV_TCP_UDP_CHKSUM_ERR_WIDTH 1
+#define	FSF_AZ_RX_EV_ETH_CRC_ERR_LBN 50
+#define	FSF_AZ_RX_EV_ETH_CRC_ERR_WIDTH 1
+#define	FSF_AZ_RX_EV_FRM_TRUNC_LBN 49
+#define	FSF_AZ_RX_EV_FRM_TRUNC_WIDTH 1
+#define	FSF_AA_RX_EV_DRIB_NIB_LBN 49
+#define	FSF_AA_RX_EV_DRIB_NIB_WIDTH 1
+#define	FSF_AZ_RX_EV_TOBE_DISC_LBN 47
+#define	FSF_AZ_RX_EV_TOBE_DISC_WIDTH 1
+#define	FSF_AZ_RX_EV_PKT_TYPE_LBN 44
+#define	FSF_AZ_RX_EV_PKT_TYPE_WIDTH 3
+#define	FSE_AZ_RX_EV_PKT_TYPE_VLAN_JUMBO 5
+#define	FSE_AZ_RX_EV_PKT_TYPE_VLAN_LLC 4
+#define	FSE_AZ_RX_EV_PKT_TYPE_VLAN 3
+#define	FSE_AZ_RX_EV_PKT_TYPE_JUMBO 2
+#define	FSE_AZ_RX_EV_PKT_TYPE_LLC 1
+#define	FSE_AZ_RX_EV_PKT_TYPE_ETH 0
+#define	FSF_AZ_RX_EV_HDR_TYPE_LBN 42
+#define	FSF_AZ_RX_EV_HDR_TYPE_WIDTH 2
+#define	FSE_AZ_RX_EV_HDR_TYPE_OTHER 3
+#define	FSE_AB_RX_EV_HDR_TYPE_IPV4_OTHER 2
+#define	FSE_CZ_RX_EV_HDR_TYPE_IPV4V6_OTHER 2
+#define	FSE_AB_RX_EV_HDR_TYPE_IPV4_UDP 1
+#define	FSE_CZ_RX_EV_HDR_TYPE_IPV4V6_UDP 1
+#define	FSE_AB_RX_EV_HDR_TYPE_IPV4_TCP 0
+#define	FSE_CZ_RX_EV_HDR_TYPE_IPV4V6_TCP 0
+#define	FSF_AZ_RX_EV_DESC_Q_EMPTY_LBN 41
+#define	FSF_AZ_RX_EV_DESC_Q_EMPTY_WIDTH 1
+#define	FSF_AZ_RX_EV_MCAST_HASH_MATCH_LBN 40
+#define	FSF_AZ_RX_EV_MCAST_HASH_MATCH_WIDTH 1
+#define	FSF_AZ_RX_EV_MCAST_PKT_LBN 39
+#define	FSF_AZ_RX_EV_MCAST_PKT_WIDTH 1
+#define	FSF_AA_RX_EV_RECOVERY_FLAG_LBN 37
+#define	FSF_AA_RX_EV_RECOVERY_FLAG_WIDTH 1
+#define	FSF_AZ_RX_EV_Q_LABEL_LBN 32
+#define	FSF_AZ_RX_EV_Q_LABEL_WIDTH 5
+#define	FSF_AZ_RX_EV_JUMBO_CONT_LBN 31
+#define	FSF_AZ_RX_EV_JUMBO_CONT_WIDTH 1
+#define	FSF_AZ_RX_EV_PORT_LBN 30
+#define	FSF_AZ_RX_EV_PORT_WIDTH 1
+#define	FSF_AZ_RX_EV_BYTE_CNT_LBN 16
+#define	FSF_AZ_RX_EV_BYTE_CNT_WIDTH 14
+#define	FSF_AZ_RX_EV_SOP_LBN 15
+#define	FSF_AZ_RX_EV_SOP_WIDTH 1
+#define	FSF_AZ_RX_EV_ISCSI_PKT_OK_LBN 14
+#define	FSF_AZ_RX_EV_ISCSI_PKT_OK_WIDTH 1
+#define	FSF_AZ_RX_EV_ISCSI_DDIG_ERR_LBN 13
+#define	FSF_AZ_RX_EV_ISCSI_DDIG_ERR_WIDTH 1
+#define	FSF_AZ_RX_EV_ISCSI_HDIG_ERR_LBN 12
+#define	FSF_AZ_RX_EV_ISCSI_HDIG_ERR_WIDTH 1
+#define	FSF_AZ_RX_EV_DESC_PTR_LBN 0
+#define	FSF_AZ_RX_EV_DESC_PTR_WIDTH 12
+
+/* RX_KER_DESC */
+#define	FSF_AZ_RX_KER_BUF_SIZE_LBN 48
+#define	FSF_AZ_RX_KER_BUF_SIZE_WIDTH 14
+#define	FSF_AZ_RX_KER_BUF_REGION_LBN 46
+#define	FSF_AZ_RX_KER_BUF_REGION_WIDTH 2
+#define	FSF_AZ_RX_KER_BUF_ADDR_LBN 0
+#define	FSF_AZ_RX_KER_BUF_ADDR_WIDTH 46
+
+/* RX_USER_DESC */
+#define	FSF_AZ_RX_USER_2BYTE_OFFSET_LBN 20
+#define	FSF_AZ_RX_USER_2BYTE_OFFSET_WIDTH 12
+#define	FSF_AZ_RX_USER_BUF_ID_LBN 0
+#define	FSF_AZ_RX_USER_BUF_ID_WIDTH 20
+
+/* TX_EV */
+#define	FSF_AZ_TX_EV_PKT_ERR_LBN 38
+#define	FSF_AZ_TX_EV_PKT_ERR_WIDTH 1
+#define	FSF_AZ_TX_EV_PKT_TOO_BIG_LBN 37
+#define	FSF_AZ_TX_EV_PKT_TOO_BIG_WIDTH 1
+#define	FSF_AZ_TX_EV_Q_LABEL_LBN 32
+#define	FSF_AZ_TX_EV_Q_LABEL_WIDTH 5
+#define	FSF_AZ_TX_EV_PORT_LBN 16
+#define	FSF_AZ_TX_EV_PORT_WIDTH 1
+#define	FSF_AZ_TX_EV_WQ_FF_FULL_LBN 15
+#define	FSF_AZ_TX_EV_WQ_FF_FULL_WIDTH 1
+#define	FSF_AZ_TX_EV_BUF_OWNER_ID_ERR_LBN 14
+#define	FSF_AZ_TX_EV_BUF_OWNER_ID_ERR_WIDTH 1
+#define	FSF_AZ_TX_EV_COMP_LBN 12
+#define	FSF_AZ_TX_EV_COMP_WIDTH 1
+#define	FSF_AZ_TX_EV_DESC_PTR_LBN 0
+#define	FSF_AZ_TX_EV_DESC_PTR_WIDTH 12
+
+/* TX_KER_DESC */
+#define	FSF_AZ_TX_KER_CONT_LBN 62
+#define	FSF_AZ_TX_KER_CONT_WIDTH 1
+#define	FSF_AZ_TX_KER_BYTE_COUNT_LBN 48
+#define	FSF_AZ_TX_KER_BYTE_COUNT_WIDTH 14
+#define	FSF_AZ_TX_KER_BUF_REGION_LBN 46
+#define	FSF_AZ_TX_KER_BUF_REGION_WIDTH 2
+#define	FSF_AZ_TX_KER_BUF_ADDR_LBN 0
+#define	FSF_AZ_TX_KER_BUF_ADDR_WIDTH 46
+
+/* TX_USER_DESC */
+#define	FSF_AZ_TX_USER_SW_EV_EN_LBN 48
+#define	FSF_AZ_TX_USER_SW_EV_EN_WIDTH 1
+#define	FSF_AZ_TX_USER_CONT_LBN 46
+#define	FSF_AZ_TX_USER_CONT_WIDTH 1
+#define	FSF_AZ_TX_USER_BYTE_CNT_LBN 33
+#define	FSF_AZ_TX_USER_BYTE_CNT_WIDTH 13
+#define	FSF_AZ_TX_USER_BUF_ID_LBN 13
+#define	FSF_AZ_TX_USER_BUF_ID_WIDTH 20
+#define	FSF_AZ_TX_USER_BYTE_OFS_LBN 0
+#define	FSF_AZ_TX_USER_BYTE_OFS_WIDTH 13
+
+/* USER_EV */
+#define	FSF_CZ_USER_QID_LBN 32
+#define	FSF_CZ_USER_QID_WIDTH 10
+#define	FSF_CZ_USER_EV_REG_VALUE_LBN 0
+#define	FSF_CZ_USER_EV_REG_VALUE_WIDTH 32
+
+/**************************************************************************
+ *
+ * Falcon B0 PCIe core indirect registers
+ *
+ **************************************************************************
+ */
+
+#define FPCR_BB_PCIE_DEVICE_CTRL_STAT 0x68
+
+#define FPCR_BB_PCIE_LINK_CTRL_STAT 0x70
+
+#define FPCR_BB_ACK_RPL_TIMER 0x700
+#define FPCRF_BB_ACK_TL_LBN 0
+#define FPCRF_BB_ACK_TL_WIDTH 16
+#define FPCRF_BB_RPL_TL_LBN 16
+#define FPCRF_BB_RPL_TL_WIDTH 16
+
+#define FPCR_BB_ACK_FREQ 0x70C
+#define FPCRF_BB_ACK_FREQ_LBN 0
+#define FPCRF_BB_ACK_FREQ_WIDTH 7
+
+/**************************************************************************
+ *
+ * Pseudo-registers and fields
+ *
+ **************************************************************************
+ */
+
+/* Interrupt acknowledge work-around register (A0/A1 only) */
+#define FR_AA_WORK_AROUND_BROKEN_PCI_READS 0x0070
+
+/* EE_SPI_HCMD_REG: SPI host command register */
+/* Values for the EE_SPI_HCMD_SF_SEL register field */
+#define FFE_AB_SPI_DEVICE_EEPROM 0
+#define FFE_AB_SPI_DEVICE_FLASH 1
+
+/* NIC_STAT_REG: NIC status register */
+#define FRF_AB_STRAP_10G_LBN 2
+#define FRF_AB_STRAP_10G_WIDTH 1
+#define FRF_AA_STRAP_PCIE_LBN 0
+#define FRF_AA_STRAP_PCIE_WIDTH 1
+
+/* FATAL_INTR_REG_KER: Fatal interrupt register for Kernel */
+#define FRF_AZ_FATAL_INTR_LBN 0
+#define FRF_AZ_FATAL_INTR_WIDTH 12
+
+/* SRM_CFG_REG: SRAM configuration register */
+/* We treat the number of SRAM banks and bank size as a single field */
+#define	FRF_AZ_SRM_NB_SZ_LBN FRF_AZ_SRM_BANK_SIZE_LBN
+#define	FRF_AZ_SRM_NB_SZ_WIDTH \
+	(FRF_AZ_SRM_BANK_SIZE_WIDTH + FRF_AZ_SRM_NUM_BANK_WIDTH)
+#define FFE_AB_SRM_NB1_SZ2M 0
+#define FFE_AB_SRM_NB1_SZ4M 1
+#define FFE_AB_SRM_NB1_SZ8M 2
+#define FFE_AB_SRM_NB_SZ_DEF 3
+#define FFE_AB_SRM_NB2_SZ4M 4
+#define FFE_AB_SRM_NB2_SZ8M 5
+#define FFE_AB_SRM_NB2_SZ16M 6
+#define FFE_AB_SRM_NB_SZ_RES 7
+
+/* RX_DESC_UPD_REGP0: Receive descriptor update register. */
+/* We write just the last dword of these registers */
+#define	FR_AZ_RX_DESC_UPD_DWORD_P0 \
+	(BUILD_BUG_ON_ZERO(FR_AA_RX_DESC_UPD_KER != FR_BZ_RX_DESC_UPD_P0) + \
+	 FR_BZ_RX_DESC_UPD_P0 + 3 * 4)
+#define	FRF_AZ_RX_DESC_WPTR_DWORD_LBN (FRF_AZ_RX_DESC_WPTR_LBN - 3 * 32)
+#define	FRF_AZ_RX_DESC_WPTR_DWORD_WIDTH FRF_AZ_RX_DESC_WPTR_WIDTH
+
+/* TX_DESC_UPD_REGP0: Transmit descriptor update register. */
+#define FR_AZ_TX_DESC_UPD_DWORD_P0 \
+	(BUILD_BUG_ON_ZERO(FR_AA_TX_DESC_UPD_KER != FR_BZ_TX_DESC_UPD_P0) + \
+	 FR_BZ_TX_DESC_UPD_P0 + 3 * 4)
+#define	FRF_AZ_TX_DESC_WPTR_DWORD_LBN (FRF_AZ_TX_DESC_WPTR_LBN - 3 * 32)
+#define	FRF_AZ_TX_DESC_WPTR_DWORD_WIDTH FRF_AZ_TX_DESC_WPTR_WIDTH
+
+/* GMF_CFG4_REG: GMAC FIFO configuration register 4 */
+#define FRF_AB_GMF_HSTFLTRFRM_PAUSE_LBN 12
+#define FRF_AB_GMF_HSTFLTRFRM_PAUSE_WIDTH 1
+
+/* GMF_CFG5_REG: GMAC FIFO configuration register 5 */
+#define FRF_AB_GMF_HSTFLTRFRMDC_PAUSE_LBN 12
+#define FRF_AB_GMF_HSTFLTRFRMDC_PAUSE_WIDTH 1
+
+/* XM_TX_PARAM_REG: XGMAC transmit parameter register */
+#define	FRF_AB_XM_MAX_TX_FRM_SIZE_LBN FRF_AB_XM_MAX_TX_FRM_SIZE_LO_LBN
+#define	FRF_AB_XM_MAX_TX_FRM_SIZE_WIDTH (FRF_AB_XM_MAX_TX_FRM_SIZE_HI_WIDTH + \
+					 FRF_AB_XM_MAX_TX_FRM_SIZE_LO_WIDTH)
+
+/* XM_RX_PARAM_REG: XGMAC receive parameter register */
+#define	FRF_AB_XM_MAX_RX_FRM_SIZE_LBN FRF_AB_XM_MAX_RX_FRM_SIZE_LO_LBN
+#define	FRF_AB_XM_MAX_RX_FRM_SIZE_WIDTH (FRF_AB_XM_MAX_RX_FRM_SIZE_HI_WIDTH + \
+					 FRF_AB_XM_MAX_RX_FRM_SIZE_LO_WIDTH)
+
+/* XX_TXDRV_CTL_REG: XAUI SerDes transmit drive control register */
+/* Default values */
+#define FFE_AB_XX_TXDRV_DEQ_DEF 0xe /* deq=.6 */
+#define FFE_AB_XX_TXDRV_DTX_DEF 0x5 /* 1.25 */
+#define FFE_AB_XX_SD_CTL_DRV_DEF 0  /* 20mA */
+
+/* XX_CORE_STAT_REG: XAUI XGXS core status register */
+/* XGXS all-lanes status fields */
+#define	FRF_AB_XX_SYNC_STAT_LBN FRF_AB_XX_SYNC_STAT0_LBN
+#define	FRF_AB_XX_SYNC_STAT_WIDTH 4
+#define	FRF_AB_XX_COMMA_DET_LBN FRF_AB_XX_COMMA_DET_CH0_LBN
+#define	FRF_AB_XX_COMMA_DET_WIDTH 4
+#define	FRF_AB_XX_CHAR_ERR_LBN FRF_AB_XX_CHAR_ERR_CH0_LBN
+#define	FRF_AB_XX_CHAR_ERR_WIDTH 4
+#define	FRF_AB_XX_DISPERR_LBN FRF_AB_XX_DISPERR_CH0_LBN
+#define	FRF_AB_XX_DISPERR_WIDTH 4
+#define	FFE_AB_XX_STAT_ALL_LANES 0xf
+#define	FRF_AB_XX_FORCE_SIG_LBN FRF_AB_XX_FORCE_SIG0_VAL_LBN
+#define	FRF_AB_XX_FORCE_SIG_WIDTH 8
+#define	FFE_AB_XX_FORCE_SIG_ALL_LANES 0xff
+
+/* RX_MAC_FILTER_TBL0 */
+/* RMFT_DEST_MAC is wider than 32 bits */
+#define FRF_CZ_RMFT_DEST_MAC_LO_LBN 12
+#define FRF_CZ_RMFT_DEST_MAC_LO_WIDTH 32
+#define FRF_CZ_RMFT_DEST_MAC_HI_LBN 44
+#define FRF_CZ_RMFT_DEST_MAC_HI_WIDTH 16
+
+/* TX_MAC_FILTER_TBL0 */
+/* TMFT_SRC_MAC is wider than 32 bits */
+#define FRF_CZ_TMFT_SRC_MAC_LO_LBN 12
+#define FRF_CZ_TMFT_SRC_MAC_LO_WIDTH 32
+#define FRF_CZ_TMFT_SRC_MAC_HI_LBN 44
+#define FRF_CZ_TMFT_SRC_MAC_HI_WIDTH 16
+
+/* TX_PACE_TBL */
+/* Values >20 are documented as reserved, but will result in a queue going
+ * into the fast bin with a pace value of zero. */
+#define FFE_BZ_TX_PACE_OFF 0
+#define FFE_BZ_TX_PACE_RESERVED 21
+
+/* DRIVER_EV */
+/* Sub-fields of an RX flush completion event */
+#define FSF_AZ_DRIVER_EV_RX_FLUSH_FAIL_LBN 12
+#define FSF_AZ_DRIVER_EV_RX_FLUSH_FAIL_WIDTH 1
+#define FSF_AZ_DRIVER_EV_RX_DESCQ_ID_LBN 0
+#define FSF_AZ_DRIVER_EV_RX_DESCQ_ID_WIDTH 12
+
+/* EVENT_ENTRY */
+/* Magic number field for event test */
+#define FSF_AZ_DRV_GEN_EV_MAGIC_LBN 0
+#define FSF_AZ_DRV_GEN_EV_MAGIC_WIDTH 32
+
+/**************************************************************************
+ *
+ * Falcon MAC stats
+ *
+ **************************************************************************
+ *
+ */
+
+#define GRxGoodOct_offset 0x0
+#define GRxGoodOct_WIDTH 48
+#define GRxBadOct_offset 0x8
+#define GRxBadOct_WIDTH 48
+#define GRxMissPkt_offset 0x10
+#define GRxMissPkt_WIDTH 32
+#define GRxFalseCRS_offset 0x14
+#define GRxFalseCRS_WIDTH 32
+#define GRxPausePkt_offset 0x18
+#define GRxPausePkt_WIDTH 32
+#define GRxBadPkt_offset 0x1C
+#define GRxBadPkt_WIDTH 32
+#define GRxUcastPkt_offset 0x20
+#define GRxUcastPkt_WIDTH 32
+#define GRxMcastPkt_offset 0x24
+#define GRxMcastPkt_WIDTH 32
+#define GRxBcastPkt_offset 0x28
+#define GRxBcastPkt_WIDTH 32
+#define GRxGoodLt64Pkt_offset 0x2C
+#define GRxGoodLt64Pkt_WIDTH 32
+#define GRxBadLt64Pkt_offset 0x30
+#define GRxBadLt64Pkt_WIDTH 32
+#define GRx64Pkt_offset 0x34
+#define GRx64Pkt_WIDTH 32
+#define GRx65to127Pkt_offset 0x38
+#define GRx65to127Pkt_WIDTH 32
+#define GRx128to255Pkt_offset 0x3C
+#define GRx128to255Pkt_WIDTH 32
+#define GRx256to511Pkt_offset 0x40
+#define GRx256to511Pkt_WIDTH 32
+#define GRx512to1023Pkt_offset 0x44
+#define GRx512to1023Pkt_WIDTH 32
+#define GRx1024to15xxPkt_offset 0x48
+#define GRx1024to15xxPkt_WIDTH 32
+#define GRx15xxtoJumboPkt_offset 0x4C
+#define GRx15xxtoJumboPkt_WIDTH 32
+#define GRxGtJumboPkt_offset 0x50
+#define GRxGtJumboPkt_WIDTH 32
+#define GRxFcsErr64to15xxPkt_offset 0x54
+#define GRxFcsErr64to15xxPkt_WIDTH 32
+#define GRxFcsErr15xxtoJumboPkt_offset 0x58
+#define GRxFcsErr15xxtoJumboPkt_WIDTH 32
+#define GRxFcsErrGtJumboPkt_offset 0x5C
+#define GRxFcsErrGtJumboPkt_WIDTH 32
+#define GTxGoodBadOct_offset 0x80
+#define GTxGoodBadOct_WIDTH 48
+#define GTxGoodOct_offset 0x88
+#define GTxGoodOct_WIDTH 48
+#define GTxSglColPkt_offset 0x90
+#define GTxSglColPkt_WIDTH 32
+#define GTxMultColPkt_offset 0x94
+#define GTxMultColPkt_WIDTH 32
+#define GTxExColPkt_offset 0x98
+#define GTxExColPkt_WIDTH 32
+#define GTxDefPkt_offset 0x9C
+#define GTxDefPkt_WIDTH 32
+#define GTxLateCol_offset 0xA0
+#define GTxLateCol_WIDTH 32
+#define GTxExDefPkt_offset 0xA4
+#define GTxExDefPkt_WIDTH 32
+#define GTxPausePkt_offset 0xA8
+#define GTxPausePkt_WIDTH 32
+#define GTxBadPkt_offset 0xAC
+#define GTxBadPkt_WIDTH 32
+#define GTxUcastPkt_offset 0xB0
+#define GTxUcastPkt_WIDTH 32
+#define GTxMcastPkt_offset 0xB4
+#define GTxMcastPkt_WIDTH 32
+#define GTxBcastPkt_offset 0xB8
+#define GTxBcastPkt_WIDTH 32
+#define GTxLt64Pkt_offset 0xBC
+#define GTxLt64Pkt_WIDTH 32
+#define GTx64Pkt_offset 0xC0
+#define GTx64Pkt_WIDTH 32
+#define GTx65to127Pkt_offset 0xC4
+#define GTx65to127Pkt_WIDTH 32
+#define GTx128to255Pkt_offset 0xC8
+#define GTx128to255Pkt_WIDTH 32
+#define GTx256to511Pkt_offset 0xCC
+#define GTx256to511Pkt_WIDTH 32
+#define GTx512to1023Pkt_offset 0xD0
+#define GTx512to1023Pkt_WIDTH 32
+#define GTx1024to15xxPkt_offset 0xD4
+#define GTx1024to15xxPkt_WIDTH 32
+#define GTx15xxtoJumboPkt_offset 0xD8
+#define GTx15xxtoJumboPkt_WIDTH 32
+#define GTxGtJumboPkt_offset 0xDC
+#define GTxGtJumboPkt_WIDTH 32
+#define GTxNonTcpUdpPkt_offset 0xE0
+#define GTxNonTcpUdpPkt_WIDTH 16
+#define GTxMacSrcErrPkt_offset 0xE4
+#define GTxMacSrcErrPkt_WIDTH 16
+#define GTxIpSrcErrPkt_offset 0xE8
+#define GTxIpSrcErrPkt_WIDTH 16
+#define GDmaDone_offset 0xEC
+#define GDmaDone_WIDTH 32
+
+#define XgRxOctets_offset 0x0
+#define XgRxOctets_WIDTH 48
+#define XgRxOctetsOK_offset 0x8
+#define XgRxOctetsOK_WIDTH 48
+#define XgRxPkts_offset 0x10
+#define XgRxPkts_WIDTH 32
+#define XgRxPktsOK_offset 0x14
+#define XgRxPktsOK_WIDTH 32
+#define XgRxBroadcastPkts_offset 0x18
+#define XgRxBroadcastPkts_WIDTH 32
+#define XgRxMulticastPkts_offset 0x1C
+#define XgRxMulticastPkts_WIDTH 32
+#define XgRxUnicastPkts_offset 0x20
+#define XgRxUnicastPkts_WIDTH 32
+#define XgRxUndersizePkts_offset 0x24
+#define XgRxUndersizePkts_WIDTH 32
+#define XgRxOversizePkts_offset 0x28
+#define XgRxOversizePkts_WIDTH 32
+#define XgRxJabberPkts_offset 0x2C
+#define XgRxJabberPkts_WIDTH 32
+#define XgRxUndersizeFCSerrorPkts_offset 0x30
+#define XgRxUndersizeFCSerrorPkts_WIDTH 32
+#define XgRxDropEvents_offset 0x34
+#define XgRxDropEvents_WIDTH 32
+#define XgRxFCSerrorPkts_offset 0x38
+#define XgRxFCSerrorPkts_WIDTH 32
+#define XgRxAlignError_offset 0x3C
+#define XgRxAlignError_WIDTH 32
+#define XgRxSymbolError_offset 0x40
+#define XgRxSymbolError_WIDTH 32
+#define XgRxInternalMACError_offset 0x44
+#define XgRxInternalMACError_WIDTH 32
+#define XgRxControlPkts_offset 0x48
+#define XgRxControlPkts_WIDTH 32
+#define XgRxPausePkts_offset 0x4C
+#define XgRxPausePkts_WIDTH 32
+#define XgRxPkts64Octets_offset 0x50
+#define XgRxPkts64Octets_WIDTH 32
+#define XgRxPkts65to127Octets_offset 0x54
+#define XgRxPkts65to127Octets_WIDTH 32
+#define XgRxPkts128to255Octets_offset 0x58
+#define XgRxPkts128to255Octets_WIDTH 32
+#define XgRxPkts256to511Octets_offset 0x5C
+#define XgRxPkts256to511Octets_WIDTH 32
+#define XgRxPkts512to1023Octets_offset 0x60
+#define XgRxPkts512to1023Octets_WIDTH 32
+#define XgRxPkts1024to15xxOctets_offset 0x64
+#define XgRxPkts1024to15xxOctets_WIDTH 32
+#define XgRxPkts15xxtoMaxOctets_offset 0x68
+#define XgRxPkts15xxtoMaxOctets_WIDTH 32
+#define XgRxLengthError_offset 0x6C
+#define XgRxLengthError_WIDTH 32
+#define XgTxPkts_offset 0x80
+#define XgTxPkts_WIDTH 32
+#define XgTxOctets_offset 0x88
+#define XgTxOctets_WIDTH 48
+#define XgTxMulticastPkts_offset 0x90
+#define XgTxMulticastPkts_WIDTH 32
+#define XgTxBroadcastPkts_offset 0x94
+#define XgTxBroadcastPkts_WIDTH 32
+#define XgTxUnicastPkts_offset 0x98
+#define XgTxUnicastPkts_WIDTH 32
+#define XgTxControlPkts_offset 0x9C
+#define XgTxControlPkts_WIDTH 32
+#define XgTxPausePkts_offset 0xA0
+#define XgTxPausePkts_WIDTH 32
+#define XgTxPkts64Octets_offset 0xA4
+#define XgTxPkts64Octets_WIDTH 32
+#define XgTxPkts65to127Octets_offset 0xA8
+#define XgTxPkts65to127Octets_WIDTH 32
+#define XgTxPkts128to255Octets_offset 0xAC
+#define XgTxPkts128to255Octets_WIDTH 32
+#define XgTxPkts256to511Octets_offset 0xB0
+#define XgTxPkts256to511Octets_WIDTH 32
+#define XgTxPkts512to1023Octets_offset 0xB4
+#define XgTxPkts512to1023Octets_WIDTH 32
+#define XgTxPkts1024to15xxOctets_offset 0xB8
+#define XgTxPkts1024to15xxOctets_WIDTH 32
+#define XgTxPkts1519toMaxOctets_offset 0xBC
+#define XgTxPkts1519toMaxOctets_WIDTH 32
+#define XgTxUndersizePkts_offset 0xC0
+#define XgTxUndersizePkts_WIDTH 32
+#define XgTxOversizePkts_offset 0xC4
+#define XgTxOversizePkts_WIDTH 32
+#define XgTxNonTcpUdpPkt_offset 0xC8
+#define XgTxNonTcpUdpPkt_WIDTH 16
+#define XgTxMacSrcErrPkt_offset 0xCC
+#define XgTxMacSrcErrPkt_WIDTH 16
+#define XgTxIpSrcErrPkt_offset 0xD0
+#define XgTxIpSrcErrPkt_WIDTH 16
+#define XgDmaDone_offset 0xD4
+#define XgDmaDone_WIDTH 32
+
+#define FALCON_STATS_NOT_DONE 0x00000000
+#define FALCON_STATS_DONE 0xffffffff
+
+/**************************************************************************
+ *
+ * Falcon non-volatile configuration
+ *
+ **************************************************************************
+ */
+
+/* Board configuration v2 (v1 is obsolete; later versions are compatible) */
+struct falcon_nvconfig_board_v2 {
+	__le16 nports;
+	u8 port0_phy_addr;
+	u8 port0_phy_type;
+	u8 port1_phy_addr;
+	u8 port1_phy_type;
+	__le16 asic_sub_revision;
+	__le16 board_revision;
+} __packed;
+
+/* Board configuration v3 extra information */
+struct falcon_nvconfig_board_v3 {
+	__le32 spi_device_type[2];
+} __packed;
+
+/* Bit numbers for spi_device_type */
+#define SPI_DEV_TYPE_SIZE_LBN 0
+#define SPI_DEV_TYPE_SIZE_WIDTH 5
+#define SPI_DEV_TYPE_ADDR_LEN_LBN 6
+#define SPI_DEV_TYPE_ADDR_LEN_WIDTH 2
+#define SPI_DEV_TYPE_ERASE_CMD_LBN 8
+#define SPI_DEV_TYPE_ERASE_CMD_WIDTH 8
+#define SPI_DEV_TYPE_ERASE_SIZE_LBN 16
+#define SPI_DEV_TYPE_ERASE_SIZE_WIDTH 5
+#define SPI_DEV_TYPE_BLOCK_SIZE_LBN 24
+#define SPI_DEV_TYPE_BLOCK_SIZE_WIDTH 5
+#define SPI_DEV_TYPE_FIELD(type, field)					\
+	(((type) >> EFX_LOW_BIT(field)) & EFX_MASK32(EFX_WIDTH(field)))
+
+#define FALCON_NVCONFIG_OFFSET 0x300
+
+#define FALCON_NVCONFIG_BOARD_MAGIC_NUM 0xFA1C
+struct falcon_nvconfig {
+	efx_oword_t ee_vpd_cfg_reg;			/* 0x300 */
+	u8 mac_address[2][8];			/* 0x310 */
+	efx_oword_t pcie_sd_ctl0123_reg;		/* 0x320 */
+	efx_oword_t pcie_sd_ctl45_reg;			/* 0x330 */
+	efx_oword_t pcie_pcs_ctl_stat_reg;		/* 0x340 */
+	efx_oword_t hw_init_reg;			/* 0x350 */
+	efx_oword_t nic_stat_reg;			/* 0x360 */
+	efx_oword_t glb_ctl_reg;			/* 0x370 */
+	efx_oword_t srm_cfg_reg;			/* 0x380 */
+	efx_oword_t spare_reg;				/* 0x390 */
+	__le16 board_magic_num;			/* 0x3A0 */
+	__le16 board_struct_ver;
+	__le16 board_checksum;
+	struct falcon_nvconfig_board_v2 board_v2;
+	efx_oword_t ee_base_page_reg;			/* 0x3B0 */
+	struct falcon_nvconfig_board_v3 board_v3;	/* 0x3C0 */
+} __packed;
+
+#endif /* EFX_REGS_H */
diff --git a/drivers/net/ethernet/sfc/rx.c b/drivers/net/ethernet/sfc/rx.c
new file mode 100644
index 0000000..62e4364
--- /dev/null
+++ b/drivers/net/ethernet/sfc/rx.c
@@ -0,0 +1,749 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2005-2006 Fen Systems Ltd.
+ * Copyright 2005-2011 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#include <linux/socket.h>
+#include <linux/in.h>
+#include <linux/slab.h>
+#include <linux/ip.h>
+#include <linux/tcp.h>
+#include <linux/udp.h>
+#include <linux/prefetch.h>
+#include <net/ip.h>
+#include <net/checksum.h>
+#include "net_driver.h"
+#include "efx.h"
+#include "nic.h"
+#include "selftest.h"
+#include "workarounds.h"
+
+/* Number of RX descriptors pushed at once. */
+#define EFX_RX_BATCH  8
+
+/* Maximum size of a buffer sharing a page */
+#define EFX_RX_HALF_PAGE ((PAGE_SIZE >> 1) - sizeof(struct efx_rx_page_state))
+
+/* Size of buffer allocated for skb header area. */
+#define EFX_SKB_HEADERS  64u
+
+/*
+ * rx_alloc_method - RX buffer allocation method
+ *
+ * This driver supports two methods for allocating and using RX buffers:
+ * each RX buffer may be backed by an skb or by an order-n page.
+ *
+ * When GRO is in use then the second method has a lower overhead,
+ * since we don't have to allocate then free skbs on reassembled frames.
+ *
+ * Values:
+ *   - RX_ALLOC_METHOD_AUTO = 0
+ *   - RX_ALLOC_METHOD_SKB  = 1
+ *   - RX_ALLOC_METHOD_PAGE = 2
+ *
+ * The heuristic for %RX_ALLOC_METHOD_AUTO is a simple hysteresis count
+ * controlled by the parameters below.
+ *
+ *   - Since pushing and popping descriptors are separated by the rx_queue
+ *     size, so the watermarks should be ~rxd_size.
+ *   - The performance win by using page-based allocation for GRO is less
+ *     than the performance hit of using page-based allocation of non-GRO,
+ *     so the watermarks should reflect this.
+ *
+ * Per channel we maintain a single variable, updated by each channel:
+ *
+ *   rx_alloc_level += (gro_performed ? RX_ALLOC_FACTOR_GRO :
+ *                      RX_ALLOC_FACTOR_SKB)
+ * Per NAPI poll interval, we constrain rx_alloc_level to 0..MAX (which
+ * limits the hysteresis), and update the allocation strategy:
+ *
+ *   rx_alloc_method = (rx_alloc_level > RX_ALLOC_LEVEL_GRO ?
+ *                      RX_ALLOC_METHOD_PAGE : RX_ALLOC_METHOD_SKB)
+ */
+static int rx_alloc_method = RX_ALLOC_METHOD_AUTO;
+
+#define RX_ALLOC_LEVEL_GRO 0x2000
+#define RX_ALLOC_LEVEL_MAX 0x3000
+#define RX_ALLOC_FACTOR_GRO 1
+#define RX_ALLOC_FACTOR_SKB (-2)
+
+/* This is the percentage fill level below which new RX descriptors
+ * will be added to the RX descriptor ring.
+ */
+static unsigned int rx_refill_threshold = 90;
+
+/* This is the percentage fill level to which an RX queue will be refilled
+ * when the "RX refill threshold" is reached.
+ */
+static unsigned int rx_refill_limit = 95;
+
+/*
+ * RX maximum head room required.
+ *
+ * This must be at least 1 to prevent overflow and at least 2 to allow
+ * pipelined receives.
+ */
+#define EFX_RXD_HEAD_ROOM 2
+
+/* Offset of ethernet header within page */
+static inline unsigned int efx_rx_buf_offset(struct efx_nic *efx,
+					     struct efx_rx_buffer *buf)
+{
+	/* Offset is always within one page, so we don't need to consider
+	 * the page order.
+	 */
+	return (((__force unsigned long) buf->dma_addr & (PAGE_SIZE - 1)) +
+		efx->type->rx_buffer_hash_size);
+}
+static inline unsigned int efx_rx_buf_size(struct efx_nic *efx)
+{
+	return PAGE_SIZE << efx->rx_buffer_order;
+}
+
+static u8 *efx_rx_buf_eh(struct efx_nic *efx, struct efx_rx_buffer *buf)
+{
+	if (buf->is_page)
+		return page_address(buf->u.page) + efx_rx_buf_offset(efx, buf);
+	else
+		return ((u8 *)buf->u.skb->data +
+			efx->type->rx_buffer_hash_size);
+}
+
+static inline u32 efx_rx_buf_hash(const u8 *eh)
+{
+	/* The ethernet header is always directly after any hash. */
+#if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) || NET_IP_ALIGN % 4 == 0
+	return __le32_to_cpup((const __le32 *)(eh - 4));
+#else
+	const u8 *data = eh - 4;
+	return ((u32)data[0]       |
+		(u32)data[1] << 8  |
+		(u32)data[2] << 16 |
+		(u32)data[3] << 24);
+#endif
+}
+
+/**
+ * efx_init_rx_buffers_skb - create EFX_RX_BATCH skb-based RX buffers
+ *
+ * @rx_queue:		Efx RX queue
+ *
+ * This allocates EFX_RX_BATCH skbs, maps them for DMA, and populates a
+ * struct efx_rx_buffer for each one. Return a negative error code or 0
+ * on success. May fail having only inserted fewer than EFX_RX_BATCH
+ * buffers.
+ */
+static int efx_init_rx_buffers_skb(struct efx_rx_queue *rx_queue)
+{
+	struct efx_nic *efx = rx_queue->efx;
+	struct net_device *net_dev = efx->net_dev;
+	struct efx_rx_buffer *rx_buf;
+	struct sk_buff *skb;
+	int skb_len = efx->rx_buffer_len;
+	unsigned index, count;
+
+	for (count = 0; count < EFX_RX_BATCH; ++count) {
+		index = rx_queue->added_count & rx_queue->ptr_mask;
+		rx_buf = efx_rx_buffer(rx_queue, index);
+
+		rx_buf->u.skb = skb = netdev_alloc_skb(net_dev, skb_len);
+		if (unlikely(!skb))
+			return -ENOMEM;
+
+		/* Adjust the SKB for padding and checksum */
+		skb_reserve(skb, NET_IP_ALIGN);
+		rx_buf->len = skb_len - NET_IP_ALIGN;
+		rx_buf->is_page = false;
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+		rx_buf->dma_addr = pci_map_single(efx->pci_dev,
+						  skb->data, rx_buf->len,
+						  PCI_DMA_FROMDEVICE);
+		if (unlikely(pci_dma_mapping_error(efx->pci_dev,
+						   rx_buf->dma_addr))) {
+			dev_kfree_skb_any(skb);
+			rx_buf->u.skb = NULL;
+			return -EIO;
+		}
+
+		++rx_queue->added_count;
+		++rx_queue->alloc_skb_count;
+	}
+
+	return 0;
+}
+
+/**
+ * efx_init_rx_buffers_page - create EFX_RX_BATCH page-based RX buffers
+ *
+ * @rx_queue:		Efx RX queue
+ *
+ * This allocates memory for EFX_RX_BATCH receive buffers, maps them for DMA,
+ * and populates struct efx_rx_buffers for each one. Return a negative error
+ * code or 0 on success. If a single page can be split between two buffers,
+ * then the page will either be inserted fully, or not at at all.
+ */
+static int efx_init_rx_buffers_page(struct efx_rx_queue *rx_queue)
+{
+	struct efx_nic *efx = rx_queue->efx;
+	struct efx_rx_buffer *rx_buf;
+	struct page *page;
+	void *page_addr;
+	struct efx_rx_page_state *state;
+	dma_addr_t dma_addr;
+	unsigned index, count;
+
+	/* We can split a page between two buffers */
+	BUILD_BUG_ON(EFX_RX_BATCH & 1);
+
+	for (count = 0; count < EFX_RX_BATCH; ++count) {
+		page = alloc_pages(__GFP_COLD | __GFP_COMP | GFP_ATOMIC,
+				   efx->rx_buffer_order);
+		if (unlikely(page == NULL))
+			return -ENOMEM;
+		dma_addr = pci_map_page(efx->pci_dev, page, 0,
+					efx_rx_buf_size(efx),
+					PCI_DMA_FROMDEVICE);
+		if (unlikely(pci_dma_mapping_error(efx->pci_dev, dma_addr))) {
+			__free_pages(page, efx->rx_buffer_order);
+			return -EIO;
+		}
+		page_addr = page_address(page);
+		state = page_addr;
+		state->refcnt = 0;
+		state->dma_addr = dma_addr;
+
+		page_addr += sizeof(struct efx_rx_page_state);
+		dma_addr += sizeof(struct efx_rx_page_state);
+
+	split:
+		index = rx_queue->added_count & rx_queue->ptr_mask;
+		rx_buf = efx_rx_buffer(rx_queue, index);
+		rx_buf->dma_addr = dma_addr + EFX_PAGE_IP_ALIGN;
+		rx_buf->u.page = page;
+		rx_buf->len = efx->rx_buffer_len - EFX_PAGE_IP_ALIGN;
+		rx_buf->is_page = true;
+		++rx_queue->added_count;
+		++rx_queue->alloc_page_count;
+		++state->refcnt;
+
+		if ((~count & 1) && (efx->rx_buffer_len <= EFX_RX_HALF_PAGE)) {
+			/* Use the second half of the page */
+			get_page(page);
+			dma_addr += (PAGE_SIZE >> 1);
+			page_addr += (PAGE_SIZE >> 1);
+			++count;
+			goto split;
+		}
+	}
+
+	return 0;
+}
+
+static void efx_unmap_rx_buffer(struct efx_nic *efx,
+				struct efx_rx_buffer *rx_buf)
+{
+	if (rx_buf->is_page && rx_buf->u.page) {
+		struct efx_rx_page_state *state;
+
+		state = page_address(rx_buf->u.page);
+		if (--state->refcnt == 0) {
+			pci_unmap_page(efx->pci_dev,
+				       state->dma_addr,
+				       efx_rx_buf_size(efx),
+				       PCI_DMA_FROMDEVICE);
+		}
+	} else if (!rx_buf->is_page && rx_buf->u.skb) {
+		pci_unmap_single(efx->pci_dev, rx_buf->dma_addr,
+				 rx_buf->len, PCI_DMA_FROMDEVICE);
+	}
+}
+
+static void efx_free_rx_buffer(struct efx_nic *efx,
+			       struct efx_rx_buffer *rx_buf)
+{
+	if (rx_buf->is_page && rx_buf->u.page) {
+		__free_pages(rx_buf->u.page, efx->rx_buffer_order);
+		rx_buf->u.page = NULL;
+	} else if (!rx_buf->is_page && rx_buf->u.skb) {
+		dev_kfree_skb_any(rx_buf->u.skb);
+		rx_buf->u.skb = NULL;
+	}
+}
+
+static void efx_fini_rx_buffer(struct efx_rx_queue *rx_queue,
+			       struct efx_rx_buffer *rx_buf)
+{
+	efx_unmap_rx_buffer(rx_queue->efx, rx_buf);
+	efx_free_rx_buffer(rx_queue->efx, rx_buf);
+}
+
+/* Attempt to resurrect the other receive buffer that used to share this page,
+ * which had previously been passed up to the kernel and freed. */
+static void efx_resurrect_rx_buffer(struct efx_rx_queue *rx_queue,
+				    struct efx_rx_buffer *rx_buf)
+{
+	struct efx_rx_page_state *state = page_address(rx_buf->u.page);
+	struct efx_rx_buffer *new_buf;
+	unsigned fill_level, index;
+
+	/* +1 because efx_rx_packet() incremented removed_count. +1 because
+	 * we'd like to insert an additional descriptor whilst leaving
+	 * EFX_RXD_HEAD_ROOM for the non-recycle path */
+	fill_level = (rx_queue->added_count - rx_queue->removed_count + 2);
+	if (unlikely(fill_level > rx_queue->max_fill)) {
+		/* We could place "state" on a list, and drain the list in
+		 * efx_fast_push_rx_descriptors(). For now, this will do. */
+		return;
+	}
+
+	++state->refcnt;
+	get_page(rx_buf->u.page);
+
+	index = rx_queue->added_count & rx_queue->ptr_mask;
+	new_buf = efx_rx_buffer(rx_queue, index);
+	new_buf->dma_addr = rx_buf->dma_addr ^ (PAGE_SIZE >> 1);
+	new_buf->u.page = rx_buf->u.page;
+	new_buf->len = rx_buf->len;
+	new_buf->is_page = true;
+	++rx_queue->added_count;
+}
+
+/* Recycle the given rx buffer directly back into the rx_queue. There is
+ * always room to add this buffer, because we've just popped a buffer. */
+static void efx_recycle_rx_buffer(struct efx_channel *channel,
+				  struct efx_rx_buffer *rx_buf)
+{
+	struct efx_nic *efx = channel->efx;
+	struct efx_rx_queue *rx_queue = efx_channel_get_rx_queue(channel);
+	struct efx_rx_buffer *new_buf;
+	unsigned index;
+
+	if (rx_buf->is_page && efx->rx_buffer_len <= EFX_RX_HALF_PAGE &&
+	    page_count(rx_buf->u.page) == 1)
+		efx_resurrect_rx_buffer(rx_queue, rx_buf);
+
+	index = rx_queue->added_count & rx_queue->ptr_mask;
+	new_buf = efx_rx_buffer(rx_queue, index);
+
+	memcpy(new_buf, rx_buf, sizeof(*new_buf));
+	rx_buf->u.page = NULL;
+	++rx_queue->added_count;
+}
+
+/**
+ * efx_fast_push_rx_descriptors - push new RX descriptors quickly
+ * @rx_queue:		RX descriptor queue
+ * This will aim to fill the RX descriptor queue up to
+ * @rx_queue->@fast_fill_limit. If there is insufficient atomic
+ * memory to do so, a slow fill will be scheduled.
+ *
+ * The caller must provide serialisation (none is used here). In practise,
+ * this means this function must run from the NAPI handler, or be called
+ * when NAPI is disabled.
+ */
+void efx_fast_push_rx_descriptors(struct efx_rx_queue *rx_queue)
+{
+	struct efx_channel *channel = efx_rx_queue_channel(rx_queue);
+	unsigned fill_level;
+	int space, rc = 0;
+
+	/* Calculate current fill level, and exit if we don't need to fill */
+	fill_level = (rx_queue->added_count - rx_queue->removed_count);
+	EFX_BUG_ON_PARANOID(fill_level > rx_queue->efx->rxq_entries);
+	if (fill_level >= rx_queue->fast_fill_trigger)
+		goto out;
+
+	/* Record minimum fill level */
+	if (unlikely(fill_level < rx_queue->min_fill)) {
+		if (fill_level)
+			rx_queue->min_fill = fill_level;
+	}
+
+	space = rx_queue->fast_fill_limit - fill_level;
+	if (space < EFX_RX_BATCH)
+		goto out;
+
+	netif_vdbg(rx_queue->efx, rx_status, rx_queue->efx->net_dev,
+		   "RX queue %d fast-filling descriptor ring from"
+		   " level %d to level %d using %s allocation\n",
+		   efx_rx_queue_index(rx_queue), fill_level,
+		   rx_queue->fast_fill_limit,
+		   channel->rx_alloc_push_pages ? "page" : "skb");
+
+	do {
+		if (channel->rx_alloc_push_pages)
+			rc = efx_init_rx_buffers_page(rx_queue);
+		else
+			rc = efx_init_rx_buffers_skb(rx_queue);
+		if (unlikely(rc)) {
+			/* Ensure that we don't leave the rx queue empty */
+			if (rx_queue->added_count == rx_queue->removed_count)
+				efx_schedule_slow_fill(rx_queue);
+			goto out;
+		}
+	} while ((space -= EFX_RX_BATCH) >= EFX_RX_BATCH);
+
+	netif_vdbg(rx_queue->efx, rx_status, rx_queue->efx->net_dev,
+		   "RX queue %d fast-filled descriptor ring "
+		   "to level %d\n", efx_rx_queue_index(rx_queue),
+		   rx_queue->added_count - rx_queue->removed_count);
+
+ out:
+	if (rx_queue->notified_count != rx_queue->added_count)
+		efx_nic_notify_rx_desc(rx_queue);
+}
+
+void efx_rx_slow_fill(unsigned long context)
+{
+	struct efx_rx_queue *rx_queue = (struct efx_rx_queue *)context;
+	struct efx_channel *channel = efx_rx_queue_channel(rx_queue);
+
+	/* Post an event to cause NAPI to run and refill the queue */
+	efx_nic_generate_fill_event(channel);
+	++rx_queue->slow_fill_count;
+}
+
+static void efx_rx_packet__check_len(struct efx_rx_queue *rx_queue,
+				     struct efx_rx_buffer *rx_buf,
+				     int len, bool *discard,
+				     bool *leak_packet)
+{
+	struct efx_nic *efx = rx_queue->efx;
+	unsigned max_len = rx_buf->len - efx->type->rx_buffer_padding;
+
+	if (likely(len <= max_len))
+		return;
+
+	/* The packet must be discarded, but this is only a fatal error
+	 * if the caller indicated it was
+	 */
+	*discard = true;
+
+	if ((len > rx_buf->len) && EFX_WORKAROUND_8071(efx)) {
+		if (net_ratelimit())
+			netif_err(efx, rx_err, efx->net_dev,
+				  " RX queue %d seriously overlength "
+				  "RX event (0x%x > 0x%x+0x%x). Leaking\n",
+				  efx_rx_queue_index(rx_queue), len, max_len,
+				  efx->type->rx_buffer_padding);
+		/* If this buffer was skb-allocated, then the meta
+		 * data at the end of the skb will be trashed. So
+		 * we have no choice but to leak the fragment.
+		 */
+		*leak_packet = !rx_buf->is_page;
+		efx_schedule_reset(efx, RESET_TYPE_RX_RECOVERY);
+	} else {
+		if (net_ratelimit())
+			netif_err(efx, rx_err, efx->net_dev,
+				  " RX queue %d overlength RX event "
+				  "(0x%x > 0x%x)\n",
+				  efx_rx_queue_index(rx_queue), len, max_len);
+	}
+
+	efx_rx_queue_channel(rx_queue)->n_rx_overlength++;
+}
+
+/* Pass a received packet up through the generic GRO stack
+ *
+ * Handles driverlink veto, and passes the fragment up via
+ * the appropriate GRO method
+ */
+static void efx_rx_packet_gro(struct efx_channel *channel,
+			      struct efx_rx_buffer *rx_buf,
+			      const u8 *eh, bool checksummed)
+{
+	struct napi_struct *napi = &channel->napi_str;
+	gro_result_t gro_result;
+
+	/* Pass the skb/page into the GRO engine */
+	if (rx_buf->is_page) {
+		struct efx_nic *efx = channel->efx;
+		struct page *page = rx_buf->u.page;
+		struct sk_buff *skb;
+
+		rx_buf->u.page = NULL;
+
+		skb = napi_get_frags(napi);
+		if (!skb) {
+			put_page(page);
+			return;
+		}
+
+		if (efx->net_dev->features & NETIF_F_RXHASH)
+			skb->rxhash = efx_rx_buf_hash(eh);
+
+		skb_shinfo(skb)->frags[0].page = page;
+		skb_shinfo(skb)->frags[0].page_offset =
+			efx_rx_buf_offset(efx, rx_buf);
+		skb_shinfo(skb)->frags[0].size = rx_buf->len;
+		skb_shinfo(skb)->nr_frags = 1;
+
+		skb->len = rx_buf->len;
+		skb->data_len = rx_buf->len;
+		skb->truesize += rx_buf->len;
+		skb->ip_summed =
+			checksummed ? CHECKSUM_UNNECESSARY : CHECKSUM_NONE;
+
+		skb_record_rx_queue(skb, channel->channel);
+
+		gro_result = napi_gro_frags(napi);
+	} else {
+		struct sk_buff *skb = rx_buf->u.skb;
+
+		EFX_BUG_ON_PARANOID(!checksummed);
+		rx_buf->u.skb = NULL;
+
+		gro_result = napi_gro_receive(napi, skb);
+	}
+
+	if (gro_result == GRO_NORMAL) {
+		channel->rx_alloc_level += RX_ALLOC_FACTOR_SKB;
+	} else if (gro_result != GRO_DROP) {
+		channel->rx_alloc_level += RX_ALLOC_FACTOR_GRO;
+		channel->irq_mod_score += 2;
+	}
+}
+
+void efx_rx_packet(struct efx_rx_queue *rx_queue, unsigned int index,
+		   unsigned int len, bool checksummed, bool discard)
+{
+	struct efx_nic *efx = rx_queue->efx;
+	struct efx_channel *channel = efx_rx_queue_channel(rx_queue);
+	struct efx_rx_buffer *rx_buf;
+	bool leak_packet = false;
+
+	rx_buf = efx_rx_buffer(rx_queue, index);
+
+	/* This allows the refill path to post another buffer.
+	 * EFX_RXD_HEAD_ROOM ensures that the slot we are using
+	 * isn't overwritten yet.
+	 */
+	rx_queue->removed_count++;
+
+	/* Validate the length encoded in the event vs the descriptor pushed */
+	efx_rx_packet__check_len(rx_queue, rx_buf, len,
+				 &discard, &leak_packet);
+
+	netif_vdbg(efx, rx_status, efx->net_dev,
+		   "RX queue %d received id %x at %llx+%x %s%s\n",
+		   efx_rx_queue_index(rx_queue), index,
+		   (unsigned long long)rx_buf->dma_addr, len,
+		   (checksummed ? " [SUMMED]" : ""),
+		   (discard ? " [DISCARD]" : ""));
+
+	/* Discard packet, if instructed to do so */
+	if (unlikely(discard)) {
+		if (unlikely(leak_packet))
+			channel->n_skbuff_leaks++;
+		else
+			efx_recycle_rx_buffer(channel, rx_buf);
+
+		/* Don't hold off the previous receive */
+		rx_buf = NULL;
+		goto out;
+	}
+
+	/* Release card resources - assumes all RX buffers consumed in-order
+	 * per RX queue
+	 */
+	efx_unmap_rx_buffer(efx, rx_buf);
+
+	/* Prefetch nice and early so data will (hopefully) be in cache by
+	 * the time we look at it.
+	 */
+	prefetch(efx_rx_buf_eh(efx, rx_buf));
+
+	/* Pipeline receives so that we give time for packet headers to be
+	 * prefetched into cache.
+	 */
+	rx_buf->len = len - efx->type->rx_buffer_hash_size;
+out:
+	if (channel->rx_pkt)
+		__efx_rx_packet(channel,
+				channel->rx_pkt, channel->rx_pkt_csummed);
+	channel->rx_pkt = rx_buf;
+	channel->rx_pkt_csummed = checksummed;
+}
+
+/* Handle a received packet.  Second half: Touches packet payload. */
+void __efx_rx_packet(struct efx_channel *channel,
+		     struct efx_rx_buffer *rx_buf, bool checksummed)
+{
+	struct efx_nic *efx = channel->efx;
+	struct sk_buff *skb;
+	u8 *eh = efx_rx_buf_eh(efx, rx_buf);
+
+	/* If we're in loopback test, then pass the packet directly to the
+	 * loopback layer, and free the rx_buf here
+	 */
+	if (unlikely(efx->loopback_selftest)) {
+		efx_loopback_rx_packet(efx, eh, rx_buf->len);
+		efx_free_rx_buffer(efx, rx_buf);
+		return;
+	}
+
+	if (!rx_buf->is_page) {
+		skb = rx_buf->u.skb;
+
+		prefetch(skb_shinfo(skb));
+
+		skb_reserve(skb, efx->type->rx_buffer_hash_size);
+		skb_put(skb, rx_buf->len);
+
+		if (efx->net_dev->features & NETIF_F_RXHASH)
+			skb->rxhash = efx_rx_buf_hash(eh);
+
+		/* Move past the ethernet header. rx_buf->data still points
+		 * at the ethernet header */
+		skb->protocol = eth_type_trans(skb, efx->net_dev);
+
+		skb_record_rx_queue(skb, channel->channel);
+	}
+
+	if (unlikely(!(efx->net_dev->features & NETIF_F_RXCSUM)))
+		checksummed = false;
+
+	if (likely(checksummed || rx_buf->is_page)) {
+		efx_rx_packet_gro(channel, rx_buf, eh, checksummed);
+		return;
+	}
+
+	/* We now own the SKB */
+	skb = rx_buf->u.skb;
+	rx_buf->u.skb = NULL;
+
+	/* Set the SKB flags */
+	skb_checksum_none_assert(skb);
+
+	/* Pass the packet up */
+	netif_receive_skb(skb);
+
+	/* Update allocation strategy method */
+	channel->rx_alloc_level += RX_ALLOC_FACTOR_SKB;
+}
+
+void efx_rx_strategy(struct efx_channel *channel)
+{
+	enum efx_rx_alloc_method method = rx_alloc_method;
+
+	/* Only makes sense to use page based allocation if GRO is enabled */
+	if (!(channel->efx->net_dev->features & NETIF_F_GRO)) {
+		method = RX_ALLOC_METHOD_SKB;
+	} else if (method == RX_ALLOC_METHOD_AUTO) {
+		/* Constrain the rx_alloc_level */
+		if (channel->rx_alloc_level < 0)
+			channel->rx_alloc_level = 0;
+		else if (channel->rx_alloc_level > RX_ALLOC_LEVEL_MAX)
+			channel->rx_alloc_level = RX_ALLOC_LEVEL_MAX;
+
+		/* Decide on the allocation method */
+		method = ((channel->rx_alloc_level > RX_ALLOC_LEVEL_GRO) ?
+			  RX_ALLOC_METHOD_PAGE : RX_ALLOC_METHOD_SKB);
+	}
+
+	/* Push the option */
+	channel->rx_alloc_push_pages = (method == RX_ALLOC_METHOD_PAGE);
+}
+
+int efx_probe_rx_queue(struct efx_rx_queue *rx_queue)
+{
+	struct efx_nic *efx = rx_queue->efx;
+	unsigned int entries;
+	int rc;
+
+	/* Create the smallest power-of-two aligned ring */
+	entries = max(roundup_pow_of_two(efx->rxq_entries), EFX_MIN_DMAQ_SIZE);
+	EFX_BUG_ON_PARANOID(entries > EFX_MAX_DMAQ_SIZE);
+	rx_queue->ptr_mask = entries - 1;
+
+	netif_dbg(efx, probe, efx->net_dev,
+		  "creating RX queue %d size %#x mask %#x\n",
+		  efx_rx_queue_index(rx_queue), efx->rxq_entries,
+		  rx_queue->ptr_mask);
+
+	/* Allocate RX buffers */
+	rx_queue->buffer = kzalloc(entries * sizeof(*rx_queue->buffer),
+				   GFP_KERNEL);
+	if (!rx_queue->buffer)
+		return -ENOMEM;
+
+	rc = efx_nic_probe_rx(rx_queue);
+	if (rc) {
+		kfree(rx_queue->buffer);
+		rx_queue->buffer = NULL;
+	}
+	return rc;
+}
+
+void efx_init_rx_queue(struct efx_rx_queue *rx_queue)
+{
+	struct efx_nic *efx = rx_queue->efx;
+	unsigned int max_fill, trigger, limit;
+
+	netif_dbg(rx_queue->efx, drv, rx_queue->efx->net_dev,
+		  "initialising RX queue %d\n", efx_rx_queue_index(rx_queue));
+
+	/* Initialise ptr fields */
+	rx_queue->added_count = 0;
+	rx_queue->notified_count = 0;
+	rx_queue->removed_count = 0;
+	rx_queue->min_fill = -1U;
+
+	/* Initialise limit fields */
+	max_fill = efx->rxq_entries - EFX_RXD_HEAD_ROOM;
+	trigger = max_fill * min(rx_refill_threshold, 100U) / 100U;
+	limit = max_fill * min(rx_refill_limit, 100U) / 100U;
+
+	rx_queue->max_fill = max_fill;
+	rx_queue->fast_fill_trigger = trigger;
+	rx_queue->fast_fill_limit = limit;
+
+	/* Set up RX descriptor ring */
+	efx_nic_init_rx(rx_queue);
+}
+
+void efx_fini_rx_queue(struct efx_rx_queue *rx_queue)
+{
+	int i;
+	struct efx_rx_buffer *rx_buf;
+
+	netif_dbg(rx_queue->efx, drv, rx_queue->efx->net_dev,
+		  "shutting down RX queue %d\n", efx_rx_queue_index(rx_queue));
+
+	del_timer_sync(&rx_queue->slow_fill);
+	efx_nic_fini_rx(rx_queue);
+
+	/* Release RX buffers NB start at index 0 not current HW ptr */
+	if (rx_queue->buffer) {
+		for (i = 0; i <= rx_queue->ptr_mask; i++) {
+			rx_buf = efx_rx_buffer(rx_queue, i);
+			efx_fini_rx_buffer(rx_queue, rx_buf);
+		}
+	}
+}
+
+void efx_remove_rx_queue(struct efx_rx_queue *rx_queue)
+{
+	netif_dbg(rx_queue->efx, drv, rx_queue->efx->net_dev,
+		  "destroying RX queue %d\n", efx_rx_queue_index(rx_queue));
+
+	efx_nic_remove_rx(rx_queue);
+
+	kfree(rx_queue->buffer);
+	rx_queue->buffer = NULL;
+}
+
+
+module_param(rx_alloc_method, int, 0644);
+MODULE_PARM_DESC(rx_alloc_method, "Allocation method used for RX buffers");
+
+module_param(rx_refill_threshold, uint, 0444);
+MODULE_PARM_DESC(rx_refill_threshold,
+		 "RX descriptor ring fast/slow fill threshold (%)");
+
diff --git a/drivers/net/ethernet/sfc/selftest.c b/drivers/net/ethernet/sfc/selftest.c
new file mode 100644
index 0000000..822f6c2
--- /dev/null
+++ b/drivers/net/ethernet/sfc/selftest.c
@@ -0,0 +1,761 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2005-2006 Fen Systems Ltd.
+ * Copyright 2006-2010 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#include <linux/netdevice.h>
+#include <linux/module.h>
+#include <linux/delay.h>
+#include <linux/kernel_stat.h>
+#include <linux/pci.h>
+#include <linux/ethtool.h>
+#include <linux/ip.h>
+#include <linux/in.h>
+#include <linux/udp.h>
+#include <linux/rtnetlink.h>
+#include <linux/slab.h>
+#include <asm/io.h>
+#include "net_driver.h"
+#include "efx.h"
+#include "nic.h"
+#include "selftest.h"
+#include "workarounds.h"
+
+/*
+ * Loopback test packet structure
+ *
+ * The self-test should stress every RSS vector, and unfortunately
+ * Falcon only performs RSS on TCP/UDP packets.
+ */
+struct efx_loopback_payload {
+	struct ethhdr header;
+	struct iphdr ip;
+	struct udphdr udp;
+	__be16 iteration;
+	const char msg[64];
+} __packed;
+
+/* Loopback test source MAC address */
+static const unsigned char payload_source[ETH_ALEN] = {
+	0x00, 0x0f, 0x53, 0x1b, 0x1b, 0x1b,
+};
+
+static const char payload_msg[] =
+	"Hello world! This is an Efx loopback test in progress!";
+
+/* Interrupt mode names */
+static const unsigned int efx_interrupt_mode_max = EFX_INT_MODE_MAX;
+static const char *efx_interrupt_mode_names[] = {
+	[EFX_INT_MODE_MSIX]   = "MSI-X",
+	[EFX_INT_MODE_MSI]    = "MSI",
+	[EFX_INT_MODE_LEGACY] = "legacy",
+};
+#define INT_MODE(efx) \
+	STRING_TABLE_LOOKUP(efx->interrupt_mode, efx_interrupt_mode)
+
+/**
+ * efx_loopback_state - persistent state during a loopback selftest
+ * @flush:		Drop all packets in efx_loopback_rx_packet
+ * @packet_count:	Number of packets being used in this test
+ * @skbs:		An array of skbs transmitted
+ * @offload_csum:	Checksums are being offloaded
+ * @rx_good:		RX good packet count
+ * @rx_bad:		RX bad packet count
+ * @payload:		Payload used in tests
+ */
+struct efx_loopback_state {
+	bool flush;
+	int packet_count;
+	struct sk_buff **skbs;
+	bool offload_csum;
+	atomic_t rx_good;
+	atomic_t rx_bad;
+	struct efx_loopback_payload payload;
+};
+
+/**************************************************************************
+ *
+ * MII, NVRAM and register tests
+ *
+ **************************************************************************/
+
+static int efx_test_phy_alive(struct efx_nic *efx, struct efx_self_tests *tests)
+{
+	int rc = 0;
+
+	if (efx->phy_op->test_alive) {
+		rc = efx->phy_op->test_alive(efx);
+		tests->phy_alive = rc ? -1 : 1;
+	}
+
+	return rc;
+}
+
+static int efx_test_nvram(struct efx_nic *efx, struct efx_self_tests *tests)
+{
+	int rc = 0;
+
+	if (efx->type->test_nvram) {
+		rc = efx->type->test_nvram(efx);
+		tests->nvram = rc ? -1 : 1;
+	}
+
+	return rc;
+}
+
+static int efx_test_chip(struct efx_nic *efx, struct efx_self_tests *tests)
+{
+	int rc = 0;
+
+	/* Test register access */
+	if (efx->type->test_registers) {
+		rc = efx->type->test_registers(efx);
+		tests->registers = rc ? -1 : 1;
+	}
+
+	return rc;
+}
+
+/**************************************************************************
+ *
+ * Interrupt and event queue testing
+ *
+ **************************************************************************/
+
+/* Test generation and receipt of interrupts */
+static int efx_test_interrupts(struct efx_nic *efx,
+			       struct efx_self_tests *tests)
+{
+	netif_dbg(efx, drv, efx->net_dev, "testing interrupts\n");
+	tests->interrupt = -1;
+
+	/* Reset interrupt flag */
+	efx->last_irq_cpu = -1;
+	smp_wmb();
+
+	efx_nic_generate_interrupt(efx);
+
+	/* Wait for arrival of test interrupt. */
+	netif_dbg(efx, drv, efx->net_dev, "waiting for test interrupt\n");
+	schedule_timeout_uninterruptible(HZ / 10);
+	if (efx->last_irq_cpu >= 0)
+		goto success;
+
+	netif_err(efx, drv, efx->net_dev, "timed out waiting for interrupt\n");
+	return -ETIMEDOUT;
+
+ success:
+	netif_dbg(efx, drv, efx->net_dev, "%s test interrupt seen on CPU%d\n",
+		  INT_MODE(efx),
+		efx->last_irq_cpu);
+	tests->interrupt = 1;
+	return 0;
+}
+
+/* Test generation and receipt of interrupting events */
+static int efx_test_eventq_irq(struct efx_channel *channel,
+			       struct efx_self_tests *tests)
+{
+	struct efx_nic *efx = channel->efx;
+	unsigned int read_ptr, count;
+
+	tests->eventq_dma[channel->channel] = -1;
+	tests->eventq_int[channel->channel] = -1;
+	tests->eventq_poll[channel->channel] = -1;
+
+	read_ptr = channel->eventq_read_ptr;
+	channel->efx->last_irq_cpu = -1;
+	smp_wmb();
+
+	efx_nic_generate_test_event(channel);
+
+	/* Wait for arrival of interrupt */
+	count = 0;
+	do {
+		schedule_timeout_uninterruptible(HZ / 100);
+
+		if (ACCESS_ONCE(channel->eventq_read_ptr) != read_ptr)
+			goto eventq_ok;
+	} while (++count < 2);
+
+	netif_err(efx, drv, efx->net_dev,
+		  "channel %d timed out waiting for event queue\n",
+		  channel->channel);
+
+	/* See if interrupt arrived */
+	if (channel->efx->last_irq_cpu >= 0) {
+		netif_err(efx, drv, efx->net_dev,
+			  "channel %d saw interrupt on CPU%d "
+			  "during event queue test\n", channel->channel,
+			  raw_smp_processor_id());
+		tests->eventq_int[channel->channel] = 1;
+	}
+
+	/* Check to see if event was received even if interrupt wasn't */
+	if (efx_nic_event_present(channel)) {
+		netif_err(efx, drv, efx->net_dev,
+			  "channel %d event was generated, but "
+			  "failed to trigger an interrupt\n", channel->channel);
+		tests->eventq_dma[channel->channel] = 1;
+	}
+
+	return -ETIMEDOUT;
+ eventq_ok:
+	netif_dbg(efx, drv, efx->net_dev, "channel %d event queue passed\n",
+		  channel->channel);
+	tests->eventq_dma[channel->channel] = 1;
+	tests->eventq_int[channel->channel] = 1;
+	tests->eventq_poll[channel->channel] = 1;
+	return 0;
+}
+
+static int efx_test_phy(struct efx_nic *efx, struct efx_self_tests *tests,
+			unsigned flags)
+{
+	int rc;
+
+	if (!efx->phy_op->run_tests)
+		return 0;
+
+	mutex_lock(&efx->mac_lock);
+	rc = efx->phy_op->run_tests(efx, tests->phy_ext, flags);
+	mutex_unlock(&efx->mac_lock);
+	return rc;
+}
+
+/**************************************************************************
+ *
+ * Loopback testing
+ * NB Only one loopback test can be executing concurrently.
+ *
+ **************************************************************************/
+
+/* Loopback test RX callback
+ * This is called for each received packet during loopback testing.
+ */
+void efx_loopback_rx_packet(struct efx_nic *efx,
+			    const char *buf_ptr, int pkt_len)
+{
+	struct efx_loopback_state *state = efx->loopback_selftest;
+	struct efx_loopback_payload *received;
+	struct efx_loopback_payload *payload;
+
+	BUG_ON(!buf_ptr);
+
+	/* If we are just flushing, then drop the packet */
+	if ((state == NULL) || state->flush)
+		return;
+
+	payload = &state->payload;
+
+	received = (struct efx_loopback_payload *) buf_ptr;
+	received->ip.saddr = payload->ip.saddr;
+	if (state->offload_csum)
+		received->ip.check = payload->ip.check;
+
+	/* Check that header exists */
+	if (pkt_len < sizeof(received->header)) {
+		netif_err(efx, drv, efx->net_dev,
+			  "saw runt RX packet (length %d) in %s loopback "
+			  "test\n", pkt_len, LOOPBACK_MODE(efx));
+		goto err;
+	}
+
+	/* Check that the ethernet header exists */
+	if (memcmp(&received->header, &payload->header, ETH_HLEN) != 0) {
+		netif_err(efx, drv, efx->net_dev,
+			  "saw non-loopback RX packet in %s loopback test\n",
+			  LOOPBACK_MODE(efx));
+		goto err;
+	}
+
+	/* Check packet length */
+	if (pkt_len != sizeof(*payload)) {
+		netif_err(efx, drv, efx->net_dev,
+			  "saw incorrect RX packet length %d (wanted %d) in "
+			  "%s loopback test\n", pkt_len, (int)sizeof(*payload),
+			  LOOPBACK_MODE(efx));
+		goto err;
+	}
+
+	/* Check that IP header matches */
+	if (memcmp(&received->ip, &payload->ip, sizeof(payload->ip)) != 0) {
+		netif_err(efx, drv, efx->net_dev,
+			  "saw corrupted IP header in %s loopback test\n",
+			  LOOPBACK_MODE(efx));
+		goto err;
+	}
+
+	/* Check that msg and padding matches */
+	if (memcmp(&received->msg, &payload->msg, sizeof(received->msg)) != 0) {
+		netif_err(efx, drv, efx->net_dev,
+			  "saw corrupted RX packet in %s loopback test\n",
+			  LOOPBACK_MODE(efx));
+		goto err;
+	}
+
+	/* Check that iteration matches */
+	if (received->iteration != payload->iteration) {
+		netif_err(efx, drv, efx->net_dev,
+			  "saw RX packet from iteration %d (wanted %d) in "
+			  "%s loopback test\n", ntohs(received->iteration),
+			  ntohs(payload->iteration), LOOPBACK_MODE(efx));
+		goto err;
+	}
+
+	/* Increase correct RX count */
+	netif_vdbg(efx, drv, efx->net_dev,
+		   "got loopback RX in %s loopback test\n", LOOPBACK_MODE(efx));
+
+	atomic_inc(&state->rx_good);
+	return;
+
+ err:
+#ifdef EFX_ENABLE_DEBUG
+	if (atomic_read(&state->rx_bad) == 0) {
+		netif_err(efx, drv, efx->net_dev, "received packet:\n");
+		print_hex_dump(KERN_ERR, "", DUMP_PREFIX_OFFSET, 0x10, 1,
+			       buf_ptr, pkt_len, 0);
+		netif_err(efx, drv, efx->net_dev, "expected packet:\n");
+		print_hex_dump(KERN_ERR, "", DUMP_PREFIX_OFFSET, 0x10, 1,
+			       &state->payload, sizeof(state->payload), 0);
+	}
+#endif
+	atomic_inc(&state->rx_bad);
+}
+
+/* Initialise an efx_selftest_state for a new iteration */
+static void efx_iterate_state(struct efx_nic *efx)
+{
+	struct efx_loopback_state *state = efx->loopback_selftest;
+	struct net_device *net_dev = efx->net_dev;
+	struct efx_loopback_payload *payload = &state->payload;
+
+	/* Initialise the layerII header */
+	memcpy(&payload->header.h_dest, net_dev->dev_addr, ETH_ALEN);
+	memcpy(&payload->header.h_source, &payload_source, ETH_ALEN);
+	payload->header.h_proto = htons(ETH_P_IP);
+
+	/* saddr set later and used as incrementing count */
+	payload->ip.daddr = htonl(INADDR_LOOPBACK);
+	payload->ip.ihl = 5;
+	payload->ip.check = htons(0xdead);
+	payload->ip.tot_len = htons(sizeof(*payload) - sizeof(struct ethhdr));
+	payload->ip.version = IPVERSION;
+	payload->ip.protocol = IPPROTO_UDP;
+
+	/* Initialise udp header */
+	payload->udp.source = 0;
+	payload->udp.len = htons(sizeof(*payload) - sizeof(struct ethhdr) -
+				 sizeof(struct iphdr));
+	payload->udp.check = 0;	/* checksum ignored */
+
+	/* Fill out payload */
+	payload->iteration = htons(ntohs(payload->iteration) + 1);
+	memcpy(&payload->msg, payload_msg, sizeof(payload_msg));
+
+	/* Fill out remaining state members */
+	atomic_set(&state->rx_good, 0);
+	atomic_set(&state->rx_bad, 0);
+	smp_wmb();
+}
+
+static int efx_begin_loopback(struct efx_tx_queue *tx_queue)
+{
+	struct efx_nic *efx = tx_queue->efx;
+	struct efx_loopback_state *state = efx->loopback_selftest;
+	struct efx_loopback_payload *payload;
+	struct sk_buff *skb;
+	int i;
+	netdev_tx_t rc;
+
+	/* Transmit N copies of buffer */
+	for (i = 0; i < state->packet_count; i++) {
+		/* Allocate an skb, holding an extra reference for
+		 * transmit completion counting */
+		skb = alloc_skb(sizeof(state->payload), GFP_KERNEL);
+		if (!skb)
+			return -ENOMEM;
+		state->skbs[i] = skb;
+		skb_get(skb);
+
+		/* Copy the payload in, incrementing the source address to
+		 * exercise the rss vectors */
+		payload = ((struct efx_loopback_payload *)
+			   skb_put(skb, sizeof(state->payload)));
+		memcpy(payload, &state->payload, sizeof(state->payload));
+		payload->ip.saddr = htonl(INADDR_LOOPBACK | (i << 2));
+
+		/* Ensure everything we've written is visible to the
+		 * interrupt handler. */
+		smp_wmb();
+
+		if (efx_dev_registered(efx))
+			netif_tx_lock_bh(efx->net_dev);
+		rc = efx_enqueue_skb(tx_queue, skb);
+		if (efx_dev_registered(efx))
+			netif_tx_unlock_bh(efx->net_dev);
+
+		if (rc != NETDEV_TX_OK) {
+			netif_err(efx, drv, efx->net_dev,
+				  "TX queue %d could not transmit packet %d of "
+				  "%d in %s loopback test\n", tx_queue->queue,
+				  i + 1, state->packet_count,
+				  LOOPBACK_MODE(efx));
+
+			/* Defer cleaning up the other skbs for the caller */
+			kfree_skb(skb);
+			return -EPIPE;
+		}
+	}
+
+	return 0;
+}
+
+static int efx_poll_loopback(struct efx_nic *efx)
+{
+	struct efx_loopback_state *state = efx->loopback_selftest;
+	struct efx_channel *channel;
+
+	/* NAPI polling is not enabled, so process channels
+	 * synchronously */
+	efx_for_each_channel(channel, efx) {
+		if (channel->work_pending)
+			efx_process_channel_now(channel);
+	}
+	return atomic_read(&state->rx_good) == state->packet_count;
+}
+
+static int efx_end_loopback(struct efx_tx_queue *tx_queue,
+			    struct efx_loopback_self_tests *lb_tests)
+{
+	struct efx_nic *efx = tx_queue->efx;
+	struct efx_loopback_state *state = efx->loopback_selftest;
+	struct sk_buff *skb;
+	int tx_done = 0, rx_good, rx_bad;
+	int i, rc = 0;
+
+	if (efx_dev_registered(efx))
+		netif_tx_lock_bh(efx->net_dev);
+
+	/* Count the number of tx completions, and decrement the refcnt. Any
+	 * skbs not already completed will be free'd when the queue is flushed */
+	for (i=0; i < state->packet_count; i++) {
+		skb = state->skbs[i];
+		if (skb && !skb_shared(skb))
+			++tx_done;
+		dev_kfree_skb_any(skb);
+	}
+
+	if (efx_dev_registered(efx))
+		netif_tx_unlock_bh(efx->net_dev);
+
+	/* Check TX completion and received packet counts */
+	rx_good = atomic_read(&state->rx_good);
+	rx_bad = atomic_read(&state->rx_bad);
+	if (tx_done != state->packet_count) {
+		/* Don't free the skbs; they will be picked up on TX
+		 * overflow or channel teardown.
+		 */
+		netif_err(efx, drv, efx->net_dev,
+			  "TX queue %d saw only %d out of an expected %d "
+			  "TX completion events in %s loopback test\n",
+			  tx_queue->queue, tx_done, state->packet_count,
+			  LOOPBACK_MODE(efx));
+		rc = -ETIMEDOUT;
+		/* Allow to fall through so we see the RX errors as well */
+	}
+
+	/* We may always be up to a flush away from our desired packet total */
+	if (rx_good != state->packet_count) {
+		netif_dbg(efx, drv, efx->net_dev,
+			  "TX queue %d saw only %d out of an expected %d "
+			  "received packets in %s loopback test\n",
+			  tx_queue->queue, rx_good, state->packet_count,
+			  LOOPBACK_MODE(efx));
+		rc = -ETIMEDOUT;
+		/* Fall through */
+	}
+
+	/* Update loopback test structure */
+	lb_tests->tx_sent[tx_queue->queue] += state->packet_count;
+	lb_tests->tx_done[tx_queue->queue] += tx_done;
+	lb_tests->rx_good += rx_good;
+	lb_tests->rx_bad += rx_bad;
+
+	return rc;
+}
+
+static int
+efx_test_loopback(struct efx_tx_queue *tx_queue,
+		  struct efx_loopback_self_tests *lb_tests)
+{
+	struct efx_nic *efx = tx_queue->efx;
+	struct efx_loopback_state *state = efx->loopback_selftest;
+	int i, begin_rc, end_rc;
+
+	for (i = 0; i < 3; i++) {
+		/* Determine how many packets to send */
+		state->packet_count = efx->txq_entries / 3;
+		state->packet_count = min(1 << (i << 2), state->packet_count);
+		state->skbs = kzalloc(sizeof(state->skbs[0]) *
+				      state->packet_count, GFP_KERNEL);
+		if (!state->skbs)
+			return -ENOMEM;
+		state->flush = false;
+
+		netif_dbg(efx, drv, efx->net_dev,
+			  "TX queue %d testing %s loopback with %d packets\n",
+			  tx_queue->queue, LOOPBACK_MODE(efx),
+			  state->packet_count);
+
+		efx_iterate_state(efx);
+		begin_rc = efx_begin_loopback(tx_queue);
+
+		/* This will normally complete very quickly, but be
+		 * prepared to wait up to 100 ms. */
+		msleep(1);
+		if (!efx_poll_loopback(efx)) {
+			msleep(100);
+			efx_poll_loopback(efx);
+		}
+
+		end_rc = efx_end_loopback(tx_queue, lb_tests);
+		kfree(state->skbs);
+
+		if (begin_rc || end_rc) {
+			/* Wait a while to ensure there are no packets
+			 * floating around after a failure. */
+			schedule_timeout_uninterruptible(HZ / 10);
+			return begin_rc ? begin_rc : end_rc;
+		}
+	}
+
+	netif_dbg(efx, drv, efx->net_dev,
+		  "TX queue %d passed %s loopback test with a burst length "
+		  "of %d packets\n", tx_queue->queue, LOOPBACK_MODE(efx),
+		  state->packet_count);
+
+	return 0;
+}
+
+/* Wait for link up. On Falcon, we would prefer to rely on efx_monitor, but
+ * any contention on the mac lock (via e.g. efx_mac_mcast_work) causes it
+ * to delay and retry. Therefore, it's safer to just poll directly. Wait
+ * for link up and any faults to dissipate. */
+static int efx_wait_for_link(struct efx_nic *efx)
+{
+	struct efx_link_state *link_state = &efx->link_state;
+	int count, link_up_count = 0;
+	bool link_up;
+
+	for (count = 0; count < 40; count++) {
+		schedule_timeout_uninterruptible(HZ / 10);
+
+		if (efx->type->monitor != NULL) {
+			mutex_lock(&efx->mac_lock);
+			efx->type->monitor(efx);
+			mutex_unlock(&efx->mac_lock);
+		} else {
+			struct efx_channel *channel = efx_get_channel(efx, 0);
+			if (channel->work_pending)
+				efx_process_channel_now(channel);
+		}
+
+		mutex_lock(&efx->mac_lock);
+		link_up = link_state->up;
+		if (link_up)
+			link_up = !efx->mac_op->check_fault(efx);
+		mutex_unlock(&efx->mac_lock);
+
+		if (link_up) {
+			if (++link_up_count == 2)
+				return 0;
+		} else {
+			link_up_count = 0;
+		}
+	}
+
+	return -ETIMEDOUT;
+}
+
+static int efx_test_loopbacks(struct efx_nic *efx, struct efx_self_tests *tests,
+			      unsigned int loopback_modes)
+{
+	enum efx_loopback_mode mode;
+	struct efx_loopback_state *state;
+	struct efx_channel *channel = efx_get_channel(efx, 0);
+	struct efx_tx_queue *tx_queue;
+	int rc = 0;
+
+	/* Set the port loopback_selftest member. From this point on
+	 * all received packets will be dropped. Mark the state as
+	 * "flushing" so all inflight packets are dropped */
+	state = kzalloc(sizeof(*state), GFP_KERNEL);
+	if (state == NULL)
+		return -ENOMEM;
+	BUG_ON(efx->loopback_selftest);
+	state->flush = true;
+	efx->loopback_selftest = state;
+
+	/* Test all supported loopback modes */
+	for (mode = LOOPBACK_NONE; mode <= LOOPBACK_TEST_MAX; mode++) {
+		if (!(loopback_modes & (1 << mode)))
+			continue;
+
+		/* Move the port into the specified loopback mode. */
+		state->flush = true;
+		mutex_lock(&efx->mac_lock);
+		efx->loopback_mode = mode;
+		rc = __efx_reconfigure_port(efx);
+		mutex_unlock(&efx->mac_lock);
+		if (rc) {
+			netif_err(efx, drv, efx->net_dev,
+				  "unable to move into %s loopback\n",
+				  LOOPBACK_MODE(efx));
+			goto out;
+		}
+
+		rc = efx_wait_for_link(efx);
+		if (rc) {
+			netif_err(efx, drv, efx->net_dev,
+				  "loopback %s never came up\n",
+				  LOOPBACK_MODE(efx));
+			goto out;
+		}
+
+		/* Test all enabled types of TX queue */
+		efx_for_each_channel_tx_queue(tx_queue, channel) {
+			state->offload_csum = (tx_queue->queue &
+					       EFX_TXQ_TYPE_OFFLOAD);
+			rc = efx_test_loopback(tx_queue,
+					       &tests->loopback[mode]);
+			if (rc)
+				goto out;
+		}
+	}
+
+ out:
+	/* Remove the flush. The caller will remove the loopback setting */
+	state->flush = true;
+	efx->loopback_selftest = NULL;
+	wmb();
+	kfree(state);
+
+	return rc;
+}
+
+/**************************************************************************
+ *
+ * Entry point
+ *
+ *************************************************************************/
+
+int efx_selftest(struct efx_nic *efx, struct efx_self_tests *tests,
+		 unsigned flags)
+{
+	enum efx_loopback_mode loopback_mode = efx->loopback_mode;
+	int phy_mode = efx->phy_mode;
+	enum reset_type reset_method = RESET_TYPE_INVISIBLE;
+	struct efx_channel *channel;
+	int rc_test = 0, rc_reset = 0, rc;
+
+	/* Online (i.e. non-disruptive) testing
+	 * This checks interrupt generation, event delivery and PHY presence. */
+
+	rc = efx_test_phy_alive(efx, tests);
+	if (rc && !rc_test)
+		rc_test = rc;
+
+	rc = efx_test_nvram(efx, tests);
+	if (rc && !rc_test)
+		rc_test = rc;
+
+	rc = efx_test_interrupts(efx, tests);
+	if (rc && !rc_test)
+		rc_test = rc;
+
+	efx_for_each_channel(channel, efx) {
+		rc = efx_test_eventq_irq(channel, tests);
+		if (rc && !rc_test)
+			rc_test = rc;
+	}
+
+	if (rc_test)
+		return rc_test;
+
+	if (!(flags & ETH_TEST_FL_OFFLINE))
+		return efx_test_phy(efx, tests, flags);
+
+	/* Offline (i.e. disruptive) testing
+	 * This checks MAC and PHY loopback on the specified port. */
+
+	/* Detach the device so the kernel doesn't transmit during the
+	 * loopback test and the watchdog timeout doesn't fire.
+	 */
+	netif_device_detach(efx->net_dev);
+
+	mutex_lock(&efx->mac_lock);
+	if (efx->loopback_modes) {
+		/* We need the 312 clock from the PHY to test the XMAC
+		 * registers, so move into XGMII loopback if available */
+		if (efx->loopback_modes & (1 << LOOPBACK_XGMII))
+			efx->loopback_mode = LOOPBACK_XGMII;
+		else
+			efx->loopback_mode = __ffs(efx->loopback_modes);
+	}
+
+	__efx_reconfigure_port(efx);
+	mutex_unlock(&efx->mac_lock);
+
+	/* free up all consumers of SRAM (including all the queues) */
+	efx_reset_down(efx, reset_method);
+
+	rc = efx_test_chip(efx, tests);
+	if (rc && !rc_test)
+		rc_test = rc;
+
+	/* reset the chip to recover from the register test */
+	rc_reset = efx->type->reset(efx, reset_method);
+
+	/* Ensure that the phy is powered and out of loopback
+	 * for the bist and loopback tests */
+	efx->phy_mode &= ~PHY_MODE_LOW_POWER;
+	efx->loopback_mode = LOOPBACK_NONE;
+
+	rc = efx_reset_up(efx, reset_method, rc_reset == 0);
+	if (rc && !rc_reset)
+		rc_reset = rc;
+
+	if (rc_reset) {
+		netif_err(efx, drv, efx->net_dev,
+			  "Unable to recover from chip test\n");
+		efx_schedule_reset(efx, RESET_TYPE_DISABLE);
+		return rc_reset;
+	}
+
+	rc = efx_test_phy(efx, tests, flags);
+	if (rc && !rc_test)
+		rc_test = rc;
+
+	rc = efx_test_loopbacks(efx, tests, efx->loopback_modes);
+	if (rc && !rc_test)
+		rc_test = rc;
+
+	/* restore the PHY to the previous state */
+	mutex_lock(&efx->mac_lock);
+	efx->phy_mode = phy_mode;
+	efx->loopback_mode = loopback_mode;
+	__efx_reconfigure_port(efx);
+	mutex_unlock(&efx->mac_lock);
+
+	netif_device_attach(efx->net_dev);
+
+	return rc_test;
+}
+
diff --git a/drivers/net/ethernet/sfc/selftest.h b/drivers/net/ethernet/sfc/selftest.h
new file mode 100644
index 0000000..dba5456
--- /dev/null
+++ b/drivers/net/ethernet/sfc/selftest.h
@@ -0,0 +1,53 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2005-2006 Fen Systems Ltd.
+ * Copyright 2006-2010 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#ifndef EFX_SELFTEST_H
+#define EFX_SELFTEST_H
+
+#include "net_driver.h"
+
+/*
+ * Self tests
+ */
+
+struct efx_loopback_self_tests {
+	int tx_sent[EFX_TXQ_TYPES];
+	int tx_done[EFX_TXQ_TYPES];
+	int rx_good;
+	int rx_bad;
+};
+
+#define EFX_MAX_PHY_TESTS 20
+
+/* Efx self test results
+ * For fields which are not counters, 1 indicates success and -1
+ * indicates failure.
+ */
+struct efx_self_tests {
+	/* online tests */
+	int phy_alive;
+	int nvram;
+	int interrupt;
+	int eventq_dma[EFX_MAX_CHANNELS];
+	int eventq_int[EFX_MAX_CHANNELS];
+	int eventq_poll[EFX_MAX_CHANNELS];
+	/* offline tests */
+	int registers;
+	int phy_ext[EFX_MAX_PHY_TESTS];
+	struct efx_loopback_self_tests loopback[LOOPBACK_TEST_MAX + 1];
+};
+
+extern void efx_loopback_rx_packet(struct efx_nic *efx,
+				   const char *buf_ptr, int pkt_len);
+extern int efx_selftest(struct efx_nic *efx,
+			struct efx_self_tests *tests,
+			unsigned flags);
+
+#endif /* EFX_SELFTEST_H */
diff --git a/drivers/net/ethernet/sfc/siena.c b/drivers/net/ethernet/sfc/siena.c
new file mode 100644
index 0000000..5735e84
--- /dev/null
+++ b/drivers/net/ethernet/sfc/siena.c
@@ -0,0 +1,676 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2005-2006 Fen Systems Ltd.
+ * Copyright 2006-2010 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#include <linux/bitops.h>
+#include <linux/delay.h>
+#include <linux/pci.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/random.h>
+#include "net_driver.h"
+#include "bitfield.h"
+#include "efx.h"
+#include "nic.h"
+#include "mac.h"
+#include "spi.h"
+#include "regs.h"
+#include "io.h"
+#include "phy.h"
+#include "workarounds.h"
+#include "mcdi.h"
+#include "mcdi_pcol.h"
+
+/* Hardware control for SFC9000 family including SFL9021 (aka Siena). */
+
+static void siena_init_wol(struct efx_nic *efx);
+
+
+static void siena_push_irq_moderation(struct efx_channel *channel)
+{
+	efx_dword_t timer_cmd;
+
+	if (channel->irq_moderation)
+		EFX_POPULATE_DWORD_2(timer_cmd,
+				     FRF_CZ_TC_TIMER_MODE,
+				     FFE_CZ_TIMER_MODE_INT_HLDOFF,
+				     FRF_CZ_TC_TIMER_VAL,
+				     channel->irq_moderation - 1);
+	else
+		EFX_POPULATE_DWORD_2(timer_cmd,
+				     FRF_CZ_TC_TIMER_MODE,
+				     FFE_CZ_TIMER_MODE_DIS,
+				     FRF_CZ_TC_TIMER_VAL, 0);
+	efx_writed_page_locked(channel->efx, &timer_cmd, FR_BZ_TIMER_COMMAND_P0,
+			       channel->channel);
+}
+
+static void siena_push_multicast_hash(struct efx_nic *efx)
+{
+	WARN_ON(!mutex_is_locked(&efx->mac_lock));
+
+	efx_mcdi_rpc(efx, MC_CMD_SET_MCAST_HASH,
+		     efx->multicast_hash.byte, sizeof(efx->multicast_hash),
+		     NULL, 0, NULL);
+}
+
+static int siena_mdio_write(struct net_device *net_dev,
+			    int prtad, int devad, u16 addr, u16 value)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	uint32_t status;
+	int rc;
+
+	rc = efx_mcdi_mdio_write(efx, efx->mdio_bus, prtad, devad,
+				 addr, value, &status);
+	if (rc)
+		return rc;
+	if (status != MC_CMD_MDIO_STATUS_GOOD)
+		return -EIO;
+
+	return 0;
+}
+
+static int siena_mdio_read(struct net_device *net_dev,
+			   int prtad, int devad, u16 addr)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	uint16_t value;
+	uint32_t status;
+	int rc;
+
+	rc = efx_mcdi_mdio_read(efx, efx->mdio_bus, prtad, devad,
+				addr, &value, &status);
+	if (rc)
+		return rc;
+	if (status != MC_CMD_MDIO_STATUS_GOOD)
+		return -EIO;
+
+	return (int)value;
+}
+
+/* This call is responsible for hooking in the MAC and PHY operations */
+static int siena_probe_port(struct efx_nic *efx)
+{
+	int rc;
+
+	/* Hook in PHY operations table */
+	efx->phy_op = &efx_mcdi_phy_ops;
+
+	/* Set up MDIO structure for PHY */
+	efx->mdio.mode_support = MDIO_SUPPORTS_C45 | MDIO_EMULATE_C22;
+	efx->mdio.mdio_read = siena_mdio_read;
+	efx->mdio.mdio_write = siena_mdio_write;
+
+	/* Fill out MDIO structure, loopback modes, and initial link state */
+	rc = efx->phy_op->probe(efx);
+	if (rc != 0)
+		return rc;
+
+	/* Allocate buffer for stats */
+	rc = efx_nic_alloc_buffer(efx, &efx->stats_buffer,
+				  MC_CMD_MAC_NSTATS * sizeof(u64));
+	if (rc)
+		return rc;
+	netif_dbg(efx, probe, efx->net_dev,
+		  "stats buffer at %llx (virt %p phys %llx)\n",
+		  (u64)efx->stats_buffer.dma_addr,
+		  efx->stats_buffer.addr,
+		  (u64)virt_to_phys(efx->stats_buffer.addr));
+
+	efx_mcdi_mac_stats(efx, efx->stats_buffer.dma_addr, 0, 0, 1);
+
+	return 0;
+}
+
+static void siena_remove_port(struct efx_nic *efx)
+{
+	efx->phy_op->remove(efx);
+	efx_nic_free_buffer(efx, &efx->stats_buffer);
+}
+
+static const struct efx_nic_register_test siena_register_tests[] = {
+	{ FR_AZ_ADR_REGION,
+	  EFX_OWORD32(0x0003FFFF, 0x0003FFFF, 0x0003FFFF, 0x0003FFFF) },
+	{ FR_CZ_USR_EV_CFG,
+	  EFX_OWORD32(0x000103FF, 0x00000000, 0x00000000, 0x00000000) },
+	{ FR_AZ_RX_CFG,
+	  EFX_OWORD32(0xFFFFFFFE, 0xFFFFFFFF, 0x0003FFFF, 0x00000000) },
+	{ FR_AZ_TX_CFG,
+	  EFX_OWORD32(0x7FFF0037, 0xFFFF8000, 0xFFFFFFFF, 0x03FFFFFF) },
+	{ FR_AZ_TX_RESERVED,
+	  EFX_OWORD32(0xFFFEFE80, 0x1FFFFFFF, 0x020000FE, 0x007FFFFF) },
+	{ FR_AZ_SRM_TX_DC_CFG,
+	  EFX_OWORD32(0x001FFFFF, 0x00000000, 0x00000000, 0x00000000) },
+	{ FR_AZ_RX_DC_CFG,
+	  EFX_OWORD32(0x00000003, 0x00000000, 0x00000000, 0x00000000) },
+	{ FR_AZ_RX_DC_PF_WM,
+	  EFX_OWORD32(0x000003FF, 0x00000000, 0x00000000, 0x00000000) },
+	{ FR_BZ_DP_CTRL,
+	  EFX_OWORD32(0x00000FFF, 0x00000000, 0x00000000, 0x00000000) },
+	{ FR_BZ_RX_RSS_TKEY,
+	  EFX_OWORD32(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF) },
+	{ FR_CZ_RX_RSS_IPV6_REG1,
+	  EFX_OWORD32(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF) },
+	{ FR_CZ_RX_RSS_IPV6_REG2,
+	  EFX_OWORD32(0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF) },
+	{ FR_CZ_RX_RSS_IPV6_REG3,
+	  EFX_OWORD32(0xFFFFFFFF, 0xFFFFFFFF, 0x00000007, 0x00000000) },
+};
+
+static int siena_test_registers(struct efx_nic *efx)
+{
+	return efx_nic_test_registers(efx, siena_register_tests,
+				      ARRAY_SIZE(siena_register_tests));
+}
+
+/**************************************************************************
+ *
+ * Device reset
+ *
+ **************************************************************************
+ */
+
+static enum reset_type siena_map_reset_reason(enum reset_type reason)
+{
+	return RESET_TYPE_ALL;
+}
+
+static int siena_map_reset_flags(u32 *flags)
+{
+	enum {
+		SIENA_RESET_PORT = (ETH_RESET_DMA | ETH_RESET_FILTER |
+				    ETH_RESET_OFFLOAD | ETH_RESET_MAC |
+				    ETH_RESET_PHY),
+		SIENA_RESET_MC = (SIENA_RESET_PORT |
+				  ETH_RESET_MGMT << ETH_RESET_SHARED_SHIFT),
+	};
+
+	if ((*flags & SIENA_RESET_MC) == SIENA_RESET_MC) {
+		*flags &= ~SIENA_RESET_MC;
+		return RESET_TYPE_WORLD;
+	}
+
+	if ((*flags & SIENA_RESET_PORT) == SIENA_RESET_PORT) {
+		*flags &= ~SIENA_RESET_PORT;
+		return RESET_TYPE_ALL;
+	}
+
+	/* no invisible reset implemented */
+
+	return -EINVAL;
+}
+
+static int siena_reset_hw(struct efx_nic *efx, enum reset_type method)
+{
+	int rc;
+
+	/* Recover from a failed assertion pre-reset */
+	rc = efx_mcdi_handle_assertion(efx);
+	if (rc)
+		return rc;
+
+	if (method == RESET_TYPE_WORLD)
+		return efx_mcdi_reset_mc(efx);
+	else
+		return efx_mcdi_reset_port(efx);
+}
+
+static int siena_probe_nvconfig(struct efx_nic *efx)
+{
+	return efx_mcdi_get_board_cfg(efx, efx->net_dev->perm_addr, NULL);
+}
+
+static int siena_probe_nic(struct efx_nic *efx)
+{
+	struct siena_nic_data *nic_data;
+	bool already_attached = 0;
+	efx_oword_t reg;
+	int rc;
+
+	/* Allocate storage for hardware specific data */
+	nic_data = kzalloc(sizeof(struct siena_nic_data), GFP_KERNEL);
+	if (!nic_data)
+		return -ENOMEM;
+	efx->nic_data = nic_data;
+
+	if (efx_nic_fpga_ver(efx) != 0) {
+		netif_err(efx, probe, efx->net_dev,
+			  "Siena FPGA not supported\n");
+		rc = -ENODEV;
+		goto fail1;
+	}
+
+	efx_reado(efx, &reg, FR_AZ_CS_DEBUG);
+	efx->net_dev->dev_id = EFX_OWORD_FIELD(reg, FRF_CZ_CS_PORT_NUM) - 1;
+
+	/* Initialise MCDI */
+	nic_data->mcdi_smem = ioremap_nocache(efx->membase_phys +
+					      FR_CZ_MC_TREG_SMEM,
+					      FR_CZ_MC_TREG_SMEM_STEP *
+					      FR_CZ_MC_TREG_SMEM_ROWS);
+	if (!nic_data->mcdi_smem) {
+		netif_err(efx, probe, efx->net_dev,
+			  "could not map MCDI at %llx+%x\n",
+			  (unsigned long long)efx->membase_phys +
+			  FR_CZ_MC_TREG_SMEM,
+			  FR_CZ_MC_TREG_SMEM_STEP * FR_CZ_MC_TREG_SMEM_ROWS);
+		rc = -ENOMEM;
+		goto fail1;
+	}
+	efx_mcdi_init(efx);
+
+	/* Recover from a failed assertion before probing */
+	rc = efx_mcdi_handle_assertion(efx);
+	if (rc)
+		goto fail2;
+
+	/* Let the BMC know that the driver is now in charge of link and
+	 * filter settings. We must do this before we reset the NIC */
+	rc = efx_mcdi_drv_attach(efx, true, &already_attached);
+	if (rc) {
+		netif_err(efx, probe, efx->net_dev,
+			  "Unable to register driver with MCPU\n");
+		goto fail2;
+	}
+	if (already_attached)
+		/* Not a fatal error */
+		netif_err(efx, probe, efx->net_dev,
+			  "Host already registered with MCPU\n");
+
+	/* Now we can reset the NIC */
+	rc = siena_reset_hw(efx, RESET_TYPE_ALL);
+	if (rc) {
+		netif_err(efx, probe, efx->net_dev, "failed to reset NIC\n");
+		goto fail3;
+	}
+
+	siena_init_wol(efx);
+
+	/* Allocate memory for INT_KER */
+	rc = efx_nic_alloc_buffer(efx, &efx->irq_status, sizeof(efx_oword_t));
+	if (rc)
+		goto fail4;
+	BUG_ON(efx->irq_status.dma_addr & 0x0f);
+
+	netif_dbg(efx, probe, efx->net_dev,
+		  "INT_KER at %llx (virt %p phys %llx)\n",
+		  (unsigned long long)efx->irq_status.dma_addr,
+		  efx->irq_status.addr,
+		  (unsigned long long)virt_to_phys(efx->irq_status.addr));
+
+	/* Read in the non-volatile configuration */
+	rc = siena_probe_nvconfig(efx);
+	if (rc == -EINVAL) {
+		netif_err(efx, probe, efx->net_dev,
+			  "NVRAM is invalid therefore using defaults\n");
+		efx->phy_type = PHY_TYPE_NONE;
+		efx->mdio.prtad = MDIO_PRTAD_NONE;
+	} else if (rc) {
+		goto fail5;
+	}
+
+	return 0;
+
+fail5:
+	efx_nic_free_buffer(efx, &efx->irq_status);
+fail4:
+fail3:
+	efx_mcdi_drv_attach(efx, false, NULL);
+fail2:
+	iounmap(nic_data->mcdi_smem);
+fail1:
+	kfree(efx->nic_data);
+	return rc;
+}
+
+/* This call performs hardware-specific global initialisation, such as
+ * defining the descriptor cache sizes and number of RSS channels.
+ * It does not set up any buffers, descriptor rings or event queues.
+ */
+static int siena_init_nic(struct efx_nic *efx)
+{
+	efx_oword_t temp;
+	int rc;
+
+	/* Recover from a failed assertion post-reset */
+	rc = efx_mcdi_handle_assertion(efx);
+	if (rc)
+		return rc;
+
+	/* Squash TX of packets of 16 bytes or less */
+	efx_reado(efx, &temp, FR_AZ_TX_RESERVED);
+	EFX_SET_OWORD_FIELD(temp, FRF_BZ_TX_FLUSH_MIN_LEN_EN, 1);
+	efx_writeo(efx, &temp, FR_AZ_TX_RESERVED);
+
+	/* Do not enable TX_NO_EOP_DISC_EN, since it limits packets to 16
+	 * descriptors (which is bad).
+	 */
+	efx_reado(efx, &temp, FR_AZ_TX_CFG);
+	EFX_SET_OWORD_FIELD(temp, FRF_AZ_TX_NO_EOP_DISC_EN, 0);
+	EFX_SET_OWORD_FIELD(temp, FRF_CZ_TX_FILTER_EN_BIT, 1);
+	efx_writeo(efx, &temp, FR_AZ_TX_CFG);
+
+	efx_reado(efx, &temp, FR_AZ_RX_CFG);
+	EFX_SET_OWORD_FIELD(temp, FRF_BZ_RX_DESC_PUSH_EN, 0);
+	EFX_SET_OWORD_FIELD(temp, FRF_BZ_RX_INGR_EN, 1);
+	/* Enable hash insertion. This is broken for the 'Falcon' hash
+	 * if IPv6 hashing is also enabled, so also select Toeplitz
+	 * TCP/IPv4 and IPv4 hashes. */
+	EFX_SET_OWORD_FIELD(temp, FRF_BZ_RX_HASH_INSRT_HDR, 1);
+	EFX_SET_OWORD_FIELD(temp, FRF_BZ_RX_HASH_ALG, 1);
+	EFX_SET_OWORD_FIELD(temp, FRF_BZ_RX_IP_HASH, 1);
+	efx_writeo(efx, &temp, FR_AZ_RX_CFG);
+
+	/* Set hash key for IPv4 */
+	memcpy(&temp, efx->rx_hash_key, sizeof(temp));
+	efx_writeo(efx, &temp, FR_BZ_RX_RSS_TKEY);
+
+	/* Enable IPv6 RSS */
+	BUILD_BUG_ON(sizeof(efx->rx_hash_key) <
+		     2 * sizeof(temp) + FRF_CZ_RX_RSS_IPV6_TKEY_HI_WIDTH / 8 ||
+		     FRF_CZ_RX_RSS_IPV6_TKEY_HI_LBN != 0);
+	memcpy(&temp, efx->rx_hash_key, sizeof(temp));
+	efx_writeo(efx, &temp, FR_CZ_RX_RSS_IPV6_REG1);
+	memcpy(&temp, efx->rx_hash_key + sizeof(temp), sizeof(temp));
+	efx_writeo(efx, &temp, FR_CZ_RX_RSS_IPV6_REG2);
+	EFX_POPULATE_OWORD_2(temp, FRF_CZ_RX_RSS_IPV6_THASH_ENABLE, 1,
+			     FRF_CZ_RX_RSS_IPV6_IP_THASH_ENABLE, 1);
+	memcpy(&temp, efx->rx_hash_key + 2 * sizeof(temp),
+	       FRF_CZ_RX_RSS_IPV6_TKEY_HI_WIDTH / 8);
+	efx_writeo(efx, &temp, FR_CZ_RX_RSS_IPV6_REG3);
+
+	/* Enable event logging */
+	rc = efx_mcdi_log_ctrl(efx, true, false, 0);
+	if (rc)
+		return rc;
+
+	/* Set destination of both TX and RX Flush events */
+	EFX_POPULATE_OWORD_1(temp, FRF_BZ_FLS_EVQ_ID, 0);
+	efx_writeo(efx, &temp, FR_BZ_DP_CTRL);
+
+	EFX_POPULATE_OWORD_1(temp, FRF_CZ_USREV_DIS, 1);
+	efx_writeo(efx, &temp, FR_CZ_USR_EV_CFG);
+
+	efx_nic_init_common(efx);
+	return 0;
+}
+
+static void siena_remove_nic(struct efx_nic *efx)
+{
+	struct siena_nic_data *nic_data = efx->nic_data;
+
+	efx_nic_free_buffer(efx, &efx->irq_status);
+
+	siena_reset_hw(efx, RESET_TYPE_ALL);
+
+	/* Relinquish the device back to the BMC */
+	if (efx_nic_has_mc(efx))
+		efx_mcdi_drv_attach(efx, false, NULL);
+
+	/* Tear down the private nic state */
+	iounmap(nic_data->mcdi_smem);
+	kfree(nic_data);
+	efx->nic_data = NULL;
+}
+
+#define STATS_GENERATION_INVALID ((__force __le64)(-1))
+
+static int siena_try_update_nic_stats(struct efx_nic *efx)
+{
+	__le64 *dma_stats;
+	struct efx_mac_stats *mac_stats;
+	__le64 generation_start, generation_end;
+
+	mac_stats = &efx->mac_stats;
+	dma_stats = efx->stats_buffer.addr;
+
+	generation_end = dma_stats[MC_CMD_MAC_GENERATION_END];
+	if (generation_end == STATS_GENERATION_INVALID)
+		return 0;
+	rmb();
+
+#define MAC_STAT(M, D) \
+	mac_stats->M = le64_to_cpu(dma_stats[MC_CMD_MAC_ ## D])
+
+	MAC_STAT(tx_bytes, TX_BYTES);
+	MAC_STAT(tx_bad_bytes, TX_BAD_BYTES);
+	mac_stats->tx_good_bytes = (mac_stats->tx_bytes -
+				    mac_stats->tx_bad_bytes);
+	MAC_STAT(tx_packets, TX_PKTS);
+	MAC_STAT(tx_bad, TX_BAD_FCS_PKTS);
+	MAC_STAT(tx_pause, TX_PAUSE_PKTS);
+	MAC_STAT(tx_control, TX_CONTROL_PKTS);
+	MAC_STAT(tx_unicast, TX_UNICAST_PKTS);
+	MAC_STAT(tx_multicast, TX_MULTICAST_PKTS);
+	MAC_STAT(tx_broadcast, TX_BROADCAST_PKTS);
+	MAC_STAT(tx_lt64, TX_LT64_PKTS);
+	MAC_STAT(tx_64, TX_64_PKTS);
+	MAC_STAT(tx_65_to_127, TX_65_TO_127_PKTS);
+	MAC_STAT(tx_128_to_255, TX_128_TO_255_PKTS);
+	MAC_STAT(tx_256_to_511, TX_256_TO_511_PKTS);
+	MAC_STAT(tx_512_to_1023, TX_512_TO_1023_PKTS);
+	MAC_STAT(tx_1024_to_15xx, TX_1024_TO_15XX_PKTS);
+	MAC_STAT(tx_15xx_to_jumbo, TX_15XX_TO_JUMBO_PKTS);
+	MAC_STAT(tx_gtjumbo, TX_GTJUMBO_PKTS);
+	mac_stats->tx_collision = 0;
+	MAC_STAT(tx_single_collision, TX_SINGLE_COLLISION_PKTS);
+	MAC_STAT(tx_multiple_collision, TX_MULTIPLE_COLLISION_PKTS);
+	MAC_STAT(tx_excessive_collision, TX_EXCESSIVE_COLLISION_PKTS);
+	MAC_STAT(tx_deferred, TX_DEFERRED_PKTS);
+	MAC_STAT(tx_late_collision, TX_LATE_COLLISION_PKTS);
+	mac_stats->tx_collision = (mac_stats->tx_single_collision +
+				   mac_stats->tx_multiple_collision +
+				   mac_stats->tx_excessive_collision +
+				   mac_stats->tx_late_collision);
+	MAC_STAT(tx_excessive_deferred, TX_EXCESSIVE_DEFERRED_PKTS);
+	MAC_STAT(tx_non_tcpudp, TX_NON_TCPUDP_PKTS);
+	MAC_STAT(tx_mac_src_error, TX_MAC_SRC_ERR_PKTS);
+	MAC_STAT(tx_ip_src_error, TX_IP_SRC_ERR_PKTS);
+	MAC_STAT(rx_bytes, RX_BYTES);
+	MAC_STAT(rx_bad_bytes, RX_BAD_BYTES);
+	mac_stats->rx_good_bytes = (mac_stats->rx_bytes -
+				    mac_stats->rx_bad_bytes);
+	MAC_STAT(rx_packets, RX_PKTS);
+	MAC_STAT(rx_good, RX_GOOD_PKTS);
+	MAC_STAT(rx_bad, RX_BAD_FCS_PKTS);
+	MAC_STAT(rx_pause, RX_PAUSE_PKTS);
+	MAC_STAT(rx_control, RX_CONTROL_PKTS);
+	MAC_STAT(rx_unicast, RX_UNICAST_PKTS);
+	MAC_STAT(rx_multicast, RX_MULTICAST_PKTS);
+	MAC_STAT(rx_broadcast, RX_BROADCAST_PKTS);
+	MAC_STAT(rx_lt64, RX_UNDERSIZE_PKTS);
+	MAC_STAT(rx_64, RX_64_PKTS);
+	MAC_STAT(rx_65_to_127, RX_65_TO_127_PKTS);
+	MAC_STAT(rx_128_to_255, RX_128_TO_255_PKTS);
+	MAC_STAT(rx_256_to_511, RX_256_TO_511_PKTS);
+	MAC_STAT(rx_512_to_1023, RX_512_TO_1023_PKTS);
+	MAC_STAT(rx_1024_to_15xx, RX_1024_TO_15XX_PKTS);
+	MAC_STAT(rx_15xx_to_jumbo, RX_15XX_TO_JUMBO_PKTS);
+	MAC_STAT(rx_gtjumbo, RX_GTJUMBO_PKTS);
+	mac_stats->rx_bad_lt64 = 0;
+	mac_stats->rx_bad_64_to_15xx = 0;
+	mac_stats->rx_bad_15xx_to_jumbo = 0;
+	MAC_STAT(rx_bad_gtjumbo, RX_JABBER_PKTS);
+	MAC_STAT(rx_overflow, RX_OVERFLOW_PKTS);
+	mac_stats->rx_missed = 0;
+	MAC_STAT(rx_false_carrier, RX_FALSE_CARRIER_PKTS);
+	MAC_STAT(rx_symbol_error, RX_SYMBOL_ERROR_PKTS);
+	MAC_STAT(rx_align_error, RX_ALIGN_ERROR_PKTS);
+	MAC_STAT(rx_length_error, RX_LENGTH_ERROR_PKTS);
+	MAC_STAT(rx_internal_error, RX_INTERNAL_ERROR_PKTS);
+	mac_stats->rx_good_lt64 = 0;
+
+	efx->n_rx_nodesc_drop_cnt =
+		le64_to_cpu(dma_stats[MC_CMD_MAC_RX_NODESC_DROPS]);
+
+#undef MAC_STAT
+
+	rmb();
+	generation_start = dma_stats[MC_CMD_MAC_GENERATION_START];
+	if (generation_end != generation_start)
+		return -EAGAIN;
+
+	return 0;
+}
+
+static void siena_update_nic_stats(struct efx_nic *efx)
+{
+	int retry;
+
+	/* If we're unlucky enough to read statistics wduring the DMA, wait
+	 * up to 10ms for it to finish (typically takes <500us) */
+	for (retry = 0; retry < 100; ++retry) {
+		if (siena_try_update_nic_stats(efx) == 0)
+			return;
+		udelay(100);
+	}
+
+	/* Use the old values instead */
+}
+
+static void siena_start_nic_stats(struct efx_nic *efx)
+{
+	__le64 *dma_stats = efx->stats_buffer.addr;
+
+	dma_stats[MC_CMD_MAC_GENERATION_END] = STATS_GENERATION_INVALID;
+
+	efx_mcdi_mac_stats(efx, efx->stats_buffer.dma_addr,
+			   MC_CMD_MAC_NSTATS * sizeof(u64), 1, 0);
+}
+
+static void siena_stop_nic_stats(struct efx_nic *efx)
+{
+	efx_mcdi_mac_stats(efx, efx->stats_buffer.dma_addr, 0, 0, 0);
+}
+
+/**************************************************************************
+ *
+ * Wake on LAN
+ *
+ **************************************************************************
+ */
+
+static void siena_get_wol(struct efx_nic *efx, struct ethtool_wolinfo *wol)
+{
+	struct siena_nic_data *nic_data = efx->nic_data;
+
+	wol->supported = WAKE_MAGIC;
+	if (nic_data->wol_filter_id != -1)
+		wol->wolopts = WAKE_MAGIC;
+	else
+		wol->wolopts = 0;
+	memset(&wol->sopass, 0, sizeof(wol->sopass));
+}
+
+
+static int siena_set_wol(struct efx_nic *efx, u32 type)
+{
+	struct siena_nic_data *nic_data = efx->nic_data;
+	int rc;
+
+	if (type & ~WAKE_MAGIC)
+		return -EINVAL;
+
+	if (type & WAKE_MAGIC) {
+		if (nic_data->wol_filter_id != -1)
+			efx_mcdi_wol_filter_remove(efx,
+						   nic_data->wol_filter_id);
+		rc = efx_mcdi_wol_filter_set_magic(efx, efx->net_dev->dev_addr,
+						   &nic_data->wol_filter_id);
+		if (rc)
+			goto fail;
+
+		pci_wake_from_d3(efx->pci_dev, true);
+	} else {
+		rc = efx_mcdi_wol_filter_reset(efx);
+		nic_data->wol_filter_id = -1;
+		pci_wake_from_d3(efx->pci_dev, false);
+		if (rc)
+			goto fail;
+	}
+
+	return 0;
+ fail:
+	netif_err(efx, hw, efx->net_dev, "%s failed: type=%d rc=%d\n",
+		  __func__, type, rc);
+	return rc;
+}
+
+
+static void siena_init_wol(struct efx_nic *efx)
+{
+	struct siena_nic_data *nic_data = efx->nic_data;
+	int rc;
+
+	rc = efx_mcdi_wol_filter_get_magic(efx, &nic_data->wol_filter_id);
+
+	if (rc != 0) {
+		/* If it failed, attempt to get into a synchronised
+		 * state with MC by resetting any set WoL filters */
+		efx_mcdi_wol_filter_reset(efx);
+		nic_data->wol_filter_id = -1;
+	} else if (nic_data->wol_filter_id != -1) {
+		pci_wake_from_d3(efx->pci_dev, true);
+	}
+}
+
+
+/**************************************************************************
+ *
+ * Revision-dependent attributes used by efx.c and nic.c
+ *
+ **************************************************************************
+ */
+
+const struct efx_nic_type siena_a0_nic_type = {
+	.probe = siena_probe_nic,
+	.remove = siena_remove_nic,
+	.init = siena_init_nic,
+	.fini = efx_port_dummy_op_void,
+	.monitor = NULL,
+	.map_reset_reason = siena_map_reset_reason,
+	.map_reset_flags = siena_map_reset_flags,
+	.reset = siena_reset_hw,
+	.probe_port = siena_probe_port,
+	.remove_port = siena_remove_port,
+	.prepare_flush = efx_port_dummy_op_void,
+	.update_stats = siena_update_nic_stats,
+	.start_stats = siena_start_nic_stats,
+	.stop_stats = siena_stop_nic_stats,
+	.set_id_led = efx_mcdi_set_id_led,
+	.push_irq_moderation = siena_push_irq_moderation,
+	.push_multicast_hash = siena_push_multicast_hash,
+	.reconfigure_port = efx_mcdi_phy_reconfigure,
+	.get_wol = siena_get_wol,
+	.set_wol = siena_set_wol,
+	.resume_wol = siena_init_wol,
+	.test_registers = siena_test_registers,
+	.test_nvram = efx_mcdi_nvram_test_all,
+	.default_mac_ops = &efx_mcdi_mac_operations,
+
+	.revision = EFX_REV_SIENA_A0,
+	.mem_map_size = FR_CZ_MC_TREG_SMEM, /* MC_TREG_SMEM mapped separately */
+	.txd_ptr_tbl_base = FR_BZ_TX_DESC_PTR_TBL,
+	.rxd_ptr_tbl_base = FR_BZ_RX_DESC_PTR_TBL,
+	.buf_tbl_base = FR_BZ_BUF_FULL_TBL,
+	.evq_ptr_tbl_base = FR_BZ_EVQ_PTR_TBL,
+	.evq_rptr_tbl_base = FR_BZ_EVQ_RPTR,
+	.max_dma_mask = DMA_BIT_MASK(FSF_AZ_TX_KER_BUF_ADDR_WIDTH),
+	.rx_buffer_hash_size = 0x10,
+	.rx_buffer_padding = 0,
+	.max_interrupt_mode = EFX_INT_MODE_MSIX,
+	.phys_addr_channels = 32, /* Hardware limit is 64, but the legacy
+				   * interrupt handler only supports 32
+				   * channels */
+	.tx_dc_base = 0x88000,
+	.rx_dc_base = 0x68000,
+	.offload_features = (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
+			     NETIF_F_RXHASH | NETIF_F_NTUPLE),
+};
diff --git a/drivers/net/ethernet/sfc/spi.h b/drivers/net/ethernet/sfc/spi.h
new file mode 100644
index 0000000..71f2e3e
--- /dev/null
+++ b/drivers/net/ethernet/sfc/spi.h
@@ -0,0 +1,99 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2005 Fen Systems Ltd.
+ * Copyright 2006-2010 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#ifndef EFX_SPI_H
+#define EFX_SPI_H
+
+#include "net_driver.h"
+
+/**************************************************************************
+ *
+ * Basic SPI command set and bit definitions
+ *
+ *************************************************************************/
+
+#define SPI_WRSR 0x01		/* Write status register */
+#define SPI_WRITE 0x02		/* Write data to memory array */
+#define SPI_READ 0x03		/* Read data from memory array */
+#define SPI_WRDI 0x04		/* Reset write enable latch */
+#define SPI_RDSR 0x05		/* Read status register */
+#define SPI_WREN 0x06		/* Set write enable latch */
+#define SPI_SST_EWSR 0x50	/* SST: Enable write to status register */
+
+#define SPI_STATUS_WPEN 0x80	/* Write-protect pin enabled */
+#define SPI_STATUS_BP2 0x10	/* Block protection bit 2 */
+#define SPI_STATUS_BP1 0x08	/* Block protection bit 1 */
+#define SPI_STATUS_BP0 0x04	/* Block protection bit 0 */
+#define SPI_STATUS_WEN 0x02	/* State of the write enable latch */
+#define SPI_STATUS_NRDY 0x01	/* Device busy flag */
+
+/**
+ * struct efx_spi_device - an Efx SPI (Serial Peripheral Interface) device
+ * @device_id:		Controller's id for the device
+ * @size:		Size (in bytes)
+ * @addr_len:		Number of address bytes in read/write commands
+ * @munge_address:	Flag whether addresses should be munged.
+ *	Some devices with 9-bit addresses (e.g. AT25040A EEPROM)
+ *	use bit 3 of the command byte as address bit A8, rather
+ *	than having a two-byte address.  If this flag is set, then
+ *	commands should be munged in this way.
+ * @erase_command:	Erase command (or 0 if sector erase not needed).
+ * @erase_size:		Erase sector size (in bytes)
+ *	Erase commands affect sectors with this size and alignment.
+ *	This must be a power of two.
+ * @block_size:		Write block size (in bytes).
+ *	Write commands are limited to blocks with this size and alignment.
+ */
+struct efx_spi_device {
+	int device_id;
+	unsigned int size;
+	unsigned int addr_len;
+	unsigned int munge_address:1;
+	u8 erase_command;
+	unsigned int erase_size;
+	unsigned int block_size;
+};
+
+static inline bool efx_spi_present(const struct efx_spi_device *spi)
+{
+	return spi->size != 0;
+}
+
+int falcon_spi_cmd(struct efx_nic *efx,
+		   const struct efx_spi_device *spi, unsigned int command,
+		   int address, const void* in, void *out, size_t len);
+int falcon_spi_wait_write(struct efx_nic *efx,
+			  const struct efx_spi_device *spi);
+int falcon_spi_read(struct efx_nic *efx,
+		    const struct efx_spi_device *spi, loff_t start,
+		    size_t len, size_t *retlen, u8 *buffer);
+int falcon_spi_write(struct efx_nic *efx,
+		     const struct efx_spi_device *spi, loff_t start,
+		     size_t len, size_t *retlen, const u8 *buffer);
+
+/*
+ * SFC4000 flash is partitioned into:
+ *     0-0x400       chip and board config (see falcon_hwdefs.h)
+ *     0x400-0x8000  unused (or may contain VPD if EEPROM not present)
+ *     0x8000-end    boot code (mapped to PCI expansion ROM)
+ * SFC4000 small EEPROM (size < 0x400) is used for VPD only.
+ * SFC4000 large EEPROM (size >= 0x400) is partitioned into:
+ *     0-0x400       chip and board config
+ *     configurable  VPD
+ *     0x800-0x1800  boot config
+ * Aside from the chip and board config, all of these are optional and may
+ * be absent or truncated depending on the devices used.
+ */
+#define FALCON_NVCONFIG_END 0x400U
+#define FALCON_FLASH_BOOTCODE_START 0x8000U
+#define EFX_EEPROM_BOOTCONFIG_START 0x800U
+#define EFX_EEPROM_BOOTCONFIG_END 0x1800U
+
+#endif /* EFX_SPI_H */
diff --git a/drivers/net/ethernet/sfc/tenxpress.c b/drivers/net/ethernet/sfc/tenxpress.c
new file mode 100644
index 0000000..7b0fd89
--- /dev/null
+++ b/drivers/net/ethernet/sfc/tenxpress.c
@@ -0,0 +1,494 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2007-2011 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#include <linux/delay.h>
+#include <linux/rtnetlink.h>
+#include <linux/seq_file.h>
+#include <linux/slab.h>
+#include "efx.h"
+#include "mdio_10g.h"
+#include "nic.h"
+#include "phy.h"
+#include "workarounds.h"
+
+/* We expect these MMDs to be in the package. */
+#define TENXPRESS_REQUIRED_DEVS (MDIO_DEVS_PMAPMD	| \
+				 MDIO_DEVS_PCS		| \
+				 MDIO_DEVS_PHYXS	| \
+				 MDIO_DEVS_AN)
+
+#define SFX7101_LOOPBACKS ((1 << LOOPBACK_PHYXS) |	\
+			   (1 << LOOPBACK_PCS) |	\
+			   (1 << LOOPBACK_PMAPMD) |	\
+			   (1 << LOOPBACK_PHYXS_WS))
+
+/* We complain if we fail to see the link partner as 10G capable this many
+ * times in a row (must be > 1 as sampling the autoneg. registers is racy)
+ */
+#define MAX_BAD_LP_TRIES	(5)
+
+/* Extended control register */
+#define PMA_PMD_XCONTROL_REG	49152
+#define PMA_PMD_EXT_GMII_EN_LBN	1
+#define PMA_PMD_EXT_GMII_EN_WIDTH 1
+#define PMA_PMD_EXT_CLK_OUT_LBN	2
+#define PMA_PMD_EXT_CLK_OUT_WIDTH 1
+#define PMA_PMD_LNPGA_POWERDOWN_LBN 8
+#define PMA_PMD_LNPGA_POWERDOWN_WIDTH 1
+#define PMA_PMD_EXT_CLK312_WIDTH 1
+#define PMA_PMD_EXT_LPOWER_LBN  12
+#define PMA_PMD_EXT_LPOWER_WIDTH 1
+#define PMA_PMD_EXT_ROBUST_LBN	14
+#define PMA_PMD_EXT_ROBUST_WIDTH 1
+#define PMA_PMD_EXT_SSR_LBN	15
+#define PMA_PMD_EXT_SSR_WIDTH	1
+
+/* extended status register */
+#define PMA_PMD_XSTATUS_REG	49153
+#define PMA_PMD_XSTAT_MDIX_LBN	14
+#define PMA_PMD_XSTAT_FLP_LBN   (12)
+
+/* LED control register */
+#define PMA_PMD_LED_CTRL_REG	49159
+#define PMA_PMA_LED_ACTIVITY_LBN	(3)
+
+/* LED function override register */
+#define PMA_PMD_LED_OVERR_REG	49161
+/* Bit positions for different LEDs (there are more but not wired on SFE4001)*/
+#define PMA_PMD_LED_LINK_LBN	(0)
+#define PMA_PMD_LED_SPEED_LBN	(2)
+#define PMA_PMD_LED_TX_LBN	(4)
+#define PMA_PMD_LED_RX_LBN	(6)
+/* Override settings */
+#define	PMA_PMD_LED_AUTO	(0)	/* H/W control */
+#define	PMA_PMD_LED_ON		(1)
+#define	PMA_PMD_LED_OFF		(2)
+#define PMA_PMD_LED_FLASH	(3)
+#define PMA_PMD_LED_MASK	3
+/* All LEDs under hardware control */
+/* Green and Amber under hardware control, Red off */
+#define SFX7101_PMA_PMD_LED_DEFAULT (PMA_PMD_LED_OFF << PMA_PMD_LED_RX_LBN)
+
+#define PMA_PMD_SPEED_ENABLE_REG 49192
+#define PMA_PMD_100TX_ADV_LBN    1
+#define PMA_PMD_100TX_ADV_WIDTH  1
+#define PMA_PMD_1000T_ADV_LBN    2
+#define PMA_PMD_1000T_ADV_WIDTH  1
+#define PMA_PMD_10000T_ADV_LBN   3
+#define PMA_PMD_10000T_ADV_WIDTH 1
+#define PMA_PMD_SPEED_LBN        4
+#define PMA_PMD_SPEED_WIDTH      4
+
+/* Misc register defines */
+#define PCS_CLOCK_CTRL_REG	55297
+#define PLL312_RST_N_LBN 2
+
+#define PCS_SOFT_RST2_REG	55302
+#define SERDES_RST_N_LBN 13
+#define XGXS_RST_N_LBN 12
+
+#define	PCS_TEST_SELECT_REG	55303	/* PRM 10.5.8 */
+#define	CLK312_EN_LBN 3
+
+/* PHYXS registers */
+#define PHYXS_XCONTROL_REG	49152
+#define PHYXS_RESET_LBN		15
+#define PHYXS_RESET_WIDTH	1
+
+#define PHYXS_TEST1         (49162)
+#define LOOPBACK_NEAR_LBN   (8)
+#define LOOPBACK_NEAR_WIDTH (1)
+
+/* Boot status register */
+#define PCS_BOOT_STATUS_REG		53248
+#define PCS_BOOT_FATAL_ERROR_LBN	0
+#define PCS_BOOT_PROGRESS_LBN		1
+#define PCS_BOOT_PROGRESS_WIDTH		2
+#define PCS_BOOT_PROGRESS_INIT		0
+#define PCS_BOOT_PROGRESS_WAIT_MDIO	1
+#define PCS_BOOT_PROGRESS_CHECKSUM	2
+#define PCS_BOOT_PROGRESS_JUMP		3
+#define PCS_BOOT_DOWNLOAD_WAIT_LBN	3
+#define PCS_BOOT_CODE_STARTED_LBN	4
+
+/* 100M/1G PHY registers */
+#define GPHY_XCONTROL_REG	49152
+#define GPHY_ISOLATE_LBN	10
+#define GPHY_ISOLATE_WIDTH	1
+#define GPHY_DUPLEX_LBN	  	8
+#define GPHY_DUPLEX_WIDTH	1
+#define GPHY_LOOPBACK_NEAR_LBN	14
+#define GPHY_LOOPBACK_NEAR_WIDTH 1
+
+#define C22EXT_STATUS_REG       49153
+#define C22EXT_STATUS_LINK_LBN  2
+#define C22EXT_STATUS_LINK_WIDTH 1
+
+#define C22EXT_MSTSLV_CTRL			49161
+#define C22EXT_MSTSLV_CTRL_ADV_1000_HD_LBN	8
+#define C22EXT_MSTSLV_CTRL_ADV_1000_FD_LBN	9
+
+#define C22EXT_MSTSLV_STATUS			49162
+#define C22EXT_MSTSLV_STATUS_LP_1000_HD_LBN	10
+#define C22EXT_MSTSLV_STATUS_LP_1000_FD_LBN	11
+
+/* Time to wait between powering down the LNPGA and turning off the power
+ * rails */
+#define LNPGA_PDOWN_WAIT	(HZ / 5)
+
+struct tenxpress_phy_data {
+	enum efx_loopback_mode loopback_mode;
+	enum efx_phy_mode phy_mode;
+	int bad_lp_tries;
+};
+
+static int tenxpress_init(struct efx_nic *efx)
+{
+	/* Enable 312.5 MHz clock */
+	efx_mdio_write(efx, MDIO_MMD_PCS, PCS_TEST_SELECT_REG,
+		       1 << CLK312_EN_LBN);
+
+	/* Set the LEDs up as: Green = Link, Amber = Link/Act, Red = Off */
+	efx_mdio_set_flag(efx, MDIO_MMD_PMAPMD, PMA_PMD_LED_CTRL_REG,
+			  1 << PMA_PMA_LED_ACTIVITY_LBN, true);
+	efx_mdio_write(efx, MDIO_MMD_PMAPMD, PMA_PMD_LED_OVERR_REG,
+		       SFX7101_PMA_PMD_LED_DEFAULT);
+
+	return 0;
+}
+
+static int tenxpress_phy_probe(struct efx_nic *efx)
+{
+	struct tenxpress_phy_data *phy_data;
+
+	/* Allocate phy private storage */
+	phy_data = kzalloc(sizeof(*phy_data), GFP_KERNEL);
+	if (!phy_data)
+		return -ENOMEM;
+	efx->phy_data = phy_data;
+	phy_data->phy_mode = efx->phy_mode;
+
+	efx->mdio.mmds = TENXPRESS_REQUIRED_DEVS;
+	efx->mdio.mode_support = MDIO_SUPPORTS_C45;
+
+	efx->loopback_modes = SFX7101_LOOPBACKS | FALCON_XMAC_LOOPBACKS;
+
+	efx->link_advertising = (ADVERTISED_TP | ADVERTISED_Autoneg |
+				 ADVERTISED_10000baseT_Full);
+
+	return 0;
+}
+
+static int tenxpress_phy_init(struct efx_nic *efx)
+{
+	int rc;
+
+	falcon_board(efx)->type->init_phy(efx);
+
+	if (!(efx->phy_mode & PHY_MODE_SPECIAL)) {
+		rc = efx_mdio_wait_reset_mmds(efx, TENXPRESS_REQUIRED_DEVS);
+		if (rc < 0)
+			return rc;
+
+		rc = efx_mdio_check_mmds(efx, TENXPRESS_REQUIRED_DEVS);
+		if (rc < 0)
+			return rc;
+	}
+
+	rc = tenxpress_init(efx);
+	if (rc < 0)
+		return rc;
+
+	/* Reinitialise flow control settings */
+	efx_link_set_wanted_fc(efx, efx->wanted_fc);
+	efx_mdio_an_reconfigure(efx);
+
+	schedule_timeout_uninterruptible(HZ / 5); /* 200ms */
+
+	/* Let XGXS and SerDes out of reset */
+	falcon_reset_xaui(efx);
+
+	return 0;
+}
+
+/* Perform a "special software reset" on the PHY. The caller is
+ * responsible for saving and restoring the PHY hardware registers
+ * properly, and masking/unmasking LASI */
+static int tenxpress_special_reset(struct efx_nic *efx)
+{
+	int rc, reg;
+
+	/* The XGMAC clock is driven from the SFX7101 312MHz clock, so
+	 * a special software reset can glitch the XGMAC sufficiently for stats
+	 * requests to fail. */
+	falcon_stop_nic_stats(efx);
+
+	/* Initiate reset */
+	reg = efx_mdio_read(efx, MDIO_MMD_PMAPMD, PMA_PMD_XCONTROL_REG);
+	reg |= (1 << PMA_PMD_EXT_SSR_LBN);
+	efx_mdio_write(efx, MDIO_MMD_PMAPMD, PMA_PMD_XCONTROL_REG, reg);
+
+	mdelay(200);
+
+	/* Wait for the blocks to come out of reset */
+	rc = efx_mdio_wait_reset_mmds(efx, TENXPRESS_REQUIRED_DEVS);
+	if (rc < 0)
+		goto out;
+
+	/* Try and reconfigure the device */
+	rc = tenxpress_init(efx);
+	if (rc < 0)
+		goto out;
+
+	/* Wait for the XGXS state machine to churn */
+	mdelay(10);
+out:
+	falcon_start_nic_stats(efx);
+	return rc;
+}
+
+static void sfx7101_check_bad_lp(struct efx_nic *efx, bool link_ok)
+{
+	struct tenxpress_phy_data *pd = efx->phy_data;
+	bool bad_lp;
+	int reg;
+
+	if (link_ok) {
+		bad_lp = false;
+	} else {
+		/* Check that AN has started but not completed. */
+		reg = efx_mdio_read(efx, MDIO_MMD_AN, MDIO_STAT1);
+		if (!(reg & MDIO_AN_STAT1_LPABLE))
+			return; /* LP status is unknown */
+		bad_lp = !(reg & MDIO_AN_STAT1_COMPLETE);
+		if (bad_lp)
+			pd->bad_lp_tries++;
+	}
+
+	/* Nothing to do if all is well and was previously so. */
+	if (!pd->bad_lp_tries)
+		return;
+
+	/* Use the RX (red) LED as an error indicator once we've seen AN
+	 * failure several times in a row, and also log a message. */
+	if (!bad_lp || pd->bad_lp_tries == MAX_BAD_LP_TRIES) {
+		reg = efx_mdio_read(efx, MDIO_MMD_PMAPMD,
+				    PMA_PMD_LED_OVERR_REG);
+		reg &= ~(PMA_PMD_LED_MASK << PMA_PMD_LED_RX_LBN);
+		if (!bad_lp) {
+			reg |= PMA_PMD_LED_OFF << PMA_PMD_LED_RX_LBN;
+		} else {
+			reg |= PMA_PMD_LED_FLASH << PMA_PMD_LED_RX_LBN;
+			netif_err(efx, link, efx->net_dev,
+				  "appears to be plugged into a port"
+				  " that is not 10GBASE-T capable. The PHY"
+				  " supports 10GBASE-T ONLY, so no link can"
+				  " be established\n");
+		}
+		efx_mdio_write(efx, MDIO_MMD_PMAPMD,
+			       PMA_PMD_LED_OVERR_REG, reg);
+		pd->bad_lp_tries = bad_lp;
+	}
+}
+
+static bool sfx7101_link_ok(struct efx_nic *efx)
+{
+	return efx_mdio_links_ok(efx,
+				 MDIO_DEVS_PMAPMD |
+				 MDIO_DEVS_PCS |
+				 MDIO_DEVS_PHYXS);
+}
+
+static void tenxpress_ext_loopback(struct efx_nic *efx)
+{
+	efx_mdio_set_flag(efx, MDIO_MMD_PHYXS, PHYXS_TEST1,
+			  1 << LOOPBACK_NEAR_LBN,
+			  efx->loopback_mode == LOOPBACK_PHYXS);
+}
+
+static void tenxpress_low_power(struct efx_nic *efx)
+{
+	efx_mdio_set_mmds_lpower(
+		efx, !!(efx->phy_mode & PHY_MODE_LOW_POWER),
+		TENXPRESS_REQUIRED_DEVS);
+}
+
+static int tenxpress_phy_reconfigure(struct efx_nic *efx)
+{
+	struct tenxpress_phy_data *phy_data = efx->phy_data;
+	bool phy_mode_change, loop_reset;
+
+	if (efx->phy_mode & (PHY_MODE_OFF | PHY_MODE_SPECIAL)) {
+		phy_data->phy_mode = efx->phy_mode;
+		return 0;
+	}
+
+	phy_mode_change = (efx->phy_mode == PHY_MODE_NORMAL &&
+			   phy_data->phy_mode != PHY_MODE_NORMAL);
+	loop_reset = (LOOPBACK_OUT_OF(phy_data, efx, LOOPBACKS_EXTERNAL(efx)) ||
+		      LOOPBACK_CHANGED(phy_data, efx, 1 << LOOPBACK_GPHY));
+
+	if (loop_reset || phy_mode_change) {
+		tenxpress_special_reset(efx);
+		falcon_reset_xaui(efx);
+	}
+
+	tenxpress_low_power(efx);
+	efx_mdio_transmit_disable(efx);
+	efx_mdio_phy_reconfigure(efx);
+	tenxpress_ext_loopback(efx);
+	efx_mdio_an_reconfigure(efx);
+
+	phy_data->loopback_mode = efx->loopback_mode;
+	phy_data->phy_mode = efx->phy_mode;
+
+	return 0;
+}
+
+static void
+tenxpress_get_settings(struct efx_nic *efx, struct ethtool_cmd *ecmd);
+
+/* Poll for link state changes */
+static bool tenxpress_phy_poll(struct efx_nic *efx)
+{
+	struct efx_link_state old_state = efx->link_state;
+
+	efx->link_state.up = sfx7101_link_ok(efx);
+	efx->link_state.speed = 10000;
+	efx->link_state.fd = true;
+	efx->link_state.fc = efx_mdio_get_pause(efx);
+
+	sfx7101_check_bad_lp(efx, efx->link_state.up);
+
+	return !efx_link_state_equal(&efx->link_state, &old_state);
+}
+
+static void sfx7101_phy_fini(struct efx_nic *efx)
+{
+	int reg;
+
+	/* Power down the LNPGA */
+	reg = (1 << PMA_PMD_LNPGA_POWERDOWN_LBN);
+	efx_mdio_write(efx, MDIO_MMD_PMAPMD, PMA_PMD_XCONTROL_REG, reg);
+
+	/* Waiting here ensures that the board fini, which can turn
+	 * off the power to the PHY, won't get run until the LNPGA
+	 * powerdown has been given long enough to complete. */
+	schedule_timeout_uninterruptible(LNPGA_PDOWN_WAIT); /* 200 ms */
+}
+
+static void tenxpress_phy_remove(struct efx_nic *efx)
+{
+	kfree(efx->phy_data);
+	efx->phy_data = NULL;
+}
+
+
+/* Override the RX, TX and link LEDs */
+void tenxpress_set_id_led(struct efx_nic *efx, enum efx_led_mode mode)
+{
+	int reg;
+
+	switch (mode) {
+	case EFX_LED_OFF:
+		reg = (PMA_PMD_LED_OFF << PMA_PMD_LED_TX_LBN) |
+			(PMA_PMD_LED_OFF << PMA_PMD_LED_RX_LBN) |
+			(PMA_PMD_LED_OFF << PMA_PMD_LED_LINK_LBN);
+		break;
+	case EFX_LED_ON:
+		reg = (PMA_PMD_LED_ON << PMA_PMD_LED_TX_LBN) |
+			(PMA_PMD_LED_ON << PMA_PMD_LED_RX_LBN) |
+			(PMA_PMD_LED_ON << PMA_PMD_LED_LINK_LBN);
+		break;
+	default:
+		reg = SFX7101_PMA_PMD_LED_DEFAULT;
+		break;
+	}
+
+	efx_mdio_write(efx, MDIO_MMD_PMAPMD, PMA_PMD_LED_OVERR_REG, reg);
+}
+
+static const char *const sfx7101_test_names[] = {
+	"bist"
+};
+
+static const char *sfx7101_test_name(struct efx_nic *efx, unsigned int index)
+{
+	if (index < ARRAY_SIZE(sfx7101_test_names))
+		return sfx7101_test_names[index];
+	return NULL;
+}
+
+static int
+sfx7101_run_tests(struct efx_nic *efx, int *results, unsigned flags)
+{
+	int rc;
+
+	if (!(flags & ETH_TEST_FL_OFFLINE))
+		return 0;
+
+	/* BIST is automatically run after a special software reset */
+	rc = tenxpress_special_reset(efx);
+	results[0] = rc ? -1 : 1;
+
+	efx_mdio_an_reconfigure(efx);
+
+	return rc;
+}
+
+static void
+tenxpress_get_settings(struct efx_nic *efx, struct ethtool_cmd *ecmd)
+{
+	u32 adv = 0, lpa = 0;
+	int reg;
+
+	reg = efx_mdio_read(efx, MDIO_MMD_AN, MDIO_AN_10GBT_CTRL);
+	if (reg & MDIO_AN_10GBT_CTRL_ADV10G)
+		adv |= ADVERTISED_10000baseT_Full;
+	reg = efx_mdio_read(efx, MDIO_MMD_AN, MDIO_AN_10GBT_STAT);
+	if (reg & MDIO_AN_10GBT_STAT_LP10G)
+		lpa |= ADVERTISED_10000baseT_Full;
+
+	mdio45_ethtool_gset_npage(&efx->mdio, ecmd, adv, lpa);
+
+	/* In loopback, the PHY automatically brings up the correct interface,
+	 * but doesn't advertise the correct speed. So override it */
+	if (LOOPBACK_EXTERNAL(efx))
+		ethtool_cmd_speed_set(ecmd, SPEED_10000);
+}
+
+static int tenxpress_set_settings(struct efx_nic *efx, struct ethtool_cmd *ecmd)
+{
+	if (!ecmd->autoneg)
+		return -EINVAL;
+
+	return efx_mdio_set_settings(efx, ecmd);
+}
+
+static void sfx7101_set_npage_adv(struct efx_nic *efx, u32 advertising)
+{
+	efx_mdio_set_flag(efx, MDIO_MMD_AN, MDIO_AN_10GBT_CTRL,
+			  MDIO_AN_10GBT_CTRL_ADV10G,
+			  advertising & ADVERTISED_10000baseT_Full);
+}
+
+const struct efx_phy_operations falcon_sfx7101_phy_ops = {
+	.probe		  = tenxpress_phy_probe,
+	.init             = tenxpress_phy_init,
+	.reconfigure      = tenxpress_phy_reconfigure,
+	.poll             = tenxpress_phy_poll,
+	.fini             = sfx7101_phy_fini,
+	.remove		  = tenxpress_phy_remove,
+	.get_settings	  = tenxpress_get_settings,
+	.set_settings	  = tenxpress_set_settings,
+	.set_npage_adv    = sfx7101_set_npage_adv,
+	.test_alive	  = efx_mdio_test_alive,
+	.test_name	  = sfx7101_test_name,
+	.run_tests	  = sfx7101_run_tests,
+};
diff --git a/drivers/net/ethernet/sfc/tx.c b/drivers/net/ethernet/sfc/tx.c
new file mode 100644
index 0000000..84eb99e
--- /dev/null
+++ b/drivers/net/ethernet/sfc/tx.c
@@ -0,0 +1,1212 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2005-2006 Fen Systems Ltd.
+ * Copyright 2005-2010 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#include <linux/pci.h>
+#include <linux/tcp.h>
+#include <linux/ip.h>
+#include <linux/in.h>
+#include <linux/ipv6.h>
+#include <linux/slab.h>
+#include <net/ipv6.h>
+#include <linux/if_ether.h>
+#include <linux/highmem.h>
+#include "net_driver.h"
+#include "efx.h"
+#include "nic.h"
+#include "workarounds.h"
+
+/*
+ * TX descriptor ring full threshold
+ *
+ * The tx_queue descriptor ring fill-level must fall below this value
+ * before we restart the netif queue
+ */
+#define EFX_TXQ_THRESHOLD(_efx) ((_efx)->txq_entries / 2u)
+
+static void efx_dequeue_buffer(struct efx_tx_queue *tx_queue,
+			       struct efx_tx_buffer *buffer)
+{
+	if (buffer->unmap_len) {
+		struct pci_dev *pci_dev = tx_queue->efx->pci_dev;
+		dma_addr_t unmap_addr = (buffer->dma_addr + buffer->len -
+					 buffer->unmap_len);
+		if (buffer->unmap_single)
+			pci_unmap_single(pci_dev, unmap_addr, buffer->unmap_len,
+					 PCI_DMA_TODEVICE);
+		else
+			pci_unmap_page(pci_dev, unmap_addr, buffer->unmap_len,
+				       PCI_DMA_TODEVICE);
+		buffer->unmap_len = 0;
+		buffer->unmap_single = false;
+	}
+
+	if (buffer->skb) {
+		dev_kfree_skb_any((struct sk_buff *) buffer->skb);
+		buffer->skb = NULL;
+		netif_vdbg(tx_queue->efx, tx_done, tx_queue->efx->net_dev,
+			   "TX queue %d transmission id %x complete\n",
+			   tx_queue->queue, tx_queue->read_count);
+	}
+}
+
+/**
+ * struct efx_tso_header - a DMA mapped buffer for packet headers
+ * @next: Linked list of free ones.
+ *	The list is protected by the TX queue lock.
+ * @dma_unmap_len: Length to unmap for an oversize buffer, or 0.
+ * @dma_addr: The DMA address of the header below.
+ *
+ * This controls the memory used for a TSO header.  Use TSOH_DATA()
+ * to find the packet header data.  Use TSOH_SIZE() to calculate the
+ * total size required for a given packet header length.  TSO headers
+ * in the free list are exactly %TSOH_STD_SIZE bytes in size.
+ */
+struct efx_tso_header {
+	union {
+		struct efx_tso_header *next;
+		size_t unmap_len;
+	};
+	dma_addr_t dma_addr;
+};
+
+static int efx_enqueue_skb_tso(struct efx_tx_queue *tx_queue,
+			       struct sk_buff *skb);
+static void efx_fini_tso(struct efx_tx_queue *tx_queue);
+static void efx_tsoh_heap_free(struct efx_tx_queue *tx_queue,
+			       struct efx_tso_header *tsoh);
+
+static void efx_tsoh_free(struct efx_tx_queue *tx_queue,
+			  struct efx_tx_buffer *buffer)
+{
+	if (buffer->tsoh) {
+		if (likely(!buffer->tsoh->unmap_len)) {
+			buffer->tsoh->next = tx_queue->tso_headers_free;
+			tx_queue->tso_headers_free = buffer->tsoh;
+		} else {
+			efx_tsoh_heap_free(tx_queue, buffer->tsoh);
+		}
+		buffer->tsoh = NULL;
+	}
+}
+
+
+static inline unsigned
+efx_max_tx_len(struct efx_nic *efx, dma_addr_t dma_addr)
+{
+	/* Depending on the NIC revision, we can use descriptor
+	 * lengths up to 8K or 8K-1.  However, since PCI Express
+	 * devices must split read requests at 4K boundaries, there is
+	 * little benefit from using descriptors that cross those
+	 * boundaries and we keep things simple by not doing so.
+	 */
+	unsigned len = (~dma_addr & 0xfff) + 1;
+
+	/* Work around hardware bug for unaligned buffers. */
+	if (EFX_WORKAROUND_5391(efx) && (dma_addr & 0xf))
+		len = min_t(unsigned, len, 512 - (dma_addr & 0xf));
+
+	return len;
+}
+
+/*
+ * Add a socket buffer to a TX queue
+ *
+ * This maps all fragments of a socket buffer for DMA and adds them to
+ * the TX queue.  The queue's insert pointer will be incremented by
+ * the number of fragments in the socket buffer.
+ *
+ * If any DMA mapping fails, any mapped fragments will be unmapped,
+ * the queue's insert pointer will be restored to its original value.
+ *
+ * This function is split out from efx_hard_start_xmit to allow the
+ * loopback test to direct packets via specific TX queues.
+ *
+ * Returns NETDEV_TX_OK or NETDEV_TX_BUSY
+ * You must hold netif_tx_lock() to call this function.
+ */
+netdev_tx_t efx_enqueue_skb(struct efx_tx_queue *tx_queue, struct sk_buff *skb)
+{
+	struct efx_nic *efx = tx_queue->efx;
+	struct pci_dev *pci_dev = efx->pci_dev;
+	struct efx_tx_buffer *buffer;
+	skb_frag_t *fragment;
+	struct page *page;
+	int page_offset;
+	unsigned int len, unmap_len = 0, fill_level, insert_ptr;
+	dma_addr_t dma_addr, unmap_addr = 0;
+	unsigned int dma_len;
+	bool unmap_single;
+	int q_space, i = 0;
+	netdev_tx_t rc = NETDEV_TX_OK;
+
+	EFX_BUG_ON_PARANOID(tx_queue->write_count != tx_queue->insert_count);
+
+	if (skb_shinfo(skb)->gso_size)
+		return efx_enqueue_skb_tso(tx_queue, skb);
+
+	/* Get size of the initial fragment */
+	len = skb_headlen(skb);
+
+	/* Pad if necessary */
+	if (EFX_WORKAROUND_15592(efx) && skb->len <= 32) {
+		EFX_BUG_ON_PARANOID(skb->data_len);
+		len = 32 + 1;
+		if (skb_pad(skb, len - skb->len))
+			return NETDEV_TX_OK;
+	}
+
+	fill_level = tx_queue->insert_count - tx_queue->old_read_count;
+	q_space = efx->txq_entries - 1 - fill_level;
+
+	/* Map for DMA.  Use pci_map_single rather than pci_map_page
+	 * since this is more efficient on machines with sparse
+	 * memory.
+	 */
+	unmap_single = true;
+	dma_addr = pci_map_single(pci_dev, skb->data, len, PCI_DMA_TODEVICE);
+
+	/* Process all fragments */
+	while (1) {
+		if (unlikely(pci_dma_mapping_error(pci_dev, dma_addr)))
+			goto pci_err;
+
+		/* Store fields for marking in the per-fragment final
+		 * descriptor */
+		unmap_len = len;
+		unmap_addr = dma_addr;
+
+		/* Add to TX queue, splitting across DMA boundaries */
+		do {
+			if (unlikely(q_space-- <= 0)) {
+				/* It might be that completions have
+				 * happened since the xmit path last
+				 * checked.  Update the xmit path's
+				 * copy of read_count.
+				 */
+				netif_tx_stop_queue(tx_queue->core_txq);
+				/* This memory barrier protects the
+				 * change of queue state from the access
+				 * of read_count. */
+				smp_mb();
+				tx_queue->old_read_count =
+					ACCESS_ONCE(tx_queue->read_count);
+				fill_level = (tx_queue->insert_count
+					      - tx_queue->old_read_count);
+				q_space = efx->txq_entries - 1 - fill_level;
+				if (unlikely(q_space-- <= 0)) {
+					rc = NETDEV_TX_BUSY;
+					goto unwind;
+				}
+				smp_mb();
+				if (likely(!efx->loopback_selftest))
+					netif_tx_start_queue(
+						tx_queue->core_txq);
+			}
+
+			insert_ptr = tx_queue->insert_count & tx_queue->ptr_mask;
+			buffer = &tx_queue->buffer[insert_ptr];
+			efx_tsoh_free(tx_queue, buffer);
+			EFX_BUG_ON_PARANOID(buffer->tsoh);
+			EFX_BUG_ON_PARANOID(buffer->skb);
+			EFX_BUG_ON_PARANOID(buffer->len);
+			EFX_BUG_ON_PARANOID(!buffer->continuation);
+			EFX_BUG_ON_PARANOID(buffer->unmap_len);
+
+			dma_len = efx_max_tx_len(efx, dma_addr);
+			if (likely(dma_len >= len))
+				dma_len = len;
+
+			/* Fill out per descriptor fields */
+			buffer->len = dma_len;
+			buffer->dma_addr = dma_addr;
+			len -= dma_len;
+			dma_addr += dma_len;
+			++tx_queue->insert_count;
+		} while (len);
+
+		/* Transfer ownership of the unmapping to the final buffer */
+		buffer->unmap_single = unmap_single;
+		buffer->unmap_len = unmap_len;
+		unmap_len = 0;
+
+		/* Get address and size of next fragment */
+		if (i >= skb_shinfo(skb)->nr_frags)
+			break;
+		fragment = &skb_shinfo(skb)->frags[i];
+		len = fragment->size;
+		page = fragment->page;
+		page_offset = fragment->page_offset;
+		i++;
+		/* Map for DMA */
+		unmap_single = false;
+		dma_addr = pci_map_page(pci_dev, page, page_offset, len,
+					PCI_DMA_TODEVICE);
+	}
+
+	/* Transfer ownership of the skb to the final buffer */
+	buffer->skb = skb;
+	buffer->continuation = false;
+
+	/* Pass off to hardware */
+	efx_nic_push_buffers(tx_queue);
+
+	return NETDEV_TX_OK;
+
+ pci_err:
+	netif_err(efx, tx_err, efx->net_dev,
+		  " TX queue %d could not map skb with %d bytes %d "
+		  "fragments for DMA\n", tx_queue->queue, skb->len,
+		  skb_shinfo(skb)->nr_frags + 1);
+
+	/* Mark the packet as transmitted, and free the SKB ourselves */
+	dev_kfree_skb_any(skb);
+
+ unwind:
+	/* Work backwards until we hit the original insert pointer value */
+	while (tx_queue->insert_count != tx_queue->write_count) {
+		--tx_queue->insert_count;
+		insert_ptr = tx_queue->insert_count & tx_queue->ptr_mask;
+		buffer = &tx_queue->buffer[insert_ptr];
+		efx_dequeue_buffer(tx_queue, buffer);
+		buffer->len = 0;
+	}
+
+	/* Free the fragment we were mid-way through pushing */
+	if (unmap_len) {
+		if (unmap_single)
+			pci_unmap_single(pci_dev, unmap_addr, unmap_len,
+					 PCI_DMA_TODEVICE);
+		else
+			pci_unmap_page(pci_dev, unmap_addr, unmap_len,
+				       PCI_DMA_TODEVICE);
+	}
+
+	return rc;
+}
+
+/* Remove packets from the TX queue
+ *
+ * This removes packets from the TX queue, up to and including the
+ * specified index.
+ */
+static void efx_dequeue_buffers(struct efx_tx_queue *tx_queue,
+				unsigned int index)
+{
+	struct efx_nic *efx = tx_queue->efx;
+	unsigned int stop_index, read_ptr;
+
+	stop_index = (index + 1) & tx_queue->ptr_mask;
+	read_ptr = tx_queue->read_count & tx_queue->ptr_mask;
+
+	while (read_ptr != stop_index) {
+		struct efx_tx_buffer *buffer = &tx_queue->buffer[read_ptr];
+		if (unlikely(buffer->len == 0)) {
+			netif_err(efx, tx_err, efx->net_dev,
+				  "TX queue %d spurious TX completion id %x\n",
+				  tx_queue->queue, read_ptr);
+			efx_schedule_reset(efx, RESET_TYPE_TX_SKIP);
+			return;
+		}
+
+		efx_dequeue_buffer(tx_queue, buffer);
+		buffer->continuation = true;
+		buffer->len = 0;
+
+		++tx_queue->read_count;
+		read_ptr = tx_queue->read_count & tx_queue->ptr_mask;
+	}
+}
+
+/* Initiate a packet transmission.  We use one channel per CPU
+ * (sharing when we have more CPUs than channels).  On Falcon, the TX
+ * completion events will be directed back to the CPU that transmitted
+ * the packet, which should be cache-efficient.
+ *
+ * Context: non-blocking.
+ * Note that returning anything other than NETDEV_TX_OK will cause the
+ * OS to free the skb.
+ */
+netdev_tx_t efx_hard_start_xmit(struct sk_buff *skb,
+				      struct net_device *net_dev)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	struct efx_tx_queue *tx_queue;
+	unsigned index, type;
+
+	EFX_WARN_ON_PARANOID(!netif_device_present(net_dev));
+
+	index = skb_get_queue_mapping(skb);
+	type = skb->ip_summed == CHECKSUM_PARTIAL ? EFX_TXQ_TYPE_OFFLOAD : 0;
+	if (index >= efx->n_tx_channels) {
+		index -= efx->n_tx_channels;
+		type |= EFX_TXQ_TYPE_HIGHPRI;
+	}
+	tx_queue = efx_get_tx_queue(efx, index, type);
+
+	return efx_enqueue_skb(tx_queue, skb);
+}
+
+void efx_init_tx_queue_core_txq(struct efx_tx_queue *tx_queue)
+{
+	struct efx_nic *efx = tx_queue->efx;
+
+	/* Must be inverse of queue lookup in efx_hard_start_xmit() */
+	tx_queue->core_txq =
+		netdev_get_tx_queue(efx->net_dev,
+				    tx_queue->queue / EFX_TXQ_TYPES +
+				    ((tx_queue->queue & EFX_TXQ_TYPE_HIGHPRI) ?
+				     efx->n_tx_channels : 0));
+}
+
+int efx_setup_tc(struct net_device *net_dev, u8 num_tc)
+{
+	struct efx_nic *efx = netdev_priv(net_dev);
+	struct efx_channel *channel;
+	struct efx_tx_queue *tx_queue;
+	unsigned tc;
+	int rc;
+
+	if (efx_nic_rev(efx) < EFX_REV_FALCON_B0 || num_tc > EFX_MAX_TX_TC)
+		return -EINVAL;
+
+	if (num_tc == net_dev->num_tc)
+		return 0;
+
+	for (tc = 0; tc < num_tc; tc++) {
+		net_dev->tc_to_txq[tc].offset = tc * efx->n_tx_channels;
+		net_dev->tc_to_txq[tc].count = efx->n_tx_channels;
+	}
+
+	if (num_tc > net_dev->num_tc) {
+		/* Initialise high-priority queues as necessary */
+		efx_for_each_channel(channel, efx) {
+			efx_for_each_possible_channel_tx_queue(tx_queue,
+							       channel) {
+				if (!(tx_queue->queue & EFX_TXQ_TYPE_HIGHPRI))
+					continue;
+				if (!tx_queue->buffer) {
+					rc = efx_probe_tx_queue(tx_queue);
+					if (rc)
+						return rc;
+				}
+				if (!tx_queue->initialised)
+					efx_init_tx_queue(tx_queue);
+				efx_init_tx_queue_core_txq(tx_queue);
+			}
+		}
+	} else {
+		/* Reduce number of classes before number of queues */
+		net_dev->num_tc = num_tc;
+	}
+
+	rc = netif_set_real_num_tx_queues(net_dev,
+					  max_t(int, num_tc, 1) *
+					  efx->n_tx_channels);
+	if (rc)
+		return rc;
+
+	/* Do not destroy high-priority queues when they become
+	 * unused.  We would have to flush them first, and it is
+	 * fairly difficult to flush a subset of TX queues.  Leave
+	 * it to efx_fini_channels().
+	 */
+
+	net_dev->num_tc = num_tc;
+	return 0;
+}
+
+void efx_xmit_done(struct efx_tx_queue *tx_queue, unsigned int index)
+{
+	unsigned fill_level;
+	struct efx_nic *efx = tx_queue->efx;
+
+	EFX_BUG_ON_PARANOID(index > tx_queue->ptr_mask);
+
+	efx_dequeue_buffers(tx_queue, index);
+
+	/* See if we need to restart the netif queue.  This barrier
+	 * separates the update of read_count from the test of the
+	 * queue state. */
+	smp_mb();
+	if (unlikely(netif_tx_queue_stopped(tx_queue->core_txq)) &&
+	    likely(efx->port_enabled) &&
+	    likely(netif_device_present(efx->net_dev))) {
+		fill_level = tx_queue->insert_count - tx_queue->read_count;
+		if (fill_level < EFX_TXQ_THRESHOLD(efx)) {
+			EFX_BUG_ON_PARANOID(!efx_dev_registered(efx));
+			netif_tx_wake_queue(tx_queue->core_txq);
+		}
+	}
+
+	/* Check whether the hardware queue is now empty */
+	if ((int)(tx_queue->read_count - tx_queue->old_write_count) >= 0) {
+		tx_queue->old_write_count = ACCESS_ONCE(tx_queue->write_count);
+		if (tx_queue->read_count == tx_queue->old_write_count) {
+			smp_mb();
+			tx_queue->empty_read_count =
+				tx_queue->read_count | EFX_EMPTY_COUNT_VALID;
+		}
+	}
+}
+
+int efx_probe_tx_queue(struct efx_tx_queue *tx_queue)
+{
+	struct efx_nic *efx = tx_queue->efx;
+	unsigned int entries;
+	int i, rc;
+
+	/* Create the smallest power-of-two aligned ring */
+	entries = max(roundup_pow_of_two(efx->txq_entries), EFX_MIN_DMAQ_SIZE);
+	EFX_BUG_ON_PARANOID(entries > EFX_MAX_DMAQ_SIZE);
+	tx_queue->ptr_mask = entries - 1;
+
+	netif_dbg(efx, probe, efx->net_dev,
+		  "creating TX queue %d size %#x mask %#x\n",
+		  tx_queue->queue, efx->txq_entries, tx_queue->ptr_mask);
+
+	/* Allocate software ring */
+	tx_queue->buffer = kzalloc(entries * sizeof(*tx_queue->buffer),
+				   GFP_KERNEL);
+	if (!tx_queue->buffer)
+		return -ENOMEM;
+	for (i = 0; i <= tx_queue->ptr_mask; ++i)
+		tx_queue->buffer[i].continuation = true;
+
+	/* Allocate hardware ring */
+	rc = efx_nic_probe_tx(tx_queue);
+	if (rc)
+		goto fail;
+
+	return 0;
+
+ fail:
+	kfree(tx_queue->buffer);
+	tx_queue->buffer = NULL;
+	return rc;
+}
+
+void efx_init_tx_queue(struct efx_tx_queue *tx_queue)
+{
+	netif_dbg(tx_queue->efx, drv, tx_queue->efx->net_dev,
+		  "initialising TX queue %d\n", tx_queue->queue);
+
+	tx_queue->insert_count = 0;
+	tx_queue->write_count = 0;
+	tx_queue->old_write_count = 0;
+	tx_queue->read_count = 0;
+	tx_queue->old_read_count = 0;
+	tx_queue->empty_read_count = 0 | EFX_EMPTY_COUNT_VALID;
+
+	/* Set up TX descriptor ring */
+	efx_nic_init_tx(tx_queue);
+
+	tx_queue->initialised = true;
+}
+
+void efx_release_tx_buffers(struct efx_tx_queue *tx_queue)
+{
+	struct efx_tx_buffer *buffer;
+
+	if (!tx_queue->buffer)
+		return;
+
+	/* Free any buffers left in the ring */
+	while (tx_queue->read_count != tx_queue->write_count) {
+		buffer = &tx_queue->buffer[tx_queue->read_count & tx_queue->ptr_mask];
+		efx_dequeue_buffer(tx_queue, buffer);
+		buffer->continuation = true;
+		buffer->len = 0;
+
+		++tx_queue->read_count;
+	}
+}
+
+void efx_fini_tx_queue(struct efx_tx_queue *tx_queue)
+{
+	if (!tx_queue->initialised)
+		return;
+
+	netif_dbg(tx_queue->efx, drv, tx_queue->efx->net_dev,
+		  "shutting down TX queue %d\n", tx_queue->queue);
+
+	tx_queue->initialised = false;
+
+	/* Flush TX queue, remove descriptor ring */
+	efx_nic_fini_tx(tx_queue);
+
+	efx_release_tx_buffers(tx_queue);
+
+	/* Free up TSO header cache */
+	efx_fini_tso(tx_queue);
+}
+
+void efx_remove_tx_queue(struct efx_tx_queue *tx_queue)
+{
+	if (!tx_queue->buffer)
+		return;
+
+	netif_dbg(tx_queue->efx, drv, tx_queue->efx->net_dev,
+		  "destroying TX queue %d\n", tx_queue->queue);
+	efx_nic_remove_tx(tx_queue);
+
+	kfree(tx_queue->buffer);
+	tx_queue->buffer = NULL;
+}
+
+
+/* Efx TCP segmentation acceleration.
+ *
+ * Why?  Because by doing it here in the driver we can go significantly
+ * faster than the GSO.
+ *
+ * Requires TX checksum offload support.
+ */
+
+/* Number of bytes inserted at the start of a TSO header buffer,
+ * similar to NET_IP_ALIGN.
+ */
+#ifdef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
+#define TSOH_OFFSET	0
+#else
+#define TSOH_OFFSET	NET_IP_ALIGN
+#endif
+
+#define TSOH_BUFFER(tsoh)	((u8 *)(tsoh + 1) + TSOH_OFFSET)
+
+/* Total size of struct efx_tso_header, buffer and padding */
+#define TSOH_SIZE(hdr_len)					\
+	(sizeof(struct efx_tso_header) + TSOH_OFFSET + hdr_len)
+
+/* Size of blocks on free list.  Larger blocks must be allocated from
+ * the heap.
+ */
+#define TSOH_STD_SIZE		128
+
+#define PTR_DIFF(p1, p2)  ((u8 *)(p1) - (u8 *)(p2))
+#define ETH_HDR_LEN(skb)  (skb_network_header(skb) - (skb)->data)
+#define SKB_TCP_OFF(skb)  PTR_DIFF(tcp_hdr(skb), (skb)->data)
+#define SKB_IPV4_OFF(skb) PTR_DIFF(ip_hdr(skb), (skb)->data)
+#define SKB_IPV6_OFF(skb) PTR_DIFF(ipv6_hdr(skb), (skb)->data)
+
+/**
+ * struct tso_state - TSO state for an SKB
+ * @out_len: Remaining length in current segment
+ * @seqnum: Current sequence number
+ * @ipv4_id: Current IPv4 ID, host endian
+ * @packet_space: Remaining space in current packet
+ * @dma_addr: DMA address of current position
+ * @in_len: Remaining length in current SKB fragment
+ * @unmap_len: Length of SKB fragment
+ * @unmap_addr: DMA address of SKB fragment
+ * @unmap_single: DMA single vs page mapping flag
+ * @protocol: Network protocol (after any VLAN header)
+ * @header_len: Number of bytes of header
+ * @full_packet_size: Number of bytes to put in each outgoing segment
+ *
+ * The state used during segmentation.  It is put into this data structure
+ * just to make it easy to pass into inline functions.
+ */
+struct tso_state {
+	/* Output position */
+	unsigned out_len;
+	unsigned seqnum;
+	unsigned ipv4_id;
+	unsigned packet_space;
+
+	/* Input position */
+	dma_addr_t dma_addr;
+	unsigned in_len;
+	unsigned unmap_len;
+	dma_addr_t unmap_addr;
+	bool unmap_single;
+
+	__be16 protocol;
+	unsigned header_len;
+	int full_packet_size;
+};
+
+
+/*
+ * Verify that our various assumptions about sk_buffs and the conditions
+ * under which TSO will be attempted hold true.  Return the protocol number.
+ */
+static __be16 efx_tso_check_protocol(struct sk_buff *skb)
+{
+	__be16 protocol = skb->protocol;
+
+	EFX_BUG_ON_PARANOID(((struct ethhdr *)skb->data)->h_proto !=
+			    protocol);
+	if (protocol == htons(ETH_P_8021Q)) {
+		/* Find the encapsulated protocol; reset network header
+		 * and transport header based on that. */
+		struct vlan_ethhdr *veh = (struct vlan_ethhdr *)skb->data;
+		protocol = veh->h_vlan_encapsulated_proto;
+		skb_set_network_header(skb, sizeof(*veh));
+		if (protocol == htons(ETH_P_IP))
+			skb_set_transport_header(skb, sizeof(*veh) +
+						 4 * ip_hdr(skb)->ihl);
+		else if (protocol == htons(ETH_P_IPV6))
+			skb_set_transport_header(skb, sizeof(*veh) +
+						 sizeof(struct ipv6hdr));
+	}
+
+	if (protocol == htons(ETH_P_IP)) {
+		EFX_BUG_ON_PARANOID(ip_hdr(skb)->protocol != IPPROTO_TCP);
+	} else {
+		EFX_BUG_ON_PARANOID(protocol != htons(ETH_P_IPV6));
+		EFX_BUG_ON_PARANOID(ipv6_hdr(skb)->nexthdr != NEXTHDR_TCP);
+	}
+	EFX_BUG_ON_PARANOID((PTR_DIFF(tcp_hdr(skb), skb->data)
+			     + (tcp_hdr(skb)->doff << 2u)) >
+			    skb_headlen(skb));
+
+	return protocol;
+}
+
+
+/*
+ * Allocate a page worth of efx_tso_header structures, and string them
+ * into the tx_queue->tso_headers_free linked list. Return 0 or -ENOMEM.
+ */
+static int efx_tsoh_block_alloc(struct efx_tx_queue *tx_queue)
+{
+
+	struct pci_dev *pci_dev = tx_queue->efx->pci_dev;
+	struct efx_tso_header *tsoh;
+	dma_addr_t dma_addr;
+	u8 *base_kva, *kva;
+
+	base_kva = pci_alloc_consistent(pci_dev, PAGE_SIZE, &dma_addr);
+	if (base_kva == NULL) {
+		netif_err(tx_queue->efx, tx_err, tx_queue->efx->net_dev,
+			  "Unable to allocate page for TSO headers\n");
+		return -ENOMEM;
+	}
+
+	/* pci_alloc_consistent() allocates pages. */
+	EFX_BUG_ON_PARANOID(dma_addr & (PAGE_SIZE - 1u));
+
+	for (kva = base_kva; kva < base_kva + PAGE_SIZE; kva += TSOH_STD_SIZE) {
+		tsoh = (struct efx_tso_header *)kva;
+		tsoh->dma_addr = dma_addr + (TSOH_BUFFER(tsoh) - base_kva);
+		tsoh->next = tx_queue->tso_headers_free;
+		tx_queue->tso_headers_free = tsoh;
+	}
+
+	return 0;
+}
+
+
+/* Free up a TSO header, and all others in the same page. */
+static void efx_tsoh_block_free(struct efx_tx_queue *tx_queue,
+				struct efx_tso_header *tsoh,
+				struct pci_dev *pci_dev)
+{
+	struct efx_tso_header **p;
+	unsigned long base_kva;
+	dma_addr_t base_dma;
+
+	base_kva = (unsigned long)tsoh & PAGE_MASK;
+	base_dma = tsoh->dma_addr & PAGE_MASK;
+
+	p = &tx_queue->tso_headers_free;
+	while (*p != NULL) {
+		if (((unsigned long)*p & PAGE_MASK) == base_kva)
+			*p = (*p)->next;
+		else
+			p = &(*p)->next;
+	}
+
+	pci_free_consistent(pci_dev, PAGE_SIZE, (void *)base_kva, base_dma);
+}
+
+static struct efx_tso_header *
+efx_tsoh_heap_alloc(struct efx_tx_queue *tx_queue, size_t header_len)
+{
+	struct efx_tso_header *tsoh;
+
+	tsoh = kmalloc(TSOH_SIZE(header_len), GFP_ATOMIC | GFP_DMA);
+	if (unlikely(!tsoh))
+		return NULL;
+
+	tsoh->dma_addr = pci_map_single(tx_queue->efx->pci_dev,
+					TSOH_BUFFER(tsoh), header_len,
+					PCI_DMA_TODEVICE);
+	if (unlikely(pci_dma_mapping_error(tx_queue->efx->pci_dev,
+					   tsoh->dma_addr))) {
+		kfree(tsoh);
+		return NULL;
+	}
+
+	tsoh->unmap_len = header_len;
+	return tsoh;
+}
+
+static void
+efx_tsoh_heap_free(struct efx_tx_queue *tx_queue, struct efx_tso_header *tsoh)
+{
+	pci_unmap_single(tx_queue->efx->pci_dev,
+			 tsoh->dma_addr, tsoh->unmap_len,
+			 PCI_DMA_TODEVICE);
+	kfree(tsoh);
+}
+
+/**
+ * efx_tx_queue_insert - push descriptors onto the TX queue
+ * @tx_queue:		Efx TX queue
+ * @dma_addr:		DMA address of fragment
+ * @len:		Length of fragment
+ * @final_buffer:	The final buffer inserted into the queue
+ *
+ * Push descriptors onto the TX queue.  Return 0 on success or 1 if
+ * @tx_queue full.
+ */
+static int efx_tx_queue_insert(struct efx_tx_queue *tx_queue,
+			       dma_addr_t dma_addr, unsigned len,
+			       struct efx_tx_buffer **final_buffer)
+{
+	struct efx_tx_buffer *buffer;
+	struct efx_nic *efx = tx_queue->efx;
+	unsigned dma_len, fill_level, insert_ptr;
+	int q_space;
+
+	EFX_BUG_ON_PARANOID(len <= 0);
+
+	fill_level = tx_queue->insert_count - tx_queue->old_read_count;
+	/* -1 as there is no way to represent all descriptors used */
+	q_space = efx->txq_entries - 1 - fill_level;
+
+	while (1) {
+		if (unlikely(q_space-- <= 0)) {
+			/* It might be that completions have happened
+			 * since the xmit path last checked.  Update
+			 * the xmit path's copy of read_count.
+			 */
+			netif_tx_stop_queue(tx_queue->core_txq);
+			/* This memory barrier protects the change of
+			 * queue state from the access of read_count. */
+			smp_mb();
+			tx_queue->old_read_count =
+				ACCESS_ONCE(tx_queue->read_count);
+			fill_level = (tx_queue->insert_count
+				      - tx_queue->old_read_count);
+			q_space = efx->txq_entries - 1 - fill_level;
+			if (unlikely(q_space-- <= 0)) {
+				*final_buffer = NULL;
+				return 1;
+			}
+			smp_mb();
+			netif_tx_start_queue(tx_queue->core_txq);
+		}
+
+		insert_ptr = tx_queue->insert_count & tx_queue->ptr_mask;
+		buffer = &tx_queue->buffer[insert_ptr];
+		++tx_queue->insert_count;
+
+		EFX_BUG_ON_PARANOID(tx_queue->insert_count -
+				    tx_queue->read_count >=
+				    efx->txq_entries);
+
+		efx_tsoh_free(tx_queue, buffer);
+		EFX_BUG_ON_PARANOID(buffer->len);
+		EFX_BUG_ON_PARANOID(buffer->unmap_len);
+		EFX_BUG_ON_PARANOID(buffer->skb);
+		EFX_BUG_ON_PARANOID(!buffer->continuation);
+		EFX_BUG_ON_PARANOID(buffer->tsoh);
+
+		buffer->dma_addr = dma_addr;
+
+		dma_len = efx_max_tx_len(efx, dma_addr);
+
+		/* If there is enough space to send then do so */
+		if (dma_len >= len)
+			break;
+
+		buffer->len = dma_len; /* Don't set the other members */
+		dma_addr += dma_len;
+		len -= dma_len;
+	}
+
+	EFX_BUG_ON_PARANOID(!len);
+	buffer->len = len;
+	*final_buffer = buffer;
+	return 0;
+}
+
+
+/*
+ * Put a TSO header into the TX queue.
+ *
+ * This is special-cased because we know that it is small enough to fit in
+ * a single fragment, and we know it doesn't cross a page boundary.  It
+ * also allows us to not worry about end-of-packet etc.
+ */
+static void efx_tso_put_header(struct efx_tx_queue *tx_queue,
+			       struct efx_tso_header *tsoh, unsigned len)
+{
+	struct efx_tx_buffer *buffer;
+
+	buffer = &tx_queue->buffer[tx_queue->insert_count & tx_queue->ptr_mask];
+	efx_tsoh_free(tx_queue, buffer);
+	EFX_BUG_ON_PARANOID(buffer->len);
+	EFX_BUG_ON_PARANOID(buffer->unmap_len);
+	EFX_BUG_ON_PARANOID(buffer->skb);
+	EFX_BUG_ON_PARANOID(!buffer->continuation);
+	EFX_BUG_ON_PARANOID(buffer->tsoh);
+	buffer->len = len;
+	buffer->dma_addr = tsoh->dma_addr;
+	buffer->tsoh = tsoh;
+
+	++tx_queue->insert_count;
+}
+
+
+/* Remove descriptors put into a tx_queue. */
+static void efx_enqueue_unwind(struct efx_tx_queue *tx_queue)
+{
+	struct efx_tx_buffer *buffer;
+	dma_addr_t unmap_addr;
+
+	/* Work backwards until we hit the original insert pointer value */
+	while (tx_queue->insert_count != tx_queue->write_count) {
+		--tx_queue->insert_count;
+		buffer = &tx_queue->buffer[tx_queue->insert_count &
+					   tx_queue->ptr_mask];
+		efx_tsoh_free(tx_queue, buffer);
+		EFX_BUG_ON_PARANOID(buffer->skb);
+		if (buffer->unmap_len) {
+			unmap_addr = (buffer->dma_addr + buffer->len -
+				      buffer->unmap_len);
+			if (buffer->unmap_single)
+				pci_unmap_single(tx_queue->efx->pci_dev,
+						 unmap_addr, buffer->unmap_len,
+						 PCI_DMA_TODEVICE);
+			else
+				pci_unmap_page(tx_queue->efx->pci_dev,
+					       unmap_addr, buffer->unmap_len,
+					       PCI_DMA_TODEVICE);
+			buffer->unmap_len = 0;
+		}
+		buffer->len = 0;
+		buffer->continuation = true;
+	}
+}
+
+
+/* Parse the SKB header and initialise state. */
+static void tso_start(struct tso_state *st, const struct sk_buff *skb)
+{
+	/* All ethernet/IP/TCP headers combined size is TCP header size
+	 * plus offset of TCP header relative to start of packet.
+	 */
+	st->header_len = ((tcp_hdr(skb)->doff << 2u)
+			  + PTR_DIFF(tcp_hdr(skb), skb->data));
+	st->full_packet_size = st->header_len + skb_shinfo(skb)->gso_size;
+
+	if (st->protocol == htons(ETH_P_IP))
+		st->ipv4_id = ntohs(ip_hdr(skb)->id);
+	else
+		st->ipv4_id = 0;
+	st->seqnum = ntohl(tcp_hdr(skb)->seq);
+
+	EFX_BUG_ON_PARANOID(tcp_hdr(skb)->urg);
+	EFX_BUG_ON_PARANOID(tcp_hdr(skb)->syn);
+	EFX_BUG_ON_PARANOID(tcp_hdr(skb)->rst);
+
+	st->packet_space = st->full_packet_size;
+	st->out_len = skb->len - st->header_len;
+	st->unmap_len = 0;
+	st->unmap_single = false;
+}
+
+static int tso_get_fragment(struct tso_state *st, struct efx_nic *efx,
+			    skb_frag_t *frag)
+{
+	st->unmap_addr = pci_map_page(efx->pci_dev, frag->page,
+				      frag->page_offset, frag->size,
+				      PCI_DMA_TODEVICE);
+	if (likely(!pci_dma_mapping_error(efx->pci_dev, st->unmap_addr))) {
+		st->unmap_single = false;
+		st->unmap_len = frag->size;
+		st->in_len = frag->size;
+		st->dma_addr = st->unmap_addr;
+		return 0;
+	}
+	return -ENOMEM;
+}
+
+static int tso_get_head_fragment(struct tso_state *st, struct efx_nic *efx,
+				 const struct sk_buff *skb)
+{
+	int hl = st->header_len;
+	int len = skb_headlen(skb) - hl;
+
+	st->unmap_addr = pci_map_single(efx->pci_dev, skb->data + hl,
+					len, PCI_DMA_TODEVICE);
+	if (likely(!pci_dma_mapping_error(efx->pci_dev, st->unmap_addr))) {
+		st->unmap_single = true;
+		st->unmap_len = len;
+		st->in_len = len;
+		st->dma_addr = st->unmap_addr;
+		return 0;
+	}
+	return -ENOMEM;
+}
+
+
+/**
+ * tso_fill_packet_with_fragment - form descriptors for the current fragment
+ * @tx_queue:		Efx TX queue
+ * @skb:		Socket buffer
+ * @st:			TSO state
+ *
+ * Form descriptors for the current fragment, until we reach the end
+ * of fragment or end-of-packet.  Return 0 on success, 1 if not enough
+ * space in @tx_queue.
+ */
+static int tso_fill_packet_with_fragment(struct efx_tx_queue *tx_queue,
+					 const struct sk_buff *skb,
+					 struct tso_state *st)
+{
+	struct efx_tx_buffer *buffer;
+	int n, end_of_packet, rc;
+
+	if (st->in_len == 0)
+		return 0;
+	if (st->packet_space == 0)
+		return 0;
+
+	EFX_BUG_ON_PARANOID(st->in_len <= 0);
+	EFX_BUG_ON_PARANOID(st->packet_space <= 0);
+
+	n = min(st->in_len, st->packet_space);
+
+	st->packet_space -= n;
+	st->out_len -= n;
+	st->in_len -= n;
+
+	rc = efx_tx_queue_insert(tx_queue, st->dma_addr, n, &buffer);
+	if (likely(rc == 0)) {
+		if (st->out_len == 0)
+			/* Transfer ownership of the skb */
+			buffer->skb = skb;
+
+		end_of_packet = st->out_len == 0 || st->packet_space == 0;
+		buffer->continuation = !end_of_packet;
+
+		if (st->in_len == 0) {
+			/* Transfer ownership of the pci mapping */
+			buffer->unmap_len = st->unmap_len;
+			buffer->unmap_single = st->unmap_single;
+			st->unmap_len = 0;
+		}
+	}
+
+	st->dma_addr += n;
+	return rc;
+}
+
+
+/**
+ * tso_start_new_packet - generate a new header and prepare for the new packet
+ * @tx_queue:		Efx TX queue
+ * @skb:		Socket buffer
+ * @st:			TSO state
+ *
+ * Generate a new header and prepare for the new packet.  Return 0 on
+ * success, or -1 if failed to alloc header.
+ */
+static int tso_start_new_packet(struct efx_tx_queue *tx_queue,
+				const struct sk_buff *skb,
+				struct tso_state *st)
+{
+	struct efx_tso_header *tsoh;
+	struct tcphdr *tsoh_th;
+	unsigned ip_length;
+	u8 *header;
+
+	/* Allocate a DMA-mapped header buffer. */
+	if (likely(TSOH_SIZE(st->header_len) <= TSOH_STD_SIZE)) {
+		if (tx_queue->tso_headers_free == NULL) {
+			if (efx_tsoh_block_alloc(tx_queue))
+				return -1;
+		}
+		EFX_BUG_ON_PARANOID(!tx_queue->tso_headers_free);
+		tsoh = tx_queue->tso_headers_free;
+		tx_queue->tso_headers_free = tsoh->next;
+		tsoh->unmap_len = 0;
+	} else {
+		tx_queue->tso_long_headers++;
+		tsoh = efx_tsoh_heap_alloc(tx_queue, st->header_len);
+		if (unlikely(!tsoh))
+			return -1;
+	}
+
+	header = TSOH_BUFFER(tsoh);
+	tsoh_th = (struct tcphdr *)(header + SKB_TCP_OFF(skb));
+
+	/* Copy and update the headers. */
+	memcpy(header, skb->data, st->header_len);
+
+	tsoh_th->seq = htonl(st->seqnum);
+	st->seqnum += skb_shinfo(skb)->gso_size;
+	if (st->out_len > skb_shinfo(skb)->gso_size) {
+		/* This packet will not finish the TSO burst. */
+		ip_length = st->full_packet_size - ETH_HDR_LEN(skb);
+		tsoh_th->fin = 0;
+		tsoh_th->psh = 0;
+	} else {
+		/* This packet will be the last in the TSO burst. */
+		ip_length = st->header_len - ETH_HDR_LEN(skb) + st->out_len;
+		tsoh_th->fin = tcp_hdr(skb)->fin;
+		tsoh_th->psh = tcp_hdr(skb)->psh;
+	}
+
+	if (st->protocol == htons(ETH_P_IP)) {
+		struct iphdr *tsoh_iph =
+			(struct iphdr *)(header + SKB_IPV4_OFF(skb));
+
+		tsoh_iph->tot_len = htons(ip_length);
+
+		/* Linux leaves suitable gaps in the IP ID space for us to fill. */
+		tsoh_iph->id = htons(st->ipv4_id);
+		st->ipv4_id++;
+	} else {
+		struct ipv6hdr *tsoh_iph =
+			(struct ipv6hdr *)(header + SKB_IPV6_OFF(skb));
+
+		tsoh_iph->payload_len = htons(ip_length - sizeof(*tsoh_iph));
+	}
+
+	st->packet_space = skb_shinfo(skb)->gso_size;
+	++tx_queue->tso_packets;
+
+	/* Form a descriptor for this header. */
+	efx_tso_put_header(tx_queue, tsoh, st->header_len);
+
+	return 0;
+}
+
+
+/**
+ * efx_enqueue_skb_tso - segment and transmit a TSO socket buffer
+ * @tx_queue:		Efx TX queue
+ * @skb:		Socket buffer
+ *
+ * Context: You must hold netif_tx_lock() to call this function.
+ *
+ * Add socket buffer @skb to @tx_queue, doing TSO or return != 0 if
+ * @skb was not enqueued.  In all cases @skb is consumed.  Return
+ * %NETDEV_TX_OK or %NETDEV_TX_BUSY.
+ */
+static int efx_enqueue_skb_tso(struct efx_tx_queue *tx_queue,
+			       struct sk_buff *skb)
+{
+	struct efx_nic *efx = tx_queue->efx;
+	int frag_i, rc, rc2 = NETDEV_TX_OK;
+	struct tso_state state;
+
+	/* Find the packet protocol and sanity-check it */
+	state.protocol = efx_tso_check_protocol(skb);
+
+	EFX_BUG_ON_PARANOID(tx_queue->write_count != tx_queue->insert_count);
+
+	tso_start(&state, skb);
+
+	/* Assume that skb header area contains exactly the headers, and
+	 * all payload is in the frag list.
+	 */
+	if (skb_headlen(skb) == state.header_len) {
+		/* Grab the first payload fragment. */
+		EFX_BUG_ON_PARANOID(skb_shinfo(skb)->nr_frags < 1);
+		frag_i = 0;
+		rc = tso_get_fragment(&state, efx,
+				      skb_shinfo(skb)->frags + frag_i);
+		if (rc)
+			goto mem_err;
+	} else {
+		rc = tso_get_head_fragment(&state, efx, skb);
+		if (rc)
+			goto mem_err;
+		frag_i = -1;
+	}
+
+	if (tso_start_new_packet(tx_queue, skb, &state) < 0)
+		goto mem_err;
+
+	while (1) {
+		rc = tso_fill_packet_with_fragment(tx_queue, skb, &state);
+		if (unlikely(rc)) {
+			rc2 = NETDEV_TX_BUSY;
+			goto unwind;
+		}
+
+		/* Move onto the next fragment? */
+		if (state.in_len == 0) {
+			if (++frag_i >= skb_shinfo(skb)->nr_frags)
+				/* End of payload reached. */
+				break;
+			rc = tso_get_fragment(&state, efx,
+					      skb_shinfo(skb)->frags + frag_i);
+			if (rc)
+				goto mem_err;
+		}
+
+		/* Start at new packet? */
+		if (state.packet_space == 0 &&
+		    tso_start_new_packet(tx_queue, skb, &state) < 0)
+			goto mem_err;
+	}
+
+	/* Pass off to hardware */
+	efx_nic_push_buffers(tx_queue);
+
+	tx_queue->tso_bursts++;
+	return NETDEV_TX_OK;
+
+ mem_err:
+	netif_err(efx, tx_err, efx->net_dev,
+		  "Out of memory for TSO headers, or PCI mapping error\n");
+	dev_kfree_skb_any(skb);
+
+ unwind:
+	/* Free the DMA mapping we were in the process of writing out */
+	if (state.unmap_len) {
+		if (state.unmap_single)
+			pci_unmap_single(efx->pci_dev, state.unmap_addr,
+					 state.unmap_len, PCI_DMA_TODEVICE);
+		else
+			pci_unmap_page(efx->pci_dev, state.unmap_addr,
+				       state.unmap_len, PCI_DMA_TODEVICE);
+	}
+
+	efx_enqueue_unwind(tx_queue);
+	return rc2;
+}
+
+
+/*
+ * Free up all TSO datastructures associated with tx_queue. This
+ * routine should be called only once the tx_queue is both empty and
+ * will no longer be used.
+ */
+static void efx_fini_tso(struct efx_tx_queue *tx_queue)
+{
+	unsigned i;
+
+	if (tx_queue->buffer) {
+		for (i = 0; i <= tx_queue->ptr_mask; ++i)
+			efx_tsoh_free(tx_queue, &tx_queue->buffer[i]);
+	}
+
+	while (tx_queue->tso_headers_free != NULL)
+		efx_tsoh_block_free(tx_queue, tx_queue->tso_headers_free,
+				    tx_queue->efx->pci_dev);
+}
diff --git a/drivers/net/ethernet/sfc/txc43128_phy.c b/drivers/net/ethernet/sfc/txc43128_phy.c
new file mode 100644
index 0000000..7c21b33
--- /dev/null
+++ b/drivers/net/ethernet/sfc/txc43128_phy.c
@@ -0,0 +1,560 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2006-2011 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+/*
+ * Driver for Transwitch/Mysticom CX4 retimer
+ * see www.transwitch.com, part is TXC-43128
+ */
+
+#include <linux/delay.h>
+#include <linux/slab.h>
+#include "efx.h"
+#include "mdio_10g.h"
+#include "phy.h"
+#include "nic.h"
+
+/* We expect these MMDs to be in the package */
+#define TXC_REQUIRED_DEVS (MDIO_DEVS_PCS |	\
+			   MDIO_DEVS_PMAPMD |	\
+			   MDIO_DEVS_PHYXS)
+
+#define TXC_LOOPBACKS ((1 << LOOPBACK_PCS) |	\
+		       (1 << LOOPBACK_PMAPMD) |	\
+		       (1 << LOOPBACK_PHYXS_WS))
+
+/**************************************************************************
+ *
+ * Compile-time config
+ *
+ **************************************************************************
+ */
+#define TXCNAME "TXC43128"
+/* Total length of time we'll wait for the PHY to come out of reset (ms) */
+#define TXC_MAX_RESET_TIME	500
+/* Interval between checks (ms) */
+#define TXC_RESET_WAIT		10
+/* How long to run BIST (us) */
+#define TXC_BIST_DURATION	50
+
+/**************************************************************************
+ *
+ * Register definitions
+ *
+ **************************************************************************
+ */
+
+/* Command register */
+#define TXC_GLRGS_GLCMD		0xc004
+/* Useful bits in command register */
+/* Lane power-down */
+#define TXC_GLCMD_L01PD_LBN	5
+#define TXC_GLCMD_L23PD_LBN	6
+/* Limited SW reset: preserves configuration but
+ * initiates a logic reset. Self-clearing */
+#define TXC_GLCMD_LMTSWRST_LBN	14
+
+/* Signal Quality Control */
+#define TXC_GLRGS_GSGQLCTL	0xc01a
+/* Enable bit */
+#define TXC_GSGQLCT_SGQLEN_LBN	15
+/* Lane selection */
+#define TXC_GSGQLCT_LNSL_LBN	13
+#define TXC_GSGQLCT_LNSL_WIDTH	2
+
+/* Analog TX control */
+#define TXC_ALRGS_ATXCTL	0xc040
+/* Lane power-down */
+#define TXC_ATXCTL_TXPD3_LBN	15
+#define TXC_ATXCTL_TXPD2_LBN	14
+#define TXC_ATXCTL_TXPD1_LBN	13
+#define TXC_ATXCTL_TXPD0_LBN	12
+
+/* Amplitude on lanes 0, 1 */
+#define TXC_ALRGS_ATXAMP0	0xc041
+/* Amplitude on lanes 2, 3 */
+#define TXC_ALRGS_ATXAMP1	0xc042
+/* Bit position of value for lane 0 (or 2) */
+#define TXC_ATXAMP_LANE02_LBN	3
+/* Bit position of value for lane 1 (or 3) */
+#define TXC_ATXAMP_LANE13_LBN	11
+
+#define TXC_ATXAMP_1280_mV	0
+#define TXC_ATXAMP_1200_mV	8
+#define TXC_ATXAMP_1120_mV	12
+#define TXC_ATXAMP_1060_mV	14
+#define TXC_ATXAMP_0820_mV	25
+#define TXC_ATXAMP_0720_mV	26
+#define TXC_ATXAMP_0580_mV	27
+#define TXC_ATXAMP_0440_mV	28
+
+#define TXC_ATXAMP_0820_BOTH					\
+	((TXC_ATXAMP_0820_mV << TXC_ATXAMP_LANE02_LBN)		\
+	 | (TXC_ATXAMP_0820_mV << TXC_ATXAMP_LANE13_LBN))
+
+#define TXC_ATXAMP_DEFAULT	0x6060 /* From databook */
+
+/* Preemphasis on lanes 0, 1 */
+#define TXC_ALRGS_ATXPRE0	0xc043
+/* Preemphasis on lanes 2, 3 */
+#define TXC_ALRGS_ATXPRE1	0xc044
+
+#define TXC_ATXPRE_NONE 0
+#define TXC_ATXPRE_DEFAULT	0x1010 /* From databook */
+
+#define TXC_ALRGS_ARXCTL	0xc045
+/* Lane power-down */
+#define TXC_ARXCTL_RXPD3_LBN	15
+#define TXC_ARXCTL_RXPD2_LBN	14
+#define TXC_ARXCTL_RXPD1_LBN	13
+#define TXC_ARXCTL_RXPD0_LBN	12
+
+/* Main control */
+#define TXC_MRGS_CTL		0xc340
+/* Bits in main control */
+#define TXC_MCTL_RESET_LBN	15	/* Self clear */
+#define TXC_MCTL_TXLED_LBN	14	/* 1 to show align status */
+#define TXC_MCTL_RXLED_LBN	13	/* 1 to show align status */
+
+/* GPIO output */
+#define TXC_GPIO_OUTPUT		0xc346
+#define TXC_GPIO_DIR		0xc348
+
+/* Vendor-specific BIST registers */
+#define TXC_BIST_CTL		0xc280
+#define TXC_BIST_TXFRMCNT	0xc281
+#define TXC_BIST_RX0FRMCNT	0xc282
+#define TXC_BIST_RX1FRMCNT	0xc283
+#define TXC_BIST_RX2FRMCNT	0xc284
+#define TXC_BIST_RX3FRMCNT	0xc285
+#define TXC_BIST_RX0ERRCNT	0xc286
+#define TXC_BIST_RX1ERRCNT	0xc287
+#define TXC_BIST_RX2ERRCNT	0xc288
+#define TXC_BIST_RX3ERRCNT	0xc289
+
+/* BIST type (controls bit patter in test) */
+#define TXC_BIST_CTRL_TYPE_LBN	10
+#define TXC_BIST_CTRL_TYPE_TSD	0	/* TranSwitch Deterministic */
+#define TXC_BIST_CTRL_TYPE_CRP	1	/* CRPAT standard */
+#define TXC_BIST_CTRL_TYPE_CJP	2	/* CJPAT standard */
+#define TXC_BIST_CTRL_TYPE_TSR	3	/* TranSwitch pseudo-random */
+/* Set this to 1 for 10 bit and 0 for 8 bit */
+#define TXC_BIST_CTRL_B10EN_LBN	12
+/* Enable BIST (write 0 to disable) */
+#define TXC_BIST_CTRL_ENAB_LBN	13
+/* Stop BIST (self-clears when stop complete) */
+#define TXC_BIST_CTRL_STOP_LBN	14
+/* Start BIST (cleared by writing 1 to STOP) */
+#define TXC_BIST_CTRL_STRT_LBN	15
+
+/* Mt. Diablo test configuration */
+#define TXC_MTDIABLO_CTRL	0xc34f
+#define TXC_MTDIABLO_CTRL_PMA_LOOP_LBN	10
+
+struct txc43128_data {
+	unsigned long bug10934_timer;
+	enum efx_phy_mode phy_mode;
+	enum efx_loopback_mode loopback_mode;
+};
+
+/* The PHY sometimes needs a reset to bring the link back up.  So long as
+ * it reports link down, we reset it every 5 seconds.
+ */
+#define BUG10934_RESET_INTERVAL (5 * HZ)
+
+/* Perform a reset that doesn't clear configuration changes */
+static void txc_reset_logic(struct efx_nic *efx);
+
+/* Set the output value of a gpio */
+void falcon_txc_set_gpio_val(struct efx_nic *efx, int pin, int on)
+{
+	efx_mdio_set_flag(efx, MDIO_MMD_PHYXS, TXC_GPIO_OUTPUT, 1 << pin, on);
+}
+
+/* Set up the GPIO direction register */
+void falcon_txc_set_gpio_dir(struct efx_nic *efx, int pin, int dir)
+{
+	efx_mdio_set_flag(efx, MDIO_MMD_PHYXS, TXC_GPIO_DIR, 1 << pin, dir);
+}
+
+/* Reset the PMA/PMD MMD. The documentation is explicit that this does a
+ * global reset (it's less clear what reset of other MMDs does).*/
+static int txc_reset_phy(struct efx_nic *efx)
+{
+	int rc = efx_mdio_reset_mmd(efx, MDIO_MMD_PMAPMD,
+				    TXC_MAX_RESET_TIME / TXC_RESET_WAIT,
+				    TXC_RESET_WAIT);
+	if (rc < 0)
+		goto fail;
+
+	/* Check that all the MMDs we expect are present and responding. */
+	rc = efx_mdio_check_mmds(efx, TXC_REQUIRED_DEVS);
+	if (rc < 0)
+		goto fail;
+
+	return 0;
+
+fail:
+	netif_err(efx, hw, efx->net_dev, TXCNAME ": reset timed out!\n");
+	return rc;
+}
+
+/* Run a single BIST on one MMD */
+static int txc_bist_one(struct efx_nic *efx, int mmd, int test)
+{
+	int ctrl, bctl;
+	int lane;
+	int rc = 0;
+
+	/* Set PMA to test into loopback using Mt Diablo reg as per app note */
+	ctrl = efx_mdio_read(efx, MDIO_MMD_PCS, TXC_MTDIABLO_CTRL);
+	ctrl |= (1 << TXC_MTDIABLO_CTRL_PMA_LOOP_LBN);
+	efx_mdio_write(efx, MDIO_MMD_PCS, TXC_MTDIABLO_CTRL, ctrl);
+
+	/* The BIST app. note lists these  as 3 distinct steps. */
+	/* Set the BIST type */
+	bctl = (test << TXC_BIST_CTRL_TYPE_LBN);
+	efx_mdio_write(efx, mmd, TXC_BIST_CTL, bctl);
+
+	/* Set the BSTEN bit in the BIST Control register to enable */
+	bctl |= (1 << TXC_BIST_CTRL_ENAB_LBN);
+	efx_mdio_write(efx, mmd, TXC_BIST_CTL, bctl);
+
+	/* Set the BSTRT bit in the BIST Control register */
+	efx_mdio_write(efx, mmd, TXC_BIST_CTL,
+		       bctl | (1 << TXC_BIST_CTRL_STRT_LBN));
+
+	/* Wait. */
+	udelay(TXC_BIST_DURATION);
+
+	/* Set the BSTOP bit in the BIST Control register */
+	bctl |= (1 << TXC_BIST_CTRL_STOP_LBN);
+	efx_mdio_write(efx, mmd, TXC_BIST_CTL, bctl);
+
+	/* The STOP bit should go off when things have stopped */
+	while (bctl & (1 << TXC_BIST_CTRL_STOP_LBN))
+		bctl = efx_mdio_read(efx, mmd, TXC_BIST_CTL);
+
+	/* Check all the error counts are 0 and all the frame counts are
+	   non-zero */
+	for (lane = 0; lane < 4; lane++) {
+		int count = efx_mdio_read(efx, mmd, TXC_BIST_RX0ERRCNT + lane);
+		if (count != 0) {
+			netif_err(efx, hw, efx->net_dev, TXCNAME": BIST error. "
+				  "Lane %d had %d errs\n", lane, count);
+			rc = -EIO;
+		}
+		count = efx_mdio_read(efx, mmd, TXC_BIST_RX0FRMCNT + lane);
+		if (count == 0) {
+			netif_err(efx, hw, efx->net_dev, TXCNAME": BIST error. "
+				  "Lane %d got 0 frames\n", lane);
+			rc = -EIO;
+		}
+	}
+
+	if (rc == 0)
+		netif_info(efx, hw, efx->net_dev, TXCNAME": BIST pass\n");
+
+	/* Disable BIST */
+	efx_mdio_write(efx, mmd, TXC_BIST_CTL, 0);
+
+	/* Turn off loopback */
+	ctrl &= ~(1 << TXC_MTDIABLO_CTRL_PMA_LOOP_LBN);
+	efx_mdio_write(efx, MDIO_MMD_PCS, TXC_MTDIABLO_CTRL, ctrl);
+
+	return rc;
+}
+
+static int txc_bist(struct efx_nic *efx)
+{
+	return txc_bist_one(efx, MDIO_MMD_PCS, TXC_BIST_CTRL_TYPE_TSD);
+}
+
+/* Push the non-configurable defaults into the PHY. This must be
+ * done after every full reset */
+static void txc_apply_defaults(struct efx_nic *efx)
+{
+	int mctrl;
+
+	/* Turn amplitude down and preemphasis off on the host side
+	 * (PHY<->MAC) as this is believed less likely to upset Falcon
+	 * and no adverse effects have been noted. It probably also
+	 * saves a picowatt or two */
+
+	/* Turn off preemphasis */
+	efx_mdio_write(efx, MDIO_MMD_PHYXS, TXC_ALRGS_ATXPRE0, TXC_ATXPRE_NONE);
+	efx_mdio_write(efx, MDIO_MMD_PHYXS, TXC_ALRGS_ATXPRE1, TXC_ATXPRE_NONE);
+
+	/* Turn down the amplitude */
+	efx_mdio_write(efx, MDIO_MMD_PHYXS,
+		       TXC_ALRGS_ATXAMP0, TXC_ATXAMP_0820_BOTH);
+	efx_mdio_write(efx, MDIO_MMD_PHYXS,
+		       TXC_ALRGS_ATXAMP1, TXC_ATXAMP_0820_BOTH);
+
+	/* Set the line side amplitude and preemphasis to the databook
+	 * defaults as an erratum causes them to be 0 on at least some
+	 * PHY rev.s */
+	efx_mdio_write(efx, MDIO_MMD_PMAPMD,
+		       TXC_ALRGS_ATXPRE0, TXC_ATXPRE_DEFAULT);
+	efx_mdio_write(efx, MDIO_MMD_PMAPMD,
+		       TXC_ALRGS_ATXPRE1, TXC_ATXPRE_DEFAULT);
+	efx_mdio_write(efx, MDIO_MMD_PMAPMD,
+		       TXC_ALRGS_ATXAMP0, TXC_ATXAMP_DEFAULT);
+	efx_mdio_write(efx, MDIO_MMD_PMAPMD,
+		       TXC_ALRGS_ATXAMP1, TXC_ATXAMP_DEFAULT);
+
+	/* Set up the LEDs  */
+	mctrl = efx_mdio_read(efx, MDIO_MMD_PHYXS, TXC_MRGS_CTL);
+
+	/* Set the Green and Red LEDs to their default modes */
+	mctrl &= ~((1 << TXC_MCTL_TXLED_LBN) | (1 << TXC_MCTL_RXLED_LBN));
+	efx_mdio_write(efx, MDIO_MMD_PHYXS, TXC_MRGS_CTL, mctrl);
+
+	/* Databook recommends doing this after configuration changes */
+	txc_reset_logic(efx);
+
+	falcon_board(efx)->type->init_phy(efx);
+}
+
+static int txc43128_phy_probe(struct efx_nic *efx)
+{
+	struct txc43128_data *phy_data;
+
+	/* Allocate phy private storage */
+	phy_data = kzalloc(sizeof(*phy_data), GFP_KERNEL);
+	if (!phy_data)
+		return -ENOMEM;
+	efx->phy_data = phy_data;
+	phy_data->phy_mode = efx->phy_mode;
+
+	efx->mdio.mmds = TXC_REQUIRED_DEVS;
+	efx->mdio.mode_support = MDIO_SUPPORTS_C45 | MDIO_EMULATE_C22;
+
+	efx->loopback_modes = TXC_LOOPBACKS | FALCON_XMAC_LOOPBACKS;
+
+	return 0;
+}
+
+/* Initialisation entry point for this PHY driver */
+static int txc43128_phy_init(struct efx_nic *efx)
+{
+	int rc;
+
+	rc = txc_reset_phy(efx);
+	if (rc < 0)
+		return rc;
+
+	rc = txc_bist(efx);
+	if (rc < 0)
+		return rc;
+
+	txc_apply_defaults(efx);
+
+	return 0;
+}
+
+/* Set the lane power down state in the global registers */
+static void txc_glrgs_lane_power(struct efx_nic *efx, int mmd)
+{
+	int pd = (1 << TXC_GLCMD_L01PD_LBN) | (1 << TXC_GLCMD_L23PD_LBN);
+	int ctl = efx_mdio_read(efx, mmd, TXC_GLRGS_GLCMD);
+
+	if (!(efx->phy_mode & PHY_MODE_LOW_POWER))
+		ctl &= ~pd;
+	else
+		ctl |= pd;
+
+	efx_mdio_write(efx, mmd, TXC_GLRGS_GLCMD, ctl);
+}
+
+/* Set the lane power down state in the analog control registers */
+static void txc_analog_lane_power(struct efx_nic *efx, int mmd)
+{
+	int txpd = (1 << TXC_ATXCTL_TXPD3_LBN) | (1 << TXC_ATXCTL_TXPD2_LBN)
+		| (1 << TXC_ATXCTL_TXPD1_LBN) | (1 << TXC_ATXCTL_TXPD0_LBN);
+	int rxpd = (1 << TXC_ARXCTL_RXPD3_LBN) | (1 << TXC_ARXCTL_RXPD2_LBN)
+		| (1 << TXC_ARXCTL_RXPD1_LBN) | (1 << TXC_ARXCTL_RXPD0_LBN);
+	int txctl = efx_mdio_read(efx, mmd, TXC_ALRGS_ATXCTL);
+	int rxctl = efx_mdio_read(efx, mmd, TXC_ALRGS_ARXCTL);
+
+	if (!(efx->phy_mode & PHY_MODE_LOW_POWER)) {
+		txctl &= ~txpd;
+		rxctl &= ~rxpd;
+	} else {
+		txctl |= txpd;
+		rxctl |= rxpd;
+	}
+
+	efx_mdio_write(efx, mmd, TXC_ALRGS_ATXCTL, txctl);
+	efx_mdio_write(efx, mmd, TXC_ALRGS_ARXCTL, rxctl);
+}
+
+static void txc_set_power(struct efx_nic *efx)
+{
+	/* According to the data book, all the MMDs can do low power */
+	efx_mdio_set_mmds_lpower(efx,
+				 !!(efx->phy_mode & PHY_MODE_LOW_POWER),
+				 TXC_REQUIRED_DEVS);
+
+	/* Global register bank is in PCS, PHY XS. These control the host
+	 * side and line side settings respectively. */
+	txc_glrgs_lane_power(efx, MDIO_MMD_PCS);
+	txc_glrgs_lane_power(efx, MDIO_MMD_PHYXS);
+
+	/* Analog register bank in PMA/PMD, PHY XS */
+	txc_analog_lane_power(efx, MDIO_MMD_PMAPMD);
+	txc_analog_lane_power(efx, MDIO_MMD_PHYXS);
+}
+
+static void txc_reset_logic_mmd(struct efx_nic *efx, int mmd)
+{
+	int val = efx_mdio_read(efx, mmd, TXC_GLRGS_GLCMD);
+	int tries = 50;
+
+	val |= (1 << TXC_GLCMD_LMTSWRST_LBN);
+	efx_mdio_write(efx, mmd, TXC_GLRGS_GLCMD, val);
+	while (tries--) {
+		val = efx_mdio_read(efx, mmd, TXC_GLRGS_GLCMD);
+		if (!(val & (1 << TXC_GLCMD_LMTSWRST_LBN)))
+			break;
+		udelay(1);
+	}
+	if (!tries)
+		netif_info(efx, hw, efx->net_dev,
+			   TXCNAME " Logic reset timed out!\n");
+}
+
+/* Perform a logic reset. This preserves the configuration registers
+ * and is needed for some configuration changes to take effect */
+static void txc_reset_logic(struct efx_nic *efx)
+{
+	/* The data sheet claims we can do the logic reset on either the
+	 * PCS or the PHYXS and the result is a reset of both host- and
+	 * line-side logic. */
+	txc_reset_logic_mmd(efx, MDIO_MMD_PCS);
+}
+
+static bool txc43128_phy_read_link(struct efx_nic *efx)
+{
+	return efx_mdio_links_ok(efx, TXC_REQUIRED_DEVS);
+}
+
+static int txc43128_phy_reconfigure(struct efx_nic *efx)
+{
+	struct txc43128_data *phy_data = efx->phy_data;
+	enum efx_phy_mode mode_change = efx->phy_mode ^ phy_data->phy_mode;
+	bool loop_change = LOOPBACK_CHANGED(phy_data, efx, TXC_LOOPBACKS);
+
+	if (efx->phy_mode & mode_change & PHY_MODE_TX_DISABLED) {
+		txc_reset_phy(efx);
+		txc_apply_defaults(efx);
+		falcon_reset_xaui(efx);
+		mode_change &= ~PHY_MODE_TX_DISABLED;
+	}
+
+	efx_mdio_transmit_disable(efx);
+	efx_mdio_phy_reconfigure(efx);
+	if (mode_change & PHY_MODE_LOW_POWER)
+		txc_set_power(efx);
+
+	/* The data sheet claims this is required after every reconfiguration
+	 * (note at end of 7.1), but we mustn't do it when nothing changes as
+	 * it glitches the link, and reconfigure gets called on link change,
+	 * so we get an IRQ storm on link up. */
+	if (loop_change || mode_change)
+		txc_reset_logic(efx);
+
+	phy_data->phy_mode = efx->phy_mode;
+	phy_data->loopback_mode = efx->loopback_mode;
+
+	return 0;
+}
+
+static void txc43128_phy_fini(struct efx_nic *efx)
+{
+	/* Disable link events */
+	efx_mdio_write(efx, MDIO_MMD_PMAPMD, MDIO_PMA_LASI_CTRL, 0);
+}
+
+static void txc43128_phy_remove(struct efx_nic *efx)
+{
+	kfree(efx->phy_data);
+	efx->phy_data = NULL;
+}
+
+/* Periodic callback: this exists mainly to poll link status as we
+ * don't use LASI interrupts */
+static bool txc43128_phy_poll(struct efx_nic *efx)
+{
+	struct txc43128_data *data = efx->phy_data;
+	bool was_up = efx->link_state.up;
+
+	efx->link_state.up = txc43128_phy_read_link(efx);
+	efx->link_state.speed = 10000;
+	efx->link_state.fd = true;
+	efx->link_state.fc = efx->wanted_fc;
+
+	if (efx->link_state.up || (efx->loopback_mode != LOOPBACK_NONE)) {
+		data->bug10934_timer = jiffies;
+	} else {
+		if (time_after_eq(jiffies, (data->bug10934_timer +
+					    BUG10934_RESET_INTERVAL))) {
+			data->bug10934_timer = jiffies;
+			txc_reset_logic(efx);
+		}
+	}
+
+	return efx->link_state.up != was_up;
+}
+
+static const char *txc43128_test_names[] = {
+	"bist"
+};
+
+static const char *txc43128_test_name(struct efx_nic *efx, unsigned int index)
+{
+	if (index < ARRAY_SIZE(txc43128_test_names))
+		return txc43128_test_names[index];
+	return NULL;
+}
+
+static int txc43128_run_tests(struct efx_nic *efx, int *results, unsigned flags)
+{
+	int rc;
+
+	if (!(flags & ETH_TEST_FL_OFFLINE))
+		return 0;
+
+	rc = txc_reset_phy(efx);
+	if (rc < 0)
+		return rc;
+
+	rc = txc_bist(efx);
+	txc_apply_defaults(efx);
+	results[0] = rc ? -1 : 1;
+	return rc;
+}
+
+static void txc43128_get_settings(struct efx_nic *efx, struct ethtool_cmd *ecmd)
+{
+	mdio45_ethtool_gset(&efx->mdio, ecmd);
+}
+
+const struct efx_phy_operations falcon_txc_phy_ops = {
+	.probe		= txc43128_phy_probe,
+	.init		= txc43128_phy_init,
+	.reconfigure	= txc43128_phy_reconfigure,
+	.poll		= txc43128_phy_poll,
+	.fini		= txc43128_phy_fini,
+	.remove		= txc43128_phy_remove,
+	.get_settings	= txc43128_get_settings,
+	.set_settings	= efx_mdio_set_settings,
+	.test_alive	= efx_mdio_test_alive,
+	.run_tests	= txc43128_run_tests,
+	.test_name	= txc43128_test_name,
+};
diff --git a/drivers/net/ethernet/sfc/workarounds.h b/drivers/net/ethernet/sfc/workarounds.h
new file mode 100644
index 0000000..99ff114
--- /dev/null
+++ b/drivers/net/ethernet/sfc/workarounds.h
@@ -0,0 +1,61 @@
+/****************************************************************************
+ * Driver for Solarflare Solarstorm network controllers and boards
+ * Copyright 2006-2010 Solarflare Communications Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation, incorporated herein by reference.
+ */
+
+#ifndef EFX_WORKAROUNDS_H
+#define EFX_WORKAROUNDS_H
+
+/*
+ * Hardware workarounds.
+ * Bug numbers are from Solarflare's Bugzilla.
+ */
+
+#define EFX_WORKAROUND_ALWAYS(efx) 1
+#define EFX_WORKAROUND_FALCON_A(efx) (efx_nic_rev(efx) <= EFX_REV_FALCON_A1)
+#define EFX_WORKAROUND_FALCON_AB(efx) (efx_nic_rev(efx) <= EFX_REV_FALCON_B0)
+#define EFX_WORKAROUND_SIENA(efx) (efx_nic_rev(efx) == EFX_REV_SIENA_A0)
+#define EFX_WORKAROUND_10G(efx) 1
+
+/* XAUI resets if link not detected */
+#define EFX_WORKAROUND_5147 EFX_WORKAROUND_ALWAYS
+/* RX PCIe double split performance issue */
+#define EFX_WORKAROUND_7575 EFX_WORKAROUND_ALWAYS
+/* Bit-bashed I2C reads cause performance drop */
+#define EFX_WORKAROUND_7884 EFX_WORKAROUND_10G
+/* TX_EV_PKT_ERR can be caused by a dangling TX descriptor
+ * or a PCIe error (bug 11028) */
+#define EFX_WORKAROUND_10727 EFX_WORKAROUND_ALWAYS
+/* Transmit flow control may get disabled */
+#define EFX_WORKAROUND_11482 EFX_WORKAROUND_FALCON_AB
+/* Truncated IPv4 packets can confuse the TX packet parser */
+#define EFX_WORKAROUND_15592 EFX_WORKAROUND_FALCON_AB
+/* Legacy ISR read can return zero once */
+#define EFX_WORKAROUND_15783 EFX_WORKAROUND_ALWAYS
+/* Legacy interrupt storm when interrupt fifo fills */
+#define EFX_WORKAROUND_17213 EFX_WORKAROUND_SIENA
+/* Write combining and sriov=enabled are incompatible */
+#define EFX_WORKAROUND_22643 EFX_WORKAROUND_SIENA
+
+/* Spurious parity errors in TSORT buffers */
+#define EFX_WORKAROUND_5129 EFX_WORKAROUND_FALCON_A
+/* Unaligned read request >512 bytes after aligning may break TSORT */
+#define EFX_WORKAROUND_5391 EFX_WORKAROUND_FALCON_A
+/* iSCSI parsing errors */
+#define EFX_WORKAROUND_5583 EFX_WORKAROUND_FALCON_A
+/* RX events go missing */
+#define EFX_WORKAROUND_5676 EFX_WORKAROUND_FALCON_A
+/* RX_RESET on A1 */
+#define EFX_WORKAROUND_6555 EFX_WORKAROUND_FALCON_A
+/* Increase filter depth to avoid RX_RESET */
+#define EFX_WORKAROUND_7244 EFX_WORKAROUND_FALCON_A
+/* Flushes may never complete */
+#define EFX_WORKAROUND_7803 EFX_WORKAROUND_FALCON_AB
+/* Leak overlength packets rather than free */
+#define EFX_WORKAROUND_8071 EFX_WORKAROUND_FALCON_A
+
+#endif /* EFX_WORKAROUNDS_H */