Merge commit 'v2.6.30-rc5' into core/iommu
Merge reason: core/iommu was on an .30-rc1 base,
update it to .30-rc5 to refresh.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug
index d8359e7..5865712 100644
--- a/arch/x86/Kconfig.debug
+++ b/arch/x86/Kconfig.debug
@@ -161,8 +161,7 @@
config IOMMU_LEAK
bool "IOMMU leak tracing"
- depends on DEBUG_KERNEL
- depends on IOMMU_DEBUG
+ depends on IOMMU_DEBUG && DMA_API_DEBUG
---help---
Add a simple leak tracer to the IOMMU code. This is useful when you
are debugging a buggy device driver that leaks IOMMU mappings.
diff --git a/arch/x86/kernel/pci-calgary_64.c b/arch/x86/kernel/pci-calgary_64.c
index 755c21e..971a3be 100644
--- a/arch/x86/kernel/pci-calgary_64.c
+++ b/arch/x86/kernel/pci-calgary_64.c
@@ -186,37 +186,6 @@
static struct calgary_bus_info bus_info[MAX_PHB_BUS_NUM] = { { NULL, 0, 0 }, };
-/* enable this to stress test the chip's TCE cache */
-#ifdef CONFIG_IOMMU_DEBUG
-static int debugging = 1;
-
-static inline unsigned long verify_bit_range(unsigned long* bitmap,
- int expected, unsigned long start, unsigned long end)
-{
- unsigned long idx = start;
-
- BUG_ON(start >= end);
-
- while (idx < end) {
- if (!!test_bit(idx, bitmap) != expected)
- return idx;
- ++idx;
- }
-
- /* all bits have the expected value */
- return ~0UL;
-}
-#else /* debugging is disabled */
-static int debugging;
-
-static inline unsigned long verify_bit_range(unsigned long* bitmap,
- int expected, unsigned long start, unsigned long end)
-{
- return ~0UL;
-}
-
-#endif /* CONFIG_IOMMU_DEBUG */
-
static inline int translation_enabled(struct iommu_table *tbl)
{
/* only PHBs with translation enabled have an IOMMU table */
@@ -228,7 +197,6 @@
{
unsigned long index;
unsigned long end;
- unsigned long badbit;
unsigned long flags;
index = start_addr >> PAGE_SHIFT;
@@ -243,14 +211,6 @@
spin_lock_irqsave(&tbl->it_lock, flags);
- badbit = verify_bit_range(tbl->it_map, 0, index, end);
- if (badbit != ~0UL) {
- if (printk_ratelimit())
- printk(KERN_ERR "Calgary: entry already allocated at "
- "0x%lx tbl %p dma 0x%lx npages %u\n",
- badbit, tbl, start_addr, npages);
- }
-
iommu_area_reserve(tbl->it_map, index, npages);
spin_unlock_irqrestore(&tbl->it_lock, flags);
@@ -326,7 +286,6 @@
unsigned int npages)
{
unsigned long entry;
- unsigned long badbit;
unsigned long badend;
unsigned long flags;
@@ -346,14 +305,6 @@
spin_lock_irqsave(&tbl->it_lock, flags);
- badbit = verify_bit_range(tbl->it_map, 1, entry, entry + npages);
- if (badbit != ~0UL) {
- if (printk_ratelimit())
- printk(KERN_ERR "Calgary: bit is off at 0x%lx "
- "tbl %p dma 0x%Lx entry 0x%lx npages %u\n",
- badbit, tbl, dma_addr, entry, npages);
- }
-
iommu_area_free(tbl->it_map, entry, npages);
spin_unlock_irqrestore(&tbl->it_lock, flags);
@@ -1488,9 +1439,8 @@
iommu_detected = 1;
calgary_detected = 1;
printk(KERN_INFO "PCI-DMA: Calgary IOMMU detected.\n");
- printk(KERN_INFO "PCI-DMA: Calgary TCE table spec is %d, "
- "CONFIG_IOMMU_DEBUG is %s.\n", specified_table_size,
- debugging ? "enabled" : "disabled");
+ printk(KERN_INFO "PCI-DMA: Calgary TCE table spec is %d\n",
+ specified_table_size);
/* swiotlb for devices that aren't behind the Calgary. */
if (max_pfn > MAX_DMA32_PFN)
diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c
index b284b58..1e8920d 100644
--- a/arch/x86/kernel/pci-gart_64.c
+++ b/arch/x86/kernel/pci-gart_64.c
@@ -144,48 +144,21 @@
}
#ifdef CONFIG_IOMMU_LEAK
-
-#define SET_LEAK(x) \
- do { \
- if (iommu_leak_tab) \
- iommu_leak_tab[x] = __builtin_return_address(0);\
- } while (0)
-
-#define CLEAR_LEAK(x) \
- do { \
- if (iommu_leak_tab) \
- iommu_leak_tab[x] = NULL; \
- } while (0)
-
/* Debugging aid for drivers that don't free their IOMMU tables */
-static void **iommu_leak_tab;
static int leak_trace;
static int iommu_leak_pages = 20;
static void dump_leak(void)
{
- int i;
static int dump;
- if (dump || !iommu_leak_tab)
+ if (dump)
return;
dump = 1;
- show_stack(NULL, NULL);
- /* Very crude. dump some from the end of the table too */
- printk(KERN_DEBUG "Dumping %d pages from end of IOMMU:\n",
- iommu_leak_pages);
- for (i = 0; i < iommu_leak_pages; i += 2) {
- printk(KERN_DEBUG "%lu: ", iommu_pages-i);
- printk_address((unsigned long) iommu_leak_tab[iommu_pages-i],
- 0);
- printk(KERN_CONT "%c", (i+1)%2 == 0 ? '\n' : ' ');
- }
- printk(KERN_DEBUG "\n");
+ show_stack(NULL, NULL);
+ debug_dma_dump_mappings(NULL);
}
-#else
-# define SET_LEAK(x)
-# define CLEAR_LEAK(x)
#endif
static void iommu_full(struct device *dev, size_t size, int dir)
@@ -248,7 +221,6 @@
for (i = 0; i < npages; i++) {
iommu_gatt_base[iommu_page + i] = GPTE_ENCODE(phys_mem);
- SET_LEAK(iommu_page + i);
phys_mem += PAGE_SIZE;
}
return iommu_bus_base + iommu_page*PAGE_SIZE + (phys_mem & ~PAGE_MASK);
@@ -294,7 +266,6 @@
npages = iommu_num_pages(dma_addr, size, PAGE_SIZE);
for (i = 0; i < npages; i++) {
iommu_gatt_base[iommu_page + i] = gart_unmapped_entry;
- CLEAR_LEAK(iommu_page + i);
}
free_iommu(iommu_page, npages);
}
@@ -377,7 +348,6 @@
pages = iommu_num_pages(s->offset, s->length, PAGE_SIZE);
while (pages--) {
iommu_gatt_base[iommu_page] = GPTE_ENCODE(addr);
- SET_LEAK(iommu_page);
addr += PAGE_SIZE;
iommu_page++;
}
@@ -801,11 +771,12 @@
#ifdef CONFIG_IOMMU_LEAK
if (leak_trace) {
- iommu_leak_tab = (void *)__get_free_pages(GFP_KERNEL|__GFP_ZERO,
- get_order(iommu_pages*sizeof(void *)));
- if (!iommu_leak_tab)
+ int ret;
+
+ ret = dma_debug_resize_entries(iommu_pages);
+ if (ret)
printk(KERN_DEBUG
- "PCI-DMA: Cannot allocate leak trace area\n");
+ "PCI-DMA: Cannot trace all the entries\n");
}
#endif
diff --git a/arch/x86/kernel/pci-swiotlb.c b/arch/x86/kernel/pci-swiotlb.c
index 221a385..a1712f2 100644
--- a/arch/x86/kernel/pci-swiotlb.c
+++ b/arch/x86/kernel/pci-swiotlb.c
@@ -28,7 +28,7 @@
return paddr;
}
-phys_addr_t swiotlb_bus_to_phys(dma_addr_t baddr)
+phys_addr_t swiotlb_bus_to_phys(struct device *hwdev, dma_addr_t baddr)
{
return baddr;
}
diff --git a/include/linux/dma-debug.h b/include/linux/dma-debug.h
index 28d53cb..171ad8a 100644
--- a/include/linux/dma-debug.h
+++ b/include/linux/dma-debug.h
@@ -32,6 +32,8 @@
extern void dma_debug_init(u32 num_entries);
+extern int dma_debug_resize_entries(u32 num_entries);
+
extern void debug_dma_map_page(struct device *dev, struct page *page,
size_t offset, size_t size,
int direction, dma_addr_t dma_addr,
@@ -91,6 +93,11 @@
{
}
+static inline int dma_debug_resize_entries(u32 num_entries)
+{
+ return 0;
+}
+
static inline void debug_dma_map_page(struct device *dev, struct page *page,
size_t offset, size_t size,
int direction, dma_addr_t dma_addr,
diff --git a/include/linux/swiotlb.h b/include/linux/swiotlb.h
index ac9ff54..cb1a663 100644
--- a/include/linux/swiotlb.h
+++ b/include/linux/swiotlb.h
@@ -29,7 +29,8 @@
extern dma_addr_t swiotlb_phys_to_bus(struct device *hwdev,
phys_addr_t address);
-extern phys_addr_t swiotlb_bus_to_phys(dma_addr_t address);
+extern phys_addr_t swiotlb_bus_to_phys(struct device *hwdev,
+ dma_addr_t address);
extern int swiotlb_arch_range_needs_mapping(phys_addr_t paddr, size_t size);
diff --git a/lib/dma-debug.c b/lib/dma-debug.c
index 69da09a..cdd205d 100644
--- a/lib/dma-debug.c
+++ b/lib/dma-debug.c
@@ -85,6 +85,7 @@
static u32 num_free_entries;
static u32 min_free_entries;
+static u32 nr_total_entries;
/* number of preallocated entries requested by kernel cmdline */
static u32 req_entries;
@@ -257,6 +258,21 @@
put_hash_bucket(bucket, &flags);
}
+static struct dma_debug_entry *__dma_entry_alloc(void)
+{
+ struct dma_debug_entry *entry;
+
+ entry = list_entry(free_entries.next, struct dma_debug_entry, list);
+ list_del(&entry->list);
+ memset(entry, 0, sizeof(*entry));
+
+ num_free_entries -= 1;
+ if (num_free_entries < min_free_entries)
+ min_free_entries = num_free_entries;
+
+ return entry;
+}
+
/* struct dma_entry allocator
*
* The next two functions implement the allocator for
@@ -276,9 +292,7 @@
goto out;
}
- entry = list_entry(free_entries.next, struct dma_debug_entry, list);
- list_del(&entry->list);
- memset(entry, 0, sizeof(*entry));
+ entry = __dma_entry_alloc();
#ifdef CONFIG_STACKTRACE
entry->stacktrace.max_entries = DMA_DEBUG_STACKTRACE_ENTRIES;
@@ -286,9 +300,6 @@
entry->stacktrace.skip = 2;
save_stack_trace(&entry->stacktrace);
#endif
- num_free_entries -= 1;
- if (num_free_entries < min_free_entries)
- min_free_entries = num_free_entries;
out:
spin_unlock_irqrestore(&free_entries_lock, flags);
@@ -310,6 +321,53 @@
spin_unlock_irqrestore(&free_entries_lock, flags);
}
+int dma_debug_resize_entries(u32 num_entries)
+{
+ int i, delta, ret = 0;
+ unsigned long flags;
+ struct dma_debug_entry *entry;
+ LIST_HEAD(tmp);
+
+ spin_lock_irqsave(&free_entries_lock, flags);
+
+ if (nr_total_entries < num_entries) {
+ delta = num_entries - nr_total_entries;
+
+ spin_unlock_irqrestore(&free_entries_lock, flags);
+
+ for (i = 0; i < delta; i++) {
+ entry = kzalloc(sizeof(*entry), GFP_KERNEL);
+ if (!entry)
+ break;
+
+ list_add_tail(&entry->list, &tmp);
+ }
+
+ spin_lock_irqsave(&free_entries_lock, flags);
+
+ list_splice(&tmp, &free_entries);
+ nr_total_entries += i;
+ num_free_entries += i;
+ } else {
+ delta = nr_total_entries - num_entries;
+
+ for (i = 0; i < delta && !list_empty(&free_entries); i++) {
+ entry = __dma_entry_alloc();
+ kfree(entry);
+ }
+
+ nr_total_entries -= i;
+ }
+
+ if (nr_total_entries != num_entries)
+ ret = 1;
+
+ spin_unlock_irqrestore(&free_entries_lock, flags);
+
+ return ret;
+}
+EXPORT_SYMBOL(dma_debug_resize_entries);
+
/*
* DMA-API debugging init code
*
@@ -439,6 +497,8 @@
return;
}
+ nr_total_entries = num_free_entries;
+
printk(KERN_INFO "DMA-API: debugging enabled by kernel config\n");
}
diff --git a/lib/swiotlb.c b/lib/swiotlb.c
index 2b0b5a7..bffe6d7 100644
--- a/lib/swiotlb.c
+++ b/lib/swiotlb.c
@@ -60,8 +60,8 @@
int swiotlb_force;
/*
- * Used to do a quick range check in swiotlb_unmap_single and
- * swiotlb_sync_single_*, to see if the memory was in fact allocated by this
+ * Used to do a quick range check in unmap_single and
+ * sync_single_*, to see if the memory was in fact allocated by this
* API.
*/
static char *io_tlb_start, *io_tlb_end;
@@ -129,7 +129,7 @@
return paddr;
}
-phys_addr_t __weak swiotlb_bus_to_phys(dma_addr_t baddr)
+phys_addr_t __weak swiotlb_bus_to_phys(struct device *hwdev, dma_addr_t baddr)
{
return baddr;
}
@@ -140,9 +140,15 @@
return swiotlb_phys_to_bus(hwdev, virt_to_phys(address));
}
-static void *swiotlb_bus_to_virt(dma_addr_t address)
+void * __weak swiotlb_bus_to_virt(struct device *hwdev, dma_addr_t address)
{
- return phys_to_virt(swiotlb_bus_to_phys(address));
+ return phys_to_virt(swiotlb_bus_to_phys(hwdev, address));
+}
+
+int __weak swiotlb_arch_address_needs_mapping(struct device *hwdev,
+ dma_addr_t addr, size_t size)
+{
+ return !is_buffer_dma_capable(dma_get_mask(hwdev), addr, size);
}
int __weak swiotlb_arch_range_needs_mapping(phys_addr_t paddr, size_t size)
@@ -309,10 +315,10 @@
return -ENOMEM;
}
-static int
+static inline int
address_needs_mapping(struct device *hwdev, dma_addr_t addr, size_t size)
{
- return !is_buffer_dma_capable(dma_get_mask(hwdev), addr, size);
+ return swiotlb_arch_address_needs_mapping(hwdev, addr, size);
}
static inline int range_needs_mapping(phys_addr_t paddr, size_t size)
@@ -341,7 +347,7 @@
unsigned long flags;
while (size) {
- sz = min(PAGE_SIZE - offset, size);
+ sz = min_t(size_t, PAGE_SIZE - offset, size);
local_irq_save(flags);
buffer = kmap_atomic(pfn_to_page(pfn),
@@ -476,7 +482,7 @@
* dma_addr is the kernel virtual address of the bounce buffer to unmap.
*/
static void
-unmap_single(struct device *hwdev, char *dma_addr, size_t size, int dir)
+do_unmap_single(struct device *hwdev, char *dma_addr, size_t size, int dir)
{
unsigned long flags;
int i, count, nslots = ALIGN(size, 1 << IO_TLB_SHIFT) >> IO_TLB_SHIFT;
@@ -560,7 +566,6 @@
size)) {
/*
* The allocated memory isn't reachable by the device.
- * Fall back on swiotlb_map_single().
*/
free_pages((unsigned long) ret, order);
ret = NULL;
@@ -568,9 +573,8 @@
if (!ret) {
/*
* We are either out of memory or the device can't DMA
- * to GFP_DMA memory; fall back on
- * swiotlb_map_single(), which will grab memory from
- * the lowest available address range.
+ * to GFP_DMA memory; fall back on map_single(), which
+ * will grab memory from the lowest available address range.
*/
ret = map_single(hwdev, 0, size, DMA_FROM_DEVICE);
if (!ret)
@@ -587,7 +591,7 @@
(unsigned long long)dev_addr);
/* DMA_TO_DEVICE to avoid memcpy in unmap_single */
- unmap_single(hwdev, ret, size, DMA_TO_DEVICE);
+ do_unmap_single(hwdev, ret, size, DMA_TO_DEVICE);
return NULL;
}
*dma_handle = dev_addr;
@@ -604,7 +608,7 @@
free_pages((unsigned long) vaddr, get_order(size));
else
/* DMA_TO_DEVICE to avoid memcpy in unmap_single */
- unmap_single(hwdev, vaddr, size, DMA_TO_DEVICE);
+ do_unmap_single(hwdev, vaddr, size, DMA_TO_DEVICE);
}
EXPORT_SYMBOL(swiotlb_free_coherent);
@@ -634,7 +638,7 @@
* physical address to use is returned.
*
* Once the device is given the dma address, the device owns this memory until
- * either swiotlb_unmap_single or swiotlb_dma_sync_single is performed.
+ * either swiotlb_unmap_page or swiotlb_dma_sync_single is performed.
*/
dma_addr_t swiotlb_map_page(struct device *dev, struct page *page,
unsigned long offset, size_t size,
@@ -642,18 +646,17 @@
struct dma_attrs *attrs)
{
phys_addr_t phys = page_to_phys(page) + offset;
- void *ptr = page_address(page) + offset;
dma_addr_t dev_addr = swiotlb_phys_to_bus(dev, phys);
void *map;
BUG_ON(dir == DMA_NONE);
/*
- * If the pointer passed in happens to be in the device's DMA window,
+ * If the address happens to be in the device's DMA window,
* we can safely return the device addr and not worry about bounce
* buffering it.
*/
if (!address_needs_mapping(dev, dev_addr, size) &&
- !range_needs_mapping(virt_to_phys(ptr), size))
+ !range_needs_mapping(phys, size))
return dev_addr;
/*
@@ -679,23 +682,35 @@
/*
* Unmap a single streaming mode DMA translation. The dma_addr and size must
- * match what was provided for in a previous swiotlb_map_single call. All
+ * match what was provided for in a previous swiotlb_map_page call. All
* other usages are undefined.
*
* After this call, reads by the cpu to the buffer are guaranteed to see
* whatever the device wrote there.
*/
+static void unmap_single(struct device *hwdev, dma_addr_t dev_addr,
+ size_t size, int dir)
+{
+ char *dma_addr = swiotlb_bus_to_virt(hwdev, dev_addr);
+
+ BUG_ON(dir == DMA_NONE);
+
+ if (is_swiotlb_buffer(dma_addr)) {
+ do_unmap_single(hwdev, dma_addr, size, dir);
+ return;
+ }
+
+ if (dir != DMA_FROM_DEVICE)
+ return;
+
+ dma_mark_clean(dma_addr, size);
+}
+
void swiotlb_unmap_page(struct device *hwdev, dma_addr_t dev_addr,
size_t size, enum dma_data_direction dir,
struct dma_attrs *attrs)
{
- char *dma_addr = swiotlb_bus_to_virt(dev_addr);
-
- BUG_ON(dir == DMA_NONE);
- if (is_swiotlb_buffer(dma_addr))
- unmap_single(hwdev, dma_addr, size, dir);
- else if (dir == DMA_FROM_DEVICE)
- dma_mark_clean(dma_addr, size);
+ unmap_single(hwdev, dev_addr, size, dir);
}
EXPORT_SYMBOL_GPL(swiotlb_unmap_page);
@@ -703,7 +718,7 @@
* Make physical memory consistent for a single streaming mode DMA translation
* after a transfer.
*
- * If you perform a swiotlb_map_single() but wish to interrogate the buffer
+ * If you perform a swiotlb_map_page() but wish to interrogate the buffer
* using the cpu, yet do not wish to teardown the dma mapping, you must
* call this function before doing so. At the next point you give the dma
* address back to the card, you must first perform a
@@ -713,13 +728,19 @@
swiotlb_sync_single(struct device *hwdev, dma_addr_t dev_addr,
size_t size, int dir, int target)
{
- char *dma_addr = swiotlb_bus_to_virt(dev_addr);
+ char *dma_addr = swiotlb_bus_to_virt(hwdev, dev_addr);
BUG_ON(dir == DMA_NONE);
- if (is_swiotlb_buffer(dma_addr))
+
+ if (is_swiotlb_buffer(dma_addr)) {
sync_single(hwdev, dma_addr, size, dir, target);
- else if (dir == DMA_FROM_DEVICE)
- dma_mark_clean(dma_addr, size);
+ return;
+ }
+
+ if (dir != DMA_FROM_DEVICE)
+ return;
+
+ dma_mark_clean(dma_addr, size);
}
void
@@ -746,13 +767,7 @@
unsigned long offset, size_t size,
int dir, int target)
{
- char *dma_addr = swiotlb_bus_to_virt(dev_addr) + offset;
-
- BUG_ON(dir == DMA_NONE);
- if (is_swiotlb_buffer(dma_addr))
- sync_single(hwdev, dma_addr, size, dir, target);
- else if (dir == DMA_FROM_DEVICE)
- dma_mark_clean(dma_addr, size);
+ swiotlb_sync_single(hwdev, dev_addr + offset, size, dir, target);
}
void
@@ -777,7 +792,7 @@
/*
* Map a set of buffers described by scatterlist in streaming mode for DMA.
- * This is the scatter-gather version of the above swiotlb_map_single
+ * This is the scatter-gather version of the above swiotlb_map_page
* interface. Here the scatter gather list elements are each tagged with the
* appropriate dma address and length. They are obtained via
* sg_dma_{address,length}(SG).
@@ -788,7 +803,7 @@
* The routine returns the number of addr/length pairs actually
* used, at most nents.
*
- * Device ownership issues as mentioned above for swiotlb_map_single are the
+ * Device ownership issues as mentioned above for swiotlb_map_page are the
* same here.
*/
int
@@ -836,7 +851,7 @@
/*
* Unmap a set of streaming mode DMA translations. Again, cpu read rules
- * concerning calls here are the same as for swiotlb_unmap_single() above.
+ * concerning calls here are the same as for swiotlb_unmap_page() above.
*/
void
swiotlb_unmap_sg_attrs(struct device *hwdev, struct scatterlist *sgl,
@@ -847,13 +862,9 @@
BUG_ON(dir == DMA_NONE);
- for_each_sg(sgl, sg, nelems, i) {
- if (sg->dma_address != swiotlb_phys_to_bus(hwdev, sg_phys(sg)))
- unmap_single(hwdev, swiotlb_bus_to_virt(sg->dma_address),
- sg->dma_length, dir);
- else if (dir == DMA_FROM_DEVICE)
- dma_mark_clean(swiotlb_bus_to_virt(sg->dma_address), sg->dma_length);
- }
+ for_each_sg(sgl, sg, nelems, i)
+ unmap_single(hwdev, sg->dma_address, sg->dma_length, dir);
+
}
EXPORT_SYMBOL(swiotlb_unmap_sg_attrs);
@@ -879,15 +890,9 @@
struct scatterlist *sg;
int i;
- BUG_ON(dir == DMA_NONE);
-
- for_each_sg(sgl, sg, nelems, i) {
- if (sg->dma_address != swiotlb_phys_to_bus(hwdev, sg_phys(sg)))
- sync_single(hwdev, swiotlb_bus_to_virt(sg->dma_address),
+ for_each_sg(sgl, sg, nelems, i)
+ swiotlb_sync_single(hwdev, sg->dma_address,
sg->dma_length, dir, target);
- else if (dir == DMA_FROM_DEVICE)
- dma_mark_clean(swiotlb_bus_to_virt(sg->dma_address), sg->dma_length);
- }
}
void