powerpc: Use 64k io pages when we never see an HEA
When we never get around to seeing an HEA ethernet adapter, there's
no point in restricting ourselves to 4k IO page size.
This speeds up IO maps when CONFIG_IBMEBUS is disabled.
[ Updated the test to also lift the restriction on arch 2.07
(Power 8) which cannot have an HEA
-- BenH ]
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
foo
diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c
index d766d6e..12f1a39 100644
--- a/arch/powerpc/mm/hash_utils_64.c
+++ b/arch/powerpc/mm/hash_utils_64.c
@@ -445,6 +445,24 @@
mmu_psize_defs[bpsize].penc[apsize] = -1;
}
+#ifdef CONFIG_PPC_64K_PAGES
+
+static bool might_have_hea(void)
+{
+ /*
+ * The HEA ethernet adapter requires awareness of the
+ * GX bus. Without that awareness we can easily assume
+ * we will never see an HEA ethernet device.
+ */
+#ifdef CONFIG_IBMEBUS
+ return !cpu_has_feature(CPU_FTR_ARCH_207S);
+#else
+ return false;
+#endif
+}
+
+#endif /* #ifdef CONFIG_PPC_64K_PAGES */
+
static void __init htab_init_page_sizes(void)
{
int rc;
@@ -499,10 +517,11 @@
mmu_linear_psize = MMU_PAGE_64K;
if (mmu_has_feature(MMU_FTR_CI_LARGE_PAGE)) {
/*
- * Don't use 64k pages for ioremap on pSeries, since
- * that would stop us accessing the HEA ethernet.
+ * When running on pSeries using 64k pages for ioremap
+ * would stop us accessing the HEA ethernet. So if we
+ * have the chance of ever seeing one, stay at 4k.
*/
- if (!machine_is(pseries))
+ if (!might_have_hea() || !machine_is(pseries))
mmu_io_psize = MMU_PAGE_64K;
} else
mmu_ci_restrictions = 1;