x86: work around Fedora-11 x86-32 kernel failures on Intel Atom CPUs

commit 211b3d03c7400f48a781977a50104c9d12f4e229 upstream

[Trivial backport to 2.6.27 by cebbert@redhat.com]

x86: work around Fedora-11 x86-32 kernel failures on Intel Atom CPUs

Impact: work around boot crash

Work around Intel Atom erratum AAH41 (probabilistically) - it's triggering
in the field.

Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Tested-by: Kyle McMartin <kyle@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cc: <cebbert@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index 7c3b8dc52dc81aab303ab166711c46fef08b7c6d..5468c195a1c93a7eb903f58b72d137c3c95e07d7 100644 (file)
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -565,6 +565,17 @@ static int split_large_page(pte_t *kpte, unsigned long address)
        ref_prot = pte_pgprot(pte_mkexec(pte_clrhuge(*kpte)));
        pgprot_val(ref_prot) |= _PAGE_PRESENT;
        __set_pmd_pte(kpte, address, mk_pte(base, ref_prot));
+
+       /*
+        * Intel Atom errata AAH41 workaround.
+        *
+        * The real fix should be in hw or in a microcode update, but
+        * we also probabilistically try to reduce the window of having
+        * a large TLB mixed with 4K TLBs while instruction fetches are
+        * going on.
+        */
+       __flush_tlb_all();
+
        base = NULL;
 
 out_unlock: