x86: AMD Athlon X2 hard hang fix

An Athlon 64 X2 test system showed hard hangs shortly after marking
the kernel text read-only, if we tried to preserve largepages and
changed the PSE entry from RW to RO. The pagetable code itself is
correct, it's the CPU that locked up hard (and not even the NMI
watchdog could punch through that hard hang).

So be conservative and always do splitups - like we did in the past.

Signed-off-by: Ingo Molnar <mingo@elte.hu>

diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index 40b7ac5..3810f7a 100644 (file)
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -243,6 +243,17 @@ static int try_preserve_large_page(pte_t *kpte, unsigned long address,
        pgprot_t old_prot, new_prot;
        int level, res = CPA_SPLIT;
 
+       /*
+        * An Athlon 64 X2 showed hard hangs if we tried to preserve
+        * largepages and changed the PSE entry from RW to RO.
+        *
+        * As AMD CPUs have a long series of erratas in this area,
+        * (and none of the known ones seem to explain this hang),
+        * disable this code until the hang can be debugged:
+        */
+       if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD)
+               return res;
+
        spin_lock_irqsave(&pgd_lock, flags);
        /*
         * Check for races, another CPU might have split this page