>On Fri, Sep 20, 2019 at 11:58 PM Aneesh Kumar K.V <aneesh.kumar@xxxxxxxxxxxxx> wrote:
>On 9/20/19 9:21 PM, Qiujun Huang wrote:
>> __get_user_pages_fast try to walk the page table but the
> >hugepage pte is replace by hwpoison swap entry by mca path.
> >...
>
>
>Can you describe this in more details. I guess you are facing the issue
>with respect PUD level PTE entry that got updated by hwpoison as a swap
>entry. Since we don't specifically check for pud_present(), we walk the
>page table with wrong values and that results in corruption?
Yes, in the case using 2G hugepage.
>> [15798.177437] mce: Uncorrected hardware memory error in
>> user-access at 224f1761c0
> >[15798.180171] MCE 0x224f176: Killing pal_main:6784 due to
> > hardware memory corruption
> >[15798.180176] MCE 0x224f176: Killing qemu-system-x86:167336
> > due to hardware memory corruption
>> ...
>> [15798.180206] BUG: unable to handle kernel
>> [15798.180226] paging request at ffff891200003000
>> [15798.180236] IP: [<ffffffff8106edae>] gup_pud_range+
>> 0x13e/0x1e0
>> ...
>>
>> We need to skip the hwpoison entry in gup_pud_range.
>>
>> Signed-off-by: Qiujun Huang <hqjagain@xxxxxxxxx>
>> ---
>> mm/gup.c | 2 ++
>> 1 file changed, 2 insertions(+)
>>
>> diff --git a/mm/gup.c b/mm/gup.c
>> index 98f13ab..6157ed9 100644
>> --- a/mm/gup.c
>> +++ b/mm/gup.c
>> @@ -2230,6 +2230,8 @@ static int gup_pud_range(p4d_t p4d, unsigned long addr, unsigned long end,
>> next = pud_addr_end(addr, end);
>> if (pud_none(pud))
>> return 0;
>> + if (unlikely(!pud_present(pud)))
>> + return 0;
>You should be able to remove that if (pud_none(pud)) check and just keep
>the pud_present() check?
indeed
>> if (unlikely(pud_huge(pud))) {
>> if (!gup_huge_pud(pud, pudp, addr, next, flags,
>> pages, nr))
>>
diff --git a/mm/gup.c b/mm/gup.c
index 98f13ab..2e3a1d3 100644
--- a/mm/gup.c
+++ b/mm/gup.c
@@ -2228,7 +2228,7 @@ static int gup_pud_range(p4d_t p4d, unsigned long addr, unsigned long end,
pud_t pud = READ_ONCE(*pudp);
next = pud_addr_end(addr, end);
- if (pud_none(pud))
+ if (!pud_present(pud))
return 0;
if (unlikely(pud_huge(pud))) {
if (!gup_huge_pud(pud, pudp, addr, next, flags,
