Re: [PATCH v2 2/4] mm/hwpoison: fix miss catch transparent huge page

[Chen Gong]

On Tue, Sep 03, 2013 at 12:18:58PM +0800, Wanpeng Li wrote:
> Date: Tue, 3 Sep 2013 12:18:58 +0800
> From: Wanpeng Li <liwanp@xxxxxxxxxxxxxxxxxx>
> To: Chen Gong <gong.chen@xxxxxxxxxxxxxxx>
> Cc: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>, Andi Kleen
>  <andi@xxxxxxxxxxxxxx>, Fengguang Wu <fengguang.wu@xxxxxxxxx>, Naoya
>  Horiguchi <n-horiguchi@xxxxxxxxxxxxx>, Tony Luck <tony.luck@xxxxxxxxx>,
>  linux-mm@xxxxxxxxx, linux-kernel@xxxxxxxxxxxxxxx
> Subject: Re: [PATCH v2 2/4] mm/hwpoison: fix miss catch transparent huge
>  page
> User-Agent: Mutt/1.5.21 (2010-09-15)
> 
> On Mon, Sep 02, 2013 at 11:15:19PM -0400, Chen Gong wrote:
> >On Tue, Sep 03, 2013 at 07:36:44AM +0800, Wanpeng Li wrote:
> >> Date: Tue,  3 Sep 2013 07:36:44 +0800
> >> From: Wanpeng Li <liwanp@xxxxxxxxxxxxxxxxxx>
> >> To: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
> >> Cc: Andi Kleen <andi@xxxxxxxxxxxxxx>, Fengguang Wu
> >>  <fengguang.wu@xxxxxxxxx>, Naoya Horiguchi <n-horiguchi@xxxxxxxxxxxxx>,
> >>  Tony Luck <tony.luck@xxxxxxxxx>, gong.chen@xxxxxxxxxxxxxxx,
> >>  linux-mm@xxxxxxxxx, linux-kernel@xxxxxxxxxxxxxxx, Wanpeng Li
> >>  <liwanp@xxxxxxxxxxxxxxxxxx>
> >> Subject: [PATCH v2 2/4] mm/hwpoison: fix miss catch transparent huge page 
> >> X-Mailer: git-send-email 1.7.5.4
> >> 
> >> Changelog:
> >>  *v1 -> v2: reverse PageTransHuge(page) && !PageHuge(page) check 
> >> 
> >> PageTransHuge() can't guarantee the page is transparent huge page since it 
> >> return true for both transparent huge and hugetlbfs pages. This patch fix 
> >> it by check the page is also !hugetlbfs page.
> >> 
> >> Before patch:
> >> 
> >> [  121.571128] Injecting memory failure at pfn 23a200
> >> [  121.571141] MCE 0x23a200: huge page recovery: Delayed
> >> [  140.355100] MCE: Memory failure is now running on 0x23a200
> >> 
> >> After patch:
> >> 
> >> [   94.290793] Injecting memory failure at pfn 23a000
> >> [   94.290800] MCE 0x23a000: huge page recovery: Delayed
> >> [  105.722303] MCE: Software-unpoisoned page 0x23a000
> >> 
> >> Signed-off-by: Wanpeng Li <liwanp@xxxxxxxxxxxxxxxxxx>
> >> ---
> >>  mm/memory-failure.c | 2 +-
> >>  1 file changed, 1 insertion(+), 1 deletion(-)
> >> 
> >> diff --git a/mm/memory-failure.c b/mm/memory-failure.c
> >> index e28ee77..b114570 100644
> >> --- a/mm/memory-failure.c
> >> +++ b/mm/memory-failure.c
> >> @@ -1349,7 +1349,7 @@ int unpoison_memory(unsigned long pfn)
> >>  	 * worked by memory_failure() and the page lock is not held yet.
> >>  	 * In such case, we yield to memory_failure() and make unpoison fail.
> >>  	 */
> >> -	if (PageTransHuge(page)) {
> >> +	if (!PageHuge(page) && PageTransHuge(page)) {
> >>  		pr_info("MCE: Memory failure is now running on %#lx\n", pfn);
> >>  			return 0;
> >>  	}
> >
> >Not sure which git tree should be used to apply this patch series? I assume
> >this patch series follows this link: https://lkml.org/lkml/2013/8/26/76.
> >
> 
> mmotm tree or linux-next. ;-)
> 
> >In unpoison_memory we already have
> >        if (PageHuge(page)) {
> >                ...
> >                return 0;
> >        }
> >so it looks like this patch is redundant.
> 
> - Do you aware there is condition before go to this check?
> - Do you also analysis why the check can't catch the hugetlbfs page
>   through the dump information?
> 

Looks like we use different trees. After checking your working tree,
your patch is right. So just ignore my words above. FWIW, please be
polite and give a positive response.

> Regards,
> Wanpeng Li 
> 
> --
> To unsubscribe, send a message with 'unsubscribe linux-mm' in
> the body to majordomo@xxxxxxxxxx  For more info on Linux MM,
> see: http://www.linux-mm.org/ .
> Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>

Attachment: signature.asc
Description: Digital signature