[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20151105085033.GB7614@node.shutemov.name>
Date: Thu, 5 Nov 2015 10:50:33 +0200
From: "Kirill A. Shutemov" <kirill@...temov.name>
To: Vlastimil Babka <vbabka@...e.cz>
Cc: yalin wang <yalin.wang2010@...il.com>, akpm@...ux-foundation.org,
kirill.shutemov@...ux.intel.com, jmarchan@...hat.com,
mgorman@...hsingularity.net, ebru.akagunduz@...il.com,
willy@...ux.intel.com, linux-mm@...ck.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH V2] mm: fix kernel crash in khugepaged thread
On Thu, Nov 05, 2015 at 09:12:34AM +0100, Vlastimil Babka wrote:
> On 10/29/2015 01:35 AM, Kirill A. Shutemov wrote:
> >> @@ -2605,9 +2603,9 @@ out_unmap:
> >> /* collapse_huge_page will return with the mmap_sem released */
> >> collapse_huge_page(mm, address, hpage, vma, node);
> >> }
> >> -out:
> >> - trace_mm_khugepaged_scan_pmd(mm, page_to_pfn(page), writable, referenced,
> >> - none_or_zero, result, unmapped);
> >> + trace_mm_khugepaged_scan_pmd(mm, pte_present(pteval) ?
> >> + pte_pfn(pteval) : -1, writable, referenced,
> >> + none_or_zero, result, unmapped);
> >
> > maybe passing down pte instead of pfn?
>
> Maybe just pass the page, and have tracepoint's fast assign check for !NULL and
> do page_to_pfn itself? That way the complexity and overhead is only in the
> tracepoint and when enabled.
Agreed.
--
Kirill A. Shutemov
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists