Messages in this thread | | | Date | Fri, 6 Apr 2012 20:25:54 -0400 | From | Chris Metcalf <> | Subject | Re: [PATCH v2] hugetlb: fix race condition in hugetlb_fault() |
| |
On 4/6/2012 7:35 PM, Hugh Dickins wrote: > On Fri, 6 Apr 2012, Andrew Morton wrote: >> On Fri, 6 Apr 2012 16:10:13 -0700 (PDT) >> Hugh Dickins <hughd@google.com> wrote: >>> The resulting patch is okay; but let's reassure Chris that his >>> original patch was better, before he conceded to make the get_page >>> and put_page unconditional, and added unnecessary detail of the race. >>> >> Yes, the v1 patch was better. No reason was given for changing it? > I think Chris was aiming to be a model citizen, and followed review > suggestions that he would actually have done better to resist.
Yes, exactly. I figure if I'm submitting patches to mm, I should defer to suggestions from someone like Hillf who has committed a lot more of them than I have. :-) Arguably the unconditional version is simpler at the source code level in any case, and I figure more is usually better when it comes to documenting race conditions, so it didn't seem necessary to push back. Frankly I'm happy to keep my sign-off on either version of the patch and defer to Andrew or whomever as to which one gets taken.
-- Chris Metcalf, Tilera Corp. http://www.tilera.com
| |