| To: | "Darrick J. Wong" <darrick.wong@xxxxxxxxxx> |
|---|---|
| Subject: | Re: aio completions vs file_accessed race, was: Re: [PATCH 7/8] xfs: split direct I/O and DAX path |
| From: | Christoph Hellwig <hch@xxxxxx> |
| Date: | Thu, 29 Sep 2016 22:18:34 +0200 |
| Cc: | Christoph Hellwig <hch@xxxxxx>, xfs@xxxxxxxxxxx, linux-fsdevel@xxxxxxxxxxxxxxx, linux-aio@xxxxxxxxx |
| Delivered-to: | xfs@xxxxxxxxxxx |
| In-reply-to: | <20160929083824.GA12799@xxxxxx> |
| References: | <1466609236-23801-1-git-send-email-hch@xxxxxx> <1466609236-23801-8-git-send-email-hch@xxxxxx> <20160929025351.GB4901@xxxxxxxxxxxxxxxx> <20160929083824.GA12799@xxxxxx> |
| User-agent: | Mutt/1.5.17 (2007-11-01) |
Can you try the patch below? That just moves the file_accessed call before the I/O, similar to how we handle timestamp updates on the write side. generic_file_read_iter will also need a similar update. |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: [PATCH v2 3/3] block: implement (some of) fallocate for block devices, Bart Van Assche |
|---|---|
| Next by Date: | Re: aio completions vs file_accessed race, was: Re: [PATCH 7/8] xfs: split direct I/O and DAX path, Christoph Hellwig |
| Previous by Thread: | aio completions vs file_accessed race, was: Re: [PATCH 7/8] xfs: split direct I/O and DAX path, Christoph Hellwig |
| Next by Thread: | Re: aio completions vs file_accessed race, was: Re: [PATCH 7/8] xfs: split direct I/O and DAX path, Christoph Hellwig |
| Indexes: | [Date] [Thread] [Top] [All Lists] |