| To: | Jeff Moyer <jmoyer@xxxxxxxxxx> |
|---|---|
| Subject: | Re: AIO read returns negative number for bytes read |
| From: | Avi Kivity <avi@xxxxxxxxxxxx> |
| Date: | Tue, 17 Nov 2015 14:52:19 +0200 |
| Cc: | xfs@xxxxxxxxxxx, linux-aio@xxxxxxxxx, Steven Whitehouse <swhiteho@xxxxxxxxxx>, Jan Kara <jack@xxxxxxx>, linux-fsdevel@xxxxxxxxxxxxxxx |
| Delivered-to: | xfs@xxxxxxxxxxx |
| Dkim-signature: | v=1; a=rsa-sha256; c=relaxed/relaxed; d=scylladb-com.20150623.gappssmtp.com; s=20150623; h=subject:to:references:cc:from:message-id:date:user-agent :mime-version:in-reply-to:content-type:content-transfer-encoding; bh=0vGFpZVSDKXZQ1d+UKUtZyKsRkM78LSxkrJyYPG7n0E=; b=i67y0J95OTvrXuy4ToWzxDaoTeLzY8Y/Selg6XUJUlQnToj5NchAJ/6KHwWDm5xv9U IT9I+Cs1X1BU9iWAZogL90ggykgOUsvkruPdN9d87xc1nKUh0f/Lhv6x1SImoMcG5GBF h4ZPHpIrQ+Kt4m+ux6Ttrg/ZiE68Xe0+9JEgTg63/Gc7+SW6UZaYI3YDMZjOD52UKZVx ZNmo8zIolyT0zSL4gZAcCcJ+1Nu+AzAjDjbv4KECOZUhU9OekfgXh1zyUcH+lM1GVwyA BkLoGvNArUaMCUaLydQBx0WVsdma/R1hGKgRxIM0/yqJ+Y52xzhJ9DIKWH2YKenrF7D+ qaHA== |
| In-reply-to: | <x497flhsq1c.fsf@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx> |
| References: | <564883BD.8070607@xxxxxxxxxxxx> <x497flhsq1c.fsf@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx> |
| User-agent: | Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.3.0 |
On 11/16/2015 09:27 PM, Jeff Moyer wrote: Hi Avi, Avi Kivity <avi@xxxxxxxxxxxx> writes:Due to a bug in my program, I initiated a read beyond eof. Specifically, the file size is 13002 bytes and the read offset is 13312 (0x3400). I would expect such a read to return 0 bytes read, but io_getevents returns -310, which is suspiciously equal to (13002 - 13312). I attach a reproducer. 4.2.5-201.fc22.x86_64 Are my expectations incorrect, or is this a bug in aio or xfs?Your expectations are correct. The bug was introduced by commit 9fe55eea7e4b4 (Fix race when checking i_size on direct i/o read). I've CC'd the patch author and linux-fsdevel. I'm not sure what the right fix is, given that the size checks were removed from the vfs to fix some race condition. Unfortunately, the commit message doesn't really do a good job of explaining the race. In order to save others time, here is a good explanation of the problem that commit is meant to fix, along with a reproducer: http://marc.info/?l=linux-fsdevel&m=138641356614458&w=2 Thanks for the great bug report, and sorry I have no solution to proffer. Thanks. I will await a fix with interest. |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: Disk error, then endless loop, Brian Foster |
|---|---|
| Next by Date: | Re: clone ioctl return values, Chris Mason |
| Previous by Thread: | Re: AIO read returns negative number for bytes read, Jeff Moyer |
| Next by Thread: | Re: AIO read returns negative number for bytes read, Jan Kara |
| Indexes: | [Date] [Thread] [Top] [All Lists] |