On Tue, Jan 12, 2021 at 10:06:29PM +0100, Bastien Traverse wrote:
> Hello everyone,
>
> A couple of weeks back I got an xfs_corruption_error stack trace on my
> rootfs on Arch Linux, a few minutes after creating a swap file an enabling
> it. Here is the process I followed to do so:
>
> fallocate -l 4G /swapfile
> chmod 600 /swapfile
> mkswap /swapfile
> swapon /swapfile
> echo "/swapfile none swap defaults 0 0" >> /etc/fstab
>
> And the trace appeared a few minutes later, without me doing much at that
> moment:
Sounds like:
commit 41663430588c737dd735bad5a0d1ba325dcabd59
Author: Gao Xiang <hsiangkao@xxxxxxxxxx>
Date: Fri Sep 25 21:19:01 2020 -0700
mm, THP, swap: fix allocating cluster for swapfile by mistake
SWP_FS is used to make swap_{read,write}page() go through the
filesystem, and it's only used for swap files over NFS. So, !SWP_FS
means non NFS for now, it could be either file backed or device backed.
Something similar goes with legacy SWP_FILE.
So in order to achieve the goal of the original patch, SWP_BLKDEV should
be used instead.
FS corruption can be observed with SSD device + XFS + fragmented
swapfile due to CONFIG_THP_SWAP=y.
.....
But I thought that was fixed in 5.9-rc7 so should be in your kernel.
Can you confirm that your kernel has this fix?
Cheers,
Dave.
--
Dave Chinner
david@xxxxxxxxxxxxx
