[Bug 215261] New: [xfstests xfs/177] test still fails after merge f38a032b165d ("xfs: fix I_DONTCACHE")

[Bug 215261] New: [xfstests xfs/177] test still fails after merge f38a032b165d ("xfs: fix I_DONTCACHE")

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=215261

            Bug ID: 215261
           Summary: [xfstests xfs/177] test still fails after merge
                    f38a032b165d ("xfs: fix I_DONTCACHE")
           Product: File System
           Version: 2.5
    Kernel Version: v5.16-rc2 + xfs-5.16-fixes-2
          Hardware: All
                OS: Linux
              Tree: Mainline
            Status: NEW
          Severity: normal
          Priority: P1
         Component: XFS
          Assignee: filesystem_xfs@kernel-bugs.kernel.org
          Reporter: zlang@redhat.com
        Regression: No

xfstests xfs/177 still fails on xfs-5.16-fixes-2 sometimes, e.g [1][2]. The
kernel has merged f38a032b165d ("xfs: fix I_DONTCACHE").

[1]
FSTYP         -- xfs (debug)
PLATFORM      -- Linux/x86_64 ibm-x3650m4-xxx-vm-xx 5.16.0-rc2+ #1 SMP PREEMPT
Sun Nov 28 06:43:30 EST 2021
MKFS_OPTIONS  -- -f -m crc=1,finobt=1,rmapbt=0,reflink=1,bigtime=1,inobtcount=1
/dev/vda3
MOUNT_OPTIONS -- -o context=system_u:object_r:root_t:s0 /dev/vda3
/mnt/xfstests/scratch

xfs/177 - output mismatch (see /var/lib/xfstests/results//xfs/177.out.bad)
    --- tests/xfs/177.out       2021-11-28 09:56:52.236830568 -0500
    +++ /var/lib/xfstests/results//xfs/177.out.bad      2021-11-30
04:21:03.282699626 -0500
    @@ -2,11 +2,14 @@
     new file count is in range
     inodes after creating files is in range
     Round 1
    -inodes after bulkstat is in range
    +inodes after bulkstat has value of 102
    +inodes after bulkstat is NOT in range 1023 .. 1033
     inodes after expire is in range
    ...
    (Run 'diff -u /var/lib/xfstests/tests/xfs/177.out
/var/lib/xfstests/results//xfs/177.out.bad'  to see the entire diff)
Ran: xfs/177
Failures: xfs/177
Failed 1 of 1 tests

[2]
FSTYP         -- xfs (debug)
PLATFORM      -- Linux/aarch64 hpe-apollo-cnxxxx-xx-vm-xx 5.16.0-rc2+ #1 SMP
Sun Nov 28 06:10:24 EST 2021
MKFS_OPTIONS  -- -f -m crc=1,finobt=1,rmapbt=0,reflink=1,bigtime=1,inobtcount=1
/dev/vda3
MOUNT_OPTIONS -- -o context=system_u:object_r:root_t:s0 /dev/vda3
/mnt/xfstests/scratch

xfs/177 - output mismatch (see /var/lib/xfstests/results//xfs/177.out.bad)
    --- tests/xfs/177.out       2021-11-28 10:04:49.415509327 -0500
    +++ /var/lib/xfstests/results//xfs/177.out.bad      2021-11-29
04:47:32.148868974 -0500
    @@ -2,7 +2,8 @@
     new file count is in range
     inodes after creating files is in range
     Round 1
    -inodes after bulkstat is in range
    +inodes after bulkstat has value of 575
    +inodes after bulkstat is NOT in range 1023 .. 1033
     inodes after expire is in range
    ...
    (Run 'diff -u /var/lib/xfstests/tests/xfs/177.out
/var/lib/xfstests/results//xfs/177.out.bad'  to see the entire diff)
Ran: xfs/177
Failures: xfs/177
Failed 1 of 1 tests

-- 
You may reply to this email to add a comment.

You are receiving this mail because:
You are watching the assignee of the bug.

Re: [Bug 215261] New: [xfstests xfs/177] test still fails after merge f38a032b165d ("xfs: fix I_DONTCACHE")

[Carlos Maiolino]

> xfs/177 - output mismatch (see /var/lib/xfstests/results//xfs/177.out.bad)
>     --- tests/xfs/177.out       2021-11-28 09:56:52.236830568 -0500
>     +++ /var/lib/xfstests/results//xfs/177.out.bad      2021-11-30
> 04:21:03.282699626 -0500
>     @@ -2,11 +2,14 @@
>      new file count is in range
>      inodes after creating files is in range
>      Round 1
>     -inodes after bulkstat is in range
>     +inodes after bulkstat has value of 102
>     +inodes after bulkstat is NOT in range 1023 .. 1033
>      inodes after expire is in range
>     ...
>     (Run 'diff -u /var/lib/xfstests/tests/xfs/177.out
> /var/lib/xfstests/results//xfs/177.out.bad'  to see the entire diff)
> Ran: xfs/177
> Failures: xfs/177
> Failed 1 of 1 tests
> 

I'm still looking into it, but I suspect this is an issue with the test itself.
In some occasions, the timer set for xfssyncd_centisecs is too short, and by the
time the test collect the amount of cached objects, the system is already
reclaiming unused objects, so, the test 'finds' fewer cached objects than it
expects.

> [2]
> FSTYP         -- xfs (debug)
> PLATFORM      -- Linux/aarch64 hpe-apollo-cnxxxx-xx-vm-xx 5.16.0-rc2+ #1 SMP
> Sun Nov 28 06:10:24 EST 2021
> MKFS_OPTIONS  -- -f -m crc=1,finobt=1,rmapbt=0,reflink=1,bigtime=1,inobtcount=1
> /dev/vda3
> MOUNT_OPTIONS -- -o context=system_u:object_r:root_t:s0 /dev/vda3
> /mnt/xfstests/scratch
> 
> xfs/177 - output mismatch (see /var/lib/xfstests/results//xfs/177.out.bad)
>     --- tests/xfs/177.out       2021-11-28 10:04:49.415509327 -0500
>     +++ /var/lib/xfstests/results//xfs/177.out.bad      2021-11-29
> 04:47:32.148868974 -0500
>     @@ -2,7 +2,8 @@
>      new file count is in range
>      inodes after creating files is in range
>      Round 1
>     -inodes after bulkstat is in range
>     +inodes after bulkstat has value of 575
>     +inodes after bulkstat is NOT in range 1023 .. 1033
>      inodes after expire is in range
>     ...
>     (Run 'diff -u /var/lib/xfstests/tests/xfs/177.out
> /var/lib/xfstests/results//xfs/177.out.bad'  to see the entire diff)
> Ran: xfs/177
> Failures: xfs/177
> Failed 1 of 1 tests
> 
> -- 
> You may reply to this email to add a comment.
> 
> You are receiving this mail because:
> You are watching the assignee of the bug.
> 

-- 
Carlos