_xfs_buf_ioapply: no ops on block 0x188e78/0x8 on vanilla 4.4.11
Hi,
another issue i'm having with 4.4.11 is the following trace after a reboot:
[ 541.661121] XFS (dm-0): _xfs_buf_ioapply: no ops on block 0x188e78/0x8 [ 541.662423] ffff88013a3b3000: 44 51 01 01 00 00 00 00 00 00 00 00 00 00 00 00 DQ.............. [ 541.663258] ffff88013a3b3010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ [ 541.664156] ffff88013a3b3020: 00 00 00 00 00 00 00 00 00 00 00 00 00 0b db fd ................ [ 541.664930] ffff88013a3b3030: 00 00 00 00 00 01 9e 82 00 00 00 00 00 00 00 00 ................ [ 541.665704] CPU: 0 PID: 489 Comm: xfsaild/dm-0 Not tainted 4.4.11 #1 [ 541.666486] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.8.2-0-g33fbe13 by qemu-project.org 04/01/2014 [ 541.667276] 0000000000000000 ffff880037a57c70 ffffffffbe3b3c33 0000000000000001 [ 541.668094] ffff880139d16900 ffff880037a57d18 ffffffffbe31c681 0000000000000000 [ 541.668896] ffff880139d16980 0000000000000000 0000000000000000 0000000000000000 [ 541.669705] Call Trace: [ 541.670491] [<ffffffffbe3b3c33>] dump_stack+0x63/0x90 [ 541.671286] [<ffffffffbe31c681>] _xfs_buf_ioapply+0x3e1/0x440 [ 541.672113] [<ffffffffbe0aaba0>] ? wake_up_q+0x70/0x70 [ 541.672901] [<ffffffffbe31df99>] ? __xfs_buf_delwri_submit+0x179/0x270 [ 541.673704] [<ffffffffbe31dcc4>] xfs_buf_submit+0x64/0x1c0 [ 541.674506] [<ffffffffbe31df99>] __xfs_buf_delwri_submit+0x179/0x270 [ 541.675321] [<ffffffffbe31eea0>] ? xfs_buf_delwri_submit_nowait+0x20/0x30 [ 541.676129] [<ffffffffbe31eea0>] xfs_buf_delwri_submit_nowait+0x20/0x30 [ 541.676930] [<ffffffffbe3495f6>] xfsaild+0x2a6/0x710 [ 541.677738] [<ffffffffbe349350>] ? xfs_trans_ail_cursor_first+0x90/0x90 [ 541.678557] [<ffffffffbe349350>] ? xfs_trans_ail_cursor_first+0x90/0x90 [ 541.679357] [<ffffffffbe09f36b>] kthread+0xdb/0x100 [ 541.680136] [<ffffffffbe09f290>] ? kthread_park+0x60/0x60 [ 541.680912] [<ffffffffbe6972cf>] ret_from_fork+0x3f/0x70 [ 541.681708] [<ffffffffbe09f290>] ? kthread_park+0x60/0x60
I found a bug report from redhat here
https://bugzilla.redhat.com/show_bug.cgi?id=1234586 pointing to this commit:
xfs: handle dquot buffer readahead in log recovery correctly
which isn't part of 4.4. Is it missing from linux-stable?
i'm using XFS V5 with root quotas on in this case.
Greets,
Stefan
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs
Quoted 39 lines · Stefan Priebe - Profihost AG
Hi,
another issue i'm having with 4.4.11 is the following trace after a reboot:
[ 541.661121] XFS (dm-0): _xfs_buf_ioapply: no ops on block 0x188e78/0x8 [ 541.662423] ffff88013a3b3000: 44 51 01 01 00 00 00 00 00 00 00 00 00 00 00 00 DQ.............. [ 541.663258] ffff88013a3b3010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ [ 541.664156] ffff88013a3b3020: 00 00 00 00 00 00 00 00 00 00 00 00 00 0b db fd ................ [ 541.664930] ffff88013a3b3030: 00 00 00 00 00 01 9e 82 00 00 00 00 00 00 00 00 ................ [ 541.665704] CPU: 0 PID: 489 Comm: xfsaild/dm-0 Not tainted 4.4.11 #1 [ 541.666486] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.8.2-0-g33fbe13 by qemu-project.org 04/01/2014 [ 541.667276] 0000000000000000 ffff880037a57c70 ffffffffbe3b3c33 0000000000000001 [ 541.668094] ffff880139d16900 ffff880037a57d18 ffffffffbe31c681 0000000000000000 [ 541.668896] ffff880139d16980 0000000000000000 0000000000000000 0000000000000000 [ 541.669705] Call Trace: [ 541.670491] [<ffffffffbe3b3c33>] dump_stack+0x63/0x90 [ 541.671286] [<ffffffffbe31c681>] _xfs_buf_ioapply+0x3e1/0x440 [ 541.672113] [<ffffffffbe0aaba0>] ? wake_up_q+0x70/0x70 [ 541.672901] [<ffffffffbe31df99>] ? __xfs_buf_delwri_submit+0x179/0x270 [ 541.673704] [<ffffffffbe31dcc4>] xfs_buf_submit+0x64/0x1c0 [ 541.674506] [<ffffffffbe31df99>] __xfs_buf_delwri_submit+0x179/0x270 [ 541.675321] [<ffffffffbe31eea0>] ? xfs_buf_delwri_submit_nowait+0x20/0x30 [ 541.676129] [<ffffffffbe31eea0>] xfs_buf_delwri_submit_nowait+0x20/0x30 [ 541.676930] [<ffffffffbe3495f6>] xfsaild+0x2a6/0x710 [ 541.677738] [<ffffffffbe349350>] ? xfs_trans_ail_cursor_first+0x90/0x90 [ 541.678557] [<ffffffffbe349350>] ? xfs_trans_ail_cursor_first+0x90/0x90 [ 541.679357] [<ffffffffbe09f36b>] kthread+0xdb/0x100 [ 541.680136] [<ffffffffbe09f290>] ? kthread_park+0x60/0x60 [ 541.680912] [<ffffffffbe6972cf>] ret_from_fork+0x3f/0x70 [ 541.681708] [<ffffffffbe09f290>] ? kthread_park+0x60/0x60
Quoted 5 lines
I found a bug report from redhat here
https://bugzilla.redhat.com/show_bug.cgi?id=1234586 pointing to this commit:
xfs: handle dquot buffer readahead in log recovery correctlywhich isn't part of 4.4. Is it missing from linux-stable?
Yes, mainline 7d6a13f02356 is missing from 4.4-stable because it applies, but doesn't compile. :)
I'll look into a fix.
-h
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs
Quoted 11 lines · Holger Hoffstätte
Quoted 5 lines · Stefan Priebe - Profihost AG
I found a bug report from redhat here
https://bugzilla.redhat.com/show_bug.cgi?id=1234586 pointing to this commit:
xfs: handle dquot buffer readahead in log recovery correctlywhich isn't part of 4.4. Is it missing from linux-stable?
Yes, mainline
7d6a13f02356is missing from 4.4-stable because it applies, but doesn't compile. :)I'll look into a fix.
Well, that was easy. It also needs 233135b763 ("xfs: print name of verifier if it fails") which added a .name to the various xfs_buf_ops types. Can someone confirm that these can both go into 4.4-stable?
-h
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs
Quoted 8 lines · Holger Hoffstätte
Quoted 7 lines · Holger Hoffstätte
Quoted 5 lines · Stefan Priebe - Profihost AG
I found a bug report from redhat here
https://bugzilla.redhat.com/show_bug.cgi?id=1234586 pointing to this commit:
xfs: handle dquot buffer readahead in log recovery correctlywhich isn't part of 4.4. Is it missing from linux-stable?
Quoted 8 lines
Quoted 4 lines
Yes, mainline
7d6a13f02356is missing from 4.4-stable because it applies, but doesn't compile. :)I'll look into a fix.
Well, that was easy. It also needs
233135b763("xfs: print name of verifier if it fails") which added a .name to the variousxfs_buf_opstypes. Can someone confirm that these can both go into 4.4-stable?
You could just remove the line:
+ .name = "xfs_dquot_ra",
from 7d6a13f02356 and that will fix the backport problem and not require any other commits to be backported. I'm fine with either solution.
Cheers,
Dave.
Signature · 7 lines
Dave Chinner [email protected] _______________________________________________ xfs mailing list [email protected] http://oss.sgi.com/mailman/listinfo/xfs
Quoted 18 lines · Dave Chinner
Quoted 8 lines · Holger Hoffstätte
Quoted 7 lines · Holger Hoffstätte
Quoted 5 lines · Stefan Priebe - Profihost AG
I found a bug report from redhat here
https://bugzilla.redhat.com/show_bug.cgi?id=1234586 pointing to this commit:
xfs: handle dquot buffer readahead in log recovery correctlywhich isn't part of 4.4. Is it missing from linux-stable?
Quoted 8 lines
Quoted 4 lines
Yes, mainline
7d6a13f02356is missing from 4.4-stable because it applies, but doesn't compile. :)I'll look into a fix.
Well, that was easy. It also needs
233135b763("xfs: print name of verifier if it fails") which added a .name to the variousxfs_buf_opstypes. Can someone confirm that these can both go into 4.4-stable?
Quoted 7 lines
You could just remove the line:
+ .name = "
xfs_dquot_ra",from
7d6a13f02356and that will fix the backport problem and not require any other commits to be backported. I'm fine with either solution.
Ha..that totally didn't occur to me..too simple. :}
The .name patch applies cleanly and can't hurt either, so I think sending both commits to Greg is easiest for everyone.
thanks!
Holger
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs
Am 03.06.2016 um 00:25 schrieb Dave Chinner:
Quoted 18 lines · Holger Hoffstätte
Quoted 8 lines · Holger Hoffstätte
Quoted 7 lines · Stefan Priebe - Profihost AG
Quoted 5 lines
I found a bug report from redhat here
https://bugzilla.redhat.com/show_bug.cgi?id=1234586 pointing to this commit:
xfs: handle dquot buffer readahead in log recovery correctlywhich isn't part of 4.4. Is it missing from linux-stable?
Quoted 8 lines
Quoted 4 lines
Yes, mainline
7d6a13f02356is missing from 4.4-stable because it applies, but doesn't compile. :)I'll look into a fix.
Well, that was easy. It also needs
233135b763("xfs: print name of verifier if it fails") which added a .name to the variousxfs_buf_opstypes. Can someone confirm that these can both go into 4.4-stable?
Quoted 8 lines
You could just remove the line:
+ .name = "
xfs_dquot_ra",from
7d6a13f02356and that will fix the backport problem and not require any other commits to be backported. I'm fine with either solution.
Great. Should be push that one to greg?
Greets,
Stefan
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs
Quoted 21 lines · Stefan Priebe - Profihost AG
Am 03.06.2016 um 00:25 schrieb Dave Chinner:
Quoted 20 lines · Holger Hoffstätte
Quoted 8 lines · Holger Hoffstätte
Quoted 7 lines · Stefan Priebe - Profihost AG
Quoted 5 lines
I found a bug report from redhat here
https://bugzilla.redhat.com/show_bug.cgi?id=1234586 pointing to this commit:
xfs: handle dquot buffer readahead in log recovery correctlywhich isn't part of 4.4. Is it missing from linux-stable?
Quoted 8 lines
Quoted 4 lines
Yes, mainline
7d6a13f02356is missing from 4.4-stable because it applies, but doesn't compile. :)I'll look into a fix.
Well, that was easy. It also needs
233135b763("xfs: print name of verifier if it fails") which added a .name to the variousxfs_buf_opstypes. Can someone confirm that these can both go into 4.4-stable?You could just remove the line:
Quoted 8 lines
Quoted 6 lines
+ .name = "
xfs_dquot_ra",from
7d6a13f02356and that will fix the backport problem and not require any other commits to be backported. I'm fine with either solution.Great. Should be push that one to greg?
You're too slow ;-)
http://www.spinics.net/lists/stable/msg134549.html
..or steal them from my patch queue.
-h
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs