> About the second one: > > > ==The Second Issue (aka "The Busy Looping sync()" case) == > > The box is different from first, so conditions are a bit different. > > - /dev/root on / type btrfs (rw,noatime,autodefrag) > > (note autodefrag!) > > - 15% full 594GB filesystem (usual nonmixed mode) > > > > $ liferea > > <wait it to calm down. it does a lot of SQLite reads/writes> > > $ sync > > Got CPU is 100% loaded <hung> > > Still reproducible with 2 patches above + $SUBJ one. strace says it hangs in > strace() syscall. Stack trace is odd: > # cat /proc/`pidof sync`/stack > [<ffffffffffffffff>] 0xffffffffffffffff
"got" stack of the looper with such hack:
$ while :; do sudo echo "-"; cat /proc/2304/stack | fgrep -v
'[<ffffffffffffffff>] 0xffffffffffffffff'; done | uniq
[<ffffffff8109d39d>] find_get_pages_tag+0x14d/0x1a0
[<ffffffff81076981>] __lock_acquire+0x5a1/0xbd0
[<ffffffff8109e7bd>] filemap_fdatawait_range+0x9d/0x1b0
[<ffffffff8109e8f2>] filemap_fdatawait+0x22/0x30
[<ffffffff81106a04>] sync_inodes_sb+0x1c4/0x250
[<ffffffff8110b398>] __sync_filesystem+0x78/0x80
[<ffffffff8110b3b7>] sync_one_sb+0x17/0x20
[<ffffffff810e3fbe>] iterate_supers+0x7e/0xe0
[<ffffffff8110b400>] sys_sync+0x40/0x60
[<ffffffff81456afb>] system_call_fastpath+0x16/0x1b
-
[<ffffffff811069f1>] sync_inodes_sb+0x1b1/0x250
[<ffffffff8110b398>] __sync_filesystem+0x78/0x80
[<ffffffff8110b3b7>] sync_one_sb+0x17/0x20
[<ffffffff810e3fbe>] iterate_supers+0x7e/0xe0
[<ffffffff8110b400>] sys_sync+0x40/0x60
[<ffffffff81456afb>] system_call_fastpath+0x16/0x1b
-
[<ffffffff811069f1>] sync_inodes_sb+0x1b1/0x250
[<ffffffff8110b398>] __sync_filesystem+0x78/0x80
[<ffffffff8110b3b7>] sync_one_sb+0x17/0x20
[<ffffffff810e3fbe>] iterate_supers+0x7e/0xe0
[<ffffffff8110b400>] sys_sync+0x40/0x60
[<ffffffff81456afb>] system_call_fastpath+0x16/0x1b
-
[<ffffffff814564b6>] retint_kernel+0x26/0x30
[<ffffffff8109d398>] find_get_pages_tag+0x148/0x1a0
[<ffffffff810a7a60>] pagevec_lookup_tag+0x20/0x30
[<ffffffff8109e7bd>] filemap_fdatawait_range+0x9d/0x1b0
[<ffffffff8109e8f2>] filemap_fdatawait+0x22/0x30
[<ffffffff81106a04>] sync_inodes_sb+0x1c4/0x250
[<ffffffff8110b398>] __sync_filesystem+0x78/0x80
[<ffffffff8110b3b7>] sync_one_sb+0x17/0x20
[<ffffffff810e3fbe>] iterate_supers+0x7e/0xe0
[<ffffffff8110b400>] sys_sync+0x40/0x60
[<ffffffff81456afb>] system_call_fastpath+0x16/0x1b
Looks like dirty inode count (or dirty page count) is affected by
the $SUBJ patch.
--
Sergei
signature.asc
Description: PGP signature
