From: Tang Junhui <tang.junhui@xxxxxxxxxx>
Hi, everyone:
bcache stucked when reboot system after high load.
root 1704 3.7 0.0 4164 360 ? D 14:07 0:09 /usr/lib/udev/bcache-register /dev/sdc
[<ffffffffa062d2f5>] closure_sync+0x25/0x90 [bcache]
[<ffffffffa062b481>] bch_btree_set_root+0x1f1/0x250 [bcache]
[<ffffffffa062bcf2>] btree_split+0x632/0x760 [bcache]
[<ffffffffa062c1fb>] bch_btree_insert_recurse+0x3db/0x500 [bcache]
[<ffffffffa062c487>] bch_btree_insert+0x167/0x360 [bcache]
[<ffffffffa062feba>] bch_journal_replay+0x1aa/0x2e0 [bcache]
[<ffffffffa0642b36>] run_cache_set+0x813/0x83e [bcache]
[<ffffffffa063aee3>] register_bcache+0xea3/0x1410 [bcache]
[<ffffffff812e453f>] kobj_attr_store+0xf/0x20
[<ffffffff81246be6>] sysfs_write_file+0xc6/0x140
[<ffffffff811cdbfd>] vfs_write+0xbd/0x1e0
[<ffffffff811ce648>] SyS_write+0x58/0xb0
[<ffffffff816306c9>] system_call_fastpath+0x16/0x1b
root 2097 0.0 0.0 0 0 ? D 14:08 0:00 [bcache_allocato]
[<ffffffffa062d2f5>] closure_sync+0x25/0x90 [bcache]
[<ffffffffa06387fe>] bch_prio_write+0x23e/0x340 [bcache]
[<ffffffffa0620e50>] bch_allocator_thread+0x340/0x350 [bcache]
[<ffffffff810990bf>] kthread+0xcf/0xe0
[<ffffffff81630618>] ret_from_fork+0x58/0x90
I try to add some debug info to the code, it seems that it always run in
journal_write_unlocked()
else if (journal_full(&c->journal)) {
journal_reclaim(c);
spin_unlock(&c->journal.lock);
btree_flush_write(c);
continue_at(cl, journal_write, system_wq);
return;
}
the condition of journal_full() always returns true, so the journal
can not finish all the time.
My code has a little difference with the upstream branch.
Could anyone give me some suggestions?
Thanks,
Tang
