From: David P. <dav...@gm...> - 2012-02-14 13:12:04
|
Hello, I'm currently working on an implementation of a target driver through scst_user. I recently encountered an issue, where my daemon hanged (or was quite slow). This occurred during a 'mkfs.ext2' on a 8TB exported block device (through scst_local). After looking through the logs, here is the (possibly relevant) kernel trace I could find : Feb 14 12:15:06 blockserver kernel: [ 5990.526260] INFO: task blockd:6267 blocked for more than 120 seconds. Feb 14 12:15:06 blockserver kernel: [ 5990.526263] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. Feb 14 12:15:06 blockserver kernel: [ 5990.526266] blockd D 0000000000000001 0 6267 6021 0x00000000 Feb 14 12:15:06 blockserver kernel: [ 5990.526270] ffff88005deb7338 0000000000000082 ffff88005deb7fd8 ffff88005deb6000 Feb 14 12:15:06 blockserver kernel: [ 5990.526276] 0000000000013d00 ffff88005deb9a98 ffff88005deb7fd8 0000000000013d00 Feb 14 12:15:06 blockserver kernel: [ 5990.526281] ffff880137720000 ffff88005deb96e0 ffff880040d7fc80 ffff8800bf233d00 Feb 14 12:15:06 blockserver kernel: [ 5990.526286] Call Trace: Feb 14 12:15:06 blockserver kernel: [ 5990.526291] [<ffffffff815c0ea0>] io_schedule+0x70/0xc0 Feb 14 12:15:06 blockserver kernel: [ 5990.526296] [<ffffffff812c2f49>] get_request_wait+0xc9/0x1a0 Feb 14 12:15:06 blockserver kernel: [ 5990.526301] [<ffffffff81087fb0>] ? autoremove_wake_function+0x0/0x40 Feb 14 12:15:06 blockserver kernel: [ 5990.526305] [<ffffffff812bc1e8>] ? elv_merge+0x108/0x120 Feb 14 12:15:06 blockserver kernel: [ 5990.526310] [<ffffffff812c3706>] __make_request+0x76/0x4c0 Feb 14 12:15:06 blockserver kernel: [ 5990.526314] [<ffffffff81038c79>] ? default_spin_lock_flags+0x9/0x10 Feb 14 12:15:06 blockserver kernel: [ 5990.526319] [<ffffffff812c0bf8>] generic_make_request+0x2d8/0x5c0 Feb 14 12:15:06 blockserver kernel: [ 5990.526324] [<ffffffff8110e7a5>] ? mempool_alloc_slab+0x15/0x20 Feb 14 12:15:06 blockserver kernel: [ 5990.526328] [<ffffffff8110eae9>] ? mempool_alloc+0x59/0x140 Feb 14 12:15:06 blockserver kernel: [ 5990.526332] [<ffffffff812c0f69>] submit_bio+0x89/0x120 Feb 14 12:15:06 blockserver kernel: [ 5990.526337] [<ffffffff8119776b>] ? bio_alloc_bioset+0x5b/0xf0 Feb 14 12:15:06 blockserver kernel: [ 5990.526342] [<ffffffff8119195b>] submit_bh+0xeb/0x120 Feb 14 12:15:06 blockserver kernel: [ 5990.526346] [<ffffffff811936a0>] __block_write_full_page+0x210/0x3a0 Feb 14 12:15:06 blockserver kernel: [ 5990.526350] [<ffffffff812df229>] ? prio_tree_next+0x199/0x1a0 Feb 14 12:15:06 blockserver kernel: [ 5990.526354] [<ffffffff81192790>] ? end_buffer_async_write+0x0/0x170 Feb 14 12:15:06 blockserver kernel: [ 5990.526359] [<ffffffff81197fc0>] ? blkdev_get_block+0x0/0x70 Feb 14 12:15:06 blockserver kernel: [ 5990.526363] [<ffffffff81197fc0>] ? blkdev_get_block+0x0/0x70 Feb 14 12:15:06 blockserver kernel: [ 5990.526367] [<ffffffff81194543>] block_write_full_page_endio+0xe3/0x120 Feb 14 12:15:06 blockserver kernel: [ 5990.526372] [<ffffffff81194595>] block_write_full_page+0x15/0x20 Feb 14 12:15:06 blockserver kernel: [ 5990.526376] [<ffffffff81198b48>] blkdev_writepage+0x18/0x20 Feb 14 12:15:06 blockserver kernel: [ 5990.526381] [<ffffffff811594c7>] fallback_migrate_page+0xc7/0x130 Feb 14 12:15:06 blockserver kernel: [ 5990.526386] [<ffffffff81159633>] move_to_new_page+0x103/0x120 Feb 14 12:15:06 blockserver kernel: [ 5990.526391] [<ffffffff81159ae6>] unmap_and_move+0x2f6/0x350 Feb 14 12:15:06 blockserver kernel: [ 5990.526396] [<ffffffff8114e7f0>] ? compaction_alloc+0x0/0x70 Feb 14 12:15:06 blockserver kernel: [ 5990.526400] [<ffffffff81159d32>] migrate_pages+0xa2/0x150 Feb 14 12:15:06 blockserver kernel: [ 5990.526405] [<ffffffff8114e9c4>] compact_zone+0xa4/0x220 Feb 14 12:15:06 blockserver kernel: [ 5990.526410] [<ffffffff8114ed0c>] compact_zone_order+0xac/0xf0 Feb 14 12:15:06 blockserver kernel: [ 5990.526415] [<ffffffff8114ee1d>] try_to_compact_pages+0xcd/0x100 Feb 14 12:15:06 blockserver kernel: [ 5990.526420] [<ffffffff8111411f>] __alloc_pages_direct_compact+0xcf/0x190 Feb 14 12:15:06 blockserver kernel: [ 5990.526425] [<ffffffff8111461d>] __alloc_pages_nodemask+0x43d/0x840 Feb 14 12:15:06 blockserver kernel: [ 5990.526431] [<ffffffff81149f85>] alloc_pages_current+0xa5/0x110 Feb 14 12:15:06 blockserver kernel: [ 5990.526435] [<ffffffff811536d5>] new_slab+0x1f5/0x290 Feb 14 12:15:06 blockserver kernel: [ 5990.526440] [<ffffffff81155172>] __slab_alloc+0x1b2/0x390 Feb 14 12:15:06 blockserver kernel: [ 5990.526465] [<ffffffffa047a4e5>] ? sgv_pool_alloc+0x455/0xa50 [scst] Feb 14 12:15:06 blockserver kernel: [ 5990.526470] [<ffffffff81155e4d>] __kmalloc+0x14d/0x160 Feb 14 12:15:06 blockserver kernel: [ 5990.526492] [<ffffffffa047a4e5>] ? sgv_pool_alloc+0x455/0xa50 [scst] Feb 14 12:15:06 blockserver kernel: [ 5990.526513] [<ffffffffa047a4e5>] sgv_pool_alloc+0x455/0xa50 [scst] Feb 14 12:15:06 blockserver kernel: [ 5990.526520] [<ffffffffa04eda4e>] ? dev_user_add_to_ready_head+0xae/0x1d0 [scst_user] Feb 14 12:15:06 blockserver kernel: [ 5990.526527] [<ffffffffa04eb606>] dev_user_alloc_sg+0x146/0x6e0 [scst_user] Feb 14 12:15:06 blockserver kernel: [ 5990.526534] [<ffffffffa04eee2e>] dev_user_alloc_space+0x6e/0x260 [scst_user] Feb 14 12:15:06 blockserver kernel: [ 5990.526541] [<ffffffffa04ef07a>] dev_user_alloc_data_buf+0x5a/0x130 [scst_user] Feb 14 12:15:06 blockserver kernel: [ 5990.526555] [<ffffffffa044b189>] scst_prepare_space+0x99/0x670 [scst] Feb 14 12:15:06 blockserver kernel: [ 5990.526570] [<ffffffffa04506a5>] scst_process_active_cmd+0x3f5/0x720 [scst] Feb 14 12:15:06 blockserver kernel: [ 5990.526577] [<ffffffffa04effa8>] dev_user_process_reply_alloc.clone.9+0xb8/0x210 [scst_user] Feb 14 12:15:06 blockserver kernel: [ 5990.526584] [<ffffffffa04f2014>] dev_user_process_reply+0x1f4/0x3e0 [scst_user] Feb 14 12:15:06 blockserver kernel: [ 5990.526592] [<ffffffffa04f33bc>] dev_user_reply_get_cmd.clone.13+0x20c/0x4b0 [scst_user] Feb 14 12:15:06 blockserver kernel: [ 5990.526599] [<ffffffffa04f4092>] dev_user_ioctl+0x222/0x6b4 [scst_user] Feb 14 12:15:06 blockserver kernel: [ 5990.526605] [<ffffffff8118332f>] ? mntput+0x1f/0x30 Feb 14 12:15:06 blockserver kernel: [ 5990.526609] [<ffffffff811764ef>] do_vfs_ioctl+0x8f/0x360 Feb 14 12:15:06 blockserver kernel: [ 5990.526614] [<ffffffff81176851>] sys_ioctl+0x91/0xa0 Feb 14 12:15:06 blockserver kernel: [ 5990.526618] [<ffffffff8100c002>] system_call_fastpath+0x16/0x1b This trace was accompanied by other traces of the same kind. As far as I understand, this is the trace at the time where the unusual wait occurred, but I have no way of identifying the cause. Is there anything you could notice that I didn't through this trace ? Thanks for the help, -- David Pineau |