> On May 2, 2016, at 4:07 PM, Mkrtchyan, Tigran <tigran.mkrtchyan@xxxxxxx> wrote: > > Hi Dros, > > as you can see from the stack trace, this is NFSv3. Oh, I didn't notice that! > For now, it happened only once. May be there is some change with > our workflow. I will let you know, if we see it again. OK. If you do, try applying that patch set and see if it’s still reproducible. -dros > > > Thanks, > Tigran. > > ----- Original Message ----- >> From: "Weston Andros Adamson" <dros@xxxxxxxxxx> >> To: "Tigran Mkrtchyan" <tigran.mkrtchyan@xxxxxxx> >> Cc: "linux-nfs list" <linux-nfs@xxxxxxxxxxxxxxx>, "Steve Dickson" <steved@xxxxxxxxxx>, "Andy Adamson" >> <william.adamson@xxxxxxxxxx> >> Sent: Monday, May 2, 2016 3:30:29 PM >> Subject: Re: blocked on the spin lock > >>> On May 2, 2016, at 3:29 AM, Mkrtchyan, Tigran <tigran.mkrtchyan@xxxxxxx> wrote: >>> >>> >>> >>> Hi Dros et. al. >>> >>> We have seen the following stack trace on one of out systems: >>> >>> >>> Apr 28 22:09:51 bird510 kernel: BUG: soft lockup - CPU#7 stuck for 67s! >>> [tee:13755] >>> Apr 28 22:09:51 bird510 kernel: Modules linked in: ipmi_devintf dell_rbu >>> nfs_layout_nfsv41_files fuse nfs lockd fscache auth_rpcgss nfs_acl vfat fat >>> usb_storage mpt3sas mpt2sas raid_class mptctl openafs(P)(U) autofs4 sunrpc >>> cpufreq_ondemand acpi_cpufreq freq_table mperf sg joy >>> dev power_meter acpi_ipmi ipmi_si ipmi_msghandler iTCO_wdt iTCO_vendor_support >>> bnx2 microcode dcdbas lpc_ich mfd_core i7core_edac edac_core shpchp ext4 jbd2 >>> mbcache sd_mod crc_t10dif mptsas mptscsih mptbase scsi_transport_sas mlx4_ib >>> ib_sa ib_mad ib_core ib_addr ipv6 mlx4_en p >>> tp pps_core mlx4_core dm_mirror dm_region_hash dm_log dm_mod [last unloaded: >>> ipmi_devintf] >>> Apr 28 22:09:51 bird510 kernel: CPU 7 >>> Apr 28 22:09:51 bird510 kernel: Modules linked in: ipmi_devintf dell_rbu >>> nfs_layout_nfsv41_files fuse nfs lockd fscache auth_rpcgss nfs_acl vfat fat >>> usb_storage mpt3sas mpt2sas raid_class mptctl openafs(P)(U) autofs4 sunrpc >>> cpufreq_ondemand acpi_cpufreq freq_table mperf sg joy >>> dev power_meter acpi_ipmi ipmi_si ipmi_msghandler iTCO_wdt iTCO_vendor_support >>> bnx2 microcode dcdbas lpc_ich mfd_core i7core_edac edac_core shpchp ext4 jbd2 >>> mbcache sd_mod crc_t10dif mptsas mptscsih mptbase scsi_transport_sas mlx4_ib >>> ib_sa ib_mad ib_core ib_addr ipv6 mlx4_en p >>> tp pps_core mlx4_core dm_mirror dm_region_hash dm_log dm_mod [last unloaded: >>> ipmi_devintf] >>> Apr 28 22:09:51 bird510 kernel: >>> Apr 28 22:09:51 bird510 kernel: Pid: 13755, comm: tee Tainted: P -- >>> ------------ 2.6.32-573.12.1.el6.x86_64 #1 Dell Inc. PowerEdge M610/0N582M >>> Apr 28 22:09:51 bird510 kernel: RIP: 0010:[<ffffffff8153bde1>] >>> [<ffffffff8153bde1>] _spin_lock+0x21/0x30 >>> Apr 28 22:09:51 bird510 kernel: RSP: 0018:ffff8803528179c8 EFLAGS: 00000297 >>> Apr 28 22:09:51 bird510 kernel: RAX: 00000000000003b1 RBX: ffff8803528179c8 RCX: >>> 0000000000000000 >>> Apr 28 22:09:51 bird510 kernel: RDX: 00000000000003b0 RSI: ffff880159aeb820 RDI: >>> ffff880159aeb8d0 >>> Apr 28 22:09:51 bird510 kernel: RBP: ffffffff8100bc0e R08: ffff880352817af0 R09: >>> 0000000000000002 >>> Apr 28 22:09:51 bird510 kernel: R10: ffffea0008efc428 R11: 0000000000000003 R12: >>> 0000000000000001 >>> Apr 28 22:09:51 bird510 kernel: R13: 001310facb5806bd R14: 0000000000000000 R15: >>> 0000000000000000 >>> Apr 28 22:09:51 bird510 kernel: FS: 00007f3e78118700(0000) >>> GS:ffff880028260000(0000) knlGS:0000000000000000 >>> Apr 28 22:09:51 bird510 kernel: CS: 0010 DS: 0000 ES: 0000 CR0: >>> 000000008005003b >>> Apr 28 22:09:51 bird510 kernel: CR2: 00007fefa21ff000 CR3: 0000000352959000 CR4: >>> 00000000000007e0 >>> Apr 28 22:09:51 bird510 kernel: DR0: 0000000000000000 DR1: 0000000000000000 DR2: >>> 0000000000000000 >>> Apr 28 22:09:51 bird510 kernel: DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: >>> 0000000000000400 >>> Apr 28 22:09:51 bird510 kernel: Process tee (pid: 13755, threadinfo >>> ffff880352814000, task ffff880636064ab0) >>> Apr 28 22:09:51 bird510 kernel: Stack: >>> Apr 28 22:09:51 bird510 kernel: ffff880352817a18 ffffffffa0490dc1 >>> ffff880159aeb8d0 ffff880159aeb770 >>> Apr 28 22:09:51 bird510 kernel: <d> 0000000000000000 0000000000000000 >>> ffffffffa04bd040 ffffffffa048f39a >>> Apr 28 22:09:51 bird510 kernel: <d> ffff8803378761c0 ffffea000a13dcc0 >>> ffff880352817a48 ffffffffa0490ec8 >>> Apr 28 22:09:51 bird510 kernel: Call Trace: >>> Apr 28 22:09:51 bird510 kernel: [<ffffffffa0490dc1>] ? >>> nfs_clear_request_commit+0xa1/0xf0 [nfs] >>> Apr 28 22:09:51 bird510 kernel: [<ffffffffa048f39a>] ? >>> nfs_page_find_request_locked+0x2a/0x40 [nfs] >>> Apr 28 22:09:51 bird510 kernel: [<ffffffffa0490ec8>] ? >>> nfs_wb_page_cancel+0xb8/0xf0 [nfs] >>> Apr 28 22:09:51 bird510 kernel: [<ffffffffa047e4f7>] ? >>> nfs_invalidate_page+0x47/0x80 [nfs] >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff8113edb5>] ? >>> do_invalidatepage+0x25/0x30 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff8113f0d2>] ? >>> truncate_inode_page+0xa2/0xc0 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff8113f47f>] ? >>> truncate_inode_pages_range+0x16f/0x500 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff81129cd7>] ? >>> mempool_free_slab+0x17/0x20 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff8113f8a5>] ? >>> truncate_inode_pages+0x15/0x20 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff8113f8ff>] ? >>> truncate_pagecache+0x4f/0x70 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffffa0480eb9>] ? >>> nfs_setattr_update_inode+0xb9/0x150 [nfs] >>> Apr 28 22:09:51 bird510 kernel: [<ffffffffa0493a6b>] ? >>> nfs3_proc_setattr+0xdb/0x120 [nfs] >>> Apr 28 22:09:51 bird510 kernel: [<ffffffffa048f6b3>] ? nfs_wb_all+0x43/0x50 >>> [nfs] >>> Apr 28 22:09:51 bird510 kernel: [<ffffffffa0482610>] ? nfs_setattr+0xf0/0x170 >>> [nfs] >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff811b1118>] ? notify_change+0x168/0x340 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffffa0480de8>] ? nfs_open+0x78/0x90 [nfs] >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff8118fcb4>] ? do_truncate+0x64/0xa0 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff81232b9f>] ? >>> security_inode_permission+0x1f/0x30 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff811a4fc1>] ? do_filp_open+0x861/0xd20 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff8129df8a>] ? >>> strncpy_from_user+0x4a/0x90 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff811b1ff2>] ? alloc_fd+0x92/0x160 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff8118e967>] ? do_sys_open+0x67/0x130 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff8118ea70>] ? sys_open+0x20/0x30 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff8100b0d2>] ? >>> system_call_fastpath+0x16/0x1b >>> Apr 28 22:09:51 bird510 kernel: Code: 01 74 05 e8 52 1c d6 ff c9 c3 55 48 89 e5 >>> 0f 1f 44 00 00 b8 00 00 01 00 f0 0f c1 07 0f b7 d0 c1 e8 10 39 c2 74 0e f3 90 >>> 0f b7 17 <eb> f5 83 3f 00 75 f4 eb df c9 c3 0f 1f 40 00 55 48 89 e5 0f 1f >>> Apr 28 22:09:51 bird510 kernel: Call Trace: >>> Apr 28 22:09:51 bird510 kernel: [<ffffffffa0490dc1>] ? >>> nfs_clear_request_commit+0xa1/0xf0 [nfs] >>> Apr 28 22:09:51 bird510 kernel: [<ffffffffa048f39a>] ? >>> nfs_page_find_request_locked+0x2a/0x40 [nfs] >>> Apr 28 22:09:51 bird510 kernel: [<ffffffffa0490ec8>] ? >>> nfs_wb_page_cancel+0xb8/0xf0 [nfs] >>> Apr 28 22:09:51 bird510 kernel: [<ffffffffa047e4f7>] ? >>> nfs_invalidate_page+0x47/0x80 [nfs] >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff8113edb5>] ? >>> do_invalidatepage+0x25/0x30 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff8113f0d2>] ? >>> truncate_inode_page+0xa2/0xc0 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff8113f47f>] ? >>> truncate_inode_pages_range+0x16f/0x500 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff81129cd7>] ? >>> mempool_free_slab+0x17/0x20 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff8113f8a5>] ? >>> truncate_inode_pages+0x15/0x20 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff8113f8ff>] ? >>> truncate_pagecache+0x4f/0x70 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffffa0480eb9>] ? >>> nfs_setattr_update_inode+0xb9/0x150 [nfs] >>> Apr 28 22:09:51 bird510 kernel: [<ffffffffa0493a6b>] ? >>> nfs3_proc_setattr+0xdb/0x120 [nfs] >>> Apr 28 22:09:51 bird510 kernel: [<ffffffffa048f6b3>] ? nfs_wb_all+0x43/0x50 >>> [nfs] >>> Apr 28 22:09:51 bird510 kernel: [<ffffffffa0482610>] ? nfs_setattr+0xf0/0x170 >>> [nfs] >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff811b1118>] ? notify_change+0x168/0x340 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffffa0480de8>] ? nfs_open+0x78/0x90 [nfs] >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff8118fcb4>] ? do_truncate+0x64/0xa0 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff81232b9f>] ? >>> security_inode_permission+0x1f/0x30 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff811a4fc1>] ? do_filp_open+0x861/0xd20 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff8129df8a>] ? >>> strncpy_from_user+0x4a/0x90 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff811b1ff2>] ? alloc_fd+0x92/0x160 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff8118e967>] ? do_sys_open+0x67/0x130 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff8118ea70>] ? sys_open+0x20/0x30 >>> Apr 28 22:09:51 bird510 kernel: [<ffffffff8100b0d2>] ? >>> system_call_fastpath+0x16/0x1b >>> >>> >>> >>> >>> The questionable spin lock was remove by commit 411a99adf. Is there was a >>> problem? >>> Do we need a packport in RHEL kernel? >>> >>> Thanks a lot, >>> Tigran. >>> >> >> Hey Tigran, >> >> It’s been a while since I looked at this, so I don’t know off the top of my >> head, but IIRC the patch series that 411a99adf is part of are some important >> bug fixes to the effort to split sub-page regions: >> >> 411a99adffb4 nfs: clear_request_commit while holding i_lock >> e6cf82d1830f pnfs: add pnfs_put_lseg_async >> 02d1426c7053 pnfs: find swapped pages on pnfs commit lists too >> b412ddf0661e nfs: fix comment and add warn_on for PG_INODE_REF >> e7029206ff43 nfs: check wait_on_bit_lock err in page_group_lock >> >> Also, if you are using the pnfs_nfs commit path, there are several recent fixes >> that are pretty important. Until recently, the only server to do NFS backed >> pnfs did only STABLE writes, so that stuff was pretty much untested… >> >> In fact, I have a patch to post today that fixes a problem in the commit path, >> although it probably isn’t an issue in RHEL as it was a regression introduced >> by a recent patch. >> >> -dros -- To unsubscribe from this list: send the line "unsubscribe linux-nfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
