Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1272018 > unrolled thread

kernel BUG at drivers/scsi/scsi_lib.c:1096!

Started byMichael Ellerman <mpe@ellerman.id.au>
First post2015-11-18 10:20 +0100
Last post2015-11-24 02:40 +0100
Articles 20 on this page of 34 — 12 participants

Back to article view | Back to linux.kernel

Contents

  kernel BUG at drivers/scsi/scsi_lib.c:1096! Michael Ellerman <mpe@ellerman.id.au> - 2015-11-18 10:20 +0100
    Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Mark Salter <msalter@redhat.com> - 2015-11-18 15:10 +0100
      Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Michael Ellerman <mpe@ellerman.id.au> - 2015-11-19 02:10 +0100
        Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Christoph Hellwig <hch@infradead.org> - 2015-11-19 09:30 +0100
          Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Hannes Reinecke <hare@suse.de> - 2015-11-19 16:40 +0100
            Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Ewan Milne <emilne@redhat.com> - 2015-11-20 15:40 +0100
              Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Hannes Reinecke <hare@suse.de> - 2015-11-20 16:10 +0100
                Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Ewan Milne <emilne@redhat.com> - 2015-11-20 16:30 +0100
                  Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Hannes Reinecke <hare@suse.de> - 2015-11-23 08:00 +0100
                  Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Hannes Reinecke <hare@suse.de> - 2015-11-25 10:10 +0100
                    Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Jens Axboe <axboe@fb.com> - 2015-11-25 19:00 +0100
                      Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Hannes Reinecke <hare@suse.de> - 2015-11-25 20:20 +0100
                        Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Jens Axboe <axboe@fb.com> - 2015-11-25 20:30 +0100
                          Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Mike Snitzer <snitzer@redhat.com> - 2015-11-25 21:30 +0100
                            Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Mike Snitzer <snitzer@redhat.com> - 2015-11-25 22:30 +0100
                    Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Mike Snitzer <snitzer@redhat.com> - 2015-11-25 19:10 +0100
                      Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Hannes Reinecke <hare@suse.de> - 2015-11-25 20:10 +0100
          Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Michael Ellerman <mpe@ellerman.id.au> - 2015-11-20 13:20 +0100
            Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Laurent Dufour <ldufour@linux.vnet.ibm.com> - 2015-11-20 14:00 +0100
              Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Mark Salter <msalter@redhat.com> - 2015-11-20 14:40 +0100
            Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Laurent Dufour <ldufour@linux.vnet.ibm.com> - 2015-11-21 12:40 +0100
              Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Ming Lei <ming.lei@canonical.com> - 2015-11-21 18:00 +0100
                Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Mark Salter <msalter@redhat.com> - 2015-11-23 00:30 +0100
                  Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Ming Lei <ming.lei@canonical.com> - 2015-11-23 01:40 +0100
                    Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Mark Salter <msalter@redhat.com> - 2015-11-23 03:00 +0100
                      Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Ming Lei <ming.lei@canonical.com> - 2015-11-23 03:50 +0100
                        Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Ming Lei <tom.leiming@gmail.com> - 2015-11-23 16:30 +0100
                          Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Alan Ott <alan@softiron.co.uk> - 2015-11-24 20:40 +0100
                  Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Laurent Dufour <ldufour@linux.vnet.ibm.com> - 2015-11-23 15:00 +0100
                    Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Pratyush Anand <panand@redhat.com> - 2015-11-23 16:20 +0100
                      Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Laurent Dufour <ldufour@linux.vnet.ibm.com> - 2015-11-23 16:30 +0100
                        Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Ming Lei <ming.lei@canonical.com> - 2015-11-23 16:30 +0100
                          Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Laurent Dufour <ldufour@linux.vnet.ibm.com> - 2015-11-23 17:30 +0100
                          Re: kernel BUG at drivers/scsi/scsi_lib.c:1096! Mark Salter <msalter@redhat.com> - 2015-11-24 02:40 +0100

Page 1 of 2  [1] 2  Next page →

#1272018 — kernel BUG at drivers/scsi/scsi_lib.c:1096!

FromMichael Ellerman <mpe@ellerman.id.au>
Date2015-11-18 10:20 +0100
Subjectkernel BUG at drivers/scsi/scsi_lib.c:1096!
Message-ID<qw74e-WO-35@gated-at.bofh.it>
Hi folks,

I'm intermittently seeing the following oops on at least one powerpc box.

The BUG_ON() is from:

static int scsi_init_sgtable(struct request *req, struct scsi_data_buffer *sdb)
{
	...
	count = blk_rq_map_sg(req->q, req, sdb->table.sgl);
	BUG_ON(count > sdb->table.nents);

Looking at the dump it looks like count was 2, I can't work out what nents was.

The machine's just a fairly boring bare metal setup, with a single IPR adapter:

0001:08:00.0 RAID bus controller: IBM PCI-E IPR SAS Adapter (ASIC) (rev 02)
	Subsystem: IBM PCIe3 x8 SAS RAID Internal Adapter 6Gb (57D7)
	Flags: bus master, fast devsel, latency 0
	Kernel driver in use: ipr


Anyone seen it before or have any ideas?

cheers


systemd[1]: Starting Uncomplicated firewall...
         Starting Uncomplicated firewall...
------------[ cut here ]------------
kernel BUG at drivers/scsi/scsi_lib.c:1096!
Oops: Exception in kernel mode, sig: 5 [#1]
SMP NR_CPUS=2048 NUMA PowerNV
Modules linked in:
CPU: 132 PID: 2699 Comm: kworker/132:1H Not tainted 4.4.0-rc1-54939-ge22a248-dirty #77
Workqueue: kblockd cfq_kick_queue
task: c000000fef147400 ti: c000000feb384000 task.ti: c000000feb384000
NIP: c0000000005ab4a8 LR: c0000000005ab490 CTR: 0000000000000000
REGS: c000000feb387620 TRAP: 0700   Not tainted  (4.4.0-rc1-54939-ge22a248-dirty)
MSR: 9000000100029033 <SF,HV,EE,ME,IR,DR,RI,LE>  CR: 24002228  XER: 00000000
CFAR: c000000000464950 SOFTE: 0 
GPR00: c0000000005ab490 c000000feb3878a0 c000000000d77d00 0000000000000002 
GPR04: c000000ff2030158 c000000ff47a0c00 0000000000000000 0000000000001000 
GPR08: 0000000000000000 0000000000000001 0000000000000000 fffffffffffffff7 
GPR12: 0000000000002200 c00000000fde5200 c0000000000c8098 c000000000b39858 
GPR16: c000000000ae12c8 c000000000b39948 0000000000000000 0000000000000000 
GPR20: c0000000005ab9c0 c0000017f5800144 0000000000000000 c000001e551a6850 
GPR24: c0000017f5800140 c0000007efb9c800 0000000000000000 c000001e551a6800 
GPR28: 0000000000000000 0000000000000000 c000000ff2030158 c000000feb420240 
NIP [c0000000005ab4a8] scsi_init_sgtable+0xa8/0x180
LR [c0000000005ab490] scsi_init_sgtable+0x90/0x180
Call Trace:
[c000000feb3878a0] [c0000000005ab490] scsi_init_sgtable+0x90/0x180 (unreliable)
[c000000feb3878e0] [c0000000005ab5d4] scsi_init_io+0x54/0x160
[c000000feb387930] [c0000000005fb43c] sd_init_command+0x6c/0xb00
[c000000feb3879f0] [c0000000005ac2f8] scsi_setup_cmnd+0x108/0x1a0
[c000000feb387a30] [c0000000005ac594] scsi_prep_fn+0x104/0x1c0
[c000000feb387a70] [c00000000045d85c] blk_peek_request+0x20c/0x390
[c000000feb387af0] [c0000000005ad6a8] scsi_request_fn+0xb8/0x7f0
[c000000feb387bf0] [c0000000004583b4] __blk_run_queue+0x54/0x80
[c000000feb387c20] [c00000000047e178] cfq_kick_queue+0x38/0xc0
[c000000feb387c50] [c0000000000c00cc] process_one_work+0x2ac/0x560
[c000000feb387ce0] [c0000000000c0510] worker_thread+0x190/0x660
[c000000feb387d80] [c0000000000c8198] kthread+0x108/0x130
[c000000feb387e30] [c0000000000095b0] ret_from_kernel_thread+0x5c/0xac
Instruction dump:
2fa30000 409e00e4 e8bf0000 e87e0030 7fc4f378 4beb91ed 60000000 813f0008 
78630020 7f891840 7d301026 5529effe <0b090000> 907f0008 38600000 813e005c 
---[ end trace 3ea632a512ff48d5 ]---

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [next] | [standalone]

#1272203

FromMark Salter <msalter@redhat.com>
Date2015-11-18 15:10 +0100
Message-ID<qwbAT-3W1-33@gated-at.bofh.it>
In reply to#1272018
On Wed, 2015-11-18 at 20:18 +1100, Michael Ellerman wrote:
> Hi folks,
> 
> I'm intermittently seeing the following oops on at least one powerpc box.
> 
> The BUG_ON() is from:
> 
> static int scsi_init_sgtable(struct request *req, struct scsi_data_buffer *sdb)
> {
> 	...
> 	count = blk_rq_map_sg(req->q, req, sdb->table.sgl);
> 	BUG_ON(count > sdb->table.nents);
> 
> Looking at the dump it looks like count was 2, I can't work out what nents was.
> 
> The machine's just a fairly boring bare metal setup, with a single IPR adapter:
> 
> 0001:08:00.0 RAID bus controller: IBM PCI-E IPR SAS Adapter (ASIC) (rev 02)
> 	Subsystem: IBM PCIe3 x8 SAS RAID Internal Adapter 6Gb (57D7)
> 	Flags: bus master, fast devsel, latency 0
> 	Kernel driver in use: ipr
> 
> 
> Anyone seen it before or have any ideas?

I'm also seeing it on arm64 in 4.4-rc1

[    6.859003] Call trace:                                                                      
[    6.861439] [<fffffe000074345c>] scsi_init_sgtable+0x84/0x88                                 
[    6.867072] [<fffffe00007434ac>] scsi_init_io+0x4c/0x1ac                                     
[    6.872358] [<fffffe000075004c>] sd_setup_read_write_cmnd+0x44/0x844                         
[    6.878682] [<fffffe0000750884>] sd_init_command+0x38/0xb0                                   
[    6.884141] [<fffffe00007436e4>] scsi_setup_cmnd+0xd8/0x13c                                  
[    6.889686] [<fffffe000074397c>] scsi_prep_fn+0xc0/0x140                                     
[    6.894973] [<fffffe00003bd0cc>] blk_peek_request+0x148/0x24c                                
[    6.900692] [<fffffe0000744754>] scsi_request_fn+0x58/0x648                                  
[    6.906237] [<fffffe00003b8e80>] __blk_run_queue+0x40/0x58                                   
[    6.911696] [<fffffe00003b91a4>] blk_run_queue+0x30/0x48                                     
[    6.916983] [<fffffe00007422e4>] scsi_run_queue+0x204/0x294                                  
[    6.922528] [<fffffe000074321c>] scsi_end_request+0x13c/0x1a0                                
[    6.928247] [<fffffe0000744ef4>] scsi_io_completion+0xf0/0x564                               
[    6.934052] [<fffffe000073beb0>] scsi_finish_command+0xe4/0x144                              
[    6.939943] [<fffffe00007446cc>] scsi_softirq_done+0x148/0x178                               
[    6.945748] [<fffffe00003c39a4>] blk_done_softirq+0x7c/0x94                                  
[    6.951295] [<fffffe00000c8f44>] __do_softirq+0x114/0x2a0                                    
[    6.956667] [<fffffe00000c9340>] irq_exit+0x8c/0xe4                                          
[    6.961522] [<fffffe000009cf00>] handle_IPI+0x170/0x228                                      
[    6.966721] [<fffffe00000904e0>] gic_handle_irq+0xa0/0xb8                                    
[    6.972093] Exception stack(0xfffffe03dc143de0 to 0xfffffe03dc143f00)                        

> 
> cheers
> 
> 
> systemd[1]: Starting Uncomplicated firewall...
>          Starting Uncomplicated firewall...
> ------------[ cut here ]------------
> kernel BUG at drivers/scsi/scsi_lib.c:1096!
> Oops: Exception in kernel mode, sig: 5 [#1]
> SMP NR_CPUS=2048 NUMA PowerNV
> Modules linked in:
> CPU: 132 PID: 2699 Comm: kworker/132:1H Not tainted 4.4.0-rc1-54939-ge22a248-dirty #77
> Workqueue: kblockd cfq_kick_queue
> task: c000000fef147400 ti: c000000feb384000 task.ti: c000000feb384000
> NIP: c0000000005ab4a8 LR: c0000000005ab490 CTR: 0000000000000000
> REGS: c000000feb387620 TRAP: 0700   Not tainted  (4.4.0-rc1-54939-ge22a248-dirty)
> MSR: 9000000100029033 <SF,HV,EE,ME,IR,DR,RI,LE>  CR: 24002228  XER: 00000000
> CFAR: c000000000464950 SOFTE: 0 
> GPR00: c0000000005ab490 c000000feb3878a0 c000000000d77d00 0000000000000002 
> GPR04: c000000ff2030158 c000000ff47a0c00 0000000000000000 0000000000001000 
> GPR08: 0000000000000000 0000000000000001 0000000000000000 fffffffffffffff7 
> GPR12: 0000000000002200 c00000000fde5200 c0000000000c8098 c000000000b39858 
> GPR16: c000000000ae12c8 c000000000b39948 0000000000000000 0000000000000000 
> GPR20: c0000000005ab9c0 c0000017f5800144 0000000000000000 c000001e551a6850 
> GPR24: c0000017f5800140 c0000007efb9c800 0000000000000000 c000001e551a6800 
> GPR28: 0000000000000000 0000000000000000 c000000ff2030158 c000000feb420240 
> NIP [c0000000005ab4a8] scsi_init_sgtable+0xa8/0x180
> LR [c0000000005ab490] scsi_init_sgtable+0x90/0x180
> Call Trace:
> [c000000feb3878a0] [c0000000005ab490] scsi_init_sgtable+0x90/0x180 (unreliable)
> [c000000feb3878e0] [c0000000005ab5d4] scsi_init_io+0x54/0x160
> [c000000feb387930] [c0000000005fb43c] sd_init_command+0x6c/0xb00
> [c000000feb3879f0] [c0000000005ac2f8] scsi_setup_cmnd+0x108/0x1a0
> [c000000feb387a30] [c0000000005ac594] scsi_prep_fn+0x104/0x1c0
> [c000000feb387a70] [c00000000045d85c] blk_peek_request+0x20c/0x390
> [c000000feb387af0] [c0000000005ad6a8] scsi_request_fn+0xb8/0x7f0
> [c000000feb387bf0] [c0000000004583b4] __blk_run_queue+0x54/0x80
> [c000000feb387c20] [c00000000047e178] cfq_kick_queue+0x38/0xc0
> [c000000feb387c50] [c0000000000c00cc] process_one_work+0x2ac/0x560
> [c000000feb387ce0] [c0000000000c0510] worker_thread+0x190/0x660
> [c000000feb387d80] [c0000000000c8198] kthread+0x108/0x130
> [c000000feb387e30] [c0000000000095b0] ret_from_kernel_thread+0x5c/0xac
> Instruction dump:
> 2fa30000 409e00e4 e8bf0000 e87e0030 7fc4f378 4beb91ed 60000000 813f0008 
> 78630020 7f891840 7d301026 5529effe <0b090000> 907f0008 38600000 813e005c 
> ---[ end trace 3ea632a512ff48d5 ]---
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1272727

FromMichael Ellerman <mpe@ellerman.id.au>
Date2015-11-19 02:10 +0100
Message-ID<qwlTA-2sv-3@gated-at.bofh.it>
In reply to#1272203
On Wed, 2015-11-18 at 09:03 -0500, Mark Salter wrote:
> On Wed, 2015-11-18 at 20:18 +1100, Michael Ellerman wrote:
> > Hi folks,
> > 
> > I'm intermittently seeing the following oops on at least one powerpc box.
> > 
> > The BUG_ON() is from:
> > 
> > static int scsi_init_sgtable(struct request *req, struct scsi_data_buffer *sdb)
> > {
> > 	...
> > 	count = blk_rq_map_sg(req->q, req, sdb->table.sgl);
> > 	BUG_ON(count > sdb->table.nents);
> > 
> > Looking at the dump it looks like count was 2, I can't work out what nents was.
> > 
> > The machine's just a fairly boring bare metal setup, with a single IPR adapter:
> > 
> > 0001:08:00.0 RAID bus controller: IBM PCI-E IPR SAS Adapter (ASIC) (rev 02)
> > 	Subsystem: IBM PCIe3 x8 SAS RAID Internal Adapter 6Gb (57D7)
> > 	Flags: bus master, fast devsel, latency 0
> > 	Kernel driver in use: ipr
> > 
> > 
> > Anyone seen it before or have any ideas?
> 
> I'm also seeing it on arm64 in 4.4-rc1

Ah thanks, that's a good data point. I was assuming it was a driver bug, but I
assume you're not using IPR :)

> [    6.859003] Call trace:                                                                      
> [    6.861439] [<fffffe000074345c>] scsi_init_sgtable+0x84/0x88                                 
> [    6.867072] [<fffffe00007434ac>] scsi_init_io+0x4c/0x1ac                                     
> [    6.872358] [<fffffe000075004c>] sd_setup_read_write_cmnd+0x44/0x844                         
> [    6.878682] [<fffffe0000750884>] sd_init_command+0x38/0xb0                                   
> [    6.884141] [<fffffe00007436e4>] scsi_setup_cmnd+0xd8/0x13c                                  
> [    6.889686] [<fffffe000074397c>] scsi_prep_fn+0xc0/0x140                                     
> [    6.894973] [<fffffe00003bd0cc>] blk_peek_request+0x148/0x24c                                
> [    6.900692] [<fffffe0000744754>] scsi_request_fn+0x58/0x648                                  
> [    6.906237] [<fffffe00003b8e80>] __blk_run_queue+0x40/0x58                                   
> [    6.911696] [<fffffe00003b91a4>] blk_run_queue+0x30/0x48                                     
> [    6.916983] [<fffffe00007422e4>] scsi_run_queue+0x204/0x294                                  
> [    6.922528] [<fffffe000074321c>] scsi_end_request+0x13c/0x1a0                                
> [    6.928247] [<fffffe0000744ef4>] scsi_io_completion+0xf0/0x564                               
> [    6.934052] [<fffffe000073beb0>] scsi_finish_command+0xe4/0x144                              
> [    6.939943] [<fffffe00007446cc>] scsi_softirq_done+0x148/0x178                               
> [    6.945748] [<fffffe00003c39a4>] blk_done_softirq+0x7c/0x94                                  
> [    6.951295] [<fffffe00000c8f44>] __do_softirq+0x114/0x2a0                                    
> [    6.956667] [<fffffe00000c9340>] irq_exit+0x8c/0xe4                                          
> [    6.961522] [<fffffe000009cf00>] handle_IPI+0x170/0x228                                      
> [    6.966721] [<fffffe00000904e0>] gic_handle_irq+0xa0/0xb8                                    
> [    6.972093] Exception stack(0xfffffe03dc143de0 to 0xfffffe03dc143f00)                        

cheers

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1272928

FromChristoph Hellwig <hch@infradead.org>
Date2015-11-19 09:30 +0100
Message-ID<qwsLo-6Qc-9@gated-at.bofh.it>
In reply to#1272727
It's pretty much guaranteed a block layer bug, most likely in the
merge bios to request infrastucture where we don't obey the merging
limits properly.

Does either of you have a known good and first known bad kernel?
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1273224

FromHannes Reinecke <hare@suse.de>
Date2015-11-19 16:40 +0100
Message-ID<qwztx-2Iv-35@gated-at.bofh.it>
In reply to#1272928
On 11/19/2015 09:23 AM, Christoph Hellwig wrote:
> It's pretty much guaranteed a block layer bug, most likely in the
> merge bios to request infrastucture where we don't obey the merging
> limits properly.
> 
> Does either of you have a known good and first known bad kernel?

Well, I have been fighting a similar issue for several months now,
albeit with multipath enabled. Haven't had much progress with this,
sadly.
Seeing that this is our distro kernel it might or might not be
related; however, as the symptoms are identical there still is a
chance that this is actually a generic block-layer problem.

Cheers,

Hannes
-- 
Dr. Hannes Reinecke		               zSeries & Storage
hare@suse.de			               +49 911 74053 688
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: F. Imendörffer, J. Smithard, J. Guild, D. Upmanyu, G. Norton
HRB 21284 (AG Nürnberg)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1274117

FromEwan Milne <emilne@redhat.com>
Date2015-11-20 15:40 +0100
Message-ID<qwV0Z-8qz-11@gated-at.bofh.it>
In reply to#1273224
On Thu, 2015-11-19 at 16:35 +0100, Hannes Reinecke wrote:
> On 11/19/2015 09:23 AM, Christoph Hellwig wrote:
> > It's pretty much guaranteed a block layer bug, most likely in the
> > merge bios to request infrastucture where we don't obey the merging
> > limits properly.
> > 
> > Does either of you have a known good and first known bad kernel?
> 
> Well, I have been fighting a similar issue for several months now,
> albeit with multipath enabled. Haven't had much progress with this,
> sadly.
> Seeing that this is our distro kernel it might or might not be
> related; however, as the symptoms are identical there still is a
> chance that this is actually a generic block-layer problem.
> 
> Cheers,
> 
> Hannes

We have seen this also.  (e.g.  req->nr_phys_segments was 3, but
blk_rq_map_sg() returned 4.)  I was suspicious of the patch:

bio: modify __bio_add_page() to accept pages that don't start a new segment

But we put some debugging code in and didn't hit it.  We haven't
found the problem yet, either, though.  We're still looking.

As Christoph said, it would seem to be a problem with the block layer
merging.

The API for this seems defective, in that blk_rq_map_sg() should
never be returning a value indicating that it overwrote past the
end of the supplied SG array and depend on the caller to check it.
(We could get data corruption on another I/O if it used adjacent
memory for a different SG list, for example.)

-Ewan





--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1274160

FromHannes Reinecke <hare@suse.de>
Date2015-11-20 16:10 +0100
Message-ID<qwVu2-q6-25@gated-at.bofh.it>
In reply to#1274117
On 11/20/2015 03:38 PM, Ewan Milne wrote:
> On Thu, 2015-11-19 at 16:35 +0100, Hannes Reinecke wrote:
>> On 11/19/2015 09:23 AM, Christoph Hellwig wrote:
>>> It's pretty much guaranteed a block layer bug, most likely in the
>>> merge bios to request infrastucture where we don't obey the merging
>>> limits properly.
>>>
>>> Does either of you have a known good and first known bad kernel?
>>
>> Well, I have been fighting a similar issue for several months now,
>> albeit with multipath enabled. Haven't had much progress with this,
>> sadly.
>> Seeing that this is our distro kernel it might or might not be
>> related; however, as the symptoms are identical there still is a
>> chance that this is actually a generic block-layer problem.
>>
>> Cheers,
>>
>> Hannes
> 
> We have seen this also.  (e.g.  req->nr_phys_segments was 3, but
> blk_rq_map_sg() returned 4.)  I was suspicious of the patch:
> 
> bio: modify __bio_add_page() to accept pages that don't start a new segment
> 
> But we put some debugging code in and didn't hit it.  We haven't
> found the problem yet, either, though.  We're still looking.
> 
Can't we have a joint effort here?
I've been spending a _LOT_ of time trying to debug things here, but
none of the ideas I've come up with have been able to fix anything.

I'm almost tempted to increase the count from scsi_alloc_sgtable()
by one and be done with ...

> As Christoph said, it would seem to be a problem with the block layer
> merging.
> 
> The API for this seems defective, in that blk_rq_map_sg() should
> never be returning a value indicating that it overwrote past the
> end of the supplied SG array and depend on the caller to check it.
> (We could get data corruption on another I/O if it used adjacent
> memory for a different SG list, for example.)
> 
Yeah, the API is bloody useless.
By the time you hit the BUG_ON you've already caused a memory
corruption, so no way you can recover there.

At the very least we should be passing in the sg list count into
blk_map_rq_sg(), but that's a core blocklayer API and changes here
would require changes by quite a set of drivers. Plus it wouldn't
help me for a distribution kernel ...

Cheers,

Hannes
-- 
Dr. Hannes Reinecke		               zSeries & Storage
hare@suse.de			               +49 911 74053 688
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: F. Imendörffer, J. Smithard, J. Guild, D. Upmanyu, G. Norton
HRB 21284 (AG Nürnberg)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1274176

FromEwan Milne <emilne@redhat.com>
Date2015-11-20 16:30 +0100
Message-ID<qwVNn-xS-1@gated-at.bofh.it>
In reply to#1274160
On Fri, 2015-11-20 at 15:55 +0100, Hannes Reinecke wrote:
> Can't we have a joint effort here?
> I've been spending a _LOT_ of time trying to debug things here, but
> none of the ideas I've come up with have been able to fix anything.

Yes.  I'm not the one primarily looking at it, and we don't have a
reproducer in-house.  We just have the one dump right now.

> 
> I'm almost tempted to increase the count from scsi_alloc_sgtable()
> by one and be done with ...
> 

That might not fix it if it is a problem with the merge code, though.


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1275063

FromHannes Reinecke <hare@suse.de>
Date2015-11-23 08:00 +0100
Message-ID<qxTgt-6Hn-5@gated-at.bofh.it>
In reply to#1274176
On 11/20/2015 04:28 PM, Ewan Milne wrote:
> On Fri, 2015-11-20 at 15:55 +0100, Hannes Reinecke wrote:
>> Can't we have a joint effort here?
>> I've been spending a _LOT_ of time trying to debug things here, but
>> none of the ideas I've come up with have been able to fix anything.
>
> Yes.  I'm not the one primarily looking at it, and we don't have a
> reproducer in-house.  We just have the one dump right now.
>
Oh, I got plenty of them :-(

>>
>> I'm almost tempted to increase the count from scsi_alloc_sgtable()
>> by one and be done with ...
>>
>
> That might not fix it if it is a problem with the merge code, though.
>
Of course not. But it'll be a band-aid to keep the customer happy.

Cheers,

Hannes
-- 
Dr. Hannes Reinecke		               zSeries & Storage
hare@suse.de			               +49 911 74053 688
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: F. Imendörffer, J. Smithard, J. Guild, D. Upmanyu, G. Norton
HRB 21284 (AG Nürnberg)

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1277113

FromHannes Reinecke <hare@suse.de>
Date2015-11-25 10:10 +0100
Message-ID<qyEfo-48i-13@gated-at.bofh.it>
In reply to#1274176
On 11/20/2015 04:28 PM, Ewan Milne wrote:
> On Fri, 2015-11-20 at 15:55 +0100, Hannes Reinecke wrote:
>> Can't we have a joint effort here?
>> I've been spending a _LOT_ of time trying to debug things here, but
>> none of the ideas I've come up with have been able to fix anything.
> 
> Yes.  I'm not the one primarily looking at it, and we don't have a
> reproducer in-house.  We just have the one dump right now.
> 
>>
>> I'm almost tempted to increase the count from scsi_alloc_sgtable()
>> by one and be done with ...
>>
> 
> That might not fix it if it is a problem with the merge code, though.
> 
And indeed, it doesn't.
Seems I finally found the culprit.

What happens is this:
We have two paths, with these seg_boundary_masks:

path-1:    seg_boundary_mask = 65535,
path-2:    seg_boundary_mask = 4294967295,

consequently the DM request queue has this:

md-1:    seg_boundary_mask = 65535,

What happens now is that a request is being formatted, and sent
to path 2. During submission req->nr_phys_segments is formatted
with the limits of path 2, arriving at a count of 3.
Now the request gets retried on path 1, but as the NOMERGE request
flag is set req->nr_phys_segments is never updated.
But blk_rq_map_sg() ignores all counters, and just uses the
bi_vec directly, resulting in a count of 4 -> boom.

So the culprit here is the NOMERGE flag, which is evaluated
via
->dm_dispatch_request()
  ->blk_insert_cloned_request()
    ->blk_rq_check_limits()

If the above assessment is correct, the following patch should
fix it:

diff --git a/block/blk-core.c b/block/blk-core.c
index 801ced7..12cccd6 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -1928,7 +1928,7 @@ EXPORT_SYMBOL(submit_bio);
  */
 int blk_rq_check_limits(struct request_queue *q, struct request *rq)
 {
-       if (!rq_mergeable(rq))
+       if (rq->cmd_type != REQ_TYPE_FS)
                return 0;

        if (blk_rq_sectors(rq) > blk_queue_get_max_sectors(q,
rq->cmd_flags)) {


Mike? Jens?
Can you comment on it?

Cheers,

Hannes
-- 
Dr. Hannes Reinecke		               zSeries & Storage
hare@suse.de			               +49 911 74053 688
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: F. Imendörffer, J. Smithard, J. Guild, D. Upmanyu, G. Norton
HRB 21284 (AG Nürnberg)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1277657

FromJens Axboe <axboe@fb.com>
Date2015-11-25 19:00 +0100
Message-ID<qyMwi-12v-7@gated-at.bofh.it>
In reply to#1277113
On 11/25/2015 02:04 AM, Hannes Reinecke wrote:
> On 11/20/2015 04:28 PM, Ewan Milne wrote:
>> On Fri, 2015-11-20 at 15:55 +0100, Hannes Reinecke wrote:
>>> Can't we have a joint effort here?
>>> I've been spending a _LOT_ of time trying to debug things here, but
>>> none of the ideas I've come up with have been able to fix anything.
>>
>> Yes.  I'm not the one primarily looking at it, and we don't have a
>> reproducer in-house.  We just have the one dump right now.
>>
>>>
>>> I'm almost tempted to increase the count from scsi_alloc_sgtable()
>>> by one and be done with ...
>>>
>>
>> That might not fix it if it is a problem with the merge code, though.
>>
> And indeed, it doesn't.
> Seems I finally found the culprit.
>
> What happens is this:
> We have two paths, with these seg_boundary_masks:
>
> path-1:    seg_boundary_mask = 65535,
> path-2:    seg_boundary_mask = 4294967295,
>
> consequently the DM request queue has this:
>
> md-1:    seg_boundary_mask = 65535,
>
> What happens now is that a request is being formatted, and sent
> to path 2. During submission req->nr_phys_segments is formatted
> with the limits of path 2, arriving at a count of 3.
> Now the request gets retried on path 1, but as the NOMERGE request
> flag is set req->nr_phys_segments is never updated.
> But blk_rq_map_sg() ignores all counters, and just uses the
> bi_vec directly, resulting in a count of 4 -> boom.
>
> So the culprit here is the NOMERGE flag, which is evaluated
> via
> ->dm_dispatch_request()
>    ->blk_insert_cloned_request()
>      ->blk_rq_check_limits()
>
> If the above assessment is correct, the following patch should
> fix it:
>
> diff --git a/block/blk-core.c b/block/blk-core.c
> index 801ced7..12cccd6 100644
> --- a/block/blk-core.c
> +++ b/block/blk-core.c
> @@ -1928,7 +1928,7 @@ EXPORT_SYMBOL(submit_bio);
>    */
>   int blk_rq_check_limits(struct request_queue *q, struct request *rq)
>   {
> -       if (!rq_mergeable(rq))
> +       if (rq->cmd_type != REQ_TYPE_FS)
>                  return 0;
>
>          if (blk_rq_sectors(rq) > blk_queue_get_max_sectors(q,
> rq->cmd_flags)) {
>
>
> Mike? Jens?
> Can you comment on it?

We only support merging on REQ_TYPE_FS already, so how is the above 
making it any different? In general, NOMERGE being set or not should not 
make a difference. It's only a hint that we need not check further if we 
should be merging on this request, since we already tried it once, found 
we'd exceed various limits, then set NOMERGE to reflect that.


-- 
Jens Axboe

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1277713

FromHannes Reinecke <hare@suse.de>
Date2015-11-25 20:20 +0100
Message-ID<qyNLH-1Zv-13@gated-at.bofh.it>
In reply to#1277657
On 11/25/2015 06:56 PM, Jens Axboe wrote:
> On 11/25/2015 02:04 AM, Hannes Reinecke wrote:
>> On 11/20/2015 04:28 PM, Ewan Milne wrote:
>>> On Fri, 2015-11-20 at 15:55 +0100, Hannes Reinecke wrote:
>>>> Can't we have a joint effort here?
>>>> I've been spending a _LOT_ of time trying to debug things here, but
>>>> none of the ideas I've come up with have been able to fix anything.
>>>
>>> Yes.  I'm not the one primarily looking at it, and we don't have a
>>> reproducer in-house.  We just have the one dump right now.
>>>
>>>>
>>>> I'm almost tempted to increase the count from scsi_alloc_sgtable()
>>>> by one and be done with ...
>>>>
>>>
>>> That might not fix it if it is a problem with the merge code, though.
>>>
>> And indeed, it doesn't.
>> Seems I finally found the culprit.
>>
>> What happens is this:
>> We have two paths, with these seg_boundary_masks:
>>
>> path-1:    seg_boundary_mask = 65535,
>> path-2:    seg_boundary_mask = 4294967295,
>>
>> consequently the DM request queue has this:
>>
>> md-1:    seg_boundary_mask = 65535,
>>
>> What happens now is that a request is being formatted, and sent
>> to path 2. During submission req->nr_phys_segments is formatted
>> with the limits of path 2, arriving at a count of 3.
>> Now the request gets retried on path 1, but as the NOMERGE request
>> flag is set req->nr_phys_segments is never updated.
>> But blk_rq_map_sg() ignores all counters, and just uses the
>> bi_vec directly, resulting in a count of 4 -> boom.
>>
>> So the culprit here is the NOMERGE flag, which is evaluated
>> via
>> ->dm_dispatch_request()
>>    ->blk_insert_cloned_request()
>>      ->blk_rq_check_limits()
>>
>> If the above assessment is correct, the following patch should
>> fix it:
>>
>> diff --git a/block/blk-core.c b/block/blk-core.c
>> index 801ced7..12cccd6 100644
>> --- a/block/blk-core.c
>> +++ b/block/blk-core.c
>> @@ -1928,7 +1928,7 @@ EXPORT_SYMBOL(submit_bio);
>>    */
>>   int blk_rq_check_limits(struct request_queue *q, struct request *rq)
>>   {
>> -       if (!rq_mergeable(rq))
>> +       if (rq->cmd_type != REQ_TYPE_FS)
>>                  return 0;
>>
>>          if (blk_rq_sectors(rq) > blk_queue_get_max_sectors(q,
>> rq->cmd_flags)) {
>>
>>
>> Mike? Jens?
>> Can you comment on it?
>
> We only support merging on REQ_TYPE_FS already, so how is the above
> making it any different? In general, NOMERGE being set or not should not
> make a difference. It's only a hint that we need not check further if we
> should be merging on this request, since we already tried it once, found
> we'd exceed various limits, then set NOMERGE to reflect that.
>
The problem is that NOMERGE does too much, as it inhibits _any_ merging.

Unfortunately, the req->nr_phys_segments value is evaluated in the final
_driver_ context _after_ the merging happend; cf 
scsi_lib.c:scsi_init_sgtable().
As nr_phys_segments is inherited from the original request (and never
recalculated with the new request queue limits) the following 
blk_rq_map_sg() call might end up at a different calculation, especially
after retrying a request on another path.

Cheers,

Hannes
-- 
Dr. Hannes Reinecke		               zSeries & Storage
hare@suse.de			               +49 911 74053 688
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: F. Imendörffer, J. Smithard, J. Guild, D. Upmanyu, G. Norton
HRB 21284 (AG Nürnberg)

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1277723

FromJens Axboe <axboe@fb.com>
Date2015-11-25 20:30 +0100
Message-ID<qyNVn-22H-17@gated-at.bofh.it>
In reply to#1277713
On 11/25/2015 12:10 PM, Hannes Reinecke wrote:
> On 11/25/2015 06:56 PM, Jens Axboe wrote:
>> On 11/25/2015 02:04 AM, Hannes Reinecke wrote:
>>> On 11/20/2015 04:28 PM, Ewan Milne wrote:
>>>> On Fri, 2015-11-20 at 15:55 +0100, Hannes Reinecke wrote:
>>>>> Can't we have a joint effort here?
>>>>> I've been spending a _LOT_ of time trying to debug things here, but
>>>>> none of the ideas I've come up with have been able to fix anything.
>>>>
>>>> Yes.  I'm not the one primarily looking at it, and we don't have a
>>>> reproducer in-house.  We just have the one dump right now.
>>>>
>>>>>
>>>>> I'm almost tempted to increase the count from scsi_alloc_sgtable()
>>>>> by one and be done with ...
>>>>>
>>>>
>>>> That might not fix it if it is a problem with the merge code, though.
>>>>
>>> And indeed, it doesn't.
>>> Seems I finally found the culprit.
>>>
>>> What happens is this:
>>> We have two paths, with these seg_boundary_masks:
>>>
>>> path-1:    seg_boundary_mask = 65535,
>>> path-2:    seg_boundary_mask = 4294967295,
>>>
>>> consequently the DM request queue has this:
>>>
>>> md-1:    seg_boundary_mask = 65535,
>>>
>>> What happens now is that a request is being formatted, and sent
>>> to path 2. During submission req->nr_phys_segments is formatted
>>> with the limits of path 2, arriving at a count of 3.
>>> Now the request gets retried on path 1, but as the NOMERGE request
>>> flag is set req->nr_phys_segments is never updated.
>>> But blk_rq_map_sg() ignores all counters, and just uses the
>>> bi_vec directly, resulting in a count of 4 -> boom.
>>>
>>> So the culprit here is the NOMERGE flag, which is evaluated
>>> via
>>> ->dm_dispatch_request()
>>>    ->blk_insert_cloned_request()
>>>      ->blk_rq_check_limits()
>>>
>>> If the above assessment is correct, the following patch should
>>> fix it:
>>>
>>> diff --git a/block/blk-core.c b/block/blk-core.c
>>> index 801ced7..12cccd6 100644
>>> --- a/block/blk-core.c
>>> +++ b/block/blk-core.c
>>> @@ -1928,7 +1928,7 @@ EXPORT_SYMBOL(submit_bio);
>>>    */
>>>   int blk_rq_check_limits(struct request_queue *q, struct request *rq)
>>>   {
>>> -       if (!rq_mergeable(rq))
>>> +       if (rq->cmd_type != REQ_TYPE_FS)
>>>                  return 0;
>>>
>>>          if (blk_rq_sectors(rq) > blk_queue_get_max_sectors(q,
>>> rq->cmd_flags)) {
>>>
>>>
>>> Mike? Jens?
>>> Can you comment on it?
>>
>> We only support merging on REQ_TYPE_FS already, so how is the above
>> making it any different? In general, NOMERGE being set or not should not
>> make a difference. It's only a hint that we need not check further if we
>> should be merging on this request, since we already tried it once, found
>> we'd exceed various limits, then set NOMERGE to reflect that.
>>
> The problem is that NOMERGE does too much, as it inhibits _any_ merging.

Right, that is the point of the flag from the block layer view, where it 
was originally added for the case mentioned.

> Unfortunately, the req->nr_phys_segments value is evaluated in the final
> _driver_ context _after_ the merging happend; cf
> scsi_lib.c:scsi_init_sgtable().
> As nr_phys_segments is inherited from the original request (and never
> recalculated with the new request queue limits) the following
> blk_rq_map_sg() call might end up at a different calculation, especially
> after retrying a request on another path.

That all sounds pretty horrible. Why is blk_rq_check_limits() checking 
for mergeable at all? If merging is disabled on the request, I'm 
assuming that's an attempt at an optimization since we know it won't 
change. But that should be tracked separately, like how it's done on the 
bio.


-- 
Jens Axboe

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1277779

FromMike Snitzer <snitzer@redhat.com>
Date2015-11-25 21:30 +0100
Message-ID<qyORt-2F7-33@gated-at.bofh.it>
In reply to#1277723
On Wed, Nov 25 2015 at  2:24pm -0500,
Jens Axboe <axboe@fb.com> wrote:

> On 11/25/2015 12:10 PM, Hannes Reinecke wrote:
> >On 11/25/2015 06:56 PM, Jens Axboe wrote:
> >>On 11/25/2015 02:04 AM, Hannes Reinecke wrote:
> >>>On 11/20/2015 04:28 PM, Ewan Milne wrote:
> >>>>On Fri, 2015-11-20 at 15:55 +0100, Hannes Reinecke wrote:
> >>>>>Can't we have a joint effort here?
> >>>>>I've been spending a _LOT_ of time trying to debug things here, but
> >>>>>none of the ideas I've come up with have been able to fix anything.
> >>>>
> >>>>Yes.  I'm not the one primarily looking at it, and we don't have a
> >>>>reproducer in-house.  We just have the one dump right now.
> >>>>
> >>>>>
> >>>>>I'm almost tempted to increase the count from scsi_alloc_sgtable()
> >>>>>by one and be done with ...
> >>>>>
> >>>>
> >>>>That might not fix it if it is a problem with the merge code, though.
> >>>>
> >>>And indeed, it doesn't.
> >>>Seems I finally found the culprit.
> >>>
> >>>What happens is this:
> >>>We have two paths, with these seg_boundary_masks:
> >>>
> >>>path-1:    seg_boundary_mask = 65535,
> >>>path-2:    seg_boundary_mask = 4294967295,
> >>>
> >>>consequently the DM request queue has this:
> >>>
> >>>md-1:    seg_boundary_mask = 65535,
> >>>
> >>>What happens now is that a request is being formatted, and sent
> >>>to path 2. During submission req->nr_phys_segments is formatted
> >>>with the limits of path 2, arriving at a count of 3.
> >>>Now the request gets retried on path 1, but as the NOMERGE request
> >>>flag is set req->nr_phys_segments is never updated.
> >>>But blk_rq_map_sg() ignores all counters, and just uses the
> >>>bi_vec directly, resulting in a count of 4 -> boom.
> >>>
> >>>So the culprit here is the NOMERGE flag, which is evaluated
> >>>via
> >>>->dm_dispatch_request()
> >>>   ->blk_insert_cloned_request()
> >>>     ->blk_rq_check_limits()
> >>>
> >>>If the above assessment is correct, the following patch should
> >>>fix it:
> >>>
> >>>diff --git a/block/blk-core.c b/block/blk-core.c
> >>>index 801ced7..12cccd6 100644
> >>>--- a/block/blk-core.c
> >>>+++ b/block/blk-core.c
> >>>@@ -1928,7 +1928,7 @@ EXPORT_SYMBOL(submit_bio);
> >>>   */
> >>>  int blk_rq_check_limits(struct request_queue *q, struct request *rq)
> >>>  {
> >>>-       if (!rq_mergeable(rq))
> >>>+       if (rq->cmd_type != REQ_TYPE_FS)
> >>>                 return 0;
> >>>
> >>>         if (blk_rq_sectors(rq) > blk_queue_get_max_sectors(q,
> >>>rq->cmd_flags)) {
> >>>
> >>>
> >>>Mike? Jens?
> >>>Can you comment on it?
> >>
> >>We only support merging on REQ_TYPE_FS already, so how is the above
> >>making it any different? In general, NOMERGE being set or not should not
> >>make a difference. It's only a hint that we need not check further if we
> >>should be merging on this request, since we already tried it once, found
> >>we'd exceed various limits, then set NOMERGE to reflect that.
> >>
> >The problem is that NOMERGE does too much, as it inhibits _any_ merging.
> 
> Right, that is the point of the flag from the block layer view,
> where it was originally added for the case mentioned.

And we really don't want _any_ merging.  The merging, if any, will have
already happened in upper DM-multipath's elevator.  So there should be
no need to have the underlying SCSI paths do any merging.

> >Unfortunately, the req->nr_phys_segments value is evaluated in the final
> >_driver_ context _after_ the merging happend; cf
> >scsi_lib.c:scsi_init_sgtable().
> >As nr_phys_segments is inherited from the original request (and never
> >recalculated with the new request queue limits) the following
> >blk_rq_map_sg() call might end up at a different calculation, especially
> >after retrying a request on another path.
> 
> That all sounds pretty horrible. Why is blk_rq_check_limits()
> checking for mergeable at all? If merging is disabled on the
> request, I'm assuming that's an attempt at an optimization since we
> know it won't change. But that should be tracked separately, like
> how it's done on the bio.

Not clear to me why it was checking for merging...
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1277810

FromMike Snitzer <snitzer@redhat.com>
Date2015-11-25 22:30 +0100
Message-ID<qyPNw-3gO-9@gated-at.bofh.it>
In reply to#1277779
On Wed, Nov 25 2015 at  3:23pm -0500,
Mike Snitzer <snitzer@redhat.com> wrote:

> On Wed, Nov 25 2015 at  2:24pm -0500,
> Jens Axboe <axboe@fb.com> wrote:
> 
> > On 11/25/2015 12:10 PM, Hannes Reinecke wrote:
> > >The problem is that NOMERGE does too much, as it inhibits _any_ merging.
> > 
> > Right, that is the point of the flag from the block layer view,
> > where it was originally added for the case mentioned.
> 
> And we really don't want _any_ merging.  The merging, if any, will have
> already happened in upper DM-multipath's elevator.  So there should be
> no need to have the underlying SCSI paths do any merging.
> 
> > >Unfortunately, the req->nr_phys_segments value is evaluated in the final
> > >_driver_ context _after_ the merging happend; cf
> > >scsi_lib.c:scsi_init_sgtable().
> > >As nr_phys_segments is inherited from the original request (and never
> > >recalculated with the new request queue limits) the following
> > >blk_rq_map_sg() call might end up at a different calculation, especially
> > >after retrying a request on another path.
> > 
> > That all sounds pretty horrible. Why is blk_rq_check_limits()
> > checking for mergeable at all? If merging is disabled on the
> > request, I'm assuming that's an attempt at an optimization since we
> > know it won't change. But that should be tracked separately, like
> > how it's done on the bio.
> 
> Not clear to me why it was checking for merging...

Ewan pointed out that blk_rq_check_limits()'s call to rq_mergable() was
introduced as part of Martin's DISCARD cleanup that prepared for
WRITE_SAME, see: commit e2a60da74 ("block: Clean up special command handling logic")

And prior to that, blk_rq_check_limits()'s (rq->cmd_flags & REQ_DISCARD)
check was introduced by some guy crazy doppelganger name "Ike Snitzer",
see commit: 3383977f ("block: update request stacking methods to support discards")

So basically we probably never needed the extra check in
blk_rq_check_limits() to begin with.. Ike was probably paranoid. ;)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1277666

FromMike Snitzer <snitzer@redhat.com>
Date2015-11-25 19:10 +0100
Message-ID<qyMFX-1lB-1@gated-at.bofh.it>
In reply to#1277113
On Wed, Nov 25 2015 at  4:04am -0500,
Hannes Reinecke <hare@suse.de> wrote:

> On 11/20/2015 04:28 PM, Ewan Milne wrote:
> > On Fri, 2015-11-20 at 15:55 +0100, Hannes Reinecke wrote:
> >> Can't we have a joint effort here?
> >> I've been spending a _LOT_ of time trying to debug things here, but
> >> none of the ideas I've come up with have been able to fix anything.
> > 
> > Yes.  I'm not the one primarily looking at it, and we don't have a
> > reproducer in-house.  We just have the one dump right now.
> > 
> >>
> >> I'm almost tempted to increase the count from scsi_alloc_sgtable()
> >> by one and be done with ...
> >>
> > 
> > That might not fix it if it is a problem with the merge code, though.
> > 
> And indeed, it doesn't.

How did you arrive at that?  Do you have a reproducer now?

> Seems I finally found the culprit.
> 
> What happens is this:
> We have two paths, with these seg_boundary_masks:
> 
> path-1:    seg_boundary_mask = 65535,
> path-2:    seg_boundary_mask = 4294967295,
> 
> consequently the DM request queue has this:
> 
> md-1:    seg_boundary_mask = 65535,
> 
> What happens now is that a request is being formatted, and sent
> to path 2. During submission req->nr_phys_segments is formatted
> with the limits of path 2, arriving at a count of 3.
> Now the request gets retried on path 1, but as the NOMERGE request
> flag is set req->nr_phys_segments is never updated.
> But blk_rq_map_sg() ignores all counters, and just uses the
> bi_vec directly, resulting in a count of 4 -> boom.
> 
> So the culprit here is the NOMERGE flag,

NOMERGE is always set in __blk_rq_prep_clone() for cloned requests.

> which is evaluated via
> ->dm_dispatch_request()
>   ->blk_insert_cloned_request()
>     ->blk_rq_check_limits()

blk_insert_cloned_request() is the only caller of blk_rq_check_limits();
anyway after reading your mail I'm still left wondering if your proposed
patch is correct.

> If the above assessment is correct, the following patch should
> fix it:
> 
> diff --git a/block/blk-core.c b/block/blk-core.c
> index 801ced7..12cccd6 100644
> --- a/block/blk-core.c
> +++ b/block/blk-core.c
> @@ -1928,7 +1928,7 @@ EXPORT_SYMBOL(submit_bio);
>   */
>  int blk_rq_check_limits(struct request_queue *q, struct request *rq)
>  {
> -       if (!rq_mergeable(rq))
> +       if (rq->cmd_type != REQ_TYPE_FS)
>                 return 0;
> 
>         if (blk_rq_sectors(rq) > blk_queue_get_max_sectors(q,
> rq->cmd_flags)) {
> 
> 
> Mike? Jens?
> Can you comment on it?

You're not explaining the actual change in the patch very well; I think
you're correct but you're leaving the justification as an exercise to 
the reviewer:

blk_rq_check_limits() will call blk_recalc_rq_segments() after the
!rq_mergeable() check but you're saying for this case in question we
never get there -- due to the cloned request having NOMERGE set.

So in blk_rq_check_limits() you've unrolled rq_mergeable() and
open-coded the lone remaining check (rq->cmd_type != REQ_TYPE_FS)

I agree that the (rq->cmd_flags & REQ_NOMERGE_FLAGS) check in
the blk_insert_cloned_request() call-chain (via rq_mergeable()) makes no
sense for cloned requests that always have NOMERGE set.

So you're saying that by having blk_rq_check_limits() go on to call
blk_recalc_rq_segments() this bug will be fixed?

BTW, I think blk_rq_check_limits()'s export should be removed and the
function made static and renamed to blk_clone_rq_check_limits(), again:
blk_insert_cloned_request() is the only caller of blk_rq_check_limits()

Seems prudent to make that change now to be clear that this code is only
used by cloned requests.

Thanks,
Mike
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1277702

FromHannes Reinecke <hare@suse.de>
Date2015-11-25 20:10 +0100
Message-ID<qyNC1-1Wc-5@gated-at.bofh.it>
In reply to#1277666
On 11/25/2015 07:01 PM, Mike Snitzer wrote:
> On Wed, Nov 25 2015 at  4:04am -0500,
> Hannes Reinecke <hare@suse.de> wrote:
>
>> On 11/20/2015 04:28 PM, Ewan Milne wrote:
>>> On Fri, 2015-11-20 at 15:55 +0100, Hannes Reinecke wrote:
>>>> Can't we have a joint effort here?
>>>> I've been spending a _LOT_ of time trying to debug things here, but
>>>> none of the ideas I've come up with have been able to fix anything.
>>>
>>> Yes.  I'm not the one primarily looking at it, and we don't have a
>>> reproducer in-house.  We just have the one dump right now.
>>>
>>>>
>>>> I'm almost tempted to increase the count from scsi_alloc_sgtable()
>>>> by one and be done with ...
>>>>
>>>
>>> That might not fix it if it is a problem with the merge code, though.
>>>
>> And indeed, it doesn't.
>
> How did you arrive at that?  Do you have a reproducer now?
>
Not a reproducer, but several dumps for analysis.

>> Seems I finally found the culprit.
>>
>> What happens is this:
>> We have two paths, with these seg_boundary_masks:
>>
>> path-1:    seg_boundary_mask = 65535,
>> path-2:    seg_boundary_mask = 4294967295,
>>
>> consequently the DM request queue has this:
>>
>> md-1:    seg_boundary_mask = 65535,
>>
>> What happens now is that a request is being formatted, and sent
>> to path 2. During submission req->nr_phys_segments is formatted
>> with the limits of path 2, arriving at a count of 3.
>> Now the request gets retried on path 1, but as the NOMERGE request
>> flag is set req->nr_phys_segments is never updated.
>> But blk_rq_map_sg() ignores all counters, and just uses the
>> bi_vec directly, resulting in a count of 4 -> boom.
>>
>> So the culprit here is the NOMERGE flag,
>
> NOMERGE is always set in __blk_rq_prep_clone() for cloned requests.
>
Yes.

>> which is evaluated via
>> ->dm_dispatch_request()
>>    ->blk_insert_cloned_request()
>>      ->blk_rq_check_limits()
>
> blk_insert_cloned_request() is the only caller of blk_rq_check_limits();
> anyway after reading your mail I'm still left wondering if your proposed
> patch is correct.
>
>> If the above assessment is correct, the following patch should
>> fix it:
>>
>> diff --git a/block/blk-core.c b/block/blk-core.c
>> index 801ced7..12cccd6 100644
>> --- a/block/blk-core.c
>> +++ b/block/blk-core.c
>> @@ -1928,7 +1928,7 @@ EXPORT_SYMBOL(submit_bio);
>>    */
>>   int blk_rq_check_limits(struct request_queue *q, struct request *rq)
>>   {
>> -       if (!rq_mergeable(rq))
>> +       if (rq->cmd_type != REQ_TYPE_FS)
>>                  return 0;
>>
>>          if (blk_rq_sectors(rq) > blk_queue_get_max_sectors(q,
>> rq->cmd_flags)) {
>>
>>
>> Mike? Jens?
>> Can you comment on it?
>
> You're not explaining the actual change in the patch very well; I think
> you're correct but you're leaving the justification as an exercise to
> the reviewer:
>
> blk_rq_check_limits() will call blk_recalc_rq_segments() after the
> !rq_mergeable() check but you're saying for this case in question we
> never get there -- due to the cloned request having NOMERGE set.
>
> So in blk_rq_check_limits() you've unrolled rq_mergeable() and
> open-coded the lone remaining check (rq->cmd_type != REQ_TYPE_FS)
>
> I agree that the (rq->cmd_flags & REQ_NOMERGE_FLAGS) check in
> the blk_insert_cloned_request() call-chain (via rq_mergeable()) makes no
> sense for cloned requests that always have NOMERGE set.
>
> So you're saying that by having blk_rq_check_limits() go on to call
> blk_recalc_rq_segments() this bug will be fixed?
>
That is the idea.

I've already established that in all instances I have seen so far
req->nr_phys_segments is _less_ than req->bio->bi_phys_segments.

As it turns out, req->nr_phys_segemnts _would_ have been updated in
blk_rq_check_limits(), but isn't due to the NOMERGE flag being set
for the cloned request.
So each cloned request inherits the values from the original request,
despite the fact that req->nr_phys_segments _has_ to be evaluated in
the final request_queue context, as the queue limits _might_ be 
different from the original (merged) queue limits of the multipath
request queue.

> BTW, I think blk_rq_check_limits()'s export should be removed and the
> function made static and renamed to blk_clone_rq_check_limits(), again:
> blk_insert_cloned_request() is the only caller of blk_rq_check_limits()
>
Actually, seeing Jens' last comment the check for REQ_TYPE_FS is 
pointless, too, so we might as well remove the entire if-clause.

> Seems prudent to make that change now to be clear that this code is only
> used by cloned requests.
>
Yeah, that would make sense. I'll be preparing a patch.
With a more detailed description :-)

Cheers,

Hannes

-- 
Dr. Hannes Reinecke		               zSeries & Storage
hare@suse.de			               +49 911 74053 688
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: F. Imendörffer, J. Smithard, J. Guild, D. Upmanyu, G. Norton
HRB 21284 (AG Nürnberg)

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1274020

FromMichael Ellerman <mpe@ellerman.id.au>
Date2015-11-20 13:20 +0100
Message-ID<qwSPw-76t-19@gated-at.bofh.it>
In reply to#1272928
On Thu, 2015-11-19 at 00:23 -0800, Christoph Hellwig wrote:

> It's pretty much guaranteed a block layer bug, most likely in the
> merge bios to request infrastucture where we don't obey the merging
> limits properly.
> 
> Does either of you have a known good and first known bad kernel?

Not me, I've only hit it one or two times. All I can say is I have hit it in
4.4-rc1.

Laurent, can you narrow it down at all?

cheers

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1274054

FromLaurent Dufour <ldufour@linux.vnet.ibm.com>
Date2015-11-20 14:00 +0100
Message-ID<qwTse-7ka-9@gated-at.bofh.it>
In reply to#1274020
On 20/11/2015 13:10, Michael Ellerman wrote:
> On Thu, 2015-11-19 at 00:23 -0800, Christoph Hellwig wrote:
> 
>> It's pretty much guaranteed a block layer bug, most likely in the
>> merge bios to request infrastucture where we don't obey the merging
>> limits properly.
>>
>> Does either of you have a known good and first known bad kernel?
> 
> Not me, I've only hit it one or two times. All I can say is I have hit it in
> 4.4-rc1.
> 
> Laurent, can you narrow it down at all?

I'm trying bisect it but it's very time consuming, and I think I missed
some bad kernel because I didn't let enough time to the system to panic.

But I'm still on...

Cheers

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1274068

FromMark Salter <msalter@redhat.com>
Date2015-11-20 14:40 +0100
Message-ID<qwU4W-7NT-11@gated-at.bofh.it>
In reply to#1274054
On Fri, 2015-11-20 at 13:56 +0100, Laurent Dufour wrote:
> On 20/11/2015 13:10, Michael Ellerman wrote:
> > On Thu, 2015-11-19 at 00:23 -0800, Christoph Hellwig wrote:
> > 
> > > It's pretty much guaranteed a block layer bug, most likely in the
> > > merge bios to request infrastucture where we don't obey the merging
> > > limits properly.
> > > 
> > > Does either of you have a known good and first known bad kernel?
> > 
> > Not me, I've only hit it one or two times. All I can say is I have hit it in
> > 4.4-rc1.
> > 
> > Laurent, can you narrow it down at all?
> 
> I'm trying bisect it but it's very time consuming, and I think I missed
> some bad kernel because I didn't let enough time to the system to panic.
> 
> But I'm still on...
> 
I usually hit it at boot time. Maybe one out of five times. I'll take a stab
at bisecting today...

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

Page 1 of 2  [1] 2  Next page →

Back to top | Article view | linux.kernel

csiph-web