linux-kernel - Re: [PATCH] scsi/mesh: Fix panic after host or bus reset

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-Id: <159564519423.31464.15591259959982994104.b4-ty@oracle.com>
Date:   Fri, 24 Jul 2020 22:50:35 -0400
From:   "Martin K. Petersen" <martin.petersen@...cle.com>
To:     "James E.J. Bottomley" <jejb@...ux.ibm.com>,
        Finn Thain <fthain@...egraphics.com.au>
Cc:     "Martin K . Petersen" <martin.petersen@...cle.com>,
        Paul Mackerras <paulus@...abs.org>,
        linux-kernel@...r.kernel.org, linux-scsi@...r.kernel.org
Subject: Re: [PATCH] scsi/mesh: Fix panic after host or bus reset

On Thu, 23 Jul 2020 09:25:51 +1000, Finn Thain wrote:

> Booting Linux with a Conner CP3200 drive attached to the MESH SCSI bus
> results in EH measures and a panic:
> 
> [   25.499838] mesh: configured for synchronous 5 MB/s
> [   25.787154] mesh: performing initial bus reset...
> [   29.867115] scsi host0: MESH
> [   29.929527] mesh: target 0 synchronous at 3.6 MB/s
> [   29.998763] scsi 0:0:0:0: Direct-Access     CONNER   CP3200-200mb-3.5 4040 PQ: 0 ANSI: 1 CCS
> [   31.989975] sd 0:0:0:0: [sda] 415872 512-byte logical blocks: (213 MB/203 MiB)
> [   32.070975] sd 0:0:0:0: [sda] Write Protect is off
> [   32.137197] sd 0:0:0:0: [sda] Mode Sense: 5b 00 00 08
> [   32.209661] sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
> [   32.332708]  sda: [mac] sda1 sda2 sda3
> [   32.417733] sd 0:0:0:0: [sda] Attached SCSI disk
> ... snip ...
> [   76.687067] mesh_abort((ptrval))
> [   76.743606] mesh: state at (ptrval), regs at (ptrval), dma at (ptrval)
> [   76.810798]     ct=6000 seq=86 bs=4017 fc= 0 exc= 0 err= 0 im= 7 int= 0 sp=85
> [   76.880720]     dma stat=84e0 cmdptr=1f73d000
> [   76.941387]     phase=4 msgphase=0 conn_tgt=0 data_ptr=24576
> [   77.005567]     dma_st=1 dma_ct=0 n_msgout=0
> [   77.065456]     target 0: req=(ptrval) goes_out=0 saved_ptr=0
> [   77.130512] mesh_abort((ptrval))
> [   77.187670] mesh: state at (ptrval), regs at (ptrval), dma at (ptrval)
> [   77.255594]     ct=6000 seq=86 bs=4017 fc= 0 exc= 0 err= 0 im= 7 int= 0 sp=85
> [   77.325778]     dma stat=84e0 cmdptr=1f73d000
> [   77.387239]     phase=4 msgphase=0 conn_tgt=0 data_ptr=24576
> [   77.453665]     dma_st=1 dma_ct=0 n_msgout=0
> [   77.515900]     target 0: req=(ptrval) goes_out=0 saved_ptr=0
> [   77.582902] mesh_host_reset
> [   88.187083] Kernel panic - not syncing: mesh: double DMA start !
> [   88.254510] CPU: 0 PID: 358 Comm: scsi_eh_0 Not tainted 5.6.13-pmac #1
> [   88.323302] Call Trace:
> [   88.378854] [e16ddc58] [c0027080] panic+0x13c/0x308 (unreliable)
> [   88.446221] [e16ddcb8] [c02b2478] mesh_start.part.12+0x130/0x414
> [   88.513298] [e16ddcf8] [c02b2fc8] mesh_queue+0x54/0x70
> [   88.577097] [e16ddd18] [c02a1848] scsi_send_eh_cmnd+0x374/0x384
> [   88.643476] [e16dddc8] [c02a1938] scsi_eh_tur+0x5c/0xb8
> [   88.707878] [e16dddf8] [c02a1ab8] scsi_eh_test_devices+0x124/0x178
> [   88.775663] [e16dde28] [c02a2094] scsi_eh_ready_devs+0x588/0x8a8
> [   88.843124] [e16dde98] [c02a31d8] scsi_error_handler+0x344/0x520
> [   88.910697] [e16ddf08] [c00409c8] kthread+0xe4/0xe8
> [   88.975166] [e16ddf38] [c000f234] ret_from_kernel_thread+0x14/0x1c
> [   89.044112] Rebooting in 180 seconds..
> 
> [...]

Applied to 5.9/scsi-queue, thanks!

[1/1] scsi: mesh: Fix panic after host or bus reset
      https://git.kernel.org/mkp/scsi/c/edd7dd2292ab

-- 
Martin K. Petersen	Oracle Linux Engineering