• James Bottomley's avatar
    [SCSI] fix WARNING: at drivers/scsi/scsi_lib.c:1704 · 4e6c82b3
    James Bottomley authored
    On Mon, 2011-11-07 at 17:24 +1100, Stephen Rothwell wrote:
    > Hi all,
    >
    > Starting some time last week I am getting the following during boot on
    > our PPC970 blade:
    >
    > calling  .ipr_init+0x0/0x68 @ 1
    > ipr: IBM Power RAID SCSI Device Driver version: 2.5.2 (April 27, 2011)
    > ipr 0000:01:01.0: Found IOA with IRQ: 26
    > ipr 0000:01:01.0: Starting IOA initialization sequence.
    > ipr 0000:01:01.0: Adapter firmware version: 06160039
    > ipr 0000:01:01.0: IOA initialized.
    > scsi0 : IBM 572E Storage Adapter
    > ------------[ cut here ]------------
    > WARNING: at drivers/scsi/scsi_lib.c:1704
    > Modules linked in:
    > NIP: c00000000053b3d4 LR: c00000000053e5b0 CTR: c000000000541d70
    > REGS: c0000000783c2f60 TRAP: 0700   Not tainted  (3.1.0-autokern1)
    > MSR: 8000000000029032 <EE,ME,CE,IR,DR>  CR: 24002024  XER: 20000002
    > TASK = c0000000783b8000[1] 'swapper' THREAD: c0000000783c0000 CPU: 0
    > GPR00: 0000000000000001 c0000000783c31e0 c000000000cf38b0 c00000000239a9d0
    > GPR04: c000000000cbe8f8 0000000000000000 c0000000783c3040 0000000000000000
    > GPR08: c000000075daf488 c000000078a3b7ff c000000000bcacc8 0000000000000000
    > GPR12: 0000000044002028 c000000007ffb000 0000000002e40000 000000000099b800
    > GPR16: 0000000000000000 c000000000bba5fc c000000000a61db8 0000000000000000
    > GPR20: 0000000001b77200 0000000000000000 c000000078990000 0000000000000001
    > GPR24: c000000002396828 0000000000000000 0000000000000000 c000000078a3b938
    > GPR28: fffffffffffffffa c0000000008ad2c0 c000000000c7faa8 c00000000239a9d0
    > NIP [c00000000053b3d4] .scsi_free_queue+0x24/0x90
    > LR [c00000000053e5b0] .scsi_alloc_sdev+0x280/0x2e0
    > Call Trace:
    > [c0000000783c31e0] [c000000000c7faa8] wireless_seq_fops+0x278d0/0x2eb88 (unreliable)
    > [c0000000783c3270] [c00000000053e5b0] .scsi_alloc_sdev+0x280/0x2e0
    > [c0000000783c3330] [c00000000053eba0] .scsi_probe_and_add_lun+0x390/0xb40
    > [c0000000783c34a0] [c00000000053f7ec] .__scsi_scan_target+0x16c/0x650
    > [c0000000783c35f0] [c00000000053fd90] .scsi_scan_channel+0xc0/0x100
    > [c0000000783c36a0] [c00000000053fefc] .scsi_scan_host_selected+0x12c/0x1c0
    > [c0000000783c3750] [c00000000083dcb4] .ipr_probe+0x2c0/0x390
    > [c0000000783c3830] [c0000000003f50b4] .local_pci_probe+0x34/0x50
    > [c0000000783c38a0] [c0000000003f5f78] .pci_device_probe+0x148/0x150
    > [c0000000783c3950] [c0000000004e1e8c] .driver_probe_device+0xdc/0x210
    > [c0000000783c39f0] [c0000000004e20cc] .__driver_attach+0x10c/0x110
    > [c0000000783c3a80] [c0000000004e1228] .bus_for_each_dev+0x98/0xf0
    > [c0000000783c3b30] [c0000000004e1bf8] .driver_attach+0x28/0x40
    > [c0000000783c3bb0] [c0000000004e07d8] .bus_add_driver+0x218/0x340
    > [c0000000783c3c60] [c0000000004e2a2c] .driver_register+0x9c/0x1b0
    > [c0000000783c3d00] [c0000000003f62d4] .__pci_register_driver+0x64/0x140
    > [c0000000783c3da0] [c000000000b99f88] .ipr_init+0x4c/0x68
    > [c0000000783c3e20] [c00000000000ad24] .do_one_initcall+0x1a4/0x1e0
    > [c0000000783c3ee0] [c000000000b512d0] .kernel_init+0x14c/0x1fc
    > [c0000000783c3f90] [c000000000022468] .kernel_thread+0x54/0x70
    > Instruction dump:
    > ebe1fff8 7c0803a6 4e800020 7c0802a6 fba1ffe8 fbe1fff8 7c7f1b78 f8010010
    > f821ff71 e8030398 3120ffff 7c090110 <0b000000> e86303b0 482de065 60000000
    > ---[ end trace 759bed76a85e8dec ]---
    > scsi 0:0:1:0: Direct-Access     IBM-ESXS MAY2036RC        T106 PQ: 0 ANSI: 5
    > ------------[ cut here ]------------
    >
    > I get lots more of these.  The obvious commit to point the finger at
    > is 3308511c ("[SCSI] Make scsi_free_queue() kill pending SCSI
    > commands") but the root cause may be something different.
    
    Caused by
    
    commit f7c9c6bb
    Author: Anton Blanchard <anton@samba.org>
    Date:   Thu Nov 3 08:56:22 2011 +1100
    
        [SCSI] Fix block queue and elevator memory leak in scsi_alloc_sdev
    
    Doesn't completely do the teardown.  The true fix is to do a proper
    teardown instead of hand rolling it
    Reported-by: default avatarStephen Rothwell <sfr@canb.auug.org.au>
    Tested-by: default avatarStephen Rothwell <sfr@canb.auug.org.au>
    Cc: stable@kernel.org	#2.6.38+
    Signed-off-by: default avatarJames Bottomley <JBottomley@Parallels.com>
    4e6c82b3
scsi_scan.c 54.2 KB