From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mike Christie Subject: Re: [PATCH, resend] Defer processing of REQ_PREEMPT requests for blocked devices Date: Mon, 06 Apr 2015 10:37:58 -0500 Message-ID: <5522A856.5000105@cs.wisc.edu> References: <54F6D103.9040709@sandisk.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Return-path: Received: from sabe.cs.wisc.edu ([128.105.6.20]:44991 "EHLO sabe.cs.wisc.edu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752891AbbDFPiU (ORCPT ); Mon, 6 Apr 2015 11:38:20 -0400 In-Reply-To: <54F6D103.9040709@sandisk.com> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Bart Van Assche , James Bottomley Cc: Max Gurtovoy , Christoph Hellwig , Sagi Grimberg , "linux-scsi@vger.kernel.org" On 03/04/2015 03:31 AM, Bart Van Assche wrote: > SCSI transport drivers and SCSI LLDs block a SCSI device if the > transport layer is not operational. This means that in this state > no requests should be processed, even if the REQ_PREEMPT flag has > been set. This patch avoids that a rescan shortly after a cable > pull sporadically triggers the following kernel oops: > > BUG: unable to handle kernel paging request at ffffc9001a6bc084 > IP: [] mlx4_ib_post_send+0xd2/0xb30 [mlx4_ib] > Process rescan-scsi-bus (pid: 9241, threadinfo ffff88053484a000, task ffff880534aae100) > Call Trace: > [] srp_post_send+0x65/0x70 [ib_srp] > [] srp_queuecommand+0x1cf/0x3e0 [ib_srp] > [] scsi_dispatch_cmd+0x101/0x280 [scsi_mod] > [] scsi_request_fn+0x411/0x4d0 [scsi_mod] > [] __blk_run_queue+0x27/0x30 > [] blk_execute_rq_nowait+0x82/0x110 > [] blk_execute_rq+0x62/0xf0 > [] scsi_execute+0xe8/0x190 [scsi_mod] > [] scsi_execute_req+0xa3/0x130 [scsi_mod] > [] scsi_probe_lun+0x17a/0x450 [scsi_mod] > [] scsi_probe_and_add_lun+0x156/0x480 [scsi_mod] > [] __scsi_scan_target+0xdf/0x1f0 [scsi_mod] > [] scsi_scan_host_selected+0x183/0x1c0 [scsi_mod] > [] scsi_scan+0xdb/0xe0 [scsi_mod] > [] store_scan+0x13/0x20 [scsi_mod] > [] sysfs_write_file+0xcb/0x160 > [] vfs_write+0xce/0x140 > [] sys_write+0x53/0xa0 > [] system_call_fastpath+0x16/0x1b > [<00007f611c9d9300>] 0x7f611c9d92ff > > Reported-by: Max Gurtuvoy > Signed-off-by: Bart Van Assche > Cc: Mike Christie > Cc: Christoph Hellwig > Cc: Sagi Grimberg > Cc: > --- > drivers/scsi/scsi_lib.c | 4 +++- > include/linux/blk_types.h | 4 +++- > 2 files changed, 6 insertions(+), 2 deletions(-) > > diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c > index 54d7a6c..b1a2631 100644 > --- a/drivers/scsi/scsi_lib.c > +++ b/drivers/scsi/scsi_lib.c > @@ -1311,9 +1311,11 @@ scsi_prep_state_check(struct scsi_device *sdev, struct request *req) > "rejecting I/O to dead device\n"); > ret = BLKPREP_KILL; > break; > - case SDEV_QUIESCE: > case SDEV_BLOCK: > case SDEV_CREATED_BLOCK: > + ret = BLKPREP_DEFER; > + break; > + case SDEV_QUIESCE: > /* > * If the devices is blocked we defer normal commands. > */ > diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h > index c294e3e..a1b25e3 100644 > --- a/include/linux/blk_types.h > +++ b/include/linux/blk_types.h > @@ -181,7 +181,9 @@ enum rq_flag_bits { > __REQ_ELVPRIV, /* elevator private data attached */ > __REQ_FAILED, /* set if the request failed */ > __REQ_QUIET, /* don't worry about errors */ > - __REQ_PREEMPT, /* set for "ide_preempt" requests */ > + __REQ_PREEMPT, /* set for "ide_preempt" requests and also > + for requests for which the SCSI "quiesce" > + state must be ignored. */ > __REQ_ALLOCED, /* request came from our alloc pool */ > __REQ_COPY_USER, /* contains copies of user pages */ > __REQ_FLUSH_SEQ, /* request for flush sequence */ > Look ok to me. Also tested here with iscsi and FC. Reviewed-by: Mike Christie