From 6f381fa344911d5a234b13574433cf23036f9467 Mon Sep 17 00:00:00 2001 From: Lin Ming Date: Thu, 12 Apr 2012 13:50:38 +0800 Subject: [SCSI] scsi_lib: use correct DMA device in __scsi_alloc_queue MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Currently, __scsi_alloc_queue uses SCSI host's parent device as DMA device to set segment boundary. But the parent device may not refer to the DMA device. For example, for ATA disk, SCSI host's parent device now refers to ATA port. Since commit d139b9b([SCSI] scsi_lib_dma: fix bug with dma maps on nested scsi objects), a new field Scsi_Host->dma_dev was introduced to refer to the real DMA device. Use ->dma_dev in __scsi_alloc_queue to correctly set segment boundary. Bug report: http://marc.info/?l=linux-ide&m=133177818318187&w=2 Reported-and-tested-by: Jörg Sommer Signed-off-by: Lin Ming Signed-off-by: James Bottomley --- drivers/scsi/scsi_lib.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/scsi/scsi_lib.c') diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index ead6405f3e51..5dfd7495d1a1 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -1638,7 +1638,7 @@ struct request_queue *__scsi_alloc_queue(struct Scsi_Host *shost, request_fn_proc *request_fn) { struct request_queue *q; - struct device *dev = shost->shost_gendev.parent; + struct device *dev = shost->dma_dev; q = blk_init_queue(request_fn, NULL); if (!q) -- cgit v1.2.3 From a7a20d103994fd760766e6c9d494daa569cbfe06 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Thu, 22 Mar 2012 17:05:11 -0700 Subject: [SCSI] sd: limit the scope of the async probe domain sd injects and synchronizes probe work on the global kernel-wide domain. This runs into conflict with PM that wants to perform resume actions in async context: [ 494.237079] INFO: task kworker/u:3:554 blocked for more than 120 seconds. [ 494.294396] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [ 494.360809] kworker/u:3 D 0000000000000000 0 554 2 0x00000000 [ 494.420739] ffff88012e4d3af0 0000000000000046 ffff88013200c160 ffff88012e4d3fd8 [ 494.484392] ffff88012e4d3fd8 0000000000012500 ffff8801394ea0b0 ffff88013200c160 [ 494.548038] ffff88012e4d3ae0 00000000000001e3 ffffffff81a249e0 ffff8801321c5398 [ 494.611685] Call Trace: [ 494.632649] [] schedule+0x5a/0x5c [ 494.674687] [] async_synchronize_cookie_domain+0xb6/0x112 [ 494.734177] [] ? __init_waitqueue_head+0x50/0x50 [ 494.787134] [] ? scsi_remove_target+0x48/0x48 [ 494.837900] [] async_synchronize_cookie+0x15/0x17 [ 494.891567] [] async_synchronize_full+0x54/0x70 <-- here we wait for async contexts to complete [ 494.943783] [] ? async_synchronize_full_domain+0x1a/0x1a [ 495.002547] [] sd_remove+0x2c/0xa2 [sd_mod] [ 495.051861] [] __device_release_driver+0x86/0xcf [ 495.104807] [] device_release_driver+0x25/0x32 <-- here we take device_lock() [ 853.511341] INFO: task kworker/u:4:549 blocked for more than 120 seconds. [ 853.568693] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [ 853.635119] kworker/u:4 D ffff88013097b5d0 0 549 2 0x00000000 [ 853.695129] ffff880132773c40 0000000000000046 ffff880130790000 ffff880132773fd8 [ 853.758990] ffff880132773fd8 0000000000012500 ffff88013288a0b0 ffff880130790000 [ 853.822796] 0000000000000246 0000000000000040 ffff88013097b5c8 ffff880130790000 [ 853.886633] Call Trace: [ 853.907631] [] schedule+0x5a/0x5c [ 853.949670] [] __mutex_lock_common+0x220/0x351 [ 854.001225] [] ? device_resume+0x58/0x1c4 [ 854.049082] [] ? device_resume+0x58/0x1c4 [ 854.097011] [] mutex_lock_nested+0x2f/0x36 <-- here we wait for device_lock() [ 854.145591] [] device_resume+0x58/0x1c4 [ 854.192066] [] async_resume+0x1e/0x45 [ 854.237019] [] async_run_entry_fn+0xc6/0x173 <-- ...while running in async context Provide a 'scsi_sd_probe_domain' so that async probe actions actions can be flushed without regard for the state of PM, and allow for the resume path to handle devices that have transitioned from SDEV_QUIESCE to SDEV_DEL prior to resume. Acked-by: Alan Stern [alan: uplevel scsi_sd_probe_domain, clarify scsi_device_resume] Signed-off-by: Dan Williams [jejb: remove unneeded config guards in include file] Signed-off-by: James Bottomley --- drivers/scsi/scsi.c | 6 ++++++ drivers/scsi/scsi_lib.c | 10 +++++++--- drivers/scsi/scsi_pm.c | 2 +- drivers/scsi/scsi_priv.h | 2 ++ drivers/scsi/sd.c | 5 +++-- 5 files changed, 19 insertions(+), 6 deletions(-) (limited to 'drivers/scsi/scsi_lib.c') diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c index 07322ecff90d..61c82a345f82 100644 --- a/drivers/scsi/scsi.c +++ b/drivers/scsi/scsi.c @@ -90,6 +90,12 @@ unsigned int scsi_logging_level; EXPORT_SYMBOL(scsi_logging_level); #endif +#if IS_ENABLED(CONFIG_PM) || IS_ENABLED(CONFIG_BLK_DEV_SD) +/* sd and scsi_pm need to coordinate flushing async actions */ +LIST_HEAD(scsi_sd_probe_domain); +EXPORT_SYMBOL(scsi_sd_probe_domain); +#endif + /* NB: These are exposed through /proc/scsi/scsi and form part of the ABI. * You may not alter any existing entry (although adding new ones is * encouraged once assigned by ANSI/INCITS T10 diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index ead6405f3e51..dbe43924d7ae 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -2348,10 +2348,14 @@ EXPORT_SYMBOL(scsi_device_quiesce); * * Must be called with user context, may sleep. */ -void -scsi_device_resume(struct scsi_device *sdev) +void scsi_device_resume(struct scsi_device *sdev) { - if(scsi_device_set_state(sdev, SDEV_RUNNING)) + /* check if the device state was mutated prior to resume, and if + * so assume the state is being managed elsewhere (for example + * device deleted during suspend) + */ + if (sdev->sdev_state != SDEV_QUIESCE || + scsi_device_set_state(sdev, SDEV_RUNNING)) return; scsi_run_queue(sdev->request_queue); } diff --git a/drivers/scsi/scsi_pm.c b/drivers/scsi/scsi_pm.c index c4670642d023..f661a41fa4c6 100644 --- a/drivers/scsi/scsi_pm.c +++ b/drivers/scsi/scsi_pm.c @@ -97,7 +97,7 @@ static int scsi_bus_prepare(struct device *dev) { if (scsi_is_sdev_device(dev)) { /* sd probing uses async_schedule. Wait until it finishes. */ - async_synchronize_full(); + async_synchronize_full_domain(&scsi_sd_probe_domain); } else if (scsi_is_host_device(dev)) { /* Wait until async scanning is finished */ diff --git a/drivers/scsi/scsi_priv.h b/drivers/scsi/scsi_priv.h index be4fa6d179b1..07ce3f51701d 100644 --- a/drivers/scsi/scsi_priv.h +++ b/drivers/scsi/scsi_priv.h @@ -163,6 +163,8 @@ static inline int scsi_autopm_get_host(struct Scsi_Host *h) { return 0; } static inline void scsi_autopm_put_host(struct Scsi_Host *h) {} #endif /* CONFIG_PM_RUNTIME */ +extern struct list_head scsi_sd_probe_domain; + /* * internal scsi timeout functions: for use by mid-layer and transport * classes. diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c index 5ba5c2a9e8e9..6f0a4c612b3b 100644 --- a/drivers/scsi/sd.c +++ b/drivers/scsi/sd.c @@ -65,6 +65,7 @@ #include #include "sd.h" +#include "scsi_priv.h" #include "scsi_logging.h" MODULE_AUTHOR("Eric Youngdale"); @@ -2722,7 +2723,7 @@ static int sd_probe(struct device *dev) dev_set_drvdata(dev, sdkp); get_device(&sdkp->dev); /* prevent release before async_schedule */ - async_schedule(sd_probe_async, sdkp); + async_schedule_domain(sd_probe_async, sdkp, &scsi_sd_probe_domain); return 0; @@ -2756,7 +2757,7 @@ static int sd_remove(struct device *dev) sdkp = dev_get_drvdata(dev); scsi_autopm_get_device(sdkp->device); - async_synchronize_full(); + async_synchronize_full_domain(&scsi_sd_probe_domain); blk_queue_prep_rq(sdkp->device->request_queue, scsi_prep_fn); blk_queue_unprep_rq(sdkp->device->request_queue, NULL); device_del(&sdkp->dev); -- cgit v1.2.3 From b7e94a1686c5daef4f649f7f4f839cc294f07710 Mon Sep 17 00:00:00 2001 From: Jun'ichi Nomura Date: Tue, 22 May 2012 18:57:17 +0900 Subject: [SCSI] Fix dm-multipath starvation when scsi host is busy block congestion control doesn't have any concept of fairness across multiple queues. This means that if SCSI reports the host as busy in the queue congestion control it can result in an unfair starvation situation in dm-mp if there are multiple multipath devices on the same host. For example: http://www.redhat.com/archives/dm-devel/2012-May/msg00123.html The fix for this is to report only the sdev busy state (and ignore the host busy state) in the block congestion control call back. The host is still congested, but the SCSI subsystem will sort out the congestion in a fair way because it knows the relation between the queues and the host. [jejb: fixed up trailing whitespace] Reported-by: Bernd Schubert Tested-by: Bernd Schubert Signed-off-by: Jun'ichi Nomura Cc: Signed-off-by: James Bottomley --- drivers/scsi/scsi_lib.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) (limited to 'drivers/scsi/scsi_lib.c') diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index 62ddfd31d4ce..6dfb9785d345 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -1378,16 +1378,19 @@ static int scsi_lld_busy(struct request_queue *q) { struct scsi_device *sdev = q->queuedata; struct Scsi_Host *shost; - struct scsi_target *starget; if (!sdev) return 0; shost = sdev->host; - starget = scsi_target(sdev); - if (scsi_host_in_recovery(shost) || scsi_host_is_busy(shost) || - scsi_target_is_busy(starget) || scsi_device_is_busy(sdev)) + /* + * Ignore host/starget busy state. + * Since block layer does not have a concept of fairness across + * multiple queues, congestion of host/starget needs to be handled + * in SCSI layer. + */ + if (scsi_host_in_recovery(shost) || scsi_device_is_busy(sdev)) return 1; return 0; -- cgit v1.2.3