kernel.git/drivers/scsi/scsi_error.c, branch linux-3.6.y

[SCSI] Fix 'Device not ready' issue on mpt2sas

2012-08-22T05:42:54Z

This is a particularly nasty SCSI ATA Translation Layer (SATL) problem. SAT-2 says (section 8.12.2) if the device is in the stopped state as the result of processing a START STOP UNIT command (see 9.11), then the SATL shall terminate the TEST UNIT READY command with CHECK CONDITION status with the sense key set to NOT READY and the additional sense code of LOGICAL UNIT NOT READY, INITIALIZING COMMAND REQUIRED; mpt2sas internal SATL seems to implement this. The result is very confusing standby behaviour (using hdparm -y). If you suspend a drive and then send another command, usually it wakes up. However, if the next command is a TEST UNIT READY, the SATL sees that the drive is suspended and proceeds to follow the SATL rules for this, returning NOT READY to all subsequent commands. This means that the ordering of TEST UNIT READY is crucial: if you send TUR and then a command, you get a NOT READY to both back. If you send a command and then a TUR, you get GOOD status because the preceeding command woke the drive. This bit us badly because commit 85ef06d1d252f6a2e73b678591ab71caad4667bb Author: Tejun Heo Date: Fri Jul 1 16:17:47 2011 +0200 block: flush MEDIA_CHANGE from drivers on close(2) Changed our ordering on TEST UNIT READY commands meaning that SATA drives connected to an mpt2sas now suspend and refuse to wake (because the mpt2sas SATL sees the suspend *before* the drives get awoken by the next ATA command) resulting in lots of failed commands. The standard is completely nuts forcing this inconsistent behaviour, but we have to work around it. The fix for this is twofold: 1. Set the allow_restart flag so we wake the drive when we see it has been suspended 2. Return all TEST UNIT READY status directly to the mid layer without any further error handling which prevents us causing error handling which may offline the device just because of a media check TUR. Reported-by: Matthias Prager Cc: stable@vger.kernel.org Signed-off-by: James Bottomley

[SCSI] cleanup setting task state in scsi_error_handler()

2012-07-20T07:58:47Z

A quick reading of scsi_error_handler() one could come away with the impression that it does its wakeup event check while the task state is TASK_RUNNING. In fact it sets TASK_INTERRUPTIBLE at the bottom of the loop, but that is ~50 lines down. Just set TASK_INTERRUPTIBLE at the top of loop and be done. Signed-off-by: Dan Williams Signed-off-by: James Bottomley

[SCSI] fix eh wakeup (scsi_schedule_eh vs scsi_restart_operations)

2012-07-20T07:58:46Z

Rapid ata hotplug on a libsas controller results in cases where libsas is waiting indefinitely on eh to perform an ata probe. A race exists between scsi_schedule_eh() and scsi_restart_operations() in the case when scsi_restart_operations() issues i/o to other devices in the sas domain. When this happens the host state transitions from SHOST_RECOVERY (set by scsi_schedule_eh) back to SHOST_RUNNING and ->host_busy is non-zero so we put the eh thread to sleep even though ->host_eh_scheduled is active. Before putting the error handler to sleep we need to check if the host_state needs to return to SHOST_RECOVERY for another trip through eh. Since i/o that is released by scsi_restart_operations has been blocked for at least one eh cycle, this implementation allows those i/o's to run before another eh cycle starts to discourage hung task timeouts. Cc: Reported-by: Tom Jackson Tested-by: Tom Jackson Signed-off-by: Dan Williams Signed-off-by: James Bottomley

Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial

2012-05-23T02:22:50Z

Pull trivial updates from Jiri Kosina: "As usual, it's mostly typo fixes, redundant code elimination and some documentation updates." * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/trivial: (57 commits) edac, mips: don't change code that has been removed in edac/mips tree xtensa: Change mail addresses of Hannes Weiner and Oskar Schirmer lib: Change mail address of Oskar Schirmer net: Change mail address of Oskar Schirmer arm/m68k: Change mail address of Sebastian Hess i2c: Change mail address of Oskar Schirmer net: Fix tcp_build_and_update_options comment in struct tcp_sock atomic64_32.h: fix parameter naming mismatch Kconfig: replace "--- help ---" with "---help---" c2port: fix bogus Kconfig "default no" edac: Fix spelling errors. qla1280: Remove redundant NULL check before release_firmware() call remoteproc: remove redundant NULL check before release_firmware() qla2xxx: Remove redundant NULL check before release_firmware() call. aic94xx: Get rid of redundant NULL check before release_firmware() call tehuti: delete redundant NULL check before release_firmware() qlogic: get rid of a redundant test for NULL before call to release_firmware() bna: remove redundant NULL test before release_firmware() tg3: remove redundant NULL test before release_firmware() call typhoon: get rid of redundant conditional before all to release_firmware() ...

scsi: fix comment spelling fix recory->recovery

2012-04-16T12:35:30Z

Signed-off-by: Santosh Y Signed-off-by: Jiri Kosina

SCSI: Fix error handling when no ULD is attached

2012-04-15T18:08:53Z

Commit 18a4d0a22ed6 ("[SCSI] Handle disk devices which can not process medium access commands") introduced a bug in which we would attempt to dereference the scsi driver even when the device had no ULD attached. Ensure that a driver is registered and make the driver accessor function more resilient to errors during device discovery. Reported-by: Elric Fu Reported-by: Bart Van Assche Signed-off-by: Martin K. Petersen Signed-off-by: Linus Torvalds

[SCSI] Handle disk devices which can not process medium access commands

2012-02-19T16:14:52Z

We have experienced several devices which fail in a fashion we do not currently handle gracefully in SCSI. After a failure these devices will respond to the SCSI primary command set (INQUIRY, TEST UNIT READY, etc.) but any command accessing the storage medium will time out. The following patch adds an callback that can be used by upper level drivers to inspect the results of an error handling command. This in turn has been used to implement additional checking in the SCSI disk driver. If a medium access command fails twice but TEST UNIT READY succeeds both times in the subsequent error handling we will offline the device. The maximum number of failed commands required to take a device offline can be tweaked in sysfs. Also add a new error flag to scsi_debug which allows this scenario to be easily reproduced. [jejb: fix up integer parsing to use kstrtouint] Signed-off-by: Martin K. Petersen Signed-off-by: James Bottomley

[SCSI] scsi_error: classify some ILLEGAL_REQUEST sense as a permanent TARGET_ERROR

2012-02-19T15:39:59Z

Permanent target failures are non-retryable and should be classified as TARGET_ERROR; otherwise dm-multipath will retry an IO request that will always fail at the target. A SCSI command that fails with ILLEGAL_REQUEST sense and Additional sense 0x20, 0x21, 0x24 or 0x26 represents a permanent TARGET_ERROR. Signed-off-by: Mike Snitzer Signed-off-by: James Bottomley

[SCSI] fix the new host byte settings (DID_TARGET_FAILURE and DID_NEXUS_FAILURE)

2012-02-19T14:08:59Z

This patch fixes the host byte settings DID_TARGET_FAILURE and DID_NEXUS_FAILURE. The function __scsi_error_from_host_byte, tries to reset the host byte to DID_OK. But that does not happen because of the OR operation. Here is the flow. scsi_softirq_done-> scsi_decide_disposition -> __scsi_error_from_host_byte Let's take an example with DID_NEXUS_FAILURE. In scsi_decide_disposition, result will be set as DID_NEXUS_FAILURE (=0x11). Then in __scsi_error_from_host_byte, when we do OR with DID_OK. Purpose is to reset it back to DID_OK. But that does not happen. This patch fixes this issue. Signed-off-by: Babu Moger Signed-off-by: James Bottomley

[SCSI] add flag to skip the runtime PM calls on the host

2012-01-09T00:14:57Z

With previous change, now the ata port runtime suspend will happen as: disk suspend --> scsi target suspend --> scsi host suspend --> ata port suspend ata port(parent device) suspend need to schedule scsi EH which will resume scsi host(child device). Then the child device resume will in turn make parent device resume first. This is kind of recursive. This patch adds a new flag Scsi_Host::eh_noresume. ata port will set this flag to skip the runtime PM calls on scsi host. Acked-by: Alan Stern Signed-off-by: Lin Ming Signed-off-by: Jeff Garzik