linux-toradex.git/drivers/scsi, branch v2.6.34

[SCSI] Retry commands with UNIT_ATTENTION sense codes to fix ext3/ext4 I/O error

2010-05-05T16:15:57+00:00

There's nastyness in the way we currently handle barriers (and
discards): They're effectively filesystem commands, but they get
processed as BLOCK_PC commands.  Unfortunately BLOCK_PC commands are
taken by SCSI to be SG_IO commands and the issuer expects to see and
handle any returned errors, however trivial.  This leads to a huge
problem, because the block layer doesn't expect this to happen and any
trivially retryable error on a barrier causes an immediate I/O error
to the filesystem.

The only real way to hack around this is to take the usual class of
offending errors (unit attentions) and make them all retryable in the
case of a REQ_HARDBARRIER.  A correct fix would involve a rework of
the entire block and SCSI submit system, and so is out of scope for a
quick fix.

Cc: Hannes Reinecke 
Cc: Stable Tree 
Signed-off-by: James Bottomley

[SCSI] Enable retries for SYNCRONIZE_CACHE commands to fix I/O error

2010-05-05T16:13:26+00:00

Some arrays are giving I/O errors with ext3 filesystems when
SYNCHRONIZE_CACHE gets a UNIT_ATTENTION.  What is happening is that
these commands have no retries, so the UNIT_ATTENTION causes the
barrier to fail.  We should be enable retries here to clear any
transient error and allow the barrier to succeed.

Signed-off-by: Hannes Reinecke 
Cc: Stable Tree 
Signed-off-by: James Bottomley

[SCSI] scsi_debug: virtual_gb ignores sector_size

2010-05-02T15:18:55+00:00

In the scsi_debug driver, the virtual_gb option ignores the
sector_size, implicitly assuming that is 512 bytes.  So if
'virtual_gb=1 sector_size=4096' the result is an 8 GB (virtual) disk.

Signed-off-by: Douglas Gilbert 
Cc: Stable Tree 
Signed-off-by: James Bottomley

[SCSI] libiscsi: regression: fix header digest errors

2010-05-02T15:16:50+00:00

This fixes a regression introduced with this commit:

commit d3305f3407fa3e9452079ec6cc8379067456e4aa
Author: Mike Christie 
Date:   Thu Aug 20 15:10:58 2009 -0500

    [SCSI] libiscsi: don't increment cmdsn if cmd is not sent

in 2.6.32.

When I moved the hdr->cmdsn after init_task, I added
a bug when header digests are used. The problem is
that the LLD may calculate the header digest in init_task,
so if we then set the cmdsn after the init_task call we
change what the digest will be calculated by the target.

Signed-off-by: Mike Christie 
Cc: Stable Tree 
Signed-off-by: James Bottomley

[SCSI] fix locking around blk_abort_request()

2010-05-01T19:17:19+00:00

blk_abort_request() expects queue lock to be held by the caller.
Grab it before calling the function.

Lack of this synchronization led to infinite loop on corrupt
q->timeout_list.

Signed-off-by: Tejun Heo 
Cc: stable@kernel.org
Signed-off-by: James Bottomley

[SCSI] advansys: fix narrow board error path

2010-05-01T14:54:22+00:00

Error handling on advansys_board_found is fixed, because it's buggy in
the case we have an ASC_NARROW_BOARD set and failure happens on
AscInitAsc1000Driver step: it was freeing items of wrong struct in the
dvc_var union of struct asc_board, which could lead to an oops in the
case we set some of the fields in struct of narrow board as code was
choosing to always freeing wide board fields, and not everything was
being freed/released properly.

Signed-off-by: Herton Ronaldo Krzesinski 
Signed-off-by: James Bottomley

scsi: fix operator precedence warning

2010-04-22T18:11:02+00:00

Fix operator precedence warning (from sparse), which results in the
data value always being 0:

drivers/scsi/qla4xxx/ql4_mbx.c:470:66: warning: right shift by bigger than source value

Signed-off-by: Randy Dunlap 
Acked-by: Ravi Anand 
Cc: David C Somayajulu 
Cc: Karen Higgins 
Cc: Vikas Chaudhary 
Signed-off-by: Linus Torvalds

Merge git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6

2010-04-22T14:50:11+00:00

* git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6:
  [SCSI] bnx2i: Bug fixes related to MTU change issue when there are active iscsi sessions
  [SCSI] ibmvscsi: fix DMA API misuse
  [SCSI] wd7000: fix reset handler typo spin_unlock_irq() => spin_lock_irq()
  [SCSI] zfcp: Fix tracing of requests with error status
  [SCSI] zfcp: Update MAINTAINERS entry
  [SCSI] iscsi_tcp: fix relogin/shutdown hang
  [SCSI] qla2xxx: fix lock imbalance
  [SCSI] lpfc: fix lock imbalances
  [SCSI] be2iscsi: fix lock imbalance
  [SCSI] dpt_i2o: several use after free issues

[SCSI] bnx2i: Bug fixes related to MTU change issue when there are active iscsi sessions

2010-04-11T18:57:11+00:00

bnx2i driver has to wait and cleanup all iscsi endpoints before
returning from bnx2i_stop(). This is to make sure all chip resources
are freed before chip is reset.

As the requirements for 1G and 10G chipsets is different, added
per-device 'hba_shutdown_tmo' parameter to adapter structure

If the connections are not torn down by the daemon within this timeout
period, 'cid's will be leaked in 10G device. 1G devices are more
flexible and do not leak any resources because the whole chip ports
gets reset when MTU is changed or ethtool selftest is run

fixed a minor issue in bnx2i_ep_poll() which unnecessarily forced
error return code when driver timed out waiting for TCP connect
request to complete

Signed-off-by: Anil Veerabhadrappa 
Reviewed-by: Mike Christie 
Signed-off-by: James Bottomley

[SCSI] ibmvscsi: fix DMA API misuse

2010-04-11T18:38:54+00:00

ibmvscsi uses dma_unmap_single() for buffers mapped via
dma_map_sg(). It works however it's the API violation. The DMA debug
facility complains about it:

http://marc.info/?l=linux-scsi&m=127018555013151&w=2

Reported-by: Sachin Sant 
Tested-by: Sachin Sant 
Signed-off-by: FUJITA Tomonori 
Signed-off-by: James Bottomley