On Tue, Feb 21, 2017 at 5:57 PM, Hannes Reinecke wrote:
> When attempting a command abort we should check the command status
> prior to sending the abort; the command might've been completed
> already.
I think we are already taking care of this. Before calling
'mpt3sas_scsih_issue_tm()' function
Martin,
can we get at least this patch still in for 4.11? I'd really like
to see as many intance of the old MSI-X allocation and IRQ affinity
mess dead as soon as possible.
On 02/22/2017 09:28 AM, Christoph Hellwig wrote:
> Martin,
>
> can we get at least this patch still in for 4.11? I'd really like
> to see as many intance of the old MSI-X allocation and IRQ affinity
> mess dead as soon as possible.
>
I'm currently reworking the patchset with the suggestions from
On Wed, Feb 22, 2017 at 1:14 AM, Jan Kara wrote:
> On Sun 19-02-17 18:19:58, Omar Sandoval wrote:
>> On Fri, Feb 17, 2017 at 04:43:56PM -0800, James Bottomley wrote:
>> > This seems to be related to a 0day test we got on the block tree,
>> > details here:
>> >
>> > http://marc.info/?t=148624068800
Cleanup the MSI-X handling allowing us to use
the PCI-layer provided vector allocation.
Signed-off-by: Hannes Reinecke
Reviewed-by: Christoph Hellwig
---
drivers/scsi/mpt3sas/mpt3sas_base.c | 105 +---
drivers/scsi/mpt3sas/mpt3sas_base.h | 2 -
2 files changed,
When attempting a command abort we should check the command status
prior to sending the abort; the command might've been completed
already.
Signed-off-by: Hannes Reinecke
Reviewed-by: Christoph Hellwig
---
drivers/scsi/mpt3sas/mpt3sas_scsih.c | 11 ---
1 file changed, 8 insertions(+), 3
No functional change.
Signed-off-by: Hannes Reinecke
Reviewed-by: Christoph Hellwig
---
drivers/scsi/mpt3sas/mpt3sas_base.c | 23 ++-
1 file changed, 14 insertions(+), 9 deletions(-)
diff --git a/drivers/scsi/mpt3sas/mpt3sas_base.c
b/drivers/scsi/mpt3sas/mpt3sas_base.c
ind
Abstract accesses to the scsi_lookup array by introducing
mpt3sas_get_st_from_smid().
Signed-off-by: Hannes Reinecke
Reviewed-by: Christoph Hellwig
---
drivers/scsi/mpt3sas/mpt3sas_base.c | 22 ++
drivers/scsi/mpt3sas/mpt3sas_base.h | 2 ++
drivers/scsi/mpt3sas/mp
One can simply check 'target_busy' or 'device_busy' when figuring
out if there are outstanding commands; no need to painstakingly
counting them by hand.
Suggested-by: Christoph Hellwig
Signed-off-by: Hannes Reinecke
---
drivers/scsi/mpt3sas/mpt3sas_scsih.c | 88 +++--
Hi all,
this is the first part of my patchset to enable scsi multiqueue for the
mpt3sas driver.
While the HBA only has a single mailbox register for submitting commands,
it does have individual receive queues per MSI-X interrupt and as such
does benefit from converting it to full multiqueue suppor
ioctl passthrough commands require a SCSIIO smid, but cannot
easily integrate with the block layer. But the driver already
has reserved some SCSIIO smids and we're only ever allowing
one ioctl command at a time we can use the first reserved smid
for ioctl commands.
Signed-off-by: Hannes Reinecke
Use 'list_splice_init()' instead of hand-crafted function.
No functional change.
Signed-off-by: Hannes Reinecke
---
drivers/scsi/mpt3sas/mpt3sas_base.c | 11 ++-
1 file changed, 2 insertions(+), 9 deletions(-)
diff --git a/drivers/scsi/mpt3sas/mpt3sas_base.c
b/drivers/scsi/mpt3sas/mpt3
Instead of holding 'struct scsiio_tracker' in its own pool we
can embed it into the payload of the scsi command. This allows
us to get rid of the lock when submitting and receiving commands
and streamline operations.
Signed-off-by: Hannes Reinecke
---
drivers/scsi/mpt3sas/mpt3sas_base.c | 1
Just a wrapper around the scsi lookup array and only used
in one place, so open-code it.
Signed-off-by: Hannes Reinecke
Reviewed-by: Christoph Hellwig
---
drivers/scsi/mpt3sas/mpt3sas_scsih.c | 15 +--
1 file changed, 1 insertion(+), 14 deletions(-)
diff --git a/drivers/scsi/mpt3sa
No functional change.
Signed-off-by: Hannes Reinecke
Reviewed-by: Christoph Hellwig
---
drivers/scsi/mpt3sas/mpt3sas_base.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/drivers/scsi/mpt3sas/mpt3sas_base.c
b/drivers/scsi/mpt3sas/mpt3sas_base.c
index 5b7aec5..3062171
Use 'host_busy' instead of counting outstanding commands by hand.
Suggested-by: Christoph Hellwig
Signed-off-by: Hannes Reinecke
---
drivers/scsi/mpt3sas/mpt3sas_base.c | 14 --
1 file changed, 4 insertions(+), 10 deletions(-)
diff --git a/drivers/scsi/mpt3sas/mpt3sas_base.c
b/dri
Move the check for outstanding commands out of the function
allowing us to simplify the overall code.
Signed-off-by: Hannes Reinecke
---
drivers/scsi/mpt3sas/mpt3sas_base.h | 6 +-
drivers/scsi/mpt3sas/mpt3sas_ctl.c | 4 +-
drivers/scsi/mpt3sas/mpt3sas_scsih.c | 106 --
The device has several reply queues, so this patch implements
correct scsi-mq interrupt steering to take advantage of full
blk-mq support.
However, as performance results are inconclusive this patch
also adds a module parameter 'enable_scsi_mq' to expose all
completion queues via scsi-mq; default i
Hi all,
this is the second part of my patchset to enable scsi multiqueue for the
mpt3sas driver.
While the HBA only has a single mailbox register for submitting commands,
it does have individual receive queues per MSI-X interrupt and as such
does benefit from converting it to full multiqueue suppo
Implement blk_mq_make_unique_tag() to generate a unique tag
value from a given hardware queue and tag value.
Signed-off-by: Hannes Reinecke
---
block/blk-mq-tag.c | 3 +--
include/linux/blk-mq.h | 5 +
2 files changed, 6 insertions(+), 2 deletions(-)
diff --git a/block/blk-mq-tag.c b/bl
Use msix_index as explicit argument for the various put_smid
callbacks.
No functional change.
Signed-off-by: Hannes Reinecke
---
drivers/scsi/mpt3sas/mpt3sas_base.c | 56 +---
drivers/scsi/mpt3sas/mpt3sas_base.h | 18 --
drivers/scsi/mpt3sas/mpt3sas_
> "Christoph" == Christoph Hellwig writes:
>> I can bring them in after Linus' initial block pull.
Christoph> Both the block and SCSI trees are now merged by Linus, and
Christoph> Jens didn't pick up patch one from this series yet - maybe
Christoph> it's best to send the whole series through
> -Original Message-
> From: linux-scsi-ow...@vger.kernel.org [mailto:linux-scsi-
> ow...@vger.kernel.org] On Behalf Of Tobin C. Harding
> Sent: Wednesday, February 22, 2017 12:13 AM
> To: Don Brace
> Cc: linux-scsi@vger.kernel.org; kernel-janit...@vger.kernel.org; Tobin C.
> Harding
> Su
> -Original Message-
> From: Tobin C. Harding [mailto:m...@tobin.cc]
> Sent: Wednesday, February 22, 2017 12:13 AM
> To: Don Brace
> Cc: linux-scsi@vger.kernel.org; kernel-janit...@vger.kernel.org; Tobin C.
> Harding
> Subject: [PATCH 2/3] cciss: Fix checkpatch OPEN_BRACE
>
> EXTERNAL EM
> -Original Message-
> From: Tobin C. Harding [mailto:m...@tobin.cc]
> Sent: Wednesday, February 22, 2017 12:13 AM
> To: Don Brace
> Cc: linux-scsi@vger.kernel.org; kernel-janit...@vger.kernel.org; Tobin C.
> Harding
> Subject: [PATCH 1/3] cciss: Fix checkpatch TRAILING_WHITESPACE
>
> EX
When a command is sent as part of the error handling there
is not point whatsoever to start EH escalation when that
command fails; we are _already_ in the error handler,
and the escalation is about to commence anyway.
So just call 'scsi_try_to_abort_cmd()' to abort outstanding
commands and let the
From: Christoph Hellwig
We now first try to call ->eh_abort_handler from a work queue, but libsas
was always failing that for no good reason. Allow async aborts.
Reviewed-by: Johannes Thumshirn
Reviewed-by: Hannes Reinecke
Signed-off-by: Christoph Hellwig
---
drivers/scsi/libsas/sas_scsi_ho
There hasn't been any reports for HBAs where asynchronous abort
would not work, so we should make it mandatory and remove
the fallback.
Signed-off-by: Hannes Reinecke
Reviewed-by: Johannes Thumshirn
---
Documentation/scsi/scsi_eh.txt | 18 --
drivers/scsi/scsi_error.c | 81
To detect if a failed command has been retried we must not
clear scmd->eh_eflags when EH finishes.
The flag should be persistent throughout the lifetime
of the command.
Reviewed-by: Johannes Thumshirn
Signed-off-by: Hannes Reinecke
---
Documentation/scsi/scsi_eh.txt | 14 +++---
drivers
Hi all,
this is a resend of a small patchset for cleaning up SCSI EH.
Primary goal is to make asynchronous aborts mandatory; there hasn't
been a single report so far where asynchronous abort won't work, so
the 'no_async_abort' flag has never been used and will be removed
with this patchset.
Additi
On 02/21/2017 11:12 PM, Tobin C. Harding wrote:
> Re-send after spell checking. :(
>
> Coccinelle warns about unnecessary cast on call to kmalloc(). Checkpatch
> emits various warnings when parsing file.
>
> Clean up two checkpatch warnings. Remove trailing whitespace and
> clean up opening brace
scsi_eh_scmd_add() currently only will fail if no
error handler thread is started (which will never be the
case) or if the state machine encounters an illegal transition.
But if we're encountering an invalid state transition
chances is we cannot fixup things with the error handler.
So better add a
add missing Kconfig NVME dependencies
Can't believe I missed posting this
-- james
Signed-off-by: James Smart
---
drivers/scsi/Kconfig | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig
index d4023bf..2558434 100644
--- a/drivers/scsi/Kconfig
+++ b
On Fri, Feb 3, 2017 at 2:49 PM, Alim Akhtar wrote:
> Hi Kishon,
>
>
> On 11/19/2015 07:09 PM, Kishon Vijay Abraham I wrote:
>>
>> Hi,
>>
>> On Tuesday 17 November 2015 01:41 PM, Alim Akhtar wrote:
>>>
>>> Hi
>>> Thanks again for looking into this.
>>>
>>> On 11/17/2015 11:46 AM, Kishon Vijay Abrah
add missing Kconfig NVME dependencies
Can't believe I missed posting this
-- james
Heh, the this sort of comment should come after
the '---' separator (below) unless you want it to live forever
in the git log...
Signed-off-by: James Smart
---
[here]
drivers/scsi/Kconfig | 1 +
1 fil
On 02/22/2017 08:07 AM, Hannes Reinecke wrote:
> this is a resend of a small patchset for cleaning up SCSI EH.
> Primary goal is to make asynchronous aborts mandatory; there hasn't
> been a single report so far where asynchronous abort won't work, so
> the 'no_async_abort' flag has never been used
https://bugzilla.kernel.org/show_bug.cgi?id=120371
Wilfried Klaebe (linux-ker...@lebenslange-mailadresse.de) changed:
What|Removed |Added
Status|NEW
add missing Kconfig NVME dependencies
-- james
Signed-off-by: James Smart
---
drivers/scsi/Kconfig | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig
index d4023bf..2558434 100644
--- a/drivers/scsi/Kconfig
+++ b/drivers/scsi/Kconfig
@@ -1240,6 +124
On 02/17/2017 01:24 AM, lixi...@cmss.chinamobile.com wrote:
> From: Xiubo Li
>
> Currently for the TCMU, the ring buffer size is fixed to 64K cmd
> area + 1M data area, and this will be bottlenecks for high iops.
Hi Xiubo, thanks for your work.
daynmic -> dynamic
Have you benchmarked this patc
On Wed, 2017-02-22 at 17:07 +0100, Hannes Reinecke wrote:
> this is a resend of a small patchset for cleaning up SCSI EH.
> Primary goal is to make asynchronous aborts mandatory; there hasn't
> been a single report so far where asynchronous abort won't work, so
> the 'no_async_abort' flag has never
Current driver Hotplug processing code skips over Enclosure channel,
therefore any addition/removal of expander enclosure is not processed.
Additionally device addition code relies on older device type, which
prevents the hotplug of adapter expanders.
Fixed by removing code that skips over Enclos
The mac_scsi driver still gets disabled when SCSI=m. This should have
been fixed back when I enabled the tristate but I didn't see the bug.
Fixes: 6e9ae6d560e1 ("[PATCH] mac_scsi: Add module option to Kconfig")
Signed-off-by: Finn Thain
---
drivers/scsi/Kconfig | 2 +-
1 file changed, 1 insertio
On 2017-02-20 23:36, Potomski, MichalX wrote:
Since in UFS 2.1 specification some of the descriptor
lengths differs from 2.0 specification and some devices,
which are reporting spec version 2.0 have different
descriptor lengths we can not rely on hardcoded values
taken from 2.0 specification. Thi
On 2017-02-20 19:06, Martin K. Petersen wrote:
"Subhash" == Subhash Jadavani writes:
Subhash,
Subhash> Dan Carpenter reported this: ---
The
Subhash> patch 9c46b8676271: "scsi: ufs-qcom: dump additional testbus
Subhash> registers" from Feb 3, 2017, leads to the following static
Subhash> che
> "James" == James Smart writes:
James> add missing Kconfig NVME dependencies
Applied to 4.11/scsi-fixes.
--
Martin K. Petersen Oracle Linux Engineering
> "Chad" == Dupuis, Chad writes:
Chad> The QLogic FastLinQ Driver for FCoE (qedf) is the FCoE specific
Chad> module for 41000 Series Converged Network Adapters by QLogic. This
Chad> patch consists of following changes:
Now that Linus pulled Dave's tree I have gone ahead and merged this into
> "Colin" == Colin King writes:
Colin> On an allocation failure of gd, the current exit path is via
Colin> out_free_devt which leaves sdpk still allocated and hence it gets
Colin> leaked. Fix this by correcting the order of resource free'ing
Colin> with a change in the error exit path labels.
> "Michal" == Potomski, MichalX writes:
Michal,
Michal> Since in UFS 2.1 specification some of the descriptor lengths
Michal> differs from 2.0 specification and some devices, which are
Michal> reporting spec version 2.0 have different descriptor lengths we
Michal> can not rely on hardcoded v
> "Himanshu" == Himanshu Madhani writes:
Himanshu,
Himanshu> min_vecs is the minimum amount of vectors needed to operate in
Himanshu> MSI-X mode which may just include the vectors that don't need
Himanshu> affinity.
Himanshu> Disabling affinity settings causes the qla2xxx driver
Himanshu> s
> "Dan" == Dan Carpenter writes:
Dan,
Dan> We accidentally return an uninitialized variable on success.
Applied to 4.11/scsi-fixes.
Thanks!
--
Martin K. Petersen Oracle Linux Engineering
> "Christoph" == Christoph Hellwig writes:
Christoph,
>> I can bring them in after Linus' initial block pull.
Christoph> Both the block and SCSI trees are now merged by Linus, and
Christoph> Jens didn't pick up patch one from this series yet - maybe
Christoph> it's best to send the whole se
> "Christoph" == Christoph Hellwig writes:
Christoph> Martin, can we get at least this patch still in for 4.11?
Christoph> I'd really like to see as many intance of the old MSI-X
Christoph> allocation and IRQ affinity mess dead as soon as possible.
Me too. I'll queue it up for 4.11 unless Br
Martin,
On 2/23/17 13:44, Damien Le Moal wrote:
> I do not see the problematic resid correction code in the mpt3sas driver
> in 4.11/scsi-queue branch. Is this expected ? Did the mpt3sas driver
> updates removed it ?
>
> It looks like that branch is based on 4.10.0-rc2. The mpt3sas was
> applied
Martin,
On 2/22/17 13:24, Martin K. Petersen wrote:
>> "Damien" == Damien Le Moal writes:
>
> Damien,
>
> Damien> I think we would still need the check for REQ_TYPE_FS to avoid
> Damien> interfering with SG_IO commands. As for the "medium access
> Damien> command" test, I am not sure if the
From: Nicholas Bellinger
When transport_clear_lun_ref() is shutting down a se_lun via
configfs with new I/O in-flight, it's possible to trigger a
NULL pointer dereference in transport_lookup_cmd_lun() due
to the fact percpu_ref_get() doesn't do any __PERCPU_REF_DEAD
checking before incrementing l
On 02/22/2017 08:44 PM, Bart Van Assche wrote:
> On 02/22/2017 08:07 AM, Hannes Reinecke wrote:
>> this is a resend of a small patchset for cleaning up SCSI EH.
>> Primary goal is to make asynchronous aborts mandatory; there hasn't
>> been a single report so far where asynchronous abort won't work,
On 02/22/2017 10:21 PM, Bart Van Assche wrote:
> On Wed, 2017-02-22 at 17:07 +0100, Hannes Reinecke wrote:
>> this is a resend of a small patchset for cleaning up SCSI EH.
>> Primary goal is to make asynchronous aborts mandatory; there hasn't
>> been a single report so far where asynchronous abort
57 matches
Mail list logo
