Currently, for SCSI LUN_RESETs the target layer can only wait on
bio/requests it has sent. This normally results in the LUN_RESET
timing out on the initiator side and that SCSI error handler
escalating to something more disruptive.
To fix this, the following patches add a block layer helper and
ca
From: Mike Christie
Currently, if the SCSI eh runs then before we do a LUN_RESET
we stop the host. This patch and the block layer one before it
begin to add infrastructure to be able to do a LUN_RESET and
eventually do a transport level recovery without having to stop the
host.
For LUn-reset, th
From: Mike Christie
Instead of waiting for commands, map the lun reset operation
to a queue reset.
We do not check the result of blk_reset_queue because if
it works then we need to wait for the bio/request completions
and if it failed we might as wait and hope like we did before.
Signed-off-by:
From: Mike Christie
This adds a request_queue/mq_ops callout which when called
should force the completion/failure of requests that have been
dequeued by the driver. Requests can be completed normally
or should be failed with -EINTR.
On success the reset callout should complete/fail dequeued
req
From: Mike Christie
Like a lot of drivers, iscsi already supported SG_SCSI_RESET_DEVICE
and did not need the host stopped, so supporting
eh_async_device_reset_handler just required some renames, and
changing the callout argument from a scsi_cmnd to a scsi_device.
Signed-off-by: Mike Christie
--
From: Mike Christie
If the __blk_mq_run_hw_queue is a result of a async run or retry
then we do not have a reference to the queue. At this time,
blk_cleanup_queue could begin tearing it down while the queue_rq
function is being run.
This patch just has us do a blk_queue_enter call.
Signed-off-b
Hi experts,
I think these patches are great, and I am ready to help in user space.
Thanks,
BR
Zhu Lingshan
On 05/09/2016 09:17 AM, Lee Duncan wrote:
On 04/14/2016 06:18 PM, Lee Duncan wrote:
These patches make the location of "/var/target" configurable,
though it still defauls to "/var/target
Hi Chaitra and Suganath,
I've got a question regarding mpt3sas' memory allocation used when doing a
firmware upgrade. Currently you're doing a pci_alloc_consitent() which tries
to allocate memory via GFP_ATOMIC. This memory then is passed as a single
element on a sg_list.
Jeff reported it returne
Hi,
On 24-05-16 14:44, James Bottomley wrote:
On Tue, 2016-05-24 at 08:53 +0200, Hans de Goede wrote:
Hi,
On 23-05-16 19:36, James Bottomley wrote:
On Mon, 2016-05-23 at 13:49 +0200, Hans de Goede wrote:
Commit 198de51dbc34 ("USB: uas: Limit qdepth at the scsi-host
level")
removed the scsi_c
What if a UAS bridge requires specific SCSI command (e.g. UNMAP) to be
issued unqueued/untagged? Would track_queue_depth help?
On 25 May 2016 at 19:04, Hans de Goede wrote:
> Hi,
>
>
> On 24-05-16 14:44, James Bottomley wrote:
>>
>> On Tue, 2016-05-24 at 08:53 +0200, Hans de Goede wrote:
>>>
>>>
This patch series addresses comments by Joe with runing checkpatch with a
--strict. It cleans up all the misc styling and removes all the forward
declarations.
The patch also addresses all of Bart's comments besides the preallocation of
buffers before IO starts and the merging of unpack_lun with s
From: bryantly
This patch removes forward declarations and re-organizes the
functions within the driver. This patch also fixes MAINTAINERS
for ibmvscsis.
Signed-off-by: bryantly
---
MAINTAINERS |4 +-
drivers/scsi/ibmvscsi/ibmvscsis.c | 2709 ++
From: bryantly
This patch contains cleaning up the code for styling and also addresses Bart's
comments.
Signed-off-by: bryantly
---
MAINTAINERS | 4 +-
drivers/scsi/Kconfig | 36 +--
drivers/scsi/Makefile | 4 +-
drivers/scsi/ibmvscsi/Makefi
On Wed, 2016-05-25 at 09:17 -0500, Bryant G. Ly wrote:
> From: bryantly
Please use your whole name here and for your sign-off like:
From: Bryant G. Ly
Signed-off-by: Bryant G. Ly
> This patch removes forward declarations and re-organizes the
> functions within the driver. This patch also fixe
This patch is based on:
git.kernel.org:/pub/scm/linux/kernel/git/hare/scsi-devel branch zbc.v6
According to the Zoned Block Commands (ZBC) Revision 5 SCSI specification,
a Host Managed and Host Aware device is limited to support at a minimum the
READ (16) and WRITE (16) read/write CDBs. This means
On Tue, May 24, 2016 at 9:40 AM, Nicholas A. Bellinger
wrote:
> Hi Or & Co,
> On Wed, 2016-05-18 at 14:45 +0300, Or Gerlitz wrote:
>> On Sat, Apr 30, 2016 at 6:54 PM, Or Gerlitz wrote:
>> > On Tue, Apr 19, 2016 at 9:30 PM, Varun Prakash wrote:
>> >> cxgbit.h - This file contains data structure
>
On 05/25/2016 12:54 AM, mchri...@redhat.com wrote:
If the __blk_mq_run_hw_queue is a result of a async run or retry
then we do not have a reference to the queue. At this time,
blk_cleanup_queue could begin tearing it down while the queue_rq
function is being run.
This patch just has us do a blk_
On 05/25/2016 12:55 AM, mchri...@redhat.com wrote:
+ spin_lock_irq(q->queue_lock);
+ wait_event_lock_irq(q->reset_wq,
+ !queue_flag_test_and_set(QUEUE_FLAG_RESETTING, q),
+ *q->queue_lock);
+ if (blk_queue_dead(q)) {
+
> From: Or Gerlitz [mailto:gerlitz...@gmail.com]
>
> On Tue, May 24, 2016 at 9:40 AM, Nicholas A. Bellinger
> wrote:
> > Hi Or & Co,
> > On Wed, 2016-05-18 at 14:45 +0300, Or Gerlitz wrote:
> >> On Sat, Apr 30, 2016 at 6:54 PM, Or Gerlitz wrote:
> >> > On Tue, Apr 19, 2016 at 9:30 PM, Varun Prak
On 05/25/2016 11:13 AM, Bart Van Assche wrote:
> On 05/25/2016 12:55 AM, mchri...@redhat.com wrote:
>> +spin_lock_irq(q->queue_lock);
>> +wait_event_lock_irq(q->reset_wq,
>> +!queue_flag_test_and_set(QUEUE_FLAG_RESETTING, q),
>> +*q->queue_lock);
>> +if (
On 05/25/2016 02:15 PM, Mike Christie wrote:
> On 05/25/2016 10:53 AM, Bart Van Assche wrote:
>> On 05/25/2016 12:54 AM, mchri...@redhat.com wrote:
>>> If the __blk_mq_run_hw_queue is a result of a async run or retry
>>> then we do not have a reference to the queue. At this time,
>>> blk_cleanup_qu
If _scsih_sas_host_add's call to mpt3sas_config_get_sas_iounit_pg0
fails, ioc->sas_hba.parent_dev may be left uninitialized. A later
device probe could invoke mpt3sas_transport_port_add which will call
sas_port_alloc_num [scsi_transport_sas] with a NULL parent_dev pointer.
Signed-off-by: Joe Lawr
In _scsih_sas_host_add, the number of HBA phys are determined and then
later used to allocate an array of struct _sas_phy's. If the routine
sets ioc->sas_hba.num_phys, but then fails to allocate the
ioc->sas_hba.phy array (by kcalloc error or other intermediate
error/exit path), ioc->sas_hba is l
On 05/25/2016 10:53 AM, Bart Van Assche wrote:
> On 05/25/2016 12:54 AM, mchri...@redhat.com wrote:
>> If the __blk_mq_run_hw_queue is a result of a async run or retry
>> then we do not have a reference to the queue. At this time,
>> blk_cleanup_queue could begin tearing it down while the queue_rq
There are many error paths in _scsih_sas_host_add that lead to an early
exit and a few that leave IOC resources uninitialized, setting the stage
for a later crash.
This can be emulated using a systemtap script like:
% stap -g -e \
'probe module("mpt3sas").function("mpt3sas_config_get_sas_ioun
Hello, Brian.
So, of all the ata drivers, ipr seems to be the only driver which
doesn't implement ata_port_opeations->error_handler and thus depends
on the old error handling path. I'm wondering whether it'd be possible
to convert ipr to implement ->error_handler and drop the old path.
Would that
From: Nicholas Bellinger
During transport_generic_free_cmd() with a concurrent TMR
ABORT_TASK and shutdown CMD_T_FABRIC_STOP bit set, the
caller will be blocked on se_cmd->cmd_wait_stop completion
until the final kref_put() -> target_release_cmd_kref()
has been invoked to call complete().
Howeve
- Original Message -
> From: "Himanshu Madhani"
> To: "Laurence Oberman" , "Nicholas A. Bellinger"
>
> Cc: "Bart Van Assche" , "linux-scsi"
> , "target-devel"
> , "Quinn Tran"
> Sent: Monday, May 9, 2016 1:08:36 PM
> Subject: Re: [PATCH] tcm_qla2xxx Add SCSI command jammer/discard c
On Wed, May 25, 2016 at 8:40 PM, Steve Wise wrote:
>> From: Or Gerlitz [mailto:gerlitz...@gmail.com]
>>
>> On Tue, May 24, 2016 at 9:40 AM, Nicholas A. Bellinger
>> wrote:
>> > Hi Or & Co,
>> > On Wed, 2016-05-18 at 14:45 +0300, Or Gerlitz wrote:
>> >> On Sat, Apr 30, 2016 at 6:54 PM, Or Gerlitz
On Thu, Apr 28 2016 at 11:40am -0400,
James Bottomley wrote:
> On Thu, 2016-04-28 at 08:11 -0400, Mike Snitzer wrote:
> > Full disclosure: I'll be looking at reinstating bio-based DM multipath to
> > regain efficiencies that now really matter when issuing IO to extremely
> > fast devices (e.g. NV
On Wed, 2016-05-25 at 16:12 -0400, Laurence Oberman wrote:
>
> Hi Nicholas,
>
> Can we pull this one in for submission now that its been acked'd by Himanshu.
> Thanks
> Laurence
This patch has been queued in target-pending/for-next for the last
weeks, and will be included in the v4.7-rc1 PULL
On Wed, 2016-05-25 at 23:41 +0300, Or Gerlitz wrote:
> On Wed, May 25, 2016 at 8:40 PM, Steve Wise
> wrote:
> >> From: Or Gerlitz [mailto:gerlitz...@gmail.com]
> >>
> .
> >
> > Hey Or and Nicholas,
> >
> > I've been staying out of this directly due to my workload. But I will now
> > take a lo
I think the function could be further cut down, given that
a) there is no need to take a lock to read a single field
b) CMD_T_FABRIC_STOP is never cleared and we can check it later
See the incremental patch below:
diff --git a/drivers/target/target_core_transport.c
b/drivers/target/target_co
33 matches
Mail list logo
