On Wed, 2017-09-27 at 21:35 -0400, Martin K. Petersen wrote:
> SBC-4 states:
>
> "A MAXIMUM UNMAP LBA COUNT field set to a non-zero value indicates
> the
> maximum number of LBAs that may be unmapped by an UNMAP command"
>
> "A MAXIMUM WRITE SAME LENGTH field set to a non-zero value
> indi
On 27.9.2017 23:29, Don Brace wrote:
> From: Kevin Barnett
>
> Update the driver’s PCI IDs
>
> Reviewed-by: Scott Benesh
> Reviewed-by: Scott Teel
> Signed-off-by: Kevin Barnett
> Signed-off-by: Don Brace
Reviewed-by: Tomas Henzl
On 27.9.2017 23:30, Don Brace wrote:
> From: Kevin Barnett
>
> Fix a small cosmetic bug in a very rarely encountered
> error message that can occur when a LD has a corrupted
> raid map.
>
> Reviewed-by: Scott Benesh
> Signed-off-by: Kevin Barnett
> Signed-off-by: Don Brace
Reviewed-by: Tomas H
On 27.9.2017 23:30, Don Brace wrote:
> Reviewed-by: Gerry Morong
> Reviewed-by: Scott Benesh
> Reviewed-by: Scott Teel
> Signed-off-by: Kevin Barnett
> Signed-off-by: Don Brace
Reviewed-by: Tomas Henzl
On Wed, 2017-09-27 at 21:38 -0400, Martin K. Petersen wrote:
> A user may lower the max_sectors_kb setting in sysfs to accommodate
> certain workloads. Previously we would always set the max I/O size to
> either the block layer default or the optional preferred I/O size
> reported by the device.
>
Laurence,
> I am testing this but its not being picked up so I want to know if I
> have the kernel command line wrong here.
>
> scsi_dev_flags=LIO-ORG:thin2:0x8000
>
> What am I doing wrong to pass the BLIST flags.
This worked for me:
[root@kvm ~]# echo "Linux:scsi_debug:0x8000" > /proc
On Mon, 2017-09-25 at 15:28 -0400, Martin K. Petersen wrote:
> Xin,
>
> > ChunYu found a kernel crash by syzkaller:
>
> [...]
>
> > It's caused by skb_shared_info at the end of sk_buff was overwritten by
> > ISCSI_KEVENT_IF_ERROR when parsing nlmsg info from skb in iscsi_if_rx.
> >
> > During th
> -Original Message-
> From: linux-scsi-ow...@vger.kernel.org [mailto:linux-scsi-
> ow...@vger.kernel.org] On Behalf Of Martin K. Petersen
> Sent: Wednesday, September 27, 2017 8:39 PM
> To: linux-scsi@vger.kernel.org
> Cc: Martin K. Petersen
> Subject: [PATCH] scsi: sd: Do not override ma
On Fri, 2017-09-29 at 09:21 -0400, Martin K. Petersen wrote:
> Laurence,
>
> > I am testing this but its not being picked up so I want to know if
> > I
> > have the kernel command line wrong here.
> >
> > scsi_dev_flags=LIO-ORG:thin2:0x8000
> >
> > What am I doing wrong to pass the BLIST fla
Ming Lei - 27.09.17, 16:27:
> On Wed, Sep 27, 2017 at 09:57:37AM +0200, Martin Steigerwald wrote:
> > Hi Ming.
> >
> > Ming Lei - 27.09.17, 13:48:
> > > Hi,
> > >
> > > The current SCSI quiesce isn't safe and easy to trigger I/O deadlock.
> > >
> > > Once SCSI device is put into QUIESCE, no new
Noticed these don't seem to be in 4.14/scsi-queue
On Tue, Aug 29, 2017 at 6:45 PM, Martin K. Petersen
wrote:
>
> Chris,
>
>> Looks good to me, fixes up the code given that the comment there about
>> calling iscsi_remove_session wasn't being followed.
>
> Applied these two to 4.14/scsi-queue.
>
>
From: Dick Kennedy
During pci hot plug, the kernel crashes in timer management code.
The sli4 remove_one handler is not stoping the timers as it starts to
remove the port so that it can be swapped.
Fix: Stop the timers early in the handler routine.
Note: Fix in SLI-4 only. SLI-3 already stoppe
This patch set provides a number of bug fixes and additions to
the driver.
The patches were cut against the Martin's 4.14/scsi-queue tree.
There are no outside dependencies.
v2:
Split patch 1 into separate patches
Added stable cc's where requested
"Fix oops of nvme host during driver unload
From: Dick Kennedy
The driver crashes when attempting to use a freed ndpl pointer.
The pci_remove_one handler runs on a separate kernel thread. The order
of the removal is starting by freeing all of the ndlps and then
disabling interrupts. In between these two events the driver can still
receive
From: Dick Kennedy
Need to make ktime samples more accurate
If ktime is turned on in the middle of an IO, the max calculation
could be misleading. Base sampling on the start time of the IO
as opposed to ktime_on.
Make ISR ktime timestamps be from when CQE is read instead of EQE.
Added additiona
From: Dick Kennedy
Local Reject/Invalid RPI errors seen during discovery.
Temporary RPI cleanup was occurring regardless of SLI rev. It's
only necessary on SLI-4.
Adjust the test for whether cleanup is necessary.
Signed-off-by: Dick Kennedy
Signed-off-by: James Smart
Reviewed-by: Johannes Th
From: Dick Kennedy
The driver is seeing a NULL pointer in lpfc_nvme_fcp_io_submit.
This was ultimately due to a transport AER being sent on a terminated
controller, thus some of the values were not set. In case we're in a
system without a corrected transport and in case a race condition
occurs wh
From: Dick Kennedy
The driver is encountering oops in lpfc_sli_calc_ring.
The driver is setting hba_wqidx for FCP based on the
policy in use for NVME. The two may not be the same.
Change to set the wqidx based on the FCP policy.
Cc: # 4.12+
Signed-off-by: Dick Kennedy
Signed-off-by: James Sm
From: Dick Kennedy
During pci hot plug, the kernel crashes in a list_add_call
The lookup by tag function will return null if the IOCB is out of
range or does not have the on txcmplq flag set.
Fix: Check for null return from lookup by tag.
Cc: # 4.12+
Signed-off-by: Dick Kennedy
Signed-off-by
From: Dick Kennedy
Warning messages when NVME_TARGET_FC not defined on ppc builds
The lpfc_nvmet_replenish_context() function is only meaningful when
NVME target mode enabled. Surround the function body with ifdefs
for target mode enablement.
Signed-off-by: Dick Kennedy
Signed-off-by: James Sm
From: Dick Kennedy
In a link bounce scenario, a condition can occur where the
discovery engine swaps an ndlp structure (address changbe for
an nport). While the swap was successfully executed by the
discovery engine, the driver did not properly detect a change in
the ndlp bound to the nvme rport.
There are several log messages that report abnormal
terminations that by default are marked warn. These are
typically the result of failures due to invalid controller
state or abort completions. They are all natural when a
controller resets.
Unfortunately, as they are logged by default, it makes t
From: Dick Kennedy
Before releasing nvme io back to the io stack for possible
retry on other paths, ensure the io termination is interlocked
with the target device by ensuring the entire ABTS-LS protocol is
complete.
Additionally, FC-NVME ABTS-LS protocol does not use RRQ. Remove RRQ
behavior fr
From: Dick Kennedy
Under heavy target nvme load duration, the lpfc irq handler
is encountering cpu lockup warnings.
Convert the driver to a shortened ISR handler which identifies
the interrupting condition then schedules a workq thread to
process the completion queue the interrupt was for. This
From: Dick Kennedy
Change version to 11.4.0.4
Signed-off-by: Dick Kennedy
Signed-off-by: James Smart
Reviewed-by: Johannes Thumshirn
---
drivers/scsi/lpfc/lpfc_version.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/scsi/lpfc/lpfc_version.h b/drivers/scsi/lpfc/l
From: Dick Kennedy
Firmware update fails with: status x17 add_status x56 on the final write
If multiple DMA buffers are used for the download, some firmware revs
have difficulty with signatures and crcs split across the dma buffer
boundaries. Resolve by making all writes be a single 4k page in
From: Dick Kennedy
Support RDP and Multiple Frames
If the remote Nport is not logged in, the driver would not populate
all the descriptors in the RDP response payload. Doing so would create
a payload length that requires multiple frames due to exceeding the
default rx buffer size without an expl
From: Dick Kennedy
When running NVME io as a NVME host, if the driver is unloaded there
would be oops in lpfc_sli4_issue_wqe.
When unloading, controllers are torn down and the transport initiates
set_property commands to reset the controller and issues aborts to
terminate existing io. The drive
Always set ctxp->state to LPFC_NVMET_STE_ABORT if ABORT op gets called
Signed-off-by: Dick Kennedy
Signed-off-by: James Smart
Reviewed-by: Johannes Thumshirn
---
drivers/scsi/lpfc/lpfc_nvmet.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/scsi/lpfc/lpfc_nvmet.c b/drivers/scsi/lpf
The internal cfg flag is actually smaller, by 1 (for a partial page sge),
than the sg list maintained by the driver. Thus the check on sg segments
errored out when it shouldn't have
Ensure the check is +1
Note: having a value that is less than what it really is is bogus.
Correcting it now would
From: Dick Kennedy
The initial implementation of NVME didn't merge with NPIV support.
As such, there are several issues if NPIV is used with NVME. For now,
ensure that if NVME is enabled then NPIV is not enabled.
Support for NPIV with NVME will be added in the near future.
Signed-off-by: Dick K
From: Dick Kennedy
if nvmet targetport registration fails, the driver encounters a
NULL pointer oops in lpfc_hb_timeout_handler.
To fix: if registration fails, ensure nvmet_support is cleared
on the port structure.
Also enhanced the log message on failure.
Cc: # 4.12+
Signed-off-by: Dick Kenn
From: Dick Kennedy
The descriptions for lpfc_xri_split and lpfc_enable_fc4_type were
poor. Revise for better understanding:
lpfc_xri_split - Percentage of FCP XRI resources versus NVME
lpfc_enable_fc4_type - Enable FC4 Protocol support - FCP / NVME
Signed-off-by: Dick Kennedy
Signed-off-by:
We need to pass PREEMPT flags to blk_queue_enter()
for allocating request with RQF_PREEMPT in the
following patch.
Tested-by: Oleksandr Natalenko
Tested-by: Martin Steigerwald
Cc: Bart Van Assche
Signed-off-by: Ming Lei
---
block/blk-core.c | 10 ++
block/blk-mq.c | 5 +
This usage is basically same with blk-mq, so that we can
support to freeze legacy queue easily.
Also 'wake_up_all(&q->mq_freeze_wq)' has to be moved
into blk_set_queue_dying() since both legacy and blk-mq
may wait on the wait queue of .mq_freeze_wq.
Tested-by: Oleksandr Natalenko
Tested-by: Mart
Hi Jens,
Please consider this patchset for V4.15, and it fixes one
kind of long-term I/O hang issue in either block legacy path
or blk-mq.
The current SCSI quiesce isn't safe and easy to trigger I/O deadlock.
Once SCSI device is put into QUIESCE, no new request except for
RQF_PREEMPT can be disp
When queue is in PREEMPT_ONLY mode, only RQF_PREEMPT request
can be allocated and dispatched, other requests won't be allowed
to enter I/O path.
This is useful for supporting safe SCSI quiesce.
Part of this patch is from Bart's '[PATCH v4 4∕7] block: Add the
QUEUE_FLAG_PREEMPT_ONLY
request queue
Simply quiesing SCSI device and waiting for completeion of IO
dispatched to SCSI queue isn't safe, it is easy to use up
request pool because all allocated requests before can't
be dispatched when device is put in QIUESCE. Then no request
can be allocated for RQF_PREEMPT, and system may hang somewhe
REQF_PREEMPT is a bit special because the request is required
to be dispatched to lld even when SCSI device is quiesced.
So this patch introduces __blk_get_request() and allows users to pass
RQF_PREEMPT flag in, then we can allow to allocate request of RQF_PREEMPT
when queue is in mode of PREEMPT
This patch just makes it explicitely.
Tested-by: Oleksandr Natalenko
Tested-by: Martin Steigerwald
Reviewed-by: Johannes Thumshirn
Cc: Bart Van Assche
Signed-off-by: Ming Lei
---
block/blk-mq.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/block/blk-mq.c b/block/blk-m
Since we have coverted all the user of UFS_BIT() macro
with the actual bit position, let remove unused UFS_BIT()macro.
Signed-off-by: Alim Akhtar
---
drivers/scsi/ufs/ufshci.h | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/scsi/ufs/ufshci.h b/drivers/scsi/ufs/ufshci.h
index 5a60a8f..
Use actaul bit position instead of UFS_BIT() macro. This
patch also changes bit-17 to meaningful #define.
This change is as per discussion here [1]
[1] -> https://lkml.org/lkml/2017/8/28/786
Signed-off-by: Alim Akhtar
Cc: Subhash Jadavani
---
This patch is only complied tested, appreciate tes
Currently UFS HCI uses UFS_BIT() macro to get various bit
position for the hardware registers status bits. Which makes
code longer instead of shorter. This macro does not improve
code readability as well.
Lets re-write these macro definition with the actual bit position.
Suggested-by: Bart Van Ass
Hi Martin,
On 09/26/2017 01:01 AM, Martin K. Petersen wrote:
>
> Alim,
>
>> Should I drop this patch and send another one which removes UFS_BIT()
>> macro?
>
> I fail to see the point of UFS_BIT(). So yes.
>
> Please make sure to CC: Subhash on ufs changes.
>
Thanks for looking into this patc
44 matches
Mail list logo
