Re: [bug report] scsi: mpt3sas: Added support for nvme encapsulated request message.

On Tue, Nov 07, 2017 at 10:29:20AM -0700, Sathya Prakash Veerichetty wrote: > Dan, > The MPI structures are of variable length and can go up to a maximum of > 128 bytes (a MPI frame size) and as MPI standard the variable length MPI > structures are left out with the last element as a single dword

Re: [PATCH] sd: Fix a disk probing hang

On 11/07/2017 11:57 PM, James Bottomley wrote: > On Tue, 2017-11-07 at 22:42 +, Bart Van Assche wrote: >> On Tue, 2017-11-07 at 10:09 -0800, James Bottomley wrote: >>> >>> but can you investigate the root cause rather than trying this >>> bandaid? >> >> Hello James, >> >> Thanks for your

[PATCH 1/2] tcmu: Fix some memory corruption

"udev->nl_reply_supported" is an int but on 64 bit arches we are writing 8 bytes of data to it so it corrupts four bytes beyond the end of the struct. Fixes: b849b4567549 ("target: Add netlink command reply supported option for each device") Signed-off-by: Dan Carpenter

[PATCH V7 3/8] block: mq-deadline: Introduce zone locking support

Introduce zone write locking to avoid write request reordering with zoned block devices. This is achieved using a finer selection of the next request to dispatch: 1) Any non-write request is always allowed to proceed. 2) Any write to a conventional zone is always allowed to proceed. 3) For a write

[PATCH V7 8/8] scsi: sd_zbc: Fix zone information initialization

Make sure that the device request queue zone information (number of zones and zone bitmaps) are reinitialized if the number of zones changes (e.g. on a drive capacity change on revalidate). Signed-off-by: Damien Le Moal --- drivers/scsi/sd_zbc.c | 36

[PATCH V7 1/8] block: introduce zoned block devices zone write locking

From: Christoph Hellwig Components relying only on the request_queue structure for accessing block devices (e.g. I/O schedulers) have a limited knowledged of the device characteristics. In particular, the device capacity cannot be easily discovered, which for a zoned block device

[PATCH V7 6/8] scsi: sd_zbc: Initialize device request queue zoned data

Initialize the seq_zone_bitmap and nr_zones fields of the disk request queue on disk revalidate. As the seq_zone_bitmap allocation is identical to the allocation of the zone write lock bitmap, introduce the helper sd_zbc_alloc_zone_bitmap(). Using this helper, wait for the disk capacity and number

[PATCH V7 7/8] scsi: sd: Remove zone write locking

The block layer now handles zone write locking. Signed-off-by: Damien Le Moal --- drivers/scsi/sd.c| 41 +++--- drivers/scsi/sd.h| 11 - drivers/scsi/sd_zbc.c| 105 +++

[PATCH V7 4/8] block: deadline-iosched: Introduce dispatch helpers

Avoid directly referencing the next_rq and fifo_list arrays using the helper functions deadline_next_request() and deadline_fifo_request() to facilitate changes in the dispatch request selection in deadline_dispatch_requests() for zoned block devices. While at it, also remove the unnecessary

[PATCH V7 2/8] blokc: mq-deadline: Introduce dispatch helpers

Avoid directly referencing the next_rq and fifo_list arrays using the helper functions deadline_next_request() and deadline_fifo_request() to facilitate changes in the dispatch request selection in __dd_dispatch_request() for zoned block devices. Signed-off-by: Damien Le Moal

[PATCH V7 0/8] blk-mq support for ZBC disks

This series, formerly titled "scsi-mq support for ZBC disks", implements support for ZBC disks for system using the scsi-mq I/O path. The current scsi level support of ZBC disks guarantees write request ordering using a per-zone write lock which prevents issuing simultaneously multiple write

[PATCH V7 5/8] block: deadline-iosched: Introduce zone locking support

Introduce zone write locking to avoid write request reordering with zoned block devices. This is achieved using a finer selection of the next request to dispatch: 1) Any non-write request is always allowed to proceed. 2) Any write to a conventional zone is always allowed to proceed. 3) For a write

[PATCH 1/2] scsi: mpt3sas: cleanup _scsih_pcie_enumeration_event()

The indenting wasn't right, because the last two prints weren't indented far enough. Also it used pr_info() where it was supposed to use pr_cont(). I reversed the if statement and pulled the code in one tab and did a couple other minor cleanups. Fixes: 4318c7347847 ("scsi: mpt3sas: Handle NVMe

[patch 2/2] scsi: mpt3sas: remove a stray KERN_INFO

pr_info() has a KERN_INFO already so the second KERN_INFO isn't needed. Signed-off-by: Dan Carpenter diff --git a/drivers/scsi/mpt3sas/mpt3sas_scsih.c b/drivers/scsi/mpt3sas/mpt3sas_scsih.c index 3a9438a1704e..b258f210120a 100644 ---

[PATCH 2/2] tcmu: Add a missing unlock on an error path

We added a new error path here but we forgot to drop the lock first before returning. Fixes: 0d44374c1aae ("tcmu: fix double se_cmd completion") Signed-off-by: Dan Carpenter diff --git a/drivers/target/target_core_user.c b/drivers/target/target_core_user.c index

[PATCH V3] scsi_debugfs: fix crash in scsi_show_rq()

cmd->cmnd can be allocated/freed dynamically in case of T10_PI_TYPE2_PROTECTION, so we should check it in scsi_show_rq() because this request may have been freed in scsi_show_rq(). This patch fixs the following kernel crash when dumping request via block's debugfs interface: [ 252.962045] BUG:

RE: [PATCH 1/2] scsi: mpt3sas: cleanup _scsih_pcie_enumeration_event()

Thanks. Acked-by: Sathya Prakash Veerichetty -Original Message- From: Dan Carpenter [mailto:dan.carpen...@oracle.com] Sent: Wednesday, November 8, 2017 1:38 AM To: Sathya Prakash; Suganath Prabu Subramani Cc: Chaitra P B; James E.J. Bottomley; Martin K.

Re: [PATCH V3] scsi_debugfs: fix crash in scsi_show_rq()

On Wed, Nov 08, 2017 at 08:01:08AM -0800, James Bottomley wrote: > On Wed, 2017-11-08 at 21:21 +0800, Ming Lei wrote: > > cmd->cmnd can be allocated/freed dynamically in case of > > T10_PI_TYPE2_PROTECTION, > > so we should check it in scsi_show_rq() because this request may have > > been freed in

Re: [PATCH V3] scsi_debugfs: fix crash in scsi_show_rq()

On Wed, 2017-11-08 at 21:21 +0800, Ming Lei wrote: > cmd->cmnd can be allocated/freed dynamically in case of > T10_PI_TYPE2_PROTECTION, > so we should check it in scsi_show_rq() because this request may have > been freed in scsi_show_rq(). > > This patch fixs the following kernel crash when

RE: [PATCH] scsi: megaraid: Convert timers to use timer_setup()

> -Original Message- > From: Sumit Saxena [mailto:sumit.sax...@broadcom.com] > Sent: Wednesday, November 1, 2017 12:18 AM > To: Kees Cook; Martin K. Petersen > Cc: Kashyap Desai; Shivasharan Srikanteshwara; James E.J. Bottomley; > PDL,MEGARAIDLINUX; linux-scsi@vger.kernel.org; linux- >

Re: [PATCH] SCSI: don't get target/host busy_count in scsi_mq_get_budget()

On Wed, Nov 08, 2017 at 02:20:41PM +0800, Ming Lei wrote: > On Tue, Nov 07, 2017 at 08:17:59PM -0700, Jens Axboe wrote: > > On 11/07/2017 08:12 PM, Ming Lei wrote: > > > On Tue, Nov 07, 2017 at 08:01:48PM -0700, Jens Axboe wrote: > > >> On 11/07/2017 06:03 PM, Ming Lei wrote: > > >>> On Tue, Nov

[PATCH 2/3] zfcp: purely mechanical update using timer API, plus blank lines

erp_memwait only occurs in seldom memory pressure situations. The typical case never uses the associated timer and thus also does not need to initialize the timer. Also, we don't want to re-initialize the timer each time we re-use an erp_action in zfcp_erp_setup_act() [see also v4.14-rc7 commit

[PATCH 3/3] zfcp: drop open coded assignments of timer_list.function

The majority of requests is regular SCSI I/O on the hot path. Since these use a timeout owned by the block layer, zfcp does not use zfcp_fsf_req.timer. Hence, the very early unconditional and even incomplete (handler function yet unknown) timer initialization in zfcp_fsf_req_create() is not

[PATCH 1/3] zfcp: convert timers to use timer_setup()

From: Kees Cook In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Cc: Steffen Maier Cc: Benjamin Block

[PATCH 0/3] zfcp: timer_setup() refactoring feature for v4.15-rc1

Hi all, here is a small series for the timer_setup() refactoring of zfcp. We target it for the merge window to land in v4.15-rc1. Unfortunately, they don't seem to apply to the current state of neither James' misc branch nor Martin's 4.15/scsi-queue branch, because they depend on: v4.14-rc3

RE: [patch 2/2] scsi: mpt3sas: remove a stray KERN_INFO

Thanks. Acked-by: Sathya Prakash Veerichetty -Original Message- From: Dan Carpenter [mailto:dan.carpen...@oracle.com] Sent: Wednesday, November 8, 2017 1:38 AM To: Sathya Prakash; Suganath Prabu Subramani Cc: Chaitra P B; James E.J. Bottomley; Martin K.

Re: [PATCH] sd: Fix a disk probing hang

On Wed, 2017-11-08 at 09:12 +0100, Hannes Reinecke wrote: > You know, this actually looks like the same issue I'm chasing with iser; > we have a customer who regularly sees lockups during scanning. > As it turns out, iser is calling scsi_device_del() from the RX thread. > Which in turn needs to

Re: [PATCH] SCSI: don't get target/host busy_count in scsi_mq_get_budget()

On Tue, 2017-11-07 at 20:06 -0700, Jens Axboe wrote: > At this point, I have no idea what Bart's setup looks like. Bart, it > would be REALLY helpful if you could tell us how you are reproducing > your hang. I don't know why this has to be dragged out. Hello Jens, It is a disappointment to me

Re: [PATCH 1/4] qla2xxx_nvmet: Add files for FC-NVMe Target support

Hi James, > On Nov 7, 2017, at 9:05 AM, James Smart wrote: > > On 11/6/2017 11:55 AM, Himanshu Madhani wrote: >> From: Anil Gurumurthy >> >> +static struct nvmet_fc_target_template qla_nvmet_fc_transport = { >> +.targetport_delete

Re: [PATCH] SCSI: don't get target/host busy_count in scsi_mq_get_budget()

On 11/08/2017 09:41 AM, Bart Van Assche wrote: > On Tue, 2017-11-07 at 20:06 -0700, Jens Axboe wrote: >> At this point, I have no idea what Bart's setup looks like. Bart, it >> would be REALLY helpful if you could tell us how you are reproducing >> your hang. I don't know why this has to be

Re: [PATCH V7 5/8] block: deadline-iosched: Introduce zone locking support

Looks good, Reviewed-by: Christoph Hellwig

Re: [PATCH V7 3/8] block: mq-deadline: Introduce zone locking support

Looks good: Reviewed-by: Christoph Hellwig

Re: [PATCH V7 4/8] block: deadline-iosched: Introduce dispatch helpers

Looks good, Reviewed-by: Christoph Hellwig

Re: [PATCH] SCSI: don't get target/host busy_count in scsi_mq_get_budget()

On Wed, 2017-11-08 at 10:57 -0700, Jens Axboe wrote: > On 11/08/2017 09:41 AM, Bart Van Assche wrote: > > On Tue, 2017-11-07 at 20:06 -0700, Jens Axboe wrote: > > > At this point, I have no idea what Bart's setup looks like. Bart, > > > it > > > would be REALLY helpful if you could tell us how you

Re: [PATCH] SCSI: don't get target/host busy_count in scsi_mq_get_budget()

On 11/08/2017 11:22 AM, Laurence Oberman wrote: > On Wed, 2017-11-08 at 10:57 -0700, Jens Axboe wrote: >> On 11/08/2017 09:41 AM, Bart Van Assche wrote: >>> On Tue, 2017-11-07 at 20:06 -0700, Jens Axboe wrote: At this point, I have no idea what Bart's setup looks like. Bart, it

Re: [PATCH] SCSI: don't get target/host busy_count in scsi_mq_get_budget()

On 11/08/2017 08:59 AM, Ming Lei wrote: > On Wed, Nov 08, 2017 at 02:20:41PM +0800, Ming Lei wrote: >> On Tue, Nov 07, 2017 at 08:17:59PM -0700, Jens Axboe wrote: >>> On 11/07/2017 08:12 PM, Ming Lei wrote: On Tue, Nov 07, 2017 at 08:01:48PM -0700, Jens Axboe wrote: > On 11/07/2017 06:03

Re: [PATCH V7 2/8] blokc: mq-deadline: Introduce dispatch helpers

Except for a bit of mis-spelling in the Subject this looks fine to me: Reviewed-by: Christoph Hellwig FYI, I would just simplify the subject to mq-deadline: Introduce dispatch helpers and similar in the other patches.

Re: [PATCH] blk-mq: Avoid that request queue removal can trigger list corruption

On 11/08/2017 11:23 AM, Bart Van Assche wrote: > Avoid that removal of a request queue sporadically triggers the > following warning: > > list_del corruption. next->prev should be 8807d649b970, but was > 6b6b6b6b6b6b6b6b > WARNING: CPU: 3 PID: 342 at lib/list_debug.c:56 >

[PATCH] scsi: ppa: mark expected switch fall-throughs

In preparation to enabling -Wimplicit-fallthrough, mark switch cases where we are expecting to fall through. Addresses-Coverity-ID: 114988 Addresses-Coverity-ID: 114989 Addresses-Coverity-ID: 114990 Addresses-Coverity-ID: 114991 Signed-off-by: Gustavo A. R. Silva ---

Re: [PATCH V7 6/8] scsi: sd_zbc: Initialize device request queue zoned data

A bit of duplication vs the existing ZBC code here, but it seems the old code is going away later in the series, so: Reviewed-by: Christoph Hellwig

[PATCH] blk-mq: Avoid that request queue removal can trigger list corruption

Avoid that removal of a request queue sporadically triggers the following warning: list_del corruption. next->prev should be 8807d649b970, but was 6b6b6b6b6b6b6b6b WARNING: CPU: 3 PID: 342 at lib/list_debug.c:56 __list_del_entry_valid+0x92/0xa0 Call Trace: process_one_work+0x11b/0x660

Re: [PATCH V7 8/8] scsi: sd_zbc: Fix zone information initialization

Looks good: Reviewed-by: Christoph Hellwig

Re: [PATCH V7 7/8] scsi: sd: Remove zone write locking

Looks good: Reviewed-by: Christoph Hellwig

Re: [PATCH] lpfc: Fix hard lock up NMI in els timeout handling.

On Tue, 2017-11-07 at 12:59 -0800, James Smart wrote: > From: Dick Kennedy > > System crashed due to a hard lockup at lpfc_els_timeout_handler+0x128. > > The els ring's txcmplq list is corrupted: the last element in the list > does not point back the the head causing

Re: [PATCH 01/17] lpfc: FLOGI failures are reported when connected to a private loop.

On 11/8/2017 1:05 AM, Hannes Reinecke wrote: Isn't that a bit excessive? Printing the same message _two_ times, and not providing additional info? Please do only duplicate message if there is new information printed; otherwise we might as well skip the second message. Cheers, Hannes yep. will

Re: [PATCH 3/6] target: Fix quiese during transport_write_pending_qf endless loop

> From: Nicholas Bellinger > > This patch fixes a potential end-less loop during QUEUE_FULL, > where cmd->se_tfo->write_pending() callback fails repeatedly > but __transport_wait_for_tasks() has already been invoked to > quiese the outstanding se_cmd descriptor. > > To

Re: [PATCH 1/6] target: Fix QUEUE_FULL + SCSI task attribute handling

> From: Nicholas Bellinger > > This patch fixes a bug during QUEUE_FULL where transport_complete_qf() > calls transport_complete_task_attr() after it's already been invoked > by target_complete_ok_work() or transport_generic_request_failure() > during initial completion,

Re: [PATCHv2] tcmu: Add fifo type waiter list support to avoid starvation

On 7/30/17 5:10 PM, Nicholas A. Bellinger wrote: > Hi Xiubo, > > Apologies for the delayed response. Comments below. > > On Wed, 2017-07-12 at 15:16 +0800, lixi...@cmss.chinamobile.com wrote: >> From: Xiubo Li >> >> The fifo type waiter list will hold the udevs

Re: [PATCH] tcmu: clean up the scatter helper

On 7/30/17 5:19 PM, Nicholas A. Bellinger wrote: > On Thu, 2017-07-13 at 14:33 +0800, lixi...@cmss.chinamobile.com wrote: >> From: Xiubo Li >> >> Add some comments to make the scatter code to be more readable. >> >> Signed-off-by: Xiubo Li

Re: [PATCH 0/3] zfcp: timer_setup() refactoring feature for v4.15-rc1

On Wed, Nov 8, 2017 at 6:17 AM, Steffen Maier wrote: > Hi all, > > here is a small series for the timer_setup() refactoring of zfcp. > We target it for the merge window to land in v4.15-rc1. > > Unfortunately, they don't seem to apply to the current state of neither >

Re: [PATCH] tcmu: Oops in unmap_thread_fn()

On 8/6/17 6:27 PM, Nicholas A. Bellinger wrote: > On Tue, 2017-08-01 at 23:09 +0300, Dan Carpenter wrote: >> Calling list_del() on the iterator pointer in list_for_each_entry() will >> cause an oops. We need to user the _safe() version for that. >> >> Fixes: c73d02f63c16 ("tcmu: Add fifo type

Re: [PATCHv2] tcmu: Add fifo type waiter list support to avoid starvation

On 11/08/2017 04:35 PM, Bryant G. Ly wrote: > > On 7/30/17 5:10 PM, Nicholas A. Bellinger wrote: > >> Hi Xiubo, >> >> Apologies for the delayed response. Comments below. >> >> On Wed, 2017-07-12 at 15:16 +0800, lixi...@cmss.chinamobile.com wrote: >>> From: Xiubo Li

Re: [patch 2/2] scsi: mpt3sas: remove a stray KERN_INFO

Dan, > pr_info() has a KERN_INFO already so the second KERN_INFO isn't needed. > > Signed-off-by: Dan Carpenter Applied to 4.15/scsi-queue. -- Martin K. Petersen Oracle Linux Engineering

Re: [PATCH] lpfc: Fix hard lock up NMI in els timeout handling.

James, > System crashed due to a hard lockup at lpfc_els_timeout_handler+0x128. > > The els ring's txcmplq list is corrupted: the last element in the list > does not point back the the head causing a loop. Issue is the > els processing path for sli4 hbas are using the hbalock instead of > the

Re: [PATCH V7 6/8] scsi: sd_zbc: Initialize device request queue zoned data

Martin, On 11/9/17 08:41, Martin K. Petersen wrote: > > Damien, > >> wait for the disk capacity and number of zones to stabilize on the >> second revalidation pass to allocate and initialize the bitmaps. > > Stabilize how? If RC_BASIS is 0, the capacity changes after the first report zones...

Re: [PATCH] tcmu: clean up the scatter helper

On 11/08/2017 04:39 PM, Bryant G. Ly wrote: > On 7/30/17 5:19 PM, Nicholas A. Bellinger wrote: > >> On Thu, 2017-07-13 at 14:33 +0800, lixi...@cmss.chinamobile.com wrote: >>> From: Xiubo Li >>> >>> Add some comments to make the scatter code to be more readable. >>>

Re: [PATCH 1/2] scsi: mpt3sas: cleanup _scsih_pcie_enumeration_event()

Dan, > The indenting wasn't right, because the last two prints weren't > indented far enough. Also it used pr_info() where it was supposed to > use pr_cont(). I reversed the if statement and pulled the code in one > tab and did a couple other minor cleanups. Applied to 4.15/scsi-queue. Thank

Re: [PATCH V7 4/8] block: deadline-iosched: Introduce dispatch helpers

Damien, > Avoid directly referencing the next_rq and fifo_list arrays using the > helper functions deadline_next_request() and deadline_fifo_request() to > facilitate changes in the dispatch request selection in > deadline_dispatch_requests() for zoned block devices. > > While at it, also remove

Re: [PATCH V7 5/8] block: deadline-iosched: Introduce zone locking support

Damien, > Introduce zone write locking to avoid write request reordering with > zoned block devices. This is achieved using a finer selection of the > next request to dispatch: > 1) Any non-write request is always allowed to proceed. > 2) Any write to a conventional zone is always allowed to

Re: [PATCH V7 3/8] block: mq-deadline: Introduce zone locking support

Damien, > Introduce zone write locking to avoid write request reordering with > zoned block devices. This is achieved using a finer selection of the > next request to dispatch: > 1) Any non-write request is always allowed to proceed. > 2) Any write to a conventional zone is always allowed to

Re: [PATCH V7 6/8] scsi: sd_zbc: Initialize device request queue zoned data

Damien, > wait for the disk capacity and number of zones to stabilize on the > second revalidation pass to allocate and initialize the bitmaps. Stabilize how? -- Martin K. Petersen Oracle Linux Engineering

Re: [PATCH] scsi_transport_fc: add 64GBIT and 128GBIT port speed definitions

James, > Add 64GBIT and 128GBIT port speed definitions. > > Upcoming hardware will reference these speeds. Applied to 4.15/scsi-queue. Thanks! -- Martin K. Petersen Oracle Linux Engineering

Re: [PATCH] aacraid: use timespec64 instead of timeval

Arnd, > aacraid passes the current time to the firmware in one of two ways, > either as year/month/day/... or as 32-bit unsigned seconds. > > The first one is broken on 32-bit architectures as it cannot go past > year 2038. Using timespec64 here makes it behave properly on both > 32-bit and

Re: [PATCH 0/3] zfcp: timer_setup() refactoring feature for v4.15-rc1

Kees, >> Kees Cook (1): >> zfcp: convert timers to use timer_setup() >> >> Steffen Maier (2): >> zfcp: purely mechanical update using timer API, plus blank lines >> zfcp: drop open coded assignments of timer_list.function > > These looks great, thanks! These look good to me too.

Re: [PATCH V7 1/8] block: introduce zoned block devices zone write locking

Damien, > + * Zoned block device information for request dispatch control. > + * nr_zones is the total number of zones of the device. This is always > + * 0 for regular block devices. seq_zones_bitmap is a bitmap of nr_zones > + * bits which indicates if a zone is

Re: [PATCH V7 7/8] scsi: sd: Remove zone write locking

Damien, > The block layer now handles zone write locking. Looks OK. Reviewed-by: Martin K. Petersen -- Martin K. Petersen Oracle Linux Engineering

Re: [PATCH V7 8/8] scsi: sd_zbc: Fix zone information initialization

Damien, > Make sure that the device request queue zone information (number of > zones and zone bitmaps) are reinitialized if the number of zones > changes (e.g. on a drive capacity change on revalidate). This should probably be part of patch 6. Reviewed-by: Martin K. Petersen

[target:for-next 17/33] drivers/target/target_core_user.c:891:2-8: preceding lock on line 791 (fwd)

Please check whether an unlock is needed before the return on line 891. julia -- Forwarded message -- Date: Wed, 8 Nov 2017 15:05:33 +0800 From: kbuild test robot To: kbu...@01.org Cc: Julia Lawall Subject: [target:for-next 17/33]

Re: [PATCH] tcmu: clean up the scatter helper

> On 11/08/2017 04:39 PM, Bryant G. Ly wrote: >> On 7/30/17 5:19 PM, Nicholas A. Bellinger wrote: >> >>> On Thu, 2017-07-13 at 14:33 +0800, lixi...@cmss.chinamobile.com wrote: From: Xiubo Li Add some comments to make the scatter code to be more

Re: [PATCH V7 2/8] blokc: mq-deadline: Introduce dispatch helpers

Damien, > Avoid directly referencing the next_rq and fifo_list arrays using the > helper functions deadline_next_request() and deadline_fifo_request() > to facilitate changes in the dispatch request selection in > __dd_dispatch_request() for zoned block devices. Reviewed-by: Martin K. Petersen

Re: [PATCH 01/17] lpfc: FLOGI failures are reported when connected to a private loop.

On 11/03/2017 11:56 PM, James Smart wrote: > When the HBA is connected to a private loop, the driver > reports FLOGI loop-open failure as functional error. This is > an expected condition. > > Mark loop-open failure as a warning instead of error. > > Signed-off-by: Dick Kennedy

Re: [PATCH 02/17] lpfc: Expand WQE capability of every NVME hardware queue

On 11/03/2017 11:56 PM, James Smart wrote: > Hardware queues are a fast staging area to push commands into the adapter. > The adapter should drain them extremely quickly. However, under heavy > io load, the host cpu is pushing commands faster than the drain rate of > the adapter causing the driver

Re: [PATCH 03/17] lpfc: Handle XRI_ABORTED_CQE in soft IRQ

On 11/03/2017 11:56 PM, James Smart wrote: > XRI_ABORTED_CQE completions were not being handled in the fast path. > They were being queued and deferred to the lpfc worker thread > for processing. This is an artifact of the driver design prior > to moving queue processing out of the isr and into a

Re: [PATCH 06/17] lpfc: Raise maximum NVME sg list size for 256 elements

On 11/03/2017 11:56 PM, James Smart wrote: > Raise the maximum NVME sg list size allowed to 256 elements. > > Signed-off-by: Dick Kennedy > Signed-off-by: James Smart > --- > drivers/scsi/lpfc/lpfc.h | 2 +- > 1 file changed, 1 insertion(+),

Re: [PATCH 05/17] lpfc: Fix NVME LS abort_xri

On 11/03/2017 11:56 PM, James Smart wrote: > performing an LS abort results in the following message being seen: > 0603 Invalid CQ subtype 6: 0300 2202 0016 d005 > and the associated exchange is not properly freed. > > The code did not recognize the exchange type that was

Re: [PATCH 04/17] lpfc: Fix crash after bad bar setup on driver attachment

On 11/03/2017 11:56 PM, James Smart wrote: > In test cases where an instance of the driver is detached and > reattached, the driver will crash on reattachment. There is a > compound if statement that will skip over the bar setup if > the pci_resource_start call is not successful. The driver >

Re: [PATCH 08/17] lpfc: Fix display for debugfs queInfo

On 11/03/2017 11:56 PM, James Smart wrote: > Display for lpfc/fnX/iDiag/queInfo isn't formatted perfectly. > Corrected the format strings for the queue info debug messages. > > Signed-off-by: Dick Kennedy > Signed-off-by: James Smart > --- >

Re: [PATCH 07/17] lpfc: Driver fails to detect direct attach storage array

On 11/03/2017 11:56 PM, James Smart wrote: > The driver does not respond to PLOGI from the direct attach target. > The driver uses incorrect S_ID in CONFIG_LINK, after FLOGI completion > > Correct by issuing CONFIG_LINK with the correct S_ID after receiving > the PLOGI from the target > >

Re: target-pending/for-next patches

On Sun, 2017-11-05 at 08:05 -0800, James Bottomley wrote: > On Sat, 2017-11-04 at 18:14 -0700, Nicholas A. Bellinger wrote: > > Hi all, > > > > Just a friendly email after catching up on patches this week, the > > majority of those outstanding on the list have been merged into > >

Re: [PATCH 09/17] lpfc: Adjust default value of lpfc_nvmet_mrq

On 11/03/2017 11:56 PM, James Smart wrote: > The current default for async hw receive queues is 1, which presents > issues under heavy load as number of queues influence the available > async receive buffer limits. > > Raise the default to the either the current hw limit (16) or the number > of

isci regression in 4.11.0-rc2 by scsi: libsas: allow async aborts

Hello, after upgrading some servers from 4.9 to a kernel > 4.11 i saw a regression where i cannot boot into the os. The kernel is crashing in the isci (NOT iscsi) driver in task abort. Message: Unable to handle kernel NULL pointer dereference at isci_task_abort_task git bisect detected the

Re: [PATCH 1/2] tcmu: Fix some memory corruption

On Wed, 2017-11-08 at 11:43 +0300, Dan Carpenter wrote: > "udev->nl_reply_supported" is an int but on 64 bit arches we are writing > 8 bytes of data to it so it corrupts four bytes beyond the end of the > struct. > > Fixes: b849b4567549 ("target: Add netlink command reply supported option for >

Re: [PATCH 2/2] tcmu: Add a missing unlock on an error path

On Wed, 2017-11-08 at 11:44 +0300, Dan Carpenter wrote: > We added a new error path here but we forgot to drop the lock first > before returning. > > Fixes: 0d44374c1aae ("tcmu: fix double se_cmd completion") > Signed-off-by: Dan Carpenter Applied to

Re: [PATCH] scsi/eh: fix hang adding ehandler wakeups after decrementing host_busy

Are there any issues with this patch (https://patchwork.kernel.org/patch/9938919/) that Pavel Tikhomirov submitted back in September? I am willing to help if there's anything I can do to help get it accepted. The failing case I'm working on involves lots of servers with disk read/write

Re: target-pending/for-next patches

On Wed, 2017-11-08 at 01:37 -0800, Nicholas A. Bellinger wrote: > On Sun, 2017-11-05 at 08:05 -0800, James Bottomley wrote: > > > > On Sat, 2017-11-04 at 18:14 -0700, Nicholas A. Bellinger wrote: > > > > > > Hi all, > > > > > > Just a friendly email after catching up on patches this week, the >

Re: [PATCH v11 0/7] block, scsi, md: Improve suspend and resume

Bart, is this something known to you, or it is just my fault applying this series to v4.13? Except having this warning, suspend/resume works for me: === [ 27.383846] sd 0:0:0:0: [sda] Starting disk [ 27.383976] sd 1:0:0:0: [sdb] Starting disk [ 27.451218] sdb: Attempt to allocate

[PATCH V8 0/7] blk-mq support for ZBC disks

This series, formerly titled "scsi-mq support for ZBC disks", implements support for ZBC disks for system using the scsi-mq I/O path. The current scsi level support of ZBC disks guarantees write request ordering using a per-zone write lock which prevents issuing simultaneously multiple write

[PATCH V8 5/7] deadline-iosched: Introduce zone locking support

Introduce zone write locking to avoid write request reordering with zoned block devices. This is achieved using a finer selection of the next request to dispatch: 1) Any non-write request is always allowed to proceed. 2) Any write to a conventional zone is always allowed to proceed. 3) For a write

[PATCH V8 1/7] block: introduce zoned block devices zone write locking

From: Christoph Hellwig Components relying only on the request_queue structure for accessing block devices (e.g. I/O schedulers) have a limited knowledged of the device characteristics. In particular, the device capacity cannot be easily discovered, which for a zoned block device

[PATCH V8 7/7] sd: Remove zone write locking

The block layer now handles zone write locking. Signed-off-by: Damien Le Moal Reviewed-by: Christoph Hellwig Reviewed-by: Martin K. Petersen --- drivers/scsi/sd.c| 41 +++- drivers/scsi/sd.h|

[PATCH V8 3/7] mq-deadline: Introduce zone locking support

Introduce zone write locking to avoid write request reordering with zoned block devices. This is achieved using a finer selection of the next request to dispatch: 1) Any non-write request is always allowed to proceed. 2) Any write to a conventional zone is always allowed to proceed. 3) For a write

Re: [PATCH] SCSI: don't get target/host busy_count in scsi_mq_get_budget()

On Wed, Nov 08, 2017 at 04:41:35PM +, Bart Van Assche wrote: > On Tue, 2017-11-07 at 20:06 -0700, Jens Axboe wrote: > > At this point, I have no idea what Bart's setup looks like. Bart, it > > would be REALLY helpful if you could tell us how you are reproducing > > your hang. I don't know why

Re: [PATCH] SCSI: don't get target/host busy_count in scsi_mq_get_budget()

On Wed, Nov 08, 2017 at 10:57:23AM -0700, Jens Axboe wrote: > On 11/08/2017 09:41 AM, Bart Van Assche wrote: > > On Tue, 2017-11-07 at 20:06 -0700, Jens Axboe wrote: > >> At this point, I have no idea what Bart's setup looks like. Bart, it > >> would be REALLY helpful if you could tell us how you

[PATCH v2 3/4] qla2xxx_nvmet: Add FC-NVMe Target LS request handling

From: Anil Gurumurthy Signed-off-by: Anil Gurumurthy Signed-off-by: Giridhar Malavali Signed-off-by: Darren Trapp Signed-off-by: Himanshu Madhani ---

[PATCH v2 0/4] qla2xxx: Add FC-NVMe Target support

Hi Christoph/Sagi, I've corrected the mailing list and maintainers for this series. This series adds support for FC-NVMe Target. Patch #1 addes new qla_nvmet files for FC-NVMe Target support. Patch #2 adds Kconfig and Makefile changes to prepare code compile. Patch #3 and #4 has bulk of

[PATCH v2 1/4] qla2xxx_nvmet: Add files for FC-NVMe Target support

From: Anil Gurumurthy Signed-off-by: Anil Gurumurthy Signed-off-by: Giridhar Malavali Signed-off-by: Darren Trapp Signed-off-by: Himanshu Madhani ---

[PATCH v2 4/4] qla2xxx_nvmet: Add FC-NVMe Target handling

From: Anil Gurumurthy This patch Adds following code in the driver to support FC-NVMe Target - Updated ql2xenablenvme to allow FC-NVMe Target operation - Added LS Request handling for NVMe Target - Added passthr IOCB for LS4 request - Added CTIO for sending response

[PATCH v2 2/4] qla2xxx_nvmet: Added Makefile and Kconfig changes

From: Anil Gurumurthy Signed-off-by: Anil Gurumurthy Signed-off-by: Giridhar Malavali Signed-off-by: Himanshu Madhani --- drivers/scsi/qla2xxx/Kconfig | 1 +

[PATCH V8 2/7] mq-deadline: Introduce dispatch helpers

Avoid directly referencing the next_rq and fifo_list arrays using the helper functions deadline_next_request() and deadline_fifo_request() to facilitate changes in the dispatch request selection in __dd_dispatch_request() for zoned block devices. Signed-off-by: Damien Le Moal

[PATCH V8 4/7] deadline-iosched: Introduce dispatch helpers

Avoid directly referencing the next_rq and fifo_list arrays using the helper functions deadline_next_request() and deadline_fifo_request() to facilitate changes in the dispatch request selection in deadline_dispatch_requests() for zoned block devices. While at it, also remove the unnecessary

  1   2   >