From: Bart Van Assche
A value is assigned to the variable 'capacity' in sd_zbc_read_zones()
but that value is never used. Hence remove the variable 'capacity'.
Signed-off-by: Bart Van Assche
The variable 'ret' is initialized in
Hi Kishon,
On 02/28/2017 09:04 AM, Kishon Vijay Abraham I wrote:
> Hi,
>
> On Monday 27 February 2017 07:40 PM, Alim Akhtar wrote:
>> Hi Kishon,
>>
>> On 02/27/2017 10:56 AM, Kishon Vijay Abraham I wrote:
>>> Hi,
>>>
>>> On Thursday 23 February 2017 12:20 AM, Alim Akhtar wrote:
On Fri, Feb
On 02/28/2017 03:11 AM, Jeff Layton wrote:
<>
>
> I'll probably have questions about the read side as well, but for now it
> looks like it's mostly used in an ad-hoc way to communicate errors
> across subsystems (block to fs layer, for instance).
If memory does not fail me it used to be checked
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
From: Bart Van Assche
Commit f2e767bb5d6e ("mpt3sas: Force request partial completion
alignment") introduced a forced alignment of resid to the device
logical block size to fix bogus HBA firmware sometimes returning an
unaligned value. This fix however did not
On 02/27/2017 08:33 PM, Ewan D. Milne wrote:
> On Thu, 2017-02-23 at 11:27 +0100, Hannes Reinecke wrote:
>> The current medium access timeout counter will be increased for
>> each command, so if there are enough failed commands we'll hit
>> the medium access timeout for even a single failure.
>>
Remove most of the zone request specific code from sd_done() and
move it to sd_zbc_complete(), which is called at the end of sd_done()
so that good_bytes is set for REQ_OP_ZONE_REPORT and REQ_OP_ZONE_RESET.
In sd_zbc_complete(), enhance checks on the value of good_bytes for
report zones to avoid
2 patches in this series.
The first one from myself, cleans up and simplify sd_done() by moving all
ZBC specific processing to sd_zbc_complete().
sd_zbc_complete() as well as sd_zbc_report_zones_complete() are improved to
be more resilient against eventual weird values of good_bytes and resid
From: Bart Van Assche
Use blk_rq_accesses_medium() instead of !blk_rq_is_passthrough() to
ensure that code that is intended for normal medium access requests,
e.g. DISCARD, READ and WRITE requests, is not applied to
REQ_OP_ZONE_REPORT requests nor to REQ_OP_ZONE_RESET
On 02/27/2017 12:57 PM, Bart Van Assche wrote:
...
How about the (untested) patch below? The approach below avoids that the check
is
duplicated and - at least in my opinion - results in code that is easier to
read.
I find lba_too_large() a little dense, but functionally OK. The "shift
>= 0"
On Mon, Feb 27, 2017 at 05:19:31PM -0800, Stephen Hemminger wrote:
> Fixes: ee5242360424 ("scsi: zero per-cmd driver data before each I/O")
>
> but that is already in linux-next.
>
> Noticed another place where memset(of the data was being done not the extra
> bits.
> Tried this, but didn't fix
This series introduces blk_rq_accesses_medium(), which is equivalent to
!blk_rq_is_passthrough() minus the zone request operations REQ_OP_ZONE_REPORT
and REQ_OP_ZONE_RESET. This new helper allows avoiding problems due to the
non-standard nature of these commands (report zones does no operate on
On 28/02/17 4:32 PM, "Colin King" wrote:
>From: Colin Ian King
>
>The call to qedi_setup_int is not updating the return code rc yet rc
>is being checked for an error. Fix this by assigning rc to the return
>code from the call to
On 02/28/2017 07:08 AM, Christoph Hellwig wrote:
> On Mon, Feb 27, 2017 at 05:19:31PM -0800, Stephen Hemminger wrote:
>> Fixes: ee5242360424 ("scsi: zero per-cmd driver data before each I/O")
>>
>> but that is already in linux-next.
>>
>> Noticed another place where memset(of the data was being
On 02/27/2017 11:15 PM, Bart Van Assche wrote:
> On Wed, 2017-02-22 at 17:07 +0100, Hannes Reinecke wrote:
>> 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.
>
> Hello
From: Bart Van Assche
A medium access request is defined as an internal regular request that
operates on a whole number of logical blocks of the storage medium.
These include REQ_OP_READ, REQ_OP_WRITE, REQ_OP_FLUSH, REQ_OP_DISCARD,
REQ_OP_SECURE_ERASE,
On 02/27/2017 11:27 PM, Bart Van Assche wrote:
> On Wed, 2017-02-22 at 17:07 +0100, Hannes Reinecke wrote:
>> spin_lock_irqsave(shost->host_lock, flags);
>> +WARN_ON(shost->shost_state != SHOST_RUNNING &&
>> +shost->shost_state != SHOST_CANCEL &&
>> +
On Tue, 2017-02-28 at 12:12 +0200, Boaz Harrosh wrote:
> On 02/28/2017 03:11 AM, Jeff Layton wrote:
> <>
> >
> > I'll probably have questions about the read side as well, but for now it
> > looks like it's mostly used in an ad-hoc way to communicate errors
> > across subsystems (block to fs
From: Colin Ian King
The call to qedi_setup_int is not updating the return code rc yet rc
is being checked for an error. Fix this by assigning rc to the return
code from the call to qedi_setup_int.
Signed-off-by: Colin Ian King
---
Hi James,
On 02/28/2017 12:28 AM, James Smart wrote:
> I'd like to announce the availability of the Broadcom (Emulex) FC Target
> driver - efct.
> This driver has been part of the Emulex OneCore Storage SDK tool kit for
> Emulex
> SLI-4 adapters. The SLI-4 adapters support 16Gb/s and higher
I don't really like this too much - this is too many SCSI specifics
for the block layer to care. Maybe using bios for the zone ops was a
mistake after all, and we should just have operations in struct block_device
instead..
On Tue, 2017-02-28 at 19:25 +0900, Damien Le Moal wrote:
> From: Bart Van Assche
>
> Use blk_rq_accesses_medium() instead of !blk_rq_is_passthrough() to
> ensure that code that is intended for normal medium access requests,
> e.g. DISCARD, READ and WRITE requests, is
On Tue, 2017-02-28 at 17:02 +0100, Christoph Hellwig wrote:
> I don't really like this too much - this is too many SCSI specifics
> for the block layer to care. Maybe using bios for the zone ops was a
> mistake after all, and we should just have operations in struct block_device
> instead..
On Tue, 2017-02-28 at 08:32 -0700, Jens Axboe wrote:
> On 02/28/2017 07:08 AM, Christoph Hellwig wrote:
> > On Mon, Feb 27, 2017 at 05:19:31PM -0800, Stephen Hemminger wrote:
> > > Fixes: ee5242360424 ("scsi: zero per-cmd driver data before each
> > > I/O")
> > >
> > > but that is already in
On 02/28/2017 10:16 AM, Stephen Hemminger wrote:
> On Tue, 28 Feb 2017 09:06:13 -0800
> James Bottomley wrote:
>
>> On Tue, 2017-02-28 at 08:32 -0700, Jens Axboe wrote:
>>> On 02/28/2017 07:08 AM, Christoph Hellwig wrote:
On Mon, Feb 27, 2017 at 05:19:31PM -0800,
On Tue, 28 Feb 2017 15:08:12 +0100
Christoph Hellwig wrote:
> On Mon, Feb 27, 2017 at 05:19:31PM -0800, Stephen Hemminger wrote:
> > Fixes: ee5242360424 ("scsi: zero per-cmd driver data before each I/O")
> >
> > but that is already in linux-next.
> >
> > Noticed another place
On 02/27/2017 07:22 PM, Xiubo Li wrote:
> Hi Mike
>
> Thanks verrry much for your work and test cases.
>
>
> 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
Arnd,
Thank you. Looks good.
-- james
Signed-off-by: James Smart
On 2/27/2017 12:31 PM, Arnd Bergmann wrote:
The new debugfs output causes a link error on 32-bit architectures:
ERROR: "__aeabi_uldivmod" [drivers/scsi/lpfc/lpfc.ko] undefined!
This code is not
On Tue, 2017-02-28 at 22:00 +, Madhani, Himanshu wrote:
> Can we get some review comments for this series.
Hello Himanshu,
The merge window opened one week ago and is still open. Anyone who contributes
to the Linux kernel should know that patches should not be posted during the
merge window
On Tue, 28 Feb 2017 09:06:13 -0800
James Bottomley wrote:
> On Tue, 2017-02-28 at 08:32 -0700, Jens Axboe wrote:
> > On 02/28/2017 07:08 AM, Christoph Hellwig wrote:
> > > On Mon, Feb 27, 2017 at 05:19:31PM -0800, Stephen Hemminger wrote:
> > > > Fixes: ee5242360424
On Mon, Feb 27 2017, Jeff Layton wrote:
> On Tue, 2017-02-28 at 10:32 +1100, NeilBrown wrote:
>> On Mon, Feb 27 2017, Andreas Dilger wrote:
>>
>> >
>> > My thought is that PG_error is definitely useful for applications to get
>> > correct errors back when doing write()/sync_file_range() so that
Hi Nic/Bart,
Can we get some review comments for this series.
Thanks,
Himanshu
> -Original Message-
> From: Himanshu Madhani [mailto:himanshu.madh...@cavium.com]
> Sent: Friday, February 24, 2017 1:37 PM
> To: target-de...@vger.kernel.org; bart.vanass...@sandisk.com;
>
If SCSI-ML has already issued abort on a command i.e
FNIC_IOREQ_ABTS_PENDING is set and we get a IO completion avoid
this being flagged as out-of-order completion by setting the
FNIC_IO_DONE flag in fnic_fcpio_icmnd_cmpl_handler
Signed-off-by: Satish Kharat
Signed-off-by:
> -Original Message-
> From: Bart Van Assche [mailto:bart.vanass...@sandisk.com]
> Sent: Tuesday, February 28, 2017 2:10 PM
> To: Madhani, Himanshu ; target-
> de...@vger.kernel.org; n...@linux-iscsi.org
> Cc: linux-scsi@vger.kernel.org; Malavali, Giridhar
>
On Tue, 28 Feb 2017 09:06:13 -0800
James Bottomley wrote:
> On Tue, 2017-02-28 at 08:32 -0700, Jens Axboe wrote:
> > On 02/28/2017 07:08 AM, Christoph Hellwig wrote:
> > > On Mon, Feb 27, 2017 at 05:19:31PM -0800, Stephen Hemminger wrote:
> > > > Fixes: ee5242360424
On Tue, 2017-02-28 at 10:57 -0800, Stephen Hemminger wrote:
> On Tue, 28 Feb 2017 09:06:13 -0800
> James Bottomley wrote:
>
> > On Tue, 2017-02-28 at 08:32 -0700, Jens Axboe wrote:
> > > On 02/28/2017 07:08 AM, Christoph Hellwig wrote:
> > > > On Mon, Feb 27, 2017 at
Consistently use types from linux/types.h to fix the following
scsi/scsi_netlink_fc.h userspace compilation errors:
/usr/include/scsi/scsi_netlink_fc.h:60:2: error: unknown type name 'uint64_t'
uint64_t seconds;
/usr/include/scsi/scsi_netlink_fc.h:61:2: error: unknown type name 'uint64_t'
Consistently use types from linux/types.h to fix the following
scsi/scsi_netlink.h userspace compilation errors:
/usr/include/scsi/scsi_netlink.h:43:2: error: unknown type name 'uint8_t'
uint8_t version;
/usr/include/scsi/scsi_netlink.h:44:2: error: unknown type name 'uint8_t'
uint8_t
Include and consistently use types it provides
to fix the following scsi/scsi_bsg_fc.h userspace compilation errors:
/usr/include/scsi/scsi_bsg_fc.h:83:2: error: unknown type name 'uint8_t'
uint8_t reserved;
/usr/include/scsi/scsi_bsg_fc.h:86:2: error: unknown type name 'uint8_t'
uint8_t
Write throughput is pretty
low at around 150 MB/s.
What's the original write throughput without this patch? Is it also
around 80 MB/s ?
It is around 20-30 MB/s. Same fio args except using --rw=write.
Got it.
Thanks.
BRs
Xiubo
The IO and Abort latency counter count the time taken to complete
the IO and abort command into broad buckets. It is not intend for
performance measurement, just a debug stat.
current_max_io_time tries to keep tarck of the maximun time an IO
has taken to complete if it > 30sec, it just another
Thanks very much Chris, really good patch. Hopefully it can reach 4.11!
Cheers,
Guilherme
Just a simple counter of number of check conditions encountered on
that host.
Signed-off-by: Satish Kharat
Signed-off-by: Sesidhar Baddela
---
drivers/scsi/fnic/fnic_scsi.c | 3 +++
drivers/scsi/fnic/fnic_stats.h | 1 +
drivers/scsi/fnic/fnic_trace.c |
In sas_rediscover_dev when we call sas_get_phy_attached_dev to find the
device is ok and when in the flutter case when we call
sas_ex_phy_discover the device is gone, the sas_addr was changed to
zero.
[300247.584696] sas: ex 500e004aaa1f phy0 originated
BROADCAST(CHANGE)
[300247.663516] sas:
I'm seeing this while testing on Linus' current master:
[ 427.814466] WARNING: CPU: 0 PID: 0 at kernel/irq/handle.c:149
__handle_irq_event_percpu+0x187/0x190
[ 427.832552] irq 116 handler _base_interrupt+0x0/0x9e0 [mpt3sas] enabled
interrupts
I tracked it down to commit 669f044170d8 ("scsi:
> Let's concentrate on INQUIRY since that's the first command in the
> probe sequence. I think it's completing successfully because your
> hyperv layer says it has 36 bytes of transfer and that's the size of a
> successful initial INQUIRY, so the fact that the code above would break
> stuff if
On Tue, 28 Feb 2017 09:06:13 -0800
James Bottomley wrote:
> On Tue, 2017-02-28 at 08:32 -0700, Jens Axboe wrote:
> > On 02/28/2017 07:08 AM, Christoph Hellwig wrote:
> > > On Mon, Feb 27, 2017 at 05:19:31PM -0800, Stephen Hemminger wrote:
> > > > Fixes: ee5242360424
Arnd,
Thank you. Looks good.
-- james
Signed-off-by: James Smart
On 2/27/2017 12:37 PM, Arnd Bergmann wrote:
dma_addr_t may be either u32 or u64, depending on the kernel configuration,
and we get a warning for the 32-bit case:
drivers/scsi/lpfc/lpfc_nvme.c: In
Included in the current storvsc driver for Hyper-V is the ability
to access luns on an FC fabric via a virtualized fiber channel
adapter exposed by the Hyper-V host. The driver also attaches to the FC
transport to allow host and port names to be published under
/sys/class/fc_host/hostX. Current
On 2/28/2017 8:34 AM, Hannes Reinecke wrote:
Can you clarify these?
Are these 'just' resource allocation problems or something else, too?
Most are resource allocation - buffer pools, dma pools, pages for
resources, and hw resource allocation splits. However, async receive RQ
policies are a
Getting rid of else case to make the flow look bit more simpler
logical
Signed-off-by: Satish Kharat
Signed-off-by: Sesidhar Baddela
---
drivers/scsi/fnic/fnic_scsi.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git
Martin,
On 3/1/17 11:52, Martin K. Petersen wrote:
>> "Christoph" == Christoph Hellwig writes:
>
> Christoph> I don't really like this too much - this is too many SCSI
> Christoph> specifics for the block layer to care. Maybe using bios for
> Christoph> the zone ops was a
> "Chris" == Chris Leech writes:
Chris> There's a rather long standing regression from the commit
Chris> "libiscsi: Reduce locking contention in fast path"
Applied to 4.11/scsi-fixes.
--
Martin K. Petersen Oracle Linux Engineering
Fixing the IO stats updation (Active IOs and IO completion) to
fix "Number of Active IOs" becoming negative in the fnistats
output.
Signed-off-by: Satish Kharat
Signed-off-by: Sesidhar Baddela
---
drivers/scsi/fnic/fnic_scsi.c | 15 ++-
1
On Tue, 2017-02-28 at 10:41 -0800, Stephen Hemminger wrote:
> [1.652279] hv_storvsc: IO cmd 0x12 0x0 0x0 scsi status 0x0 srb
> status 0x20 length 36
> [1.652297] scsi host1: scsi scan: INQUIRY result too short (5),
> using 36
This is definitive. We sent the Inquiry command, we got 36
On Tue, 2017-02-28 at 16:25 -0800, Omar Sandoval wrote:
> I'm seeing this while testing on Linus' current master:
>
> [ 427.814466] WARNING: CPU: 0 PID: 0 at kernel/irq/handle.c:149
> __handle_irq_event_percpu+0x187/0x190
> [ 427.832552] irq 116 handler _base_interrupt+0x0/0x9e0 [mpt3sas]
> "Christoph" == Christoph Hellwig writes:
Christoph> I don't really like this too much - this is too many SCSI
Christoph> specifics for the block layer to care. Maybe using bios for
Christoph> the zone ops was a mistake after all, and we should just have
Christoph> operations
On Tue, 28 Feb 2017 22:20:58 -0800
James Bottomley wrote:
> On Tue, 2017-02-28 at 17:25 -0800, Stephen Hemminger wrote:
> > [1.346023] hv_storvsc: IO cmd 0x12 0x0 0x0 scsi status 0x0 srb
> > status 0x20 length 36
> > [1.352913] inquiry data: : 00 aa be
From: Satish Kharat
This is to avoid the log from being filled with vlan discovery
messages when there is no vlan configured on the switch.
Signed-off-by: Satish Kharat
Signed-off-by: Sesidhar Baddela
---
> "Colin" == Colin King writes:
Colin> The call to qedi_setup_int is not updating the return code rc yet
Colin> rc is being checked for an error. Fix this by assigning rc to the
Colin> return code from the call to qedi_setup_int.
Applied to 4.11/scsi-fixes.
--
On Wed, Mar 01, 2017 at 01:07:12AM +, Bart Van Assche wrote:
> On Tue, 2017-02-28 at 16:25 -0800, Omar Sandoval wrote:
> > I'm seeing this while testing on Linus' current master:
> >
> > [ 427.814466] WARNING: CPU: 0 PID: 0 at kernel/irq/handle.c:149
> >
On Tue, 2017-02-28 at 17:25 -0800, Stephen Hemminger wrote:
> [1.346023] hv_storvsc: IO cmd 0x12 0x0 0x0 scsi status 0x0 srb
> status 0x20 length 36
> [1.352913] inquiry data: : 00 aa be f1 5c 98 ff ff f0 64
> 02 89 ff ff ff ff
> [1.356543] inquiry data: 0010: 00 00 00 00
Hi Damien,
[auto build test WARNING on scsi/for-next]
[also build test WARNING on v4.10 next-20170228]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system]
url:
https://github.com/0day-ci/linux/commits/Damien-Le-Moal/Improve-ZBC-integration
> "Damien" == Damien Le Moal writes:
Damien,
Damien> The problem remains that the mpt3sas driver needs fixing. As you
Damien> suggest, we can do that in sd, or directly in mpt3sas. I tried
Damien> to do a clean fix in sd, but always end up consuming a lot of
Damien>
64 matches
Mail list logo