On 04/12/2016 09:51 AM, Christoph Hellwig wrote:
On Tue, Apr 12, 2016 at 08:34:03AM -0700, Bart Van Assche wrote:
ib_sg_to_pages() can handle segments that are larger than mr->page_size.
The interface handles it fine, but we'll still end up using a segment
per (MR) page.
Have you considered
On Tue, Apr 12, 2016 at 08:34:03AM -0700, Bart Van Assche wrote:
> Hello Christoph,
>
> ib_sg_to_pages() can handle segments that are larger than mr->page_size.
The interface handles it fine, but we'll still end up using a segment
per (MR) page.
> Have you considered to set
On 04/08/2016 12:06 PM, Keith Busch wrote:
On Fri, Apr 08, 2016 at 01:40:06PM -0400, Matthew Wilcox wrote:
- Inability to use all queues supported by a device. Intel's P3700
supports 31 queues, but block-mq insists on assigning an even multiple
of CPUs to each queue. So if you have
> "Hannes" == Hannes Reinecke writes:
Hannes> zfcp has its own mechanism for selective scanning, so revert to
Hannes> the original scanning behaviour to not confuse users.
Applied to 4.7/scsi-queue.
--
Martin K. Petersen Oracle Linux Engineering
--
To unsubscribe from
>Hey Willy,
>
>> - Interrupt steering needs to be controlled by block-mq instead of
>> the driver. It's pointless to have each driver implement its own
>> policies on interrupt steering, irqbalanced remains a source of
>> end-user frustration, and block-mq can change the
On 04/12/2016 01:42 PM, Martin K. Petersen wrote:
>> "Michel" == Michel Meyers writes:
>
>>> Shouldn't we just quirk SYNOLOGY, then?
>
> Michel> Could be worth it. As far as I can see, they manufacture no
> Michel> directly attached drives (only NAS devices, which I
On Tue, Apr 12, 2016 at 1:47 PM, Hannes Reinecke wrote:
> ZAC drives implement a 'ZAC Management Out' command template,
> which maps onto the ZBC OUT command.
>
> Signed-off-by: Hannes Reinecke
> ---
> drivers/ata/libata-eh.c | 1 +
> drivers/ata/libata-scsi.c
On Tue, Apr 12, 2016 at 11:13:52AM -0700, Bart Van Assche wrote:
> >That's what NVMe does, but I don't think it's a good idea. Because
> >of the unaligned start into the page this means you have to set the limit
> >to one lower than the actual hardware limit.
>
> I think this means that there is
> "Michel" == Michel Meyers writes:
>> Shouldn't we just quirk SYNOLOGY, then?
Michel> Could be worth it. As far as I can see, they manufacture no
Michel> directly attached drives (only NAS devices, which I presume are
Michel> all affected by the same problem as mine).
> "Mike" == Mike Christie writes:
>> Mike: Any preference?
Mike> I think quirking SYNOLOGY is best. It will be the safest route and
Mike> work for all devices.
Michel: Please spin a new version of the patch.
Thanks!
--
Martin K. Petersen Oracle Linux
> "Ming" == Ming Lin writes:
Ming> Are we ready to merge it?
I have queued your patches in my for-next branch. Will let them simmer
there for a bit so they can get tested in linux-next. Barring any
problems I will shuffle them over to 4.7/scsi-queue at the end of the
week.
On 2016/4/12 20:45, Hannes Reinecke wrote:
> On 04/12/2016 02:35 PM, Xinwei Kong wrote:
>> some usb disk such as Manufacturer Lenovo will have some problem in
>> init process.
>>
>> when it run "scsi_dh_find_driver" function, because this parameter of
>> "sdev->inquiry[5]" is 115, it will return
On 04/12/2016 08:47 AM, Hannes Reinecke wrote:
> Update libsas and dependent drivers to handle FPDMA
> SEND/RECEIVE correctly.
>
> Signed-off-by: Hannes Reinecke
> ---
> drivers/scsi/hisi_sas/hisi_sas_v2_hw.c | 2 ++
> drivers/scsi/isci/request.c| 4 +++-
>
Hi Hannes,
[auto build test ERROR on tj-libata/for-next]
[also build test ERROR on next-20160412]
[cannot apply to v4.6-rc3]
[if your patch is applied to the wrong git tree, please drop us a note to help
improving the system]
url:
https://github.com/0day-ci/linux/commits/Hannes-Reinecke/ZAC
Hi Hannes,
[auto build test WARNING on tj-libata/for-next]
[also build test WARNING on next-20160412]
[cannot apply to v4.6-rc3]
[if your patch is applied to the wrong git tree, please drop us a note to help
improving the system]
url:
https://github.com/0day-ci/linux/commits/Hannes-Reinecke
libata device disabling is ... curious. So add the correct
definitions that we can disable ZAC devices properly.
Signed-off-by: Hannes Reinecke
---
include/linux/libata.h | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/include/linux/libata.h
Device-managed ZAC devices just set the zoned capabilities field
in INQUIRY byte 69 (cf ACS-4). This corresponds to the 'zoned'
field in the block device characteristics VPD page.
As this is only defined in SPC-5/SBC-4 we also need to update
the supported SCSI version descriptor.
Signed-off-by:
When reading the NCQ Send/Recv log it might actually not
supported, thereby causing irritating messages
'READ LOG DMA EXT failed'.
Instead we should be reading the log directory first to
figure out if the log is actually supported before trying
to access it.
Signed-off-by: Hannes Reinecke
Update libsas and dependent drivers to handle FPDMA
SEND/RECEIVE correctly.
Signed-off-by: Hannes Reinecke
---
drivers/scsi/hisi_sas/hisi_sas_v2_hw.c | 2 ++
drivers/scsi/isci/request.c| 4 +++-
drivers/scsi/libsas/sas_ata.c | 6 --
Some commands like FPDMA RECEIVE or NCQ NON DATA can encapsulate
other commands to NCQ transport. So decode the subcmds, too.
Signed-off-by: Hannes Reinecke
---
drivers/ata/libata-trace.c| 43 +++
include/linux/ata.h | 17
Recent devices can use NCQ encapsulation for READ LOG DMA EXT,
so we should be using it if available.
Signed-off-by: Hannes Reinecke
---
drivers/ata/libata-core.c | 5 ++---
drivers/ata/libata-eh.c | 30 +-
drivers/ata/libata.h | 5 +++--
ZAC drives implement a 'ZAC Management Out' command template,
which maps onto the ZBC OUT command.
Signed-off-by: Hannes Reinecke
---
drivers/ata/libata-eh.c | 1 +
drivers/ata/libata-scsi.c | 67 +++
drivers/ata/libata-trace.c
Add NCQ encapsulation for ZAC MANAGEMENT OUT and evaluate
NCQ Non-Data log pages to figure out if NCQ encapsulation
is supported.
Signed-off-by: Hannes Reinecke
---
drivers/ata/libata-core.c | 43 ---
drivers/ata/libata-scsi.c | 18
Do not call ata_request_sense() if the sense code is already
present.
Signed-off-by: Hannes Reinecke
---
drivers/ata/libata-eh.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index 5b340ce..e816619 100644
ZAC drives implement a 'ZAC Management In' command template,
which maps onto the ZBC IN command.
Signed-off-by: Hannes Reinecke
---
drivers/ata/libata-eh.c | 1 +
drivers/ata/libata-scsi.c | 157 ++
drivers/ata/libata-trace.c
Byte 69 bits 0:1 in the IDENTIFY DEVICE data indicate a
host-aware ZAC device.
Host-managed ZAC devices have their own individual signature,
and to not set the bits in the IDENTIFY DEVICE data.
And whenever we detect a ZAC-compatible device we should
be displaying the zoned block characteristics
If a device is disabled after error recovery it doesn't make
any sense to generate an ATA sense, but we should rather
return a generic sense code indicating the device is gone.
Signed-off-by: Hannes Reinecke
---
drivers/ata/libata-scsi.c | 6 ++
1 file changed, 6
Move NCQ SEND/RECEIVE checks into a separate function.
Signed-off-by: Hannes Reinecke
---
drivers/ata/libata-core.c | 47 +++
1 file changed, 27 insertions(+), 20 deletions(-)
diff --git a/drivers/ata/libata-core.c
Hi all,
here's a patchset implementing ZAC support for libata.
This is the second part of a larger patchset for ZAC/ZBC support;
it requires the scsi trace fixes queued for in mkp/4.7/scsi-queue and
the patchset 'libata: SATL update' I've posted earlier.
The full patchset can be found at:
So that we don't overflow the number of MR segments allocated because
we have to split on SGL segment into multiple MR segments.
Signed-off-by: Christoph Hellwig
---
drivers/infiniband/ulp/iser/iscsi_iser.c | 1 +
1 file changed, 1 insertion(+)
diff --git
From: Ming Lin
Rename SCSI specific struct and functions to more genenic names.
Reviewed-by: Christoph Hellwig
Signed-off-by: Ming Lin
Reviewed-by: Sagi Grimberg
--
To unsubscribe from this list: send the line
On 2016-04-12 4:20, Martin K. Petersen wrote:
>> "Michel" == Michel Meyers writes:
>
> Michel> SYNOLOGY iSCSI devices do not work with large I/O and SYNOLOGY
> Michel> changed the identifier of the devices
>
> Shouldn't we just quirk SYNOLOGY, then?
Could be worth it.
From: Ming Lin
Parameter "bool mq" is block driver specific.
Change it to "first_chunk" to make it more generic.
Reviewed-by: Christoph Hellwig
Signed-off-by: Ming Lin
Reviewed-by: Sagi Grimberg
--
To
On 12/04/2016 07:47, Hannes Reinecke wrote:
Define the NCQ NON DATA command and update libsas to handle it
correctly.
Signed-off-by: Hannes Reinecke
---
drivers/scsi/hisi_sas/hisi_sas_v2_hw.c | 1 +
drivers/scsi/isci/request.c| 3 ++-
Add a new NO_REPORT_LUNS quirk and set it for Seagate drives with
an usb-id of: 0bc2:331a, as these will fail to respond to a
REPORT_LUNS command.
Cc: sta...@vger.kernel.org
Reported-and-tested-by: David Webb
Signed-off-by: Hans de Goede
---
Changes in v2:
Commit 64d513ac31bd ("scsi: use host wide tags by default") causes
the SCSI core to queue more commands then we can handle on devices with
multiple LUNs, limit the queue depth at the scsi-host level instead of
per slave to fix this.
BugLink: https://bugzilla.redhat.com/show_bug.cgi?id=1315013
Cc:
From: Ming Lin
Replace parameter "struct scsi_data_buffer" with "struct sg_table" in
SG alloc/free functions to make them generic.
Reviewed-by: Christoph Hellwig
Signed-off-by: Ming Lin
Reviewed-by: Sagi Grimberg
We're still missing an ack from Sagi.
Oops, wasn't aware that mine was needed.
I reviewed these on the nvme-fabrics project.
I'll add my review tags too.
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majord...@vger.kernel.org
More
From: Ming Lin
Now it's ready to move the mempool based SG chained allocator code from
SCSI driver to lib/sg_pool.c, which will be compiled only based on a Kconfig
symbol CONFIG_SG_POOL.
SCSI selects CONFIG_SG_POOL.
Reviewed-by: Christoph Hellwig
wrote:
"Ming" == Ming Lin writes:
Ming> Are we ready to merge it?
We're still missing an ack from Sagi.
Thought we already had a ack from Bart.
OK, let's get one more from Sagi.
Hmm ... this patch doesn't touch any code for which Sagi is
On 04/12/2016 12:03 PM, John Garry wrote:
> On 12/04/2016 07:47, Hannes Reinecke wrote:
>> Define the NCQ NON DATA command and update libsas to handle it
>> correctly.
>>
>> Signed-off-by: Hannes Reinecke
>> ---
>> drivers/scsi/hisi_sas/hisi_sas_v2_hw.c | 1 +
>>
some usb disk such as Manufacturer Lenovo will have some problem in
init process.
when it run "scsi_dh_find_driver" function, because this parameter of
"sdev->inquiry[5]" is 115, it will return "alua" in "scsi_dh_find_driver"
function and will have some error log (alua: Attach failed (-22)). some
Modified patch to use min_not_zero()
Ran a number of tests overnight on F/C, SCSI/SAS and SRP (RDMA) and no issues
found.
Tested-by: Laurence Oberman
Laurence Oberman
Principal Software Maintenance Engineer
Red Hat Global Support Services
- Original Message -
How about this?
drivers/scsi/scsi_scan: mark STARGET_REMOVE state before destroy
Signed-off-by: Xiong Zhou
---
drivers/scsi/scsi_scan.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c
index 27df7e7..21092e5
On 22:14 Mon 11 Apr , Martin K. Petersen wrote:
> > "Hannes" == Hannes Reinecke writes:
>
> Hannes> zfcp has its own mechanism for selective scanning, so revert to
> Hannes> the original scanning behaviour to not confuse users.
>
> Benjamin?
>
Sry, I meant to quickly
On 04/12/2016 02:35 PM, Xinwei Kong wrote:
> some usb disk such as Manufacturer Lenovo will have some problem in
> init process.
>
> when it run "scsi_dh_find_driver" function, because this parameter of
> "sdev->inquiry[5]" is 115, it will return "alua" in "scsi_dh_find_driver"
> function and
So that we don't overflow the number of MR segments allocated because
we have to split on SGL segment into multiple MR segments.
Signed-off-by: Christoph Hellwig
---
drivers/infiniband/ulp/iser/iscsi_iser.c | 1 +
1 file changed, 1 insertion(+)
diff --git
RDMA drivers need segments that aren't larger than a single HCA page
for memory registrations to work properly, so wire up this limitation
in the host.
While we could just call blk_queue_max_segment_size from ->slave_configure,
that would override the global limit based on the DMA device, so
Since iSER was converted to use the generic virt_boundary mechanism
(which was called something else in 4.4), it didn't handle the case
where a request is using up the full size of max_hw_segments, but
not actually aligned to the virt boundary. This series sets the
maximum segment size limit to
On Tue, 2016-04-12 at 13:48 +0300, Sagi Grimberg wrote:
> > So that we don't overflow the number of MR segments allocated
> > because
> > we have to split on SGL segment into multiple MR segments.
> >
> > Signed-off-by: Christoph Hellwig
> > ---
> >
On 04/12/2016 07:13 AM, Christoph Hellwig wrote:
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 8106515..ad79372 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -2120,7 +2120,8 @@ static void __scsi_init_queue(struct Scsi_Host *shost,
struct
The driver creates its own per-CPU threads which are updated based on
CPU hotplug events. It is also possible to use kworkers and remove some
of the kthread infrastrucure.
The code checked ->thread to decide if there is an active per-CPU
thread. By using the kworker infrastructure this is no
On 04/12/2016 07:13 AM, Christoph Hellwig wrote:
So that we don't overflow the number of MR segments allocated because
we have to split on SGL segment into multiple MR segments.
Signed-off-by: Christoph Hellwig
---
drivers/infiniband/ulp/iser/iscsi_iser.c | 1 +
1 file changed,
Other than adding the patch and rebuilding the kernel and testing regular
stuff, which I had to do anyway, that was the extent of testing.
I did not see where it was used to be honest other than adding the structure
member.
I wanted to test the simple change because it was in scsi_lib.c which
On Tue, Apr 12, 2016 at 07:27:36AM -0700, James Bottomley wrote:
> > In iser we sorta rely on 4k pages so we avoid
> > PAGE_SIZE but rather set SIZE_4K for these sort
> > of things (like we did in the virt_boundary).
>
> So you still want only 4k segments even on PPC where the PAGE_SIZE is
> 16k?
55 matches
Mail list logo