thanks, applied all 4
--
Roland Dreier || For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/index.html
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majord...@vger.kernel.org
More majordomo info at
When transitioning a QP to the error state, in progress RWQEs need
to be marked complete. This also involves releasing the reference
count to the memory regions referenced in the SGEs. The locking in
the receive packet processing wasn't sufficient to prevent
qib_error_qp() from modifying the r_sge
Don't processes too many packets without allowing other IRQ
functions a chance to run. Otherwise, there is a chance of
getting "soft lockup" messages and poor application response times.
Signed-off-by: Ralph Campbell
---
0 files changed, 0 insertions(+), 0 deletions(-)
diff --git a/drivers/inf
On Mon, 2010-08-02 at 17:32 +0200, Bart Van Assche wrote:
> Exports req_lim via sysfs, which is convenient for debugging purposes.
>
> Signed-off-by: Bart Van Assche
> Cc: Roland Dreier
> Cc: David Dillow
Acked-by: David Dillow
--
To unsubscribe from this list: send the line "unsubscribe linu
On Mon, 2010-08-02 at 17:32 +0200, Bart Van Assche wrote:
> Makes sure that the SCSI mid-layer never tries to queue more than
> (SRP request limit) - 1 SCSI commands. Improves performance for targets whose
> request limit is less than or equal to SRP_SQ_REQ_SIZE (63) by reducing the
> number of BUS
On Mon, 2010-08-02 at 17:32 +0200, Bart Van Assche wrote:
> This patch enables allocation of request and response information units on the
> send queue instead of only requests, and implements processing of SRP_CRED_REQ
> information units. Also, declarations have been added to include/scsi/srp.h
>
On Mon, 2010-08-02 at 17:32 +0200, Bart Van Assche wrote:
> The patch below realizes the following transformations on the symbolic
> constants defined in ib_srp.h and used in ib_srp.h and ib_srp.c:
> * Added the constants SRP_RQ_MASK and SRP_SQ_MASK.
> * Renamed SRP_SQ_SIZE into SRP_REQ_SQ_SIZE.
>
Bart Van Assche, on 08/02/2010 10:40 PM wrote:
On Mon, Aug 2, 2010 at 8:36 PM, David Dillow wrote:
On Mon, 2010-08-02 at 22:16 +0400, Vladislav Bolkhovitin wrote:
Bart Van Assche, on 08/02/2010 07:57 PM wrote:
block size number ofIOPSIOPS IOPS
in bytesthreads wi
On Mon, 2010-08-02 at 22:16 +0400, Vladislav Bolkhovitin wrote:
> Bart Van Assche, on 08/02/2010 07:57 PM wrote:
> >>>
> >>> block size number ofIOPSIOPS IOPS
> >>> in bytesthreads without with with
> >>>($bs) ($numjobs) this patch thread=n thread=y
>
On Mon, Aug 2, 2010 at 8:36 PM, David Dillow wrote:
>
> On Mon, 2010-08-02 at 22:16 +0400, Vladislav Bolkhovitin wrote:
> > Bart Van Assche, on 08/02/2010 07:57 PM wrote:
> > >>>
> > >>> block size number of IOPS IOPS IOPS
> > >>> in bytes threads without with wit
Add cleanup to remove CR from SP and EP
during DTO errors in dapli_evd_cqe_to_event.
dapl_sp_remove_ep needs to remove cr_ptr
reference from EP before freeing cr object.
Signed-off-by: Arlin Davis
---
dapl/common/dapl_evd_util.c |1 +
dapl/common/dapl_sp_util.c |2 ++
2 files changed
Signed-off-by: Arlin Davis
---
dapl/openib_ucm/cm.c | 20 +++-
1 files changed, 11 insertions(+), 9 deletions(-)
diff --git a/dapl/openib_ucm/cm.c b/dapl/openib_ucm/cm.c
index b874c8b..05b5af5 100644
--- a/dapl/openib_ucm/cm.c
+++ b/dapl/openib_ucm/cm.c
@@ -165,9 +165,9 @@ sta
Convert consumer input to network order before verbs query pkey check.
Change appropriate debug messages, increase level of bad key message.
Signed-off-by: Arlin Davis
---
dapl/openib_common/util.c | 14 +++---
1 files changed, 7 insertions(+), 7 deletions(-)
diff --git a/dapl/openib
Bart Van Assche, on 08/02/2010 07:57 PM wrote:
SRP I/O with small block sizes causes a high CPU load. Processing IB
completions on the context of a kernel thread instead of in interrupt context
allows to process up to 25% more I/O operations per second. This patch does
add a kernel parameter 'thr
Tziporet Koren wrote:
Hi Tom,
What is the purpose of this?
Is there a reason you did it only for mthca and not mlx4?
Tziporet
Hi Tziporet,
I just picked mthca arbitrarily to demonstrate how to do it. If people
like the verb, then I'll do it for all the devices, but I didn't want to
do all
On Mon, Aug 2, 2010 at 3:08 PM, Vladislav Bolkhovitin wrote:
>
> Bart Van Assche, on 08/02/2010 12:15 PM wrote:
>>
>> SRP I/O with small block sizes causes a high CPU load. Processing IB
>> completions on the context of a kernel thread instead of in interrupt context
>> allows to process up to 25%
Exports req_lim via sysfs, which is convenient for debugging purposes.
Signed-off-by: Bart Van Assche
Cc: Roland Dreier
Cc: David Dillow
---
drivers/infiniband/ulp/srp/ib_srp.c | 14 ++
1 files changed, 14 insertions(+), 0 deletions(-)
diff --git a/drivers/infiniband/ulp/srp/ib
Makes sure that the SCSI mid-layer never tries to queue more than
(SRP request limit) - 1 SCSI commands. Improves performance for targets whose
request limit is less than or equal to SRP_SQ_REQ_SIZE (63) by reducing the
number of BUSY responses.
Signed-off-by: Bart Van Assche
Cc: Roland Dreier
C
This patch enables allocation of request and response information units on the
send queue instead of only requests, and implements processing of SRP_CRED_REQ
information units. Also, declarations have been added to include/scsi/srp.h
for the SRP_CRED_REQ and SRP_CRED_RSP information units.
Signed-
The patch below realizes the following transformations on the symbolic
constants defined in ib_srp.h and used in ib_srp.h and ib_srp.c:
* Added the constants SRP_RQ_MASK and SRP_SQ_MASK.
* Renamed SRP_SQ_SIZE into SRP_REQ_SQ_SIZE.
* Changed the value of SRP_SQ_SIZE from 63 to 64.
Note: this patch
This series of four patches adds SRP_CRED_REQ support in ib_srp, which is a
feature defined by the SRP (draft) standard. Compared to an earlier patch for
implementing this functionality that I posted a few months ago I have done my
best to address all comments that were posted to this mailing list,
Bart Van Assche, on 08/02/2010 12:15 PM wrote:
SRP I/O with small block sizes causes a high CPU load. Processing IB
completions on the context of a kernel thread instead of in interrupt context
allows to process up to 25% more I/O operations per second. This patch does
add a kernel parameter 'thr
Hi Tom,
What is the purpose of this?
Is there a reason you did it only for mthca and not mlx4?
Tziporet
SRP I/O with small block sizes causes a high CPU load. Processing IB
completions on the context of a kernel thread instead of in interrupt context
allows to process up to 25% more I/O operations per second. This patch does
add a kernel parameter 'thread' that allows to specify whether to process IB
On Mon, 19 Jul 2010 13:34:29 +0200
Alexander Schmidt wrote:
> Hi Roland,
>
> this is the fourth version of the huge pages support, please see below for
> change log + description, we appreciate your comments.
Hi Roland,
did you have a chance to take a look at the code? We would like to get thi
25 matches
Mail list logo