Hi Roland,
There's a typo in Jack's email address which is our mistake, was in V3
9/9, please fix
it to be Jack Morgenstein ja...@dev.mellanox.co.il (the error is
missing l in mellanox)
thanks,
Or.
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
the body of a
On 08/07/2013 12:40, Fengguang Wu wrote:
Hi Eli,
FYI, there are coccinelle warnings in
tree: git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband.git
for-next
head: 1af1abad19f6a40d8822cb7a35736e9e102fade6
commit: 809d3a921f9047bf575488f410ed12b365fe5cd7 [769/772] mlx5: Add
On Mon, Jul 08, 2013 at 12:58:10PM +0300, Or Gerlitz wrote:
On 08/07/2013 12:40, Fengguang Wu wrote:
Hi Eli,
FYI, there are coccinelle warnings in
tree: git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband.git
for-next
head: 1af1abad19f6a40d8822cb7a35736e9e102fade6
On 08/07/2013 13:58, Fengguang Wu wrote:
Roland, you may need to replace some with \t in the patches.
Thanks, Fengguang
Roland, I have prepared these patches properly and will send them to
you, they can be just squashed into the initial commit
or come as add on patches.
Or.
--
To
All Reported-by Fengguang Wu fengguang...@intel.com
Or Gerlitz (2):
mlx5: Use simple_open when possible
IB/mlx5: Removes unneeded semicolons
drivers/infiniband/hw/mlx5/mr.c | 15 ---
drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 13 +++--
Found by coccinelle
Generated by: coccinelle/misc/semicolon.cocci
Reported-by: Fengguang Wu fengguang...@intel.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/hw/mlx5/mr.c |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git
If the open entry for char-device just does
file-private_data = inode-i_private;
we can use simple_open instead.
Generated by: coccinelle/api/simple_open.cocci
Reported-by: Fengguang Wu fengguang...@intel.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
Hi Or,
static const struct file_operations limit_fops = {
.owner = THIS_MODULE,
- .open = file_open,
+ .open = simple_open,
.write = limit_write,
.read = limit_read,
static const struct file_operations dfops = {
.owner = THIS_MODULE,
- .open
On 08/07/2013 15:34, Fengguang Wu wrote:
The above chunks will need to fix alignments.
sure
--
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 http://vger.kernel.org/majordomo-info.html
All Reported-by Fengguang Wu fengguang...@intel.com
changes from V0:
- replaced space with tabs in few places
Or Gerlitz (2):
mlx5: Use simple_open when possible
IB/mlx5: Removes unneeded semicolons
drivers/infiniband/hw/mlx5/mr.c | 15 ---
If the open entry for char-device just does
file-private_data = inode-i_private;
we can use simple_open instead.
Generated by: coccinelle/api/simple_open.cocci
Reported-by: Fengguang Wu fengguang...@intel.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
Found by coccinelle
Generated by: coccinelle/misc/semicolon.cocci
Reported-by: Fengguang Wu fengguang...@intel.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/hw/mlx5/mr.c |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git
Hi Roland,
Indeed late, but these are some iser initiator updates for 3.11
We have another batch coming which changes the initiator driver such that
it can use FRWRs (Fast-Reg-Work-Requests) when FMRs aren't available, e.g
on ConnectX (mlx4) VF or Connect-IB (mlx5) - the other batch should
be
From: Shlomo Pongratz shlo...@mellanox.com
Use cmds_max passed from user space to be the number of PDUs to be
supported for the session instead of hard-coded ISCSI_DEF_XMIT_CMDS_MAX.
Specifically, this allows to control the max number of SCSI commands
for the seesion. Also don't ignore the qdepth
From: Shlomo Pongratz shlo...@mellanox.com
This is a preparation step to a patch that accepts the number of max SCSI
commands to be supported for the session from the user space iSCSI tools.
Move the allocation of the login buffer, FMR pool and its associated
page vector from
Commit 4f363882612 IB/iser: Move informational messages from error to info
level
was setting info prints to require lower value for the debug level vs warning
prints
which isn't the common convention, fix that. Also move the prints on unaligned
SG
from warning to debug level.
Signed-off-by: Or
On 07/08/13 15:19, Or Gerlitz wrote:
+ iser_err(FMR alloction failed, err %d\n, ret);
I see alloction instead of allocation - this looks like an
(unimportant) typo ?
Bart.
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
the body of a message to
On 07/08/13 15:19, Or Gerlitz wrote:
+#define ISER_DEF_XMIT_CMDS_DEFUALT 512
+#if ISCSI_DEF_XMIT_CMDS_MAX ISER_DEF_XMIT_CMDS_DEFUALT
This looks like another spelling issue - shouldn't DEFUALT be changed
into DEFAULT ?
Bart.
--
To unsubscribe from this list: send the line
On 08/07/2013 17:00, Bart Van Assche wrote:
On 07/08/13 15:19, Or Gerlitz wrote:
+#define ISER_DEF_XMIT_CMDS_DEFUALT 512
+#if ISCSI_DEF_XMIT_CMDS_MAX ISER_DEF_XMIT_CMDS_DEFUALT
This looks like another spelling issue - shouldn't DEFUALT be changed
into DEFAULT ?
Bart.
thanks for the
Commit 4f363882612 IB/iser: Move informational messages from error to info
level
was setting info prints to require lower value for the debug level vs warning
prints
which isn't the common convention, fix that. Also move the prints on unaligned
SG
from warning to debug level.
Signed-off-by: Or
Hi Roland,
Indeed late, but these are some iser initiator updates for 3.11
We have another batch coming which changes the initiator driver such that
it can use FRWRs (Fast-Reg-Work-Requests) when FMRs aren't available, e.g
on ConnectX (mlx4) VF or Connect-IB (mlx5) - the other batch should
be
From: Shlomo Pongratz shlo...@mellanox.com
This is a preparation step to a patch that accepts the number of max SCSI
commands to be supported for the session from the user space iSCSI tools.
Move the allocation of the login buffer, FMR pool and its associated
page vector from
From: Shlomo Pongratz shlo...@mellanox.com
Use cmds_max passed from user space to be the number of PDUs to be
supported for the session instead of hard-coded ISCSI_DEF_XMIT_CMDS_MAX.
Specifically, this allows to control the max number of SCSI commands
for the seesion. Also don't ignore the qdepth
On Jul 5, 2013, at 3:11 PM, Roland Dreier rol...@purestorage.com wrote:
So what happens if I have an old application binary, and I run against
a new libibverbs without recompiling?
Also it seems that I'm forced to change my source code to be able to
compile against new libibverbs?
I
[resending to reply-all, sorry Jeff]
On Mon, Jul 8, 2013 at 9:26 AM, Jeff Squyres (jsquyres)
jsquy...@cisco.com wrote:
So what happens if I have an old application binary, and I run against
a new libibverbs without recompiling?
Also it seems that I'm forced to change my source code to be able
On Mon, Jul 08, 2013 at 10:19:33AM -0700, Roland Dreier wrote:
[resending to reply-all, sorry Jeff]
On Mon, Jul 8, 2013 at 9:26 AM, Jeff Squyres (jsquyres)
jsquy...@cisco.com wrote:
So what happens if I have an old application binary, and I run against
a new libibverbs without
Though, now that I've unpacked it -- I don't think it is OK for
dev_loss_tmo to be off, but fast IO to be on? That drops another
conditional.
The combination of dev_loss_tmo off and reconnect_delay 0 worked fine
in my tests. An I/O failure was detected shortly after the cable to the
Thanks, I just applied a patch to convert to
get_unused_fd_flags(O_CLOEXEC) in uverbs, since there isn't anything
useful that can be done with uverbs fds across an exec.
- R.
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
the body of a message to
On 07/08/13 19:26, Vu Pham wrote:
After running cable pull test on two local IB links for several hrs,
I/Os got stuck.
Further commands multipath -ll or fdisk -l got stuck and never return
Here are the stack dump for srp-x kernel threads.
I'll run with #DEBUG to get more debug info on scsi
On Thu, 2013-07-04 at 10:01 +0200, Bart Van Assche wrote:
On 07/03/13 20:57, David Dillow wrote:
And I'm getting the strong sense that the answer to my question about
fast_io_fail_tmo = 0 when dev_loss_tmo is that we should not allow that
combination, even if it doesn't break the kernel. If
Le 08.07.2013 20:23, Roland Dreier a écrit :
Thanks, I just applied a patch to convert to
get_unused_fd_flags(O_CLOEXEC) in uverbs, since there isn't anything
useful that can be done with uverbs fds across an exec.
Thanks.
In fact, InfiniBand was my main target and I kept this change
31 matches
Mail list logo