Dotan,
Enforcing the MTU to 1KB did the trick. Thanks for you precious help !
Xavier
On Wed, Oct 24, 2012 at 10:19 PM, Xavier Guérin wrote:
> Hello Dotan,
>
>> I'm trying to understand exactly what you did...
>>
>> I'm guessing here the scenario, please correct me if I'm wrong:
>> You've tried
In the previous commit, lines which seemed not destined for
upstream prevented sim_cmd from building. Remove them.
Signed-off-by: Jon Stanley
---
ibsim/sim_cmd.c | 23 ---
1 files changed, 0 insertions(+), 23 deletions(-)
diff --git a/ibsim/sim_cmd.c b/ibsim/sim_cmd.c
inde
Hello Dotan,
> I'm trying to understand exactly what you did...
>
> I'm guessing here the scenario, please correct me if I'm wrong:
> You've tried to work with RoCE over ConnectX-3 using RC QP.
> You were able to send messages up to 1024 (including this value) and it
> failed.
>
> You've to execu
Signed-off-by: Yishai Hadas
Signed-off-by: Tzahi Oved
---
changes from V4:
increase MLX4_UVERBS_MIN_ABI_VERSION to be 3
- version 2 is not supported any more.
fix error message as part of mlx4_driver_init
---
src/mlx4-abi.h |2 +-
src/mlx4.c | 82 ++---
Signed-off-by: Alex Netes
---
opensm/osm_subnet.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/opensm/osm_subnet.c b/opensm/osm_subnet.c
index b9a491b..541105a 100644
--- a/opensm/osm_subnet.c
+++ b/opensm/osm_subnet.c
@@ -2168,7 +2168,7 @@ int osm_subn_rescan_conf_files(IN
Hi Bart,
On 16:40 Wed 24 Oct , Bart Van Assche wrote:
> On 10/24/12 15:27, Alex Netes wrote:
> > On 16:44 Fri 21 Sep , Bart Van Assche wrote:
> >> +default_rdma_service=openibd
> >> +AC_ARG_WITH([rdma_service],
> >> +AC_HELP_STRING([--with-rdma-service=name],
> >> + [
On 10/24/2012 10:45 AM, Bart Van Assche wrote:
> On 10/24/12 16:33, Doug Ledford wrote:
>> On 10/24/2012 9:27 AM, Alex Netes wrote:
>>> On 16:44 Fri 21 Sep , Bart Van Assche wrote:
start () {
+if [ -e $pidfile ]; then
>>>
>>> On opensm segfault (happens one in a while :), pidfil
On 10/24/12 16:33, Doug Ledford wrote:
On 10/24/2012 9:27 AM, Alex Netes wrote:
On 16:44 Fri 21 Sep , Bart Van Assche wrote:
start () {
+if [ -e $pidfile ]; then
On opensm segfault (happens one in a while :), pidfile won't be removed, so
you won't be able to start the opensm again.
On 10/24/12 15:27, Alex Netes wrote:
> On 16:44 Fri 21 Sep , Bart Van Assche wrote:
>> +default_rdma_service=openibd
>> +AC_ARG_WITH([rdma_service],
>> +AC_HELP_STRING([--with-rdma-service=name],
>> + [name of the RDMA service: "rdma" when using
>> /etc/init.d/rdma to sta
On 10/24/2012 9:27 AM, Alex Netes wrote:
> Hi Bart,
>
> On 16:44 Fri 21 Sep , Bart Van Assche wrote:
>> On recent SLES and openSUSE systems it is necessary to read the
>> file /lib/lsb/init-functions instead of /etc/rc.status for proper
>> integration with systemd. Certain implementations of k
Signed-off-by: Hal Rosenstock
---
opensm/osm_vl15intf.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/opensm/osm_vl15intf.c b/opensm/osm_vl15intf.c
index 253578b..f85252c 100644
--- a/opensm/osm_vl15intf.c
+++ b/opensm/osm_vl15intf.c
@@ -346,7 +346,7 @@ void osm_vl15_
Trap represses (responses) never get to osm_trap_rcv_process.
They are handled in osm_sm_mad_ctrl.c:sm_mad_ctrl_rcv_callback
which invokes sm_mad_ctrl_process_trap_repress.
Signed-off-by: Hal Rosenstock
---
opensm/osm_trap_rcv.c | 24 +++-
1 files changed, 3 insertions(+),
Should subtract rather than add decrement
Routine was same as cl_atomic_add other than parameter name
being decrement instead of increment so assume this is
cut 'n paste error
Nothing in opensm used this routine up to now
Signed-off-by: Hal Rosenstock
---
include/complib/cl_atomic_osd.h |2
On 10/24/12 15:19, Alex Netes wrote:
Hi Bart,
On 16:43 Fri 21 Sep , Bart Van Assche wrote:
"chkconfig $service on" enables a service in the runlevels mentioned
next to "Default-Start" and only in those runlevels.
Signed-off-by: Bart Van Assche
Cc: Doug Ledford
---
configure.in
Hi Bart,
On 16:44 Fri 21 Sep , Bart Van Assche wrote:
> On recent SLES and openSUSE systems it is necessary to read the
> file /lib/lsb/init-functions instead of /etc/rc.status for proper
> integration with systemd. Certain implementations of killproc
> need a pidfile so make sure that one get
Hi Bart,
On 16:43 Fri 21 Sep , Bart Van Assche wrote:
> "chkconfig $service on" enables a service in the runlevels mentioned
> next to "Default-Start" and only in those runlevels.
>
> Signed-off-by: Bart Van Assche
> Cc: Doug Ledford
> ---
> configure.in | 10 ++
16 matches
Mail list logo