On Thu, Feb 17, 2011 at 8:19 PM, Hal Rosenstock
wrote:
>> This looks like a reasonable approach to me, although of course the SM
>> has no way of knowing it should update a port's node description if a
>> hostname changes.
> It does; There's an enhanced trap for this now.
Right; sorry for being
On Thu, Feb 17, 2011 at 6:20 PM, Roland Dreier wrote:
> On Thu, Feb 17, 2011 at 1:30 PM, Michael Heinz
> wrote:
>> This patch addresses the problem by providing a function to build the node
>> description. If the provided source string for the description contains an
>> '@' character, the functi
We too have had installability, perhaps associated with these lists, but it has
been difficult to diagnose.
We duplicate it by forcing dropped packets and seeing the QP's come/go at the
rate of 1000s a second because of the 0 rnr_retry and retry counts. This
analysis is in line behind other bu
> Yes, that is the crux of the issue. I had missed that ipoib_mcast_free() is
> only called on remove_list.
So do we have any idea of what this patch is fixing? Any thoughts from
the qlogic people involved in this patch?
> While we are discussing IPoIB issues, how about the two other issues that
On Thu, Feb 17, 2011 at 1:30 PM, Michael Heinz wrote:
> This patch addresses the problem by providing a function to build the node
> description. If the provided source string for the description contains an
> '@' character, the function will substitute the current utsname.
>
> This ensures that e
Linus, please pull from
master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband.git for-linus
This tree is also available from kernel.org mirrors at:
git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband.git
for-linus
This will get a few small things:
- Don't send even
The common practice in IB fabrics is to set the description of an HCA to be
the hostname of the machine plus a description (i.e., "myhost hca-1",
"myhost hca-2", etc..)
This has a limitation, however. The first is that if the machine's
hostname is set via DHCP, the HCA description may be set befo
Adds support for ib_build_node_desc() to the HCAs.
Signed-off-by: Michael Heinz
---
drivers/infiniband/hw/ipath/ipath_mad.c |2 +-
drivers/infiniband/hw/mlx4/mad.c|3 ++-
drivers/infiniband/hw/mthca/mthca_mad.c |3 ++-
drivers/infiniband/hw/qib/qib_mad.c |2 +-
4 fil
The common practice in IB fabrics is to set the description of an HCA to be
the hostname of the machine plus a description (i.e., "myhost hca-1",
"myhost hca-2", etc..)
This has a limitation, however. The first is that if the machine's
hostname is set via DHCP, the HCA description may be set befo
Yevgeny,
I've got users complaining about ibdiagpath failing. When I dug into it I
found that patch f3cf1f7c15ca24598fdf68b9ba71788b386b2f14 is syntactically
incorrect and needs to be rolled back.
The problem is that the patch contains lines like
set entry [lindex $values $i+1]
these lin
Mike,
Yevgeny is the maintainer and that is the master branch, so submitting patches
to him is the correct person.
Did you submit your patch on linux-rdma as well? I am sure he just missed the
patch.
Ira
On Thu, 17 Feb 2011 08:47:05 -0800
Mike Heinz wrote:
> Ira,
>
> What's the correct gi
On 02/16/2011 05:15 PM, Roland Dreier wrote:
On Wed, Feb 16, 2011 at 4:51 PM, Pradeep Satyanarayana
wrote:
The list_move_tail() in ipoib_mcast_restart_task() is conditional. So it
feasible that mcast->list is not moved to the remove_list, but the
ipoib_neigh structure is freed. A subsequent ca
Ira,
What's the correct git repo for ibutils? The most recently updated version
appears to be:
http://git.openfabrics.org/git?p=~kliteyn/ibutils.git;a=summary
I submitted a report that a patch needs to be rolled back to Yevgeny but he
never replied and you seem to be the last person to commit
Dear all,
I try to install OFED stacks on the host. Since my kernel version is 2.6.36-
UL4-x86_64, which is not supported by OFED-1.5.2. So, I tried OFED-1.5.3-rc4.
In the process of installation I got the error message of
Build ofa_kernel RPM
Running rpmbuild --rebuild --define '_topdir
14 matches
Mail list logo