Mike> Thanks. So there should be no limit to how many CQ's you can
Mike> create under a single completion channel and still receive
Mike> events for any of the CQ's?
Right, there is no limit other than the total number of CQs you can
create (independent of completion channels).
- R.
Roland Dreier wrote:
Mike> Given that context->num_comp_vectors always appears to be
Mike> one, does it mean you can only have a single CQ generating
Mike> events on a comp_channel at any point in time?
No, completion vectors and completion channels are orthogonal.
Thanks. So th
> Second, in file ofa_kernel-1.2/drivers/infiniband/ulp/vnic/vnic_main.c,
> failed definition SPIN_LOCK_UNLOCKED. Seems to be, "spinlock.h" include is
> missed in this file.
this is a problem with the vnic code that needs to be cleaned up --
SPIN_LOCK_UNLOCKED is not really supposed to be used
Mike> Can someone explain or point me to documentation on what the
Mike> completion vector argument to ibv_create_cq() is for?
In the future, it will be used to support adapters that have multiple
completion event vectors (via MSI-X or similar).
Mike> Given that context->num_comp_vect
Can someone explain or point me to documentation on what the completion
vector argument to ibv_create_cq() is for? Given that
context->num_comp_vectors always appears to be one, does it mean you can
only have a single CQ generating events on a comp_channel at any point
in time?
Also, how do
Dear OFA Member,
The OpenFabrics Alliance is proud to announce:
OFA Interoperability Event #3
April 16 - 20, 2007
University of New Hampshire (UNH) Interoperability Lab (IOL)
Durham, NH.
This event will provide an opportunity for participants to measure their
products for
Interoperabi
On Fri, 2007-03-23 at 10:28, Michael S. Tsirkin wrote:
> > Quoting Hal Rosenstock <[EMAIL PROTECTED]>:
> > Subject: Re: Re: [PATCH] IB/umad: fix GRH handling
> >
> > On Wed, 2007-03-21 at 18:22, Hal Rosenstock wrote:
> > > On Wed, 2007-03-21 at 01:52, Sean Hefty wrote:
> > > > >> Unfortunately, at
https://bugs.openfabrics.org/show_bug.cgi?id=488
[EMAIL PROTECTED] changed:
What|Removed |Added
CC||[EMAIL PROTECTED]
--
Configure bug
https://bugs.openfabrics.org/show_bug.cgi?id=488
Summary: user_mad GRH handling on receive side is incomplete
Product: OpenFabrics Linux
Version: 1.2beta1
Platform: All
OS/Version: All
Status: NEW
Severity: normal
P
We have been getting some interesting failures with ibv_reg_mr..
The max_mr number is on the order of 100K regions, yet we are only
able to register under 2500 regions on a mellanox card, and around
4800 regions on an ehca.
A testcase is available here:
http://source.scl.ameslab.gov/hg/ibv
Overall, looks safe.
If you want the fix in OFED 1.2, file a bug in the bugzilla.
I've made one adjustment to the original patch to set the hop_limit to 0xff on
receives. The updated patch is in the ib_router branch of my git tree.
But - is this patch going into 2.6.21? And if not, why does
https://bugs.openfabrics.org/show_bug.cgi?id=485
[EMAIL PROTECTED] changed:
What|Removed |Added
CC||[EMAIL PROTECTED]
--
Configure bug
> Quoting Hal Rosenstock <[EMAIL PROTECTED]>:
> Subject: Re: Re: [PATCH] IB/umad: fix GRH handling
>
> On Wed, 2007-03-21 at 18:22, Hal Rosenstock wrote:
> > On Wed, 2007-03-21 at 01:52, Sean Hefty wrote:
> > > >> Unfortunately, at least opensm cannot respond to SA queries issued
> > > >> from a
[EMAIL PROTECTED] wrote on Wed, 21 Mar 2007 11:23 -0700:
> > I've wondered about this for a while. In ib_umem_get, there is a
> > call to get_user_pages that does the work of virtual to physical
> > translation and increasing the ref counts. It is always invoked
> > with write == 1, even if c
Hi
I trying to compile OFED 1.2 beta OFED-1.2-20070322-0837.tgz
on SuSE9 SP3 x86_64 with 2.6.19 Kernel and Real Time PREEMPT patch.
I get two problems.
First, in file ofa_user-1.2/src/userspace/sdpnetstat/lib/fdd.c.
It has #include "if_fddi.h", and in "linux/if_fddi.h" I saw error message, that
> What about mthca QP reset issues?
I'm still thinking about synchronizing with the completion EQ's irq.
___
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit h
Hi Tzahi,
On 18:56 Wed 21 Feb , Tzachi Dar wrote:
>
> To be on the practical side, I have read the introduction to pthreads in
> the past and from what I saw it was relatively easy to implement that on
> Win32. I want to look at the functions that were mentioned before in
> this thread and se
On Wed, 2007-03-21 at 18:22, Hal Rosenstock wrote:
> On Wed, 2007-03-21 at 01:52, Sean Hefty wrote:
> > >> Unfortunately, at least opensm cannot respond to SA queries issued from a
> > >> remote subnet. I'm not sure how much work this would take to fix, or if
> > >> other SAs have this issue. Hal
This email was generated automatically, please do not reply
Common build parameters: --with-ipoib-mod --with-sdp-mod --with-srp-mod
--with-user_mad-mod --with-user_access-mod --with-mthca-mod --with-core-mod
--with-addr_trans-mod --with-rds-mod --with-cxgb3-mod
Passed:
Passed on i686 with 2.
> Quoting Roland Dreier <[EMAIL PROTECTED]>:
> Subject: [GIT PULL] please pull infiniband.git
>
> 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.o
20 matches
Mail list logo
