On 13 Jul 2005 10:34:36 -0400, Hal Rosenstock <[EMAIL PROTECTED]> wrote:
> On Wed, 2005-07-13 at 10:21, William Jordan wrote:
> > On 7/13/05, Talal jaafar <[EMAIL PROTECTED]> wrote:
> > > Thanks for the help, I was able to load the infiniband drivers using
> &
On 7/9/05, Roland Dreier <[EMAIL PROTECTED]> wrote:
> I'm attaching some (very hacky) python scripts to this email that can
> be used as a device management client to discover storage. Just untar
> the package and follow the README to use it.
The README appears to be missing from the PyOpenIB tar
On 7/13/05, Talal jaafar <[EMAIL PROTECTED]> wrote:
> Thanks for the help, I was able to load the infiniband drivers using
> your suggestion. However, I am still stuck at one point. I do not
> understand why is it happening but here what I am getting when I try
> to run "service network restart"
>
On 7/10/05, Talal jaafar <[EMAIL PROTECTED]> wrote:
> hi all,
>
> I am fairly new to OpenIB and I have downloaded and built the
> 2.6.12 kernel that has OpenIB support. My question now is how to
> compile the OpenIB modules in that kernel; There is no documentation,
> and all what I need to do i
On 6/20/05, Grant Grundler <[EMAIL PROTECTED]> wrote:
> On Mon, Jun 20, 2005 at 06:14:04PM +0300, Michael S. Tsirkin wrote:
> > Reducing the rx ring size gives me bandwidth win of about 1%.
>
> Michael,
> Doesn't this sound counter-intuitive to you?
>
> Can you share which configuration (chipset,
If you specify a service id of IB_CM_ASSIGN_SERVICE_ID, how do you
determine the service ID assigned by the SM? In the kernel, I suppose
you could just dereference the ib_cm_id struct, but what about from
the userspace cm?
--
Bill Jordan
SilverStorm Technologies
__
On 6/14/05, James Lentini <[EMAIL PROTECTED]> wrote:
>
> Sounds like I need to understand the difference between the
> ib_cm_req_param's retry_count and max_cm_retries fields. We set the
> former to 0 and the later to 4.
The retry_count is the number of retries you want to configure for
data on y
On 6/10/05, Roland Dreier <[EMAIL PROTECTED]> wrote:
>Hal> dev/core.c netdev_wait_allrefs says: * Any protocol or device
>Hal> that holds a reference should register * for netdevice
>Hal> notification, and cleanup and put back the * reference if
>Hal> they receive an UNREGISTER even
On 6/8/05, Libor Michalek <[EMAIL PROTECTED]> wrote:
> On Wed, Jun 08, 2005 at 01:29:42PM -0400, William Jordan wrote:
> > I'm trying to run netperf over SDP using libsdp. I'm having
> > intermittent success, but usually, I get a failure. The client does a
> &g
I'm trying to run netperf over SDP using libsdp. I'm having
intermittent success, but usually, I get a failure. The client does a
shutdown (SHUT_WR), but the server gets an ECONNRESET, "Connection
reset by peer", on the recv instead of a zero length recv.
I'm running 2568.
Anyone seen similar/dif
On 6/3/05, Roland Dreier <[EMAIL PROTECTED]> wrote:
>William> Should the device nodes /dev/infiniband_mad and
>William> /dev/infiniband_verbs exist at all? Are they artifacts of
>William> the way the classes work, or a setup problem on my
>William> system? Do other people have these
On 6/2/05, Libor Michalek <[EMAIL PROTECTED]> wrote:
> On Thu, Jun 02, 2005 at 03:43:47PM -0400, William Jordan wrote:
> >
> > The udev support/naming works fine. Regarding naming, the ib_uverbs
> > and ib_umad drivers only put port specific devices in the /dev/infiniba
Libor,
Ignore this patch. I'll submit a different one to change
the ib_ucm device node from /dev/infiniband/cm
to /dev/infiniband_cm. This makes the udev setup
unnecessary.
And, this probably wasn't the right README file for the udev
rules anyway. They belong with the module, not with the library
>Has anyone seen ib_send_cm_req() return -22?
I'm not sure what you are testing with, Jeff, but I ran into the same
problem the first time I tried to use ucm_simple. The source and
destination lid and guid are embedded in the source, and need to be
modified to fit the test systems. Otherwise you w
Why does the userlevel CM use devfs to create device nodes? Userlevel
verbs and mad layers appear to rely on udev.
--
Bill Jordan
SilverStorm Technologies
___
openib-general mailing list
openib-general@openib.org
http://openib.org/mailman/listinfo/openi
Sean,
You mentioned that the cm_id leak was fixed, but I don't see a patch
for it, so I'm
submitting one.
Fixed leak of cm_ids in server side of cmpost. Fixed an Oops if number
of connections
on server and client side don't match.
Signed-off-by: Bill Jordan <[EMAIL PROTECTED]>
Index: cmpost.c
On 5/7/05, Hugh Dickins <[EMAIL PROTECTED]> wrote:
> > My understanding is that mlock() could in theory allow the page to be moved,
> > but that currently nothing in the kernel would actually move it. However,
> > that could change in the future to allow hot-swapping of RAM.
>
> That's my underst
Patch to respond to invalid REQs (no listener) with a REJ message.
Signed-off-by: Bill Jordan <[EMAIL PROTECTED]>
Index: cm.c
===
--- cm.c(revision 2293)
+++ cm.c(working copy)
@@ -70,6 +70,7 @@ static struct ib_cm {
On 5/3/05, Andy Isaacson <[EMAIL PROTECTED]> wrote:
> Rather than replacing the fully-registered pages with pages of zeros,
> you could simply unmap them.
I don't like this option. It is nearly free to map all of the pages to
the zero-page. You never have to allocate a page if the user never
write
19 matches
Mail list logo