Roland Dreier wrote:
> > I tested it before on the Roland tree ( iboe branch ) and it fails,
> > because it writen in the way suitable for OFED. If adapt the patch to
> > the Roland tree, then appling Mellanox OFED patches will fail, because
> > it changes the same functions in the code.
> > H
Moni Shoua wrote:
>
>
>> It doesn't even look like this patch and the mlx4 patch were ever posted
>> to linux-rdma. Only to the EWG list.
>>
>>
> Not 100% correct. See thread from April 30.
> Patches to core, libibverbs and NES driver were presented there.
>
>
I'll go look, but I thou
Jason Gunthorpe wrote:
> On Wed, Jun 16, 2010 at 09:09:59AM -0500, Steve Wise wrote:
>
>
>> Granted our dev process may not be documented, but I always assumed the
>> general idea was to get changes accepted upstream, then pull into ofed.
>> OFED is just a mechanis
Jason Gunthorpe wrote:
> On Wed, Jun 16, 2010 at 12:12:06PM -0500, Steve Wise wrote:
>
>
>>>> Granted our dev process may not be documented, but I always assumed
>>>> the general idea was to get changes accepted upstream, then pull into
>>>> ofed.
Hey Joachim,
This bug is actually a duplicate of an old one I have open. You can
work around the problem by configuring in cxgb3 and iw_cxgb3 in your
kernel. Then building/installing ofed will overlay these two modules,
but the genalloc code will still have been pulled into the kernel. Does
Joachim Fenkes wrote:
> Steve Wise wrote on 20.07.2010 20:14:08:
>
>
>> This bug is actually a duplicate of an old one I have open. You can
>> work around the problem by configuring in cxgb3 and iw_cxgb3 in your
>> kernel. Then building/installing ofed wil
On 08/02/2010 10:47 AM, Tziporet Koren wrote:
> Agenda for the meeting today (Aug 2):
>
> 1. OFED 1.5.2 status& bugs:
>
> 767 maj sw...@opengridcomputing.com Non backport Kernels that don't
> build in genalloc cause c...
>
Do you want me to fix this issue for ofed-1.5.2? The fix
: Steve Wise
Date: Thu Aug 5 13:45:24 2010 -0500
Move genalloc services to iw_cxgb3.
Since iw_cxgb3 is the only module using the genalloc services, move
them
to iw_cxgb3 as static functions. This resolves bug 767.
Changes:
- Remove genalloc from all backports and ib_core
Vlad,
Please pull from
ssh://v...@sofa.openfabrics.org/~swise/scm/ofed_kernel ofed_1_5
This commit adds the cxgb3 backport for 2.6.33.
Thanks,
Steve.
RDMA/cxgb3: Add 2.6.33 backport.
Signed-off-by: Steve Wise
---
.../backport/2.6.33/iw_cxgb3_to_2.6.33.patch | 238
Hey Vlad,
Please pull these two critical bug fixes for iw_cxgb3 from:
ssh://v...@sofa.openfabrics.org/~swise/scm/ofed_kernel.git ofed_1_5
The are needed in ofed-1.5.2 for large cluster stability.
Thanks
Steve.
commit d3a225ba07fdd7bf4eccaab383818dde41231166
Author: Steve Wise
---
cxgb3_release_notes.txt | 20 +++-
1 files changed, 3 insertions(+), 17 deletions(-)
diff --git a/cxgb3_release_notes.txt b/cxgb3_release_notes.txt
index 30feda5..9a993c8 100644
--- a/cxgb3_release_notes.txt
+++ b/cxgb3_release_notes.txt
@@ -1,6 +1,6 @@
Open F
I just opened 2 critical bugs that I'd like to get into ofed-1.5.2.
They are needed for larger MPI cluster using iWARP. The bugs are 2131
and 2132. The fix for 2131 hits iw_cxgb3 only and is a small change to
turn off rx coalescing for iwarp connections. The fix for 2132 is a 1
liner to
Here are the fixes for bugs 2131 and 2132. You can pull these from:
ssh://v...@sofa.openfabrics.org/~swise/scm/ofed_kernel ofed_1_5
It would be great if these can get into ofed-1.5.2.
Thanks,
Steve.
Steve Wise (2):
cxgb3: Turn off RX coalescing for iwarp connections.
rdma_ucm
This is needed for NP256 sized iwarp clusters.
Signed-off-by: Steve Wise
---
.../fixes/core_0590_ucma_backlog_1024.patch| 13 +
1 files changed, 13 insertions(+), 0 deletions(-)
create mode 100644 kernel_patches/fixes/core_0590_ucma_backlog_1024.patch
diff --git a
This drops the connect latency from 200ms to 16. Needed for large
NP clusters.
Signed-off-by: Steve Wise
---
.../fixes/iw_cxgb3_1300_rx_coalesce_off.patch | 24
1 files changed, 24 insertions(+), 0 deletions(-)
create mode 100644 kernel_patches/fixes
Hello,
Looks like the latest ofa_kernel tree fails to run
ofed_scripts/ofed_makedist.sh for the 2.6.36 back/forward port. Looks
like an SDP issue:
git clone -q -s -n /home/swise/newgit/linux-2.6
/tmp/build-ofed_kernel-w25011/ofed_kernel-2.6.36
pushd /tmp/build-ofed_kernel-w25011/ofed_kerne
Vlad,
Please pull from
git://git.openfabrics.org/~swise/ofed_kernel.git ofed_1_5
This tree has a handful of cxgb3 upstream fixes as well as 7.11.0 firmware.
Thanks,
Steve.
___
ewg mailing list
ewg@lists.openfabrics.org
http://lists.openfabri
On 11/17/2010 03:31 PM, Steve Wise wrote:
Vlad,
Please pull from
git://git.openfabrics.org/~swise/ofed_kernel.git ofed_1_5
This tree has a handful of cxgb3 upstream fixes as well as 7.11.0
firmware.
Thanks,
Steve.
Forgot to include the commit logs.
--
commit
On 11/29/2010 10:31 AM, Robert D. Russell wrote:
Hi:
I have hit the following strange situation and would appreciate an
explanation as to how to fix it.
When calling ibv_create_cq() to create a queue pair, I set sq_sig_all
= 0.
When calling ibv_post_send() I set send_flags in the work reques
all out.
There should be a similar requirement in the IBTA IB Verbs specification...
Steve.
Thanks,
Bob Russell
r...@iol.unh.edu
On Mon, 29 Nov 2010, Steve Wise wrote:
On 11/29/2010 10:31 AM, Robert D. Russell wrote:
Hi:
I have hit the following strange situation and would appreciate an
Hey ofa developers,
My QA group is seeing problems running ofed-1.5.3-rc1 on a 32b system with 2.6.34 kernel. It seems the core drivers
aren't setting up the abi version correctly. The library wants ABI version 6, but when they cat
/sys/class/infiniband_verbs/abi_version it returns nothing.
Vlad,
Please pull from:
ssh://v...@www.openfabrics.org/~swise/scm/ofed_kernel.git ofed_1_5
Thanks,
Steve.
commit 21beaae06cea6cb10a3e8c626089b53d8ee6d5f6
Author: Steve Wise
Date: Mon Jan 31 12:32:34 2011 -0600
cxgb3: 2.6.36 backports.
Signed-off-by: Steve Wise
diff
I know we removed NFSRDMA for being enabled by default, but did
someone intentionally remove it altogether? Or is this a valid bug?
Vlad, do you know?
Original Message
Subject:
[Bug 2224] New: OFED-1
On 02/02/2011 02:47 AM, Vladimir Sokolovsky wrote:
On 02/01/2011 05:38 PM, Steve Wise wrote:
I know we removed NFSRDMA for being enabled by default, but did someone
intentionally remove it altogether? Or is this a valid bug?
Vlad, do you know?
NFSoRDMA was removed from the list of
Vlad,
Please pull from:
ssh://v...@www.openfabrics.org/~swise/scm/ofed_kernel.git ofed_1_5
Thanks,
Steve.
commit b56ea1597bcc9baf0f538a974aa6c50ca72e83a4
Author: Steve Wise
Date: Wed Feb 2 16:01:55 2011 -0600
cxgb3: 2.6.35 backports.
Signed-off-by: Steve Wise
diff --git
Hey Vlad,
Please pull 1.3 libcxgb3 for ofed-1.6:
http://www.openfabrics.org/downloads/cxgb3/libcxgb3-1.3.0.tar.gz
Thanks,
Steve.
___
ewg mailing list
ewg@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
In addition, I'll be adding the cxgb4 support. The kernel drivers will be in 2.6.37, but I'll need to add libcxgb4.
How should I proceed with that? IE what git repos should I be modifying to add libcxgb4?
Thanks,
Steve.
On 02/14/2011 11:19 AM, Steve Wise wrote:
Hey Vlad,
P
Hey Sean,
Can you peruse:
http://bugs.openfabrics.org/bugzilla/show_bug.cgi?id=2230
The changes I added to rping to fix problems I was seeing when running rping over the upstream iw_cxgb3 have been
included in OFED-1.5.3. That change, however breaks rping over iw_cxgb3 in 1.5.3. It causes a
d exits before destroying the CQ
It is possible for the CQ event thread to poll the CQ after it has been
destroyed which can result in a seg fault on T3 interfaces. This patch
waits for the thread to exit before destroying the CQ.
Signed-off-by: Steve Wise
Signed-off-by:
? If so, I'll ask to pull it into 1.5.3
-Original Message-
From: Steve Wise [mailto:sw...@opengridcomputing.com]
Sent: Tuesday, February 15, 2011 10:37 AM
To: Hefty, Sean
Cc: OpenFabrics EWG; Tziporet Koren
Subject: Re: rping/cxgb3 regression
On 02/15/2011 12:18 PM, Hefty, Sean
Hey Vlad,
I no longer see the rnfs-utils src rpm in the OFED-1.5.3 kit. Shouldn't it be
there?
___
ewg mailing list
ewg@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
Ignore this. I was mistaken.
On 02/17/2011 11:38 AM, Steve Wise wrote:
Hey Vlad,
I no longer see the rnfs-utils src rpm in the OFED-1.5.3 kit. Shouldn't it be
there?
___
ewg mailing list
ewg@lists.openfabrics.org
http://lists.openfabric
Signed-off-by: Steve Wise
---
release_notes/cxgb3_release_notes.txt | 10 +-
1 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/release_notes/cxgb3_release_notes.txt
b/release_notes/cxgb3_release_notes.txt
index 9a993c8..b1fcf4b 100644
--- a/release_notes
On 02/24/2011 09:32 AM, Vladimir Sokolovsky wrote:
Hi,
The following git trees will be used for OFED-1.6:
Kernel:
---
git://git.openfabrics.org/ofed_1_6/linux-2.6.git
branch: ofed_kernel_1_6
This tree is based on kernel 2.6.37.
Note:
No more "kernel_patches/fixes" directory. All changes (b
Unlike Infiniband, iWARP runs on top of TCP/IP and thus is routable.
On 03/03/2011 07:45 AM, Tanin wrote:
Thanks, Richard.
So, is that also apply to iWARP? If there is an IP router in between the path,
iWARP would not work, right?
On Thu, Mar 3, 2011 at 4:21 AM, rich...@informatix-sol.com
Hey Vlad,
If you haven't already, please pull in libcxgb3-1.3.0 into OFED-1.6.
Thanks,
Steve.
On 12/10/2010 01:11 PM, Steve Wise wrote:
The libcxgb3 package is a userspace driver for Chelsio T3 iWARP RNICs. It is a plug-in module for libibverbs that
allows programs to use Chelsio
On 03/15/2011 05:31 AM, daliu_771 wrote:
Hello, ewg
In the current OFED distribution package, can SDP over iwarp work ? And can SDP
over InfiniBand work ?
At this time, SDP only runs over Infiniband.
Steve.
___
ewg mailing list
ewg@lists.openfa
aging soon to finish cxgb4 inclusion in OFED-1.6.
Thanks,
Steve.
-
commit cb075d05851e4f5db08be91652e252ac9a425074
Author: Steve Wise
Date: Thu Mar 10 12:19:13 2011 -0600
Add cxgb4 drivers.
Signed-off-by: Steve Wise
commit 656f75cc4615e35936764e0060e1624c99d2648c
Author: Steve
On 03/28/2011 11:14 AM, Tziporet Koren wrote:
EWG/OFED Agenda for today:
1. OFED 1.6 schedule:
--
- Move to kernel 2.6.38 (since its GA already, and we have not started
backports)
- Ongoing work on backports - during Q2
- First RC - end of June
RCs every 2 weeks
On 03/29/2011 12:08 PM, Tziporet Koren wrote:
EWG/OFED meeting minutes for 28-Mar, 2011:
1. OFED 1.6 schedule:
---
- Move to kernel 2.6.38 (since its GA already, and we have not started
backports)
Hey Vlad,
Has this been completed yet? If not, when do you think
ib_write_bw is an Infiniband-only application. Try rdma_bw and rdma_lat and
use the --cma flag.
Steve.
On 04/21/2011 05:39 PM, ren yufei wrote:
Hi,
Two hosts connected through both IB and iWarp.
1 'ib_write_bw' success via IB connection, while failure via iWarp connection.
2 Both clinet and
This was the nice feature of OFED-1.6 (plus a cleaned up kernel bldenv with no
kernel_patches/fixes)...
On 06/13/2011 11:52 AM, Syoyo Fujita wrote:
Hello Tziporet,
1. Cancel OFED 1.6 and instead have OFED 2.0 in 2012 - work will start after
XRC will be accepted to the kernel
2. Add OFED 1.
Vlad,
Please pull these three commits from:
ssh://v...@sofa.openfabrics.org/home/vipulp/scm/ofed_kernel.git ofed_1_5
These patches add the upstream cxgb4 and iw_cxgb4 modules, as well as the backport for 2.6.39. We'll be pushing more
backport patches as we get them developed. For now, please
I think these patches break the RHEL5 NFSRDMA backports. I'll
adjust them.
Steve.
On 07/25/2011 04:10 PM, Jeff Becker wrote:
Hi Vlad
These patches comprise an initial cleanup for NFSRDMA (e.g., we
don't check out all
On 07/27/2011 02:13 PM, Jeff Becker wrote:
Hi Steve. I only tested these against 2.6.30 (OFED 1.5.4 base). I haven't
gotten to backports yet. The main change is not to check out and build NFS,
since that conflicted with the distros. Anyway, any help is appreciated. Thanks.
Jeff
Steve
On 07/29/2011 01:42 PM, Jeff Becker wrote:
When Tziporet and I discussed this for OFED 1.5.4, we agreed that NFSRDMA would be supported on newer distros (RHEL 6,
SLES 11) which do not need the fs/nfs stuff. If we need to support the older distros, we probably do need fs/nfs
since their kernels
Hey Vlad,
The release notes for OFED-1.5.3.1 only claim upstream kernels from 2.6.29 and above. Should we really bother with
building against older upstream kernels?
Snipit from the release notes:
o Linux Operating Systems:
- RedHat EL5 up42.6.18-164.el5
- RedHat
Vlad,
Please pull these commits from:
ssh://v...@sofa.openfabrics.org/home/vipulp/scm/ofed_kernel.git ofed_1_5
They add Chelsio T4 driver backports for 2.6.38-2.6.35 plus fix a problem with
the previous 2.6.39 backport.
Thanks,
Steve.
commit 88bfe7ef260c44e78acec0bde47432a27921635e
A
Since the OFED-1.5 kernel sources are based on 2.6.30, there is an assumption in the configure logic that no backports
are needed for 2.6.30. The cxgb4 src base, however, is from the 3.0 kernel. So yes, you need to add a 2.6.30 (and i
guess a 2.6.29) backports tree in kernel_patches/backports and
Anybody have a comment on this? Should we be maintaining backports on kernels
we don't support?
On 08/01/2011 10:17 AM, Steve Wise wrote:
Hey Vlad,
The release notes for OFED-1.5.3.1 only claim upstream kernels from 2.6.29 and above. Should we really bother with
building against
Vlad,
Please pull these commits from:
ssh://v...@sofa.openfabrics.org/home/vipulp/scm/ofed_kernel.git ofed_1_5
They add more cxgb4 backports, plus fix a cxgb3 build issue.
Thanks,
Steve.
-
Kumar Sanghvi (10):
cxgb4: Updates to 2.6.34 backport
iw_cxgb4: Add 2.6.34 backport
On 08/09/2011 06:58 AM, Tziporet Koren wrote:
Hi All
In our last EWG meeting there was a request to delay OFED 1.5.4 release
Original plan:
First RC: 8/15
GA: 10/11
New suggestion:
First RC: 8/30
GA: 11/1
I had actually requested a 3 week push. So that would be 9/5.
Plea
Hey all,
I can't make the call today, comments in-line below.
Steve.
On 08/22/2011 10:13 AM, Vladimir Sokolovsky wrote:
EWG/OFED agenda for - Aug 22, 2011:
1. OFED 1.5.4 releases:
- Features:
- FDR support - should be submitted next week
- New OpenSM with 3D torus - done
Vlad,
Please pull these commits from:
ssh://v...@sofa.openfabrics.org/home/vipulp/scm/ofed_kernel.git ofed_1_5
They add more cxgb4 backports.
Thanks,
Steve.
---
commit 6128a5d33674df915a196f6deb3b741706b8f7c0
Author: Kumar Sanghvi
Date: Sat Aug 27 12:15:49 2011 +0530
iw_cxgb4/cxgb
Vlad/Tziporet,
We have a large patch that creates a new backports tree in kernel_addons/backport/2.6.16_sles10_sp4 and
kernel_patches/backport/2.6.16_sles10_sp4. This was done to get cxgb4 backported. We've included all the sles10sp3
patches, and then modified them and added cxgb4-specific pat
On 08/30/2011 09:50 AM, Steve Wise wrote:
Vlad/Tziporet,
We have a large patch that creates a new backports tree in kernel_addons/backport/2.6.16_sles10_sp4 and
kernel_patches/backport/2.6.16_sles10_sp4. This was done to get cxgb4 backported. We've included all the sles10sp3
patches
On 08/30/2011 10:44 AM, Vladimir Sokolovsky wrote:
On 08/30/2011 05:50 PM, Steve Wise wrote:
Vlad/Tziporet,
We have a large patch that creates a new backports tree in kernel_addons/backport/2.6.16_sles10_sp4 and
kernel_patches/backport/2.6.16_sles10_sp4. This was done to get cxgb4 backported
Here are the details:
We need this iw_cxg3 upstream fix. Chelsio will do this asap.
commit b955150ea784af4c193b708a2e8091673bf23004
Author: Steve Wise
Date: Thu Oct 21 12:37:06 2010 +
RDMA/cxgb3: When a user QP is marked in error, also mark the CQs in error
Also we need these
Original Message-----
From: Steve Wise [mailto:sw...@opengridcomputing.com]
Sent: Wednesday, September 14, 2011 7:55 AM
To: Hefty, Sean
Cc: Vladimir Sokolovsky; Tziporet Koren; OpenFabrics EWG; Divy Le Ray; Kumar A
S; Vipul Pandya
Subject: T3 change needed for OFED-1.5.4
Hey Sean,
Do you remember we p
Vlad,
Please pull this commit from:
ssh://v...@sofa.openfabrics.org/~swise/scm/ofed_kernel.git ofed_kernel
This pulls in 2 RDS iWARP fixes that are heading upstream.
Thanks,
Steve.
-
commit 43bc1e73f2357f38b0b9c73317c0eb7913f2472f
Author: Steve Wise
Date: Wed Sep 28 10:05:20 2011
Vlad, please pull libcxgb3-1.3.1 into ofed-1.5.4.
Thanks,
Steve.
Original Message
Subject:[ANNOUNCE] libcxgb3 version 1.3.1 released
Date: Tue, 11 Oct 2011 09:38:43 -0500
From: Steve Wise
To: linux-rdma
Hello,
Version 1.3.1 of libcxgb3 is available for
in upstream CQE flush fixes.
Bring in changes fixing iw_cxgb4 cqe logic into OFED.
These changes fix RQ QID / SQ QID mixups in insert/count
functions in the t4_cqe.
This fixes RDS crashes using iw_cxgb4.
Signed-off by: Jonathan Lallinger
Signed-off by: Steve Wise
commit
That's strange.
Maybe I sent the wrong patch. I better re-test things.
Vlad, can you please remove this commit? I'll re-submit this patch later.
Thanks Jeff,
Steve.
On 10/25/2011 12:41 PM, Jeff Becker wrote:
Hi Steve. Looks like your patch is DOS formatted. I can remove the carriage
ret
Vlad, please pull libcxgb4-1.2.0 into OFED-1.5.4.
Thanks,
Steve.
On 11/10/2011 11:09 AM, Steve Wise wrote:
Hello,
Version 1.2.0 of libcxgb4 is now available for download at:
http://www.openfabrics.org/downloads/cxgb4/libcxgb4-1.2.0.tar.gz
with md5sum:
9e26aee8ec9975336e27439791bf9615
On 11/29/2011 05:32 AM, Tziporet Koren wrote:
EWG/OFED meeting agenda for today:
1. OFED 1.5.4:
Status:
RC4 is available since Nov 10
There are still some critical issues that need to be fixed
We wish to build final RC this week so we can have the GA next week
We will review the list of critic
On 12/05/2011 01:05 AM, Or Gerlitz wrote:
To be concrete/constructive here, a per IB stack module individual has to be assigned for that backporting, which
doesn't mean "make IB code from kernel X to build under kernel Y" - lets
see if we have people to actually do that.
For example, on the i
In case EWG missed this, note Roland's pointer below.
Original Message
Subject:Re: [RFC] – Proposal for new process for OFED releases
Date: Tue, 6 Dec 2011 10:59:13 -0800
From: Roland Dreier
To: Or Gerlitz
CC: Hefty, Sean , Tziporet Koren , ewg@lists.openfabri
On 12/22/2011 02:26 PM, Edward Mascarenhas wrote:
There is a suggestion from Eduard from Qlogic to have the backports ad
different branches and not as today.
He will start a mail thread on his suggestion.
The attached file in pdf format has the proposal. It is also inlined in
text below so its
Fine with me.
Steve.
On 02/21/2012 10:29 AM, Tziporet Koren wrote:
Hi All
Vlad will send the new OFED proposal only tomorrow and I think we should have
the meeting after everybody will try and review it.
We can do it next week the same time if it's good for everyone
If yes I will send a new
Hey vlad, I ran the admin_rdma.sh script and it logged a few fatal errors:
fatal: No annotated tags can describe
'6343adaefb8a7a21d9ae1a018159f90f482cac61'.
However, there were unannotated tags: try --tags.
Updated from local tree: /tmp/git/linux-3.2/
Origin remote URL: git://git.openfabrics.or
I cannot make today's meeting. I think we should go forward with the new plan.
Steve.
On 03/06/2012 10:17 AM, Tziporet Koren wrote:
Agenda for today:
- Decide if we move forward with the new suggestion for the OFED package and
backports method.
Tziporet& Vlad
___
Hey Tziporet,
Chelsio is definitely committed to OFED and we'll get cxgb3/4 backports
done soon. We'll discuss timelines in the next call.
Thanks,
Steve.
On 4/18/2012 10:04 AM, Tziporet Koren wrote:
Hi All,
We started to work on OFED 3.2 and for now we already have the base code
working
Hey vipul, I recall us having to make a change to the cxgb3/cxgb4
makefiles so that they included the correct header files, because they
both have the same names, like l2t.h.If you look at the ofed-1.5.4
tree I think you'll see this.
On 5/3/2012 1:50 PM, Vipul Pandya wrote:
On 03-May-2
I cannot attend today's ewg call, but I wanted to give you a status
update on the chelsio backports:
- the cxgb3/4 backports are code-complete for 3.0, 3.1, and rhel6.2.
- testing and code review under way
- will submit these this week
Thanks,
Steve.
On 05/15/2012 10:12 AM, Woodruff, Robe
On 05/17/2012 11:43 AM, Woodruff, Robert J wrote:
We discussed people's progress towards our goal of getting the initial
backports for OFED-3.3 complete. The status is:
NES driver: Patches have been submitted but have not yet been applied.
Intel IB QIB Driver:
SLES 11 SP2; no backport
This series brings forth the discussion we had tuesday in the EWG meeting.
Chelsio is proposing ofed developers use the linux-next-cherry-picks
directory for patches upstream and critical to a driver's functionality,
yet too big for the stable flow. Once OFED releases with a new kernel
that i
On 6/13/2012 7:55 AM, Tziporet Koren wrote:
2.2 Next step:
Discussion if to move to kernel 3.5-rc2 (instead of GA kernel)
Pros:
- Issues we will fix can be submitted to the same kernel
- Have all latest and greatest fixes and features already in the kernel
On 6/17/2012 9:23 AM, Tziporet Koren wrote:
Q: So are we still planning to release ofed-3.2? Do we have a projected
release date for this or whatever we're releasing next?
A: We will not have 3.2 GA. We will start working on 3.5 this week and will see
progress.
I hope we will have dates in
The oldstyle=y logic in ibdev2netdev to match netdevs to ibdevs didn't
work for RNIC devices since their mac address is left justified in
the gid. This patch uses the type attribute to do the right thing for
RNIC devices. Note: the oldstyle=n logic works.
Signed-off-by: Steve
Hey Jeff,
Can you tell us the status of NFSRDMA in OFED-3.x? IE what backports,
if any, are complete, and what testing have you done on various distros?
Thanks,
Steve.
On 7/5/2012 11:40 AM, Jeff Becker wrote:
Hi Vlad. Please apply the attached patch to compat-rdma. It alllows
NFSRDMA to b
I cannot make today's call, but Chelsio will begin the backports asap.
STeve.
On 7/10/2012 10:28 AM, Tziporet Koren wrote:
1. OFED 3.5. status:
- Base kernel package should be ready today
- RHEL 6.3/6.2 should be ready this week
- SLES11 SP2 should be ready next week
All asked
Hey Tziporet,
Sorry I missed today's call. Chelsio's QA is running now and they've
completed around 30%. There are some outstanding bugs blocking progress
that we're working these actively. We'll have code drops for these in
the pipeline.
Steve.
On 09/04/2012 10:53 AM, Tziporet Koren
Hey Tziporet,
I cannot make today's meeting. I'll email Chelsio's OFED-3.5 status asap.
___
ewg mailing list
ewg@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
On 10/16/2012 11:28 AM, Tziporet Koren wrote:
Meeting Summary:
---
1. OFED 5.3:
RC2 was published
o Linux Operating Systems:
- RedHat EL6.2 2.6.32-220.el6
- RedHat EL6.3 2.6.32-279.el6
- SLES11 SP2 3.0.13-0.27
-boun...@lists.openfabrics.org] On Behalf Of Steve Wise
Sent: Tuesday, October 16, 2012 7:31 PM
To: ewg@lists.openfabrics.org
Cc: Vipul Pandya
Subject: Re: [ewg] EWG/OFED meeting minutes for Oct 16, 2012
On 10/16/2012 11:28 AM, Tziporet Koren wrote:
Meeting Summary:
---
1
On 10/24/2012 6:31 AM, Vipul Pandya wrote:
Hi All,
As many of us might be aware that kernel.org maintains a separate repo for the
adapter firmware. Chelsio keeps updating its latest firmware to this repo. For
our T3 and T4 adapter's robust functionality and for ease of use for the
customer, w
I cannot attend today's meeting again. Actually I have a emergency
daily meeting that conflicts with this on.
Here is Chelsio's status:
- we have a patch series that resolves scalability bugs that we need in
OFED-3.5. The patch is tested and being reviewed now
- we also would like to ship FW
poret
*From:*Steve Wise [mailto:sw...@opengridcomputing.com]
*Sent:* Tuesday, October 30, 2012 5:01 PM
*To:* Tziporet Koren
*Cc:* ewg@lists.openfabrics.org; Divy Le Ray; Vipul Pandya
*Subject:* Re: EWG/OFED bi-weekly meeting
I cannot attend today's meeting again. Actually I have a emerge
On 11/13/2012 10:27 AM, Tziporet Koren wrote:
Hi All
Unfortunately I cannot attend the meeting today
Woody - I hope you can replace me
All we need to close today is whether Chelsio and Intel drivers are ready so
that Vlad can build RC3
I cannot attend today either. Vipul is out for Diwali.
Hey Tziporet,
Chelsio is happy with the current OFED-3.5 bits at this point. There are
a few outstanding issues, but none to stop ship for.
We're ready to release.
Thanks,
Steve.
___
ewg mailing list
ewg@lists.openfabrics.org
http://lists.openfabri
We have a critical regression that we need fixed that hits libcxgb4.
We'd like to push a new libcxgb4 release for OFED-3.5. Should be today
or tomorrow.
___
ewg mailing list
ewg@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listi
http://bugs.openfabrics.org/bugzilla/show_bug.cgi?id=2408
On 12/13/2012 9:43 AM, Steve Wise wrote:
We have a critical regression that we need fixed that hits libcxgb4.
We'd like to push a new libcxgb4 release for OFED-3.5. Should be today
or tom
It turns out this libcxgb4 problem is best resolved for OFED-3.5 by
removing the kernel change that introduced this issue.
Vipul will be posting a patch soon.
Steve.
On 12/13/2012 9:53 AM, Steve Wise wrote:
http://bugs.openfabrics.org/bugzilla/show_bug.cgi?id=2408
On 12/13/2012 9:43 AM
On 12/18/2012 10:30 AM, Tziporet Koren wrote:
Agenda for today:
1. OFED 3.5:
- RC3 - was done on 12/11/12
- A blocker issue was found in Chelsio - a fix is in latest daily build
Need Chelsio approval and then we can build RC4
I confirmed Vipul's fix is in the latest daily build, so we're re
t for the new T5
hardware.
Shortlog from 1.2.0:
Steve Wise (11):
Disable strict aliasing warnings.
Remove useless dma_addr_t define.
Silently eat unsignaled read response cqes.
flush bug can cause application crash
Add new T4 device IDs.
zero out r3 and r4 fields
Vlad, this should be pulled into the next OFED release.
Thanks!
Steve.
Original Message
Subject:[ewg] [ANNOUNCE] libcxgb4 1.3.0 is released
Date: Wed, 03 Apr 2013 10:18:42 -0500
From: Steve Wise
To: linux-r...@vger.kernel.org
CC: Vipul Pandya , &quo
...@lists.openfabrics.org] *On Behalf Of *Tziporet Koren
*Sent:* Sunday, April 14, 2013 12:18 PM
*To:* ewg@lists.openfabrics.org; Steve Wise
*Subject:* Re: [ewg] EWG/OFED bi-weekly meetings
Tomorrow is Israel Independence day
Thus I will not be able to attend
Woody will run the meeting or mail all if it is
Hey Rupert, I just read the minutes, and I wanted to let folks know that FBSD
already has a port of
OFED. And I believe mlx and cxgb are there.
From: ewg-boun...@lists.openfabrics.org
[mailto:ewg-boun...@lists.openfabrics.org] On Behalf Of
Rupert Dance
Sent: Saturday, August 10, 2013 7:4
plus
copyright updates.
Shortlog from 1.3.0:
Steve Wise (12):
fixed false uninitialized var warning.
Handle qp flushes with pending sw cqes.
accounting for unsignled sq wrs fails to deal with wrap.
clean up queue dump logic.
flush problem with unsignaled wrs.
e
o*Vlad*- 9/30/2013
§Review all of the bugs assigned to Mellanox developers with his team
o*Steve Wise*- 9/30/2013
§Steve Wise agreed to review bugs 2402 and 2445 and try to get them
resolved
o*Woody*- 9/30/2013
§Check with Arlin regarding the status of the bugs in OFED 3.5 and 3.5-x
1 - 100 of 630 matches
Mail list logo