Re: [ewg] Fixes for bug 1596 (for 2.6.25, 2.6.22 and RHEL5.3)

2009-05-18 Thread Vladimir Sokolovsky
Jeff Becker wrote: Hi Vlad. Please pull from: git://git.openfabrics.org/~jbecker/nfs_rdma_ofed_1_4.git Thanks. -jeff ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg Done, Regards, Vladimir _

[ewg] [PATCH OFED-1.4.1] NFSRDMA: RHEL5.1 and RHEL5.2 fixes for bug 1596

2009-05-18 Thread Jon Mason
openibd cannot be stopped if NFS is running due to a dependency to code added to ib_core. This dependency code does not need to be added to ib_core and can be added to the NFS module to remove the need for ib_core. By modifying it in this way, openibd can be stopped without issue. The patch belo

[ewg] Fixes for bug 1596 (for 2.6.25, 2.6.22 and RHEL5.3)

2009-05-18 Thread Jeff Becker
Hi Vlad. Please pull from: git://git.openfabrics.org/~jbecker/nfs_rdma_ofed_1_4.git Thanks. -jeff ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg

RE: [ewg] update our bugzilla with components and Default Assignee

2009-05-18 Thread Tung, Chien Tin
>This is the list of components we have so far. If someone wish to add >more please let me know Please add nes driver: NetEffect iWARP driver and I will be the default assignee, Chien Tung . Thanks, Chien ___ ewg mailing list ewg@lists.openfabrics.or

[ewg] Re: update our bugzilla with components and Default Assignee

2009-05-18 Thread Johann George
Hello Tziporet. I'll continue to maintain qperf so you can keep me as the Default Assignee. I do know that the following are incorrect: >* ipath driver (its jeremy.br...@qlogic.com) >* Also - does anyone know what is: utils Random userspace tools > maintained by b...@pathscale.com

[ewg] Re: EWG/OFED meeting agenda for today (May 18)

2009-05-18 Thread Steve Wise
Tziporet Koren wrote: This is the agenda for today's EWG/OFED meeting: 1. OFED 1.4.1 bugs status and decision on RC6 date 1628 blo andy.gro...@oracle.com RDS in 1.4.1 cannot connect to RDS in 1.3.1 - I think Andy sent a fix for this 1596 cri jeffrey.c.bec...@nasa.gov openibd stop failed whe

RE: [ewg] Bug in IMB included in OFED

2009-05-18 Thread Supalov, Alexander
Hi, The big fix has been forwarded to the IMB maintainer for review. Best regards. Alexander -Original Message- From: Pavel Shamis (Pasha) [mailto:pash...@gmail.com] Sent: Sunday, May 17, 2009 10:38 AM To: Jeff Squyres; Supalov, Alexander Cc: OpenFabrics EWG Subject: Re: [ewg] Bug in

[ewg] EWG/OFED meeting agenda for today (May 18)

2009-05-18 Thread Tziporet Koren
This is the agenda for today's EWG/OFED meeting: 1. OFED 1.4.1 bugs status and decision on RC6 date 1628blo andy.gro...@oracle.com RDS in 1.4.1 cannot connect to RDS in 1.3.1 - I think Andy sent a fix for this 1596cri jeffrey.c.bec...@nasa.gov openibd stop

[ewg] OFED teleconf today

2009-05-18 Thread Jeff Squyres
HIT "3" WHEN YOU DIAL IN TO GET TO WEBEX!! HIT "3" WHEN YOU DIAL IN TO GET TO WEBEX!! HIT "3" WHEN YOU DIAL IN TO GET TO WEBEX!! Friendly reminder: the OFED teleconference is *today* (18 May 2009). Next few scheduled teleconferences: 1. Monday, May 18, noon US Eastern / 9am US

[ewg] [PATCH 2/2] ib_mthca: Use module parameter for number of MTTs per segment

2009-05-18 Thread Eli Cohen
The current MTTs allocator uses kmalloc to allocate a buffer for it's buddy system implementation and thus is limited by the amount of MTT segments that it can control. As a result, the size of memory that can be registered is limited too. This patch uses a module parameter to control the number of

[ewg] [PATCH 1/2] mlx4_core: Use module parameter for number of MTTs per segment

2009-05-18 Thread Eli Cohen
The current MTTs allocator uses kmalloc to allocate a buffer for it's buddy system implementation and thus is limited by the amount of MTT segments that it can control. As a result, the size of memory that can be registered is limited too. This patch uses a module parameter to control the number of

Re: [ewg] Bug in IMB included in OFED

2009-05-18 Thread Tziporet Koren
Jeff Squyres wrote: There is a bug in the IMB MPI test suite that is included in OFED (the mpitests SRPM). Who is the maintainer for that? I would file a bug on bugzilla, but there's no category for the MPI tests. I added MPI tests to bugzilla with Pasha as a maintainer Tziporet _

[ewg] update our bugzilla with components and Default Assignee

2009-05-18 Thread Tziporet Koren
Hi, I wish to update our bugzilla with missing components and default assignees (so it will not be people who are not working on OFED any more, or bugzilla) I have some questions on who should be the Default Assignee for: * ipath driver (its jeremy.br...@qlogic.com) * iSER (its dor...@v

[ewg] [PATCH] Perftest: attach QP to a multicast group in INIT state rather then RESET state

2009-05-18 Thread orenmeron
commit 95a83ad74db6a9b996ec3602e02d3c4cd5ff648a Author: Hoang-Nam Nguyen Date: Mon May 18 09:14:43 2009 +0300 Perftest: attach QP to a multicast group in INIT state rather then RESET state Signed-off-by: Oren Meron diff --git a/send_bw.c b/send_bw.c index afabfa4..fdb88a1 100755 ---