[ofa-general] Agenda for the OFED meeting today (Jan 5, 09)

2009-01-04 Thread Tziporet Koren
Hello all, I hope we all had nice holidays and vacations, and now it's the time to get back to business. Agenda for OFED meeting today: 1. Conclusions from OFED 1.4 release: Open discussion 2. Do we wish to have OFED 1.4.: Please send pros & cons before the meeting 3. OFED 1.5: Schedule and f

Re: [ofa-general] [PATCH] ipoib: failure during startup wiith non-default pkey set.

2009-01-04 Thread Roland Dreier
a) checkpatch.pl says: ERROR: trailing whitespace #11: FILE: drivers/infiniband/ulp/ipoib/ipoib_ib.c:724: +^I$ ERROR: trailing whitespace #13: FILE: drivers/infiniband/ulp/ipoib/ipoib_ib.c:726: +^I^I$ WARNING: braces {} are not necessary for single statement blocks #19: FILE: drivers/infiniband/

Re: [ofa-general] Re: [PATCH] mlx4_ib: fix for bugzilla 1383 (LSO packet processing)

2009-01-04 Thread Roland Dreier
So do you think my patch (which avoids all the code duplication and goto) is OK, or is that still too much overhead? Keep in mind the global impact of a larger I-cache footprint because of the code duplication... - R. ___ general mailing list general@l

[ofa-general] Re: [PATCH v3] ipoib: do not join broadcast group if interface is brought down

2009-01-04 Thread Roland Dreier
So what protects priv->broadcast? It seems that the only lock taken when setting broadcast to NULL is priv->lock. But eg here: - priv->mcast_mtu = IPOIB_UD_MTU(ib_mtu_enum_to_int(priv->broadcast->mcmember.mtu)); + if (priv->broadcast) + priv->mcast_mtu = IPOIB_UD_MTU(

[ofa-general] I'm not going to fix stupid problems in patches any more

2009-01-04 Thread Roland Dreier
Maybe I'm just in a cranky mood after the holidays, but I'm fed up with spending my time fixing stupid problems with patches sent to me. It seems the only way that I can teach people is to reject patches. So if you send me a patch that I can't apply because of some formatting problem that you sho

***SPAM*** Re: [ofa-general] [RDMA CM IPv6 PATCHv7 2/2] RDMA CM

2009-01-04 Thread Venkat Venkatsubra
To return  the error on the active side that "this iWARP device doesn't support IPv6", the following places are the possibility -rdma_connect() -rdma_resolve_addr() -qp setup time   rdma_resolve_addr() would probably be the earliest to handle this. For the driver to expose this lack of capability

[ofa-general] [openib-general] Infiniband on Debian etch RC1

2009-01-04 Thread songzhonglei
hi, i wanted to enable ipoib on debian,so recompiled kernel with infiniband supported. after reboot i modprobe ib_ipoib lsmod : ib_ipoib,ib_sa,ib_mthca,ib_mad,ib_core. and then ifconfig ib0 x.x.x.x.->ifconfig ib0 up,but failed. dmesg showed: ADDRCONF(NETDEV_UP):ib0:link is not ready ib0:multic

[ofa-general] ofa_1_4_kernel 20090104-0200 daily build status

2009-01-04 Thread Vladimir Sokolovsky (Mellanox)
This email was generated automatically, please do not reply git_url: git://git.openfabrics.org/ofed_1_4/linux-2.6.git git_branch: ofed_kernel Common build parameters: Passed: Passed on i686 with linux-2.6.16 Passed on i686 with linux-2.6.19 Passed on i686 with linux-2.6.18 Passed on i686 with

[ofa-general] infiniband on debian

2009-01-04 Thread songzhonglei
hi, i wanted to enable ipoib on debian,so recompiled kernel with infiniband supported. after reboot i modprobe ib_ipoib lsmod : ib_ipoib,ib_sa,ib_mthca,ib_mad,ib_core. and then ifconfig ib0 x.x.x.x.->ifconfig ib0 up,but failed. dmesg showed: ADDRCONF(NETDEV_UP):ib0:link is not ready ib0:multi