Re: [openib-general] [RFC/PATCH] librdmacm: use the ipoib broadcast group qkey

Sean Hefty wrote: >> Maybe just ask user to always call rdma_join_multicast after rdma_create_qp? >> Joins are now properly reference counted, so it shouldn't be a problem >> to repeat this any number of times. Right? > > This is the solution for now, and it should work fine. I don't think it wou

Re: [openib-general] [PATCH 2/2] librdmacm: add support to join IPOIB multicast groups

Sean Hefty wrote: > Add to the librdmacm an IPOIB port space that allows interoperability with > IPoIB multicast traffic. Use of the RDMA_PS_IPOIB is limited to multicast > join/leave. the two patches seems fine, however i will not be able to test them today being out of the office all the day,

Re: [openib-general] [RFC/PATCH v3] rdma/cma: add RDMA_PS_IPOIB port space

Sean Hefty wrote: > I was thinking of SIDR, but what about connected mode ipoib? This could > make the ipoib port space interesting, or require breaking it into two > separate port spaces, or... I'm only going to worry about multicast for > now, unless there's a reason to consider other use.

Re: [openib-general] win related [was: Re: [PATCH 1/2] opensm: sigusr1: syslog() fixes]

Hi Hal, Hal Rosenstock wrote: > Hi again Yevgeny, > > On Tue, 2007-01-23 at 11:20, Yevgeny Kliteynik wrote: >> Hi Hal, >> >> Hal Rosenstock wrote: >>> Hi Yevgeny, >>> >>> On Sun, 2007-01-21 at 04:17, Yevgeny Kliteynik wrote: Sasha Khapyorsky wrote: > On 18:24 Thu 11 Jan , Yevgeny Kli

Re: [openib-general] [PATCH 1/2] rdma_cm: add support to join IPOIB multicast groups

Sean Hefty wrote: > Add to the rdma_cm an IPOIB port space that allows interoperability with > IPoIB multicast traffic. Use of the RDMA_PS_IPOIB is limited to multicast > join/leave. OK, Sean the patch looks perfectly fine for allowing multicast interoperability with IPoIB. However, it will not

[openib-general] nightly osm_sim report 2007-01-24:normal completion

OSM Simulation Regression Summary OpenSM rev = Tue_Jan_23_10:07:29_2007 5becf9 ibutils rev = Wed_Jan_3_11:42:12_2007 913448 Total=410 Pass=410 Fail=0 Pass: 30 Stability IS1-16.topo 30 Pkey IS1-16.topo 30 OsmTest IS1-16.topo 30 OsmStress IS1-16.topo 30 Multicast IS1-16.topo 30 LidMgr IS1-16.topo

Re: [openib-general] [PATCH] libibverbs: should not access NULL pointer.

Yes, you have. I had a few days old bits when I saw this, sorry. Michael's suggestion is good though. thanks, - KK Roland Dreier <[EMAIL PROTECTED]> wrote on 01/23/2007 11:19:17 AM: > I already fixed this in commit 789728f7, right? ___ openib-gener

[openib-general] [PATCH] ib_addr: Handle Ethernet neighbour updates during route resolution.

Handle Ethernet neighbour updates during route resolution. The IWCM uses the ib_addr services to do route resolution (neighbour discovery in the IP world). The ib_addr netevent callback routine, however, currently only acts on Inifininband neighbour updates. It needs to act on ethernet neighbou

Re: [openib-general] RDMA CM multicast

> rdma_join_multicast(0.0.0.0, port 0) <- exchange group info out of band Trying to work through this more, having the first node join seems trivial. Getting additional nodes to join the same group through the rdma_cm is proving more difficult... The MGID of the group would need to be treated

[openib-general] [RFT] [PATCH] Add ABI compatibility for apps linked against libibverbs 1.0

Here is a patch that attempts to make it possible use libibverbs 1.1 (ie the libibverbs.git master branch) to run applications compiled and linked against libibverbs 1.0. I would appreciate it if people who have such applications could test this out. In particular, I would like to know the result

Re: [openib-general] [PATCH 2/2 vex branch] IB/VNIC Fix failover delay issue

thanks, I (finally) rolled these into my vex branch. ___ openib-general mailing list openib-general@openib.org http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general

Re: [openib-general] OFED release testing Task force meeting minutes

You're correct - it is 8:30 PDT Nimrod Gindi Mellanox Technologies Ltd. mail: [EMAIL PROTECTED] Cellular: +1-408-750-4801 Office:+1-347-342-0011 Fax:+1-212-987-0275 - Original Message - From: Jeremy Brown <[EMAIL PROTECTED]> To: Ni

Re: [openib-general] [PATCH 2/2] opensm: cleanup sa related _ctrl layer

On Fri, 2007-01-19 at 21:57, Sasha Khapyorsky wrote: > Cleanup SA related *_ctrl layer. > > Signed-off-by: Sasha Khapyorsky <[EMAIL PROTECTED]> Thanks. Applied. -- Hal ___ openib-general mailing list openib-general@openib.org http://openib.org/mailma

Re: [openib-general] [PATCH 1/2] opensm: cleanup sm related _ctrl layer

On Fri, 2007-01-19 at 21:56, Sasha Khapyorsky wrote: > This cleanups sm related _ctrl layer. > > Signed-off-by: Sasha Khapyorsky <[EMAIL PROTECTED]> Thanks. Applied. -- Hal ___ openib-general mailing list openib-general@openib.org http://openib.org/m

Re: [openib-general] OFED release testing Task force meeting minutes

On Tue, 2007-01-23 at 10:56 -0800, Nimrod Gindi wrote: > Follow-up meeting will be scheduled for 31st January 2007 8:30am > PDT=11am EDT=6pm Israel. Just a double-check: Shouldn't this read "8:30 am PDT = 11:30 am EDT = 6:30 pm Israel?" I want to make sure I don't show up half an hour late. :)

[openib-general] [PATCH 2.6.21 4/4] ehca: remove obsolete prototypes

Here is a patch for ehca_classes.h that removes obsolete prototypes. Thanks Nam Signed-off-by: Hoang-Nam Nguyen <[EMAIL PROTECTED]> --- ehca_classes.h | 14 -- 1 files changed, 14 deletions(-) diff -Nurp infiniband_orig/drivers/infiniband/hw/ehca/ehca_classes.h infiniband_work

[openib-general] [PATCH 2.6.21 3/4] ehca: remove do_mmap()

This patch removes do_mmap() from ehca: - Call remap_pfn_range() for hardware register block - Use vm_insert_page() to register memory allocated for completion queues and queue pairs - The actual mmap() call/trigger is now controlled by user space, ie. libehca Thanks Nam Signed-off-by: Hoang-Nam

[openib-general] [PATCH 2.6.21 2/4] ehca: fix mismatched use of spin_unlock in irq handler

Here is a patch for ehca_irq.c that fixes mismatched use of spin_unlock in irq handler. Thanks Nam Signed-off-by: Hoang-Nam Nguyen <[EMAIL PROTECTED]> --- ehca_irq.c |3 ++- 1 files changed, 2 insertions(+), 1 deletion(-) diff -Nurp infiniband_orig/drivers/infiniband/hw/ehca/ehca_irq.c

[openib-general] [PATCH 2.6.21 1/4] ehca: fix improper use of yield with spinlock held

Here is a patch for ehca_cq.c that fixes improper use of yield with spinlock held. Thanks Nam Signed-off-by: Hoang-Nam Nguyen <[EMAIL PROTECTED]> --- ehca_cq.c |5 - 1 files changed, 4 insertions(+), 1 deletion(-) diff -Nurp infiniband_orig/drivers/infiniband/hw/ehca/ehca_cq.c infin

[openib-general] [PATCH 2.6.21 0/4] ehca: remove do_mmap() and some bug fixes

Hello Roland! Here is a patch set for ehca as a result of previous disscussions and comments: 1. fix improper use of yield within spinlock context 2. fix mismatched use of spin_unlock in irq handler 3. remove do_mmap() 4. remove obsolete prototypes PS: I've sent the first two recently for 2.6.20, b

Re: [openib-general] RDMA CM multicast

Posting to openib-general list... >RDMA CM has multicast of course, though it seems no means of preventing >address collisions (to me, that means two separate MPI jobs using the >same multicast address). I know that part of the new multicast support >you had developed a few months ago was the abi

[openib-general] [PATCH RFC ] ofed_1_2 simulate neighbour update events by snooping ARP packets

OFED/iWARP Developers, Here is a proposal for supporting the minimum required neighbour update event notifications needed for iwarp devices on the older kernels supported by ofed. This patch is a request for comments. Please review. If you think it looks ok, then I'll provide patches to all t

[openib-general] [PATCH 2/2] librdmacm: add support to join IPOIB multicast groups

Add to the librdmacm an IPOIB port space that allows interoperability with IPoIB multicast traffic. Use of the RDMA_PS_IPOIB is limited to multicast join/leave. Rename the RDMA_UD_QKEY to RDMA_UDP_QKEY to signify that the qkey is only used with the RDMA_PS_UDP port space. Update mckey to allow t

[openib-general] [PATCH 1/2] rdma_cm: add support to join IPOIB multicast groups

Add to the rdma_cm an IPOIB port space that allows interoperability with IPoIB multicast traffic. Use of the RDMA_PS_IPOIB is limited to multicast join/leave. Rename the RDMA_UD_QKEY to RDMA_UDP_QKEY to signify that the qkey is only used with the RDMA_PS_UDP port space. Signed-off-by: Sean Hefty

Re: [openib-general] [RFC/PATCH] librdmacm: use the ipoib broadcast group qkey

>Maybe just ask user to always call rdma_join_multicast after rdma_create_qp? >Joins are now properly reference counted, so it shouldn't be a problem >to repeat this any number of times. Right? This is the solution for now, and it should work fine. I don't think it would be hard to support creati

Re: [openib-general] [RFC/PATCH] librdmacm: use the ipoib broadcast group qkey

> Quoting Sean Hefty <[EMAIL PROTECTED]>: > Subject: Re: [RFC/PATCH] librdmacm: use the ipoib broadcast group qkey > > Or Gerlitz wrote: > > Modify librdmacm use a qkey for its UD IDs/QPs delivered to it by the > > rdma cm kernel code instead the a hard coded RDMA_UD_QKEY. For RDMA_PS_UDP > > ID,

[openib-general] Sonoma developers conference - can we move it to after OFED 1.2 release?

Hi All, I just saw in the openfabrics web page this new message: ***UPDATE*** The OpenFabrics Alliance will host the third annual Developer's Workshop at the Lodge in Sonoma - March 11-14, 2007. Click here for details.

Re: [openib-general] [RFC/PATCH] librdmacm: use the ipoib broadcast group qkey

Or Gerlitz wrote: > Modify librdmacm use a qkey for its UD IDs/QPs delivered to it by the > rdma cm kernel code instead the a hard coded RDMA_UD_QKEY. For RDMA_PS_UDP > ID, the qkey is provided by the kernel in ADDR_RESOLVED and CONNECT_REQUEST > events and is stored by the library in struct cma_id

[openib-general] OFED release testing Task force meeting minutes

Meeting took place on Thursday - Jan. 18th, 20078:30AM (PST) Agenda: 1. Review report summary (sent from Amit K.- Mellanox) and agree of common structure to be used 2. Open discussion Attending companies: Mellanox, NetEffect, Qlogic, Voltaire, SystemFabricWorks Discussion Items and Action It

Re: [openib-general] [PATCH] IB/core - ib_umad can cause address alignment fault on ia64

> Quoting Roland Dreier <[EMAIL PROTECTED]>: > Subject: Re: [openib-general] [PATCH] IB/core - ib_umad can cause address > alignment fault on ia64 > > Thanks, I queued this for 2.6.21. I think the tradeoff between risk > and return is not a good one for 2.6.20 -- I'm irrationally afraid > this i

Re: [openib-general] [libibverbs/examples] [PATCH] Added resource cleaning before end of pingpong tests + ack to CQ events

> Quoting Roland Dreier <[EMAIL PROTECTED]>: > Subject: Re: [libibverbs/examples] [PATCH] Added resource cleaning before end > of pingpong tests + ack to CQ events > > > I tried to apply the patch locally with -p1, and everything was fine > > I don't see how that could work... the patch he

Re: [openib-general] [libibverbs/examples] [PATCH] Added resource cleaning before end of pingpong tests + ack to CQ events

> Subject: Re: [libibverbs/examples] [PATCH] Added resource cleaning before end > of pingpong tests + ack to CQ events > > Thanks, applied after s/num_of_events/num_cq_events/ > > > --- gen2_devel_user.orig/src/userspace/libibverbs/examples/rc_pingpong.c > 2007-01-17 10:58:02.0 +0200

Re: [openib-general] [RFC/PATCH v3] rdma/cma: add RDMA_PS_IPOIB port space

> I was thinking of SIDR, but what about connected mode ipoib? Note IPoIB CM does not follow the SDP/CMA model for listening on multiple Service IDs at all. There's a single well known Service ID and that is all. -- MST ___ openib-general mailing list

Re: [openib-general] [PATCH] IB/ipoib: Add field dev to struct ipoib_neigh

> Quoting Moni Shoua <[EMAIL PROTECTED]>: > Subject: [PATCH] IB/ipoib: Add field dev to struct ipoib_neigh > > IPoIB uses a two layer neighboring scheme, such that for each struct neighbour > whose device is an ipoib one, there is a struct ipoib_neigh buddy which is > created on demand at the tx f

Re: [openib-general] [PATCH] IB/core - ib_umad can cause address alignment fault on ia64

Thanks, I queued this for 2.6.21. I think the tradeoff between risk and return is not a good one for 2.6.20 -- I'm irrationally afraid this is going to break the user-kernel ABI in some screwy case, and I think the impact of the problem itself is not that great (annoying kernel messages for the ti

Re: [openib-general] [RFC/PATCH v3] rdma/cma: add RDMA_PS_IPOIB port space

> I think we want to limit the ipoib port space to join/leave multicast groups. > Would there be any reason to support listen/connect/accept (i.e. SIDR) > operations? I was thinking of SIDR, but what about connected mode ipoib? This could make the ipoib port space interesting, or require break

Re: [openib-general] [libibverbs/examples] [PATCH] Added resource cleaning before end of pingpong tests + ack to CQ events

> I tried to apply the patch locally with -p1, and everything was fine I don't see how that could work... the patch headers have pathnames like gen2_devel_user.orig/src/userspace/libibverbs/examples/rc_pingpong.c in them, so stripping just one directory would leave something like src/userspa

Re: [openib-general] [RFC/PATCH v3] rdma/cma: add RDMA_PS_IPOIB port space

Or Gerlitz wrote: > Add to the RDMA CM an IPoIB port space (RDMA_PS_IPOIB) whose semantics are > similar > to those of RDMA_PS_UDP where RDMA_PS_IPOIB IDs allow for inter operability > with > IPoIB on some traffic patterns. I think we want to limit the ipoib port space to join/leave multicast gr

[openib-general] [PATCH] IB/ipoib: Add field dev to struct ipoib_neigh

IPoIB uses a two layer neighboring scheme, such that for each struct neighbour whose device is an ipoib one, there is a struct ipoib_neigh buddy which is created on demand at the tx flow by an ipoib_neigh_alloc(skb->dst->neighbour) call. When using the bonding driver, neighbours are created by the

[openib-general] Add bonding suuport to OFED

Originally, bonding is a High Availability solution for Ethernet network interfaces. It is a module that implements a virtual network device (not bounded to hardware) and enslaves "real" devices. Bonding device controls its slaves according to the bonding policy and the slave's health. I am add

Re: [openib-general] [libibverbs/examples] [PATCH] Added resource cleaning before end of pingpong tests + ack to CQ events

> Thanks, applied after s/num_of_events/num_cq_events/ > > > --- > gen2_devel_user.orig/src/userspace/libibverbs/examples/rc_pingpong.c > 2007-01-17 > 10:58:02.0 +0200 > > +++ > gen2_devel_user/src/userspace/libibverbs/examples/rc_pingpong.c > 2007-01-17 > 11:09:59.0 +020

Re: [openib-general] win related [was: Re: [PATCH 1/2] opensm: sigusr1: syslog() fixes]

Hi again Yevgeny, On Tue, 2007-01-23 at 11:20, Yevgeny Kliteynik wrote: > Hi Hal, > > Hal Rosenstock wrote: > > Hi Yevgeny, > > > > On Sun, 2007-01-21 at 04:17, Yevgeny Kliteynik wrote: > >> Sasha Khapyorsky wrote: > >>> On 18:24 Thu 11 Jan , Yevgeny Kliteynik wrote: > As for the mailin

Re: [openib-general] ib_local_sa workqueue creation crash with ofed 1.2 + sles9sp3

I guess I was all wet...the latest code works with the name "local_sa"... Sorry for the noise... Steve. On Sat, 2007-01-20 at 09:44 -0600, Steve WIse wrote: > On Sat, 2007-01-20 at 09:27 -0600, Steve WIse wrote: > > BTW: "local_sa" is still too long. I reduced it to "lsa" and it > > worked. >

Re: [openib-general] [PATCH 0/2] opensm: _ctrl layer cleanup

Hi Sasha, On Fri, 2007-01-19 at 21:54, Sasha Khapyorsky wrote: > Hi Hal, > > I found that opensm mad processing _ctrl layer is non-functional, OTOH > it costs us 11k lines of code and 17k in run-time. So there are two > patches which clean this up (for SM and SA related MADs). > > Any real objec

Re: [openib-general] [PATCH 0/6] osm: QoS policy parser

Hi Sasha, Sasha Khapyorsky wrote: > On 10:46 Sun 21 Jan , Yevgeny Kliteynik wrote: >> Hi Sasha. >> >> Sasha Khapyorsky wrote: >>> Hi Yevgeny, >>> >>> On 17:01 Wed 17 Jan , Yevgeny Kliteynik wrote: Hi Hal The following series of six patches implements QoS policy file parser:

Re: [openib-general] win related [was: Re: [PATCH 1/2] opensm: sigusr1: syslog() fixes]

Hi Hal, Hal Rosenstock wrote: > Hi Yevgeny, > > On Sun, 2007-01-21 at 04:17, Yevgeny Kliteynik wrote: >> Sasha Khapyorsky wrote: >>> On 18:24 Thu 11 Jan , Yevgeny Kliteynik wrote: As for the mailing list it's [EMAIL PROTECTED] You can access it here: http://openib.org/mailman/list

Re: [openib-general] librdmacm and udapl: Which git branch to use in ofed_1_2 build

Tziporet Koren wrote: > Sean Hefty wrote: >>>multicast >>> >> >> This goes with the multicast branch of my rdma-dev git tree. >> >> IMO, OFED should determine which features they want and pull in the >> appropriate branch. I know that Voltaire would like the multicast >> feature, but re

Re: [openib-general] [RFC/PATCH] librdmacm: modify multicast code for RDMA_PS_IPOIB port space

Sean, with one host being both the client and the server, mckey does not work for me even without the IPOIB PS changes, it used to work between two hosts with the below patch that forces the sender to generate and poll completions on its TX packets. please let me know how its going with mckey o

[openib-general] [RFC/PATCH] librdmacm: modify multicast code for RDMA_PS_IPOIB port space

Enhance the mckey test program to work in either of the port spaces. Signed-off-by: Or Gerlitz <[EMAIL PROTECTED]> Index: librdmacm/examples/mckey.c === --- librdmacm.orig/examples/mckey.c 2007-01-23 16:52:16.0 +0200 +++

[openib-general] [RFC/PATCH v2] librdmacm: add RDMA_PS_IPOIB port space

Add to librdmacm an IPoIB port space (RDMA_PS_IPOIB) whose semantics are similar to those of RDMA_PS_UDP where RDMA_PS_IPOIB IDs allow for inter operability with IPoIB on some traffic patterns. For RDMA_PS_UDP and RDMA_PS_IPOIB IDs, the qkey is provided by the kernel in ADDR_RESOLVED and CONNECT_R

[openib-general] [RFC/PATCH] rdma/cma: port rdma_cm multicast code to the UDP/IPOIB port space framework

Allow rdma_cm/ipoib multicast inter operability for RDMA_PS_IPOIB IDs. This is implemented by having the rdma cm use the --same-- qkey and multicast gid used by ipoib where for RDMA_UD_UDP IDs the rdma cm uses a qkey of its own and adds a signature byte to the multicast gid. Signed-off-by: Or G

[openib-general] [RFC/PATCH v3] rdma/cma: add RDMA_PS_IPOIB port space

Add to the RDMA CM an IPoIB port space (RDMA_PS_IPOIB) whose semantics are similar to those of RDMA_PS_UDP where RDMA_PS_IPOIB IDs allow for inter operability with IPoIB on some traffic patterns. For RDMA_PS_UDP and RDMA_PS_IPOIB IDs, the qkey is stored in struct rdma_id_private and delivered al

[openib-general] SLES9SP3 and the RDMA-CM

Has anyone tested the RDMA CM on SLES9SP3? ip_dev_find() is returning the loopback dev ptr for lookups on the local ip address of the RNIC or ipoib device. This causes problems. Anybody seen this? Thanks, Steve. ___ openib-general mailing list o

Re: [openib-general] [PATCH] IB/SRP add need_reset

Thanks. I queued this for 2.6.21, since I think it's too late for this in the 2.6.20 cycle. ___ openib-general mailing list openib-general@openib.org http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailm

[openib-general] [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.org/pub/scm/linux/kernel/git/roland/infiniband.git for-linus This will pick up three small fixes: Hoang-Nam Nguy

Re: [openib-general] [libibverbs/examples] [PATCH] Added resource cleaning before end of pingpong tests + ack to CQ events

Thanks, applied after s/num_of_events/num_cq_events/ > --- gen2_devel_user.orig/src/userspace/libibverbs/examples/rc_pingpong.c > 2007-01-17 10:58:02.0 +0200 > +++ gen2_devel_user/src/userspace/libibverbs/examples/rc_pingpong.c > 2007-01-17 11:09:59.0 +0200 BTW, can you

Re: [openib-general] How to get NFS with RDMA on OFED-1.1?

On Tue, 23 Jan 2007, Bub Thomas wrote: > I know this might end up as an RTFM but before reading the wrong manual > I'd rather ask > We are currently using OFED-1.1 on SLES-10 doing RDMA via libibverbs and > lnow ike to utilize NFS with RDMA over IB for another project. > Any help, RTFM etc.

Re: [openib-general] ofa_1_2_kernel 20070123-0346 daily build status

> Quoting [EMAIL PROTECTED] <[EMAIL PROTECTED]>: > Subject: ofa_1_2_kernel 20070123-0346 daily build status > > This email was generated automatically, please do not reply OK, mail to openib-general seems to be finally working

[openib-general] ofa_1_2_kernel 20070123-0346 daily build status

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-cxgb3-mod Passed: Passed on i686 with 2.6.15-23-server

Re: [openib-general] rdma/cma: use the ipoib broadcast group qkey - linux capabilities

Or Gerlitz wrote: >> This checks prevents applications from trying to use port numbers below 1024 >> without unless they possess the net bind service capability. A similar check >> could just be: >> >> if (ps == RDMA_PS_IPOIB && !capable(CAP_NET_BIND_SERVICE)) >> return -EACCES; > > OK, lets