Re: ib_srpt and mellanox mthca device

Hi Roland! I made tracing for srpt_add_one function. Trace log attached to mail Roland Dreier писал 2012-04-11 02:09: On Tue, Apr 10, 2012 at 7:52 AM, Alexey Shvetsov wrote: I'm  trying to use ib_srpt from mainline kernel (3.3.x and 3.4-rcX) But right after loading ib_srpt i always get error

[PATCH] Timeout on VLARB and SL2VL SET operation MADs should cause heavy-sweep

This change re-adds send error handling for VLArbitration and SL2VLMappingTable SM attributes which was last removed by commit 2a424a7e4fab93aad4781c3abad39be3015fee56 No additional option is believed to be needed as send errors are timeouts and cancellations from the OpenSM vendor layer and not

RE: [PATCH 02/11] ipath: replace open-coded ARRAY_SIZE with macro

> From: Jim Cromie [mailto:jim.cro...@gmail.com] > Subject: [PATCH 02/11] ipath: replace open-coded ARRAY_SIZE with macro > > > Signed-off-by: Jim Cromie I noticed a similar issue in ipath_iba6110.c around line 599. Do you want me to correct or do you want to catch it with a single patch? Mik

Re: [PATCH 02/11] ipath: replace open-coded ARRAY_SIZE with macro

On Wed, Apr 11, 2012 at 11:13 AM, Marciniszyn, Mike wrote: >> From: Jim Cromie [mailto:jim.cro...@gmail.com] >> Subject: [PATCH 02/11] ipath: replace open-coded ARRAY_SIZE with macro >> >> >> Signed-off-by: Jim Cromie > > I noticed a similar issue in ipath_iba6110.c around line 599. > > Do you wa

[RFC] [PATCH 4/4] rsocket: Allow use of LD_PRELOAD to intercept socket calls

Intercept socket calls and convert TCP socket operation to streaming over RDMA. Allow falling back from rsockets to normal sockets on error or when trying to bind/connect to a reserved port. This is needed to handle MPI job startup, where MPI should use rsockets, but mpiexect needs to communicate

[RFC] [PATCH 2/4] rsocket: Add example program that uses rsocket

rstream provides an example that uses either rsocket or socket APIs. The latter allows rstream to be used to verify rsocket behavior compared to socket. Signed-off-by: Sean Hefty --- Makefile.am|5 examples/rstream.c | 570 man/

[RFC] [PATCH 0/4] librdmacm: Rsockets API and implementation

The following patch set contains an initial implementation of rsockets as presented at the 2012 OpenFabrics Workshop. A copy of that presentation is available at: https://www.openfabrics.org/downloads/rdmacm/rsockets-ofa12.pptx and a video of the presentation can be found here:

[RFC] [PATCH 3/4] rsocket: Add sample application to copy files over rsockets

rcopy will copy files from a source system to a specified remote server. It's essentially a really dumb FTP type program that can be used to quickly transfer files between systems, which can be useful to verify data integrity. (It was easier to create this program than modify an existing FTP clie

Re: ib_srpt and mellanox mthca device

On Wed, Apr 11, 2012 at 2:21 AM, Alexey Shvetsov wrote: > I made tracing for srpt_add_one function. Trace log attached to mail It looks like the call to ib_create_srq() is failing. What specific HCA type do you have? Do you have up-to-date firmware? I thought all mthca hardware is SRQ-capable,

[GIT PULL] please pull infiniband.git

Hi Linus, Please pull from git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband.git tags/rdma-for-linus Fix regression in the /sys/class/infiniband/.../rate attribute -- old kernels used to just return something, ev

Re: ib_srpt and mellanox mthca device

Hi Roland! Actual IB HBA is store ~ # lspci -v -s 4b:00.0 4b:00.0 InfiniBand: Mellanox Technologies MT25208 InfiniHost III Ex (Tavor compatibility mode) (rev 20) Subsystem: Mellanox Technologies MT25208 InfiniHost III Ex (Tavor compatibility mode) Physical Slot: 2-1 Fl

Re: [RFC] [PATCH 0/4] librdmacm: Rsockets API and implementation

On 04/11/12 18:29, Hefty, Sean wrote: > The following patch set contains an initial implementation of rsockets as > presented at the 2012 OpenFabrics Workshop. A copy of that presentation > is available at: > > https://www.openfabrics.org/downloads/rdmacm/rsockets-ofa12.pptx > > and a vid

[PATCH] IB: mlx4: Fix mem leaks in ib_link_query_port()

If, in ib_link_query_port(), the call to mlx4_MAD_IFC() fails we will currently do 'return err;' which will leak 'in_mad' and 'out_mad' that we previously allocated. I believe we should instead do 'goto out;' where we'll properly free the memory we previously allocated. Signed-off-by: Jesper Juhl

RE: [RFC] [PATCH 0/4] librdmacm: Rsockets API and implementation

> Interesting work. Regarding direct data placement: have the io_submit() > and io_getevents() system calls been considered ? Those are the > foundation of libaio. I did look at aio and other interfaces, but I didn't try to move beyond the idea phase. With aio, I didn't see how the calls could e

Re: ib_srpt and mellanox mthca device

Oh, I may see. Can you try the following? (maybe whitespace damaged, I'm pasting into a web form, but should be easy to apply by hand) drivers/infiniband/ulp/srpt/ib_srpt.c |1 + 1 file changed, 1 insertion(+) diff --git a/drivers/infiniband/ulp/srpt/ib_srpt.c b/drivers/infiniband/ulp/srpt

Re: ib_srpt and mellanox mthca device

Hi Roland! I tryed this patch but it doesnt work i still get ib_mthca: Initializing :4b:00.0 ib_mthca :4b:00.0: irq 49 for MSI/MSI-X ib_mthca :4b:00.0: irq 50 for MSI/MSI-X ib_mthca :4b:00.0: irq 51 for MSI/MSI-X srpt_add_one(mthca0) failed. New trace attached Roland Dreier писа

Re: ib_srpt and mellanox mthca device

> I tryed this patch but it doesnt work i still get > ib_mthca: Initializing :4b:00.0 > ib_mthca :4b:00.0: irq 49 for MSI/MSI-X > ib_mthca :4b:00.0: irq 50 for MSI/MSI-X > ib_mthca :4b:00.0: irq 51 for MSI/MSI-X > srpt_add_one(mthca0) failed. Hmm, in srpt_add_one(), where it says

Re: ib_srpt and mellanox mthca device

Hi Roland! Seems like this card doesnt support srq store linux # ibv_srq_pingpong Couldn't get local LID PS on nodes with mlx4 devices ibv_srq_pingpong works also with yout change it doesnt report anything new Loading kernel module for a network device with CAP_SYS_MODULE (deprecated). Use CA