Re: [ANNOUNCE] libnes-1.0.0 release

Tung, Chien Tin wrote: New release of libnes library (1.0.0) is available at: http://www.openfabrics.org/downloads/nes/libnes-1.0.0.tar.gz sha1sum: e9d73b46cbc6c000f9904874fb1aba874b6ab8cf libnes-1.0.0.tar.gz Changes since last release: Chien Tung (2): libnes: sync up with function pro

Re: [PATCH] opensm/libvendor//osm_vendor_ibumad_sa.c: uninitialized fields

Hi Sasha, On 09/Dec/09 02:14, Yevgeny Kliteynik wrote: Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-Path: klit...@dev.mellanox.co.il X-OriginalArrivalTime: 09 Dec 2009 00:05:42.0455 (UTC) FILETIME=[590C3C70:01CA7863] No idea where did this header came fr

[PATCH] opensm/libvendor//osm_vendor_ibumad_sa.c: uninitialized fields

Taking care of couple of uninitialized fields. Discovered with 'osmtest -f a' - bad things start to happen when a packet is dropped, and osmtest should really get a timeout. Signed-off-by: Yevgeny Kliteynik --- opensm/libvendor/osm_vendor_ibumad_sa.c |2 ++ 1 files changed, 2 insertions(+),

Re: Sun V880 + Infiniband?

From: Patrick Finnegan Date: Thu, 3 Dec 2009 01:44:18 -0500 > On Wednesday 02 December 2009, David Miller wrote: >> If there's something Linux isn't doing right, those dumps will help >> me spot it. > > Thanks again! Please give this patch a try: sparc64: Fix overly strict range type matching

End user documentation for Infiniband?

Hi guys, Sorry if this is a bit of a newbie question, or posted to the wrong place, but I've been struggling to find any documentation for configuring infiniband, and the openfabrics mailing lists appear well and truly dead. I have a very basic network running now, with three hosts, ipoib and ope

[PATCH] mlx4_core: Cleanup bug in __mlx4_init_one()

If mlx4_init_port_info() fails, cleanup the initialized ports only. Signed-off-by: Eli Cohen --- drivers/net/mlx4/main.c |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c index 6f5a3cf..0c868c9 100644 --- a/drivers/net/mlx

Re: [Patch] Init ipoib_neigh.dgid

thanks, applied at last -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: [PATCH] ipath: Use bitmap_weight

thanks, applied. -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: [PATCH 9/9] RDMA/nes: Update copyright and branding string

thanks applied all 9. Nice work, I'm glad to see continuing cleanups in there. -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: [PATCH resend] ib/ehca: rework destroy_eq()

thanks, applied -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: RDMAoE verbs questions

> It looks good to me. Thanks, I will take it for RDMAoE. Great... as Jason suggested, please also add in the appropriate reserved fields to pad the struct to a 32 bit boundary and zero them in the wrapper. So if there is a next time we don't have this problem again. - R. -- To unsubscribe fr

Re: Sun V880 + Infiniband?

On Wednesday 09 December 2009, David Miller wrote: > Please give this patch a try: > > sparc64: Fix overly strict range type matching for PCI devices. This appears to work for me. Thanks! Pat -- Purdue University Research Computing --- http://www.rcac.purdue.edu/ The Computer Refuge

[PATCH 01/11] RDMA/nes: change MAX_CM_BUFFER define

Changed MAX_CM_BUFFER for MPA frame to be conformant to RFC 5044. This will change it to 512 + 20 instead of 512. Signed-off-by: Faisal Latif diff --git a/drivers/infiniband/hw/nes/nes_cm.h b/drivers/infiniband/hw/nes/nes_cm.h index 90e8e4d..2d867a6 100644 --- a/drivers/infiniband/hw/nes/nes

[PATCH 02/11] RDMA/nes: fix query of ORD values

The ORD size needs updating as we are supporting more inbound READ resources per connection. Signed-off-by: Faisal Latif diff --git a/drivers/infiniband/hw/nes/nes_verbs.c b/drivers/infiniband/hw/nes/nes_verbs.c index 7a77450..0e3eee2 100644 --- a/drivers/infiniband/hw/nes/nes_verbs.c +++ b

Re: [PATCH 02/11] RDMA/nes: fix query of ORD values

I only got patches 1 and 2 here (and patchwork also seems to only have those two)... am I missing 9 patches, or is the subject line wrong? - R. -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at ht

Re: RDMAoE verbs questions

On Wed, Dec 09, 2009 at 11:06:41AM -0800, Roland Dreier wrote: > > > It looks good to me. Thanks, I will take it for RDMAoE. > > Great... as Jason suggested, please also add in the appropriate reserved > fields to pad the struct to a 32 bit boundary and zero them in the > wrapper. So if there i

Re: [PATCH] IB/core: clarify the documentation of ib_post_send()

thanks, applied. -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: IB/documentation: fix grammar in ipoib.txt

thanks, applied -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: RDMAoE verbs questions

On Wed, Dec 09, 2009 at 02:48:43PM -0700, Jason Gunthorpe wrote: > > Also, if we really must do this, can you please send a patch to Roland > that at least adds the constants for IB, ASAP. Ideally to be included > in OFED 1.5. > > At least that way people can make updates to check it prior to RDM

Re: [PATCH] IB/uverbs: PTR_ERR return of wrong pointer in ib_uverbs_get_context()

Well spotted, applied thanks. -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: [PATCH 3/5] drivers/infiniband: correct size computation

applied, thanks everyone. -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: [ewg] [PATCH] mlx4: Fix bug in mlx4_ib_mcg_attach

This bug doesn't seem to ever have been present in the upstream kernel -- what are you generating this patch against? - R. -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/

Re: [PATCH] mlx4_core: Cleanup bug in __mlx4_init_one()

Looks like a valid fix but your patch doesn't apply. I don't seem to have mlx4_cleanup_counters_table() in my tree. -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordom

[PATCH 01/11] RDMA/nes: change MAX_CM_BUFFER define

Changed MAX_CM_BUFFER for MPA frame to be conformant to RFC 5044. This will change it to 512 + 20 instead of 512. Signed-off-by: Faisal Latif --- drivers/infiniband/hw/nes/nes_cm.h |4 +++- 1 files changed, 3 insertions(+), 1 deletions(-) diff --git a/drivers/infiniband/hw/nes/nes_cm.h b/d

[PATCH 02/11] RDMA/nes: fix query of ORD values

The ORD size needs updating as we are supporting more inbound READ resources per connection. Signed-off-by: Faisal Latif --- drivers/infiniband/hw/nes/nes_verbs.c |8 1 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/infiniband/hw/nes/nes_verbs.c b/drivers/infi

[PATCH 03/11] RDMA/nes: MPA request/response error checking

During Xansation testing, error handling of MPA frame msg/response is not handled properly. Signed-off-by: Faisal Latif --- drivers/infiniband/hw/nes/nes_cm.c | 31 +-- 1 files changed, 29 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/hw/nes/nes_cm.

Re: [ewg] [PATCH] IPoIB: synchronize timer deletion with completion handler

> When calling del_timer_sync on priv->poll_timer, it is necessary to prevent > farther arming of the timer which can be done by a completion handler. Though > it is harmless since the timer will eventually stop being rearmed, it is > better > practice to avoid calling the timer function aft

[PATCH 04/11] RDMA/nes: Resource not freed for REJECT connection

During testing of REJECT connection error handling, the cm_id resources are not release. When retransmit timer expires, need to send reset message to remote node before sending ABORTED event. Signed-off-by: Faisal Latif --- drivers/infiniband/hw/nes/nes_cm.c |2 ++ 1 files changed, 2 inserti

Re: [ewg] Re: [PATCH] mlx4_core: Cleanup bug in __mlx4_init_one()

On Wed, Dec 09, 2009 at 02:35:18PM -0800, Roland Dreier wrote: > Looks like a valid fix but your patch doesn't apply. I don't seem to > have mlx4_cleanup_counters_table() in my tree. I noticed that already. I'll send a new patch in my next working day. -- To unsubscribe from this list: send the l

[PATCH 05/11] RDMA/nes: crash in nes_accept

While running IMP_EXT's window test, the node crashed in nes_accept(). Here is the sequence of what happened. (1) In MVAPICH2, connect request is received for port #0. FIX:For this, nes_connect() check to make sure local or remote tcp port is not 0. (2) Remote node's (passive) tcp sta

[PATCH 06/11] RDMA/nes: abnormal listener termination causes loopback node crash

when listener is destroyed for loopback connection, listener node is getting reset event. This causes crash as listener is not expecting reset event. Code review of cm_event_reset() durring debugging showed the cm_id ref count in incrementing after calling its event handler and not before. Signed-

[PATCH 07/11] RDMA/nes: Xansation test crash on cm_node ref_count

Running Xansation test, active side node crashed. The problem started from passive side. It generated stag which was 0. The passive side sent TERMINATE instead of MPA REJECT msg. On the active side, it receives TERMINATE and sends connect_err() and set the cm_node state to CLOSED. The passive side

[PATCH 08/11] RDMA/nes: check for zero stag

STags are generated randomly but the driver does not correctly prevent a zero STag. Using STag zero is privileged and causes a user space application to fail. This change prevents the driver from trying to allocate a zero STag. Signed-off-by: Faisal Latif --- drivers/infiniband/hw/nes/nes_hw.c

[PATCH 09/11] RDMA/nes: free kmap() resources

We fail in creating many qps as kmap() fails for sq_vbase. Now we do kunmap() as soon as we are done with sq_vbase. We do kunmap() in one of the locations below. (1) nes_destroy_qp() (2) nes_accept() (3) nes_connect_event We set a flag, to not cause multiple calls to kunmap(). Signed-off-by: Faisa

[PATCH 10/11] RDMA/nes: FIN during MPA startup causes timeout

A FIN that is received during an MPA start up sequence causes a timeout in iwcm.c. The connection has not been completely closed so the iwcm code is waiting for resources to be cleaned up. This closes the connection so everything cleans up correctly. Signed-off-by: Don Wood Signed-off-by: Faisa

[PATCH 11/11] RDMA/nes: stale arp issue fix

When the remote node's mac addr changes, the connection keeps trying to connect using old mac address. The connection wil continue failing until the driver is unloaded and loaded again (eiter reboot or rmmod). Now it is checked before starting a connection if the mac address of the remote node is s

Re: [PATCH v2 0/51] updated patches for adding QIB driver

have not had a chance to look at the new submission yet but I think you need at least the following rolled in, to avoid calling __exit code that might have been thrown away from the error path of your init function. Building with CONFIG_DEBUG_SECTION_MISMATCH=y when you see that the kernel build t

Re: [ewg] [PATCH] mlx4: Fix bug in mlx4_ib_mcg_attach

On Wed, Dec 09, 2009 at 02:33:31PM -0800, Roland Dreier wrote: > This bug doesn't seem to ever have been present in the upstream > kernel -- what are you generating this patch against? > I think it came from your for-next branch. -- To unsubscribe from this list: send the line "unsubscribe linux-

Re: [ewg] [PATCH] mlx4: Fix bug in mlx4_ib_mcg_attach

> > This bug doesn't seem to ever have been present in the upstream > > kernel -- what are you generating this patch against? > I think it came from your for-next branch. I don't see anything touching this code there. The patch that introduced this code upstream, 521e575b ("IB/mlx4: Add supp

Re: [patch 1/6] RDMA/amso1100 - ib_post_send fix

Somehow you managed to line-wrap the patch, and format it so patchwork didn't catch it (maybe the "patch" in the subject needs to be "PATCH"?). But applied anyway... -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More

Re: [PATCH v2 0/51] updated patches for adding QIB driver

OK. Done. On Wed, 2009-12-09 at 14:47 -0800, Roland Dreier wrote: > have not had a chance to look at the new submission yet but I think you > need at least the following rolled in, to avoid calling __exit code that > might have been thrown away from the error path of your init function. > > Build

Re: [patch 2/6] RDMA/ehca - ib_post_send fixes, v2

thanks, applied... same complaints about formatting plus: > +while(send_wr) { checkpatch.pl warns about the missing space after "while" here. -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at

[infiniband-diags] [PATCH] [0/2] Support libibnetdiscover caching

Hi Sasha, I've added two new functions into libibnetdisc, ibnd_cache_fabric() and ibnd_load_fabric(). As the names suggest, these functions allow the user to store the ibnd_fabric_t fabric to disk and read it back later on. The patches add the functions and the base functionality to ibnetdiscove

[infiniband-diags] [PATCH] [1/2] add libibnetdisc caching to libibnetdiscover

Hi Sahsa, This patch adds caching functionality to libibnetdisc through the new functions ibnd_cache_fabric() and ibnd_load_fabric(). Al -- Albert Chu ch...@llnl.gov Computer Scientist High Performance Systems Division Lawrence Livermore National Laboratory From: Albert Chu Date: Wed, 9 Dec 20

[infiniband-diags] [PATCH] [2/2] support --cache and --load-cache options in ibnetdiscover

Hi Sasha, This adds the --cache and --load-cache options to ibnetdiscover. Al -- Albert Chu ch...@llnl.gov Computer Scientist High Performance Systems Division Lawrence Livermore National Laboratory From: Albert Chu Date: Wed, 9 Dec 2009 15:19:47 -0800 Subject: [PATCH] support --cache and --loa

Re: [PATCH 11/11] RDMA/nes: stale arp issue fix

thanks, applied all 11 -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: [PATCH v3 23/27] misc: use helpers for rlimits

Not sure it matters at this part, but for the drivers/infiniband part, you can add: Acked-by: Roland Dreier wrapping all the current->signal->rlim[] stuff in a helper function looks like a nice cleanup independent of the writable limit stuff. - R. -- To unsubscribe from this list: send the lin

Re: [PATCH v2 12/51] IB/qib: Add qib_dma.c

> +struct ib_dma_mapping_ops qib_dma_mapping_ops = { > +qib_mapping_error, > +qib_dma_map_single, > +qib_dma_unmap_single, > +qib_dma_map_page, > +qib_dma_unmap_page, > +qib_map_sg, > +qib_unmap_sg, > +qib_sg_dma_address, > +qib_sg_dma_len, > +qib

Re: Sun V880 + Infiniband?

From: Patrick Finnegan Date: Wed, 9 Dec 2009 14:29:56 -0500 > On Wednesday 09 December 2009, David Miller wrote: >> Please give this patch a try: >> >> sparc64: Fix overly strict range type matching for PCI devices. > > This appears to work for me. > > Thanks! Thanks for your report and testi

Re: [PATCH 06/11] RDMA/nes: abnormal listener termination causes loopback node crash

Faisal Latif wrote: when listener is destroyed for loopback connection Does the upstream iwarp stack supports loopback connections? does it apply to all vendors? Or. -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org M