On Mon, Feb 3, 2014 at 9:27 AM, Vasiliy Tolstov v.tols...@selfip.ru wrote:
2014-02-03 Or Gerlitz or.gerl...@gmail.com:
What kernel you used before? can you please send here the dmesg output when
the mlx4_core driver loads with debug_level=1
Before that i'm using 2.6.32 from debian squeeze
On Mon, Feb 3, 2014, Shachar Raindel rain...@mellanox.com wrote:
From: Upinder Malhi (umalhi) [mailto:uma...@cisco.com]
Any particular reason that the code in usNIC that walks the chunk sg's isn't
ported to the linear SG table?
As your code didn't use the API which we changed, we preferred
On 03/02/2014 21:12, Hefty, Sean wrote:
I just noted you've tagged librdmacm 1.0.18 last week, but there was no
announce email, was it just forgotten or something is stillmissing there?
I was waiting for a note from Vlad that he had pulled it into OFED, and that it
was working fine.
...@mellanox.com
Signed-off-by: Roi Dayan r...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/ulp/iser/iser_verbs.c | 10 +++---
1 files changed, 7 insertions(+), 3 deletions(-)
diff --git a/drivers/infiniband/ulp/iser/iser_verbs.c
b/drivers/infiniband/ulp/iser
.
Signed-off-by: Moni Shoua mo...@mellanox.co.il
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/hw/mlx4/main.c | 92 +++--
1 files changed, 68 insertions(+), 24 deletions(-)
diff --git a/drivers/infiniband/hw/mlx4/main.c
b/drivers
. As such, remove the
locking to take place in mlx4_ib_add() such that both cases are covered.
Signed-off-by: Moni Shoua mo...@mellanox.co.il
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/hw/mlx4/main.c |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff
Hi Roland,
This batch contain few small fixes to the mlx4 logic that deals with
managing the RoCE ports GID table management under IP addressing, which
needs to go into 3.14 along with the other two patches I sent you earlier
this week.
Or.
Moni Shoua (5):
IB/mlx4: Make sure GID index 0 is
or INET6 events
This patch makes sure that the flow of initializing the GID table is
consistent with the other two flow w.r.t on what step the lock is taken.
Signed-off-by: Moni Shoua mo...@mellanox.co.il
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/hw/mlx4/main.c |6
so will cause failure with IB core
flows that selects ports by GID.
Signed-off-by: Moni Shoua mo...@mellanox.co.il
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/hw/mlx4/main.c | 41 +---
1 files changed, 37 insertions(+), 4 deletions
flow to act accordingly.
Signed-off-by: Moni Shoua mo...@mellanox.co.il
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/hw/mlx4/main.c | 35 ++-
1 files changed, 18 insertions(+), 17 deletions(-)
diff --git a/drivers/infiniband/hw/mlx4
On 05/02/2014 15:12, Or Gerlitz wrote:
From: Moni Shouamo...@mellanox.co.il
Make sure that for Ethernet ports, the port GID table index 0 is always
occupied with a default GID of the relevant IPv6 link-local adderss.
This provides better user-experience for legacy applications that don't use
provider_out_words;
__u32 cmd_hdr_reserved;
__u32 comp_mask;
The above fields will be initialized by IBV_INIT_CMD_EX and IBV_INIT_CMD_RESP_EX
new macros.
Signed-off-by: Matan Barak mat...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
include/infiniband/kern-abi.h | 36
Hi Roland,
These two patches add flow-steering support to libibverbs
The patches are against your tree + the IP addressing UD patches
sent earlier this week.
Or.
Matan Barak (2):
Add general definitions to support uverbs extensions
Add receive flow steering support
for Ethernet link type.
Signed-off-by: Hadar Hen Zion had...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
Signed-off-by: Matan Barak mat...@mellanox.com
---
include/infiniband/driver.h |4 +
include/infiniband/kern-abi.h | 99 +++-
include/infiniband
From: Matan Barak mat...@mellanox.com
The implementation uses the default commands from libibverbs.
Signed-off-by: Matan Barak mat...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
src/mlx4.c |2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/src/mlx4.c
On Tue, Feb 4, 2014 Devesh Sharma devesh.sha...@emulex.com wrote:
From: Selvin Xavier selvin.xav...@emulex.com
Issue mailbox command to get the statistics counters from the skyhawk
hardware
Reading from the HW the RC transport stats/errors is OK, but there's no
justification to display them
...@mellanox.co.il
Signed-off-by: Matan Barak mat...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
V0 -- V1 changes:
- make sure ocrdma advertizes the new bit too
drivers/infiniband/hw/mlx4/main.c |2 +-
drivers/infiniband/hw/ocrdma/ocrdma_verbs.c |2 +-
include/rdma
On 08/02/2014 06:26, Devesh Sharma wrote:
We are having some 200 Odds vendor specific stats to be displayed. With sysfs
infrastructure for IB devices [..] We had a discussion with Roland on this, and
he gave us a go-ahead on this as well.
Roland, what's wrong with either plugging into the
Hi Roland,
This series adds support for Ethernet L2 address resolution for UD QPs, whose
L2 address-handles, unlike RC/UC/XRC/etc QPs are set from user space without
going through uverbs and the kernel IB core.
Matan and Or.
changes from V0:
- don't change struct ibv_port_attr flags field from
...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
include/infiniband/verbs.h | 22 ++
1 files changed, 22 insertions(+), 0 deletions(-)
diff --git a/include/infiniband/verbs.h b/include/infiniband/verbs.h
index 5064636..a79a1de 100644
--- a/include/infiniband
layer's type in mlx4_context.
Caching will allow us to avoid ibv_query_port calls and save time
in ibv_create_ah.
Signed-off-by: Matan Barak mat...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
src/mlx4.c |4 +++
src/mlx4.h |9 +++
src/verbs.c | 73
.
Signed-off-by: Matan Barak mat...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
include/infiniband/verbs.h | 102
src/device.c |2 +
src/verbs.c|7 ++-
3 files changed, 109 insertions(+), 2 deletions
Hi Roland and Yishai,
This series adds support for Ethernet L2 address resolution for UD QPs, whose
L2 address-handles, unlike RC/UC/XRC/etc QPs are set from user space without
going through uverbs and the kernel IB core.
Matan and Or
- better checking for invalid port numbers
- create_ah_ex
and sets
mlx4_ah relevant attributes.
Signed-off-by: Matan Barak mat...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
src/mlx4.c |5 ++-
src/mlx4.h |2 +
src/verbs.c | 90 +++---
3 files changed, 85 insertions
the neigh table
c. query neigh table in order to know if the neigh has shown up between
query until we started monitoring it
6. query vlan id from the interface
This solution depends on libnl v1.
Signed-off-by: Matan Barak mat...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
On 11/02/2014 15:53, Yann Droneaud wrote:
The last part is no more true.
sure, will fix that for V2, thanks for spotting this!
--
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
From: Matan Barak mat...@mellanox.com
Add an enum that describes ibv_port_cap_flags that complies
with the respective kernel enum.
This value could be fetched when using ibv_query_port in
port_cap_flags.
Signed-off-by: Matan Barak mat...@mellanox.com
Signed-off-by: Or Gerlitz ogerl
.
Signed-off-by: Matan Barak mat...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
include/infiniband/verbs.h | 102
src/device.c |2 +
src/verbs.c|7 ++-
3 files changed, 109 insertions(+), 2 deletions
Hi Roland,
This series adds support for Ethernet L2 address resolution for RoCE UD QPs,
whose L2 address-handles, unlike RC/UC/XRC/etc QPs are set from user space
without going through uverbs and the kernel IB core. The code is also
compatible both with old kernels that don't run IP based
the neigh table
c. query neigh table in order to know if the neigh has shown up between
query until we started monitoring it
6. query vlan id from the interface
This solution depends on libnl-3 with backports to libnl-1.
Signed-off-by: Matan Barak mat...@mellanox.com
Signed-off-by: Or Gerlitz
On Sun, Feb 23, 2014 at 2:19 PM, Sagi Grimberg sa...@mellanox.com wrote:
When unmapping request data, it is unsafe automatically
decrement req-nfmr regardless of it's value since IO and
reconnect flow may run concurrently resulting in req-nfmr = -1
and falsely call ib_fmr_pool_unmap.
Hi Sean,
We came across a pretty deadly situation with rdma-cm based
client/server application where the client set their RC QP to send to
HCA X on the server node but the server app opened their QP on HCA Y.
The result was un-acked RC packets and RC session failure.
This happened because
On Fri, Feb 7, 2014 at 2:04 AM, Roland Dreier rol...@kernel.org wrote:
On Thu, Feb 6, 2014 at 4:02 PM, Nicholas A. Bellinger
n...@linux-iscsi.org wrote:
Can you give us an estimate of when you'll have some time to give
feedback on the outstanding patches..?
I hope to get to it in the next
On 02/03/2014 01:50, Hefty, Sean wrote:
Such situation can happen in the following cases:
1. net.ipv4.conf.default.arp_ignore equals 0 (the default)
2. server side bonding/teaming fail-over when the Gratitous ARP sent was
lost
3. re-order of ibM net-devices mapping to HCA PCI devices after
On Mon, Mar 3, 2014 at 4:46 PM, Hefty, Sean sean.he...@intel.com wrote:
I'm not sure that this results in a single error case.
Sorry... I'm not sure to follow, can you elaborate a bit more?
We don't know what type of device responds to the ARP query. It could come
from an ethernet device.
On 04/03/2014 02:01, Nicholas A. Bellinger wrote:
This is necessary because even though IB_SEND_SIGNALED is
not set for RDMA WRITEs + READs, during a QP failure event
the work requests will be returned with exception status
from the TX completion queue.
Impossible... for rdma reads we must ask
On 03/03/2014 06:44, Mike Christie wrote:
The xmit_task callout does handle failures like EINVAL. If the above map
calls fail then you would get infinite retries. You would currently want
to do the mapping in the init_task callout instead.
If it makes it easier on the driver implementation then
On 04/03/2014 11:59, Sagi Grimberg wrote:
On 3/4/2014 11:38 AM, Or Gerlitz wrote:
On 03/03/2014 06:44, Mike Christie wrote:
The xmit_task callout does handle failures like EINVAL. If the above
map
calls fail then you would get infinite retries. You would currently
want
to do the mapping
On 04/03/2014 16:44, Sagi Grimberg wrote:
@@ -1707,10 +1707,17 @@ int iscsi_queuecommand(struct Scsi_Host
*host, struct scsi_cmnd *sc)
goto prepd_fault;
}
}
- if (session-tt-xmit_task(task)) {
-
On Mon, Mar 3, 2014 at 7:15 PM, Hefty, Sean sean.he...@intel.com wrote:
[...]
If a CM REQ actually gets to the remote side, it could be rejected as an
invalid GID, and the client could retry the request.
Retrying the request is practically calling
1. rdma_resolve_addr
2. rdma_resolve_route
3.
On 04/03/2014 14:41, Hariprasad Shenai wrote:
Hariprasad Shenai (1):
Revert cxgb4: Don't assume LSO only uses SGL path in t4_eth_xmit()
Kumar Sanghvi (5):
cxgb4: Fix some small bugs in t4_sge_init_soft() when our Page Size is
64KB
cxgb4: Add code to dump SGE registers when hitting
On 05/03/2014 17:18, Roland Dreier wrote:
On Wed, Mar 5, 2014 at 1:54 AM, Nicholas A. Bellinger
n...@linux-iscsi.org wrote:
That all said, do you have an objection wrt taking this bits through
target-pending..? Given the dependencies involved, that would seem the
most logical path to take.
On Wed, Mar 5, 2014 at 9:15 PM, Casey Leedom lee...@chelsio.com wrote:
Yes, thanks mightily for your help and advice and sorry for the size of the
updates. Hari has taken on the very difficult task of synchronizing our
out-of-kernel development branch with the in-kernel code. These two
On Tue, Mar 4, 2014 at 11:31 PM, Hefty, Sean sean.he...@intel.com wrote:
Clients should not be made aware of how the resolution was done. The RDMA CM
needs to abstract that problem. Alternate mechanisms may be usable, but
there aren't exactly a whole lot of options available. The client
patches are available in target-pending git repo (branch rdma-dif) and
passed 0-DAY testing.
Roland, I would like to hear your feedback on this.
Acked-By: Or Gerlitz ogerl...@mellanox.com
Roland, given that Sagi addressed all the feedback from the reviewers,
specifically from Mike Christie
-positives stack traces.
Signed-off-by: Yan Burman y...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/core/mad.c | 14 ++
1 files changed, 14 insertions(+), 0 deletions(-)
diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c
On 23/01/2014 12:31, Or Gerlitz wrote:
From: Sagi Grimberg sa...@mellanox.com
In case iSER uses fast registration method, it should not
request for successful completions on fast registration nor
local invalidate. We color wr_id with ISER_FRWR_LI_WRID in
order to correctly consume error
On Mon, Mar 17, 2014, Sagi Grimberg sa...@dev.mellanox.co.il wrote:
On 3/17/2014 6:59 PM, Mike Christie wrote:
This patch is ok, but happened to the patch for the xmit task path?
Hey Mike, Thanks for your Ack on this.
The xmit_task fix was posted as a separate patch since it is not specific
On Tue, Mar 11, 2014 at 2:41 PM, Or Gerlitz ogerl...@mellanox.com wrote:
From: Yan Burman y...@mellanox.com
Running over a kernel built with DMA_API_DEBUG and not checking for
DMA mapping error triggers a kernel stack trace with DMA-API: device
driver failed to check map error message. So
Hi Roland,
From time to time I get a customer case which goes through something
like the below trace which steps on a design limitation of the
upstream IB stack -- namely, if you have a process with open uverbs
reference -- device removal flow hangs and this would happen with any
device/driver,
module. It was buggy b/c
this callback also deals with UD SIDR exchange under which we wrongly
looked on the REQ member of the CM event and dereferenced a random value.
Signed-off-by: Moni Shoua mo...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
Hi Roland, we're post 3.14-rc8
On Wed, Mar 26, 2014 at 6:23 PM, Or Gerlitz ogerl...@mellanox.com wrote:
---
Hi Roland, we're post 3.14-rc8 and hence I assume will go to 3.15-rc1
and once there we will post it to -stable to it gets into 3.14.y
So we have here some warnings on unused variables, I forgot to amend
that, sorry
On Wed, Mar 26, 2014 at 7:10 PM, Yann Droneaud ydrone...@opteya.com wrote:
Is that a revert of some patch ?
Its not a full revents, but I will pick what you are probably
suggesting and mention
in the V1 change-log which commit this fixes, thanks
Just add Cc: sta...@vger.kernel.org
I know
module. It was buggy b/c
this callback also deals with UD SIDR exchange under which we wrongly
looked on the REQ member of the CM event and dereferenced a random value.
Signed-off-by: Moni Shoua mo...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
Hi Roland, we're post 3.14-rc8
On 27/03/2014 10:23, Shamir Rabinovith wrote:
So what exact capability is under the spot here? and how it's related
to a certain HCA driver and not to the kernel uverbs layer?
- End forwarded message -
ibv_open_device fail when opening the hca many times.
sorry, but you didn't provide
-by: Or Gerlitz ogerl...@mellanox.com
---
Hi Roland, we're post 3.14-rc8 and hence I assume will go to 3.15-rc1
and once there we will post it to -stable to it gets into 3.14.y
That nasty random deref wasn't causing instant crashes, but rather only
when running on lengthy QA loops, so we missed
On Sun, Mar 30, 2014 at 1:27 PM, Shamir Rabinovith
shamir.rabinovi...@oracle.com wrote:
On Thu, Mar 27, 2014 at 10:38:19AM +0200, Or Gerlitz wrote:
sorry, but you didn't provide an answer to my question - why is
ibv_open_device failing if called from N M (what's
N and what's M) processes
Update Mellanox copyrights for 2014 on the iser initiator driver.
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/ulp/iser/iscsi_iser.c |2 +-
drivers/infiniband/ulp/iser/iscsi_iser.h |2 +-
drivers/infiniband/ulp/iser/iser_initiator.c |2 +-
drivers
-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/ulp/iser/iscsi_iser.c | 45 +-
drivers/infiniband/ulp/iser/iscsi_iser.h | 10 +
drivers/infiniband/ulp/iser/iser_initiator.c | 54 --
drivers/infiniband/ulp/iser
Add an iser info print with the local/remote QP information
carried out when the connection is established. While on that,
fix little leftover from the T10 work, and set a debug print
to be carried in debug and not info level.
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers
the connection will never terminate from the iscsi stack point of view,
and we hang.
To resolve this race, do the draining of the tx cq before the loop on the rx cq.
Issue: 375623
Change-Id: I38952f738ac0e3373f6651987136bcc0fed18b7c
Signed-off-by: Roi Dayan r...@mellanox.com
Signed-off-by: Or Gerlitz
Set the driver version to be 1.3
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/ulp/iser/iscsi_iser.h |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.h
b/drivers/infiniband/ulp/iser/iscsi_iser.h
index
was queued internally behind the large T10 patch set and hence
arrives a bit late, still, taking into account it's relatively small volume,
we'd like to get it in for 3.15.
Or.
Ariel Nahum (1):
IB/iser: Remove struct iscsi_iser_conn
Or Gerlitz (3):
IB/iser: Print QP information once connection
Add an iser info print with the local/remote QP information
carried out when the connection is established. While on that,
fix little leftover from the T10 work, and set a debug print
to be carried in debug and not info level.
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers
Set the driver version to be 1.3
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/ulp/iser/iscsi_iser.h |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.h
b/drivers/infiniband/ulp/iser/iscsi_iser.h
index
the connection will never terminate from the iscsi stack point of view,
and we hang.
To resolve this race, do the draining of the tx cq before the loop on the rx cq.
Issue: 375623
Change-Id: I38952f738ac0e3373f6651987136bcc0fed18b7c
Signed-off-by: Roi Dayan r...@mellanox.com
Signed-off-by: Or Gerlitz
Update Mellanox copyrights for 2014 on the iser initiator driver.
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/ulp/iser/iscsi_iser.c |2 +-
drivers/infiniband/ulp/iser/iscsi_iser.h |2 +-
drivers/infiniband/ulp/iser/iser_initiator.c |2 +-
drivers
-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/ulp/iser/iscsi_iser.c | 45 +-
drivers/infiniband/ulp/iser/iscsi_iser.h | 10 +
drivers/infiniband/ulp/iser/iser_initiator.c | 54 --
drivers/infiniband/ulp/iser
On 01/04/2014 16:28, Or Gerlitz wrote:
Issue: 375623
Change-Id: I38952f738ac0e3373f6651987136bcc0fed18b7c
Roland, I see that in patches 1 and 2 I screwed it up and left notes
from our internal system, care to clean this up before applying or you
want me to re-send?
--
To unsubscribe from
On Tue, Apr 1, 2014 at 8:05 PM, Weiny, Ira ira.we...@intel.com wrote:
diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.h
b/drivers/infiniband/ulp/iser/iscsi_iser.h
index a4e4876..324129f 100644
--- a/drivers/infiniband/ulp/iser/iscsi_iser.h
+++ b/drivers/infiniband/ulp/iser/iscsi_iser.h
On Thu, Mar 27, 2014 at 11:52 AM, Or Gerlitz ogerl...@mellanox.com wrote:
From: Moni Shoua mo...@mellanox.com
The code that resolves the passive side source mac within the rdma_cm
connection request handler was both redundant and buggy, remove it.
Hi Roland,
I see that you've picked the iser
From: Eli Cohen e...@dev.mellanox.co.il
Add support for the block multicast loopback QP creation flag along
the proper firmware API for that.
Signed-off-by: Eli Cohen e...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/hw/mlx5/main.c |2 ++
drivers
On Wed, Jan 22, 2014 at 3:43 AM, Roland Dreier rol...@kernel.org wrote:
[...]
I'd really rather spend my time on something actually useful like cleaning up
softroce.
Hi Roland,
Were you referring to the code that was posted here few years ago, or
you're working on something new? two comments
From: Matan Barak mat...@mellanox.com
When using libnl3, libibverbs should be linked with
libnl3 and not the utilities.
Signed-off-by: Matan Barak mat...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
Makefile.am | 28 ++--
1 files changed, 14
On Mon, Apr 21, 2014 at 9:09 PM, Christoph Lameter c...@linux.com wrote:
We need this option for many hosts to avoid backflow of multicast
packages. Could we get that merged?
[...]
Currently, there's no way for user-space applications to specify
the IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK QP
On Tue, Apr 22, 2014 at 12:06 AM, Christoph Lameter c...@linux.com wrote:
On Mon, 21 Apr 2014, Or Gerlitz wrote:
Currently, there's no way for user-space applications to specify
the IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK QP creation flags
defined by commit 47ee1b9 IB/core: Add support
On 22/04/2014 16:48, Christoph Lameter wrote:
Note that I am pretty unhappy about Mellanox right now. 8 year old patches
that have never been refreshed? If I submit patchsets then I will post
these things regularly until they are accepted. You need to have a patch
queue or a git tree with up to
On 23/04/2014 16:44, Hefty, Sean wrote:
Regarding SubnetTimeout changes: the code in
drivers/infiniband/core/cache.c already queues a work request after each
port state change. Inside that work request e.g. the P_Key cache is
updated. Would it be acceptable to modify ib_cache_update() such that
On 02/03/2014 12:49, Haggai Eran wrote:
The following set of patches implements on-demand paging (ODP) support
in the RDMA stack and in the mlx5_ib Infiniband driver.
I've placed the latest cut of the ODP patches on my public git tree @
git://beany.openfabrics.org/~ogerlitz/linux-2.6.git odp
On Wed, Apr 30, 2014 at 10:16 PM, Shirley Ma shirley...@oracle.com wrote:
[...]
3. Upstream NFSoRDMA status:
So does it currently works...? I understand that Yan tried it out
today, and @ least one side just crashed.
Chuck, I assume there is a configuration which basically works for you
and
On Tue, Apr 29, 2014 at 8:32 PM, Christoph Lameter c...@linux.com wrote:
Ok I saw the ODP patches. But where are the multicast block patches? Is
there any way to move this forward?
We can try, earlier on this thread I wrote
Roland, how you prefer to move on here? we can come up with quick
On Wed, Apr 30, 2014 at 10:47 PM, Chuck Lever chuck.le...@oracle.com
If I understood Yan, he is trying to use NFS/RDMA in guests (kvm?). We
are pretty sure that is not working at the moment,
can you provide a short 1-2 liner why/what is broken there? the only
thing which I can think of to be
On 01/05/2014 02:28, Doug Ledford wrote:
This seems like driving a thumb tack with a sledge hammer. No doubt
the correct path here is to get the proper flag added to verbs multicast
attach command. Or, do you still have a patch that does this the
right way? Did the one from 2008 do it the
On Mon, May 5, 2014 at 10:19 PM, Roland Dreier rol...@kernel.org wrote:
libibverbs is a library that allows programs to use RDMA verbs for
direct access to RDMA (currently InfiniBand and iWARP) hardware from
userspace.
The new stable release, 1.1.8, is available from
Hi Roland, as you've
On 05/05/2014 22:19, Roland Dreier wrote:
Matan Barak (1):
Add general definitions to support uverbs extensions
Or Gerlitz (1):
Add receive flow steering support
Roland, due to a mistake I made in the rebased flow-steering patch I
handed you -- I was nominated as the author
From: Matan Barak mat...@mellanox.com
Add an enum that describes ibv_port_cap_flags that complies
with the respective kernel enum.
This value could be fetched when using ibv_query_port in
port_cap_flags.
Signed-off-by: Matan Barak mat...@mellanox.com
Signed-off-by: Or Gerlitz ogerl
Hi Roland,
This series adds support for Ethernet L2 address resolution for RoCE UD QPs,
whose L2 address-handles, unlike RC/UC/XRC/etc QPs are set from user space
without going through uverbs and the kernel IB core. The code is also
compatible both with old kernels that don't run IP based
.
Signed-off-by: Matan Barak mat...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
include/infiniband/verbs.h | 105 +++-
src/device.c |2 +
src/verbs.c|7 ++-
3 files changed, 111 insertions(+), 3 deletions
the neigh table
c. query neigh table in order to know if the neigh has shown up between
query until we started monitoring it
6. query vlan id from the interface
This solution depends on libnl-3 with backports to libnl-1.
Signed-off-by: Matan Barak mat...@mellanox.com
Signed-off-by: Or Gerlitz
Hi Yishai,
This is just a rebased-port of the series which adds support for Ethernet L2
address
resolution for UD QPs, whose L2 address-handles, unlike RC/UC/XRC/etc QPs are
set
from user space without going through uverbs and the kernel IB core.
Matan and Or.
changes from V1:
- rebased
and sets
mlx4_ah relevant attributes.
Signed-off-by: Matan Barak mat...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
src/mlx4.c |4 ++-
src/mlx4.h |2 +
src/verbs.c | 90 +++---
3 files changed, 84 insertions
layer's type in mlx4_context.
Caching will allow us to avoid ibv_query_port calls and save time
in ibv_create_ah.
Signed-off-by: Matan Barak mat...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
src/mlx4.c |4 +++
src/mlx4.h |9 +++
src/verbs.c | 73
On Thu, May 8, 2014 at 10:29 PM, Jason Gunthorpe
jguntho...@obsidianresearch.com wrote:
On Thu, May 08, 2014 at 09:51:22AM +0300, Or Gerlitz wrote:
In order to implement RoCE IP based addressing for UD QPs, without
introducing uverbs changes, we need a way to resolve the L2 Ethernet
On Thu, May 8, 2014 at 11:15 PM, Jason Gunthorpe
jguntho...@obsidianresearch.com wrote:
On Thu, May 08, 2014 at 11:05:41PM +0300, Or Gerlitz wrote:
Thanks for looking commenting, Matan and myself will digest your
feedback and respond/fix what's needed. BTW - note that this is V3
where V0/1/2
verbs
IB_QP_CREATE_USE_GFP_NOIO QP creation flag is provided.
Signed-off-by: Mel Gorman mgor...@suse.de
Signed-off-by: Jiri Kosina jkos...@suse.cz
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/hw/mlx4/cq.c|6 ++--
drivers/infiniband/hw/mlx4/mlx4_ib.h
the new flag in the ipoib connected mode path, and if the driver
doesn't support that, re-issue the QP creation w.o this specification.
Signed-off-by: Mel Gorman mgor...@suse.de
Signed-off-by: Jiri Kosina jkos...@suse.cz
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/ulp/ipoib
Fix the usnic and thw qib driver to err when QP creation flags are provided.
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/hw/qib/qib_qp.c |3 ++-
drivers/infiniband/hw/usnic/usnic_ib_verbs.c |3 +++
2 files changed, 5 insertions(+), 1 deletions
GFP_NOIO for the QP creation and fallback to GFP_KERNEL
(as before) if the HW driver doesn't support that. This approach will let
Jiri Kosina (1):
mlx4: Enhance the QP creation path to use a given GFP directive
Or Gerlitz (2):
IB: Return error when QP creation are provided for driver
On 09/05/2014 21:10, Jason Gunthorpe wrote:
On Fri, May 09, 2014 at 07:01:36AM -0700, Roland Dreier wrote:
On Thu, May 8, 2014 at 12:09 PM, Jason Gunthorpe
jguntho...@obsidianresearch.com wrote:
struct verbs_context {
/* grows up - new fields go here */
+ int
-by: Matan Barak mat...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
Makefile.am|6 ++-
include/infiniband/verbs.h | 29 +++
man/ibv_create_flow.3 | 86
src/device.c |4 --
4 files
1301 - 1400 of 2157 matches
Mail list logo