From: Moni Shoua mo...@mellanox.co.il
This patch is similar in spirit to the IB/mlx4: Use RoCE IP based GIDs
in the port GID table patch for this series.
Changes to inet4 and inet6 addresses for the host are monitored and if the
address is associated with an ocrdma device then a gid is added or
From: Matan Barak mat...@mellanox.com
mlx4_ib driver should indicate that it supports
MODIFY_QP_EX user verbs extended command.
Signed-off-by: Matan Barak mat...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/hw/mlx4/main.c |3 ++-
1 files changed, 2
On 06/08/2013 14:22, Or Gerlitz wrote:
On 06/08/2013 12:38, Greg KH wrote:
On Tue, Aug 06, 2013 at 11:38:38AM +0300, Or Gerlitz wrote:
Hi Greg,
This series contains few fixes to the IB core, mlx4 IB driver and
IPoIB.
The patches enable working properly with IPoIB devices set over SRIOV
On 07/08/2013 14:01, Or Gerlitz wrote:
Hi Roland,
V4 addresses further comments made by Sean.
There are still some concerns/questions posed by Roland on the uverbs
extensions element of the series. Both Tzahi Oved and myself have
posted replies for them, but so far no further comments from you
On 12/08/2013 10:01, Ernst Gunnar Gran wrote:
I’m about to start experimenting with SR-IOV and IB (ConnectX-2 G2, HP DL360p
Gen8, E5-2609). Is it OK to just download the latest OFED release and start
playing, or are there any known issues that I should know of when it comes to
SR-IOV with the
On 09/08/2013 03:44, Jim Foraker wrote:
In several places, this snippet is used when removing neigh entries:
list_del(neigh-list);
ipoib_neigh_free(neigh);
The list_del() removes neigh from the associated struct ipoib_path, while
ipoib_neigh_free() removes neigh from the
On 13/08/2013 22:18, Roland Dreier wrote:
+struct ib_uverbs_cmd_hdr_ex {
+ __u32 command;
+ __u16 in_words;
+ __u16 out_words;
+ __u16 provider_in_words;
+ __u16 provider_out_words;
+ __u32 cmd_hdr_reserved;
+};
+
If I understand the vague explanations and
On 14/08/2013 12:06, Or Gerlitz wrote:
On 13/08/2013 22:18, Roland Dreier wrote:
+struct ib_uverbs_cmd_hdr_ex {
+ __u32 command;
+ __u16 in_words;
+ __u16 out_words;
+ __u16 provider_in_words;
+ __u16 provider_out_words;
+ __u32 cmd_hdr_reserved;
+};
+
If I
On 13/08/2013 22:18, Roland Dreier wrote:
Yes, you have tried to answer those questions separately but do you
think git writes changelogs magically when a patch is applied? If you
don't supply a good enough changelog, you're expecting me to write it
for you.
I never expected you to write our
From: Hadar Hen Zion had...@mellanox.com
Implement ib_uverbs_create_flow and ib_uverbs_destroy_flow to
support flow steering for user space applications.
Signed-off-by: Hadar Hen Zion had...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/core/uverbs.h
-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/core/verbs.c | 27 +
include/rdma/ib_verbs.h | 121 ++-
2 files changed, 146 insertions(+), 2 deletions(-)
diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c
Hi Roland,
V5 addresses further comments made by you re patch #2, hope the extension
scheme documentation is OK by now.
V5 changes:
- Addressed comments from Roland on patch #2 -- enhanced the change-log
V4 changes:
- Addressed comments from Sean on the uverbs code that copies the
and
ib_uverbs_cmd_hdr.provider_in_words that will carry the provider (mlx4_ib)
input argument size.
Signed-off-by: Igor Ivanov igor.iva...@itseez.com
Signed-off-by: Hadar Hen Zion had...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/core/uverbs_main.c | 29
On 14/08/2013 13:58, Or Gerlitz wrote:
From: Igor Ivanov igor.iva...@itseez.com
Add infra-structure to support extended uverbs capabilities in a
forward/backward
manner. Uverbs command opcodes which are based on the verbs extensions approach
should
be greater or equal
On 23/08/2013 01:36, Roland Dreier wrote:
Sorry for not replying. My feeling about these patches is basically meh.
English (please)?
They seem to be correct fixes but I can't convince myself the impact merits
applying them to stable.
Let it be for now, if it turns out that 3.10.y becomes
Hi folks,
I am attempting to implement SMB Direct (aka SMB over RDMA) for Samba.
For historical, protocol and performance reasons I believe that I need
to write a character driver that offloads RDMA stuff to the kernel.
Briefly, these reasons are:
1. Samba forks a new smbd when each
On 25/08/2013 19:19, Greg KH wrote:
On Sun, Aug 25, 2013 at 11:10:36AM +0300, Or Gerlitz wrote:
On 23/08/2013 01:36, Roland Dreier wrote:
Sorry for not replying. My feeling about these patches is basically meh.
English (please)?
That was English:
http://www.urbandictionary.com
uses ib_user_path_rec_ex
Signed-off-by: Matan Barak mat...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/core/ucma.c | 175 -
include/uapi/rdma/rdma_user_cm.h | 29 ++-
2 files changed, 198 insertions(+), 6
of parsing the dgid to
get the MAC and VLAN, take them from the address handle attributes.
Cc: Naresh Gottumukkala bgottumukk...@emulex.com
Signed-off-by: Moni Shoua mo...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/hw/ocrdma/ocrdma.h| 12
on top of that port so this should be considered when building and
maintaining the gid table.
Signed-off-by: Moni Shoua mo...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/hw/mlx4/main.c| 474 --
drivers/infiniband/hw/mlx4
drivers are modified to support the new function signature.
Signed-off-by: Matan Barak mat...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/core/addr.c | 97 ++-
drivers/infiniband/core/cm.c| 55
From: Matan Barak mat...@mellanox.com
mlx4_ib driver should indicate that it supports
MODIFY_QP_EX user verbs extended command.
Signed-off-by: Matan Barak mat...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/hw/mlx4/main.c |3 ++-
1 files changed, 2
parameters are not longer encoded
within gids, had to extend the Infiniband address structures (e.g.
ib_ah_attr) with layer 2 address parameters, namely mac and vlan.
Signed-off-by: Moni Shoua mo...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/core/cma.c | 22
, instead of
parsing the dgid to get the MAC and VLAN, take them from the
address handle attributes.
Signed-off-by: Moni Shoua mo...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/hw/mlx4/ah.c | 40 +++-
drivers/infiniband/hw/mlx4/cq.c
changes from V3:
- dropped the uverbs infrastructure patch for extensions which is now upstream
400dbc9 IB/core: Infrastructure for extensible uverbs commands
- added ocrdma patch to handle Ethernet L2 parameters, similar to the mlx4
patch.
- removed the assumption that the low
CREATE_AH_EX command that uses a new struct
ib_uverbs_create_ah_ex.
struct ib_user_path_rec was extended too, to contain source and destination
MAC and VLAN ID, this structure is of use by the rdma_ucm driver.
Signed-off-by: Matan Barak mat...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
On 11/09/2013 12:52, Yann Droneaud wrote:
Le 10.09.2013 16:41, Or Gerlitz a écrit :
+static ssize_t ucma_init_qp_attr_ex(struct ucma_file *file,
+const char __user *inbuf,
+int in_len, int out_len)
+{
+struct rdma_ucm_init_qp_attr_ex cmd;
+struct
On 11/09/2013 13:06, m...@meuh.org wrote:
@@ -672,6 +821,15 @@ struct ib_uverbs_create_ah {
struct ib_uverbs_ah_attr attr;
};
+struct ib_uverbs_create_ah_ex {
+__u32 comp_mask;
This create a hole or introduced unaligned access depending on the
64bits ABI.
Matan is OOO for couple
From: Eli Cohen e...@mellanox.com
Add missing entry in the table for UC transport.
Signed-off-by: Eli Cohen e...@mellanox.com
---
drivers/infiniband/hw/mlx5/qp.c |1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/drivers/infiniband/hw/mlx5/qp.c
From: Eli Cohen e...@mellanox.com
The layout of struct health_buffer was not according to firmware specification.
Fix it to comply.
Signed-off-by: Eli Cohen e...@mellanox.com
---
include/linux/mlx5/device.h |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git
From: Eli Cohen e...@mellanox.com
Checksum calculations consume CPU resources and can be significant to the rate
of resource creation/destruction.
Signed-off-by: Eli Cohen e...@mellanox.com
---
drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 26 ---
From: Eli Cohen e...@mellanox.com
Currently Atomic operations don't work properly. Disable them for the time
being.
Signed-off-by: Eli Cohen e...@mellanox.com
---
drivers/infiniband/hw/mlx5/main.c |5 +--
drivers/infiniband/hw/mlx5/qp.c | 48 +++-
2 files
From: Eli Cohen e...@mellanox.com
Call mlx5_ib_populate_pas() before mapping the DMA buffer to ensure the
hardware reads the values written by the CPU.
Found by: Haggai Eran hagg...@mellanox.com
Signed-off-by: Eli Cohen e...@mellanox.com
---
drivers/infiniband/hw/mlx5/mr.c |7 ---
1
From: Moshe Lazer mos...@mellanox.com
Destroying the workqueue without flushing it first can lead to a case in which
the kernel tries to push a delayed work to the workqueue which does not exist
anymore.
Signed-off-by: Moshe Lazer mos...@mellanox.com
Signed-off-by: Eli Cohen e...@mellanox.com
From: Eli Cohen e...@mellanox.com
The hardware requires that gather buffers for UMR work requests be aligned to
2K.
Signed-off-by: Eli Cohen e...@mellanox.com
---
drivers/infiniband/hw/mlx5/mr.c | 11 ---
1 files changed, 8 insertions(+), 3 deletions(-)
diff --git
From: Eli Cohen e...@mellanox.com
1. Make sure wqe_cnt does not exceed the limit published by firmware.
2. There is no requirement that the number of outstanding work requests will be
a power of two. Remove the ilog2 in the calculation of sq.max_post to fix
that.
3. Add case for
From: Eli Cohen e...@mellanox.com
On a single ported Connect-IB, its possible for the firmware to issue events on
the non-existing 2nd port. Make sure to ignore events generated for such ports.
Signed-off-by: Eli Cohen e...@mellanox.com
---
drivers/infiniband/hw/mlx5/main.c |5 +
1
From: Eli Cohen e...@mellanox.com
Change mlx5_reclaim_startup_pages() to keep polling while any pages are
returned. If none are returned, keep polling for five more seconds before
exiting with an error message.
Signed-off-by: Eli Cohen e...@mellanox.com
---
From: Eli Cohen e...@mellanox.com
Limits exceeded should be translated to ENOMEM.
Signed-off-by: Eli Cohen e...@mellanox.com
---
drivers/net/ethernet/mellanox/mlx5/core/cmd.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
From: Moshe Lazer mos...@mellanox.com
The patch fixes the rollback in case of failure in creating SRQ.
Signed-off-by: Moshe Lazer mos...@mellanox.com
Signed-off-by: Eli Cohen e...@mellanox.com
---
drivers/infiniband/hw/mlx5/srq.c |4 +++-
1 files changed, 3 insertions(+), 1 deletions(-)
From: Sagi Grimberg sa...@mellanox.com
It is normal for a device to display the pci slot name along with it's IRQs, so
the user which will check /proc/interrupts will see the pci slot of the device.
Signed-off-by: Sagi Grimberg sa...@mellanox.com
Signed-off-by: Eli Cohen e...@mellanox.com
---
From: Eli Cohen e...@mellanox.com
Change the logic so we do not allocate memory nor map the device before
actually posting to the REG_UMR QP. In addition, unmap and free the memory
after we get completion.
Signed-off-by: Eli Cohen e...@mellanox.com
---
drivers/infiniband/hw/mlx5/mr.c | 57
From: Eli Cohen e...@mellanox.com
Failed to configure opt mask to configure rre from init to rtr.
Signed-off-by: Eli Cohen e...@mellanox.com
---
drivers/infiniband/hw/mlx5/qp.c |9 +
1 files changed, 9 insertions(+), 0 deletions(-)
diff --git a/drivers/infiniband/hw/mlx5/qp.c
Hi Roland,
This series from Eli and Co contains batch of bug fixes to mlx5 issues we've
stumbled
on during various regression tests. Few other fixes which are more of cleanups
and
enhancements will be sent for-next/3.13 once this series gets into 3.12
One or two patches need to go into
From: Alex Tabachnik al...@mellanox.com
Allow to build on latest upstream kernels
Signed-off-by: Alex Tabachnik al...@mellanox.com
---
krping.c | 33 +++--
1 files changed, 19 insertions(+), 14 deletions(-)
diff --git a/krping.c b/krping.c
index 67381af..59900ab
From: Alex Tabachnik al...@mellanox.com
To allow standalone compilation with latest 3.x kernels
Signed-off-by: Alex Tabachnik al...@mellanox.com
---
Makefile | 33 +++--
1 files changed, 3 insertions(+), 30 deletions(-)
diff --git a/Makefile b/Makefile
index
Hi Steve,
This little series from Alex makes krping from
git://git.openfabrics.org/~swise/krping.git
usable for latest upstream kernels, something we find very useful for testing
the upstream RDMA stack when developing variuos features.
Or.
Alex Tabachnik (2):
Fixes and changes to support
On 11/09/2013 17:04, Yann Droneaud wrote:
[...]
+ int i;
+ int kern_attr_size;
+
+ if (out_len sizeof(resp))
+ return -ENOSPC;
+
+ if (copy_from_user(cmd, buf, sizeof(cmd)))
+ return -EFAULT;
+
+ if (cmd.comp_mask)
+
On 11/09/2013 17:32, Steve Wise wrote:
What HW did you test on?
Connect-IB HW, mlx5 driver
--
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
:19 PM, Or Gerlitz or.gerl...@gmail.com wrote:
So you are supporting UD QPs but these QPs don't generate IBTA UD
headers nor iWARP headers, correct?
--
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
On 12/09/2013 14:31, Devesh Sharma wrote:
On Thu, Sep 12, 2013 at 4:15 PM, Or Gerlitz ogerl...@mellanox.com wrote:
We've posted the kernel patches, that should be enough for the review. If you
have any specific questions re user space aspects of this series, feel free to
send them now.
Yes
On Wed, Sep 11, 2013 at 4:35 PM, Or Gerlitz ogerl...@mellanox.com wrote:
Hi Roland,
This series from Eli and Co contains batch of bug fixes to mlx5 issues we've
stumbled
on during various regression tests. Few other fixes which are more of
cleanups and
enhancements will be sent for-next
On Tue, Sep 17, 2013 at 8:50 PM, Roland Dreier wrote:
On Thu, Sep 12, 2013 at 10:22 AM, Jason Gunthorpe wrote:
On Thu, Sep 12, 2013 at 03:24:46PM +0300, Or Gerlitz wrote:
Let me clarify this. The idea is that current RoCE applications will
run as is after they update their librdmacm, since
On Wed, Sep 18, 2013 at 2:10 AM, Roland Dreier rol...@kernel.org wrote:
On Tue, Sep 17, 2013 at 1:49 PM, Or Gerlitz or.gerl...@gmail.com wrote:
To begin with, we don't break the whole stack -- using the current
patch set, for ports whose link is IB, all biz as usual
On Tue, Sep 24, 2013 at 6:13 AM, Richard Sharpe
realrichardsha...@gmail.com wrote:
On Mon, Sep 23, 2013 at 10:55 AM, Richard Sharpe wrote:
On Mon, Sep 23, 2013 at 5:02 AM, Richard Sharpe wrote:
Hi folks,
I am trying to figure out the reason for the latest hang in my SMB
Direct driver for
On Wed, Sep 11, 2013 at 4:35 PM, Or Gerlitz ogerl...@mellanox.com wrote:
This series from Eli and Co contains batch of bug fixes to mlx5 issues we've
stumbled
on during various regression tests. Few other fixes which are more of
cleanups and
enhancements will be sent for-next/3.13 once
On Wed, Oct 2, 2013 at 6:09 PM, Devesh Sharma desh...@gmail.com wrote:
One more point I have is, since current application like
perftest/qperf/rping/krping do not have code to receive ipv6 address,
do you have plans to modify these?
rping supports ipv6, as for krping, yes, it needs to be
Somnath Kotur somnath.ko...@emulex.com wrote:
empty change-log? please write something...
Signed-off-by: Somnath Kotur somnath.ko...@emulex.com
---
krping.c | 55 ---
1 files changed, 40 insertions(+), 15 deletions(-)
diff --git
On Fri, Oct 4, 2013 at 7:33 AM, Vasiliy Tolstov v.tols...@selfip.ru wrote:
Hi All.
I'm interesting of mellanox ConnectX-3 cards that support VPI.
As a read from
http://www.openfabrics.org/downloads/OFED/ofed-1.4/OFED-1.4-docs/mlx4_release_notes.txt
VPI provide ability to autodetect link type
On 04/10/2013 09:21, Vasiliy Tolstov wrote:
Where i can find docs about this or can somebody says me how can i do that?
# lspci | grep Mell
06:00.0 InfiniBand: Mellanox Technologies MT26428 [ConnectX VPI PCIe 2.0
5GT/s - IB QDR / 10GigE] (rev b0)
07:00.0 Network controller: Mellanox
Yan, 3.12-rc4 should be out today or tomorrow, if you still want to put
such volume of change into the uverbs user/kernel extensions API, you
should
post you final patch in the coming 1-2 days. This patch is too big for
-stable and we don't want to branch in user space depending if we work
On 06/10/2013 12:14, Vasiliy Tolstov wrote:
As I see this card has two ports. How about eth and ib on the same
port in the same time?
not supported, to switch kind of switch you wanted to plug that creature?
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
the body of a
On 08/10/2013 01:44, Joe Perches wrote:
Logging messages need terminating newlines to avoid
possible message interleaving. Add them.
Signed-off-by: Joe Perchesj...@perches.com
---
Acked-by: Or Gerlitz ogerl...@mellanox.com
--
To unsubscribe from this list: send the line unsubscribe linux-rdma
On 08/10/2013 14:58, Matan Barak wrote:
Regarding the last patch, you are right that it simplifies things for
creating new uverbs where command parts are in-lined one after
another, but the infrastructure got a bit more complex.
If we're going to this direction, I think the we should also
On 09/10/2013 14:29, Yann Droneaud wrote:
The other solution is reverting the current patches (22878db, 436f2ad,
400dbc9) and continue to work on this subject for 3.13: I've proposed
a slightly different scheme, but it share the same spirit. In the same
time I feel like we haven't find
On 10/10/2013 00:12, Yann Droneaud wrote:
I've suggested reverting temporarily commit 22878db, 436f2ad, 400dbc9
for v3.12 in a previous message [1]. This proposition doesn't sound
acceptable for Or.
So I continue to improve them to be in a better shape for v3.12.
Yan, we are 3-4 days before
On 22/09/2013 16:06, Matan Barak wrote:
This patch fixes the following issues:
1. Unneeded checks were removed
2. Removed the fixed size out of flow_attr.size and by thus
simplifying the checks.
3. Remove a 32bit hole on 64bit systems with strict alignment
in struct ib_kern_flow_att by
On 10/10/2013 11:32, Yann Droneaud wrote:
Le 10.10.2013 10:02, Or Gerlitz a écrit :
On 10/10/2013 00:12, Yann Droneaud wrote:
I've suggested reverting temporarily commit 22878db, 436f2ad, 400dbc9
for v3.12 in a previous message [1]. This proposition doesn't sound
acceptable for Or.
So I
On 10/10/2013 12:10, Yann Droneaud wrote:
create_flow/destroy_flow uverbs are under review and should
be considered too experimental at this point.
To not expose to userspace such experimental feature,
the data structure and code is temporarily disabled for v3.12.
The feature will be enabled
On Sun, Sep 29, 2013 at 1:48 PM, Or Gerlitz ogerl...@mellanox.com wrote:
On 17/09/2013 23:49, Or Gerlitz wrote:
On Tue, Sep 17, 2013 at 8:50 PM, Roland Dreier wrote:
On Thu, Sep 12, 2013 at 10:22 AM, Jason Gunthorpe wrote:
On Thu, Sep 12, 2013 at 03:24:46PM +0300, Or Gerlitz wrote:
Let me
On 11/10/2013 20:18, Yann Droneaud wrote:
In patch IB/core: clarify overflow/underflow checks on ib_create/destroy_flow,
the meaning of the size field was modified to only represent
the size of the flow_spec appended to the flow_attr structure.
The size of the flow_attr structure must be added
On 11/10/2013 20:18, Yann Droneaud wrote:
As requested by Or Gerlitz, I've rebased my previous fixes [1]
against Matan Barak patch [2]. I've added two patches to fix issues
left or introduced by this patch.
Regarding the previous patchset [1], I've kept the public data
structure improving
On Tue, Oct 15, 2013 at 7:44 PM, Roland Dreier rol...@purestorage.com wrote:
I agree. So as I said before, I think the best idea is just to
disable the half-baked command extension stuff before 3.12, and get
things right for 3.13.
Roland, after applying a small fix which was sent to you by
, wait_for_completion() is called only for multicast entries that
started the join process, meaning that ib_sa_join_multicast() finished.
Signed-off-by: Erez Shitrit ere...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/ulp/ipoib/ipoib.h |2
Markuze mark...@mellanox.com
Signed-off-by: Erez Shitrit ere...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/ulp/ipoib/ipoib_ib.c |5 +
1 files changed, 5 insertions(+), 0 deletions(-)
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c
b/drivers
-by: Erez Shitrit ere...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/ulp/ipoib/ipoib.h |2 +-
drivers/infiniband/ulp/ipoib/ipoib_ib.c |4 ++--
drivers/infiniband/ulp/ipoib/ipoib_main.c| 10 +-
drivers/infiniband/ulp/ipoib
.
Signed-off-by: Erez Shitrit ere...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/ulp/ipoib/ipoib_multicast.c |8
1 files changed, 8 insertions(+), 0 deletions(-)
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
b/drivers/infiniband/ulp
From: Erez Shitrit ere...@mellanox.com
The driver should not flush the whole workqueue when only one work (the
pkey poll one) needs to be cancelled, use cancel_delayed_work_sync instead.
Signed-off-by: Erez Shitrit ere...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers
Hi Roland,
Here's a set of IPoIB fixes which came through regression case failures
from the team here targeted to 3.13.
Or.
Erez Shitrit (6):
IB/ipoib: Fix crash in dev_open error flow
IB/ipoib: Fix deadlock between dev_change_flags to __ipoib_dev_flush
IB/ipoib: Avoid flushing the
-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/ulp/ipoib/ipoib_ib.c | 11 +++
1 files changed, 7 insertions(+), 4 deletions(-)
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c
b/drivers/infiniband/ulp/ipoib/ipoib_ib.c
index 196b1d1..c14f949 100644
--- a/drivers/infiniband
...@mellanox.com
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/ulp/ipoib/ipoib_cm.c | 14 +-
1 files changed, 9 insertions(+), 5 deletions(-)
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_cm.c
b/drivers/infiniband/ulp/ipoib/ipoib_cm.c
index 7a31754..1377f85
On 06/08/2013 23:32, Or Gerlitz wrote:
On Tue, Aug 6, 2013 at 8:40 PM, Roland Dreier rol...@kernel.org wrote:
On Tue, Aug 6, 2013 at 8:03 AM, Or Gerlitz or.gerl...@gmail.com wrote:
Could you finally-long-lastish post some comments on how you see this?
any bugs or flames you find here
On 06/08/2013 23:32, Or Gerlitz wrote:
On Tue, Aug 6, 2013 at 8:40 PM, Roland Dreier rol...@kernel.org wrote:
On Tue, Aug 6, 2013 at 8:03 AM, Or Gerlitz or.gerl...@gmail.com wrote:
Could you finally-long-lastish post some comments on how you see this?
any bugs or flames you find here
On 16/10/2013 03:04, Roland Dreier wrote:
On Tue, Oct 15, 2013 at 2:34 PM, Or Gerlitz or.gerl...@gmail.com wrote:
Roland, after applying a small fix which was sent to you by Matan on
Sep 22nd + another tiny fix by Yann to Matan's patch (the two can be
surely squashed into one small patch
Hi Dan,
With the latest smatch I still see these hits on the IB core, basically,
I would be happy to see the IB stack free from such warnings and then we
can easily require each new patch not to introduce them... can you shed
some light on the problem and possible solutions?
On 28/10/2013 18:32, Hefty, Sean wrote:
Visually inspecting the code for both of these, I don't see anything wrong. In
both cases the listen mutex is acquired near the top of the function and
released at the end. I don't see how you exit either function with the mutex
locked.
In the top
On 28/10/2013 23:22, Hefty, Sean wrote:
Assuming that flags will be a bitwise OR of values, they should be an int, not
an enum.
Hi Sean, anything else from your side re the IB core patches and this
series in general?
Or.
--
To unsubscribe from this list: send the line unsubscribe
On 28/10/2013 21:34, Markus Stockhausen wrote:
After some quite hard test iterations the problem seems to come from the
IPoIB switch from LRO to GRO between kernels 2.6.37 and 2.6.38.
I built a test setup with a 2.6.38 kernel and additionaly compiled a 2.6.37
ib_ipoib module against it. This
On 29/10/2013 13:10, Markus Stockhausen wrote:
Just to be on the right way: What are the basics to get GRO working
with a ConnectX (not 2 or 3) card in 2044 MTU datagram mode?
- enable GRO with ethtool.
- Activate Coalescing with ethtool? If yes how?
GRO is SW element of the network stack, so
On 29/10/2013 14:55, Erez Shitrit wrote:
In addition to what Or just wrote,
GRO currently doesn't work on ipoib interfaces, that according to bad
handling mac address that are not 6 bytes (we have plans to fix that
in the near future), that is the reason you don't see 64k packets on
tcpdump
On 29/10/2013 17:54, Markus Stockhausen wrote:
Should I exepect GRO to work on Mellanox IB cards with Linux 3.12 in general?
YES
--
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
On Tue, Oct 29, 2013 at 9:34 PM, Hefty, Sean sean.he...@intel.com wrote:
Hi Sean, anything else from your side re the IB core patches and this
series in general?
Do we have a user of these changes?
Oh sure, we didn't do that just for the fun of it... Sagi is working
to enhance the upstream
On 28/10/2013 16:26, Sagi Grimberg wrote:
This patchset Introduces Verbs level support for signature handover
feature. Siganture is intended to implement end-to-end data integrity
on a transactional basis in a completely offloaded manner.
Hi Roland, Sean
These patches are around for few
On 30/10/2013 17:20, Hefty, Sean wrote:
The team here enhanced krping to fully cover (and test...) the proposed
API and driver implementation, its (free) under
git://beany.openfabrics.org/~sgrimberg/krping.git
We'd like to see this landing in 3.13 such that the development of the
upper layers
Fix the below make W=1 build warning we have on the mlx4_ib
drivers/infiniband/hw/mlx4/qp.c: In function ‘mlx4_ib_post_send’:
drivers/infiniband/hw/mlx4/qp.c:2463: warning: comparison of unsigned
expression 0 is always false
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers
Hi Roland,
This little patch set removes the current make W=1 build warnings from the IB
core
and the mlx4_ib driver, which would help when we examine new patches to make
sure they
don't add any warnings..
Or.
Or Gerlitz (2):
IB/core: Fix build warnings
IB/mlx4: Fix build warnings
: variable ‘ret’ set but not used
Signed-off-by: Or Gerlitz ogerl...@mellanox.com
---
drivers/infiniband/core/iwcm.c |2 --
drivers/infiniband/core/sysfs.c |2 +-
drivers/infiniband/core/verbs.c |3 +--
3 files changed, 2 insertions(+), 5 deletions(-)
diff --git a/drivers/infiniband
On Thu, Oct 31, 2013 at 7:59 PM, Hefty, Sean sean.he...@intel.com wrote:
removing 'ret' is one thing, removing the call to destroy_listen() is another.
sure, I will fix and resubmit
--
To unsubscribe from this list: send the line unsubscribe linux-rdma in
the body of a message to
On Fri, Nov 1, 2013 at 10:37 PM, Bart Van Assche bvanass...@acm.org wrote:
On 31/10/2013 5:24, Sagi Grimberg wrote:
This patch implements IB_WR_REG_SIG_MR posted by the user.
Baisically this WR involvs 3 WQEs in order to prepare and properly
register the signature layout:
1. post UMR WR to
,
spotted by Sean.
Or.
Or Gerlitz (2):
IB/core: Fix build warnings
IB/mlx4: Fix build warnings
drivers/infiniband/core/iwcm.c |3 +--
drivers/infiniband/core/sysfs.c |2 +-
drivers/infiniband/core/verbs.c |3 +--
drivers/infiniband/hw/mlx4/qp.c |2 +-
4 files changed, 4
1101 - 1200 of 2157 matches
Mail list logo