On Thu, Jun 11, 2015 at 8:55 PM, Jason Gunthorpe
wrote:
> On Thu, Jun 11, 2015 at 04:35:20PM +0300, Or Gerlitz wrote:
>> From: Matan Barak
>>
>> Add a new ib_cq_init_attr structure which contains the
>> previous cqe (minimum number of CQ entries) and comp_vecto
On Thu, Jun 11, 2015 at 8:53 PM, Jason Gunthorpe
wrote:
> On Thu, Jun 11, 2015 at 04:35:19PM +0300, Or Gerlitz wrote:
>> Jason had another comment calling for re-thinking / questioning the
>> need for a dedicated uverbs_ex_cmd_mask in the IB device. This goes
>> beyond the s
On Thu, Jun 11, 2015 at 1:01 AM, Jason Gunthorpe
wrote:
> On Wed, Jun 10, 2015 at 11:19:03PM +0300, Matan Barak wrote:
>
>> > Sure gid_type is gone, but I didn't say roceve2 specific, I said
>> > latent elements. ie I'm assuming reasons for the scary locking are
>> > because the ripped out rocev2
On Sat, Jun 13, 2015 at 8:35 AM, Doug Ledford wrote:
> I ran across a problem today when I went to do some run tests of my
> for-4.2 tree. For a second there, I was about to seriously have a
> conniption fit. But, after about 6 hours of work bisecting and
> debugging, I've come to find that I wa
On Sat, Jun 13, 2015 at 3:50 PM, Doug Ledford wrote:
> On 06/13/2015 03:18 AM, Or Gerlitz wrote:
>> On Sat, Jun 13, 2015 at 8:35 AM, Doug Ledford wrote:
>>> I ran across a problem today when I went to do some run tests of my
>>> for-4.2 tree. For a second there, I wa
utilize this capability.
Signed-off-by: Moshe Lazer
Signed-off-by: Or Gerlitz
---
This is a rpost of the code we sent earlier on January,
I understand that people's focus nowadays is around
the few threads that deal with kernel changes, but
let's get it in to sit in the patchw
Signed-off-by: Or Gerlitz
---
src/qp.c | 11 ++-
src/wqe.h |2 ++
2 files changed, 12 insertions(+), 1 deletions(-)
diff --git a/src/qp.c b/src/qp.c
index 721bed4..5b800e7 100644
--- a/src/qp.c
+++ b/src/qp.c
@@ -286,6 +286,10 @@ int mlx4_post_send(struct ibv_qp *ibqp, struct
On Sat, Jun 13, 2015 at 8:35 AM, Doug Ledford wrote:
> I ran across a problem today when I went to do some run tests of my
> for-4.2 tree. For a second there, I was about to seriously have a
> conniption fit. But, after about 6 hours of work bisecting and
> debugging, I've come to find that I wa
On Fri, Jun 12, 2015 at 2:09 AM, Mike Marciniszyn
wrote:
> +++ b/drivers/infiniband/hw/hfi1/device.c
> +int __init dev_init(void)
> +{
> + int ret;
> +
> + ret = alloc_chrdev_region(&hfi1_dev, 0, HFI1_NMINORS, DRIVER_NAME);
> + if (ret < 0) {
> + pr_err("Could not
On Fri, Jun 12, 2015 at 2:11 AM, Mike Marciniszyn
wrote:
> +++ b/drivers/infiniband/hw/hfi1/user_pages.c
> +/**
> + * hfi1_get_user_pages - lock user pages into memory
> + * @start_page: the start page
> + * @num_pages: the number of pages
> + * @p: the output page structures
> + *
> + * This fu
On Mon, Jun 15, 2015 at 4:29 PM, Mike Marciniszyn
wrote:
58K LOC with a single non-empty change log line saying "This patch
series adds the OPA gen1 driver" is surely a nice candidate for all
times record of how not to submit upstream drivers. Why not sparing
few words for the reviewers what is O
On Mon, Jun 15, 2015 at 10:14 PM, Or Gerlitz wrote:
> On Mon, Jun 15, 2015 at 4:29 PM, Mike Marciniszyn
> wrote:
>
> 58K LOC with a single non-empty change log line saying "This patch
> series adds the OPA gen1 driver" is surely a nice candidate for all
> times
On Wed, Jun 17, 2015 at 2:48 PM, Marciniszyn, Mike
wrote:
>> what's SC and what's SC-to-VL? aren't you using SM to do such mappings? if
>> this
>> and the below sysfs work-around/s for the maybe current lack of SM to
>> configure the port? why do the upstream kernel has to carry ~hundreds of
>>
On Fri, Jun 19, 2015 at 3:57 AM, Doug Ledford wrote:
>> On Jun 14, 2015, at 10:31 AM, Or Gerlitz wrote:
>> The fix is easy, disable Granular VF QoS in that VPI config, I tested
>> it and sent that now to net [1]
> Did this fix the mlx4_en shutdown issue too, or is there
On Wed, Jun 17, 2015, Mike Marciniszyn wrote:
> This patch series adds the OPA gen1 driver.
> IB/hfi1: add qp handling
> IB/hfi1: add RC QP handling
> IB/hfi1: add routines for RC/UC
> IB/hfi1: add UC QP handling
> IB/hfi1: add UD QP handling
Mike,
This is the 3rd
On Tue, May 19, 2015 at 12:48 PM, Sagi Grimberg wrote:
> +++ b/drivers/infiniband/core/cma.c
> +const char *rdma_event_msg(enum rdma_cm_event_type event)
> +{
> + size_t index = event;
> +
> + return (index < ARRAY_SIZE(cma_events) && cma_events[index]) ?
> + cm
On Wed, Jun 24, 2015 at 3:59 PM, Matan Barak wrote:
[...]
> Changes from V5:
> (1) Incoporate the changes to cache.c so we use the same infrastructure
> to manage both IB and RoCE (per Doug's request)
> (2) Replace the locking mechanism in the IB core GID cache from seqcount +
> rcu to rwl
On 6/25/2015 4:50 AM, ira.we...@intel.com wrote:
From: Ira Weiny
commit 97f229a8515f932e5adc6cdfa18cc1440235b9fd
IB/mad: Add support for additional MAD info to/from drivers
added BUG_ON's which were inappropriate for a condition which should never
happen. Change these to be WARN_ON_ONCE as a d
On Thu, Jun 25, 2015 at 4:52 AM, Weiny, Ira wrote:
> Linus,
>
>>
>> On the *other* side of the same conflict, I find an even more offensive
>> commit,
>> namely commit 4cd7c9479aff ("IB/mad: Add support for additional MAD info
>> to/from drivers") which adds a BUG_ON() for a sanity check, rather
For IB links, reading HCA flow counters through iboe_process_mad() should
be used when mlx4_ib_process_mad() is invoked only for VFs PMA queries and
exactly nothing else.
Fixes: 7193a141eb74 ('IB/mlx4: Set VF to read from QP counters')
Reported-by: Linus Torvalds
Signed-off-by:
On 6/25/2015 4:52 PM, ira.we...@intel.com wrote:
From: Ira Weiny
We recently added BUG_ON's which were inappropriate for a condition which
should never happen. Change these to be WARN_ON_ONCE as a debugging aid.
Fixes: 4cd7c9479aff ('IB/mad: Add support for additional MAD info to/from
drivers'
On Thu, Jun 25, 2015 at 5:45 PM, Or Gerlitz wrote:
> For IB links, reading HCA flow counters through iboe_process_mad() should
> be used when mlx4_ib_process_mad() is invoked only for VFs PMA queries and
> exactly nothing else.
>
> Fixes: 7193a141eb74 ('IB/mlx4: Set VF to r
On Thu, Jun 25, 2015 at 9:13 PM, Jason Gunthorpe
wrote:
> On Thu, Jun 25, 2015 at 11:34:43AM +0300, Or Gerlitz wrote:
>
>> So... are we finally OK wrt the feedback you provided?
>
> I've been looking at Yishai's series, I though it was almost good to
> go, but
On 6/30/2015 12:36 AM, Steve Wise wrote:
The semantics for MR access flags are not consistent across RDMA
protocols. So rather than have applications try and glean what they
need, have them pass in the intended roles and attributes for the MR to
be allocated and let the RDMA core select the appr
On 6/28/2015 1:33 PM, Or Gerlitz wrote:
On Thu, Jun 25, 2015 at 5:45 PM, Or Gerlitz wrote:
For IB links, reading HCA flow counters through iboe_process_mad() should
be used when mlx4_ib_process_mad() is invoked only for VFs PMA queries and
exactly nothing else.
Fixes: 7193a141eb74 ('IB
On 6/30/2015 7:42 PM, Jason Gunthorpe wrote:
NFSRDMA currently checks the transport type to decide how to set the
>access flags for memory registration. With the new services
>exported in this series, we can change/simplify NFSRDMA to not have
>to know the transport type.
It would be excellent
On 6/30/2015 9:42 PM, Steve Wise wrote:
I prefer to decouple the iSER changes with this core work.
Jason/Sean... thoughts? I could do the iSER w/o patch 3, and the
follow up with a series that includes our final solution on
transport independent memory registration and change all the TI
kernel u
On 6/25/2015 6:23 PM, Weiny, Ira wrote:
On 6/25/2015 4:52 PM, ira.we...@intel.com wrote:
From: Ira Weiny
We recently added BUG_ON's which were inappropriate for a condition
which should never happen. Change these to be WARN_ON_ONCE as a
debugging aid.
Fixes: 4cd7c9479aff ('IB/mad: Add support
On 6/25/2015 8:28 PM, Bart Van Assche wrote:
On 06/25/2015 09:04 AM, ira.we...@intel.com wrote:
From: Ira Weiny
The define OPA_LID_PERMISSIVE is big endian and was compared to cpu
value
opa_drslid.
0-day build caught this while building with the OPA (hfi1) driver
which was
recently sent t
Hi Doug,
The mlx4 bug (VF query_device fails) was introduced with the time-stamping
changes,
the ipoib bug is something we stepped on while having the mlx4 issue, now fixed.
Or.
Amir Vadai (1):
IB/IPoIB: Fix bad error flow in ipoib_add_port()
Matan Barak (1):
IB/mlx4: Do not attemp to re
Fixes: 4b664c4355b2 ('IB/mlx4: Add support for CQ time-stamping')
Signed-off-by: Matan Barak
Signed-off-by: Or Gerlitz
---
drivers/infiniband/hw/mlx4/main.c | 11 ++-
1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/drivers/infiniband/hw/mlx4/main.c
b/drivers/inf
From: Amir Vadai
Error values of ib_query_port() and ib_query_device() weren't propagated
correctly. Because of that, ipoib_add_port() could return NULL value,
which escaped the IS_ERR() check in ipoib_add_one() and we crashed.
Signed-off-by: Amir Vadai
Signed-off-by: Or Gerlitz
---
dr
On Wed, Jul 1, 2015 at 7:30 PM, Steve Wise wrote:
please no empty change-logs for this driver (and elsewhere), find
something that fills @ least one sentence, it's easy.
Also, we are aiming to initiate (Sagi please make sure you follow on
this practice too) commit titles for the iser initiator w
On Wed, Jul 1, 2015 at 7:30 PM, Steve Wise wrote:
> Memory regions that are the target of an iWARP RDMA READ RESPONSE need
> REMOTE_WRITE access rights. So enable REMOTE_WRITE for iWARP devices.
I don't see the point to mess the code with this branching on the
iwarp specific diffs from IB/RoCE -
On Wed, Jul 1, 2015 at 11:53 PM, Steve Wise wrote:
>> From: Or Gerlitz [mailto:gerlitz...@gmail.com]
> Yes, the MR is a local MR, but it is used for REMOTE access for iWARP, but
> not IB. It think the reason is that in iWARP there is no distinction between
> local and remote k
On 7/7/2015 4:59 PM, Steve Wise wrote:
> >>>diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.c
b/drivers/infiniband/ulp/iser/iscsi_iser.c
> >>>index 6a594aa..de8730d 100644
> >>>--- a/drivers/infiniband/ulp/iser/iscsi_iser.c
> >>>+++ b/drivers/infiniband/ulp/iser/iscsi_iser.c
> >>>@@ -640,6 +
On 7/8/2015 12:42 PM, Jack Wang wrote:
We're using MLX OFED 2.4-1.0.4 together on top of 3.18.14.
So this list is for upstream things.. still, let's see
We hit bug below spontaneously, our test trigger this bug around 1 in 5 times.
and what is your test if I may ask?!
HCA 'mlx4_0'
CA t
On 7/8/2015 3:47 PM, Jack Wang wrote:
static void mlx4_ib_cq_comp(struct mlx4_cq *cq)
47 {
48 struct ib_cq *ibcq = &to_mibcq(cq)->ibcq;
49 ibcq->comp_handler(ibcq, ibcq->cq_context);
50 }
Looks like cq use-after-free? I have no idea where.
see if you have in the code base you're using (why not
On Wed, Jul 8, 2015 at 5:07 PM, Jack Wang wrote:
> Looks the last one is missing in OFED 2.4 driver, I just checked the
> history of mainline
>
> commit bf1bac5b7882daa41249f85fbc97828f0597de5c
> Author: Eli Cohen
> Date: Thu Oct 23 15:57:27 2014 +0300
>
> net/mlx4_core: Call synchronize_irq(
On 7/9/2015 1:30 AM, Doug Ledford wrote:
In mlx4_main.c:do_slave_init(), the function is supposed to queue up
each work struct. However, it checks to make sure the sriov support
isn't going down first. When it is going down, it doesn't queue up the
work struct, which results in us leaking the w
On Thu, Jul 9, 2015 at 1:36 AM, Doug Ledford wrote:
> Thanks, both applied.
to where?! Doug, I didn't see any update on your gitbub nor your
kernel.org tree, and again, it's a bit annoying to chase two trees w.o
clear branching policy... can you use something with clear name (say)
4.2-fixes on yo
On 7/9/2015 2:14 PM, Jack Wang wrote:
I managed to update the kernel to OFED 3.0 to verify the bug, but I
can still produce the bug, maybe there're still some synchronice_irq
is missing?
Again, even if you don't use the upstream kernel for production, I
suggest you
try to reproduce the bug the
On 7/9/2015 4:35 PM, Jack Wang wrote:
We have other kernel modules together also the autotest
infrastructure. It's not that easy to install a 3.18.14 kernel.
you said you are running on 3.18.14 and just replaced their stock RDMA
stack with MLNX OFED
I look into the code a little bit. I thi
On 7/9/2015 3:31 PM, Doug Ledford wrote:
i Doug,
>
>I'd like Jack to review this before we ack, not sure if he's in today,
>so he might get to look on that only on Sunday.
Try to get it reviewed before then please. If it passes my
build/functional tests (which I want to get to today), it will g
On Fri, Jul 10, 2015 at 10:15 PM, andrew banman wrote:
> I'm seeing a large number of allocation errors originating from the Mellanox
> IB
> driver when booting the 4.2-rc1 kernel on a 4096cpu 32TB memory system:
Just to make sure, mlx4 works fine on this small (...) system with 4.1
and 4.2-rc1
On Tue, Jul 14, 2015 at 9:48 PM, Alex Thorlton wrote:
> On Tue, Jul 14, 2015 at 01:22:34PM -0500, andrew banman wrote:
>> On Sat, Jul 11, 2015 at 11:20:19PM +0300, Or Gerlitz wrote:
>> > On Fri, Jul 10, 2015 at 10:15 PM, andrew banman wrote:
>> > > I'm seeing
On 7/14/2015 11:28 PM, Alex Thorlton wrote:
We see the same exact messages on 4.1-rc8.
does this solves the problem?
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index ad31e47..c8ae3b9 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -
On Sun, Jul 19, 2015 at 7:07 PM, Sagi Grimberg wrote:
> On 7/16/2015 6:25 PM, Bart Van Assche wrote:
> I agree it would definitely help as the lack of immediate data
> emphasizes the additional latency of doing rdma reads.
Sagi, do we have black box evidence from iSER showing notable
(results? s
On 7/20/2015 12:44 PM, Sagi Grimberg wrote:
On 7/20/2015 12:43 AM, Or Gerlitz wrote:
On Sun, Jul 19, 2015 at 7:07 PM, Sagi Grimberg
wrote:
On 7/16/2015 6:25 PM, Bart Van Assche wrote:
I agree it would definitely help as the lack of immediate data
emphasizes the additional latency of doing
On Thu, Jul 30, 2015 at 11:06 AM, Sagi Grimberg wrote:
> We don't update those anywhere in the code and they
> seem pretty useless (no one seem to care about those).
>
> qp_tx_queue_full: We never should get this
why?
> fmr_map_not_avail: We can never get to this
why? if for some reason fmr poo
On Thu, Jul 30, 2015 at 11:06 AM, Sagi Grimberg wrote:
> iser support up to 512KB data transfer in a single scsi
> command. In order to support up to 8MB, iser needs to pre-allocate
> larger memory regions and larger page vectors.
We should be doing things for a reason, and we are following that
On Thu, Jul 30, 2015 at 11:06 AM, Sagi Grimberg wrote:
> If iser_initialize_task_headers() routine failed before
> dma mapping, we should not attempt to unmap in cleanup_task().
This fixes some specific commit? if yes, add Fixes: line, if not, is
that from day-0?
--
To unsubscribe from this list:
On Thu, Jul 30, 2015 at 11:06 AM, Sagi Grimberg wrote:
> Concatination of send work requests benefits performance
> by reducing the send queue lock contention (acquired in
> ib_post_send) and saves us HW doorbells which is posted
> only once.
s/Concatination/Concatenation/
AFAIK, do we today! i
On Thu, Jul 30, 2015 at 11:06 AM, Sagi Grimberg wrote:
> In the past the we always tried to allocate an fmr_pool
> and if it failed on ENOSYS (not supported) then we continued
> with dma mr. This is not the case anymore and if we tried to
> allocate an fmr_pool then it is supported and we expect t
On 7/30/2015 3:09 PM, Sagi Grimberg wrote:
I'll add the Fixes tag.
don't forget to use --abbrev=12 for the Fixes: tag
--
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/maj
On Thu, Jul 30, 2015 at 3:23 PM, Sagi Grimberg wrote:
> Today, iser is not usable with no FRWR and no FMR support. (it once was
> when we bounced to higher-order allocations but we don't do that
> anymore). Memory registration is a requirement support for iser today.
OK, sure, we now have FRWR su
Hi Doug,
Some fixes included, none of them accounts for regression introduced in 4.2-rc1,
so all can go to 4.3 -- genetated them again 4.2-rc4
Or.
Jack Morgenstein (3):
IB/mlx4: Fix potential deadlock when sending mad to wire
IB/mlx4: Deprecate mcast group warning message to debug because of
From: Jack Morgenstein
send_mad_to_wire takes the same spinlock that is taken in
the interrupt context. Therefore, it needs irqsave/restore.
Fixes: b9c5d6a64358 ('IB/mlx4: Add multicast group (MCG) paravirtualization for
SR-IOV')
Signed-off-by: Jack Morgenstein
Signed-off-by:
From: Jack Morgenstein
The mcg "too many pending requests" warning message fills the log
when OpenSM is downed. Deprecate the warning to be debug output.
Signed-off-by: Jack Morgenstein
Signed-off-by: Or Gerlitz
---
drivers/infiniband/hw/mlx4/mcg.c | 8 ++--
1 file changed, 6
: c1e7e466120b ('IB/mlx4: Add iov directory in sysfs under the ib device')
Signed-off-by: Jack Morgenstein
Signed-off-by: Or Gerlitz
---
drivers/infiniband/hw/mlx4/sysfs.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/infiniband/hw/mlx4/sysfs.c
b/drivers/infiniba
Signed-off-by: Or Gerlitz
---
drivers/infiniband/hw/mlx4/ah.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/drivers/infiniband/hw/mlx4/ah.c b/drivers/infiniband/hw/mlx4/ah.c
index f50a546..33fdd50 100644
--- a/drivers/infiniband/hw/mlx4/ah.c
+++ b/drivers/infiniband/hw
On Thu, Jul 30, 2015 at 10:17 PM, Mike Marciniszyn
wrote:
> The following patch series adds the OPA device driver.
[...]
> IB/hfi1: add qp handling
> IB/hfi1: add RC QP handling
> IB/hfi1: add UC QP handling
> IB/hfi1: add UD QP handling
On Wed, Jun 17, 2015, Mike Marcini
On Thu, Jul 30, 2015 at 6:33 PM, Matan Barak wrote:
[...]
> Changes from V6:
> (1) Addressed Jason's comments:
> (a) Cache is no longer a client but part of IB infrastructure
> (b) No need for READ_ONCE and flush_workqueue when tearing down
> the cache
Doug
So... ar
On Fri, Jul 31, 2015 at 8:41 PM, Doug Ledford wrote:
> On 07/31/2015 12:32 PM, Jason Gunthorpe wrote:
>> On Fri, Jul 31, 2015 at 08:50:24AM -0400, Doug Ledford wrote:
So... are we ready to go with V7 upstream?
>>> Yes. I've already pulled it in.
>> You are taking the netdev stuff without an
On Sat, Aug 1, 2015 at 11:34 PM, Doug Ledford wrote:
> On 07/31/2015 03:34 AM, Christoph Hellwig wrote:
>> On Fri, Jul 31, 2015 at 02:05:06AM +0300, Or Gerlitz wrote:
>>> So... enough is enough, please put it in a kernel module residing in
>>> the IB core and use it in
On Sat, Aug 1, 2015 at 1:01 AM, Jason Gunthorpe
wrote:
> On Sat, Aug 01, 2015 at 12:24:23AM +0300, Or Gerlitz wrote:
>
>> addressed in incremental patch, as Doug suggested. Jason, it's wrong
>> to send developers again and again to fix things which were
>> perfe
On Tue, Aug 4, 2015 at 8:10 PM, Sagi Grimberg wrote:
>
Why SIZE_4K and not PAGE_SIZE?
>>>
>>>
>>> Yes, I'll change that to PAGE_SIZE.
>>>
>>> Thanks.
>>
>>
>> Would non-4KB pages (e.g. PPC 64KB) be an issue? Would this work between
>> hosts with different page sizes?
>
>
> iser was always usi
On Wed, Aug 5, 2015 at 8:16 PM, Jason Gunthorpe
wrote:
> On Wed, Aug 05, 2015 at 06:34:26PM +0300, Amir Vadai wrote:
>> struct ib_uverbs_ex_query_device {
>> __u32 comp_mask;
>> + __u32 csum_caps;
>> __u32 reserved;
>> };
>
> Uh no.
>
>> @@ -221,6 +222,7 @@ struct ib_uverbs_odp_c
On Wed, Aug 5, 2015 at 7:17 PM, Hefty, Sean wrote:
> TPC and UDP should be separate flags.
Sean,
I don't think we should over-complex things vs. what the network stack
does for many (since kernel 2.4?!) years. They have basically three
flags
NETIF_F_IP_CSUM - device can checksum TCP/UDP over I
On Thu, Aug 6, 2015 at 3:00 AM, Jason Gunthorpe
wrote:
[...]
> The participating we are sorely lacking right now is on the review
> side, which is like most of the kernel, unfortunately.
I agree, if a proper internal review was taking place here, it
wouldn't been sent this way. But you know,
it
On Thu, Aug 6, 2015 at 2:23 AM, Hefty, Sean wrote:
>> I don't think we should over-complex things vs. what the network stack
>> does for many (since kernel 2.4?!) years. They have basically three
>> flags
>>
>> NETIF_F_IP_CSUM - device can checksum TCP/UDP over IPv4
>> NETIF_F_IP6_CSUM - device c
On Fri, Aug 7, 2015 at 7:43 PM, Marciniszyn, Mike
wrote:
>> Could you get some experienced engineers to look at the driver internally to
>> Intel before publishing? There are numerous other drivers in the kernel by
>> Intel that do the right thing.
> I will try to get a review from internal engi
On Thu, Jun 18, 2015 at 7:38 PM, Jason Gunthorpe
wrote:
> On Sun, Jun 14, 2015 at 01:13:04PM +0300, Or Gerlitz wrote:
>> From: Moshe Lazer
>>
>> Add a device capability flag IB_DEVICE_IP_SUM to denote checksum offload
>> support. Devices should set this flag if they s
On Fri, Aug 28, 2015 at 10:27 PM, Doug Ledford wrote:
> I'm seeing this with rc8 on a dual port mlx4 adapter set to IB/Eth mode:
mmm, both Amir and myself are just finishing vacations... so WB notes
are not always lovely as you want them to be, life
>
> [ 77.883513] IPv6: ADDRCONF(NETDEV_UP):
On Fri, Aug 28, 2015 at 9:02 PM, Doug Ledford wrote:
> On 08/28/2015 09:10 PM, ira.weiny wrote:
>> On Tue, Aug 18, 2015 at 12:22:10PM +0300, Dan Carpenter wrote:
>>> Smatch says that, based on the indenting, we should probably add curly
>>> braces here.
>>>
>>> Fixes: 230145ff8124 ('IB/core: Add R
On Thu, Aug 27, 2015 at 5:55 AM, Haggai Eran wrote:
> When no matching listening ID is found for a given request, the net_dev
> that was used to find the request isn't released.
>
> Fixes: 20c36836ecad ("IB/cma: Use found net_dev for passive connections")
same here, Doug, if this ID isn't the one
On Mon, Aug 31, 2015 at 6:24 PM, Sagi Grimberg wrote:
> Since patch series "Demux IB CM requests in the rdma_cm module" the
> P_Key index is taken from the work completion rather than the message itself
so prior to this series nobody in the IB core (and maybe across the
whole upstream kernel) use
On Mon, Aug 31, 2015 at 4:02 PM, Doug Ledford wrote:
> On 08/31/2015 03:09 AM, Matan Barak wrote:
>> Actually, it looks like the dump stack we've got before [1] was fixed.
>> This happens when the mlx4 driver is used in setups where number of
>> cores >= 32.
>> Doug, is that the case?
> Indeed,
On Tue, Sep 1, 2015 at 1:13 AM, Doug Ledford wrote:
> On 08/31/2015 04:21 PM, Or Gerlitz wrote:
>> On Mon, Aug 31, 2015 at 4:02 PM, Doug Ledford wrote:
>>> On 08/31/2015 03:09 AM, Matan Barak wrote:
>>
>>>> Actually, it looks like the dump stack we've go
On 9/1/2015 11:08 AM, Sagi Grimberg wrote:
(see http://www.spinics.net/lists/netdev/msg335599.html).
better to have pointer here to upstream commit and not to an archive
URL which is possibly gonna die some day
This commit is not upstream yet.
I know, but but under the way the net/ne
On Tue, Sep 1, 2015 at 11:42 AM, Matan Barak wrote:
> If I recall, this code was sent through linux-net mailing list. So
> it's possible the different branches weren't rebased, isn't it?
but the code was merged for 4.2-rc1 -- so Doug, this means that when
you did the 4.2-rc work you haven't reba
On 9/2/2015 5:32 PM, Christoph Lameter wrote:
On Wed, 5 Aug 2015, Amir Vadai wrote:
This will be used by a revised version of the IP checksum patches [1], that
will be sent later on.
Ok when can we get the full set for testing? Seems that the libibverbs and
libmlx4 portions are missing?
not
On Sat, Sep 5, 2015 at 2:41 AM, Doug Ledford wrote:
> On 08/17/2015 11:50 AM, Bodong Wang wrote:
>> Add a device capability flag IBV_DEVICE_IP_CSUM to denote IPv4 checksum
>> offload support. Devices should set this flag if they support
>> insertion/verification of IPv4, TCP and UDP checksums on
>
On Sat, Sep 5, 2015 at 2:43 AM, Doug Ledford wrote:
> Series applied, thanks.
Doug, a user-space related note, in LPC you made a comment that the
user space RoCE UD patches are broken with libnl3, well, talking to
Matan, he said this is indeed the case with the RHEL provided libnl3
but not with t
On 9/9/2015 11:09 AM, Matan Barak wrote:
On 9/9/2015 9:41 AM, Jiri Pirko wrote:
Wed, Sep 09, 2015 at 05:33:28AM CEST,
torvalds@linux-foundation.orgwrote:
On Tue, Sep 8, 2015 at 8:08 PM, Doug Ledford
wrote:
With a comment that said "I can carry this merge forward, no further
action is necessa
On 9/10/2015 2:04 AM, Nicholas Krause wrote:
This fixes a incorrect assumation that ib_get_cached_pkey always runs
successfully in the function build_mlx_header by checking if the calls
to this particular function return the error code, -EINVAL in order to
signal they failed to grap the public ke
On 9/9/2015 6:20 PM, Doug Ledford wrote:
On 09/05/2015 03:59 PM, Or Gerlitz wrote:
--> they must not put any
additional bits on the wire
Maybe. For base level interop, sure, but for enhanced service in a homogeneous
environment, not necessarily true.
--> RC isn't an option
No
On 9/5/2015 2:43 AM, Doug Ledford wrote:
On 09/03/2015 10:56 AM, Haggai Eran wrote:
This series adds userspace support for on-demand paging. The first patch adds
support for the new extended query device verb. Patch 2 adds the capability and
interface bits related to on-demand paging, and patch
On 9/10/2015 7:53 PM, Doug Ledford wrote:
>I don't see it in your kernel.org tree [1]
Because I didn't push it yet. I can push it now
Push complete.
So we should keep chasing every commit you took and have humans look
for it among bunch of trees/branches, it's so tired-ing and annoyin
On 8/7/2015 4:00 PM, Matan Barak wrote:
This purpose of this series is to add usage of the GID cache to
the CMA and IB stack. Instead of passing Ethernet L2 attributes
via QP attributes, we could just use the GID cache that's already
points to a ndev and thus to all required L2 attributes.
[...
On Wed, Sep 16, 2015 at 7:31 PM, Christoph Lameter wrote:
> Another approach may be to tie the unsub from sendonly multicast joins to
> the expiration of the layer 2 addresses in IPoIB. F.e. add code to
> __ipoib_reap_ah() to detect if the handle was used for a sendonly
> multicast join. If so un
On Wed, Sep 16, 2015 at 11:17 PM, Christoph Lameter wrote:
> On Wed, 16 Sep 2015, Or Gerlitz wrote:
>
>> On Wed, Sep 16, 2015 at 7:31 PM, Christoph Lameter wrote:
>> > Another approach may be to tie the unsub from sendonly multicast joins to
>> > the expiration of
notification
mechanism initiated from __ipoib_reap_neigh().
Signed-off-by: Erez Shitrit
Signed-off-by: Or Gerlitz
---
drivers/infiniband/ulp/ipoib/ipoib.h | 6 ++
drivers/infiniband/ulp/ipoib/ipoib_multicast.c | 87 ++
2 files changed, 93 insertions(+)
diff --git
Patches from Erez, to be used for cleaning up send-only objects and multicast
group SM registrations.
Or.
Erez Shitrit (2):
IB/ipoib: Add mechanism for ipoib neigh state change notifications
IB/ipoib: Add cleanup to sendonly multicast objects
drivers/infiniband/ulp/ipoib/ipoib.h
the change in the neigh state.
The callback should pay attention to the context it runs on, and should
act/run according to that context limitation, for example on the neigh
reap flow, the neigh calls the callback under spinlock etc.
Signed-off-by: Erez Shitrit
Signed-off-by: Or Gerlitz
On 9/17/2015 3:48 AM, Christoph Lameter wrote:
On Wed, 16 Sep 2015, Or Gerlitz wrote:
Could you please post here a few (say 2-4) liner summary of what is
still missing or done wrong in 4.3-rc1 and what is your suggestion how
to resolve that.
With Doug's patch here the only thing that is
On Thu, Sep 17, 2015 at 1:38 PM, Or Gerlitz wrote:
> From: Erez Shitrit
>
> Sendonly multicast group entries are potentially created by the driver during
> the xmit flow. Their objects remain in the driver memory, plus the related
> group
> existing in the SM and the fabric ti
On Thu, Sep 17, 2015 at 5:42 PM, Christoph Lameter wrote:
> Could we simplify it a bit. This compiles but avoids all the
> generalizations and workqueues.
Do you find some over complexity in Erez's implementation? what? as I
said, he's pretty busy, but I hope he can get to review your proposal
On Thu, Sep 17, 2015 at 5:42 PM, Christoph Lameter wrote:
> Could we simplify it a bit. [...] but avoids all the
> generalizations and workqueues. Had to export two new functions from
> ipoib_multicast.c though.
Do you find some over complexity in Erez's implementation? what? as I
said, he's pret
On Mon, Sep 21, 2015 at 11:03 PM, Or Gerlitz wrote:
> On Mon, Sep 21, 2015 at 7:41 PM, Sagi Grimberg wrote:
>> Changes from v0:
>> - Replace xprtrdma patch to Chuck's one
>> - Fixed typo in iser modparam description
>
> The change log of patch #3 needs isn&
1 - 100 of 2427 matches
Mail list logo