On Tue, Sep 8, 2015 at 1:54 PM, Haggai Eran wrote:
> On 08/09/2015 10:04, Parav Pandit wrote:
>> On Tue, Sep 8, 2015 at 11:18 AM, Haggai Eran wrote:
>>> On 07/09/2015 23:38, Parav Pandit wrote:
@@ -2676,7 +2686,7 @@ static inline int
On 07/09/2015 23:38, Parav Pandit wrote:
> +void devcgroup_rdma_uncharge_resource(struct ib_ucontext *ucontext,
> + enum devcgroup_rdma_rt type, int num)
> +{
> + struct dev_cgroup *dev_cg, *p;
> + struct task_struct *ctx_task;
> +
> + if (!num)
> +
On 07/09/2015 23:38, Parav Pandit wrote:
> +static void init_ucontext_lists(struct ib_ucontext *ucontext)
> +{
> + INIT_LIST_HEAD(>pd_list);
> + INIT_LIST_HEAD(>mr_list);
> + INIT_LIST_HEAD(>mw_list);
> + INIT_LIST_HEAD(>cq_list);
> + INIT_LIST_HEAD(>qp_list);
> +
On Tue, Sep 8, 2015 at 11:18 AM, Haggai Eran wrote:
> On 07/09/2015 23:38, Parav Pandit wrote:
>> @@ -2676,7 +2686,7 @@ static inline int thread_group_empty(struct
>> task_struct *p)
>> * Protects ->fs, ->files, ->mm, ->group_info, ->comm, keyring
>> * subscriptions and
On Tue, Sep 8, 2015 at 11:01 AM, Haggai Eran wrote:
> On 07/09/2015 23:38, Parav Pandit wrote:
>> diff --git a/include/linux/device_cgroup.h b/include/linux/device_cgroup.h
>> index 8b64221..cdbdd60 100644
>> --- a/include/linux/device_cgroup.h
>> +++
On 07/09/2015 23:38, Parav Pandit wrote:
> +/* RDMA resources from device cgroup perspective */
> +enum devcgroup_rdma_rt {
> + DEVCG_RDMA_RES_TYPE_UCTX,
> + DEVCG_RDMA_RES_TYPE_CQ,
> + DEVCG_RDMA_RES_TYPE_PD,
> + DEVCG_RDMA_RES_TYPE_AH,
> + DEVCG_RDMA_RES_TYPE_MR,
> +
On 08/09/2015 10:04, Parav Pandit wrote:
> On Tue, Sep 8, 2015 at 11:18 AM, Haggai Eran wrote:
>> On 07/09/2015 23:38, Parav Pandit wrote:
>>> @@ -2676,7 +2686,7 @@ static inline int thread_group_empty(struct
>>> task_struct *p)
>>> * Protects ->fs, ->files, ->mm,
On 9/6/2015 9:45 AM, Christoph Hellwig wrote:
Hi All,
right now RDMA/CM works on a QP basis, but seems very awakward if you
want multiple QPs as part of a single logical device, which will be
useful for a lot of modern protocols. For example we will need to check
in the CM handler that we're
On Tue, Sep 8, 2015 at 2:06 PM, Haggai Eran wrote:
> On 07/09/2015 23:38, Parav Pandit wrote:
>> +void devcgroup_rdma_uncharge_resource(struct ib_ucontext *ucontext,
>> + enum devcgroup_rdma_rt type, int num)
>> +{
>> + struct dev_cgroup
On Tue, Sep 8, 2015 at 2:10 PM, Haggai Eran wrote:
> On 07/09/2015 23:38, Parav Pandit wrote:
>> +static void init_ucontext_lists(struct ib_ucontext *ucontext)
>> +{
>> + INIT_LIST_HEAD(>pd_list);
>> + INIT_LIST_HEAD(>mr_list);
>> + INIT_LIST_HEAD(>mw_list);
>> +
On 07/09/2015 23:38, Parav Pandit wrote:
> Currently user space applications can easily take away all the rdma
> device specific resources such as AH, CQ, QP, MR etc. Due to which other
> applications in other cgroup or kernel space ULPs may not even get chance
> to allocate any rdma resources.
>
On Tue, Sep 8, 2015 at 1:52 PM, Haggai Eran wrote:
> On 07/09/2015 23:38, Parav Pandit wrote:
>> +/* RDMA resources from device cgroup perspective */
>> +enum devcgroup_rdma_rt {
>> + DEVCG_RDMA_RES_TYPE_UCTX,
>> + DEVCG_RDMA_RES_TYPE_CQ,
>> +
> This fixes the incorrect assumation that the function ib_send_cm_drep
> always runs successfully and never fails in the function rdma_disconnect
> by making this call be assigned the variable used to return to callers
> of rdma_disconnect in order to make sure that a error code is returned
> if
Signed-off-by: Steve Wise
> -Original Message-
> From: Hariprasad Shenai [mailto:haripra...@chelsio.com]
> Sent: Monday, September 07, 2015 11:27 PM
> To: dledf...@redhat.com
> Cc: linux-rdma@vger.kernel.org; sw...@opengridcomputing.com; Hariprasad Shenai
>
On 9/6/2015 11:15 AM, Bart Van Assche wrote:
On 09/06/15 00:50, Christoph Hellwig wrote:
Note that the SRP driver already in tree is a good example for this,
although it doesn't use RDMA/CM and thus already operates on a
per-ib_device level.
The challenges with regard to adding RDMA/CM
On Tue, Sep 08, 2015 at 03:32:11PM +0300, Sagi Grimberg wrote:
> The CM is responsible of establishing an RDMA channel. What you are
> referring to is a concept of a session. I'm not entirely sure how we can
> fit a model where the CM establishes a multi-channel session as the
> CM request
On Tue, Sep 8, 2015 at 7:20 PM, Haggai Eran wrote:
> On 08/09/2015 13:18, Parav Pandit wrote:
>>> >
>> + * RDMA resource limits are hierarchical, so the highest configured
>> limit of
>> + * the hierarchy is enforced. Allowing resource limit configuration to
On 08/09/2015 13:22, Parav Pandit wrote:
> On Tue, Sep 8, 2015 at 2:10 PM, Haggai Eran wrote:
>> On 07/09/2015 23:38, Parav Pandit wrote:
>>> +static void init_ucontext_lists(struct ib_ucontext *ucontext)
>>> +{
>>> + INIT_LIST_HEAD(>pd_list);
>>> +
On 08/09/2015 13:50, Parav Pandit wrote:
> On Tue, Sep 8, 2015 at 2:06 PM, Haggai Eran wrote:
>> On 07/09/2015 23:38, Parav Pandit wrote:
>>> +void devcgroup_rdma_uncharge_resource(struct ib_ucontext *ucontext,
>>> + enum devcgroup_rdma_rt
On 08/09/2015 13:18, Parav Pandit wrote:
>> >
>>> >> + * RDMA resource limits are hierarchical, so the highest configured
>>> >> limit of
>>> >> + * the hierarchy is enforced. Allowing resource limit configuration to
>>> >> default
>>> >> + * cgroup allows fair share to kernel space ULPs as
On 09/08/2015 06:57 AM, Tom Talpey wrote:
On 9/6/2015 11:15 AM, Bart Van Assche wrote:
On 09/06/15 00:50, Christoph Hellwig wrote:
Note that the SRP driver already in tree is a good example for this,
although it doesn't use RDMA/CM and thus already operates on a
per-ib_device level.
The
On Tue, Sep 8, 2015 at 9:24 AM, Doug Ledford wrote:
>
> Matan Barak (5):
> net: Add info for NETDEV_CHANGEUPPER event
Why the hell is this coming in through the infiniband tree? Especially
with the networking tree having done the same thing differently?
I see that
On Tue, Sep 8, 2015 at 5:21 PM, Linus Torvalds
wrote:
>
> This needs a good explanation, because for now this pull request is dead to
> me.
Just to clarify: it's really the combination of:
(a) you were told about this
(b) you rebased your commit series after
On Tue, Sep 8, 2015 at 7:50 PM, Doug Ledford wrote:
>
> Because the tree isn't buildable, let alone testable, without it.
You're missing the ENTIRE POINT.
> Well, I expected it to be handled without much effort on your part and
> the existing network infrastructure to be
On 09/08/2015 11:08 PM, Linus Torvalds wrote:
> On Tue, Sep 8, 2015 at 7:50 PM, Doug Ledford wrote:
>>
>> Because the tree isn't buildable, let alone testable, without it.
>
> You're missing the ENTIRE POINT.
>
>> Well, I expected it to be handled without much effort on
On Tue, Sep 8, 2015 at 8:53 PM, Tejun Heo wrote:
> Hello, Parav.
>
> On Tue, Sep 08, 2015 at 02:08:16AM +0530, Parav Pandit wrote:
>> Currently user space applications can easily take away all the rdma
>> device specific resources such as AH, CQ, QP, MR etc. Due to which other
>>
On 09/08/2015 08:21 PM, Linus Torvalds wrote:
> On Tue, Sep 8, 2015 at 9:24 AM, Doug Ledford wrote:
>>
>> Matan Barak (5):
>> net: Add info for NETDEV_CHANGEUPPER event
>
> Why the hell is this coming in through the infiniband tree? Especially
> with the networking
On 09/08/2015 08:35 PM, Linus Torvalds wrote:
> On Tue, Sep 8, 2015 at 5:21 PM, Linus Torvalds
> wrote:
>>
>> This needs a good explanation, because for now this pull request is dead to
>> me.
>
> Just to clarify: it's really the combination of:
>
> (a) you were
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 necessary on your part". That combined with my lack of deep
> internal knowledge of what it is that Stephen is doing made me go "Ok,
> he
Hi,
OFED-3.18-1-rc1 is available at:
https://openfabrics.org/downloads/OFED/ofed-3.18-1/OFED-3.18-1-rc1.tgz
To get BUILD_ID run ofed_info
Please report any issues in bugzilla
http://bugs.openfabrics.org/bugzilla/ for OFED 3.18-1
Release notes:
> This fixes mutex locking to lock before unlocking in the function
> cma_ib_mc_handler for the mutex handler_mutex as part of the pointer
> id_priv before unlocking it after the critical region for event handler
> work and execution in order to have actual proper concurrent execution
> protection
Hi Linus,
This is a fairly sizeable set of changes. I've put them through a
decent amount of testing prior to sending the pull request due to
that. There are still a few fixups that I know are coming, but I
wanted to go ahead and get the big, sizable chunk into your hands
sooner rather than
> This fixes error handling in the function cm_init_av_by_path
> to properly check if the internal call to the function
> ib_init_ah_from_path has failed by returning a error code and
> if so return immediately to the caller with this error code in
> order to signal/allow the caller to handle the
Hello, Parav.
On Tue, Sep 08, 2015 at 02:08:16AM +0530, Parav Pandit wrote:
> Currently user space applications can easily take away all the rdma
> device specific resources such as AH, CQ, QP, MR etc. Due to which other
> applications in other cgroup or kernel space ULPs may not even get chance
> diff --git a/drivers/infiniband/core/multicast.c
> b/drivers/infiniband/core/multicast.c
> index 2cb865c..9284337 100644
> --- a/drivers/infiniband/core/multicast.c
> +++ b/drivers/infiniband/core/multicast.c
> @@ -526,8 +526,9 @@ static void join_handler(int status, struct
> ib_sa_mcmember_rec
From: Ira Weiny
Byteswap link_width_downgrade_*_active values before sending on the wire. In
addition properly define the Port State Info structure.
Reviewed-by: Dennis Dalessandro
Reviewed-by: Christian Gomez
36 matches
Mail list logo