Subject: Re: [PATCH 2/9] IB: add a proper completion queue abstraction
To: Sagi Grimberg <sagig@dev.mellanox.co.il>, Christoph Hellwig <hch@lst.de>,
        "linux-rdma@vger.kernel.org" <linux-rdma@vger.kernel.org>
References: <1447422410-20891-1-git-send-email-hch@lst.de>
 <1447422410-20891-3-git-send-email-hch@lst.de> <564B697A.2020601@sandisk.com>
 <564C2F01.6020407@dev.mellanox.co.il>
CC: "axboe@fb.com" <axboe@fb.com>,
        "linux-scsi@vger.kernel.org" <linux-scsi@vger.kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
From: Bart Van Assche <bart.vanassche@sandisk.com>
Message-ID: <564CC15E.7030602@sandisk.com>
Date: Wed, 18 Nov 2015 10:20:14 -0800
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101
 Thunderbird/38.3.0
MIME-Version: 1.0
In-Reply-To: <564C2F01.6020407@dev.mellanox.co.il>
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Transfer-Encoding: 7bit
SpamDiagnosticOutput: 1:23
SpamDiagnosticMetadata: NSPM
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 18 Nov 2015 18:20:16.6437
 (UTC)
X-MS-Exchange-CrossTenant-Id: fcd9ea9c-ae8c-460c-ab3c-3db42d7ac64d
X-MS-Exchange-CrossTenant-OriginalAttributedTenantConnectingIp: TenantId=fcd9ea9c-ae8c-460c-ab3c-3db42d7ac64d;Ip=[63.163.107.173];Helo=[milsmgep12.sandisk.com]
X-MS-Exchange-CrossTenant-FromEntityHeader: HybridOnPrem
X-MS-Exchange-Transport-CrossTenantHeadersStamped: SN1PR02MB1389
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org
Content-Length: 2110
Lines: 51

On 11/17/2015 11:55 PM, Sagi Grimberg wrote:
>>> +static void ib_cq_poll_work(struct work_struct *work)
>>> +{
>>> +    struct ib_cq *cq = container_of(work, struct ib_cq, work);
>>> +    int completed;
>>> +
>>> +    completed = __ib_process_cq(cq, IB_POLL_BUDGET_WORKQUEUE);
>>> +    if (completed >= IB_POLL_BUDGET_WORKQUEUE ||
>>> +        ib_req_notify_cq(cq, IB_POLL_FLAGS) > 0)
>>> +        queue_work(ib_comp_wq, &cq->work);
>>> +}
>>> +
>>> +static void ib_cq_completion_workqueue(struct ib_cq *cq, void *private)
>>> +{
>>> +    queue_work(ib_comp_wq, &cq->work);
>>> +}
>>
>> The above code will cause all polling to occur on the context of the CPU
>> that received the completion interrupt. This approach is not powerful
>> enough. For certain workloads throughput is higher if work completions
>> are processed by another CPU core on the same CPU socket. Has it been
>> considered to make the CPU core on which work completions are processed
>> configurable ?
>
> The workqueue is unbound. This means that the functionality you are
> you are asking for exists.

Hello Sagi,

Are you perhaps referring to the sysfs CPU mask that allows to control 
workqueue affinity ? I expect that setting the CPU mask for an entire 
pool through sysfs will lead to suboptimal results. What I have learned 
by tuning target systems is that there is a significant performance 
difference (> 30% IOPS) between a configuration where each completion 
thread is pinned to exactly one CPU compared to allowing the scheduler 
to choose a CPU.

Controlling the CPU affinity of worker threads with the taskset command 
is not possible since the function create_worker() in kernel/workqueue.c 
calls kthread_bind_mask(). That function sets PF_NO_SETAFFINITY. From 
sched.h:

#define PF_NO_SETAFFINITY 0x04000000 /* Userland is not allowed to 
meddle with cpus_allowed */

Bart.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/