linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Bart Van Assche <bart.vanassche@sandisk.com>
To: Christoph Hellwig <hch@lst.de>, <linux-rdma@vger.kernel.org>
Cc: <sagig@dev.mellanox.co.il>, <bart.vanassche@sandisk.com>,
	<axboe@fb.com>, <linux-scsi@vger.kernel.org>,
	<linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 07/13] IB: add a proper completion queue abstraction
Date: Thu, 10 Dec 2015 10:42:22 -0800	[thread overview]
Message-ID: <5669C78E.6070302@sandisk.com> (raw)
In-Reply-To: <1449521512-22921-8-git-send-email-hch@lst.de>

On 12/07/2015 12:51 PM, Christoph Hellwig wrote:
> This adds an abstraction that allows ULP to simply pass a completion
                                        ^^^

I think this should either be changed into either "an ULP" or "ULPs".

> +/**
> + * ib_process_direct_cq - process a CQ in caller context
> + * @cq:		CQ to process
> + * @budget:	number of CQEs to poll for
> + *
> + * This function is used to process all outstanding CQ entries on a
> + * %IB_POLL_DIRECT CQ.  It does not offload CQ processing to a different
> + * context and does not ask from completion interrupts from the HCA.
                                ^^^^

Should this perhaps be changed into "for" ?

> + *
> + * Note: for compatibility reasons -1 can be passed in %budget for unlimited
> + * polling.  Do not use this feature in new code, it will be remove soon.
                                                                 ^^^^^^

Did you perhaps intend "removed" ?


> +struct ib_cq *ib_alloc_cq(struct ib_device *dev, void *private,
> +		int nr_cqe, int comp_vector, enum ib_poll_context poll_ctx)
> +{
 > [ ... ]
> +	cq->wc = kmalloc_array(IB_POLL_BATCH, sizeof(*cq->wc), GFP_KERNEL);

Why is the wc array allocated separately instead of being embedded in 
struct ib_cq ? I think the faster completion queues can be created the 
better so if it is possible to eliminate the above kmalloc() call I 
would prefer that.

> --- a/drivers/infiniband/ulp/srp/ib_srp.c
> +++ b/drivers/infiniband/ulp/srp/ib_srp.c
> @@ -457,10 +457,11 @@ static struct srp_fr_pool *srp_alloc_fr_pool(struct srp_target_port *target)
>   static void srp_destroy_qp(struct srp_rdma_ch *ch)
>   {
>   	static struct ib_qp_attr attr = { .qp_state = IB_QPS_ERR };
> -	static struct ib_recv_wr wr = { .wr_id = SRP_LAST_WR_ID };
> +	static struct ib_recv_wr wr = { 0 };
>   	struct ib_recv_wr *bad_wr;
>   	int ret;

Is explicit initialization to "{ 0 }" really needed for static structures ?

Bart.

  reply	other threads:[~2015-12-10 18:42 UTC|newest]

Thread overview: 44+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-12-07 20:51 completion queue abstraction V2 Christoph Hellwig
2015-12-07 20:51 ` [PATCH 01/13] irq_poll: make blk-iopoll available outside the block layer Christoph Hellwig
2015-12-10 18:41   ` Bart Van Assche
2015-12-07 20:51 ` [PATCH 02/13] irq_poll: don't disable new irq_poll instances Christoph Hellwig
2015-12-10 18:41   ` Bart Van Assche
2015-12-07 20:51 ` [PATCH 03/13] irq_poll: fold irq_poll_sched_prep into irq_poll_sched Christoph Hellwig
2015-12-10 18:41   ` Bart Van Assche
2015-12-29  9:54   ` Bart Van Assche
2015-12-30  9:42     ` Christoph Hellwig
2016-01-20  7:02       ` Andrew Donnellan
2016-01-20  7:15         ` Andrew Donnellan
2015-12-07 20:51 ` [PATCH 04/13] irq_poll: fold irq_poll_disable_pending into irq_poll_softirq Christoph Hellwig
2015-12-10 18:41   ` Bart Van Assche
2015-12-07 20:51 ` [PATCH 05/13] irq_poll: mark __irq_poll_complete static Christoph Hellwig
2015-12-10 18:42   ` Bart Van Assche
2015-12-07 20:51 ` [PATCH 06/13] irq_poll: remove unused data and max fields Christoph Hellwig
2015-12-10 18:42   ` Bart Van Assche
2015-12-07 20:51 ` [PATCH 07/13] IB: add a proper completion queue abstraction Christoph Hellwig
2015-12-10 18:42   ` Bart Van Assche [this message]
2015-12-11 14:17     ` Christoph Hellwig
2016-01-15 13:54   ` Parav Pandit
2016-01-17  9:24     ` Sagi Grimberg
2016-01-17 11:01       ` Parav Pandit
2016-01-17 11:06         ` Sagi Grimberg
2016-01-17 11:09           ` Parav Pandit
2015-12-07 20:51 ` [PATCH 08/13] IB/srpt: chain RDMA READ/WRITE requests Christoph Hellwig
2015-12-10 18:42   ` Bart Van Assche
2015-12-29  9:58   ` Bart Van Assche
2015-12-30  9:43     ` Christoph Hellwig
2015-12-07 20:51 ` [PATCH 09/13] IB/srpt: use the new CQ API Christoph Hellwig
2015-12-10 18:42   ` Bart Van Assche
2015-12-07 20:51 ` [PATCH 10/13] IB/srp: " Christoph Hellwig
2015-12-10 18:42   ` Bart Van Assche
2015-12-11 14:22     ` Christoph Hellwig
2015-12-11 17:59       ` Doug Ledford
2015-12-12  8:08         ` Christoph Hellwig
2015-12-13 10:26           ` Sagi Grimberg
2015-12-14 16:26             ` Doug Ledford
2015-12-07 20:51 ` [PATCH 11/13] IB/iser: Use a dedicated descriptor for login Christoph Hellwig
2015-12-07 20:51 ` [PATCH 12/13] IB/iser: Use helper for container_of Christoph Hellwig
2015-12-07 20:51 ` [PATCH 13/13] IB/iser: Convert to CQ abstraction Christoph Hellwig
2015-12-13 10:25 ` completion queue abstraction V2 Sagi Grimberg
2015-12-23 19:44 ` Doug Ledford
2015-12-29  9:51 ` Bart Van Assche

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=5669C78E.6070302@sandisk.com \
    --to=bart.vanassche@sandisk.com \
    --cc=axboe@fb.com \
    --cc=hch@lst.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-rdma@vger.kernel.org \
    --cc=linux-scsi@vger.kernel.org \
    --cc=sagig@dev.mellanox.co.il \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).