From: Yanjun Zhu <yanjun.zhu@linux.dev>
To: "matsuda-daisuke@fujitsu.com" <matsuda-daisuke@fujitsu.com>,
'Bart Van Assche' <bvanassche@acm.org>,
"linux-rdma@vger.kernel.org" <linux-rdma@vger.kernel.org>,
"leonro@nvidia.com" <leonro@nvidia.com>,
"jgg@nvidia.com" <jgg@nvidia.com>,
"zyjzyj2000@gmail.com" <zyjzyj2000@gmail.com>
Cc: "nvdimm@lists.linux.dev" <nvdimm@lists.linux.dev>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"rpearsonhpe@gmail.com" <rpearsonhpe@gmail.com>,
"yangx.jy@fujitsu.com" <yangx.jy@fujitsu.com>,
"lizhijian@fujitsu.com" <lizhijian@fujitsu.com>,
"y-goto@fujitsu.com" <y-goto@fujitsu.com>
Subject: Re: [RFC PATCH 2/7] RDMA/rxe: Convert the triple tasklets to workqueues
Date: Mon, 12 Sep 2022 16:29:51 +0800 [thread overview]
Message-ID: <7d30b552-a737-9f19-dea8-26978a6601ea@linux.dev> (raw)
In-Reply-To: <TYCPR01MB8455D739FC9FB034E3485C87E5449@TYCPR01MB8455.jpnprd01.prod.outlook.com>
在 2022/9/12 15:58, matsuda-daisuke@fujitsu.com 写道:
> On Mon, Sep 12, 2022 12:09 AM Bart Van Assche wrote:
>> On 9/11/22 00:10, Yanjun Zhu wrote:
>>> I also implemented a workqueue for rxe. IMO, can we add a variable to
>>> decide to use tasklet or workqueue?
>>>
>>> If user prefer using tasklet, he can set the variable to use
>>> tasklet. And the default is tasklet. Set the variable to another
>>> value to use workqueue.
> That's an interesting idea, but I am not sure how users specify it.
> IIRC, tasklets are generated when rdma link is added, typically by
> executing ' rdma link add' command. I don't think we can add
> an device specific option to the utility(iproute2/rdma).
In my workqueue implementation, I used a sysctl variable to set a value
to choose tasklet or workqueue.
You can use sysctl or driver parameter to decide to use tasklet or
workqueue.
And finally the Macros like #if ... #else ... #endif also acts as a method.
Zhu Yanjun
>
>> I'm in favor of removing all uses of the tasklet mechanism because of
>> the disadvantages of that mechanism. See also:
>> * "Eliminating tasklets" (https://lwn.net/Articles/239633/).
>> * "Modernizing the tasklet API" (https://lwn.net/Articles/830964/).
>> * Sebastian Andrzej Siewior's opinion about tasklets
>> (https://lore.kernel.org/all/YvovfXMJQAUBsvBZ@linutronix.de/).
> I am also in favor of using workqueues alone not only because of the
> disadvantages above but also to avoid complexity. I would like to know
> if there is anybody who will bothered by the change especially in terms
> of performance.
>
> Thanks,
> Daisuke
>
>> Thanks,
>>
>> Bart.
next prev parent reply other threads:[~2022-09-12 8:30 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-09-07 2:42 [RFC PATCH 0/7] RDMA/rxe: On-Demand Paging on SoftRoCE Daisuke Matsuda
2022-09-07 2:42 ` [RFC PATCH 1/7] IB/mlx5: Change ib_umem_odp_map_dma_single_page() to retain umem_mutex Daisuke Matsuda
2022-09-07 2:43 ` [RFC PATCH 2/7] RDMA/rxe: Convert the triple tasklets to workqueues Daisuke Matsuda
2022-09-09 19:39 ` Bob Pearson
2022-09-12 8:27 ` matsuda-daisuke
2022-09-11 7:10 ` Yanjun Zhu
2022-09-11 15:08 ` Bart Van Assche
2022-09-12 7:58 ` matsuda-daisuke
2022-09-12 8:29 ` Yanjun Zhu [this message]
2022-09-12 19:52 ` Bob Pearson
2022-09-28 6:40 ` matsuda-daisuke
2022-09-12 8:25 ` Yanjun Zhu
2022-09-07 2:43 ` [RFC PATCH 3/7] RDMA/rxe: Cleanup code for responder Atomic operations Daisuke Matsuda
2022-09-07 2:43 ` [RFC PATCH 4/7] RDMA/rxe: Add page invalidation support Daisuke Matsuda
2022-09-07 2:43 ` [RFC PATCH 5/7] RDMA/rxe: Allow registering MRs for On-Demand Paging Daisuke Matsuda
2022-09-08 16:57 ` Haris Iqbal
2022-09-09 0:55 ` matsuda-daisuke
2022-09-07 2:43 ` [RFC PATCH 6/7] RDMA/rxe: Add support for Send/Recv/Write/Read operations with ODP Daisuke Matsuda
2022-09-08 8:29 ` Leon Romanovsky
2022-09-09 2:45 ` matsuda-daisuke
2022-09-07 2:43 ` [RFC PATCH 7/7] RDMA/rxe: Add support for the traditional Atomic " Daisuke Matsuda
2022-09-08 8:40 ` [RFC PATCH 0/7] RDMA/rxe: On-Demand Paging on SoftRoCE Zhu Yanjun
2022-09-08 10:25 ` matsuda-daisuke
2022-09-09 3:07 ` Li Zhijian
2022-09-12 9:21 ` matsuda-daisuke
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=7d30b552-a737-9f19-dea8-26978a6601ea@linux.dev \
--to=yanjun.zhu@linux.dev \
--cc=bvanassche@acm.org \
--cc=jgg@nvidia.com \
--cc=leonro@nvidia.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=lizhijian@fujitsu.com \
--cc=matsuda-daisuke@fujitsu.com \
--cc=nvdimm@lists.linux.dev \
--cc=rpearsonhpe@gmail.com \
--cc=y-goto@fujitsu.com \
--cc=yangx.jy@fujitsu.com \
--cc=zyjzyj2000@gmail.com \
/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).