Linux block layer
 help / color / mirror / Atom feed
From: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
To: "Hunter, Adrian" <adrian.hunter@intel.com>
Cc: Linus Walleij <linus.walleij@linaro.org>,
	Ulf Hansson <ulf.hansson@linaro.org>,
	linux-mmc <linux-mmc@vger.kernel.org>,
	linux-block <linux-block@vger.kernel.org>,
	linux-kernel <linux-kernel@vger.kernel.org>,
	Bough Chen <haibo.chen@nxp.com>,
	Alex Lemberg <alex.lemberg@sandisk.com>,
	"Nowak, Mateusz" <mateusz.nowak@intel.com>,
	Yuliy Izrailov <Yuliy.Izrailov@sandisk.com>,
	Jaehoon Chung <jh80.chung@samsung.com>,
	Dong Aisheng <dongas86@gmail.com>,
	Das Asutosh <asutoshd@codeaurora.org>,
	Zhangfei Gao <zhangfei.gao@gmail.com>,
	Sahitya Tummala <stummala@codeaurora.org>,
	Harjani Ritesh <riteshh@codeaurora.org>,
	Venu Byravarasu <vbyravarasu@nvidia.com>,
	Shawn Lin <shawn.lin@rock-chips.com>,
	Christoph Hellwig <hch@lst.de>
Subject: Re: [PATCH V9 13/15] mmc: block: Add CQE and blk-mq support
Date: Thu, 05 Oct 2017 14:00:48 +0200	[thread overview]
Message-ID: <1591139.5Ihy91Pq6o@amdc3058> (raw)
In-Reply-To: <363DA0ED52042842948283D2FC38E4649BF33EFD@IRSMSX106.ger.corp.intel.com>


Hi,

On Wednesday, October 04, 2017 07:23:07 PM Hunter, Adrian wrote:
> > -----Original Message-----
> > From: Bartlomiej Zolnierkiewicz [mailto:b.zolnierkie@samsung.com]
> > Sent: Wednesday, October 4, 2017 12:40 PM
> > To: Linus Walleij <linus.walleij@linaro.org>
> > Cc: Hunter, Adrian <adrian.hunter@intel.com>; Ulf Hansson
> > <ulf.hansson@linaro.org>; linux-mmc <linux-mmc@vger.kernel.org>; linux-
> > block <linux-block@vger.kernel.org>; linux-kernel <linux-
> > kernel@vger.kernel.org>; Bough Chen <haibo.chen@nxp.com>; Alex
> > Lemberg <alex.lemberg@sandisk.com>; Nowak, Mateusz
> > <mateusz.nowak@intel.com>; Yuliy Izrailov <Yuliy.Izrailov@sandisk.com>;
> > Jaehoon Chung <jh80.chung@samsung.com>; Dong Aisheng
> > <dongas86@gmail.com>; Das Asutosh <asutoshd@codeaurora.org>;
> > Zhangfei Gao <zhangfei.gao@gmail.com>; Sahitya Tummala
> > <stummala@codeaurora.org>; Harjani Ritesh <riteshh@codeaurora.org>;
> > Venu Byravarasu <vbyravarasu@nvidia.com>; Shawn Lin <shawn.lin@rock-
> > chips.com>; Christoph Hellwig <hch@lst.de>
> > Subject: Re: [PATCH V9 13/15] mmc: block: Add CQE and blk-mq support
> > 
> > 
> > Hi,
> > 
> > On Wednesday, October 04, 2017 09:39:45 AM Linus Walleij wrote:
> > > On Fri, Sep 22, 2017 at 2:37 PM, Adrian Hunter <adrian.hunter@intel.com>
> > wrote:
> > >
> > > > Add CQE support to the block driver, including:
> > > >     - optionally using DCMD for flush requests
> > > >     - "manually" issuing discard requests
> > > >     - issuing read / write requests to the CQE
> > > >     - supporting block-layer timeouts
> > > >     - handling recovery
> > > >     - supporting re-tuning
> > > >
> > > > CQE offers 25% - 50% better random multi-threaded I/O.  There is a
> > > > slight (e.g. 2%) drop in sequential read speed but no observable
> > > > change to sequential write.
> > > >
> > > > CQE automatically sends the commands to complete requests.  However
> > > > it only supports reads / writes and so-called "direct commands"
> > > > (DCMD).  Furthermore DCMD is limited to one command at a time, but
> > discards require 3 commands.
> > > > That makes issuing discards through CQE very awkward, but some CQE's
> > > > don't support DCMD anyway.  So for discards, the existing non-CQE
> > > > approach is taken, where the mmc core code issues the 3 commands one
> > at a time i.e.
> > > > mmc_erase(). Where DCMD is used, is for issuing flushes.
> > > >
> > > > For host controllers without CQE support, blk-mq support is extended
> > > > to synchronous reads/writes or, if the host supports
> > > > CAP_WAIT_WHILE_BUSY, asynchonous reads/writes.  The advantage of
> > > > asynchronous reads/writes is that it allows the preparation of the
> > > > next request while the current request is in progress.
> > > >
> > > > Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
> > >
> > > I am trying to wrap my head around this large patch. The size makes it
> > > hard but I am doing my best.
> > 
> > I also think that this patch should be split on two patches. The 1st one
> > introducing blk-mq and the 2nd one adding CQE support.
> > 
> > [ I don't agree that they make more sense together, on the contrary,
> >   it is very difficult to properly analyze blk-mq changes on their
> >   own while there are mixed with CQE related ones. ]
> 
> The CQE and non-CQE code paths are clearly marked.  But maybe you

The combined patch is > 1 kLOC which is a lot and since the CQE and
non-CQE code paths are clearly marked it should be really easy to
split them.

> are asking what the code would look like if we *never* had to support CQE.
> And my point is we *do* have to support CQE. 

We *do* but not in the same step, it is just normal kernel engineering
practice to split patches on logical changes. This is not asking about
something extraordinary.

> Do you have any questions about the code?

I would really hope to see more documentation about blk-mq changes
first before starting to review it. The patch description documents
CQE changes but not blk-mq ones and the code itself is also lacking
in-code documentation describing new functions added for blk-mq.

Best regards,
--
Bartlomiej Zolnierkiewicz
Samsung R&D Institute Poland
Samsung Electronics

  reply	other threads:[~2017-10-05 12:00 UTC|newest]

Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-09-22 12:36 [PATCH V9 00/15] mmc: Add Command Queue support Adrian Hunter
2017-09-22 12:36 ` [PATCH V9 01/15] mmc: core: Remove unnecessary host claim Adrian Hunter
2017-09-22 14:03   ` Linus Walleij
2017-09-22 12:36 ` [PATCH V9 02/15] mmc: core: Introduce host claiming by context Adrian Hunter
2017-09-22 12:36 ` [PATCH V9 03/15] mmc: core: Add support for handling CQE requests Adrian Hunter
2017-09-26 15:31   ` Linus Walleij
2017-09-22 12:36 ` [PATCH V9 04/15] mmc: mmc: Enable Command Queuing Adrian Hunter
2017-09-26 15:33   ` Linus Walleij
2017-09-22 12:36 ` [PATCH V9 05/15] mmc: mmc: Enable CQE's Adrian Hunter
2017-09-26 15:33   ` Linus Walleij
2017-09-22 12:36 ` [PATCH V9 06/15] mmc: block: Use local variables in mmc_blk_data_prep() Adrian Hunter
2017-09-26 15:34   ` Linus Walleij
2017-09-22 12:36 ` [PATCH V9 07/15] mmc: block: Prepare CQE data Adrian Hunter
2017-09-26 15:37   ` Linus Walleij
2017-09-22 12:36 ` [PATCH V9 08/15] mmc: block: Factor out mmc_setup_queue() Adrian Hunter
2017-09-26 15:38   ` Linus Walleij
2017-09-22 12:36 ` [PATCH V9 09/15] mmc: core: Add parameter use_blk_mq Adrian Hunter
2017-09-26 23:42   ` Linus Walleij
2017-09-27 12:02     ` Adrian Hunter
2017-09-27 19:49       ` Linus Walleij
2017-09-27 12:58     ` Avri Altman
2017-10-02  8:30   ` Ulf Hansson
2017-09-22 12:36 ` [PATCH V9 10/15] mmc: core: Export mmc_start_bkops() Adrian Hunter
2017-09-26 23:46   ` Linus Walleij
2017-09-22 12:37 ` [PATCH V9 11/15] mmc: core: Export mmc_start_request() Adrian Hunter
2017-09-26 23:47   ` Linus Walleij
2017-09-22 12:37 ` [PATCH V9 12/15] mmc: core: Export mmc_retune_hold_now() and mmc_retune_release() Adrian Hunter
2017-09-26 23:49   ` Linus Walleij
2017-10-02  8:30   ` Ulf Hansson
2017-09-22 12:37 ` [PATCH V9 13/15] mmc: block: Add CQE and blk-mq support Adrian Hunter
2017-10-02  8:32   ` Ulf Hansson
2017-10-06  8:03     ` Adrian Hunter
2017-10-04  7:39   ` Linus Walleij
2017-10-04  9:39     ` Bartlomiej Zolnierkiewicz
2017-10-04  9:48       ` Ulf Hansson
2017-10-05 10:55         ` Bartlomiej Zolnierkiewicz
2017-10-05 11:12           ` Ulf Hansson
2017-10-04 19:23       ` Hunter, Adrian
2017-10-05 12:00         ` Bartlomiej Zolnierkiewicz [this message]
2017-10-05 14:14           ` Bartlomiej Zolnierkiewicz
2017-10-04 13:27     ` Adrian Hunter
2017-10-05  8:59   ` [PATCH V10 " Adrian Hunter
2017-09-22 12:37 ` [PATCH V9 14/15] mmc: cqhci: support for command queue enabled host Adrian Hunter
2017-09-25  2:39   ` Bough Chen
2017-09-25  6:32     ` Adrian Hunter
2017-09-25  7:27   ` [PATCH V10 " Adrian Hunter
2017-09-22 12:37 ` [PATCH V9 15/15] mmc: sdhci-pci: Add CQHCI support for Intel GLK Adrian Hunter
2017-09-26 22:25 ` [PATCH V9 00/15] mmc: Add Command Queue support Ulf Hansson
2017-10-11 13:24   ` Ulf Hansson

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=1591139.5Ihy91Pq6o@amdc3058 \
    --to=b.zolnierkie@samsung.com \
    --cc=Yuliy.Izrailov@sandisk.com \
    --cc=adrian.hunter@intel.com \
    --cc=alex.lemberg@sandisk.com \
    --cc=asutoshd@codeaurora.org \
    --cc=dongas86@gmail.com \
    --cc=haibo.chen@nxp.com \
    --cc=hch@lst.de \
    --cc=jh80.chung@samsung.com \
    --cc=linus.walleij@linaro.org \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mmc@vger.kernel.org \
    --cc=mateusz.nowak@intel.com \
    --cc=riteshh@codeaurora.org \
    --cc=shawn.lin@rock-chips.com \
    --cc=stummala@codeaurora.org \
    --cc=ulf.hansson@linaro.org \
    --cc=vbyravarasu@nvidia.com \
    --cc=zhangfei.gao@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