RE: [PATCH V9 13/15] mmc: block: Add CQE and blk-mq support
From: Hunter, Adrian
Date: Wed Oct 04 2017 - 15:23:17 EST
> -----Original Message-----
> From: Bartlomiej Zolnierkiewicz [mailto:b.zolnierkie@xxxxxxxxxxx]
> Sent: Wednesday, October 4, 2017 12:40 PM
> To: Linus Walleij <linus.walleij@xxxxxxxxxx>
> Cc: Hunter, Adrian <adrian.hunter@xxxxxxxxx>; Ulf Hansson
> <ulf.hansson@xxxxxxxxxx>; linux-mmc <linux-mmc@xxxxxxxxxxxxxxx>; linux-
> block <linux-block@xxxxxxxxxxxxxxx>; linux-kernel <linux-
> kernel@xxxxxxxxxxxxxxx>; Bough Chen <haibo.chen@xxxxxxx>; Alex
> Lemberg <alex.lemberg@xxxxxxxxxxx>; Nowak, Mateusz
> <mateusz.nowak@xxxxxxxxx>; Yuliy Izrailov <Yuliy.Izrailov@xxxxxxxxxxx>;
> Jaehoon Chung <jh80.chung@xxxxxxxxxxx>; Dong Aisheng
> <dongas86@xxxxxxxxx>; Das Asutosh <asutoshd@xxxxxxxxxxxxxx>;
> Zhangfei Gao <zhangfei.gao@xxxxxxxxx>; Sahitya Tummala
> <stummala@xxxxxxxxxxxxxx>; Harjani Ritesh <riteshh@xxxxxxxxxxxxxx>;
> Venu Byravarasu <vbyravarasu@xxxxxxxxxx>; Shawn Lin <shawn.lin@rock-
> chips.com>; Christoph Hellwig <hch@xxxxxx>
> Subject: Re: [PATCH V9 13/15] mmc: block: Add CQE and blk-mq support
> 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@xxxxxxxxx>
> > > 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@xxxxxxxxx>
> > 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
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.
Do you have any questions about the code?