linux-spi.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Mark Brown <broonie@kernel.org>
To: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
Cc: "linux-spi@vger.kernel.org" <linux-spi@vger.kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Fabio Estevam <festevam@gmail.com>,
	Rasmus Villemoes <Rasmus.Villemoes@prevas.se>
Subject: Re: [RFC PATCH 0/4] spi: spi-fsl-spi: try to make cpu-mode transfers faster
Date: Tue, 2 Apr 2019 16:10:05 +0700	[thread overview]
Message-ID: <20190402091005.GC2059@sirena.org.uk> (raw)
In-Reply-To: <ce9e8ca8-6db7-1be8-5ea3-c098b1c1c813@prevas.dk>

[-- Attachment #1: Type: text/plain, Size: 858 bytes --]

On Tue, Apr 02, 2019 at 08:43:51AM +0000, Rasmus Villemoes wrote:

> Thanks! There's one other option I can think of: don't do the interrupts
> at all, but just busy-wait for the completion of each word transfer (in
> a cpu_relax() loop). That could be guarded by something like
> 1000000*bits_per_word < hz (roughly, the word transfer takes less than 1
> us). At least on -rt, having the interrupt thread scheduled in and out
> again easily takes more than 1us of cpu time, and AFAIU we'd still be
> preemptible throughout - and/or one can throw in a cond_resched() every
> nnn words. But this might be a bit -rt specific, and the 1us threshold
> is rather arbitrary.

Yeah, that's definitely worth exploring as a mitigation but obviously
with things like flash I/O that gets a bit rude.  Hopefully what's there
at the minute turns out to be robust enough.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

      reply	other threads:[~2019-04-02  9:10 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-03-27 14:30 [RFC PATCH 0/4] spi: spi-fsl-spi: try to make cpu-mode transfers faster Rasmus Villemoes
2019-03-27 14:30 ` [RFC PATCH 1/4] spi: spi-fsl-spi: remove always-true conditional in fsl_spi_do_one_msg Rasmus Villemoes
2019-04-01  8:54   ` Applied "spi: spi-fsl-spi: remove always-true conditional in fsl_spi_do_one_msg" to the spi tree Mark Brown
2019-03-27 14:30 ` [RFC PATCH 3/4] spi: spi-fsl-spi: allow changing bits_per_word while CS is still active Rasmus Villemoes
2019-03-27 14:30 ` [RFC PATCH 2/4] spi: spi-fsl-spi: relax message sanity checking a little Rasmus Villemoes
2019-04-01  8:54   ` Applied "spi: spi-fsl-spi: relax message sanity checking a little" to the spi tree Mark Brown
2019-03-27 14:30 ` [RFC PATCH 4/4] spi: spi-fsl-spi: automatically adapt bits-per-word in cpu mode Rasmus Villemoes
2019-04-01  7:34 ` [RFC PATCH 0/4] spi: spi-fsl-spi: try to make cpu-mode transfers faster Mark Brown
2019-04-02  8:43   ` Rasmus Villemoes
2019-04-02  9:10     ` Mark Brown [this message]

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=20190402091005.GC2059@sirena.org.uk \
    --to=broonie@kernel.org \
    --cc=Rasmus.Villemoes@prevas.se \
    --cc=festevam@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-spi@vger.kernel.org \
    --cc=rasmus.villemoes@prevas.dk \
    /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).