linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: Mikhail Rudenko <mike.rudenko@gmail.com>
To: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Cc: Dafna Hirschfeld <dafna@fastmail.com>,
	Mauro Carvalho Chehab <mchehab@kernel.org>,
	Heiko Stuebner <heiko@sntech.de>,
	linux-media@vger.kernel.org, linux-rockchip@lists.infradead.org,
	linux-arm-kernel@lists.infradead.org,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH] media: rkisp1: allow non-coherent video capture buffers
Date: Tue, 14 Jan 2025 19:00:39 +0300	[thread overview]
Message-ID: <87bjw9s4s3.fsf@gmail.com> (raw)
In-Reply-To: <20250103152326.GP554@pendragon.ideasonboard.com>


Hi Laurent,

On 2025-01-03 at 17:23 +02, Laurent Pinchart <laurent.pinchart@ideasonboard.com> wrote:

> On Thu, Jan 02, 2025 at 06:35:00PM +0300, Mikhail Rudenko wrote:
>> Currently, the rkisp1 driver always uses coherent DMA allocations for
>> video capture buffers. However, on some platforms, using non-coherent
>> buffers can improve performance, especially when CPU processing of
>> MMAP'ed video buffers is required.
>>
>> For example, on the Rockchip RK3399 running at maximum CPU frequency,
>> the time to memcpy a frame from a 1280x720 XRGB32 MMAP'ed buffer to a
>> malloc'ed userspace buffer decreases from 7.7 ms to 1.1 ms when using
>> non-coherent DMA allocation. CPU usage also decreases accordingly.
>
> What's the time taken by the cache management operations ?

Sorry for the late reply, your question turned out a little more
interesting than I expected initially. :)

When capturing using Yavta with MMAP buffers under the conditions mentioned
in the commit message, ftrace gives 437.6 +- 1.1 us for
dma_sync_sgtable_for_cpu and 409 +- 14 us for
dma_sync_sgtable_for_device. Thus, it looks like using non-coherent
buffers in this case is more CPU-efficient even when considering cache
management overhead.

When trying to do the same measurements with libcamera, I failed. In a
typical libcamera use case when MMAP buffers are allocated from a
device, exported as dmabufs and then used for capture on the same device
with DMABUF memory type, cache management in kernel is skipped [1]
[2]. Also, vb2_dc_dmabuf_ops_{begin,end}_cpu_access are no-ops [3], so
DMA_BUF_IOCTL_SYNC from userspace does not work either.

So it looks like to make this change really useful, the above issue of
cache management for libcamera/DMABUF/videobuf2-dma-contig has to be
solved. I'm not an expert in this area, so any advice is kindly welcome. :)

[1] https://git.linuxtv.org/media.git/tree/drivers/media/common/videobuf2/videobuf2-core.c?id=94794b5ce4d90ab134b0b101a02fddf6e74c437d#n411
[2] https://git.linuxtv.org/media.git/tree/drivers/media/common/videobuf2/videobuf2-core.c?id=94794b5ce4d90ab134b0b101a02fddf6e74c437d#n829
[3] https://git.linuxtv.org/media.git/tree/drivers/media/common/videobuf2/videobuf2-dma-contig.c?id=94794b5ce4d90ab134b0b101a02fddf6e74c437d#n426

--
Best regards,
Mikhail Rudenko


  reply	other threads:[~2025-01-14 17:09 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-01-02 15:35 [PATCH] media: rkisp1: allow non-coherent video capture buffers Mikhail Rudenko
2025-01-03 15:23 ` Laurent Pinchart
2025-01-14 16:00   ` Mikhail Rudenko [this message]
2025-01-15  8:31     ` Tomasz Figa
2025-01-15 13:24       ` Mikhail Rudenko
2025-01-15 14:46         ` Tomasz Figa
2025-01-15 17:29           ` Mikhail Rudenko
2025-01-15 19:13     ` Nicolas Dufresne
2025-02-27 17:05     ` Jacopo Mondi
2025-02-27 20:46       ` Mikhail Rudenko
2025-02-28  2:58         ` Nicolas Dufresne
2025-02-28  9:54           ` Hans Verkuil
2025-02-28 10:00       ` Tomasz Figa
2025-02-28 10:18         ` Jacopo Mondi
2025-02-28 10:28           ` Tomasz Figa
2025-02-28 10:48             ` Jacopo Mondi
2025-02-28 11:19               ` Tomasz Figa

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=87bjw9s4s3.fsf@gmail.com \
    --to=mike.rudenko@gmail.com \
    --cc=dafna@fastmail.com \
    --cc=heiko@sntech.de \
    --cc=laurent.pinchart@ideasonboard.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-media@vger.kernel.org \
    --cc=linux-rockchip@lists.infradead.org \
    --cc=mchehab@kernel.org \
    /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).