From: Boris Brezillon <boris.brezillon@collabora.com>
To: Steven Price <steven.price@arm.com>
Cc: dri-devel@lists.freedesktop.org, Rob Herring <robh+dt@kernel.org>,
Robin Murphy <robin.murphy@arm.com>,
Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Subject: Re: [RFC PATCH 0/7] drm/panfrost: Add a new submit ioctl
Date: Fri, 12 Mar 2021 15:15:04 +0100 [thread overview]
Message-ID: <20210312151504.6234b437@collabora.com> (raw)
In-Reply-To: <86062831-6608-9ae7-c513-e74d3a9c2e33@arm.com>
On Thu, 11 Mar 2021 12:16:33 +0000
Steven Price <steven.price@arm.com> wrote:
> Also the current code completely ignores PANFROST_BO_REF_READ. So either
> that should be defined as 0, or even better we support 3 modes:
>
> * Exclusive ('write' access)
> * Shared ('read' access)
> * No fence - ensures the BO is mapped, but doesn't add any implicit
> fences.
I looked at other drivers and they seem to have this
NO_FENCE/NO_IMPLICIT flag at the job/submit level and conditioned on
the presence of a in_sync_fd file. I can see per-BO NO_FENCE being
useful for sub-buffer accesses, assuming we don't want to deal with
other deps explicitly, but we could also force the NO_IMPLICIT behavior
for the whole job and let the user deal with all deps explicitly. TBH,
it's a bit unclear to me how that feature would be used by the vulkan
driver, so I'd rather leave that NO_FENCE use case for later.
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
prev parent reply other threads:[~2021-03-12 14:15 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-11 9:25 [RFC PATCH 0/7] drm/panfrost: Add a new submit ioctl Boris Brezillon
2021-03-11 9:25 ` [RFC PATCH 1/7] drm/panfrost: Pass a job to panfrost_{acquire, attach_object_fences}() Boris Brezillon
2021-03-11 9:25 ` [RFC PATCH 2/7] drm/panfrost: Collect implicit and explicit deps in an XArray Boris Brezillon
2021-03-11 9:25 ` [RFC PATCH 3/7] drm/panfrost: Move the mappings collection out of panfrost_lookup_bos() Boris Brezillon
2021-03-11 9:25 ` [RFC PATCH 4/7] drm/panfrost: Add BO access flags to relax dependencies between jobs Boris Brezillon
2021-03-11 9:25 ` [RFC PATCH 5/7] drm/panfrost: Add a new ioctl to submit batches Boris Brezillon
2021-03-11 9:25 ` [RFC PATCH 6/7] drm/panfrost: Advertise the SYNCOBJ_TIMELINE feature Boris Brezillon
2021-03-11 9:25 ` [RFC PATCH 7/7] drm/panfrost: Bump minor version to reflect the feature additions Boris Brezillon
2021-03-11 12:16 ` [RFC PATCH 0/7] drm/panfrost: Add a new submit ioctl Steven Price
2021-03-11 13:00 ` Boris Brezillon
2021-03-11 16:58 ` Jason Ekstrand
2021-03-11 17:24 ` Boris Brezillon
2021-03-11 18:11 ` Jason Ekstrand
2021-03-11 22:38 ` Alyssa Rosenzweig
2021-03-12 7:31 ` Boris Brezillon
2021-03-12 15:37 ` Jason Ekstrand
2021-03-12 18:25 ` Boris Brezillon
2021-03-12 20:06 ` Boris Brezillon
2021-03-12 21:48 ` Alyssa Rosenzweig
2021-03-12 14:15 ` Boris Brezillon [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=20210312151504.6234b437@collabora.com \
--to=boris.brezillon@collabora.com \
--cc=alyssa.rosenzweig@collabora.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=robh+dt@kernel.org \
--cc=robin.murphy@arm.com \
--cc=steven.price@arm.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