From: Dmitry Osipenko <digetx@gmail.com>
To: Mikko Perttunen <cyndis@kapsi.fi>,
Thierry Reding <thierry.reding@gmail.com>,
Jon Hunter <jonathanh@nvidia.com>,
David Airlie <airlied@linux.ie>, Daniel Vetter <daniel@ffwll.ch>,
sumit.semwal@linaro.org, gustavo@padovan.org
Cc: "linux-tegra@vger.kernel.org" <linux-tegra@vger.kernel.org>,
talho@nvidia.com, bhuntsman@nvidia.com,
dri-devel <dri-devel@lists.freedesktop.org>
Subject: Re: [RFC] Host1x/TegraDRM UAPI (drm_tegra_submit_command)
Date: Sun, 28 Jun 2020 01:32:00 +0300 [thread overview]
Message-ID: <f31fbe3b-3fc5-2ece-4c2c-9ff4e99995d6@gmail.com> (raw)
In-Reply-To: <9b06b7ec-f952-2561-7afb-5653514cd5d3@kapsi.fi>
23.06.2020 15:09, Mikko Perttunen пишет:
> /* Command is an opcode gather from a GEM handle */
> #define DRM_TEGRA_SUBMIT_COMMAND_GATHER 0
> /* Command is an opcode gather from a user pointer */
> #define DRM_TEGRA_SUBMIT_COMMAND_GATHER_UPTR 1
> /* Command is a wait for syncpt fence completion */
> #define DRM_TEGRA_SUBMIT_COMMAND_WAIT_SYNCPT 2
> /* Command is a wait for SYNC_FILE FD completion */
> #define DRM_TEGRA_SUBMIT_COMMAND_WAIT_SYNC_FILE 3
> /* Command is a wait for DRM syncobj completion */
> #define DRM_TEGRA_SUBMIT_COMMAND_WAIT_SYNCOBJ 4
>
> /*
> * Allow driver to skip command execution if engine
> * was not accessed by another channel between
> * submissions.
> */
> #define DRM_TEGRA_SUBMIT_CONTEXT_SETUP (1<<0)
>
> struct drm_tegra_submit_command {
> __u16 type;
> __u16 flags;
Shouldn't the "packed" attribute be needed if a non-32bit aligned fields
are used?
> union {
> struct {
> /* GEM handle */
> __u32 handle;
>
> /*
> * Offset into GEM object in bytes.
> * Must be aligned by 4.
> */
> __u64 offset;
64bits for a gather offset is a bit too much, in most cases gathers are
under 4K.
u32 should be more than enough (maybe even u16 if offset is given in a
dword granularity).
> /*
> * Length of gather in bytes.
> * Must be aligned by 4.
> */
> __u64 length;
u32/16
> } gather;
> struct {
> __u32 reserved[1];
>
> /*
> * Pointer to gather data.
> * Must be aligned by 4 bytes.
> */
> __u64 base;
> /*
> * Length of gather in bytes.
> * Must be aligned by 4.
> */
> __u64 length;
> } gather_uptr;
What about to inline the UPTR gather data and relocs into the
drm_tegra_submit_command[] buffer:
struct drm_tegra_submit_command {
struct {
u16 num_words;
u16 num_relocs;
gather_data[];
drm_tegra_submit_relocation relocs[];
} gather_uptr;
};
struct drm_tegra_channel_submit {
__u32 num_syncpt_incrs;
__u32 syncpt_idx;
__u64 commands_ptr;
__u32 commands_size;
...
};
struct drm_tegra_submit_command example[] = {
cmd.gather_uptr{},
...
gather_data[],
gather_relocs[],
cmd.wait_syncpt{},
...
};
This way we will have only a single copy_from_user() for the whole
cmdstream, which should be more efficient to do and nicer from both
userspace and kernel perspectives in regards to forming and parsing the
commands.
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
next prev parent reply other threads:[~2020-06-28 20:36 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-06-23 12:09 [RFC] Host1x/TegraDRM UAPI Mikko Perttunen
2020-06-24 20:55 ` Dmitry Osipenko
2020-06-25 9:17 ` Mikko Perttunen
2020-06-24 22:33 ` Dmitry Osipenko
2020-06-25 9:27 ` Mikko Perttunen
2020-06-25 22:50 ` [RFC] Host1x/TegraDRM UAPI (drm_tegra_submit_relocation) Dmitry Osipenko
2020-06-26 9:01 ` Mikko Perttunen
2020-06-24 23:11 ` [RFC] Host1x/TegraDRM UAPI Dmitry Osipenko
2020-06-25 9:16 ` Mikko Perttunen
2020-06-25 23:24 ` Dmitry Osipenko
2020-06-26 9:05 ` Mikko Perttunen
2020-06-24 23:18 ` Dmitry Osipenko
2020-06-25 0:59 ` Dmitry Osipenko
2020-06-24 23:23 ` Dmitry Osipenko
2020-06-25 9:19 ` Mikko Perttunen
2020-06-25 0:47 ` Dmitry Osipenko
2020-06-25 9:23 ` Mikko Perttunen
2020-06-25 22:47 ` [RFC] Host1x/TegraDRM UAPI (drm_tegra_channel_map) Dmitry Osipenko
2020-06-26 7:34 ` Thierry Reding
2020-06-26 16:35 ` Dmitry Osipenko
2020-06-28 11:16 ` Mikko Perttunen
2020-06-28 22:59 ` Dmitry Osipenko
2020-06-30 10:55 ` Mikko Perttunen
2020-06-30 19:48 ` Dmitry Osipenko
2020-06-26 11:06 ` [RFC] Host1x/TegraDRM UAPI Karol Herbst
2020-06-26 11:13 ` Mikko Perttunen
2020-06-26 11:16 ` Karol Herbst
2020-06-26 11:40 ` Thierry Reding
2020-06-26 13:38 ` Daniel Vetter
2020-06-26 13:59 ` Dmitry Osipenko
2020-06-30 9:09 ` Daniel Vetter
2020-06-27 21:47 ` [RFC] Host1x/TegraDRM UAPI (drm_tegra_submit_syncpt_incr) Dmitry Osipenko
2020-06-28 11:10 ` Mikko Perttunen
2020-06-27 22:32 ` Dmitry Osipenko [this message]
2020-06-28 10:28 ` [RFC] Host1x/TegraDRM UAPI (drm_tegra_submit_command) Mikko Perttunen
2020-06-29 0:00 ` Dmitry Osipenko
2020-06-30 10:40 ` Mikko Perttunen
2020-06-27 23:27 ` [RFC] Host1x/TegraDRM UAPI (sync points) Dmitry Osipenko
2020-06-28 9:44 ` Mikko Perttunen
2020-06-29 2:36 ` Dmitry Osipenko
2020-06-29 10:27 ` Mikko Perttunen
2020-06-29 19:28 ` Dmitry Osipenko
2020-06-29 19:33 ` Dmitry Osipenko
2020-06-29 19:42 ` Dmitry Osipenko
2020-06-30 10:26 ` Mikko Perttunen
2020-07-01 0:22 ` Dmitry Osipenko
2020-07-02 12:10 ` Mikko Perttunen
2020-07-07 11:06 ` Dmitry Osipenko
2020-07-08 10:06 ` Mikko Perttunen
2020-07-09 9:28 ` Dmitry Osipenko
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=f31fbe3b-3fc5-2ece-4c2c-9ff4e99995d6@gmail.com \
--to=digetx@gmail.com \
--cc=airlied@linux.ie \
--cc=bhuntsman@nvidia.com \
--cc=cyndis@kapsi.fi \
--cc=daniel@ffwll.ch \
--cc=dri-devel@lists.freedesktop.org \
--cc=gustavo@padovan.org \
--cc=jonathanh@nvidia.com \
--cc=linux-tegra@vger.kernel.org \
--cc=sumit.semwal@linaro.org \
--cc=talho@nvidia.com \
--cc=thierry.reding@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