From: "Christian König via dri-devel" <dri-devel@lists.freedesktop.org>
To: Alex Deucher <alexdeucher@gmail.com>, Rob Herring <robh@kernel.org>
Cc: "Christian König" <ckoenig.leichtzumerken@gmail.com>,
"Marek Vasut" <marex@denx.de>,
"Simon Shields" <simon@lineageos.org>,
lima@lists.freedesktop.org,
"Neil Armstrong" <narmstrong@baylibre.com>,
"Maxime Ripard" <maxime.ripard@bootlin.com>,
dri-devel <dri-devel@lists.freedesktop.org>,
"Vasily Khoruzhick" <anarsoul@gmail.com>,
"David Airlie" <airlied@linux.ie>, "Qiang Yu" <yuq825@gmail.com>,
"Sean Paul" <sean@poorly.run>,
"Andreas Baierl" <ichgeh@imkreisrum.de>,
"Erico Nunes" <nunes.erico@gmail.com>
Subject: Re: [PATCH 2/2] drm/lima: driver for ARM Mali4xx GPUs
Date: Thu, 14 Feb 2019 10:12:15 +0100 [thread overview]
Message-ID: <8be4643d-f6ba-3ba9-e0b1-ec0bd8119deb@gmail.com> (raw)
In-Reply-To: <CADnq5_OSjAqiDXXFfPE9sAQ5vsE4oBj-Y0tFO=jFwA9zpOE_Qw@mail.gmail.com>
Am 14.02.19 um 03:52 schrieb Alex Deucher via dri-devel:
> [SNIP]
>>>>> +static int lima_ioctl_gem_va(struct drm_device *dev, void *data, struct drm_file *file)
>>>>> +{
>>>>> + struct drm_lima_gem_va *args = data;
>>>>> +
>>>>> + switch (args->op) {
>>>>> + case LIMA_VA_OP_MAP:
>>>>> + return lima_gem_va_map(file, args->handle, args->flags, args->va);
>>>>> + case LIMA_VA_OP_UNMAP:
>>>>> + return lima_gem_va_unmap(file, args->handle, args->va);
>>>> These are mapping to GPU VA. Why not do that on GEM object creation or
>>>> import or when the objects are submitted with cmd queue as other
>>>> drivers do?
>>>>
>>>> To put it another way, These ioctls look different than what other
>>>> drivers do. Why do you need to do things differently? My understanding
>>>> is best practice is to map and return the GPU offset when the GEM
>>>> object is created. This is what v3d does. I think Intel is moving to
>>>> that. And panfrost will do that.
>>> I think it would be a good idea to look at the amdgpu driver. This
>>> driver is heavily modeled after it. Basically the GEM VA ioctl allows
>>> userspace to manage per process (per fd really) virtual addresses.
>> Why do you want userspace to manage assigning VAs versus the kernel to
>> do so? Exposing that detail to userspace means the driver must support
>> a per process address space. Letting the kernel assign addresses means
>> it can either be a single address space or be a per process address
>> space. It seems to me more flexible to allow the kernel driver to
>> evolve without that ABI.
> Having it in userspace provides a lot more flexibility and makes it
> easier to support things like unified address space between CPU and
> GPU. I guess it depends on the hw as to what is the right choice.
To summarize we actually have tried this approach with the radeon and it
turned out to be a really bad mistake.
To implement features like partial residential textures and shared
virtual address space you absolutely need userspace to be in charge of
allocating virtual addresses.
Regards,
Christian.
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
next prev parent reply other threads:[~2019-02-14 9:12 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-02-06 13:14 [PATCH 0/2] Lima DRM driver Qiang Yu
2019-02-06 13:14 ` [PATCH 1/2] drm/fourcc: add ARM tiled format modifier Qiang Yu
2019-02-14 15:26 ` Brian Starkey
2019-02-14 15:34 ` Daniel Vetter
2019-02-15 1:48 ` Qiang Yu via dri-devel
2019-02-15 11:30 ` Brian Starkey
2019-02-06 13:14 ` [PATCH 2/2] drm/lima: driver for ARM Mali4xx GPUs Qiang Yu
2019-02-06 19:17 ` Eric Anholt
2019-02-07 8:27 ` Qiang Yu
2019-02-07 19:38 ` Eric Anholt
2019-02-12 15:46 ` Rob Herring via dri-devel
2019-02-12 16:23 ` Alex Deucher via dri-devel
2019-02-12 20:04 ` Rob Herring via dri-devel
2019-02-13 1:13 ` Qiang Yu via dri-devel
2019-02-14 2:52 ` Alex Deucher via dri-devel
2019-02-14 9:12 ` Christian König via dri-devel [this message]
2019-02-14 10:15 ` Daniel Vetter
2019-03-14 20:44 ` Dave Airlie
2019-03-14 21:44 ` Rob Herring
2019-03-14 22:28 ` Eric Anholt
2019-03-15 8:06 ` Koenig, Christian
2019-03-15 16:05 ` Eric Anholt
2019-03-15 16:19 ` Koenig, Christian
2019-02-13 0:56 ` Qiang Yu via dri-devel
2019-02-13 1:46 ` Rob Herring via dri-devel
2019-02-07 9:09 ` [PATCH 0/2] Lima DRM driver Daniel Vetter
2019-02-07 9:39 ` Christian König
2019-02-07 15:33 ` Qiang Yu
2019-02-07 19:14 ` Koenig, Christian
2019-02-07 15:21 ` Qiang Yu
2019-02-07 15:44 ` Alyssa Rosenzweig
2019-02-07 15:51 ` Daniel Vetter
2019-02-11 18:11 ` Rob Herring
2019-02-13 1:00 ` Eric Anholt
2019-02-13 1:44 ` Rob Herring via dri-devel
2019-02-13 7:59 ` Daniel Vetter
2019-02-13 8:35 ` Christian König via dri-devel
2019-02-13 9:38 ` Daniel Vetter
2019-02-13 10:09 ` Christian König via dri-devel
2019-02-26 15:58 ` Daniel Vetter
2019-02-26 16:23 ` Christian König
2019-02-14 21:15 ` Noralf Trønnes
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=8be4643d-f6ba-3ba9-e0b1-ec0bd8119deb@gmail.com \
--to=dri-devel@lists.freedesktop.org \
--cc=airlied@linux.ie \
--cc=alexdeucher@gmail.com \
--cc=anarsoul@gmail.com \
--cc=christian.koenig@amd.com \
--cc=ckoenig.leichtzumerken@gmail.com \
--cc=ichgeh@imkreisrum.de \
--cc=lima@lists.freedesktop.org \
--cc=marex@denx.de \
--cc=maxime.ripard@bootlin.com \
--cc=narmstrong@baylibre.com \
--cc=nunes.erico@gmail.com \
--cc=robh@kernel.org \
--cc=sean@poorly.run \
--cc=simon@lineageos.org \
--cc=yuq825@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;
as well as URLs for NNTP newsgroup(s).