From: "Christian König" <ckoenig.leichtzumerken@gmail.com>
To: "Marek Olšák" <maraeo@gmail.com>
Cc: amd-gfx mailing list <amd-gfx@lists.freedesktop.org>
Subject: Re: [PATCH 2/2] drm/amdgpu: add AMDGPU_INFO_VM_STAT to return GPU VM
Date: Wed, 4 Jan 2023 15:51:18 +0100 [thread overview]
Message-ID: <d1463910-1eab-2dac-a633-812ada011cc4@gmail.com> (raw)
In-Reply-To: <CAAxE2A69e+rHQJP+wHYOxywB0+B4Vp4XsO429euoGE=H-VRsPw@mail.gmail.com>
[-- Attachment #1: Type: text/plain, Size: 3025 bytes --]
Am 04.01.23 um 00:08 schrieb Marek Olšák:
> I see about the access now, but did you even look at the patch?
I did look at the patch, but I haven't fully understood yet what you are
trying to do here.
> Because what the patch does isn't even exposed to common drm code,
> such as the preferred domain and visible VRAM placement, so it can't
> be in fdinfo right now.
>
> Or do you even know what fdinfo contains? Because it contains nothing
> useful. It only has VRAM and GTT usage, which we already have in the
> INFO ioctl, so it has nothing that we need. We mainly need the
> eviction information and visible VRAM information now. Everything else
> is a bonus.
Well the main question is what are you trying to get from that
information? The eviction list for example is completely meaningless to
userspace, that stuff is only temporary and will be cleared on the next
CS again.
What we could expose is the VRAM over-commit value, e.g. how much BOs
which where supposed to be in VRAM are in GTT now. I think that's what
you are looking for here, right?
> Also, it's undesirable to open and parse a text file if we can just
> call an ioctl.
Well I see the reasoning for that, but I also see why other drivers do a
lot of the stuff we have as IOCTL as separate files in sysfs, fdinfo or
debugfs.
Especially repeating all the static information which were already
available under sysfs in the INFO IOCTL was a design mistake as far as I
can see. Just compare what AMDGPU and the KFD code is doing to what for
example i915 is doing.
Same for things like debug information about a process. The fdinfo stuff
can be queried from external tools (gdb, gputop, umr etc...) as well
which makes that interface more preferred.
>
> So do you want me to move it into amdgpu_vm.c? Because you could have
> just said: Let's move it into amdgpu_vm.c. :)
>
> Thanks,
> Marek
>
> On Tue, Jan 3, 2023 at 3:33 AM Christian König
> <ckoenig.leichtzumerken@gmail.com> wrote:
>
> Take a look at /proc/self/fdinfo/$fd.
>
> The Intel guys made that vendor agnostic and are using it within
> their IGT gpu top tool.
>
> Christian.
>
> Am 02.01.23 um 18:57 schrieb Marek Olšák:
>> What are you talking about? Is fdinfo in sysfs? Userspace drivers
>> can't access sysfs.
>>
>> Marek
>>
>> On Mon, Jan 2, 2023, 10:56 Christian König
>> <ckoenig.leichtzumerken@gmail.com> wrote:
>>
>> Well first of all don't mess with the VM internals outside of
>> the VM code.
>>
>> Then why would we want to expose this through the IOCTL
>> interface? We already have this in the fdinfo.
>>
>> Christian.
>>
>> Am 30.12.22 um 23:07 schrieb Marek Olšák:
>>> To give userspace a detailed view about its GPU memory usage
>>> and evictions.
>>> This will help performance investigations.
>>>
>>> Signed-off-by: Marek Olšák <marek.olsak@amd.com>
>>>
>>> The patch is attached.
>>>
>>> Marek
>>
>
[-- Attachment #2: Type: text/html, Size: 6718 bytes --]
next prev parent reply other threads:[~2023-01-04 14:51 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-12-30 22:07 [PATCH 2/2] drm/amdgpu: add AMDGPU_INFO_VM_STAT to return GPU VM Marek Olšák
2022-12-30 22:10 ` Marek Olšák
2023-01-02 15:56 ` Christian König
2023-01-02 17:57 ` Marek Olšák
2023-01-03 8:33 ` Christian König
2023-01-03 23:08 ` Marek Olšák
2023-01-04 14:51 ` Christian König [this message]
2023-01-10 15:28 ` Marek Olšák
2023-01-10 16:23 ` Christian König
2023-01-10 16:55 ` Marek Olšák
2023-01-21 0:45 ` Marek Olšák
2023-01-23 9:31 ` Christian König
2023-01-24 7:37 ` Marek Olšák
2023-01-24 7:58 ` Christian König
2023-01-24 8:13 ` Marek Olšák
2023-01-24 8:27 ` Marek Olšák
2023-01-24 8:29 ` Christian König
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=d1463910-1eab-2dac-a633-812ada011cc4@gmail.com \
--to=ckoenig.leichtzumerken@gmail.com \
--cc=amd-gfx@lists.freedesktop.org \
--cc=maraeo@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