From: Nirmoy Das <nirmoy.das@intel.com>
To: Francois Dugast <francois.dugast@intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>,
<intel-xe@lists.freedesktop.org>,
<joonas.lahtinen@linux.intel.com>, <mateusz.jablonski@intel.com>
Subject: Re: [RFC PATCH 2/2] drm/xe/uapi: Expose wmtp as engine capability
Date: Mon, 15 Jul 2024 10:21:14 +0200 [thread overview]
Message-ID: <eaca3c8c-88b6-4edc-9308-92308fac62b2@intel.com> (raw)
In-Reply-To: <ZpFVZT-OHmcIsm9j@fdugast-desk>
On 7/12/2024 6:10 PM, Francois Dugast wrote:
> Hi,
>
> On Mon, Feb 19, 2024 at 11:09:13AM +0100, Nirmoy Das wrote:
>> On 2/16/2024 8:52 PM, Matt Roper wrote:
>>> On Sat, Feb 03, 2024 at 07:05:34AM +0100, Nirmoy Das wrote:
>>>> Add capabilities to engine struct which can be used to
>>>> expose various capabilities of each engines.
>>>>
>>>> Add wmtp as engine capability which is retrieve from
>>>> GT info.
>>>>
>>>> Take a field from the reserved for that purpose.
>>>>
>>>> Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
>>> We'll need a userspace consumer for the uapi and some simple IGT tests
>>> (e.g., making sure this flag never shows up on the wrong engine types or
>>> platforms), but once those are available,
>> Thanks Matt, I wanted to get initial feedback. I will now sync up with UMD
>> friends to create a pull request for this, and also work on IGT tests.
>>
>>
>>> Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
>>>
>>>
>>> Matt
>>>
>>>> ---
>>>> drivers/gpu/drm/xe/xe_query.c | 5 +++++
>>>> include/uapi/drm/xe_drm.h | 6 ++++--
>>>> 2 files changed, 9 insertions(+), 2 deletions(-)
>>>>
>>>> diff --git a/drivers/gpu/drm/xe/xe_query.c b/drivers/gpu/drm/xe/xe_query.c
>>>> index 7e924faeeea0..10981b46c56f 100644
>>>> --- a/drivers/gpu/drm/xe/xe_query.c
>>>> +++ b/drivers/gpu/drm/xe/xe_query.c
>>>> @@ -215,6 +215,11 @@ static int query_engines(struct xe_device *xe,
>>>> engines->engines[i].instance.pad = 0;
>>>> memset(engines->engines[i].reserved, 0,
>>>> sizeof(engines->engines[i].reserved));
>>>> + if (gt->info.has_wmtp &&
>>>> + (hwe->class == XE_ENGINE_CLASS_COMPUTE ||
>>>> + hwe->class == XE_ENGINE_CLASS_RENDER))
>>>> + engines->engines[i].capabilities |=
>>>> + DRM_XE_ENGINE_CAPABILITY_WMTP;
>>>> i++;
>>>> }
>>>> diff --git a/include/uapi/drm/xe_drm.h b/include/uapi/drm/xe_drm.h
>>>> index 50bbea0992d9..5dac079e64de 100644
>>>> --- a/include/uapi/drm/xe_drm.h
>>>> +++ b/include/uapi/drm/xe_drm.h
>>>> @@ -247,9 +247,11 @@ struct drm_xe_engine_class_instance {
>>>> struct drm_xe_engine {
>>>> /** @instance: The @drm_xe_engine_class_instance */
>>>> struct drm_xe_engine_class_instance instance;
>>>> -
>>>> +#define DRM_XE_ENGINE_CAPABILITY_WMTP BIT(0)
> Where does user space get the BIT() macro definition from? We should
> probably use an explicit value for this constant, as it is done in
> the rest of this header.
I was unaware of it. I will remove the BIT macro.
Thanks,
Nirmoy
>
>>>> + /** @capabilities: Capabilities of this engine. */
>>>> + __u64 capabilities;
>>>> /** @reserved: Reserved */
>>>> - __u64 reserved[3];
>>>> + __u64 reserved[2];
>>>> };
>>>> /**
>>>> --
>>>> 2.42.0
>>>>
next prev parent reply other threads:[~2024-07-15 8:21 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-02-03 6:05 [RFC PATCH 1/2] drm/xe: Add has_wmpt GT info Nirmoy Das
2024-02-03 6:05 ` [RFC PATCH 2/2] drm/xe/uapi: Expose wmtp as engine capability Nirmoy Das
2024-02-16 19:52 ` Matt Roper
2024-02-19 10:09 ` Nirmoy Das
2024-07-12 16:10 ` Francois Dugast
2024-07-15 8:21 ` Nirmoy Das [this message]
2024-07-13 13:55 ` Lucas De Marchi
2024-07-15 8:29 ` Nirmoy Das
2024-02-03 6:22 ` ✓ CI.Patch_applied: success for series starting with [RFC,1/2] drm/xe: Add has_wmpt GT info Patchwork
2024-02-03 6:22 ` ✓ CI.checkpatch: " Patchwork
2024-02-03 6:23 ` ✓ CI.KUnit: " Patchwork
2024-02-03 6:30 ` ✓ CI.Build: " Patchwork
2024-02-03 6:31 ` ✗ CI.Hooks: failure " Patchwork
2024-02-03 6:32 ` ✓ CI.checksparse: success " Patchwork
2024-02-03 6:56 ` ✗ CI.BAT: failure " Patchwork
2024-02-16 18:47 ` [RFC PATCH 1/2] " Matt Roper
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=eaca3c8c-88b6-4edc-9308-92308fac62b2@intel.com \
--to=nirmoy.das@intel.com \
--cc=francois.dugast@intel.com \
--cc=intel-xe@lists.freedesktop.org \
--cc=joonas.lahtinen@linux.intel.com \
--cc=mateusz.jablonski@intel.com \
--cc=matthew.d.roper@intel.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