Commit 9bb66c179f50 ("drm/i915: Reserve some kernel space per
vm") reduces the available VM space of one page in order to apply
Wa_16018031267 and Wa_16018063123.
This page was reserved indiscrimitely in all platforms even when not needed.
Limit it to DG2 onwards.
Fixes: 9bb66c179f50 ("drm/i915:
On Wed, Mar 13, 2024 at 09:19:48PM +0100, Andi Shyti wrote:
> Hi,
>
> this series does basically two things:
>
> 1. Disables automatic load balancing as adviced by the hardware
>workaround.
>
> 2. Assigns all the CCS slices to one single user engine. The user
>will then be able to query
> If we provide the total GTT size we will have one page that will be contended
> between kernel and userspace and, if userspace is unaware that the page
> belongs to the > kernel, we might step on each other toe.
That's fine, Compute needs to know total GTT size.
Not available GTT size.
> > Lionel, Michal, thoughts?
Compute UMD needs to know exact GTT total size.
Commit 9bb66c179f50 ("drm/i915: Reserve some kernel space per
vm") has reserved an object for kernel space usage.
Userspace, though, needs to know the full address range.
Fixes: 9bb66c179f50 ("drm/i915: Reserve some kernel space per vm")
Signed-off-by: Andi Shyti
Cc: Andrzej Hajda
Cc: Chris Wil
Acked-by: Michal Mrozek
-Original Message-
From: Harrison, John C
Sent: Thursday, March 3, 2022 11:38 PM
To: Intel-GFX@Lists.FreeDesktop.Org
Cc: dri-de...@lists.freedesktop.org; Harrison, John C
; Ceraolo Spurio, Daniele
; Mrozek, Michal
Subject: [PATCH v3 4/4] drm/i915: Improve
> -Original Message-
> From: Tvrtko Ursulin
> Sent: Thursday, March 12, 2020 12:58 PM
> To: Intel-gfx@lists.freedesktop.org
> Cc: Ursulin, Tvrtko ; Chris Wilson wilson.co.uk>; Joonas Lahtinen ; Mrozek,
> Michal ; sta...@vger.kernel.org
> Subject: [PATCH v3]
>> We need confirmation and acks from all three userspace components here.
>> Especially since my impression was some are for and some were against
>> whitelisting this one.
Acked-by: Michal Mrozek
___
Intel-gfx mailing list
Intel-gfx@lists.freedeskto
>>With "disable it completely" you mean disable preemption completely - go
>>lower than thread-group in granularity?
Yes, disable it completely.
Michal
___
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/lis
>>But he asked whether it's possible for Media and OpenCL drivers to also
>>disable mid-thread preemption through the INTERFACE_DESCRIPTOR_DATA, instead
>>of from the >>kernel side, so we could try to experiment with it in the
>>future.
Interface Descriptor setting only switches the preemption
10 matches
Mail list logo