Intel-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Andi Shyti <andi.shyti@linux.intel.com>
To: Jonathan Cavitt <jonathan.cavitt@intel.com>
Cc: chris.p.wilson@linux.intel.com, intel-gfx@lists.freedesktop.org,
	matthew.d.roper@intel.com, nirmoy.das@intel.com
Subject: Re: [Intel-gfx] [PATCH 2/3] drm/i915: Make i915_coherent_map_type GT-centric
Date: Thu, 27 Jul 2023 14:16:40 +0200	[thread overview]
Message-ID: <ZMJgKO9DbXg4cob4@ashyti-mobl2.lan> (raw)
In-Reply-To: <20230726155356.1652979-2-jonathan.cavitt@intel.com>

Hi Jonathan,

On Wed, Jul 26, 2023 at 08:53:55AM -0700, Jonathan Cavitt wrote:
> Refactor i915_coherent_map_type to be GT-centric rather than
> device-centric.  Each GT may require different coherency
> handling due to hardware workarounds.
> 
> Since the function now takes a GT instead of the i915, the function is
> renamed and moved to the gt folder.

[...]

> --- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
> @@ -465,21 +465,6 @@ void *i915_gem_object_pin_map_unlocked(struct drm_i915_gem_object *obj,
>  	return ret;
>  }
>  
> -enum i915_map_type i915_coherent_map_type(struct drm_i915_private *i915,
> -					  struct drm_i915_gem_object *obj,
> -					  bool always_coherent)
> -{
> -	/*
> -	 * Wa_22016122933: always return I915_MAP_WC for MTL
> -	 */
> -	if (i915_gem_object_is_lmem(obj) || IS_METEORLAKE(i915))
> -		return I915_MAP_WC;
> -	if (HAS_LLC(i915) || always_coherent)
> -		return I915_MAP_WB;
> -	else
> -		return I915_MAP_WC;
> -}
> -

[...]

> --- a/drivers/gpu/drm/i915/gt/intel_gt.c
> +++ b/drivers/gpu/drm/i915/gt/intel_gt.c
> @@ -1134,6 +1134,21 @@ void intel_gt_invalidate_tlb(struct intel_gt *gt, u32 seqno)
>  	}
>  }
>  
> +enum i915_map_type intel_gt_coherent_map_type(struct intel_gt *gt,
> +					      struct drm_i915_gem_object *obj,
> +					      bool always_coherent)
> +{
> +	/*
> +	 * Wa_22016122933: always return I915_MAP_WC for MTL
> +	 */
> +	if (i915_gem_object_is_lmem(obj) || IS_METEORLAKE(gt->i915))
> +		return I915_MAP_WC;
> +	if (HAS_LLC(gt->i915) || always_coherent)
> +		return I915_MAP_WB;
> +	else
> +		return I915_MAP_WC;
> +}
> +

yep! That's how it should be! Thanks for moving it.

I'm just thinking whether intel_gt.c is the right place to put
it... I'm anyway OK with it.


> --- a/drivers/gpu/drm/i915/gt/intel_ring.c
> +++ b/drivers/gpu/drm/i915/gt/intel_ring.c
> @@ -13,6 +13,7 @@
>  #include "intel_engine_regs.h"
>  #include "intel_gpu_commands.h"
>  #include "intel_ring.h"
> +#include "intel_gt.h"

mmhhh... pity... this required this extra include because it
needs the prototype of intel_gt_coherent_map_type();

Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>  

Andi

  parent reply	other threads:[~2023-07-27 12:16 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-07-26 15:53 [Intel-gfx] [PATCH 1/3] drm/i915/gt: Simplify shmem_create_from_object map_type selection Jonathan Cavitt
2023-07-26 15:53 ` [Intel-gfx] [PATCH 2/3] drm/i915: Make i915_coherent_map_type GT-centric Jonathan Cavitt
2023-07-27 11:35   ` Tvrtko Ursulin
2023-07-28  1:34     ` Andi Shyti
2023-07-28  8:07       ` Tvrtko Ursulin
2023-07-28  8:18         ` Andi Shyti
2023-07-27 12:16   ` Andi Shyti [this message]
2023-08-01 14:32   ` Andi Shyti
2023-07-26 15:53 ` [Intel-gfx] [PATCH 3/3] drm/i915/gt: Apply workaround 22016122933 correctly Jonathan Cavitt
2023-07-27 12:20   ` Andi Shyti
2023-07-26 16:37 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for series starting with [1/3] drm/i915/gt: Simplify shmem_create_from_object map_type selection Patchwork
2023-07-26 16:37 ` [Intel-gfx] ✗ Fi.CI.SPARSE: " Patchwork
2023-07-26 16:51 ` [Intel-gfx] ✓ Fi.CI.BAT: success " Patchwork
2023-07-26 20:30 ` [Intel-gfx] ✗ Fi.CI.IGT: failure " Patchwork
2023-07-27 11:27 ` [Intel-gfx] [PATCH 1/3] " Tvrtko Ursulin
2023-07-27 12:23 ` Andi Shyti
2023-08-01 12:17 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for series starting with [1/3] drm/i915/gt: Simplify shmem_create_from_object map_type selection (rev2) Patchwork
2023-08-01 12:17 ` [Intel-gfx] ✗ Fi.CI.SPARSE: " Patchwork
2023-08-01 12:31 ` [Intel-gfx] ✓ Fi.CI.BAT: success " Patchwork
2023-08-01 15:38 ` [Intel-gfx] ✗ Fi.CI.IGT: failure " Patchwork
  -- strict thread matches above, loose matches on Subject: below --
2023-08-01 15:32 [Intel-gfx] [PATCH 0/3] drm/i915: Fix Wa_22016122933 implementation Jonathan Cavitt
2023-08-01 15:32 ` [Intel-gfx] [PATCH 2/3] drm/i915: Make i915_coherent_map_type GT-centric Jonathan Cavitt

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=ZMJgKO9DbXg4cob4@ashyti-mobl2.lan \
    --to=andi.shyti@linux.intel.com \
    --cc=chris.p.wilson@linux.intel.com \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=jonathan.cavitt@intel.com \
    --cc=matthew.d.roper@intel.com \
    --cc=nirmoy.das@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