Intel-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
* [Intel-gfx] [PATCH i-g-t 1/2] tests/i915/gem_caching: handle discrete
@ 2022-05-04 15:08 Matthew Auld
  2022-05-04 15:08 ` [Intel-gfx] [PATCH i-g-t 2/2] tests/i915/gem_workarounds: " Matthew Auld
  2022-05-04 15:22 ` [Intel-gfx] [PATCH i-g-t 1/2] tests/i915/gem_caching: " Das, Nirmoy
  0 siblings, 2 replies; 3+ messages in thread
From: Matthew Auld @ 2022-05-04 15:08 UTC (permalink / raw)
  To: igt-dev; +Cc: intel-gfx

Test should still be valid, even if we can't explicitly control the PTE
caching bits, like on discrete, where the caching should already be
enabled by default for system memory objects.

References: https://gitlab.freedesktop.org/drm/intel/-/issues/4873
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Nirmoy Das <nirmoy.das@linux.intel.com>
---
 tests/i915/gem_caching.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/tests/i915/gem_caching.c b/tests/i915/gem_caching.c
index 4e844952..eb0170ab 100644
--- a/tests/i915/gem_caching.c
+++ b/tests/i915/gem_caching.c
@@ -147,7 +147,8 @@ igt_main
 
 		igt_require_gem(data.fd);
 		gem_require_blitter(data.fd);
-		gem_require_caching(data.fd);
+		if (!gem_has_lmem(data.fd))
+			gem_require_caching(data.fd);
 
 		data.devid = intel_get_drm_devid(data.fd);
 		if (IS_GEN2(data.devid)) /* chipset only handles cached -> uncached */
@@ -162,7 +163,8 @@ igt_main
 		scratch_buf = intel_buf_create(data.bops, BO_SIZE/4, 1,
 					       32, 0, I915_TILING_NONE, 0);
 
-		gem_set_caching(data.fd, scratch_buf->handle, 1);
+		if (!gem_has_lmem(data.fd))
+			gem_set_caching(data.fd, scratch_buf->handle, 1);
 
 		staging_buf = intel_buf_create(data.bops, BO_SIZE/4, 1,
 					       32, 0, I915_TILING_NONE, 0);
-- 
2.34.1


^ permalink raw reply related	[flat|nested] 3+ messages in thread

* [Intel-gfx] [PATCH i-g-t 2/2] tests/i915/gem_workarounds: handle discrete
  2022-05-04 15:08 [Intel-gfx] [PATCH i-g-t 1/2] tests/i915/gem_caching: handle discrete Matthew Auld
@ 2022-05-04 15:08 ` Matthew Auld
  2022-05-04 15:22 ` [Intel-gfx] [PATCH i-g-t 1/2] tests/i915/gem_caching: " Das, Nirmoy
  1 sibling, 0 replies; 3+ messages in thread
From: Matthew Auld @ 2022-05-04 15:08 UTC (permalink / raw)
  To: igt-dev; +Cc: intel-gfx

On discrete the object should already be using I915_CACHING_CACHED, by
default, for system memory objects, although we can no longer explicitly
control the PTE caching bits.

References: https://gitlab.freedesktop.org/drm/intel/-/issues/4873
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Nirmoy Das <nirmoy.das@linux.intel.com>
---
 tests/i915/gem_workarounds.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tests/i915/gem_workarounds.c b/tests/i915/gem_workarounds.c
index 3d185127..70967b3f 100644
--- a/tests/i915/gem_workarounds.c
+++ b/tests/i915/gem_workarounds.c
@@ -107,7 +107,8 @@ static int workaround_fail_count(int i915, const intel_ctx_t *ctx)
 
 	memset(obj, 0, sizeof(obj));
 	obj[0].handle = gem_create(i915, result_sz);
-	gem_set_caching(i915, obj[0].handle, I915_CACHING_CACHED);
+	if (!gem_has_lmem(i915))
+		gem_set_caching(i915, obj[0].handle, I915_CACHING_CACHED);
 	obj[1].handle = gem_create(i915, batch_sz);
 	obj[1].relocs_ptr = to_user_pointer(reloc);
 	obj[1].relocation_count = !ahnd ? num_wa_regs : 0;
-- 
2.34.1


^ permalink raw reply related	[flat|nested] 3+ messages in thread

* Re: [Intel-gfx] [PATCH i-g-t 1/2] tests/i915/gem_caching: handle discrete
  2022-05-04 15:08 [Intel-gfx] [PATCH i-g-t 1/2] tests/i915/gem_caching: handle discrete Matthew Auld
  2022-05-04 15:08 ` [Intel-gfx] [PATCH i-g-t 2/2] tests/i915/gem_workarounds: " Matthew Auld
@ 2022-05-04 15:22 ` Das, Nirmoy
  1 sibling, 0 replies; 3+ messages in thread
From: Das, Nirmoy @ 2022-05-04 15:22 UTC (permalink / raw)
  To: Matthew Auld, igt-dev; +Cc: intel-gfx

The series is Reviewed-by: Nirmoy Das <nirmoy.das@intel.com>

On 5/4/2022 5:08 PM, Matthew Auld wrote:
> Test should still be valid, even if we can't explicitly control the PTE
> caching bits, like on discrete, where the caching should already be
> enabled by default for system memory objects.
>
> References: https://gitlab.freedesktop.org/drm/intel/-/issues/4873
> Signed-off-by: Matthew Auld <matthew.auld@intel.com>
> Cc: Nirmoy Das <nirmoy.das@linux.intel.com>
> ---
>   tests/i915/gem_caching.c | 6 ++++--
>   1 file changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/tests/i915/gem_caching.c b/tests/i915/gem_caching.c
> index 4e844952..eb0170ab 100644
> --- a/tests/i915/gem_caching.c
> +++ b/tests/i915/gem_caching.c
> @@ -147,7 +147,8 @@ igt_main
>   
>   		igt_require_gem(data.fd);
>   		gem_require_blitter(data.fd);
> -		gem_require_caching(data.fd);
> +		if (!gem_has_lmem(data.fd))
> +			gem_require_caching(data.fd);
>   
>   		data.devid = intel_get_drm_devid(data.fd);
>   		if (IS_GEN2(data.devid)) /* chipset only handles cached -> uncached */
> @@ -162,7 +163,8 @@ igt_main
>   		scratch_buf = intel_buf_create(data.bops, BO_SIZE/4, 1,
>   					       32, 0, I915_TILING_NONE, 0);
>   
> -		gem_set_caching(data.fd, scratch_buf->handle, 1);
> +		if (!gem_has_lmem(data.fd))
> +			gem_set_caching(data.fd, scratch_buf->handle, 1);
>   
>   		staging_buf = intel_buf_create(data.bops, BO_SIZE/4, 1,
>   					       32, 0, I915_TILING_NONE, 0);

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2022-05-04 15:22 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2022-05-04 15:08 [Intel-gfx] [PATCH i-g-t 1/2] tests/i915/gem_caching: handle discrete Matthew Auld
2022-05-04 15:08 ` [Intel-gfx] [PATCH i-g-t 2/2] tests/i915/gem_workarounds: " Matthew Auld
2022-05-04 15:22 ` [Intel-gfx] [PATCH i-g-t 1/2] tests/i915/gem_caching: " Das, Nirmoy

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox