[PATCH igt 0/8] Non-Intel test suite fixes

[PATCH igt 0/8] Non-Intel test suite fixes

[Ulrich Hecht]

Hi!

I have run the tests on a Renesas R-Car M3-W's DU device, and have found a
number of false negatives that mostly stem from use of Intel-specifics
without checking if that makes sense first. So here's a bunch of fixes for
those, hope they are generic enough for upstreaming.

CU
Uli


Ulrich Hecht (8):
  tests/kms_addfb_basic: skip i915-specific tests on other platforms
  tests/kms_panel_fitting: check for i915 before checking version
  lib/igt_gt: has_gpu_reset(): fix failed assertion on non-i915
    platforms
  lib/igt_gt: check for presence of GPU reset before using it
  tests/kms_plane_lowres: skip i915-specific tests on other platforms
  lib/igt_pm: turn absence of autosuspend_delay_ms from fail to skip
  tests/kms_addfb_basic: size_tests(): reduce test buffer size
  test/kms_addfb_basic: tolerate absence of 8-bit format

 lib/igt_gt.c              | 24 ++++++++++++++----------
 lib/igt_pm.c              |  2 +-
 tests/kms_addfb_basic.c   | 33 ++++++++++++++++++---------------
 tests/kms_panel_fitting.c |  1 +
 tests/kms_plane_lowres.c  |  1 +
 5 files changed, 35 insertions(+), 26 deletions(-)

-- 
2.7.4

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH igt 1/8] tests/kms_addfb_basic: skip i915-specific tests on other platforms

[Ulrich Hecht]

Add is_i915_device() requirement to tests using Intel-specific APIs.

Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>
---
 tests/kms_addfb_basic.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/tests/kms_addfb_basic.c b/tests/kms_addfb_basic.c
index 7d8852f..cf9ba37 100644
--- a/tests/kms_addfb_basic.c
+++ b/tests/kms_addfb_basic.c
@@ -104,6 +104,7 @@ static void invalid_tests(int fd)
 	}
 
 	igt_subtest("clobberred-modifier") {
+		igt_require(is_i915_device(fd));
 		f.flags = 0;
 		f.modifier[0] = 0;
 		gem_set_tiling(fd, gem_bo, I915_TILING_X, 512*4);
@@ -318,6 +319,7 @@ static void size_tests(int fd)
 	}
 
 	igt_subtest("bo-too-small-due-to-tiling") {
+		igt_require(is_i915_device(fd));
 		gem_set_tiling(fd, gem_bo_small, I915_TILING_X, 1024*4);
 		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f) == -1 &&
 			   errno == EINVAL);
@@ -369,6 +371,7 @@ static void addfb25_tests(int fd)
 
 	igt_subtest_group {
 		igt_fixture {
+			igt_require(is_i915_device(fd));
 			gem_set_tiling(fd, gem_bo, I915_TILING_X, 1024*4);
 			igt_require_fb_modifiers(fd);
 		}
-- 
2.7.4

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH igt 1/8] tests/kms_addfb_basic: skip i915-specific tests on other platforms

[Daniel Vetter]

On Thu, Mar 15, 2018 at 03:45:37PM +0100, Ulrich Hecht wrote:
> Add is_i915_device() requirement to tests using Intel-specific APIs.
> 
> Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>

Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>

> ---
>  tests/kms_addfb_basic.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/tests/kms_addfb_basic.c b/tests/kms_addfb_basic.c
> index 7d8852f..cf9ba37 100644
> --- a/tests/kms_addfb_basic.c
> +++ b/tests/kms_addfb_basic.c
> @@ -104,6 +104,7 @@ static void invalid_tests(int fd)
>  	}
>  
>  	igt_subtest("clobberred-modifier") {
> +		igt_require(is_i915_device(fd));
>  		f.flags = 0;
>  		f.modifier[0] = 0;
>  		gem_set_tiling(fd, gem_bo, I915_TILING_X, 512*4);
> @@ -318,6 +319,7 @@ static void size_tests(int fd)
>  	}
>  
>  	igt_subtest("bo-too-small-due-to-tiling") {
> +		igt_require(is_i915_device(fd));
>  		gem_set_tiling(fd, gem_bo_small, I915_TILING_X, 1024*4);
>  		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f) == -1 &&
>  			   errno == EINVAL);
> @@ -369,6 +371,7 @@ static void addfb25_tests(int fd)
>  
>  	igt_subtest_group {
>  		igt_fixture {
> +			igt_require(is_i915_device(fd));
>  			gem_set_tiling(fd, gem_bo, I915_TILING_X, 1024*4);
>  			igt_require_fb_modifiers(fd);
>  		}
> -- 
> 2.7.4
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH igt 1/8] tests/kms_addfb_basic: skip i915-specific tests on other platforms

[Laurent Pinchart]

Hi Ulrich,

Thank you for the patch.

On Thursday, 15 March 2018 16:45:37 EET Ulrich Hecht wrote:
> Add is_i915_device() requirement to tests using Intel-specific APIs.
> 
> Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>

Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>

> ---
>  tests/kms_addfb_basic.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/tests/kms_addfb_basic.c b/tests/kms_addfb_basic.c
> index 7d8852f..cf9ba37 100644
> --- a/tests/kms_addfb_basic.c
> +++ b/tests/kms_addfb_basic.c
> @@ -104,6 +104,7 @@ static void invalid_tests(int fd)
>  	}
> 
>  	igt_subtest("clobberred-modifier") {
> +		igt_require(is_i915_device(fd));
>  		f.flags = 0;
>  		f.modifier[0] = 0;
>  		gem_set_tiling(fd, gem_bo, I915_TILING_X, 512*4);
> @@ -318,6 +319,7 @@ static void size_tests(int fd)
>  	}
> 
>  	igt_subtest("bo-too-small-due-to-tiling") {
> +		igt_require(is_i915_device(fd));
>  		gem_set_tiling(fd, gem_bo_small, I915_TILING_X, 1024*4);
>  		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f) == -1 &&
>  			   errno == EINVAL);
> @@ -369,6 +371,7 @@ static void addfb25_tests(int fd)
> 
>  	igt_subtest_group {
>  		igt_fixture {
> +			igt_require(is_i915_device(fd));
>  			gem_set_tiling(fd, gem_bo, I915_TILING_X, 1024*4);
>  			igt_require_fb_modifiers(fd);
>  		}


-- 
Regards,

Laurent Pinchart

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH igt 2/8] tests/kms_panel_fitting: check for i915 before checking version

[Ulrich Hecht]

Fixes false negatives on non-i915 platforms.

Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>
---
 tests/kms_panel_fitting.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tests/kms_panel_fitting.c b/tests/kms_panel_fitting.c
index b3cee22..6d0be50 100644
--- a/tests/kms_panel_fitting.c
+++ b/tests/kms_panel_fitting.c
@@ -243,6 +243,7 @@ static void test_atomic_fastset(igt_display_t *display)
 		igt_set_module_param_int("fastboot", 1);
 
 	igt_require(display->is_atomic);
+	igt_require(is_i915_device(display->drm_fd));
 	igt_require(intel_gen(intel_get_drm_devid(display->drm_fd)) >= 5);
 
 	for_each_pipe_with_valid_output(display, pipe, output) {
-- 
2.7.4

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH igt 2/8] tests/kms_panel_fitting: check for i915 before checking version

[Daniel Vetter]

On Thu, Mar 15, 2018 at 03:45:38PM +0100, Ulrich Hecht wrote:
> Fixes false negatives on non-i915 platforms.
> 
> Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>

Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>

> ---
>  tests/kms_panel_fitting.c | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/tests/kms_panel_fitting.c b/tests/kms_panel_fitting.c
> index b3cee22..6d0be50 100644
> --- a/tests/kms_panel_fitting.c
> +++ b/tests/kms_panel_fitting.c
> @@ -243,6 +243,7 @@ static void test_atomic_fastset(igt_display_t *display)
>  		igt_set_module_param_int("fastboot", 1);
>  
>  	igt_require(display->is_atomic);
> +	igt_require(is_i915_device(display->drm_fd));
>  	igt_require(intel_gen(intel_get_drm_devid(display->drm_fd)) >= 5);
>  
>  	for_each_pipe_with_valid_output(display, pipe, output) {
> -- 
> 2.7.4
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH igt 2/8] tests/kms_panel_fitting: check for i915 before checking version

[Laurent Pinchart]

Hi Ulrich,

Thank you for the patch.

On Thursday, 15 March 2018 16:45:38 EET Ulrich Hecht wrote:
> Fixes false negatives on non-i915 platforms.
> 
> Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>
> ---
>  tests/kms_panel_fitting.c | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/tests/kms_panel_fitting.c b/tests/kms_panel_fitting.c
> index b3cee22..6d0be50 100644
> --- a/tests/kms_panel_fitting.c
> +++ b/tests/kms_panel_fitting.c
> @@ -243,6 +243,7 @@ static void test_atomic_fastset(igt_display_t *display)
>  		igt_set_module_param_int("fastboot", 1);
> 
>  	igt_require(display->is_atomic);
> +	igt_require(is_i915_device(display->drm_fd));
>  	igt_require(intel_gen(intel_get_drm_devid(display->drm_fd)) >= 5);

I'm fine with this patch as a quick fix, but what in this test is Intel-
specific ? Can't we replace the Intel generation check with a different 
feature check ?

>  	for_each_pipe_with_valid_output(display, pipe, output) {

-- 
Regards,

Laurent Pinchart

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH igt 2/8] tests/kms_panel_fitting: check for i915 before checking version

[Daniel Vetter]

On Tue, Mar 20, 2018 at 01:24:09PM +0200, Laurent Pinchart wrote:
> Hi Ulrich,
> 
> Thank you for the patch.
> 
> On Thursday, 15 March 2018 16:45:38 EET Ulrich Hecht wrote:
> > Fixes false negatives on non-i915 platforms.
> > 
> > Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>
> > ---
> >  tests/kms_panel_fitting.c | 1 +
> >  1 file changed, 1 insertion(+)
> > 
> > diff --git a/tests/kms_panel_fitting.c b/tests/kms_panel_fitting.c
> > index b3cee22..6d0be50 100644
> > --- a/tests/kms_panel_fitting.c
> > +++ b/tests/kms_panel_fitting.c
> > @@ -243,6 +243,7 @@ static void test_atomic_fastset(igt_display_t *display)
> >  		igt_set_module_param_int("fastboot", 1);
> > 
> >  	igt_require(display->is_atomic);
> > +	igt_require(is_i915_device(display->drm_fd));
> >  	igt_require(intel_gen(intel_get_drm_devid(display->drm_fd)) >= 5);
> 
> I'm fine with this patch as a quick fix, but what in this test is Intel-
> specific ? Can't we replace the Intel generation check with a different 
> feature check ?

There's some checks in there that we can do certain panel fitter mode
changes without a modeset (throught ALLOW_MODESET for atomic commits).
That's 100% encoding intel hw constraints: Our hw can disable the panel
fitter without a modest (so going from upscaled -> native resolution), but
not any of the other changes (native -> upscaled or 2 different upscaled
versions).
-Daniel

> 
> >  	for_each_pipe_with_valid_output(display, pipe, output) {
> 
> -- 
> Regards,
> 
> Laurent Pinchart
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH igt 2/8] tests/kms_panel_fitting: check for i915 before checking version

[Laurent Pinchart]

Hi Daniel,

On Wednesday, 21 March 2018 10:34:33 EET Daniel Vetter wrote:
> On Tue, Mar 20, 2018 at 01:24:09PM +0200, Laurent Pinchart wrote:
> > Hi Ulrich,
> > 
> > Thank you for the patch.
> > 
> > On Thursday, 15 March 2018 16:45:38 EET Ulrich Hecht wrote:
> > > Fixes false negatives on non-i915 platforms.
> > > 
> > > Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>
> > > ---
> > > 
> > >  tests/kms_panel_fitting.c | 1 +
> > >  1 file changed, 1 insertion(+)
> > > 
> > > diff --git a/tests/kms_panel_fitting.c b/tests/kms_panel_fitting.c
> > > index b3cee22..6d0be50 100644
> > > --- a/tests/kms_panel_fitting.c
> > > +++ b/tests/kms_panel_fitting.c
> > > @@ -243,6 +243,7 @@ static void test_atomic_fastset(igt_display_t
> > > *display)
> > > 
> > >  		igt_set_module_param_int("fastboot", 1);
> > >  	
> > >  	igt_require(display->is_atomic);
> > > 
> > > +	igt_require(is_i915_device(display->drm_fd));
> > > 
> > >  	igt_require(intel_gen(intel_get_drm_devid(display->drm_fd)) >= 5);
> > 
> > I'm fine with this patch as a quick fix, but what in this test is Intel-
> > specific ? Can't we replace the Intel generation check with a different
> > feature check ?
> 
> There's some checks in there that we can do certain panel fitter mode
> changes without a modeset (throught ALLOW_MODESET for atomic commits).
> That's 100% encoding intel hw constraints: Our hw can disable the panel
> fitter without a modest (so going from upscaled -> native resolution), but
> not any of the other changes (native -> upscaled or 2 different upscaled
> versions).

OK, thank you for the information.

Would it make sense to rename the Intel-specific test files to start with i915 
(or any other Intel prefix) to make this clear ?

> > >  	for_each_pipe_with_valid_output(display, pipe, output) {

-- 
Regards,

Laurent Pinchart

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH igt 2/8] tests/kms_panel_fitting: check for i915 before checking version

[Daniel Vetter]

On Wed, Mar 21, 2018 at 10:52:19AM +0200, Laurent Pinchart wrote:
> Hi Daniel,
> 
> On Wednesday, 21 March 2018 10:34:33 EET Daniel Vetter wrote:
> > On Tue, Mar 20, 2018 at 01:24:09PM +0200, Laurent Pinchart wrote:
> > > Hi Ulrich,
> > > 
> > > Thank you for the patch.
> > > 
> > > On Thursday, 15 March 2018 16:45:38 EET Ulrich Hecht wrote:
> > > > Fixes false negatives on non-i915 platforms.
> > > > 
> > > > Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>
> > > > ---
> > > > 
> > > >  tests/kms_panel_fitting.c | 1 +
> > > >  1 file changed, 1 insertion(+)
> > > > 
> > > > diff --git a/tests/kms_panel_fitting.c b/tests/kms_panel_fitting.c
> > > > index b3cee22..6d0be50 100644
> > > > --- a/tests/kms_panel_fitting.c
> > > > +++ b/tests/kms_panel_fitting.c
> > > > @@ -243,6 +243,7 @@ static void test_atomic_fastset(igt_display_t
> > > > *display)
> > > > 
> > > >  		igt_set_module_param_int("fastboot", 1);
> > > >  	
> > > >  	igt_require(display->is_atomic);
> > > > 
> > > > +	igt_require(is_i915_device(display->drm_fd));
> > > > 
> > > >  	igt_require(intel_gen(intel_get_drm_devid(display->drm_fd)) >= 5);
> > > 
> > > I'm fine with this patch as a quick fix, but what in this test is Intel-
> > > specific ? Can't we replace the Intel generation check with a different
> > > feature check ?
> > 
> > There's some checks in there that we can do certain panel fitter mode
> > changes without a modeset (throught ALLOW_MODESET for atomic commits).
> > That's 100% encoding intel hw constraints: Our hw can disable the panel
> > fitter without a modest (so going from upscaled -> native resolution), but
> > not any of the other changes (native -> upscaled or 2 different upscaled
> > versions).
> 
> OK, thank you for the information.
> 
> Would it make sense to rename the Intel-specific test files to start with i915 
> (or any other Intel prefix) to make this clear ?

We prefer to not rename test names too much, fairly disruptive to our CI.
We're working on some new infrastructure which should make this easier at
least.

But yeah if you bring a qualition, then we can make this happen. Need to
coordinate with igt maintainers on #intel-gfx (since we also need to pull
in CI folks, which hang out there too).
-Daniel

> 
> > > >  	for_each_pipe_with_valid_output(display, pipe, output) {
> 
> -- 
> Regards,
> 
> Laurent Pinchart
> 

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH igt 3/8] lib/igt_gt: has_gpu_reset(): fix failed assertion on non-i915 platforms

[Ulrich Hecht]

Checks if we have an i915 device before using intel_get_drm_devid().

Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>
---
 lib/igt_gt.c | 19 +++++++++++--------
 1 file changed, 11 insertions(+), 8 deletions(-)

diff --git a/lib/igt_gt.c b/lib/igt_gt.c
index e630550..9cb07c2 100644
--- a/lib/igt_gt.c
+++ b/lib/igt_gt.c
@@ -59,14 +59,17 @@ static bool has_gpu_reset(int fd)
 		struct drm_i915_getparam gp;
 		int val = 0;
 
-		memset(&gp, 0, sizeof(gp));
-		gp.param = 35; /* HAS_GPU_RESET */
-		gp.value = &val;
-
-		if (ioctl(fd, DRM_IOCTL_I915_GETPARAM, &gp))
-			once = intel_gen(intel_get_drm_devid(fd)) >= 5;
-		else
-			once = val > 0;
+		if (is_i915_device(fd)) {
+			memset(&gp, 0, sizeof(gp));
+			gp.param = 35; /* HAS_GPU_RESET */
+			gp.value = &val;
+
+			if (ioctl(fd, DRM_IOCTL_I915_GETPARAM, &gp))
+				once = intel_gen(intel_get_drm_devid(fd)) >= 5;
+			else
+				once = val > 0;
+		} else
+			once = 0;
 	}
 	return once;
 }
-- 
2.7.4

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH igt 3/8] lib/igt_gt: has_gpu_reset(): fix failed assertion on non-i915 platforms

[Ville Syrjälä]

On Thu, Mar 15, 2018 at 03:45:39PM +0100, Ulrich Hecht wrote:
> Checks if we have an i915 device before using intel_get_drm_devid().
> 
> Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>
> ---
>  lib/igt_gt.c | 19 +++++++++++--------
>  1 file changed, 11 insertions(+), 8 deletions(-)
> 
> diff --git a/lib/igt_gt.c b/lib/igt_gt.c
> index e630550..9cb07c2 100644
> --- a/lib/igt_gt.c
> +++ b/lib/igt_gt.c

I would think igt_gt as a whole is pretty much i915 specific.
So feels to me like we should not have gotten this deep when
using another driver.

> @@ -59,14 +59,17 @@ static bool has_gpu_reset(int fd)
>  		struct drm_i915_getparam gp;
>  		int val = 0;
>  
> -		memset(&gp, 0, sizeof(gp));
> -		gp.param = 35; /* HAS_GPU_RESET */
> -		gp.value = &val;
> -
> -		if (ioctl(fd, DRM_IOCTL_I915_GETPARAM, &gp))
> -			once = intel_gen(intel_get_drm_devid(fd)) >= 5;
> -		else
> -			once = val > 0;
> +		if (is_i915_device(fd)) {
> +			memset(&gp, 0, sizeof(gp));
> +			gp.param = 35; /* HAS_GPU_RESET */
> +			gp.value = &val;
> +
> +			if (ioctl(fd, DRM_IOCTL_I915_GETPARAM, &gp))
> +				once = intel_gen(intel_get_drm_devid(fd)) >= 5;
> +			else
> +				once = val > 0;
> +		} else
> +			once = 0;
>  	}
>  	return once;
>  }
> -- 
> 2.7.4
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Ville Syrjälä
Intel OTC
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH igt 3/8] lib/igt_gt: has_gpu_reset(): fix failed assertion on non-i915 platforms

[Daniel Vetter]

On Thu, Mar 15, 2018 at 07:28:53PM +0200, Ville Syrjälä wrote:
> On Thu, Mar 15, 2018 at 03:45:39PM +0100, Ulrich Hecht wrote:
> > Checks if we have an i915 device before using intel_get_drm_devid().
> > 
> > Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>
> > ---
> >  lib/igt_gt.c | 19 +++++++++++--------
> >  1 file changed, 11 insertions(+), 8 deletions(-)
> > 
> > diff --git a/lib/igt_gt.c b/lib/igt_gt.c
> > index e630550..9cb07c2 100644
> > --- a/lib/igt_gt.c
> > +++ b/lib/igt_gt.c
> 
> I would think igt_gt as a whole is pretty much i915 specific.
> So feels to me like we should not have gotten this deep when
> using another driver.

Yeah all the gt stuff should be protected with igt_require(i915).
-Daniel

> 
> > @@ -59,14 +59,17 @@ static bool has_gpu_reset(int fd)
> >  		struct drm_i915_getparam gp;
> >  		int val = 0;
> >  
> > -		memset(&gp, 0, sizeof(gp));
> > -		gp.param = 35; /* HAS_GPU_RESET */
> > -		gp.value = &val;
> > -
> > -		if (ioctl(fd, DRM_IOCTL_I915_GETPARAM, &gp))
> > -			once = intel_gen(intel_get_drm_devid(fd)) >= 5;
> > -		else
> > -			once = val > 0;
> > +		if (is_i915_device(fd)) {
> > +			memset(&gp, 0, sizeof(gp));
> > +			gp.param = 35; /* HAS_GPU_RESET */
> > +			gp.value = &val;
> > +
> > +			if (ioctl(fd, DRM_IOCTL_I915_GETPARAM, &gp))
> > +				once = intel_gen(intel_get_drm_devid(fd)) >= 5;
> > +			else
> > +				once = val > 0;
> > +		} else
> > +			once = 0;
> >  	}
> >  	return once;
> >  }
> > -- 
> > 2.7.4
> > 
> > _______________________________________________
> > Intel-gfx mailing list
> > Intel-gfx@lists.freedesktop.org
> > https://lists.freedesktop.org/mailman/listinfo/intel-gfx
> 
> -- 
> Ville Syrjälä
> Intel OTC
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH igt 4/8] lib/igt_gt: check for presence of GPU reset before using it

[Ulrich Hecht]

Fixes failed assertion on non-i915 devices.

Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>
---
 lib/igt_gt.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/lib/igt_gt.c b/lib/igt_gt.c
index 9cb07c2..825ea52 100644
--- a/lib/igt_gt.c
+++ b/lib/igt_gt.c
@@ -166,14 +166,15 @@ igt_hang_t igt_allow_hang(int fd, unsigned ctx, unsigned flags)
 	struct drm_i915_gem_context_param param;
 	unsigned ban;
 
+	if (!igt_check_boolean_env_var("IGT_HANG_WITHOUT_RESET", false))
+		igt_require(has_gpu_reset(fd));
+
 	igt_assert(igt_sysfs_set_parameter
 		   (fd, "reset", "%d", INT_MAX /* any reset method */));
 
 	if (!igt_check_boolean_env_var("IGT_HANG", true))
 		igt_skip("hang injection disabled by user");
 	gem_context_require_bannable(fd);
-	if (!igt_check_boolean_env_var("IGT_HANG_WITHOUT_RESET", false))
-		igt_require(has_gpu_reset(fd));
 
 	param.ctx_id = ctx;
 	param.size = 0;
-- 
2.7.4

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH igt 5/8] tests/kms_plane_lowres: skip i915-specific tests on other platforms

[Ulrich Hecht]

Add is_i915_device() requirement to tests using Intel-specific APIs.

Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>
---
 tests/kms_plane_lowres.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tests/kms_plane_lowres.c b/tests/kms_plane_lowres.c
index d1e4b3c..8fc7654 100644
--- a/tests/kms_plane_lowres.c
+++ b/tests/kms_plane_lowres.c
@@ -270,6 +270,7 @@ run_tests_for_pipe(data_t *data, enum pipe pipe)
 		igt_skip_on(pipe >= data->display.n_pipes);
 
 		igt_display_require_output_on_pipe(&data->display, pipe);
+		igt_require(is_i915_device(data->drm_fd));
 	}
 
 	igt_subtest_f("pipe-%s-tiling-none",
-- 
2.7.4

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH igt 5/8] tests/kms_plane_lowres: skip i915-specific tests on other platforms

[Ville Syrjälä]

On Thu, Mar 15, 2018 at 03:45:41PM +0100, Ulrich Hecht wrote:
> Add is_i915_device() requirement to tests using Intel-specific APIs.
> 
> Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>
> ---
>  tests/kms_plane_lowres.c | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/tests/kms_plane_lowres.c b/tests/kms_plane_lowres.c
> index d1e4b3c..8fc7654 100644
> --- a/tests/kms_plane_lowres.c
> +++ b/tests/kms_plane_lowres.c
> @@ -270,6 +270,7 @@ run_tests_for_pipe(data_t *data, enum pipe pipe)
>  		igt_skip_on(pipe >= data->display.n_pipes);
>  
>  		igt_display_require_output_on_pipe(&data->display, pipe);
> +		igt_require(is_i915_device(data->drm_fd));

What's the i915 specific thing here? The tiling formats? You should
still be able to do the linear tests.

We probably want to utilize https://patchwork.freedesktop.org/patch/210341/
to skip any test that is trying to use an unsupported modifier. Looks
like I didn't account for drivers that don't support blobifiers.
Shouldn't be too difficult to fix that up though.

>  	}
>  
>  	igt_subtest_f("pipe-%s-tiling-none",
> -- 
> 2.7.4
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Ville Syrjälä
Intel OTC
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH igt 6/8] lib/igt_pm: turn absence of autosuspend_delay_ms from fail to skip

[Ulrich Hecht]

Fixes false negatives on everything that doesn't happen to be at a
specific hard-coded sysfs path...

Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>
---
 lib/igt_pm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/lib/igt_pm.c b/lib/igt_pm.c
index 5bf5b2e..641157b 100644
--- a/lib/igt_pm.c
+++ b/lib/igt_pm.c
@@ -262,7 +262,7 @@ bool igt_setup_runtime_pm(void)
 	 * suite goes faster and we have a higher probability of triggering race
 	 * conditions. */
 	fd = open(POWER_DIR "/autosuspend_delay_ms", O_WRONLY);
-	igt_assert_f(fd >= 0,
+	igt_require_f(fd >= 0,
 		     "Can't open " POWER_DIR "/autosuspend_delay_ms\n");
 
 	/* If we fail to write to the file, it means this system doesn't support
-- 
2.7.4

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH igt 6/8] lib/igt_pm: turn absence of autosuspend_delay_ms from fail to skip

[Ville Syrjälä]

On Thu, Mar 15, 2018 at 03:45:42PM +0100, Ulrich Hecht wrote:
> Fixes false negatives on everything that doesn't happen to be at a
> specific hard-coded sysfs path...
> 
> Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>
> ---
>  lib/igt_pm.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/lib/igt_pm.c b/lib/igt_pm.c
> index 5bf5b2e..641157b 100644
> --- a/lib/igt_pm.c
> +++ b/lib/igt_pm.c
> @@ -262,7 +262,7 @@ bool igt_setup_runtime_pm(void)
>  	 * suite goes faster and we have a higher probability of triggering race
>  	 * conditions. */
>  	fd = open(POWER_DIR "/autosuspend_delay_ms", O_WRONLY);

The hardocded path should probably go then.

igt_sysfs_path() + "/device/power" looks like it should dtrt.
Would need to plumb the fd down though. Hopefully every caller
has it handy.

> -	igt_assert_f(fd >= 0,
> +	igt_require_f(fd >= 0,
>  		     "Can't open " POWER_DIR "/autosuspend_delay_ms\n");
>  
>  	/* If we fail to write to the file, it means this system doesn't support
> -- 
> 2.7.4
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Ville Syrjälä
Intel OTC
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH igt 7/8] tests/kms_addfb_basic: size_tests(): reduce test buffer size

[Ulrich Hecht]

Fixes fails on low-memory devices.

Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>
---
 tests/kms_addfb_basic.c | 26 +++++++++++++-------------
 1 file changed, 13 insertions(+), 13 deletions(-)

diff --git a/tests/kms_addfb_basic.c b/tests/kms_addfb_basic.c
index cf9ba37..d1da718 100644
--- a/tests/kms_addfb_basic.c
+++ b/tests/kms_addfb_basic.c
@@ -238,26 +238,26 @@ static void size_tests(int fd)
 	struct drm_mode_fb_cmd2 f_16 = {};
 	struct drm_mode_fb_cmd2 f_8 = {};
 
-	f.width = 1024;
-	f.height = 1024;
+	f.width = 512;
+	f.height = 512;
 	f.pixel_format = DRM_FORMAT_XRGB8888;
-	f.pitches[0] = 1024*4;
+	f.pitches[0] = 512*4;
 
-	f_16.width = 1024;
-	f_16.height = 1024*2;
+	f_16.width = 512;
+	f_16.height = 512*2;
 	f_16.pixel_format = DRM_FORMAT_RGB565;
-	f_16.pitches[0] = 1024*2;
+	f_16.pitches[0] = 512*2;
 
-	f_8.width = 1024*2;
-	f_8.height = 1024*2;
+	f_8.width = 512*2;
+	f_8.height = 512*2;
 	f_8.pixel_format = DRM_FORMAT_C8;
-	f_8.pitches[0] = 1024*2;
+	f_8.pitches[0] = 512*2;
 
 	igt_fixture {
-		gem_bo = igt_create_bo_with_dimensions(fd, 1024, 1024,
+		gem_bo = igt_create_bo_with_dimensions(fd, 512, 512,
 			DRM_FORMAT_XRGB8888, 0, 0, NULL, NULL, NULL);
 		igt_assert(gem_bo);
-		gem_bo_small = igt_create_bo_with_dimensions(fd, 1024, 1023,
+		gem_bo_small = igt_create_bo_with_dimensions(fd, 512, 511,
 			DRM_FORMAT_XRGB8888, 0, 0, NULL, NULL, NULL);
 		igt_assert(gem_bo_small);
 	}
@@ -311,7 +311,7 @@ static void size_tests(int fd)
 	}
 
 	/* Just to check that the parameters would work. */
-	f.height = 1020;
+	f.height = 510;
 	igt_subtest("small-bo") {
 		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f) == 0);
 		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_RMFB, &f.fb_id) == 0);
@@ -320,7 +320,7 @@ static void size_tests(int fd)
 
 	igt_subtest("bo-too-small-due-to-tiling") {
 		igt_require(is_i915_device(fd));
-		gem_set_tiling(fd, gem_bo_small, I915_TILING_X, 1024*4);
+		gem_set_tiling(fd, gem_bo_small, I915_TILING_X, 512*4);
 		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f) == -1 &&
 			   errno == EINVAL);
 	}
-- 
2.7.4

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH igt 7/8] tests/kms_addfb_basic: size_tests(): reduce test buffer size

[Daniel Vetter]

On Thu, Mar 15, 2018 at 03:45:43PM +0100, Ulrich Hecht wrote:
> Fixes fails on low-memory devices.
> 
> Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>

That's really low memory ... how low are we talking about here? Even 15
year old intel integrated gpus can do 4M buffers (just barely).

Just want to make sure we're not missing anything here, patch itself looks
ok.
-Daniel

> ---
>  tests/kms_addfb_basic.c | 26 +++++++++++++-------------
>  1 file changed, 13 insertions(+), 13 deletions(-)
> 
> diff --git a/tests/kms_addfb_basic.c b/tests/kms_addfb_basic.c
> index cf9ba37..d1da718 100644
> --- a/tests/kms_addfb_basic.c
> +++ b/tests/kms_addfb_basic.c
> @@ -238,26 +238,26 @@ static void size_tests(int fd)
>  	struct drm_mode_fb_cmd2 f_16 = {};
>  	struct drm_mode_fb_cmd2 f_8 = {};
>  
> -	f.width = 1024;
> -	f.height = 1024;
> +	f.width = 512;
> +	f.height = 512;
>  	f.pixel_format = DRM_FORMAT_XRGB8888;
> -	f.pitches[0] = 1024*4;
> +	f.pitches[0] = 512*4;
>  
> -	f_16.width = 1024;
> -	f_16.height = 1024*2;
> +	f_16.width = 512;
> +	f_16.height = 512*2;
>  	f_16.pixel_format = DRM_FORMAT_RGB565;
> -	f_16.pitches[0] = 1024*2;
> +	f_16.pitches[0] = 512*2;
>  
> -	f_8.width = 1024*2;
> -	f_8.height = 1024*2;
> +	f_8.width = 512*2;
> +	f_8.height = 512*2;
>  	f_8.pixel_format = DRM_FORMAT_C8;
> -	f_8.pitches[0] = 1024*2;
> +	f_8.pitches[0] = 512*2;
>  
>  	igt_fixture {
> -		gem_bo = igt_create_bo_with_dimensions(fd, 1024, 1024,
> +		gem_bo = igt_create_bo_with_dimensions(fd, 512, 512,
>  			DRM_FORMAT_XRGB8888, 0, 0, NULL, NULL, NULL);
>  		igt_assert(gem_bo);
> -		gem_bo_small = igt_create_bo_with_dimensions(fd, 1024, 1023,
> +		gem_bo_small = igt_create_bo_with_dimensions(fd, 512, 511,
>  			DRM_FORMAT_XRGB8888, 0, 0, NULL, NULL, NULL);
>  		igt_assert(gem_bo_small);
>  	}
> @@ -311,7 +311,7 @@ static void size_tests(int fd)
>  	}
>  
>  	/* Just to check that the parameters would work. */
> -	f.height = 1020;
> +	f.height = 510;
>  	igt_subtest("small-bo") {
>  		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f) == 0);
>  		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_RMFB, &f.fb_id) == 0);
> @@ -320,7 +320,7 @@ static void size_tests(int fd)
>  
>  	igt_subtest("bo-too-small-due-to-tiling") {
>  		igt_require(is_i915_device(fd));
> -		gem_set_tiling(fd, gem_bo_small, I915_TILING_X, 1024*4);
> +		gem_set_tiling(fd, gem_bo_small, I915_TILING_X, 512*4);
>  		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f) == -1 &&
>  			   errno == EINVAL);
>  	}
> -- 
> 2.7.4
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH igt 8/8] test/kms_addfb_basic: tolerate absence of 8-bit format

[Ulrich Hecht]

Ignores failure to add DRM_FORMAT_C8 frame buffer; some devices do not
support any 8-bit format.

Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>
---
 tests/kms_addfb_basic.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/kms_addfb_basic.c b/tests/kms_addfb_basic.c
index d1da718..db79827 100644
--- a/tests/kms_addfb_basic.c
+++ b/tests/kms_addfb_basic.c
@@ -273,8 +273,8 @@ static void size_tests(int fd)
 		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f_16) == 0);
 		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_RMFB, &f_16.fb_id) == 0);
 		f.fb_id = 0;
-		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f_8) == 0);
-		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_RMFB, &f_8.fb_id) == 0);
+		if (drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f_8) == 0)
+			igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_RMFB, &f_8.fb_id) == 0);
 		f.fb_id = 0;
 	}
 
-- 
2.7.4

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH igt 8/8] test/kms_addfb_basic: tolerate absence of 8-bit format

[Daniel Vetter]

On Thu, Mar 15, 2018 at 03:45:44PM +0100, Ulrich Hecht wrote:
> Ignores failure to add DRM_FORMAT_C8 frame buffer; some devices do not
> support any 8-bit format.
> 
> Signed-off-by: Ulrich Hecht <ulrich.hecht+renesas@gmail.com>

I think the proper fix for this would be to split it out into a new
subtest, and then check that you have universal planes (all modern planes
have that) and skip the test if no plane advertises C8.
-Daniel

> ---
>  tests/kms_addfb_basic.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/tests/kms_addfb_basic.c b/tests/kms_addfb_basic.c
> index d1da718..db79827 100644
> --- a/tests/kms_addfb_basic.c
> +++ b/tests/kms_addfb_basic.c
> @@ -273,8 +273,8 @@ static void size_tests(int fd)
>  		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f_16) == 0);
>  		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_RMFB, &f_16.fb_id) == 0);
>  		f.fb_id = 0;
> -		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f_8) == 0);
> -		igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_RMFB, &f_8.fb_id) == 0);
> +		if (drmIoctl(fd, DRM_IOCTL_MODE_ADDFB2, &f_8) == 0)
> +			igt_assert(drmIoctl(fd, DRM_IOCTL_MODE_RMFB, &f_8.fb_id) == 0);
>  		f.fb_id = 0;
>  	}
>  
> -- 
> 2.7.4
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

✗ Fi.CI.BAT: failure for Non-Intel test suite fixes

[Patchwork]

== Series Details ==

Series: Non-Intel test suite fixes
URL   : https://patchwork.freedesktop.org/series/40038/
State : failure

== Summary ==

Applying: tests/kms_addfb_basic: skip i915-specific tests on other platforms
Applying: tests/kms_panel_fitting: check for i915 before checking version
Applying: lib/igt_gt: has_gpu_reset(): fix failed assertion on non-i915 platforms
Applying: lib/igt_gt: check for presence of GPU reset before using it
Patch failed at 0004 lib/igt_gt: check for presence of GPU reset before using it
The copy of the patch that failed is found in: .git/rebase-apply/patch
When you have resolved this problem, run "git am --continue".
If you prefer to skip this patch, run "git am --skip" instead.
To restore the original branch and stop patching, run "git am --abort".

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH igt 0/8] Non-Intel test suite fixes

[Daniel Vetter]

On Thu, Mar 15, 2018 at 03:45:36PM +0100, Ulrich Hecht wrote:
> Hi!
> 
> I have run the tests on a Renesas R-Car M3-W's DU device, and have found a
> number of false negatives that mostly stem from use of Intel-specifics
> without checking if that makes sense first. So here's a bunch of fixes for
> those, hope they are generic enough for upstreaming.

Nice, other people using this! Do you plan to maintain this actively going
forward, or is this more a one-off effort? If you're planning to stick
around we'd be happy to fix you up with commit rights to make this all a
bit smoother (after a few of your patches have been merged ofc).

Cheers, Daniel
> 
> CU
> Uli
> 
> 
> Ulrich Hecht (8):
>   tests/kms_addfb_basic: skip i915-specific tests on other platforms
>   tests/kms_panel_fitting: check for i915 before checking version
>   lib/igt_gt: has_gpu_reset(): fix failed assertion on non-i915
>     platforms
>   lib/igt_gt: check for presence of GPU reset before using it
>   tests/kms_plane_lowres: skip i915-specific tests on other platforms
>   lib/igt_pm: turn absence of autosuspend_delay_ms from fail to skip
>   tests/kms_addfb_basic: size_tests(): reduce test buffer size
>   test/kms_addfb_basic: tolerate absence of 8-bit format
> 
>  lib/igt_gt.c              | 24 ++++++++++++++----------
>  lib/igt_pm.c              |  2 +-
>  tests/kms_addfb_basic.c   | 33 ++++++++++++++++++---------------
>  tests/kms_panel_fitting.c |  1 +
>  tests/kms_plane_lowres.c  |  1 +
>  5 files changed, 35 insertions(+), 26 deletions(-)
> 
> -- 
> 2.7.4
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH igt 0/8] Non-Intel test suite fixes

[Ulrich Hecht]

On Fri, Mar 16, 2018 at 9:55 AM, Daniel Vetter <daniel@ffwll.ch> wrote:
> On Thu, Mar 15, 2018 at 03:45:36PM +0100, Ulrich Hecht wrote:
>> Hi!
>>
>> I have run the tests on a Renesas R-Car M3-W's DU device, and have found a
>> number of false negatives that mostly stem from use of Intel-specifics
>> without checking if that makes sense first. So here's a bunch of fixes for
>> those, hope they are generic enough for upstreaming.
>
> Nice, other people using this! Do you plan to maintain this actively going
> forward, or is this more a one-off effort?

For now, this is just an attempt at evaluating if this works for us.
It has caught a few things that look like legitimate bugs to me,
though...

CU
Uli
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH igt 0/8] Non-Intel test suite fixes

[Laurent Pinchart]

Hello,

On Monday, 19 March 2018 18:41:05 EET Ulrich Hecht wrote:
> On Fri, Mar 16, 2018 at 9:55 AM, Daniel Vetter <daniel@ffwll.ch> wrote:
> > On Thu, Mar 15, 2018 at 03:45:36PM +0100, Ulrich Hecht wrote:
> >> Hi!
> >> 
> >> I have run the tests on a Renesas R-Car M3-W's DU device, and have found
> >> a number of false negatives that mostly stem from use of Intel-specifics
> >> without checking if that makes sense first. So here's a bunch of fixes
> >> for those, hope they are generic enough for upstreaming.
> > 
> > Nice, other people using this! Do you plan to maintain this actively going
> > forward, or is this more a one-off effort?
> 
> For now, this is just an attempt at evaluating if this works for us.
> It has caught a few things that look like legitimate bugs to me,
> though...

That's good news ! (Not that I'm happy that we have bugs, but catching them 
shows that igt is useful for us). I hope this will help convincing management 
that we should keep contributing to igt going forward.

Ulrich, for the record, I agree with all of Daniel's and Ville's comments 
posted so far.

-- 
Regards,

Laurent Pinchart

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH igt 0/8] Non-Intel test suite fixes

[Daniel Vetter]

On Tue, Mar 20, 2018 at 01:32:17PM +0200, Laurent Pinchart wrote:
> Hello,
> 
> On Monday, 19 March 2018 18:41:05 EET Ulrich Hecht wrote:
> > On Fri, Mar 16, 2018 at 9:55 AM, Daniel Vetter <daniel@ffwll.ch> wrote:
> > > On Thu, Mar 15, 2018 at 03:45:36PM +0100, Ulrich Hecht wrote:
> > >> Hi!
> > >> 
> > >> I have run the tests on a Renesas R-Car M3-W's DU device, and have found
> > >> a number of false negatives that mostly stem from use of Intel-specifics
> > >> without checking if that makes sense first. So here's a bunch of fixes
> > >> for those, hope they are generic enough for upstreaming.
> > > 
> > > Nice, other people using this! Do you plan to maintain this actively going
> > > forward, or is this more a one-off effort?
> > 
> > For now, this is just an attempt at evaluating if this works for us.
> > It has caught a few things that look like legitimate bugs to me,
> > though...
> 
> That's good news ! (Not that I'm happy that we have bugs, but catching them 
> shows that igt is useful for us). I hope this will help convincing management 
> that we should keep contributing to igt going forward.

Yeah I'm really hoping other vendors could join the fun, and long-term
we'd have a real kms validation suite. There's always going to be a need
for vendor-specific tests (and we're happy to merge them, see e.g. vc4),
but having a test suite that tries to be generic as much as possible, for
an uapi that tries to be generic too, seems like a really good idea.

Very much welcome on board!

Aside: If there's anything we can do to help convince your management that
this is good idea (like the rename from intel-gpu-tools to igt gpu tests
we've done), please bring it up.

Cheers, Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH igt 0/8] Non-Intel test suite fixes

[Laurent Pinchart]

Hi Ulrich,

On Thursday, 15 March 2018 16:45:36 EEST Ulrich Hecht wrote:
> Hi!
> 
> I have run the tests on a Renesas R-Car M3-W's DU device, and have found a
> number of false negatives that mostly stem from use of Intel-specifics
> without checking if that makes sense first. So here's a bunch of fixes for
> those, hope they are generic enough for upstreaming.

I'm looking for instructions on how to compile and use igt on elinux.org but 
can't find them. Could you please point me to the relevant page ?

Also, what are your plans to get those patches merged upstream ?

> Ulrich Hecht (8):
>   tests/kms_addfb_basic: skip i915-specific tests on other platforms
>   tests/kms_panel_fitting: check for i915 before checking version
>   lib/igt_gt: has_gpu_reset(): fix failed assertion on non-i915
>     platforms
>   lib/igt_gt: check for presence of GPU reset before using it
>   tests/kms_plane_lowres: skip i915-specific tests on other platforms
>   lib/igt_pm: turn absence of autosuspend_delay_ms from fail to skip
>   tests/kms_addfb_basic: size_tests(): reduce test buffer size
>   test/kms_addfb_basic: tolerate absence of 8-bit format
> 
>  lib/igt_gt.c              | 24 ++++++++++++++----------
>  lib/igt_pm.c              |  2 +-
>  tests/kms_addfb_basic.c   | 33 ++++++++++++++++++---------------
>  tests/kms_panel_fitting.c |  1 +
>  tests/kms_plane_lowres.c  |  1 +
>  5 files changed, 35 insertions(+), 26 deletions(-)

-- 
Regards,

Laurent Pinchart



_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH igt 0/8] Non-Intel test suite fixes

[Ulrich Hecht]

On Fri, Apr 27, 2018 at 6:03 PM, Laurent Pinchart
<laurent.pinchart@ideasonboard.com> wrote:
> Hi Ulrich,
>
> On Thursday, 15 March 2018 16:45:36 EEST Ulrich Hecht wrote:
>> Hi!
>>
>> I have run the tests on a Renesas R-Car M3-W's DU device, and have found a
>> number of false negatives that mostly stem from use of Intel-specifics
>> without checking if that makes sense first. So here's a bunch of fixes for
>> those, hope they are generic enough for upstreaming.
>
> I'm looking for instructions on how to compile and use igt on elinux.org but
> can't find them. Could you please point me to the relevant page ?

Here's a wikified version of what I have sent to our QA people:

https://elinux.org/R-Car/Tests:igt

CU
Uli
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx