[igt-dev] [PATCH i-g-t 08/11] tests/kms_addfb_basic: Add invalid buffer object test for discrete

[Petri Latvala]

On Thu, Jun 10, 2021 at 03:30:46PM +0300, Ruhl, Michael J wrote:
> >-----Original Message-----
> >From: Patnana, Venkata Sai <venkata.sai.patnana at intel.com>
> >Sent: Wednesday, June 9, 2021 8:18 AM
> >To: igt-dev at lists.freedesktop.org
> >Cc: Patnana, Venkata Sai <venkata.sai.patnana at intel.com>; Mohammed
> >Khajapasha <mohammed.khajapasha at intel.com>; Latvala; Latvala, Petri
> ><petri.latvala at intel.com>; Ruhl, Michael J <michael.j.ruhl at intel.com>
> >Subject: [PATCH i-g-t 08/11] tests/kms_addfb_basic: Add invalid buffer object
> >test for discrete
> >
> >From: Mohammed Khajapasha <mohammed.khajapasha at intel.com>
> >
> >Add invalid buffer object test case for discrete, to validate
> >IOCTL_MODE_ADDFB2 ioctl for LMEM, the framebuffer backing
> >gem objects should be from local memory for discrete.
> >
> >Cc: Latvala, Petri <petri.latvala at intel.com>
> >Cc: Michael J. Ruhl <michael.j.ruhl at intel.com>
> >Signed-off-by: Mohammed Khajapasha
> ><mohammed.khajapasha at intel.com>
> >---
> > tests/kms_addfb_basic.c | 16 ++++++++++++++++
> > 1 file changed, 16 insertions(+)
> >
> >diff --git a/tests/kms_addfb_basic.c b/tests/kms_addfb_basic.c
> >index 1fb18d668f..aac74a7568 100644
> >--- a/tests/kms_addfb_basic.c
> >+++ b/tests/kms_addfb_basic.c
> >@@ -40,6 +40,7 @@
> >
> > #include "igt_rand.h"
> > #include "igt_device.h"
> >+#include "i915/intel_memory_region.h"
> >
> > uint32_t gem_bo;
> > uint32_t gem_bo_small;
> >@@ -146,6 +147,21 @@ static void invalid_tests(int fd)
> > 		igt_assert(f.modifier[0] == 0);
> > 	}
> >
> >+	igt_subtest("invalid-smem-bo-on-discrete") {
> >+		int devid;
> >+		uint32_t handle, stride;
> >+		uint64_t size;
> >+
> >+		igt_require_intel(fd);
> >+		devid = intel_get_drm_devid(fd);
> 
> This will only work on if the discrete card is the only card on the system.
> 
> Is there a better way to:
> 
> 1) identify discrete cards and/or
> 2) specifically select a discrete card for testing?

The test does not select the device to use. The admin _running_ the
test does that, with IGT_DEVICE.

With that done, mapping of results will be

non-intel: skip
intel integrated: skip
intel discrete: actually tested

which looks correct. We're trying to test what happens specifically
if the used device is intel discrete, skip otherwise.


-- 
Petri Latvala