This looks to be recently broken in: 5389b3f3
("tests/i915/i915_module_load: Use GEM_BUSY instead of SET_CACHE for
sanity check").
Which fails on discrete platforms, since we switched over to the
gem_busy ioctl, from the set_caching ioctl, which no longer gives
-ENODEV as the expected error.
Signed-off-by: Matthew Auld <matthew.auld@xxxxxxxxx>
Cc: Janga Rahul Kumar <janga.rahul.kumar@xxxxxxxxx>
Cc: Priyanka Dandamudi <priyanka.dandamudi@xxxxxxxxx>
---
tests/i915/i915_module_load.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/i915/i915_module_load.c b/tests/i915/i915_module_load.c
index eebb3167..d3a86b11 100644
--- a/tests/i915/i915_module_load.c
+++ b/tests/i915/i915_module_load.c
@@ -228,7 +228,7 @@ static void gem_sanitycheck(void)
{
struct drm_i915_gem_busy args = {};
int i915 = __drm_open_driver(DRIVER_INTEL);
- int expected = gem_has_lmem(i915) ? -ENODEV : -ENOENT;
+ int expected = -ENOENT;
int err;
err = 0;
--
2.38.1
