Quoting Tvrtko Ursulin (2018-11-15 18:37:07) > From: Lionel Landwerlin <lionel.g.landwerlin@xxxxxxxxx> > > Verify that the per-context dynamic SSEU uAPI works as expected. <SNIP> > +static bool > +kernel_has_per_context_sseu_support(int fd) > +{ > + struct drm_i915_gem_context_param_sseu sseu = { }; > + struct drm_i915_gem_context_param arg = > + { > + .param = I915_CONTEXT_PARAM_SSEU, > + .size = sizeof(sseu), > + .value = to_user_pointer(&sseu), > + }; This ... > + int ret; > + > + if (__gem_context_get_param(fd, &arg)) > + return false; > + > + arg.value = to_user_pointer(&sseu); > + > + ret = __gem_context_set_param(fd, &arg); > + > + igt_assert(ret == 0 || ret == -ENODEV || ret == -EINVAL); > + > + return ret == 0; > +} <SNIP> > +static void test_engines(int fd) > +{ > + struct drm_i915_gem_context_param_sseu sseu = { }; > + struct drm_i915_gem_context_param arg = > + { .param = I915_CONTEXT_PARAM_SSEU, > + .ctx_id = gem_context_create(fd), > + .size = sizeof(sseu), > + .value = to_user_pointer(&sseu) }; ... and this don't seem to be consitent with kernel coding style, which I thought we try to use in IGT? There seems to be some other strangely indented structs too, with those fixed this is; Reviewed-by: Joonas Lahtinen <joonas.lahtinen@xxxxxxxxxxxxxxx> Regards, Joonas _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx