Hello Arthur, Thanks a lot for your patch! Arthur Grillo <arthurgrillo@xxxxxxxxxx> writes: > Extend the existing test cases to test the conversion from XRGB8888 to > monochromatic. > > Signed-off-by: Arthur Grillo <arthurgrillo@xxxxxxxxxx> > --- [...] > +static size_t conversion_buf_size_mono(unsigned int dst_pitch, const struct drm_rect *clip) > +{ > + if (!dst_pitch) { > + unsigned int linepixels = drm_rect_width(clip) * 1; > + > + dst_pitch = DIV_ROUND_UP(linepixels, 8); > + } > + > + return dst_pitch * drm_rect_height(clip); > +} > + I don't think you need a new helper only for this. There are other formats that have sub-byte pixels, so you may want to instead make the existing conversion_buf_size() function more general. Could you please base on the following patch that I just posted? https://lists.freedesktop.org/archives/dri-devel/2023-March/394466.html I believe with that you should be able to drop this format specific helper and just use the fourcc DRM_FORMAT_C1 instead. [...] > > +static void drm_test_fb_xrgb8888_to_mono(struct kunit *test) > +{ > + const struct convert_xrgb8888_case *params = test->param_value; > + const struct convert_to_mono_result *result = ¶ms->mono_result; > + size_t dst_size; > + u8 *buf = NULL; > + __le32 *xrgb8888 = NULL; > + struct iosys_map dst, src; > + > + struct drm_framebuffer fb = { > + .format = drm_format_info(DRM_FORMAT_XRGB8888), > + .pitches = { params->pitch, 0, 0 }, > + }; > + > + dst_size = conversion_buf_size_mono(result->dst_pitch, ¶ms->clip); > + Then here you could just do: dst_size = conversion_buf_size(DRM_FORMAT_C1, result->dst_pitch, ¶ms->clip); If you do that, feel free to add: Reviewed-by: Javier Martinez Canillas <javierm@xxxxxxxxxx> -- Best regards, Javier Martinez Canillas Core Platforms Red Hat