On Wed, Nov 13, 2019 at 05:44:32PM +0100, Hans de Goede wrote: > Instead of only setting mode->specified on false on an early exit and > leaving e.g. mode->bpp_specified and mode->refresh_specified as is, > lets be consistent and just zero out the entire passed in struct at > the top of drm_mode_parse_command_line_for_connector() > > Signed-off-by: Hans de Goede <hdego...@redhat.com> > --- > drivers/gpu/drm/drm_modes.c | 5 ++--- > 1 file changed, 2 insertions(+), 3 deletions(-) > > diff --git a/drivers/gpu/drm/drm_modes.c b/drivers/gpu/drm/drm_modes.c > index beb764efe6b3..1fee4a71eff7 100644 > --- a/drivers/gpu/drm/drm_modes.c > +++ b/drivers/gpu/drm/drm_modes.c > @@ -1745,12 +1745,11 @@ bool drm_mode_parse_command_line_for_connector(const > char *mode_option, > char *bpp_end_ptr = NULL, *refresh_end_ptr = NULL; > int i, len, ret; > > + memset(mode, 0, sizeof(*mode)); > mode->panel_orientation = DRM_MODE_PANEL_ORIENTATION_UNKNOWN;
The reported error by kbuild rings a bell. I think I tried to do this, saw that error, and then forgot about it. Looking more at the code now, I don't see any in bochs that looks really wrong. Either way, we should either fix bochs, or add a unit-test to have a test for the bochs case so that we don't have that issue sneaking around. (btw, thanks for adding more tests, that's really appreciated) Maxime
signature.asc
Description: PGP signature
_______________________________________________ dri-devel mailing list dri-devel@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/dri-devel
