Hi Maxime,
Thanks for your patch!
On Mon, Mar 4, 2024 at 10:12 AM Maxime Ripard <mripard@xxxxxxxxxx> wrote:
> Commit 358e76fd613a ("drm/sun4i: hdmi: Consolidate atomic_check and
> mode_valid") changed the clock rate from an unsigned long to an unsigned
> long long resulting in a a 64-bit division that might not be supported
> on all platforms.
Why was this changed to unsigned long long?
Can a valid pixel clock really not fit in 32-bit?
> The resulted in compilation being broken at least for m68k, xtensa and
> some arm configurations, at least.
>
> Fixes: 358e76fd613a ("drm/sun4i: hdmi: Consolidate atomic_check and mode_valid")
> Reported-by: Geert Uytterhoeven <geert@xxxxxxxxxxxxxx>
> Reported-by: Naresh Kamboju <naresh.kamboju@xxxxxxxxxx>
> Closes: https://lore.kernel.org/r/CA+G9fYvG9KE15PGNoLu+SBVyShe+u5HBLQ81+kK9Zop6u=ywmw@xxxxxxxxxxxxxx/
> Reported-by: kernel test robot <lkp@xxxxxxxxx>
> Closes: https://lore.kernel.org/oe-kbuild-all/202403011839.KLiXh4wC-lkp@xxxxxxxxx/
> Signed-off-by: Maxime Ripard <mripard@xxxxxxxxxx>
> --- a/drivers/gpu/drm/sun4i/sun4i_hdmi_enc.c
> +++ b/drivers/gpu/drm/sun4i/sun4i_hdmi_enc.c
> @@ -163,11 +163,11 @@ static enum drm_mode_status
> sun4i_hdmi_connector_clock_valid(const struct drm_connector *connector,
> const struct drm_display_mode *mode,
> unsigned long long clock)
> {
> const struct sun4i_hdmi *hdmi = drm_connector_to_sun4i_hdmi(connector);
> - unsigned long diff = clock / 200; /* +-0.5% allowed by HDMI spec */
> + unsigned long diff = div_u64(clock, 200); /* +-0.5% allowed by HDMI spec */
I'd rather see clock changed back to unsigned long.
> long rounded_rate;
>
> if (mode->flags & DRM_MODE_FLAG_DBLCLK)
> return MODE_BAD;
>
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
