drm/rockchip: Properly adjust to a true clock in adjusted_mode
Subsystems (4)
PATCH 1/2 drm/rockchip: Properly adjust to a true clock in adjusted_mode
When fixing up the clock in vop_crtc_mode_fixup() we're not doing it quite correctly. Specifically if we've got the true clock 266666667 Hz, we'll perform this calculation:
266666667 / 1000 => 266666
Later when we try to set the clock we'll do clk_set_rate(266666 * 1000). The common clock framework won't actually pick the proper clock in this case since it always wants clocks <= the specified one.
Let's solve this by using DIV_ROUND_UP.
---
Back in 2016 Mark Yao said he applied this to his drm fixes [1], but it's 2019 and it's still missing so I'm posting again.
[1] https://patchwork.freedesktop.org/patch/103872/
drivers/gpu/drm/rockchip/rockchip_drm_vop.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpu/drm/rockchip/rockchip_drm_vop.c +2 -1
@@ -1006,7 +1006,8 @@ static bool vop_crtc_mode_fixup(struct drm_crtc *crtc,struct vop *vop =to_vop (crtc); adjusted_mode->clock = -clk_round_rate (vop->dclk, mode->clock *1000 ) /1000 ; +DIV_ROUND_UP (clk_round_rate (vop->dclk, mode->clock *1000 ), +1000 );return true; }
Signature · 6 lines
2.22.0.410.gd8fdbe21b5-goog _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
PATCH 2/2 drm/rockchip: Base adjustments of the mode based on prev adjustments
In vop_crtc_mode_fixup() we fixup the mode to show what we actually will be able to achieve. However we should base our adjustments on any previous adjustments that were made.
As an example, the dw_hdmi driver may wish to make some small adjustments to clock rates in its atomic_check() function. If it does, it will update the adjusted_mode. We shouldn't throw away those adjustments.
NOTE: the version of the dw_hdmi driver upstream doesn't _actually_ make such adjustments, but downstream in Chrome OS it does. It is plausible that one day we'll figure out how to cleanly make that happen in an upstream-friendly way, so we should prepare by using the right mode.
---
drivers/gpu/drm/rockchip/rockchip_drm_vop.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/gpu/drm/rockchip/rockchip_drm_vop.c +2 -2
@@ -1006,8 +1006,8 @@ static bool vop_crtc_mode_fixup(struct drm_crtc *crtc,struct vop *vop =to_vop (crtc); adjusted_mode->clock = -DIV_ROUND_UP (clk_round_rate (vop->dclk, mode->clock *1000 ), -1000 ); +DIV_ROUND_UP (clk_round_rate (vop->dclk, + adjusted_mode->clock *1000 ),1000 );return true; }
Signature · 1 line
2.22.0.410.gd8fdbe21b5-goog
Heiko Stübner
maintainerAm Samstag, 15. Juni 2019, 00:47:29 CEST schrieb Douglas Anderson:
Quoted 9 lines
When fixing up the clock in
vop_crtc_mode_fixup()we're not doing it quite correctly. Specifically if we've got the true clock 266666667 Hz, we'll perform this calculation:266666667 / 1000 => 266666Later when we try to set the clock we'll do
clk_set_rate(266666 * 1000). The common clock framework won't actually pick the proper clock in this case since it always wants clocks <= the specified one.
Quoted 6 lines
Let's solve this by using
DIV_ROUND_UP.Fixes b59b8de31497 ("drm/rockchip: return a true clock rate to adjusted_mode")Signed-off-by Douglas Anderson <[email protected]>Signed-off-by Sean Paul <[email protected]>Reviewed-by Yakir Yang <[email protected]>
I gave both patches a testrun on rk3288, rk3328 and rk3399 and applied them to drm-misc-next thereafter
Thanks
Heiko
_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel