Am Montag, 25. November 2019, 10:28:12 CET schrieb Neil Armstrong:
Hi,
On 23/11/2019 21:51, Heiko Stuebner wrote:
Hi Neil,
Am Samstag, 23. November 2019, 13:38:41 CET schrieb Neil Armstrong:
Le 23/11/2019 à 11:39, Heiko Stuebner a écrit :
From: Heiko Stuebner heiko.stuebner@theobroma-systems.com
The timing values for dw-dsi are often dependent on the used display and according to Philippe Cornu will most likely also depend on the used phy technology in the soc-specific implementation.
To solve this and allow specific implementations to define them as needed add a new get_timing callback to phy_ops and call this from the dphy_timing function to retrieve the necessary values for the specific mode.
Right now this handles the hs2lp + lp2hs where Rockchip SoCs need handling according to the phy speed, while STM seems to be ok with static values.
changes in v3:
- check existence of phy_ops->get_timing in __dw_mipi_dsi_probe()
- emit actual error when get_timing() call fails
- add tags from Philippe and Yannick
changes in v2:
- add driver-specific handling, don't force all bridge users to use the same timings, as suggested by Philippe
Suggested-by: Philippe Cornu philippe.cornu@st.com Signed-off-by: Heiko Stuebner heiko.stuebner@theobroma-systems.com Reviewed-by: Philippe Cornu philippe.cornu@st.com Tested-by: Yannick Fertre yannick.fertre@st.com
You forgot to Cc dri-devel, so I can't apply it.
oh dang ... resend coming shortly
Heiko
Could you resent it ?
Neil
drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c | 27 +++++-- drivers/gpu/drm/rockchip/Kconfig | 1 + .../gpu/drm/rockchip/dw-mipi-dsi-rockchip.c | 78 +++++++++++++++++++ drivers/gpu/drm/stm/dw_mipi_dsi-stm.c | 13 ++++ include/drm/bridge/dw_mipi_dsi.h | 9 +++ 5 files changed, 121 insertions(+), 7 deletions(-)
[...]
Looks good, Reviewed-by: Neil Armstrong narmstrong@baylibre.com
I can apply it on monday if you want.
that would be very cool. Looking forward to monday then
Thanks Heiko