On Thu, 7 May 2020 at 13:29, Vinay Simha B N simhavcs@gmail.com wrote:
Emil,
Reply inline
On Tue, 5 May 2020 at 9:35 PM, Emil Velikov emil.l.velikov@gmail.com wrote:
From: Emil Velikov emil.velikov@collabora.com
The helper uses the MIPI_DCS_SET_TEAR_SCANLINE, although it's currently using the generic write. This does not look right.
Perhaps some platforms don't distinguish between the two writers?
Cc: Robert Chiras robert.chiras@nxp.com Cc: Vinay Simha BN simhavcs@gmail.com Cc: Jani Nikula jani.nikula@intel.com Cc: Thierry Reding treding@nvidia.com Fixes: e83950816367 ("drm/dsi: Implement set tear scanline") Signed-off-by: Emil Velikov emil.velikov@collabora.com
Robert, can you please test this against the only user - the Raydium RM67191 panel driver that you introduced.
Thanks
Vinay, can you confirm if this is a genuine typo or there's something really subtle happening.
this has been tested on nexus 7 with jdi panel.
The jdi panel (JDI LT070ME05000 I believe) does not use the function, hmm.
Looking through the ML archive - the call in the first 4 revisions of the patch. Then with v5 it has magically disappeared alongside mipi_dsi_dcs_set_tear_on().
No comment explaining why though - does the driver work w/o both of those?
I did not understand what is the typo here? We need to use DC’s write instead of generic write?
I believe the clue is in the command name - MIPI_DSI_DCS. I was going to double-check with the spec although it's members only :-\ Based on the usage in DRM, all DCS commands are issued via mipi_dsi_dcs_{read,write}
Thanks Emil