The LogicPD Type28 display used by several Logic PD products has not worked since v5.5.
The connector type for the LogicPD Type 28 display is missing and drm_panel_bridge_add() requires connector type to be set.
Signed-off-by: Adam Ford aford173@gmail.com CC: stable@vger.kernel.org #v5.5+
diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c index b6ecd1552132..334e9de5b2c8 100644 --- a/drivers/gpu/drm/panel/panel-simple.c +++ b/drivers/gpu/drm/panel/panel-simple.c @@ -2495,6 +2495,7 @@ static const struct panel_desc logicpd_type_28 = { .bus_format = MEDIA_BUS_FMT_RGB888_1X24, .bus_flags = DRM_BUS_FLAG_DE_HIGH | DRM_BUS_FLAG_PIXDATA_DRIVE_POSEDGE | DRM_BUS_FLAG_SYNC_DRIVE_NEGEDGE, + .connector_type = DRM_MODE_CONNECTOR_DPI, };
static const struct panel_desc mitsubishi_aa070mc01 = {
On Mon, Jun 15, 2020 at 10:19 AM Adam Ford aford173@gmail.com wrote:
The LogicPD Type28 display used by several Logic PD products has not worked since v5.5.
Maybe you could tell which commit exactly and then put a Fixes tag?
On Mon, Jun 15, 2020 at 9:46 AM Fabio Estevam festevam@gmail.com wrote:
On Mon, Jun 15, 2020 at 10:19 AM Adam Ford aford173@gmail.com wrote:
The LogicPD Type28 display used by several Logic PD products has not worked since v5.5.
Maybe you could tell which commit exactly and then put a Fixes tag?
I honestly don't know. I reached out to the omap mailing list, because I noted this issue. Tomi V from TI responded with a link that he posted which fixes this for another display.
https://www.mail-archive.com/dri-devel@lists.freedesktop.org/msg312208.html
I tested that patch and it worked for a different LCD, so I did the same thing to the Logic PD Type 28 display as well.
My patch and commit message were modeled after his, and his commit CC's stable with a note about being required for v5.5+
I added him to the CC list, so maybe he knows which hash needs to be referenced from a fixes tag. I was hoping to not have to go back and bisect if it's not required, but I will if necessary.
adam
On 15/06/2020 17:53, Adam Ford wrote:
On Mon, Jun 15, 2020 at 9:46 AM Fabio Estevam festevam@gmail.com wrote:
On Mon, Jun 15, 2020 at 10:19 AM Adam Ford aford173@gmail.com wrote:
The LogicPD Type28 display used by several Logic PD products has not worked since v5.5.
Maybe you could tell which commit exactly and then put a Fixes tag?
I honestly don't know. I reached out to the omap mailing list, because I noted this issue. Tomi V from TI responded with a link that he posted which fixes this for another display.
https://www.mail-archive.com/dri-devel@lists.freedesktop.org/msg312208.html
I tested that patch and it worked for a different LCD, so I did the same thing to the Logic PD Type 28 display as well.
My patch and commit message were modeled after his, and his commit CC's stable with a note about being required for v5.5+
I added him to the CC list, so maybe he knows which hash needs to be referenced from a fixes tag. I was hoping to not have to go back and bisect if it's not required, but I will if necessary.
No, I didn't check when exactly it broke. connector_type was added in v5.5, and my patch applies to v5.5, so I set that as stable version. But the WARN comes from panel bridge. Possibly 89958b7cd9555a5d82556cc9a1f4c62fffda6f96 is the one that adds requirement to have connector_type.
Tomi
Hi Adam.
On Mon, Jun 15, 2020 at 09:53:45AM -0500, Adam Ford wrote:
On Mon, Jun 15, 2020 at 9:46 AM Fabio Estevam festevam@gmail.com wrote:
On Mon, Jun 15, 2020 at 10:19 AM Adam Ford aford173@gmail.com wrote:
The LogicPD Type28 display used by several Logic PD products has not worked since v5.5.
Maybe you could tell which commit exactly and then put a Fixes tag?
I honestly don't know. I reached out to the omap mailing list, because I noted this issue. Tomi V from TI responded with a link that he posted which fixes this for another display.
https://www.mail-archive.com/dri-devel@lists.freedesktop.org/msg312208.html
I tested that patch and it worked for a different LCD, so I did the same thing to the Logic PD Type 28 display as well.
My patch and commit message were modeled after his, and his commit CC's stable with a note about being required for v5.5+
I added him to the CC list, so maybe he knows which hash needs to be referenced from a fixes tag. I was hoping to not have to go back and bisect if it's not required, but I will if necessary.
git blame is your friend - the panel was added here: 0d35408afbeb6 (Adam Ford 2019-10-16 08:51:45 -0500 2469) static const struct panel_desc logicpd_type_28 = { 0d35408afbeb6 (Adam Ford 2019-10-16 08:51:45 -0500 2470) .modes = &logicpd_type_28_mode, 0d35408afbeb6 (Adam Ford 2019-10-16 08:51:45 -0500 2471) .num_modes = 1, 0d35408afbeb6 (Adam Ford 2019-10-16 08:51:45 -0500 2472) .bpc = 8, 0d35408afbeb6 (Adam Ford 2019-10-16 08:51:45 -0500 2473) .size = { 0d35408afbeb6 (Adam Ford 2019-10-16 08:51:45 -0500 2474) .width = 105, 0d35408afbeb6 (Adam Ford 2019-10-16 08:51:45 -0500 2475) .height = 67, 0d35408afbeb6 (Adam Ford 2019-10-16 08:51:45 -0500 2476) },
So this gives us fllowing fixes info: Fixes: 0d35408afbeb ("drm/panel: simple: Add Logic PD Type 28 display support") Cc: Adam Ford aford173@gmail.com Cc: Sam Ravnborg sam@ravnborg.org Cc: Thierry Reding thierry.reding@gmail.com Cc: dri-devel@lists.freedesktop.org Cc: stable@vger.kernel.org # v5.6+
I have adjusted the changelog to say 5.6 and applied to drm-misc-fixes
Sam
dri-devel@lists.freedesktop.org
-
Adam Ford -
Fabio Estevam -
Sam Ravnborg -
Tomi Valkeinen