lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <d6c4a63fb6d51c23db50bf15335d059208aeb3a4.camel@gmail.com>
Date:   Tue, 26 Jan 2021 21:38:21 +0200
From:   aleksandr.o.makarov@...il.com
To:     Philippe CORNU <philippe.cornu@...com>,
        Andrzej Hajda <a.hajda@...sung.com>,
        Neil Armstrong <narmstrong@...libre.com>,
        Laurent Pinchart <Laurent.pinchart@...asonboard.com>,
        Jonas Karlman <jonas@...boo.se>,
        Jernej Skrabec <jernej.skrabec@...l.net>,
        David Airlie <airlied@...ux.ie>,
        Daniel Vetter <daniel@...ll.ch>,
        Sam Ravnborg <sam@...nborg.org>,
        Antonio BORNEO <antonio.borneo@...com>,
        "<dri-devel@...ts.freedesktop.org>" <linux-kernel@...r.kernel.org>
Subject: Re: panel: panel-raydium68200 driver fails to write MIPI DSI init
 commands

В Вт, 19/01/2021 в 16:24 +0000, Philippe CORNU пишет:
> Dear Aleksandr,
> 
> We use the raydium68200 ic driver in a dsi 720p 2dl panel module mounted on the MB1230 board [1], mounted on the STM32MP157 eval board [2].
> 
> According to your email, you are using the EDT ETML0500F3DHA panel module, probably composed of a raydium68200+a touchscreen+a glass+backlight+.... 
> 
> Could you please double check if your panel module has the same characteristics as the one described in panel-raydium68200.c (pixel clock, blanking values, resolutions, number of dsi data lanes, enable & reset gpios, backlight...).
> 
> Moreover, maybe your panel embeds a non-volatile ram which contains nice default values ("fused" during production) allowing to reduce a lot the panel init sequence... allowing then to use panel-simple.c instead of panel-raydium68200.c (that could explain why you can see "colors" without sending any init sequence).
> 
> The issues you encountered may come from (starting with the highest probability):
> * bad lcd hw vs sw configuration (see description above).
> * bad pixel clock frequency, bad blanking values...
> * bad dsi internal Rockchip ip programming (pll and clock trees in dt...)
> 
> Hope it helps,
> 
> [1] https://wiki.st.com/stm32mpu/wiki/MB1230
> [2] https://www.st.com/en/evaluation-tools/stm32mp157a-ev1.html
> 
> Philippe :-)
> 
I've fixed errors related to writing to DSI after applying this [1]
patch.

With the panel timings and other parameters all left intact in device
tree, I only had to apply that patch to start to see video.

After doing some testing, I found an issue with patch. With it applied,
you must keep your driver as module with DRM_PANEL_RAYDIUM_RM68200=m.
Otherwise, if DRM_PANEL_RAYDIUM_RM68200=y is set, the issues with
writing to DSI from prepare() shall persist. 

[1] https://patchwork.kernel.org/project/dri-
devel/patch/95f16906d654057c912f089d286bd51856ee3bdf.1607892237.git.tom
myhebb@...il.com/
> -----Original Message-----
> From: aleksandr.o.makarov@...il.com <aleksandr.o.makarov@...il.com> 
> Sent: Saturday, January 16, 2021 12:40
> To: Andrzej Hajda <a.hajda@...sung.com>; Neil Armstrong <narmstrong@...libre.com>; Laurent Pinchart <Laurent.pinchart@...asonboard.com>; Jonas Karlman <jonas@...boo.se>; Jernej Skrabec <jernej.skrabec@...l.net>; David Airlie <airlied@...ux.ie>; Daniel Vetter <daniel@...ll.ch>; Sam Ravnborg <sam@...nborg.org>; Philippe CORNU <philippe.cornu@...com>; Antonio BORNEO <antonio.borneo@...com>; <dri-devel@...ts.freedesktop.org> <linux-kernel@...r.kernel.org>
> Subject: drm: panel: panel-raydium68200 driver fails to write MIPI DSI init commands
> 
> I need to bring up my MIPI DSI 1280x720 EDT ETML0500F3DHA panel on a
> RockPro64 V2.1 board.
> 
> There is no completely suitable in-tree driver for that panel yet, but for the purpose of reproducing the issue that I face, the gpu/drm/panel/panel-raydium-rm68200.c can do just fine.
> 
> To reproduce:
> 
> - Get the same Linux 5.9.14 as on my RockPro64 board (with Armbian
> 20.11.6 on it)
> 
> - Patch the rk3399-rockpro64.dts to add a panel node that is compatible with "raydium,rm68200" driver on MIPI interface (rockpro64- rm68200.patch attached)
> 
> - Compile and put the resulting rk3399-rockpro64.dtb on the target system. The panel driver shall then get probed at next boot.
> 
> The kernel log shall contain following errors:
> 
> [   10.139957] dw-mipi-dsi-rockchip ff960000.mipi: failed to write
> command FIFO
> [   10.139988] [drm:rm68200_dcs_write_cmd.isra.4
> [panel_raydium_rm68200]] *ERROR* MIPI DSI DCS write failed: -110
> [   10.160972] dw-mipi-dsi-rockchip ff960000.mipi: failed to write
> command FIFO
> [   10.161000] [drm:rm68200_dcs_write_buf.isra.5
> [panel_raydium_rm68200]] *ERROR* MIPI DSI DCS write buffer failed: -110
> [   10.181929] dw-mipi-dsi-rockchip ff960000.mipi: failed to write
> command FIFO
> [   10.181953] [drm:rm68200_dcs_write_buf.isra.5
> [panel_raydium_rm68200]] *ERROR* MIPI DSI DCS write buffer failed: -110
> [   10.202923] dw-mipi-dsi-rockchip ff960000.mipi: failed to write
> command FIFO
> [   10.202947] [drm:rm68200_dcs_write_buf.isra.5
> [panel_raydium_rm68200]] *ERROR* MIPI DSI DCS write buffer failed: -110
> [   10.223064] dw-mipi-dsi-rockchip ff960000.mipi: failed to write
> command FIFO
> [   10.223094] [drm:rm68200_dcs_write_buf.isra.5
> [panel_raydium_rm68200]] *ERROR* MIPI DSI DCS write buffer failed: -110
> [   10.226104] zram1: detected capacity change from 0 to 52428800
> [   10.244027] dw-mipi-dsi-rockchip ff960000.mipi: failed to write
> command FIFO
> [   10.244073] [drm:rm68200_dcs_write_buf.isra.5
> [panel_raydium_rm68200]] *ERROR* MIPI DSI DCS write buffer failed: -110
> [   10.265024] dw-mipi-dsi-rockchip ff960000.mipi: failed to write
> command FIFO
> [   10.265064] [drm:rm68200_dcs_write_buf.isra.5
> [panel_raydium_rm68200]] *ERROR* MIPI DSI DCS write buffer failed: -110
> [   10.285711] dw-mipi-dsi-rockchip ff960000.mipi: failed to write
> command FIFO
> [   10.285746] [drm:rm68200_dcs_write_buf.isra.5
> [panel_raydium_rm68200]] *ERROR* MIPI DSI DCS write buffer failed: -110
> [   10.305926] dw-mipi-dsi-rockchip ff960000.mipi: failed to write
> command FIFO
> [   10.305955] [drm:rm68200_dcs_write_buf.isra.5
> [panel_raydium_rm68200]] *ERROR* MIPI DSI DCS write buffer failed: -110
> [   10.326996] dw-mipi-dsi-rockchip ff960000.mipi: failed to write
> command FIFO
> [   10.327039] [drm:rm68200_dcs_write_buf.isra.5
> [panel_raydium_rm68200]] *ERROR* MIPI DSI DCS write buffer failed: -110
> [   10.348030] dw-mipi-dsi-rockchip ff960000.mipi: failed to write
> command FIFO
> [   10.348074] [drm:rm68200_dcs_write_buf.isra.5
> [panel_raydium_rm68200]] *ERROR* MIPI DSI DCS write buffer failed: -110
> 
> It's remarkable that if to pull the module panel-rm682000 out and then back in, there are no errors mentioned. I can for sure say that those commands become effective - I start seeing colourful stripes on the display after. That is, if I would send the correct command set to the panel, then it would bring up just fine. the panel, then it would bring up just fine.


Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ