add support for xiaomi-ido (Xiaomi Redmi 3 [Pro])
Created by: lawl
- Boots
- USB Networking
- osk-sdl works to unlock root partition
- battery
- weston starts and displays shit, couldn't be fucked to configure eudev for it yet, so touchscreen works only in osk-sdl, not weston yet
Merge request reports
Merged by
(Mar 29, 2025 6:54pm UTC)
Activity
Last reply by Administrator- Last reply by Administrator
- Last reply by Administrator
- Last reply by Administrator
-
Created by: lawl
good points, fixed. would it make sense to add all of these checks as kind of a 'linter' to
pmbootstrap?I'll fully admit this is/was a "fuck all of this" PR that wasn't/isn't ready for upstreaming. Yes, I'm still pissed because you know why, I'll try to not let it influence a great project.
By * postmarketOS Bot on 2017-12-26T23:33:38
Edited by Ghost User 1 #!/bin/sh 2 3 # postmarketos-mkinitfs's init script does already have a framebuffer workaround script 4 # but that script checks if fb0/mode is empty. in our case, it already has the correct value 5 # but the framebuffer is nevertheless not doing anything. 6 # additionally if we set the mode again the at time the postmarketos init script does, 7 # absolutely nothing happens. It seems to work if we sleep 5 seconds to give whatever needs 8 # more time, more time and then set the framebuffer mode again. 9 10 11 set_xiaomi_ido_framebuffer_mode() { -
@lawl you could source
init_functions.shfile and callset_framebuffer_modefunction again instead of creating a new functionBy Daniele Debernardi on 2017-12-26T23:27:57
Edited by Ghost User
1 #!/bin/sh 2 3 # postmarketos-mkinitfs's init script does already have a framebuffer workaround script 4 # but that script checks if fb0/mode is empty. in our case, it already has the correct value 5 # but the framebuffer is nevertheless not doing anything. 6 # additionally if we set the mode again the at time the postmarketos init script does, 7 # absolutely nothing happens. It seems to work if we sleep 5 seconds to give whatever needs 8 # more time, more time and then set the framebuffer mode again. 9 10 11 set_xiaomi_ido_framebuffer_mode() { -
Created by: ollieparanoid
Read the comment above, he tried that already
By * postmarketOS Bot on 2017-12-26T23:27:57
Edited by Ghost User
1 #!/bin/sh 2 3 # postmarketos-mkinitfs's init script does already have a framebuffer workaround script 4 # but that script checks if fb0/mode is empty. in our case, it already has the correct value 5 # but the framebuffer is nevertheless not doing anything. 6 # additionally if we set the mode again the at time the postmarketos init script does, 7 # absolutely nothing happens. It seems to work if we sleep 5 seconds to give whatever needs 8 # more time, more time and then set the framebuffer mode again. 9 10 11 set_xiaomi_ido_framebuffer_mode() { -
Created by: lawl
nope. it's subtly different and has to be.
[ -e "/sys/class/graphics/fb0/mode" ]vs.
[ -z "$(cat /sys/class/graphics/fb0/mode)" ]otherwise
set_framebuffer_modeactually does nothing here./sys/class/graphics/fb0/modeDOES have the correct value already, we still need to rewrite it because I don't fucking know why. No output otherwise.By * postmarketOS Bot on 2017-12-26T23:27:57
Edited by Ghost User
1 #!/bin/sh 2 3 # postmarketos-mkinitfs's init script does already have a framebuffer workaround script 4 # but that script checks if fb0/mode is empty. in our case, it already has the correct value 5 # but the framebuffer is nevertheless not doing anything. 6 # additionally if we set the mode again the at time the postmarketos init script does, 7 # absolutely nothing happens. It seems to work if we sleep 5 seconds to give whatever needs 8 # more time, more time and then set the framebuffer mode again. 9 10 11 set_xiaomi_ido_framebuffer_mode() { -
Created by: lawl
arguably we could remove all checks and just cat ..> mode without any checks, which is what i had in the beginning, but changed it to this. not sure if it makes too much of a difference really. i guess it kind of highlights the difference on why
set_framebuffer_modedoesn't work for this device?By * postmarketOS Bot on 2017-12-26T23:27:57
Edited by Ghost User
-
Created by: ollieparanoid
good points, fixed. would it make sense to add all of these checks as kind of a 'linter' to
pmbootstrap?Yes it would! #854 would have catched the pkgver at least (this PR only needs to be reviewed before it can be merged). In the future we might also be able to do linting checks for shell scripts/APKBUILDs like we do it now with PEP8 for python (#110).
I'll fully admit this is/was a "fuck all of this" PR that wasn't/isn't ready for upstreaming. Yes, I'm still pissed because you know why, I'll try to not let it influence a great project.
I hope that you still enjoy the project and I wouldn't mind discussing this more (#postmarketOS-offtopic or a private message chat?).
By * postmarketOS Bot on 2017-12-26T23:34:30
-
Created by: ollieparanoid
To upstream it into
master, it only needs to boot. As said before, the device page would be nice if you have time for it. Otherwise just tell me and I'll add it.So... shall we merge it?
By * postmarketOS Bot on 2017-12-26T23:40:26
-
Created by: lawl
Device page added and added into the feature matrix.
By * postmarketOS Bot on 2017-12-27T00:32:41