<gavodavo> <DavidHeidelberg "at least serial console? :D"> Nah havent
<gavodavo> > <@okias:matrix.org> at least serial console? :D
<gavodavo> * Nah haven't even tested booting
<gavodavo> However even if I do, I don't have a soldering iron for making a uart cable
marvin24 has quit [Ping timeout: 272 seconds]
marvin24_ has joined #tegra
tommyg has quit [Quit: Client closed]
<lahvuun> I tried booting the latest grate kernel with fastboot on nexus 7 grouper and it doesn't work. There is nothing on the screen, no backlight either. Am I missing something?
<DavidHeidelberg> lahvuun can you do this and reboot into some working grate kernel and check ramoops?
<DavidHeidelberg> you'll see in dmesg, what's wrong
<lahvuun> sure
<DavidHeidelberg> s/ramoops/pstore/ :)
<DavidHeidelberg> (old Android habbits :P )
<lahvuun> it's odd, looks like it's booting fine
<lahvuun> I don't see any errors and the log ends with "No errors detected"
<lahvuun> am I supposed to copy the kernel modules to /lib/modules for the screen to work?
<lahvuun> so yeah, it actually boots properly but there I think the screen is off
<DavidHeidelberg> lahvuun so no additional errors/warnings against latest kernel?
<lahvuun> looks like it
<lahvuun> there isn't anything related to the screen in dmesg
<lahvuun> or at least I don't see it
<lahvuun> is there any way I can try to turn the screen on?
<jenneron[m]> lahvuun: is drm probed?
<jenneron[m]> host1x*
<lahvuun> I'm not sure what you mean. How do I check it?
<digetx> lahvuun: which kernel .config did you use?
<digetx> or defconfig
<lahvuun> I used tegra_defconfig
<digetx> lahvuun: I got the same result using tegra_defconfig, will take a closer look later today
<lahvuun> thanks!
<digetx> CONFIG_FB is disabled now by default
<digetx> I think CONFIG_FRAMEBUFFER_CONSOLE previous was selecting CONFIG_FB
<digetx> now framebuffer is completely gone in tegra_defconfig
<lahvuun> I enabled CONFIG_FRAMEBUFFER_CONSOLE and that fixed it
<digetx> ok
<lahvuun> thank you again!
<digetx> I'll make patch to fix it later today, thank you for catching this
<leanderglanda[m]> I dont think you saw my message earlier, as it was too long
<leanderglanda[m]> I'm just giving this info, maybe someone else has issues:
<leanderglanda[m]> jenneron, crackthesurface and me all compiled grate-linux, so kernel version 5.13.something, and both of the other guys complained that the kernel didn't work.
<leanderglanda[m]> So I investigated with UART. Things I noticed from the log:
<leanderglanda[m]> no mention of host1x at all. Screen just shuts off (from boot logo) (config contains CONFIG_GRATE_HOST1X=y)
<leanderglanda[m]> rootwait needs to be added, because mmc gets detected so late. Before only rootwait for USB as rootfs was required.
<leanderglanda[m]> * So I investigated with UART. Things I noticed from the log:
<leanderglanda[m]> * no mention of host1x at all. Screen just shuts off (from boot logo) (config contains CONFIG_GRATE_HOST1X=y)
<leanderglanda[m]> * rootwait needs to be added, because mmc gets detected so late. Before only rootwait for USB as rootfs was required.
<leanderglanda[m]> * - no mention of host1x at all. Screen just shuts off (from boot logo) (config contains CONFIG_GRATE_HOST1X=y & CONFIG_FB=y & CONFIG_DRM_PANEL_LVDS=y)
<leanderglanda[m]> Something broke the display. 🤷
<lahvuun> isn't this what my issue was?
<lahvuun> in any case, CONFIG_FRAMEBUFFER_CONSOLE fixed it for me
<leanderglanda[m]> CONFIG_FRAMEBUFFER_CONSOLE=y
<leanderglanda[m]> Already tried, doesn't work either
<leanderglanda[m]> Maybe this helps
<digetx> failed to get resets for 3d1: -16
<leanderglanda[m]> but why is that, display worked previously 🤔
<digetx> when
<leanderglanda[m]> 5.12.1, not sure when the last grate linux worked, but that was the last kernel I compiled and worked without issues
<digetx> upstream kernel doesn't use memory resets yet
<leanderglanda[m]> OK, I see. So grate-linux won't work for surfaceRT, yet?
<leanderglanda[m]> But any release from kernel.org should work then. Did I understand that correctly?
<jenneron[m]> any except 5.13+ I assume
<leanderglanda[m]> OK, thank you very much 👍️
<digetx> I'm wondering whether powergates could be controlled by firmware somehow
<digetx> but unlikely
<digetx> leanderglanda[m]: please try https://dpaste.com/534QAWNMQ
torez has quit [Quit: torez]
torez has joined #tegra
<digetx> some odd tegra3 variants may have a single 3d unit, but I'm pretty that's not the case of the surfaceRT
<leanderglanda[m]> <digetx "leanderglanda: please try https:"> This is with this patch
<leanderglanda[m]> I will modify my device tree quickly, jenneron did some things with it, maybe that resovles the lvds regulator issue
<jenneron[m]> I can update my PR
<jenneron[m]> once we find a fix
<leanderglanda[m]> This is with my "improved" device tree
<leanderglanda[m]> jenneron: I think something is wrong with the tegra30-microsoft-surfacert-efi.dts
<jenneron[m]> just replaces PMIC with fixed regulator and rtc
<jenneron[m]> leanderglanda anything else I need to change?
<digetx> leanderglanda[m]: you should update grate kernel to fix the usb irq
<jenneron[m]> well, update your local git to my PR (it is rebased to latest grate) and try digetx's patches
<leanderglanda[m]> I'm up-to-date with your pr
<digetx> the tps65911 node in device tree should describe all tps regulators, or driver won't load
<leanderglanda[m]> Problem is that I have to remove this node for EFI boot. Firmware blocks pmic. ACPI is required for that.
<digetx> it looks like the node is node removed then
<digetx> *is not
<digetx> yeah, it doesn't remove anything
<digetx> iiuc
<jenneron[m]> well, earlier logs from Leander don't have tps probing
<leanderglanda[m]> This is the dts I used with my latest log
<jenneron[m]> can you upload on dpaste
<jenneron[m]> I can't open it on phone
<digetx> ah
<jenneron[m]> doesn't seem to be music :P
<leanderglanda[m]> https://dpaste.org/v31B
<digetx> so you're not using jenneron's DT
<leanderglanda[m]> With the latest 2 attemps no, as the dt doesn't seem correct. I will help him update, once I know what to change
<jenneron[m]> yeah this one includes PMIC
<jenneron[m]> btw I tested my dtb with 5.12 kernel: display, emmc, usb work
<jenneron[m]> so dts is correct
<jenneron[m]> you can test it yourself as well
<digetx> it should work with the fixed core regulator and MC 3d reset
<jenneron[m]> I assume it's not general issue, but EFI booting issue?
<jenneron[m]> once we are sure I can update my -efi dts
<jenneron[m]> but I can't test anything right now
<jenneron[m]> *myself
<digetx> I'm not sure about the 3d1, I had a lot of problems with it before in general
<digetx> but we can fix it anyways, it won't be a problem
<digetx> I'll pushes fixes to grate kernel later today
<digetx> maybe power domains are turned off by default on surfaceRT, that could make a difference
<digetx> but 5.12 worked for you, so probably not
<digetx> we will need to check the boot state of 3d domain
<jenneron[m]> I don't have soldered UART, so only Leander can test until we have working display
gouchi has joined #tegra
<leanderglanda[m]> digetx: I got screen to work. Applied your two patches, and used the most recent version of jenneron's pull request
<leanderglanda[m]> https://dpaste.org/X5we
<leanderglanda[m]> This is the log
<digetx> ok
gouchi has quit [Remote host closed the connection]
ServerStatsDisco has quit [Write error: Connection reset by peer]
kusma has quit [Remote host closed the connection]
pangelo[m] has quit [Remote host closed the connection]
leanderglanda[m] has quit [Remote host closed the connection]
maxnet[m] has quit [Remote host closed the connection]
maxim[m] has quit [Remote host closed the connection]
nergzd723[m] has quit [Remote host closed the connection]
gavodavo has quit [Remote host closed the connection]
DavidHeidelberg has quit [Remote host closed the connection]
ivoszbg[m] has quit [Remote host closed the connection]
jenneron[m] has quit [Remote host closed the connection]
aat596[m] has quit [Remote host closed the connection]
DavidHeidelberg has joined #tegra
ServerStatsDisco has joined #tegra
nergzd723[m] has joined #tegra
ivoszbg[m] has joined #tegra
jenneron[m] has joined #tegra
pangelo[m] has joined #tegra
maxim[m] has joined #tegra
maxnet[m] has joined #tegra
leanderglanda[m] has joined #tegra
aat596[m] has joined #tegra
kusma has joined #tegra
gavodavo has joined #tegra