**** BEGIN LOGGING AT Fri Apr 13 03:00:03 2018 Apr 13 08:04:15 cyb0hrg was added by: cyb0hrg Apr 13 08:05:38 Hello! :) So, there's a hack out there I can get grub running on the most up to date surface RT, however, I compiled a nexus 7 kernel to try, and I'm trying almost every device tree blob with no luck. I always hang (with a devicetree) and initrd, at "Jumping to entrypoint" in grub lol Apr 13 08:08:08 @mayltonfnds, it will download around 25 gb Apr 13 08:10:38 @cyb0hrg Apr 13 08:10:57 Your trying to boot a totally unrelated device with n7 kernel? Apr 13 08:12:29 True, however. I figured the vexpress devicetree or versatilepb would sort of have some sort of devugging output. Even with tegra30 device trees. I'm new to kernels an stuff. But the nexus 7 has the same processor (t30L) Apr 13 08:16:47 @cyb0hrg, Same processor is probably not enough. Mobile platforms in general and ARM more specific have no real device detection, all drivers must be present in the kernel and will probe certain addresses for hardware. There is no PCI or other hardware inventory which initializes devices assigns address and memory etc Apr 13 08:18:41 https://www.xda-developers.com/paul-jacobs-reportedly-assembling-group-buyers-take-qualcomm-private/ Apr 13 08:22:16 Gotcha, interesting. This is what I assumed the original issue was, so I started compiling a few kernels. One of them I did manually without the menuconfig. haha Apr 13 08:24:23 It's suprising that you can boot grub on the surface RT at all, especially since the secureboot hack was supposedly patched but was not. My surface (1516) fully up to date, latest image. All it takes is installing the debug policy, then, entering test signing mode, however, you hafto break the chain of trust on a usb stick, with som Apr 13 08:24:23 e type of windows shim, then it falls back to a BOOT.efi that contains grub. Apr 13 08:27:49 Hmm Apr 13 08:28:03 Do you have an existing Linux port Apr 13 08:32:53 If you can figure out the pinmux it should be doable Apr 13 08:39:57 Not sure what pinmux is. Lol that would be cool though. No existing port, however someone else was trying to compile for qemu with the torvalds main kernel branch with vexpress. Apr 13 08:42:17 As for a serial console I don't know that a pcb diagram exists. Apr 13 08:48:03 Why vexpresss Apr 13 08:48:18 You have tegra30 according to you Apr 13 08:48:41 A serial should be visible on board I imagine Apr 13 08:55:20 That's what I was wondering. I kno tegra 30 dtbs are provided in the latest kernels. I can easily disassemble the thing, it was only $30 lol The largest internal connector goes to the battery which has some sort of control board attached. That's the only place I could assume the large serial plug would be, unless, microsoft has some Apr 13 08:55:20 fancy way of flashing them using the bottom keyboard outlet Apr 13 08:57:05 Uhh Apr 13 08:57:11 Serial is 3 pins Apr 13 08:57:21 Usually test points Apr 13 08:57:37 Do you have grub source Apr 13 09:00:16 I got the grub and the debug policy from here: https://forum.xda-developers.com/windows-8-rt/rt-development/wip-secure-boot-linux-surface-rt-t3653848 Apr 13 09:04:31 The tool to enable test mode is here: https://forum.xda-developers.com/windows-8-rt/rt-development/windows-8-1-rt-jailbreak-exploit-t3226835 Apr 13 09:04:33 I suggest first locating a uart Apr 13 09:05:08 Ok cool. I might check out and see if I can find diagrams online of the board Apr 13 09:54:27 I did stir this up: … ```The shared USB Client/Host port can be used to download new software. This is normally only … required if the Bootloader does not boot anymore. … To enter the recovery mode, either connect the recovery mode pads on the front of the module … together (see picture below), pull SODIMM pin 91 to GND with a 10KOh Apr 13 09:54:27 m resistor while booting or … use a bootloader command``` … for tegra t30 colibri this makes me wonder if the tegra cardhu developer tablet has something similar. Apr 13 10:09:22 I disassembled it, there are two what look like u.fl connectors disconnected, and then there are three chips that look odd, like exposed silicon kind of that might be some type of eproom Apr 13 10:09:29 [Edit] I disassembled it, there are two what look like u.fl connectors disconnected, and then there are three chips that look odd, like exposed silicon kind of that might be some type of eprom Apr 13 10:10:34 I do see one disconnected ribbon cable that was never connected to anything, I'll see if I can grab a picture Apr 13 10:18:16 @cyb0hrg, Some chips aren't packaged but only bumped. Its normal. Apr 13 10:19:01 (Photo, 1280x960) https://irc.ubports.com/avzMYsEN.png This is really the only place I think you'd get serial, unless there is that UART jumper all the way to the left where the exposed copper is near the usb port Apr 13 10:20:59 It would be good if you could send the image as a file so it doesn't get compression artifacts Apr 13 10:27:51 Here's a shot of the entire board, there's a red blinking diode near the right-most u-fl looking port marked above Apr 13 10:27:54 (Document) https://irc.ubports.com/bweHIXQO.jpeg Apr 13 10:28:48 I might have spotted the recovery pins possibly in this one, if that's what that exposed lead is Apr 13 10:28:52 (Document) https://irc.ubports.com/v4lgZMaR.jpeg Apr 13 10:37:54 U69 pins look kinda suspicious Apr 13 10:38:07 I know lol Apr 13 10:39:31 There is one other diode on the other side of the board, its kind of hidden. But not sure what makes it light up haha Apr 13 10:43:36 There's also a bunch of smaller connections at the center of this picture towards the right, next to what looks like a diode. The fact that there's a diode on the board makes me wonder if it was used to confirm a successful programming Apr 13 10:43:40 (Document) https://irc.ubports.com/PuoTjUax.jpeg Apr 13 11:38:53 I would probe with multimeter to check voltage when booting windows Apr 13 11:39:07 Assuming that outputs to serial Apr 14 02:15:23 Could someone help me please? Apr 14 02:17:23 (Photo, 1366x768) https://irc.ubports.com/xZxRtlTV.png Apr 14 02:18:56 it's stucked here Apr 14 02:19:15 How long? Apr 14 02:19:35 around 10 minutes Apr 14 02:20:30 Check task manager, if its using resources, or its stucked Apr 14 02:20:45 I know that 2 of my repositories couldn't be synced Apr 14 02:20:53 I don't know why Apr 14 02:21:04 @gouster4, ok Apr 14 02:23:04 there are 2 processes: git and git-remote Apr 14 02:23:11 Because sometimes happends that its working on background, but output is stucked until end of process. Apr 14 02:23:22 I wonder if they are doing something Apr 14 02:23:35 @gouster4, I see Apr 14 02:23:40 @mayltonfnds, Are they using some resources (cpu/hdd/netwprk)? Apr 14 02:24:02 If you enable `--force-sync` like it says, you will overwrite your local work Apr 14 02:24:03 Ram doesnt matter, its used even when its stuck Apr 14 02:24:05 But it should continue Apr 14 02:24:28 Actually it tells you exactly what command to use to fix the problem Apr 14 02:24:40 @mayltonfnds, [Edit] Are they using some resources (cpu/hdd/network)? Apr 14 02:24:58 @gouster4, only 1% of cpu (not all the time) Apr 14 02:25:07 @UniversalSuperBox, Problem is, when he dont want loose local work. Apr 14 02:25:48 Then commit the local work and then force the sync? Apr 14 02:25:54 @UniversalSuperBox, is it dangerous? I haven't made any local change. I just synced halium n-7.1 Apr 14 02:26:23 @mayltonfnds, When u dont done local change, go for it Apr 14 02:26:56 I'm new at this thing Apr 14 02:27:10 please be patient with me guys Apr 14 02:28:35 so I have to use ./halium/devices/setup x2 --force-sync Apr 14 02:28:41 am i right? Apr 14 02:29:07 Run exactly the command that it says to run Apr 14 02:29:46 now i got you Apr 14 02:33:13 I think now everything is okay Apr 14 02:33:22 I'll go to next step Apr 14 02:33:34 thank you guys **** ENDING LOGGING AT Sat Apr 14 03:00:00 2018