jakeday / linux-surface

Linux Kernel for Surface Devices
2.59k stars 242 forks source link

[SB2] 4.16.x not loading properly #176

Closed versipellis closed 6 years ago

versipellis commented 6 years ago

I'm having this issue 90% of the time with kernel versions 4.16.x. 4.15.16 was my last working version with 100% success rate. Upon selecting the kernel, I get the Ubuntu loading screen with the five progress dots below it, there's the audio hiss sound.. and the screen goes to black. Keyboard lights up, and I have to hard reset. Occasionally, I'll get to the login page, type in my credentials, and have it go to black. Other times, I'll boot in fine, and then it'll go to black the moment I open up something.

I tried removing all previous kernel versions, /lib/firmware/intel/ipts, /lib/firmware/i915, /lib/firmware/nvidia/gp108, and /lib/firmware/mrvl before running the sudo setup.sh and sudo dpkg -i commands. I'm guessing it's probably a display firmware issue somewhere, but can't figure out what. Been happening on Ubuntu 17.10 and now 18.04, both Wayland and xorg.

calidor commented 6 years ago

run dmesg, its intel_power_well_enable error, exact same symptoms as on my sb2:-)

versipellis commented 6 years ago

Thanks @calidor, I'll take a look later tonight. I don't seem to recall seeing intel_power_well_enable, but maybe now that I know what to look for I might spot it :)

calidor commented 6 years ago

oh and thanks for hint, ill try 4.15.16..

versipellis commented 6 years ago

So, I get the same output as in your dmesg in #173, even for 4.15.16. It loads fine, however. I get the same for 4.16.7 when it loads properly. and when it doesn't load properly.

calidor commented 6 years ago

yeah, i had to stop using it becouse after boot windows shows thermometer on boot. sb2 is not hot.. and well my guess error hes something to do with power management .. and for me its dangerous, i dont want to brick this expensive machine:)

npjohnson commented 6 years ago

@versipellis which device?

calidor commented 6 years ago

ill look tomorow at this: https://gist.github.com/Brainiarc7/aa43570f512906e882ad6cdd835efe57

versipellis commented 6 years ago

@npjohnson SB2. Sorry, forgot to mention.

calidor commented 6 years ago

mine is sb2 15inch version

versipellis commented 6 years ago

I am unable to reliably reproduce this``` I've had a 90% success rate for booting into 4.16.x this entire morning, and then this afternoon it started tanking again. I suspected that it might be due to external devices (USB, surface dock) but when I tried testing with it, it doesn't seem to improve success rates noticeably (I am suspecting mental bias in my observations)

dmesg from a failed boot has this, which is exactly as #173 - will close this issue as duplicate.

May 15 12:33:06 Hyde kernel: ------------[ cut here ]------------ May 15 12:33:06 Hyde kernel: WARN_ON(intel_cdclk_needs_modeset(&dev_priv->cdclk.hw, &cdclk_state)) May 15 12:33:06 Hyde kernel: WARNING: CPU: 2 PID: 1232 at drivers/gpu/drm/i915/intel_runtime_pm.c:720 gen9_dc_off_power_well_enable+0x1ed/0x240 [i915] May 15 12:33:06 Hyde kernel: Modules linked in: bnep btusb btrtl btbcm btintel bluetooth ecdh_generic nouveau mxm_wmi wmi ttm binfmt_misc nls_iso8859_1 dm_crypt 8250_dw in May 15 12:33:06 Hyde kernel: snd_timer hid_sensor_iio_common industrialio snd intel_lpss_pci soundcore processor_thermal_device mei_me shpchp intel_lpss intel_soc_dts_ios May 15 12:33:06 Hyde kernel: CPU: 2 PID: 1232 Comm: gnome-shell Tainted: G W 4.16.7-surface-linux-surface #63 May 15 12:33:06 Hyde kernel: Hardware name: Microsoft Corporation Surface Book 2/Surface Book 2, BIOS 387.1879.769 10/23/2017 May 15 12:33:06 Hyde kernel: RIP: 0010:gen9_dc_off_power_well_enable+0x1ed/0x240 [i915] May 15 12:33:06 Hyde kernel: RSP: 0018:ffffa93243a0b9e0 EFLAGS: 00010286 May 15 12:33:06 Hyde kernel: RAX: 0000000000000000 RBX: ffff9849d05d8000 RCX: ffffffffb2e62f08 May 15 12:33:06 Hyde kernel: RDX: 0000000000000001 RSI: 0000000000000096 RDI: 0000000000000247 May 15 12:33:06 Hyde kernel: RBP: ffffa93243a0ba00 R08: 00000000000004f4 R09: ffffffffb33484c0 May 15 12:33:06 Hyde kernel: R10: 0000000000000040 R11: 0000000000000000 R12: ffff9849d05d8000 May 15 12:33:06 Hyde kernel: R13: 0000000400000000 R14: aaaaaaaaaaaaaaab R15: 0000000000000022 May 15 12:33:06 Hyde kernel: FS: 00007feaf8e13ac0(0000) GS:ffff9849ef480000(0000) knlGS:0000000000000000 May 15 12:33:06 Hyde kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 May 15 12:33:06 Hyde kernel: CR2: 00002e0cd2017003 CR3: 000000045762e004 CR4: 00000000003606e0 May 15 12:33:06 Hyde kernel: DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 May 15 12:33:06 Hyde kernel: DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 May 15 12:33:06 Hyde kernel: Call Trace: May 15 12:33:06 Hyde kernel: intel_power_well_enable+0x40/0x50 [i915] May 15 12:33:06 Hyde kernel: __intel_display_power_get_domain+0x82/0xa0 [i915] May 15 12:33:06 Hyde kernel: intel_display_power_get+0x33/0x50 [i915] May 15 12:33:06 Hyde kernel: i915_gem_request_alloc+0x30f/0x420 [i915] May 15 12:33:06 Hyde kernel: i915_gem_do_execbuffer+0x633/0x1100 [i915] May 15 12:33:06 Hyde kernel: ? __enqueue_entity+0x5c/0x60 May 15 12:33:06 Hyde kernel: ? get_page_from_freelist+0xf14/0x1400 May 15 12:33:06 Hyde kernel: i915_gem_execbuffer2+0x1f8/0x3a0 [i915] May 15 12:33:06 Hyde kernel: ? i915_gem_execbuffer+0x2d0/0x2d0 [i915] May 15 12:33:06 Hyde kernel: drm_ioctl_kernel+0x5f/0xb0 [drm] May 15 12:33:06 Hyde kernel: drm_ioctl+0x31b/0x3d0 [drm] May 15 12:33:06 Hyde kernel: ? i915_gem_execbuffer+0x2d0/0x2d0 [i915] May 15 12:33:06 Hyde kernel: do_vfs_ioctl+0xa8/0x620 May 15 12:33:06 Hyde kernel: SyS_ioctl+0x79/0x90 May 15 12:33:06 Hyde kernel: do_syscall_64+0x73/0x130 May 15 12:33:06 Hyde kernel: entry_SYSCALL_64_after_hwframe+0x3d/0xa2 May 15 12:33:06 Hyde kernel: RIP: 0033:0x7feaf5ede5d7 May 15 12:33:06 Hyde kernel: RSP: 002b:00007ffc4a3cd3c8 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 May 15 12:33:06 Hyde kernel: RAX: ffffffffffffffda RBX: 0000561c73a01cc0 RCX: 00007feaf5ede5d7 May 15 12:33:06 Hyde kernel: RDX: 00007ffc4a3cd410 RSI: 0000000040406469 RDI: 000000000000000b May 15 12:33:06 Hyde kernel: RBP: 00007ffc4a3cd410 R08: 0000000000000001 R09: 0000561c73890400 May 15 12:33:06 Hyde kernel: R10: 0000000000000000 R11: 0000000000000246 R12: 0000000040406469 May 15 12:33:06 Hyde kernel: R13: 000000000000000b R14: 00007fead2567598 R15: 0000000000000000 May 15 12:33:06 Hyde kernel: Code: 85 d2 0f 8e d7 fe ff ff 8b 71 24 48 89 df e8 4b fa 06 00 e9 c7 fe ff ff 48 c7 c6 40 2a 2c c0 48 c7 c7 02 31 2b c0 e8 63 16 8d f1 <0f> 0b May 15 12:33:06 Hyde kernel: ---[ end trace 7f17dd1ff1f105bd ]---

jakeday commented 6 years ago

This is fixed in the 4.17.x series, which will upload here shortly.

versipellis commented 6 years ago

Ah, got it, my apologies!