search

DisplayLink / evdi

Extensible Virtual Display Interface
MIT License
702 stars 184 forks source link

Random desktop environment crash #197

Open UnoSD opened 4 years ago

UnoSD commented 4 years ago

Initially please make sure the problem hasn't been already addressed:

Some information for baseline would help a lot:

Please provide good title for the issue and description.

I refrained from updating to kernel 5.5 for a while for the well know issue (package in Arch Linux AUR still does not have the patch: https://aur.archlinux.org/packages/evdi/)

Since it was taking a while, I decided to try and update the kernel to 5.5 and used the git version of the Arch package (https://aur.archlinux.org/packages/evdi-git/).

After updating, everything seemed to work, but, more than once during the same hour of usage, the desktop crashed killing the whole session and going back to the lightdm login screen. This never happened with the old evdi package and kernel 5.4; Unplugged from the docking station with evdi and dispalylink process running, this does not happen.

If you can give us logs they will be welcome but please don't paste long listings as they will make it hard to read the thread, instead you can use other services for sharing raw text like pastebin or gist.

can you please suggest what logs to share? Xorg.log? dmesg?

Thank you

grant-traynor commented 4 years ago

+1

monofox commented 4 years ago

Have same problem on Gentoo while having connected two displaylink devices. With Linux 5.4.24-ck (ck-patches, will try stock Gentoo kernel now with 5.5.13). Xorg-server 1.20.7 In Xorg-Logs there is nothing. But i've a kernel trace (journalctl -b).

Happens especially under having loads (e.g. looking HD videos or compiling in parallel):

Mär 25 23:27:17 smallLukasGentoo kernel: ------------[ cut here ]------------
Mär 25 23:27:17 smallLukasGentoo kernel: WARN_ON(bound & ((((1UL))) << (8)))
Mär 25 23:27:17 smallLukasGentoo kernel: WARNING: CPU: 1 PID: 423226 at drivers/gpu/drm/i915/i915_vma.c:722 __i915_vma_do_pin+0x32/0x3b0
Mär 25 23:27:17 smallLukasGentoo kernel: Modules linked in: md4 cifs fscache btusb btrtl btbcm btintel uinput rfcomm nft_chain_nat nf_nat nf_log_ipv4 nf_log_common nft_reject_ipv4 nft_log nft_reject_ipv6 nft_reject nft_counter nft_ct nf_t>
Mär 25 23:27:17 smallLukasGentoo kernel: CPU: 1 PID: 423226 Comm: Xorg Tainted: G        W  O      5.4.24-ck #3
Mär 25 23:27:17 smallLukasGentoo kernel: Hardware name: TUXEDO N24_25BU/N24_25BU, BIOS 5.12 02/23/2018
Mär 25 23:27:17 smallLukasGentoo kernel: RIP: 0010:__i915_vma_do_pin+0x32/0x3b0
Mär 25 23:27:17 smallLukasGentoo kernel: Code: 55 48 89 fd 53 48 83 ec 20 4c 8b af 08 01 00 00 41 0f ba e5 08 73 20 48 c7 c6 cc e3 49 82 48 c7 c7 ad aa 4c 82 e8 1d c0 b9 ff <0f> 0b 41 bc f0 ff ff ff e9 de 00 00 00 4c 89 e8 48 89 cb 25 00 >
Mär 25 23:27:17 smallLukasGentoo kernel: RSP: 0018:ffffc90000e1ba08 EFLAGS: 00010286
Mär 25 23:27:17 smallLukasGentoo kernel: RAX: 0000000000000000 RBX: ffffc90000e1bb90 RCX: 0000000000000000
Mär 25 23:27:17 smallLukasGentoo kernel: RDX: 0000000000000000 RSI: ffffffff82c2a583 RDI: ffffffff82c2812c
Mär 25 23:27:17 smallLukasGentoo kernel: RBP: ffff8882a9a40240 R08: 000138e7ae8b728e R09: 0000000000000023
Mär 25 23:27:17 smallLukasGentoo kernel: R10: 0000000000000000 R11: 0000000000000023 R12: ffffc90000e1bb30
Mär 25 23:27:17 smallLukasGentoo kernel: R13: 00000000000003ff R14: ffff8882a9a40240 R15: 000000000000001c
Mär 25 23:27:17 smallLukasGentoo kernel: FS:  00007fc9a18f7940(0000) GS:ffff888464080000(0000) knlGS:0000000000000000
Mär 25 23:27:17 smallLukasGentoo kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Mär 25 23:27:17 smallLukasGentoo kernel: CR2: 00007f4bd888c008 CR3: 00000002165b6001 CR4: 00000000003606e0
Mär 25 23:27:17 smallLukasGentoo kernel: Call Trace:
Mär 25 23:27:17 smallLukasGentoo kernel:  ? kfree+0x17d/0x19b
Mär 25 23:27:17 smallLukasGentoo kernel:  ? do_raw_spin_lock+0x2b/0x52
Mär 25 23:27:17 smallLukasGentoo kernel:  eb_lookup_vmas+0x435/0x88b
Mär 25 23:27:17 smallLukasGentoo kernel:  ? intel_gt_terminally_wedged+0x1e/0xeb
Mär 25 23:27:17 smallLukasGentoo kernel:  i915_gem_do_execbuffer+0x65c/0x10d1
Mär 25 23:27:17 smallLukasGentoo kernel:  ? pollwake+0x56/0x70
Mär 25 23:27:17 smallLukasGentoo kernel:  ? kfree+0x17d/0x19b
Mär 25 23:27:17 smallLukasGentoo kernel:  ? slab_free_freelist_hook+0x39/0xa0
Mär 25 23:27:17 smallLukasGentoo kernel:  ? kfree+0x111/0x19b
Mär 25 23:27:17 smallLukasGentoo kernel:  ? drm_atomic_state_default_release+0x1f/0x29
Mär 25 23:27:17 smallLukasGentoo kernel:  ? slab_free_freelist_hook+0x39/0xa0
Mär 25 23:27:17 smallLukasGentoo kernel:  ? kfree+0x111/0x19b
Mär 25 23:27:17 smallLukasGentoo kernel:  ? drm_atomic_helper_dirtyfb+0x1cd/0x208
Mär 25 23:27:17 smallLukasGentoo kernel:  i915_gem_execbuffer2_ioctl+0x21d/0x2b7
Mär 25 23:27:17 smallLukasGentoo kernel:  ? i915_gem_execbuffer_ioctl+0x25e/0x25e
Mär 25 23:27:17 smallLukasGentoo kernel:  drm_ioctl_kernel+0x8d/0xe1
Mär 25 23:27:17 smallLukasGentoo kernel:  drm_ioctl+0x1f9/0x2d3
Mär 25 23:27:17 smallLukasGentoo kernel:  ? i915_gem_execbuffer_ioctl+0x25e/0x25e
Mär 25 23:27:17 smallLukasGentoo kernel:  ? selinux_file_ioctl+0x154/0x1a1
Mär 25 23:27:17 smallLukasGentoo kernel:  vfs_ioctl+0x19/0x26
Mär 25 23:27:17 smallLukasGentoo kernel:  do_vfs_ioctl+0x52a/0x554
Mär 25 23:27:17 smallLukasGentoo kernel:  ksys_ioctl+0x49/0x68
Mär 25 23:27:17 smallLukasGentoo kernel:  __x64_sys_ioctl+0x11/0x14
Mär 25 23:27:17 smallLukasGentoo kernel:  do_syscall_64+0x48/0x55
Mär 25 23:27:17 smallLukasGentoo kernel:  entry_SYSCALL_64_after_hwframe+0x44/0xa9
Mär 25 23:27:17 smallLukasGentoo kernel: RIP: 0033:0x7fc9a1fa3557
Mär 25 23:27:17 smallLukasGentoo kernel: Code: 00 00 00 75 0c 48 c7 c0 ff ff ff ff 48 83 c4 18 c3 e8 4d be 01 00 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 09 29 0d 00 f7 d8 64 89 01 >
Mär 25 23:27:17 smallLukasGentoo kernel: RSP: 002b:00007fff15b06948 EFLAGS: 00000202 ORIG_RAX: 0000000000000010
Mär 25 23:27:17 smallLukasGentoo kernel: RAX: ffffffffffffffda RBX: 00000000ffffffff RCX: 00007fc9a1fa3557
Mär 25 23:27:17 smallLukasGentoo kernel: RDX: 00007fff15b06990 RSI: 0000000040406469 RDI: 000000000000000d
Mär 25 23:27:17 smallLukasGentoo kernel: RBP: 00007fff15b06990 R08: 00000000ffffffff R09: 0000000000000000
Mär 25 23:27:17 smallLukasGentoo kernel: R10: 00007fff15b06ac0 R11: 0000000000000202 R12: 0000000040406469
Mär 25 23:27:17 smallLukasGentoo kernel: R13: 000000000000000d R14: 00005645bdde2b28 R15: 0000000000000000
Mär 25 23:27:17 smallLukasGentoo kernel: ---[ end trace 6447685242e0923b ]---

As there is so much i915 and some kernel bug trackers mention heavy issues with Kernel 5.4.x and i915

@UnoSD : may you able to check if you have also kernel traces in your kernel log? As stated above, can check in terminal with journalctl -b and than search for the word trace or scroll by date/time of crash.

grant-traynor commented 4 years ago 
Mar 26 07:09:51 GrantsThinkpad kernel: ------------[ cut here ]------------
Mar 26 07:09:51 GrantsThinkpad kernel: refcount_t: underflow; use-after-free.
Mar 26 07:09:51 GrantsThinkpad kernel: WARNING: CPU: 7 PID: 263 at lib/refcount.c:87 refcount_dec_not_one+0x50/0x60
Mar 26 07:09:51 GrantsThinkpad kernel: Modules linked in: rfcomm cmac algif_hash algif_skcipher af_alg hid_generic usbhid hid joydev mousedev evdi(OE) bnep cdc_ether usbnet snd_hda_codec_hdmi x86_pkg_temp_thermal snd_sof_pci intel_powercl>
Mar 26 07:09:51 GrantsThinkpad kernel:  intel_uncore amdgpu intel_spi snd_hwdep iwlmvm intel_rapl_perf spi_nor i915 psmouse input_leds pcspkr snd_pcm mtd i2c_i801 mac80211 snd_timer r8169 gpu_sched uvcvideo ttm realtek libarc4 libphy i2c_>
Mar 26 07:09:51 GrantsThinkpad kernel: CPU: 7 PID: 263 Comm: kworker/7:2 Tainted: G           OE     5.5.11-arch1-1 #1
Mar 26 07:09:51 GrantsThinkpad kernel: Hardware name: LENOVO 20RD0011AU/20RD0011AU, BIOS R16ET21W (1.07 ) 11/29/2019
Mar 26 07:09:51 GrantsThinkpad kernel: Workqueue: events delayed_fput
Mar 26 07:09:51 GrantsThinkpad kernel: RIP: 0010:refcount_dec_not_one+0x50/0x60
Mar 26 07:09:51 GrantsThinkpad kernel: Code: 74 f3 83 f8 01 74 27 8d 50 ff 39 c2 76 e1 80 3d 6a ea 09 01 00 75 de 48 c7 c7 c8 3b f4 95 c6 05 5a ea 09 01 01 e8 72 63 c1 ff <0f> 0b eb c7 31 c0 c3 66 0f 1f 84 00 00 00 00 00 41 55 49 89 f5 41
Mar 26 07:09:51 GrantsThinkpad kernel: RSP: 0018:ffff9f3f0044fdd0 EFLAGS: 00010286
Mar 26 07:09:51 GrantsThinkpad kernel: RAX: 0000000000000000 RBX: 0000000000080005 RCX: 0000000000000000
Mar 26 07:09:51 GrantsThinkpad kernel: RDX: 0000000000000001 RSI: 0000000000000096 RDI: 00000000ffffffff
Mar 26 07:09:51 GrantsThinkpad kernel: RBP: ffff9ac3a154c800 R08: 00000000000004ca R09: 0000000000000001
Mar 26 07:09:51 GrantsThinkpad kernel: R10: 0000000000000000 R11: 0000000000000001 R12: 0000000000000000
Mar 26 07:09:51 GrantsThinkpad kernel: R13: ffff9ac3b585e868 R14: ffff9ac3ce565a60 R15: ffff9ac3a7247a80
Mar 26 07:09:51 GrantsThinkpad kernel: FS:  0000000000000000(0000) GS:ffff9ac3d15c0000(0000) knlGS:0000000000000000
Mar 26 07:09:51 GrantsThinkpad kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Mar 26 07:09:51 GrantsThinkpad kernel: CR2: 000055d57a96a0e8 CR3: 0000000437bd2001 CR4: 00000000003606e0
Mar 26 07:09:51 GrantsThinkpad kernel: Call Trace:
Mar 26 07:09:51 GrantsThinkpad kernel:  refcount_dec_and_mutex_lock+0x13/0x70
Mar 26 07:09:51 GrantsThinkpad kernel:  drm_gem_object_put_unlocked+0x2e/0x80 [drm]
Mar 26 07:09:51 GrantsThinkpad kernel:  drm_gem_dmabuf_release+0x16/0x20 [drm]
Mar 26 07:09:51 GrantsThinkpad kernel:  dma_buf_release+0x4e/0xd0
Mar 26 07:09:51 GrantsThinkpad kernel:  __fput+0xae/0x230
Mar 26 07:09:51 GrantsThinkpad kernel:  delayed_fput+0x1f/0x30
Mar 26 07:09:51 GrantsThinkpad kernel:  process_one_work+0x1e1/0x3d0
Mar 26 07:09:51 GrantsThinkpad kernel:  worker_thread+0x4a/0x3d0
Mar 26 07:09:51 GrantsThinkpad kernel:  kthread+0xfb/0x130
Mar 26 07:09:51 GrantsThinkpad kernel:  ? process_one_work+0x3d0/0x3d0
Mar 26 07:09:51 GrantsThinkpad kernel:  ? kthread_park+0x90/0x90
Mar 26 07:09:51 GrantsThinkpad kernel:  ret_from_fork+0x1f/0x40
Mar 26 07:09:51 GrantsThinkpad kernel: ---[ end trace 2138961c7247be62 ]---
grant-traynor commented 4 years ago

Are these helpful?

Mar 26 11:55:18 GrantsThinkpad kernel: ------------[ cut here ]------------
Mar 26 11:55:18 GrantsThinkpad kernel: refcount_t: underflow; use-after-free.
Mar 26 11:55:18 GrantsThinkpad kernel: WARNING: CPU: 6 PID: 528 at lib/refcount.c:87 refcount_dec_not_one+0x50/0x60
Mar 26 11:55:18 GrantsThinkpad kernel: Modules linked in: rfcomm cmac algif_hash algif_skcipher af_alg joydev mousedev bnep hid_generic uvcvideo btusb videobuf2_vmalloc videobuf2_memops btrtl videobuf2_v4l2 btbcm videobuf2_common usbhid b>
Mar 26 11:55:18 GrantsThinkpad kernel:  intel_cstate nls_iso8859_1 intel_uncore snd_hda_core nls_cp437 ofpart vfat fat iwlmvm i915 intel_rapl_perf cmdlinepart snd_hwdep intel_spi_pci intel_spi snd_pcm spi_nor psmouse input_leds pcspkr mac>
Mar 26 11:55:18 GrantsThinkpad kernel: CPU: 6 PID: 528 Comm: kworker/6:3 Tainted: G           OE     5.5.11-arch1-1 #1
Mar 26 11:55:18 GrantsThinkpad kernel: Hardware name: LENOVO 20RD0011AU/20RD0011AU, BIOS R16ET21W (1.07 ) 11/29/2019
Mar 26 11:55:18 GrantsThinkpad kernel: Workqueue: events delayed_fput
Mar 26 11:55:18 GrantsThinkpad kernel: RIP: 0010:refcount_dec_not_one+0x50/0x60
Mar 26 11:55:18 GrantsThinkpad kernel: Code: 74 f3 83 f8 01 74 27 8d 50 ff 39 c2 76 e1 80 3d 6a ea 09 01 00 75 de 48 c7 c7 c8 3b b4 a7 c6 05 5a ea 09 01 01 e8 72 63 c1 ff <0f> 0b eb c7 31 c0 c3 66 0f 1f 84 00 00 00 00 00 41 55 49 89 f5 41
Mar 26 11:55:18 GrantsThinkpad kernel: RSP: 0018:ffffaa2b80a63dd0 EFLAGS: 00010286
Mar 26 11:55:18 GrantsThinkpad kernel: RAX: 0000000000000000 RBX: 0000000000080005 RCX: 0000000000000000
Mar 26 11:55:18 GrantsThinkpad kernel: RDX: 0000000000000001 RSI: 0000000000000096 RDI: 00000000ffffffff
Mar 26 11:55:18 GrantsThinkpad kernel: RBP: ffff95af1fe0a400 R08: 00000000000004c6 R09: 0000000000000001
Mar 26 11:55:18 GrantsThinkpad kernel: R10: 0000000000000000 R11: 0000000000000001 R12: 0000000000000000
Mar 26 11:55:18 GrantsThinkpad kernel: R13: ffff95af4d422868 R14: ffff95af4e4a0660 R15: ffff95af272d9000
Mar 26 11:55:18 GrantsThinkpad kernel: FS:  0000000000000000(0000) GS:ffff95af51580000(0000) knlGS:0000000000000000
Mar 26 11:55:18 GrantsThinkpad kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Mar 26 11:55:18 GrantsThinkpad kernel: CR2: 00007f2be4ff6d68 CR3: 000000033760a002 CR4: 00000000003606e0
Mar 26 11:55:18 GrantsThinkpad kernel: Call Trace:
Mar 26 11:55:18 GrantsThinkpad kernel:  refcount_dec_and_mutex_lock+0x13/0x70
Mar 26 11:55:18 GrantsThinkpad kernel:  drm_gem_object_put_unlocked+0x2e/0x80 [drm]
Mar 26 11:55:18 GrantsThinkpad kernel:  drm_gem_dmabuf_release+0x16/0x20 [drm]
Mar 26 11:55:18 GrantsThinkpad kernel:  dma_buf_release+0x4e/0xd0
Mar 26 11:55:18 GrantsThinkpad kernel:  __fput+0xae/0x230
Mar 26 11:55:18 GrantsThinkpad kernel:  delayed_fput+0x1f/0x30
Mar 26 11:55:18 GrantsThinkpad kernel:  process_one_work+0x1e1/0x3d0
Mar 26 11:55:18 GrantsThinkpad kernel:  worker_thread+0x4a/0x3d0
Mar 26 11:55:18 GrantsThinkpad kernel:  kthread+0xfb/0x130
Mar 26 11:55:18 GrantsThinkpad kernel:  ? process_one_work+0x3d0/0x3d0
Mar 26 11:55:18 GrantsThinkpad kernel:  ? kthread_park+0x90/0x90
Mar 26 11:55:18 GrantsThinkpad kernel:  ret_from_fork+0x1f/0x40
Mar 26 11:55:18 GrantsThinkpad kernel: ---[ end trace 6924403fbc83f670 ]---
grant-traynor commented 4 years ago

Would love to help if I can get an idea of the best place to start? :-)

UnoSD commented 4 years ago

@UnoSD : may you able to check if you have also kernel traces in your kernel log? As stated above, can check in terminal with journalctl -b and than search for the word trace or scroll by date/time of crash.

@monofox I just tried, but unfortunately it has been purged from the logs, it happened last time on the 23rd and then I decided to roll back to 5.4.

monofox commented 4 years ago

All given stack traces here have to do with DRM so likeley evdi has some incompatibilities with kernel? Although don't get any traces with kernel 5.5.13, but had a smaller interval until next crash of display manager.

So switched back to kernel 5.2.3-ck to see if the crashes appears there as well for me.

@grant-traynor : does the issues appear for you on earlier kernel releases? You're right now using 5.5.11.

grant-traynor commented 4 years ago

Hi @monofox,

I'm happy to give it a crack. I've been using arch for about 5 years now and have not had a need to build a kernel since using it. So I'm pretty well out of practice in general, and have never done it on arch at all. Do you have a recommendation for the best method to do that? https://wiki.archlinux.org/index.php/Kernel/Arch_Build_System ?

monofox commented 4 years ago

Hi @grant-traynor, i know only the gentoo way, which mostly should work for all operation systems until some specific routines regarding initramfs is required. Download the kernel from https://www.kernel.org/ in the wanted version, take my current kernel configuration file, make oldconfig, make and make modules_install / make install, provide the initramfs file in /boot and regenerate the grub configuration with grub-mkconfig (the "easy" way of gentoo are two commands [emerge to download the sources and genkernel to do the rest). For Arch seems best to go the way, as described in your linked page. Of course, afterwards evdi must be recompiled.

I tested during the day my old kernel 5.2.3-ck which worked without any crash with same xorg, same evdi version. There was couple of minutes ago only one issue, that displaylink docking station freezed (no ethernet anymore, no usb, images were frozen). So this actually does not have to do with evdi.

grant-traynor commented 4 years ago

OK. I'm running 5.2.3 now. I'll let you know how it goes.

 ~  $  uname -a
Linux GrantsThinkpad 5.2.3-arch1-1-gkt-5-2-3 #1 SMP PREEMPT Sat, 28 Mar 2020 23:31:38 +0000 x86_64 GNU/Linux
grant-traynor commented 4 years ago

So I've been running in the 5.2.3 kernel all day and it seems "rock solid". No issues. Good in fact. I'll stick on this version and report back if it fails, but I think you're onto something @monofox

abucodonosor commented 4 years ago

@monofox @UnoSD @grant-traynor

Are you all on Intel iGPUs?

monofox commented 4 years ago

@abucodonosor , yes: 00:02.0 VGA compatible controller: Intel Corporation HD Graphics 620 (rev 02)

UnoSD commented 4 years ago

@monofox Same here: Intel Corporation UHD Graphics 620 (Whiskey Lake)

grant-traynor commented 4 years ago

@abucodonosor Yes: Me too.

 ~  $  lspci
00:00.0 Host bridge: Intel Corporation Device 9b61 (rev 0c)
00:02.0 VGA compatible controller: Intel Corporation UHD Graphics (rev 02)
00:04.0 Signal processing controller: Intel Corporation Xeon E3-1200 v5/E3-1500 v5/6th Gen Core Processor Thermal Subsystem (rev 0c)
00:08.0 System peripheral: Intel Corporation Xeon E3-1200 v5/v6 / E3-1500 v5 / 6th/7th/8th Gen Core Processor Gaussian Mixture Model
00:12.0 Signal processing controller: Intel Corporation Comet Lake Thermal Subsytem
00:14.0 USB controller: Intel Corporation Device 02ed
00:14.2 RAM memory: Intel Corporation Device 02ef
00:14.3 Network controller: Intel Corporation Wireless-AC 9462
00:16.0 Communication controller: Intel Corporation Comet Lake Management Engine Interface
00:17.0 SATA controller: Intel Corporation Comet Lake SATA AHCI Controller
00:1c.0 PCI bridge: Intel Corporation Device 02b8 (rev f0)
00:1c.4 PCI bridge: Intel Corporation Device 02bc (rev f0)
00:1d.0 PCI bridge: Intel Corporation Device 02b0 (rev f0)
00:1d.4 PCI bridge: Intel Corporation Device 02b4 (rev f0)
00:1f.0 ISA bridge: Intel Corporation Device 0284
00:1f.3 Audio device: Intel Corporation Device 02c8
00:1f.4 SMBus: Intel Corporation Device 02a3
00:1f.5 Serial bus controller [0c80]: Intel Corporation Comet Lake SPI (flash) Controller
03:00.0 Display controller: Advanced Micro Devices, Inc. [AMD/ATI] Lexa [Radeon 540X/550X/630 / RX 640 / E9171 MCM] (rev c0)
04:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller (rev 10)
07:00.0 Non-Volatile memory controller: Samsung Electronics Co Ltd Device a809
 ~  $  lsusb
Bus 002 Device 005: ID 17ef:a357 Lenovo 
Bus 002 Device 004: ID 17ef:a359 Lenovo 
Bus 002 Device 003: ID 17e9:6015 DisplayLink USB3.1 Hub             
Bus 002 Device 002: ID 17ef:a356 Lenovo USB3.1 Hub             
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 001 Device 004: ID 27c6:55a4 Shenzhen Goodix Technology Co.,Ltd. Goodix FingerPrint Device
Bus 001 Device 003: ID 5986:2130 Acer, Inc Integrated Camera
Bus 001 Device 008: ID 046d:0a4d Logitech, Inc. G430 Surround Sound Gaming Headset
Bus 001 Device 010: ID 17ef:609b Lenovo USB2.0 Hub             
Bus 001 Device 009: ID 17ef:a354 Lenovo 
Bus 001 Device 007: ID 1a40:0101 Terminus Technology Inc. Hub
Bus 001 Device 005: ID 17ef:1029 Lenovo 
Bus 001 Device 002: ID 17ef:1028 Lenovo USB2.0 Hub             
Bus 001 Device 006: ID 8087:0026 Intel Corp. 
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
grant-traynor commented 4 years ago

No crashes today also.

Linux GrantsThinkpad 5.2.3-arch1-1-gkt-5-2-3 #1 SMP PREEMPT Sat, 28 Mar 2020 23:31:38 +0000 x86_64 GNU/Linux
abucodonosor commented 4 years ago

Those bugs look to me like Intel drm bugs, especially after they did the eb_* rework. There are still bug reported all over the place about gnome-shell and gdm breaking.

Is someone willing to test a vanilla 5.6 kernel? If so let me know since evdi need a patch for 5.6 to work.

loiselqw commented 4 years ago

Same here at 5.4.0-4-amd64 #1 SMP Debian 5.4.19-1 (2020-02-13) x86_64 GNU/Linux

X-Session (xfce4 and lxqt) crashed randomly - I suppose especially in Browser activity (heavily load) and brought back to lightdm login screen.

HW is thinkpad x260 with following lspci-output:

00:00.0 Host bridge: Intel Corporation Xeon E3-1200 v5/E3-1500 v5/6th Gen Core Processor Host Bridge/DRAM Registers (rev 08) 00:02.0 VGA compatible controller: Intel Corporation Skylake GT2 [HD Graphics 520] (rev 07) 00:14.0 USB controller: Intel Corporation Sunrise Point-LP USB 3.0 xHCI Controller (rev 21) 00:14.2 Signal processing controller: Intel Corporation Sunrise Point-LP Thermal subsystem (rev 21) 00:16.0 Communication controller: Intel Corporation Sunrise Point-LP CSME HECI #1 (rev 21) 00:17.0 SATA controller: Intel Corporation Sunrise Point-LP SATA Controller [AHCI mode] (rev 21) 00:1c.0 PCI bridge: Intel Corporation Sunrise Point-LP PCI Express Root Port #1 (rev f1) 00:1c.2 PCI bridge: Intel Corporation Sunrise Point-LP PCI Express Root Port #3 (rev f1) 00:1f.0 ISA bridge: Intel Corporation Sunrise Point-LP LPC Controller (rev 21) 00:1f.2 Memory controller: Intel Corporation Sunrise Point-LP PMC (rev 21) 00:1f.3 Audio device: Intel Corporation Sunrise Point-LP HD Audio (rev 21) 00:1f.4 SMBus: Intel Corporation Sunrise Point-LP SMBus (rev 21) 00:1f.6 Ethernet controller: Intel Corporation Ethernet Connection I219-LM (rev 21) 02:00.0 Unassigned class [ff00]: Realtek Semiconductor Co., Ltd. RTS522A PCI Express Card Reader (rev 01) 04:00.0 Network controller: Intel Corporation Wireless 8260 (rev 3a) logs after crash are attached:

user.log Mar 27 10:03:26 foo xfce4-notifyd[4507]: xfce4-notifyd: Fatal IO error 11 (Die Ressource ist zur Zeit nicht verfügbar) on X server :0. Mar 27 10:03:28 foo lightdm[36441]: Error getting user list from org.freedesktop.Accounts: GDBus.Error:org.freedesktop.DBus.Error.ServiceUnknown: The name org.freedesktop.Accounts was not provided by any .service files Mar 27 10:03:43 foo lightdm[36632]: Error getting user list from org.freedesktop.Accounts: GDBus.Error:org.freedesktop.DBus.Error.ServiceUnknown: The name org.freedesktop.Accounts was not provided by any .service files Mar 27 10:03:43 foo lightdm[801]: g_dbus_connection_call_sync_internal: assertion 'G_IS_DBUS_CONNECTION (connection)' failed Mar 27 10:40:22 foo org.freedesktop.thumbnails.Thumbnailer1[47642]: Registered thumbnailer ffmpegthumbnailer -i %i -o %o -s %s -f

If I could do something more, pls let me know.

monofox commented 4 years ago

@abucodonosor : sure, i'm willing to test.

abucodonosor commented 4 years ago

@monofox I noticed @displaylink-dkurek already added a fix for 5.6 kernels in git.

This commit https://github.com/DisplayLink/evdi/commit/34d92e8a78a7a3b8df848b61469c783f352e360c. It is similar to my fix just with <= logic :-)

monofox commented 4 years ago

@abucodonosor : Kernel is configured, compiled, installed: Linux smallLukasGentoo 5.6.0 #1 SMP Tue Mar 31 14:51:37 CEST 2020 x86_64 Intel(R) Core(TM) i7-7500U CPU @ 2.70GHz GenuineIntel GNU/Linux Corresponding devel-branch of evdi is compiled and installed. Both evdi connected monitors are attached and working. So lets see how stable it is :)

monofox commented 4 years ago

Was a short test. Desktop environment crashed after round about 30 minutes. Pasted log to https://gist.github.com/monofox/96e62d9373a10db636c41b41978567ec Xorg.0.log at https://gist.github.com/monofox/953c891f59375d4c150fac88c6cfdc45

abucodonosor commented 4 years ago

@monofox

nothing interesting in the logs. Can you maybe use journalctl or similar to get a full dmesg?

abucodonosor commented 4 years ago

@monofox

also have a look at https://github.com/DisplayLink/evdi/issues/179 and https://github.com/DisplayLink/evdi/issues/173

monofox commented 4 years ago

@abucodonosor

179 does not affect me (messages and behaviour does not fit)

173 also referes to some hang or freeze, whereas messages and behaviour do not fit to my system.

Full log is here: https://gist.github.com/monofox/9199cb2a5a844c9c706c621b60f6c91b Relevant part starts at Mär 31 16:31:58 where the kernel module was loaded and Xserver restarted. On 16:32:08 display server was terminated and restarted intentionally by me. On 16:32:36 i've attached the DisplayLink docking station. You can ignore messages regarding NetworkManager - wifi fails always due to broken Intel wifi support in 5.6.0 and finally at 16:56:49 the xsession crashed. I don't have any coredump in coredumpctl list from today.

I didn't worked last two hours (so in last two hours nothing happened). But i'll stick a while for 5.6 and check how its going.

abucodonosor commented 4 years ago

@monofox

Still nothing interesting from edvi there. But the interesting part starts with you running out-of-date firmware for your iGPU :-). Please update linux-firmware package.

Well, #179 & 173 are all about those bugs, regarding the messages.

grant-traynor commented 4 years ago

Putting this here for future reference ==> dkms install evdi/1.6.2 -k 5.6.0-arch1-1-gkt-5-6

grant-traynor commented 4 years ago

Ok. I'm using this version today. Did not get time with it yesterday due to work commitments. I'll let you know how it goes.

Linux GrantsThinkpad 5.6.0-arch1-1-gkt-5-6 #2 SMP PREEMPT Tue, 31 Mar 2020 21:06:06 +0000 x86_64 GNU/Linux
monofox commented 4 years ago

@abucodonosor : Package was up-to-date. Just my initramfs not. I've updated it. And made two tests: one with GUC enable and one with GUC disabled. In both cases, crash happens: https://gist.github.com/monofox/5a15cd0672dc6eacc4e0c1a987aceeb0 (same behaviour and unfortunately, the logs do not reveal any (for me) helpful).

grant-traynor commented 4 years ago

OK. Might be a bit premature, but I've been using this version for the whole morning with no issues, doing the sorts of things that would normally cause the crash (Videoconferencing, flowcharting and graphical stuff, Moving windows between the laptop and the two attached screens). It seems a little slower though?

For reference:

  1. Using awesomewm with predominantly chromium, Zoom, slack (thick client) and termite / terminals
  2. Installed evdi as a dkms module using AUR evdi-git
  3. Built the 5.6 kernel using the methods defined in https://wiki.archlinux.org/index.php/Kernel/Arch_Build_System yesterday afternoon.
Linux GrantsThinkpad 5.6.0-arch1-1-gkt-5-6 #2 SMP PREEMPT Tue, 31 Mar 2020 21:06:06 +0000 x86_64 GNU/Linux

dkms install evdi/1.6.2 -k 5.6.0-arch1-1-gkt-5-6
grant-traynor commented 4 years ago

A couple of crashes now. I'm going to head back to 5.2.3 to see if it remains solid. Snips from journalctl -b below. I've kept the log and can dig more out of it if it helps. These don't look the same as previous logs, maybe not evdi any more?

dump1.txt dump2.txt

It definitely feels slower too on 5.6.

grant-traynor commented 4 years ago

Back on 5.2.3 and no problems. Snappier also. I haven't really been exposed to this driver infrastructure, would be interested to hear from someone who know's what's going on in kernel land as to what's going on. A theory?

abucodonosor commented 4 years ago

Yes buggy Intel driver. And yes 5.2.x seems to be a working kernel for Intel iGPUs.

See as example: https://marc.info/?l=linux-kernel&m=158551437904739&w=2 https://gitlab.freedesktop.org/drm/intel/-/issues/1585

Ofc lots more there, https://gitlab.freedesktop.org/drm/intel/issues/

Your crashes are from iwlwifi which is fixed in 5.6.2 and Xorg crashing in modeseting driver.

In any way, I do not own such HW, and I do not think is related to evdi. Either you guys report these to Intel people or someone is trying to bisect 5.2 -> 5.3++ kernel and find the broken commits.

Zetabite commented 4 years ago

Nope it still crashes to login screen for me. 5.6.2-arch1-2

grant-traynor commented 4 years ago

https://gitlab.freedesktop.org/drm/intel/-/issues/1667

I raised this to give the intel/drm guys a heads up. They're probably already across it.

Trying to help, tell me to go away if not.

monofox commented 4 years ago

Tested earlier also with 5.6.3.; after 20 minutes, its crashing. Based on the intel gitlab issue tracker, i've figured out some debug parameters drm.debug=0x1e log_buf_len=4M, producing huge log, but actually did not really helped me. I will later provide it to the ticket created by @grant-traynor . I'm working in parallel two ways right now: (1) part about drm to figure out, to get any kind of kernel messages, error messages through proc, etc. and (2) figure out on XServer side, why its terminating, but not directly "crashing" (there is no coredump, or similar and based on logs, no graceful termination).

grant-traynor commented 4 years ago

Yeah, just to put the icing on this, 5.2.3 doesn't appear to crash at all. It's more performant, and rock solid stable.

grant-traynor commented 4 years ago

I've been running this version all day today.

5.6.4-arch1-1 #1 SMP PREEMPT Mon, 13 Apr 2020 12:21:19 +0000 x86_64 GNU/Linux

Things have definitely improved, not sure if I'm jumping the gun. But it seems to be back to it's good old 5.2.3 self again, maybe even a little tiny bit better. No crashes so far, not holding back in terms of workload. Hope that this is the light at the end of the tunnel.

deimosfr commented 4 years ago

I wanted to wait a little bit more before posting a comment but I agree that it's really better. Almost 2 days without any crash as well on my side with 5.6.4

grant-traynor commented 4 years ago

So I've gone back to 5.3.2 for now. 5.6.4 is definitely better, but still not really done. I've also noticed that DisplayLinkManager hangs on reboot, possibly as a result of a previous error the reason for which I rebooted.

@deimosfr are you still on 5.6.4?

I get the impression that there's a lot of work going on, and that things might naturally settle down as that work matures?

deimosfr commented 4 years ago

@grant-traynor since 5 days, I got only 1 crash. So I agree the work is not completed. To confirm my setup:

0k00l commented 4 years ago

@grant-traynor since 5 days, I got only 1 crash. So I agree the work is not completed. To confirm my setup:

* I recently updated to 5.6.5-arch3-1 (and saw 5.6.6 is out as well)

* Installed evdi from devel branch
  Hope this can help

I've tried this - it wont help you

Zetabite commented 4 years ago

can confirm this for pre 5.6.4 kernel and maybe even add that it freezed with 5.6.4 and some before.

loiselqw commented 4 years ago

I am running 5.5.0-2-amd64 on debian testing (bullseye). Any recommondations from you to run the fix on such distribution?

grant-switchdin commented 4 years ago

Just tried 5.6.7 and it crashed within about 15 minutes. Didn't grab the log, work is killing me at the moment. Arrggh. What on earth is going on in the intel driver space?

0k00l commented 4 years ago

Just tried 5.6.7 and it crashed within about 15 minutes. Didn't grab the log, work is killing me at the moment. Arrggh. What on earth is going on in the intel driver space?

As far as i know logs for displaylink hardware are binary :) - happy reading ;)

Zetabite commented 4 years ago

@grant-switchdin i have no idea either. DisplayLink doesnt work at all for me now.

grant-switchdin commented 4 years ago

@0k00l .. honestly dude ... you think you're the only person who knows what's going on here? How does this help?

0k00l commented 4 years ago

@0k00l .. honestly dude ... you think you're the only person who knows what's going on here? How does this help?

Saving your time (i just informed you that logs are binary )

0k00l commented 4 years ago

Some improvements with newest evdi(devel branch) with displaylink 5.3.1. It take much more time to crash and crash itself doesn't logout it shutdown screen (screen make itself black) . Only restart helps ... so it still sucks but a little bit less.