PDA

View Full Version : "EDID Invalid" On Xubuntu 16.04


OIKgnPdgNWHf
05-18-2016, 01:10 PM
Display worked at first, but after a few reboots now dmesg reports this error and the display is no longer working.


[D] evdi_detect (dev=1) Painter is connected
[D] evdi_painter_get_edid (dev=1) 54 c4 ff
evdi evdi.0: DVI-I-1: EDID invalid.


This is an HP Envy laptop. While I appreciate there is a support tool to dump more info, this is a work computer which cannot have its debug info publicly disclosed.

mlukaszek
05-18-2016, 01:49 PM
The information you provided indicates that the EDID is indeed invalid. The first three bytes that are printed out should be 00 ff ff.
What monitor is it? Does it work without problems if it's connected directly to any PC? Can you dump an EDID from it and attach here?

Cheers,
Michal

OIKgnPdgNWHf
05-19-2016, 06:00 PM
Monitor is an Acer R240HY which works fine connected directly to the HDMI port on the same computer. FWIW, now dmesg reports 3 different bytes for the EDID: "54 64 52".

mlukaszek
05-20-2016, 05:24 PM
This is interesting. Any chance you could send us at least DL logs from /var/log/displaylink?

Cheers,
Michal

OIKgnPdgNWHf
06-03-2016, 12:04 PM
Today when I booted the driver produced a SIGSEGV ("Cannot access memory at address") error without a monitor plugged into the adapter. So I tried a different monitor of the same model with the adapter and got the following EDID:

Raw EDID:
54 d4 7f 0b 2e 7f 00 00 00 01 00 00 2d 7f 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00


The other monitor also produced a mostly null/zero EDID like the one above.

DisplayLink version is 1.1.68. I recall trying 1.0.335 first; not sure if that is a factor.

OIKgnPdgNWHf
06-03-2016, 12:14 PM
This is interesting. Any chance you could send us at least DL logs from /var/log/displaylink?

Sent over PM

mlukaszek
06-08-2016, 09:34 AM
Thanks, I've raised this internally for analysis.

Cheers,
Michal

TomStaels
09-26-2016, 11:30 AM
having the same issue...
any solution to this yet ?
[ 49.032846] usb 2-1.1: new high-speed USB device number 3 using ehci-pci
[ 49.128660] usb 2-1.1: New USB device found, idVendor=17e9, idProduct=0033
[ 49.128672] usb 2-1.1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[ 49.128677] usb 2-1.1: Product: K33907
[ 49.128682] usb 2-1.1: Manufacturer: DisplayLink
[ 49.128686] usb 2-1.1: SerialNumber: 0033-100831
[ 50.148085] [drm] vendor descriptor length:e0 data:00 00 00 00 00 00 00 00 00 00 00
[ 50.148095] [drm:udl_driver_load [udl]] *ERROR* Unrecognized vendor firmware descriptor
[ 50.196976] EDID block is all zeroes
[ 50.196991] udl 2-1.1:1.0: DVI-I-1: EDID invalid.
[ 50.197561] udl 2-1.1:1.0: fb1: udldrmfb frame buffer device
[ 50.197564] [drm] Supports vblank timestamp caching Rev 2 (21.10.2013).
[ 50.197564] [drm] No driver support for vblank timestamp query.
[ 50.197566] [drm] Initialized udl on minor 1
[ 50.197591] usbcore: registered new interface driver udl
[ 50.390848] EDID block is all zeroes
[ 50.390864] udl 2-1.1:1.0: DVI-I-1: EDID invalid.
[ 50.438966] EDID block is all zeroes
[ 50.438977] udl 2-1.1:1.0: DVI-I-1: EDID invalid.
[ 120.941994] EDID block is all zeroes
[ 120.942011] udl 2-1.1:1.0: DVI-I-1: EDID invalid.
[ 121.081365] EDID block is all zeroes
[ 121.081383] udl 2-1.1:1.0: DVI-I-1: EDID invalid.
[ 306.286639] EDID block is all zeroes
[ 306.286655] udl 2-1.1:1.0: DVI-I-1: EDID invalid.
[ 758.287223] EDID block is all zeroes
[ 758.287230] udl 2-1.1:1.0: DVI-I-1: EDID invalid.
[ 758.364453] EDID block is all zeroes
[ 758.364460] udl 2-1.1:1.0: DVI-I-1: EDID invalid.
[ 758.441228] EDID block is all zeroes
[ 758.441235] udl 2-1.1:1.0: DVI-I-1: EDID invalid.
[ 758.459631] evdi: module verification failed: signature and/or required key missing - tainting kernel
[ 758.461167] evdi: [I] Initialising logging on level 5
[ 759.439734] EDID block is all zeroes
[ 759.439761] udl 2-1.1:1.0: DVI-I-1: EDID invalid.
[ 759.488616] EDID block is all zeroes
[ 759.488628] udl 2-1.1:1.0: DVI-I-1: EDID invalid.

mlukaszek
09-27-2016, 07:20 AM
Don't use udl and evdi together. For devices requiring evdi, make sure udl is blacklisted. For USB 2.0 devices using DL-1xx chips, stick to udl, and don't install evdi.

Cheers,
Michal

TomStaels
09-27-2016, 04:30 PM
Thanks for helping out Michal!

uninstalled evdi
un-blacklisted udlfb
screen works, but can't set native resolution (1280x1024), only options are 800x600 and 1024x768

help?

mlukaszek
09-28-2016, 09:09 AM
DRM driver for USB 2.0 DisplayLink Devices is udl, a rewrite of the original udlfb driver (which you shouldn't use). Which one have you unblacklisted?

Thanks,
Michla

TomStaels
09-28-2016, 12:27 PM
i was trying to get it to work with udlfb, since this seems to be a "Bad Idea" (tm), I tried udl :


$ lsmod | grep "udl"
udl 32768 1
drm_kms_helper 147456 2 udl,i915
drm 364544 7 udl,i915,drm_kms_helper


Turning ON the "Unknown Display" in Displays, the system becomes nearly unresponsive, and the display never actually turns on.
syslog is spammed with :
[ 104.344500] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 105.344655] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 106.344680] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 107.356719] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 107.456677] EDID block is all zeroes
[ 107.456687] udl 2-1.1:1.0: DVI-I-1: EDID invalid.
[ 108.564776] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 109.656798] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 110.696848] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 111.712881] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 112.728887] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 113.748959] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 114.776999] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 115.797050] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 116.817072] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 117.837127] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 118.857171] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 119.873186] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 120.893225] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 121.925288] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 122.945319] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 123.973343] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 124.431887] EDID block is all zeroes
[ 124.431904] udl 2-1.1:1.0: DVI-I-1: EDID invalid.

mlukaszek
09-28-2016, 12:42 PM
This looks like a genuine problem with finding out what the EDID for the monitor is. Note the messages telling that the block is all zeros. How is the monitor connected to the device? Do you use an adapter (DVI-VGA) - some of them could block lines reponsible for EDID read?

Thanks,
Michal

TomStaels
09-28-2016, 01:11 PM
the cable is DVI-DVI, no adapter used...
The system dual boots Win/Linux, all screens work just fine in windows.
I will try to get my hands on a 2nd DVI cable for testing.

TomStaels
09-29-2016, 01:45 PM
tried with another DVI-D Single link cable, no joy
tried another screen...no joy.
Every test with udl results in a near unresponsive system, and syslog being spammed as before.

This is a fresh install of Ubuntu 16.04, there's nothing else on there.
Anything else I could try ?

Gabor
10-08-2016, 09:13 AM
Hi,

I have an old usb 2.0 displaylink device with hdmi, also tried with different displays and have the same behaviour and outputs, my session is also crashes, when I disconnect the device, and drops me to login screen.

I know I should use fbl, but before I knew I have tried EVDI, and it doesn't work as well, but I had not all zeros invalid EDID
kernel: [ 1002.378434] [drm:drm_edid_block_valid [drm]] *ERROR* EDID checksum is invalid, remainder is184
kernel: [ 1002.378438] Raw EDID:
kernel: [ 1002.378440] 00 ff ff ff ff ff ff 00 41 2f 00 00 01 01 01 01
kernel: [ 1002.378442] 00 15 01 03 80 00 00 78 0a da ff a3 58 4a a2 29
kernel: [ 1002.378443] 17 49 4b 00 00 00 01 01 01 01 01 01 01 01 01 01
kernel: [ 1002.378444] 01 01 01 01 01 01 01 1d 00 bc 52 d0 1e 20 b8 28
kernel: [ 1002.378446] 55 40 ba 88 21 00 00 1e 01 1d 00 72 51 d0 1e 20
kernel: [ 1002.378447] 6e 28 55 00 ba 88 21 00 00 1e 00 00 00 00 00 00
kernel: [ 1002.378448] 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
kernel: [ 1002.378450] 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
kernel: [ 1002.378453] udl 2-3.1:1.0: DVI-I-2: EDID invalid.
kernel: [ 1002.379077] udl 2-3.1:1.0: fb2: udldrmfb frame buffer device

Without evdi:

[ 2617.838969] usb 2-2.1: new high-speed USB device number 19 using xhci_hcd
[ 2617.946762] usb 2-2.1: New USB device found, idVendor=17e9, idProduct=019d
[ 2617.946766] usb 2-2.1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[ 2617.946769] usb 2-2.1: Product: USB TO HDMI
[ 2617.946771] usb 2-2.1: Manufacturer: DisplayLink
[ 2617.946773] usb 2-2.1: SerialNumber: 003142
[ 2617.948083] [drm] vendor descriptor length:23 data:23 5f 01 00 21 00 04 04 07 00 01
[ 2617.986830] EDID block is all zeroes
[ 2617.986837] udl 2-2.1:1.0: DVI-I-1: EDID invalid.
[ 2617.987493] udl 2-2.1:1.0: fb1: udldrmfb frame buffer device
[ 2617.987497] [drm] Supports vblank timestamp caching Rev 2 (21.10.2013).
[ 2617.987498] [drm] No driver support for vblank timestamp query.
[ 2617.987500] [drm] Initialized udl on minor 1
[ 2618.074972] usb 2-2.2: new full-speed USB device number 20 using xhci_hcd
[ 2618.180003] usb 2-2.2: New USB device found, idVendor=0d8c, idProduct=000c
[ 2618.180008] usb 2-2.2: New USB device strings: Mfr=0, Product=1, SerialNumber=0
[ 2618.180010] usb 2-2.2: Product: C-Media USB Headphone Set
[ 2618.184149] input: C-Media USB Headphone Set as /devices/pci0000:00/0000:00:14.0/usb2/2-2/2-2.2/2-2.2:1.2/0003:0D8C:000C.0007/input/input25
[ 2618.239182] hid-generic 0003:0D8C:000C.0007: input,hidraw2: USB HID v1.00 Device [C-Media USB Headphone Set ] on usb-0000:00:14.0-2.2/input2
[ 2619.091410] EDID block is all zeroes
[ 2619.091431] udl 2-2.1:1.0: DVI-I-1: EDID invalid.
[ 2619.131242] EDID block is all zeroes
[ 2619.131247] udl 2-2.1:1.0: DVI-I-1: EDID invalid.
[ 2655.928327] [drm] write mode info 153
[ 2656.932604] [drm] wait for urb interrupted: ffffffc2 available: 0
...
[ 2870.401817] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 2871.401847] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 2872.401860] [drm] wait for urb interrupted: ffffffc2 available: 0
[ 2872.677582] [drm:udl_urb_completion [udl]] *ERROR* udl_urb_completion - nonzero write bulk status received: -71
[ 2872.678194] usb 2-2: USB disconnect, device number 18
[ 2872.678200] usb 2-2.1: USB disconnect, device number 19
[ 2872.681277] usb 2-2.2: USB disconnect, device number 20
[ 2873.307596] konsole[7540]: segfault at 7f8cffaa6608 ip 00007f8d0ec5c371 sp 00007ffed4c340b0 error 4 in libQt5Core.so.5.5.1[7f8d0e99b000+4c5000]
[ 2874.301959] [drm] wait for urb interrupted: ffffffc2 available: 4
[ 2875.201910] usb 1-1.6: reset full-speed USB device number 5 using ehci-pci

$ lsmod |grep udl
udl 32768 0
drm_kms_helper 147456 2 udl,i915
drm 364544 7 udl,i915,drm_kms_helper
$ uname -a
Linux grey 4.4.0-38-generic #57-Ubuntu SMP Tue Sep 6 15:42:33 UTC 2016 x86_64 x86_64 x86_64 GNU/Linux