Feedback send 0809 GMT 5 Dec with tag:

#PixelbookExtendedFailsWhenAsleep

I also have this issue with USB-C display LG 27UD88W using Pixelbook

ChromeOS Version 64.0.3282.41

Sounds like type-c hpd doesn't always make it to the device when we resume. Todd, is that expected?

No its not expected.  Testing using hoho (Google USB-C -> hdmi) works correctly.

It may be related to what type of display sink it is and whether pixelbook is power source or sink.  Unfortunately I can't find the feedback using,

 https://listnr.corp.google.com/product/208/reports?searchText=%23PixelbookExtendedFailsWhenAsleep&filter=1&dateRange=30

ATM.

I'll try the LG 27UD88W next time I'm in front of one (hopefully tomorrow).

I am seeing this multiple times a day with this release.

I have just submitted feedback with the hashtag #pixel-extended-display-bug

May be related/dupe of  crbug.com/644663  (discover identity timing) as when I was able to create the failure

ectool gpioget | grep HPD 

should that the HPD had been raised but only after I did

ectool pdsetmode 0 0xff01 1 0
ectool pdsetmode 0 0xff01 1 1

Did it successfully start driving the monitor.

Will try this patch, https://chromium-review.googlesource.com/c/chromiumos/platform/ec/+/860283, to see if it improves.

CL:860283 unfortunately didn't resolve the problem.  Will have a look at the USBPD traffic next to see if there's something peculiar there.

One thing that looks peculiar is that while it appears the USB-C HPD have arrived (ectool gpioget | grep HPD) I never find i915 driver to notice it (only the edp one)

[    0.247619] [drm:drm_helper_hpd_irq_event] [CONNECTOR:58:eDP-1] status updated from unknown to connected
[    0.247649] [drm:drm_helper_hpd_irq_event] [CONNECTOR:66:DP-1] status updated from unknown to disconnected
[    0.522183] [drm:drm_helper_hpd_irq_event] [CONNECTOR:71:DP-2] status updated from unknown to disconnected
[    3.446724] [drm:intel_get_hpd_pins] hotplug event received, stat 0x01000000, dig 0x12101010, pins 0x00000010
[    3.446742] [drm:intel_hpd_irq_handler] digital hpd port A - long
[    3.446752] [drm:intel_hpd_irq_handler] Received HPD interrupt on PIN 4 - cnt: 0
[    3.446782] [drm:intel_dp_hpd_pulse] ignoring long hpd on eDP port A
[    4.082088] [drm:intel_get_hpd_pins] hotplug event received, stat 0x01000000, dig 0x12101010, pins 0x00000010
[    4.082116] [drm:intel_hpd_irq_handler] digital hpd port A - long
[    4.082130] [drm:intel_hpd_irq_handler] Received HPD interrupt on PIN 4 - cnt: 1
[    4.082194] [drm:intel_dp_hpd_pulse] ignoring long hpd on eDP port A

I'd expect to see something like,

[  236.480565] [drm:intel_get_hpd_pins] hotplug event received, stat 0x00400000, dig 0x10101210, pins 0x00000040
[  236.480574] [drm:intel_hpd_irq_handler] digital hpd port C - long
[  236.480579] [drm:intel_hpd_irq_handler] Received HPD interrupt on PIN 6 - cnt: 0
[  236.480615] [drm:intel_dp_hpd_pulse] got hpd irq on port C - long

No progress w/ fix, but an update.

repro:
Yes: type-C monitor LG 27UD88W
No : hoho
No : tech-sparta typeC hub w/ power + HDMI

Next steps would be to re-confirm failure to see HPD on the AP (i915) side (#comment8) and then to determine if I see any transition physically on the wire.  That will require opening device, rework and a scope.

Also tested w/ HP elite USB-C docking station.  
Repro'd 0/3 by opening lid from S5
Repro'd 1/3 by pressing power button from S5