This is not reproducible 100% of the time, but it is fairly consistent for me. If I go to a website like imgur.com, or the facebook homepage (after logging in) and just continuously scroll down eventually the I receive the blue screen of death and the whole system crashes. After reboot, if I open nightly again and do the same thing then it will crash the whole system again.

How much do you have to scroll? I've scrolled about 100 pages down Facebook's newsfeed a couple of times and down Imgur's front page about the same and cannot reproduce the crash with the latest nightly on the Yoga c630.

I only need to scroll maybe 5-10 pages and then it crashes nearly every time. Unfortunately it's not 100% consistent, and it doesn't happen on every session or else I could come up with exact instructions.

Is the stop code for the crash usually "VIDEO_SCHEDULER_INTERNAL_ERROR"?

(In reply to dsstorefile+ from comment #4)

Is the stop code for the crash usually "VIDEO_SCHEDULER_INTERNAL_ERROR"?

Yup, that's the one.

Denis, would you have a regression range for this?

Can you reproduce if you set gfx.direct2d.disabled=true

(In reply to Jeff Muizelaar [:jrmuizel] from comment #7)

Can you reproduce if you set gfx.direct2d.disabled=true

Yes, it still crashes.

(In reply to Miko Mynttinen [:miko] from comment #6)

Denis, would you have a regression range for this?

I actually just tried the very first build I could find for aarch64 windows (Dec. 19), and I was not able to reproduce the crash. So, I'll try and bisect this tomorrow.

I tried bisecting this, but couldn't narrow it down to any specific nightly. I think it's quite intermittent and because of this it's hard to bisect. The first build does consistently seem to work, however.

Can you follow the instructions at https://mywindowshub.com/how-to-configure-windows-10-to-create-dump-files-on-bsod/ to capture a dump file the next time this happens? Then we can pass that dump to the vendor.

see comment 11

I should also mention I don't hit this problem with Edge.

We still need to get a dump file.

FWIW, even if we had reliable steps to reproduce this issue, it would likely be a lot of work to figure out what exactly is going on. We're much better off having Qualcomm try to figure it out first, or at least give us some hints into the area that's causing the problem. It's possible we can also get some help from Microsoft. In either case, we're going to need get a dump for them for it to be actionable.

Memory dumps sent to Jeff.

And sent on to Qualcomm.

Mass-removing myself from cc; search for 12b9dfe4-ece3-40dc-8d23-60e179f64ac1 or any reasonable part thereof, to mass-delete these notifications (and sorry!)