Summary: | Screen blank after sleep/resume | ||||||||
---|---|---|---|---|---|---|---|---|---|
Product: | Base System | Reporter: | Tobias <tobias> | ||||||
Component: | kern | Assignee: | freebsd-bugs (Nobody) <bugs> | ||||||
Status: | New --- | ||||||||
Severity: | Affects Some People | CC: | eadler, grahamperrin, grahamperrin, jhb, johalun0, miner.ben07 | ||||||
Priority: | --- | Keywords: | regression | ||||||
Version: | 11.0-STABLE | ||||||||
Hardware: | amd64 | ||||||||
OS: | Any | ||||||||
See Also: | https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=260994 | ||||||||
Attachments: |
|
Description
Tobias
2016-10-15 15:20:11 UTC
Created attachment 175780 [details]
dmesg
Created attachment 175781 [details]
syslog sleep/resume
I'm gonna revive this one. On my Broadwell laptop it works as expected IF I have i915kms (drm-stable-kmod package) loaded. However, when suspending from console (with no drm driver loaded), the screen does not come back on after resume. The system is fully operational though. In resumed state with black screen (from ssh): startx with scfb driver won't make the screen come on. kldload i915kms will make the screen come on. UEFI amd64 boot with kernel version r333603 I can confirm the same behaviour on an AMD based system with radeon graphics. There's no practical way to fix resume without a GPU driver. There's simply no GPU-independent method of turning the display back on. A GPU driver is required to enable the screen on modern systems (on older systems the BIOS sometimes turned the screen back on for you, but that hasn't been true for several years now). However, the original bug report does have the GPU driver it seems (since it is suspending in X) so is different from the report in comment #4 (which we can never fix). (In reply to John Baldwin from comment #5) John, I see that the original report mentions X but I do not see any mention of a GPU driver (attachments have nothing too). I guess that X with vesa driver would behave the same as any other case where a GPU driver is not loaded? (In reply to Andriy Gapon from comment #6) Yes, X with VESA will most likely not work aside from an older BIOS. kldload radeonkms fixed it for me. It seems to have something to do with the DRM driver, I'm not sure why though. (In reply to miner.ben07 from comment #8) > … seems to have something to do with the DRM driver, … Bug 260994 key phrase: Kernel Mode Setting. <https://bsd-hardware.info/?probe=e2d694053a#pci:10de-0ff6-103c-2256> nvidia, more specfically <https://bsd-hardware.info/?probe=e2d694053a&log=kldstat> nvidia-modeset and <https://bsd-hardware.info/?probe=e2d694053a&log=pkglist> x11/nvidia-driver-470. Re: <https://github.com/grahamperrin/freebsd-ports/pull/2/>, x11/nvidia-driver-470 lacks what's required for nvidia-drm. This HP ZBook 17 G2: * does wake from sleep with nvidia-driver-470 _without_ DRM * failed to wake from sleep when I used the system console frame buffer (an X.Org configuration) instead of nvidia-modeset (the kernel module). The failure is consistent with <https://wiki.freebsd.org/Graphics#hardware>, > Features that will not work with the system console frame buffer include: > > sleep and wake (suspend and resume) of the computer. (In reply to Tobias from comment #2 (2016-10-15, not seen in Bugzilla since then)) I assume that this bug can be closed as either: a) overcome by events; or b) working as expected (from the attachments, I can't tell which kernel module, if any, was loaded for the NVIDIA Quadro K2100M). Thanks |