Bug 44851
Summary: | resume to black screen | ||
---|---|---|---|
Product: | Drivers | Reporter: | Peter (energonic) |
Component: | Video(DRI - non Intel) | Assignee: | drivers_video-dri |
Status: | NEW --- | ||
Severity: | normal | CC: | alan, alexdeucher |
Priority: | P1 | ||
Hardware: | All | ||
OS: | Linux | ||
Kernel Version: | 3.9 | Subsystem: | |
Regression: | No | Bisected commit-id: | |
Attachments: |
X windows log
dmesg Xorg.0.log file dmesg.txt file kdm.log file |
Description
Peter
2012-07-16 09:26:55 UTC
Please attach your xorg log and dmesg output. What do you mean by wrong display port? Is this a hybrid laptop with multiple GPUs? Created attachment 75481 [details]
X windows log
This log is from when the system is started using nomodeset and X video driver is fbdev.
Created attachment 75491 [details]
dmesg
dmesg log from when kernel starts with nomodeset and X video driver is fbdev.
Sorry if I've used the wrong terminology. This is a standard iMac. When I used the AMD proprietary fglrx driver some time ago on ubuntu 11.04 it showed two display screens detected. The main one is the iMac display and the other is for the display port to attach an external monitor. Now I use the radeon kernel driver at boot time and the xf86 fbdev module for X. If I do a text-only boot without telling the kernel 'nomodeset', I get a black screen; but it has booted - I presume the other 'display port' (or 'connector'? not my main screen) has been chosen for text output. Assuming I boot with nomodeset then I can start X OK, provided I use the xf86 fbdev driver (the xf86 ati driver just gives me a black screen - but rebootable, etc). Then I can suspend the system and it goes quiet. If I then awaken it using a keypress, I can hear it starting up, but the screen stays black. I can reboot from here with Ctl-Alt-Del. My assumption is that, on resume, the system tries to run a modeset from the kernel module and decides to turn off the main screen. I would like to avoid it doing that because essentially I cannot use suspend/resume. Just upgraded my arch installation on iMac from kernel 3.8.13 to 3.9.2. Was working, still with the fbdev driver, but now cannot boot properly, either by grub efi or by pure efi (image selected by rEFInd). Behaves as if the 'nomodeset' parameter is ignored; results in all terminal screens black (but sounds as if it has booted). An arch image running in virtualbox (on mac) has upgraded to 3.9.2 without problem, but that does not need 'nomodeset'. Does it work ok if you boot with bootcamp? Also, you might try a 3.10 kernel there were some fixes for EFI systems. Can you attach your xorg log and dmesg output with the radeon driver (without nomodeset)? Created attachment 101291 [details]
Xorg.0.log file
Xorg.0.log obtained from attempt to boot my iMac by efi (directly from kernel image, and without the 'nomodeset' parameter).
Created attachment 101301 [details]
dmesg.txt file
Booted system from arch boot CD. Mounted main mac disk partition for arch (that doesn't boot since upgrade to kernel 3.9.2). chroot to that partition. Run dmesg and pipe output to external disk.
Created attachment 101311 [details]
kdm.log file
File from arch partition on main mac disk. Seems to show problem with kdm_greet after failing to find a dbus session (which might be a result of a graphics problem earlier).
Still on iMac, installed fedora 18 on partition previously used by Arch (would boot using KMS into black vts). fedora using kernel 3.6. Behaved like Arch did previously; had to use kernel option 'nomodeset' to get any vt screens and that stopped radeon module from being loaded by X (Xorg.0.log had: '[KMS] drm report modesetting isn't supported' and this was followed by 'No screens found') Now upgraded fedora to use kernel 3.9.2 and, as with Arch, get black vts even if 'nomodeset' on kernel boot line. The current kernel 3.9.2 is of no use on my system; I have no way to boot it and run vt screens. Others who might need nomodeset may also be affected. |