I'm getting a machine freeze with the intel driver, if it's not the primary device. I'm using an Intel 915G machine with a Radeon ES1000/RN50 in the PCI slot. The latter is configured as primary graphics device via the system BIOS. The driver works fine if the onboard graphics are configured as primary graphics device. I debugged the driver to see where the freeze happens. i830_display.c: i830_crtc_mode_set() [...] i830WaitForVblank(pScrn); --> OUTREG(dspcntr_reg, dspcntr); /* hard freeze */ /* Flush the plane changes */ i830PipeSetBase(crtc, x, y); [...] (gdb) p /x dspcntr_reg $6 = 0x70180 (gdb) p /x dspcntr $7 = 0xd8000000 I'll attach config and log files as well.
Created attachment 11306 [details] xorg.conf
Created attachment 11307 [details] Xorg.0.log.intel-primary
Created attachment 11308 [details] Xorg.0.log.intel-non_primary
As you can see Xorg.0.log.intel-non_primary is truncated after (II) intel(0): I2C bus "SDVOCTRL_E for SDVOC" removed. This is where the machine freezes.
Just an idea. The Intel VGA BIOS gets overwritten by ATI Radeon BIOS during system start, if the ATI card is the primary device. During Xserver start the intel driver posts the wrong BIOS, the ATI one --> machine freeze Is there any VGA initialization in intel driver missing, so using Intel onboard graphics as non-primary graphics currently cannot work at all?
I think something like that is happening. It's likely none of the Intel pipes are enabled and we program the display controller registers incorrectly at startup for that case...
what kind of mobo do you use? it seems that we can't reproduce this here, as the machine disabled the intel gfx if it's not set as primary...( can't find it through lspci) :(
bug reporter's respone timeout...reject the bug..
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.