[Bug 117131] vga_switcheroo does not switch IGP -> DIS ( IGP == i915 , DIS == radeon )

Mon Apr 25 19:02:25 UTC 2016

https://bugzilla.kernel.org/show_bug.cgi?id=117131

--- Comment #11 from Alex Deucher <alexdeucher at gmail.com> ---
(In reply to Jason Vas Dias from comment #10)
> OK, fine, I have to use FGLRX to use the Radeon as the primary display 
> as I do under OEL7 (Linux 3.10) - 
> what a terrific new open source radeon driver Linux 4.5 has.

fglrx still uses the intel card for display, it just hacks a bunch of stuff
under the covers rather than using the PRIME infrastructure.

> 
> But I cannot even get the Intel card to go into graphics mode under Linux
> 4.5.0, 
> owing to the failure of the radeon module to call that ACPI method .
> 
> My xorg.conf now has no references to radeon driver :
> 

Please remove your xorg config completely.  You shouldn't need one, and some
older xservers didn't properly handle multi-GPU with one in place.

> 
> 
> Now, I boot only with the 'i915.modeset=1' kernel boot parameter .
> 
> I do :
> 
> $ dmesg -c >/var/log/dmesg-4.5.0.boot.log # available on request
> $ export DISPLAY=:0
> $ Xorg -logverbose 7 :0 vt04 & (sleep 2; xterm &)
> 
> Now the X-server gets a segmentation violation in the xf86-video-intel
> driver,
> and the following dmesgs are generated :
> 
> $ dmesg
> [  121.166547] [drm] probing gen 2 caps for device 8086:c01 = 261ad03/e
> [  121.167141] [drm] PCIE gen 3 link speeds already enabled
> [  121.171391] [drm] PCIE GART of 2048M enabled (table at
> 0x00000000002E8000).
> [  121.172068] radeon 0000:01:00.0: WB enabled
> [  121.172638] radeon 0000:01:00.0: fence driver on ring 0 use gpu addr
> 0x0000000100000c00 and cpu addr 0xffff880414eb5c00
> [  121.173232] radeon 0000:01:00.0: fence driver on ring 1 use gpu addr
> 0x0000000100000c04 and cpu addr 0xffff880414eb5c04
> [  121.173821] radeon 0000:01:00.0: fence driver on ring 2 use gpu addr
> 0x0000000100000c08 and cpu addr 0xffff880414eb5c08
> [  121.174409] radeon 0000:01:00.0: fence driver on ring 3 use gpu addr
> 0x0000000100000c0c and cpu addr 0xffff880414eb5c0c
> [  121.174990] radeon 0000:01:00.0: fence driver on ring 4 use gpu addr
> 0x0000000100000c10 and cpu addr 0xffff880414eb5c10
> [  121.175941] radeon 0000:01:00.0: fence driver on ring 5 use gpu addr
> 0x0000000000075a18 and cpu addr 0xffffc90000435a18
> [  121.196617] radeon 0000:01:00.0: fence driver on ring 6 use gpu addr
> 0x0000000100000c18 and cpu addr 0xffff880414eb5c18
> [  121.197169] radeon 0000:01:00.0: fence driver on ring 7 use gpu addr
> 0x0000000100000c1c and cpu addr 0xffff880414eb5c1c
> [  121.344986] [drm] ring test on 0 succeeded in 2 usecs
> [  121.345536] [drm] ring test on 1 succeeded in 1 usecs
> [  121.346090] [drm] ring test on 2 succeeded in 1 usecs
> [  121.346659] [drm] ring test on 3 succeeded in 4 usecs
> [  121.347202] [drm] ring test on 4 succeeded in 2 usecs
> [  121.524819] [drm] ring test on 5 succeeded in 2 usecs
> [  121.525345] [drm] UVD initialized successfully.
> [  121.635907] [drm] ring test on 6 succeeded in 13 usecs
> [  121.636449] [drm] ring test on 7 succeeded in 4 usecs
> [  121.636978] [drm] VCE initialized successfully.
> [  121.637570] f 0#2: signaled from fence_wait
> [  121.638109] [drm] ib test on ring 0 succeeded in 0 usecs
> [  121.638641] f 1#2: signaled from fence_wait
> [  121.639152] [drm] ib test on ring 1 succeeded in 0 usecs
> [  121.639722] f 2#2: signaled from fence_wait
> [  121.640233] [drm] ib test on ring 2 succeeded in 0 usecs
> [  121.640801] f 3#2: signaled from fence_wait
> [  121.641313] [drm] ib test on ring 3 succeeded in 0 usecs
> [  121.641878] f 4#2: signaled from fence_wait
> [  121.642404] [drm] ib test on ring 4 succeeded in 0 usecs
> [  122.295105] f 5#4: signaled from fence_wait
> [  122.295672] [drm] ib test on ring 5 succeeded
> [  122.797067] f 6#4: signaled from fence_wait
> [  122.797648] [drm] ib test on ring 6 succeeded
> [  123.299048] f 7#4: signaled from fence_wait
> [  123.299662] [drm] ib test on ring 7 succeeded
> [  123.387817] device: 'vcs4': device_add
> [  123.387832] PM: Adding info for No Bus:vcs4
> [  123.387875] device: 'vcsa4': device_add
> [  123.387893] PM: Adding info for No Bus:vcsa4
> [  123.400978] Xorg (4385) used greatest stack depth: 12016 bytes left
> [  129.273612] ACPI Error: No handler for Region [EC81] (ffff88041d8d9bd0)
> [EmbeddedControl] (20160108/evregion-166)
> [  129.274317] ACPI Error: Region EmbeddedControl (ID=3) has no handler
> (20160108/exfldio-299)
> [  129.275027] ACPI Error: Method parse/execution failed
> [\_SB.PCI0.PEG0.PEGP.SGOF] (Node ffff88041d8fca50), AE_NOT_EXIST
> (20160108/psparse-542)
> [  129.275779] ACPI Error: Method parse/execution failed
> [\_SB.PCI0.PEG0.PEGP._OFF] (Node ffff88041d8fce88), AE_NOT_EXIST
> (20160108/psparse-542)
> [  129.276495] ACPI Error: Method parse/execution failed
> [\_SB.PCI0.GFX0.ATPX] (Node ffff88041d9034b0), AE_NOT_EXIST
> (20160108/psparse-542)
> [  129.277229] failed to evaluate ATPX got AE_NOT_EXIST
> 
> It seems even when I specify Intel driver should be used, the Intel driver
> is still hooking into the Radeon driver which initiates its incorrect ACPI
> transaction.
> 
> Must I remove the radeon driver from the kernel completely in order to get
> my display into graphics mode, in order to work around the buggy radeon
> driver ?

Please attach your dmesg output from boot without starting X.

You may be hitting a regression in 4.5 that was recently fixed.  Please try
appending radeon.runpm=0 to the kernel command line in grub.

-- 
You are receiving this mail because:
You are watching the assignee of the bug.