bbswitch problems with Schenker XMG P503 and nvidia GTX 770M (Linux 3.8 regression) #65

Closed
karolherbst opened this Issue Jun 11, 2013 · 93 comments

Comments

Projects
None yet
@karolherbst
Contributor

karolherbst commented Jun 11, 2013

Fixed with Linux 3.11.


bbswitch won't load for me throwing out this error: "bbswitch: cannot find ACPI handle for VGA device 0000:01:00.0"

dmesg | grep -C 10 bbswitch:

[   13.020762] vboxpci: IOMMU not found (not compiled)
[   13.600954] EXT4-fs (sda2): re-mounted. Opts: errors=remount-ro
[   14.919612] EXT4-fs (sdb2): ext4_orphan_cleanup: deleting unreferenced inode 40247083
[   14.919707] EXT4-fs (sdb2): ext4_orphan_cleanup: deleting unreferenced inode 40247014
[   14.953091] EXT4-fs (sdb2): ext4_orphan_cleanup: deleting unreferenced inode 38146274
[   14.953130] EXT4-fs (sdb2): ext4_orphan_cleanup: deleting unreferenced inode 39321621
[   14.968645] EXT4-fs (sdb2): 4 orphan inodes deleted
[   14.968649] EXT4-fs (sdb2): recovery complete
[   15.162609] EXT4-fs (sdb2): mounted filesystem with ordered data mode. Opts: errors=remount-ro
[   22.200104] r8168: eth0: link down
[   22.810778] bbswitch: version 0.7
[   22.810789] bbswitch: Found integrated VGA device 0000:00:02.0: \_SB_.PCI0.GFX0
[   22.810799] bbswitch: cannot find ACPI handle for VGA device 0000:01:00.0
[   22.810805] bbswitch: No discrete VGA device found
[   23.709854] r8168: eth0: link up
[   24.217956] r8168: eth0: link up
[   32.952788] ata1.00: configured for UDMA/133
[   32.952797] ata1: EH complete
[   32.996465] ata4.00: configured for UDMA/133
[   32.996474] ata4: EH complete
[   33.919047] EXT4-fs (sda2): re-mounted. Opts: errors=remount-ro,commit=0
[   33.926407] EXT4-fs (sdb2): re-mounted. Opts: errors=remount-ro,commit=0
[   51.423756] nm-applet[4611]: segfault at 10 ip 00007f059fa23b47 sp 00007fff35594260 error 4 in libcairo.so.2.11200.14[7f059f962000+141000]
[   71.727811] r8168: eth0: link down
--
[  107.539726] r8168: eth0: link up
[  107.604191] r8168: eth0: link down
[  109.197598] r8168: eth0: link up
[  109.607915] r8168: eth0: link up
[  117.201909] usb 3-6: new full-speed USB device number 4 using xhci_hcd
[  117.218414] usb 3-6: New USB device found, idVendor=046d, idProduct=0a1b
[  117.218423] usb 3-6: New USB device strings: Mfr=1, Product=2, SerialNumber=0
[  117.218429] usb 3-6: Product: Logitech H555 Headset
[  117.218434] usb 3-6: Manufacturer: Logitech
[  117.521008] usbcore: registered new interface driver snd-usb-audio
[  543.974279] bbswitch: version 0.7
[  543.974298] bbswitch: Found integrated VGA device 0000:00:02.0: \_SB_.PCI0.GFX0
[  543.974307] bbswitch: cannot find ACPI handle for VGA device 0000:01:00.0
[  543.974317] bbswitch: No discrete VGA device found

uname -a:

Linux pingu 3.10.0-rc5 #1 SMP PREEMPT Tue Jun 11 23:13:51 CEST 2013 x86_64 Intel(R) Core(TM) i7-4700MQ CPU @ 2.40GHz GenuineIntel GNU/Linux

Distribution:

Gentoo Linux

xorg-server: 1.13.3
xf86-video-intel: 2.21.9

https://bugs.launchpad.net/lpbugreporter/+bug/752542/comments/606
"XMG P503" is the attachment:
https://bugs.launchpad.net/lpbugreporter/+bug/752542/+attachment/3701104/+files/Notebook-P15SM.tar.gz

@Lekensteyn

This comment has been minimized.

Show comment Hide comment
@Lekensteyn

Lekensteyn Jun 13, 2013

Owner

Could you also add the output of:

ls -l /sys/bus/acpi/drivers/video/*/physical_node
grep . /sys/bus/acpi/drivers/video/*/path

and include a full dmesg? Do older kernels (say, 3.9) work?

Owner

Lekensteyn commented Jun 13, 2013

Could you also add the output of:

ls -l /sys/bus/acpi/drivers/video/*/physical_node
grep . /sys/bus/acpi/drivers/video/*/path

and include a full dmesg? Do older kernels (say, 3.9) work?

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Jun 13, 2013

Contributor

ls -l /sys/bus/acpi/drivers/video/*/physical_node

lrwxrwxrwx 1 root root 0 Jun 13 20:40 /sys/bus/acpi/drivers/video/LNXVIDEO:01/physical_node -> ../../../../pci0000:00/0000:00:02.0

grep . /sys/bus/acpi/drivers/video/*/path

/sys/bus/acpi/drivers/video/LNXVIDEO:00/path:\_SB_.PCI0.PEG0.PEGP
/sys/bus/acpi/drivers/video/LNXVIDEO:01/path:\_SB_.PCI0.GFX0

dmesg: http://bpaste.net/show/106852/

this is until boot is complete. If you need more just say it. I have to reboot to give you that, but I think this should be enough.

I have the same problem with 3.9 kernel. But this wouldn't be an option anyway, because of haswell. I have some Intel gpu problems on 3.9

Contributor

karolherbst commented Jun 13, 2013

ls -l /sys/bus/acpi/drivers/video/*/physical_node

lrwxrwxrwx 1 root root 0 Jun 13 20:40 /sys/bus/acpi/drivers/video/LNXVIDEO:01/physical_node -> ../../../../pci0000:00/0000:00:02.0

grep . /sys/bus/acpi/drivers/video/*/path

/sys/bus/acpi/drivers/video/LNXVIDEO:00/path:\_SB_.PCI0.PEG0.PEGP
/sys/bus/acpi/drivers/video/LNXVIDEO:01/path:\_SB_.PCI0.GFX0

dmesg: http://bpaste.net/show/106852/

this is until boot is complete. If you need more just say it. I have to reboot to give you that, but I think this should be enough.

I have the same problem with 3.9 kernel. But this wouldn't be an option anyway, because of haswell. I have some Intel gpu problems on 3.9

@Lekensteyn

This comment has been minimized.

Show comment Hide comment
@Lekensteyn

Lekensteyn Jun 13, 2013

Owner

Can the nouveau driver detect this device?

Could you build and retrieve information from https://github.com/Lekensteyn/acpi-stuff/tree/master/acpi_dump_info? After building, run cat /proc/acpi/dump_info.

Owner

Lekensteyn commented Jun 13, 2013

Can the nouveau driver detect this device?

Could you build and retrieve information from https://github.com/Lekensteyn/acpi-stuff/tree/master/acpi_dump_info? After building, run cat /proc/acpi/dump_info.

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Jun 13, 2013

Contributor

yes, nouveau is detecting it, but still same error with bbswitch. I am currently building mesa and X driver to test if it works with bumblebee.

cat /proc/acpi/dump_info

0000:00:00.0 060000 
0000:00:01.0 060400 \_SB_.PCI0.P0P2
0000:00:02.0 030000 \_SB_.PCI0.GFX0
0000:00:03.0 040300 \_SB_.PCI0.B0D3
0000:00:14.0 0c0330 \_SB_.PCI0.XHC_
0000:00:16.0 078000 
0000:00:1a.0 0c0320 \_SB_.PCI0.EHC2
0000:00:1b.0 040300 \_SB_.PCI0.HDEF
0000:00:1c.0 060400 \_SB_.PCI0.RP01
0000:00:1c.1 060400 \_SB_.PCI0.RP02
0000:00:1c.2 060400 \_SB_.PCI0.RP03
0000:00:1c.3 060400 \_SB_.PCI0.RP04
0000:00:1d.0 0c0320 \_SB_.PCI0.EHC1
0000:00:1f.0 060100 \_SB_.PCI0.LPCB
0000:00:1f.2 010601 \_SB_.PCI0.SAT0
0000:00:1f.3 0c0500 \_SB_.PCI0.SBUS
0000:01:00.0 030000 
0000:03:00.0 060400 \_SB_.PCI0.RP02.PXSX
0000:04:00.0 0c0010 
0000:05:00.0 ff0000 \_SB_.PCI0.RP03.PXSX
0000:05:00.2 020000 \_SB_.PCI0.RP03.RLAN
0000:06:00.0 028000 \_SB_.PCI0.RP04.PXSX
Contributor

karolherbst commented Jun 13, 2013

yes, nouveau is detecting it, but still same error with bbswitch. I am currently building mesa and X driver to test if it works with bumblebee.

cat /proc/acpi/dump_info

0000:00:00.0 060000 
0000:00:01.0 060400 \_SB_.PCI0.P0P2
0000:00:02.0 030000 \_SB_.PCI0.GFX0
0000:00:03.0 040300 \_SB_.PCI0.B0D3
0000:00:14.0 0c0330 \_SB_.PCI0.XHC_
0000:00:16.0 078000 
0000:00:1a.0 0c0320 \_SB_.PCI0.EHC2
0000:00:1b.0 040300 \_SB_.PCI0.HDEF
0000:00:1c.0 060400 \_SB_.PCI0.RP01
0000:00:1c.1 060400 \_SB_.PCI0.RP02
0000:00:1c.2 060400 \_SB_.PCI0.RP03
0000:00:1c.3 060400 \_SB_.PCI0.RP04
0000:00:1d.0 0c0320 \_SB_.PCI0.EHC1
0000:00:1f.0 060100 \_SB_.PCI0.LPCB
0000:00:1f.2 010601 \_SB_.PCI0.SAT0
0000:00:1f.3 0c0500 \_SB_.PCI0.SBUS
0000:01:00.0 030000 
0000:03:00.0 060400 \_SB_.PCI0.RP02.PXSX
0000:04:00.0 0c0010 
0000:05:00.0 ff0000 \_SB_.PCI0.RP03.PXSX
0000:05:00.2 020000 \_SB_.PCI0.RP03.RLAN
0000:06:00.0 028000 \_SB_.PCI0.RP04.PXSX
@Lekensteyn

This comment has been minimized.

Show comment Hide comment
@Lekensteyn

Lekensteyn Jun 13, 2013

Owner

Can you paste dmesg from where you load nouveau?

Owner

Lekensteyn commented Jun 13, 2013

Can you paste dmesg from where you load nouveau?

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Jun 13, 2013

Contributor
[  717.793800] MXM: GUID detected in BIOS
[  717.793875] nouveau 0000:01:00.0: enabling device (0006 -> 0007)
[  717.794999] nouveau  [  DEVICE][0000:01:00.0] BOOT0  : 0x0e6200a1
[  717.795007] nouveau  [  DEVICE][0000:01:00.0] Chipset: GK106 (NVE6)
[  717.795011] nouveau  [  DEVICE][0000:01:00.0] Family : NVE0
[  717.803542] nouveau  [   VBIOS][0000:01:00.0] checking PRAMIN for image...
[  717.803556] nouveau  [   VBIOS][0000:01:00.0] ... signature not found
[  717.803561] nouveau  [   VBIOS][0000:01:00.0] checking PROM for image...
[  717.980198] nouveau  [   VBIOS][0000:01:00.0] ... appears to be valid
[  717.980213] nouveau  [   VBIOS][0000:01:00.0] using image from PROM
[  717.980640] nouveau  [   VBIOS][0000:01:00.0] BIT signature found
[  717.980653] nouveau  [   VBIOS][0000:01:00.0] version 80.06.51.00.0c
[  717.981082] nouveau E[   VBIOS][0000:01:00.0] 0x9d8d[ ]: unknown opcode 0xaa
[  717.981107] nouveau  [     MXM][0000:01:00.0] BIOS version 3.0
[  717.983314] nouveau  [     MXM][0000:01:00.0] failed to locate valid SIS
[  717.983365] nouveau  [     PFB][0000:01:00.0] RAM type: GDDR5
[  717.983377] nouveau  [     PFB][0000:01:00.0] RAM size: 3072 MiB
[  717.983381] nouveau  [     PFB][0000:01:00.0]    ZCOMP: 0 tags
[  718.021150] nouveau  [  PTHERM][0000:01:00.0] FAN control: PWM
[  718.021178] nouveau  [  PTHERM][0000:01:00.0] fan management: disabled
[  718.021184] nouveau  [  PTHERM][0000:01:00.0] internal sensor: yes
[  718.134635] [TTM] Zone  kernel: Available graphics memory: 8149418 kiB
[  718.134640] [TTM] Zone   dma32: Available graphics memory: 2097152 kiB
[  718.134642] [TTM] Initializing pool allocator
[  718.134649] [TTM] Initializing DMA pool allocator
[  718.134673] mtrr: type mismatch for c0000000,10000000 old: write-back new: write-combining
[  718.134677] nouveau  [     DRM] VRAM: 3072 MiB
[  718.134680] nouveau  [     DRM] GART: 512 MiB
[  718.134688] nouveau  [     DRM] TMDS table version 2.0
[  718.134690] nouveau  [     DRM] DCB version 4.0
[  718.134695] nouveau  [     DRM] DCB outp 00: 01000f13 00010034
[  718.134699] nouveau  [     DRM] DCB outp 01: 01001f12 0f420010
[  718.134701] nouveau  [     DRM] DCB outp 02: 01002f12 00020030
[  718.134704] nouveau  [     DRM] DCB outp 03: 08015fd6 0f420020
[  718.134707] nouveau  [     DRM] DCB outp 04: 08015f92 00020020
[  718.134709] nouveau  [     DRM] DCB outp 05: 08014fc6 0f420010
[  718.134712] nouveau  [     DRM] DCB outp 06: 08014f82 00020010
[  718.134715] nouveau  [     DRM] DCB outp 07: 08013f82 0f420030
[  718.134717] nouveau  [     DRM] DCB outp 08: 04038fb6 0f420010
[  718.134720] nouveau  [     DRM] DCB outp 09: 04038f72 00020010
[  718.134723] nouveau  [     DRM] DCB outp 10: 02027fa6 0f420010
[  718.134726] nouveau  [     DRM] DCB outp 11: 02027f62 00020010
[  718.134728] nouveau  [     DRM] DCB outp 13: 02049f00 00000000
[  718.134732] nouveau  [     DRM] DCB conn 00: 08000040
[  718.134736] nouveau  [     DRM] DCB conn 01: 00001161
[  718.134739] nouveau  [     DRM] DCB conn 02: 00001231
[  718.134742] nouveau  [     DRM] DCB conn 03: 01000331
[  718.134745] nouveau  [     DRM] DCB conn 04: 01000446
[  718.134747] nouveau  [     DRM] DCB conn 05: 02000546
[  718.134750] nouveau  [     DRM] DCB conn 07: 00010746
[  718.134753] nouveau  [     DRM] DCB conn 08: 00020847
[  718.134756] nouveau  [     DRM] DCB conn 09: 00000900
[  718.136929] [drm] Supports vblank timestamp caching Rev 1 (10.10.2010).
[  718.136938] [drm] No driver support for vblank timestamp query.
[  718.136964] nouveau W[     DRM] voltage table 0x50 unknown
[  718.136987] nouveau  [     DRM] 3 available performance level(s)
[  718.137000] nouveau  [     DRM] 0: core 202MHz shader 405MHz memory 405MHz voltage 100mV
[  718.137008] nouveau  [     DRM] 1: core 405MHz shader 810MHz memory 1080MHz voltage 80mV
[  718.137015] nouveau  [     DRM] 3: core 1002MHz shader 2004MHz memory 1080MHz voltage 40mV
[  718.137021] nouveau  [     DRM] c: fanspeed 100%
[  718.171908] nouveau  [     DRM] MM: using COPY for buffer copies
[  718.391148] nouveau 0000:01:00.0: No connectors reported connected with modes
[  718.391162] [drm] Cannot find any crtc or sizes - going 1024x768
[  718.393106] nouveau E[    PBUS][0000:01:00.0] MMIO write of 0x00000000 FAULT at 0x418880 [ IBUS ]
[  718.393131] nouveau E[   PIBUS][0000:01:00.0] GPC1: 0x419eb4 0xbadf1000 (0x3800820c)
[  718.393147] nouveau E[   PIBUS][0000:01:00.0] GPC2: 0x419eb4 0xbadf1000 (0x3800820c)
[  720.395018] nouveau E[  PGRAPH][0000:01:00.0] HUB_INIT timed out
[  720.395027] nouveau E[  PGRAPH][0000:01:00.0] 409000 - done 0x00000244
[  720.395036] nouveau E[  PGRAPH][0000:01:00.0] 409000 - stat 0x000000e6 0x00000000 0x00000000 0x00000000
[  720.395045] nouveau E[  PGRAPH][0000:01:00.0] 409000 - stat 0x00000000 0x00000000 0x00000002 0x00000009
[  720.395049] nouveau E[  PGRAPH][0000:01:00.0] 502000 - done 0x00000300
[  720.395062] nouveau E[  PGRAPH][0000:01:00.0] 502000 - stat 0x00000000 0x00000000 0x00000000 0x00000000
[  720.395075] nouveau E[  PGRAPH][0000:01:00.0] 502000 - stat 0x00000000 0x00000000 0x00000000 0x00000000
[  720.395079] nouveau E[  PGRAPH][0000:01:00.0] 50a000 - done 0x00000300
[  720.395091] nouveau E[  PGRAPH][0000:01:00.0] 50a000 - stat 0x00000000 0x00000000 0x00000000 0x00000000
[  720.395105] nouveau E[  PGRAPH][0000:01:00.0] 50a000 - stat 0x00000000 0x00000000 0x00000000 0x00000000
[  720.395110] nouveau E[  PGRAPH][0000:01:00.0] 512000 - done 0x00000300
[  720.395122] nouveau E[  PGRAPH][0000:01:00.0] 512000 - stat 0x00000000 0x00000000 0x00000000 0x00000000
[  720.395136] nouveau E[  PGRAPH][0000:01:00.0] 512000 - stat 0x00000000 0x00000000 0x00000000 0x00000000
[  720.395139] nouveau E[  PGRAPH][0000:01:00.0] init failed, -16
[  720.396187] nouveau  [     DRM] allocated 1024x768 fb: 0x80000, bo ffff88043cad4400
[  720.396237] nouveau 0000:01:00.0: fb1: nouveaufb frame buffer device
[  720.396240] [drm] Initialized nouveau 1.1.1 20120801 for 0000:01:00.0 on minor 1
Contributor

karolherbst commented Jun 13, 2013

[  717.793800] MXM: GUID detected in BIOS
[  717.793875] nouveau 0000:01:00.0: enabling device (0006 -> 0007)
[  717.794999] nouveau  [  DEVICE][0000:01:00.0] BOOT0  : 0x0e6200a1
[  717.795007] nouveau  [  DEVICE][0000:01:00.0] Chipset: GK106 (NVE6)
[  717.795011] nouveau  [  DEVICE][0000:01:00.0] Family : NVE0
[  717.803542] nouveau  [   VBIOS][0000:01:00.0] checking PRAMIN for image...
[  717.803556] nouveau  [   VBIOS][0000:01:00.0] ... signature not found
[  717.803561] nouveau  [   VBIOS][0000:01:00.0] checking PROM for image...
[  717.980198] nouveau  [   VBIOS][0000:01:00.0] ... appears to be valid
[  717.980213] nouveau  [   VBIOS][0000:01:00.0] using image from PROM
[  717.980640] nouveau  [   VBIOS][0000:01:00.0] BIT signature found
[  717.980653] nouveau  [   VBIOS][0000:01:00.0] version 80.06.51.00.0c
[  717.981082] nouveau E[   VBIOS][0000:01:00.0] 0x9d8d[ ]: unknown opcode 0xaa
[  717.981107] nouveau  [     MXM][0000:01:00.0] BIOS version 3.0
[  717.983314] nouveau  [     MXM][0000:01:00.0] failed to locate valid SIS
[  717.983365] nouveau  [     PFB][0000:01:00.0] RAM type: GDDR5
[  717.983377] nouveau  [     PFB][0000:01:00.0] RAM size: 3072 MiB
[  717.983381] nouveau  [     PFB][0000:01:00.0]    ZCOMP: 0 tags
[  718.021150] nouveau  [  PTHERM][0000:01:00.0] FAN control: PWM
[  718.021178] nouveau  [  PTHERM][0000:01:00.0] fan management: disabled
[  718.021184] nouveau  [  PTHERM][0000:01:00.0] internal sensor: yes
[  718.134635] [TTM] Zone  kernel: Available graphics memory: 8149418 kiB
[  718.134640] [TTM] Zone   dma32: Available graphics memory: 2097152 kiB
[  718.134642] [TTM] Initializing pool allocator
[  718.134649] [TTM] Initializing DMA pool allocator
[  718.134673] mtrr: type mismatch for c0000000,10000000 old: write-back new: write-combining
[  718.134677] nouveau  [     DRM] VRAM: 3072 MiB
[  718.134680] nouveau  [     DRM] GART: 512 MiB
[  718.134688] nouveau  [     DRM] TMDS table version 2.0
[  718.134690] nouveau  [     DRM] DCB version 4.0
[  718.134695] nouveau  [     DRM] DCB outp 00: 01000f13 00010034
[  718.134699] nouveau  [     DRM] DCB outp 01: 01001f12 0f420010
[  718.134701] nouveau  [     DRM] DCB outp 02: 01002f12 00020030
[  718.134704] nouveau  [     DRM] DCB outp 03: 08015fd6 0f420020
[  718.134707] nouveau  [     DRM] DCB outp 04: 08015f92 00020020
[  718.134709] nouveau  [     DRM] DCB outp 05: 08014fc6 0f420010
[  718.134712] nouveau  [     DRM] DCB outp 06: 08014f82 00020010
[  718.134715] nouveau  [     DRM] DCB outp 07: 08013f82 0f420030
[  718.134717] nouveau  [     DRM] DCB outp 08: 04038fb6 0f420010
[  718.134720] nouveau  [     DRM] DCB outp 09: 04038f72 00020010
[  718.134723] nouveau  [     DRM] DCB outp 10: 02027fa6 0f420010
[  718.134726] nouveau  [     DRM] DCB outp 11: 02027f62 00020010
[  718.134728] nouveau  [     DRM] DCB outp 13: 02049f00 00000000
[  718.134732] nouveau  [     DRM] DCB conn 00: 08000040
[  718.134736] nouveau  [     DRM] DCB conn 01: 00001161
[  718.134739] nouveau  [     DRM] DCB conn 02: 00001231
[  718.134742] nouveau  [     DRM] DCB conn 03: 01000331
[  718.134745] nouveau  [     DRM] DCB conn 04: 01000446
[  718.134747] nouveau  [     DRM] DCB conn 05: 02000546
[  718.134750] nouveau  [     DRM] DCB conn 07: 00010746
[  718.134753] nouveau  [     DRM] DCB conn 08: 00020847
[  718.134756] nouveau  [     DRM] DCB conn 09: 00000900
[  718.136929] [drm] Supports vblank timestamp caching Rev 1 (10.10.2010).
[  718.136938] [drm] No driver support for vblank timestamp query.
[  718.136964] nouveau W[     DRM] voltage table 0x50 unknown
[  718.136987] nouveau  [     DRM] 3 available performance level(s)
[  718.137000] nouveau  [     DRM] 0: core 202MHz shader 405MHz memory 405MHz voltage 100mV
[  718.137008] nouveau  [     DRM] 1: core 405MHz shader 810MHz memory 1080MHz voltage 80mV
[  718.137015] nouveau  [     DRM] 3: core 1002MHz shader 2004MHz memory 1080MHz voltage 40mV
[  718.137021] nouveau  [     DRM] c: fanspeed 100%
[  718.171908] nouveau  [     DRM] MM: using COPY for buffer copies
[  718.391148] nouveau 0000:01:00.0: No connectors reported connected with modes
[  718.391162] [drm] Cannot find any crtc or sizes - going 1024x768
[  718.393106] nouveau E[    PBUS][0000:01:00.0] MMIO write of 0x00000000 FAULT at 0x418880 [ IBUS ]
[  718.393131] nouveau E[   PIBUS][0000:01:00.0] GPC1: 0x419eb4 0xbadf1000 (0x3800820c)
[  718.393147] nouveau E[   PIBUS][0000:01:00.0] GPC2: 0x419eb4 0xbadf1000 (0x3800820c)
[  720.395018] nouveau E[  PGRAPH][0000:01:00.0] HUB_INIT timed out
[  720.395027] nouveau E[  PGRAPH][0000:01:00.0] 409000 - done 0x00000244
[  720.395036] nouveau E[  PGRAPH][0000:01:00.0] 409000 - stat 0x000000e6 0x00000000 0x00000000 0x00000000
[  720.395045] nouveau E[  PGRAPH][0000:01:00.0] 409000 - stat 0x00000000 0x00000000 0x00000002 0x00000009
[  720.395049] nouveau E[  PGRAPH][0000:01:00.0] 502000 - done 0x00000300
[  720.395062] nouveau E[  PGRAPH][0000:01:00.0] 502000 - stat 0x00000000 0x00000000 0x00000000 0x00000000
[  720.395075] nouveau E[  PGRAPH][0000:01:00.0] 502000 - stat 0x00000000 0x00000000 0x00000000 0x00000000
[  720.395079] nouveau E[  PGRAPH][0000:01:00.0] 50a000 - done 0x00000300
[  720.395091] nouveau E[  PGRAPH][0000:01:00.0] 50a000 - stat 0x00000000 0x00000000 0x00000000 0x00000000
[  720.395105] nouveau E[  PGRAPH][0000:01:00.0] 50a000 - stat 0x00000000 0x00000000 0x00000000 0x00000000
[  720.395110] nouveau E[  PGRAPH][0000:01:00.0] 512000 - done 0x00000300
[  720.395122] nouveau E[  PGRAPH][0000:01:00.0] 512000 - stat 0x00000000 0x00000000 0x00000000 0x00000000
[  720.395136] nouveau E[  PGRAPH][0000:01:00.0] 512000 - stat 0x00000000 0x00000000 0x00000000 0x00000000
[  720.395139] nouveau E[  PGRAPH][0000:01:00.0] init failed, -16
[  720.396187] nouveau  [     DRM] allocated 1024x768 fb: 0x80000, bo ffff88043cad4400
[  720.396237] nouveau 0000:01:00.0: fb1: nouveaufb frame buffer device
[  720.396240] [drm] Initialized nouveau 1.1.1 20120801 for 0000:01:00.0 on minor 1
@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Jun 14, 2013

Contributor

If you want me to try something out (in the kernel for example) this won't be no problem, because I am building and configuring the kernel on my own.

Contributor

karolherbst commented Jun 14, 2013

If you want me to try something out (in the kernel for example) this won't be no problem, because I am building and configuring the kernel on my own.

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Jun 19, 2013

Contributor

reading this commit https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=448bd857d48e69b33ef323739dc6d8ca20d4cda7
I think it makes a difference, that this gpu is connected via MXM:

"Some devices, such as those in plug-in slots, have no direct platform
logic. For example, there is usually no ACPI _PR3 for them. D3cold
support for these devices can be done via the PCIe Downstream Port leading
to the device. When the PCIe port is powered on/off, the device is powered
on/off too. Wakeup events from the device will be notified to the
corresponding PCIe port."

Contributor

karolherbst commented Jun 19, 2013

reading this commit https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=448bd857d48e69b33ef323739dc6d8ca20d4cda7
I think it makes a difference, that this gpu is connected via MXM:

"Some devices, such as those in plug-in slots, have no direct platform
logic. For example, there is usually no ACPI _PR3 for them. D3cold
support for these devices can be done via the PCIe Downstream Port leading
to the device. When the PCIe port is powered on/off, the device is powered
on/off too. Wakeup events from the device will be notified to the
corresponding PCIe port."

@Lekensteyn

This comment has been minimized.

Show comment Hide comment
@Lekensteyn

Lekensteyn Jun 19, 2013

Owner

There exist an ACPI handle for LNXVIDEO:00. If you want to try saving power with PM features on the PCIe port, try enable device power management for the PCI port using sysfs (also possible using a program like powertop).

Owner

Lekensteyn commented Jun 19, 2013

There exist an ACPI handle for LNXVIDEO:00. If you want to try saving power with PM features on the PCIe port, try enable device power management for the PCI port using sysfs (also possible using a program like powertop).

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Jun 19, 2013

Contributor

yeah, but I think I have to disable the port fully to power off the nvidia card (and need some PCI hotplugging option in the kernel).

I am getting 2,5h to 3h battery lifetime through laptop-mode-tools, which is putting nearly everything into powersave, validated with powertop). But I am sure, that the nvidia card is at d3hot state or maybe something higher.

Contributor

karolherbst commented Jun 19, 2013

yeah, but I think I have to disable the port fully to power off the nvidia card (and need some PCI hotplugging option in the kernel).

I am getting 2,5h to 3h battery lifetime through laptop-mode-tools, which is putting nearly everything into powersave, validated with powertop). But I am sure, that the nvidia card is at d3hot state or maybe something higher.

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Jun 19, 2013

Contributor

Is there anything that can be done on this topic or is it a kernel side issue, because of ACPI and MXM? I would really like to shutdwon this gpu, because of the fan and power usage :(

Contributor

karolherbst commented Jun 19, 2013

Is there anything that can be done on this topic or is it a kernel side issue, because of ACPI and MXM? I would really like to shutdwon this gpu, because of the fan and power usage :(

@amonakov

This comment has been minimized.

Show comment Hide comment
@amonakov

amonakov Jun 20, 2013

Can you try using nouveau+vga_switcheroo? I don't see anything related in your pasted dmesg — is your kernel built without switcheroo enabled?

Can you try using nouveau+vga_switcheroo? I don't see anything related in your pasted dmesg — is your kernel built without switcheroo enabled?

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Jun 20, 2013

Contributor

I am not quite sure, but I guess I have disabled it, because I didn't use it on my old system. I will try it out later.

Contributor

karolherbst commented Jun 20, 2013

I am not quite sure, but I guess I have disabled it, because I didn't use it on my old system. I will try it out later.

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Jun 20, 2013

Contributor

I've enabled switcheroo, but it doesn't seem to work.

I add the lspci output of the nvidia card: http://bpaste.net/show/108521/

Contributor

karolherbst commented Jun 20, 2013

I've enabled switcheroo, but it doesn't seem to work.

I add the lspci output of the nvidia card: http://bpaste.net/show/108521/

@mrozanc

This comment has been minimized.

Show comment Hide comment
@mrozanc

mrozanc Jun 26, 2013

Hello,
I have the same issue with the same laptop and GTX 780M. I use the official nvidia driver 319.32 released yesterday http://www.nvidia.fr/object/linux-display-amd64-319.32-driver.html.
My OS is Arch Linux with the default kernel.

mrozanc commented Jun 26, 2013

Hello,
I have the same issue with the same laptop and GTX 780M. I use the official nvidia driver 319.32 released yesterday http://www.nvidia.fr/object/linux-display-amd64-319.32-driver.html.
My OS is Arch Linux with the default kernel.

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Jun 26, 2013

Contributor

I noticed, that even on windows 8 the nvidia gpu is not turning off (these "speed" LED is the indicator for this, which is always on for me).

By the way, @mrozanc I did not try the newest driver, but do you have the same performance issue: https://devtalk.nvidia.com/default/topic/547066/linux/gainward-geforce-gtx-770-phantom-2gb-half-clock-speed-/post/3830531/#3830531 ?

Contributor

karolherbst commented Jun 26, 2013

I noticed, that even on windows 8 the nvidia gpu is not turning off (these "speed" LED is the indicator for this, which is always on for me).

By the way, @mrozanc I did not try the newest driver, but do you have the same performance issue: https://devtalk.nvidia.com/default/topic/547066/linux/gainward-geforce-gtx-770-phantom-2gb-half-clock-speed-/post/3830531/#3830531 ?

@mrozanc

This comment has been minimized.

Show comment Hide comment
@mrozanc

mrozanc Jun 26, 2013

I did not notice anything particular about this issue and on Windows 8 my card is turning off.
Do you use the drivers provided by clevo?

mrozanc commented Jun 26, 2013

I did not notice anything particular about this issue and on Windows 8 my card is turning off.
Do you use the drivers provided by clevo?

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Jun 26, 2013

Contributor

yes (I can't install drivers from the nvdidia site anyway). I noticed, that the card was sometimes off, maybe an application is still using the card

Contributor

karolherbst commented Jun 26, 2013

yes (I can't install drivers from the nvdidia site anyway). I noticed, that the card was sometimes off, maybe an application is still using the card

@mrozanc

This comment has been minimized.

Show comment Hide comment
@mrozanc

mrozanc Jun 26, 2013

Maybe you did not disable the card for Steam in the settings? By default it was on and Steam was running in background.

mrozanc commented Jun 26, 2013

Maybe you did not disable the card for Steam in the settings? By default it was on and Steam was running in background.

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Jun 26, 2013

Contributor

ahh, yeah, I thought about that, but never checked anything related in the settings. Yes steam is running all the time, because I won't use windows for anything else than steam and some development.
I am sure, that steam is using the card only for the system information panel

Contributor

karolherbst commented Jun 26, 2013

ahh, yeah, I thought about that, but never checked anything related in the settings. Yes steam is running all the time, because I won't use windows for anything else than steam and some development.
I am sure, that steam is using the card only for the system information panel

@amonakov

This comment has been minimized.

Show comment Hide comment
@amonakov

amonakov Jun 26, 2013

@Lekensteyn, the problem seems to be that DEVICE_ACPI_HANDLE returns NULL, right? Do you know why?

@Lekensteyn, the problem seems to be that DEVICE_ACPI_HANDLE returns NULL, right? Do you know why?

@Lekensteyn

This comment has been minimized.

Show comment Hide comment
@Lekensteyn

Lekensteyn Jun 26, 2013

Owner

@amonakov Correct, there is no ACPI handle found for the PCI device. I cannot see the obvious wrong here.

@karolherbst Can you rebuild your kernel with the following patch:

diff --git a/drivers/acpi/glue.c b/drivers/acpi/glue.c
index 40a84cc..f7ab220 100644
--- a/drivers/acpi/glue.c
+++ b/drivers/acpi/glue.c
@@ -16,7 +16,7 @@

 #include "internal.h"

-#define ACPI_GLUE_DEBUG    0
+#define ACPI_GLUE_DEBUG    1
 #if ACPI_GLUE_DEBUG
 #define DBG(fmt, ...)                      \
    printk(KERN_DEBUG PREFIX fmt, ##__VA_ARGS__)

After doing so, please paste a new dmesg.

Note to self: the following patch helps me track what paths set the ACPI handle.

diff --git a/include/linux/device.h b/include/linux/device.h
index c0a1261..945ea87 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -729,7 +729,17 @@ static inline struct device *kobj_to_dev(struct kobject *kobj)

 #ifdef CONFIG_ACPI
 #define ACPI_HANDLE(dev)   ((dev)->acpi_node.handle)
-#define ACPI_HANDLE_SET(dev, _handle_) (dev)->acpi_node.handle = (_handle_)
+#define ACPI_HANDLE_SET(dev, _handle_) { \
+   struct acpi_buffer buf = { ACPI_ALLOCATE_BUFFER, NULL }; \
+   acpi_get_name(_handle_, ACPI_FULL_PATHNAME, &buf); \
+   if (buf.pointer) { \
+       pr_warn("ACPI_HANDLE_SET(%p, %s\n", dev, (char *)buf.pointer); \
+       kfree(buf.pointer); \
+   } else { \
+       pr_warn("ACPI_HANDLE_SET(%p, NULL)\n", dev); \
+   } \
+   dump_stack(); \
+}; (dev)->acpi_node.handle = (_handle_)
 #else
 #define ACPI_HANDLE(dev)   (NULL)
 #define ACPI_HANDLE_SET(dev, _handle_) do { } while (0)

Ran in QEMU, this gave me one including:

[    0.234569] pci 0000:00:03.0: reg 14: [mem 0xfebf1000-0xfebf10ff]
[    0.235806] pci 0000:00:03.0: reg 30: [mem 0xfebe0000-0xfebeffff pref]
[    0.236127] ACPI_HANDLE_SET(ffff88001751d890, \_SB_.PCI0.S3__
[    0.236881] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 3.10.0-rc7+ #2
[    0.237672] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2007
[    0.238440]  00000000ffffffea ffff88001744fa38 ffffffff8135eb09 ffff88001744fa98
[    0.240595]  ffffffff8119487f ffff88001744fa70 0000000000000010 ffff88001750b110
[    0.242012]  0000000000030000 00000000000080d0 ffff8800174e5800 ffff88001744fa98
[    0.243479] Call Trace:
[    0.243925]  [<ffffffff8135eb09>] dump_stack+0x19/0x1b
[    0.244006]  [<ffffffff8119487f>] acpi_bind_one+0x1ae/0x2ea
[    0.244787]  [<ffffffff81194a00>] acpi_platform_notify+0x45/0x67
[    0.245574]  [<ffffffff811f29b3>] device_add+0x103/0x670
[    0.246280]  [<ffffffff81181110>] pci_device_add+0xf0/0x130
[    0.247011]  [<ffffffff81351ad2>] pci_scan_single_device+0x92/0xc0
[    0.248007]  [<ffffffff81181199>] pci_scan_slot+0x49/0x100
[    0.248716]  [<ffffffff81181f7d>] pci_scan_child_bus+0x2d/0xc0
[    0.250018]  [<ffffffff81257f42>] pci_acpi_scan_root+0x302/0x3b0
[    0.250807]  [<ffffffff81198b3a>] acpi_pci_root_add+0x1ec/0x32e
[    0.252007]  [<ffffffff81195946>] ? acpi_scan_match_handler+0x3d/0x6c
[    0.252806]  [<ffffffff81195b40>] acpi_bus_device_attach+0x6f/0xbf
[    0.253622]  [<ffffffff811adcf4>] acpi_ns_walk_namespace+0xbb/0x17b
[    0.254411]  [<ffffffff81195ad1>] ? acpi_bus_type_and_status+0x91/0x91
[    0.255276]  [<ffffffff81195ad1>] ? acpi_bus_type_and_status+0x91/0x91
[    0.256007]  [<ffffffff811ae2a7>] acpi_walk_namespace+0x8e/0xc8
[    0.256754]  [<ffffffff81196aab>] acpi_bus_scan+0x86/0x9a
[    0.257457]  [<ffffffff8168f1d7>] ? acpi_sleep_init+0xaf/0xaf
[    0.258199]  [<ffffffff8168f1d7>] ? acpi_sleep_init+0xaf/0xaf
[    0.258946]  [<ffffffff8168f5e0>] acpi_scan_init+0x52/0x144
[    0.259672]  [<ffffffff8168f1d7>] ? acpi_sleep_init+0xaf/0xaf
[    0.260005]  [<ffffffff8168f3fb>] acpi_init+0x224/0x27e
[    0.260703]  [<ffffffff81671e50>] do_one_initcall+0x7b/0x114
[    0.261457]  [<ffffffff81671ffd>] kernel_init_freeable+0x114/0x197
[    0.262300]  [<ffffffff8167182d>] ? do_early_param+0x8a/0x8a
[    0.263052]  [<ffffffff813508d0>] ? rest_init+0x130/0x130
[    0.264008]  [<ffffffff813508d9>] kernel_init+0x9/0x180
[    0.264723]  [<ffffffff81363c2c>] ret_from_fork+0x7c/0xb0
[    0.265419]  [<ffffffff813508d0>] ? rest_init+0x130/0x130
[    0.266387] pci_bus 0000:00: on NUMA node 0

Owner

Lekensteyn commented Jun 26, 2013

@amonakov Correct, there is no ACPI handle found for the PCI device. I cannot see the obvious wrong here.

@karolherbst Can you rebuild your kernel with the following patch:

diff --git a/drivers/acpi/glue.c b/drivers/acpi/glue.c
index 40a84cc..f7ab220 100644
--- a/drivers/acpi/glue.c
+++ b/drivers/acpi/glue.c
@@ -16,7 +16,7 @@

 #include "internal.h"

-#define ACPI_GLUE_DEBUG    0
+#define ACPI_GLUE_DEBUG    1
 #if ACPI_GLUE_DEBUG
 #define DBG(fmt, ...)                      \
    printk(KERN_DEBUG PREFIX fmt, ##__VA_ARGS__)

After doing so, please paste a new dmesg.

Note to self: the following patch helps me track what paths set the ACPI handle.

diff --git a/include/linux/device.h b/include/linux/device.h
index c0a1261..945ea87 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -729,7 +729,17 @@ static inline struct device *kobj_to_dev(struct kobject *kobj)

 #ifdef CONFIG_ACPI
 #define ACPI_HANDLE(dev)   ((dev)->acpi_node.handle)
-#define ACPI_HANDLE_SET(dev, _handle_) (dev)->acpi_node.handle = (_handle_)
+#define ACPI_HANDLE_SET(dev, _handle_) { \
+   struct acpi_buffer buf = { ACPI_ALLOCATE_BUFFER, NULL }; \
+   acpi_get_name(_handle_, ACPI_FULL_PATHNAME, &buf); \
+   if (buf.pointer) { \
+       pr_warn("ACPI_HANDLE_SET(%p, %s\n", dev, (char *)buf.pointer); \
+       kfree(buf.pointer); \
+   } else { \
+       pr_warn("ACPI_HANDLE_SET(%p, NULL)\n", dev); \
+   } \
+   dump_stack(); \
+}; (dev)->acpi_node.handle = (_handle_)
 #else
 #define ACPI_HANDLE(dev)   (NULL)
 #define ACPI_HANDLE_SET(dev, _handle_) do { } while (0)

Ran in QEMU, this gave me one including:

[    0.234569] pci 0000:00:03.0: reg 14: [mem 0xfebf1000-0xfebf10ff]
[    0.235806] pci 0000:00:03.0: reg 30: [mem 0xfebe0000-0xfebeffff pref]
[    0.236127] ACPI_HANDLE_SET(ffff88001751d890, \_SB_.PCI0.S3__
[    0.236881] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 3.10.0-rc7+ #2
[    0.237672] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2007
[    0.238440]  00000000ffffffea ffff88001744fa38 ffffffff8135eb09 ffff88001744fa98
[    0.240595]  ffffffff8119487f ffff88001744fa70 0000000000000010 ffff88001750b110
[    0.242012]  0000000000030000 00000000000080d0 ffff8800174e5800 ffff88001744fa98
[    0.243479] Call Trace:
[    0.243925]  [<ffffffff8135eb09>] dump_stack+0x19/0x1b
[    0.244006]  [<ffffffff8119487f>] acpi_bind_one+0x1ae/0x2ea
[    0.244787]  [<ffffffff81194a00>] acpi_platform_notify+0x45/0x67
[    0.245574]  [<ffffffff811f29b3>] device_add+0x103/0x670
[    0.246280]  [<ffffffff81181110>] pci_device_add+0xf0/0x130
[    0.247011]  [<ffffffff81351ad2>] pci_scan_single_device+0x92/0xc0
[    0.248007]  [<ffffffff81181199>] pci_scan_slot+0x49/0x100
[    0.248716]  [<ffffffff81181f7d>] pci_scan_child_bus+0x2d/0xc0
[    0.250018]  [<ffffffff81257f42>] pci_acpi_scan_root+0x302/0x3b0
[    0.250807]  [<ffffffff81198b3a>] acpi_pci_root_add+0x1ec/0x32e
[    0.252007]  [<ffffffff81195946>] ? acpi_scan_match_handler+0x3d/0x6c
[    0.252806]  [<ffffffff81195b40>] acpi_bus_device_attach+0x6f/0xbf
[    0.253622]  [<ffffffff811adcf4>] acpi_ns_walk_namespace+0xbb/0x17b
[    0.254411]  [<ffffffff81195ad1>] ? acpi_bus_type_and_status+0x91/0x91
[    0.255276]  [<ffffffff81195ad1>] ? acpi_bus_type_and_status+0x91/0x91
[    0.256007]  [<ffffffff811ae2a7>] acpi_walk_namespace+0x8e/0xc8
[    0.256754]  [<ffffffff81196aab>] acpi_bus_scan+0x86/0x9a
[    0.257457]  [<ffffffff8168f1d7>] ? acpi_sleep_init+0xaf/0xaf
[    0.258199]  [<ffffffff8168f1d7>] ? acpi_sleep_init+0xaf/0xaf
[    0.258946]  [<ffffffff8168f5e0>] acpi_scan_init+0x52/0x144
[    0.259672]  [<ffffffff8168f1d7>] ? acpi_sleep_init+0xaf/0xaf
[    0.260005]  [<ffffffff8168f3fb>] acpi_init+0x224/0x27e
[    0.260703]  [<ffffffff81671e50>] do_one_initcall+0x7b/0x114
[    0.261457]  [<ffffffff81671ffd>] kernel_init_freeable+0x114/0x197
[    0.262300]  [<ffffffff8167182d>] ? do_early_param+0x8a/0x8a
[    0.263052]  [<ffffffff813508d0>] ? rest_init+0x130/0x130
[    0.264008]  [<ffffffff813508d9>] kernel_init+0x9/0x180
[    0.264723]  [<ffffffff81363c2c>] ret_from_fork+0x7c/0xb0
[    0.265419]  [<ffffffff813508d0>] ? rest_init+0x130/0x130
[    0.266387] pci_bus 0000:00: on NUMA node 0

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Jun 26, 2013

Contributor

yeah, I will post the output in about 3 or 4 hours, depends on when I'll come home today.

Contributor

karolherbst commented Jun 26, 2013

yeah, I will post the output in about 3 or 4 hours, depends on when I'll come home today.

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Jun 26, 2013

Contributor
@Lekensteyn

This comment has been minimized.

Show comment Hide comment
@Lekensteyn

Lekensteyn Jun 27, 2013

Owner

This method fails to find a handle:

static int acpi_pci_find_device(struct device *dev, acpi_handle *handle)
{
        struct pci_dev * pci_dev;
        u64     addr;

        pci_dev = to_pci_dev(dev);
        /* Please ref to ACPI spec for the syntax of _ADR */
        addr = (PCI_SLOT(pci_dev->devfn) << 16) | PCI_FUNC(pci_dev->devfn);
        *handle = acpi_get_child(DEVICE_ACPI_HANDLE(dev->parent), addr);
        if (!*handle)
                return -ENODEV;
        return 0;
}

Can you run lspci -tnnv to find its parent device? You can paste it here.

(The fact that this card sits in the MXM slot is probably indeed related)

Owner

Lekensteyn commented Jun 27, 2013

This method fails to find a handle:

static int acpi_pci_find_device(struct device *dev, acpi_handle *handle)
{
        struct pci_dev * pci_dev;
        u64     addr;

        pci_dev = to_pci_dev(dev);
        /* Please ref to ACPI spec for the syntax of _ADR */
        addr = (PCI_SLOT(pci_dev->devfn) << 16) | PCI_FUNC(pci_dev->devfn);
        *handle = acpi_get_child(DEVICE_ACPI_HANDLE(dev->parent), addr);
        if (!*handle)
                return -ENODEV;
        return 0;
}

Can you run lspci -tnnv to find its parent device? You can paste it here.

(The fact that this card sits in the MXM slot is probably indeed related)

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Jun 27, 2013

Contributor
-[0000:00]-+-00.0  Intel Corporation Xeon E3-1200 v3/4th Gen Core Processor DRAM Controller [8086:0c04]
           +-01.0-[01]----00.0  NVIDIA Corporation GK106M [GeForce GTX 770M] [10de:11e0]
           +-02.0  Intel Corporation 4th Gen Core Processor Integrated Graphics Controller [8086:0416]
           +-03.0  Intel Corporation Xeon E3-1200 v3/4th Gen Core Processor HD Audio Controller [8086:0c0c]
           +-14.0  Intel Corporation 8 Series/C220 Series Chipset Family USB xHCI [8086:8c31]
           +-16.0  Intel Corporation 8 Series/C220 Series Chipset Family MEI Controller #1 [8086:8c3a]
           +-1a.0  Intel Corporation 8 Series/C220 Series Chipset Family USB EHCI #2 [8086:8c2d]
           +-1b.0  Intel Corporation 8 Series/C220 Series Chipset High Definition Audio Controller [8086:8c20]
           +-1c.0-[02]--
           +-1c.1-[03-04]----00.0-[04]----00.0  Texas Instruments XIO2213A/B/XIO2221 IEEE-1394b OHCI Controller [Cheetah Express] [104c:823f]
           +-1c.2-[05]--+-00.0  Realtek Semiconductor Co., Ltd. Device [10ec:5289]
           |            \-00.2  Realtek Semiconductor Co., Ltd. RTL8111/8168 PCI Express Gigabit Ethernet controller [10ec:8168]
           +-1c.3-[06]----00.0  Intel Corporation Centrino Wireless-N 2230 [8086:0887]
           +-1d.0  Intel Corporation 8 Series/C220 Series Chipset Family USB EHCI #1 [8086:8c26]
           +-1f.0  Intel Corporation HM87 Express LPC Controller [8086:8c4b]
           +-1f.2  Intel Corporation 8 Series/C220 Series Chipset Family 6-port SATA Controller 1 [AHCI mode] [8086:8c03]
           \-1f.3  Intel Corporation 8 Series/C220 Series Chipset Family SMBus Controller [8086:8c22]
Contributor

karolherbst commented Jun 27, 2013

-[0000:00]-+-00.0  Intel Corporation Xeon E3-1200 v3/4th Gen Core Processor DRAM Controller [8086:0c04]
           +-01.0-[01]----00.0  NVIDIA Corporation GK106M [GeForce GTX 770M] [10de:11e0]
           +-02.0  Intel Corporation 4th Gen Core Processor Integrated Graphics Controller [8086:0416]
           +-03.0  Intel Corporation Xeon E3-1200 v3/4th Gen Core Processor HD Audio Controller [8086:0c0c]
           +-14.0  Intel Corporation 8 Series/C220 Series Chipset Family USB xHCI [8086:8c31]
           +-16.0  Intel Corporation 8 Series/C220 Series Chipset Family MEI Controller #1 [8086:8c3a]
           +-1a.0  Intel Corporation 8 Series/C220 Series Chipset Family USB EHCI #2 [8086:8c2d]
           +-1b.0  Intel Corporation 8 Series/C220 Series Chipset High Definition Audio Controller [8086:8c20]
           +-1c.0-[02]--
           +-1c.1-[03-04]----00.0-[04]----00.0  Texas Instruments XIO2213A/B/XIO2221 IEEE-1394b OHCI Controller [Cheetah Express] [104c:823f]
           +-1c.2-[05]--+-00.0  Realtek Semiconductor Co., Ltd. Device [10ec:5289]
           |            \-00.2  Realtek Semiconductor Co., Ltd. RTL8111/8168 PCI Express Gigabit Ethernet controller [10ec:8168]
           +-1c.3-[06]----00.0  Intel Corporation Centrino Wireless-N 2230 [8086:0887]
           +-1d.0  Intel Corporation 8 Series/C220 Series Chipset Family USB EHCI #1 [8086:8c26]
           +-1f.0  Intel Corporation HM87 Express LPC Controller [8086:8c4b]
           +-1f.2  Intel Corporation 8 Series/C220 Series Chipset Family 6-port SATA Controller 1 [AHCI mode] [8086:8c03]
           \-1f.3  Intel Corporation 8 Series/C220 Series Chipset Family SMBus Controller [8086:8c22]
@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Jun 28, 2013

Contributor

I also noticed, that this MXM slot is supported by ACPI hotplug. Would it be possible to handle the power management this way?

Contributor

karolherbst commented Jun 28, 2013

I also noticed, that this MXM slot is supported by ACPI hotplug. Would it be possible to handle the power management this way?

@wngreene

This comment has been minimized.

Show comment Hide comment
@wngreene

wngreene Jul 7, 2013

Hi. Has there been any progress on this issue? I'm having similar problems trying to get Bumblebee running on my Razer Blade Pro (Haswell i7, Nvidia GTX 765m, Xubuntu 13.04, kernel 3.9.9-030909).

wngreene commented Jul 7, 2013

Hi. Has there been any progress on this issue? I'm having similar problems trying to get Bumblebee running on my Razer Blade Pro (Haswell i7, Nvidia GTX 765m, Xubuntu 13.04, kernel 3.9.9-030909).

@MorphBonehunter

This comment has been minimized.

Show comment Hide comment
@MorphBonehunter

MorphBonehunter Jul 11, 2013

Hello
same problem exists here. Like karolherbst i have an Schenker XMG P503. The GFX Card is an nvidia GTX780M and i can not get bbswitch to work. I tried this with Gentoo (3.8.x and 3.9.x kernel) and at the Moment with ARCH Linux (3.9.9-1-ARCH kernel). I also tried to installed the latest BETA driver from NVIDIA Site (NVIDIA-Linux-x86_64-325.08.run) but it makes no difference:

dmsg:

[ 1017.619649] bbswitch: version 0.7
[ 1017.619661] bbswitch: Found integrated VGA device 0000:00:02.0: \_SB_.PCI0.GFX0
[ 1017.619669] bbswitch: cannot find ACPI handle for VGA device 0000:01:00.0
[ 1017.619674] bbswitch: No discrete VGA device found

lspci -tnnv:

-[0000:00]-+-00.0  Intel Corporation Xeon E3-1200 v3/4th Gen Core Processor DRAM Controller [8086:0c04]
           +-01.0-[01]----00.0  NVIDIA Corporation GK104M [GeForce GTX 780M] [10de:119f]
           +-02.0  Intel Corporation 4th Gen Core Processor Integrated Graphics Controller [8086:0416]
           +-03.0  Intel Corporation Xeon E3-1200 v3/4th Gen Core Processor HD Audio Controller [8086:0c0c]
           +-14.0  Intel Corporation 8 Series/C220 Series Chipset Family USB xHCI [8086:8c31]
           +-16.0  Intel Corporation 8 Series/C220 Series Chipset Family MEI Controller #1 [8086:8c3a]
           +-1a.0  Intel Corporation 8 Series/C220 Series Chipset Family USB EHCI #2 [8086:8c2d]
           +-1b.0  Intel Corporation 8 Series/C220 Series Chipset High Definition Audio Controller [8086:8c20]
           +-1c.0-[02]--
           +-1c.1-[03-04]----00.0-[04]----00.0  Texas Instruments XIO2213A/B/XIO2221 IEEE-1394b OHCI Controller [Cheetah Express] [104c:823f]
           +-1c.2-[05]--+-00.0  Realtek Semiconductor Co., Ltd. Device [10ec:5289]
           |            \-00.2  Realtek Semiconductor Co., Ltd. RTL8111/8168 PCI Express Gigabit Ethernet controller [10ec:8168]
           +-1c.3-[06]----00.0  Intel Corporation Centrino Advanced-N 6235 [8086:088e]
           +-1d.0  Intel Corporation 8 Series/C220 Series Chipset Family USB EHCI #1 [8086:8c26]
           +-1f.0  Intel Corporation HM87 Express LPC Controller [8086:8c4b]
           +-1f.2  Intel Corporation 8 Series/C220 Series Chipset Family 6-port SATA Controller 1 [AHCI mode] [8086:8c03]
           \-1f.3  Intel Corporation 8 Series/C220 Series Chipset Family SMBus Controller [8086:8c22]

cat /proc/acpi/dump_info:

0000:00:00.0 060000
0000:00:01.0 060400 \_SB_.PCI0.P0P2
0000:00:02.0 030000 \_SB_.PCI0.GFX0
0000:00:03.0 040300 \_SB_.PCI0.B0D3
0000:00:14.0 0c0330 \_SB_.PCI0.XHC_
0000:00:16.0 078000
0000:00:1a.0 0c0320 \_SB_.PCI0.EHC2
0000:00:1b.0 040300 \_SB_.PCI0.HDEF
0000:00:1c.0 060400 \_SB_.PCI0.RP01
0000:00:1c.1 060400 \_SB_.PCI0.RP02
0000:00:1c.2 060400 \_SB_.PCI0.RP03
0000:00:1c.3 060400 \_SB_.PCI0.RP04
0000:00:1d.0 0c0320 \_SB_.PCI0.EHC1
0000:00:1f.0 060100 \_SB_.PCI0.LPCB
0000:00:1f.2 010601 \_SB_.PCI0.SAT0
0000:00:1f.3 0c0500 \_SB_.PCI0.SBUS
0000:01:00.0 030000
0000:03:00.0 060400 \_SB_.PCI0.RP02.PXSX
0000:04:00.0 0c0010
0000:05:00.0 ff0000 \_SB_.PCI0.RP03.PXSX
0000:05:00.2 020000 \_SB_.PCI0.RP03.RLAN
0000:06:00.0 028000 \_SB_.PCI0.RP04.PXSX

there is no ACPI Handle for the GFX Card.

Hello
same problem exists here. Like karolherbst i have an Schenker XMG P503. The GFX Card is an nvidia GTX780M and i can not get bbswitch to work. I tried this with Gentoo (3.8.x and 3.9.x kernel) and at the Moment with ARCH Linux (3.9.9-1-ARCH kernel). I also tried to installed the latest BETA driver from NVIDIA Site (NVIDIA-Linux-x86_64-325.08.run) but it makes no difference:

dmsg:

[ 1017.619649] bbswitch: version 0.7
[ 1017.619661] bbswitch: Found integrated VGA device 0000:00:02.0: \_SB_.PCI0.GFX0
[ 1017.619669] bbswitch: cannot find ACPI handle for VGA device 0000:01:00.0
[ 1017.619674] bbswitch: No discrete VGA device found

lspci -tnnv:

-[0000:00]-+-00.0  Intel Corporation Xeon E3-1200 v3/4th Gen Core Processor DRAM Controller [8086:0c04]
           +-01.0-[01]----00.0  NVIDIA Corporation GK104M [GeForce GTX 780M] [10de:119f]
           +-02.0  Intel Corporation 4th Gen Core Processor Integrated Graphics Controller [8086:0416]
           +-03.0  Intel Corporation Xeon E3-1200 v3/4th Gen Core Processor HD Audio Controller [8086:0c0c]
           +-14.0  Intel Corporation 8 Series/C220 Series Chipset Family USB xHCI [8086:8c31]
           +-16.0  Intel Corporation 8 Series/C220 Series Chipset Family MEI Controller #1 [8086:8c3a]
           +-1a.0  Intel Corporation 8 Series/C220 Series Chipset Family USB EHCI #2 [8086:8c2d]
           +-1b.0  Intel Corporation 8 Series/C220 Series Chipset High Definition Audio Controller [8086:8c20]
           +-1c.0-[02]--
           +-1c.1-[03-04]----00.0-[04]----00.0  Texas Instruments XIO2213A/B/XIO2221 IEEE-1394b OHCI Controller [Cheetah Express] [104c:823f]
           +-1c.2-[05]--+-00.0  Realtek Semiconductor Co., Ltd. Device [10ec:5289]
           |            \-00.2  Realtek Semiconductor Co., Ltd. RTL8111/8168 PCI Express Gigabit Ethernet controller [10ec:8168]
           +-1c.3-[06]----00.0  Intel Corporation Centrino Advanced-N 6235 [8086:088e]
           +-1d.0  Intel Corporation 8 Series/C220 Series Chipset Family USB EHCI #1 [8086:8c26]
           +-1f.0  Intel Corporation HM87 Express LPC Controller [8086:8c4b]
           +-1f.2  Intel Corporation 8 Series/C220 Series Chipset Family 6-port SATA Controller 1 [AHCI mode] [8086:8c03]
           \-1f.3  Intel Corporation 8 Series/C220 Series Chipset Family SMBus Controller [8086:8c22]

cat /proc/acpi/dump_info:

0000:00:00.0 060000
0000:00:01.0 060400 \_SB_.PCI0.P0P2
0000:00:02.0 030000 \_SB_.PCI0.GFX0
0000:00:03.0 040300 \_SB_.PCI0.B0D3
0000:00:14.0 0c0330 \_SB_.PCI0.XHC_
0000:00:16.0 078000
0000:00:1a.0 0c0320 \_SB_.PCI0.EHC2
0000:00:1b.0 040300 \_SB_.PCI0.HDEF
0000:00:1c.0 060400 \_SB_.PCI0.RP01
0000:00:1c.1 060400 \_SB_.PCI0.RP02
0000:00:1c.2 060400 \_SB_.PCI0.RP03
0000:00:1c.3 060400 \_SB_.PCI0.RP04
0000:00:1d.0 0c0320 \_SB_.PCI0.EHC1
0000:00:1f.0 060100 \_SB_.PCI0.LPCB
0000:00:1f.2 010601 \_SB_.PCI0.SAT0
0000:00:1f.3 0c0500 \_SB_.PCI0.SBUS
0000:01:00.0 030000
0000:03:00.0 060400 \_SB_.PCI0.RP02.PXSX
0000:04:00.0 0c0010
0000:05:00.0 ff0000 \_SB_.PCI0.RP03.PXSX
0000:05:00.2 020000 \_SB_.PCI0.RP03.RLAN
0000:06:00.0 028000 \_SB_.PCI0.RP04.PXSX

there is no ACPI Handle for the GFX Card.

@amonakov

This comment has been minimized.

Show comment Hide comment
@amonakov

amonakov Jul 11, 2013

Could you please try building older kernels to see if e.g. on 3.5 the ACPI handle was detected?

Could you please try building older kernels to see if e.g. on 3.5 the ACPI handle was detected?

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Jul 11, 2013

Contributor

@amonakov I guess that we could run into haswell problems with this. It's worth a try though

Contributor

karolherbst commented Jul 11, 2013

@amonakov I guess that we could run into haswell problems with this. It's worth a try though

@amonakov

This comment has been minimized.

Show comment Hide comment
@amonakov

amonakov Jul 11, 2013

Sure, just arrange it to boot in text mode and check cat /proc/acpi/dump_info without starting an X session.

Sure, just arrange it to boot in text mode and check cat /proc/acpi/dump_info without starting an X session.

@MorphBonehunter

This comment has been minimized.

Show comment Hide comment
@MorphBonehunter

MorphBonehunter Jul 11, 2013

Ok, thats strange in my eyes. I installed an "old" kernel from "Arch Rollback Machine". I run now on 3.4.7-1-ARCH and the cat /proc/acpi/dump_info tells me:

0000:00:00.0 060000
0000:00:01.0 060400 \_SB_.PCI0.PEG0
0000:00:02.0 030000 \_SB_.PCI0.GFX0
0000:00:03.0 040300 \_SB_.PCI0.B0D3
0000:00:14.0 0c0330 \_SB_.PCI0.XHC_
0000:00:16.0 078000
0000:00:1a.0 0c0320 \_SB_.PCI0.EHC2
0000:00:1b.0 040300 \_SB_.PCI0.HDEF
0000:00:1c.0 060400 \_SB_.PCI0.RP01
0000:00:1c.1 060400 \_SB_.PCI0.RP02
0000:00:1c.2 060400 \_SB_.PCI0.RP03
0000:00:1c.3 060400 \_SB_.PCI0.RP04
0000:00:1d.0 0c0320 \_SB_.PCI0.EHC1
0000:00:1f.0 060100 \_SB_.PCI0.LPCB
0000:00:1f.2 010601 \_SB_.PCI0.SAT0
0000:00:1f.3 0c0500 \_SB_.PCI0.SBUS
0000:01:00.0 030000 \_SB_.PCI0.PEG0.PEGP
0000:03:00.0 060400 \_SB_.PCI0.RP02.PXSX
0000:04:00.0 0c0010
0000:05:00.0 ff0000 \_SB_.PCI0.RP03.PXSX
0000:05:00.2 020000 \_SB_.PCI0.RP03.RLAN
0000:06:00.0 028000 \_SB_.PCI0.RP04.PXSX

So as you can see there is an ACPI handle on 01:00.0 VGA compatible controller: NVIDIA Corporation GK104M [GeForce GTX 780M] (rev a1)

so i checkout the latestes bbswitch from git here (as there is no bbswitch in my used Arch Rollback Machine snapshot) and compile it. now i have in dmesg:

[  386.958014] bbswitch: version 0.7
[  386.958032] bbswitch: Found integrated VGA device 0000:00:02.0: \_SB_.PCI0.GFX0
[  386.958057] bbswitch: Found discrete VGA device 0000:01:00.0: \_SB_.PCI0.PEG0.PEGP
[  386.958324] bbswitch: detected an Optimus _DSM function
[  386.958345] bbswitch: Succesfully loaded. Discrete card 0000:01:00.0 is on

so final testing:

cat /proc/acpi/bbswitch
0000:01:00.0 ON
echo OFF > /proc/acpi/bbswitch
cat /proc/acpi/bbswitch
0000:01:00.0 OFF
dmesg:
[  682.642631] bbswitch: disabling discrete graphics
[  683.079495] pci 0000:01:00.0: power state changed by ACPI to D3

Whoaps...fan silenced and "speed" LED on Laptop ist off!
so ist a kernel problem? ( i have no clue whats broken then...)

Ok, thats strange in my eyes. I installed an "old" kernel from "Arch Rollback Machine". I run now on 3.4.7-1-ARCH and the cat /proc/acpi/dump_info tells me:

0000:00:00.0 060000
0000:00:01.0 060400 \_SB_.PCI0.PEG0
0000:00:02.0 030000 \_SB_.PCI0.GFX0
0000:00:03.0 040300 \_SB_.PCI0.B0D3
0000:00:14.0 0c0330 \_SB_.PCI0.XHC_
0000:00:16.0 078000
0000:00:1a.0 0c0320 \_SB_.PCI0.EHC2
0000:00:1b.0 040300 \_SB_.PCI0.HDEF
0000:00:1c.0 060400 \_SB_.PCI0.RP01
0000:00:1c.1 060400 \_SB_.PCI0.RP02
0000:00:1c.2 060400 \_SB_.PCI0.RP03
0000:00:1c.3 060400 \_SB_.PCI0.RP04
0000:00:1d.0 0c0320 \_SB_.PCI0.EHC1
0000:00:1f.0 060100 \_SB_.PCI0.LPCB
0000:00:1f.2 010601 \_SB_.PCI0.SAT0
0000:00:1f.3 0c0500 \_SB_.PCI0.SBUS
0000:01:00.0 030000 \_SB_.PCI0.PEG0.PEGP
0000:03:00.0 060400 \_SB_.PCI0.RP02.PXSX
0000:04:00.0 0c0010
0000:05:00.0 ff0000 \_SB_.PCI0.RP03.PXSX
0000:05:00.2 020000 \_SB_.PCI0.RP03.RLAN
0000:06:00.0 028000 \_SB_.PCI0.RP04.PXSX

So as you can see there is an ACPI handle on 01:00.0 VGA compatible controller: NVIDIA Corporation GK104M [GeForce GTX 780M] (rev a1)

so i checkout the latestes bbswitch from git here (as there is no bbswitch in my used Arch Rollback Machine snapshot) and compile it. now i have in dmesg:

[  386.958014] bbswitch: version 0.7
[  386.958032] bbswitch: Found integrated VGA device 0000:00:02.0: \_SB_.PCI0.GFX0
[  386.958057] bbswitch: Found discrete VGA device 0000:01:00.0: \_SB_.PCI0.PEG0.PEGP
[  386.958324] bbswitch: detected an Optimus _DSM function
[  386.958345] bbswitch: Succesfully loaded. Discrete card 0000:01:00.0 is on

so final testing:

cat /proc/acpi/bbswitch
0000:01:00.0 ON
echo OFF > /proc/acpi/bbswitch
cat /proc/acpi/bbswitch
0000:01:00.0 OFF
dmesg:
[  682.642631] bbswitch: disabling discrete graphics
[  683.079495] pci 0000:01:00.0: power state changed by ACPI to D3

Whoaps...fan silenced and "speed" LED on Laptop ist off!
so ist a kernel problem? ( i have no clue whats broken then...)

@Lekensteyn

This comment has been minimized.

Show comment Hide comment
@Lekensteyn

Lekensteyn Jul 11, 2013

Owner

Of course it is a kernel problem, ACPI is a kernel subsystem. Can you do a bisect? Meanwhile I will try to use karolherbst information to debug this.

Owner

Lekensteyn commented Jul 11, 2013

Of course it is a kernel problem, ACPI is a kernel subsystem. Can you do a bisect? Meanwhile I will try to use karolherbst information to debug this.

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Jul 11, 2013

Contributor

now that's kind of weird. All in all linux-3.4 was the kernel with the merged haswell stuff ;) Happy as I am I have a linux git checked out somewhere on ym machine and bisected the kernel more than once.

Contributor

karolherbst commented Jul 11, 2013

now that's kind of weird. All in all linux-3.4 was the kernel with the merged haswell stuff ;) Happy as I am I have a linux git checked out somewhere on ym machine and bisected the kernel more than once.

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Jul 11, 2013

Contributor

@MorphBonehunter if you never did that, I will do that. But I can only get home in about 2 hours. If you want to help though, you could report the highest kernel version with working bbswitch :)

Then I would start bisecting from this version

Contributor

karolherbst commented Jul 11, 2013

@MorphBonehunter if you never did that, I will do that. But I can only get home in about 2 hours. If you want to help though, you could report the highest kernel version with working bbswitch :)

Then I would start bisecting from this version

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Jul 11, 2013

Contributor

And by the way: I am happy to know, that this "power" LED is hardware switched and not software (same with the camera fn key ;) ).

I tried days to get the webcam working and found out, that you can enable/disable it with a fn key :D

Contributor

karolherbst commented Jul 11, 2013

And by the way: I am happy to know, that this "power" LED is hardware switched and not software (same with the camera fn key ;) ).

I tried days to get the webcam working and found out, that you can enable/disable it with a fn key :D

@MorphBonehunter

This comment has been minimized.

Show comment Hide comment
@MorphBonehunter

MorphBonehunter Jul 11, 2013

@Lekensteyn: "Of course it is a kernel problem" yes, thats right, was the wrong punctuation character, should be an exclamation mark than an interrogation mark :)

@karolherbst: I'm at work at the moment, but i can try to to figure out the highest kernel version with working bbswitch as i never do an bisect :)

@Lekensteyn: "Of course it is a kernel problem" yes, thats right, was the wrong punctuation character, should be an exclamation mark than an interrogation mark :)

@karolherbst: I'm at work at the moment, but i can try to to figure out the highest kernel version with working bbswitch as i never do an bisect :)

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Jul 11, 2013

Contributor

thanks :), but I will drive home now, so I will just write if it makes still sense or not

Contributor

karolherbst commented Jul 11, 2013

thanks :), but I will drive home now, so I will just write if it makes still sense or not

@amonakov

This comment has been minimized.

Show comment Hide comment
@amonakov

amonakov Jul 11, 2013

Of course it is a kernel problem, ACPI is a kernel subsystem.

Hm, could it be a firmware problem, theoretically?

I wonder if the bisect ends up pointing to the commit that fixed the Lenovo problem :)

Of course it is a kernel problem, ACPI is a kernel subsystem.

Hm, could it be a firmware problem, theoretically?

I wonder if the bisect ends up pointing to the commit that fixed the Lenovo problem :)

@MorphBonehunter

This comment has been minimized.

Show comment Hide comment
@MorphBonehunter

MorphBonehunter Jul 17, 2013

so i tried the commit with kernel 3.9.10 patch 'c7d9ca90aa9497f0b6e301ec67c52dd4b57a7852' and on top of this the patch from https://bugzilla.kernel.org/show_bug.cgi?id=60561#c2
sorry, but i do not have an acpi-handle after that:

0000:00:00.0 060000
0000:00:01.0 060400 \_SB_.PCI0.P0P2
0000:00:02.0 030000 \_SB_.PCI0.GFX0
0000:00:03.0 040300 \_SB_.PCI0.B0D3
0000:00:14.0 0c0330 \_SB_.PCI0.XHC_
0000:00:16.0 078000
0000:00:1a.0 0c0320 \_SB_.PCI0.EHC2
0000:00:1b.0 040300 \_SB_.PCI0.HDEF
0000:00:1c.0 060400 \_SB_.PCI0.RP01
0000:00:1c.1 060400 \_SB_.PCI0.RP02
0000:00:1c.2 060400 \_SB_.PCI0.RP03
0000:00:1c.3 060400 \_SB_.PCI0.RP04
0000:00:1d.0 0c0320 \_SB_.PCI0.EHC1
0000:00:1f.0 060100 \_SB_.PCI0.LPCB
0000:00:1f.2 010601 \_SB_.PCI0.SAT0
0000:00:1f.3 0c0500 \_SB_.PCI0.SBUS
0000:01:00.0 030000
0000:03:00.0 060400 \_SB_.PCI0.RP02.PXSX
0000:04:00.0 0c0010
0000:05:00.0 ff0000 \_SB_.PCI0.RP03.PXSX
0000:05:00.2 020000 \_SB_.PCI0.RP03.RLAN
0000:06:00.0 028000 \_SB_.PCI0.RP04.PXSX

so i tried the commit with kernel 3.9.10 patch 'c7d9ca90aa9497f0b6e301ec67c52dd4b57a7852' and on top of this the patch from https://bugzilla.kernel.org/show_bug.cgi?id=60561#c2
sorry, but i do not have an acpi-handle after that:

0000:00:00.0 060000
0000:00:01.0 060400 \_SB_.PCI0.P0P2
0000:00:02.0 030000 \_SB_.PCI0.GFX0
0000:00:03.0 040300 \_SB_.PCI0.B0D3
0000:00:14.0 0c0330 \_SB_.PCI0.XHC_
0000:00:16.0 078000
0000:00:1a.0 0c0320 \_SB_.PCI0.EHC2
0000:00:1b.0 040300 \_SB_.PCI0.HDEF
0000:00:1c.0 060400 \_SB_.PCI0.RP01
0000:00:1c.1 060400 \_SB_.PCI0.RP02
0000:00:1c.2 060400 \_SB_.PCI0.RP03
0000:00:1c.3 060400 \_SB_.PCI0.RP04
0000:00:1d.0 0c0320 \_SB_.PCI0.EHC1
0000:00:1f.0 060100 \_SB_.PCI0.LPCB
0000:00:1f.2 010601 \_SB_.PCI0.SAT0
0000:00:1f.3 0c0500 \_SB_.PCI0.SBUS
0000:01:00.0 030000
0000:03:00.0 060400 \_SB_.PCI0.RP02.PXSX
0000:04:00.0 0c0010
0000:05:00.0 ff0000 \_SB_.PCI0.RP03.PXSX
0000:05:00.2 020000 \_SB_.PCI0.RP03.RLAN
0000:06:00.0 028000 \_SB_.PCI0.RP04.PXSX
@lantianyu

This comment has been minimized.

Show comment Hide comment
@lantianyu

lantianyu Jul 17, 2013

Hi, please try this patch on the top of commit c7d9ca90a.
https://bugzilla.kernel.org/attachment.cgi?id=106909

Hi, please try this patch on the top of commit c7d9ca90a.
https://bugzilla.kernel.org/attachment.cgi?id=106909

@MorphBonehunter

This comment has been minimized.

Show comment Hide comment
@MorphBonehunter

MorphBonehunter Jul 17, 2013

hi @lantianyu, i tried your patch top of commit 'c7d9ca90aa9497f0b6e301ec67c52dd4b57a7852'.
now it looks much better:

0000:00:00.0 060000
0000:00:01.0 060400 \_SB_.PCI0.PEG0
0000:00:02.0 030000 \_SB_.PCI0.GFX0
0000:00:03.0 040300 \_SB_.PCI0.B0D3
0000:00:14.0 0c0330 \_SB_.PCI0.XHC_
0000:00:16.0 078000
0000:00:1a.0 0c0320 \_SB_.PCI0.EHC2
0000:00:1b.0 040300 \_SB_.PCI0.HDEF
0000:00:1c.0 060400 \_SB_.PCI0.RP01
0000:00:1c.1 060400 \_SB_.PCI0.RP02
0000:00:1c.2 060400 \_SB_.PCI0.RP03
0000:00:1c.3 060400 \_SB_.PCI0.RP04
0000:00:1d.0 0c0320 \_SB_.PCI0.EHC1
0000:00:1f.0 060100 \_SB_.PCI0.LPCB
0000:00:1f.2 010601 \_SB_.PCI0.SAT0
0000:00:1f.3 0c0500 \_SB_.PCI0.SBUS
0000:01:00.0 030000 \_SB_.PCI0.PEG0.PEGP
0000:03:00.0 060400 \_SB_.PCI0.RP02.PXSX
0000:04:00.0 0c0010
0000:05:00.0 ff0000 \_SB_.PCI0.RP03.PXSX
0000:05:00.2 020000 \_SB_.PCI0.RP03.RLAN
0000:06:00.0 028000 \_SB_.PCI0.RP04.PXSX

so bbswitch works with that:

[   45.439939] bbswitch: version 0.7
[   45.439943] bbswitch: Found integrated VGA device 0000:00:02.0: \_SB_.PCI0.GFX0
[   45.439945] bbswitch: Found discrete VGA device 0000:01:00.0: \_SB_.PCI0.PEG0.PEGP
[   45.440002] bbswitch: detected an Optimus _DSM function
[   45.440008] bbswitch: Succesfully loaded. Discrete card 0000:01:00.0 is on
[   45.442073] bbswitch: disabling discrete graphics 

hi @lantianyu, i tried your patch top of commit 'c7d9ca90aa9497f0b6e301ec67c52dd4b57a7852'.
now it looks much better:

0000:00:00.0 060000
0000:00:01.0 060400 \_SB_.PCI0.PEG0
0000:00:02.0 030000 \_SB_.PCI0.GFX0
0000:00:03.0 040300 \_SB_.PCI0.B0D3
0000:00:14.0 0c0330 \_SB_.PCI0.XHC_
0000:00:16.0 078000
0000:00:1a.0 0c0320 \_SB_.PCI0.EHC2
0000:00:1b.0 040300 \_SB_.PCI0.HDEF
0000:00:1c.0 060400 \_SB_.PCI0.RP01
0000:00:1c.1 060400 \_SB_.PCI0.RP02
0000:00:1c.2 060400 \_SB_.PCI0.RP03
0000:00:1c.3 060400 \_SB_.PCI0.RP04
0000:00:1d.0 0c0320 \_SB_.PCI0.EHC1
0000:00:1f.0 060100 \_SB_.PCI0.LPCB
0000:00:1f.2 010601 \_SB_.PCI0.SAT0
0000:00:1f.3 0c0500 \_SB_.PCI0.SBUS
0000:01:00.0 030000 \_SB_.PCI0.PEG0.PEGP
0000:03:00.0 060400 \_SB_.PCI0.RP02.PXSX
0000:04:00.0 0c0010
0000:05:00.0 ff0000 \_SB_.PCI0.RP03.PXSX
0000:05:00.2 020000 \_SB_.PCI0.RP03.RLAN
0000:06:00.0 028000 \_SB_.PCI0.RP04.PXSX

so bbswitch works with that:

[   45.439939] bbswitch: version 0.7
[   45.439943] bbswitch: Found integrated VGA device 0000:00:02.0: \_SB_.PCI0.GFX0
[   45.439945] bbswitch: Found discrete VGA device 0000:01:00.0: \_SB_.PCI0.PEG0.PEGP
[   45.440002] bbswitch: detected an Optimus _DSM function
[   45.440008] bbswitch: Succesfully loaded. Discrete card 0000:01:00.0 is on
[   45.442073] bbswitch: disabling discrete graphics 
@lantianyu

This comment has been minimized.

Show comment Hide comment
@lantianyu

lantianyu Jul 18, 2013

@MorphBonehunter Thanks for test.

@MorphBonehunter Thanks for test.

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Jul 18, 2013

Contributor

thanks guys for the work on this. I can't help here for the next 8 days, because I am on vacation

Contributor

karolherbst commented Jul 18, 2013

thanks guys for the work on this. I can't help here for the next 8 days, because I am on vacation

@amonakov amonakov referenced this issue in Bumblebee-Project/Bumblebee Jul 19, 2013

Closed

Nvidia GT750M in ASUS N550JV #446

@Lekensteyn

This comment has been minimized.

Show comment Hide comment
@MorphBonehunter

This comment has been minimized.

Show comment Hide comment
@MorphBonehunter

MorphBonehunter Jul 28, 2013

@Lekensteyn mhh...it seems it's something wrong with the patch on my system, patch could not be applied on top of 3.9.11 without any other patches (vanilla version, not gentoo version):

cassandra linux-3.9.11 # patch -p1 < ../acpi-glue-find-child-heuristics.patch 
patching file drivers/acpi/glue.c
Hunk #1 FAILED at 78.
1 out of 1 hunk FAILED -- saving rejects to file drivers/acpi/glue.c.rej
patching file include/acpi/acpi_bus.h
Hunk #1 succeeded at 429 (offset -14 lines).
patching file drivers/pci/pci-acpi.c
Hunk #1 succeeded at 297 (offset -18 lines).

sorry, but i can not say what the problem should be, as from my sight after looking in the C file, the patch should work i think but my skills with this are very limited

@Lekensteyn mhh...it seems it's something wrong with the patch on my system, patch could not be applied on top of 3.9.11 without any other patches (vanilla version, not gentoo version):

cassandra linux-3.9.11 # patch -p1 < ../acpi-glue-find-child-heuristics.patch 
patching file drivers/acpi/glue.c
Hunk #1 FAILED at 78.
1 out of 1 hunk FAILED -- saving rejects to file drivers/acpi/glue.c.rej
patching file include/acpi/acpi_bus.h
Hunk #1 succeeded at 429 (offset -14 lines).
patching file drivers/pci/pci-acpi.c
Hunk #1 succeeded at 297 (offset -18 lines).

sorry, but i can not say what the problem should be, as from my sight after looking in the C file, the patch should work i think but my skills with this are very limited

@Lekensteyn

This comment has been minimized.

Show comment Hide comment
@Lekensteyn

Lekensteyn Jul 28, 2013

Owner

Try applying it on top of 3.11 which has commit b842701f3d800a0fba3cee5ec4163a2356f3aaa5, otherwise try to manually change the contents. Basically, the acpi_get_child function is being replaced for a more extensive function. Let me know if you still cannot figure out, then I'll prepare a patch for 3.9.

Owner

Lekensteyn commented Jul 28, 2013

Try applying it on top of 3.11 which has commit b842701f3d800a0fba3cee5ec4163a2356f3aaa5, otherwise try to manually change the contents. Basically, the acpi_get_child function is being replaced for a more extensive function. Let me know if you still cannot figure out, then I'll prepare a patch for 3.9.

@MorphBonehunter

This comment has been minimized.

Show comment Hide comment
@MorphBonehunter

MorphBonehunter Jul 29, 2013

Ahh...Ok missunderstood this.
I apply the patch now to 3.11-rc3. Applying works but i got no acpi handle after running the kernel:

0000:00:00.0 060000 
0000:00:01.0 060400 \_SB_.PCI0.P0P2
0000:00:02.0 030000 \_SB_.PCI0.GFX0
0000:00:03.0 040300 \_SB_.PCI0.B0D3
0000:00:14.0 0c0330 \_SB_.PCI0.XHC_
0000:00:16.0 078000 
0000:00:1a.0 0c0320 \_SB_.PCI0.EHC2
0000:00:1b.0 040300 \_SB_.PCI0.HDEF
0000:00:1c.0 060400 \_SB_.PCI0.RP01
0000:00:1c.1 060400 \_SB_.PCI0.RP02
0000:00:1c.2 060400 \_SB_.PCI0.RP03
0000:00:1c.3 060400 \_SB_.PCI0.RP04
0000:00:1d.0 0c0320 \_SB_.PCI0.EHC1
0000:00:1f.0 060100 \_SB_.PCI0.LPCB
0000:00:1f.2 010601 \_SB_.PCI0.SAT0
0000:00:1f.3 0c0500 \_SB_.PCI0.SBUS
0000:01:00.0 030000 
0000:03:00.0 060400 \_SB_.PCI0.RP02.PXSX
0000:04:00.0 0c0010 
0000:05:00.0 ff0000 \_SB_.PCI0.RP03.PXSX
0000:05:00.2 020000 \_SB_.PCI0.RP03.RLAN
0000:06:00.0 028000 \_SB_.PCI0.RP04.PXSX

Ahh...Ok missunderstood this.
I apply the patch now to 3.11-rc3. Applying works but i got no acpi handle after running the kernel:

0000:00:00.0 060000 
0000:00:01.0 060400 \_SB_.PCI0.P0P2
0000:00:02.0 030000 \_SB_.PCI0.GFX0
0000:00:03.0 040300 \_SB_.PCI0.B0D3
0000:00:14.0 0c0330 \_SB_.PCI0.XHC_
0000:00:16.0 078000 
0000:00:1a.0 0c0320 \_SB_.PCI0.EHC2
0000:00:1b.0 040300 \_SB_.PCI0.HDEF
0000:00:1c.0 060400 \_SB_.PCI0.RP01
0000:00:1c.1 060400 \_SB_.PCI0.RP02
0000:00:1c.2 060400 \_SB_.PCI0.RP03
0000:00:1c.3 060400 \_SB_.PCI0.RP04
0000:00:1d.0 0c0320 \_SB_.PCI0.EHC1
0000:00:1f.0 060100 \_SB_.PCI0.LPCB
0000:00:1f.2 010601 \_SB_.PCI0.SAT0
0000:00:1f.3 0c0500 \_SB_.PCI0.SBUS
0000:01:00.0 030000 
0000:03:00.0 060400 \_SB_.PCI0.RP02.PXSX
0000:04:00.0 0c0010 
0000:05:00.0 ff0000 \_SB_.PCI0.RP03.PXSX
0000:05:00.2 020000 \_SB_.PCI0.RP03.RLAN
0000:06:00.0 028000 \_SB_.PCI0.RP04.PXSX
@Lekensteyn

This comment has been minimized.

Show comment Hide comment
@Lekensteyn

Lekensteyn Jul 29, 2013

Owner

Thanks for testing, how unfortunate that it didn't work. Can you double-check that you have applied the patch and booted the patched kernel?

Owner

Lekensteyn commented Jul 29, 2013

Thanks for testing, how unfortunate that it didn't work. Can you double-check that you have applied the patch and booted the patched kernel?

@MorphBonehunter

This comment has been minimized.

Show comment Hide comment
@MorphBonehunter

MorphBonehunter Jul 29, 2013

Ok, i checked it and run the whole precedure again just for sure (dl kernel, apply patch, build kernel, reboot with new kernel, rebuild and load acpi_dump_info).
Unfortunaly the same result:

Linux cassandra 3.11.0-rc3 #1 SMP Mon Jul 29 12:36:34 CEST 2013 x86_64 Intel(R) Core(TM) i7-4700MQ CPU @ 2.40GHz GenuineIntel GNU/Linux
0000:00:00.0 060000 
0000:00:01.0 060400 \_SB_.PCI0.P0P2
0000:00:02.0 030000 \_SB_.PCI0.GFX0
0000:00:03.0 040300 \_SB_.PCI0.B0D3
0000:00:14.0 0c0330 \_SB_.PCI0.XHC_
0000:00:16.0 078000 
0000:00:1a.0 0c0320 \_SB_.PCI0.EHC2
0000:00:1b.0 040300 \_SB_.PCI0.HDEF
0000:00:1c.0 060400 \_SB_.PCI0.RP01
0000:00:1c.1 060400 \_SB_.PCI0.RP02
0000:00:1c.2 060400 \_SB_.PCI0.RP03
0000:00:1c.3 060400 \_SB_.PCI0.RP04
0000:00:1d.0 0c0320 \_SB_.PCI0.EHC1
0000:00:1f.0 060100 \_SB_.PCI0.LPCB
0000:00:1f.2 010601 \_SB_.PCI0.SAT0
0000:00:1f.3 0c0500 \_SB_.PCI0.SBUS
0000:01:00.0 030000 
0000:03:00.0 060400 \_SB_.PCI0.RP02.PXSX
0000:04:00.0 0c0010 
0000:05:00.0 ff0000 \_SB_.PCI0.RP03.PXSX
0000:05:00.2 020000 \_SB_.PCI0.RP03.RLAN
0000:06:00.0 028000 \_SB_.PCI0.RP04.PXSX

Ok, i checked it and run the whole precedure again just for sure (dl kernel, apply patch, build kernel, reboot with new kernel, rebuild and load acpi_dump_info).
Unfortunaly the same result:

Linux cassandra 3.11.0-rc3 #1 SMP Mon Jul 29 12:36:34 CEST 2013 x86_64 Intel(R) Core(TM) i7-4700MQ CPU @ 2.40GHz GenuineIntel GNU/Linux
0000:00:00.0 060000 
0000:00:01.0 060400 \_SB_.PCI0.P0P2
0000:00:02.0 030000 \_SB_.PCI0.GFX0
0000:00:03.0 040300 \_SB_.PCI0.B0D3
0000:00:14.0 0c0330 \_SB_.PCI0.XHC_
0000:00:16.0 078000 
0000:00:1a.0 0c0320 \_SB_.PCI0.EHC2
0000:00:1b.0 040300 \_SB_.PCI0.HDEF
0000:00:1c.0 060400 \_SB_.PCI0.RP01
0000:00:1c.1 060400 \_SB_.PCI0.RP02
0000:00:1c.2 060400 \_SB_.PCI0.RP03
0000:00:1c.3 060400 \_SB_.PCI0.RP04
0000:00:1d.0 0c0320 \_SB_.PCI0.EHC1
0000:00:1f.0 060100 \_SB_.PCI0.LPCB
0000:00:1f.2 010601 \_SB_.PCI0.SAT0
0000:00:1f.3 0c0500 \_SB_.PCI0.SBUS
0000:01:00.0 030000 
0000:03:00.0 060400 \_SB_.PCI0.RP02.PXSX
0000:04:00.0 0c0010 
0000:05:00.0 ff0000 \_SB_.PCI0.RP03.PXSX
0000:05:00.2 020000 \_SB_.PCI0.RP03.RLAN
0000:06:00.0 028000 \_SB_.PCI0.RP04.PXSX
@MorphBonehunter

This comment has been minimized.

Show comment Hide comment
@MorphBonehunter

MorphBonehunter Aug 6, 2013

So today I saw by accident a new debug patch on this case on the kernel bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=60561#c22
So I give them a try and apply the pci_is_bridge.patch above the acpi-glue-find-child-heuristics.patch on kernel 3.11-rc4.
After that i have an ACPI handle on the device

0000:00:00.0 060000 
0000:00:01.0 060400 \_SB_.PCI0.PEG0
0000:00:02.0 030000 \_SB_.PCI0.GFX0
0000:00:03.0 040300 \_SB_.PCI0.B0D3
0000:00:14.0 0c0330 \_SB_.PCI0.XHC_
0000:00:16.0 078000 
0000:00:1a.0 0c0320 \_SB_.PCI0.EHC2
0000:00:1b.0 040300 \_SB_.PCI0.HDEF
0000:00:1c.0 060400 \_SB_.PCI0.RP01
0000:00:1c.1 060400 \_SB_.PCI0.RP02
0000:00:1c.2 060400 \_SB_.PCI0.RP03
0000:00:1c.3 060400 \_SB_.PCI0.RP04
0000:00:1d.0 0c0320 \_SB_.PCI0.EHC1
0000:00:1f.0 060100 \_SB_.PCI0.LPCB
0000:00:1f.2 010601 \_SB_.PCI0.SAT0
0000:00:1f.3 0c0500 \_SB_.PCI0.SBUS
0000:01:00.0 030000 \_SB_.PCI0.PEG0.PEGP
0000:03:00.0 060400 \_SB_.PCI0.RP02.PXSX
0000:04:00.0 0c0010 
0000:05:00.0 ff0000 \_SB_.PCI0.RP03.PXSX
0000:05:00.2 020000 \_SB_.PCI0.RP03.RLAN
0000:06:00.0 028000 \_SB_.PCI0.RP04.PXSX

So I can use bbswitch to disable the NVIDIA card ( Yeah! :) )
The two strange thing are that bbswitch doesn't turn off the card per default (i have to echo OFF > /proc/acpi/bbswitch) and i have an ACPI Warning in kernel log:

Aug  6 13:33:24 cassandra kernel: [   18.485425] bbswitch: version 0.7
Aug  6 13:33:24 cassandra kernel: [   18.485429] bbswitch: Found integrated VGA device 0000:00:02.0: \_SB_.PCI0.GFX0
Aug  6 13:33:24 cassandra kernel: [   18.485432] bbswitch: Found discrete VGA device 0000:01:00.0: \_SB_.PCI0.PEG0.PEGP
Aug  6 13:33:24 cassandra kernel: [   18.485439] ACPI Warning: \_SB_.PCI0.PEG0.PEGP._DSM: Argument #4 type mismatch - Found [Buffer], ACPI requires [Package] (20130517/nsarguments-95)
Aug  6 13:33:24 cassandra kernel: [   18.485496] bbswitch: detected an Optimus _DSM function
Aug  6 13:33:24 cassandra kernel: [   18.485504] pci 0000:01:00.0: enabling device (0006 -> 0007)
Aug  6 13:33:24 cassandra kernel: [   18.485538] bbswitch: Succesfully loaded. Discrete card 0000:01:00.0 is on

So today I saw by accident a new debug patch on this case on the kernel bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=60561#c22
So I give them a try and apply the pci_is_bridge.patch above the acpi-glue-find-child-heuristics.patch on kernel 3.11-rc4.
After that i have an ACPI handle on the device

0000:00:00.0 060000 
0000:00:01.0 060400 \_SB_.PCI0.PEG0
0000:00:02.0 030000 \_SB_.PCI0.GFX0
0000:00:03.0 040300 \_SB_.PCI0.B0D3
0000:00:14.0 0c0330 \_SB_.PCI0.XHC_
0000:00:16.0 078000 
0000:00:1a.0 0c0320 \_SB_.PCI0.EHC2
0000:00:1b.0 040300 \_SB_.PCI0.HDEF
0000:00:1c.0 060400 \_SB_.PCI0.RP01
0000:00:1c.1 060400 \_SB_.PCI0.RP02
0000:00:1c.2 060400 \_SB_.PCI0.RP03
0000:00:1c.3 060400 \_SB_.PCI0.RP04
0000:00:1d.0 0c0320 \_SB_.PCI0.EHC1
0000:00:1f.0 060100 \_SB_.PCI0.LPCB
0000:00:1f.2 010601 \_SB_.PCI0.SAT0
0000:00:1f.3 0c0500 \_SB_.PCI0.SBUS
0000:01:00.0 030000 \_SB_.PCI0.PEG0.PEGP
0000:03:00.0 060400 \_SB_.PCI0.RP02.PXSX
0000:04:00.0 0c0010 
0000:05:00.0 ff0000 \_SB_.PCI0.RP03.PXSX
0000:05:00.2 020000 \_SB_.PCI0.RP03.RLAN
0000:06:00.0 028000 \_SB_.PCI0.RP04.PXSX

So I can use bbswitch to disable the NVIDIA card ( Yeah! :) )
The two strange thing are that bbswitch doesn't turn off the card per default (i have to echo OFF > /proc/acpi/bbswitch) and i have an ACPI Warning in kernel log:

Aug  6 13:33:24 cassandra kernel: [   18.485425] bbswitch: version 0.7
Aug  6 13:33:24 cassandra kernel: [   18.485429] bbswitch: Found integrated VGA device 0000:00:02.0: \_SB_.PCI0.GFX0
Aug  6 13:33:24 cassandra kernel: [   18.485432] bbswitch: Found discrete VGA device 0000:01:00.0: \_SB_.PCI0.PEG0.PEGP
Aug  6 13:33:24 cassandra kernel: [   18.485439] ACPI Warning: \_SB_.PCI0.PEG0.PEGP._DSM: Argument #4 type mismatch - Found [Buffer], ACPI requires [Package] (20130517/nsarguments-95)
Aug  6 13:33:24 cassandra kernel: [   18.485496] bbswitch: detected an Optimus _DSM function
Aug  6 13:33:24 cassandra kernel: [   18.485504] pci 0000:01:00.0: enabling device (0006 -> 0007)
Aug  6 13:33:24 cassandra kernel: [   18.485538] bbswitch: Succesfully loaded. Discrete card 0000:01:00.0 is on
@Lekensteyn

This comment has been minimized.

Show comment Hide comment
@Lekensteyn

Lekensteyn Aug 6, 2013

Owner

@MorphBonehunter Thanks for testing, can you test another patch, https://bugzilla.kernel.org/attachment.cgi?id=107119&format=raw ?

About that _DSM type warning, that cannot be "fixed" without breaking machines, see ee0591b

Owner

Lekensteyn commented Aug 6, 2013

@MorphBonehunter Thanks for testing, can you test another patch, https://bugzilla.kernel.org/attachment.cgi?id=107119&format=raw ?

About that _DSM type warning, that cannot be "fixed" without breaking machines, see ee0591b

@MorphBonehunter

This comment has been minimized.

Show comment Hide comment
@MorphBonehunter

MorphBonehunter Aug 6, 2013

Ok, I tryed the last patch from above on an clean 3.11rc4 and it works.
I do have an ACPI handle and can disable the card:

0000:00:00.0 060000 
0000:00:01.0 060400 \_SB_.PCI0.PEG0
0000:00:02.0 030000 \_SB_.PCI0.GFX0
0000:00:03.0 040300 \_SB_.PCI0.B0D3
0000:00:14.0 0c0330 \_SB_.PCI0.XHC_
0000:00:16.0 078000 
0000:00:1a.0 0c0320 \_SB_.PCI0.EHC2
0000:00:1b.0 040300 \_SB_.PCI0.HDEF
0000:00:1c.0 060400 \_SB_.PCI0.RP01
0000:00:1c.1 060400 \_SB_.PCI0.RP02
0000:00:1c.2 060400 \_SB_.PCI0.RP03
0000:00:1c.3 060400 \_SB_.PCI0.RP04
0000:00:1d.0 0c0320 \_SB_.PCI0.EHC1
0000:00:1f.0 060100 \_SB_.PCI0.LPCB
0000:00:1f.2 010601 \_SB_.PCI0.SAT0
0000:00:1f.3 0c0500 \_SB_.PCI0.SBUS
0000:01:00.0 030000 \_SB_.PCI0.PEG0.PEGP
0000:03:00.0 060400 \_SB_.PCI0.RP02.PXSX
0000:04:00.0 0c0010 
0000:05:00.0 ff0000 \_SB_.PCI0.RP03.PXSX
0000:05:00.2 020000 \_SB_.PCI0.RP03.RLAN
0000:06:00.0 028000 \_SB_.PCI0.RP04.PXSX

@Lekensteyn: i was wondering about the warning, because i do not have this one before, ony after the patching session from today. So i guess it's because of the different detection behavior caused by the patch?

Ok, I tryed the last patch from above on an clean 3.11rc4 and it works.
I do have an ACPI handle and can disable the card:

0000:00:00.0 060000 
0000:00:01.0 060400 \_SB_.PCI0.PEG0
0000:00:02.0 030000 \_SB_.PCI0.GFX0
0000:00:03.0 040300 \_SB_.PCI0.B0D3
0000:00:14.0 0c0330 \_SB_.PCI0.XHC_
0000:00:16.0 078000 
0000:00:1a.0 0c0320 \_SB_.PCI0.EHC2
0000:00:1b.0 040300 \_SB_.PCI0.HDEF
0000:00:1c.0 060400 \_SB_.PCI0.RP01
0000:00:1c.1 060400 \_SB_.PCI0.RP02
0000:00:1c.2 060400 \_SB_.PCI0.RP03
0000:00:1c.3 060400 \_SB_.PCI0.RP04
0000:00:1d.0 0c0320 \_SB_.PCI0.EHC1
0000:00:1f.0 060100 \_SB_.PCI0.LPCB
0000:00:1f.2 010601 \_SB_.PCI0.SAT0
0000:00:1f.3 0c0500 \_SB_.PCI0.SBUS
0000:01:00.0 030000 \_SB_.PCI0.PEG0.PEGP
0000:03:00.0 060400 \_SB_.PCI0.RP02.PXSX
0000:04:00.0 0c0010 
0000:05:00.0 ff0000 \_SB_.PCI0.RP03.PXSX
0000:05:00.2 020000 \_SB_.PCI0.RP03.RLAN
0000:06:00.0 028000 \_SB_.PCI0.RP04.PXSX

@Lekensteyn: i was wondering about the warning, because i do not have this one before, ony after the patching session from today. So i guess it's because of the different detection behavior caused by the patch?

@erinaceous

This comment has been minimized.

Show comment Hide comment
@erinaceous

erinaceous Aug 7, 2013

Clevo W350STQ laptop with GTX765M gpu. Another laptop where the GPU is connected via MXM. Was getting the "cannot find ACPI handle for VGA device 0000:01:00.0" error (exactly the same) when modprobe'ing bbswitch:

uname

3.10.3-1-ARCH

dmesg | fgrep -i bbswitch

[36679.390237] bbswitch: version 0.7
[36679.390246] bbswitch: Found integrated VGA device 0000:00:02.0: \_SB_.PCI0.GFX0
[36679.390251] bbswitch: cannot find ACPI handle for VGA device 0000:01:00.0
[36679.390254] bbswitch: No discrete VGA device found

pacman -Q xorg-server xf86-video-intel nvidia

xorg-server 1.14.2-2
xf86-video-intel 2.21.14-1
nvidia 319.32-4

ls -l /sys/bus/acpi/drivers/video/*/physical_node

lrwxrwxrwx 1 root root 0 Aug  7 02:21 /sys/bus/acpi/drivers/video/LNXVIDEO:01/physical_node -> ../../../../pci0000:00/0000:00:02.0/

grep . /sys/bus/acpi/drivers/video/*/path

/sys/bus/acpi/drivers/video/LNXVIDEO:00/path:\_SB_.PCI0.PEG0.PEGP
/sys/bus/acpi/drivers/video/LNXVIDEO:01/path:\_SB_.PCI0.GFX0

Both intel and nvidia cards confirmed working properly, can use rest of bumblebee, ran some Steam games using primusrun and it used the nvidia gpu, so all fine there. Not being able to switch off the nvidia gpu is the only hurdle here.

Same as @MorphBonehunter, tested this patch (the https://bugzilla.kernel.org/attachment.cgi?id=107119&format=raw one) on clean 3.11-rc4 - it works :D I haven't seen any warnings either:

dmesg | fgrep -i bbswitch, on 3.11-rc4 with that patch applied:

[  668.921711] bbswitch: version 0.7
[  668.921725] bbswitch: Found integrated VGA device 0000:00:02.0: \_SB_.PCI0.GFX0
[  668.921737] bbswitch: Found discrete VGA device 0000:01:00.0: \_SB_.PCI0.PEG0.PEGP
[  668.921961] bbswitch: detected an Optimus _DSM function
[  668.922075] bbswitch: Succesfully loaded. Discrete card 0000:01:00.0 is on

Upon running bumblebeed, it detects bbswitch and switches off the card. The little GPU indicator LED on my laptop switches off, so I guess that's verification that it's working :)

Clevo W350STQ laptop with GTX765M gpu. Another laptop where the GPU is connected via MXM. Was getting the "cannot find ACPI handle for VGA device 0000:01:00.0" error (exactly the same) when modprobe'ing bbswitch:

uname

3.10.3-1-ARCH

dmesg | fgrep -i bbswitch

[36679.390237] bbswitch: version 0.7
[36679.390246] bbswitch: Found integrated VGA device 0000:00:02.0: \_SB_.PCI0.GFX0
[36679.390251] bbswitch: cannot find ACPI handle for VGA device 0000:01:00.0
[36679.390254] bbswitch: No discrete VGA device found

pacman -Q xorg-server xf86-video-intel nvidia

xorg-server 1.14.2-2
xf86-video-intel 2.21.14-1
nvidia 319.32-4

ls -l /sys/bus/acpi/drivers/video/*/physical_node

lrwxrwxrwx 1 root root 0 Aug  7 02:21 /sys/bus/acpi/drivers/video/LNXVIDEO:01/physical_node -> ../../../../pci0000:00/0000:00:02.0/

grep . /sys/bus/acpi/drivers/video/*/path

/sys/bus/acpi/drivers/video/LNXVIDEO:00/path:\_SB_.PCI0.PEG0.PEGP
/sys/bus/acpi/drivers/video/LNXVIDEO:01/path:\_SB_.PCI0.GFX0

Both intel and nvidia cards confirmed working properly, can use rest of bumblebee, ran some Steam games using primusrun and it used the nvidia gpu, so all fine there. Not being able to switch off the nvidia gpu is the only hurdle here.

Same as @MorphBonehunter, tested this patch (the https://bugzilla.kernel.org/attachment.cgi?id=107119&format=raw one) on clean 3.11-rc4 - it works :D I haven't seen any warnings either:

dmesg | fgrep -i bbswitch, on 3.11-rc4 with that patch applied:

[  668.921711] bbswitch: version 0.7
[  668.921725] bbswitch: Found integrated VGA device 0000:00:02.0: \_SB_.PCI0.GFX0
[  668.921737] bbswitch: Found discrete VGA device 0000:01:00.0: \_SB_.PCI0.PEG0.PEGP
[  668.921961] bbswitch: detected an Optimus _DSM function
[  668.922075] bbswitch: Succesfully loaded. Discrete card 0000:01:00.0 is on

Upon running bumblebeed, it detects bbswitch and switches off the card. The little GPU indicator LED on my laptop switches off, so I guess that's verification that it's working :)

@saschagehlich

This comment has been minimized.

Show comment Hide comment
@saschagehlich

saschagehlich Aug 7, 2013

@doomcat: Will I have to recompile the whole kernel or will it be sufficient to only build and modprobe the acpi and pci modules? And do I have to install bumblebee to see a handle or will it show up without bumblebee as well (in case the patch fixes the issue)?

@doomcat: Will I have to recompile the whole kernel or will it be sufficient to only build and modprobe the acpi and pci modules? And do I have to install bumblebee to see a handle or will it show up without bumblebee as well (in case the patch fixes the issue)?

@Lekensteyn

This comment has been minimized.

Show comment Hide comment
@Lekensteyn

Lekensteyn Aug 7, 2013

Owner

@saschagehlich ACPI and PCI are core components that are compiled into the kernel, you need to rebuild your whole kernel. Bumblebee is not needed to get the handle right, that is the task of ACPI/PCI.

Owner

Lekensteyn commented Aug 7, 2013

@saschagehlich ACPI and PCI are core components that are compiled into the kernel, you need to rebuild your whole kernel. Bumblebee is not needed to get the handle right, that is the task of ACPI/PCI.

@saschagehlich

This comment has been minimized.

Show comment Hide comment
@saschagehlich

saschagehlich Aug 8, 2013

Now I managed to patch my 10.0 kernel and I get a proper ACPI handle. But when I run optirun glxspheres I get the following error:

[ 1308.038044] [ERROR]Cannot access secondary GPU - error: [XORG] (EE) Server terminated successfully (0). Closing log file.
[ 1308.038094] [ERROR]Aborting because fallback start is disabled.

The next time I run it, I get:

/bin/sh: error while loading shared libraries: libturbojpeg.so: cannot open shared object file: No such file or directory

The next time I run it, I get the other message again.

dmesg:

[ 1428.913423] bbswitch: enabling discrete graphics
[ 1429.398249] pci 0000:01:00.0: power state changed by ACPI to D0
[ 1429.408186] [drm] Initialized nvidia-drm 0.0.0 20130102 for 0000:01:00.0 on minor 1
[ 1429.408191] NVRM: loading NVIDIA UNIX x86_64 Kernel Module  325.15  Wed Jul 31 18:50:56 PDT 2013
[ 1429.422162] vgaarb: this pci device is not a vga device
[ 1429.422606] nvidia 0000:01:00.0: irq 50 for MSI/MSI-X
[ 1432.856324] xhci_queue_intr_tx: 72 callbacks suppressed
[ 1433.493016] ACPI Error: Field [TMPB] at 286720 exceeds Buffer [ROM1] size 262144 (bits) (20130328/dsopcode-236)
[ 1433.493032] ACPI Error: Method parse/execution failed [\_SB_.PCI0.PEG0.PEGP._ROM] (Node ffff8802234eb500), AE_AML_BUFFER_LIMIT (20130328/psparse-537)
[ 1433.520516] NVRM: RmInitAdapter failed! (0x24:0x38:1131)
[ 1433.520524] NVRM: rm_init_adapter(0) failed
[ 1436.299443] vgaarb: this pci device is not a vga device
[ 1436.299869] nvidia 0000:01:00.0: irq 50 for MSI/MSI-X
[ 1436.470296] vgaarb: this pci device is not a vga device
[ 1437.162439] [drm] Module unloaded
[ 1437.163540] bbswitch: disabling discrete graphics
[ 1437.276134] pci 0000:01:00.0: power state changed by ACPI to D3cold
[ 1437.980473] xhci_queue_intr_tx: 74 callbacks suppressed

Now I managed to patch my 10.0 kernel and I get a proper ACPI handle. But when I run optirun glxspheres I get the following error:

[ 1308.038044] [ERROR]Cannot access secondary GPU - error: [XORG] (EE) Server terminated successfully (0). Closing log file.
[ 1308.038094] [ERROR]Aborting because fallback start is disabled.

The next time I run it, I get:

/bin/sh: error while loading shared libraries: libturbojpeg.so: cannot open shared object file: No such file or directory

The next time I run it, I get the other message again.

dmesg:

[ 1428.913423] bbswitch: enabling discrete graphics
[ 1429.398249] pci 0000:01:00.0: power state changed by ACPI to D0
[ 1429.408186] [drm] Initialized nvidia-drm 0.0.0 20130102 for 0000:01:00.0 on minor 1
[ 1429.408191] NVRM: loading NVIDIA UNIX x86_64 Kernel Module  325.15  Wed Jul 31 18:50:56 PDT 2013
[ 1429.422162] vgaarb: this pci device is not a vga device
[ 1429.422606] nvidia 0000:01:00.0: irq 50 for MSI/MSI-X
[ 1432.856324] xhci_queue_intr_tx: 72 callbacks suppressed
[ 1433.493016] ACPI Error: Field [TMPB] at 286720 exceeds Buffer [ROM1] size 262144 (bits) (20130328/dsopcode-236)
[ 1433.493032] ACPI Error: Method parse/execution failed [\_SB_.PCI0.PEG0.PEGP._ROM] (Node ffff8802234eb500), AE_AML_BUFFER_LIMIT (20130328/psparse-537)
[ 1433.520516] NVRM: RmInitAdapter failed! (0x24:0x38:1131)
[ 1433.520524] NVRM: rm_init_adapter(0) failed
[ 1436.299443] vgaarb: this pci device is not a vga device
[ 1436.299869] nvidia 0000:01:00.0: irq 50 for MSI/MSI-X
[ 1436.470296] vgaarb: this pci device is not a vga device
[ 1437.162439] [drm] Module unloaded
[ 1437.163540] bbswitch: disabling discrete graphics
[ 1437.276134] pci 0000:01:00.0: power state changed by ACPI to D3cold
[ 1437.980473] xhci_queue_intr_tx: 74 callbacks suppressed
@amonakov

This comment has been minimized.

Show comment Hide comment
@amonakov

amonakov Aug 8, 2013

nVidia drivers currently have problems on the 3.10 kernel. This may explain the NVRM errors.

amonakov commented Aug 8, 2013

nVidia drivers currently have problems on the 3.10 kernel. This may explain the NVRM errors.

@saschagehlich

This comment has been minimized.

Show comment Hide comment
@saschagehlich

saschagehlich Aug 8, 2013

@amonakov AFAIK the 3.10 support has been fixed with NVIDIA's 325.15 drivers (released 2 days ago), right?

@amonakov AFAIK the 3.10 support has been fixed with NVIDIA's 325.15 drivers (released 2 days ago), right?

@saschagehlich

This comment has been minimized.

Show comment Hide comment
@saschagehlich

saschagehlich Aug 8, 2013

Woops, it's 325.25 that fixed it.

Woops, it's 325.25 that fixed it.

@amonakov

This comment has been minimized.

Show comment Hide comment
@amonakov

amonakov Aug 8, 2013

The compilation failure was fixed, but the "fallen off the bus" issue still remains.

amonakov commented Aug 8, 2013

The compilation failure was fixed, but the "fallen off the bus" issue still remains.

@fyksen

This comment has been minimized.

Show comment Hide comment
@fyksen

fyksen Aug 10, 2013

I got the same "fallen off the bus" issue. What is the easiest way to fix it? Wil it get fixed with next kernel? I'm on Manjaro, and not confortable compiling my own kernel, without anyone spoonfeading me. Great work buy you guis doh!

fyksen commented Aug 10, 2013

I got the same "fallen off the bus" issue. What is the easiest way to fix it? Wil it get fixed with next kernel? I'm on Manjaro, and not confortable compiling my own kernel, without anyone spoonfeading me. Great work buy you guis doh!

@Lekensteyn

This comment has been minimized.

Show comment Hide comment
@Lekensteyn

Lekensteyn Sep 9, 2013

Owner

Affected users of this bug, can you please test the patch in https://bugzilla.kernel.org/show_bug.cgi?id=60829#c12? It can be applied on top of 3.10.10+ or 3.11-rc1+.

Owner

Lekensteyn commented Sep 9, 2013

Affected users of this bug, can you please test the patch in https://bugzilla.kernel.org/show_bug.cgi?id=60829#c12? It can be applied on top of 3.10.10+ or 3.11-rc1+.

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Sep 9, 2013

Contributor

It works for me out of the box with 3.11

Contributor

karolherbst commented Sep 9, 2013

It works for me out of the box with 3.11

@Lekensteyn

This comment has been minimized.

Show comment Hide comment
@Lekensteyn

Lekensteyn Sep 9, 2013

Owner

@karolherbst Yes, this bug is fixed, but the new patch I linked fixes another bug. Can you test the patch for regressions?

Owner

Lekensteyn commented Sep 9, 2013

@karolherbst Yes, this bug is fixed, but the new patch I linked fixes another bug. Can you test the patch for regressions?

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Sep 9, 2013

Contributor

I see. Yes I will test it later this day and report back to you

Contributor

karolherbst commented Sep 9, 2013

I see. Yes I will test it later this day and report back to you

@karolherbst

This comment has been minimized.

Show comment Hide comment
@karolherbst

karolherbst Sep 9, 2013

Contributor

@Lekensteyn no regression here.

Contributor

karolherbst commented Sep 9, 2013

@Lekensteyn no regression here.

@Lekensteyn

This comment has been minimized.

Show comment Hide comment
@Lekensteyn

Lekensteyn Sep 9, 2013

Owner

Thank you for confirming.

Owner

Lekensteyn commented Sep 9, 2013

Thank you for confirming.

@vlalov

This comment has been minimized.

Show comment Hide comment
@vlalov

vlalov Sep 14, 2013

@Lekensteyn Just tried it on top of 3.10.11 and not only there's no regression, but the nasty ACPI warning about the mismatched parameters is gone! So I guess it works even better.

vlalov commented Sep 14, 2013

@Lekensteyn Just tried it on top of 3.10.11 and not only there's no regression, but the nasty ACPI warning about the mismatched parameters is gone! So I guess it works even better.

@Lekensteyn

This comment has been minimized.

Show comment Hide comment
@Lekensteyn

Lekensteyn Sep 14, 2013

Owner

@vlalov There shouldn't be changes related to the mismatching DSM parameters, I think that you haven't rebuild bbswitch for your kernel, could that be the case?

Owner

Lekensteyn commented Sep 14, 2013

@vlalov There shouldn't be changes related to the mismatching DSM parameters, I think that you haven't rebuild bbswitch for your kernel, could that be the case?

@vlalov

This comment has been minimized.

Show comment Hide comment
@vlalov

vlalov Sep 14, 2013

I'm sure I did. I just tried the older kernel of mine 3.10.10 without this change and the warning is gone as well there. Could it be due to the update of BIOS and EC I did yesterday?

vlalov commented Sep 14, 2013

I'm sure I did. I just tried the older kernel of mine 3.10.10 without this change and the warning is gone as well there. Could it be due to the update of BIOS and EC I did yesterday?

@Lekensteyn

This comment has been minimized.

Show comment Hide comment
@Lekensteyn

Lekensteyn Sep 14, 2013

Owner

It's possible that you do not need this ACPI patch after a BIOS update, but the DSM warning should still be present (or not, if it was only added in 3.11. My memory fails here.)

Owner

Lekensteyn commented Sep 14, 2013

It's possible that you do not need this ACPI patch after a BIOS update, but the DSM warning should still be present (or not, if it was only added in 3.11. My memory fails here.)

@vlalov

This comment has been minimized.

Show comment Hide comment
@vlalov

vlalov Sep 14, 2013

I'll build 3.11 and check. Thank you.

vlalov commented Sep 14, 2013

I'll build 3.11 and check. Thank you.

@ArchangeGabriel

This comment has been minimized.

Show comment Hide comment
@ArchangeGabriel

ArchangeGabriel Dec 27, 2015

Owner

I’m closing this since it has been fixed for a long time now.

Owner

ArchangeGabriel commented Dec 27, 2015

I’m closing this since it has been fixed for a long time now.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment