search

Bumblebee-Project / bbswitch

Disable discrete graphics (currently nvidia only)
GNU General Public License v2.0
487 stars 78 forks source link

No suitable _DSM call found #160

Open runningyz opened 7 years ago

runningyz commented 7 years ago

There are some problems when I start up my ubuntu 16.04 ,I can't enter into GUI,the screen keeps printing 'bbswitch no suitable _DSM call found',how to fix this.

[ 1843.188259] bbswitch: No suitable _DSM call found.
[ 1843.448777] bbswitch: version 0.8
[ 1843.448782] bbswitch: Found integrated VGA device 0000:00:02.0: \_SB_.PCI0.GFX0
[ 1843.448787] bbswitch: Found discrete VGA device 0000:01:00.0: \_SB_.PCI0.PEG0.PEGP
[ 1843.448798] bbswitch: failed to evaluate \_SB_.PCI0.PEG0.PEGP._DSM {0xF8,0xD8,0x86,0xA4,0xDA,0x0B,0x1B,0x47,0xA7,0x2B,0x60,0x42,0xA6,0xB5,0xBE,0xE0} 0x100 0x0 {0x00,0x00,0x00,0x00}: AE_NOT_FOUND
[ 1843.448803] bbswitch: failed to evaluate \_SB_.PCI0.PEG0.PEGP._DSM {0xA0,0xA0,0x95,0x9D,0x60,0x00,0x48,0x4D,0xB3,0x4D,0x7E,0x5F,0xEA,0x12,0x9F,0xD4} 0x102 0x0 {0x00,0x00,0x00,0x00}: AE_NOT_FOUND
[ 1843.448808] ACPI Warning: \_SB_.PCI0.GFX0._DSM: Argument #4 type mismatch - Found [Buffer], ACPI requires [Package] (20150930/nsarguments-95)
Lekensteyn commented 7 years ago

Please see the README, you need to provide more details like kernel version, lspci, laptop model, etc.

mirh commented 7 years ago

This smells like this

Lekensteyn commented 7 years ago

The last message can be ignored for the above reason, but the other "failed to evaluate" warnings are more problematic.

LukeC8 commented 7 years ago

Same bug here.

Lekensteyn commented 7 years ago

@LukeC8 It should have invoked \_SB.PCI0.RP01.PEGP._DSM. It probably picked \_SB.PCI0.RP01.D055 because that is device 0, function 0 under PCI device 00:1c.0. I think this was a problem that got fixed in later kernel versions, can you try upgrading your kernel? Ubuntu 14.04 is also the previous LTS, consider upgrading to 16.04.

rdjondo commented 7 years ago

I have the same issue with 16.04

rdjondo commented 7 years ago

I have the same issue with 16.04 since I upgraded to Linux 4.4.0-97. Linux 4.4.0-93 seems fine but booting on Linux 4.4.0-97 booting fails on my machine.

dominiksp commented 7 years ago

I have the same issue as runningyz and rdjondo. Since upgrading on Ubuntu 16.04 to Linux 4.4.0-98, I cannot boot into GUI and get the error message 'bbswitch no suitable _DSM call found'. Booting 4.4.0-97 does not work either. However, booting 4.4.0-83 does work. My hardware is an Acer Aspire V5-573G with NVIDIA GeForce GTX 850M.

geoffmelnick commented 7 years ago

I have the same issue,using GT1060 and Ubuntu 16.04.3 lts

geoffmelnick commented 7 years ago

Sorry, lost most of that last message. Installed Ubuntu 16.04.03 and it worked fine. Then I looked for a driver for the new GPU, which is a GTX 1060. I used the one suggested by the software installer, and then all reboots failed, just getting a screen full of the error message from BBSwitch. The message log indicated that the GPU had been found but then it couldn't find a corresponding display device so it deinstalled it. Any suggestions would be gratefully received. I can provide more information but only to a limited extent, as the thing won't boot.

tobixx commented 7 years ago

Same here. Last working Ubuntu 16.04 kernel 4.4.0.93 with upstart.

(Installed nvidia-384, breaks running version -93, purged it, removed everything nvidia related - including bbswitch: now running successfully on newest kernel with intel)

mbacilieri commented 7 years ago

Same here, on Acer Aspire E5-771G-51HQ with nvidia GeForce 820M and nvidia drivers 384. Kernel 4.4.0-98 and Ubuntu 16.04: when I try to start in nvidia mode, a list of error "bbswitch: no suitable _DKMS call found" appears then the system start in graphic recovery mode. The series 4.x seems to be very problematic on Acers also for suspend; this is a further issue. For now I'm stuck on kernel 3.19.0-71 that seems to work good.

cesarizu commented 7 years ago

I'm having a similar problem using the open source nouveau driver: https://bugs.launchpad.net/lpbugreporter/+bug/752542/comments/829

shuklaayush commented 7 years ago

Updating to the HWE kernel (4.10) seems to resolve the issue for me. Try running: sudo apt install --install-recommends linux-image-generic-hwe-16.04 xserver-xorg-hwe-16.04 Ubuntu 16.04.3 ships with the updated kernel.

mx1010 commented 7 years ago

Same thing, but only after install drivers for 1070 MSI Aero GPU. I've tried the PPA drivers and Nvidia's latest. I'd be happy to provide any diagnostics data, if anyone is interested.

shuklaayush commented 7 years ago

@mx1010 Did you try updating the kernel as I have described? It fixed the issue for me.

mx1010 commented 7 years ago

@shuklaayush Yeah, I was still seeing the message in 4.10, depending on which TTY I chose. It was odd bc I could goto F7 and get Unity with no issue but when I went into F1-F5 it'd spam console with "bbswitch: No suitable _DSM call found".

I finally ended up just doing a fresh install on a different storage device, reseating all devices, and it seems to have disappeared, so it was possibly just a bad config setting, or something else ¯_(ツ)_/¯

mbacilieri commented 7 years ago

Well, without changing anything, 4.4.0-101 seems to work properly in my Acer E5-771G.

EDIT I sadly discovered it's not true. It appeared to work only because for unknown reason Xorg tryed to load the nouveau driver which was blacklisted. Finally, no discrete card driver was loaded (neither nouveau than nvidia) and at least Xorg started. Forcing to use the nvidia modules, the system hangs "as usual".

Luckyly, kernel 3.19 continues to work...

mbacilieri commented 6 years ago

More details: now I'm running kernel 4.4-0.103, and in dmesg I see:

[   20.682682] bbswitch: failed to evaluate \_SB_.PCI0.RP05.PXSX._DSM {0xF8,0xD8,0x86,0xA4,0xDA,0x0B,0x1B,0x47,0xA7,0x2B,0x60,0x42,0xA6,0xB5,0xBE,0xE0} 0x100 0x0 {0x00,0x00,0x00,0x00}: AE_NOT_FOUND
[   20.682689] bbswitch: failed to evaluate \_SB_.PCI0.RP05.PXSX._DSM {0xA0,0xA0,0x95,0x9D,0x60,0x00,0x48,0x4D,0xB3,0x4D,0x7E,0x5F,0xEA,0x12,0x9F,0xD4} 0x102 0x0 {0x00,0x00,0x00,0x00}: AE_NOT_FOUND
[   20.682697] ACPI Warning: \_SB_.PCI0.GFX0._DSM: Argument #4 type mismatch - Found [Buffer], ACPI requires [Package] (20150930/nsarguments-95)
[   20.682739] bbswitch: No suitable _DSM call found.

Listing the provider for xrandr, there is no nvidia (it should be):

$ xrandr --listproviders
Providers: number : 1
Provider 0: id: 0x47 cap: 0x9, Source Output, Sink Offload crtcs: 4 outputs: 4 associated providers: 0 name:Intel

The nvidia card is recognized in pci device list:

$ lspci
00:00.0 Host bridge: Intel Corporation Haswell-ULT DRAM Controller (rev 0b)
00:02.0 VGA compatible controller: Intel Corporation Haswell-ULT Integrated Graphics Controller (rev 0b)
00:03.0 Audio device: Intel Corporation Haswell-ULT HD Audio Controller (rev 0b)
00:14.0 USB controller: Intel Corporation 8 Series USB xHCI HC (rev 04)
00:16.0 Communication controller: Intel Corporation 8 Series HECI #0 (rev 04)
00:1b.0 Audio device: Intel Corporation 8 Series HD Audio Controller (rev 04)
00:1c.0 PCI bridge: Intel Corporation 8 Series PCI Express Root Port 1 (rev e4)
00:1c.2 PCI bridge: Intel Corporation 8 Series PCI Express Root Port 3 (rev e4)
00:1c.3 PCI bridge: Intel Corporation 8 Series PCI Express Root Port 4 (rev e4)
00:1c.4 PCI bridge: Intel Corporation 8 Series PCI Express Root Port 5 (rev e4)
00:1d.0 USB controller: Intel Corporation 8 Series USB EHCI #1 (rev 04)
00:1f.0 ISA bridge: Intel Corporation 8 Series LPC Controller (rev 04)
00:1f.2 SATA controller: Intel Corporation 8 Series SATA Controller 1 [AHCI mode] (rev 04)
00:1f.3 SMBus: Intel Corporation 8 Series SMBus Controller (rev 04)
01:00.0 3D controller: NVIDIA Corporation GF117M [GeForce 610M/710M/810M/820M / GT 620M/625M/630M/720M] (rev a1)
04:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller (rev 0c)
05:00.0 Network controller: Realtek Semiconductor Co., Ltd. RTL8723BE PCIe Wireless Network Adapter

If I start in "intel" mode (prime-select intel) I can boot and system works, although I can't use 3d card at all:

$ optirun glxgears
[  775.358917] [ERROR]Cannot access secondary GPU - error: [XORG] (EE) Failed to load module "mouse" (module does not exist, 0)
[  775.358962] [ERROR]Aborting because fallback start is disabled.

If I run "glxgears" in "intel" mode, it runs. Obviously it's not using the nvidia card. If I try to boot/login after "prime-select nvidia" (in nvidia mode) the system hangs in console repeating "No suitable _DSM call found.".

I tryed with nvidia driver 384 and 340, with no difference: strangely, for some reason 340 seems the only "officially recognized" by Ubuntu, for in "software and updates / additional drivers" the only nvidia option is "340". Using the 384, Ubuntu think I'm using the nouveau: that should be impossibile for I have blacklisted nouveau everywhere, starting from grub start line:

GRUB_CMDLINE_LINUX_DEFAULT="quiet splash modprobe.blacklist=nouveau"

added to the blacklisting in nvidia modprobe configuration files. Despite this, Xorg after fails to load nvidia module, try to load nouveau (from Xorg.0.log):

[    36.576] (II) LoadModule: "nvidia"
[    36.576] (WW) Warning, couldn't open module nvidia
[    36.576] (II) UnloadModule: "nvidia"
[    36.576] (II) Unloading nvidia
[    36.576] (EE) Failed to load module "nvidia" (module does not exist, 0)
[    36.576] (II) LoadModule: "nouveau"
[    36.576] (II) Loading /usr/lib/xorg/modules/drivers/nouveau_drv.so
[    36.960] (II) Module nouveau: vendor="X.Org Foundation"
[    36.960]    compiled for 1.18.1, module version = 1.0.12
[    36.960]    Module class: X.Org Video Driver
[    36.960]    ABI class: X.Org Video Driver, version 20.0

It's a very strange behaviour: kernel 3.19 - with the same configuration (I mean: just choosing the kernel at grub menu) - works properly and recognize the nvidia discrete card.

mirh commented 6 years ago

It's a very strange behaviour: kernel 3.19 - with the same configuration works

There's nothing strange about some tens of thousands of lines of code more or less producing.. an impact. And bugs/logs should just be reported against latest kernel (given it's the only 'point' where you could pretend a fix to be added) Otherwise discussing about older releases may still make sense, if somebody wants something that just works™... But there's not point in pretending something as massive as acpi (what causes these problems) to be reworked there.

mbacilieri commented 6 years ago

And bugs/logs should just be reported against latest kernel (given it's the only 'point' where you could pretend a fix to be added)

I understand the point and I agree such policy, although the involved kernel is the 4.4.0-103 which is for now the latest officially shipped to Ubuntu 16.04 LTS. So it's difficult to me to consider it an obsolete kernel.

mirh commented 6 years ago

You should consider ubuntu as obsolete 😄

mbacilieri commented 6 years ago

No way, also the last update of yesterday (kernel 4.4.0-109) didn't change the issue. There were a nvidia driver update too. Anyway, the discrete card is still unaccessible due to the bbswitch error. It begins to be a serious issue, for the kernel 4.4.0-109 patches the Meltdown and Spectre bugs, and being forced to use an old kernel 3.19 means to run unpatched.

anneb commented 6 years ago

For me kernel 4.4.0-83 does not have this issue, the next version I have is 4.4.0-104 which has the issue. Version 4.4.0-109 still has the issue.

My graphics card: sudo lspci -v | grep VGA -A 2 00:02.0 VGA compatible controller: Intel Corporation HD Graphics 5500 (rev 09) (prog-if 00 [VGA controller]) DeviceName: Onboard IGD Subsystem: Dell Broadwell-U Integrated Graphics

Edit: Solved by removing the NVIDIA driver (I do not have an NVIDIA card): https://askubuntu.com/questions/206283/how-can-i-uninstall-a-nvidia-driver-completely

dominiksp commented 6 years ago

I can confirm that the issue still exists with kernel version 4.4.0-109.

I would really appreciate a solution.

Summary: 4.4.0-98 does not work 4.4.0-97 does not work 4.4.0-83 still works My output of dmesg | grep -C 10 bbswitch: [ 3.369457] audit: type=1400 audit(1515833541.740:10): apparmor="STATUS" operation="profile_load" profile="unconfined" name="/usr/lib/snapd/snap-confine" pid=824 comm="apparmor_parser" [ 3.406481] usb 2-5.3: new low-speed USB device number 8 using xhcihcd [ 3.438555] cgroup: new mount options do not match the existing superblock, will be ignored [ 3.502407] usb 2-5.3: New USB device found, idVendor=10d5, idProduct=000d [ 3.502412] usb 2-5.3: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [ 3.502415] usb 2-5.3: Product: HA2-A3 [ 3.502417] usb 2-5.3: Manufacturer: No brand [ 3.502419] usb 2-5.3: SerialNumber: 0\xffffffc2\xffffff92 [ 3.502564] usb 2-5.3: ep 0x2 - rounding interval to 64 microframes, ep desc says 80 microframes [ 3.535528] hidraw: raw HID events driver (C) Jiri Kosina [ 3.535599] bbswitch: version 0.8 [ 3.535604] bbswitch: Found integrated VGA device 0000:00:02.0: _SB.PCI0.GFX0 [ 3.535611] bbswitch: Found discrete VGA device 0000:01:00.0: _SB.PCI0.RP05.PEGP [ 3.535620] ACPI Warning: _SB.PCI0.RP05.PEGP._DSM: Argument #4 type mismatch - Found [Buffer], ACPI requires [Package] (20150930/nsarguments-95) [ 3.535695] bbswitch: detected an Optimus _DSM function [ 3.535703] bbswitch: Succesfully loaded. Discrete card 0000:01:00.0 is on [ 3.560040] usbhid 2-5.3:1.1: couldn't find an input interrupt endpoint [ 3.560061] usbcore: registered new interface driver usbhid [ 3.560063] usbhid: USB HID core driver [ 3.563334] input: Microsoft Microsoft® Nano Transceiver v2.1 as /devices/pci0000:00/0000:00:14.0/usb2/2-5/2-5.1/2-5.1:1.0/0003:045E:07A5.0001/input/input16 [ 3.618614] hid-generic 0003:045E:07A5.0001: input,hidraw0: USB HID v1.11 Keyboard [Microsoft Microsoft® Nano Transceiver v2.1] on usb-0000:00:14.0-5.1/input0 [ 3.618827] input: Microsoft Microsoft® Nano Transceiver v2.1 as /devices/pci0000:00/0000:00:14.0/usb2/2-5/2-5.1/2-5.1:1.1/0003:045E:07A5.0002/input/input17 [ 3.618932] hid-generic 0003:045E:07A5.0002: input,hidraw1: USB HID v1.11 Mouse [Microsoft Microsoft® Nano Transceiver v2.1] on usb-0000:00:14.0-5.1/input1 [ 3.631048] input: Microsoft Microsoft® Nano Transceiver v2.1 as /devices/pci0000:00/0000:00:14.0/usb2/2-5/2-5.1/2-5.1:1.2/0003:045E:07A5.0003/input/input18 [ 3.709575] NET: Registered protocol family 4 [ 3.715732] NET: Registered protocol family 5

Last running kernel version uname -a 4.4.0-83-generic #106-Ubuntu SMP Mon Jun 26 17:54:43 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux

Distribution and version: Ubuntu 16.04.3 LTS

Xorg: unable to identify xorg version

TomLisankie commented 6 years ago

My issue is similar. I will be using my computer (specs here) for no more than 10 minutes. Then all graphics on the screen completely freeze for about 30 seconds to a minute. My screen then completely goes black except for a single underscore in the top left corner. A couple of lines of text quickly flash and then this black screen starts outputting bbswitch: No suitable _DSM call found. repeatedly.

Upgrading the kernel did not help either.

mbacilieri commented 6 years ago

SOLVED The issue is due to the upgrade to 16.04.3 that requires a 4.13 kernel "hwe". When updating the 16.04 at a certain point you must install che "hwe" kernel. First make a full update: # apt-get update && apt-get full-upgrade then install hwe kernel and hwe xorg: # sudo apt-get install --install-recommends linux-generic-hwe-16.04 xserver-xorg-hwe-16.04 Please pay attention for it will uninstall the old xorg and, for nvidia users, the bumblebee/primusrun system. If you have nvidia proprietary drivers, do not forget a: # dpkg-reconfigure nvidia-384 (or whatever version you have), just to ensure the dkms-modules will be configured against new xorg/kernel version. After all, you will reboot with the 4.13 kernel that works.

TomLisankie commented 6 years ago

That worked! Thank you!

dominiksp commented 6 years ago

Thanks a lot, this seems to have solved the issue. I am now booting into 4.13.0-26-generic #29~16.04.2-Ubuntu SMP Tue Jan 9 22:00:44 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux By the way, for some of the commands given, I needed an additional 'sudo'.

mbacilieri commented 6 years ago

Anyway, it wasn't correct that kernel hwe works. After a while, I realized that:

TomLisankie commented 6 years ago

@mbacilieri I'm getting these same issues all over again. What a shame it's not 4.x suitable. But you're running 16.04 with kernel 3.19?

mbacilieri commented 6 years ago

@TomLisankie I ran 16.04 with 3.19 until last days, when I relized I couldn't be stuck for more. So I began to test all combination of kernel, parameters and drivers I could. Yesterday I found a working setup with kernel 4.13 "hwe", with xorg "hwe". I removed all nvidia drivers, using nouveau. The trouble looked in the interaction between nvidia (also newest 390) and kernel of 4.x series. Then I added "bbswitch" in /etc/modules, which for unknown reason it's not automatically loaded at boot time. bbswitch is needed to switch on the discrete card at shutdown, otherwise system hangs instead to power off. With this configuration, all looks ok, both 3D with DRI_PRIME=1 and power management with suspend/shutdown. It's not the best performance, but at least it works properly.

Trevortds commented 6 years ago

This issue may have reoccurred in the release version of 18.04. The HWE kernel no longer fixes it. See #169

whatisor commented 6 years ago

In my case, power of graphic card is problematic.