You are not logged in.
[phas@vostok ~]$ sudo systemctl status bumblebeed
● bumblebeed.service - Bumblebee C Daemon
Loaded: loaded (/usr/lib/systemd/system/bumblebeed.service; disabled; vendor>
Active: active (running) since Thu 2019-09-26 10:31:13 CEST; 5s ago
Main PID: 5520 (bumblebeed)
Tasks: 1
Memory: 1.2M
CGroup: /system.slice/bumblebeed.service
└─5520 /usr/bin/bumblebeed
Sep 26 10:31:13 vostok systemd[1]: Started Bumblebee C Daemon.
Sep 26 10:31:13 vostok bumblebeed[5520]: [ 215.966190] [INFO]/usr/bin/bumblebe>
bumblebeed is started.
[phas@vostok ~]$ dmesg | tail
[ 215.791110] bbswitch: version 0.8
[ 215.791115] bbswitch: Found integrated VGA device 0000:00:02.0: \_SB_.PCI0.GFX0
[ 215.791122] bbswitch: Found discrete VGA device 0000:01:00.0: \_SB_.PCI0.PEG0.PEGP
[ 215.791131] ACPI Warning: \_SB.PCI0.PEG0.PEGP._DSM: Argument #4 type mismatch - Found [Buffer], ACPI requires [Package] (20190703/nsarguments-59)
[ 215.791239] bbswitch: detected an Optimus _DSM function
[ 215.791248] pci 0000:01:00.0: enabling device (0006 -> 0007)
[ 215.791354] bbswitch: Succesfully loaded. Discrete card 0000:01:00.0 is on
[ 215.793616] bbswitch: disabling discrete graphics
[phas@vostok ~]$
bbswitch reports working correctly
100.0% PCI Device: NVIDIA Corporation GP107M [GeForce GTX 1050 Mobile]
But powertop still reports the discrete card at full power and, seconds after bbswitch, the cpu fan ramps up to 100% until the computer is turned off (not rebooted)
Offline
cat /proc/acpi/bbswitch
dmes | grep bbswitch
If we ignore powertop and focus on the fan: is there high CPU load as well (to be the direct cause)?
If so: in what process?
Offline
cat /proc/acpi/bbswitch dmes | grep bbswitch
If we ignore powertop and focus on the fan: is there high CPU load as well (to be the direct cause)?
If so: in what process?
[phas@vostok ~]$ cat /proc/acpi/bbswitch
0000:01:00.0 OFF
[phas@vostok ~]$ dmesg | grep bbswitch
[ 11.541286] bbswitch: version 0.8
[ 11.541290] bbswitch: Found integrated VGA device 0000:00:02.0: \_SB_.PCI0.GFX0
[ 11.541296] bbswitch: Found discrete VGA device 0000:01:00.0: \_SB_.PCI0.PEG0.PEGP
[ 11.541402] bbswitch: detected an Optimus _DSM function
[ 11.541478] bbswitch: Succesfully loaded. Discrete card 0000:01:00.0 is on
[ 11.541820] bbswitch: disabling discrete graphics
Summary: 2281.8 wakeups/second, 0.0 GPU ops/seconds, 0.0 VFS ops/sec and 44.4% CPU use
Summary: 2281.8 wakeups/second, 0.0 GPU ops/seconds, 0.0 VFS ops/sec and 44.4% CPU use
Offline
The last bit only partly answers my question: it looks like half the cores are one fire, so check "top" which process loads the CPU.
However, that's probably the cause for the symptoms and there's no problem w/ bbswitch.
Offline
The last bit only partly answers my question: it looks like half the cores are one fire, so check "top" which process loads the CPU.
However, that's probably the cause for the symptoms and there's no problem w/ bbswitch.
Sorry I took the data in a moment where I had load because I was doing stuff.
Package | Core | CPU 0 CPU 6
| | Average 900 MHz 1089 MHz
Idle | Idle | Idle
| Core | CPU 1 CPU 7
| | Average 930 MHz 1017 MHz
| Idle | Idle
| Core | CPU 2 CPU 8
| | Average 931 MHz 1044 MHz
| Idle | Idle
| Core | CPU 3 CPU 9
| | Average 908 MHz 921 MHz
| Idle | Idle
| Core | CPU 4 CPU 10
| | Average 912 MHz 1229 MHz
| Idle | Idle
| Core | CPU 5 CPU 11
| | Average 959 MHz 1016 MHz
| Idle | Idle
Pkg(HW) | Core(HW) | CPU(OS) 0 CPU(OS) 6
| | C0 active 1.8% 1.2%
| | POLL 0.0% 0.1 ms 0.0% 0.0 ms
| | C1 11.8% 0.6 ms 0.6% 1.5 ms
C2 (pc2) 0.0% | |
C3 (pc3) 0.0% | C3 (cc3) 0.1% | C3 0.1% 0.1 ms 0.1% 0.1 ms
C6 (pc6) 0.0% | C6 (cc6) 1.0% | C6 0.7% 0.2 ms 1.4% 0.5 ms
C7 (pc7) 0.0% | C7 (cc7) 67.4% | C7s 0.0% 0.0 ms 0.0% 0.0 ms
C8 (pc8) 0.0% | | C8 4.7% 0.7 ms 5.6% 1.2 ms
C9 (pc9) 0.0% | | C9 2.6% 0.4 ms 0.0% 0.3 ms
C10 (pc10) 0.0% | |
| | C10 65.1% 2.7 ms 89.1% 12.5 ms
| | C1E 10.9% 1.1 ms 0.3% 0.4 ms
| Core(HW) | CPU(OS) 1 CPU(OS) 7
| | C0 active 0.9% 0.8%
| | POLL 0.0% 0.0 ms 0.0% 0.0 ms
| | C1 0.1% 0.3 ms 0.6% 0.5 ms
| |
| C3 (cc3) 0.0% | C3 0.0% 0.1 ms 0.0% 0.1 ms
| C6 (cc6) 1.2% | C6 0.7% 0.5 ms 0.8% 0.6 ms
| C7 (cc7) 92.9% | C7s 0.0% 0.1 ms 0.0% 0.2 ms
| | C8 6.9% 1.7 ms 2.5% 1.1 ms
| | C9 0.2% 2.4 ms 0.1% 2.5 ms
| |
| | C10 90.0% 12.8 ms 94.0% 16.0 ms
| | C1E 0.1% 0.4 ms 0.1% 0.3 ms
| Core(HW) | CPU(OS) 2 CPU(OS) 8
| | C0 active 0.7% 0.9%
| | POLL 0.0% 0.0 ms 0.0% 0.0 ms
| | C1 0.0% 0.4 ms 0.2% 1.3 ms
| |
| C3 (cc3) 0.0% | C3 0.0% 0.1 ms 0.0% 0.2 ms
| C6 (cc6) 0.9% | C6 0.8% 0.6 ms 0.3% 0.4 ms
| C7 (cc7) 94.5% | C7s 0.0% 0.0 ms 0.0% 0.0 ms
| | C8 2.0% 1.4 ms 3.4% 1.4 ms
| | C9 0.0% 0.9 ms 0.2% 3.8 ms
| |
| | C10 95.7% 18.8 ms 94.0% 19.1 ms
| | C1E 0.1% 0.5 ms 0.1% 0.4 ms
| Core(HW) | CPU(OS) 3 CPU(OS) 9
| | C0 active 0.7% 0.9%
| | POLL 0.0% 0.0 ms 0.0% 0.0 ms
| | C1 0.0% 0.4 ms 0.1% 0.2 ms
| |
| C3 (cc3) 0.0% | C3 0.0% 0.0 ms 0.0% 0.1 ms
| C6 (cc6) 0.4% | C6 0.3% 0.6 ms 0.3% 0.6 ms
| C7 (cc7) 95.0% | C7s 0.0% 0.0 ms 0.0% 0.0 ms
| | C8 1.4% 1.1 ms 2.1% 1.5 ms
| | C9 0.0% 0.0 ms 0.0% 0.8 ms
Summary: 890.6 wakeups/second, 0.0 GPU ops/seconds, 0.0 VFS ops/sec and 9.7% CPU use
Usage Events/s Category Description
3.2 ms/s 159.6 Process [PID 1852] [irq/95-ELAN1200]
1.1 ms/s 166.0 Interrupt [17] idma64.1
2.0 ms/s 136.1 Timer tick_sched_timer
100.0% Device Audio codec hwC0D0: Conexant
17.4 ms/s 104.9 Process [PID 4019] /usr/bin/gnome-shell
20.6 ms/s 75.6 Process [PID 4492] /usr/lib/gnome-terminal-server
11.3 ms/s 64.8 Process [PID 7228] /usr/lib/chromium/chromium --type=renderer --field-trial-handle=1522552500291817644,17934281172205053309,131072 --la
20.2 ms/s 50.8 Process [PID 3867] /usr/lib/Xorg -nolisten tcp -auth /var/run/sddm/{76c92d7d-e60e-4cee-b114-232709a37cfb} -background none -noreset -di
2.3 ms/s 35.2 Process [PID 4977] /usr/lib/chromium/chromium --ppapi-flash-version=32.0.0.255 --ppapi-flash-path=/usr/lib/PepperFlash/libpepflashplaye
0.8 ms/s 14.9 Process [PID 343] /usr/sbin/haveged -w 1024 -v 1 --Foreground
76.0 µs/s 8.4 kWork psi_avgs_work
59.5 µs/s 7.9 kWork flush_to_ldisc
97.6 µs/s 5.8 Process [PID 11] [rcu_preempt]
13.5 µs/s 4.9 kWork intel_atomic_helper_free_state_
68.6 µs/s 4.8 kWork intel_atomic_cleanup_work
0.0 µs/s 4.8 kWork intel_atomic_commit_work
1.8 ms/s 2.5 Process [PID 4160] /usr/share/skypeforlinux/skypeforlinux --executed-from=/home/phas --pid=4146
34.9 µs/s 2.7 Timer watchdog_timer_fn
123.1 µs/s 2.6 Interrupt [0] HI_SOFTIRQ
261.7 µs/s 2.5 Process [PID 4367] /usr/share/skypeforlinux/skypeforlinux --type=renderer --autoplay-policy=no-user-gesture-required --ms-disable-index
404.0 µs/s 2.4 Process [PID 7311] powertop
1.9 ms/s 1.6 Interrupt [7] sched(softirq)
89.6 µs/s 2.1 Process [PID 4657] /usr/share/skypeforlinux/skypeforlinux --type=renderer --autoplay-policy=no-user-gesture-required --ms-disable-index
171.8 µs/s 1.8 Process [PID 4116] /usr/lib/at-spi2-registryd --use-gnome-session
63.4 µs/s 2.0 Process [PID 4667] /usr/share/skypeforlinux/skypeforlinux --type=renderer --autoplay-policy=no-user-gesture-required --ms-disable-index
234.9 µs/s 1.6 Process [PID 4989] /usr/lib/chromium/chromium --ppapi-flash-version=32.0.0.255 --ppapi-flash-path=/usr/lib/PepperFlash/libpepflashplaye
259.0 µs/s 1.4 Process [PID 6871] /usr/lib/chromium/chromium --ppapi-flash-version=32.0.0.255 --ppapi-flash-path=/usr/lib/PepperFlash/libpepflashplaye
1.5 ms/s 0.5 Process [PID 7440] top
41.4 µs/s 1.0 Interrupt [143] i915
25.7 µs/s 1.0 Process [PID 4669] /usr/share/skypeforlinux/skypeforlinux --type=renderer --autoplay-policy=no-user-gesture-required --ms-disable-index
58.0 µs/s 1.0 Process [PID 6020] /usr/share/skypeforlinux/skypeforlinux --type=renderer --autoplay-policy=no-user-gesture-required --ms-disable-index
12.8 µs/s 1.0 kWork pci_pme_list_scan
2.3 ms/s 0.00 Interrupt [17] i2c_designware.1
18.5 µs/s 0.9 Process [PID 3778] /usr/lib/systemd/systemd-logind
74.4 µs/s 0.8 Process [PID 312] [jbd2/sdb2-8]
13.0 µs/s 0.8 kWork blk_mq_requeue_work
1.8 ms/s 0.00 Timer hrtimer_wakeup
13.6 µs/s 0.7 Timer intel_uncore_fw_release_timer
171.4 µs/s 0.6 Process [PID 3511] /usr/bin/NetworkManager --no-daemon
657.2 µs/s 0.4 Process [PID 7439] bash
59.4 µs/s 0.5 kWork i915_hangcheck_elapsed
27.4 µs/s 0.5 Process [PID 2153] [irq/129-iwlwifi]
0.9 ms/s 0.15 Interrupt [1] timer(softirq)
135.2 µs/s 0.4 kWork handle_update
60.1 µs/s 0.4 Interrupt [4] block(softirq)
29.5 µs/s 0.4 kWork retire_work_handler
24.3 µs/s 0.4 Process [PID 4353] /usr/share/skypeforlinux/skypeforlinux --type=gpu-process --no-sandbox --gpu-preferences=KAAAAAAAAACAAACAAQAAAAAAAAA
103.7 µs/s 0.3 kWork __i915_gem_free_work
58.0 µs/s 0.5 kWork i915_hangcheck_elapsed
Usage Device name
31.4% CPU misc
31.4% CPU core
31.4% DRAM
100.0% Audio codec hwC0D0: Conexant
40.6% Display backlight
0.0% USB device: usb-device-8087-0aaa
0.0% USB device: xHCI Host Controller
100.0% I2C Device (i2c-ELAN1200:00): ELAN1200:00
100.0% Radio device: iwlwifi
100.0% PCI Device: Realtek Semiconductor Co., Ltd. RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller
100.0% Radio device: btusb
100.0% PCI Device: Intel Corporation Device a30d
100.0% PCI Device: Intel Corporation Xeon E3-1200 v5/v6 / E3-1500 v5 / 6th/7th Gen Core Processor Gaussian Mixture Model
100.0% PCI Device: Intel Corporation Xeon E3-1200 v5/E3-1500 v5/6th Gen Core Processor PCIe Controller (x16)
100.0% PCI Device: Intel Corporation Xeon E3-1200 v5/E3-1500 v5/6th Gen Core Processor Thermal Subsystem
100.0% PCI Device: Intel Corporation Wireless-AC 9560 [Jefferson Peak]
100.0% PCI Device: Intel Corporation 82801 Mobile SATA Controller [RAID mode]
100.0% PCI Device: Intel Corporation Cannon Lake PCH SMBus Controller
100.0% PCI Device: Intel Corporation UHD Graphics 630 (Mobile)
100.0% PCI Device: Intel Corporation Cannon Lake PCH Thermal Controller
100.0% PCI Device: Intel Corporation Cannon Lake PCH Shared SRAM
100.0% PCI Device: Intel Corporation 8th Gen Core Processor Host Bridge/DRAM Registers
100.0% PCI Device: Intel Corporation Cannon Lake PCH cAVS
100.0% PCI Device: Intel Corporation Cannon Lake PCH USB 3.1 xHCI Host Controller
100.0% PCI Device: Intel Corporation Cannon Lake PCH PCI Express Root Port #15
100.0% PCI Device: Intel Corporation Cannon Lake PCH SPI Controller
100.0% PCI Device: NVIDIA Corporation GP107M [GeForce GTX 1050 Mobile]
42.0% PCI Device: Intel Corporation Cannon Lake PCH Serial IO I2C Controller #1
41.3% runtime-i2c_designware.1
7.5 pkts/s Network interface: wlo1 (iwlwifi)
0.0% PCI Device: Intel Corporation Cannon Lake PCH HECI Controller
0.0% PCI Device: Intel Corporation Device a32a
0.0% runtime-ACPI0003:00
0.0% runtime-INT33A1:00
0.0% PCI Device: Intel Corporation Cannon Lake PCH Serial IO I2C Controller #0
0.0% PCI Device: Intel Corporation Cannon Lake PCH PCI Express Root Port #9
0.0% runtime-PNP0C14:00
0.0% runtime-iTCO_wdt
0.0% runtime-PNP0C0D:00
0.0% runtime-rtsx_usb_ms.1.auto
0.0% runtime-INT3450:00
0.0% runtime-microcode
0.0% runtime-ASUS7000:00
0.0% runtime-intel_rapl_msr.0
0.0% runtime-ATK4002:00
0.0% runtime-PNP0C0C:00
0.0% runtime-reg-dummy
0.0 pkts/s Network interface: eno2 (r8169)
0.0% runtime-PNP0C14:03
Offline
Powertop cannot accurately asses power draw for most single devices, as the relevant sensors aren't available on most hardware (or share relevant information with other parts of the system) so it will "look" like everything is enabled while it actually isn't. If your overall power draw is within reason that's what you should be looking at.
(And if these outputs are while on AC you also won't see the relevant information, try on battery)
Last edited by V1del (2019-09-26 17:05:55)
Offline
Powertop cannot accurately asses power draw for most single devices, as the relevant sensors aren't available on most hardware (or share relevant information with other parts of the system) so it will "look" like everything is enabled while it actually isn't. If your overall power draw is within reason that's what you should be looking at.
(And if these outputs are while on AC you also won't see the relevant information, try on battery)
Ok but brand new high level laptop shouldn't do 2 hours on battery. Especially when was doing 6 on windows
Offline
Maybe, i'd still say look at the actual data. FWIW there have been reports of an issue with 5.3 and newer intel chipsets presumably solved with
i915.enable_guc=0
on your kernel parameters or try the LTS kernel for a test.
I'm still fairly certain your GPU is off and you are actually seeing a different manifestation here. You might also want to set up udev rules or enable TLP or so which triggers a few power saving knobs that aren't automatically triggered otherwise.
Last edited by V1del (2019-09-26 18:29:34)
Offline
We still have no usable data to reflect the CPU load when "the cpu fan ramps up to 100% until the computer is turned off"
Also try to drop bbswitch. If eg. the GPU reacts badly to the deactivation, this should lower the load and increase the runtime.
Especially when was doing 6 on windows
Do you still have a windows installation?
https://wiki.archlinux.org/index.php/Du … t_Start-Up
Offline
We still have no usable data to reflect the CPU load when "the cpu fan ramps up to 100% until the computer is turned off"
I tried and checked multiple times. The cpu load is reasonable and battery life
Now I'm at the airport it's kinda difficoult to do thets but I will be back.
Also try to drop bbswitch. If eg. the GPU reacts badly to the deactivation, this should lower the load and increase the runtime.
tried before didn't work.
Do you still have a windows installation?
Nope.
Offline
If leaving the GPU under control of the driver w/o actually using it does not increase (but maybe even lower?) the battery life, your problem is somewhere else.
This sentence seems capped:
The cpu load is reasonable and battery lifeThe cpu load is reasonable and battery life
I'd try to cut out gnome (or KDE as well) because the indexers of those DEs have a nasty habit of loading the system and draining the battery (though I seem to recall that one can inhibit them when the system is on battery)
Offline
If leaving the GPU under control of the driver w/o actually using it does not increase (but maybe even lower?) the battery life, your problem is somewhere else.
This sentence seems capped:The cpu load is reasonable and battery lifeThe cpu load is reasonable and battery life
I'd try to cut out gnome (or KDE as well) because the indexers of those DEs have a nasty habit of loading the system and draining the battery (though I seem to recall that one can inhibit them when the system is on battery)
I tried to disable gdm and run "powertop" without any guy. The overall power draw is still around 15W
Offline
Did you try
FWIW there have been reports of an issue with 5.3 and newer intel chipsets presumably solved with
i915.enable_guc=0
on your kernel parameters or try the LTS kernel for a test.
?
Offline
I did it. I had to add the following to boot options
acpi_osi=! acpi_osi="!Windows 2009"
(notice the second "!" before "Windows 2009")
After that, bumblebee and bbswitch started to work correctly and the average power usage dropped well below 10W.
The only problem now is that I have to
tee /proc/acpi/bbswitch <<< ON
before launching GNOME and only after that i can do
tee /proc/acpi/bbswitch <<< OFF
otherwise, I will get a black screen after login. But it seems to be a known issue. As soon as I find a proper workaround/fix I'll post it here.
Last edited by riccardocagnasso (2019-10-06 10:50:32)
Offline