coretemp widget failing about lain HOT 17 CLOSED

You're not missing depencencies.

You should have temp file in some subdir of /sys/class/thermal.

Give me the output of

ls -R /sys/class/thermal

from lain.

┌─(t0m5k1@b0x)-(3056/pts/1)-(09:09pm:20/01/14)-
└─>(%:~)
└─>> ls /sys/class/thermal
cooling_device0@ cooling_device1@

On 20 January 2014 23:40, Luke Bonham [email protected] wrote:

You're not missing depencencies.

You should have temp file in some subdir of /sys/class/thermal.

Give me the output of

ls /sys/class/thermal

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/20#issuecomment-32799652
.

Regards
Thomas Rand

from lain.

Are you running Arch Linux? Which kernel?
Do you have acpi installed?
Maybe the temp file is in /proc/acpi/thermal_zone ?

from lain.

Kernel3.12.7-2-ARCH x86_64
never needed acpi:
┌─(t0m5k1@b0x)-(3056/pts/0)-(11:35am:21/01/14)-
└─>(%:)
└─>> pacs acpi
[sudo] password for t0m5k1:
community/acpi 1.7-1
Client for battery, power, and thermal readings
community/acpid 2.0.21-1
A daemon for delivering ACPI power management events with netlink
support
community/iasl 20130823-1
Intel ACPI Source Language compiler
community/libacpi 0.2-4
general purpose lib to gather ACPI data

On 21 January 2014 11:25, bakoenig [email protected] wrote:

Are you running Arch Linux? Which kernel?
Do you have acpi installed?
Maybe the temp file is in /proc/acpi/thermal_zone ?

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/20#issuecomment-32832322
.

Regards
Thomas Rand

from lain.

just installed acpi & acpid, rebooted & now:
┌─(t0m5k1@b0x)-(3069/pts/0)-(11:55am:21/01/14)-
└─>(%:)
└─>> cat /sys/class/thermal/thermal_zone0/temp; acpi -V
cat: /sys/class/thermal/thermal_zone0/temp: No such file or directory
No support for device type: power_supply
No support for device type: power_supply
Cooling 0: Processor 0 of 0
Cooling 1: Processor 0 of 0
┌─(t0m5k1@b0x)-(3070/pts/0)-(11:56am:21/01/14)-
└─>(%:)
└─>> cat /proc/acpi/thermal_zone
cat: /proc/acpi/thermal_zone: No such file or directory

So I guess acpi is a prerequisite for this!
ok so now I have to configure this lol
Why not parse the /proc tree as that is present on every machine & gives temps & fan speeds etc

from lain.

Yes, I guess acpi is needed. The temp file has recently been moved from /proc/acpi to /sys/class/thermal (see https://bugs.launchpad.net/ubuntu/+source/linux/+bug/756472). That is why there is some confusion.

From your answer, I do not understand if you have the file thermal_zone0/temp or not. If not, maybe you have a file called thermal_zone1/temp or something similar?

from lain.

I just discovered this:

┌─(t0m5k1@b0x)-(3075/pts/0)-(12:07pm:21/01/14)-
└─>(%:~)
└─>> cat /sys/class/hwmon/hwmon1/device/temp
completing files
temp1_alarm temp1_input temp1_min temp1_type temp2_beep temp2_max temp2_offset temp3_alarm temp3_input temp3_min temp3_type
temp1_beep temp1_max temp1_offset temp2_alarm temp2_input temp2_min temp2_type temp3_beep temp3_max temp3_offset

from lain.

I know that finding those sensor files can be a pain sometimes: different machines with different hardware store them at different places and sometimes they even change names after reboot! Usually /sys/class/thermal works.

Maybe this helps: https://bbs.archlinux.org/viewtopic.php?pid=1125622

from lain.

ok progress, I got my hands dirty & made the following changes to temp.lua:

function update()
    --[[
    local f = io.open("/sys/class/thermal/thermal_zone0/temp")
    --]]
    local f = io.open("/sys/class/hwmon/hwmon0/device/temp1_input")
    if f ~= nil
    then

it appears to be working but it is only the temp of 1 core & not an average across both (i think)

from lain.

I can't control now, but I don't recall having installed acpi.

It's easy to modify the widget in order for the user to set his own temp file path, but it seems your computer stores a temp file for each core. Is it a laptop? If yes, which model?

And give me the output of

ls -R /sys/class/

don't paste it here, try gist or pastebin.

from lain.

result of the command:
http://bpaste.net/show/170455/

I have a pc, here is some info:
System: Host: b0x Kernel: 3.12.7-2-ARCH x86_64 (64 bit) Desktop: Awesome 3.5.2 Distro: Arch Linux
Machine: Mobo: Gigabyte model: M720-US3 version: x.x Bios: Award version: F3m date: 04/24/2009
CPU: Dual core AMD Athlon 64 X2 4200+ (-MCP-) cache: 1024 KB flags: (lm nx sse sse2 sse3 svm)
Clock Speeds: 1: 2357.129 MHz 2: 2357.129 MHz
Graphics: Card: NVIDIA GT218 [GeForce 210] X.Org: 1.15.0 driver: nvidia Resolution: [email protected]
GLX Renderer: GeForce 210/PCIe/SSE2 GLX Version: 3.3.0 NVIDIA 331.38
Audio: Card-1: NVIDIA MCP72XE/MCP72P/MCP78U/MCP78S High Definition Audio driver: snd_hda_intel
Card-2: NVIDIA High Definition Audio Controller driver: snd_hda_intel
Sound: Advanced Linux Sound Architecture ver: k3.12.7-2-ARCH
Network: Card: Realtek RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller driver: r8169
IF: eth0 state: up speed: 100 Mbps duplex: full
Drives: HDD Total Size: 1830.4GB (43.5% used) 1: id: /dev/sda model: ST380817AS size: 80.0GB
2: id: /dev/sdd model: ST3500514NS size: 500.1GB 3: id: /dev/sdb model: WDC_WD2500BB size: 250.1GB
4: id: /dev/sdc model: ST3500514NS size: 500.1GB 5: USB id: /dev/sde model: 0NS size: 500.1GB
Partition: ID: / size: 22G used: 13G (62%) fs: ext4 ID: /home size: 20G used: 12G (61%) fs: ext4
ID: swap-1 size: 1.08GB used: 0.00GB (0%) fs: swap
Sensors: System Temperatures: cpu: 54.0C mobo: 45.0C gpu: 60C
Fan Speeds (in rpm): cpu: 2657 fan-1: 2454 fan-3: 0 fan-4: 0
Info: Processes: 144 Uptime: 1:48 Memory: 931.5/5973.0MB Client: Shell (zsh) inxi: 1.9.18

from lain.

i have no clue why there is so much garbage in the paste's meh

from lain.

You flagged as solved, but didn't explain how you solved.

from lain.

I changed the path that it is parsing:

function update()
--[[
local f = io.open("/sys/class/thermal/thermal_zone0/temp")
--]]
local f = io.open("/sys/class/hwmon/hwmon0/device/temp1_input")
if f ~= nil
then

it's good enough for me even though it is not an average

I would say this is a bug however because you cannot be sure that the original path is going to be present on every machine that uses this.
I installed acpi, mode probed the modules I wanted & blacklisted the ones I did not, rebooted & yet that path was still not present.
My suggestion would be to look for a more predictable & reliable sensor output if there is one!) to ensure all machines are the same.

from lain.

acpi is not required, this is a machine related filesystem issue.

I'll wrote a commit looking for any sensor output, if any, instead of hardcoding to a path which I wrongly thought was universal.

from lain.

cheers luke :)

from lain.

Since I see this can change from machine to machine, regardless of linux filesystem, I chose the simple way: there's a new variable called tempfile, which stores the path of the coretemp file. It can be redefined if it doesn't match your real file path.

In your case, you could use 3 temp widgets, instead of 1 with the average, or you can have 1 widget, but defining in tempfile the path of a file where you periodically write with a script the average value of temp*_input files.

from lain.