dr-noob / gpufetch Goto Github PK

View Code? Open in Web Editor NEW

130.0 130.0 14.0 2.06 MB

Simple yet fancy GPU architecture fetching tool

License: GNU General Public License v2.0

C++ 94.47% Roff 1.06% CMake 3.56% Shell 0.91%

cuda gpu igpu intel nvidia

gpufetch's Introduction

GitHub Stats

Repositories

gpufetch's People

Contributors

Stargazers

Watchers

Forkers

pronoy2108 vic4key kz6fittycent 1div0 trevor-vincent codepurble grahamperrin clayne rongou oliveiradan darkbuck mrmelon54 bkzhn

gpufetch's Issues

[ERROR]: Unkown Intel device id: 0x9A49

Hi!
So basically I built gpufetch in source and once I did ./gpufetch; It immediately got an error which is this: https://i.imgur.com/vGWW55q.png

Even though I set my NVIDIA graphics card as my main GPU, not my integrated graphics card.

pci.cpp spelling correction: Unkown, Unknown

#21 intended to fix a typo that originated in two places:

a397eb3#diff-b52e39c94a6052ba322bd29245cba0cf7fbe55ae881900245fc886eaba0f5f5fR11

a397eb3#diff-8d90022bc2916457a50021b9cc669857bea6f47dbc917cf0b1cc8bacaef951a1R11

Add Raspberry Pi support

Since the GPU of the Raspberry Pi is VideoCore instead of NVIDIA, CUDA cannot be installed.

Add support for generic UHD/HD graphics

Intel iGPUs with generic "UHD Graphics" or "HD Graphics" name are not fully supported by gpufetch

Do not print "No GPU was detected" if CUDA backend is disabled

Motivation: #51

Weird cuda error prevents gpufetch from working

[gaming123@gamingmachine ~]$ gpufetch
[ERROR]: cudaErrorUnknown: unknown error
[ERROR]: No GPU was detected! Available GPUs are:
- GPU 0:
  * Vendor: NVIDIA
  * PCI id: 10de:2520
- GPU 1:
  * Vendor: AMD
  * PCI id: 1002:1638
Please, make sure that the appropiate backend is enabled:
- CUDA backend:  ON
- Intel backend: ON

Unknown Intel device id: 0xA788 on System76 Adder WS

Hi!

Got this error message on a System76 Adder WS.

Related issue: #45

Extract from lshw:

*-display
        description: VGA compatible controller
        product: Intel Corporation
        vendor: Intel Corporation
        physical id: 2
        bus info: pci@0000:00:02.0
        logical name: /dev/fb0
        version: 04
        width: 64 bits
        clock: 33MHz
        capabilities: pciexpress msi pm vga_controller bus_master cap_list rom fb
        configuration: depth=32 driver=i915 latency=0 resolution=1920,1080
        resources: irq:200 memory:83000000-83ffffff memory:90000000-9fffffff ioport:1000(size=64) memory:c0000-dffff memory:11a0000000-11a6ffffff memory:10c0000000-119fffffff

Full cpufetch output:

[ERROR]: Unknown CUDA device id: 0x2820
Please, create a new issue with this error message on https://github.com/Dr-Noob/gpufetch/issues
[ERROR]: Unknown Intel device id: 0xA788
Please, create a new issue with this error message on https://github.com/Dr-Noob/gpufetch/issues

               'cccccccccccccccccccccccccc   
               ;oooooooooooooooooooooooool   
           .:::.     .oooooooooooooooooool   Name:                   NVIDIA GeForce RTX 4070 Laptop GPU
      .:cll;   ,c:::.     cooooooooooooool   GPU processor:          Unknown
   ,clo'      ;.   oolc:     ooooooooooool   Microarchitecture:      Unknown (8.9)
.cloo    ;cclo .      .olc.    coooooooool   Technology:             Unknown
oooo   :lo,    ;ll;    looc    :oooooooool   Max Frequency:          2175 MHz
 oooc   ool.   ;oooc;clol    :looooooooool   SMs:                    36
  :ooc   ,ol;  ;oooooo.   .cloo;     loool   Cores/SM:               128
    ool;   .olc.       ,:lool        .lool   CUDA Cores:             4608
      ool:.    ,::::ccloo.        :clooool   Tensor Cores:           144
         oolc::.            ':cclooooooool   Memory:                 8 GiB Unknown
               ;oooooooooooooooooooooooool   Memory frequency:       8001 MHz
                                             Bus width:              128 bit
                                             L2 Size:                32 MiB
######.  ##   ##  ##  ######   ##    ###     Peak Performance:       20.04 TFLOP/s
##   ##  ##   ##  ##  ##   ##  ##   #: :#    Peak Performance (MMA): 160.36 TFLOP/s
##   ##   ## ##   ##  ##   ##  ##  #######   
##   ##    ###    ##  ######   ## ##     ##

Please let me know if additional info is needed.

Initially

nproc not found.txt

% which nproc
nproc: Command not found.
% file /compat/linux/usr/bin/nproc
/compat/linux/usr/bin/nproc: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked, interpreter /lib64/ld-linux-x86-64.so.2, for GNU/Linux 2.6.32, BuildID[sha1]=16cdfd7d28a3b43b37e5bfb9662687502f710ad4, stripped
% echo $PATH
/sbin:/bin:/usr/sbin:/usr/bin:/usr/local/sbin:/usr/local/bin:/home/grahamperrin/bin:/home/grahamperrin/.local/bin
%

Extraordinary, I never had to add /compat/linux/usr/bin to my $PATH before now.

With nproc found

% pwd
/usr/home/grahamperrin/dev/gpufetch
% which nproc
/compat/linux/usr/bin/nproc
% sh ./build.sh 
-- The CXX compiler identification is Clang 13.0.0
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Check for working CXX compiler: /usr/bin/c++ - skipped
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- Looking for a CUDA compiler
-- Looking for a CUDA compiler - NOTFOUND
-- Found PCIUTILS: /usr/local/lib/libpci.so  
-- ----------------------
-- gpufetch build report:
-- CUDA backend: OFF
-- Intel backend: ON
-- ----------------------
-- Configuring done
-- Generating done
-- Build files have been written to: /home/grahamperrin/dev/gpufetch/build
[ 12%] Building CXX object CMakeFiles/intel_backend.dir/src/intel/uarch.cpp.o
[ 12%] Building CXX object CMakeFiles/intel_backend.dir/src/intel/udev.cpp.o
[ 25%] Building CXX object CMakeFiles/intel_backend.dir/src/intel/intel.cpp.o
[ 25%] Building CXX object CMakeFiles/intel_backend.dir/src/intel/pci.cpp.o
/home/grahamperrin/dev/gpufetch/src/intel/uarch.cpp:54:27: warning: suspicious concatenation of string literals in an array initialization; did you mean to separate the elements with a comma? [-Wstring-concatenation]
  /*[ARCH_GEN12_TGL] = */ "Gen12"
                          ^
/home/grahamperrin/dev/gpufetch/src/intel/uarch.cpp:53:27: note: place parentheses around the string literal to silence warning
  /*[ARCH_GEN12_RKL] = */ "Gen12"
                          ^
[ 31%] Building CXX object CMakeFiles/intel_backend.dir/src/intel/cpuid.cpp.o
1 warning generated.
[ 37%] Linking CXX static library libintel_backend.a
[ 37%] Built target intel_backend
[ 56%] Building CXX object CMakeFiles/gpufetch.dir/src/common/args.cpp.o
[ 62%] Building CXX object CMakeFiles/gpufetch.dir/src/common/gpu.cpp.o
[ 62%] Building CXX object CMakeFiles/gpufetch.dir/src/common/pci.cpp.o
[ 62%] Building CXX object CMakeFiles/gpufetch.dir/src/common/main.cpp.o
In file included from /home/grahamperrin/dev/gpufetch/src/common/main.cpp:5:
/home/grahamperrin/dev/gpufetch/src/common/args.hpp:5:3: error: unknown type name 'int32_t'; did you mean 'uint32_t'?
  int32_t R;
  ^~~~~~~
  uint32_t
/usr/include/sys/endian.h:49:20: note: 'uint32_t' declared here
typedef __uint32_t      uint32_t;
                        ^
In file included from /home/grahamperrin/dev/gpufetch/src/common/main.cpp:5:
/home/grahamperrin/dev/gpufetch/src/common/args.hpp:6:3: error: unknown type name 'int32_t'; did you mean 'uint32_t'?
  int32_t G;
  ^~~~~~~
  uint32_t
/usr/include/sys/endian.h:49:20: note: 'uint32_t' declared here
typedef __uint32_t      uint32_t;
                        ^
In file included from /home/grahamperrin/dev/gpufetch/src/common/main.cpp:5:
/home/grahamperrin/dev/gpufetch/src/common/args.hpp:7:3: error: unknown type name 'int32_t'; did you mean 'uint32_t'?
  int32_t B;
  ^~~~~~~
  uint32_t
/usr/include/sys/endian.h:49:20: note: 'uint32_t' declared here
typedef __uint32_t      uint32_t;
                        ^
In file included from /home/grahamperrin/dev/gpufetch/src/common/args.cpp:7:
/home/grahamperrin/dev/gpufetch/src/common/args.hpp:5:3: error: unknown type name 'int32_t'; did you mean 'uint32_t'?
  int32_t R;
  ^~~~~~~
  uint32_t
/usr/include/sys/endian.h:49:20: note: 'uint32_t' declared here
typedef __uint32_t      uint32_t;
                        ^
In file included from /home/grahamperrin/dev/gpufetch/src/common/args.cpp:7:
/home/grahamperrin/dev/gpufetch/src/common/args.hpp:6:3: error: unknown type name 'int32_t'; did you mean 'uint32_t'?
  int32_t G;
  ^~~~~~~
  uint32_t
/usr/include/sys/endian.h:49:20: note: 'uint32_t' declared here
typedef __uint32_t      uint32_t;
                        ^
In file included from /home/grahamperrin/dev/gpufetch/src/common/args.cpp:7:
/home/grahamperrin/dev/gpufetch/src/common/args.hpp:7:3: error: unknown type name 'int32_t'; did you mean 'uint32_t'?
  int32_t B;
  ^~~~~~~
  uint32_t
/usr/include/sys/endian.h:49:20: note: 'uint32_t' declared here
typedef __uint32_t      uint32_t;
                        ^
3 errors generated.
--- CMakeFiles/gpufetch.dir/src/common/args.cpp.o ---
*** [CMakeFiles/gpufetch.dir/src/common/args.cpp.o] Error code 1

make[2]: stopped in /usr/home/grahamperrin/dev/gpufetch/build
3 errors generated.
--- CMakeFiles/gpufetch.dir/src/common/main.cpp.o ---
*** [CMakeFiles/gpufetch.dir/src/common/main.cpp.o] Error code 1

make[2]: stopped in /usr/home/grahamperrin/dev/gpufetch/build
2 errors

make[2]: stopped in /usr/home/grahamperrin/dev/gpufetch/build
--- CMakeFiles/gpufetch.dir/all ---
*** [CMakeFiles/gpufetch.dir/all] Error code 2

make[1]: stopped in /usr/home/grahamperrin/dev/gpufetch/build
1 error

make[1]: stopped in /usr/home/grahamperrin/dev/gpufetch/build
*** [all] Error code 2

make: stopped in /usr/home/grahamperrin/dev/gpufetch/build
1 error

make: stopped in /usr/home/grahamperrin/dev/gpufetch/build
%

Remove CUDA dependency

Is it possible to make the NVIDIA backend work without CUDA with reasonable effort/added complexity?

Maybe this will never be implemented but can be a good enhancement for future work.

Add --logo-short and --logo-long

Implement two options:

--logo-short      Show the short version of the logo
--logo-long       Show the long version of the logo

A few nitpicks (very minor changes)

These are a few tiny changes that I am "reporting" here because of the no-PR policy. Takes seconds to implement.

I found out that the CUDA_DRIVER_START_WARNING message does not end with a newline, causing any subsequent error messages to start on the same line as the warning. Adding a newline would just even things out.

Also, adding build/ to the gitignore is another thing you can consider.

Thanks and regards

`build.sh` broken

On ubuntu 22.04, running ./build.sh produces some errors:

./build.sh: 10: [: unexpected operator
...
./build.sh: 32: [: Linux==: unexpected operator
./build.sh: 34: [: Linux: unexpected operator

Looks like the sh version only supports single equal signs.

Unknown Intel device id

[ERROR]: Unkown Intel device id: 0x9BC4

inxi -G reports it as Intel CometLake-H GT2 [UHD Graphics]

hwinfo --gfxcard reports it as the following:

30: PCI 02.0: 0300 VGA compatible controller (VGA)
  [Created at pci.386]
  Unique ID: _Znp.FfPWVlpwdi4
  SysFS ID: /devices/pci0000:00/0000:00:02.0
  SysFS BusID: 0000:00:02.0
  Hardware Class: graphics card
  Model: "Intel CometLake-H GT2 [UHD Graphics]"
  Vendor: pci 0x8086 "Intel Corporation"
  Device: pci 0x9bc4 "CometLake-H GT2 [UHD Graphics]"
  SubVendor: pci 0x1025 "Acer Incorporated [ALI]"
  SubDevice: pci 0x147d 
  Revision: 0x05
  Driver: "i915"
  Driver Modules: "i915"
  Memory Range: 0xc3000000-0xc3ffffff (rw,non-prefetchable)
  Memory Range: 0xa0000000-0xafffffff (ro,non-prefetchable)
  I/O Ports: 0x5000-0x503f (rw)
  Memory Range: 0x000c0000-0x000dffff (rw,non-prefetchable,disabled)
  IRQ: 168 (93118 events)
  Module Alias: "pci:v00008086d00009BC4sv00001025sd0000147Dbc03sc00i00"
  Driver Info #0:
    Driver Status: i915 is active
    Driver Activation Cmd: "modprobe i915"
  Config Status: cfg=new, avail=yes, need=no, active=unknown

The secondary NVIDIA GPU identifies properly. Thanks for cpufetch & gpufetch!

[Feature] Support Windows x86 & x64

Hello,
@Dr-Noob , you might know me already. I was so kind and built the source code under Windows and provide here the 64Bit and 32Bit variant.

Of course I would like to share my experiences with you. CUDA & NVML can only be compiled under Windows with Microsoft's BuildTools (aka MSVC); MinGW doesn't seem to support Nvidia's libraries … Therefore I have rebuilt the source code a bit and compiled it so far and tested it myself (see photo).

For further questions I am at your disposal.

HowTo use `gpufetch` under Windows

64 bit:
Unpack the 64 bit archive and execute gpufetch. Be aware that the DLL is required!
32 bit:
Currently, Nvidia's libraries do not support 32 bit Windows architecture.

A big thank you to @Dr-Noob for developing such great tools & improving them more and more

gpufetch_x64.zip

Support for AMD GPUs (hardware needed!)

AMD GPUs support could be added to gpufetch, but I do not have access to any relatively new AMD GPU, so it's impossible for me to develop the port.

If anybody is interested, I need ssh access to a machine with an AMD GPU and ROCm installed.

[ERROR]: Unknown CUDA device id: 0x2487

I am getting this error when running gpufetch (Btw the error says "Unkown", so there is a typo)

I am using a NVIDIA RTX 3060 (01:00.0 VGA compatible controller [0300]: NVIDIA Corporation GA104 [GeForce RTX 3060] [10de:2487] (rev a1))

Show new Intel logo in new iGPUs

Like cpufetch does, and also add relevant flags like cpufetch has

The backend is enabled, but gpufetch is unable to detect my GPU

Installed using ./build.sh. It failed as it couldnt find a GPU, while it is enabled:

:~/Downloads/gpufetch$ ./gpufetch 
[ERROR]: cudaErrorSystemDriverMismatch: system has unsupported display driver / cuda driver combination
[ERROR]: No GPU was detected! Available GPUs are:
- GPU 0:
  * Vendor: NVIDIA
  * PCI id: 10de:21c4
Please, make sure that the appropiate backend is enabled:
- CUDA backend:  ON
- Intel backend: ON
Visit https://github.com/Dr-Noob/gpufetch#2-backends for more information
:~/Downloads/gpufetch$ lspci -nn | grep VGA
01:00.0 VGA compatible controller [0300]: NVIDIA Corporation TU116 [GeForce GTX 1660 SUPER] [10de:21c4] (rev a1)

[ERROR]: Unknown CUDA device id: 0x28A0

$ gpufetch
[ERROR]: Unknown CUDA device id: 0x28A0
Please, create a new issue with this error message on https://github.com/Dr-Noob/gpufetch/issues

                  MMMMMMMMMMMMMMMMMMMMMMMMMMMMMM  Name:                   NVIDIA GeForce RTX 4060 Laptop GPU
                  MMMMMMMMMMMMMMMMMMMMMMMMMMMMMM  GPU processor:          Unknown
                .::   'MMMMMMMMMMMMMMMMMMMMMMMMM  Microarchitecture:      Unknown (8.9)
           ccllooo;:;.       ;MMMMMMMMMMMMMMMMMM  Technology:             Unknown
       cloc       :ooollcc:     :MMMMMMMMMMMMMMM  Max Frequency:          2370 MHz
    cloc      :ccl;      lolc,     ;MMMMMMMMMMMM  SMs:                    24
.cloo:    :clo    ;c:      .ool;     MMMMMMMMMMM  Cores/SM:               128
  ooo:    ooo     :ool,  .cloo.    ;lMMMMMMMMMMM  CUDA Cores:             3072
   ooo:    ooc    :ooooccooo.    :MMMM  lMMMMMMM  Tensor Cores:           96
     ooc.   ool:  :oooooo'    ,cloo.        MMMM  Memory:                 8 GiB Unknown
      ool:.    olc:       .:cloo.          :MMMM  Memory frequency:       8001 MHz
         olc,     ;:::cccloo.          :MMMMMMMM  Bus width:              128 bit
            olcc::;              ,:ccloMMMMMMMMM  L2 Size:                32 MiB
                  :......oMMMMMMMMMMMMMMMMMMMMMM  Peak Performance:       14.56 TFLOP/s
                  :lllMMMMMMMMMMMMMMMMMMMMMMMMMM  Peak Performance (MMA): 116.49 TFLOP/s

Unknown Intel device id: 0x2A42

grahamperrin@fuji:~/dev/gpufetch % ./gpufetch
[ERROR]: Unkown Intel device id: 0x2A42
Please, create a new issue with this error message on https://github.com/Dr-Noob/gpufetch/issues
[ERROR]: No GPU was detected! Available GPUs are:
- GPU 0:
  * Vendor: Intel
  * PCI id: 8086:2a42
Please, make sure that the appropiate backend is enabled:
- CUDA backend:  OFF
- Intel backend: ON
Visit https://github.com/Dr-Noob/gpufetch#2-backends for more information
grahamperrin@fuji:~/dev/gpufetch % freebsd-version -kru
13.1-RELEASE
13.1-RELEASE
13.1-RELEASE
grahamperrin@fuji:~/dev/gpufetch % uname -aKU
FreeBSD fuji 13.1-RELEASE FreeBSD 13.1-RELEASE releng/13.1-n250148-fc952ac2212 GENERIC amd64 1301000 1301000
grahamperrin@fuji:~/dev/gpufetch %

Unknown Intel device id: 0x22B0

[ERROR]: Unknown Intel device id: 0x22B0
Please, create a new issue with this error message on https://github.com/Dr-Noob/gpufetch/issues
[ERROR]: No GPU was detected! Available GPUs are:

GPU 0:
- Vendor: Intel
- PCI id: 8086:22b0
  Please, make sure that the appropiate backend is enabled:
CUDA backend: OFF
Intel backend: ON
Visit https://github.com/Dr-Noob/gpufetch#2-backends for more information

My GPU is based on Intel® HD Graphics -Gen8 - Cherryview on my Atomic PI

MapSMtoCores for SM 8.9 is undefined

Running with 4090 FE:

$ gpufetch
MapSMtoCores for SM 8.9 is undefined.  Default to use 128 Cores/SM
[NOTE]: gpufetch is in beta. The provided information may be incomplete or wrong.
If you want to help to improve gpufetch, please compare the output of the program
with a reliable source which you know is right (e.g, techpowerup.com) and report
any inconsistencies to https://github.com/Dr-Noob/gpufetch/issues
                                   
                  MMMMMMMMMMMMMMMMMMMMMMMMMMMMMM  Name:                   NVIDIA Graphics Device
                  MMMMMMMMMMMMMMMMMMMMMMMMMMMMMM  GPU processor:          AD102
                .::   'MMMMMMMMMMMMMMMMMMMMMMMMM  Microarchitecture:      Ada Lovelace (8.9)
           ccllooo;:;.       ;MMMMMMMMMMMMMMMMMM  Technology:             4nm
       cloc       :ooollcc:     :MMMMMMMMMMMMMMM  Max Frequency:          2520 MHz
    cloc      :ccl;      lolc,     ;MMMMMMMMMMMM  SMs:                    128
.cloo:    :clo    ;c:      .ool;     MMMMMMMMMMM  Cores/SM:               128
  ooo:    ooo     :ool,  .cloo.    ;lMMMMMMMMMMM  CUDA Cores:             16384
   ooo:    ooc    :ooooccooo.    :MMMM  lMMMMMMM  Tensor Cores:           512
     ooc.   ool:  :oooooo'    ,cloo.        MMMM  Memory:                 24 GiB GDDR6X
      ool:.    olc:       .:cloo.          :MMMM  Memory frequency:       1312 MHz
         olc,     ;:::cccloo.          :MMMMMMMM  Bus width:              384 bit
            olcc::;              ,:ccloMMMMMMMMM  L2 Size:                72 MiB
                  :......oMMMMMMMMMMMMMMMMMMMMMM  Peak Performance:       82.58 TFLOP/s
                  :lllMMMMMMMMMMMMMMMMMMMMMMMMMM  Peak Performance (MMA): 660.60 TFLOP/s

Unknown CUDA device id: 0x28E0

[ERROR]: Unknown CUDA device id: 0x28E0
Please, create a new issue with this error message on https://github.com/Dr-Noob/gpufetch/issues

                  MMMMMMMMMMMMMMMMMMMMMMMMMMMMMM  Name:                   NVIDIA GeForce RTX 4060 Laptop GPU
                  MMMMMMMMMMMMMMMMMMMMMMMMMMMMMM  GPU processor:          Unknown
                .::   'MMMMMMMMMMMMMMMMMMMMMMMMM  Microarchitecture:      Unknown (8.9)
           ccllooo;:;.       ;MMMMMMMMMMMMMMMMMM  Technology:             Unknown
       cloc       :ooollcc:     :MMMMMMMMMMMMMMM  Max Frequency:          2010 MHz
    cloc      :ccl;      lolc,     ;MMMMMMMMMMMM  SMs:                    24
.cloo:    :clo    ;c:      .ool;     MMMMMMMMMMM  Cores/SM:               128
  ooo:    ooo     :ool,  .cloo.    ;lMMMMMMMMMMM  CUDA Cores:             3072
   ooo:    ooc    :ooooccooo.    :MMMM  lMMMMMMM  Tensor Cores:           96
     ooc.   ool:  :oooooo'    ,cloo.        MMMM  Memory:                 8 GiB Unknown
      ool:.    olc:       .:cloo.          :MMMM  Memory frequency:       8001 MHz
         olc,     ;:::cccloo.          :MMMMMMMM  Bus width:              128 bit
            olcc::;              ,:ccloMMMMMMMMM  L2 Size:                32 MiB
                  :......oMMMMMMMMMMMMMMMMMMMMMM  Peak Performance:       12.35 TFLOP/s
                  :lllMMMMMMMMMMMMMMMMMMMMMMMMMM  Peak Performance (MMA): 98.80 TFLOP/s

Unable to find memory type in some Ampere GPUs

As shown in #28 and #29

[ERROR]: Unknown Intel device id: 0x9A68

I've got an error while running program:
[ERROR]: Unknown Intel device id: 0x9A68
My processor is NVIDIA GeForce RTX 3050 Ti Mobile

Output to png/jpg

Thank you for your awesome project Dr-Noob! It would be nice to have a png or jpg image we can share with people. I made a similar request on cpufetch.

The package from aur is not being built (Arch Linux)

(I'm writing because your Upstream URL is listed here: https://aur.archlinux.org/packages/gpufetch-git/)

Unknown Intel device id: 0x4626

Executing gpu-fetch (from the AUR) on my laptop with intel and nvidia gpu results in this

[ERROR]: Unknown Intel device id: 0x4626
Please, create a new issue with this error message on https://github.com/Dr-Noob/gpufetch/issues

                  MMMMMMMMMMMMMMMMMMMMMMMMMMMMMM  Name:                   NVIDIA GeForce RTX 3060 Laptop GPU
                  MMMMMMMMMMMMMMMMMMMMMMMMMMMMMM  GPU processor:          GA106
                .::   'MMMMMMMMMMMMMMMMMMMMMMMMM  Microarchitecture:      Ampere (8.6)
           ccllooo;:;.       ;MMMMMMMMMMMMMMMMMM  Technology:             8nm
       cloc       :ooollcc:     :MMMMMMMMMMMMMMM  Max Frequency:          1425 MHz
    cloc      :ccl;      lolc,     ;MMMMMMMMMMMM  SMs:                    30
.cloo:    :clo    ;c:      .ool;     MMMMMMMMMMM  Cores/SM:               128
  ooo:    ooo     :ool,  .cloo.    ;lMMMMMMMMMMM  CUDA Cores:             3840
   ooo:    ooc    :ooooccooo.    :MMMM  lMMMMMMM  Tensor Cores:           120
     ooc.   ool:  :oooooo'    ,cloo.        MMMM  Memory:                 6 GiB GDDR6
      ool:.    olc:       .:cloo.          :MMMM  Memory frequency:       1750 MHz
         olc,     ;:::cccloo.          :MMMMMMMM  Bus width:              192 bit
            olcc::;              ,:ccloMMMMMMMMM  L2 Size:                3 MiB
                  :......oMMMMMMMMMMMMMMMMMMMMMM  Peak Performance:       10.94 TFLOP/s
                  :lllMMMMMMMMMMMMMMMMMMMMMMMMMM  Peak Performance (MMA): 87.55 TFLOP/s

The CPU is a 12th gen intel core i7 12700H with an integrated gpu listed as "Intel Alder Lake-P Integrated Graphics Controller" from hardinfo and as "Intel® Iris® Xe Graphics eligible" from the official intel specific processor page.

The nvidia gpu is correctly identified

Hope this helps

[ERROR]: Unkown CUDA device id: 0x24A0, [ERROR]: Unkown Intel device id: 0x46A6

wang@pop-os:~/Documents/git/gpufetch$ ./gpufetch 
[ERROR]: Unkown CUDA device id: 0x24A0
Please, create a new issue with this error message on https://github.com/Dr-Noob/gpufetch/issues
[ERROR]: Unkown Intel device id: 0x46A6
Please, create a new issue with this error message on https://github.com/Dr-Noob/gpufetch/issues
[NOTE]: gpufetch is in beta. The provided information may be incomplete or wrong.
If you want to help to improve gpufetch, please compare the output of the program
with a reliable source which you know is right (e.g, techpowerup.com) and report
any inconsistencies to https://github.com/Dr-Noob/gpufetch/issues

                  MMMMMMMMMMMMMMMMMMMMMMMMMMMMMM  Name:                   NVIDIA GeForce RTX 3070 Ti Laptop GPU
                  MMMMMMMMMMMMMMMMMMMMMMMMMMMMMM  GPU processor:          Unknown
                .::   'MMMMMMMMMMMMMMMMMMMMMMMMM  Microarchitecture:      Unknown (8.6)
           ccllooo;:;.       ;MMMMMMMMMMMMMMMMMM  Technology:             Unknown
       cloc       :ooollcc:     :MMMMMMMMMMMMMMM  Max Frequency:          1485 MHz
    cloc      :ccl;      lolc,     ;MMMMMMMMMMMM  SMs:                    46
.cloo:    :clo    ;c:      .ool;     MMMMMMMMMMM  Cores/SM:               128
  ooo:    ooo     :ool,  .cloo.    ;lMMMMMMMMMMM  CUDA Cores:             5888
   ooo:    ooc    :ooooccooo.    :MMMM  lMMMMMMM  Tensor Cores:           184
     ooc.   ool:  :oooooo'    ,cloo.        MMMM  Memory:                 8 GiB Unknown
      ool:.    olc:       .:cloo.          :MMMM  Memory frequency:       7001 MHz
         olc,     ;:::cccloo.          :MMMMMMMM  Bus width:              256 bit
            olcc::;              ,:ccloMMMMMMMMM  L2 Size:                4 MiB
                  :......oMMMMMMMMMMMMMMMMMMMMMM  Peak Performance:       17.49 TFLOP/s
                  :lllMMMMMMMMMMMMMMMMMMMMMMMMMM  Peak Performance (MMA): 139.90 TFLOP/s

Seems to get most of the info. Note, I'm running on NVIDIA discrete graphics mode on this laptop, so I don't really know if that affects the visibility of the Intel iGPU.

Remove NVML dependency to allow static compilation

Matrox is NOT supported

support Matrox graphics!!!

Look, it may be really old, and they may be unusable... but it would be GREAT!

Add video encode/decode capabilities

Such as H265, VP9, etc

[ERROR] Unknown Intel device id: 0x9BC4

Host machine is an Acer Nitro 5 AN515-55 with a Comet Lake I5-10300H with Comet Lake GT2 UHD Graphics

Error: cudaErrorCompatNotSupportedOnDevice

I am trying to run gpufetch but get the error below.

OS: Kubuntu 22.04 (Linux)
GPU: NVIDIA Geforce GTX 970

The GPU does show up when running lspci -nn | grep VGA

Error:

[ERROR]: cudaErrorCompatNotSupportedOnDevice: forward compatibility was attempted on non supported HW
[ERROR]: No GPU was detected! Available GPUs are:
- GPU 0:
  * Vendor: NVIDIA
  * PCI id: 10de:13c2
Please, make sure that the appropiate backend is enabled:
- CUDA backend:  ON
- Intel backend: ON
Visit https://github.com/Dr-Noob/gpufetch#2-backends for more information

Error building gpufetch (undefined references) on Ubuntu 21.10

I was trying to build gpufetch on my machine, but keep on getting undefined references errors. I am running Ubuntu 21.10 and using an NVIDIA card. I am, unfortunately, not really an expert in this, but the output of running ./build.sh can be found below.

-- The CXX compiler identification is GNU 11.2.0
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Check for working CXX compiler: /usr/bin/c++ - skipped
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- The CUDA compiler identification is NVIDIA 11.5.119
-- Detecting CUDA compiler ABI info
-- Detecting CUDA compiler ABI info - done
-- Check for working CUDA compiler: /usr/local/cuda/bin/nvcc - skipped
-- Detecting CUDA compile features
-- Detecting CUDA compile features - done
-- Could NOT find PCIUTILS (missing: PCIUTILS_LIBRARIES PCIUTILS_INCLUDE_DIR) 
-- pciutils not found, downloading and building a local copy...
-- ----------------------
-- gpufetch build report:
-- CUDA backend: ON
-- Intel backend: ON
-- ----------------------
-- Configuring done
-- Generating done
-- Build files have been written to: /home/mjo/gpufetch/build
-- Could NOT find PCIUTILS (missing: PCIUTILS_LIBRARIES PCIUTILS_INCLUDE_DIR) 
-- pciutils not found, downloading and building a local copy...
-- ----------------------
-- gpufetch build report:
-- CUDA backend: ON
-- Intel backend: ON
-- ----------------------
-- Configuring done
-- Generating done
-- Build files have been written to: /home/mjo/gpufetch/build
Scanning dependencies of target pciutils
[  3%] Creating directories for 'pciutils'
[  7%] Performing download step (git clone) for 'pciutils'
Cloning into 'pciutils'...
Already on 'master'
Your branch is up to date with 'origin/master'.
[ 11%] Performing update step for 'pciutils'
[ 14%] No patch step for 'pciutils'
[ 18%] No configure step for 'pciutils'
[ 22%] Performing build step for 'pciutils'
make[3]: warning: jobserver unavailable: using -j1.  Add '+' to parent make rule.
Configuring libpci for your system... x86_64--linux 5.13.0-23-generic x86_64 linux
Looking for access methods... sysfs proc i386-ports dump
Checking for zlib support... yes (auto-detected)
Checking for DNS support... yes (auto-detected)
Checking for libkmod... no (auto-detected)
Checking for udev hwdb support... yes (auto-detected)
Checking whether to build a shared library... no (set manually)
make[3]: warning: jobserver unavailable: using -j1.  Add '+' to parent make rule.
[ 25%] Performing install step for 'pciutils'
make[3]: warning: jobserver unavailable: using -j1.  Add '+' to parent make rule.
[ 29%] Completed 'pciutils'
[ 29%] Built target pciutils
Scanning dependencies of target cuda_backend
Scanning dependencies of target intel_backend
[ 40%] Building CXX object CMakeFiles/intel_backend.dir/src/intel/uarch.cpp.o
[ 40%] Building CXX object CMakeFiles/intel_backend.dir/src/intel/intel.cpp.o
[ 44%] Building CXX object CMakeFiles/intel_backend.dir/src/intel/pci.cpp.o
[ 44%] Building CXX object CMakeFiles/intel_backend.dir/src/intel/udev.cpp.o
[ 48%] Building CXX object CMakeFiles/intel_backend.dir/src/intel/cpuid.cpp.o
[ 51%] Building CXX object CMakeFiles/cuda_backend.dir/src/cuda/cuda.cpp.o
[ 55%] Building CXX object CMakeFiles/cuda_backend.dir/src/cuda/pci.cpp.o
[ 59%] Building CXX object CMakeFiles/cuda_backend.dir/src/cuda/uarch.cpp.o
[ 62%] Linking CXX static library libintel_backend.a
[ 62%] Built target intel_backend
[ 66%] Linking CXX static library libcuda_backend.a
[ 66%] Built target cuda_backend
Scanning dependencies of target gpufetch
[ 77%] Building CXX object CMakeFiles/gpufetch.dir/src/common/gpu.cpp.o
[ 77%] Building CXX object CMakeFiles/gpufetch.dir/src/common/pci.cpp.o
[ 77%] Building CXX object CMakeFiles/gpufetch.dir/src/common/printer.cpp.o
[ 81%] Building CXX object CMakeFiles/gpufetch.dir/src/common/args.cpp.o
[ 85%] Building CXX object CMakeFiles/gpufetch.dir/src/common/global.cpp.o
[ 88%] Building CXX object CMakeFiles/gpufetch.dir/src/common/uarch.cpp.o
[ 92%] Building CXX object CMakeFiles/gpufetch.dir/src/common/main.cpp.o
[ 96%] Building CXX object CMakeFiles/gpufetch.dir/src/common/master.cpp.o
[100%] Linking CXX executable gpufetch
/usr/bin/ld: /home/mjo/gpufetch/build/pciutils-install/lib/libpci.a(names-hwdb.o): in function `pci_id_hwdb_lookup':
names-hwdb.c:(.text+0xe4): undefined reference to `udev_hwdb_get_properties_list_entry'
/usr/bin/ld: names-hwdb.c:(.text+0xfc): undefined reference to `udev_list_entry_get_next'
/usr/bin/ld: names-hwdb.c:(.text+0x110): undefined reference to `udev_list_entry_get_name'
/usr/bin/ld: names-hwdb.c:(.text+0x12c): undefined reference to `udev_list_entry_get_value'
/usr/bin/ld: names-hwdb.c:(.text+0x19f): undefined reference to `udev_new'
/usr/bin/ld: names-hwdb.c:(.text+0x1ac): undefined reference to `udev_hwdb_new'
/usr/bin/ld: /home/mjo/gpufetch/build/pciutils-install/lib/libpci.a(names-hwdb.o): in function `pci_id_hwdb_free':
names-hwdb.c:(.text+0x2d2): undefined reference to `udev_hwdb_unref'
/usr/bin/ld: names-hwdb.c:(.text+0x2e8): undefined reference to `udev_unref'
collect2: error: ld returned 1 exit status
make[2]: *** [CMakeFiles/gpufetch.dir/build.make:210: gpufetch] Error 1
make[1]: *** [CMakeFiles/Makefile2:156: CMakeFiles/gpufetch.dir/all] Error 2
make: *** [Makefile:149: all] Error 2

gpufetch does not detect the GPU

If there is a NVIDIA GPU or Intel iGPU in the system and the appropiate backend is enabled but gpufetch does not detect the GPU, please create a new issue with the provided error message (in the gpufetch output) on the issues page.

$ gpufetch
[ERROR]: cudaErrorInsufficientDriver: CUDA driver version is insufficient for CUDA runtime version
[ERROR]: No GPU was detected! Available GPUs are:
- GPU 0:
  * Vendor: NVIDIA
  * PCI id: 10de:0640
Please, make sure that the appropiate backend is enabled:
- CUDA backend:  ON
- Intel backend: ON
Visit https://github.com/Dr-Noob/gpufetch#2-backends for more information
$ lspci -nn | grep VGA
01:00.0 VGA compatible controller [0300]: NVIDIA Corporation G96C [GeForce 9500 GT] [10de:0640] (rev a1)

Wrong RAM size detect.

Hi!

I have an MSI GeForce GTX 1650 Super card with 4GB RAM. However, the program only says 3GB, which is wrong.

Gpufetch

nvidia-smi

The program is the latest git version (gpufetch v0.21).

TU116 still shows Tensor Cores

This is in relation to d3aaf7c.

My Nvidia Geforce GTX 1660 Ti still seems to show that I have 192 Tensor Cores despite being a TU116 chip. This PR can sustain the dialog of making sure it handles this particular card since I personally have one as in one of my headless linux machines.

% gpufetch -V
gpufetch v0.22
% gpufetch
[NOTE]: gpufetch is in beta. The provided information may be incomplete or wrong.
If you want to help to improve gpufetch, please compare the output of the program
with a reliable source which you know is right (e.g, techpowerup.com) and report
any inconsistencies to https://github.com/Dr-Noob/gpufetch/issues

                  MMMMMMMMMMMMMMMMMMMMMMMMMMMMMM  Name:                   NVIDIA GeForce GTX 1660 Ti
                  MMMMMMMMMMMMMMMMMMMMMMMMMMMMMM  GPU processor:          TU116
                .::   'MMMMMMMMMMMMMMMMMMMMMMMMM  Microarchitecture:      Turing (7.5)
           ccllooo;:;.       ;MMMMMMMMMMMMMMMMMM  Technology:             12nm
       cloc       :ooollcc:     :MMMMMMMMMMMMMMM  Max Frequency:          1770 MHz
    cloc      :ccl;      lolc,     ;MMMMMMMMMMMM  SMs:                    24
.cloo:    :clo    ;c:      .ool;     MMMMMMMMMMM  Cores/SM:               64
  ooo:    ooo     :ool,  .cloo.    ;lMMMMMMMMMMM  CUDA Cores:             1536
   ooo:    ooc    :ooooccooo.    :MMMM  lMMMMMMM  Tensor Cores:           192
     ooc.   ool:  :oooooo'    ,cloo.        MMMM  Memory:                 6 GiB GDDR6
      ool:.    olc:       .:cloo.          :MMMM  Memory frequency:       1500 MHz
         olc,     ;:::cccloo.          :MMMMMMMM  Bus width:              192 bit
            olcc::;              ,:ccloMMMMMMMMM  L2 Size:                2 MiB
                  :......oMMMMMMMMMMMMMMMMMMMMMM  Peak Performance:       5.44 TFLOP/s
                  :lllMMMMMMMMMMMMMMMMMMMMMMMMMM  Peak Performance (MMA): 0.00 MFLOP/s

Unknown Intel device id: 0x46A6

output:

[ERROR]: Unknown Intel device id: 0x46A6
Please, create a new issue with this error message on https://github.com/Dr-Noob/gpufetch/issues
[ERROR]: No GPU was detected! Available GPUs are:
- GPU 0:
  * Vendor: Intel
  * PCI id: 8086:46a6
Please, make sure that the appropiate backend is enabled:
- CUDA backend:  OFF
- Intel backend: ON
Visit https://github.com/Dr-Noob/gpufetch#2-backends for more information

I think my GPU is an Intel Arc a770 of some form

[WARNING TO PEOPLE RUNNING WSL] (Windows Sub-system for Linux)

Gpufetch is unable to identify gpu's through WSL, I don't know why, but it can't. I'm assuming this isn't an issue the developer of Gpufetch can fix as its on WSL's side, but this is just a fair warning to either use the windows version, or just use Linux

Picture of Terminal:

Can't detect my gpu [nvidia]

Add PCIe version info in NVIDIA GPUs

[ERROR]: No GPU was detected!

./gpufetch -v
[WARNING]: Unable to find a valid device for vendor id 0x8086 using pciutils
[ERROR]: No GPU was detected! Available GPUs are:
- GPU 0:
  * Vendor: NVIDIA
  * PCI id: 10de:128b
Please, make sure that the appropiate backend is enabled:
- CUDA backend:  OFF
- Intel backend: ON
Visit https://github.com/Dr-Noob/gpufetch#2-backends for more information

lsb_release -a
No LSB modules are available.
Distributor ID: Ubuntu
Description:    Ubuntu 18.04.6 LTS
Release:        18.04
Codename:       bionic

lspci -nn | grep VGA
02:00.0 VGA compatible controller [0300]: NVIDIA Corporation GK208B [GeForce GT 710] [10de:128b] (rev a1)

nvidia-smi
Thu May  9 04:24:19 2024
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 470.239.06   Driver Version: 470.239.06   CUDA Version: 11.4     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                               |                      |               MIG M. |
|===============================+======================+======================|
|   0  NVIDIA GeForce ...  Off  | 00000000:02:00.0 N/A |                  N/A |
| 50%   58C    P0    N/A /  N/A |    137MiB /   976MiB |     N/A      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
|  No running processes found                                                 |
+-----------------------------------------------------------------------------+

Unknown CUDA device id: 0x2820 on System76 Adder WS

Hi!

Got this error message on a System76 Adder WS with an RTX 4070.

Related issue: #46

Extract from lshw:

*-display
     description: VGA compatible controller
     product: NVIDIA Corporation
     vendor: NVIDIA Corporation
     physical id: 0
     bus info: pci@0000:01:00.0
     logical name: /dev/fb0
     version: a1
     width: 64 bits
     clock: 33MHz
     capabilities: pm msi pciexpress vga_controller bus_master cap_list rom fb
     configuration: depth=32 driver=nvidia latency=0 mode=1920x1080 visual=truecolor xres=1920 yres=1080
     resources: iomemory:120-11f iomemory:140-13f irq:201 memory:81000000-81ffffff memory:1200000000-13ffffffff memory:1400000000-1401ffffff ioport:6000(size=128) memory:82000000-8207ffff

Full cpufetch output:

[ERROR]: Unknown CUDA device id: 0x2820
Please, create a new issue with this error message on https://github.com/Dr-Noob/gpufetch/issues
[ERROR]: Unknown Intel device id: 0xA788
Please, create a new issue with this error message on https://github.com/Dr-Noob/gpufetch/issues

               'cccccccccccccccccccccccccc   
               ;oooooooooooooooooooooooool   
           .:::.     .oooooooooooooooooool   Name:                   NVIDIA GeForce RTX 4070 Laptop GPU
      .:cll;   ,c:::.     cooooooooooooool   GPU processor:          Unknown
   ,clo'      ;.   oolc:     ooooooooooool   Microarchitecture:      Unknown (8.9)
.cloo    ;cclo .      .olc.    coooooooool   Technology:             Unknown
oooo   :lo,    ;ll;    looc    :oooooooool   Max Frequency:          2175 MHz
 oooc   ool.   ;oooc;clol    :looooooooool   SMs:                    36
  :ooc   ,ol;  ;oooooo.   .cloo;     loool   Cores/SM:               128
    ool;   .olc.       ,:lool        .lool   CUDA Cores:             4608
      ool:.    ,::::ccloo.        :clooool   Tensor Cores:           144
         oolc::.            ':cclooooooool   Memory:                 8 GiB Unknown
               ;oooooooooooooooooooooooool   Memory frequency:       8001 MHz
                                             Bus width:              128 bit
                                             L2 Size:                32 MiB
######.  ##   ##  ##  ######   ##    ###     Peak Performance:       20.04 TFLOP/s
##   ##  ##   ##  ##  ##   ##  ##   #: :#    Peak Performance (MMA): 160.36 TFLOP/s
##   ##   ## ##   ##  ##   ##  ##  #######   
##   ##    ###    ##  ######   ## ##     ##

Please let me know if additional info is needed.

gpufetch does not detect Intel Geminilake integrated graphics

I have a Lenovo 500e 2nd gen running Kubuntu 23.04. Gpufetch is unable to detect any GPU (there must be some Intel iGPU).

The gpufetch output is,

[ERROR]: No GPU was detected! Available GPUs are:
Please, make sure that the appropiate backend is enabled:
- CUDA backend:  OFF
- Intel backend: ON
Visit https://github.com/Dr-Noob/gpufetch#2-backends for more information

Here is the cpufetch output.

                               ###############@               
                       ######@                ######@         
                  ###@                              ###@      
              ##@                                     ###@    
         ##@                                             ##@  Name:                Intel Celeron N4100
         ##@                                             ##@  Microarchitecture:   Goldmont Plus
      @                    ##@                ##@        ##@  Technology:          14nm
    #@   ##@   ########@   #####@   #####@    ##@        ##@  Max Frequency:       2.400 GHz
   #@    ##@   ##@    ##@  ##@    ###@  ###@  ##@        ##@  Cores:               4 cores
  #@     ##@   ##@    ##@  ##@    ##@    ##@  ##@       ##@   AVX:                 No
 #@      ##@   ##@    ##@  ##@    #########@  ##@     ###@    FMA:                 No
 #@      ##@   ##@    ##@  ##@    ##@         ##@   ####@     L1i Size:            32KB (128KB Total)
 #@       #@   ##@    ##@   ####@  ########@   #@  ##@        L1d Size:            24KB (96KB Total)
 ##@                                                          L2 Size:             4MB
  ##@                                                         Peak Performance:    38.40 GFLOP/s
  ###@                                        ###@            
    ####@                               #########@            
      #########@               ###############@               
          ##############################@

The output of lspci -nn | grep VGA is blank.

However, the output of
lspci -nn | grep Graphics is

00:02.0 Display controller [0380]: Intel Corporation GeminiLake [UHD Graphics 600] [8086:3185] (rev 03)

Intel Pentium G3420 igpu won't work w/ gpufetch

``sh
[ERROR]: Unkown Intel device id: 0x0402
Please, create a new issue with this error message on https://github.com/Dr-Noob/gpufetch/issues
[ERROR]: No GPU was detected, or the detected GPU is not supported by gpufetch
Please, make sure that the appropiate backend is enabled:

CUDA backend: OFF
Intel backend: ON
Visit https://github.com/Dr-Noob/gpufetch#2-backends for more information``

CMakeLists.txt incorrectly sets the samples directory for >= 11.6

Hi, thank you for making this awesome tool!

gpufetch would not compile on my computer because the samples directory no longer exists in v11.6, it has moved to here:

https://github.com/NVIDIA/cuda-samples

So you need to fix the CMakeLists.txt to this:

  # https://docs.nvidia.com/cuda/cuda-samples/index.html#new-features-in-cuda-toolkit-11-6
  # Not sure about this. Why the heck did they change this?
  if(${CMAKE_CUDA_COMPILER_VERSION} VERSION_GREATER_EQUAL "11.6")
    include(FetchContent)
    FetchContent_Declare(
      cudasamples
      GIT_REPOSITORY https://github.com/NVIDIA/cuda-samples
    )
    FetchContent_MakeAvailable(cudasamples)
    set(CUDA_SAMPLES_PATH ${cudasamples_SOURCE_DIR}/Common)
  else()
    set(CUDA_SAMPLES_PATH ${CMAKE_CUDA_COMPILER_TOOLKIT_ROOT}/samples/common/inc)
  endif()

After this change it compiles fine.

[ERROR]: Unknown Intel device id: 0x3EA0

I got the error message

[ERROR]: Unknown Intel device id: 0x3EA0

My processor is Intel® Core™ i5-8265U

Support printing all the GPUs

Maybe something like gpufetch -g a?