github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-08 08:59:13 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #8730] cmake build failure on AMD CPUs #5664

New Issue
Open
opened 2026-04-12 16:57:41 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-12 16:57:41 -05:00
Owner
Copy Link

Originally created by @garrettbyrd on GitHub (Jan 31, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/8730

Originally assigned to: @mxyng on GitHub.

What is the issue?

Versions:

gcc   12.4.0
cmake 3.31.2
go    1.23.3
rocm  6.1.2

Previously, I was able to build/run the ollama binary on the same systems/versions (tested on commit 2ef3c803a151a0a9b1776c9ebe6a7e86b3971660). I.e., I was able to build/run when make was the build system.

Currently, now testing on 3f0cb36bdbe4dd6ac35a39598c7253cb0cf1349a, which uses cmake, I am hitting an error when building.

This has been tested on the following hardware configurations:
System 1 (gfx942):

          __wgliliiligw_,              garrett@nicholson
       _williiiiiiliilililw,           -----------------
     _%iiiiiilililiiiiiiiiiii_         OS: Rocky Linux 9.5 (Blue Onyx) x86_64
   .Qliiiililiiiiiiililililiilm.       Host: Super Server (0123456789)
  _iiiiiliiiiiililiiiiiiiiiiliil,      Kernel: Linux 5.14.0-503.19.1.el9_5.x86_64
 .lililiiilililiiiilililililiiiii,     Uptime: 19 days, 23 hours, 38 mins
_liiiiiiliiiiiiiliiiiiF{iiiiiilili,    Packages: 2117 (rpm)
jliililiiilililiiili@`  ~ililiiiiiL    Shell: bash 5.1.8
iiiliiiiliiiiiiili>`      ~liililii    Display (Virtual-2): 1600x1200 @ 60 Hz
liliiiliiilililii`         -9liiiil    Cursor: Adwaita
iiiiiliiliiiiii~             "4lili    Terminal: slurmstepd: [344.interactive]
4ililiiiiilil~|      -w,       )4lf    CPU: 4 x AMD Instinct MI300A Accelerator (192) @ 3.70 GHz
-liiiiililiF'       _liig,       )'    GPU: ASPEED Technology, Inc. ASPEED Graphics Family
 )iiiliii@`       _QIililig,           Memory: 7.89 GiB / 501.75 GiB (2%)
  )iiii>`       .Qliliiiililw          Swap: 40.00 KiB / 4.00 GiB (0%)
   )<>~       .mliiiiiliiiiiil,        
            _gllilililiililii~        
           giliiiiiiiiiiiiT`           
          -^~$ililili@~~'

System 2 (gfx90a):

          __wgliliiligw_,              garrett@noether
       _williiiiiiliilililw,           ---------------
     _%iiiiiilililiiiiiiiiiii_         OS: Rocky Linux 9.5 (Blue Onyx) x86_64
   .Qliiiililiiiiiiililililiilm.       Host: AS -4124GS-TNR+ (0123456789)
  _iiiiiliiiiiililiiiiiiiiiiliil,      Kernel: Linux 5.14.0-503.16.1.el9_5.x86_64
 .lililiiilililiiiilililililiiiii,     Uptime: 21 days, 19 hours, 23 mins
_liiiiiiliiiiiiiliiiiiF{iiiiiilili,    Packages: 2112 (rpm)
jliililiiilililiiili@`  ~ililiiiiiL    Shell: bash 5.1.8
iiiliiiiliiiiiiili>`      ~liililii    Display (VGA-1): 1024x768 @ 60 Hz
liliiiliiilililii`         -9liiiil    Cursor: Adwaita
iiiiiliiliiiiii~             "4lili    Terminal: slurmstepd: [343.interactive]
4ililiiiiilil~|      -w,       )4lf    CPU: 2 x AMD EPYC 7313 (64) @ 3.73 GHz
-liiiiililiF'       _liig,       )'    GPU 1: AMD Instinct MI210 [Discrete]
 )iiiliii@`       _QIililig,           GPU 2: AMD Instinct MI210 [Discrete]
  )iiii>`       .Qliliiiililw          GPU 3: AMD Instinct MI210 [Discrete]
   )<>~       .mliiiiiliiiiiil,        GPU 4: AMD Instinct MI210 [Discrete]
            _gllilililiililii~         GPU 5: ASPEED Technology, Inc. ASPEED Graphics Family
           giliiiiiiiiiiiiT`           Memory: 7.53 GiB / 235.49 GiB (3%)
          -^~$ililili@~~'              Swap: 3.00 MiB / 4.00 GiB (0%)

cmake -B build runs fine. Actually building with cmake --build build outputs the same output for both systems:

[  5%] Built target ggml-base
[  6%] Built target ggml-cpu-sandybridge-feats
[ 13%] Built target ggml-cpu-sandybridge
[ 14%] Built target ggml-cpu-haswell-feats
[ 21%] Built target ggml-cpu-haswell
[ 22%] Built target ggml-cpu-skylakex-feats
[ 29%] Built target ggml-cpu-skylakex
[ 29%] Built target ggml-cpu-icelake-feats
[ 36%] Built target ggml-cpu-icelake
[ 36%] Built target ggml-cpu-alderlake-feats
[ 37%] Building CXX object ml/backend/ggml/ggml/src/CMakeFiles/ggml-cpu-alderlake.dir/ggml-cpu/ggml-cpu-aarch64.cpp.o
In function 'block_q4_0x4 make_block_q4_0x4(block_q4_0*, unsigned int)',
    inlined from 'int repack_q4_0_to_q4_0_4_bl(ggml_tensor*, int, const void*, size_t)' at [path/to/ollama]/ml/backend/ggml/ggml/src/ggml-cpu/ggml-cpu-aarch64.cpp:3685:39:
[path/to/ollama]/ml/backend/ggml/ggml/src/ggml-cpu/ggml-cpu-aarch64.cpp:3614:19: warning: writing 32 bytes into a region of size 0 [-Wstringop-overflow=]
 3614 |             memcpy(&out.qs[dst_offset], &elems, sizeof(uint64_t));
      |             ~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
[path/to/ollama]/ml/backend/ggml/ggml/src/ggml-cpu/ggml-cpu-aarch64.cpp: In function 'int repack_q4_0_to_q4_0_4_bl(ggml_tensor*, int, const void*, size_t)':
[path/to/ollama]/ml/backend/ggml/ggml/src/ggml-cpu/ggml-cpu-aarch64.cpp:3685:20: note: at offset 72 into destination object '<anonymous>' of size 72
 3685 |             *dst++ = make_block_q4_0x4(dst_tmp, interleave_block);
      |             ~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In function 'block_q4_0x4 make_block_q4_0x4(block_q4_0*, unsigned int)',
    inlined from 'int repack_q4_0_to_q4_0_4_bl(ggml_tensor*, int, const void*, size_t)' at [path/to/ollama]/ml/backend/ggml/ggml/src/ggml-cpu/ggml-cpu-aarch64.cpp:3685:39:
[path/to/ollama]/ml/backend/ggml/ggml/src/ggml-cpu/ggml-cpu-aarch64.cpp:3614:19: warning: writing 32 bytes into a region of size 0 [-Wstringop-overflow=]
 3614 |             memcpy(&out.qs[dst_offset], &elems, sizeof(uint64_t));
      |             ~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
[path/to/ollama]/ml/backend/ggml/ggml/src/ggml-cpu/ggml-cpu-aarch64.cpp: In function 'int repack_q4_0_to_q4_0_4_bl(ggml_tensor*, int, const void*, size_t)':
[path/to/ollama]/ml/backend/ggml/ggml/src/ggml-cpu/ggml-cpu-aarch64.cpp:3685:20: note: at offset 104 into destination object '<anonymous>' of size 72
 3685 |             *dst++ = make_block_q4_0x4(dst_tmp, interleave_block);
      |             ~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/tmp/ccVIOheg.s: Assembler messages:
/tmp/ccVIOheg.s:731: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:742: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:747: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:755: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:760: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:765: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:771: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:777: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:785: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:788: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:791: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:798: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:806: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:810: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:814: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:829: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:834: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:835: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:844: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:848: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:850: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:854: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:856: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:865: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:882: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:886: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:888: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:894: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:898: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:902: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:907: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:911: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1268: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1275: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1283: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1288: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1296: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1300: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1303: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1308: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1317: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1322: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1326: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1333: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1339: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1343: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1350: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1355: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1359: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1364: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1373: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1378: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1383: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1387: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1392: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1399: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1408: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1414: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1422: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1428: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1432: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1436: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1441: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:1444: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:4135: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:4143: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:4152: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:4155: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:4163: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:4170: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:4181: Error: unsupported instruction `vpdpbusd'
/tmp/ccVIOheg.s:4185: Error: unsupported instruction `vpdpbusd'
gmake[2]: *** [ml/backend/ggml/ggml/src/CMakeFiles/ggml-cpu-alderlake.dir/build.make:107: ml/backend/ggml/ggml/src/CMakeFiles/ggml-cpu-alderlake.dir/ggml-cpu/ggml-cpu-aarch64.cpp.o] Error 1
gmake[1]: *** [CMakeFiles/Makefile2:558: ml/backend/ggml/ggml/src/CMakeFiles/ggml-cpu-alderlake.dir/all] Error 2
gmake: *** [Makefile:136: all] Error 2

OS

Linux

GPU

AMD

CPU

AMD

Ollama version

commit 3f0cb36

Originally created by @garrettbyrd on GitHub (Jan 31, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/8730 Originally assigned to: @mxyng on GitHub. ### What is the issue? Versions: ``` gcc 12.4.0 cmake 3.31.2 go 1.23.3 rocm 6.1.2 ``` Previously, I was able to build/run the `ollama` binary on the same systems/versions (tested on commit [`2ef3c803a151a0a9b1776c9ebe6a7e86b3971660`](https://github.com/ollama/ollama/tree/2ef3c803a151a0a9b1776c9ebe6a7e86b3971660)). I.e., I was able to build/run when `make` was the build system. Currently, now testing on [`3f0cb36bdbe4dd6ac35a39598c7253cb0cf1349a`](https://github.com/ollama/ollama/tree/3f0cb36bdbe4dd6ac35a39598c7253cb0cf1349a), which uses `cmake`, I am hitting an error when building. This has been tested on the following hardware configurations: System 1 (`gfx942`): ``` __wgliliiligw_, garrett@nicholson _williiiiiiliilililw, ----------------- _%iiiiiilililiiiiiiiiiii_ OS: Rocky Linux 9.5 (Blue Onyx) x86_64 .Qliiiililiiiiiiililililiilm. Host: Super Server (0123456789) _iiiiiliiiiiililiiiiiiiiiiliil, Kernel: Linux 5.14.0-503.19.1.el9_5.x86_64 .lililiiilililiiiilililililiiiii, Uptime: 19 days, 23 hours, 38 mins _liiiiiiliiiiiiiliiiiiF{iiiiiilili, Packages: 2117 (rpm) jliililiiilililiiili@` ~ililiiiiiL Shell: bash 5.1.8 iiiliiiiliiiiiiili>` ~liililii Display (Virtual-2): 1600x1200 @ 60 Hz liliiiliiilililii` -9liiiil Cursor: Adwaita iiiiiliiliiiiii~ "4lili Terminal: slurmstepd: [344.interactive] 4ililiiiiilil~| -w, )4lf CPU: 4 x AMD Instinct MI300A Accelerator (192) @ 3.70 GHz -liiiiililiF' _liig, )' GPU: ASPEED Technology, Inc. ASPEED Graphics Family )iiiliii@` _QIililig, Memory: 7.89 GiB / 501.75 GiB (2%) )iiii>` .Qliliiiililw Swap: 40.00 KiB / 4.00 GiB (0%) )<>~ .mliiiiiliiiiiil, _gllilililiililii~ giliiiiiiiiiiiiT` -^~$ililili@~~' ``` System 2 (`gfx90a`): ``` __wgliliiligw_, garrett@noether _williiiiiiliilililw, --------------- _%iiiiiilililiiiiiiiiiii_ OS: Rocky Linux 9.5 (Blue Onyx) x86_64 .Qliiiililiiiiiiililililiilm. Host: AS -4124GS-TNR+ (0123456789) _iiiiiliiiiiililiiiiiiiiiiliil, Kernel: Linux 5.14.0-503.16.1.el9_5.x86_64 .lililiiilililiiiilililililiiiii, Uptime: 21 days, 19 hours, 23 mins _liiiiiiliiiiiiiliiiiiF{iiiiiilili, Packages: 2112 (rpm) jliililiiilililiiili@` ~ililiiiiiL Shell: bash 5.1.8 iiiliiiiliiiiiiili>` ~liililii Display (VGA-1): 1024x768 @ 60 Hz liliiiliiilililii` -9liiiil Cursor: Adwaita iiiiiliiliiiiii~ "4lili Terminal: slurmstepd: [343.interactive] 4ililiiiiilil~| -w, )4lf CPU: 2 x AMD EPYC 7313 (64) @ 3.73 GHz -liiiiililiF' _liig, )' GPU 1: AMD Instinct MI210 [Discrete] )iiiliii@` _QIililig, GPU 2: AMD Instinct MI210 [Discrete] )iiii>` .Qliliiiililw GPU 3: AMD Instinct MI210 [Discrete] )<>~ .mliiiiiliiiiiil, GPU 4: AMD Instinct MI210 [Discrete] _gllilililiililii~ GPU 5: ASPEED Technology, Inc. ASPEED Graphics Family giliiiiiiiiiiiiT` Memory: 7.53 GiB / 235.49 GiB (3%) -^~$ililili@~~' Swap: 3.00 MiB / 4.00 GiB (0%) ``` `cmake -B build` runs fine. Actually building with `cmake --build build` outputs the same output for both systems: ``` [ 5%] Built target ggml-base [ 6%] Built target ggml-cpu-sandybridge-feats [ 13%] Built target ggml-cpu-sandybridge [ 14%] Built target ggml-cpu-haswell-feats [ 21%] Built target ggml-cpu-haswell [ 22%] Built target ggml-cpu-skylakex-feats [ 29%] Built target ggml-cpu-skylakex [ 29%] Built target ggml-cpu-icelake-feats [ 36%] Built target ggml-cpu-icelake [ 36%] Built target ggml-cpu-alderlake-feats [ 37%] Building CXX object ml/backend/ggml/ggml/src/CMakeFiles/ggml-cpu-alderlake.dir/ggml-cpu/ggml-cpu-aarch64.cpp.o In function 'block_q4_0x4 make_block_q4_0x4(block_q4_0*, unsigned int)', inlined from 'int repack_q4_0_to_q4_0_4_bl(ggml_tensor*, int, const void*, size_t)' at [path/to/ollama]/ml/backend/ggml/ggml/src/ggml-cpu/ggml-cpu-aarch64.cpp:3685:39: [path/to/ollama]/ml/backend/ggml/ggml/src/ggml-cpu/ggml-cpu-aarch64.cpp:3614:19: warning: writing 32 bytes into a region of size 0 [-Wstringop-overflow=] 3614 | memcpy(&out.qs[dst_offset], &elems, sizeof(uint64_t)); | ~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ [path/to/ollama]/ml/backend/ggml/ggml/src/ggml-cpu/ggml-cpu-aarch64.cpp: In function 'int repack_q4_0_to_q4_0_4_bl(ggml_tensor*, int, const void*, size_t)': [path/to/ollama]/ml/backend/ggml/ggml/src/ggml-cpu/ggml-cpu-aarch64.cpp:3685:20: note: at offset 72 into destination object '<anonymous>' of size 72 3685 | *dst++ = make_block_q4_0x4(dst_tmp, interleave_block); | ~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In function 'block_q4_0x4 make_block_q4_0x4(block_q4_0*, unsigned int)', inlined from 'int repack_q4_0_to_q4_0_4_bl(ggml_tensor*, int, const void*, size_t)' at [path/to/ollama]/ml/backend/ggml/ggml/src/ggml-cpu/ggml-cpu-aarch64.cpp:3685:39: [path/to/ollama]/ml/backend/ggml/ggml/src/ggml-cpu/ggml-cpu-aarch64.cpp:3614:19: warning: writing 32 bytes into a region of size 0 [-Wstringop-overflow=] 3614 | memcpy(&out.qs[dst_offset], &elems, sizeof(uint64_t)); | ~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ [path/to/ollama]/ml/backend/ggml/ggml/src/ggml-cpu/ggml-cpu-aarch64.cpp: In function 'int repack_q4_0_to_q4_0_4_bl(ggml_tensor*, int, const void*, size_t)': [path/to/ollama]/ml/backend/ggml/ggml/src/ggml-cpu/ggml-cpu-aarch64.cpp:3685:20: note: at offset 104 into destination object '<anonymous>' of size 72 3685 | *dst++ = make_block_q4_0x4(dst_tmp, interleave_block); | ~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ /tmp/ccVIOheg.s: Assembler messages: /tmp/ccVIOheg.s:731: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:742: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:747: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:755: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:760: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:765: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:771: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:777: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:785: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:788: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:791: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:798: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:806: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:810: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:814: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:829: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:834: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:835: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:844: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:848: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:850: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:854: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:856: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:865: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:882: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:886: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:888: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:894: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:898: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:902: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:907: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:911: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1268: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1275: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1283: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1288: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1296: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1300: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1303: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1308: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1317: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1322: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1326: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1333: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1339: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1343: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1350: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1355: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1359: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1364: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1373: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1378: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1383: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1387: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1392: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1399: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1408: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1414: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1422: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1428: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1432: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1436: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1441: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:1444: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:4135: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:4143: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:4152: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:4155: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:4163: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:4170: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:4181: Error: unsupported instruction `vpdpbusd' /tmp/ccVIOheg.s:4185: Error: unsupported instruction `vpdpbusd' gmake[2]: *** [ml/backend/ggml/ggml/src/CMakeFiles/ggml-cpu-alderlake.dir/build.make:107: ml/backend/ggml/ggml/src/CMakeFiles/ggml-cpu-alderlake.dir/ggml-cpu/ggml-cpu-aarch64.cpp.o] Error 1 gmake[1]: *** [CMakeFiles/Makefile2:558: ml/backend/ggml/ggml/src/CMakeFiles/ggml-cpu-alderlake.dir/all] Error 2 gmake: *** [Makefile:136: all] Error 2 ``` ### OS Linux ### GPU AMD ### CPU AMD ### Ollama version commit `3f0cb36`
GiteaMirror added the bug label 2026-04-12 16:57:41 -05:00
GiteaMirror commented 2026-04-12 16:57:42 -05:00
Author
Owner
Copy Link

@mxyng commented on GitHub (Jan 31, 2025):

Thanks for the issue, I'll take a closer look at this but my initial suspicion is the compiler lacks support for AVX VNNI which requires GCC >= 11.1.

Can you attach the contents of your cmake -B build? Since you mentioned using GCC 12.4.0, perhaps cmake is using an different compiler.

You can also try clang

<!-- gh-comment-id:2628004183 --> @mxyng commented on GitHub (Jan 31, 2025): Thanks for the issue, I'll take a closer look at this but my initial suspicion is the compiler lacks support for AVX VNNI which requires GCC >= 11.1. Can you attach the contents of your `cmake -B build`? Since you mentioned using GCC 12.4.0, perhaps `cmake` is using an different compiler. You can also try `clang`
GiteaMirror commented 2026-04-12 16:57:43 -05:00
Author
Owner
Copy Link

@garrettbyrd commented on GitHub (Jan 31, 2025):

Do you want the contents of cmake_install.cmake, CMakeCache.txt, or Makefile?

Or just this output?

$ cmake -B build
-- Warning: ccache not found - consider installing it for faster compilation or disable this warning with GGML_CCACHE=OFF
-- CMAKE_SYSTEM_PROCESSOR: x86_64
-- Including CPU backend
-- x86 detected
-- Adding CPU backend variant ggml-cpu-sandybridge: -msse4.2;-mavx GGML_SSE42;GGML_AVX
-- x86 detected
-- Adding CPU backend variant ggml-cpu-haswell: -msse4.2;-mf16c;-mfma;-mavx;-mavx2 GGML_SSE42;GGML_F16C;GGML_FMA;GGML_AVX;GGML_AVX2
-- x86 detected
-- Adding CPU backend variant ggml-cpu-skylakex: -msse4.2;-mf16c;-mfma;-mavx;-mavx2;-mavx512f;-mavx512cd;-mavx512vl;-mavx512dq;-mavx512bw GGML_SSE42;GGML_F16C;GGML_FMA;GGML_AVX;GGML_AVX2;GGML_AVX512
-- x86 detected
-- Adding CPU backend variant ggml-cpu-icelake: -msse4.2;-mf16c;-mfma;-mavx;-mavx2;-mavx512f;-mavx512cd;-mavx512vl;-mavx512dq;-mavx512bw;-mavx512vbmi;-mavx512vnni GGML_SSE42;GGML_F16C;GGML_FMA;GGML_AVX;GGML_AVX2;GGML_AVX512;GGML_AVX512_VBMI;GGML_AVX512_VNNI
-- x86 detected
-- Adding CPU backend variant ggml-cpu-alderlake: -msse4.2;-mf16c;-mfma;-mavx;-mavx2;-mavxvnni GGML_SSE42;GGML_F16C;GGML_FMA;GGML_AVX;GGML_AVX2;GGML_AVX_VNNI
-- x86 detected
-- Adding CPU backend variant ggml-cpu-sapphirerapids: -msse4.2;-mf16c;-mfma;-mavx;-mavx2;-mavx512f;-mavx512cd;-mavx512vl;-mavx512dq;-mavx512bw;-mavx512vbmi;-mavx512vnni;-mavx512bf16;-mamx-tile;-mamx-int8 GGML_SSE42;GGML_F16C;GGML_FMA;GGML_AVX;GGML_AVX2;GGML_AVX512;GGML_AVX512_VBMI;GGML_AVX512_VNNI;GGML_AVX512_BF16;GGML_AMX_TILE;GGML_AMX_INT8
-- HIP and hipBLAS found
-- Configuring done (0.2s)
-- Generating done (0.5s)
-- Build files have been written to: /home/garrett/projects/ai-playground/ollama-cmake/build
<!-- gh-comment-id:2628102633 --> @garrettbyrd commented on GitHub (Jan 31, 2025): Do you want the contents of `cmake_install.cmake`, `CMakeCache.txt`, or `Makefile`? Or just this output? ``` $ cmake -B build -- Warning: ccache not found - consider installing it for faster compilation or disable this warning with GGML_CCACHE=OFF -- CMAKE_SYSTEM_PROCESSOR: x86_64 -- Including CPU backend -- x86 detected -- Adding CPU backend variant ggml-cpu-sandybridge: -msse4.2;-mavx GGML_SSE42;GGML_AVX -- x86 detected -- Adding CPU backend variant ggml-cpu-haswell: -msse4.2;-mf16c;-mfma;-mavx;-mavx2 GGML_SSE42;GGML_F16C;GGML_FMA;GGML_AVX;GGML_AVX2 -- x86 detected -- Adding CPU backend variant ggml-cpu-skylakex: -msse4.2;-mf16c;-mfma;-mavx;-mavx2;-mavx512f;-mavx512cd;-mavx512vl;-mavx512dq;-mavx512bw GGML_SSE42;GGML_F16C;GGML_FMA;GGML_AVX;GGML_AVX2;GGML_AVX512 -- x86 detected -- Adding CPU backend variant ggml-cpu-icelake: -msse4.2;-mf16c;-mfma;-mavx;-mavx2;-mavx512f;-mavx512cd;-mavx512vl;-mavx512dq;-mavx512bw;-mavx512vbmi;-mavx512vnni GGML_SSE42;GGML_F16C;GGML_FMA;GGML_AVX;GGML_AVX2;GGML_AVX512;GGML_AVX512_VBMI;GGML_AVX512_VNNI -- x86 detected -- Adding CPU backend variant ggml-cpu-alderlake: -msse4.2;-mf16c;-mfma;-mavx;-mavx2;-mavxvnni GGML_SSE42;GGML_F16C;GGML_FMA;GGML_AVX;GGML_AVX2;GGML_AVX_VNNI -- x86 detected -- Adding CPU backend variant ggml-cpu-sapphirerapids: -msse4.2;-mf16c;-mfma;-mavx;-mavx2;-mavx512f;-mavx512cd;-mavx512vl;-mavx512dq;-mavx512bw;-mavx512vbmi;-mavx512vnni;-mavx512bf16;-mamx-tile;-mamx-int8 GGML_SSE42;GGML_F16C;GGML_FMA;GGML_AVX;GGML_AVX2;GGML_AVX512;GGML_AVX512_VBMI;GGML_AVX512_VNNI;GGML_AVX512_BF16;GGML_AMX_TILE;GGML_AMX_INT8 -- HIP and hipBLAS found -- Configuring done (0.2s) -- Generating done (0.5s) -- Build files have been written to: /home/garrett/projects/ai-playground/ollama-cmake/build ```
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#5664
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?