Hi there,
I have been trying to build the akmod-nvidia package for Fedora 44 on kernel 7.1.x and have had no success. I have tried forcing the build and rebuild with no luck.
I have attached the log of the most recent try: Failed akmod build · GitHub
A few steps I’ve done is increasing the file limits, disabling my dkms config, switching to nvidia-open, and limiting the threads it can use while building. Everything doesn’t work and I am unable to build for 7.1.3-201
make: *** [Makefile:115: modules] Segmentation fault (core dumped)
error: Bad exit status from /var/tmp/rpm-tmp.VXj5Xi (%build)
usually , this means that your system is not stable. Is your CPU overclocked?
My CPU is not overclocked!
I have a Ryzen 9 5900X, 64GB of DDR4 RAM, and an RTX 3090. These modules built just fine on kernel 6.x and 7.0.x
could be RAM issue. Can you recompile for an older kernel?
I just tried compiling for the current kernel version I am on and it passed.
hamcore@hatcore:~$ sudo akmods --rebuild --force
[sudo] password for hamcore:
Checking kmods exist for 7.1.3-201.fc44.x86_64 [ OK ]
Building and installing nvidia-kmod [FAILED]
Building rpms failed; see /var/cache/akmods/nvidia/610.43.03-1-for-7.1.3-201.fc44.x86_64.failed.log for details
Building and installing v4l2loopback-kmod [FAILED]
Building rpms failed; see /var/cache/akmods/v4l2loopback/0.15.4-1-for-7.1.3-201.fc44.x86_64.failed.log for details
Hint: Some kmods were ignored or failed to build or install.
You can try to rebuild and install them by by calling
'/usr/sbin/akmods --force' as root.
Checking kmods exist for 7.0.14-201.fc44.x86_64 [ OK ]
Building and installing nvidia-kmod [ OK ]
Building and installing v4l2loopback-kmod [ OK ]
strange. maybe verify rpm -V the new kernel rpms. Did the v4l2loopback module also failed compiling with a segmentation fault?
v4l2loopback did also have the segmentation fault. I am unsure what you mean about the rpm -V command
something like rpm -V $( rpm -qa kernel\* |grep 7.1.3 ) should not give any output.
you could also try reinstalling the kernel sudo dnf reinstall kernel\*
I’ve just installed the new 7.1.3-201 kernel and the nvidia kernel modules were successfully build.
2026/07/15 01:57:44 akmodsbuild: Executing(rmbuild): /bin/sh -e /var/tmp/rpm-tmp.tDvgFz
2026/07/15 01:57:44 akmodsbuild: + umask 022
2026/07/15 01:57:44 akmodsbuild: + cd /tmp/akmodsbuild.dJyy35TJ//BUILD/nvidia-kmod-610.43.03-build
2026/07/15 01:57:44 akmodsbuild: + test -d /tmp/akmodsbuild.dJyy35TJ//BUILD/nvidia-kmod-610.43.03-build
2026/07/15 01:57:44 akmodsbuild: + /usr/bin/chmod -Rf a+rX,u+w,g-w,o-w /tmp/akmodsbuild.dJyy35TJ//BUILD/nvidia-kmod-610.43.03-build
2026/07/15 01:57:44 akmodsbuild: + rm -rf /tmp/akmodsbuild.dJyy35TJ//BUILD/nvidia-kmod-610.43.03-build
2026/07/15 01:57:44 akmodsbuild: + RPM_EC=0
2026/07/15 01:57:44 akmodsbuild: ++ jobs -p
2026/07/15 01:57:44 akmodsbuild: + exit 0
2026/07/15 01:57:45 akmods: Installing newly built rpms
2026/07/15 01:57:45 akmods: DNF detected
Updating and loading repositories:
Repositories loaded.
Package Arch Version Repository Size
Installing:
kmod-nvidia-7.1.3-201.fc44.x86_64 x86_64 3:610.43.03-1.fc44 @commandline 11.8 MiB
Transaction Summary:
Installing: 1 package
Total size of inbound packages is 10 MiB. Need to download 0 B.
After this operation, 12 MiB extra will be used (install 12 MiB, remove 0 B).
Running transaction
[1/3] Verify package files 100% | 23.0 B/s | 1.0 B | 00m00s
[2/3] Prepare transaction 100% | 3.0 B/s | 1.0 B | 00m00s
[3/3] Installing kmod-nvidia-7.1.3-201. 100% | 1.4 MiB/s | 11.8 MiB | 00m08s
Warning: skipped OpenPGP checks for 1 package from repository: @commandline
Complete!
2026/07/15 01:57:54 akmods: Successful.
Hi All, do You try with command -v (verbose) ? in step’s of compilation kernel in verbobe may see whot is wrong in build (MOK key, permission or missing file or path or wrong old drivers for module). Try to get module with verbose in compilation and build without any errors in log. >> akmods --verbose --rebuild --force
Full disclosure,
This entire thing frustrated me enough that I started fresh. Unfortunately I could never narrow down the issue in the end, which will make this topic stale. My apologies. I have moved on to another flavor of Linux that manages GPU drivers a bit differently to challenge myself. I appreciate all of the help.
