aboutsummaryrefslogtreecommitdiff
path: root/system/nvidia-kernel/README
diff options
context:
space:
mode:
Diffstat (limited to 'system/nvidia-kernel/README')
-rw-r--r--system/nvidia-kernel/README44
1 files changed, 44 insertions, 0 deletions
diff --git a/system/nvidia-kernel/README b/system/nvidia-kernel/README
index 2af2536f501b5..f2c3bbc467d5a 100644
--- a/system/nvidia-kernel/README
+++ b/system/nvidia-kernel/README
@@ -5,6 +5,10 @@ To build the package for a kernel different from the running one,
start the script setting the KERNEL variable as in
KERNEL=4.6.3 ./nvidia-kernel.SlackBuild
+This script now includes the option to build the open kernel module
+instead of the default proprietary module. To build the open module,
+pass "OPEN=yes" to the script. (See notes below.)
+
A default config file is placed at
/usr/share/X11/xorg.conf.d/10-nvidia.conf
to make sure that X loads the nvidia module. If you need to make
@@ -13,3 +17,43 @@ You do not need this file at all if you have a proper and complete
xorg.conf.
The xf86-video-nouveau-blacklist package from /extra is required.
+
+After installation, you will need to reboot your computer for the
+changes to take effect.
+
+NOTES ON THE OPEN KERNEL MODULE (quoted from the Nvidia driver README)
+
+"The proprietary flavor supports the GPU architectures Maxwell, Pascal,
+Volta, Turing, and later GPUs.
+
+"The open flavor of kernel modules supports Turing and later GPUs. The
+open kernel modules cannot support GPUs before Turing, because the open
+kernel modules depend on the GPU System Processor (GSP) first
+introduced in Turing.
+
+"Most features of the Linux GPU driver are supported with the open
+flavor of kernel modules, including CUDA, Vulkan, OpenGL, OptiX, and
+X11. We eventually intend for the open kernel modules to replace the
+proprietary kernel modules.
+
+"The following features will only work with the open kernel modules
+flavor of the driver:
+
+ o NVIDIA Confidential Computing
+
+ o Magnum IO GPUDirect Storage (GDS)
+
+ o Heterogeneous Memory Management (HMM)
+
+ o CPU affinity for GPU fault handlers
+
+ o DMABUF support for CUDA allocations
+
+G-sync on notebooks is not yet supported by the open kernel modules.
+
+Use of the open kernel modules on GeForce and Workstation GPUs should
+be considered Beta quality in this release and no longer requires
+setting of the "NVreg_OpenRmEnableUnsupportedGpus" nvidia.ko kernel
+module parameter. The open kernel modules are suitable for broad
+usage, and NVIDIA requests feedback on any issues encountered that are
+specific to them."