Merge pull request #320830 from philiptaron/nixos-nvidia-change-datacenter-defaults

nixos/nvidia: move TOPOLOGY_FILE_PATH and DATABASE_PATH into nvidia-fabricmanager service definition
This commit is contained in:
Someone 2024-06-27 15:56:29 +00:00 committed by GitHub
commit 35a472db57
No known key found for this signature in database
GPG key ID: B5690EEEBB952194

View file

@ -46,8 +46,6 @@ in
TRUNK_LINK_FAILURE_MODE = 0; TRUNK_LINK_FAILURE_MODE = 0;
NVSWITCH_FAILURE_MODE = 0; NVSWITCH_FAILURE_MODE = 0;
ABORT_CUDA_JOBS_ON_FM_EXIT = 1; ABORT_CUDA_JOBS_ON_FM_EXIT = 1;
TOPOLOGY_FILE_PATH = "${nvidia_x11.fabricmanager}/share/nvidia-fabricmanager/nvidia/nvswitch";
DATABASE_PATH = "${nvidia_x11.fabricmanager}/share/nvidia-fabricmanager/nvidia/nvswitch";
}; };
defaultText = lib.literalExpression '' defaultText = lib.literalExpression ''
{ {
@ -69,8 +67,6 @@ in
TRUNK_LINK_FAILURE_MODE=0; TRUNK_LINK_FAILURE_MODE=0;
NVSWITCH_FAILURE_MODE=0; NVSWITCH_FAILURE_MODE=0;
ABORT_CUDA_JOBS_ON_FM_EXIT=1; ABORT_CUDA_JOBS_ON_FM_EXIT=1;
TOPOLOGY_FILE_PATH="''${nvidia_x11.fabricmanager}/share/nvidia-fabricmanager/nvidia/nvswitch";
DATABASE_PATH="''${nvidia_x11.fabricmanager}/share/nvidia-fabricmanager/nvidia/nvswitch";
} }
''; '';
description = '' description = ''
@ -628,7 +624,14 @@ in
TimeoutStartSec = 240; TimeoutStartSec = 240;
ExecStart = ExecStart =
let let
nv-fab-conf = settingsFormat.generate "fabricmanager.conf" cfg.datacenter.settings; # Since these rely on the `nvidia_x11.fabricmanager` derivation, they're
# unsuitable to be mentioned in the configuration defaults, but they _can_
# be overridden in `cfg.datacenter.settings` if needed.
fabricManagerConfDefaults = {
TOPOLOGY_FILE_PATH = "${nvidia_x11.fabricmanager}/share/nvidia-fabricmanager/nvidia/nvswitch";
DATABASE_PATH = "${nvidia_x11.fabricmanager}/share/nvidia-fabricmanager/nvidia/nvswitch";
};
nv-fab-conf = settingsFormat.generate "fabricmanager.conf" (fabricManagerConfDefaults // cfg.datacenter.settings);
in in
"${lib.getExe nvidia_x11.fabricmanager} -c ${nv-fab-conf}"; "${lib.getExe nvidia_x11.fabricmanager} -c ${nv-fab-conf}";
LimitCORE = "infinity"; LimitCORE = "infinity";