Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

adjust SitePackage.lua for multiple CUDA/cu* modules #798

Open
wants to merge 1 commit into
base: 2023.06-software.eessi.io
Choose a base branch
from
Open

adjust SitePackage.lua for multiple CUDA/cu* modules #798

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
38 changes: 22 additions & 16 deletions create_lmodsitepackage.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -107,36 +107,42 @@
end


local function eessi_cuda_enabled_load_hook(t)
local function eessi_cuda_and_libraries_enabled_load_hook(t)
local frameStk = require("FrameStk"):singleton()
local mt = frameStk:mt()
local simpleName = string.match(t.modFullName, "(.-)/")
-- If we try to load CUDA itself, check if the full CUDA SDK was installed on the host in host_injections.
-- This is required for end users to build additional CUDA software. If the full SDK isn't present, refuse
-- to load the CUDA module and print an informative message on how to set up GPU support for EESSI
local packagesList = { ["CUDA"] = true, ["cuDNN"] = true }
-- If we try to load any of the modules in packagesList, we check if the
-- full package was installed on the host in host_injections.
-- This is required for end users to build additional software that depends
-- on the package. If the full SDK isn't present, refuse
-- to load the module and print an informative message on how to set up GPU support for EESSI
local refer_to_docs = "For more information on how to do this, see https://www.eessi.io/docs/gpu/.\\n"
if simpleName == 'CUDA' then
if packagesList[simpleName] then
-- simpleName is a module in packagesList
-- get the full host_injections path
local hostInjections = string.gsub(os.getenv('EESSI_SOFTWARE_PATH') or "", 'versions', 'host_injections')
-- build final path where the CUDA software should be installed
local cudaEasyBuildDir = hostInjections .. "/software/" .. t.modFullName .. "/easybuild"
local cudaDirExists = isDir(cudaEasyBuildDir)
if not cudaDirExists then

-- build final path where the software should be installed
local packageEasyBuildDir = hostInjections .. "/software/" .. t.modFullName .. "/easybuild"
local packageDirExists = isDir(packageEasyBuildDir)
if not packageDirExists then
local advice = "but while the module file exists, the actual software is not entirely shipped with EESSI "
advice = advice .. "due to licencing. You will need to install a full copy of the CUDA SDK where EESSI "
advice = advice .. "due to licencing. You will need to install a full copy of the " .. simpleName .. " package where EESSI "
advice = advice .. "can find it.\\n"
advice = advice .. refer_to_docs
LmodError("\\nYou requested to load ", simpleName, " ", advice)
end
end
-- when loading CUDA enabled modules check if the necessary driver libraries are accessible to the EESSI linker,
-- when loading CUDA (and cu*) enabled modules check if the necessary driver libraries are accessible to the EESSI linker,
-- otherwise, refuse to load the requested module and print error message
local checkGpu = mt:haveProperty(simpleName,"arch","gpu")
local overrideGpuCheck = os.getenv("EESSI_OVERRIDE_GPU_CHECK")
if checkGpu and (overrideGpuCheck == nil) then
local arch = os.getenv("EESSI_CPU_FAMILY") or ""
local cudaVersionFile = "/cvmfs/software.eessi.io/host_injections/nvidia/" .. arch .. "/latest/cuda_version.txt"
local cudaDriverFile = "/cvmfs/software.eessi.io/host_injections/nvidia/" .. arch .. "/latest/libcuda.so"
local cvmfs_repo = os.getenv("EESSI_CVMFS_REPO") or ""
local cudaVersionFile = cvmfs_repo .. "/host_injections/nvidia/" .. arch .. "/latest/cuda_version.txt"
local cudaDriverFile = cvmfs_repo .. "/host_injections/nvidia/" .. arch .. "/latest/libcuda.so"
local cudaDriverExists = isFile(cudaDriverFile)
local singularityCudaExists = isFile("/.singularity.d/libs/libcuda.so")
if not (cudaDriverExists or singularityCudaExists) then
Expand Down Expand Up @@ -194,10 +200,10 @@
-- Also: make it non-local, so it can be imported and extended by other lmodrc files if needed
function eessi_load_hook(t)
eessi_espresso_deprecated_message(t)
-- Only apply CUDA hooks if the loaded module is in the EESSI prefix
-- This avoids getting an Lmod Error when trying to load a CUDA module from a local software stack
-- Only apply CUDA and cu*-library hooks if the loaded module is in the EESSI prefix
-- This avoids getting an Lmod Error when trying to load a CUDA or cu* module from a local software stack
if from_eessi_prefix(t) then
eessi_cuda_enabled_load_hook(t)
eessi_cuda_and_libraries_enabled_load_hook(t)
end
end

Expand Down