Merge branch 'main' into extrae_5.0.0

Author: bedroge

EESSI-extend-easybuild.eb

@@ -93,27 +93,31 @@ working_dir = os.getenv("WORKING_DIR") or pathJoin("/tmp", os.getenv("USER"))
 -- Gather the EPREFIX to use as a sysroot
 sysroot = os.getenv("EESSI_EPREFIX")
 
--- Check if we have GPU capabilities and configure CUDA compute capabilities
+-- Check if we have GPU capabilities and configure AMD/CUDA compute capabilities
 eessi_accelerator_target = os.getenv("EESSI_ACCELERATOR_TARGET")
 if (eessi_accelerator_target ~= nil) then
   cuda_compute_capability = string.match(eessi_accelerator_target, "^accel/nvidia/cc([0-9]+)$")
+  amd_compute_capability = string.match(eessi_accelerator_target, "^accel/amd/(gfx[0-9a-f]+)$")
+
   if (cuda_compute_capability ~= nil) then
     -- The last digit should be the minor version, insert a dot in the one-but-last position
     major_version = cuda_compute_capability:sub(1, #cuda_compute_capability - 1)
     minor_version = cuda_compute_capability:sub(#cuda_compute_capability)
     easybuild_cuda_compute_capabilities = string.format("%s.%s", major_version, minor_version)
+
+    -- If architectures are 9.0, 10.0 or 12.0, enable architecture or family-specific optimizations
+    if easybuild_cuda_compute_capabilities == '9.0' then
+      easybuild_cuda_compute_capabilities = '9.0a'
+    elseif easybuild_cuda_compute_capabilities == '10.0' then
+      easybuild_cuda_compute_capabilities = '10.0f'
+    elseif easybuild_cuda_compute_capabilities == '12.0' then
+      easybuild_cuda_compute_capabilities = '12.0f'
+    end
+  elseif (amd_compute_capability ~= nil) then
+    easybuild_amdgcn_compute_capabilities = amd_compute_capability
   else
     LmodError("Incorrect value for $EESSI_ACCELERATOR_TARGET: " .. eessi_accelerator_target)
   end
-
-  -- If architectures are 9.0, 10.0 or 12.0, enable architecture or family-specific optimizations
-  if easybuild_cuda_compute_capabilities == '9.0' then
-    easybuild_cuda_compute_capabilities = '9.0a'
-  elseif easybuild_cuda_compute_capabilities == '10.0' then
-    easybuild_cuda_compute_capabilities = '10.0f'
-  elseif easybuild_cuda_compute_capabilities == '12.0' then
-    easybuild_cuda_compute_capabilities = '12.0f'
-  end
 end
 
 -- Some environment variables affect behaviour, let's gather them once
@@ -130,7 +134,7 @@ if eessi_cvmfs_install then
   end
   easybuild_installpath = os.getenv("EESSI_SOFTWARE_PATH")
   -- enforce accelerator subdirectory usage for CVMFS installs (only if an accelerator install is requested)
-  if (eessi_accelerator_target ~= nil) and (cuda_compute_capability ~= nil) and (os.getenv("EESSI_ACCELERATOR_INSTALL") ~= nil) then
+  if (eessi_accelerator_target ~= nil) and (cuda_compute_capability ~= nil or amd_compute_capability ~= nil) and (os.getenv("EESSI_ACCELERATOR_INSTALL") ~= nil) then
       easybuild_installpath = pathJoin(easybuild_installpath, eessi_accelerator_target)
   end
 elseif eessi_site_install then
@@ -202,6 +206,11 @@ if (easybuild_cuda_compute_capabilities ~= nil) then
   setenv ("EASYBUILD_CUDA_COMPUTE_CAPABILITIES", easybuild_cuda_compute_capabilities)
 end
 
+-- Set environment variables if building for AMDGCN compute capabilities
+if (easybuild_amdgcn_compute_capabilities ~= nil) then
+  setenv ("EASYBUILD_AMDGCN_CAPABILITIES", easybuild_amdgcn_compute_capabilities)
+end
+
 -- Set all related environment variables if we have project or user installations (including extending MODULEPATH)
 if (user_modulepath ~= nil) then
   -- Use a more restrictive umask for this case

eb_hooks.py

@@ -717,13 +717,18 @@ def pre_fetch_hook(self, *args, **kwargs):
 def pre_fetch_hook_check_installation_path(self, *args, **kwargs):
     # When we know the CUDA status, we will need to verify the installation path
     # if we are doing an EESSI or host_injections installation
-    accelerator_deps = ['CUDA']
+    accelerator_deps = ['CUDA', 'ROCm-LLVM']
+    accelerator_toolchains = ['rocm-compilers', 'rompi', 'rfbf', 'rfoss']
     strict_eessi_installation = (
         bool(re.search(EESSI_INSTALLATION_REGEX, self.installdir)) or
         self.installdir.startswith(HOST_INJECTIONS_LOCATION))
     if strict_eessi_installation and not os.getenv("EESSI_OVERRIDE_STRICT_INSTALLPATH_CHECK"):
         dependency_names = self.cfg.dependency_names()
-        if self.cfg.name in accelerator_deps or any(dep in dependency_names for dep in accelerator_deps):
+        if (
+            self.cfg.name in accelerator_deps
+            or any(dep in dependency_names for dep in accelerator_deps)
+            or self.toolchain.name in accelerator_toolchains
+        ):
             # Make sure the path is an accelerator location
             if "/accel/" not in self.installdir:
                 raise EasyBuildError(

init/eessi_archdetect.sh

@@ -179,13 +179,14 @@ accelpath() {
     # If EESSI_ACCELERATOR_TARGET_OVERRIDE is set, use it
     log "DEBUG" "accelpath: Override variable set as '$EESSI_ACCELERATOR_TARGET_OVERRIDE' "
     if [ ! -z $EESSI_ACCELERATOR_TARGET_OVERRIDE ]; then
-        if [[ "$EESSI_ACCELERATOR_TARGET_OVERRIDE" =~ ^accel/nvidia/cc[0-9]+$ ]]; then
-            echo ${EESSI_ACCELERATOR_TARGET_OVERRIDE}
+        # Regex that allows both NVIDIA and AMD overrides
+        if [[ "$EESSI_ACCELERATOR_TARGET_OVERRIDE" =~ ^accel/(nvidia/cc[0-9]+|amd/gfx[0-9a-f]+)$ ]]; then
+            echo "$EESSI_ACCELERATOR_TARGET_OVERRIDE"
             return 0
         else
-            log "ERROR" "Value of \$EESSI_ACCELERATOR_TARGET_OVERRIDE should match 'accel/nvidia/cc[0-9]+', but it does not: '$EESSI_ACCELERATOR_TARGET_OVERRIDE'"
+            log "ERROR" "Value of \$EESSI_ACCELERATOR_TARGET_OVERRIDE should match 'accel/nvidia/cc[0-9]+' or 'accel/amd/gfx[0-9a-f]+', but it does not: '$EESSI_ACCELERATOR_TARGET_OVERRIDE'"
+            return 1
         fi
-        return 0
     fi
 
     # check for NVIDIA GPUs via nvidia-smi command

scripts/gpu_support/nvidia/install_cuda_and_libraries.sh

@@ -93,7 +93,7 @@ echo "Created temporary directory '${tmpdir}'"
 SAVE_MODULEPATH=${MODULEPATH}
 
 for EASYSTACK_FILE in ${TOPDIR}/easystacks/eessi-*CUDA*.yml; do
-    echo -e "Processing easystack file ${easystack_file}...\n\n"
+    echo -e "Processing easystack file ${EASYSTACK_FILE}...\n\n"
 
     # determine version of EasyBuild module to load based on EasyBuild version included in name of easystack file
     eb_version=$(echo ${EASYSTACK_FILE} | sed 's/.*eb-\([0-9.]*\).*.yml/\1/g')
@@ -104,7 +104,7 @@ for EASYSTACK_FILE in ${TOPDIR}/easystacks/eessi-*CUDA*.yml; do
     if [[ $? -eq 0 ]]; then
         echo_green ">> Found an EasyBuild/${eb_version} module"
     else
-        echo_yellow ">> No EasyBuild/${eb_version} module found: skipping step to install easystack file ${easystack_file} (see output in ${module_avail_out})"
+        echo_yellow ">> No EasyBuild/${eb_version} module found: skipping step to install easystack file ${EASYSTACK_FILE} (see output in ${module_avail_out})"
         continue
     fi
     # Safer to unload EESSI-extend before loading an EasyBuild version, in case unload behavior ever becomes dependent on EasyBuild version