Skip to content

Reapply [Offload][lit] Link against SPIR-V DeviceRTL if present#180231

Merged
sarnex merged 1 commit intollvm:mainfrom
sarnex:litagainrtl
Feb 9, 2026
Merged

Reapply [Offload][lit] Link against SPIR-V DeviceRTL if present#180231
sarnex merged 1 commit intollvm:mainfrom
sarnex:litagainrtl

Conversation

@sarnex
Copy link
Copy Markdown
Member

@sarnex sarnex commented Feb 6, 2026 •
edited
Loading

I'll merge this at the same time as some llvm-zorg changes that start building the DeviceRTL.

We only see one new test passing because everything still fails because of the issue described in #178980

Once a fix for that issue is merged we will see many new passes.

@sarnex sarnex marked this pull request as ready for review February 6, 2026 18:53
@llvmbot llvmbot added the offload label Feb 6, 2026
@llvmbot
Copy link
Copy Markdown
Member

llvmbot commented Feb 6, 2026

@llvm/pr-subscribers-offload

Author: Nick Sarnie (sarnex)

Changes

I'll merge this at the same time as some llvm-zorg changes that start building the DeviceRTL.

We only see one new test passing because everything still fails because of the issue described in #178980

Once a fix for that issue is merged we will see many new passes.

Full diff: https://github.com/llvm/llvm-project/pull/180231.diff

2 Files Affected:

  • (modified) offload/test/lit.cfg (+3-2)
  • (modified) offload/test/mapping/firstprivate_aligned.cpp (-1)
diff --git a/offload/test/lit.cfg b/offload/test/lit.cfg
index 0d5a9c95c1d95..898258f3e3cd0 100644
--- a/offload/test/lit.cfg
+++ b/offload/test/lit.cfg
@@ -166,7 +166,8 @@ elif config.operating_system == 'Darwin':
     config.test_flags += " -Wl,-rpath," + config.library_dir
     config.test_flags += " -Wl,-rpath," + config.omp_host_rtl_directory
 else: # Unices
-    if config.libomptarget_current_target != "nvptx64-nvidia-cuda":
+    if config.libomptarget_current_target != "nvptx64-nvidia-cuda" and \
+       not config.libomptarget_current_target.startswith('spirv'):
         config.test_flags += " -nogpulib"
     config.test_flags += " -Wl,-rpath," + config.library_dir
     config.test_flags += " -Wl,-rpath," + config.omp_host_rtl_directory
@@ -214,7 +215,7 @@ def add_libraries(source):
     if "gpu" not in config.available_features:
         return source
     if "intelgpu" in config.available_features:
-        # There is no DeviceRTL for Intel yet and libc doesn't work.
+        # SPIR-V uses an out-of-tree linker and libc doesn't work.
         return source
     if config.libomptarget_has_libc:
         return source + " -Xoffload-linker -lc " + \
diff --git a/offload/test/mapping/firstprivate_aligned.cpp b/offload/test/mapping/firstprivate_aligned.cpp
index bf09eb0b0f589..ae6be0f0c07f4 100644
--- a/offload/test/mapping/firstprivate_aligned.cpp
+++ b/offload/test/mapping/firstprivate_aligned.cpp
@@ -1,5 +1,4 @@
 // RUN: %libomptarget-compilexx-generic -O3 && %libomptarget-run-generic
-// XFAIL: intelgpu
 
 #include <stdio.h>

@sarnex sarnex requested a review from jhuber6 February 6, 2026 18:53
@sarnex sarnex merged commit 7fcc12c into llvm:main Feb 9, 2026
14 checks passed
@llvm-ci
Copy link
Copy Markdown

llvm-ci commented Feb 9, 2026

LLVM Buildbot has detected a new failure on builder intel-sycl-gpu running on intel-sycl-gpu-01 while building offload at step 6 "test-build-unified-tree-check-all".

Full details are available at: https://lab.llvm.org/buildbot/#/builders/225/builds/2309

Here is the relevant piece of the build log for the reference 
Step 6 (test-build-unified-tree-check-all) failure: test (failure)
******************** TEST 'libomptarget :: spirv64-intel :: mapping/declare_mapper_api.cpp' FAILED ********************
Exit Code: 1

Command Output (stdout):
--
# RUN: at line 1
/home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/./bin/clang++ -fopenmp    -I /home/test-user/llvm-buildbot-worker/intel-sycl-gpu/llvm-project/offload/test -I /home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/runtimes/runtimes-bins/openmp/runtime/src -L /home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/runtimes/runtimes-bins/offload -L /home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/./lib -L /home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/./lib -L /home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/runtimes/runtimes-bins/openmp/runtime/src  -Wl,-rpath,/home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/runtimes/runtimes-bins/offload -Wl,-rpath,/home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/runtimes/runtimes-bins/openmp/runtime/src -Wl,-rpath,/home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/./lib -Wl,-rpath,/home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/./lib  -fopenmp-targets=spirv64-intel /home/test-user/llvm-buildbot-worker/intel-sycl-gpu/llvm-project/offload/test/mapping/declare_mapper_api.cpp -o /home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/runtimes/runtimes-bins/offload/test/spirv64-intel/mapping/Output/declare_mapper_api.cpp.tmp && /home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/runtimes/runtimes-bins/offload/test/spirv64-intel/mapping/Output/declare_mapper_api.cpp.tmp | /home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/./bin/FileCheck /home/test-user/llvm-buildbot-worker/intel-sycl-gpu/llvm-project/offload/test/mapping/declare_mapper_api.cpp
# executed command: /home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/./bin/clang++ -fopenmp -I /home/test-user/llvm-buildbot-worker/intel-sycl-gpu/llvm-project/offload/test -I /home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/runtimes/runtimes-bins/openmp/runtime/src -L /home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/runtimes/runtimes-bins/offload -L /home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/./lib -L /home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/./lib -L /home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/runtimes/runtimes-bins/openmp/runtime/src -Wl,-rpath,/home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/runtimes/runtimes-bins/offload -Wl,-rpath,/home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/runtimes/runtimes-bins/openmp/runtime/src -Wl,-rpath,/home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/./lib -Wl,-rpath,/home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/./lib -fopenmp-targets=spirv64-intel /home/test-user/llvm-buildbot-worker/intel-sycl-gpu/llvm-project/offload/test/mapping/declare_mapper_api.cpp -o /home/test-user/llvm-buildbot-worker/intel-sycl-gpu/build/runtimes/runtimes-bins/offload/test/spirv64-intel/mapping/Output/declare_mapper_api.cpp.tmp
# .---command stderr------------
# | clang++: error: no library 'libomptarget-spirv.bc' found in the default clang lib directory or in LIBRARY_PATH; use '--libomptarget-spirv-bc-path' to specify spirv bitcode library
# `-----------------------------
# error: command failed with exit status: 1

--

********************

@sarnex
Copy link
Copy Markdown
Member Author

sarnex commented Feb 9, 2026

Can ignore, just gotta wait for llvm-zorg change which i already merged to take effect

@sarnex
Copy link
Copy Markdown
Member Author

sarnex commented Feb 10, 2026

Reverting again, the buildbot infra seems to have not updated yet

@sarnex sarnex mentioned this pull request Feb 10, 2026
Merged
sarnex added a commit that referenced this pull request Feb 10, 2026
@sarnex
Revert "Reapply [Offload][lit] Link against SPIR-V DeviceRTL if prese…
6c0ff8d 
…nt" (#180743)

Reverts #180231
llvm-sync Bot pushed a commit to arm/arm-toolchain that referenced this pull request Feb 10, 2026
@sarnex @github-actions
Automerge: Revert "Reapply [Offload][lit] Link against SPIR-V DeviceR…
2181d3a 
…TL if present" (#180743)

Reverts llvm/llvm-project#180231
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@jhuber6 jhuber6 jhuber6 approved these changes

Assignees

No one assigned

Labels

offload

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

4 participants

@sarnex @llvmbot @llvm-ci @jhuber6