[LinkerWrapper] Allow 'all' as a generic bundled architecture

[jhuber6]

Summary:
Currently, the linker wrapper sorts input files into different link
jobs according to their architectures. Here we assume each architecture
is a unique and incompatible link job unless they are specifically
marked compatible. This patch simply adds an all target to represent
an architecture that should be linked against every single other
architecture.

This will be useful for modelling generic IR such as the ROCm device
libraries or the NVPTX libdevice.

[llvmbot]

@llvm/pr-subscribers-llvm-binary-utilities
@llvm/pr-subscribers-clang

@llvm/pr-subscribers-clang-driver

Author: Joseph Huber (jhuber6)

Changes

Summary:
Currently, the linker wrapper sourts input files into different link
jobs according to their architectures. Here we assume each architecture
is a unique and incompatible link job unless they are specifically
marked compatible. This patch simply adds an all target to represent
an architecture that should be linked against every single other
architecture.

This will be useful for modelling generic IR such as the ROCm device
libraries or the NVPTX libdevice.

---

Full diff: https://github.com/llvm/llvm-project/pull/81193.diff

2 Files Affected:

• (modified) clang/test/Driver/linker-wrapper.c (+16)
• (modified) llvm/lib/Object/OffloadBinary.cpp (+4)

diff --git a/clang/test/Driver/linker-wrapper.c b/clang/test/Driver/linker-wrapper.c
index 010001b83d7c2d..647629a5969bdc 100644
--- a/clang/test/Driver/linker-wrapper.c
+++ b/clang/test/Driver/linker-wrapper.c
@@ -172,6 +172,22 @@ __attribute__((visibility("protected"), used)) int x;
 // AMD-TARGET-ID: clang{{.*}} -o {{.*}}.img --target=amdgcn-amd-amdhsa -mcpu=gfx90a:xnack+ -O2 -Wl,--no-undefined {{.*}}.o {{.*}}.o
 // AMD-TARGET-ID: clang{{.*}} -o {{.*}}.img --target=amdgcn-amd-amdhsa -mcpu=gfx90a:xnack- -O2 -Wl,--no-undefined {{.*}}.o {{.*}}.o
 
+// RUN: clang-offload-packager -o %t-lib.out \
+// RUN:   --image=file=%t.elf.o,kind=openmp,triple=amdgcn-amd-amdhsa,arch=all
+// RUN: %clang -cc1 %s -triple x86_64-unknown-linux-gnu -emit-obj -o %t.o -fembed-offload-object=%t-lib.out
+// RUN: llvm-ar rcs %t.a %t.o
+// RUN: clang-offload-packager -o %t1.out \
+// RUN:   --image=file=%t.elf.o,kind=openmp,triple=amdgcn-amd-amdhsa,arch=gfx90a
+// RUN: %clang -cc1 %s -triple x86_64-unknown-linux-gnu -emit-obj -o %t1.o -fembed-offload-object=%t1.out
+// RUN: clang-offload-packager -o %t2.out \
+// RUN:   --image=file=%t.elf.o,kind=openmp,triple=amdgcn-amd-amdhsa,arch=gfx908
+// RUN: %clang -cc1 %s -triple x86_64-unknown-linux-gnu -emit-obj -o %t2.o -fembed-offload-object=%t2.out
+// RUN: clang-linker-wrapper --host-triple=x86_64-unknown-linux-gnu --dry-run \
+// RUN:   --linker-path=/usr/bin/ld -- %t1.o %t2.o %t.a -o a.out 2>&1 | FileCheck %s --check-prefix=ARCH-ALL
+
+// ARCH-ALL: clang{{.*}} -o {{.*}}.img --target=amdgcn-amd-amdhsa -mcpu=gfx908 -O2 -Wl,--no-undefined {{.*}}.o {{.*}}.o
+// ARCH-ALL: clang{{.*}} -o {{.*}}.img --target=amdgcn-amd-amdhsa -mcpu=gfx90a -O2 -Wl,--no-undefined {{.*}}.o {{.*}}.o
+
 // RUN: clang-offload-packager -o %t.out \
 // RUN:   --image=file=%t.elf.o,kind=openmp,triple=x86_64-unknown-linux-gnu \
 // RUN:   --image=file=%t.elf.o,kind=openmp,triple=x86_64-unknown-linux-gnu
diff --git a/llvm/lib/Object/OffloadBinary.cpp b/llvm/lib/Object/OffloadBinary.cpp
index 22d604b125c583..58b9b39e0d2721 100644
--- a/llvm/lib/Object/OffloadBinary.cpp
+++ b/llvm/lib/Object/OffloadBinary.cpp
@@ -355,6 +355,10 @@ bool object::areTargetsCompatible(const OffloadFile::TargetID &LHS,
   if (LHS.first != RHS.first)
     return false;
 
+  // If the architecture is "all" we assume it is always compatible.
+  if (LHS.second.equals("all") || RHS.second.equals("all"))
+    return true;
+
   // Only The AMDGPU target requires additional checks.
   llvm::Triple T(LHS.first);
   if (!T.isAMDGPU())