[PATCH] D38910: [AMDGPU/LLD] Add RELATIVE64 relocation record.
Tony Tye via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Fri Oct 13 16:32:07 PDT 2017
t-tye created this revision.
Herald added subscribers: Anastasia, tpr, dstuttard, yaxunl, nhaehnle, wdng, emaste.
Add base relative relocation record that can be used for the following case (OpenCL example):
static __global int Var = 0;
__global int* Ptr[] = {&Var};
...
In this case Var is a non premptable symbol and so its address can be used as the value of Ptr, with a base relative relocation that will add the delta between the ELF address and the actual load address. Such relocations do not require a symbol.
This also fixes LLD which was incorrectly generating a PCREL64 for this case.
https://reviews.llvm.org/D38910
Files:
ELF/Arch/AMDGPU.cpp
test/ELF/amdgpu-relocs.s
Index: test/ELF/amdgpu-relocs.s
===================================================================
--- test/ELF/amdgpu-relocs.s
+++ test/ELF/amdgpu-relocs.s
@@ -65,10 +65,23 @@
.quad temp
.size ptr, 8
+# R_AMDGPU_RELATIVE64:
+ .type temp2, @object
+ .local temp2
+ .size temp2, 4
+
+ .type ptr2, @object
+ .globl ptr2
+ .size ptr2, 8
+ .p2align 3
+ptr2:
+ .quad temp2
+
# The relocation for local_var{0, 1, 2} and var should be resolved by the
# linker.
# CHECK: Relocations [
# CHECK: .rela.dyn {
+# CHECK-NEXT: R_AMDGPU_RELATIVE64 - 0x0
# CHECK-NEXT: R_AMDGPU_ABS64 common_var0 0x0
# CHECK-NEXT: R_AMDGPU_ABS64 common_var1 0x0
# CHECK-NEXT: R_AMDGPU_ABS64 common_var2 0x0
Index: ELF/Arch/AMDGPU.cpp
===================================================================
--- ELF/Arch/AMDGPU.cpp
+++ ELF/Arch/AMDGPU.cpp
@@ -32,7 +32,7 @@
} // namespace
AMDGPU::AMDGPU() {
- RelativeRel = R_AMDGPU_REL64;
+ RelativeRel = R_AMDGPU_RELATIVE64;
GotRel = R_AMDGPU_ABS64;
GotEntrySize = 8;
}
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D38910.118987.patch
Type: text/x-patch
Size: 1020 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20171013/2ff244f7/attachment.bin>
More information about the llvm-commits
mailing list