maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/lld/test/MachO/compact-unwind.s

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

172 lines
7.0 KiB
ArmAsm
Raw Normal View History

[lld-macho] Add test for a variety of arm64 relocations Reviewed By: #lld-macho, smeenai Differential Revision: https://reviews.llvm.org/D97104
2021-02-28 01:30:17 +08:00
# REQUIRES: x86, aarch64
[lld-macho] Better deduplication of personality pointers {D95809} introduced a mechanism for synthetic symbol creation of personality pointers. When multiple section relocations referred to the same personality pointer, it would deduplicate them. However, it neglected to consider that we could have symbol relocations that also refer to the same personality pointer. This diff fixes it. In practice, this mix of relocations arises when there is a statically-linked personality routine that is referenced from multiple object files. Within the same object file, it will be referred to via section relocations, but (obviously) other object files will refer to it via symbol relocations. Failing to deduplicate these references resulted in us going over the 3-personality-pointer limit when linking some larger applications. Fixes llvm.org/PR48389. Reviewed By: #lld-macho, thakis, alexshap Differential Revision: https://reviews.llvm.org/D97245
2021-02-24 10:42:02 +08:00
# RUN: rm -rf %t; split-file %s %t
[lld-macho] Add test for a variety of arm64 relocations Reviewed By: #lld-macho, smeenai Differential Revision: https://reviews.llvm.org/D97104
2021-02-28 01:30:17 +08:00
# RUN: llvm-mc -filetype=obj -triple=x86_64-apple-darwin19.0.0 %t/my-personality.s -o %t/x86_64-my-personality.o
# RUN: llvm-mc -filetype=obj -triple=x86_64-apple-darwin19.0.0 %t/main.s -o %t/x86_64-main.o
[lld-macho] Have tests default to targeting macos 10.15 D101114 enforced proper version checks, which exposed a variety of version mismatch issues in our tests. We previously changed the test inputs to target 10.0, which was the simpler thing to do, but we should really just have our lit.local.cfg default to targeting 10.15, which is what is done here. We're not likely to ever have proper support for the older versions anyway, as that would require more work for unclear benefit; for instance, llvm-mc seems to generate a different compact unwind format for older macOS versions, which would cause our compact-unwind.s test to fail. Targeting 10.15 by default causes the following behavioral changes: * `__mh_execute_header` is now a section symbol instead of an absolute symbol * LC_BUILD_VERSION gets emitted instead of LC_VERSION_MIN_MACOSX. The former is 32 bytes in size whereas the latter is 16 bytes, so a bunch of hardcoded address offsets in our tests had to be updated. * >= 10.6 executables are PIE by default Note that this diff was stacked atop of a local revert of most of the test changes in rG8c17a875150f8e736e8f9061ddf084397f45f4c5, to make review easier. Reviewed By: #lld-macho, oontvoo Differential Revision: https://reviews.llvm.org/D101119
2021-04-23 07:37:47 +08:00
# RUN: %lld -arch x86_64 -pie -lSystem -lc++ %t/x86_64-my-personality.o %t/x86_64-main.o -o %t/x86_64-personality-first
[lld-macho] Add 32-bit compact unwind support This could probably have been part of D99633, but I split it up to make things a bit more reviewable. I also fixed some bugs in the implementation that were masked through integer underflows when operating in 64-bit mode. Reviewed By: #lld-macho, gkm Differential Revision: https://reviews.llvm.org/D99823
2021-04-16 09:14:33 +08:00
# RUN: llvm-objdump --macho --unwind-info --syms --indirect-symbols --rebase %t/x86_64-personality-first | FileCheck %s --check-prefixes=FIRST,CHECK -D#%x,BASE=0x100000000
[lld/mac] Don't crash on 32-bit output binaries when dead-stripping Fixes PR50974. Differential Revision: https://reviews.llvm.org/D105399
2021-07-04 09:57:43 +08:00
# RUN: %lld -dead_strip -arch x86_64 -pie -lSystem -lc++ %t/x86_64-main.o %t/x86_64-my-personality.o -o %t/x86_64-personality-second
[lld-macho] Add 32-bit compact unwind support This could probably have been part of D99633, but I split it up to make things a bit more reviewable. I also fixed some bugs in the implementation that were masked through integer underflows when operating in 64-bit mode. Reviewed By: #lld-macho, gkm Differential Revision: https://reviews.llvm.org/D99823
2021-04-16 09:14:33 +08:00
# RUN: llvm-objdump --macho --unwind-info --syms --indirect-symbols --rebase %t/x86_64-personality-second | FileCheck %s --check-prefixes=SECOND,CHECK -D#%x,BASE=0x100000000
[lld-macho] Add test for a variety of arm64 relocations Reviewed By: #lld-macho, smeenai Differential Revision: https://reviews.llvm.org/D97104
2021-02-28 01:30:17 +08:00
# RUN: llvm-mc -filetype=obj -triple=arm64-apple-darwin19.0.0 %t/my-personality.s -o %t/arm64-my-personality.o
# RUN: llvm-mc -filetype=obj -triple=arm64-apple-darwin19.0.0 %t/main.s -o %t/arm64-main.o
[lld-macho] Have tests default to targeting macos 10.15 D101114 enforced proper version checks, which exposed a variety of version mismatch issues in our tests. We previously changed the test inputs to target 10.0, which was the simpler thing to do, but we should really just have our lit.local.cfg default to targeting 10.15, which is what is done here. We're not likely to ever have proper support for the older versions anyway, as that would require more work for unclear benefit; for instance, llvm-mc seems to generate a different compact unwind format for older macOS versions, which would cause our compact-unwind.s test to fail. Targeting 10.15 by default causes the following behavioral changes: * `__mh_execute_header` is now a section symbol instead of an absolute symbol * LC_BUILD_VERSION gets emitted instead of LC_VERSION_MIN_MACOSX. The former is 32 bytes in size whereas the latter is 16 bytes, so a bunch of hardcoded address offsets in our tests had to be updated. * >= 10.6 executables are PIE by default Note that this diff was stacked atop of a local revert of most of the test changes in rG8c17a875150f8e736e8f9061ddf084397f45f4c5, to make review easier. Reviewed By: #lld-macho, oontvoo Differential Revision: https://reviews.llvm.org/D101119
2021-04-23 07:37:47 +08:00
# RUN: %lld -arch arm64 -pie -lSystem -lc++ %t/arm64-my-personality.o %t/arm64-main.o -o %t/arm64-personality-first
[lld-macho] Add 32-bit compact unwind support This could probably have been part of D99633, but I split it up to make things a bit more reviewable. I also fixed some bugs in the implementation that were masked through integer underflows when operating in 64-bit mode. Reviewed By: #lld-macho, gkm Differential Revision: https://reviews.llvm.org/D99823
2021-04-16 09:14:33 +08:00
# RUN: llvm-objdump --macho --unwind-info --syms --indirect-symbols --rebase %t/arm64-personality-first | FileCheck %s --check-prefixes=FIRST,CHECK -D#%x,BASE=0x100000000
[lld/mac] Don't crash on 32-bit output binaries when dead-stripping Fixes PR50974. Differential Revision: https://reviews.llvm.org/D105399
2021-07-04 09:57:43 +08:00
# RUN: %lld -dead_strip -arch arm64 -pie -lSystem -lc++ %t/arm64-main.o %t/arm64-my-personality.o -o %t/arm64-personality-second
[lld-macho] Add 32-bit compact unwind support This could probably have been part of D99633, but I split it up to make things a bit more reviewable. I also fixed some bugs in the implementation that were masked through integer underflows when operating in 64-bit mode. Reviewed By: #lld-macho, gkm Differential Revision: https://reviews.llvm.org/D99823
2021-04-16 09:14:33 +08:00
# RUN: llvm-objdump --macho --unwind-info --syms --indirect-symbols --rebase %t/arm64-personality-second | FileCheck %s --check-prefixes=SECOND,CHECK -D#%x,BASE=0x100000000
# RUN: llvm-mc -filetype=obj -triple=arm64_32-apple-watchos %t/my-personality.s -o %t/arm64-32-my-personality.o
# RUN: llvm-mc -filetype=obj -triple=arm64_32-apple-watchos %t/main.s -o %t/arm64-32-main.o
# RUN: %lld-watchos -pie -lSystem -lc++ %t/arm64-32-my-personality.o %t/arm64-32-main.o -o %t/arm64-32-personality-first
# RUN: llvm-objdump --macho --unwind-info --syms --indirect-symbols --rebase %t/arm64-32-personality-first | FileCheck %s --check-prefixes=FIRST,CHECK -D#%x,BASE=0x4000
[lld/mac] Don't crash on 32-bit output binaries when dead-stripping Fixes PR50974. Differential Revision: https://reviews.llvm.org/D105399
2021-07-04 09:57:43 +08:00
# RUN: %lld-watchos -dead_strip -pie -lSystem -lc++ %t/arm64-32-main.o %t/arm64-32-my-personality.o -o %t/arm64-32-personality-second
[lld-macho] Add 32-bit compact unwind support This could probably have been part of D99633, but I split it up to make things a bit more reviewable. I also fixed some bugs in the implementation that were masked through integer underflows when operating in 64-bit mode. Reviewed By: #lld-macho, gkm Differential Revision: https://reviews.llvm.org/D99823
2021-04-16 09:14:33 +08:00
# RUN: llvm-objdump --macho --unwind-info --syms --indirect-symbols --rebase %t/arm64-32-personality-second | FileCheck %s --check-prefixes=SECOND,CHECK -D#%x,BASE=0x4000
[lld-macho] Emit personalities in compact unwind Note that there is a triple indirection involved with personalities and compact unwind: 1. Two bits of each CU encoding are used as an offset into the personality array. 2. Each entry of the personality array is an offset from the image base. The resulting address (after adding the image base) should point within the GOT. 3. The corresponding GOT entry contains the actual pointer to the personality function. To further complicate things, when the personality function is in the object file (as opposed to a dylib), its references in `__compact_unwind` may refer to it via a section + offset relocation instead of a symbol relocation. Since our GOT implementation can only create entries for symbols, we have to create a synthetic symbol at the given section offset. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95809
2021-02-09 02:47:33 +08:00
[lld-macho] Better deduplication of personality pointers {D95809} introduced a mechanism for synthetic symbol creation of personality pointers. When multiple section relocations referred to the same personality pointer, it would deduplicate them. However, it neglected to consider that we could have symbol relocations that also refer to the same personality pointer. This diff fixes it. In practice, this mix of relocations arises when there is a statically-linked personality routine that is referenced from multiple object files. Within the same object file, it will be referred to via section relocations, but (obviously) other object files will refer to it via symbol relocations. Failing to deduplicate these references resulted in us going over the 3-personality-pointer limit when linking some larger applications. Fixes llvm.org/PR48389. Reviewed By: #lld-macho, thakis, alexshap Differential Revision: https://reviews.llvm.org/D97245
2021-02-24 10:42:02 +08:00
# FIRST: Indirect symbols for (__DATA_CONST,__got)
# FIRST-NEXT: address index name
# FIRST-DAG: 0x[[#%x,GXX_PERSONALITY:]] [[#]] ___gxx_personality_v0
# FIRST-DAG: 0x[[#%x,MY_PERSONALITY:]] LOCAL
# SECOND: Indirect symbols for (__DATA_CONST,__got)
# SECOND-NEXT: address index name
# SECOND-DAG: 0x[[#%x,GXX_PERSONALITY:]] [[#]] ___gxx_personality_v0
# SECOND-DAG: 0x[[#%x,MY_PERSONALITY:]] [[#]] _my_personality
[lld-macho] Emit personalities in compact unwind Note that there is a triple indirection involved with personalities and compact unwind: 1. Two bits of each CU encoding are used as an offset into the personality array. 2. Each entry of the personality array is an offset from the image base. The resulting address (after adding the image base) should point within the GOT. 3. The corresponding GOT entry contains the actual pointer to the personality function. To further complicate things, when the personality function is in the object file (as opposed to a dylib), its references in `__compact_unwind` may refer to it via a section + offset relocation instead of a symbol relocation. Since our GOT implementation can only create entries for symbols, we have to create a synthetic symbol at the given section offset. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95809
2021-02-09 02:47:33 +08:00
[lld-macho] Emit LSDA info in compact unwind The LSDA pointers are encoded as offsets from the image base, and arranged in one big contiguous array. Each second-level page records the offset within that LSDA array which corresponds to the LSDA for its first CU entry. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95810
2021-02-09 02:47:34 +08:00
# CHECK: SYMBOL TABLE:
# CHECK-DAG: [[#%x,MAIN:]] g F __TEXT,__text _main
[lld/mac] Add explicit "no unwind info" entries for functions without unwind info Fixes PR50529. With this, lld-linked Chromium base_unittests passes on arm macs. Surprisingly, no measurable impact on link time. Differential Revision: https://reviews.llvm.org/D104681
2021-06-22 10:29:11 +08:00
# CHECK-DAG: [[#%x,QUUX:]] g F __TEXT,__text _quux
[lld-macho] Better deduplication of personality pointers {D95809} introduced a mechanism for synthetic symbol creation of personality pointers. When multiple section relocations referred to the same personality pointer, it would deduplicate them. However, it neglected to consider that we could have symbol relocations that also refer to the same personality pointer. This diff fixes it. In practice, this mix of relocations arises when there is a statically-linked personality routine that is referenced from multiple object files. Within the same object file, it will be referred to via section relocations, but (obviously) other object files will refer to it via symbol relocations. Failing to deduplicate these references resulted in us going over the 3-personality-pointer limit when linking some larger applications. Fixes llvm.org/PR48389. Reviewed By: #lld-macho, thakis, alexshap Differential Revision: https://reviews.llvm.org/D97245
2021-02-24 10:42:02 +08:00
# CHECK-DAG: [[#%x,FOO:]] l F __TEXT,__text _foo
[lld/mac] Add explicit "no unwind info" entries for functions without unwind info Fixes PR50529. With this, lld-linked Chromium base_unittests passes on arm macs. Surprisingly, no measurable impact on link time. Differential Revision: https://reviews.llvm.org/D104681
2021-06-22 10:29:11 +08:00
# CHECK-DAG: [[#%x,BAZ:]] l F __TEXT,__text _baz
[lld-macho] Emit LSDA info in compact unwind The LSDA pointers are encoded as offsets from the image base, and arranged in one big contiguous array. Each second-level page records the offset within that LSDA array which corresponds to the LSDA for its first CU entry. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95810
2021-02-09 02:47:34 +08:00
# CHECK-DAG: [[#%x,EXCEPTION0:]] g O __TEXT,__gcc_except_tab _exception0
# CHECK-DAG: [[#%x,EXCEPTION1:]] g O __TEXT,__gcc_except_tab _exception1
[lld-macho] Emit personalities in compact unwind Note that there is a triple indirection involved with personalities and compact unwind: 1. Two bits of each CU encoding are used as an offset into the personality array. 2. Each entry of the personality array is an offset from the image base. The resulting address (after adding the image base) should point within the GOT. 3. The corresponding GOT entry contains the actual pointer to the personality function. To further complicate things, when the personality function is in the object file (as opposed to a dylib), its references in `__compact_unwind` may refer to it via a section + offset relocation instead of a symbol relocation. Since our GOT implementation can only create entries for symbols, we have to create a synthetic symbol at the given section offset. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95809
2021-02-09 02:47:33 +08:00
# CHECK: Contents of __unwind_info section:
# CHECK: Personality functions: (count = 2)
[lld-macho] Add 32-bit compact unwind support This could probably have been part of D99633, but I split it up to make things a bit more reviewable. I also fixed some bugs in the implementation that were masked through integer underflows when operating in 64-bit mode. Reviewed By: #lld-macho, gkm Differential Revision: https://reviews.llvm.org/D99823
2021-04-16 09:14:33 +08:00
# CHECK-DAG: personality[{{[0-9]+}}]: 0x{{0*}}[[#MY_PERSONALITY-BASE]]
# CHECK-DAG: personality[{{[0-9]+}}]: 0x{{0*}}[[#GXX_PERSONALITY-BASE]]
[lld/mac] Fix function offset on 1st-level unwind table sentinel Two bugs: 1. This tries to take the address of the last symbol plus the length of the last symbol. However, the sorted vector is cuPtrVector, not cuVector. Also, cuPtrVector has tombstone values removed and cuVector doesn't. If there was a stripped value at the end, the "last" element's value was UINT64_MAX, which meant the sentinel value was one less than the length of that "last" dead symbol. 2. We have to subtract in.header->addr. For 64-bit binaries that's (1 << 32) and functionAddress is 32-bit so this is a no-op, but for 32-bit binaries the sentinel's value was too large. I believe this has no effect in practice since the first-level binary search code in libunwind (in UnwindCursor.hpp) does: uint32_t low = 0; uint32_t high = sectionHeader.indexCount(); uint32_t last = high - 1; while (low < high) { uint32_t mid = (low + high) / 2; if ((mid == last) || (topIndex.functionOffset(mid + 1) > targetFunctionOffset)) { low = mid; break; } else { low = mid + 1; } So the address of the last entry in the first-level table isn't really checked -- except for the very end, but the check against `last` means we just run the loop once more than necessary. But it makes `unwinddump` output look less confusing, and it's what it looks was the intention here. (No test since I can't think of a way to make FileCheck check that one number is larger than another.) Differential Revision: https://reviews.llvm.org/D105404
2021-07-04 10:23:42 +08:00
# CHECK: Top level indices: (count = 2)
# CHECK-DAG: [0]: function offset={{.*}}, 2nd level page offset=0x[[#%x,PAGEOFF:]],
# CHECK-DAG: [1]: function offset={{.*}}, 2nd level page offset=0x00000000,
[lld-macho] Emit LSDA info in compact unwind The LSDA pointers are encoded as offsets from the image base, and arranged in one big contiguous array. Each second-level page records the offset within that LSDA array which corresponds to the LSDA for its first CU entry. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95810
2021-02-09 02:47:34 +08:00
# CHECK: LSDA descriptors:
[lld-macho] Add 32-bit compact unwind support This could probably have been part of D99633, but I split it up to make things a bit more reviewable. I also fixed some bugs in the implementation that were masked through integer underflows when operating in 64-bit mode. Reviewed By: #lld-macho, gkm Differential Revision: https://reviews.llvm.org/D99823
2021-04-16 09:14:33 +08:00
# CHECK-DAG: function offset=0x[[#%.8x,FOO-BASE]], LSDA offset=0x[[#%.8x,EXCEPTION0-BASE]]
# CHECK-DAG: function offset=0x[[#%.8x,MAIN-BASE]], LSDA offset=0x[[#%.8x,EXCEPTION1-BASE]]
[lld/mac] Add explicit "no unwind info" entries for functions without unwind info Fixes PR50529. With this, lld-linked Chromium base_unittests passes on arm macs. Surprisingly, no measurable impact on link time. Differential Revision: https://reviews.llvm.org/D104681
2021-06-22 10:29:11 +08:00
# CHECK: Second level indices:
[lld/mac] Fix function offset on 1st-level unwind table sentinel Two bugs: 1. This tries to take the address of the last symbol plus the length of the last symbol. However, the sorted vector is cuPtrVector, not cuVector. Also, cuPtrVector has tombstone values removed and cuVector doesn't. If there was a stripped value at the end, the "last" element's value was UINT64_MAX, which meant the sentinel value was one less than the length of that "last" dead symbol. 2. We have to subtract in.header->addr. For 64-bit binaries that's (1 << 32) and functionAddress is 32-bit so this is a no-op, but for 32-bit binaries the sentinel's value was too large. I believe this has no effect in practice since the first-level binary search code in libunwind (in UnwindCursor.hpp) does: uint32_t low = 0; uint32_t high = sectionHeader.indexCount(); uint32_t last = high - 1; while (low < high) { uint32_t mid = (low + high) / 2; if ((mid == last) || (topIndex.functionOffset(mid + 1) > targetFunctionOffset)) { low = mid; break; } else { low = mid + 1; } So the address of the last entry in the first-level table isn't really checked -- except for the very end, but the check against `last` means we just run the loop once more than necessary. But it makes `unwinddump` output look less confusing, and it's what it looks was the intention here. (No test since I can't think of a way to make FileCheck check that one number is larger than another.) Differential Revision: https://reviews.llvm.org/D105404
2021-07-04 10:23:42 +08:00
# CHECK-NEXT: Second level index[0]: offset in section=0x[[#%.8x,PAGEOFF]]
[lld/mac] Add explicit "no unwind info" entries for functions without unwind info Fixes PR50529. With this, lld-linked Chromium base_unittests passes on arm macs. Surprisingly, no measurable impact on link time. Differential Revision: https://reviews.llvm.org/D104681
2021-06-22 10:29:11 +08:00
# CHECK-DAG: function offset=0x[[#%.8x,MAIN-BASE]], encoding
# CHECK-DAG: function offset=0x[[#%.8x,FOO-BASE]], encoding
# CHECK-DAG: function offset=0x[[#%.8x,BAZ-BASE]], encoding
# CHECK-DAG: function offset=0x[[#%.8x,QUUX-BASE]], encoding{{.*}}=0x00000000
[lld-macho] Emit personalities in compact unwind Note that there is a triple indirection involved with personalities and compact unwind: 1. Two bits of each CU encoding are used as an offset into the personality array. 2. Each entry of the personality array is an offset from the image base. The resulting address (after adding the image base) should point within the GOT. 3. The corresponding GOT entry contains the actual pointer to the personality function. To further complicate things, when the personality function is in the object file (as opposed to a dylib), its references in `__compact_unwind` may refer to it via a section + offset relocation instead of a symbol relocation. Since our GOT implementation can only create entries for symbols, we have to create a synthetic symbol at the given section offset. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95809
2021-02-09 02:47:33 +08:00
## Check that we do not add rebase opcodes to the compact unwind section.
# CHECK: Rebase table:
# CHECK-NEXT: segment section address type
[lld-macho] Add 32-bit compact unwind support This could probably have been part of D99633, but I split it up to make things a bit more reviewable. I also fixed some bugs in the implementation that were masked through integer underflows when operating in 64-bit mode. Reviewed By: #lld-macho, gkm Differential Revision: https://reviews.llvm.org/D99823
2021-04-16 09:14:33 +08:00
# CHECK-NEXT: __DATA_CONST __got 0x{{[0-9A-F]*}} pointer
[lld-macho] Better deduplication of personality pointers {D95809} introduced a mechanism for synthetic symbol creation of personality pointers. When multiple section relocations referred to the same personality pointer, it would deduplicate them. However, it neglected to consider that we could have symbol relocations that also refer to the same personality pointer. This diff fixes it. In practice, this mix of relocations arises when there is a statically-linked personality routine that is referenced from multiple object files. Within the same object file, it will be referred to via section relocations, but (obviously) other object files will refer to it via symbol relocations. Failing to deduplicate these references resulted in us going over the 3-personality-pointer limit when linking some larger applications. Fixes llvm.org/PR48389. Reviewed By: #lld-macho, thakis, alexshap Differential Revision: https://reviews.llvm.org/D97245
2021-02-24 10:42:02 +08:00
# CHECK-NOT: __TEXT
[lld-macho] Emit personalities in compact unwind Note that there is a triple indirection involved with personalities and compact unwind: 1. Two bits of each CU encoding are used as an offset into the personality array. 2. Each entry of the personality array is an offset from the image base. The resulting address (after adding the image base) should point within the GOT. 3. The corresponding GOT entry contains the actual pointer to the personality function. To further complicate things, when the personality function is in the object file (as opposed to a dylib), its references in `__compact_unwind` may refer to it via a section + offset relocation instead of a symbol relocation. Since our GOT implementation can only create entries for symbols, we have to create a synthetic symbol at the given section offset. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95809
2021-02-09 02:47:33 +08:00
[lld/mac] Don't crash when dead-stripping removes all unwind info If the input has compact unwind info but all of it is removed after dead stripping, we would crash. Now we don't write any __unwind_info section at all, like ld64. This is a bit awkward to implement because we only know the final state of unwind info after UnwindInfoSectionImpl<Ptr>::finalize(), which is called after sections are added. So add a small amount of bookkeeping to relocateCompactUnwind() instead (which runs earlier) so that we can predict what finalize() will do before it runs. Fixes PR51010. Differential Revision: https://reviews.llvm.org/D105557
2021-07-07 23:28:27 +08:00
## Check that we don't create an __unwind_info section if no unwind info
## remains after dead-stripping.
# RUN: llvm-mc -filetype=obj -triple=x86_64-apple-darwin19.0.0 \
# RUN: %t/empty-after-dead-strip.s -o %t/x86_64-empty-after-dead-strip.o
# RUN: %lld -dylib -dead_strip -arch x86_64 -lSystem \
# RUN: %t/x86_64-empty-after-dead-strip.o -o %t/x86_64-empty-after-strip.dylib
# RUN: llvm-objdump --macho --unwind-info %t/x86_64-empty-after-strip.dylib | \
# RUN: FileCheck %s --check-prefixes=NOUNWIND --allow-empty
# NOUNWIND-NOT: Contents of __unwind_info section:
[lld-macho] Add test for a variety of arm64 relocations Reviewed By: #lld-macho, smeenai Differential Revision: https://reviews.llvm.org/D97104
2021-02-28 01:30:17 +08:00
#--- my-personality.s
[lld-macho] Better deduplication of personality pointers {D95809} introduced a mechanism for synthetic symbol creation of personality pointers. When multiple section relocations referred to the same personality pointer, it would deduplicate them. However, it neglected to consider that we could have symbol relocations that also refer to the same personality pointer. This diff fixes it. In practice, this mix of relocations arises when there is a statically-linked personality routine that is referenced from multiple object files. Within the same object file, it will be referred to via section relocations, but (obviously) other object files will refer to it via symbol relocations. Failing to deduplicate these references resulted in us going over the 3-personality-pointer limit when linking some larger applications. Fixes llvm.org/PR48389. Reviewed By: #lld-macho, thakis, alexshap Differential Revision: https://reviews.llvm.org/D97245
2021-02-24 10:42:02 +08:00
.globl _my_personality, _exception0
[lld-macho] Emit personalities in compact unwind Note that there is a triple indirection involved with personalities and compact unwind: 1. Two bits of each CU encoding are used as an offset into the personality array. 2. Each entry of the personality array is an offset from the image base. The resulting address (after adding the image base) should point within the GOT. 3. The corresponding GOT entry contains the actual pointer to the personality function. To further complicate things, when the personality function is in the object file (as opposed to a dylib), its references in `__compact_unwind` may refer to it via a section + offset relocation instead of a symbol relocation. Since our GOT implementation can only create entries for symbols, we have to create a synthetic symbol at the given section offset. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95809
2021-02-09 02:47:33 +08:00
.text
[lld-macho] Add test for a variety of arm64 relocations Reviewed By: #lld-macho, smeenai Differential Revision: https://reviews.llvm.org/D97104
2021-02-28 01:30:17 +08:00
.p2align 2
[lld-macho] Emit personalities in compact unwind Note that there is a triple indirection involved with personalities and compact unwind: 1. Two bits of each CU encoding are used as an offset into the personality array. 2. Each entry of the personality array is an offset from the image base. The resulting address (after adding the image base) should point within the GOT. 3. The corresponding GOT entry contains the actual pointer to the personality function. To further complicate things, when the personality function is in the object file (as opposed to a dylib), its references in `__compact_unwind` may refer to it via a section + offset relocation instead of a symbol relocation. Since our GOT implementation can only create entries for symbols, we have to create a synthetic symbol at the given section offset. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95809
2021-02-09 02:47:33 +08:00
_foo:
.cfi_startproc
[lld-macho] Better deduplication of personality pointers {D95809} introduced a mechanism for synthetic symbol creation of personality pointers. When multiple section relocations referred to the same personality pointer, it would deduplicate them. However, it neglected to consider that we could have symbol relocations that also refer to the same personality pointer. This diff fixes it. In practice, this mix of relocations arises when there is a statically-linked personality routine that is referenced from multiple object files. Within the same object file, it will be referred to via section relocations, but (obviously) other object files will refer to it via symbol relocations. Failing to deduplicate these references resulted in us going over the 3-personality-pointer limit when linking some larger applications. Fixes llvm.org/PR48389. Reviewed By: #lld-macho, thakis, alexshap Differential Revision: https://reviews.llvm.org/D97245
2021-02-24 10:42:02 +08:00
## This will generate a section relocation.
[lld-macho] Emit personalities in compact unwind Note that there is a triple indirection involved with personalities and compact unwind: 1. Two bits of each CU encoding are used as an offset into the personality array. 2. Each entry of the personality array is an offset from the image base. The resulting address (after adding the image base) should point within the GOT. 3. The corresponding GOT entry contains the actual pointer to the personality function. To further complicate things, when the personality function is in the object file (as opposed to a dylib), its references in `__compact_unwind` may refer to it via a section + offset relocation instead of a symbol relocation. Since our GOT implementation can only create entries for symbols, we have to create a synthetic symbol at the given section offset. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95809
2021-02-09 02:47:33 +08:00
.cfi_personality 155, _my_personality
[lld-macho] Emit LSDA info in compact unwind The LSDA pointers are encoded as offsets from the image base, and arranged in one big contiguous array. Each second-level page records the offset within that LSDA array which corresponds to the LSDA for its first CU entry. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95810
2021-02-09 02:47:34 +08:00
.cfi_lsda 16, _exception0
[lld-macho] Emit personalities in compact unwind Note that there is a triple indirection involved with personalities and compact unwind: 1. Two bits of each CU encoding are used as an offset into the personality array. 2. Each entry of the personality array is an offset from the image base. The resulting address (after adding the image base) should point within the GOT. 3. The corresponding GOT entry contains the actual pointer to the personality function. To further complicate things, when the personality function is in the object file (as opposed to a dylib), its references in `__compact_unwind` may refer to it via a section + offset relocation instead of a symbol relocation. Since our GOT implementation can only create entries for symbols, we have to create a synthetic symbol at the given section offset. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95809
2021-02-09 02:47:33 +08:00
.cfi_def_cfa_offset 16
[lld-macho] Add test for a variety of arm64 relocations Reviewed By: #lld-macho, smeenai Differential Revision: https://reviews.llvm.org/D97104
2021-02-28 01:30:17 +08:00
ret
[lld-macho] Emit personalities in compact unwind Note that there is a triple indirection involved with personalities and compact unwind: 1. Two bits of each CU encoding are used as an offset into the personality array. 2. Each entry of the personality array is an offset from the image base. The resulting address (after adding the image base) should point within the GOT. 3. The corresponding GOT entry contains the actual pointer to the personality function. To further complicate things, when the personality function is in the object file (as opposed to a dylib), its references in `__compact_unwind` may refer to it via a section + offset relocation instead of a symbol relocation. Since our GOT implementation can only create entries for symbols, we have to create a synthetic symbol at the given section offset. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95809
2021-02-09 02:47:33 +08:00
.cfi_endproc
[lld-macho] Add test for a variety of arm64 relocations Reviewed By: #lld-macho, smeenai Differential Revision: https://reviews.llvm.org/D97104
2021-02-28 01:30:17 +08:00
.p2align 2
[lld-macho] Emit personalities in compact unwind Note that there is a triple indirection involved with personalities and compact unwind: 1. Two bits of each CU encoding are used as an offset into the personality array. 2. Each entry of the personality array is an offset from the image base. The resulting address (after adding the image base) should point within the GOT. 3. The corresponding GOT entry contains the actual pointer to the personality function. To further complicate things, when the personality function is in the object file (as opposed to a dylib), its references in `__compact_unwind` may refer to it via a section + offset relocation instead of a symbol relocation. Since our GOT implementation can only create entries for symbols, we have to create a synthetic symbol at the given section offset. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95809
2021-02-09 02:47:33 +08:00
_bar:
.cfi_startproc
## Check that we dedup references to the same statically-linked personality.
.cfi_personality 155, _my_personality
[lld-macho] Emit LSDA info in compact unwind The LSDA pointers are encoded as offsets from the image base, and arranged in one big contiguous array. Each second-level page records the offset within that LSDA array which corresponds to the LSDA for its first CU entry. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95810
2021-02-09 02:47:34 +08:00
.cfi_lsda 16, _exception0
[lld-macho] Emit personalities in compact unwind Note that there is a triple indirection involved with personalities and compact unwind: 1. Two bits of each CU encoding are used as an offset into the personality array. 2. Each entry of the personality array is an offset from the image base. The resulting address (after adding the image base) should point within the GOT. 3. The corresponding GOT entry contains the actual pointer to the personality function. To further complicate things, when the personality function is in the object file (as opposed to a dylib), its references in `__compact_unwind` may refer to it via a section + offset relocation instead of a symbol relocation. Since our GOT implementation can only create entries for symbols, we have to create a synthetic symbol at the given section offset. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95809
2021-02-09 02:47:33 +08:00
.cfi_def_cfa_offset 16
[lld-macho] Add test for a variety of arm64 relocations Reviewed By: #lld-macho, smeenai Differential Revision: https://reviews.llvm.org/D97104
2021-02-28 01:30:17 +08:00
ret
[lld-macho] Emit personalities in compact unwind Note that there is a triple indirection involved with personalities and compact unwind: 1. Two bits of each CU encoding are used as an offset into the personality array. 2. Each entry of the personality array is an offset from the image base. The resulting address (after adding the image base) should point within the GOT. 3. The corresponding GOT entry contains the actual pointer to the personality function. To further complicate things, when the personality function is in the object file (as opposed to a dylib), its references in `__compact_unwind` may refer to it via a section + offset relocation instead of a symbol relocation. Since our GOT implementation can only create entries for symbols, we have to create a synthetic symbol at the given section offset. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95809
2021-02-09 02:47:33 +08:00
.cfi_endproc
[lld-macho] Add test for a variety of arm64 relocations Reviewed By: #lld-macho, smeenai Differential Revision: https://reviews.llvm.org/D97104
2021-02-28 01:30:17 +08:00
.p2align 2
[lld-macho] Better deduplication of personality pointers {D95809} introduced a mechanism for synthetic symbol creation of personality pointers. When multiple section relocations referred to the same personality pointer, it would deduplicate them. However, it neglected to consider that we could have symbol relocations that also refer to the same personality pointer. This diff fixes it. In practice, this mix of relocations arises when there is a statically-linked personality routine that is referenced from multiple object files. Within the same object file, it will be referred to via section relocations, but (obviously) other object files will refer to it via symbol relocations. Failing to deduplicate these references resulted in us going over the 3-personality-pointer limit when linking some larger applications. Fixes llvm.org/PR48389. Reviewed By: #lld-macho, thakis, alexshap Differential Revision: https://reviews.llvm.org/D97245
2021-02-24 10:42:02 +08:00
_my_personality:
[lld-macho] Add test for a variety of arm64 relocations Reviewed By: #lld-macho, smeenai Differential Revision: https://reviews.llvm.org/D97104
2021-02-28 01:30:17 +08:00
ret
[lld-macho] Better deduplication of personality pointers {D95809} introduced a mechanism for synthetic symbol creation of personality pointers. When multiple section relocations referred to the same personality pointer, it would deduplicate them. However, it neglected to consider that we could have symbol relocations that also refer to the same personality pointer. This diff fixes it. In practice, this mix of relocations arises when there is a statically-linked personality routine that is referenced from multiple object files. Within the same object file, it will be referred to via section relocations, but (obviously) other object files will refer to it via symbol relocations. Failing to deduplicate these references resulted in us going over the 3-personality-pointer limit when linking some larger applications. Fixes llvm.org/PR48389. Reviewed By: #lld-macho, thakis, alexshap Differential Revision: https://reviews.llvm.org/D97245
2021-02-24 10:42:02 +08:00
.section __TEXT,__gcc_except_tab
_exception0:
.space 1
#--- main.s
[lld/mac] Add explicit "no unwind info" entries for functions without unwind info Fixes PR50529. With this, lld-linked Chromium base_unittests passes on arm macs. Surprisingly, no measurable impact on link time. Differential Revision: https://reviews.llvm.org/D104681
2021-06-22 10:29:11 +08:00
.globl _main, _quux, _my_personality, _exception1
[lld-macho] Better deduplication of personality pointers {D95809} introduced a mechanism for synthetic symbol creation of personality pointers. When multiple section relocations referred to the same personality pointer, it would deduplicate them. However, it neglected to consider that we could have symbol relocations that also refer to the same personality pointer. This diff fixes it. In practice, this mix of relocations arises when there is a statically-linked personality routine that is referenced from multiple object files. Within the same object file, it will be referred to via section relocations, but (obviously) other object files will refer to it via symbol relocations. Failing to deduplicate these references resulted in us going over the 3-personality-pointer limit when linking some larger applications. Fixes llvm.org/PR48389. Reviewed By: #lld-macho, thakis, alexshap Differential Revision: https://reviews.llvm.org/D97245
2021-02-24 10:42:02 +08:00
.text
[lld-macho] Add test for a variety of arm64 relocations Reviewed By: #lld-macho, smeenai Differential Revision: https://reviews.llvm.org/D97104
2021-02-28 01:30:17 +08:00
.p2align 2
[lld-macho] Emit personalities in compact unwind Note that there is a triple indirection involved with personalities and compact unwind: 1. Two bits of each CU encoding are used as an offset into the personality array. 2. Each entry of the personality array is an offset from the image base. The resulting address (after adding the image base) should point within the GOT. 3. The corresponding GOT entry contains the actual pointer to the personality function. To further complicate things, when the personality function is in the object file (as opposed to a dylib), its references in `__compact_unwind` may refer to it via a section + offset relocation instead of a symbol relocation. Since our GOT implementation can only create entries for symbols, we have to create a synthetic symbol at the given section offset. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95809
2021-02-09 02:47:33 +08:00
_main:
.cfi_startproc
.cfi_personality 155, ___gxx_personality_v0
[lld-macho] Emit LSDA info in compact unwind The LSDA pointers are encoded as offsets from the image base, and arranged in one big contiguous array. Each second-level page records the offset within that LSDA array which corresponds to the LSDA for its first CU entry. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95810
2021-02-09 02:47:34 +08:00
.cfi_lsda 16, _exception1
[lld-macho] Emit personalities in compact unwind Note that there is a triple indirection involved with personalities and compact unwind: 1. Two bits of each CU encoding are used as an offset into the personality array. 2. Each entry of the personality array is an offset from the image base. The resulting address (after adding the image base) should point within the GOT. 3. The corresponding GOT entry contains the actual pointer to the personality function. To further complicate things, when the personality function is in the object file (as opposed to a dylib), its references in `__compact_unwind` may refer to it via a section + offset relocation instead of a symbol relocation. Since our GOT implementation can only create entries for symbols, we have to create a synthetic symbol at the given section offset. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95809
2021-02-09 02:47:33 +08:00
.cfi_def_cfa_offset 16
[lld-macho] Add test for a variety of arm64 relocations Reviewed By: #lld-macho, smeenai Differential Revision: https://reviews.llvm.org/D97104
2021-02-28 01:30:17 +08:00
ret
[lld-macho] Emit personalities in compact unwind Note that there is a triple indirection involved with personalities and compact unwind: 1. Two bits of each CU encoding are used as an offset into the personality array. 2. Each entry of the personality array is an offset from the image base. The resulting address (after adding the image base) should point within the GOT. 3. The corresponding GOT entry contains the actual pointer to the personality function. To further complicate things, when the personality function is in the object file (as opposed to a dylib), its references in `__compact_unwind` may refer to it via a section + offset relocation instead of a symbol relocation. Since our GOT implementation can only create entries for symbols, we have to create a synthetic symbol at the given section offset. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95809
2021-02-09 02:47:33 +08:00
.cfi_endproc
[lld/mac] Add explicit "no unwind info" entries for functions without unwind info Fixes PR50529. With this, lld-linked Chromium base_unittests passes on arm macs. Surprisingly, no measurable impact on link time. Differential Revision: https://reviews.llvm.org/D104681
2021-06-22 10:29:11 +08:00
## _quux has no unwind information.
## (In real life, it'd be part of a separate TU that was built with
## -fno-exceptions, while the previous and next TU might be Objective-C++
## which has unwind info for Objective-C).
.p2align 2
[lld/mac] Don't crash on 32-bit output binaries when dead-stripping Fixes PR50974. Differential Revision: https://reviews.llvm.org/D105399
2021-07-04 09:57:43 +08:00
.no_dead_strip _quux
[lld/mac] Add explicit "no unwind info" entries for functions without unwind info Fixes PR50529. With this, lld-linked Chromium base_unittests passes on arm macs. Surprisingly, no measurable impact on link time. Differential Revision: https://reviews.llvm.org/D104681
2021-06-22 10:29:11 +08:00
_quux:
ret
[lld/mac] Don't crash on absolute symbols in unwind info generation Fixes a regression from d6565a2dbcbe and PR50820.
2021-06-24 02:25:08 +08:00
.globl _abs
[lld/mac] Don't crash on 32-bit output binaries when dead-stripping Fixes PR50974. Differential Revision: https://reviews.llvm.org/D105399
2021-07-04 09:57:43 +08:00
.no_dead_strip _abs
[lld/mac] Don't crash on absolute symbols in unwind info generation Fixes a regression from d6565a2dbcbe and PR50820.
2021-06-24 02:25:08 +08:00
_abs = 4
[lld-macho] Add test for a variety of arm64 relocations Reviewed By: #lld-macho, smeenai Differential Revision: https://reviews.llvm.org/D97104
2021-02-28 01:30:17 +08:00
.p2align 2
[lld/mac] Don't crash on 32-bit output binaries when dead-stripping Fixes PR50974. Differential Revision: https://reviews.llvm.org/D105399
2021-07-04 09:57:43 +08:00
.no_dead_strip _baz
[lld-macho] Better deduplication of personality pointers {D95809} introduced a mechanism for synthetic symbol creation of personality pointers. When multiple section relocations referred to the same personality pointer, it would deduplicate them. However, it neglected to consider that we could have symbol relocations that also refer to the same personality pointer. This diff fixes it. In practice, this mix of relocations arises when there is a statically-linked personality routine that is referenced from multiple object files. Within the same object file, it will be referred to via section relocations, but (obviously) other object files will refer to it via symbol relocations. Failing to deduplicate these references resulted in us going over the 3-personality-pointer limit when linking some larger applications. Fixes llvm.org/PR48389. Reviewed By: #lld-macho, thakis, alexshap Differential Revision: https://reviews.llvm.org/D97245
2021-02-24 10:42:02 +08:00
_baz:
.cfi_startproc
## This will generate a symbol relocation. Check that we reuse the personality
## referenced by the section relocation in my_personality.s.
.cfi_personality 155, _my_personality
.cfi_lsda 16, _exception1
.cfi_def_cfa_offset 16
[lld-macho] Add test for a variety of arm64 relocations Reviewed By: #lld-macho, smeenai Differential Revision: https://reviews.llvm.org/D97104
2021-02-28 01:30:17 +08:00
ret
[lld-macho] Better deduplication of personality pointers {D95809} introduced a mechanism for synthetic symbol creation of personality pointers. When multiple section relocations referred to the same personality pointer, it would deduplicate them. However, it neglected to consider that we could have symbol relocations that also refer to the same personality pointer. This diff fixes it. In practice, this mix of relocations arises when there is a statically-linked personality routine that is referenced from multiple object files. Within the same object file, it will be referred to via section relocations, but (obviously) other object files will refer to it via symbol relocations. Failing to deduplicate these references resulted in us going over the 3-personality-pointer limit when linking some larger applications. Fixes llvm.org/PR48389. Reviewed By: #lld-macho, thakis, alexshap Differential Revision: https://reviews.llvm.org/D97245
2021-02-24 10:42:02 +08:00
.cfi_endproc
[lld/mac] Don't crash on 32-bit output binaries when dead-stripping Fixes PR50974. Differential Revision: https://reviews.llvm.org/D105399
2021-07-04 09:57:43 +08:00
.globl _stripped
_stripped:
.cfi_startproc
.cfi_personality 155, ___gxx_personality_v0
.cfi_lsda 16, _exception1
.cfi_def_cfa_offset 16
ret
.cfi_endproc
[lld-macho] Emit LSDA info in compact unwind The LSDA pointers are encoded as offsets from the image base, and arranged in one big contiguous array. Each second-level page records the offset within that LSDA array which corresponds to the LSDA for its first CU entry. Reviewed By: clayborg Differential Revision: https://reviews.llvm.org/D95810
2021-02-09 02:47:34 +08:00
.section __TEXT,__gcc_except_tab
_exception1:
.space 1
[lld/mac] Don't crash on 32-bit output binaries when dead-stripping Fixes PR50974. Differential Revision: https://reviews.llvm.org/D105399
2021-07-04 09:57:43 +08:00
.subsections_via_symbols
[lld/mac] Don't crash when dead-stripping removes all unwind info If the input has compact unwind info but all of it is removed after dead stripping, we would crash. Now we don't write any __unwind_info section at all, like ld64. This is a bit awkward to implement because we only know the final state of unwind info after UnwindInfoSectionImpl<Ptr>::finalize(), which is called after sections are added. So add a small amount of bookkeeping to relocateCompactUnwind() instead (which runs earlier) so that we can predict what finalize() will do before it runs. Fixes PR51010. Differential Revision: https://reviews.llvm.org/D105557
2021-07-07 23:28:27 +08:00
#--- empty-after-dead-strip.s
.text
## Local symbol with unwind info.
## The symbol is removed by -dead_strip.
_foo :
.cfi_startproc
.cfi_def_cfa_offset 16
retq
.cfi_endproc
.subsections_via_symbols