This is an archive of the discontinued LLVM Phabricator instance.

Differential D128025

[lld-macho][nfc] Tests for -force_load + regular archive load combinations
ClosedPublic

Authored by int3 on Jun 16 2022, 8:22 PM.

Details

Reviewers
thakis
Group Reviewers
Restricted Project
Commits
rG8eeede973c69: [lld-macho][nfc] Tests for -force_load + regular archive load combinations
Summary

I realized we'd forgotten to cover this case (though our existing
behavior is indeed correct / matches ld64's).

Diff Detail

Event Timeline

int3 created this revision.Jun 16 2022, 8:22 PM
Herald added projects: Restricted Project, Restricted Project. · View Herald TranscriptJun 16 2022, 8:22 PM
int3 requested review of this revision.Jun 16 2022, 8:22 PM
Herald added a project: Restricted Project. · View Herald TranscriptJun 16 2022, 8:22 PM
Herald added a subscriber: llvm-commits. · View Herald Transcript
Harbormaster completed remote builds in B170424: Diff 437782.Jun 16 2022, 8:33 PM
thakis accepted this revision.Jun 16 2022, 8:43 PM
thakis added a subscriber: thakis.

Seems possibly worth emitting a warning for, seems like surprising behavior...

This revision is now accepted and ready to land.Jun 16 2022, 8:43 PM
int3 added a comment.Jun 16 2022, 8:47 PM

Maybe after D127941: [lld-macho] Fix an issue where Objective-C symbols where not force-loaded due to LC_LINKER_OPTION lands

Closed by commit rG8eeede973c69: [lld-macho][nfc] Tests for -force_load + regular archive load combinations (authored by int3). · Explain WhyJun 16 2022, 8:50 PM
This revision was automatically updated to reflect the committed changes.
int3 added a commit: rG8eeede973c69: [lld-macho][nfc] Tests for -force_load + regular archive load combinations.
PRESIDENT810 mentioned this in D129556: [lld-macho] Fix loading same libraries from both LC_LINKER_OPTION and command line.Jul 12 2022, 3:27 AM
int3 added inline comments.Jul 12 2022, 2:57 PM
lld/test/MachO/force-load.s
25

... whoops I forgot to change this line after copypasting. Fixed in 61ace8f78b1c

int3 mentioned this in rGdd5635541cd7: [lld-macho] Fix loading same libraries from both LC_LINKER_OPTION and command….Jul 19 2022, 2:46 PM

Revision Contents

PathSize
lld/
test/
MachO/
10 lines

Diff 437782

lld/test/MachO/force-load.s

Show All 9 Lines
# RUN: %lld -lSystem -force_load %t/foo.a %t/foo.o %t/test.o -o %t/test-force-load-first # RUN: %lld -lSystem -force_load %t/foo.a %t/foo.o %t/test.o -o %t/test-force-load-first
# FORCE-LOAD-FIRST: __TEXT,archive _foo # FORCE-LOAD-FIRST: __TEXT,archive _foo
# RUN: llvm-objdump --syms %t/test-force-load-first | FileCheck %s --check-prefix=FORCE-LOAD-FIRST # RUN: llvm-objdump --syms %t/test-force-load-first | FileCheck %s --check-prefix=FORCE-LOAD-FIRST
# RUN: %lld %t/foo.o -lSystem -force_load %t/foo.a %t/test.o -o %t/test-force-load-second # RUN: %lld %t/foo.o -lSystem -force_load %t/foo.a %t/test.o -o %t/test-force-load-second
# RUN: llvm-objdump --syms %t/test-force-load-second | FileCheck %s --check-prefix=FORCE-LOAD-SECOND # RUN: llvm-objdump --syms %t/test-force-load-second | FileCheck %s --check-prefix=FORCE-LOAD-SECOND
# FORCE-LOAD-SECOND: __TEXT,obj _foo # FORCE-LOAD-SECOND: __TEXT,obj _foo
## If an archive has already been loaded w/o -force_load earlier in the command
## line, a later -force_load argument will not have an effect.
# RUN: %lld -lSystem %t/foo.a -force_load %t/foo.a %t/test.o -o %t/test-regular-then-force
# RUN: llvm-objdump --syms %t/test-regular-then-force | FileCheck %s --check-prefix=REGULAR-THEN-FORCE
# REGULAR-THEN-FORCE-NOT: _foo
## If the -force_load comes first, then the second load will just be a no-op.
# RUN: %lld -lSystem -force_load %t/foo.a %t/foo.a %t/test.o -o %t/test-force-then-regular
# RUN: llvm-objdump --syms %t/test-regular-then-force | FileCheck %s --check-prefix=REGULAR-THEN-FORCE
int3AuthorUnsubmitted
Done
ReplyInline Actions

... whoops I forgot to change this line after copypasting. Fixed in 61ace8f78b1c

int3: ... whoops I forgot to change this line after copypasting. Fixed in 61ace8f78b1c
# FORCE-THEN-REGULAR: _foo
## Force-loading the same path twice is fine ## Force-loading the same path twice is fine
# RUN: %lld -lSystem %t/foo.o -force_load %t/foo.a -force_load %t/foo.a %t/test.o -o /dev/null # RUN: %lld -lSystem %t/foo.o -force_load %t/foo.a -force_load %t/foo.a %t/test.o -o /dev/null
## Note that we do not call realpath() before dedup'ing the force-load ## Note that we do not call realpath() before dedup'ing the force-load
## arguments, so this is an error. ## arguments, so this is an error.
# RUN: cd %t; not %lld -lSystem %t/foo.o -force_load %t/foo.a -force_load foo.a \ # RUN: cd %t; not %lld -lSystem %t/foo.o -force_load %t/foo.a -force_load foo.a \
# RUN: %t/test.o -o /dev/null 2>&1 | FileCheck %s --check-prefix=DUP # RUN: %t/test.o -o /dev/null 2>&1 | FileCheck %s --check-prefix=DUP
Show All 31 Lines