For debugging dylib loading, it's useful to have some insight into what
the linker is doing.
ld64 has the undocumented RC_TRACE_DYLIB_SEARCHING env var
for this printing dylib search candidates.
This adds a flag --print-dylib-search to make lld print the seame information.
It's useful for users, but also for writing tests. The output is formatted
slightly differently than ld64, but we still support RC_TRACE_DYLIB_SEARCHING
to offer at least a compatible way to trigger this.
ld64 has both `-print_statistics` and `-trace_symbol_output` to enable
diagnostics output. I went with "print" since that seems like a more
straightforward name.
Differential Revision: https://reviews.llvm.org/D103985
It causes libraries whose names start with "swift" to be force-loaded.
Note that unlike the more general `-force_load`, this flag only applies
to libraries specified via LC_LINKER_OPTIONS, and not those passed on
the command-line. This is what ld64 does.
Reviewed By: #lld-macho, thakis
Differential Revision: https://reviews.llvm.org/D103709
Our implementation draws heavily from LLD-ELF's, which in turn delegates
its string deduplication to llvm-mc's StringTableBuilder. The messiness of
this diff is largely due to the fact that we've previously assumed that
all InputSections get concatenated together to form the output. This is
no longer true with CStringInputSections, which split their contents into
StringPieces. StringPieces are much more lightweight than InputSections,
which is important as we create a lot of them. They may also overlap in
the output, which makes it possible for strings to be tail-merged. In
fact, the initial version of this diff implemented tail merging, but
I've dropped it for reasons I'll explain later.
**Alignment Issues**
Mergeable cstring literals are found under the `__TEXT,__cstring`
section. In contrast to ELF, which puts strings that need different
alignments into different sections, clang's Mach-O backend puts them all
in one section. Strings that need to be aligned have the `.p2align`
directive emitted before them, which simply translates into zero padding
in the object file.
I *think* ld64 extracts the desired per-string alignment from this data
by preserving each string's offset from the last section-aligned
address. I'm not entirely certain since it doesn't seem consistent about
doing this; but perhaps this can be chalked up to cases where ld64 has
to deduplicate strings with different offset/alignment combos -- it
seems to pick one of their alignments to preserve. This doesn't seem
correct in general; we can in fact can induce ld64 to produce a crashing
binary just by linking in an additional object file that only contains
cstrings and no code. See PR50563 for details.
Moreover, this scheme seems rather inefficient: since unaligned and
aligned strings are all put in the same section, which has a single
alignment value, it doesn't seem possible to tell whether a given string
doesn't have any alignment requirements. Preserving offset+alignments
for strings that don't need it is wasteful.
In practice, the crashes seen so far seem to stem from x86_64 SIMD
operations on cstrings. X86_64 requires SIMD accesses to be
16-byte-aligned. So for now, I'm thinking of just aligning all strings
to 16 bytes on x86_64. This is indeed wasteful, but implementation-wise
it's simpler than preserving per-string alignment+offsets. It also
avoids the aforementioned crash after deduplication of
differently-aligned strings. Finally, the overhead is not huge: using
16-byte alignment (vs no alignment) is only a 0.5% size overhead when
linking chromium_framework.
With these alignment requirements, it doesn't make sense to attempt tail
merging -- most strings will not be eligible since their overlaps aren't
likely to start at a 16-byte boundary. Tail-merging (with alignment) for
chromium_framework only improves size by 0.3%.
It's worth noting that LLD-ELF only does tail merging at `-O2`. By
default (at `-O1`), it just deduplicates w/o tail merging. @thakis has
also mentioned that they saw it regress compressed size in some cases
and therefore turned it off. `ld64` does not seem to do tail merging at
all.
**Performance Numbers**
CString deduplication reduces chromium_framework from 250MB to 242MB, or
about a 3.2% reduction.
Numbers for linking chromium_framework on my 3.2 GHz 16-Core Intel Xeon W:
N Min Max Median Avg Stddev
x 20 3.91 4.03 3.935 3.95 0.034641016
+ 20 3.99 4.14 4.015 4.0365 0.0492336
Difference at 95.0% confidence
0.0865 +/- 0.027245
2.18987% +/- 0.689746%
(Student's t, pooled s = 0.0425673)
As expected, cstring merging incurs some non-trivial overhead.
When passing `--no-literal-merge`, it seems that performance is the
same, i.e. the refactoring in this diff didn't cost us.
N Min Max Median Avg Stddev
x 20 3.91 4.03 3.935 3.95 0.034641016
+ 20 3.89 4.02 3.935 3.9435 0.043197831
No difference proven at 95.0% confidence
Reviewed By: #lld-macho, gkm
Differential Revision: https://reviews.llvm.org/D102964
Also adds support for live_support sections, no_dead_strip sections,
.no_dead_strip symbols.
Chromium Framework 345MB unstripped -> 250MB stripped
(vs 290MB unstripped -> 236M stripped with ld64).
Doing dead stripping is a bit faster than not, because so much less
data needs to be processed:
% ministat lld_*
x lld_nostrip.txt
+ lld_strip.txt
N Min Max Median Avg Stddev
x 10 3.929414 4.07692 4.0269079 4.0089678 0.044214794
+ 10 3.8129408 3.9025559 3.8670411 3.8642573 0.024779651
Difference at 95.0% confidence
-0.144711 +/- 0.0336749
-3.60967% +/- 0.839989%
(Student's t, pooled s = 0.0358398)
This interacts with many parts of the linker. I tried to add test coverage
for all added `isLive()` checks, so that some test will fail if any of them
is removed. I checked that the test expectations for the most part match
ld64's behavior (except for live-support-iterations.s, see the comment
in the test). Interacts with:
- debug info
- export tries
- import opcodes
- flags like -exported_symbol(s_list)
- -U / dynamic_lookup
- mod_init_funcs, mod_term_funcs
- weak symbol handling
- unwind info
- stubs
- map files
- -sectcreate
- undefined, dylib, common, defined (both absolute and normal) symbols
It's possible it interacts with more features I didn't think of,
of course.
I also did some manual testing:
- check-llvm check-clang check-lld work with lld with this patch
as host linker and -dead_strip enabled
- Chromium still starts
- Chromium's base_unittests still pass, including unwind tests
Implemenation-wise, this is InputSection-based, so it'll work for
object files with .subsections_via_symbols (which includes all
object files generated by clang). I first based this on the COFF
implementation, but later realized that things are more similar to ELF.
I think it'd be good to refactor MarkLive.cpp to look more like the ELF
part at some point, but I'd like to get a working state checked in first.
Mechanical parts:
- Rename canOmitFromOutput to wasCoalesced (no behavior change)
since it really is for weak coalesced symbols
- Add noDeadStrip to Defined, corresponding to N_NO_DEAD_STRIP
(`.no_dead_strip` in asm)
Fixes PR49276.
Differential Revision: https://reviews.llvm.org/D103324
This omits load commands for unreferenced dylibs if:
- the dylib was loaded implicitly,
- it is marked MH_DEAD_STRIPPABLE_DYLIB
- or -dead_strip_dylibs is passed
This matches ld64.
Currently, the "is dylib referenced" state is computed before dead code
stripping and is not updated after dead code stripping. This too matches ld64.
We should do better here.
With this, clang-format linked with lld (like with ld64) no longer has
libobjc.A.dylib in `otool -L` output. (It was implicitly loaded as a reexport
of CoreFoundation.framework, but it's not needed.)
Differential Revision: https://reviews.llvm.org/D103430
Extend the range of calls beyond an architecture's limited branch range by first calling a thunk, which loads the far address into a scratch register (x16 on ARM64) and branches through it.
Other ports (COFF, ELF) use multiple passes with successively-refined guesses regarding the expansion of text-space imposed by thunk-space overhead. This MachO algorithm places thunks during MergedOutputSection::finalize() in a single pass using exact thunk-space overheads. Thunks are kept in a separate vector to avoid the overhead of inserting into the `inputs` vector of `MergedOutputSection`.
FIXME:
* arm64-stubs.s test is broken
* add thunk tests
* Handle thunks to DylibSymbol in MergedOutputSection::finalize()
Differential Revision: https://reviews.llvm.org/D100818
ld64 automatically renames many sections depending on output type and assorted flags. Here, we implement the most common configs. We can add more obscure flags and behaviors as needed.
Depends on D101393
Differential Revision: https://reviews.llvm.org/D101395
Modern versions of macOS (>= 10.7) and in general all modern Mach-O
target archs want PIEs by default. ld64 defaults to PIE for iOS >= 4.3,
as well as for all versions of watchOS and simulators. Basically all the
platforms LLD is likely to target want PIE. So instead of cluttering LLD's
code with legacy version checks, I think it's simpler to just default to
PIE for everything.
Note that `-no_pie` still works, so users can still opt out of it.
Reviewed By: #lld-macho, thakis, MaskRay
Differential Revision: https://reviews.llvm.org/D101513
Add option to limit (or remove limits) on the number of errors printed before exiting. This option exists in the other lld ports: COFF & ELF.
Differential Revision: https://reviews.llvm.org/D101274
This load command records a range spanning from the end of the load
commands to the end of the `__TEXT` segment. Presumably the kernel will encrypt
all this data.
Reviewed By: #lld-macho, thakis
Differential Revision: https://reviews.llvm.org/D100973
This diff creates an empty XAR file and copies it into
`__LLVM,__bundle`. Follow-up work will actually populate the contents of
that XAR.
Reviewed By: #lld-macho, gkm
Differential Revision: https://reviews.llvm.org/D100650
Swift builds seem to use it. All it requires is emitting the
corresponding paths as STABS.
Fixes llvm.org/PR49385.
Reviewed By: clayborg
Differential Revision: https://reviews.llvm.org/D100076
The test is loosely based off LLD-ELF's `thinlto.ll`. However, I
found that test questionable because the the -save_temps behavior it
checks for is identical regardless of whether we are running in single-
or multi-threaded mode. I tried writing a test based on `--time-trace`
but couldn't get it to run deterministically... so I've opted to just
skip checking that behavior for now.
Reviewed By: #lld-macho, gkm
Differential Revision: https://reviews.llvm.org/D99356
Summary: We needed to use `Joined` instead of `Flag`. This wasn't caught
because the relevant test that was copied from LLD-ELF was still
invoking LLD-ELF instead of LLD-MachO...
Differential Revision: https://reviews.llvm.org/D99313
Pretty simple code-wise. Also threw in some refactoring:
* Put the functionStartSection under Writer instead of InStruct, since
it doesn't need to be accessed outside of Writer
* Adjusted the test to put all files under the temp dir instead of at
the top-level
* Added some CHECK-LABELs to make it clearer where the function starts
data is
Differential Revision: https://reviews.llvm.org/D99112
This reverts commit 4876ba5b2d.
Third-attemp relanding D98559, new change:
- explicitly cast enum to underlying type to avoid ambiguity (workaround to clang's bug).
This reverts commit 3c21166a94.
The build is broken (clang-8 host compiler):
lld/MachO/DriverUtils.cpp:271:8: error: use of overloaded operator '<<' is ambiguous (with operand types 'llvm::raw_fd_ostream' and 'lld::macho::DependencyTracker::DepOpCode')
os << opcode;
~~ ^ ~~~~~~
This reverts commit 9670d2e4af.
Second attemp to reland D98559. New changes:
- inline functions removed from cpp file.
- updated tests to use CHECK-DAG instead of CHECK-NEXT
- fixed ambiguous "<<" operator by switching `char` to uint8_t
This reverts commit 2554b95db5.
Relanding [lld-macho] Implement -dependency_info (D98559) with changes:
- inline functions removed from cpp file.
- updated tests to not check libSystem.tbd with other input files (because of possible indeterministic ordering)
This reverts commit c53a1322f3.
Test only passes depending on build dir having a lexicographically later name
than the source dir, and doesn't link on mac/win. See
https://reviews.llvm.org/D98559#2640265 onward.
Bug: https://bugs.llvm.org/show_bug.cgi?id=49278
The flag is not well documented, so this implementation is based on observed behaviour.
When specified, `-dependency_info <path>` produced a text file containing information pertaining to the current linkage, such as input files, output file, linker version, etc.
This file's layout is also not documented, but it seems to be a series of null ('\0') terminated strings in the form `<op code><path>`
`<op code>` could be:
`0x00` : linker version
`0x10` : input
`0x11` : files not found(??)
`0x40` : output
`<path>` : is the file path, except for the linker-version case.
(??) This part is a bit unclear. I think it means all the files the linker attempted to look at, but could not find.
Differential Revision: https://reviews.llvm.org/D98559
This is the minimal port from ELF. Any extension should easy from here
Test plan: ninja check-all-macho
Reviewed By: #lld-macho, thakis
Differential Revision: https://reviews.llvm.org/D98419
lld doesn't read MH_DEAD_STRIPPABLE_DYLIB to strip dead dylibs yet,
but now it can produce dylibs with it set.
While here, also switch an existing test that looks only at the main Mach-O
header from --all-headers to --private-header.
Differential Revision: https://reviews.llvm.org/D98262
Implement command-line options to alter a dylib's exported-symbols list:
* `-exported_symbol*` options override the default export list. The export list is compiled according to the command-line option(s) only.
* `-unexported_symbol*` options hide otherwise public symbols.
* `-*exported_symbol PATTERN` options specify a single literal or glob pattern.
* `-*exported_symbols_list FILE` options specify a file containing a series of lines containing symbol literals or glob patterns. Whitespace and `#`-prefix comments are stripped.
Note: This is a simple implementation of the primary use case. ld64 has much more complexity surrounding interactions with other options, many of which are obscure and undocumented. We will start simple and complexity as necessary.
Differential Revision: https://reviews.llvm.org/D98223
These are mostly things that ld64 has itself marked obsolete.
In the case of `-sectorder`, it's suggested in ld64's manpage that it
could be deprecated, so let's skip implementing it for now.
Reviewed By: #lld-macho, thakis
Differential Revision: https://reviews.llvm.org/D98066
Previously, lld/mac only ad-hoc codesigned executables on arm64.
Matches ld64 behavior. Part of PR49443. Fixes 14 of 17 failures when running
check-llvm with lld as host linker on an M1 MBP.
Differential Revision: https://reviews.llvm.org/D97994
-flat_namespace makes lld emit binaries that use name lookup that's more in
line with other POSIX systems: Instead of looking up symbols as (dylib,name)
pairs by dyld, they're instead looked up just by name.
-flat_namespace has three effects:
1. MH_TWOLEVEL and MH_NNOUNDEFS are no longer set in the Mach-O header
2. All symbols use BIND_SPECIAL_DYLIB_FLAT_LOOKUP as ordinal
3. When a dylib is added to the link, its dependent dylibs are also added,
so that lld can verify that no undefined symbols remain at the end of
a link with -flat_namespace. These transitive dylibs are added for symbol
resolution, but they are not emitted in LC_LOAD_COMMANDs.
-undefined with -flat_namespace still isn't implemented. Before this change,
it was impossible to hit that combination because -flat_namespace caused a
diagnostic. Now that it no longer does, emit a dedicated temporary diagnostic
when both flags are used.
Differential Revision: https://reviews.llvm.org/D97641
Dynamic lookup symbols are symbols that work like dynamic symbols
in ELF: They're not bound to a dylib like normal Mach-O twolevel lookup
symbols, but they live in a global pool and dyld resolves them against
exported symbols from all loaded dylibs.
This adds support for dynamical lookup symbols to lld/mac. They are
represented as DylibSymbols with file set to nullptr.
This also uses this support to implement the -U flag, which makes
a specific symbol that's undefined at the end of the link a
dynamic lookup symbol.
For -U, it'd be sufficient to just to a pass over remaining undefined symbols
at the end of the link and to replace them with dynamic lookup symbols then.
But I'd like to use this code to implement flat_namespace too, and that will
require real support for resolving dynamic lookup symbols in SymbolTable. So
this patch adds this now already.
While writing tests for this, I noticed that we didn't set N_WEAK_DEF in the
symbol table for DylibSymbols, so this fixes that too.
Differential Revision: https://reviews.llvm.org/D97521
See discussion on https://reviews.llvm.org/D93263
-flat_namespace isn't implemented yet, and neither is -undefined dynamic,
so this makes -undefined pretty pointless in lld/MachO for now. But once
we implement -flat_namespace (which we need to do anyways to get check-llvm
to pass with lld as host linker), the code's already there.
Follow-up to https://reviews.llvm.org/D93263#2491865
Differential Revision: https://reviews.llvm.org/D96963
Differential Revision: https://reviews.llvm.org/D95913
Usage: -bundle_loader <executable>
This option specifies the executable that will load the build output file being linked.
When building a bundle, users can use the --bundle_loader to specify an executable
that contains symbols referenced, but not implemented in the bundle.
Since we emit diagnostics for undefineds in Writer::scanRelocations()
and symbols referenced by -u flags aren't referenced by any relocations,
this needs some manual code (similar to the entry point).
Differential Revision: https://reviews.llvm.org/D94371
I noticed that this option was not appearing at all in the `--help`
messages for `wasm-ld` or `ld.lld`.
Add help text and make it consistent across all ports.
Differential Revision: https://reviews.llvm.org/D94925
Just getting rid of some logspew as I test LLD under existing build
systems.
Reviewed By: #lld-macho, smeenai
Differential Revision: https://reviews.llvm.org/D95213
Run the ObjCARCContractPass during LTO. The legacy LTO backend (under
LTO/ThinLTOCodeGenerator.cpp) already does this; this diff just adds that
behavior to the new LTO backend. Without that pass, the objc.clang.arc.use
intrinsic will get passed to the instruction selector, which doesn't know how to
handle it.
In order to test both the new and old pass managers, I've also added support for
the `--[no-]lto-legacy-pass-manager` flags.
P.S. Not sure if the ordering of the pass within the pipeline matters...
Reviewed By: fhahn
Differential Revision: https://reviews.llvm.org/D94547
Nico Weber