https://discourse.llvm.org/t/parallel-input-file-parsing/60164
initializeSymbols currently sets Defined::section and handles non-prevailing
COMDAT groups. Move the code to the parallel postParse to reduce work from the
single-threading code path and make parallel section initialization infeasible.
Postpone reporting duplicate symbol errors so that the messages have the
section information. (`Defined::section` is assigned in postParse and another
thread may not have the information).
* duplicated-synthetic-sym.s: BinaryFile duplicate definition (very rare) now
has no section information
* comdat-binding: `%t/w.o %t/g.o` leads to an undesired undefined symbol. This
is not ideal but we report a diagnostic to inform that this is unsupported.
(See release note)
* comdat-discarded-lazy.s: %tdef.o is unextracted. The new behavior (discarded
section error) makes more sense
* i386-comdat.s: switched to a better approach working around
.gnu.linkonce.t.__x86.get_pc_thunk.bx in glibc<2.32 for x86-32.
Drop the ancient no-longer-relevant workaround for __i686.get_pc_thunk.bx
Depends on D120640
Differential Revision: https://reviews.llvm.org/D120626
In particular we use these in two places:
1. When building PIC code we no longer need to combine output segments
into a single segment that can be initialized at `__memory_base`.
Instead each segment can encode its offset from `__memory_base` in
its initializer. e.g.
```
(i32.add (global.get __memory_base) (i32.const offset)
```
2. When building PIC code we no longer need to relocation internalized
global addresses. We can just initialize them with their correct
offsets.
Differential Revision: https://reviews.llvm.org/D121420
Since Mach-O has a two-level namespace (unlike ELF), we can usually set
this property to true.
(I believe this setting is only available in the new LTO backend, so I
can't really use ld64 / libLTO's behavior as a reference here... I'm
just doing what I think is correct.)
See {D119294} for the work done to calculate the `interposable` used in
this diff.
Reviewed By: MaskRay
Differential Revision: https://reviews.llvm.org/D119506
This is a new mode for handling unresolved symbols that allows all
symbols to be imported in the same that they would be in the case of
`-fpie` or `-shared`, but generting an otherwise fixed/non-relocatable
binary.
Code linked in this way should still be compiled with `-fPIC` so that
data symbols can be resolved via imports.
This essentially allows the building of static binaries that have
dynamic imports. See:
https://github.com/emscripten-core/emscripten/issues/12682
As with other uses of the experimental dynamic linking ABI, this
behaviour will produce a warning unless run with `--experimental-pic`.
Differential Revision: https://reviews.llvm.org/D91577
All references to interposable symbols can be redirected at runtime to
point to a different symbol definition (with the same name). For
example, if both dylib A and B define symbol _foo, and we load A before
B at runtime, then all references to _foo within dylib B will point to
the definition in dylib A.
ld64 makes all extern symbols interposable when linking with
`-flat_namespace`.
TODO 1: Support `-interposable` and `-interposable_list`, which should
just be a matter of parsing those CLI flags and setting the
`Defined::interposable` bit.
TODO 2: Set Reloc::FinalDefinitionInLinkageUnit correctly with this info
(we are currently not setting it at all, so we're erring on the
conservative side, but we should help the LTO backend generate more
optimal code.)
Reviewed By: modimo, MaskRay
Differential Revision: https://reviews.llvm.org/D119294
Previously, we only allowed this for DylibSymbols. However, in order to
properly support `-flat_namespace` as well as `-interposable`, we need
to allow this for Defined symbols too. Therefore we hoist the
`lazyBindOffset` and the `stubsHelperIndex` into the parent Symbol
class.
The actual change to support interposition under `-flat_namespace` is in
{D119294}; the NFC changes here have been split out for easier review.
Perf regression isn't stat sig on my 3.2 GHz 16-Core Intel Xeon W linking
chromium_framework:
base diff difference (95% CI)
sys_time 1.227 ± 0.021 1.234 ± 0.031 [ -0.3% .. +1.5%]
user_time 3.665 ± 0.036 3.674 ± 0.035 [ -0.2% .. +0.7%]
wall_time 4.596 ± 0.055 4.609 ± 0.064 [ -0.3% .. +0.9%]
samples 34 47
Max RSS regression is barely stat sig:
base diff difference (95% CI)
time 1003664356.324 ± 15404053.912 1010380403.613 ± 10578309.455 [ +0.0% .. +1.3%]
samples 37 31
Reviewed By: modimo
Differential Revision: https://reviews.llvm.org/D121351
This reverts commit e049a87f04.
That commit breaks the build with errors of the form:
/usr/local/google/home/saugustine/llvm/llvm-project/lld/MachO/ExportTrie.cpp:148:11: error: definition of implicitly declared destructor
TrieNode::~TrieNode() {
The code can be used in multi-threads and the allocator is not thread safe.
fixes PR/54378
Reviewed By: int3, #lld-macho
Differential Revision: https://reviews.llvm.org/D121638
https://discourse.llvm.org/t/parallel-input-file-parsing/60164
initializeSymbols currently sets Defined::section and handles non-prevailing
COMDAT groups. Move the code to the parallel postParse to reduce work from the
single-threading code path and make parallel section initialization infeasible.
Postpone reporting duplicate symbol errors so that the messages have the
section information. (`Defined::section` is assigned in postParse and another
thread may not have the information).
* duplicated-synthetic-sym.s: BinaryFile duplicate definition (very rare) now
has no section information
* comdat-binding: `%t/w.o %t/g.o` leads to an undesired undefined symbol. This
is not ideal but we report a diagnostic to inform that this is unsupported.
(See release note)
* comdat-discarded-lazy.s: %tdef.o is unextracted. The new behavior (discarded
section error) makes more sense
Depends on D120640
Reviewed By: peter.smith
Differential Revision: https://reviews.llvm.org/D120626
In GCC -fgnu-unique output, STB_GNU_UNIQUE symbols are always defined relative
to a section in a COMDAT group. Currently `other` cannot be STB_GNU_UNIQUE for
valid input, so this patch is NFC.
If we switch to the model that ignores COMDAT resolution when performing symbol
resolution (D120626), this will fix bogus `relocation refers to a symbol in a
discarded section` errors when mixing -fno-gnu-unique objects with -fgnu-unique
objects.
Differential Revision: https://reviews.llvm.org/D120640
This change continues to lay the ground work for supporting extended
const expressions in the linker.
The included test covers object file reading and writing and the YAML
representation.
Differential Revision: https://reviews.llvm.org/D121349
Previously, the test checked for a "undefined symbol" error
(instead of the "could not open std*.lib" which would happen without
the flag).
Instead, use /entry: so that the link succeeds.
No behavior change, but maybe makes the test a bit easier to understand.
Differential Revision: https://reviews.llvm.org/D121553
This clarifies that this is an LLVM specific variable and avoids
potential conflicts with other projects.
Differential Revision: https://reviews.llvm.org/D119918
GNU ld 2.38 added -z pack-relative-relocs which is similar to
--pack-dyn-relocs=relr but synthesizes the `GLIBC_ABI_DT_RELR` version
dependency if a shared object named `libc.so.*` has a `GLIBC_2.*` version
dependency.
This is used to implement the (as some glibc folks call) version lockout
mechanism. Add this option, because glibc does not want to support
--pack-dyn-relocs=relr which does not add `GLIBC_ABI_DT_RELR`.
See https://maskray.me/blog/2021-10-31-relative-relocations-and-relr for
detail.
Close https://github.com/llvm/llvm-project/issues/53775
Reviewed By: peter.smith
Differential Revision: https://reviews.llvm.org/D120701
Previously, we aligned every cstring to 16 bytes as a temporary hack to
deal with https://github.com/llvm/llvm-project/issues/50135. However, it
was highly wasteful in terms of binary size.
To recap, 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. In other words, we have to infer the alignment of
the cstrings from their addresses.
We differ slightly from ld64 in how we've chosen to align these
cstrings. Both LLD and ld64 preserve the number of trailing zeros in
each cstring's address in the input object files. When deduplicating
identical cstrings, both linkers pick the cstring whose address has more
trailing zeros, and preserve the alignment of that address in the final
binary. However, ld64 goes a step further and also preserves the offset
of the cstring from the last section-aligned address. I.e. if a cstring
is at offset 18 in the input, with a section alignment of 16, then both
LLD and ld64 will ensure the final address is 2-byte aligned (since
`18 == 16 + 2`). But ld64 will also ensure that the final address is of
the form 16 * k + 2 for some k (which implies 2-byte alignment).
Note that ld64's heuristic means that a dedup'ed cstring's final address is
dependent on the order of the input object files. E.g. if in addition to the
cstring at offset 18 above, we have a duplicate one in another file with a
`.cstring` section alignment of 2 and an offset of zero, then ld64 will pick
the cstring from the object file earlier on the command line (since both have
the same number of trailing zeros in their address). So the final cstring may
either be at some address `16 * k + 2` or at some address `2 * k`.
I've opted not to follow this behavior primarily for implementation
simplicity, and secondarily to save a few more bytes. It's not clear to me
that preserving the section alignment + offset is ever necessary, and there
are many cases that are clearly redundant. In particular, if an x86_64 object
file contains some strings that are accessed via SIMD instructions, then the
.cstring section in the object file will be 16-byte-aligned (since SIMD
requires its operand addresses to be 16-byte aligned). However, there will
typically also be other cstrings in the same file that aren't used via SIMD
and don't need this alignment. They will be emitted at some arbitrary address
`A`, but ld64 will treat them as being 16-byte aligned with an offset of
`16 % A`.
I have verified that the two repros in https://github.com/llvm/llvm-project/issues/50135
work well with the new alignment behavior.
Fixes https://github.com/llvm/llvm-project/issues/54036.
Reviewed By: #lld-macho, oontvoo
Differential Revision: https://reviews.llvm.org/D121342
Combined with the previous change, lld executable is ~2K smaller and some code
paths using InputSection::getParent are more efficient.
The fragmented headers lead to a design limitation that OutputSection has to be
incomplete, so we cannot use static_cast.
Add an OutputDesc class inheriting from SectionCommand. An OutputDesc wraps an
OutputSection. This change allows InputSection::getParent to be inlined.
Differential Revision: https://reviews.llvm.org/D120650
ld64 breaks down `__objc_classrefs` on a per-word level and deduplicates
them. This greatly reduces the number of bind entries emitted (and
therefore the amount of work `dyld` has to do at runtime). For
chromium_framework, this change to LLD cuts the number of (non-lazy)
binds from 912 to 190, getting us to parity with ld64 in this aspect.
Reviewed By: #lld-macho, thakis
Differential Revision: https://reviews.llvm.org/D121053
`__cfstring` has embedded addends that foil ICF's hashing / equality
checks. (We can ignore embedded addends when doing ICF because the same
information gets recorded in our Reloc structs.) Therefore, in order to
properly dedup CFStrings, we create a mutable copy of the CFString and
zero out the embedded addends before performing any hashing / equality
checks.
(We did in fact have a partial implementation of CFString deduplication
already. However, it only worked when the cstrings they point to are at
identical offsets in their object files.)
I anticipate this approach can be extended to other similar
statically-allocated struct sections in the future.
In addition, we previously treated all references with differing addends
as unequal. This is not true when the references are to literals:
different addends may point to the same literal in the output binary. In
particular, `__cfstring` has such references to `__cstring`. I've
adjusted ICF's `equalsConstant` logic accordingly, and I've added a few
more tests to make sure the addend-comparison code path is adequately
covered.
Fixes https://github.com/llvm/llvm-project/issues/51281.
Reviewed By: #lld-macho, Roger
Differential Revision: https://reviews.llvm.org/D120137
... from a `uint64_t` to a `uint32_t`. (LLD-ELF uses a `uint32_t` too.)
About a 1.7% reduction in peak RSS when linking chromium_framework on my
3.2 GHz 16-Core Intel Xeon W Mac Pro, and no stat sig change in wall
time.
</Users/jezng/test2.sh ["before"]> </Users/jezng/test2.sh ["after"]> difference (95% CI)
RSS 1003036672.000 ± 9891065.259 985539505.231 ± 10272748.749 [ -2.3% .. -1.2%]
samples 27 26
base diff difference (95% CI)
sys_time 1.277 ± 0.023 1.277 ± 0.024 [ -0.9% .. +0.9%]
user_time 6.682 ± 0.046 6.598 ± 0.043 [ -1.6% .. -0.9%]
wall_time 5.904 ± 0.062 5.895 ± 0.063 [ -0.7% .. +0.4%]
samples 46 28
No appreciable change (~0.01%) in number of `equals` comparisons either:
Before:
ld64.lld: ICF needed 8 iterations
ld64.lld: equalsConstant() called 701643 times
ld64.lld: equalsVariable() called 3438526 times
After:
ld64.lld: ICF needed 8 iterations
ld64.lld: equalsConstant() called 701729 times
ld64.lld: equalsVariable() called 3438526 times
Reviewed By: #lld-macho, MaskRay, thakis
Differential Revision: https://reviews.llvm.org/D121052
The existing hashing of stubsHelperIndex has mostly been a no-op* for
some time now (ever since we made ICF run before dylib symbols get their
stubs indices assigned). I guess we could consider hashing the name +
filename of the DylibSymbol instead, but I'm not sure the overhead's
worth it... moreover, LLD/ELF only hashes their Defined symbols as well.
*: Technically it does change the hash value since stubsHelperIndex is
initialized to `UINT32_MAX` by default. But since all stubsHelperIndex
values are the same at when ICF runs, they don't add any useful
information to the hash.
This is debug code that is disabled by default. It'll provide a easy way
to figure out the impact (if any) of tweaking ICF's hashing algorithm
(since a poor quality hash will result in many more `equals*` calls).
Reviewed By: #lld-macho, oontvoo
Differential Revision: https://reviews.llvm.org/D121051
ObjFile::parse combines symbol initialization and resolution. Many tasks
unrelated to symbol resolution can be postponed and parallelized. This patch
extracts local symbol initialization and parallelizes it.
Technically the new function initializeLocalSymbols can be merged into
ObjFile::postParse, but functions like getSrcMsg may access the
uninitialized (all nullptr) local part of InputFile::symbols.
Linking chrome: 1.02x as fast with glibc malloc, 1.04x as fast with mimalloc
Depends on f456c3ae3f and D119908
Reviewed By: ikudrin
Differential Revision: https://reviews.llvm.org/D119909
addWrappedSymbols may trigger archive extraction: split stack implementation
uses --wrap=pthread_create, which extracts libgcc.a(generic-morestack-thread.o).
This fixes the regression caused by 09602d3b47 by
making the invariant satisfied: no more non-compileBitcodeFiles object file is
produced at postParseObjectFile.
LLD (and ld64) emits uppercase hex addresses in the mapfile. The
map-file.s test passes right now because the addresses we emit happen
not to include any alphabets, but that can easily change.
I noticed this while dealing with
https://github.com/llvm/llvm-project/issues/54184.
Reviewed By: #lld-macho, thakis
Differential Revision: https://reviews.llvm.org/D120941
Now all the tests that cover symbol resolution / precedence have
"resolution" in their filename.
I also added a couple of extra comments.
Reviewed By: #lld-macho, thakis
Differential Revision: https://reviews.llvm.org/D120938
If we fix https://github.com/llvm/llvm-project/issues/54184, we will end
up including libSystem in every %lld invocation, which would break
tapi-link.s as it assumes that libSystem isn't directly linked (instead
it goes through libReexportSystem).
Let's remove this unnecessary coupling, as well as use `split-file`
instead of having a separate file under `Inputs`.
Reviewed By: #lld-macho, oontvoo
Differential Revision: https://reviews.llvm.org/D120939
Per discussion on
https://reviews.llvm.org/D59709#inline-1148734, this seems like the
right course of action. `canBeOmittedFromSymbolTable()` subsumes and
generalizes the previous logic. In addition to handling `linkonce_odr`
`unnamed_addr` globals, we now also internalize `linkonce_odr` +
`local_unnamed_addr` constants.
Reviewed By: tejohnson
Differential Revision: https://reviews.llvm.org/D120173
Previously, we were using a syslibroot that pointed to macos while
linking against arch arm64_32, which didn't really make sense. It isn't
currently an issue, but will be if we add the `-lSystem` as part of
dealing with https://github.com/llvm/llvm-project/issues/54184.
If we fix https://github.com/llvm/llvm-project/issues/54184, the
`dyld_stub_binder` symbol will get included in every output dylib. This
would cause the addresses of the other symbols to shift, breaking the
test as it currently stands. Let's make the test more flexible.
Reviewed By: lgrey
Differential Revision: https://reviews.llvm.org/D120940
So far, we sort all discardable sections at the end, with only some
extra logic to make sure that the .reloc section is at the start
of that group of sections. But if there are other discardable
sections, other than .reloc, they must also be ordered before
.debug_* sections, to avoid leaving gaps if the executable is
stripped.
(Stripping executables doesn't remove all discardable sections,
only the ones named .debug_*).
Rust binaries seem to include a .rmeta section, which is marked
discardable. This fixes stripping such binaries if built with
dwarf debug info included.
This fixes issues observed in MSYS2 in
https://github.com/msys2/MINGW-packages/pull/10555.
Differential Revision: https://reviews.llvm.org/D120805
Show the name of of the archive in the error message as well as the name
of the object within it.
Differential Revision: https://reviews.llvm.org/D120689
This does tail merging (and deduplication) of the strings.
On a statically linked clang.exe, this shrinks the ~17 MB string
table by around 0.5 MB. This adds ~160 ms to the linking time
which originally was around 950 ms.
For cases where `-debug:symtab` or `-debug:dwarf` isn't set, the
string table is only used for long section names, where this
shouldn't make any difference at all.
Differential Revision: https://reviews.llvm.org/D120677
The previous code used an unbounded sprintf, which in theory can
overflow, writing either the null terminator or the last digits
into the next struct member.
In practice, in LLD, all long section names are written sequentially
first at the start of the string table, followed by all the long
symbol names. Due to this, even if the total string table would
end up large, the long section names have fairly short offsets,
which is why this hasn't been an issue in practice.
I don't think it's worth trying to write a test that produces an
executable with enough long section names to make the section names
themselves exceed 10^6 bytes, which is currently necessary to trigger
faults with the previous form.
Differential Revision: https://reviews.llvm.org/D120676
https://github.com/ClangBuiltLinux/linux/issues/1606
When GNU_PROPERTY_X86_FEATURE_1_IBT is enabled, ld.lld will create .plt output
section even if there is no PLT entry. Fix this by implementing
IBTPltSection::isNeeded instead of using the default code path (which always
returns true).
Reviewed By: MaskRay
Differential Revision: https://reviews.llvm.org/D120600
ObjFile::parse combines symbol initialization and resolution. Many tasks
unrelated to symbol resolution can be postponed and parallelized. This patch
extracts local symbol initialization and parallelizes it.
Technically the new function initializeLocalSymbols can be merged into
ObjFile::postParse, but functions like getSrcMsg may access the
uninitialized (all nullptr) local part of InputFile::symbols.
Linking chrome: 1.02x as fast with glibc malloc, 1.04x as fast with mimalloc
Reviewed By: ikudrin
Differential Revision: https://reviews.llvm.org/D119909
This was based off @thakis' draft in {D103517}. I employed templates to ensure
the support for `-why_live` wouldn't slow down the regular non-why-live code
path.
No stat sig perf difference on my 3.2 GHz 16-Core Intel Xeon W:
base diff difference (95% CI)
sys_time 1.195 ± 0.015 1.199 ± 0.022 [ -0.4% .. +1.0%]
user_time 3.716 ± 0.022 3.701 ± 0.025 [ -0.7% .. -0.1%]
wall_time 4.606 ± 0.034 4.597 ± 0.046 [ -0.6% .. +0.2%]
samples 44 37
Reviewed By: #lld-macho, thakis
Differential Revision: https://reviews.llvm.org/D120377
* detect `def_tls.o undef_nontls.o` violation
* place error checking code (checking duplicate symbol) together
* allow `--defsym tls1=tls2 def_tls.o`
As a degraded error checking, `--defsym tls1=42` violation will not be detected.
Symbol.h depends on InputFiles.h. This change moves us toward dropping the
weird dependency.
The call sites will become slightly uglier (`cast<SharedFile>(s->file)`), but
the compromise is acceptable.
In GNU ld, the definition precedence is: regular symbol assignment > relocatable object definition > `PROVIDE` symbol assignment.
GNU ld's internal linker scripts define the non-reserved (by C and C++)
edata/end/etext with `PROVIDE` so the relocatable object definition takes
precedence. This makes sense because `int end;` is valid.
We currently redefine such symbols if they are COMMON, but not if they are
regular definitions, so `int end;` with -fcommon is essentially a UB in ld.lld.
Fix this (also improve consistency and match GNU ld) by using the
`isDefined` code path for `isCommon`. In GNU ld, reserved identifiers like
`__ehdr_start` do not use `PROVIDE`, while we treat them all as `PROVIDE`, this
seems fine.
Reviewed By: peter.smith
Differential Revision: https://reviews.llvm.org/D120389
This mirrors the code structure in `lld/ELF`. It also paves the way for
an upcoming diff where I templatize things.
Reviewed By: #lld-macho, thakis
Differential Revision: https://reviews.llvm.org/D120376
This is what the Clang standalone build does too. And setting this
seems to be required to get the standalone build to work on my Mac.
Reviewed By: #lld-macho, MaskRay, Ericson2314, smeenai
Differential Revision: https://reviews.llvm.org/D120269
https://discourse.llvm.org/t/parallel-input-file-parsing/60164
To decouple symbol initialization and section initialization, `Defined::section`
assignment should be postponed after input file parsing. To avoid spurious
duplicate definition error due to two definitions in COMDAT groups of the same
signature, we should postpone the duplicate symbol check.
The function is called postScan instead of a more specific name like
checkDuplicateSymbols, because we may merge Symbol::mergeProperties into
postScan. It is placed after compileBitcodeFiles to apply to ET_REL files
produced by LTO. This causes minor diagnostic regression
for skipLinkedOutput configurations: ld.lld --thinlto-index-only a.bc b.o
(bitcode definition prevails) won't detect duplicate symbol error. I think this
is an acceptable compromise. The important cases where (a) both files are
bitcode or (b) --thinlto-index-only is unused are still detected.
Reviewed By: ikudrin
Differential Revision: https://reviews.llvm.org/D119908
In many call sites we know uncompression cannot happen (non-SHF_ALLOC, or the
data (even if compressed) must have been uncompressed by a previous pass).
Prefer rawData in these cases. data() increases code size and prevents
optimization on rawData.
D118577: the 0.1~1.1% .strtab size reduction does not justify the 3~6%
link time increase. Just remove it even for -O2. release/14.x
has D118577 and the release note mentioned that this may be removed.
Fix https://github.com/ClangBuiltLinux/linux/issues/1578
caused by D118577 (empty string not in stringMap).
Making a (NOLOAD) section SHT_PROGBITS is fishy (the user may expect all-zero
content, but the linker does not check that), but some projects (e.g. Linux
kernel https://github.com/ClangBuiltLinux/linux/issues/1597) traditionally rely
on the behavior. Issue a warning to not break them.
Symbols for which `canBeOmittedFromSymbolTable()` is true should be
treated as private externs. This diff tries to do that by unsetting the
ExportDynamic bit. It seems to mostly work with the FullLTO backend, but
with the ThinLTO backend, the `local_unnamed_addr` symbols still fail to
be properly hidden. Nonetheless, this is a step in the right direction.
I've documented all the remaining differences between our behavior and
LD64's in the lto-internalized-unnamed-addr.ll test.
See also https://discourse.llvm.org/t/mach-o-lto-handling-of-linkonce-odr-unnamed-addr/60015
Reviewed By: #lld-macho, thevinster
Differential Revision: https://reviews.llvm.org/D119767
The current output section type allows to set the ELF section type to
SHT_PROGBITS or SHT_NOLOAD. This patch allows an arbitrary section value
to be specified. Some common SHT_* literal names are supported as well.
```
SECTIONS {
note (TYPE=SHT_NOTE) : { BYTE(8) *(note) }
init_array ( TYPE=14 ) : { QUAD(14) }
fini_array (TYPE = SHT_FINI_ARRAY) : { QUAD(15) }
}
```
When `sh_type` is specified, it is an error if an input section has a different type.
Our syntax is compatible with GNU ld 2.39 (https://sourceware.org/bugzilla/show_bug.cgi?id=28841).
Reviewed By: peter.smith
Differential Revision: https://reviews.llvm.org/D118840
The inline `lld::error` expands to two function calls `errorHandler` and `error`
where the latter is opaque. Move the functions to .cpp files to decrease code
size.
My x86-64 lld executable is 9KiB smaller.
Reviewed By: #lld-macho, thakis
Differential Revision: https://reviews.llvm.org/D120002
If both an order file and a call graph profile are present, the edges of the
call graph which use symbols present in the order file are not used. All of
the symbols in the order file will appear at the beginning of the section just
as they do currently. In other words, the highest priority derived from the
call graph will be below the lowest priority derived from the order file.
Practically, this change renames CallGraphSort.{h,cpp} to SectionPriorities.{h,cpp},
and most order file and call graph profile related code is moved into the new
file to reduce duplication.
Differential Revision: https://reviews.llvm.org/D117354
Main motivation: including `llvm/CodeGen/CommandFlags.h` in
`CommonLinkerContext.h` means that the declaration of `llvm::Reloc` is
visible in any file that includes `CommonLinkerContext.h`. Since our
cpp files have both `using namespace llvm` and `using namespace
lld::macho`, this results in conflicts with `lld::macho::Reloc`.
I suppose we could put `llvm::Reloc` into a nested namespace, but in general,
I think we should avoid transitively including too many header files in
a very widely used header like `CommonLinkerContext.h`.
RegisterCodeGenFlags' ctor initializes a bunch of function-`static`
structures and does nothing else, so it should be fine to "initialize"
it as a temporary stack variable rather than as a file static.
Reviewed By: aganea
Differential Revision: https://reviews.llvm.org/D119913
This relands 73e585e44d (and 0574b5fc65), with a fix for
the failing test (by using Optional<StringRef>s instead of
making StringRef::empty() mean absence of value).
Differential Revision: https://reviews.llvm.org/D118070
The added run lines build a bitcode file for x86 and an object file
for whatever the default target is that is running the test. This
causes an incompatibility between the files.
Add the triple to the llvm-mc invocation.
`parseSections()` is a getting a bit large unwieldy, let's factor out
logic where we can.
Other minor changes in this diff:
* `"__cg_profile"` is now a global constexpr
* We now use `checkError()` instead of `fatal()`-ing without handling
the Error
* Check for `callGraphProfileSort` before checking the section name,
since the boolean comparison is likely cheaper
Reviewed By: #lld-macho, lgrey, oontvoo
Differential Revision: https://reviews.llvm.org/D119892
Symbols with regular GOT entries do need to be exported, but those that
are internalized (and have dymmy/internal GOT entries) need not be
exported.
This happens to fix the failures on the emscripten waterfall where extra
symbols were being exported by the linker (and then later removed by
wasm-opt).
Differential Revision: https://reviews.llvm.org/D119902
Reported by Stefan Pintilie in D119773.
For a branch to a hidden undefined weak symbol, there is an
`assert(sym->getVA());` failure in PPC64LongBranchTargetSection::writeTo for a
-no-pie link. The root cause is that we unnecessarily create the thunk for the
-no-pie link.
Fix this by changing the condition to just `s.isUndefined()`. See the inline
comment.
Rename ppc64-weak-undef-call.s to ppc64-undefined-weak.s to be consistent with
other architectures.
Reviewed By: sfertile, stefanp
Differential Revision: https://reviews.llvm.org/D119787
https://maskray.me/blog/2022-01-16-archives-and-start-lib
For every definition in an extracted archive member, we intern the symbol twice,
once for the archive index entry, once for the .o symbol table after extraction.
This is inefficient.
Symbols in a --start-lib ObjFile/BitcodeFile are only interned once because the
result is cached in symbols[i].
Just handle an archive using the --start-lib code path. We can therefore remove
ArchiveFile and LazyArchive. For many projects, archive member extraction ratio
is high and it is a net performance win. Linking a Release build of clang is
1.01x as fast.
Note: --start-lib scans symbols in the same order that llvm-ar adds them to the
index, so in the common case the semantics should be identical. If the archive
symbol table was created in a different order, or is incomplete, this strategy
may have different semantics. Such cases are considered user error.
The `is neither ET_REL nor LLVM bitcode` error is changed to a warning.
Previously an archive may have such members without a diagnostic. Using a
warning prevents breakage.
* For some tests, the diagnostics get improved where we did not consider
the archive member name: `b.a:` => `b.a(b.o):`.
* `no-obj.s`: the link is now allowed, matching GNU ld
* `archive-no-index.s`: the `is neither ET_REL nor LLVM bitcode` diagnostic is
demoted to a warning.
* `incompatible.s`: even when an archive is unextracted, we may report an
"incompatible with" error.
---
I recently decreased sizeof(SymbolUnion) by 8 and decreased memory usage quite a
bit, so retaining `symbols` for un-extracted archive members should not cause a
memory usage problem.
Reviewed By: peter.smith
Differential Revision: https://reviews.llvm.org/D119074
These global TLS symbols are shared across all shared libraries and
therefor should not be assumed to be local to the current module.
Also add new error in the linker when TLS relocations are used against
undefined symbols. TLS relocations are offsets into the current modules
tls data segment, and don't make sense for undefined symbols which are
modeled as global imports.
Fixes: https://github.com/emscripten-core/emscripten/issues/13398
Differential Revision: https://reviews.llvm.org/D119630
By unsetting this property, we are now able to internalize more symbols
during LTO. I compared the output of `-save-temps` for both LLD and
ld64, and we now match ld64's behavior as far as `lto-internalize.ll` is
concerned.
(Thanks @smeenai for working on an initial version of this diff!)
Fixes https://github.com/llvm/llvm-project/issues/50574.
Reviewed By: #lld-macho, thakis
Differential Revision: https://reviews.llvm.org/D119372
Makes lld-link work in a non-MSVC shell by autodetecting MSVC toolchain. Also
adds support for /winsysroot and a few other switches.
All this is done by refactoring to share code with clang-cl's existing support
for the same.
Differential Revision: https://reviews.llvm.org/D118070
... to use hyphens instead of underscores, making it consistent with
our other substitutions like %no-arg-lld and %lld-watchos.
Reviewed By: keith
Differential Revision: https://reviews.llvm.org/D119513
We have a mix of substituted lld (`%lld`) and hard-coded lld (`ld64.lld`) commands.
When testing with different versions of LLD, this would require going into every place
where lld is hard-coded and changing that. If we centralize it, this'll only require us
to modify it in only one place and will make it easy to run the same test suite. Plus,
this will make it be consistent with how we write other tests.
Reviewed By: #lld-macho, int3, oontvoo
Differential Revision: https://reviews.llvm.org/D119394
Having clarified that executing the SerializeToHsaco pass can
depend on a ROCm installation, switch from calling lld as a library to
using the copy of lld guaranteed to be included in a ROCm install.
This removes the workaround introduced in D119277
Reviewed By: whchung
Differential Revision: https://reviews.llvm.org/D119463
https://maskray.me/blog/2022-02-06-all-about-common-symbols#no-define-common
In GNU ld, -dc only affects -r links and causes COMMON symbols to be allocated.
--no-define-common is defined to make COMMON symbols undefined for -shared.
AIUI --no-define-common is a workaround around glibc 2.1 time and not really useful.
gold confuses --define-common with -d/FORCE_COMMON_ALLOCATION and implements
--define-common with -d semantics. Its --no-define-common is incompatible with
GNU ld.
In ld.lld, b2a23cf3c0 fixed the default -r
behavior for COMMON symbols but ported the incompatible gold
--[no-]define-common. To the best of my knowledge, no project uses -dp
--[no-]define-common. So just remove these options.
-d/-dc are used by the following projects:
* grub grub-core/genmod.sh.in uses -Wl,-r,-d (https://lists.gnu.org/archive/html/grub-devel/2022-02/msg00088.html)
* FreeBSD crunchgen uses -Wl,-dc (https://reviews.freebsd.org/D34215)
A no-op implementation works for them. Only when a program inspects relocatable
output by itself and does not recognize COMMON symbols, there may be a problem.
This is an extremely unlikely case.
Reviewed By: peter.smith
Differential Revision: https://reviews.llvm.org/D119108
This fixes the issue raised in https://reviews.llvm.org/D108850#3303452
Before C++17, the function object is evaluated in a unspecified order. In the following example: https://godbolt.org/z/8ao4vdsr7 the function object is either evaluated before or after the arguments, depending on the compiler. With MSVC and /std:c++14 the function object is evaluated after the arguments; with clang and gcc, it is evaluated before. With C++17, the function object is guaranteed to be evaluated before the arguments, see: https://riptutorial.com/cplusplus/example/19369/evaluation-order-of-function-arguments
In our case, the issue was that the `args` conversion to `ArrayRef` was evaluated before the lambda call `link`, which internally was calling `parseFlavor()`, which in turned modified `args`. We ended with an `ArrayRef` argument that reflected the previous contents of `args`.
Add coverage for `-flavor` which we didn't have before.
Differential Revision: https://reviews.llvm.org/D119278
This makes it easier to pinpoint the source of the problem.
TODO: Have more relocation error messages make use of this
functionality.
Reviewed By: #lld-macho, oontvoo
Differential Revision: https://reviews.llvm.org/D118798
to decrease sizeof(SymbolUnion) from 72 to 64 on ELF64 platforms.
Use a dummy `Undefined` to prevent null pointer dereference (though unused)
`*rel.sym` in InputSectionBase::relocateAlloc.
The relocation order may shuffle a bit, but otherwise there is no behavior
difference.
lld/ELF/OutputSections.cpp includes llvm/Config/config.h for
LLVM_ENABLE_ZLIB definition, but llvm/Config/config.h doesn't exist in
standalone build.
To fix this, this patch moves LLVM_ENABLE_ZLIB from config.h to
llvm-config.h and updates OutputSections.cpp to include llvm-config.h
instead of config.h
Reviewed By: MaskRay, mgorny
Differential Revision: https://reviews.llvm.org/D119058
Adds `-pagezero_size`. `-pagezero_size` commonly used for kernel development.
`-pagezero_size` changes the `__PAGEZERO` size, removing that segment if it is set to zero.
One of the four flags from {D118570}
Now with error messages and tests.
Differential Revision: https://reviews.llvm.org/D118724
SharedSymbol::SharedSymbol initializes verdefIndex and Symbol::replace
copies verdefIndex.
By move verdefIndex assignment outside of ctor, Symbol::replace can be changed
to not copy verdefIndex. This can be used to decrease work for for
ObjKind/BitcodeKind.
Currently `this->getName() == newSym.getName()`.
By keeping the old nameData/nameSize, newSym's nameData/nameSize will be
ignored. The call sites can avoid calling getName().
printTraceSymbol needs to take the symbol name since `other`'s name is empty.
* partition and isPreemptible are frequently used. Move it to the front
* move used beside isUsedInRegularObj. They are similar and accessed together in .symtab finalizing
* move auxIdx/dynsymIndex/verdefIndex to the end.
This decreases code size.
For -no-pie/-pie, when `__real_foo` is interposable in a shared object, `foo` is
exported. This rule does not match GNU ld and is unneeded because:
* the exported `foo` does not interpose `__real_foo` at run-time
* the similar `__wrap_foo` <-> `foo` relation does not have the rule
A STV_PROTECTED shared definition does not set exportDynamic of a defined
symbol. This is on the basis that a protected definition cannot be preempted so
the export is unnecessary. However, the condition is imperfect because we don't
know whether the shared object was built with a symbolic option. Since dropping
the condition simplifies code and matches GNU ld, let's do it.
There's a couple of motivations here:
* LLD 12 (which I was originally testing with) was adding an undefined
symbol to the symbol table if you attempted to wrap an unreferenced
lazy symbol, which would later break `--no-allow-shlib-undefined`. LLD
on main actually produces a weak undefined symbol, so this doesn't
break anyway, but it's cleaner to not have the weak undefined symbol
as well. The new behavior also matches bfd and gold.
* PROVIDE in a linker script referencing a wrapped symbol would think
that an otherwise-unreferenced lazy symbol which was wrapped was
actually referenced, and therefore proceed with the definition, which
goes against expectations. The new behavior also matches bfd and gold.
Reviewed By: MaskRay
Differential Revision: https://reviews.llvm.org/D118756
Xcode 13 comes with a mismatched platform in libcompiler_rt.dylib,
so this creates a linker error on mac catalyst.
Fix it by adding it to the skip list.
Reviewed By: MaskRay, #lld-macho, int3
Differential Revision: https://reviews.llvm.org/D117925
-fprofile-use=/-fprofile-sample-use= compiles may produce REL-format
.rel.llvm.call-graph-profile even if the prevailing format is RELA on AArch64.
Add R_AARCH64_NONE to getImplicitAddend to fix this linker error:
```
ld.lld: error: internal linker error: cannot read addend for relocation R_AARCH64_NONE
PLEASE submit a bug report to https://crbug.com and run tools/clang/scripts/process_crashreports.py (only works inside Google) which will upload a report and include the crash backtrace.
```
Earlier in LLD's evolution, I tried to create the illusion that
subsections were indistinguishable from "top-level" sections. Thus, even
though the subsections shared many common field values, I hid those
common values away in a private Shared struct (see D105305). More
recently, however, @gkm added a public `Section` struct in D113241 that
served as an explicit way to store values that are common to an entire
set of subsections (aka InputSections). Now that we have another "common
value" struct, `Shared` has been rendered redundant. All its fields can
be moved into `Section` instead, and the pointer to `Shared` can be replaced
with a pointer to `Section`.
This `Section` pointer also has the advantage of letting us inspect other
subsections easily, simplifying the implementation of {D118798}.
P.S. I do think that having both `Section` and `InputSection` makes for
a slightly confusing naming scheme. I considered renaming `InputSection`
to `Subsection`, but that would break the symmetry with `OutputSection`.
It would also make us deviate from LLD-ELF's naming scheme.
This change is perf-neutral on my 3.2 GHz 16-Core Intel Xeon W machine:
base diff difference (95% CI)
sys_time 1.258 ± 0.031 1.248 ± 0.023 [ -1.6% .. +0.1%]
user_time 3.659 ± 0.047 3.658 ± 0.041 [ -0.5% .. +0.4%]
wall_time 4.640 ± 0.085 4.625 ± 0.063 [ -1.0% .. +0.3%]
samples 49 61
There's also no stat sig change in RSS (as measured by `time -l`):
base diff difference (95% CI)
time 998038627.097 ± 13567305.958 1003327715.556 ± 15210451.236 [ -0.2% .. +1.2%]
samples 31 36
Reviewed By: #lld-macho, oontvoo
Differential Revision: https://reviews.llvm.org/D118797
- If not using `llvm-config`, `LLVM_MAIN_SRC_DIR` now has a sane default
- `LLVM_CONFIG_PATH` will continue to work for LLD for back compat.
- More quoting of paths in an abundance of caution.
Reviewed By: nikic
Differential Revision: https://reviews.llvm.org/D118792
In the case your framework bundles contain relocatable objects, and your
objects include LC_LINKER_OPTIONs for the framework, previously they
would not be deduplicated like they would have if they were static
archives. This was also the case if you passed `-framework` for the
framework as well.
Reviewed By: #lld-macho, thakis, oontvoo
Differential Revision: https://reviews.llvm.org/D114841
This updates all the non-runtime project release notes to use the
version number from CMake instead of the hard-coded version numbers
in conf.py.
It also hides warnings about pre-releases when the git suffix
is dropped from the LLVM version in CMake.
Reviewed By: MaskRay
Differential Revision: https://reviews.llvm.org/D112181
This diff adds support for ADRP+ADD optimization for AArch64 described in
d2ca58c54b
i.e. under appropriate constraints
ADRP x0, symbol
ADD x0, x0, :lo12: symbol
can be turned into
NOP
ADR x0, symbol
Test plan: make check-all
Differential revision: https://reviews.llvm.org/D117614
Added some comments (particularly around finalize() and
finalizeContents()) as well as doing some rephrasing / grammar fixes for
existing comments.
Also did some minor style fixups, such as by putting methods together in
a class definition and having fields of similar types next to each
other.
Reviewed By: #lld-macho, oontvoo
Differential Revision: https://reviews.llvm.org/D118714
See the updated insert-before.test for the effects: many synthetic
sections are SHF_ALLOC|SHF_WRITE. If they are discarded, we don't want
to propagate their flags to subsequent output section descriptions.
`getFirstInputSection(sec) == nullptr` can technically be merged into
`isDiscardable` but I'd like to postpone that as not sharing code may give more
refactoring opportunity.
Depends on D118529.
Reviewed By: peter.smith, bluca
Differential Revision: https://reviews.llvm.org/D118530
adjustSectionsBeforeSorting updates some output section attributes
(alignment/flags) and removes discardable empty sections. When it is called,
INSERT commands have not been processed. Therefore the flags propagation rule
may not affect output sections defined in an INSERT command properly.
Fix this by moving processInsertCommands before adjustSectionsBeforeSorting.
adjustSectionsBeforeSorting is somewhat misnamed. The order between it and
sortInputSections does not matter. With the pass shuffle, the name of
adjustSectionsBeforeSorting becomes wrong. Therefore rename it. The new
name is not set into stone. The function mixes several tasks and the
code may be refactored in a way that we may give them more meaningful
names.
With this patch, I think the behavior of attribute propagation becomes more
reasonable. In particular, in the absence of non-INSERT SECTIONS,
inserting a section after a SHF_ALLOC one will give us a SHF_ALLOC section,
not a non-SHF_ALLOC one (see linkerscript/insert-after.test).
Reviewed By: peter.smith, bluca
Differential Revision: https://reviews.llvm.org/D118529
The deduplication requires a DenseMap of the same size of the local part of
.strtab . I optimized it in e205445434 but it is
still quite slow.
For Release build of clang, deduplication makes .strtab 1.1% smaller and makes the link 3% slower.
For chrome, deduplication makes .strtab 0.1% smaller and makes the link 6% slower.
I suggest that we only perform the optimization with -O2 (default is -O1).
Not deduplicating local symbol names will simplify parallel symbol table write.
Reviewed By: peter.smith
Differential Revision: https://reviews.llvm.org/D118577
Replace `f<ELFT>(x)` with `InvokeELFT(f, x)`.
The size reduction comes from turning `link` from 4 specializations into 1.
My x86-64 lld executable is 26KiB smaller.
Reviewed By: ikudrin
Differential Revision: https://reviews.llvm.org/D118551
Previously an InputSectionBase is dead (`partition==0`) by default.
SyntheticSection calls markLive and BssSection overrides that with markDead.
It is more natural to make InputSectionBase live by default and let
--gc-sections mark InputSectionBase dead.
When linking a Release build of clang:
* --no-gc-sections:, the removed `inputSections` loop decreases markLive time from 4ms to 1ms.
* --gc-sections: the extra `inputSections` loop increases markLive time from 0.181296s to 0.188526s.
This is as of we lose the removing one `inputSections` loop optimization (4374824ccf).
I believe the loss can be mitigated if we refactor markLive.
This is a ld64 option equivalent to `-sectcreate seg sect /dev/null`
that's useful for creating sections like the RESTRICT section.
Differential Revision: https://reviews.llvm.org/D117749
Previously functions that aren't included in the symtab were also
excluded from the function starts. Symbols missing from function starts
degrades the debugger experience in the case you don't have debug info
for them.
Differential Revision: https://reviews.llvm.org/D114275
Fangrui Song