DefinedElf was a superclass of DefinedRegular and SharedSymbol classes
and represented the notion of defined symbols created for ELF symbols.
It turned out that we didn't use that class often. We had only two
occurrences of dyn_cast'ing to DefinedElf, and both were easily
rewritten without it.
The class was also a bit confusing. The concept of "created for ELF
symbol" is orthogonal to defined/undefined types. However, we had
two distinct classes, DefinedElf and UndefinedElf.
This patch simply removes the class. Now the class hierarchy is one
level shallower.
llvm-svn: 265234
Some targets might require creation of thunks. For example, MIPS targets
require stubs to call PIC code from non-PIC one. The patch implements
infrastructure for thunk code creation and provides support for MIPS
LA25 stubs. Any MIPS PIC code function is invoked with its address
in register $t9. So if we have a branch instruction from non-PIC code
to the PIC one we cannot make the jump directly and need to create a small
stub to save the target function address.
See page 3-38 ftp://www.linux-mips.org/pub/linux/mips/doc/ABI/mipsabi.pdf
- In relocation scanning phase we ask target about thunk creation necessity
by calling `TagetInfo::needsThunk` method. The `InputSection` class
maintains list of Symbols requires thunk creation.
- Reassigning offsets performed for each input sections after relocation
scanning complete because position of each section might change due
thunk creation.
- The patch introduces new dedicated value for DefinedSynthetic symbols
DefinedSynthetic::SectionEnd. Synthetic symbol with that value always
points to the end of the corresponding output section. That allows to
escape updating synthetic symbols if output sections sizes changes after
relocation scanning due thunk creation.
- In the `InputSection::writeTo` method we write thunks after corresponding
input section. Each thunk is written by calling `TargetInfo::writeThunk` method.
- The patch supports the only type of thunk code for each target. For now,
it is enough.
Differential Revision: http://reviews.llvm.org/D17934
llvm-svn: 265059
Some optimizations, e.g. SimplifyLibCalls, can replace functions with
others as part of the lowering, e.g. printf => puts.
The new symbols don't have the isUsedInRegularObj flag set so they
don't get included in the final symbol table (and dynamic symbol
table), and the dynamic linker gets confused. Include them as a fix.
Differential Revision: http://reviews.llvm.org/D18357
llvm-svn: 264688
IPO doesn't work very well across symbols referenced
by others TUs. The linker here tries to evaluate
which symbols are safe to internalize and switches
their linkage.
Differential Revision: http://reviews.llvm.org/D18415
llvm-svn: 264585
We want to make SymbolBody the central place to query symbol information.
This patch also renames canBePreempted to isPreemptible because I feel that
the latter is slightly better (the former is three words and the latter
is two words.)
llvm-svn: 263386
We can argue about a maximum alignment of a group of symbols,
but for each symbol, there is only one alignment.
So it is a bit weird that each symbol has a "maximum alignment".
llvm-svn: 263151
It looks a bit wierd that we have to initialize symbols for all ELFT
types when we use only one ELFT for link. We can only init those
that we need. Patch fixes it.
Differential revision: http://reviews.llvm.org/D18047
llvm-svn: 263133
Get rid of few accessors in that class, and replace
them with direct fields access.
Differential revision: http://reviews.llvm.org/D17879
llvm-svn: 262796
SymbolBody constructor and friends take isFunc and isTLS boolean arguments.
ELF symbols have already a type so than be easily passed as argument.
If we want to support another type, this scheme is not good enough, that is,
the current code logic would require passing another `bool isObject` around.
Up to two argument, this stretching exercise was a little bit goofy but
still acceptable, but with more types to support, is just too much, IMHO.
Change the code so that the type is passed instead.
Differential Revision: http://reviews.llvm.org/D17871
llvm-svn: 262684
https://docs.oracle.com/cd/E53394_01/html/E54766/u-etext-3c.html
It is said that:
_etext - The address of _etext is the first
location after the last read-only loadable segment.
_edata - The address of _edata is the first
location after the last read-write loadable segment.
_end - If the address of _edata is greater than the address
of _etext, the address of _end is same as the address of _edata.
In real life _end and _edata has different values for that case.
Both gold/bfd set _edata to the end of the last non SHT_NOBITS section.
This patch do the same for consistency.
It should fix the https://llvm.org/bugs/show_bug.cgi?id=26729.
Differential revision: http://reviews.llvm.org/D17601
llvm-svn: 262019
This patch implements the same algorithm as LLD/COFF's ICF. I'm
not going to repeat the same description about how it works, so you
want to read the comment in ICF.cpp in this patch if you want to know
the details. This algorithm should be more powerful than the ICF
algorithm implemented in GNU gold. It can even merge mutually-recursive
functions (which is harder than one might think).
ICF is a fairly effective size optimization. Here are some examples.
LLD: 37.14 MB -> 35.80 MB (-3.6%)
Clang: 59.41 MB -> 57.80 MB (-2.7%)
The lacking feature is "safe" version of ICF. This merges all
identical sections. That is not compatible with a C/C++ language
requirement that two distinct functions must have distinct addresses.
But as long as your program do not rely on the pointer equality
(which is in many cases true), your program should work with the
feature. LLD works fine for example.
GNU gold implements so-called "safe ICF" that identifies functions
that are safe to merge by heuristics -- for example, gold thinks
that constructors are safe to merge because there is no way to
take an address of a constructor in C++. We have a different idea
which David Majnemer suggested that we add NOPs at beginning of
merged functions so that two or more pointers can have distinct
values. We can do whichever we want, but this patch does not
include neither.
http://reviews.llvm.org/D17529
llvm-svn: 261912
This is the function equivalent of a copy relocation.
Since functions are expected to change sizes, we cannot use copy
relocations. In situations where one would be needed, what is done
instead is:
* Create a plt entry
* Output an undefined symbol whose addr is the plt entry.
The dynamic linker makes sure any shared library uses the plt entry as
the function address.
llvm-svn: 260224
The variable was marking various cases where a symbol must be included
in the dynamic symbol table. Being used by a dynamic relocation was only
one of them.
llvm-svn: 259889
-Bsymbolic-functions:
When creating a shared library, bind references to global
function symbols to the definition within the shared library, if any.
This patch also fixed behavior of already existent -Bsymbolic:
previously PLT entries were created even if -Bsymbolic was specified.
Differential revision: http://reviews.llvm.org/D16411
llvm-svn: 259481
Previously, the methods to get symbol addresses were somewhat scattered
in many places. You can use getEntryAddr returns the address of the symbol,
but if you want to get the GOT address for the symbol, you needed to call
Out<ELFT>::Got->getEntryAddr(Sym). This change adds new functions, getVA,
getGotVA, getGotPltVA, and getPltVA to SymbolBody, so that you can use
SymbolBody as the central place to ask about symbols.
http://reviews.llvm.org/D16710
llvm-svn: 259404
On MIPS O32 ABI, _gp_disp is a magic symbol designates offset between
start of function and gp pointer into GOT. To make seal with such symbol
we add new method addIgnoredStrong(). It adds ignored symbol with global
binding to prevent the symbol substitution. The addIgnored call is not
enough here because this call adds a weak symbol which might be
substituted by symbol from shared library.
Differential Revision: http://reviews.llvm.org/D16084
llvm-svn: 257449
In this patch, all symbols are resolved normally and then wrap options
are applied. Renaming is implemented by mutating `Body` pointers of
Symbols. (As a result, Symtab.find(SymbolName)->getName() may return
a string that's different from SymbolName, but that is by design.
I designed the symbol and the symbol table to allow this kind of
operations.)
http://reviews.llvm.org/D15896
llvm-svn: 257075
There are 3 symbol types that a .bc can provide during lto: defined,
undefined, common.
Defined and undefined symbols have already been refactored. I was
working on common and noticed that absolute symbols would become an
oddity: They would be the only symbol type present in a .o but not in
a.bc.
Looking a bit more, other than the special section number they were only
used for special rules for computing values. In that way they are
similar to TLS, and we don't have a DefinedTLS.
This patch deletes it. With it we have a reasonable rule of the thumb
for having a symbol kind: It exists if it has special resolution
semantics.
llvm-svn: 256383
I am working on adding LTO support to the new ELF lld.
In order to do that, it will be necessary to represent defined and
undefined symbols that are not from ELF files. One way to do it is to
change the symbol hierarchy to look like
Defined : SymbolBody
Undefined : SymbolBody
DefinedElf<ELFT> : Defined
UndefinedElf<ELFT> : Undefined
Another option would be to use bogus Elf_Sym, but I think that is
getting a bit too hackish.
This patch does the Undefined/UndefinedElf. Split. The next one
will do the Defined/DefinedElf split.
llvm-svn: 256289
This relocation is similar to R_*_RELATIVE except that the value used in this relocation is the program address returned by the function, which takes no arguments, at the address of
the result of the corresponding R_*_RELATIVE relocation as specified in the processor-specific ABI. The purpose of this relocation to avoid name lookup for locally defined STT_GNU_IFUNC symbols at load-time.
More info can be found in ifunc.txt from https://sites.google.com/site/x32abi/documents.
Differential revision: http://reviews.llvm.org/D15235
llvm-svn: 256144
Previously, OffsetInBSS is -1 if it has no information about copy
relocation, 0 if it needs a copy relocation, and >0 if its offset
in BSS has been assigned. These flags were too subtle. This patch
adds a new flag, NeedsCopy, to carry information about whether
a shared symbol needs a copy relocation or not.
llvm-svn: 255865
Combination of @tlsgd and @gottpoff at the same time leads to miss of R_X86_64_TPOFF64 dynamic relocation. Patch fixes that.
@tlsgd(%rip) - Allocate two contiguous entries in the GOT to hold a tls index
structure (for passing to tls get addr).
@gottpoff(%rip) - Allocate one GOT entry to hold a variable offset in initial TLS
block (relative to TLS block end, %fs:0).
The same situation can be observed for x86 (probably others too, not sure) with corresponding for that target relocations: @tlsgd, @gotntpoff.
Differential revision: http://reviews.llvm.org/D15105
llvm-svn: 254443
This patch implements R_MIPS_GOT16 relocation for global symbols in order to
generate some entries in GOT. Only reserved and global entries are supported
for now. For the detailed description about GOT in MIPS, see "Global Offset
Table" in Chapter 5 in the followin document:
ftp://www.linux-mips.org/pub/linux/mips/doc/ABI/mipsabi.pdf
In addition, the platform specific symbol "_gp" is added, see "Global Data
Symbols" in Chapter 6 in the aforementioned document.
Differential revision: http://reviews.llvm.org/D14211
llvm-svn: 252275
Section garbage collection is a feature to remove unused sections
from outputs. Unused sections are sections that cannot be reachable
from known GC-root symbols or sections. Naturally the feature is
implemented as a mark-sweep garbage collector.
In this patch, I added Live bit to InputSectionBase. If and only
if Live bit is on, the section will be written to the output.
Starting from GC-root symbols or sections, a new function, markLive(),
visits all reachable sections and sets their Live bits. Writer then
ignores sections whose Live bit is off, so that such sections are
excluded from the output.
This change has small negative impact on performance if you use
the feature because making sections means more work. The time to
link Clang changes from 0.356s to 0.386s, or +8%.
It reduces Clang size from 57,764,984 bytes to 55,296,600 bytes.
That is 4.3% reduction.
http://reviews.llvm.org/D13950
llvm-svn: 251043
Target has supportsLazyRelocations() method which can switch lazy relocations on/off (currently all targets are OFF except x64 which is ON). So no any other targets are affected now.
Differential Revision: http://reviews.llvm.org/D13856?id=37726
llvm-svn: 250808
This patch is to use ELFT instead of Is64Bits to template OutputSection
and its subclasses. This increases code size slightly because it creates
two identical functions for some classes, but that's only 20 KB out of
33 MB, so it's negligible.
This is as per discussion with Rafael. He's not fan of the idea but OK
with this. We'll revisit later to this topic.
llvm-svn: 250466
What was done:
1) .got.plt section is created for functions that requires PLT. .got.plt has 3 predefined empty entries now that are required for dynamic linker.
Also other new items created are configured to have correct jump to PLT[N].
2) PLT section now has PLT[0] entry, also others ones are configured to support PLT->GOT(.got.plt) calls.
3) Implemented .rel[a].plt sections (based on patch http://reviews.llvm.org/D13569).
4) Fixed plt relocations types (based on patch http://reviews.llvm.org/D13589).
NOTES:
The .plt.got zero entry is still empty now. According to ELF specification it should hold the address of the dynamic structure, referenced with the symbol
_DYNAMIC. The _DYNAMIC entry points to the .dynamic section which contains information used by the ELF interpreter to setup the binary.
Differential Revision: http://reviews.llvm.org/D13651
llvm-svn: 250169
This patch adds AsNeeded and IsUsed bool fields to SharedFile. AsNeeded bit
is set if the DSO is enclosed with --as-needed and --no-as-needed. IsUsed
bit is off by default. When we adds a symbol to the symbol table for dynamic
linking, we set its SharedFile's IsUsed bit.
If AsNeeded is set but IsUsed is not set, we don't want to write that
file's SO name to DT_NEEDED field.
http://reviews.llvm.org/D13579
llvm-svn: 249998
This is a case that requires --start-group --end-group with regular ELF
linkers. Fortunately it is still possible to handle it with lazy symbols without
taking a second look at archives.
Thanks to Michael Spencer for the bug report.
llvm-svn: 249406
Add symbol specified with -u as undefined which may cause additional
object files from archives to be linked into the resulting binary.
Differential Revision: http://reviews.llvm.org/D13345
llvm-svn: 249295
This is more consistent with OutputSection. This is also part of removing
the "Chunk" term from the ELF linker, since we just have input/output sections
and program headers.
llvm-svn: 248183
The ELF spec says:
... if any reference to or definition of a name is a symbol with a
non-default visibility attribute, the visibility attribute must be
propagated to the resolving symbol in the linked object. If different
visibility attributes are specified for distinct references to or
definitions of a symbol, the most constraining visibility attribute
must be propagated to the resolving symbol in the linked object. The
attributes, ordered from least to most constraining, are:
STV_PROTECTED, STV_HIDDEN and STV_INTERNAL.
llvm-svn: 246603
Now that we print a symbol table and all symbol kinds are at least declared,
we can test all combinations that don't produce an error.
This also includes a few fixes to keep the test passing:
* Keep the strong symbol in a weak X strong pair
* Handle common symbols.
The common X common case will be finished in a followup patch.
llvm-svn: 246401
This is a direct port of the new PE/COFF linker to ELF.
It can take a single object file and generate a valid executable that executes at the first byte in the text section.
llvm-svn: 242088
Rui Ueyama