Previously, mergeable section's constructors did more than just
setting member variables; it split section contents into small
pieces. It is not always computationally cheap task because if
the section is a mergeable string section, it needs to scan the
entire section to split them by NUL characters.
If a section would be thrown away by GC, that cost ended up
being a waste of time. It is going to be larger problem if the
section is compressed -- the whole time to uncompress it and
split it up is going to be a waste.
Luckily, we can defer section splitting after GC. We just have
to remember which offsets are in use during GC and apply that later.
This patch implements it.
Differential Revision: http://reviews.llvm.org/D20516
llvm-svn: 270455
This patch adds Size member to SectionPiece so that getRangeAndSize
can just return a SectionPiece instead of a std::pair<SectionPiece *, uint_t>.
Also renamed the function.
llvm-svn: 270346
We were using std::pair to represents pieces of splittable section
contents. It hurt readability because "first" and "second" are not
meaningful. This patch give them names.
One more thing is that piecewise liveness information is stored to
the second element of the pair as a special value of output section
offset. It was confusing, so I defiend a new bit, "Live", in the
new struct.
llvm-svn: 270340
This makes it explicit that each R_RELAX_TLS_* is equivalent to some
other expression.
With this I think we are at a sweet spot for how much is done in
Target.cpp. I did experiment with moving *all* the value math out of it.
It has the advantage that we know the final value in target independent
code, but it gets quite verbose.
llvm-svn: 270277
This adds direct support for computing offsets from the thread pointer
for both variants. Of the architectures we support, variant 1 is used
only by aarch64 (but that doesn't seem to be documented anywhere.)
llvm-svn: 270243
New names reflect purpose of corresponding GOT entries better.
Both expression types related to entries allocated in the 'local'
part of MIPS GOT. R_MIPS_GOT_LOCAL_PAGE is for entries contain 'page'
addresses. R_MIPS_GOT_LOCAL is for entries contain 'full' address.
llvm-svn: 269597
We were previously using an output offset of -1 for both GC'd and tail
merged pieces. We need to distinguish these two cases in order to filter
GC'd symbols from the symbol table -- we were previously asserting when we
asked for the VA of a symbol pointing into a dead piece, which would end
up asking the tail merging string table for an offset even though we hadn't
initialized it properly.
This patch fixes the bug by using an offset of -1 to exclusively mean GC'd
pieces, using 0 for tail merges, and distinguishing the tail merge case from
an offset of 0 by asking the output section whether it is tail merge.
Differential Revision: http://reviews.llvm.org/D19953
llvm-svn: 268604
MIPS N64 ABI introduces .MIPS.options section which specifies miscellaneous
options to be applied to an object/shared/executable file. LLVM as well as
modern versions of GNU tools read and write the only type of the options -
ODK_REGINFO. It is exact copy of .reginfo section used by O32 ABI.
llvm-svn: 268485
Relocations against sections with no SHF_ALLOC bit are R_ABS relocations.
Currently we are creating Relocations vector for them, but that is wasteful.
This patch is to skip vector construction and to directly apply relocations
in place.
This patch seems to be pretty effective for large executables with debug info.
r266158 (Rafael's patch to change the way how we apply relocations) caused a
temporary performance degradation for such executables, but this patch makes
it even faster than before.
Time to link clang with debug info (output size is 1070 MB):
before r266158: 15.312 seconds (0%)
r266158: 17.301 seconds (+13.0%)
Head: 16.484 seconds (+7.7%)
w/patch: 13.166 seconds (-14.0%)
Differential Revision: http://reviews.llvm.org/D19645
llvm-svn: 267917
The fix is to handle local symbols referring to SHF_MERGE sections.
Original message:
GC entries of SHF_MERGE sections.
It is a fairly direct extension of the gc algorithm. For merge sections
instead of remembering just a live bit, we remember which offsets
were used.
This reduces the .rodata sections in chromium from 9648861 to 9477472
bytes.
llvm-svn: 267233
It is a fairly direct extension of the gc algorithm. For merge sections
instead of remembering just a live bit, we remember which offsets were
used.
This reduces the .rodata sections in chromium from 9648861 to 9477472
bytes.
llvm-svn: 267164
It turns out that this will read data from the section to properly
handle Elf_Rel implicit addends.
Sorry for the noise.
Original messages:
Try to fix Windows lld build.
Move getRelocTarget to ObjectFile.
It doesn't use anything from the InputSection.
llvm-svn: 267163
This requires adding a few more expression types, but is already a small
simplification. Having Writer.cpp know the exact expression will also
allow further simplifications.
llvm-svn: 266604
With this patch we use the first scan over the relocations to remember
the information we found about them: will them be relaxed, will a plt be
used, etc.
With that the actual relocation application becomes much simpler. That
is particularly true for the interfaces in Target.h.
This unfortunately means that we now do two passes over relocations for
non SHF_ALLOC sections. I think this can be solved by factoring out the
code that scans a single relocation. It can then be used both as a scan
that record info and for a dedicated direct relocation of non SHF_ALLOC
sections.
I also think it is possible to reduce the number of enum values by
representing a target with just an OutputSection and an offset (which
can be from the start or end).
This should unblock adding features like relocation optimizations.
llvm-svn: 266158
Our symbol representation was redundant, and some times would get out of
sync. It had an Elf_Sym, but some fields were copied to SymbolBody.
Different parts of the code were checking the bits in SymbolBody and
others were checking Elf_Sym.
There are two general approaches to fix this:
* Copy the required information and don't store and Elf_Sym.
* Don't copy the information and always use the Elf_Smy.
The second way sounds tempting, but has a big problem: we would have to
template SymbolBody. I started doing it, but it requires templeting
*everything* and creates a bit chicken and egg problem at the driver
where we have to find ELFT before we can create an ArchiveFile for
example.
As much as possible I compared the test differences with what gold and
bfd produce to make sure they are still valid. In most cases we are just
adding hidden visibility to a local symbol, which is harmless.
In most tests this is a small speedup. The only slowdown was scylla
(1.006X). The largest speedup was clang with no --build-id, -O3 or
--gc-sections (i.e.: focus on the relocations): 1.019X.
llvm-svn: 265293
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
The original comments were separated by new code that is irrelevant to
the comment. This patch moves the comment to the right place and update it.
llvm-svn: 264816
This simplifies a few things
* Read the value as early as possible, instead of passing a pointer to
the location.
* Print the warning for missing pair close to where we find out it is
missing.
* Don't pass the value to relocateOne.
llvm-svn: 264802
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
The patch does not reduce the size of the code but makes
InputSectionBase::relocate cleaner a bit.
Differential Revision: http://reviews.llvm.org/D18119
llvm-svn: 263381
which was reverted because included
unrelative changes by mistake.
Original commit message:
[ELF] - Change all messages to lowercase to be consistent.
That is directly opposite to http://reviews.llvm.org/D18045,
which was reverted.
This patch changes all messages to start from lowercase letter if
they were not before.
That is done to be consistent with clang.
Differential revision: http://reviews.llvm.org/D18085
llvm-svn: 263337
That is directly opposite to http://reviews.llvm.org/D18045,
which was reverted.
This patch changes all messages to start from lowercase letter if
they were not before.
That is done to be consistent with clang.
Differential revision: http://reviews.llvm.org/D18085
llvm-svn: 263252
It is really odd that Mips differentiates symbols that are born local
and those that become local because of hidden visibility. I don't know
enough mips to known if this is a bug or not.
llvm-svn: 263228
It was a badly specified hack for when a tls relocation should be
propagated to the dynamic relocation table.
This replaces it with a not as bad hack of saying that a local dynamic
tls relocation is never preempted.
I will try to remove even that second hack in the next patch.
llvm-svn: 262955
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
The rules for when we can relax tls relocations are target independent.
The only things that are target dependent are the relocation values.
llvm-svn: 262748
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
There was a known limitation for -r option:
relocations against local symbols were not supported.
For example rel[a].eh_frame sections contained relocations against sections
and that was not supported for -r before. Patch fixes that.
Differential review: http://reviews.llvm.org/D17813
llvm-svn: 262590
For shared libraries we allow any weak undefined symbol to eventually be
resolved, even if we never see a definition in another .so. This matches
the behavior when handling other undefined symbols in a shared library.
For executables, we require seeing a definition in a .so or resolve it
to zero. This is also similar to how non weak symbols are handled.
llvm-svn: 262017
R_MIPS_GOT16 relocation against local symbol requires index of a local
GOT entry which contains page address corresponds to sum of the symbol
address and addend. The addend in that case is calculated using addends
from the R_MIPS_GOT16 and paired R_MIPS_LO16 relocations.
Differential Revision: http://reviews.llvm.org/D17610
llvm-svn: 261930
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
-r, -relocatable - Generate relocatable output
Currently does not have support for files containing
relocation sections with entries that refer to local
symbols (like rel[a].eh_frame which refer to sections
and not to symbols)
Differential revision: http://reviews.llvm.org/D14382
llvm-svn: 261838
"Discarded" section is a marker for discarded sections, and we do not
use the instance except for checking its identity. In that sense, it
is just another type of a "null" pointer for InputSectionBase. So,
it doesn't have to be a real instance of InputSectionBase class.
In this patch, we no longer instantiate Discarded section but instead
use -1 as a pointer value. This eliminates a global variable which
needed initialization at startup.
llvm-svn: 261761
This is a preparation for ICF. If we merge two sections, we want to
align the merged section at the largest alignment requirement.
That means we want to update the alignment value, which was
impossible before this patch because Header is a const value.
llvm-svn: 261712
The previous names contained "Local" and "Current", but what we
are handling is always local and current, so they were redundant.
TlsIndex comes from "tls_index" struct that Ulrich Drepper is using
in this document to describe this data structure in GOT.
llvm-svn: 259852
Symbol does not need an entry i the 'global' part of GOT if it cannot be
preempted. So canBePreempted fully satisfies us at least for now.
llvm-svn: 259779
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
Rafael Espindola