.ARM.exidx sections have a reverse dependency on the section they have
a SHF_LINK_ORDER dependency on. In other words a .ARM.exidx section is
live only if the executable section it describes is live. We implement
this with a reverse dependency field in InputSection.
Adding the dependency to InputSection is the simplest implementation
but it could be moved out to a separate map if it were found to decrease
performance for non ARM targets.
Differential revision: https://reviews.llvm.org/D25234
llvm-svn: 283734
The .ARM.exidx sections contain a table. Each entry has two fields:
- PREL31 offset to the function the table entry describes
- Action to take, either cantunwind, inline unwind, or PREL31 offset to
.ARM.extab section
The table entries must be sorted in order of the virtual addresses the
first entry of the table describes. Traditionally this is implemented by
the SHF_LINK_ORDER dependency. Instead of implementing this directly we
sort the table entries post relocation.
The .ARM.exidx OutputSection is described by the PT_ARM_EXIDX program
header
Differential revision: https://reviews.llvm.org/D25127
llvm-svn: 283730
This spreads out computing the hash and using it in a hash table. The
speedups are:
firefox
master 6.811232891
patch 6.559280249 1.03841162939x faster
chromium
master 4.369323666
patch 4.33171853 1.00868134338x faster
chromium fast
master 1.856679971
patch 1.850617741 1.00327578725x faster
the gold plugin
master 0.32917962
patch 0.325711944 1.01064645023x faster
clang
master 0.558015452
patch 0.550284165 1.01404962652x faster
llvm-as
master 0.032563515
patch 0.032152077 1.01279662275x faster
the gold plugin fsds
master 0.356221362
patch 0.352772162 1.00977741549x faster
clang fsds
master 0.635096494
patch 0.627249229 1.01251060127x faster
llvm-as fsds
master 0.030183188
patch 0.029889544 1.00982430511x faster
scylla
master 3.071448906
patch 2.938484138 1.04524944215x faster
This seems to be because we don't stall as much. When linking firefox
stalled-cycles-frontend goes from 57.56% to 55.55%.
With -O2 the difference is even more significant since we avoid
recomputing the hash. For firefox we go from 9.990295265 to
9.149627521 seconds (1.09x faster).
llvm-svn: 283367
It is pretty easy to get the data from the InputSection, so we don't
have to store it.
This opens the way for storing the hash instead.
llvm-svn: 283357
This simplifies error handling as there is now only one place in the
code that needs to consider the possibility that the name is
corrupted. Before we would do it in every access.
llvm-svn: 280937
Previously we used LayoutInputSection class to correctly assign
symbols defined in linker script. This patch removes it and uses
pointer to preceding input section in SymbolAssignment class instead.
Differential revision: https://reviews.llvm.org/D23661
llvm-svn: 280348
This section supersedes .reginfo and .MIPS.options sections. But for now
we have to support all three sections for ABI transition period.
llvm-svn: 278482
All other singleton instances are accessible globally.
CommonInputSection shouldn't be an exception.
Differential Revision: https://reviews.llvm.org/D22935
llvm-svn: 277034
Not all relocations from a .eh_frame that point to an executable
section should be ignored. In particular, the relocation finding the
personality function should not.
This is a reduction from trying to bootstrap a static lld on linux.
llvm-svn: 276329
We no longer need it for relocations in .eh_frame.
The only relocations that point to .eh_frame are the ones trying to
find the output .eh_frame.
This actually fixes a bug in the symbol value code. It was not
handling -1 as an indicator for a piece not being included in the
output.
llvm-svn: 276175
Creating sections on linkerscript side requires some methods
that can be reused if are exported from writer.
Patch implements that change.
Differential revision: http://reviews.llvm.org/D20104
llvm-svn: 275162
The TinyPtrVector of const Thunk<ELFT>* in InputSections.h can cause
build failures on certain compiler/library combinations when Thunk<ELFT>
is not a complete type or is an abstract class. Fixed by making Thunk<ELFT>
non Abstract.
type or is an abstract class
llvm-svn: 274863
Generalise the Mips LA25 Thunk code and implement ARM and Thumb
interworking Thunks.
- Introduce a new module Thunks.cpp to store the Target Specific Thunk
implementations.
- DefinedRegular and Shared have a ThunkData field to record Thunk.
- A Target can have more than one type of Thunk.
- Support PC-relative calls to Thunks.
- Support Thunks to PLT entries.
- Existing Mips LA25 Thunk code integrated.
- Support for ARMv7A interworking Thunks.
Limitations:
- Only one Thunk per SymbolBody, this is sufficient for all currently
implemented Thunks.
- ARM thunks assume presence of V6T2 MOVT and MOVW instructions.
Differential revision: http://reviews.llvm.org/D21891
llvm-svn: 274836
Previously, ch_size was read in host byte order, so if a host and
a target are different in byte order, we would produce a corrupted
output.
llvm-svn: 274729
Patch implements support of zlib style compressed sections.
SHF_COMPRESSED flag is used to recognize that decompression is required.
After that decompression is performed and flag is removed from output.
Differential revision: http://reviews.llvm.org/D20272
llvm-svn: 273661
Peter Smith found while trying to support thunk creation for ARM that
LLD sometimes creates broken thunks for MIPS. The cause of the bug is
that we assign file offsets to input sections too early. We need to
create all sections and then assign section offsets because appending
thunks changes file offsets for all following sections.
This patch separates the pass to assign file offsets from thunk
creation pass. This effectively reverts r265673.
Differential Revision: http://reviews.llvm.org/D21598
llvm-svn: 273532
I think it is me who named these variables, but I always find that
they are slightly confusing because align is a verb.
Adding four letters is worth it.
llvm-svn: 272984
MergedInputSection::getOffset is the busiest function in LLD if string
merging is enabled and input files have lots of mergeable sections.
It is usually the case when creating executable with debug info,
so it is pretty common.
The reason why it is slow is because it has to do faily complex
computations. For non-mergeable sections, section contents are
contiguous in output, so in order to compute an output offset,
we only have to add the output section's base address to an input
offset. But for mergeable strings, section contents are split for
merging, so they are not contigous. We've got to do some lookups.
We used to do binary search on the list of section pieces.
It is slow because I think it's hostile to branch prediction.
This patch replaces it with hash table lookup. Seems it's working
pretty well. Below is "perf stat -r10" output when linking clang
with debug info. In this case this patch speeds up about 4%.
Before:
6584.153205 task-clock (msec) # 1.001 CPUs utilized ( +- 0.09% )
238 context-switches # 0.036 K/sec ( +- 6.59% )
0 cpu-migrations # 0.000 K/sec ( +- 50.92% )
1,067,675 page-faults # 0.162 M/sec ( +- 0.15% )
18,369,931,470 cycles # 2.790 GHz ( +- 0.09% )
9,640,680,143 stalled-cycles-frontend # 52.48% frontend cycles idle ( +- 0.18% )
<not supported> stalled-cycles-backend
21,206,747,787 instructions # 1.15 insns per cycle
# 0.45 stalled cycles per insn ( +- 0.04% )
3,817,398,032 branches # 579.786 M/sec ( +- 0.04% )
132,787,249 branch-misses # 3.48% of all branches ( +- 0.02% )
6.579106511 seconds time elapsed ( +- 0.09% )
After:
6312.317533 task-clock (msec) # 1.001 CPUs utilized ( +- 0.19% )
221 context-switches # 0.035 K/sec ( +- 4.11% )
1 cpu-migrations # 0.000 K/sec ( +- 45.21% )
1,280,775 page-faults # 0.203 M/sec ( +- 0.37% )
17,611,539,150 cycles # 2.790 GHz ( +- 0.19% )
10,285,148,569 stalled-cycles-frontend # 58.40% frontend cycles idle ( +- 0.30% )
<not supported> stalled-cycles-backend
18,794,779,900 instructions # 1.07 insns per cycle
# 0.55 stalled cycles per insn ( +- 0.03% )
3,287,450,865 branches # 520.799 M/sec ( +- 0.03% )
72,259,605 branch-misses # 2.20% of all branches ( +- 0.01% )
6.307411828 seconds time elapsed ( +- 0.19% )
Differential Revision: http://reviews.llvm.org/D20645
llvm-svn: 270999
This patch makes SectionPiece class 8 bytes smaller on platforms
on which pointer size is 8 bytes. Sean suggested in a post commit
review for r270340 that this could make a differentce, and it
actually is. Time to link clang (with debug info) improved from
6.725 seconds to 6.589 seconds or by about 2%.
Differential Revision: http://reviews.llvm.org/D20613
llvm-svn: 270717
scanReloc and the functions on which scanReloc depends is in total
more than 600 lines of code. Since scanReloc does not depend on Writer,
it is better to move it into a separate file.
Differential Revision: http://reviews.llvm.org/D20554
llvm-svn: 270606
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
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
Peter Smith