Skip to content

S
swiftshader
Switch branch/tag
  1. 24 Feb, 2015 4 commits
  3. 20 Feb, 2015 2 commits
  5. 19 Feb, 2015 1 commit
  7. 13 Feb, 2015 5 commits
  9. 12 Feb, 2015 2 commits
    • Subzero: Use -filetype instead of -ias and -elf-writer. · d442e7ee
      Jim Stichnoth authored 
      This matches LLVM's -filetype={obj,asm} options, and also adds -filetype=iasm to get textual rendering with the integrated assembler.

It also removes the old illegal combination of -ias=0 -elf-writer.

BUG= none
R=jvoung@chromium.org, kschimpf@google.com

Review URL: https://codereview.chromium.org/920953002
    • Subzero: Emit functions and global initializers in a separate thread. · bbca754a
      Jim Stichnoth authored 
      (This is a continuation of https://codereview.chromium.org/876083007/ .)

Emission is done in a separate thread when -threads=N with N>0 is specified.  This includes both functions and global initializers.

Emission is deterministic.  The parser assigns sequence numbers, and the emitter thread reassembles work units into their original order, regardless of the number of threads.

Dump output, however, is not intended to be in deterministic, reassembled order.  As such, lit tests that test dump output (i.e., '-verbose inst') are explicitly run with -threads=0.

For -elf-writer and -ias=1, the translator thread invokes Cfg::emitIAS() and the assembler buffer is passed to the emitter thread.  For -ias=0, the translator thread passed the Cfg to the emitter thread which then invokes Cfg::emit() to produce the textual asm.

Minor cleanup along the way:
  * Removed Flags from the Ice::Translator object and ctor, since it was redundant with Ctx->getFlags().
  * Cfg::getAssembler<> is the same as Cfg::getAssembler<Assembler> and is useful for just passing the assembler around.
  * Removed the redundant Ctx argument from TargetDataLowering::lowerConstants() .

BUG= https://code.google.com/p/nativeclient/issues/detail?id=4075
R=jvoung@chromium.org

Review URL: https://codereview.chromium.org/916653004
  11. 10 Feb, 2015 1 commit
    • Fix PNaCl bitcode reader to release global variables to emitter. · 6ca7d2b6
      Karl Schimpf authored 
      Fixes the PNaCl bitcode reader to maintain two lists of global
variables. The first, VariableDeclarations, is the list of
variable declarations to be lowered by the emitter. The second,
ValueIDConstants, is the corresponding constant symbol to use
when references to the corresponding global variable declaration
is referenced when processing functions.

BUG=None
R=jvoung@chromium.org, stichnot@chromium.org

Review URL: https://codereview.chromium.org/883673005
  13. 09 Feb, 2015 1 commit
  15. 06 Feb, 2015 1 commit
  17. 05 Feb, 2015 1 commit
  19. 04 Feb, 2015 1 commit
    • Add comment for the forked Dart revision for the assembler code. · 33a5f41d
      Jan Voung authored 
      Also note to keep that up to date.
See also Patch set 1 of https://codereview.chromium.org/574133002/,
vs later patch sets.

Some things that were changed:
(*) Headers / constants use Ice version (RegX8632::Encoded_Reg_eax vs EAX), (KB / MB -> other...)
(*) Use llvm/Subzero allocator instead of Dart one.
(*) Class/Field/On-stack-replacement/Dart runtime stuff is removed
(*) Relocation/Fixups are now POD -- rather than a class
with a virtual method for fixup. For now, we write out
an ELF relocation, but later we may do a target pass
to handle function calls within the same section, etc.
(*) ASSERT -> assert
(*) uword -> uintptr_t (should check).
(*) clang-format
(*) ???

BUG=none
R=stichnot@chromium.org

Review URL: https://codereview.chromium.org/901453002
  21. 03 Feb, 2015 4 commits
  23. 01 Feb, 2015 2 commits
  25. 31 Jan, 2015 2 commits
    • Subzero: Fix stats collection and output for multithreading. · a1dd3cc8
      Jim Stichnoth authored 
      Updates of current-function and cumulative stats are done entirely in TLS.  At the end, cumulative stats are merged across all threads' TLS into the global cumulative stats.

Printing of cumulative stats after every function is removed, since there's very little value from that.  It was probably done in the first place just to give partial cumulative information in the face of crashes or assertion failures.

BUG= none
R=jfb@chromium.org

Review URL: https://codereview.chromium.org/887213002
    • Fix subzero Windows build · ae6e12ca
      JF Bastien authored 
      MinGW's GCC 4.8.1 was sad because SectionType was shadowing the other SectionType. Also, the enum's values are in the ELFObjectWriter namespace, not ELFObjectWriter::SectionType.

R=stichnot@chromium.org, jvoung@chromium.org
BUG= Windows build is sad

Review URL: https://codereview.chromium.org/891953002
  27. 30 Jan, 2015 2 commits
    • Subzero: Fix timers for multithreaded translation. · 380d7b96
      Jim Stichnoth authored 
      Now that multithreaded parsing and translation is in place, timer operations have to be made thread-local.  After the non-main threads end, their thread-local timer data needs to be merged into the global timer data, which resides in the GlobalContext object.  The merge is a bit tricky because the internal timer stack structure is built up dynamically as items are pushed and popped.  Two threads may have radically different timing data:

1. The parser thread profile is completely different from a translator thread.

2. For -timing-funcs, two translator threads hold data for entirely different sets of functions.

A bit more tweaking will need to be done to make the timing output fully usable in a multithreaded run.  Because of multiple threads, times may add up to >100%.  Also, time spent blocked is being "unfairly" attributed to the caller of the blocking operation - we should either count the user time instead of wall-clock time, or add a special timer marker for blocking locking operations.

BUG= none
R=jvoung@chromium.org

Review URL: https://codereview.chromium.org/878383004
    • Subzero: Minor Makefile fix. · 51d00936
      Jim Stichnoth authored 
      The problem showed up after the link step failed, in which case $(OBJDIR)/llvm2ice was deleted but the ./llvm2ice symlink still existed.  A subsequent "make check-lit" or "make check" would fail, so the basic "make" would have to be done first.

BUG= none
R=jvoung@chromium.org

Review URL: https://codereview.chromium.org/887873002
  29. 29 Jan, 2015 1 commit
    • Write out global initializers and data rel directly to ELF file. · 72984d88
      Jan Voung authored 
      The local symbol relocations are a bit different from
llvm-mc, which are section-relative. E.g., instead "bytes",
it will be ".data + offsetof(bytes, .data)". So the
contents of the text/data/rodata sections can also differ
since the offsets written in place are different.

Still need to fill the symbol table with undefined
symbols (e.g., memset, and szrt lib functions) before
trying to link.

BUG=none
R=kschimpf@google.com, stichnot@chromium.org

Review URL: https://codereview.chromium.org/874353006
  31. 28 Jan, 2015 6 commits
  33. 27 Jan, 2015 3 commits
    • Fix pedantic build warnings; · 8427ea2b
      JF Bastien authored 
      GCC 4.8.1 is sad;

There are extra semicolons in Subzero;

It removes the semicolons or it gets the build warning hose again;^H

R=stichnot@chromium.org
BUG= none

Review URL: https://codereview.chromium.org/882743003
    • Subzero: Use a "known" version of clang-format. · dd842dbb
      Jim Stichnoth authored 
      There are two problems with "make format" and "make format-diff" in
Makefile.standalone:

1. You have to make sure clang-format and clang-format-diff.py are
available in $PATH.

2. Different users may have different versions installed (even for the
same user on different machines), leading to whitespace wars.  Can't we
all just get along?

Since the normal LLVM build that Subzero depends on also exposes and
builds clang-format and friends, we might as well use it.  The
clang-format binary is found in $LLVM_BIN_PATH, and clang-format-diff.py
is found relative to $LLVM_SRC_PATH.  As long as the user's LLVM build
is fairly up to date, whitespace wars are unlikely.

Given this, there's a much higher incentive to use "make format"
regularly instead of "make format-diff".  In particular, inline comments
on variable/field declaration lists can get lined up more nicely by
looking at the entire context, rather than the small diff window.

BUG= none
R=jvoung@chromium.org

Review URL: https://codereview.chromium.org/877003003
    • Subzero: Initial implementation of multithreaded translation. · fa4efea5
      Jim Stichnoth authored 
      Provides a single-producer, multiple-consumer translation queue where the number of translation threads is given by the -threads=N argument.  The producer (i.e., bitcode parser) blocks if the queue size is >=N, in order to control the memory footprint.  If N=0 (which is the default), execution is purely single-threaded.  If N=1, there is a single translation thread running in parallel with the parser thread.  "make check" succeeds with the default changed to N=1.

Currently emission is also done by the translation thread, which limits scalability since the emit stream has to be locked.  Also, since the ELF writer stream is not locked, it won't be safe to use N>1 with the ELF writer.  Furthermore, for N>1, emitted function ordering is nondeterministic and needs to be recombobulated.  This will all be fixed in a follow-on CL.

The -timing option is broken for N>0.  This will be fixed in a follow-on CL.

Verbose flags are now managed in the Cfg instead of (or in addition to) the GlobalContext, due to the -verbose-focus option which wants to temporarily change the verbose level for a particular function.

TargetLowering::emitConstants() and related methods are changed to be static, so that a valid TargetLowering object isn't required.  This is because the TargetLowering object wants to hold a valid Cfg, and none really exists after all functions are translated and the constant pool is ready for emission.

The Makefile.standalone now has a TSAN=1 option to enable ThreadSanitizer.

BUG= none
R=jfb@chromium.org

Review URL: https://codereview.chromium.org/870653002
  35. 26 Jan, 2015 1 commit