Skip to content
Commits on May 30, 2016
  1. Reduce diff with mainstream.

    In the new sljt version, ppc_cache_flush() is guarded by
    SLJIT_CACHE_FLUSH_OWN_IMPL. We can keep is as long as we
    don't define SLJIT_CACHE_FLUSH_OWN_IMPL.
    alnsn committed May 30, 2016
  2. Adapt to new sljit.

    nakayama committed May 30, 2016
  3. regen

    nonaka committed May 30, 2016
  4. Added some Intel wireless devices from OpenBSD.

    nonaka committed May 30, 2016
  5. fix r1.409 mistake

    knakahara committed May 30, 2016
  6. Note that i386 needs a PIE-related cleandir. Should avoid having PR 5…

    …1174
    
    happen to anyone else. :-/
    dholland committed May 30, 2016
  7. further whitespace

    dholland committed May 30, 2016
  8. KNF

    dholland committed May 30, 2016
  9. Reindent.

    dholland committed May 30, 2016
  10. PR 51182 David Binderman: fix redundant conditional

    dholland committed May 30, 2016
Commits on May 29, 2016
  1. Update sljit entry.

    alnsn committed May 29, 2016
  2. Call the path for makewhatis _PATH_MAKEWHATIS instead of _PATH_WHATIS,

    for clarity.
    dholland committed May 29, 2016
  3. Add notes on how to lift the BUGS entry in the man page (about not

    supporting hardlinks) in case anyone thinks it's worth doing sometime.
    dholland committed May 29, 2016
  4. Add missing rcsid.

    dholland committed May 29, 2016
  5. Update version in the comment too.

    alnsn committed May 29, 2016
  6. Version bump after sljit import.

    New version of sljit is not binary compatible with the previous
    version.
    alnsn committed May 29, 2016
  7. Adapt to the new version of sljit@r313.

    alnsn committed May 29, 2016
  8. Resolve conflicts.

    alnsn committed May 29, 2016
  9. Switch to elf notes for amd64 instead of the old key=value list to de…

    …scribe the
    
      guest requirements and support.
    Add infrastructure to query the hypervisor about features support.
    For verbose boot, print the features suppoted by the hypervisor for this
      guest.
    bouyer committed May 29, 2016
  10. Import sljit version r313.

    Changes since the last import:
    
    r313 Fix a racing condition reported by Mozilla.
    r312 Support Android ARM64 cacheflush. Patch by Tavian Barnes.
    r311 Update opcode dump.
    r310 Better type specification for integer operations and conditional types.
    r309 Better type specification for long mul and divide operators.
    r308 Better type specification for floating point operations.
    r307 Change d and s postfixes to f64 and f32.
    r306 Change b,h,i postifxes to 8,16,32.
    r305 Fix a wrong immediate number. Reported by Michael McConville.
    r304 Make file patch by Michael McConville.
    r303 Landed changes created by Michael McConville.
    r302 Add support for cmov on x86.
    r301 Tile-Gx fixes by Walter Lee
    r300 Add WinCE support.
    r299 Add a tutorial for SLJIT. Contributed by Wen Xichang.
    r298 Improve integer division.
    r297 Start reworking the integer division.
    r296 Add an overview for SLJIT.
    r295 Small optimizations for x86 and ARM64.
    r294 Fix ARM64 SP alignment.
    r293 Add a new sljit_set_compiler_memory_error function.
    r292 Add missing argument on ARM32.
    r291 Minor fixes and NOINLINE support.
    r290 Support custom memory allocators.
    r289 Add SLJIT_DOUBLE_ALIGNMENT support and emit_op_custom size check.
    r288 Add suport for SLJIT_ARGUMENT_CHECKS on all architectures.
    r287 Add SLJIT_ARGUMENT_CHECKS compiler option and lots of cleanups.
    r286 Support the new naming on PPC and MIPS.
    r285 Support the new naming on ARM and SPARC.
    r284 Move the type letter to the beggining of the floating point opcodes.
    r283 Add i_ s_ d_ froms to compare types.
    r282 Fix an incorrect form of mul on x86-64.
    r281 Refactoring sljitConfigInternal.h header.
    r280 Remove variable locals offset.
    r279 Test skips are not reported if verbose is disabled.
    r278 Add options to sljit_emit_enter and sljit_set_context.
    r277 Reindexing opX opcodes.
    r276 Some comments are fixed and minor refactors were done.
    r275 Minor optimizations.
    r274 Extend the register set on ARM.
    r273 Extend the register set on PPC, MIPS, SPARC.
    r272 Allow remapping of any registers: not just scratch->saved, but saved->scratch as well.
    r271 Renaming floating point registers to support more of them.
    r270 Extend the register set on x86.
    r269 Renaming integer registers for preparing the support of any all machine registers.
    r268 Refactor a macro to work when debug is disabled.
    r267 Introduce architecture macros without 32/64 bit postfix.
    r266 The floating point conversion operators are finished on PPC, and they are supported on all architectures now.
    r265 FPU operations support locals access now. CONW is supported on PPC now.
    r264 Continue code refactoring, MOVS/MOVD are optimized on all architectures.
    r263 SPARC implementation of floating point operators.
    r262 Small refactorings.
    r261 MIPS implementation of floating point conversion operators.
    r260 ARM implementation of floating point operators.
    r259 x86 implementation of floating point conversion operators.
    r258 Refactoring sljit_emit_fop1 opcodes, inserting placeholders for new ones.
    alnsn committed May 29, 2016
  11. Import sljit version r313.

    Changes since the last import:
    
    r313 Fix a racing condition reported by Mozilla.
    r312 Support Android ARM64 cacheflush. Patch by Tavian Barnes.
    r311 Update opcode dump.
    r310 Better type specification for integer operations and conditional types.
    r309 Better type specification for long mul and divide operators.
    r308 Better type specification for floating point operations.
    r307 Change d and s postfixes to f64 and f32.
    r306 Change b,h,i postifxes to 8,16,32.
    r305 Fix a wrong immediate number. Reported by Michael McConville.
    r304 Make file patch by Michael McConville.
    r303 Landed changes created by Michael McConville.
    r302 Add support for cmov on x86.
    r301 Tile-Gx fixes by Walter Lee
    r300 Add WinCE support.
    r299 Add a tutorial for SLJIT. Contributed by Wen Xichang.
    r298 Improve integer division.
    r297 Start reworking the integer division.
    r296 Add an overview for SLJIT.
    r295 Small optimizations for x86 and ARM64.
    r294 Fix ARM64 SP alignment.
    r293 Add a new sljit_set_compiler_memory_error function.
    r292 Add missing argument on ARM32.
    r291 Minor fixes and NOINLINE support.
    r290 Support custom memory allocators.
    r289 Add SLJIT_DOUBLE_ALIGNMENT support and emit_op_custom size check.
    r288 Add suport for SLJIT_ARGUMENT_CHECKS on all architectures.
    r287 Add SLJIT_ARGUMENT_CHECKS compiler option and lots of cleanups.
    r286 Support the new naming on PPC and MIPS.
    r285 Support the new naming on ARM and SPARC.
    r284 Move the type letter to the beggining of the floating point opcodes.
    r283 Add i_ s_ d_ froms to compare types.
    r282 Fix an incorrect form of mul on x86-64.
    r281 Refactoring sljitConfigInternal.h header.
    r280 Remove variable locals offset.
    r279 Test skips are not reported if verbose is disabled.
    r278 Add options to sljit_emit_enter and sljit_set_context.
    r277 Reindexing opX opcodes.
    r276 Some comments are fixed and minor refactors were done.
    r275 Minor optimizations.
    r274 Extend the register set on ARM.
    r273 Extend the register set on PPC, MIPS, SPARC.
    r272 Allow remapping of any registers: not just scratch->saved, but saved->scratch as well.
    r271 Renaming floating point registers to support more of them.
    r270 Extend the register set on x86.
    r269 Renaming integer registers for preparing the support of any all machine registers.
    r268 Refactor a macro to work when debug is disabled.
    r267 Introduce architecture macros without 32/64 bit postfix.
    r266 The floating point conversion operators are finished on PPC, and they are supported on all architectures now.
    r265 FPU operations support locals access now. CONW is supported on PPC now.
    r264 Continue code refactoring, MOVS/MOVD are optimized on all architectures.
    r263 SPARC implementation of floating point operators.
    r262 Small refactorings.
    r261 MIPS implementation of floating point conversion operators.
    r260 ARM implementation of floating point operators.
    r259 x86 implementation of floating point conversion operators.
    r258 Refactoring sljit_emit_fop1 opcodes, inserting placeholders for new ones.
    alnsn committed May 29, 2016
  12. Import sljit version r313.

    Changes since the last import:
    
    r313 Fix a racing condition reported by Mozilla.
    r312 Support Android ARM64 cacheflush. Patch by Tavian Barnes.
    r311 Update opcode dump.
    r310 Better type specification for integer operations and conditional types.
    r309 Better type specification for long mul and divide operators.
    r308 Better type specification for floating point operations.
    r307 Change d and s postfixes to f64 and f32.
    r306 Change b,h,i postifxes to 8,16,32.
    r305 Fix a wrong immediate number. Reported by Michael McConville.
    r304 Make file patch by Michael McConville.
    r303 Landed changes created by Michael McConville.
    r302 Add support for cmov on x86.
    r301 Tile-Gx fixes by Walter Lee
    r300 Add WinCE support.
    r299 Add a tutorial for SLJIT. Contributed by Wen Xichang.
    r298 Improve integer division.
    r297 Start reworking the integer division.
    r296 Add an overview for SLJIT.
    r295 Small optimizations for x86 and ARM64.
    r294 Fix ARM64 SP alignment.
    r293 Add a new sljit_set_compiler_memory_error function.
    r292 Add missing argument on ARM32.
    r291 Minor fixes and NOINLINE support.
    r290 Support custom memory allocators.
    r289 Add SLJIT_DOUBLE_ALIGNMENT support and emit_op_custom size check.
    r288 Add suport for SLJIT_ARGUMENT_CHECKS on all architectures.
    r287 Add SLJIT_ARGUMENT_CHECKS compiler option and lots of cleanups.
    r286 Support the new naming on PPC and MIPS.
    r285 Support the new naming on ARM and SPARC.
    r284 Move the type letter to the beggining of the floating point opcodes.
    r283 Add i_ s_ d_ froms to compare types.
    r282 Fix an incorrect form of mul on x86-64.
    r281 Refactoring sljitConfigInternal.h header.
    r280 Remove variable locals offset.
    r279 Test skips are not reported if verbose is disabled.
    r278 Add options to sljit_emit_enter and sljit_set_context.
    r277 Reindexing opX opcodes.
    r276 Some comments are fixed and minor refactors were done.
    r275 Minor optimizations.
    r274 Extend the register set on ARM.
    r273 Extend the register set on PPC, MIPS, SPARC.
    r272 Allow remapping of any registers: not just scratch->saved, but saved->scratch as well.
    r271 Renaming floating point registers to support more of them.
    r270 Extend the register set on x86.
    r269 Renaming integer registers for preparing the support of any all machine registers.
    r268 Refactor a macro to work when debug is disabled.
    r267 Introduce architecture macros without 32/64 bit postfix.
    r266 The floating point conversion operators are finished on PPC, and they are supported on all architectures now.
    r265 FPU operations support locals access now. CONW is supported on PPC now.
    r264 Continue code refactoring, MOVS/MOVD are optimized on all architectures.
    r263 SPARC implementation of floating point operators.
    r262 Small refactorings.
    r261 MIPS implementation of floating point conversion operators.
    r260 ARM implementation of floating point operators.
    r259 x86 implementation of floating point conversion operators.
    r258 Refactoring sljit_emit_fop1 opcodes, inserting placeholders for new ones.
    alnsn committed May 29, 2016
  13. Import sljit version r313.

    Changes since the last import:
    
    r313 Fix a racing condition reported by Mozilla.
    r312 Support Android ARM64 cacheflush. Patch by Tavian Barnes.
    r311 Update opcode dump.
    r310 Better type specification for integer operations and conditional types.
    r309 Better type specification for long mul and divide operators.
    r308 Better type specification for floating point operations.
    r307 Change d and s postfixes to f64 and f32.
    r306 Change b,h,i postifxes to 8,16,32.
    r305 Fix a wrong immediate number. Reported by Michael McConville.
    r304 Make file patch by Michael McConville.
    r303 Landed changes created by Michael McConville.
    r302 Add support for cmov on x86.
    r301 Tile-Gx fixes by Walter Lee
    r300 Add WinCE support.
    r299 Add a tutorial for SLJIT. Contributed by Wen Xichang.
    r298 Improve integer division.
    r297 Start reworking the integer division.
    r296 Add an overview for SLJIT.
    r295 Small optimizations for x86 and ARM64.
    r294 Fix ARM64 SP alignment.
    r293 Add a new sljit_set_compiler_memory_error function.
    r292 Add missing argument on ARM32.
    r291 Minor fixes and NOINLINE support.
    r290 Support custom memory allocators.
    r289 Add SLJIT_DOUBLE_ALIGNMENT support and emit_op_custom size check.
    r288 Add suport for SLJIT_ARGUMENT_CHECKS on all architectures.
    r287 Add SLJIT_ARGUMENT_CHECKS compiler option and lots of cleanups.
    r286 Support the new naming on PPC and MIPS.
    r285 Support the new naming on ARM and SPARC.
    r284 Move the type letter to the beggining of the floating point opcodes.
    r283 Add i_ s_ d_ froms to compare types.
    r282 Fix an incorrect form of mul on x86-64.
    r281 Refactoring sljitConfigInternal.h header.
    r280 Remove variable locals offset.
    r279 Test skips are not reported if verbose is disabled.
    r278 Add options to sljit_emit_enter and sljit_set_context.
    r277 Reindexing opX opcodes.
    r276 Some comments are fixed and minor refactors were done.
    r275 Minor optimizations.
    r274 Extend the register set on ARM.
    r273 Extend the register set on PPC, MIPS, SPARC.
    r272 Allow remapping of any registers: not just scratch->saved, but saved->scratch as well.
    r271 Renaming floating point registers to support more of them.
    r270 Extend the register set on x86.
    r269 Renaming integer registers for preparing the support of any all machine registers.
    r268 Refactor a macro to work when debug is disabled.
    r267 Introduce architecture macros without 32/64 bit postfix.
    r266 The floating point conversion operators are finished on PPC, and they are supported on all architectures now.
    r265 FPU operations support locals access now. CONW is supported on PPC now.
    r264 Continue code refactoring, MOVS/MOVD are optimized on all architectures.
    r263 SPARC implementation of floating point operators.
    r262 Small refactorings.
    r261 MIPS implementation of floating point conversion operators.
    r260 ARM implementation of floating point operators.
    r259 x86 implementation of floating point conversion operators.
    r258 Refactoring sljit_emit_fop1 opcodes, inserting placeholders for new ones.
    alnsn committed May 29, 2016
  14. Import sljit version r313.

    Changes since the last import:
    
    r313 Fix a racing condition reported by Mozilla.
    r312 Support Android ARM64 cacheflush. Patch by Tavian Barnes.
    r311 Update opcode dump.
    r310 Better type specification for integer operations and conditional types.
    r309 Better type specification for long mul and divide operators.
    r308 Better type specification for floating point operations.
    r307 Change d and s postfixes to f64 and f32.
    r306 Change b,h,i postifxes to 8,16,32.
    r305 Fix a wrong immediate number. Reported by Michael McConville.
    r304 Make file patch by Michael McConville.
    r303 Landed changes created by Michael McConville.
    r302 Add support for cmov on x86.
    r301 Tile-Gx fixes by Walter Lee
    r300 Add WinCE support.
    r299 Add a tutorial for SLJIT. Contributed by Wen Xichang.
    r298 Improve integer division.
    r297 Start reworking the integer division.
    r296 Add an overview for SLJIT.
    r295 Small optimizations for x86 and ARM64.
    r294 Fix ARM64 SP alignment.
    r293 Add a new sljit_set_compiler_memory_error function.
    r292 Add missing argument on ARM32.
    r291 Minor fixes and NOINLINE support.
    r290 Support custom memory allocators.
    r289 Add SLJIT_DOUBLE_ALIGNMENT support and emit_op_custom size check.
    r288 Add suport for SLJIT_ARGUMENT_CHECKS on all architectures.
    r287 Add SLJIT_ARGUMENT_CHECKS compiler option and lots of cleanups.
    r286 Support the new naming on PPC and MIPS.
    r285 Support the new naming on ARM and SPARC.
    r284 Move the type letter to the beggining of the floating point opcodes.
    r283 Add i_ s_ d_ froms to compare types.
    r282 Fix an incorrect form of mul on x86-64.
    r281 Refactoring sljitConfigInternal.h header.
    r280 Remove variable locals offset.
    r279 Test skips are not reported if verbose is disabled.
    r278 Add options to sljit_emit_enter and sljit_set_context.
    r277 Reindexing opX opcodes.
    r276 Some comments are fixed and minor refactors were done.
    r275 Minor optimizations.
    r274 Extend the register set on ARM.
    r273 Extend the register set on PPC, MIPS, SPARC.
    r272 Allow remapping of any registers: not just scratch->saved, but saved->scratch as well.
    r271 Renaming floating point registers to support more of them.
    r270 Extend the register set on x86.
    r269 Renaming integer registers for preparing the support of any all machine registers.
    r268 Refactor a macro to work when debug is disabled.
    r267 Introduce architecture macros without 32/64 bit postfix.
    r266 The floating point conversion operators are finished on PPC, and they are supported on all architectures now.
    r265 FPU operations support locals access now. CONW is supported on PPC now.
    r264 Continue code refactoring, MOVS/MOVD are optimized on all architectures.
    r263 SPARC implementation of floating point operators.
    r262 Small refactorings.
    r261 MIPS implementation of floating point conversion operators.
    r260 ARM implementation of floating point operators.
    r259 x86 implementation of floating point conversion operators.
    r258 Refactoring sljit_emit_fop1 opcodes, inserting placeholders for new ones.
    alnsn committed May 29, 2016
  15. Don't make crunch binaries PIE

    christos committed May 29, 2016
  16. Don't build PIE binaries for install media

    christos committed May 29, 2016
Something went wrong with that request. Please try again.