Programming Languages Research Group: Git - oota-llvm.git/history

[oota-llvm.git] / lib / Target / X86 / X86InstrCompiler.td

2014-10-08	Robin Morisset	[X86] Avoid generating inc/dec when slow for x.atomic_s...	blob \| commitdiff \| raw
2014-09-22	Pavel Chupin	[x32] Fix segmented stacks support	blob \| commitdiff \| raw \| diff to current
2014-09-02	Robin Morisset	[X86] Allow atomic operations using immediates to avoid...	blob \| commitdiff \| raw \| diff to current
2014-08-04	Reid Kleckner	Fix failure to invoke exception handler on Win64	blob \| commitdiff \| raw \| diff to current
2014-08-01	Akira Hatanaka	[X86] Simplify X87 stackifier pass.	blob \| commitdiff \| raw \| diff to current
2014-07-15	Cameron McInally	Revert r213070. It's breaking the build in MCELFStreame...	blob \| commitdiff \| raw \| diff to current
2014-07-15	Cameron McInally	Add x86 patterns to match a specific add-with-carry.	blob \| commitdiff \| raw \| diff to current
2014-07-01	Tim Northover	X86: expand atomics in IR instead of as MachineInstrs.	blob \| commitdiff \| raw \| diff to current
2014-06-25	NAKAMURA Takumi	Re-apply r211399, "Generate native unwind info on Win64...	blob \| commitdiff \| raw \| diff to current
2014-06-25	NAKAMURA Takumi	Reformat.	blob \| commitdiff \| raw \| diff to current
2014-06-22	NAKAMURA Takumi	Revert r211399, "Generate native unwind info on Win64"	blob \| commitdiff \| raw \| diff to current
2014-06-20	Reid Kleckner	Generate native unwind info on Win64	blob \| commitdiff \| raw \| diff to current
2014-06-09	Alexey Volkov	[X86] Use ADD/SUB instead of INC/DEC for Silvermont	blob \| commitdiff \| raw \| diff to current
2014-05-14	Jay Foad	Rename ComputeMaskedBits to computeKnownBits. "Masked...	blob \| commitdiff \| raw \| diff to current
2014-03-12	Adam Nemet	[X86] Add peephole for masked rotate amount	blob \| commitdiff \| raw \| diff to current
2014-03-12	Adam Nemet	[X86] Refactor peepholes for masked shift amount into...	blob \| commitdiff \| raw \| diff to current
2014-03-11	Jim Grosbach	X86: Enable ISel of 16-bit MOVBE instructions.	blob \| commitdiff \| raw \| diff to current
2014-02-02	Craig Topper	Merge x86 HasOpSizePrefix/HasOpSize16Prefix into a...	blob \| commitdiff \| raw \| diff to current
2014-01-08	David Woodhouse	[x86] Remove OpSize16 flag from MOV32r0	blob \| commitdiff \| raw \| diff to current
2014-01-08	David Woodhouse	[x86] Add OpSize16 to instructions that need it	blob \| commitdiff \| raw \| diff to current
2014-01-05	Craig Topper	Remove opcode from MOV32r0 that I accidentally left...	blob \| commitdiff \| raw \| diff to current
2013-12-31	Craig Topper	Handle MOV32r0 in expandPostRAPseudo instead of MCInst...	blob \| commitdiff \| raw \| diff to current
2013-12-20	Eric Christopher	[x86] Rename In32BitMode predicate to Not64BitMode	blob \| commitdiff \| raw \| diff to current
2013-12-17	Duncan P. N. Exon...	Revert "Revert "Mark vastart_save_xmm_regs as changing...	blob \| commitdiff \| raw \| diff to current
2013-12-17	Duncan P. N. Exon...	Revert "Mark vastart_save_xmm_regs as changing EFLAGS"	blob \| commitdiff \| raw \| diff to current
2013-12-17	Duncan P. N. Exon...	Mark vastart_save_xmm_regs as changing EFLAGS	blob \| commitdiff \| raw \| diff to current
2013-10-31	Elena Demikhovsky	AVX-512: Implemented CMOV for 512-bit vectors	blob \| commitdiff \| raw \| diff to current
2013-10-14	Eric Christopher	Revert part of a fix from 2010, changes since then:	blob \| commitdiff \| raw \| diff to current
2013-10-14	Eric Christopher	Remove some extraneous whitespace.	blob \| commitdiff \| raw \| diff to current
2013-07-21	Craig Topper	Mark that the _ftol2 function used by windows on x86...	blob \| commitdiff \| raw \| diff to current
2013-06-01	Tim Northover	X86: change MOV64ri64i32 into MOV32ri64	blob \| commitdiff \| raw \| diff to current
2013-05-31	Eric Christopher	Temporarily Revert "X86: change MOV64ri64i32 into MOV32...	blob \| commitdiff \| raw \| diff to current
2013-05-31	Tim Northover	X86: change MOV64ri64i32 into MOV32ri64	blob \| commitdiff \| raw \| diff to current
2013-05-30	Tim Northover	X86: use sub-register sequences for MOV*r0 operations	blob \| commitdiff \| raw \| diff to current
2013-05-30	Tim Northover	X86: change zext moves to use sub-register infrastructure.	blob \| commitdiff \| raw \| diff to current
2013-03-25	Jakob Stoklund Olesen	Annotate X86InstrCompiler.td with SchedRW lists.	blob \| commitdiff \| raw \| diff to current
2013-03-19	Jakob Stoklund Olesen	Annotate X86InstrCompiler.td with SchedRW lists.	blob \| commitdiff \| raw \| diff to current
2013-03-19	Ulrich Weigand	Remove an invalid and unnecessary Pat pattern from...	blob \| commitdiff \| raw \| diff to current
2013-02-23	Benjamin Kramer	X86: Disable cmov-memory patterns on subtargets without...	blob \| commitdiff \| raw \| diff to current
2013-01-22	Michael Liao	Fix an issue of pseudo atomic instruction DAG schedule	blob \| commitdiff \| raw \| diff to current
2013-01-07	Craig Topper	Remove # from the beginning and end of def names.	blob \| commitdiff \| raw \| diff to current
2012-12-26	Craig Topper	Add hasSideEffects=0 to some atomic instructions.	blob \| commitdiff \| raw \| diff to current
2012-10-15	Michael Liao	Add __builtin_setjmp/_longjmp supprt in X86 backend	blob \| commitdiff \| raw \| diff to current
2012-10-07	Benjamin Kramer	X86: fcmov doesn't handle all possible EFLAGS, fall...	blob \| commitdiff \| raw \| diff to current
2012-10-05	Craig Topper	Remove some encoding bits I forgot to remove from SETB_...	blob \| commitdiff \| raw \| diff to current
2012-10-05	Craig Topper	Move expansion of SETB_C(8/16/32/64)r from MCInstLower...	blob \| commitdiff \| raw \| diff to current
2012-09-26	Michael Liao	Add 'lock' prefix output support in assembly printer	blob \| commitdiff \| raw \| diff to current
2012-09-22	Michael Liao	Fix 16-bit atomic inst encoding and keep pseudo-inst...	blob \| commitdiff \| raw \| diff to current
2012-09-22	Michael Liao	Fix typo in r164357	blob \| commitdiff \| raw \| diff to current
2012-09-21	Michael Liao	Fix a typo in r164357	blob \| commitdiff \| raw \| diff to current
2012-09-21	Michael Liao	Revise td of X86 atomic instructions	blob \| commitdiff \| raw \| diff to current
2012-09-20	Michael Liao	Re-work X86 code generation of atomic ops with spin...	blob \| commitdiff \| raw \| diff to current
2012-09-13	Jakob Stoklund Olesen	Fix the TCRETURNmi64 bug differently.	blob \| commitdiff \| raw \| diff to current
2012-09-13	Jakob Stoklund Olesen	Revert r163761 "Don't fold indexed loads into TCRETURNm...	blob \| commitdiff \| raw \| diff to current
2012-09-13	Jakob Stoklund Olesen	Don't fold indexed loads into TCRETURNmi64.	blob \| commitdiff \| raw \| diff to current
2012-06-01	Hans Wennborg	Implement the local-dynamic TLS model for x86 (PR3985)	blob \| commitdiff \| raw \| diff to current
2012-05-09	Jakob Stoklund Olesen	Use ptr_rc_tailcall instead of GR32_TC.	blob \| commitdiff \| raw \| diff to current
2012-05-07	Manman Ren	X86: optimization for -(x != 0)	blob \| commitdiff \| raw \| diff to current
2012-04-04	Rafael Espindola	Always compute all the bits in ComputeMaskedBits.	blob \| commitdiff \| raw \| diff to current
2012-03-29	Lang Hames	Make x86 REP_MOV* and REP_STO instructions use the...	blob \| commitdiff \| raw \| diff to current
2012-03-19	Preston Gurd	This patch adds X86 instruction itineraries for non...	blob \| commitdiff \| raw \| diff to current
2012-02-24	Michael J. Spencer	Add WIN_FTOL_* psudo-instructions to model the unique...	blob \| commitdiff \| raw \| diff to current
2012-02-16	Jakob Stoklund Olesen	Use the same CALL instructions for Windows as for every...	blob \| commitdiff \| raw \| diff to current
2012-01-16	Eli Friedman	Make sure the non-SSE lowering for fences correctly...	blob \| commitdiff \| raw \| diff to current
2012-01-16	Eli Friedman	Get rid of unused codegen-only instruction.	blob \| commitdiff \| raw \| diff to current
2012-01-12	Benjamin Kramer	X86: Generalize the x << (y & const) optimization to...	blob \| commitdiff \| raw \| diff to current
2011-12-24	Chandler Carruth	Switch the lowering of CTLZ_ZERO_UNDEF from a .td patte...	blob \| commitdiff \| raw \| diff to current
2011-12-20	Chandler Carruth	Begin teaching the X86 target how to efficiently codege...	blob \| commitdiff \| raw \| diff to current
2011-10-26	Rafael Espindola	Fixes an issue reported by -verify-machineinstrs.	blob \| commitdiff \| raw \| diff to current
2011-10-26	Rafael Espindola	This commit introduces two fake instructions MORESTACK_...	blob \| commitdiff \| raw \| diff to current
2011-09-13	Eli Friedman	Fix the assembler strings for a couple of atomic instru...	blob \| commitdiff \| raw \| diff to current
2011-09-07	Eli Friedman	Fix atomic load and store on x86 to pass -verify-machin...	blob \| commitdiff \| raw \| diff to current
2011-09-02	Jakob Stoklund Olesen	Pseudo CMOV instructions don't clobber EFLAGS.	blob \| commitdiff \| raw \| diff to current
2011-08-30	Rafael Espindola	Adds a SelectionDAG node X86SegAlloca which will be...	blob \| commitdiff \| raw \| diff to current
2011-08-26	Eli Friedman	Add support for generating CMPXCHG16B on x86-64 for...	blob \| commitdiff \| raw \| diff to current
2011-08-24	Eli Friedman	Basic x86 code generation for atomic load and store...	blob \| commitdiff \| raw \| diff to current
2011-08-09	Bruno Cardoso Lopes	Add 256-bit support for v8i32, v4i64 and v4f64 ISD...	blob \| commitdiff \| raw \| diff to current
2011-08-09	Eli Friedman	Fix a couple ridiculous copy-paste errors. rdar:/...	blob \| commitdiff \| raw \| diff to current
2011-07-27	Eli Friedman	X86ISD::MEMBARRIER does not require SSE2; it doesn...	blob \| commitdiff \| raw \| diff to current
2011-06-16	Dan Gohman	Add a comment describing why transforming (shl x, 1...	blob \| commitdiff \| raw \| diff to current
2011-05-21	Benjamin Kramer	X86: smulo -> add is now done target-independently...	blob \| commitdiff \| raw \| diff to current
2011-05-20	Stuart Hastings	Re-commit 131641 with fixes; de-pseudoize MOVSX16rr8...	blob \| commitdiff \| raw \| diff to current
2011-05-19	Stuart Hastings	Reverting 131641 to investigate 'bot complaint.	blob \| commitdiff \| raw \| diff to current
2011-05-19	Stuart Hastings	Revise MOVSX16rr8/MOVZX16rr8 (and rm variants) to no...	blob \| commitdiff \| raw \| diff to current
2011-05-17	Eric Christopher	Support XOR and AND optimization with no return value.	blob \| commitdiff \| raw \| diff to current
2011-05-10	Eric Christopher	Optimize atomic lock or that doesn't use the result...	blob \| commitdiff \| raw \| diff to current
2011-05-10	Eric Christopher	Refactor lock versions of binary operators to be a...	blob \| commitdiff \| raw \| diff to current
2011-05-08	Benjamin Kramer	X86: Add a bunch of peeps for add and sub of SETB.	blob \| commitdiff \| raw \| diff to current
2011-02-17	Dan Gohman	The labyrinthine X86 backend no longer appears to require	blob \| commitdiff \| raw \| diff to current
2011-01-26	NAKAMURA Takumi	Target/X86: Tweak win64's tailcall.	blob \| commitdiff \| raw \| diff to current
2011-01-26	NAKAMURA Takumi	Fix whitespace.	blob \| commitdiff \| raw \| diff to current
2011-01-18	Eric Christopher	The stub routine that we're calling uses test and so...	blob \| commitdiff \| raw \| diff to current
2010-12-20	Chris Lattner	We lower setb to sbb with the hope that the and will...	blob \| commitdiff \| raw \| diff to current
2010-12-19	Chris Lattner	improve the setcc -> setcc_carry optimization to happen...	blob \| commitdiff \| raw \| diff to current
2010-12-15	Evan Cheng	Only rr forms of ADD*_DB are commutable.	blob \| commitdiff \| raw \| diff to current
2010-12-09	Eric Christopher	Add rsp to the uses for the same reason as 32-bit.	blob \| commitdiff \| raw \| diff to current
2010-11-28	Rafael Espindola	Move lowering of TLS_addr32 and TLS_addr64 to X86MCInst...	blob \| commitdiff \| raw \| diff to current
2010-11-27	Rafael Espindola	Lower TLS_addr32 and TLS_addr64.	blob \| commitdiff \| raw \| diff to current
2010-11-01	Chris Lattner	reject instructions that contain a \n in their asmstrin...	blob \| commitdiff \| raw \| diff to current
2010-10-31	Chris Lattner	two changes: make the asmmatcher generator ignore ARM...	blob \| commitdiff \| raw \| diff to current
next

C/C++ LLVM-based compilers that forbids OOTA behaviors

RSS Atom