aboutsummaryrefslogtreecommitdiff
path: root/lib/Target/X86
AgeCommit message (Expand)Author
2009-03-19Remove strange extra semicolons.Nick Lewycky
2009-03-18Disable the "call to immediate" optimization on x86-64. It isChris Lattner
2009-03-17Recognize bswapl as bswap too.Dan Gohman
2009-03-17Recognize "bswapq" as an alternate spelling for the bswap instruction.Dan Gohman
2009-03-14Use %rip-relative addressing on x86-64 whenever practical, asDan Gohman
2009-03-14Don't forego folding of loads into 64-bit adds when the otherDan Gohman
2009-03-13Improve FastISel's handling of truncates to i1, and implementDan Gohman
2009-03-13Fix FastISel's assumption that i1 values are always zero-extendedDan Gohman
2009-03-13add 8 and 16 bit TLS moves.Rafael Espindola
2009-03-13Improve sext and zext of TLS variables.Rafael Espindola
2009-03-13generalize this code so that fast isel handles integer truncates to i1, whichChris Lattner
2009-03-13These instructions have special lowering that may lower them to SSEBill Wendling
2009-03-13Fix some significant problems with constant pools that resulted in unnecessar...Evan Cheng
2009-03-13generalize the previous code to use the full generality of LEAChris Lattner
2009-03-13optimize the case of cond ? 42 : 41 and friends. This compiles theChris Lattner
2009-03-13Enhance address-mode folding of ISD::ADD to handle cases where theDan Gohman
2009-03-12Re-apply 66024 with fixes: 1. Fixed indirect call to immediate address assemb...Evan Cheng
2009-03-12Move 3 "(add (select cc, 0, c), x) -> (select cc, x, (add, x, c))"Chris Lattner
2009-03-12improve comment.Chris Lattner
2009-03-12On x86, if the only use of a i64 load is a i64 store, generate a pair of doub...Evan Cheng
2009-03-11Revert r66024. The JIT encoding for CALLpcrel32 is wrong -- see PR3773, and theDan Gohman
2009-03-11optimize i8 and i16 tls values.Rafael Espindola
2009-03-11Add a -no-implicit-float flag. This acts like -soft-float, but may generateBill Wendling
2009-03-11It makes no sense to have a ODR version of commonDuncan Sands
2009-03-11For yonah, fix a vector shuffle case for v16i8 where we didn't properly clear...Mon P Wang
2009-03-11Fixed a v8i16 shuffle case that should generate a pshufb instead of a pshuflw...Mon P Wang
2009-03-11formatting change, reduce indentation. No functionality change.Chris Lattner
2009-03-10Add more information to the EFLAGS note.Dan Gohman
2009-03-09Add a note about EFLAGS optimization.Dan Gohman
2009-03-08do not export all the X86FastISel symbols, ever.Chris Lattner
2009-03-08add a note.Chris Lattner
2009-03-08add a note.Chris Lattner
2009-03-07Introduce new linkage types linkonce_odr, weak_odr, common_odrDuncan Sands
2009-03-07Arithmetic instructions don't set EFLAGS bits OF and CF bitsDan Gohman
2009-03-05Don't use plain INC32 and DEC32 on x86-64; it needsDan Gohman
2009-03-05When creating X86ISD::INC and X86ISD::DEC nodes, only add one operand.Dan Gohman
2009-03-05Fix the "test" optimization to recognize "dec" as an add ofDan Gohman
2009-03-04Re-apply 66008, now that the unfoldMemoryOperand bug is fixed.Dan Gohman
2009-03-04Correct this comment.Dan Gohman
2009-03-04When using MachineInstr operand indices on SDNodes, the numberDan Gohman
2009-03-04Fix PR3666: isel calls to constant addresses.Evan Cheng
2009-03-04Revert r66004 for now; it's causing a variety of test failures.Dan Gohman
2009-03-04Teach the x86 backend to eliminate "test" instructions by using the EFLAGSDan Gohman
2009-03-04Fix PR3701. 1. X86 target renamed eflags register to flags. This matches what...Evan Cheng
2009-03-03Add '(implicit EFLAGS)' for AND, OR, XOR, NEG, INC, and DECDan Gohman
2009-03-03Fix a bunch of Doxygen syntax issues. Escape special characters,Dan Gohman
2009-02-28Added another darwin subtargetMon P Wang
2009-02-27Refactor TLS code and add some tests. The tests and expected results are:Rafael Espindola
2009-02-26ADDS{D|S}rr_Int and MULS{D|S}rr_Int are not commutable. The users of these in...Evan Cheng
2009-02-25Revert BuildVectorSDNode related patches: 65426, 65427, and 65296.Evan Cheng