aboutsummaryrefslogtreecommitdiff
AgeCommit message (Expand)Author
2006-04-06POR encoded as PAND, yikes.Evan Cheng
2006-04-05An entry about comi / ucomi intrinsics.Evan Cheng
2006-04-05Support for comi / ucomi intrinsics.Evan Cheng
2006-04-05Added comi and ucomi SSE intrinsics.Evan Cheng
2006-04-05Add all of the data stream intrinsics and instructions. wooChris Lattner
2006-04-05add altivec ds* intrinsicsChris Lattner
2006-04-05Fix a typoChris Lattner
2006-04-05Fix CodeGen/PowerPC/2006-04-05-splat-ish.llChris Lattner
2006-04-05new testcase that was miscompiled to vspltisbChris Lattner
2006-04-05Handle canonical form of e.g.Evan Cheng
2006-04-05Make a vector live across blocks have the correct Vec type. This fixesChris Lattner
2006-04-05new testcaseChris Lattner
2006-04-05Bogus assertEvan Cheng
2006-04-05Fallthrough to expand if a VECTOR_SHUFFLE cannot be custom lowered.Evan Cheng
2006-04-05Exapnd a VECTOR_SHUFFLE to a BUILD_VECTOR if target asks for it to be expandedEvan Cheng
2006-04-05Fix more tablegen depedency issues in Visual Studio.Jeff Cohen
2006-04-05Separate out to 2 test casesEvan Cheng
2006-04-05make this test less exactingAndrew Lenharth
2006-04-05revert this, this is safe, if conservative. leave a note to that effectAndrew Lenharth
2006-04-05Handle v8i16 shuffle that must be broken into a pair of pshufhw / pshuflw.Evan Cheng
2006-04-05Add a new shuffle test case that requires pshuflw / pshufhw pair.Evan Cheng
2006-04-05add vslChris Lattner
2006-04-05Get the types right, third time is the charm. Add vsl.Chris Lattner
2006-04-05add vmladduhmChris Lattner
2006-04-05correct the type of two intrinsics, add int_ppc_altivec_vmladduhmChris Lattner
2006-04-05Add m[tf]vscr instructions.Chris Lattner
2006-04-05Add m[tf]vscr intrinsics.Chris Lattner
2006-04-04add a noteChris Lattner
2006-04-04Add missing byte merges.Chris Lattner
2006-04-04Add FP -> Int ConversionsChris Lattner
2006-04-04add average intrinsicsChris Lattner
2006-04-04add average intrinsics.Chris Lattner
2006-04-04add a noteChris Lattner
2006-04-04Fix some broken logic that would cause us to codegen {2147483647,2147483647,2...Chris Lattner
2006-04-04Added pslldq and psrldq.Evan Cheng
2006-04-04Added intrinsics to match __builtin_ia32_pslldqi128 andEvan Cheng
2006-04-04Minor fixes + naming changes.Evan Cheng
2006-04-04How could this ever have worked?Chris Lattner
2006-04-04PSHUF* encoding bugs.Evan Cheng
2006-04-04Make sure to consider alignment of variable sized objects.Chris Lattner
2006-04-04Do not create ZEXTLOAD's unless we are before legalize or the operation isChris Lattner
2006-04-04New testcaseChris Lattner
2006-04-04Ask legalize to promote all vector shuffles to be v16i8 instead of having toChris Lattner
2006-04-04* Add supprot for SCALAR_TO_VECTOR operations where the input needs to beChris Lattner
2006-04-04Move isShuffleLegal from TLI to Legalize.Chris Lattner
2006-04-04new testcaseChris Lattner
2006-04-04Signed shr by a constant is not the same as sdiv by 2^kChris Lattner
2006-04-04cmpps / cmppd encoding bugEvan Cheng
2006-04-04Fix the types for these intrinsics.Chris Lattner
2006-04-04Constant fold bitconvert(undef)Chris Lattner