Stackmap shadows should consider call returns a branch target.
[oota-llvm.git] / test / CodeGen / X86 / avx-splat.ll
2014-10-04 Chandler Carruth[x86] Enable the new vector shuffle lowering by default.
2014-09-15 Chandler Carruth[x86] Undo a flawed transform I added to form UNPCK...
2014-09-14 Chandler Carruth[x86] Teach the vector combiner that picks a canonical...
2014-08-02 Chandler Carruth[x86] Teach the target shuffle mask extraction to recog...
2014-07-27 Chandler Carruth[x86] Add a much more powerful framework for combining...
2014-07-09 Chandler Carruth[x86] Re-apply a variant of the x86 side of r212324...
2013-10-08 Craig TopperRemove some instructions that existed to provide aliase...
2013-01-26 Benjamin KramerX86: Do splat promotion later, so the optimizer can...
2012-12-07 Nadav RotemX86: Prefer using VPSHUFD over VPERMIL because it has...
2012-11-26 Jakub StaszakNormalize splat 256bit vectors with 8 elements.
2012-02-07 Craig TopperAdd instruction selection for 256-bit VPSHUFD and 128...
2011-11-29 Craig TopperFix VINSERTF128/VEXTRACTF128 to be marked as FP instruc...
2011-09-27 Jakob Stoklund OlesenRemove X86-dependent stuff from SSEDomainFix.
2011-08-23 Bruno Cardoso LopesFix a nasty bug where a v4i64 was being wrong emitted...
2011-08-23 Craig TopperAdd support for breaking 256-bit v16i16 and v32i8 VSETC...
2011-08-17 Bruno Cardoso LopesUpdate test to not use the scalar type to splat from...
2011-08-17 Bruno Cardoso LopesNow that we have a canonical way to handle 256-bit...
2011-08-11 Bruno Cardoso LopesSplats for v8i32/v8f32 can be handled by VPERMILPSY...
2011-08-11 Bruno Cardoso LopesUse the splat index to generate the desired shuffle...
2011-08-09 Bruno Cardoso LopesRename and tidy up tests