| Age | Commit message (Expand) | Author | Files | Lines |
|---|
| 2013-02-08 | sse.md (FMAMODEM): New mode iterator. | Uros Bizjak | 1 | -27/+31 |
| 2013-01-20 | Add AMD btver2 pipeline descriptions | Venkataramanan Kumar | 1 | -0/+44 |
| 2013-01-10 | Update copyright years in gcc/ | Richard Sandiford | 1 | -2/+1 |
| 2013-01-09 | sse.md (*vec_interleave_highv2df): Change mode attribute of movddup insn to DF. | Uros Bizjak | 1 | -3/+3 |
| 2012-11-25 | sse.md (<sse>_loadu<ssemodesuffix><avxsizesuffix>): Do not depend on TARGET_S... | Uros Bizjak | 1 | -12/+2 |
| 2012-11-20 | sse.md (*<ssse3_avx2>_pmulhrsw<mode>3): Merge *avx2_pmulhrswv16hi3 and... | Andrey Turetskiy | 1 | -35/+14 |
| 2012-11-16 | AMD bdver3 enablement | Ganesh Gopalasubramanian | 1 | -6/+6 |
| 2012-11-06 | * config/i386/sse.md | Uros Bizjak | 1 | -1/+1 |
| 2012-11-06 | i386-protos.h (emit_i387_cw_initialization): Deleted. | Vladimir Yakovlev | 1 | -2/+1 |
| 2012-11-06 | i386.c (bdesc_args): Rename CODE_FOR_avx2_umulhrswv16hi3 to CODE_FOR_avx2_pmu... | Andrey Turetskiy | 1 | -269/+56 |
| 2012-10-15 | sse.md (UNSPEC_MOVU): Remove. | Uros Bizjak | 1 | -17/+86 |
| 2012-10-08 | re PR middle-end/54400 (recognize vector reductions) | Marc Glisse | 1 | -5/+85 |
| 2012-09-28 | re PR target/54716 (Select best typed instruction for bitwise operations) | Jakub Jelinek | 1 | -3/+6 |
| 2012-09-14 | re PR target/54564 (Broken __builtin_ia32_vfmadds[sd]3) | Jakub Jelinek | 1 | -29/+25 |
| 2012-09-05 | sse.md (<sse4_1>_blendv<ssemodesuffix><avxsizesuffix>): Use register_operand ... | Uros Bizjak | 1 | -12/+12 |
| 2012-08-11 | sse.md (xop integer multiply/add insns): Use register_operand for operand 3 p... | Uros Bizjak | 1 | -16/+13 |
| 2012-08-11 | i386.md (isa): Add fma and fma4. | Uros Bizjak | 1 | -138/+60 |
| 2012-08-10 | * config/i386/sse.md (*fma_fmadd_<mode>, *fma_fmsub_<mode>, | Uros Bizjak | 1 | -172/+168 |
| 2012-07-10 | i386: Rename patterns for vec_widen_<s>mult_even/odd_<mode> | Richard Henderson | 1 | -12/+6 |
| 2012-06-26 | i386: Implement widen_smul_*_v4si for plain sse2 | Richard Henderson | 1 | -11/+18 |
| 2012-06-26 | i386: Expand mul<VI8_AVX2> earlier | Richard Henderson | 1 | -81/+3 |
| 2012-06-25 | i386: Delete sse sdot_prod; add sdot_prodv4si for xop | Richard Henderson | 1 | -54/+8 |
| 2012-06-25 | i386: Add widen_multiply_even/odd builtins and hooks | Richard Henderson | 1 | -0/+14 |
| 2012-06-25 | i386: Use vpmacsdd for mulv4si3 | Richard Henderson | 1 | -0/+6 |
| 2012-06-25 | i386: Cleanup and unify widening multiply patterns | Richard Henderson | 1 | -188/+18 |
| 2012-06-25 | re PR target/53759 (gcc -mavx emits vshufps for __builtin_ia32_loadlps) | Jakub Jelinek | 1 | -1/+1 |
| 2012-06-23 | re PR target/53749 (ice in expand_shift_1) | Richard Henderson | 1 | -47/+29 |
| 2012-06-23 | i386: Pass ix86_expand_sse_unpack operands by value | Richard Henderson | 1 | -4/+4 |
| 2012-06-23 | i386: Extract the guts of mulv16qi3 to ix86_expand_vecop_qihi | Richard Henderson | 1 | -60/+3 |
| 2012-06-19 | Handle const_vector in mulv4si3 for pre-sse4.1. | Richard Henderson | 1 | -59/+13 |
| 2012-06-18 | re PR target/53712 (Does not combine unaligned load with _mm_cmpistri, redun... | Uros Bizjak | 1 | -0/+115 |
| 2012-06-17 | sse.md (vcvtph2ps): Fix vec_select selector. | Uros Bizjak | 1 | -1/+1 |
| 2012-05-14 | re PR target/46098 (ICE: in extract_insn, at recog.c:2100 with -msse3 -ffloat... | Uros Bizjak | 1 | -23/+2 |
| 2012-05-13 | * config/i386/sse.md: The other part of my previous commit. | Uros Bizjak | 1 | -14/+28 |
| 2012-05-13 | sse.md (<sse>_andnot<mode>3): Handle TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL. | Uros Bizjak | 1 | -65/+179 |
| 2012-05-11 | i386.md (*movti_internal_rex64): Avoid MOVAPS size optimization for TARGET_AVX. | Uros Bizjak | 1 | -0/+3 |
| 2012-05-10 | i386.md (*movoi_internal_avx): Handle TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL a... | Uros Bizjak | 1 | -29/+32 |
| 2012-05-09 | re PR target/52908 (xop-mul-1:f9 miscompiled on bulldozer (-mxop)) | Uros Bizjak | 1 | -82/+14 |
| 2012-05-09 | re PR target/44141 (Redundant loads and stores generated for AMD bdver1 target) | Uros Bizjak | 1 | -16/+58 |
| 2012-04-14 | * config/i386/sse.md: Fix whitespace and formatting. | Uros Bizjak | 1 | -416/+224 |
| 2012-04-14 | sse.md (ssse3_plusminus): New code iterator. | Uros Bizjak | 1 | -1011/+139 |
| 2012-04-12 | re PR target/52932 (AVX2 intrinsic _mm256_permutevar8x32_ps has wrong paramet... | Uros Bizjak | 1 | -21/+8 |
| 2012-03-29 | sse.md (avx_h<plusminus_insn>v4df3): Fix results crossing 128bit lane boundary. | Uros Bizjak | 1 | -5/+5 |
| 2012-03-28 | re PR target/52736 (miscompilation: store to aliased __m128d is 8 Bytes off) | Jakub Jelinek | 1 | -1/+1 |
| 2012-03-20 | re PR target/52607 (v4df __builtin_shuffle with {0,2,1,3} or {1,3,0,2}) | Jakub Jelinek | 1 | -4/+38 |
| 2012-03-18 | i386.md: Remove empty predicates and/or constraints. | Uros Bizjak | 1 | -616/+616 |
| 2012-03-16 | re PR target/52568 (suboptimal __builtin_shuffle on cycles with AVX) | Jakub Jelinek | 1 | -28/+14 |
| 2012-03-12 | Properly set ix86_gen_xxx function pointers | H.J. Lu | 1 | -2/+2 |
| 2012-03-01 | re PR target/52437 (internal compiler error: in spill_failure, at reload1.c:2... | Jakub Jelinek | 1 | -5/+5 |
| 2012-01-02 | sse.md (*avx2_gatherdi<mode>_3, [...]): New patterns. | Jakub Jelinek | 1 | -1/+47 |
