| Age | Commit message (Expand) | Author | Files | Lines |
|---|
| 11 days | x86: Implement Fast-Math Float Truncation to BF16 via PSRLD Instruction | Levy Hsu | 1 | -7/+9 |
| 13 days | ssa-math-opts, i386: Improve spaceship expansion [PR116896] | Jakub Jelinek | 1 | -6/+70 |
| 2024-09-19 | i386: Add ssemov2, sseicvt2 for some load instructions that use memory on ope... | Hu, Lin1 | 1 | -4/+7 |
| 2024-09-12 | i386: Use offsetable address constraint for double-word memory operands, part 2 | Uros Bizjak | 1 | -1/+1 |
| 2024-09-09 | i386: Use offsetable address constraint for double-word memory operands | Uros Bizjak | 1 | -3/+3 |
| 2024-09-02 | i386: Optimize generate insn for AVX10.2 compare | Hu, Lin1 | 1 | -2/+29 |
| 2024-08-26 | AVX10.2: Support compare instructions | Zhang, Jun | 1 | -0/+1 |
| 2024-08-26 | AVX10.2: Support vector copy instructions | Zhang, Jun | 1 | -1/+2 |
| 2024-08-15 | i386: Improve split of *extendv2di2_highpart_stv_noavx512vl. | Roger Sayle | 1 | -2/+30 |
| 2024-08-15 | Movement between GENERAL_REGS and SSE_REGS for TImode doesn't need secondary ... | liuhongt | 1 | -0/+19 |
| 2024-08-14 | i386: Optimization for APX NDD is always zero-uppered for shift | Lingling Kong | 1 | -0/+66 |
| 2024-08-14 | i386: Optimization for APX NDD is always zero-uppered for logic | Lingling Kong | 1 | -0/+94 |
| 2024-08-14 | i386: Optimization for APX NDD is always zero-uppered for sub/adc/sbb | Lingling Kong | 1 | -8/+236 |
| 2024-08-14 | i386: Optimization for APX NDD is always zero-uppered for ADD | Lingling Kong | 1 | -0/+80 |
| 2024-08-12 | PR target/116275: Handle STV of *extenddi2_doubleword_highpart on i386. | Roger Sayle | 1 | -0/+18 |
| 2024-08-02 | i386: Fix memory constraint for APX NF | Lingling Kong | 1 | -2/+3 |
| 2024-08-01 | Fix mismatch between constraint and predicate for ashl<mode>3_doubleword. | liuhongt | 1 | -1/+1 |
| 2024-08-01 | i386: Remove ndd support for *add<mode>_4 [PR113744] | Lingling Kong | 1 | -25/+15 |
| 2024-07-26 | i386: Use BLKmode for {ld,st}tilecfg | Haochen Jiang | 1 | -7/+5 |
| 2024-07-23 | i386: Change prefetchi output template | Haochen Jiang | 1 | -1/+1 |
| 2024-07-15 | [APX NF] Add a pass to convert legacy insn to NF insns | Hongyu Wang | 1 | -5/+62 |
| 2024-07-10 | i386: Swap compare operands in ustrunc patterns | Uros Bizjak | 1 | -3/+3 |
| 2024-07-09 | i386: Implement .SAT_TRUNC for unsigned integers | Uros Bizjak | 1 | -2/+110 |
| 2024-07-08 | i386: Promote {QI,HI}mode x86_mov<mode>cc_0_m1_neg to SImode | Uros Bizjak | 1 | -6/+19 |
| 2024-07-04 | i386: Add additional variant of bswaphisi2_lowpart peephole2. | Roger Sayle | 1 | -0/+24 |
| 2024-07-02 | i386: Support APX NF and NDD for imul/mul | Lingling Kong | 1 | -45/+53 |
| 2024-07-01 | i386: Additional peephole2 to use lea in round-up integer division. | Roger Sayle | 1 | -0/+15 |
| 2024-07-01 | Extend lshifrtsi3_1_zext to ?k alternative. | liuhongt | 1 | -6/+13 |
| 2024-06-28 | i386: Handle sign_extend like zero_extend in *concatditi3_[346] | Roger Sayle | 1 | -3/+3 |
| 2024-06-13 | [APX CCMP] Use ctestcc when comparing to const 0 | Hongyu Wang | 1 | -4/+7 |
| 2024-06-13 | [APX ZU] Support APX zero-upper | Lingling Kong | 1 | -2/+23 |
| 2024-06-11 | i386: Use CMOV in .SAT_{ADD|SUB} expansion for TARGET_CMOV [PR112600] | Uros Bizjak | 1 | -9/+53 |
| 2024-06-09 | i386: Implement .SAT_SUB for unsigned scalar integers [PR112600] | Uros Bizjak | 1 | -1/+30 |
| 2024-06-08 | i386: Implement .SAT_ADD for unsigned scalar integers [PR112600] | Uros Bizjak | 1 | -2/+22 |
| 2024-06-06 | [APX CCMP] Support APX CCMP | Hongyu Wang | 1 | -2/+33 |
| 2024-06-03 | i386: Force operand 1 of bswapsi2 to a register for !TARGET_BSWAP [PR115321] | Uros Bizjak | 1 | -10/+11 |
| 2024-06-03 | [APX NF] Support APX NF for lzcnt/tzcnt/popcnt | Lingling Kong | 1 | -11/+113 |
| 2024-06-03 | [APX NF] Support APX NF for mul/div | Lingling Kong | 1 | -17/+30 |
| 2024-06-03 | [APX NF] Support APX NF for shld/shrd | Lingling Kong | 1 | -81/+308 |
| 2024-06-03 | [APX NF] Support APX NF for rotate insns | Lingling Kong | 1 | -21/+38 |
| 2024-06-03 | [APX NF] Support APX NF for right shift insns | Lingling Kong | 1 | -36/+46 |
| 2024-06-03 | [APX NF] Support APX NF for left shift insns | Lingling Kong | 1 | -26/+70 |
| 2024-06-03 | [APX NF] Support APX NF for {sub/and/or/xor/neg} | Lingling Kong | 1 | -82/+91 |
| 2024-06-03 | [APX NF] Support APX NF add | Lingling Kong | 1 | -44/+93 |
| 2024-05-31 | i386: Rewrite bswaphi2 handling [PR115102] | Uros Bizjak | 1 | -27/+50 |
| 2024-05-29 | Align tight&hot loop without considering max skipping bytes. | liuhongt | 1 | -4/+6 |
| 2024-05-20 | i386: Remove Xeon Phi ISA support | Haochen Jiang | 1 | -16/+3 |
| 2024-04-28 | Adjust alternative *k to ?k for avx512 mask in zero_extend patterns | liuhongt | 1 | -8/+8 |
| 2024-04-23 | i386: Avoid =&r,r,r andn double-word alternative for ia32 [PR114810] | Jakub Jelinek | 1 | -4/+5 |
| 2024-04-15 | x86: Allow TImode offsettable memory only with 8-bit constant | H.J. Lu | 1 | -17/+19 |
