| Age | Commit message (Expand) | Author | Files | Lines |
|---|
| 2019-01-07 | Investigating PR target/86891 revealed a number of issues with the way the... | Richard Earnshaw | 3 | -64/+246 |
| 2019-01-07 | [nvptx] Force vl32 if calling vector-partitionable routines | Tom de Vries | 1 | -0/+45 |
| 2019-01-07 | sse.md (vec_extract<mode><ssehalfvecmodelower>): Use V_256_512 iterator inste... | Jakub Jelinek | 1 | -2/+2 |
| 2019-01-07 | [nvptx] Handle large vector reductions | Tom de Vries | 4 | -29/+140 |
| 2019-01-07 | [nvptx] Don't emit barriers for empty loops -- fix | Tom de Vries | 1 | -4/+11 |
| 2019-01-07 | [nvptx] Add support for a per-worker broadcast buffer and barrier | Tom de Vries | 2 | -19/+138 |
| 2019-01-07 | [nvptx] Allow larger PTX_MAX_VECTOR_LENGTH in nvptx_goacc_validate_dims_1 | Tom de Vries | 1 | -4/+27 |
| 2019-01-07 | [nvptx] Postpone warnings in nvptx_goacc_validate_dims_1 | Tom de Vries | 1 | -13/+25 |
| 2019-01-07 | [nvptx] Eliminate PTX_VECTOR_LENGTH | Tom de Vries | 1 | -7/+8 |
| 2019-01-07 | [nvptx] Add asserts in nvptx_goacc_validate_dims | Tom de Vries | 1 | -0/+4 |
| 2019-01-07 | re PR target/88521 (GCC from r266355 miscompiles x265 for mingw-w64 target) | Mateusz B | 1 | -0/+7 |
| 2019-01-05 | re PR debug/88635 (Assembler error when building with "-g -O2 -m32") | Jakub Jelinek | 1 | -12/+0 |
| 2019-01-04 | re PR target/88594 (ICE in int_mode_for_mode, at stor-layout.c:403) | Jakub Jelinek | 1 | -3/+1 |
| 2019-01-04 | x86: relax mask register constraints | Jan Beulich | 1 | -26/+26 |
| 2019-01-03 | rs6000-string.c (expand_block_move): Don't use unaligned vsx and avoid lxvd2x... | Aaron Sawdey | 1 | -2/+21 |
| 2019-01-03 | [nvptx] Add nvptx_mach_vector_length, nvptx_mach_max_workers | Tom de Vries | 2 | -0/+44 |
| 2019-01-03 | [nvptx] Factor out populate_offload_attrs | Tom de Vries | 1 | -13/+50 |
| 2019-01-03 | [nvptx] Add early-out cases in nvptx_goacc_validate_dims | Tom de Vries | 1 | -12/+50 |
| 2019-01-03 | [nvptx] Eliminate changed local var in nvptx_goacc_validate_dims | Tom de Vries | 1 | -12/+26 |
| 2019-01-02 | re PR target/87545 (-mtune=intel regressions) | Richard Biener | 1 | -1/+1 |
| 2019-01-01 | Update copyright years. | Jakub Jelinek | 1476 | -1491/+1491 |
| 2018-12-31 | re PR target/86814 (xtensa port needs updating for CVE-2017-5753) | Max Filippov | 1 | -0/+3 |
| 2018-12-31 | * x86-tune.def: Enable inter_unit_moves_to_vec for generic. | Jan Hubicka | 1 | -2/+6 |
| 2018-12-29 | msp430.c (msp430_attr): Warn when the critical and interrupt function attribu... | Jozef Lawrynowicz | 1 | -0/+9 |
| 2018-12-28 | i386.h (ADDITIONAL_REGISTER_NAMES): Add sil, dil, bpl and spl aliases. | Uros Bizjak | 1 | -14/+7 |
| 2018-12-26 | aarch64-cores.def (octeontx): New. | Andrew Pinski | 2 | -1/+7 |
| 2018-12-23 | [RS6000] PR88346, Inconsistent list of CPUs supported after r266502 | Alan Modra | 4 | -10/+12 |
| 2018-12-23 | Fix ident tests by Enabling ident output for X86 Darwin. | Iain Sandoe | 1 | -0/+5 |
| 2018-12-21 | S/390: Add support for double<->long vector converts | Andreas Krebbel | 1 | -0/+52 |
| 2018-12-21 | re PR target/88522 (Error: operand size mismatch for `vpgatherqq') | Jakub Jelinek | 1 | -28/+36 |
| 2018-12-21 | re PR target/88556 (Inline built-in sinh, cosh, tanh for -ffast-math) | Uros Bizjak | 3 | -1/+203 |
| 2018-12-21 | re PR target/88547 (missed optimization for vector comparisons) | Jakub Jelinek | 1 | -0/+98 |
| 2018-12-21 | x86: VAESDEC{,LAST} allow memory inputs | Jan Beulich | 1 | -2/+2 |
| 2018-12-20 | [AArch64][SVE] Add ABS support | Richard Sandiford | 1 | -2/+3 |
| 2018-12-20 | [AArch64][SVE] Fix IFN_COND_FMLA movprfx alternative | Richard Sandiford | 1 | -1/+1 |
| 2018-12-20 | re PR target/88547 (missed optimization for vector comparisons) | Jakub Jelinek | 1 | -116/+199 |
| 2018-12-20 | [RS6000] Fix non-linux powerpc breakage | Alan Modra | 1 | -0/+4 |
| 2018-12-19 | i386.md (SWI1248_AVX512BWDQ_64): Rename from SWI1248_AVX512BWDQ2_64. | Uros Bizjak | 1 | -6/+6 |
| 2018-12-19 | [ARM] Optimize executable size when using softfloat fmul/dmul | Thomas Preud'homme | 1 | -1/+1 |
| 2018-12-19 | [nvptx] Add PTX_CTA_SIZE | Tom de Vries | 1 | -0/+5 |
| 2018-12-19 | [rs6000] Fix x86 SSSE3 compatibility implementations and testcases | Paul A. Clarke | 1 | -11/+11 |
| 2018-12-19 | [ARM] Do softfloat when -mfpu set, -mfloat-abi=softfp | Thomas Preud'homme | 2 | -2/+3 |
| 2018-12-19 | Restrict a VSX extract to TARGET_POWERPC64 (PR88213) | Segher Boessenkool | 1 | -1/+1 |
| 2018-12-19 | [nvptx] Use MAX, MIN, ROUND_UP macros | Tom de Vries | 1 | -18/+10 |
| 2018-12-19 | [nvptx] Make nvptx state propagation function names more generic | Tom de Vries | 1 | -42/+54 |
| 2018-12-19 | [nvptx] Rename worker_bcast variables to oacc_bcast | Tom de Vries | 1 | -29/+30 |
| 2018-12-19 | [nvptx] Generalize bar.sync instruction | Tom de Vries | 2 | -3/+9 |
| 2018-12-19 | [nvptx] Only use one logical barrier resource | Tom de Vries | 1 | -2/+2 |
| 2018-12-19 | [nvptx] Use TARGET_SET_CURRENT_FUNCTION | Tom de Vries | 1 | -0/+14 |
| 2018-12-19 | [aarch64] Correct architecture for tsv110. | Shaokun Zhang | 1 | -3/+3 |
