| Age | Commit message (Expand) | Author | Files | Lines |
|---|
| 2024-05-29 | libgomp: Enable USM for AMD APUs and MI200 devices | Tobias Burnus | 1 | -1/+1 |
| 2024-04-26 | amdgcn: Add gfx90c target | Frederik Harwath | 7 | -2/+26 |
| 2024-04-08 | GCN: '--param=gcn-preferred-vectorization-factor=[default,32,64]' | Thomas Schwinge | 2 | -1/+29 |
| 2024-04-03 | GCN: Fix --with-arch= handling in mkoffload [PR111966] | Tobias Burnus | 1 | -18/+72 |
| 2024-03-25 | amdgcn: Add gfx1036 target | Richard Biener | 7 | -5/+27 |
| 2024-03-22 | amdgcn: Adjust GFX10/GFX11 cache coherency | Andrew Stubbs | 1 | -31/+55 |
| 2024-03-22 | amdgcn: Prefer V32 on RDNA devices | Andrew Stubbs | 1 | -0/+26 |
| 2024-03-22 | amdgcn: Add gfx1103 target | Andrew Stubbs | 7 | -7/+29 |
| 2024-03-21 | amdgcn: Comment correction | Andrew Stubbs | 1 | -2/+2 |
| 2024-03-21 | amdgcn: Ensure gfx11 is running in cumode | Andrew Stubbs | 1 | -0/+1 |
| 2024-03-21 | amdgcn: Clean up device memory in gcn-run | Andrew Stubbs | 1 | -1/+7 |
| 2024-03-14 | gcn: Fix a comment typo | Jakub Jelinek | 1 | -1/+1 |
| 2024-02-19 | GCN: Conditionalize 'define_expand "reduc_<fexpander>_scal_<mode>"' on '!TARG... | Thomas Schwinge | 2 | -1/+5 |
| 2024-02-19 | GCN: Restore lost '__gfx90a__' target CPU definition | Thomas Schwinge | 1 | -0/+6 |
| 2024-02-15 | amdgcn: Disallow unsupported permute on RDNA devices | Andrew Stubbs | 2 | -8/+14 |
| 2024-02-01 | GCN: Don't hard-code number of SGPR/VGPR/AVGPR registers | Thomas Schwinge | 2 | -6/+15 |
| 2024-02-01 | GCN: Remove 'FIRST_{SGPR,VGPR,AVGPR}_REG', 'LAST_{SGPR,VGPR,AVGPR}_REG' from ... | Thomas Schwinge | 1 | -8/+2 |
| 2024-02-01 | GCN: Remove 'SGPR_OR_VGPR_REGNO_P' definition | Thomas Schwinge | 1 | -1/+0 |
| 2024-02-01 | GCN, RDNA 3: Adjust 'sync_compare_and_swap<mode>_lds_insn' | Thomas Schwinge | 1 | -1/+6 |
| 2024-01-29 | gcn/gcn-valu.md: Disable fold_left_plus for TARGET_RDNA2_PLUS [PR113615] | Tobias Burnus | 1 | -1/+2 |
| 2024-01-29 | gcn/mkoffload.cc: Fix SRAM_ECC and XNACK handling [PR111966] | Tobias Burnus | 1 | -18/+32 |
| 2024-01-26 | amdgcn: additional gfx1030/gfx1100 support | Andrew Stubbs | 4 | -16/+62 |
| 2024-01-26 | gcn/gcn-hsa.h: Always pass --amdhsa-code-object-version= in ASM_SPEC | Tobias Burnus | 1 | -1/+16 |
| 2024-01-25 | gcn: Add missing space to ASM_SPEC in gcn-hsa.h | Tobias Burnus | 1 | -1/+1 |
| 2024-01-23 | gcn: Fix a warning | Jakub Jelinek | 1 | -1/+2 |
| 2024-01-22 | [gcn] mkoffload: Fix linking with "-g"; fix file deletion; improve diagnostic... | Tobias Burnus | 1 | -16/+16 |
| 2024-01-08 | amdgcn: Add gfx1100 to new XNACK defaults in mkoffload | Tobias Burnus | 1 | -0/+1 |
| 2024-01-08 | GCN: Add pre-initial support for gfx1100 | Tobias Burnus | 9 | -36/+65 |
| 2024-01-08 | amdgcn: Match new XNACK defaults in mkoffload | Andrew Stubbs | 1 | -2/+21 |
| 2024-01-08 | amdgcn: Don't double-count AVGPRs | Andrew Stubbs | 1 | -29/+0 |
| 2024-01-05 | Implement ASM_DECLARE_FUNCTION_NAME using ASM_OUTPUT_FUNCTION_LABEL | Ilya Leoshkevich | 1 | -3/+2 |
| 2024-01-04 | Add generated .opt.urls files | David Malcolm | 1 | -0/+23 |
| 2024-01-03 | Update copyright years. | Jakub Jelinek | 17 | -17/+17 |
| 2023-12-20 | No libstdc++ for GCN | Thomas Schwinge | 1 | -0/+4 |
| 2023-12-13 | amdgcn: Work around XNACK register allocation problem | Andrew Stubbs | 4 | -4/+33 |
| 2023-12-13 | amdgcn: Support XNACK mode | Andrew Stubbs | 5 | -136/+179 |
| 2023-12-06 | amdgcn, libgomp: low-latency allocator | Andrew Stubbs | 2 | -1/+17 |
| 2023-12-02 | Allow target attributes in non-gnu namespaces | Richard Sandiford | 1 | -5/+3 |
| 2023-11-27 | GCN: Remove 'last_arg' spec function | Thomas Schwinge | 3 | -43/+1 |
| 2023-11-27 | GCN: Tag '-march=[...]', '-mtune=[...]' as 'Negative' of themselves [PR112669] | Thomas Schwinge | 1 | -2/+2 |
| 2023-11-27 | amdgcn: Disallow TImode vector permute | Andrew Stubbs | 1 | -1/+3 |
| 2023-11-22 | amdgcn: Fix vector TImode reload loop | Andrew Stubbs | 1 | -4/+4 |
| 2023-11-15 | amdgcn: Add Accelerator VGPR registers | Andrew Stubbs | 7 | -191/+533 |
| 2023-11-15 | amdgcn: simplify secondary reload patterns | Andrew Stubbs | 2 | -90/+4 |
| 2023-11-10 | amdgcn: Fix v_add constraints (pr112308) | Andrew Stubbs | 1 | -50/+68 |
| 2023-11-10 | amdgcn: Fix vector min/max ICE | Andrew Stubbs | 1 | -1/+9 |
| 2023-11-07 | openmp: Add support for the 'indirect' clause in C/C++ | Kwok Cheung Yeung | 1 | -2/+27 |
| 2023-10-27 | amdgcn: Fix bug in gfx1030 support patch | Andrew Stubbs | 1 | -4/+2 |
| 2023-10-27 | amdgcn: silence warnings | Andrew Stubbs | 2 | -3/+5 |
| 2023-10-20 | amdgcn: add -march=gfx1030 EXPERIMENTAL | Andrew Stubbs | 9 | -87/+183 |
