aboutsummaryrefslogtreecommitdiff
path: root/llvm/lib/Target/AMDGPU/SIInsertWaitcnts.cpp
AgeCommit message (Expand)AuthorFilesLines
5 days[AMDGPU] introduce S_WAITCNT_LDS_DIRECT in the memory legalizer (#150887)Sameer Sahasrabuddhe1-0/+20
7 days[AMDGPU][gfx1250] Use SCOPE_SE for stores that may hit scratch (#150586)Pierre van Houtryve1-28/+1
10 days[AMDGPU] gfx1250 vmem prefetch target intrinsics and builtins (#150466)Stanislav Mekhanoshin1-5/+3
14 days[AMDGPU] ISel & PEI for whole wave functions (#145858)Diana Picus1-0/+1
2025-07-17[AMDGPU] Move common fields out of WaitcntBrackets. NFC. (#148864)Jay Foad1-63/+59
2025-07-17[AMDGPU] Move class WaitcntBrackets after class SIInsertWaitcnts. NFC.Jay Foad1-234/+236
2025-07-02[AMDGPU] Remove HasSampler variable. NFC. (#146682)Jay Foad1-3/+8
2025-07-02[AMDGPU] Use isImage. NFC. (#146677)Jay Foad1-2/+1
2025-07-01[AMDGPU][SIInsertWaitCnts] skip meta instructions early (#145720)Sameer Sahasrabuddhe1-2/+5
2025-06-26[AMDGPU][NFC] rename some constants for readability (#145870)Sameer Sahasrabuddhe1-9/+9
2025-06-25[AMDGPU] Simplify S_WAIT_XCNT insertion. NFC. (#145682)Jay Foad1-16/+5
2025-06-25[AMDGPU][SIInsertWaitcnts] don't crash when printing messages at end of block...Sameer Sahasrabuddhe1-2/+14
2025-06-25[AMDGPU][GFX1250] Insert S_WAIT_XCNT for SMEM and VMEM load-stores (#145566)Christudasan Devadasan1-16/+121
2025-06-24[AMDGPU] Replace dynamic VGPR feature with attribute (#133444)Diana Picus1-4/+6
2025-06-20[AMDGPU] Omit image waits in function prologue on gfx1250 (#145097)Stanislav Mekhanoshin1-0/+4
2025-06-19[NFC][AMDGPU] print more info when debugging SIInsertWaitcnts pass (#144629)Sameer Sahasrabuddhe1-32/+80
2025-05-28[AMDGPU] Treat image_msaa_load as a sampler operation (#141726)Carl Ritson1-3/+3
2025-05-22[llvm] Use *Map::try_emplace (NFC) (#141190)Kazu Hirata1-1/+1
2025-05-07[AMDGPU] Classify FLAT instructions as VMEM (#137148)Robert Imschweiler1-7/+8
2025-04-22[AMDGPU][InsertWaitCnts] Track global_wb/inv/wbinv (#135340)Pierre van Houtryve1-3/+15
2025-04-10[AMDGPU] Add support for point sample accel out of order returns (#127991)David Stuttard1-1/+41
2025-04-08[AMDGPU] Fix excessive stack usage in SIInsertWaitcnts::run (#134835)Jay Foad1-4/+9
2025-04-08[AMDGPU] Make a few WaitcntBrackets methods const. NFC. (#134824)Jay Foad1-8/+8
2025-03-24[AMDGPU] Relax lds dma waitcnt with no aliasing pair (#131842)Austin Kerbow1-6/+3
2025-03-24[AMDGPU][NPM] Port SIInsertWaitcnts to NPM (#130061)Akshat Oke1-30/+63
2025-03-23[AMDGPU] Avoid repeated hash lookups (NFC) (#132657)Kazu Hirata1-4/+6
2025-03-21[AMDGPU] Insert before and after instructions that always use GDS (#131338)Stephen Thomas1-7/+84
2025-03-19[AMDGPU] Deallocate VGPRs before exiting in dynamic VGPR mode (#130037)Diana Picus1-23/+37
2025-03-11[AMDGPU][True16][CodeGen] update waitcnt for true16 (#128927)Brox Chen1-13/+18
2025-03-06[AMDGPU] Reduce size of SGPR arrays in SIInsertWaitcnts. NFC. (#130097)Jay Foad1-7/+6
2025-03-06[llvm] Fix an unused variable warningDmitri Gribenko1-2/+2
2025-03-06[AMDGPU] Remove RegisterEncoding from SIInsertWaitcnts. NFC. (#130056)Jay Foad1-30/+9
2025-03-04[AMDGPU] Remove unused s_barrier_{init,join,leave} instructions (#129548)Mariusz Sikora1-1/+0
2025-02-12[TableGen] Emit OpName as an enum class instead of a namespace (#125313)Rahul Joshi1-1/+1
2025-01-30[AMDGPU] Create new option for force flush load counter (#124974)Stanislav Mekhanoshin1-0/+8
2024-11-13[AMDGPU] Remove unused includes (NFC) (#116154)Kazu Hirata1-1/+0
2024-10-21[AMDGPU] Skip VGPR deallocation for waveslot limited kernels (#112765)Stanislav Mekhanoshin1-8/+17
2024-10-14[AMDGPU] Skip terminators when forcing emit zero flag (#112116)Shilei Tian1-1/+3
2024-10-14[AMDGPU] Use MachineInstr::mayLoadOrStore. NFC.Jay Foad1-1/+1
2024-10-14[NFC] clang-format llvm/lib/Target/AMDGPU/SIInsertWaitcnts.cppShilei Tian1-10/+9
2024-10-14[NFC][AMDGPU] Remove unnecessary member `ForceEmitZeroWaitcnts` (#112114)Shilei Tian1-5/+1
2024-10-03[AMDGPU] Qualify auto. NFC. (#110878)Jay Foad1-2/+2
2024-10-02[AMDGPU] Remember to clear a DenseMap between runs of SIInsertWaitcnts (#110650)Jay Foad1-0/+1
2024-10-01[AMDGPU] Refactor several functions for merging with downstream work. (#110562)Gang Chen1-111/+133
2024-09-25[AMDGPU] Do not count implicit VGPRs in SIInsertWaitcnts (#109049)Stanislav Mekhanoshin1-0/+8
2024-09-18[AMDGPU] Omit isReg() check for all_uses() in SIInsertWaitcnts. NFC. (#109041)Stanislav Mekhanoshin1-3/+3
2024-09-17[AMDGPU] Refine operand iterators in the SIInsertWaitcnts. NFCI. (#108884)Stanislav Mekhanoshin1-108/+79
2024-09-14[AMDGPU] Avoid unneeded waitcounts before spill stores (#108303)Stanislav Mekhanoshin1-1/+10
2024-08-27[AMDGPU] Use range-based for loops (NFC) (#106184)Kazu Hirata1-4/+2
2024-08-23[AMDGPU] Remove one case of vmcnt loop header flushing for GFX12 (#105550)Jay Foad1-1/+1
generated by cgit v1.1 at 2025-08-04 03:20:03 +0000