aboutsummaryrefslogtreecommitdiff
path: root/backend/lib
Commit message (Collapse)AuthorAgeFilesLines
* [GlobalISel] Propagate no-fp-except flagsRichard-Qin-X2026-08-162-0/+52
|
* [RISCV][GlobalISel] Select mixed-width copy-signRichard-Qin-X2026-08-162-4/+44
|
* [GlobalISel] Delete dead instructions before selectionRichard-Qin-X2026-08-161-11/+1
|
* [RISCV][GlobalISel] Select word division pseudosRichard-Qin-X2026-08-161-1/+10
|
* [RISCV][GlobalISel] Select RV64 word shiftsRichard-Qin-X2026-08-161-0/+69
|
* [RISCV][GlobalISel] Select constant operands as immediatesRichard-Qin-X2026-08-163-30/+138
|
* [RISCV][GlobalISel] Select word shift pseudosRichard-Qin-X2026-08-151-2/+8
|
* [GlobalISel] Clear virtual register types after selectionRichard-Qin-X2026-08-152-2/+6
|
* [RISCV][GlobalISel] Match LLVM constant materializationRichard-Qin-X2026-08-155-196/+246
|
* [GlobalISel] Preserve MIR live-ins during selectionRichard-Qin-X2026-08-151-20/+0
|
* [RISCV][GlobalISel] Constrain selected builder operandsRichard-Qin-X2026-08-151-1/+73
|
* [GlobalISel] Legalize RISC-V va_startRichard-Qin-X2026-08-152-0/+87
|
* [GlobalISel] Lower inline memory copiesRichard-Qin-X2026-08-151-6/+104
|
* [GlobalISel] Legalize RISC-V jump tablesRichard-Qin-X2026-08-152-0/+206
|
* [GlobalISel] Legalize scalar memory operationsRichard-Qin-X2026-08-151-0/+759
|
* [GlobalISel] Legalize saturating arithmeticRichard-Qin-X2026-08-151-0/+362
|
* [GlobalISel] Legalize signed extended add/subRichard-Qin-X2026-08-151-16/+178
|
* [GlobalISel] Legalize multiply with overflowRichard-Qin-X2026-08-152-0/+203
|
* [GlobalISel] Legalize multi-precision multiplicationRichard-Qin-X2026-08-151-0/+268
|
* [RISCV][GlobalISel] Legalize floating-point classesRichard-Qin-X2026-08-151-1/+67
|
* [GlobalISel] Narrow integer comparisonsRichard-Qin-X2026-08-151-0/+195
|
* [GlobalISel] Lower funnel shiftsRichard-Qin-X2026-08-151-0/+121
|
* [RISCV][GlobalISel] Legalize floating-point remainderRichard-Qin-X2026-08-1511-26/+318
|
* [RISCV][GlobalISel] Widen integer-to-floating conversions to XLENRichard-Qin-X2026-08-151-0/+11
|
* [RISCV][GlobalISel] Legalize floating-point to integer conversionsRichard-Qin-X2026-08-153-39/+68
|
* [GlobalISel] Lower floating-point conversion libcallsRichard-Qin-X2026-08-153-88/+262
|
* [GlobalISel] Lower stack operationsRichard-Qin-X2026-08-153-3/+170
|
* [GlobalISel] Legalize scalar PHI operationsRichard-Qin-X2026-08-152-0/+191
|
* [RISCV][GlobalISel] Preserve Zbb sign extensionsRichard-Qin-X2026-08-152-1/+6
|
* [RISCV][GlobalISel] Legalize scalar bit count operationsRichard-Qin-X2026-08-155-14/+783
|
* [GlobalISel] Lower byte-swap operationsRichard-Qin-X2026-08-152-1/+133
|
* [GlobalISel] Lower atomic subtractionRichard-Qin-X2026-08-151-0/+38
|
* [GlobalISel] Use generic compare-exchange legalizationRichard-Qin-X2026-08-152-295/+46
|
* [GlobalISel] Lower combined division and remainderRichard-Qin-X2026-08-151-0/+29
|
* [GlobalISel] Lower integer operations to libcallsRichard-Qin-X2026-08-153-6/+227
|
* [RISCV] Match explicit target feature semanticsRichard-Qin-X2026-08-151-5/+14
|
* [GlobalISel] Lower integer rotate operationsRichard-Qin-X2026-08-152-0/+134
|
* [GlobalISel] Preserve worklists when moving CSE instructionsRichard-Qin-X2026-08-152-7/+29
|
* [GlobalISel] Lower three-way integer comparisonsRichard-Qin-X2026-08-151-16/+102
|
* [GlobalISel] Lower integer min/max operationsRichard-Qin-X2026-08-151-0/+66
|
* [GlobalISel] Widen scalar memory operationsRichard-Qin-X2026-08-151-0/+39
|
* [GlobalISel] Legalize scalar floating comparisonsRichard-Qin-X2026-08-151-0/+26
|
* [GlobalISel] Legalize scalar pointer castsRichard-Qin-X2026-08-151-0/+57
|
* [GlobalISel] Narrow scalar shiftsRichard-Qin-X2026-08-151-2/+913
|
* [MIR] Print target generic operand typesRichard-Qin-X2026-08-151-2/+2
| | | | Use the instruction descriptor's Generic flag when deduplicating LLT annotations so target-specific generic pseudos follow LLVM MIR spelling.
* [GlobalISel] Narrow scalar selectsRichard-Qin-X2026-08-151-10/+86
| | | | Match LLVM's narrowScalarSelect lowering for wide scalar values. Build widened sources from LLT destinations so CSE hits reuse the actual result register instead of introducing copies.
* [GlobalISel] Narrow scalar bitwise operationsRichard-Qin-X2026-08-151-0/+78
|
* [GlobalISel] Legalize constants and implicit definitionsRichard-Qin-X2026-08-153-34/+195
|
* [GlobalISel] Match LLVM CSE builder result semanticsRichard-Qin-X2026-08-1510-252/+1198
|
* [GlobalISel] Legalize G_ABDS and G_ABDURichard-Qin-X2026-08-144-178/+168
|
generated by cgit v1.2.3 (git 2.46.0) at 2026-10-29 14:47:40 +0000