forked from OSchip/llvm-project
4ff4e7ea30
The code size cost model for most targets uses the legalization cost for the type of the pointer of a load. If this load is followed directly by a trunc instruction, and is the only use of the result of the load, only one instruction is generated in the target assembly language. This adds a check for this case, and uses the target type of the trunc instruction if so. This did not show any changes in CTMark code size benchmarks. Reviewers: paquette, samparker, dmgreen Differential Revision: https://reviews.llvm.org/D109388 |
||
|---|---|---|
| .. | ||
| cmp-ext-01.ll | ||
| cmp-ext-02.ll | ||
| cmp-mem.ll | ||
| cmp-tofp-scalar.ll | ||
| cmp-tofp.ll | ||
| cmpsel.ll | ||
| divrem-const.ll | ||
| divrem-pow2.ll | ||
| divrem-reg.ll | ||
| ext-load.ll | ||
| ext-of-icmp-cost.ll | ||
| fp-arith.ll | ||
| fp-cast.ll | ||
| huge-immediates.ll | ||
| int-arith.ll | ||
| int-cast.ll | ||
| int-operands-extcost.ll | ||
| intrinsic-cost-crash.ll | ||
| intrinsics.ll | ||
| lit.local.cfg | ||
| load-and-test.ll | ||
| load-ptr-cmp-fold.ll | ||
| load-to-trunc.ll | ||
| load_store.ll | ||
| logic-miscext3.ll | ||
| logical.ll | ||
| logicalop.ll | ||
| memop-folding-int-arith.ll | ||
| numvectorregs.ll | ||
| oldarch-vectors.ll | ||
| reduce-and.ll | ||
| reduce-or.ll | ||
| scalar-cmp-cmp-log-sel.ll | ||
| shuffle.ll | ||
| vectorinstrs-inseltpoison.ll | ||
| vectorinstrs.ll | ||