[llvm-branch-commits] [lld] [lld][MachO] Batch command-line object files and archive member scanning in parallel (PR #240984)

Liza Burakova via llvm-branch-commits llvm-branch-commits at lists.llvm.org
Tue Nov 10 12:07:11 PST 2026


liza371 wrote:

Some benchmark numbers on mac_rel_static_sym_1

Note: my new baseline is a bit higher because some endpoint security software on my corp machine got an update and now uses more CPU when running benchmarks.

Baseline
  Time (mean ± σ):      5.600 s ±  0.069 s    [User: 6.247 s, System: 3.546 s]
  Range (min … max):    5.478 s …  5.747 s    20 runs
 
Optimized
  Time (mean ± σ):      5.113 s ±  0.062 s    [User: 6.656 s, System: 4.008 s]
  Range (min … max):    5.013 s …  5.272 s    20 runs
 
Baseline
  Time (mean ± σ):      5.585 s ±  0.058 s    [User: 6.262 s, System: 3.562 s]
  Range (min … max):    5.508 s …  5.746 s    20 runs

https://github.com/llvm/llvm-project/pull/240984


More information about the llvm-branch-commits mailing list