[v2,0/8] x86: a few more optimizations

Message ID 72726722-9d28-4d82-84ef-320e1786b0e4@suse.com
Headers
Series x86: a few more optimizations |

Message

Jan Beulich June 21, 2024, 12:47 p.m. UTC
  APX {nf} insn forms present a number of interesting optimization
opportunities, often mostly for size. There are a few more that I'm
aware of, but where I'm less convinced that input code would really
ever be written in a way triggering the (hypothetical) code.

Two non-{nf} ones and a change limiting when a certain optimization
is done are here in addition, as I came to think of them while doing
the preparations / work.

1: optimize {nf} forms of ADD/SUB with specific immediates
2: optimize {nf}-form rotate-by-width-less-1
3: optimize certain {nf}-form insns to LEA
4: restrict by-imm31 optimization
5: extend TEST-by-imm7 optimization to CTESTcc
6: optimize {nf}-form IMUL-by-power-of-2 to SHL
7: optimize certain {nf}-form insns to BMI2 ones
8: apply NDD-to-legacy transformation to further CMOVcc forms

Jan