nasm: update to 3.01

[MIvanchev]

Testing the changes

• I tested the changes in this PR: YES

Local build testing

• I built this PR locally for my native architecture, (X86_64-GLIBC)

[MIvanchev]

CC @Rutpiv

[MIvanchev]

There seems to be a big performance difference though: let's wait for the outcome netwide-assembler/nasm#253

[Rutpiv]

Hi @MIvanchev, thanks for the ping. I have this in flight in #57682, currently draft because I'm splitting the findings on libx86emu/stepmania/pcsxr/libretro-mupen64plus into separate PRs for a better reviewer experience.

Quick status across the four: libx86emu already has a fix in the PR that I'm fairly confident in. stepmania has local template work that still needs refining. pcsxr and libretro-mupen64plus I haven't gotten to yet due to dayjob constraints.

On the template side, #57682 does a refactor relative to my earlier work in #55489. That one added a dedicated build option just for docs, which in retrospect didn't really make sense. The current template lets documentation follow the standard build flow.

Got curious about your comment on netwide-assembler/nasm#253. Across the asm files I exercised through the dependent packages, none of them showed signs of a perf regression. Could you point me at the specific file you compiled when you saw the difference? I'd like to try to reproduce it on my side.

[MIvanchev]

Thanks for the update @Rutpiv, I can't quite get you my files (test suite for a client that basically generates random opcodes and uses a lib to emit NASM compilable ASM) but I can try to generate you something similar. It only gets noticeable after thousands of lines though.