Re: arcbios_calls.S data hazards

[Steve Rumble <stephen.m.rumble%gmail.com@localhost>]

On Mon, Dec 8, 2025 at 11:23 AM Maciej W. Rozycki <macro%orcam.me.uk@localhost> wrote:
>  I guess you need to figure out what's wrong with your build system that
> prevents the assembler from doing its job.  I'm only familiar with the
> compiler tool chain and not the NetBSD compilation setup.

Thanks a lot for the detailed response!

This ended up just me being daft. Given the data dependencies, the
assembler can't move anything into the BDS I was looking at without
creating a new data hazard.

Also, since we're invoking via gcc (-x assembler-with-cpp) and aren't
specifying an optimisation level, it looks like gcc passes its -O0
default into as, overriding as' -O2 default. If I pass -O2 to either
gcc or into as specifically (via -Wa), as does appear to fill other
delay slots where it can. I imagine we're producing suboptimal output
where reordering is currently being used.

Thanks again,
Steve