IRC channel logs

<Guest74>This must be a cc_amd64 bug, no?

<stikonas>or its limitation

<stikonas>so the "fix" could be workaround in M2-Planet too

<stikonas>in any case, we don't know what causes it, so we need to gather more info first

<stikonas>root cause it

<stikonas>and then we can decide on the best solution

<stikonas>(not doing anything is also an option if it is too hard)

<stikonas>since #if __M2__ works, only #if !__M2__ doesn't

<Guest74>Is there a reason cc_amd64 wouldn't be able to correctly parse `return !macro_primary_expr();`?

<stikonas>hmm, don't remember if it correctly handles non 1 true values

<stikonas>Guest74: looks that this might be the culprit

<stikonas>I'm trying #if !A

<stikonas>./AMD64/artifact/M2 -D A=1 -f test2.c -E and ./AMD64/artifact/M2 -D A=42 -f test2.c -E give different results

<stikonas>M2-Planet has better handling of !

<stikonas>I think I actually fixed that in M2-Planet a couple of years ago

<stikonas>so we either need to fix all cc_* compiler assembly output

<stikonas>there almost 10 of them

<stikonas>there are almost 10 of them

<stikonas>or we workaround that line that you mentioned with !macro_primary_expr()

<stikonas>Guest74: could you also open an issue in M2-Planet repo

<stikonas>(or perhaps issue in stage0-posix repo is better...)

<stikonas>(since it's not really M2-Planet bug)

<stikonas>personally I think fixing all cc_* compilers would be cleaner

<stikonas>yes, this function...

2025-06-12.log