This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Inline sqrt for ia64
"Zack Weinberg" <zack@codesourcery.com> writes:
| Gabriel Dos Reis <gdr@integrable-solutions.net> writes:
|
| > "Zack Weinberg" <zack@codesourcery.com> writes:
| >
| > | Gabriel Dos Reis <gdr@integrable-solutions.net> writes:
| > |
| > | > "Zack Weinberg" <zack@codesourcery.com> writes:
| > | >
| > | > | + if (TARGET_INLINE_SQRT_LAT)
| > | > | + {
| > | > | + warning ("not yet implemented: latency-optimized inline square root");
| > | >
| > | > Is a reason why this isn't a call to sorry()?
| > | > (sorry has the prepending text "not yet implemented").
| > |
| > | Because it's a warning, not a hard error.
| >
| > a sorry does not count as an error.
|
| Yes it does:
well, from the diagnostic machinary point of view, it is not an error.
It looks like other part of the compiler decides to test *also* for
occurance of sorry. Thanks for elaborating.
| toplev_main(...)
| {
| ...
| if (errorcount || sorrycount)
| return (FATAL_EXIT_CODE);
| }
|
| That's not the behavior I wanted.
-- Gaby