[Bug math/34319] fromfp(INFINITY, ...) not NAN

https://sourceware.org/bugzilla/show_bug.cgi?id=34319

Adhemerval Zanella <adhemerval.zanella at linaro dot org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |adhemerval.zanella at linaro dot o
                   |                            |rg
         Resolution|---                         |DUPLICATE
             Status|UNCONFIRMED                 |RESOLVED

--- Comment #1 from Adhemerval Zanella <adhemerval.zanella at linaro dot org> ---
The fromfp before 2.43 followed the TS 18661-1 signature that returned intmax_t
and that's why you are seeing this value:

$ gcc -Wall bz34319.c -o bz34319 -std=c2x -lm
bz34319.c:7: warning: ignoring ‘#pragma STDC FENV_ACCESS’ [-Wunknown-pragmas]
    7 | #pragma STDC FENV_ACCESS ON
      |
$ objdump -t bz34319 | grep fromfp
0000000000000000       F *UND*  0000000000000000              fromfp@GLIBC_2.25

It binds to the old/compat implementation.  If you build against a new glibc,
it works as expected:

$ x86_64-glibc-linux-gnu-gcc -Wall bz34319.c -o bz34319 -std=c23 -lm
bz34319.c:7: warning: ignoring ‘#pragma STDC FENV_ACCESS’ [-Wunknown-pragmas]
    7 | #pragma STDC FENV_ACCESS ON
$ objdump -t bz34319 | grep fromfp
0000000000000000       F *UND*  0000000000000000              fromfp@GLIBC_2.43
$ elf/ld.so --library-path .:math ./bz34319
-nan

*** This bug has been marked as a duplicate of bug 28327 ***

-- 
You are receiving this mail because:
You are on the CC list for the bug.