diff options
| author | Alan Modra <amodra@bigpond.net.au> | 2004-03-04 10:23:23 +0000 |
|---|---|---|
| committer | Alan Modra <amodra@gcc.gnu.org> | 2004-03-04 20:53:23 +1030 |
| commit | ddc6856430bcef3e68563b08cc513f794a5cb57e (patch) | |
| tree | 2a5b34507a773fc68de86a6f3a19dec1a8df7b3e /gcc/real.c | |
| parent | ca87076c9bfac0090fcc26c06dd1c24dc5e72bc9 (diff) | |
| download | gcc-ddc6856430bcef3e68563b08cc513f794a5cb57e.zip gcc-ddc6856430bcef3e68563b08cc513f794a5cb57e.tar.gz gcc-ddc6856430bcef3e68563b08cc513f794a5cb57e.tar.bz2 | |
real.c (encode_ibm_extended): Don't bother rounding low double.
* real.c (encode_ibm_extended): Don't bother rounding low double.
* c-cppbuiltin.c (builtin_define_float_constants): Tweak MAX
when fmt->pnan < fmt->p.
From-SVN: r78900
Diffstat (limited to 'gcc/real.c')
| -rw-r--r-- | gcc/real.c | 4 |
1 files changed, 3 insertions, 1 deletions
@@ -3248,7 +3248,9 @@ encode_ibm_extended (const struct real_format *fmt, long *buf, if (u.class == rvc_normal) { do_add (&v, &normr, &u, 1); - round_for_format (base_fmt, &v); + /* The low double won't need rounding, since we round to a 106 bit + mantissa before calling this function, and we've just + subtracted off the top 54 bits. (53+1 because u is rounded.) */ encode_ieee_double (base_fmt, &buf[2], &v); } else |