diff options
| author | Albert Cheng <acheng@hdfgroup.org> | 2000-03-08 23:25:15 (GMT) |
|---|---|---|
| committer | Albert Cheng <acheng@hdfgroup.org> | 2000-03-08 23:25:15 (GMT) |
| commit | a1b8768af219ad9e38f3f2c92b9b7c586a285a7f (patch) | |
| tree | dc66e3ad4166a61ad4f63479c217b297371af112 /src | |
| parent | d216d71c168a46087c2334f07fadfa779ef288dc (diff) | |
| download | hdf5-a1b8768af219ad9e38f3f2c92b9b7c586a285a7f.zip hdf5-a1b8768af219ad9e38f3f2c92b9b7c586a285a7f.tar.gz hdf5-a1b8768af219ad9e38f3f2c92b9b7c586a285a7f.tar.bz2 | |
[svn-r2009] Purpose
Bug fix.
Description
When converting a floating point zero to a different format (e.g. Cray),
the mantissa part was not set to zeros complete. Cray does not
seem to mind since the exponent part was set to all 0's. Somehow,
the HDF4 does not like it and convert that pattern to a non-zero
value in IEEE.
Solution
corrected the code to set the whole mantissa to zeros completely.
Platform tested:
J90 & Solaris 2.7.
Diffstat (limited to 'src')
| -rw-r--r-- | src/H5Tconv.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/src/H5Tconv.c b/src/H5Tconv.c index 6ec11df..3d29e7d 100644 --- a/src/H5Tconv.c +++ b/src/H5Tconv.c @@ -2496,7 +2496,7 @@ H5T_conv_f_f (hid_t src_id, hid_t dst_id, H5T_cdata_t *cdata, size_t nelmts, /* +0 or -0 */ H5T_bit_copy (d, dst.u.f.sign, s, src.u.f.sign, 1); H5T_bit_set (d, dst.u.f.epos, dst.u.f.esize, FALSE); - H5T_bit_set (d, dst.u.f.mpos, dst.u.f.esize, FALSE); + H5T_bit_set (d, dst.u.f.mpos, dst.u.f.msize, FALSE); goto padding; } else if (H5T_bit_find (s, src.u.f.epos, src.u.f.esize, H5T_BIT_LSB, FALSE)<0) { |