philipp/hdf5
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

[svn-r12199] Purpose:

Browse Source 
bug fix

Description:
the percent error calculation in h5diff using double precision floating point were causing different results
on the release and debug versions of Visual Studio version 6, due to a compiler bug

Solution:
used single precision (float)

Platforms tested:
windows
linux
solaris
AIX

Misc. update:
This commit is contained in:
Pedro Vicente Nunes
2006-04-04 16:54:48 -05:00
parent cfa54ffdc1
commit 12f01d9cd4
2 changed files with 4 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
tools/h5diff/h5diff_common.c
View File

@@ -96,7 +96,7 @@ void parse_input(int argc, const char* argv[], const char** fname1, const char**
printf("<-d %s> is not a valid option\n", argv[i+1] );
usage();
}
options->delta = atof(argv[i+1]);
options->delta = (float) atof(argv[i+1]);
i++; /* go to next */
}
else
@@ -114,7 +114,7 @@ void parse_input(int argc, const char* argv[], const char** fname1, const char**
printf("<-p %s> is not a valid option\n", argv[i+1] );
usage();
}
options->percent = atof(argv[i+1]);
options->percent = (float) atof(argv[i+1]);
i++; /* go to next */
}
else