This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [patch] gcc.dg/compat: use fp values from arrays, not expressions
- From: Janis Johnson <janis187 at us dot ibm dot com>
- To: Zack Weinberg <zack at codesourcery dot com>
- Cc: Janis Johnson <janis187 at us dot ibm dot com>, gcc-patches at gcc dot gnu dot org
- Date: Tue, 27 Apr 2004 15:00:04 -0700
- Subject: Re: [patch] gcc.dg/compat: use fp values from arrays, not expressions
- References: <20040426235531.GA5665@us.ibm.com> <87brle6z1r.fsf@egil.codesourcery.com>
On Mon, Apr 26, 2004 at 05:32:16PM -0700, Zack Weinberg wrote:
> Janis Johnson <janis187@us.ibm.com> writes:
>
> > This patch changes the gcc.dg/compat tests that compare floating point
> > values to use values from arrays rather than recomputing them each time
> > they are used. Zack said this should make it easier to hunt down the
> > causes of test failures.
>
> Why not use static initialization? And why keep the g* variables
> around?
I've now got a version that uses static initialization for the arrays
of values but still uses init functions for the g* variables. There
are 16 global variables (g*) for each of 16 struct types, and each
variable of the same type has a different value. Static
initialization of those would be a lot of work if they continued to
have unique values, and those seem useful for making sure that the
correct values are passed.
Janis