This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH, middle-end]: Optimize a/sqrt(b/c) into a*sqrt(c/b).


On 6/11/07, Andrew Pinski <pinskia@gmail.com> wrote:

>         PR middle-end/32279
>         * gcc.dg/builtins-11.c: Check folding of a/sqrt(b/c).

Please add a new testcase instead of modifying an old one.  I think
there are only a couple of reasons why you should ever be editing a

Please note that this is _not_ a change, just an addition. It won't interfer with current testcase functionality in any way. It was a common practice in the past that missing folds were just added to various builtin-*.c file, because these subtests were clearly separated, and were pass/fail type. An example of this practice is:

2005-11-28  Roger Sayle  <roger@eyesopen.com>
	    Uros Bizjak  <uros@kss-loka.si>

	PR middle-end/20219
	* gcc.dg/builtins-20.c: Add checks for sin(x)/tan(x) and
	tan(x)/sin(x) transformations.

That said, it is no problem for me to add new file. I'm just worried
of dozens of little files with trivial folding checks.

Uros.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]