Re: compile time testsuite [Re: alias slowdown?]

[Diego Novillo <dnovillo at redhat dot com>]

Benjamin Kosnik wrote on 11/21/06 08:54:
> > Perhaps we need a webpage with multiple compile times being tracked,
> > or simply a standard compile time regression suite from existing PRs
> > to avoid this kind of thing in the future.
>
> I think it is past time for a formal compile-time testsuite.
>
> My suggestion would be to check in 
>
> all.i
> bug2.ii
> cpgram.ii
>
> And whatever stuff there is for measuring slowdowns, and then create
> an optional "make check-compile" rule for gcc/Makefile that runs
> through the tests. I believe that you will have to make this testing
> always optional, since there will no doubt be pathological behavior
> involved.
>
> Then you could do
>
> time make check-compile
Yup, I like this idea.  I would only avoid adding .ii files, as they are 
very brittle.  I use this sandbox:

DLV (Gerald's)
MICO (www.mico.org)
cc1-i-files (.i and .ii files from a full bootstrap)
tramp3d (Richard G's)
SPEC2000
two .ii files from Bugzilla PRs (cpgram.ii from 15855 and rt-3.4.ii from 
some other PR I can't remember ATM)

With the exception of DLV and SPEC2000, we could probably add most of 
them to the check-compile testsuite.  I don't know if Gerald would be 
able to allow DLV to be included.