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: [tree-ssa] New tests


In message <Pine.BSF.4.55.0305311808320.57679@acrux.dbai.tuwien.ac.at>, Gerald 
Pfeifer writes:
 >On Fri, 30 May 2003 law@redhat.com wrote:
 >> These test two bugs found while working on the tree-ssa code.
 >> [...]
 >> I've checked these into the branch.
 >
 >Wouldn't it be better to check them into mainline?
I've got no strong opinions here.

 >The rationale being that,
 > - this will help avoid merge conflicts with duplicate names of tests
 >   (which we had with some other branch IIRC),
I try pretty hard to make sure this doesn't happen.  And it's pretty trivial
to identify them later.

 > - it will help test mainline right away, and
Possibly, though the tests are highly specific to problems with tree-ssa.
ie, I don't see any value in these tests for the mainline.

 > - the tests won't get lost should the branch not get merged for some
 >   reason (though I don't consider this a serious option for tree-ssa <g>).
Given the tests are for tree-ssa specific issues, if we never merged
tree-ssa, then we would have lost nothing.

Also note that at least one of the tests is likely to be converted to
dump it's tree-ssa optimized output and check that output for properly
optimized code (since I know we miss the trivial redundancy right now).
That will preclude the test from being merged into the mainline.

jeff


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