Re: [tree-ssa] New regressions as of 2003-11-04

[law at redhat dot com]

In message <20031110191320.GA11846@atrey.karlin.mff.cuni.cz>, Zdenek Dvorak wri
tes:
 >>  >Concerning compile time -- there is nothing much to worry about, since
 >>  >no transformation done by remove_useless_stmts_and_vars does anything
 >>  >important with it.
 >> My timings showed quite the opposite.
 >
 >Huh??? Timings of what?
When I wrote remove_useless_stmts_and_vars I actually took the time to measure
how it impacted compile-time performance.



 >>  >http://gcc.gnu.org/ml/gcc-patches/2003-11/msg00388.html.
 >> And have you compared the timings before and after with your patch?
 >> 
 >> Have you verified that it covers all the cases that we were previously
 >> handling (by looking at the dumps before/after)
 >
 >I have just copied everything that currently indeed does anything to the
 >function, so it should.
I want you to sit down and look at the dumps.  Just saying you copied
everything isn't sufficient since you don't actually know if it's working.

You didn't do this kind of analysis with the COND_EXPR lowering code
for example, and if you had done so you probably would have noticed that
were were missing many transformations that were working with the old
COND_EXPR code, but were not working with the new COND_EXPR code.


 >> Have you verifed that it does the right thing when the CFG isn't available
 >> since this code is called before we build the CFG?
 >
 >No it won't -- I plan to leave the remove_useless_stmts_and_vars call
 >before cfg creation as it is (trusting you that it is useful in some
 >way), and just replace the call to it done at the end of tree-ssa
 >optimizations.
OK.
jeff