------- Comment #21 from zadeck at naturalbridge dot com 2008-01-21 13:25 ------- Subject: Re: [4.3 Regression] gfortran.dg/array_constructor_9.f90
bonzini at gnu dot org wrote: > ------- Comment #20 from bonzini at gnu dot org 2008-01-21 13:21 ------- > Subject: Re: [4.3 Regression] gfortran.dg/array_constructor_9.f90 > > > >> I am testing a patch that add eq_notes to the lr problem if you specify >> DF_EQ_NOTES. >> If you and steven still want me to abandon this, i will, but in my mind, >> it seems that if the pass wants chains built with eq_notes, then the >> rest of the df information should match this decision. >> > > Chains is one thing, liveness is another. EQ notes are there as > "reminders of useful information", I don't want them to affect register > allocation (via liveness) if the information was not used. > > > I understand that, that is why if the pass does not specify DF_EQ_NOTES, the lr (and the rest of the info) stays as it is now. But if you are building chains out of them, then you are treating them as live anyway. kenny -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34884
