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] Removing redundant loads


In message <wvl8ysp3umv.fsf@prospero.boston.redhat.com>, Jason Merrill writes:
 >Do we think that assign_vars crashes indicate something other than "the
 >ssa->normal pass isn't done yet"?
It can indicate that an earlier pass set up overlapping lifetimes as
well.  That's what we're running into on the rs6000 (and presumably
x86 with checking disabled).

I've got a preliminary change I'm poking at which appears to fix the 
rs6000 problem (though it will be several hours before I know if it's
good enough to bootstrap).  I'll be testing it on a non-checking x86
tree momentarily.


jeff


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