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]

Re: [new-regalloc] Problem on x86 with initialization of ra_reg_info




On Fri, 2 Feb 2001, Richard Henderson wrote:

> On Fri, Feb 02, 2001 at 02:25:21PM -0500, Daniel Berlin wrote:
> > > On Fri, 2 Feb 2001 13:37:43 -0500 (EST), Daniel Berlin wrote:
> > > ``Yes it does. There is no point in considering non hard, non fixed
> > > registers, that are never set, anywhere.''
>
> Yes there is, because they _do_ appear.  Write broken code and
> you'll see them a lot.

Okay, so you want i should change it to consider any register that is
def'd or used to be a candidate (which of course, fixes the problem)
Or just give an error, since at this point we know for a fact that it's
used, and *never* initialized (and isn't a hard/fixed/etc register).


>
> > I'd like to find the cause, rather than the symptom.
>
> That is the cause.  I pointed this out the other day.  I still see
> that segfault on alpha for gcc.c-torture/compile/20000728-1.c.
>
I don't:
% ./cc1 ../../gcc/testsuite/gcc.c-torture/compile/20000728-1.c
 foo
Execution times (seconds)
 preprocessing         :   0.00 ( 6%) usr   0.00 ( 0%) sys   0.00 ( 4%)
wall
 lexical analysis      :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.00 ( 0%)
wall
 parser                :   0.00 ( 6%) usr   0.00 ( 0%) sys   0.00 ( 4%)
wall
 varconst              :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.00 ( 0%)
wall
 jump                  :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.00 ( 0%)
wall
 flow analysis         :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.00 ( 0%)
wall
 global alloc          :   0.00 (11%) usr   0.00 ( 0%) sys   0.00 ( 9%)
wall
 flow 2                :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.00 ( 0%)
wall
 shorten branches      :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.00 ( 0%)
wall
 final                 :   0.00 ( 6%) usr   0.00 ( 0%) sys   0.00 ( 4%)
wall
 symout                :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.00 ( 0%)
wall
 rest of compilation   :   0.00 ( 6%) usr   0.00 ( 0%) sys   0.00 ( 4%)
wall
 TOTAL                 :   0.01             0.00             0.02
<dberlin@castro.cygnus.com> {/horton/dberlin/gcc/build/gcc}
%



Oh, wait, i get it with -O1



>
> r~
>


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