This is the mail archive of the gcc@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: reload problem:


Thanks Alexandre,

Well, just to check myself:
I missed a fact, that during instruction recognition, gcc does not look at the condition.
So, during reload pass it faces with a condition which cannot be satisfied. So, it fails to find reloads. So, do I understand gcc issues right, that to meet a condition during instruction recognition I have to write named instructions?


Thanks,
Dmitry

On 22 Jan 2002 12:45:10 -0200
Alexandre Oliva <aoliva@redhat.com> wrote:

> On Jan 21, 2002, dimmy <diwil@mail.ru> wrote:
> 
> > So, the question is: Why gcc does not follow '0' constraint and tries
> > to combine 3 different operands into one insn?
> 
> Because combine does not even look at constraints.  That's regalloc's
> business.  combine uses only predicates and conditions to validate an
> instruction.
> 
> -- 
> Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
> Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
> CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
> Free Software Evangelist    *Please* write to mailing lists, not to me
> 


*********************************************************************
   ("`-''-/").___..--''"`-._     (\       Dimmy the Wild      UA1ACZ
    `6_ 6  )   `-.  (     ).`-.__.`)      Enterprise Information Sys 
    (_Y_.)'  ._   )  `._ `. ``-..-'       Nevsky prospekt,   20 / 44
  _..`--'_..-_/  /--'_.' ,'               Saint Petersburg,   Russia
 (il),-''  (li),'  ((!.-'                 +7 (812) 314-8860, 5585314
*********************************************************************


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