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: [patch] Partial fix for PR 18048


On Oct 27, 2004, at 11:22 AM, Zdenek Dvorak wrote:

Hello,

here is the patch with the Zack's comment reflected.  Bootstrapped &
regtested on x86_64.


This should also help Java code where we get:
<L1>:;
  i.3 = (unsigned int) i;
  if ((unsigned int) D.376 > i.3) goto <L5>; else goto <L2>;

<L2>:;
  D.385 = _Jv_ThrowBadArrayIndex (i);

<L5>:;
  *(&a->data[0] + (int *) i.3 * 4B) = 0;

;
  i = (int) (i.3 + 1);
  D.376 = a->length;
  if (i >= D.376) goto <L8>; else goto <L1>;

<L8>:;

But the real question is why is &a->data[0] not pulled out of the loop
in the first place.

-- Pinski

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