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]

Re: sparc-sun-solaris2.8 bootstrap: ICE in bb-reorder.c


> Jan, I'm afraid this is an outfall of your patch(es) yesterday:
> 
> /sw/test/gcc/cvs/gcc/libgcc2.c: In function `__lshrdi3':
> /sw/test/gcc/cvs/gcc/libgcc2.c:266: Internal compiler error in emit_jump_to_block_after, at bb-reorder.c:546
> Please submit a full bug report,
> with preprocessed source if appropriate.
> See <URL:http://www.gnu.org/software/gcc/bugs.html> for instructions.
> gmake[3]: *** [libgcc/./_lshrdi3.o] Error 1
> gmake[2]: *** [stmp-multilib] Error 2
Apparently, following check got lost.
The patch compiles fine on i386, but does nothing, solves the crash and I can't
access sparc machine till next week, as our laboratory is relocating.

Could you please test whether it works for you?

Honza


Tue Sep 11 17:36:13 CEST 2001  Jan Hubicka  <jh@suse.cz>

	* bb-reorder.c (fixup_reorder_chain): Fallthru edge to exit block
	is OK.

Index: bb-reorder.c
===================================================================
RCS file: /cvs/gcc/egcs/gcc/bb-reorder.c,v
retrieving revision 1.36
diff -c -3 -p -r1.36 bb-reorder.c
*** bb-reorder.c	2001/09/11 09:39:11	1.36
--- bb-reorder.c	2001/09/11 15:35:38
*************** fixup_reorder_chain ()
*** 695,700 ****
--- 695,704 ----
  	  if (RBI (bb)->next == e_fall->dest)
  	    continue;
  
+ 	  /* An fallthru to exit block.  */
+ 	  if (!RBI (bb)->next && e_fall->dest == EXIT_BLOCK_PTR)
+ 	    continue;
+ 
  	  /* We need a new jump insn.  If the block has only one outgoing
  	     edge, then we can stuff the new jump insn in directly.  */
  	  if (bb->succ->succ_next == NULL)


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