This is the mail archive of the gcc-bugs@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: bug in propagate_one_insn


  In message <20010731145826.B4357@redhat.com>you write:
  > On Tue, Jul 31, 2001 at 03:01:37PM -0600, law@redhat.com wrote:
  > > Err, how did the arguments end up getting set in a different block from
  > > the libcall itself?
  > 
  > Region scheduling moved them.
  > 
  > As far as I can tell, nothing after flow1 uses LIBCALL/RETVAL at all.
  > Anyone got a reason not to remove these notes before combine rather
  > than waiting until after reload?
Seems reasonable.  I think we zap them after reload right now, but if
they're not used between flow1 & reload, we might as well zap them
after flow1.

jeff



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