On Sep 12, 2005, at 10:27 PM, GCC regression checker wrote:
With your recent patch, GCC HEAD has some regression test failures,
which used to pass. There are 1 new failures, and 13
failures that existed before and after that patch; 3 failures
have been fixed.
The new failures are:
native gcc.sum gcc.c-torture/execute/960512-1.c
This is a latent bug in the darwin's backend, darwin.md.
What is going on is that we moving a restore the PIC register
before are done with it but since this was not described in the
RTL, the scheduler thinks it is not a live at this point.
I will be testing a fix tonight for this by change movdf_low_si
into an define_insn_and_split.