Re: i586-pc-linux-gnulibc1 regressions

[Jeffrey A Law <law at cygnus dot com>]

  In message <199906020457.VAA07594@atrus.synopsys.com>you write:
  > +FAIL: gcc.c-torture/compile/990519-1.c,  -O1  
  > +FAIL: gcc.c-torture/compile/990519-1.c,  -O2  
  > +FAIL: gcc.c-torture/compile/990519-1.c,  -O3 -fomit-frame-pointer  
  > +FAIL: gcc.c-torture/compile/990519-1.c,  -O3 -g  
  > +FAIL: gcc.c-torture/compile/990519-1.c,  -Os  
  > 
  > The above all fail with an ICE:
  > .../gcc/testsuite/gcc.c-torture/compile/990519-1.c:9:
  > Internal compiler error in `fold_truthop', at fold-const.c:3813
Yup.  Already fixed with Andreas's latest patch to fold-const.c :-)

  > +FAIL: gcc.c-torture/execute/ieee/980619-1.c execution,  -O1 
I've added 980619-1.c to the regression list for this target.  It may be
the case that this test triggers the extended precision problems on the x86.
I'll have to investigate.

  > +FAIL: gcc.dg/noreturn-1.c detect falling off end of noreturn (test for
  > warnings, line 9)
  > +FAIL: gcc.dg/noreturn-1.c this function should not get any warnings (test
  > for bogus messages, line 22)
  > +FAIL: gcc.dg/return-type-1.c warning for falling off end of non-void
  > function (test for warnings, line 9)
These should all be fixed by my latest jump.c change :-)

  > For g++ there do not appear to be any regressions; the snapshot has
  > only the following failures:
  > 
  > FAIL: g++.dg/special/conpr-1.C execution test
  > FAIL: g++.dg/special/conpr-2.C execution test
  > FAIL: g++.dg/special/conpr-3.C execution test
  > FAIL: g++.dg/special/conpr-3.C execution test
I believe those are constructor priority tests.  I do not think egcs-1.1.x
supported that feature (though I will be verifying).  I would not be surprised
if they fail on several systems since they are dependent on specific versions
of the assembler & linker.

Thanks!