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]

Re: [patch] New flag -Wsystem-headers


 > From: Benjamin Kosnik <bkoz@redhat.com>
 > 
 > > >gcc/testsuite/gcc.sum:FAIL: gcc.dg/cpp/syshdr.c (test for excess errors)
 > > >gcc/testsuite/gcc.sum:FAIL: gcc.dg/wtr-int-type-1.c decimal constant (test for warnings, line 40)
 > 
 > Not quite sure if they are related to -Wsystem-headers. Perhaps
 > -Wno-system-headers could be used for these files...
 > 
 > from gcc.dg/wtr-int-type-1.c:
 > /* Test for -Wtraditional warnings on integer constant types.
 >    Note, gcc should omit these warnings in system header files.
 >    By Kaveh R. Ghazi <ghazi@caip.rutgers.edu> 8/22/2000.  */
 > /* { dg-do compile } */
 > /* { dg-options "-Wtraditional" } */


I believe the wtr-int-type-1.c regression is harmless.  It stems from
a missing warning in system headers peripheral to what I was testing.

I told dejagnu to expect and ignore that warning, since I couldn't
write the test without triggering the extra noise.  Now with the
default being that no warnings from system headers occur, the
"expected" warning doesn't happen and so we get a "failure".

Since the diagnostic isn't what I was testing for, simply not
expecting it any longer should fix it.  I'll try to take care of it
tonight.

		--Kaveh
--
Kaveh R. Ghazi			Engagement Manager / Project Services
ghazi@caip.rutgers.edu		Qwest Internet Solutions

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