This is the mail archive of the libstdc++@gcc.gnu.org mailing list for the libstdc++ project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] New *-*-solaris2.x cross configury, including looserwchar_t requirements for all targets


>I always checkout with CVS_READ=y, so that probably forced them to be
>read-only.  Perhaps the make rule or Deja script should chmod them
>before running the tests?

Try:

Index: testsuite/lib/libstdc++.exp
===================================================================
RCS file: /cvs/gcc/gcc/libstdc++-v3/testsuite/lib/libstdc++.exp,v
retrieving revision 1.11
diff -c -p -r1.11 libstdc++.exp
*** testsuite/lib/libstdc++.exp 5 Aug 2003 01:34:28 -0000       1.11
--- testsuite/lib/libstdc++.exp 26 Sep 2003 21:28:25 -0000
*************** proc v3-init { args } {
*** 58,63 ****
--- 58,65 ----
      # Copy any required data files.
      v3-copy-files [glob -nocomplain "$srcdir/data/*.tst"] $outdir
      v3-copy-files [glob -nocomplain "$srcdir/data/*.txt"] $outdir
+     chmod +w $outdir/*.tst
+     chmod +w $outdir/*.txt
   
      # set LD_LIBRARY_PATH so that libgcc_s, libstdc++ binaries can be found.
      # locate libgcc.a so we don't need to account for different values of
<

> 
>> Odd, as there is a using directive in scope. However, I've checked in
>> your patch to explicitly qualify, and removed the directive.
>
>Hmmm.  I didn't notice the directive at first.  I can't explain that.
>
>Which "4 odd fails on Solaris" were you referring to in your earlier
>message?  

These ones.... it'll be interesting to see if they pass now.

-benjamin


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