This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [v3] Enable four 27_io tests
- From: Paolo Carlini <pcarlini at unitus dot it>
- To: Ulrich Weigand <Ulrich dot Weigand at de dot ibm dot com>
- Cc: gcc-patches at gcc dot gnu dot org, bkoz at redhat dot com
- Date: Fri, 25 Jan 2002 21:25:08 +0100
- Subject: Re: [v3] Enable four 27_io tests
- References: <OFC1C0CD92.61267148-ONC1256B4C.006E8058@de.ibm.com>
Ulrich Weigand wrote:
> long l = -1;
> [...]
> o << oct << l << ' ' << hex << l;
> VERIFY( strbuf.str() == "37777777777 ffffffff" );
> strbuf.str(str_blank);
>
> Shouldn't the test take the size of the long type into account here?
You are correct.
Indeed, the whole test03 of ostream_inserter_arith.cc blindly assumes way too
much (16 bit shorts, 32 bit ints and longs).
I'm not sure however if at this point is really worth properly fixing the test,
making it platform independent, or we can simply disable it again...
Benjamin?
Thanks for your feeback,
Paolo.