[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: integer types on HP NonStop
From: |
Schmitz, Joachim |
Subject: |
RE: integer types on HP NonStop |
Date: |
Sat, 2 Oct 2010 17:27:23 +0000 |
The platform used to support long long, but for some stupid reason not unsigned
long long (not sure about signed long long), this has been solved meanwhile
(and quite a while ago). As of newest we even have a full c99 compiler (called
c99). We also have a -Wc99lite switch to our c89 compiler since a while (I
guess that is when the unsigned long long issue got solved), which enables many
c99 features, but is of course not fully compliant. Both exist only on TNS/E
the Itanium based systems, not on TNS/R the older, MIPS based systems.
That sample code compiles with a gazillon warnings "implicit conversion from
"int" to "unsigned char": rounding, sign extension, or loss of accuracy may
result" and the likes. These and it's output attached...
Bye, Jojo
-----Original Message-----
From: Bruno Haible [mailto:address@hidden
Sent: Saturday, October 02, 2010 1:56 PM
To: Schmitz, Joachim; address@hidden
Subject: integer types on HP NonStop
Hi Joachim,
Have the severe compiler bugs regarding integer type operations
that were reported about Tandem/NSK in 2006 [1][2] been fixed by now?
Can you please try the test program from [1], compile it with
-D__CHAR_UNSIGNED__ and -O, and post its output?
Bruno
[1] http://lists.gnu.org/archive/html/bug-gnulib/2006-10/msg00135.html
[2] http://lists.gnu.org/archive/html/bug-gnulib/2006-10/msg00142.html
intparam.out
Description: intparam.out