[Qemu-devel] Comment for Solaris fix for the HPTC

[Qemu-devel] Comment for Solaris fix for the HPTC

[Ben Taylor]

Compiling the latest CVS code, I found that a function in vl.c uses a define
called ULONG_LONG_MAX.  The value is defined in solaris's sys/types.h,
but only on _LP64 builds, so 32-bit builds won't pickup if I redefine it as 
ULONG_MAX.

I'm looking for comments on this patch.

--- qemu.ORIG/vl.h      2007-08-26 13:46:00.000000000 -0400
+++ qemu/vl.h   2007-09-15 17:59:24.179299000 -0400
@@ -49,6 +49,12 @@
 #define ENOMEDIUM ENODEV
 #endif

+#ifdef __sun__
+/* Have to define this for Solaris as ULONG_LONG_MAX is not defined
+   anywhere.  ULONG_MAX is correct only on _LP64 systems */
+#define ULONG_LONG_MAX 18446744073709551615UL
+#endif
+
 #ifdef _WIN32
 #include <windows.h>
 #define fsync _commit

Re: [Qemu-devel] Comment for Solaris fix for the HPTC

[Johannes Schindelin]

Hi Ben,

On Sat, 15 Sep 2007, Ben Taylor wrote:

> +#ifdef __sun__
> +/* Have to define this for Solaris as ULONG_LONG_MAX is not defined
> +   anywhere.  ULONG_MAX is correct only on _LP64 systems */
> +#define ULONG_LONG_MAX 18446744073709551615UL
> +#endif
> +

These constants are always defined, right?  So why not "#ifndef 
ULONG_LONG_MAX"?

Ciao,
Dscho

Re: [Qemu-devel] Comment for Solaris fix for the HPTC

[Ben Taylor]

---- Johannes Schindelin <Johannes.Schindelin@gmx.de> wrote: 
> Hi Ben,
> 
> On Sat, 15 Sep 2007, Ben Taylor wrote:
> 
> > +#ifdef __sun__
> > +/* Have to define this for Solaris as ULONG_LONG_MAX is not defined
> > +   anywhere.  ULONG_MAX is correct only on _LP64 systems */
> > +#define ULONG_LONG_MAX 18446744073709551615UL
> > +#endif
> > +
> 
> These constants are always defined, right?  So why not "#ifndef 
> ULONG_LONG_MAX"?

I fear I may not have explained this well.

On the Solaris systems I have access to, I find no system include file
that defines "ULONG_LONG_MAX".

There are defines for ULONG_MAX which is equivilent to the value that
ULONG_LONG_MAX is, but only when the system is being compiled on
an "_LP64" (ie x86_64) system.  When the system is 32-bit, ULONG_MAX
defaults to a typical max value for a 32-bit unsigned int.

That help?

Ben

Re: [Qemu-devel] Comment for Solaris fix for the HPTC

[Andreas Schwab]

Ben Taylor <sol10x86@cox.net> writes:

> ---- Johannes Schindelin <Johannes.Schindelin@gmx.de> wrote: 
>> Hi Ben,
>> 
>> On Sat, 15 Sep 2007, Ben Taylor wrote:
>> 
>> > +#ifdef __sun__
>> > +/* Have to define this for Solaris as ULONG_LONG_MAX is not defined
>> > +   anywhere.  ULONG_MAX is correct only on _LP64 systems */
>> > +#define ULONG_LONG_MAX 18446744073709551615UL
>> > +#endif
>> > +
>> 
>> These constants are always defined, right?  So why not "#ifndef 
>> ULONG_LONG_MAX"?
>
> I fear I may not have explained this well.
>
> On the Solaris systems I have access to, I find no system include file
> that defines "ULONG_LONG_MAX".
>
> There are defines for ULONG_MAX which is equivilent to the value that
> ULONG_LONG_MAX is, but only when the system is being compiled on
> an "_LP64" (ie x86_64) system.  When the system is 32-bit, ULONG_MAX
> defaults to a typical max value for a 32-bit unsigned int.

Your reference to ULONG_MAX is a red herring.  ULONG_MAX is the limit
for unsigned long, and ULONG_LONG_MAX is the limit for unsigned long
long.  If your compiler does not support the long long type then
ULONG_LONG_MAX should not be defined either.  Instead, vl.c should use
UINT64_MAX.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: [Qemu-devel] Comment for Solaris fix for the HPTC

[Johannes Schindelin]

Hi,

On Sun, 16 Sep 2007, Ben Taylor wrote:

> ---- Johannes Schindelin <Johannes.Schindelin@gmx.de> wrote: 
> 
> > On Sat, 15 Sep 2007, Ben Taylor wrote:
> > 
> > > +#ifdef __sun__
> > > +/* Have to define this for Solaris as ULONG_LONG_MAX is not defined
> > > +   anywhere.  ULONG_MAX is correct only on _LP64 systems */
> > > +#define ULONG_LONG_MAX 18446744073709551615UL
> > > +#endif
> > > +
> > 
> > These constants are always defined, right?  So why not "#ifndef 
> > ULONG_LONG_MAX"?
> 
> I fear I may not have explained this well.

I fear _I_ may not have made myself very clear.  What I meant was: why not 
check if ULONG_LONG_MAX is undefined, instead of checking if __sun__ is 
defined?  That way, more than one platform might benefit from this change.

Just an idea,
Dscho

Re: [Qemu-devel] Comment for Solaris fix for the HPTC

[Juergen Keil]

Andreas Schwab wrote:

> Your reference to ULONG_MAX is a red herring.  ULONG_MAX is the limit
> for unsigned long, and ULONG_LONG_MAX is the limit for unsigned long
> long.  If your compiler does not support the long long type then
> ULONG_LONG_MAX should not be defined either.  Instead, vl.c should use
> UINT64_MAX.

Looking at a (draft) c99 standard document, I don't find any references
for an ULONG_LONG_MAX macro, anyway.  The c99 <limits.h> header is supposed 
to define LLONG_MIN, LLONG_MAX and ULLONG_MAX for the "long long" and
"unsigned long long" type limits...