Re: CVS commit: src/crypto/external/bsd/netpgp/dist/src/lib
On Sat, Jun 26, 2010 at 06:18:11AM +0200, Alistair Crooks wrote: > It would be better to make this a check which is size_t dependent, > rather than platform-dependent. The idea is to black list platforms that don't do %zu and there is no way to do that without breaking cross-compilation. It is still preferable to use that if it is available, e.g. to help format string checks. > I still don't understand why autoconf is passing C pre-processor > directives down. > > #if sizeof(size_t) == sizeof(int) > ... > #elif sizeof(size_t) == sizeof(long) > ... > #endif > > directly in the code is much more readable. Doesn't work as the preprocessor doesn't know what sizeof(size_t) is. The macros are pre-computed similar to what genassym is doing for the NetBSD kernel build. Joerg
Re: CVS commit: src/crypto/external/bsd/netpgp/dist/src/lib
On Sat, Jun 26, 2010 at 05:25:31AM +0200, Joerg Sonnenberger wrote: > On Sat, Jun 26, 2010 at 05:11:39AM +0200, Alistair Crooks wrote: > > On Sat, Jun 26, 2010 at 01:32:05AM +0200, Joerg Sonnenberger wrote: > > > On Fri, Jun 25, 2010 at 11:54:32PM +0200, Alistair Crooks wrote: > > > > Even in C99, the "%lu" method will work unless size_t is bigger > > > > than > > > > unsigned long *and* the value being printed exceeds ULONG_MAX, > > > > which > > > > is unlikely to happen in practice. > > > > Please get the attributions right - I was quoting that text. > > > > > Actually, it doesn't. This method breaks as soon as size_t != u_long and > > > might only work in a few edge cases like the size_t being the last > > > argument and the byte order is Little Endian. This is worse because IIRC > > > Microsoft decided to use IL32LLP64 or something similarly fancy. > > > > Can you give us a reference to this, please? > > E.g. > http://stackoverflow.com/questions/384502/what-is-the-bit-size-of-long-on-64-bit-windows > and the MSDN reference inside. > > > > A more portable approach with autoconf can be found in pkg_install, look > > > for MISSING_SIZE_T_SUPPORT and the corresponding AC_CHECK_SIZEOF calls > > > in configure.ac. > > > > Hmmm, I see this in configure.ac - > > > > AC_CHECK_SIZEOF(int) > > AC_CHECK_SIZEOF(long) > > AC_CHECK_SIZEOF(long long) > > AC_CHECK_SIZEOF(size_t, [#include ]) > > ...compute the sizes to not depend on SIZE_MAX (which would simplify the > logic a lot). > > > and > > > > case $host in > > *-*-hpux*) > > AC_DEFINE(MISSING_SIZE_T_SUPPORT) > > AH_TEMPLATE([MISSING_SIZE_T_SUPPORT], [ > > Define to 1 if the `z' modifider for printf is missing. > > ]) > > ;; > > esac > > The only platform for pkgsrc purposes ATM which lacks the %z support. It would be better to make this a check which is size_t dependent, rather than platform-dependent. > > and > > > > #ifndef MISSING_SIZE_T_SUPPORT > > # define PRIzu "zu" > > #elif SIZEOF_SIZE_T == SIZEOF_INT > > # define PRIzu "u" > > #elif SIZEOF_SIZE_T == SIZEOF_LONG > > # define PRIzu "lu" > > #elif SIZEOF_SIZE_T == SIZEOF_LONG_LONG > > # define PRIzu "llu" > > #else > > # errror "Unknown size_t size" > > #endif > > > > Not quite what I'd been expecting, though, from the glowing description > > above. > > It would be simpler if SIZE_MAX support can be assumed. In that case it > would boil down to > #if SIZE_MAX == INT_MAX > #define PRIzu "u" > #elif SIZE_MAX == LONG_MAX > #define PRIzu "lu" > #else SIZE_MAX == LLONG_MAX > #define PRIzu "llu" > #endif That is even worse, though. I still don't understand why autoconf is passing C pre-processor directives down. #if sizeof(size_t) == sizeof(int) ... #elif sizeof(size_t) == sizeof(long) ... #endif directly in the code is much more readable. Regards, Alistair
Re: CVS commit: src/crypto/external/bsd/netpgp/dist/src/lib
On Fri, Jun 25, 2010 at 09:34:12PM -0600, M. Warner Losh wrote:
> You could easily enough have something like the following in autoconf
> to generate that:
>
> #include
> #include
>
> int main(int argc, char **argv)
> {
> size_t foo = ~0;
> printf("#ifndef SIZE_MAX\n#define SIZE_MAX %llu\n#endif\n",
> (unsigned long long)foo);
> return (0);
> }
There is one problem with this approach. It is not cross-compiling safe.
The code using the sizeof macros from autoconf is.
Joerg
Re: CVS commit: src/crypto/external/bsd/netpgp/dist/src/lib
In message: <20100626032531.ga14...@britannica.bec.de>
Joerg Sonnenberger writes:
: On Sat, Jun 26, 2010 at 05:11:39AM +0200, Alistair Crooks wrote:
: > On Sat, Jun 26, 2010 at 01:32:05AM +0200, Joerg Sonnenberger wrote:
: > > On Fri, Jun 25, 2010 at 11:54:32PM +0200, Alistair Crooks wrote:
: > > > Even in C99, the "%lu" method will work unless size_t is bigger
than
: > > > unsigned long *and* the value being printed exceeds ULONG_MAX,
which
: > > > is unlikely to happen in practice.
: >
: > Please get the attributions right - I was quoting that text.
: >
: > > Actually, it doesn't. This method breaks as soon as size_t != u_long and
: > > might only work in a few edge cases like the size_t being the last
: > > argument and the byte order is Little Endian. This is worse because IIRC
: > > Microsoft decided to use IL32LLP64 or something similarly fancy.
: >
: > Can you give us a reference to this, please?
:
: E.g.
:
http://stackoverflow.com/questions/384502/what-is-the-bit-size-of-long-on-64-bit-windows
: and the MSDN reference inside.
:
: > > A more portable approach with autoconf can be found in pkg_install, look
: > > for MISSING_SIZE_T_SUPPORT and the corresponding AC_CHECK_SIZEOF calls
: > > in configure.ac.
: >
: > Hmmm, I see this in configure.ac -
: >
: > AC_CHECK_SIZEOF(int)
: > AC_CHECK_SIZEOF(long)
: > AC_CHECK_SIZEOF(long long)
: > AC_CHECK_SIZEOF(size_t, [#include ])
:
: ...compute the sizes to not depend on SIZE_MAX (which would simplify the
: logic a lot).
:
: > and
: >
: > case $host in
: > *-*-hpux*)
: > AC_DEFINE(MISSING_SIZE_T_SUPPORT)
: > AH_TEMPLATE([MISSING_SIZE_T_SUPPORT], [
: > Define to 1 if the `z' modifider for printf is missing.
: > ])
: > ;;
: > esac
:
: The only platform for pkgsrc purposes ATM which lacks the %z support.
:
: > and
: >
: > #ifndef MISSING_SIZE_T_SUPPORT
: > # define PRIzu "zu"
: > #elif SIZEOF_SIZE_T == SIZEOF_INT
: > # define PRIzu "u"
: > #elif SIZEOF_SIZE_T == SIZEOF_LONG
: > # define PRIzu "lu"
: > #elif SIZEOF_SIZE_T == SIZEOF_LONG_LONG
: > # define PRIzu "llu"
: > #else
: > # errror "Unknown size_t size"
: > #endif
: >
: > Not quite what I'd been expecting, though, from the glowing description
: > above.
:
: It would be simpler if SIZE_MAX support can be assumed. In that case it
: would boil down to
: #if SIZE_MAX == INT_MAX
: #define PRIzu "u"
: #elif SIZE_MAX == LONG_MAX
: #define PRIzu "lu"
: #else SIZE_MAX == LLONG_MAX
: #define PRIzu "llu"
: #endif
You could easily enough have something like the following in autoconf
to generate that:
#include
#include
int main(int argc, char **argv)
{
size_t foo = ~0;
printf("#ifndef SIZE_MAX\n#define SIZE_MAX %llu\n#endif\n",
(unsigned long long)foo);
return (0);
}
Warner
Re: CVS commit: src/crypto/external/bsd/netpgp/dist/src/lib
On Sat, Jun 26, 2010 at 05:11:39AM +0200, Alistair Crooks wrote: > On Sat, Jun 26, 2010 at 01:32:05AM +0200, Joerg Sonnenberger wrote: > > On Fri, Jun 25, 2010 at 11:54:32PM +0200, Alistair Crooks wrote: > > > Even in C99, the "%lu" method will work unless size_t is bigger than > > > unsigned long *and* the value being printed exceeds ULONG_MAX, which > > > is unlikely to happen in practice. > > Please get the attributions right - I was quoting that text. > > > Actually, it doesn't. This method breaks as soon as size_t != u_long and > > might only work in a few edge cases like the size_t being the last > > argument and the byte order is Little Endian. This is worse because IIRC > > Microsoft decided to use IL32LLP64 or something similarly fancy. > > Can you give us a reference to this, please? E.g. http://stackoverflow.com/questions/384502/what-is-the-bit-size-of-long-on-64-bit-windows and the MSDN reference inside. > > A more portable approach with autoconf can be found in pkg_install, look > > for MISSING_SIZE_T_SUPPORT and the corresponding AC_CHECK_SIZEOF calls > > in configure.ac. > > Hmmm, I see this in configure.ac - > > AC_CHECK_SIZEOF(int) > AC_CHECK_SIZEOF(long) > AC_CHECK_SIZEOF(long long) > AC_CHECK_SIZEOF(size_t, [#include ]) ...compute the sizes to not depend on SIZE_MAX (which would simplify the logic a lot). > and > > case $host in > *-*-hpux*) > AC_DEFINE(MISSING_SIZE_T_SUPPORT) > AH_TEMPLATE([MISSING_SIZE_T_SUPPORT], [ > Define to 1 if the `z' modifider for printf is missing. > ]) > ;; > esac The only platform for pkgsrc purposes ATM which lacks the %z support. > and > > #ifndef MISSING_SIZE_T_SUPPORT > # define PRIzu "zu" > #elif SIZEOF_SIZE_T == SIZEOF_INT > # define PRIzu "u" > #elif SIZEOF_SIZE_T == SIZEOF_LONG > # define PRIzu "lu" > #elif SIZEOF_SIZE_T == SIZEOF_LONG_LONG > # define PRIzu "llu" > #else > # errror "Unknown size_t size" > #endif > > Not quite what I'd been expecting, though, from the glowing description > above. It would be simpler if SIZE_MAX support can be assumed. In that case it would boil down to #if SIZE_MAX == INT_MAX #define PRIzu "u" #elif SIZE_MAX == LONG_MAX #define PRIzu "lu" #else SIZE_MAX == LLONG_MAX #define PRIzu "llu" #endif Joerg
Re: CVS commit: src/crypto/external/bsd/netpgp/dist/src/lib
On Sat, Jun 26, 2010 at 01:32:05AM +0200, Joerg Sonnenberger wrote: > On Fri, Jun 25, 2010 at 11:54:32PM +0200, Alistair Crooks wrote: > > Even in C99, the "%lu" method will work unless size_t is bigger than > > unsigned long *and* the value being printed exceeds ULONG_MAX, which > > is unlikely to happen in practice. Please get the attributions right - I was quoting that text. > Actually, it doesn't. This method breaks as soon as size_t != u_long and > might only work in a few edge cases like the size_t being the last > argument and the byte order is Little Endian. This is worse because IIRC > Microsoft decided to use IL32LLP64 or something similarly fancy. Can you give us a reference to this, please? > A more portable approach with autoconf can be found in pkg_install, look > for MISSING_SIZE_T_SUPPORT and the corresponding AC_CHECK_SIZEOF calls > in configure.ac. Hmmm, I see this in configure.ac - AC_CHECK_SIZEOF(int) AC_CHECK_SIZEOF(long) AC_CHECK_SIZEOF(long long) AC_CHECK_SIZEOF(size_t, [#include ]) and case $host in *-*-hpux*) AC_DEFINE(MISSING_SIZE_T_SUPPORT) AH_TEMPLATE([MISSING_SIZE_T_SUPPORT], [ Define to 1 if the `z' modifider for printf is missing. ]) ;; esac and #ifndef MISSING_SIZE_T_SUPPORT # define PRIzu "zu" #elif SIZEOF_SIZE_T == SIZEOF_INT # define PRIzu "u" #elif SIZEOF_SIZE_T == SIZEOF_LONG # define PRIzu "lu" #elif SIZEOF_SIZE_T == SIZEOF_LONG_LONG # define PRIzu "llu" #else # errror "Unknown size_t size" #endif Not quite what I'd been expecting, though, from the glowing description above. Am I missing a revision of that file or something? It took me ages to find it, since I was looking under src/external first of all - and cvs status does say it's up to date. Thanks, Alistair
Re: CVS commit: src/sys/arch/hpcarm/conf
> > Modified Files: > > src/sys/arch/hpcarm/conf: WZERO3 > > > > Log Message: > > Set options RTC_OFFSET=-540 since Windows Mobile stores localtime > > into the RTC and we can assume most W-ZERO3 users live in JST timezone. > > > > XXX: Probably it would be better to allow a kernel getting RTC_OFFSET value > > XXX: via bootinfo set by boot(8) or boot.cfg(5). > > Hmm, hpcboot does pass rtc offset in bootinfo->timezone. > See e.g. hpcsh/machdep.c Oh, I thought hpcsh had RTC_OFFSET=-540. I'll fix hpcarm to use bootinfo->timezone (it just works [tm]). Anyway, it's still better to make NetBSD/i386 also use timezone passed from boot.cfg(5) via bootloader so that all people can use pre-built GENERIC without patching binary. --- Izumi Tsutsui
Re: CVS commit: src/crypto/external/bsd/netpgp/dist/src/lib
On Fri, Jun 25, 2010 at 11:54:32PM +0200, Alistair Crooks wrote: > Even in C99, the "%lu" method will work unless size_t is bigger than > unsigned long *and* the value being printed exceeds ULONG_MAX, which > is unlikely to happen in practice. Actually, it doesn't. This method breaks as soon as size_t != u_long and might only work in a few edge cases like the size_t being the last argument and the byte order is Little Endian. This is worse because IIRC Microsoft decided to use IL32LLP64 or something similarly fancy. A more portable approach with autoconf can be found in pkg_install, look for MISSING_SIZE_T_SUPPORT and the corresponding AC_CHECK_SIZEOF calls in configure.ac. Joerg
Re: CVS commit: src/crypto/external/bsd/netpgp/dist/src/lib
On Fri, Jun 25, 2010 at 08:40:26PM +, Christos Zoulas wrote:
> In article <20100625183016.ac0be17...@cvs.netbsd.org>,
> Alistair G. Crooks wrote:
> >-=-=-=-=-=-
> >
> >Module Name: src
> >Committed By:agc
> >Date:Fri Jun 25 18:30:16 UTC 2010
> >
> >Modified Files:
> > src/crypto/external/bsd/netpgp/dist/src/lib: misc.c
> >
> >Log Message:
> >Fix build problems on LP64 platforms - thanks to Paul Goyette for the nudge.
>
> That should be %zu really; it is more readable.
%zu is more readble, but it's C99, and even some C99 implementations
don't define it (I have in my mind Microsoft from the time that I brought
up the iSCSI target on Windows, but I may be wrong there)
The following quote is apt:
C99 adds a 'z' modifier specifically for size_t:
printf("Total buffer size: %zu bytes\n", buffer_size);
but many printf implementations don't support it. (Even if your
compiler supports C99 and defines __STDC_VERSION__ appropriately,
that's not, practically speaking, a guarantee that the library also
conforms to C99.)
Even in C99, the "%lu" method will work unless size_t is bigger than
unsigned long *and* the value being printed exceeds ULONG_MAX, which
is unlikely to happen in practice.
--
Keith Thompson (The_Other_Keith) ks...@mib.org
San Diego Supercomputer Center <*>
referenced in:
http://bytes.com/topic/c/answers/221867-portable-way-printf-size_t-instance
Oh, and yes, this is not a "we need to conform to c89 as LCD" post -
this is 2010 after all - it's an "I want netpgp to be as portable as
possible" post.
Regards,
Alistair
Re: CVS commit: src/crypto/external/bsd/netpgp/dist/src/lib
In article <20100625183016.ac0be17...@cvs.netbsd.org>, Alistair G. Crooks wrote: >-=-=-=-=-=- > >Module Name: src >Committed By: agc >Date: Fri Jun 25 18:30:16 UTC 2010 > >Modified Files: > src/crypto/external/bsd/netpgp/dist/src/lib: misc.c > >Log Message: >Fix build problems on LP64 platforms - thanks to Paul Goyette for the nudge. That should be %zu really; it is more readable. christos
Re: CVS commit: src/sys/arch/hpcarm/conf
On Fri, Jun 25, 2010 at 17:40:33 +, Izumi Tsutsui wrote: > Module Name: src > Committed By: tsutsui > Date: Fri Jun 25 17:40:33 UTC 2010 > > Modified Files: > src/sys/arch/hpcarm/conf: WZERO3 > > Log Message: > Set options RTC_OFFSET=-540 since Windows Mobile stores localtime > into the RTC and we can assume most W-ZERO3 users live in JST timezone. > > XXX: Probably it would be better to allow a kernel getting RTC_OFFSET value > XXX: via bootinfo set by boot(8) or boot.cfg(5). Hmm, hpcboot does pass rtc offset in bootinfo->timezone. See e.g. hpcsh/machdep.c SY, Uwe -- u...@stderr.spb.ru | Zu Grunde kommen http://snark.ptc.spbu.ru/~uwe/ | Ist zu Grunde gehen
Re: CVS commit: src/sys/dev/pci
On Fri, Jun 25, 2010 at 01:10:13PM +0900, masan...@iij.ad.jp wrote: > > From: SAITOH Masanobu > Subject: CVS commit: src/sys/dev/pci > Date: Fri, 25 Jun 2010 03:47:58 + > > > Module Name:src > > Committed By: msaitoh > > Date: Fri Jun 25 03:47:57 UTC 2010 > > > > Modified Files: > > src/sys/dev/pci: if_wmreg.h > > > > Log Message: > > The GIO master enable bit in STATUS register is not bit 16 but bit 18. > > It will fix a problem on 82580 SGMII system. > > This is wrong. I've cvs admined with the following message: > > The GIO master enable bit in STATUS register is not bit 16 but bit 18. > It will fix a problem in the reset sequence on PCI-e chips. Is it 18 or 19? -#defineSTATUS_GIO_M_ENA (1U << 16) /* PCIX master enable */ +#defineSTATUS_GIO_M_ENA (1U << 19) /* GIO master enable */ Use __BIT(3) ? Dave -- David Young OJC Technologies dyo...@ojctech.com Urbana, IL * (217) 278-3933
Re: CVS commit: src/sys
On Fri, 25 Jun 2010, Izumi Tsutsui wrote: Module Name:src Committed By: tsutsui Date: Fri Jun 25 15:10:43 UTC 2010 Modified Files: src/sys/kern: init_main.c subr_autoconf.c src/sys/sys: device.h Log Message: Add config_mountroot(9), which defers device configuration after mountroot(), like config_interrupt(9) that defers configuration after interrupts are enabled. This will be used for devices that require firmware loaded from the root file system by firmload(9) to complete device initialization (getting MAC address etc). This could also possibly help us move forward with modularizing some device drivers which might need to be loaded from the file system. - | Paul Goyette | PGP Key fingerprint: | E-mail addresses: | | Customer Service | FA29 0E3B 35AF E8AE 6651 | paul at whooppee.com| | Network Engineer | 0786 F758 55DE 53BA 7731 | pgoyette at juniper.net | | Kernel Developer | | pgoyette at netbsd.org | -
