headers
Bucky Katz | 2 Aug 2007 22:16

dumb question: tracking gcc in netbsd-4


I should be able to find the answer to this in documentation but I'm
clearly not looking with the right keywords, sorry for asking the
list:

As 4.0 makes its way through beta, gcc has, of course, gotten
ahead. What is the policy for dealing with this? Say frozen on the
current gcc 4.1 snapshot or update to a gcc 4.2?

Thanks,

Bucky
Permalink | Reply | 1 comment |
headers
matthew green | 3 Aug 2007 03:25
Picon
Favicon

re: dumb question: tracking gcc in netbsd-4


   I should be able to find the answer to this in documentation but I'm
   clearly not looking with the right keywords, sorry for asking the
   list:

   As 4.0 makes its way through beta, gcc has, of course, gotten
   ahead. What is the policy for dealing with this? Say frozen on the
   current gcc 4.1 snapshot or update to a gcc 4.2?

gcc 4.1 will ship with netbsd 4.x.

it's not yet clear what will happen for 5.x tho i am hoping
we will have gcc 4.2.  stay tuned for more details...

.mrg.
Permalink | Reply | 0 comments |
headers
Kurt Lidl | 6 Aug 2007 18:16

cross-building from macos x

I recently attempted to cross-build netbsd on my Mac.

I ran into a problem that has already been examined by
another user.

Is there anybody on this mailing list that can comment
on the changes suggested in this posting:

http://mail-index.netbsd.org/netbsd-bugs/2007/07/29/0002.html

Thanks!

-Kurt
Permalink | Reply | 5 comments |
headers
Joerg Sonnenberger | 6 Aug 2007 18:20
Picon

Re: cross-building from macos x

On Mon, Aug 06, 2007 at 12:16:51PM -0400, Kurt Lidl wrote:
> Is there anybody on this mailing list that can comment
> on the changes suggested in this posting:

It should only include the minimal version from the #else case.
It doesn't have to be fast, so just providing the default version in C
is good enough.

Joerg
Permalink | Reply | 4 comments |
headers
Alan Barrett | 6 Aug 2007 22:09
Gravatar

Re: cross-building from macos x

On Mon, 06 Aug 2007, Joerg Sonnenberger wrote:
> On Mon, Aug 06, 2007 at 12:16:51PM -0400, Kurt Lidl wrote:
> > http://mail-index.netbsd.org/netbsd-bugs/2007/07/29/0002.html
> 
> It should only include the minimal version from the #else case.
> It doesn't have to be fast, so just providing the default version in C
> is good enough.

I agree.  I will do some testing and then commit a fix along those
lines, unless somebody else beats me to it.

--apb (Alan Barrett)
Permalink | Reply | 3 comments |
headers
Kurt J. Lidl | 7 Aug 2007 22:10

Re: cross-building from macos x

On Mon, Aug 06, 2007 at 10:09:19PM +0200, Alan Barrett wrote:
> On Mon, 06 Aug 2007, Joerg Sonnenberger wrote:
> > On Mon, Aug 06, 2007 at 12:16:51PM -0400, Kurt Lidl wrote:
> > > http://mail-index.netbsd.org/netbsd-bugs/2007/07/29/0002.html
> > 
> > It should only include the minimal version from the #else case.
> > It doesn't have to be fast, so just providing the default version in C
> > is good enough.
> 
> I agree.  I will do some testing and then commit a fix along those
> lines, unless somebody else beats me to it.

The following patch (derived from the original one) progresses
beyond the very early failure that I saw before.  Hopefully this
qualifies as "the minimal version", and can be commited.  Thanks!

-Kurt

Index: tools/compat/configure.ac
===================================================================
RCS file: /cvsroot/src/tools/compat/configure.ac,v
retrieving revision 1.60
diff -u -r1.60 configure.ac
--- tools/compat/configure.ac	9 Oct 2006 14:59:05 -0000	1.60
+++ tools/compat/configure.ac	6 Aug 2007 19:22:53 -0000
 <at>  <at>  -142,6 +142,8  <at>  <at> 

 AC_CHECK_DECLS([bswap16, bswap32, bswap64],,, [#include <machine/bswap.h>])

+AC_CHECK_DECLS([be16enc, le16enc, be16dec, le16dec, be32enc, le32enc, be32dec, le32dec, be64enc,
(Continue reading)

Permalink | Reply | 2 comments |
headers
Alan Barrett | 7 Aug 2007 23:40
Gravatar

Re: cross-building from macos x

On Tue, 07 Aug 2007, Kurt J. Lidl wrote:
> The following patch (derived from the original one) progresses
> beyond the very early failure that I saw before.  Hopefully this
> qualifies as "the minimal version", and can be commited.  Thanks!

I committed my own version earlier today.  See revision 1.60 of
src/tools/compat/compat_defs.h.

--apb (Alan Barrett)
Permalink | Reply | 1 comment |
headers
Kurt J. Lidl | 8 Aug 2007 05:15

Re: cross-building from macos x

On Tue, Aug 07, 2007 at 11:40:53PM +0200, Alan Barrett wrote:
> On Tue, 07 Aug 2007, Kurt J. Lidl wrote:
> > The following patch (derived from the original one) progresses
> > beyond the very early failure that I saw before.  Hopefully this
> > qualifies as "the minimal version", and can be commited.  Thanks!
> 
> I committed my own version earlier today.  See revision 1.60 of
> src/tools/compat/compat_defs.h.

Thanks!  Can this change be requested as a pullup to the netbsd-4 branch too?
That would be greatly appreciated.

-Kurt
Permalink | Reply | 0 comments |
headers
HQVG1 | 10 Aug 2007 23:28

Email not allowed

A message sent by this account comes from
a domain or host not allowed by this mail server.

From: tech-toolchain <at> netbsd.org
To: tech-tipsmailgeek <at> computergeeks.com
Permalink | Reply | 0 comments |
headers
Matthias Drochner | 13 Aug 2007 16:39
Picon
Picon
Favicon

accuracy of "long double"


Inspired by a selftest failure of sqlite3, I tried to
understand how "long double" math is supposed to work
if the FPU is set to 53-bit rounding.
My findings are posted as PR port-i386/36770 and a reply.

Before I take further steps (like the gcc configuration
change mentioned there) I'd like to get a second opinion.

Could someone understanding the matter have a look at
that PR and tell whether I'm correct?

thanks & best regards
Matthias

Forschungszentrum Juelich GmbH
52425 Juelich

Sitz der Gesellschaft: Juelich
Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498
Vorsitzende des Aufsichtsrats: MinDirig'in Baerbel Brumme-Bothe
Vorstand: Prof. Dr. Achim Bachem (Vorsitzender), Dr. Ulrich Krafft (stellv. 
Vorsitzender)
Permalink | Reply | 22 comments |

Gmane