headers
Havard Eidnes | 2 Oct 2006 17:14
Picon

Re: Package binaries for NetBSD/sparc64 4.0 / pkgsrc-2006Q2

Hi,

my sparc64 bulk build machine running 4.0_BETA just completed its
third round of bulk building of pkgsrc-2006Q2.

I've uploaded and merged the result into

   ftp://ftp.NetBSD.org/pub/pkgsrc/packages-2006Q2/NetBSD-4.0/sparc64/

A total of 869MB of packages were uploaded.

The source this was built from was updated Sep 19 2006.  Among
the upgrades present in this round are: SDL (which drags along
quite a few new packages), BIND, firefox, KDE (a few new binary
packages), mailman, mysql, nas, openldap, openssl, seamonkey, and
xine.

As already noted, due to the upgraded g++ compiler in 4.0_BETA,
the C++ ABI in NetBSD 4.0 has changed compared to earlier NetBSD
releases.  This means that mixing C++ binaries and libraries
across the version boundary is not going to work.

This bulk package build host is now proceeding to the first build on
the pkgsrc-2006Q3 branch, so this will be the last update along
pkgsrc-2006Q2 from this corner for this mix of CPU architecture, OS
version and pkgsrc version.

Regards,

- Håvard
(Continue reading)

Permalink | Reply | 0 comments |
headers
Julian Coleman | 4 Oct 2006 17:52
Picon

Binary packages for NetBSD-3.0/sparc64 updated

The results of a pkgsrc bulk build update of pkgsrc-2006Q2 on
NetBSD-3.0.1/sparc64 have been uploaded and the packages are now available
at:

  ftp://ftp.NetBSD.org/pub/NetBSD/packages/packages-2006Q2/NetBSD-3.0/sparc64/

and its mirrors.

The packages were built from sources updated on September 14, 2006.  2305
packages were built in this update.

The next 3.0.1 bulk build will be with the new 2006Q3 branch.

J

--

-- 
  My other computer also runs NetBSD    /        Sailing at Newbiggin
        http://www.netbsd.org/        /   http://www.newbigginsailingclub.org/
Permalink | Reply | 0 comments |
headers
Ben Collver | 5 Oct 2006 14:08

4.0 BETA succeeds

I installed 4.0 BETA from releng 20061001 on my Ultra 10 without
problem.  I am using LFS on SATA and it is going fairly zippy.  When
compiling things, it seems CPU bound.

Ben
Permalink | Reply | 0 comments |
headers
John Nemeth | 6 Oct 2006 10:54
Picon
Favicon

floppy support

     I have committed initial support for floppy drives.  The code has
been working on my Ultra 2 for some time now.  Unfortunately, the ebus
support doesn't work and I don't have hardware.  If anybody can help
with testing and fixing ebus support, that would be great.  Given that
this is working on my systems, hopefully nothing broke.  However, I
have started a full build on my Ultra 2.  If the build did break, I
will catch it; but, it could take a couple of days for my build to
complete.  Here is the commit message:

On Jan 21,  8:52pm, John Nemeth wrote:
} Subject: CVS commit: src/sys/arch/sparc64
} 
} Module Name:	src
} Committed By:	jnemeth
} Date:		Fri Oct  6 08:44:59 UTC 2006
} 
} Modified Files:
} 	src/sys/arch/sparc64/conf: GENERIC files.sparc64
} 	src/sys/arch/sparc64/dev: auxio.c auxioreg.h
} 	src/sys/arch/sparc64/include: intr.h
} Added Files:
} 	src/sys/arch/sparc64/dev: auxiovar.h fdc.c fdcreg.h fdcvar.h
} 
} Log Message:
} Initial support for floppy drives on sparc64.  This has been tested
} on an Ultra 2 and works fine, apart from formatting which is known
} to be broken.  It failed to work on an ebus machine.  The ebus
} support compiles fine, but I don't have hardware for testing.  This
} code is based on the sparc driver with hints from OpenBSD on how
} to do the sbus and ebus attachments, along with help from martin <at>
(Continue reading)

Permalink | Reply | 4 comments |
headers
Jochen Kunz | 6 Oct 2006 17:21
Picon

Re: floppy support

On Fri, 6 Oct 2006 01:54:36 -0700
jnemeth <at> victoria.tc.ca (John Nemeth) wrote:

>      I have committed initial support for floppy drives.  The code has
> been working on my Ultra 2 for some time now.  Unfortunately, the ebus
> support doesn't work and I don't have hardware.  If anybody can help
> with testing and fixing ebus support, that would be great.
I have a spare Ultra10 that I can use for tests.
cvs update is already running.
What needs to be done for ebus?
--

-- 

tschüß,
       Jochen

Homepage: http://www.unixag-kl.fh-kl.de/~jkunz/
Permalink | Reply | 0 comments |
headers
John Nemeth | 6 Oct 2006 21:16
Picon
Favicon

Re: floppy support

On Feb 26, 11:57am, Jochen Kunz wrote:
} On Fri, 6 Oct 2006 01:54:36 -0700 jnemeth <at> victoria.tc.ca (John Nemeth) wrote:
} 
} >      I have committed initial support for floppy drives.  The code has
} > been working on my Ultra 2 for some time now.  Unfortunately, the ebus
} > support doesn't work and I don't have hardware.  If anybody can help
} > with testing and fixing ebus support, that would be great.
} I have a spare Ultra10 that I can use for tests.
} cvs update is already running.
} What needs to be done for ebus?

     First uncomment the "fdc0 at ebus?" line and build a kernel.
After that, test and debug.  The report I have, said that the
controller attaches but the drive fails to attach.  I based the ebus
attachment on another NetBSD driver (I think it was auxio) and the
OpenBSD floppy driver (their ebus code is considerably different then
ours so I couldn't port the ebus attachment directly).  Thanks for
taking a look at this.  Please let me know what you find.

}-- End of excerpt from Jochen Kunz
Permalink | Reply | 0 comments |
headers
Jochen Kunz | 6 Oct 2006 21:40
Picon

Re: floppy support

On Fri, 6 Oct 2006 01:54:36 -0700
jnemeth <at> victoria.tc.ca (John Nemeth) wrote:

> Unfortunately, the ebus
> support doesn't work and I don't have hardware.  If anybody can help
> with testing and fixing ebus support, that would be great.
GENERIC kernel from todays source with "fdc0 at ebus?" enabled:

# dmesg | grep ^fd
fdc0 at ebus0 addr 3023f0-3023f7, 706000-70600f, 720000-720003 ipl 39 softpri 4: chip 82077: manual eject
fd0 at fdc0 drive 0: 1.44MB 80 cyl, 2 head, 18 sec

# dd if=/dev/rfd0c of=/tmp/rfd0c   
fd0: timeout: state 10
fd0: timeout: state 13
fd0: timeout: state 13
fd0: timeout: state 13
fd0: timeout: state 13
fd0: timeout: state 13
fd0c: hard error reading fsbn 0
fd0: controller status: state 13
fd0: timeout: state 20
fd0: timeout: state 13
fd0: timeout: state 13
fd0: timeout: state 13
fd0: timeout: state 13
fd0: timeout: state 13
fd0c: hard error reading fsbn 0
fd0: controller status: state 13
dd: /dev/rfd0c: Input/output error
(Continue reading)

Permalink | Reply | 0 comments |
headers
John Nemeth | 6 Oct 2006 22:20
Picon
Favicon

Re: floppy support

On Feb 26,  4:16pm, Jochen Kunz wrote:
} On Fri, 6 Oct 2006 01:54:36 -0700 jnemeth <at> victoria.tc.ca (John Nemeth) wrote:
} 
} > Unfortunately, the ebus
} > support doesn't work and I don't have hardware.  If anybody can help
} > with testing and fixing ebus support, that would be great.
} GENERIC kernel from todays source with "fdc0 at ebus?" enabled:
} 
} # dmesg | grep ^fd
} fdc0 at ebus0 addr 3023f0-3023f7, 706000-70600f, 720000-720003 ipl 39 softpri 4: chip 82077: manual eject
} fd0 at fdc0 drive 0: 1.44MB 80 cyl, 2 head, 18 sec

     Hmm, a floppy drive attachment; that's a good start.

} # dd if=/dev/rfd0c of=/tmp/rfd0c   

     Use rfd0a.  Floppy drives are different from hard drives in that
the final letter is used to indicated the type of floppy and not a
partition.  By using rfd0c you told it that you have a 720KB floppy.

}-- End of excerpt from Jochen Kunz
Permalink | Reply | 1 comment |
headers
Kurt Schreiner | 6 Oct 2006 22:42
Picon
Picon
Favicon

crashing kernel on ultra2

Hi,

just got this crash on an oldish Ultra2. Kernel -ajusted to the available
hardware and needed features- compiled from sources cvs update'd an hour or
so ago. The system crashed while trying to compile a new userland.
Src- and objdirs are come over nfs. The crash is easily reproduceable by just
calling build.sh.  Kernel from Tue Oct  3 19:38:04 MEST 2006 works fine...

Kurt

cpu0: data fault: pc=10fd2b0 addr=1b000000
kernel trap 30: data access exception
Stopped in pid 3582.1 (nbmake) at       netbsd:uvm_pagealloc_strat+0x290:       s
tx              %i0, [%g3 + %g0]
db> bt
uvn_findpage(df0b790, 0, cb10e00, 0, 700, 11853fc) at netbsd:uvn_findpage+0x30
uvn_findpages(0, 0, cb10ea4, cb10e00, 0, 1fff) at netbsd:uvn_findpages+0x68
genfs_getpages(cb11048, d, 1, 0, 0, 0) at netbsd:genfs_getpages+0x41c
nfs_getpages(cb11048, cb106d1, df0b790, cb10f70, 1, 0) at netbsd:nfs_getpages+0x
78
VOP_GETPAGES(df0b790, 0, cb11150, cb1116c, 0, 1) at netbsd:VOP_GETPAGES+0x40
ubc_fault(cb11378, c0f4000, cb108a1, 0, 1, cb11378) at netbsd:ubc_fault+0x144
uvm_fault_internal(1, 1, 0, 7, cb112f8, 1832aa0) at netbsd:uvm_fault_internal+0x
6e8
data_access_fault(cb11590, 30, 1009e80, c0f4000, c0f4000, 400) at netbsd:data_ac
cess_fault+0xd4
?(c0f4000, 238000, 1d7f, cb0e000, 1009f18, 800000001e564034) at 0x100870c
copyout_vmspace(ca3d350, c0f4000, 238000, 1d87, 4, bf38000) at netbsd:copyout_vm
space+0x68
uiomove(c0f4000, 1d87, cb11bd8, 0, 1, 0) at netbsd:uiomove+0xb0
(Continue reading)

Permalink | Reply | 0 comments |
headers
Jochen Kunz | 6 Oct 2006 22:58
Picon

Re: floppy support

On Fri, 6 Oct 2006 13:20:07 -0700
jnemeth <at> victoria.tc.ca (John Nemeth) wrote:

>      Use rfd0a.  Floppy drives are different from hard drives in that
> the final letter is used to indicated the type of floppy and not a
> partition.  By using rfd0c you told it that you have a 720KB floppy.
Ahh, yes, of course. Forgot that detail. (I haven't used floppies for
some years now...)

[man fdc]

No luck:
# dd if=/dev/rfd0c of=/tmp/rfd0a 
fd0: timeout: state 10
fd0: timeout: state 13
fd0: timeout: state 13
fd0: timeout: state 13
fd0: timeout: state 13
fd0: timeout: state 13
fd0c: hard error reading fsbn 0
fd0: controller status: state 13
fd0: timeout: state 10
fd0: timeout: state 13
fd0: timeout: state 13
fd0: timeout: state 13
fd0: timeout: state 13
fd0: timeout: state 13
fd0c: hard error reading fsbn 0
fd0: controller status: state 13
dd: /dev/rfd0c: Input/output error
(Continue reading)

Permalink | Reply | 0 comments |

Gmane