Subject: Re: openafs on netbsd/sparc64?
To: None <port-sparc64@netbsd.org>
From: Sean Caron <caron.sean@gmail.com>
List: port-sparc64
Date: 06/09/2006 14:58:21
------=_Part_11595_29320035.1149879501121
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
Ehh, sorry -- I meant NetBSD/sparc64 in a 32-bit-on-UltraSPARC sense of the
word. Sorry for the
false alarm :)
I'm using NetBSD-3.0 on all systems -- servers are Sun Ultra 1 and Ultra 10
systems running NetBSD
3.0 and CODA 6.0.14, clients are Sun Ultra 10 systems running NetBSD/3.0 and
CODA 6.0.14, Apple
PMG3 B&W systems running NetBSD/3.0 and CODA 6.0.14*, and Sun SPARCstation
10 systems
running NetBSD/3.0 and CODA 6.0.14. Ever since I patched up that one bug
with the help of yourself
and others on the Coda list, and figured out a silly mis-made volume problem
that was tripping up venus,
I have been running venus and codasrv on all these architectures and systems
with absolutely no trouble
at all. I feel like I've tried a pretty good mix of operations so far --
small reads/writes, larger bulk reads/writes,
volume creation, volume deletion, manipulating ACLs and mountpoints, etc. It
works great.. I'm digging CODA.
Regards, Sean
scaron@umich.edu
scaron@diablonet.net
* CODA support doesnt come mixed in with the NetBSD/macppc kernel for some
reason; I just added the requisite lines to the kernel config file and
modified the major/minor number database -- I preempted some stupid cyclades
driver that uses major 47 on macppc and ripped it out and replaced that with
vcoda just like on my SPARC systems. Deleted all the cyclades cruft in /dev,
mknod /dev/cfs0 c 47 0, fire up venus and it seems happy.
On 6/8/06, Greg Troxel <gdt@ir.bbn.com> wrote:
>
>
> Are you really running both venus and codasrv on sparc64 as native
> 64-bit binaries? That's very cool if so - it' been a long time
> coming.
>
> Are you using 3.0ish or current? There are some locking bugs that I fixed
> in
> current and I think haven't pulled up to 3.0 yet.
>
> (I'm sort of the netbsd coda kernel bits maintainer.)
>
> --
> Greg Troxel <gdt@ir.bbn.com>
>
>
>
------=_Part_11595_29320035.1149879501121
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
Ehh, sorry -- I meant NetBSD/sparc64 in a 32-bit-on-UltraSPARC sense of the word. Sorry for the<br>false alarm :)<br><br>I'm using NetBSD-3.0 on all systems -- servers are Sun Ultra 1 and Ultra 10 systems running NetBSD<br>
3.0 and CODA 6.0.14, clients are Sun Ultra 10 systems running NetBSD/3.0 and CODA 6.0.14, Apple<br>PMG3 B&W systems running NetBSD/3.0 and CODA 6.0.14*, and Sun SPARCstation 10 systems<br>running NetBSD/3.0 and CODA 6.0.14
. Ever since I patched up that one bug with the help of yourself<br>and others on the Coda list, and figured out a silly mis-made volume problem that was tripping up venus,<br>I have been running venus and codasrv on all these architectures and systems with absolutely no trouble
<br>at all. I feel like I've tried a pretty good mix of operations so far -- small reads/writes, larger bulk reads/writes,<br>volume creation, volume deletion, manipulating ACLs and mountpoints, etc. It works great.. I'm digging CODA.
<br><br>Regards, Sean<br><a href="mailto:scaron@umich.edu">scaron@umich.edu</a><br><a href="mailto:scaron@diablonet.net">scaron@diablonet.net</a><br><br>* CODA support doesnt come mixed in with the NetBSD/macppc kernel for some reason; I just added the requisite lines to the kernel config file and modified the major/minor number database -- I preempted some stupid cyclades driver that uses major 47 on macppc and ripped it out and replaced that with vcoda just like on my SPARC systems. Deleted all the cyclades cruft in /dev, mknod /dev/cfs0 c 47 0, fire up venus and it seems happy.
<br><br><div><span class="gmail_quote">On 6/8/06, <b class="gmail_sendername">Greg Troxel</b> <<a href="mailto:gdt@ir.bbn.com">gdt@ir.bbn.com</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<br>Are you really running both venus and codasrv on sparc64 as native<br>64-bit binaries? That's very cool if so - it' been a long time<br>coming.<br><br>Are you using 3.0ish or current? There are some locking bugs that I fixed in
<br>current and I think haven't pulled up to 3.0 yet.<br><br>(I'm sort of the netbsd coda kernel bits maintainer.)<br><br>--<br> Greg Troxel <<a href="mailto:gdt@ir.bbn.com">gdt@ir.bbn.com</a>><br><br><br></blockquote>
</div><br>
------=_Part_11595_29320035.1149879501121--