headers
Chris Double | 11 Jun 2012 02:37
Picon
Gravatar

Re: Inferno on Android

This post documents the steps I took to build the Android port of
Inferno from source to run on an ICS based Nexus S:

<http://www.bluishcoder.co.nz/2012/06/11/building-inferno-os-for-android-phones.html>

Chris.
--

-- 
http://www.bluishcoder.co.nz
Permalink | Reply | 0 comments |
headers
Alex Efros | 15 Jun 2012 20:03
Favicon
Gravatar

inferno build on FreeBSD-9.0 64-bit

Hi!

Is it possible to run Inferno on 64-bit FreeBSD using /lib32 instead of
32-bit chroot to complete 32-bit system?

[powerman <at> freebsd90-64 ~/inferno]$ file FreeBSD/386/bin/emu-g 
FreeBSD/386/bin/emu-g: ELF 32-bit LSB executable, Intel 80386, version 1 (FreeBSD), dynamically
linked (uses shared libs), for FreeBSD 9.0 (900044), not stripped

[powerman <at> freebsd90-64 ~/inferno]$ emu-g
SYS: process main faults: Segmentation violation
Killed: 9

[powerman <at> freebsd90-64 ~/inferno]$ ldd FreeBSD/386/bin/emu-g 
FreeBSD/386/bin/emu-g:
        libm.so.5 => /usr/lib32/libm.so.5 (0x2810c000)
        libc.so.7 => /usr/lib32/libc.so.7 (0x28126000)

Other binaries in FreeBSD/386/bin/ works fine, but they all linked only
with libc.so.7, while emu also linked with libm.so.5 - maybe it crash
because of this?

Only change I made is in mkfiles/mkfile-FreeBSD-386:
-AS=            cc -c
+AS=            cc -c -m32
-CC=            cc -c
+CC=            cc -c -m32
-LD=            cc
+LD=            cc -m32
(Continue reading)

Permalink | Reply | 0 comments |
headers
Alex Efros | 19 Jun 2012 13:02
Favicon
Gravatar

retrymount code review request

Hi!

I'd like to ask for code review for my retrymount tool:
http://code.google.com/p/inferno-contrib-retrymount/source/browse/appl/cmd/retrymount.b

It act as wrapper/proxy for mount to be able to detect broken connection
and restart mount.

I don't wanna overcomplicate things trying to hide the fact of broken
connection from applications which use files mounted using this connection -
it's fine for me if application will receive file i/o errors in this case.
But if application able to recover after such error, I'd like to let it
try to re-open these files again and again until success and continue i/o
with these files. I.e. application shouldn't bother about remounting
resources - actually it shouldn't even know these resources was mounted
from somewhere. It just work with files while it's possible and if error
happens it just try to recover after errors, reopen files and continue
it's work, if possible.

Problem is, I've seen few times on production servers the connection
wasn't remounted by this tool after error. This happens just a few times
in last year, and I was unable to catch this bug. Looking at code I can't
even imagine how this may happens. But it happened, so probably I miss
something. Or this happens because of bug in Inferno, not in my code - but
I hope this isn't the case.

So, if anyone will have spare time, please take a look at the code, maybe
you'll got idea how this bug may happens. Maybe I just misunderstand
something important and trying to solve this issue in wrong way.
(Continue reading)

Permalink | Reply | 0 comments |
headers
Pete Elmore | 20 Jun 2012 07:27
Favicon
Gravatar

Re: Authentication basics

> Whenever you get the info into your blog, please post a url in this
> thread.
It took me forever, but here's the link:
http://debu.gs/entries/inferno-part-2-let-s-make-a-cluster

Any corrections anyone has are welcome.  Writing the previous entry and
not knowing about unicode(1), I had to do some acrobatics to get a \r
into a shell script.  :)  If I've got something wrong in this one, feel
free to let me know; I'm happy to update the entry.

Thanks!

Pete
Permalink | Reply | 1 comment |

Gmane