Package: gnumach
Version: 2:1.3.99.dfsg.cvs20070211-1
Severity:
When the files are downloaded with apt-get install, the system reboots
occasionally.
When the package of gnumach is replaced with version: 2:1.3.99.dfsg.1-1+b1,
the problem seem not to be occur.
Because I am confirming it only on the
Hi,
YAMAMOTO, Hiroyuki, le Sat 03 Mar 2007 11:47:00 +0900, a écrit :
> When the files are downloaded with apt-get install, the system reboots
> occasionally.
> When the package of gnumach is replaced with version: 2:1.3.99.dfsg.1-1+b1,
> the problem seem not to be occur.
BTW, what is +b1 at the e
Hi,
Samuel Thibault wrote:
BTW, what is +b1 at the end of your version? debian doesn't provide such
version.
Yes, it is the one officially provided as debian-K14-hurd-i386-CD1.iso.
It is possible to download it on the following WEB site:
http://snapshot.debian.net/archive/2007/02/12/debian/
YAMAMOTO, Hiroyuki, le Sat 03 Mar 2007 21:06:05 +0900, a écrit :
> Samuel Thibault wrote:
> >BTW, what is +b1 at the end of your version? debian doesn't provide such
> >version.
> >
>
> Yes, it is the one officially provided as debian-K14-hurd-i386-CD1.iso.
> It is possible to download it on the
On Sat, 2007-03-03 at 21:06 +0900, YAMAMOTO, Hiroyuki wrote:
> Hi,
>
> Samuel Thibault wrote:
> > BTW, what is +b1 at the end of your version? debian doesn't provide such
> > version.
> >
>
> Yes, it is the one officially provided as debian-K14-hurd-i386-CD1.iso.
> It is possible to download
Hi,
Samuel Thibault wrote:
For this bug and the mouse one, could you try compiling from
upstream repository to see whether it fails too?
cvs -d :pserver:[EMAIL PROTECTED]:/cvsroot/hurd co -r gnumach-1-branch gnumach
The cvs files were downloaded by your instruction and I ran the aclocal,
Hi,
YAMAMOTO, Hiroyuki, le Sun 04 Mar 2007 01:41:40 +0900, a écrit :
> The cvs files were downloaded by your instruction and I ran the aclocal,
> the autoconf and the automake.
> But the automake didn't finish making Makefile.in, because there is no
> build-aux/ directory.
>
> How can I downloa
On Sat, Mar 03, 2007 at 11:47:00AM +0900, YAMAMOTO, Hiroyuki wrote:
> Package: gnumach
> Version: 2:1.3.99.dfsg.cvs20070211-1
> Severity:
>
> When the files are downloaded with apt-get install, the system reboots
> occasionally.
> When the package of gnumach is replaced with version: 2:1.3.99.dfsg
Samuel Thibault wrote:
For this bug and the mouse one, could you try compiling from
upstream repository to see whether it fails too?
cvs -d :pserver:[EMAIL PROTECTED]:/cvsroot/hurd co -r
gnumach-1-branch gnumach
In case it fails too, could you try checking out older CVS revisions, so
as to fi
YAMAMOTO, Hiroyuki, le Sun 04 Mar 2007 09:13:48 +0900, a écrit :
> I compiled the upsteam CVS source with gcc-4.0, installed it, and
> verified the phenomenon.
> As a result, I have understood happening similarly as for the source of
> the upstream.
Ok, so this is a different bug from the mouse
Michael Banck wrote:
Can you install the corresponding version of gnumach-dbg, and change
your Grub entry to boot that? If you catch the Mach bug the next time,
it should drop you to the kernel debugger; please tell us the address of
eip for further information.
I was not able to reproduce
YAMAMOTO, Hiroyuki, le Sun 04 Mar 2007 16:05:01 +0900, a écrit :
> I suggest that this phenomenon doesn't occur in the gnumach-dbg kernel
> in fact.
Eergl :(
Guillem, can you confirm that gnumach and gnumach-dbg are compiled
exactly the same way (-O2, notably) except --enable-kdb?
Samuel
Samuel Thibault wrote:
YAMAMOTO, Hiroyuki, le Sun 04 Mar 2007 16:05:01 +0900, a écrit :
I suggest that this phenomenon doesn't occur in the gnumach-dbg kernel
in fact.
Eergl :(
Guillem, can you confirm that gnumach and gnumach-dbg are compiled
exactly the same way (-O2, notably) excep
Hi,
The system rebooted still though I tested with not `apt-get install' but
`wget'.
It seems to be sure to be a bug related to communication I/O.
Hiroyuki
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Hi,
YAMAMOTO, Hiroyuki, le Mon 05 Mar 2007 17:47:17 +0900, a écrit :
> The system rebooted still though I tested with not `apt-get install' but
> `wget'.
> It seems to be sure to be a bug related to communication I/O.
How fast is your network connection? There's a known bug here:
http://bugs.d
Hi,
Samuel Thibault wrote:
YAMAMOTO, Hiroyuki, le Mon 05 Mar 2007 17:47:17 +0900, a écrit :
The system rebooted still though I tested with not `apt-get install' but
`wget'.
It seems to be sure to be a bug related to communication I/O.
How fast is your network connection? There's a kn
YAMAMOTO, Hiroyuki, le Mon 05 Mar 2007 18:33:50 +0900, a écrit :
> Samuel Thibault wrote:
> >YAMAMOTO, Hiroyuki, le Mon 05 Mar 2007 17:47:17 +0900, a écrit :
> >
> >>The system rebooted still though I tested with not `apt-get install' but
> >>`wget'.
> >>It seems to be sure to be a bug related t
YAMAMOTO, Hiroyuki, le Mon 05 Mar 2007 19:00:36 +0900, a écrit :
> Samuel Thibault wrote:
> >YAMAMOTO, Hiroyuki, le Mon 05 Mar 2007 18:33:50 +0900, a écrit :
> >
> >>Samuel Thibault wrote:
> >>
> >>>YAMAMOTO, Hiroyuki, le Mon 05 Mar 2007 17:47:17 +0900, a écrit :
> >>>
> >>>
> The
Samuel Thibault wrote:
YAMAMOTO, Hiroyuki, le Mon 05 Mar 2007 18:33:50 +0900, a écrit :
Samuel Thibault wrote:
YAMAMOTO, Hiroyuki, le Mon 05 Mar 2007 17:47:17 +0900, a écrit :
The system rebooted still though I tested with not `apt-get install' but
`wget'.
It seems to be sure
Hi,
Samuel Thibault wrote:
Ah, yes, the report is older than that, but let's see if fixing #396741
fixes yours.
I compiled today's CVS version with --enable-kdb, and succeeded in the
reproduction of the phenomenon.
db>trace
0x10d577(2edabc,10379000,2eecc4,173b99,ff007b8)
0x133dae(ff007b8
Hi,
YAMAMOTO, Hiroyuki, le Mon 05 Mar 2007 23:48:37 +0900, a écrit :
> Samuel Thibault wrote:
> >Ah, yes, the report is older than that, but let's see if fixing #396741
> >fixes yours.
> >
>
> I compiled today's CVS version with --enable-kdb, and succeeded in the
> reproduction of the phenomen
On Mon, Mar 05, 2007 at 11:48:37PM +0900, YAMAMOTO, Hiroyuki wrote:
> Samuel Thibault wrote:
> >Ah, yes, the report is older than that, but let's see if fixing #396741
> >fixes yours.
>
> I compiled today's CVS version with --enable-kdb, and succeeded in the
> reproduction of the phenomenon.
Exc
Hi,
Samuel Thibault wrote:
YAMAMOTO, Hiroyuki, le Mon 05 Mar 2007 23:48:37 +0900, a écrit :
Samuel Thibault wrote:
Ah, yes, the report is older than that, but let's see if fixing #396741
fixes yours.
I compiled today's CVS version with --enable-kdb, and succeeded in the
repro
YAMAMOTO, Hiroyuki, le Tue 06 Mar 2007 00:57:13 +0900, a écrit :
> >>0x133dae(ff007b8,10379000,588,1,1080)
> >>0x173b99(572c8,4,2eecf4,202,2b7480)
> >>0x1b4b18(562c0,0,9,2eed3c,2eed8c)
> >>0x1b557b(9,2b7480,2eed64,11f6f4,fef5ca0)
> >>0x15a3b0(9)
> >>0x1377f5(2edabc,1036e000,11a86ee0,1738bb,ff007b8)
Hi,
I don't have the time to test myself, maybe it's just completely bogus:
could you try, in device/net_io.c, in the net_deliver() function, to
move the call to net_filter() _above_
simple_unlock(&net_queue_lock);
(void) spl0();
And in linux/dev/glue/net.c, in device_write(), m
YAMAMOTO, Hiroyuki, le Tue 06 Mar 2007 01:44:15 +0900, a écrit :
> $ addr2line -f -e gnumach 0x1377f5
> interrupt
> i386/i386at/interrupt.S:44
> $ addr2line -f -e gnumach 0x133dae
> net_packet
> device/net_io.c:674
> $ addr2line -f -e gnumach 0x1738bb
> device_write
> linux/dev/glue/net.c:522
That
Samuel Thibault wrote:
YAMAMOTO, Hiroyuki, le Tue 06 Mar 2007 00:57:13 +0900, a écrit :
0x133dae(ff007b8,10379000,588,1,1080)
0x173b99(572c8,4,2eecf4,202,2b7480)
0x1b4b18(562c0,0,9,2eed3c,2eed8c)
0x1b557b(9,2b7480,2eed64,11f6f4,fef5ca0)
0x15a3b0(9)
0x1377f5(2edabc,1036e000,11a86ee0,1738bb,ff0
Hi,
Samuel Thibault wrote:
I don't have the time to test myself, maybe it's just completely bogus:
could you try, in device/net_io.c, in the net_deliver() function, to
move the call to net_filter() _above_
simple_unlock(&net_queue_lock);
(void) spl0();
And in linux/dev/glue/ne
YAMAMOTO, Hiroyuki, le Tue 06 Mar 2007 04:03:45 +0900, a écrit :
> Hi,
>
> Samuel Thibault wrote:
> >I don't have the time to test myself, maybe it's just completely bogus:
> >could you try, in device/net_io.c, in the net_deliver() function, to
> >move the call to net_filter() _above_
> >
> >
Hi,
I commited the fix in CVS.
Samuel
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Hi,
On Sat, Mar 03, 2007 at 01:16:43PM +0100, Samuel Thibault wrote:
> YAMAMOTO, Hiroyuki, le Sat 03 Mar 2007 21:06:05 +0900, a écrit :
> > Samuel Thibault wrote:
> > >BTW, what is +b1 at the end of your version? debian doesn't provide
> > >such version.
> > >
> >
> > Yes, it is the one officia
[EMAIL PROTECTED] wrote:
Hi,
On Sat, Mar 03, 2007 at 01:16:43PM +0100, Samuel Thibault wrote:
YAMAMOTO, Hiroyuki, le Sat 03 Mar 2007 21:06:05 +0900, a écrit :
Samuel Thibault wrote:
BTW, what is +b1 at the end of your version? debian doesn't provide
such version.
Ye
32 matches
Mail list logo
