Ok, I found the issue as being in gnumach: it was reading from a freed
ipc_port structure, and it happens that the payload was overwritten by
the free slot linked-list pointer. This happens when the destination
port gets down to zero reference right when delivering the message.
Thomas Schwinge, le
Samuel Thibault, le Wed 22 Apr 2015 22:59:30 +0200, a écrit :
> I'm rebuilding an ext2fs.static with -fno-omit-frame-pointer to get more
> backtrace.
db> show task $task1
TASKTHREADS
1 ext2fs (f4d68e40): 15 threads:
0 (f4d693c0) ..SO.F(walking_zombie)
1 (f
Justus Winter, le Wed 22 Apr 2015 01:58:33 +0200, a écrit :
> Quoting Samuel Thibault (2015-04-20 04:17:53)
> > Boot it, use all the default parameters, and just after partition
> > reloading from the kernel (which prints the new partition layout), the
> > system is completely frozen.
>
> Please c
Justus Winter, le Wed 22 Apr 2015 13:53:12 +0200, a écrit :
> Anyhow, reloading the partition tables seems to work as expected. I
> don't think it's related.
Ok, I just mentioned it because it was giving a very precise timing
along the whole d-i process.
Samuel
Justus Winter, le Wed 22 Apr 2015 13:53:12 +0200, a écrit :
> I created a program to reload the partition table. That wasn't as
> smooth as it should be, and I didn't manage to get it done in a posixy
> way.
That doesn't exist in POSIX indeed.
> I wonder which component of d-i does that, and how
Quoting Samuel Thibault (2015-04-20 04:17:53)
> Boot it, use all the default parameters, and just after partition
> reloading from the kernel (which prints the new partition layout), the
> system is completely frozen.
I created a program to reload the partition table. That wasn't as
smooth as it
Hi Samuel :)
Quoting Samuel Thibault (2015-04-20 04:17:53)
> Boot it, use all the default parameters, and just after partition
> reloading from the kernel (which prints the new partition layout), the
> system is completely frozen.
Please check whether or not the root filesystem in fact dies.
/hur
Thomas Schwinge, le Mon 20 Apr 2015 13:25:06 +0200, a écrit :
> As the reversion of this patch is the only change from Debian hurd 0.6-1
> to 0.6-2 packages, I assume it's a related issue:
That's possible. Another candidate could be
fb8c531ddeef55c8581d3882bd1b1f501d980a30
libports: avoid acquirin
Hi!
As the reversion of this patch is the only change from Debian hurd 0.6-1
to 0.6-2 packages, I assume it's a related issue:
On Mon, 20 Apr 2015 04:17:53 +0200, Samuel Thibault
wrote:
> Samuel Thibault, le Mon 20 Apr 2015 04:10:36 +0200, a écrit :
> > Mmm, unfortunately it seems that this bri
Samuel Thibault, le Mon 20 Apr 2015 04:17:53 +0200, a écrit :
> Boot it, use all the default parameters, and just after partition
> reloading from the kernel (which prints the new partition layout), the
> system is completely frozen.
(using a preseed file can be useful to get that done automatical
Samuel Thibault, le Mon 20 Apr 2015 04:10:36 +0200, a écrit :
> Mmm, unfortunately it seems that this brings a hang inside the debian
> installer, so I'll have to revert it for our Jessie release.
Yes, this is confirmed.
An easy way to reproduce and experiment is to take the broken image I
have g
Mmm, unfortunately it seems that this brings a hang inside the debian
installer, so I'll have to revert it for our Jessie release.
Samuel
Justus Winter, le Sat 04 Apr 2015 18:38:53 +0200, a écrit :
> * NEWS: Mention protected payloads.
> * libports/Makefile (SRCS): Add `port-deref-deferred.c'.
> * libports/create-internal.c (_ports_create_port_internal): Set the
> protected payload to the objects address.
> * libports/import-port.c (
* NEWS: Mention protected payloads.
* libports/Makefile (SRCS): Add `port-deref-deferred.c'.
* libports/create-internal.c (_ports_create_port_internal): Set the
protected payload to the objects address.
* libports/import-port.c (ports_import_port): Likewise.
* libports/reallocate-from-external.c (p
14 matches
Mail list logo
