On Tue, Sep 11, 2007 at 12:12:27AM -0400, Michael Casadevall wrote:
Ever since applying this patch, mach panics for me on startup. It gets
as far as starting sshd, then it panics with
panic: zalloc: zone i386 pcb state exhausted
kernel breakpoint trap, eip 0x1179d7
stopped at 0x1179d6: int
Oops, yes, sorry, please use the attached patch instead.
Samuel
Index: kern/zalloc.c
===
RCS file: /cvsroot/hurd/gnumach/kern/zalloc.c,v
retrieving revision 1.3.2.8
diff -u -p -r1.3.2.8 zalloc.c
--- kern/zalloc.c 30 Apr 2007
On Tue, 11 Sep 2007 11:05:51 +0200, Samuel Thibault
[EMAIL PROTECTED] wrote:
Oops, yes, sorry, please use the attached patch instead.
i got gnumach package source, patched manually and rebuilt it. it worked on
qemu with and without kqemu.
pietro.
Ever since applying this patch, mach panics for me on startup. It gets
as far as starting sshd, then it panics with
panic: zalloc: zone i386 pcb state exhausted
kernel breakpoint trap, eip 0x1179d7
stopped at 0x1179d6: int $3
Any ideas why I'm getting a panic while others aren't? (I haven't had
Hi,
Samuel Thibault, le Wed 29 Aug 2007 16:50:20 +0200, a écrit :
Michael Banck, le Wed 29 Aug 2007 16:13:20 +0200, a écrit :
eax 0x107502f8
This is the culprit: fxsave needs the buffer to be 16 bytes-aligned. It
looks like there is a bug somewhere in the alignment support of
zalloc().
Hi,
Thomas Schwinge, le Thu 30 Aug 2007 01:11:23 +0200, a écrit :
Hello!
On Wed, Aug 29, 2007 at 11:12:14PM +0200, Samuel Thibault wrote:
Thomas Schwinge, le Wed 29 Aug 2007 22:44:06 +0200, a écrit :
Why isn't that patch in upsteam cvs? Does it need more testing (as this
bug report
package: gnumach
version: 2:1.3.99.dfsg.cvs20070809-1
I caught this GPT twice now, so filing here:
The error message was (something like): General Protection Trap (13)
code=0
This is the trace:
0x100d60
fp_save
/build/buildd/gnumach-1.3.99.dfsg.cvs20070809/build-dbg/../i386/i386/fpu.c:675
Hi,
Michael Banck, le Wed 29 Aug 2007 16:13:20 +0200, a écrit :
eax 0x107502f8
This is the culprit: fxsave needs the buffer to be 16 bytes-aligned. It
looks like there is a bug somewhere in the alignment support of
zalloc().
Samuel
[Cced to [EMAIL PROTECTED] as it's a GNU Mach bug.]
Hello!
On Wed, Aug 29, 2007 at 04:50:20PM +0200, Samuel Thibault wrote:
Michael Banck, le Wed 29 Aug 2007 16:13:20 +0200, a écrit :
eax 0x107502f8
This is the culprit: fxsave needs the buffer to be 16 bytes-aligned. It
fnsave, you mean,
Thomas Schwinge, le Wed 29 Aug 2007 21:39:36 +0200, a écrit :
This is the culprit: fxsave needs the buffer to be 16 bytes-aligned. It
fnsave, you mean, I guess.
No, really fxsave (the MMX version). fnsave would be fine.
looks like there is a bug somewhere in the alignment support of
Thomas Schwinge, le Wed 29 Aug 2007 22:44:06 +0200, a écrit :
Why isn't that patch in upsteam cvs? Does it need more testing (as this
bug report might indicate)?
Well, the buildd tested it pretty well, and yes, this bug needs to be
fixed. Then it should be ok to commit. The MMX support is not
11 matches
Mail list logo
