* [Qemu-devel] strange crash on FreeBSD-current/amd64 (pointer truncation?)
@ 2007-01-24 20:00 Juergen Lock
2007-02-01 20:00 ` [Qemu-devel] " Juergen Lock
0 siblings, 1 reply; 11+ messages in thread
From: Juergen Lock @ 2007-01-24 20:00 UTC (permalink / raw)
To: qemu-devel
Hi!
I got a report of qemu segfaulting here on FreeBSD-current/amd64:
> #0 main_loop () at /usr/ports-cvs/emulators/qemu/work/qemu-snapshot-2007-01-11_05/vl.c:6125
> 6125 env = env->next_cpu;
> [New Thread 0x801e10190 (LWP 100214)]
> (gdb) print env
> $1 = (CPUX86State *) 0xac10000
> (gdb) print first_cpu
> $2 = (CPUX86State *) 0x80ac10000
This works on earlier FreeBSD branches, and the build logs,
http://pointyhat.freebsd.org/errorlogs/amd64-7-latest-logs/qemu-0.8.2s.20070111.log
http://pointyhat.freebsd.org/errorlogs/amd64-6-latest-logs/qemu-0.8.2s.20070111.log
(7 is the -current branch and 6 is the branch that the latest release,
6.2, was branched off) show no differences in compiler warnings, so
I'm a bit at a loss where to look for the problem. So the qeustion is:
Has anyone seen something like this or has a helpful idea?
Thanx,
Juergen
^ permalink raw reply [flat|nested] 11+ messages in thread
* [Qemu-devel] Re: strange crash on FreeBSD-current/amd64 (pointer truncation?)
2007-01-24 20:00 [Qemu-devel] strange crash on FreeBSD-current/amd64 (pointer truncation?) Juergen Lock
@ 2007-02-01 20:00 ` Juergen Lock
2007-02-02 4:02 ` Thiemo Seufer
0 siblings, 1 reply; 11+ messages in thread
From: Juergen Lock @ 2007-02-01 20:00 UTC (permalink / raw)
To: qemu-devel
On Wed, Jan 24, 2007 at 09:00:19PM +0100, Juergen Lock wrote:
> Hi!
>
> I got a report of qemu segfaulting here on FreeBSD-current/amd64:
>
> > #0 main_loop () at /usr/ports-cvs/emulators/qemu/work/qemu-snapshot-2007-01-11_05/vl.c:6125
> > 6125 env = env->next_cpu;
> > [New Thread 0x801e10190 (LWP 100214)]
> > (gdb) print env
> > $1 = (CPUX86State *) 0xac10000
> > (gdb) print first_cpu
> > $2 = (CPUX86State *) 0x80ac10000
Ok Jung-uk Kim found the following fix: (Thanx!)
--- qemu/cpu-exec.c.orig Wed Jan 31 16:58:03 2007
+++ qemu/cpu-exec.c Wed Jan 31 17:08:11 2007
@@ -226,9 +226,9 @@
int cpu_exec(CPUState *env1)
{
- int saved_T0, saved_T1;
+ long saved_T0, saved_T1;
#if defined(reg_T2)
- int saved_T2;
+ long saved_T2;
#endif
CPUState *saved_env;
#if defined(TARGET_I386)
It's funny that the upper half of rbx (which holds env above)
apparently only gets spilled on FreeBSD-current, even tho on amd64
T0 (== AREG1 == rbx) certainly doesn't fit into an int...
Juergen
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [Qemu-devel] Re: strange crash on FreeBSD-current/amd64 (pointer truncation?)
2007-02-01 20:00 ` [Qemu-devel] " Juergen Lock
@ 2007-02-02 4:02 ` Thiemo Seufer
2007-02-02 21:45 ` Juergen Lock
0 siblings, 1 reply; 11+ messages in thread
From: Thiemo Seufer @ 2007-02-02 4:02 UTC (permalink / raw)
To: Juergen Lock; +Cc: qemu-devel
Juergen Lock wrote:
> On Wed, Jan 24, 2007 at 09:00:19PM +0100, Juergen Lock wrote:
> > Hi!
> >
> > I got a report of qemu segfaulting here on FreeBSD-current/amd64:
> >
> > > #0 main_loop () at /usr/ports-cvs/emulators/qemu/work/qemu-snapshot-2007-01-11_05/vl.c:6125
> > > 6125 env = env->next_cpu;
> > > [New Thread 0x801e10190 (LWP 100214)]
> > > (gdb) print env
> > > $1 = (CPUX86State *) 0xac10000
> > > (gdb) print first_cpu
> > > $2 = (CPUX86State *) 0x80ac10000
>
> Ok Jung-uk Kim found the following fix: (Thanx!)
>
> --- qemu/cpu-exec.c.orig Wed Jan 31 16:58:03 2007
> +++ qemu/cpu-exec.c Wed Jan 31 17:08:11 2007
> @@ -226,9 +226,9 @@
>
> int cpu_exec(CPUState *env1)
> {
> - int saved_T0, saved_T1;
> + long saved_T0, saved_T1;
> #if defined(reg_T2)
> - int saved_T2;
> + long saved_T2;
I used target_ulong instead.
Thiemo
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [Qemu-devel] Re: strange crash on FreeBSD-current/amd64 (pointer truncation?)
2007-02-02 4:02 ` Thiemo Seufer
@ 2007-02-02 21:45 ` Juergen Lock
2007-02-02 22:24 ` Thiemo Seufer
2007-02-02 22:27 ` andrzej zaborowski
0 siblings, 2 replies; 11+ messages in thread
From: Juergen Lock @ 2007-02-02 21:45 UTC (permalink / raw)
To: Thiemo Seufer; +Cc: qemu-devel
On Fri, Feb 02, 2007 at 04:02:00AM +0000, Thiemo Seufer wrote:
> Juergen Lock wrote:
> > On Wed, Jan 24, 2007 at 09:00:19PM +0100, Juergen Lock wrote:
> > > Hi!
> > >
> > > I got a report of qemu segfaulting here on FreeBSD-current/amd64:
> > >
> > > > #0 main_loop () at /usr/ports-cvs/emulators/qemu/work/qemu-snapshot-2007-01-11_05/vl.c:6125
> > > > 6125 env = env->next_cpu;
> > > > [New Thread 0x801e10190 (LWP 100214)]
> > > > (gdb) print env
> > > > $1 = (CPUX86State *) 0xac10000
> > > > (gdb) print first_cpu
> > > > $2 = (CPUX86State *) 0x80ac10000
> >
> > Ok Jung-uk Kim found the following fix: (Thanx!)
> >
> > --- qemu/cpu-exec.c.orig Wed Jan 31 16:58:03 2007
> > +++ qemu/cpu-exec.c Wed Jan 31 17:08:11 2007
> > @@ -226,9 +226,9 @@
> >
> > int cpu_exec(CPUState *env1)
> > {
> > - int saved_T0, saved_T1;
> > + long saved_T0, saved_T1;
> > #if defined(reg_T2)
> > - int saved_T2;
> > + long saved_T2;
>
> I used target_ulong instead.
Isn't that 32 bit for 32 bit targets? Then it wouldn't fix the bug,
the problem happened for 32 bit guests (i386-softmmu) on 64 bit hosts
(FreeBSD-current/amd64), there the upper half of rbx which held a
pointer in the calling function was lost...
Juergen
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [Qemu-devel] Re: strange crash on FreeBSD-current/amd64 (pointer truncation?)
2007-02-02 21:45 ` Juergen Lock
@ 2007-02-02 22:24 ` Thiemo Seufer
2007-02-03 1:43 ` Juergen Lock
2007-02-02 22:27 ` andrzej zaborowski
1 sibling, 1 reply; 11+ messages in thread
From: Thiemo Seufer @ 2007-02-02 22:24 UTC (permalink / raw)
To: Juergen Lock; +Cc: qemu-devel
Juergen Lock wrote:
> > > Ok Jung-uk Kim found the following fix: (Thanx!)
> > >
> > > --- qemu/cpu-exec.c.orig Wed Jan 31 16:58:03 2007
> > > +++ qemu/cpu-exec.c Wed Jan 31 17:08:11 2007
> > > @@ -226,9 +226,9 @@
> > >
> > > int cpu_exec(CPUState *env1)
> > > {
> > > - int saved_T0, saved_T1;
> > > + long saved_T0, saved_T1;
> > > #if defined(reg_T2)
> > > - int saved_T2;
> > > + long saved_T2;
> >
> > I used target_ulong instead.
>
> Isn't that 32 bit for 32 bit targets? Then it wouldn't fix the bug,
> the problem happened for 32 bit guests (i386-softmmu) on 64 bit hosts
> (FreeBSD-current/amd64), there the upper half of rbx which held a
> pointer in the calling function was lost...
Apparently I miss something here. If T0 is a (32bit) target_ulong,
why does saving it to a (32bit) target_ulong and restoring it from
there cause any trouble?
Thiemo
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [Qemu-devel] Re: strange crash on FreeBSD-current/amd64 (pointer truncation?)
2007-02-02 21:45 ` Juergen Lock
2007-02-02 22:24 ` Thiemo Seufer
@ 2007-02-02 22:27 ` andrzej zaborowski
2007-02-02 22:40 ` Paul Brook
1 sibling, 1 reply; 11+ messages in thread
From: andrzej zaborowski @ 2007-02-02 22:27 UTC (permalink / raw)
To: qemu-devel
On 02/02/07, Juergen Lock <nox@jelal.kn-bremen.de> wrote:
> On Fri, Feb 02, 2007 at 04:02:00AM +0000, Thiemo Seufer wrote:
> > Juergen Lock wrote:
> > > On Wed, Jan 24, 2007 at 09:00:19PM +0100, Juergen Lock wrote:
> > > > Hi!
> > > >
> > > > I got a report of qemu segfaulting here on FreeBSD-current/amd64:
> > > >
> > > > > #0 main_loop () at /usr/ports-cvs/emulators/qemu/work/qemu-snapshot-2007-01-11_05/vl.c:6125
> > > > > 6125 env = env->next_cpu;
> > > > > [New Thread 0x801e10190 (LWP 100214)]
> > > > > (gdb) print env
> > > > > $1 = (CPUX86State *) 0xac10000
> > > > > (gdb) print first_cpu
> > > > > $2 = (CPUX86State *) 0x80ac10000
> > >
> > > Ok Jung-uk Kim found the following fix: (Thanx!)
> > >
> > > --- qemu/cpu-exec.c.orig Wed Jan 31 16:58:03 2007
> > > +++ qemu/cpu-exec.c Wed Jan 31 17:08:11 2007
> > > @@ -226,9 +226,9 @@
> > >
> > > int cpu_exec(CPUState *env1)
> > > {
> > > - int saved_T0, saved_T1;
> > > + long saved_T0, saved_T1;
> > > #if defined(reg_T2)
> > > - int saved_T2;
> > > + long saved_T2;
> >
> > I used target_ulong instead.
>
> Isn't that 32 bit for 32 bit targets? Then it wouldn't fix the bug,
Well, it just happens that target_ulong is 64 bit even for 32 bit
target on a 64 bit host (as a workaround for something I think, was
wondering about it too. I asked on IRC once but no one knew the actual
reason). This however doesn't change the fact that target_ulong
shouldn't be used here.
Regards,
Andrew
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [Qemu-devel] Re: strange crash on FreeBSD-current/amd64 (pointer truncation?)
2007-02-02 22:27 ` andrzej zaborowski
@ 2007-02-02 22:40 ` Paul Brook
2007-02-02 22:57 ` andrzej zaborowski
0 siblings, 1 reply; 11+ messages in thread
From: Paul Brook @ 2007-02-02 22:40 UTC (permalink / raw)
To: qemu-devel, balrogg
> > Isn't that 32 bit for 32 bit targets? Then it wouldn't fix the bug,
>
> Well, it just happens that target_ulong is 64 bit even for 32 bit
> target on a 64 bit host (as a workaround for something I think, was
> wondering about it too. I asked on IRC once but no one knew the actual
> reason). This however doesn't change the fact that target_ulong
> shouldn't be used here.
Are you sure you're not confusing it with target_phys_address_t?
Paul
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [Qemu-devel] Re: strange crash on FreeBSD-current/amd64 (pointer truncation?)
2007-02-02 22:40 ` Paul Brook
@ 2007-02-02 22:57 ` andrzej zaborowski
0 siblings, 0 replies; 11+ messages in thread
From: andrzej zaborowski @ 2007-02-02 22:57 UTC (permalink / raw)
To: Paul Brook; +Cc: qemu-devel
On 02/02/07, Paul Brook <paul@codesourcery.com> wrote:
> > > Isn't that 32 bit for 32 bit targets? Then it wouldn't fix the bug,
> >
> > Well, it just happens that target_ulong is 64 bit even for 32 bit
> > target on a 64 bit host (as a workaround for something I think, was
> > wondering about it too. I asked on IRC once but no one knew the actual
> > reason). This however doesn't change the fact that target_ulong
> > shouldn't be used here.
>
> Are you sure you're not confusing it with target_phys_address_t?
Oops, right, TARGET_PHYS_ADDR_BITS and TARGET_LONG_BITS are different.
Sorry for the confusion.
Regards,
Andrew
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [Qemu-devel] Re: strange crash on FreeBSD-current/amd64 (pointer truncation?)
2007-02-02 22:24 ` Thiemo Seufer
@ 2007-02-03 1:43 ` Juergen Lock
2007-02-03 18:12 ` Gwenole Beauchesne
0 siblings, 1 reply; 11+ messages in thread
From: Juergen Lock @ 2007-02-03 1:43 UTC (permalink / raw)
To: Thiemo Seufer; +Cc: qemu-devel
On Fri, Feb 02, 2007 at 10:24:28PM +0000, Thiemo Seufer wrote:
> Juergen Lock wrote:
> > > > Ok Jung-uk Kim found the following fix: (Thanx!)
> > > >
> > > > --- qemu/cpu-exec.c.orig Wed Jan 31 16:58:03 2007
> > > > +++ qemu/cpu-exec.c Wed Jan 31 17:08:11 2007
> > > > @@ -226,9 +226,9 @@
> > > >
> > > > int cpu_exec(CPUState *env1)
> > > > {
> > > > - int saved_T0, saved_T1;
> > > > + long saved_T0, saved_T1;
> > > > #if defined(reg_T2)
> > > > - int saved_T2;
> > > > + long saved_T2;
> > >
> > > I used target_ulong instead.
> >
> > Isn't that 32 bit for 32 bit targets? Then it wouldn't fix the bug,
> > the problem happened for 32 bit guests (i386-softmmu) on 64 bit hosts
> > (FreeBSD-current/amd64), there the upper half of rbx which held a
> > pointer in the calling function was lost...
>
> Apparently I miss something here. If T0 is a (32bit) target_ulong,
> why does saving it to a (32bit) target_ulong and restoring it from
> there cause any trouble?
Hmm. All I can say is the upper half of rbx (which holds T0) gets
spilled on FreeBSD-current/amd64 hosts unless saving and restoring
the full 64 bit of it...
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [Qemu-devel] Re: strange crash on FreeBSD-current/amd64 (pointer truncation?)
2007-02-03 1:43 ` Juergen Lock
@ 2007-02-03 18:12 ` Gwenole Beauchesne
2007-02-04 13:38 ` Paul Brook
0 siblings, 1 reply; 11+ messages in thread
From: Gwenole Beauchesne @ 2007-02-03 18:12 UTC (permalink / raw)
To: qemu-devel
[-- Attachment #1: Type: text/plain, Size: 1050 bytes --]
Hi,
> Hmm. All I can say is the upper half of rbx (which holds T0) gets
> spilled on FreeBSD-current/amd64 hosts unless saving and restoring
> the full 64 bit of it...
That's also what I got with VirtualBox on x86_64. Here is an update to
the patch I posted yesterday and that applies to current QEMU CVS
instead.
Please note this one applies unconditionnally of the host and target
combinations. Besides, I only tested the following combinations:
x86_64/x86_64-softmmu, x86_64/i386-softmmu, i586/x86_64-softmmu,
i586/i386-softmmu. And I verified that the generated code for
cpu_exec() with both gcc 3.4 and gcc 4.1 actually does the host
registers save/restore correctly.
BTW, for e.g. SPARC hosts with %l<R> registers, we could define
something like DONT_PRESERVE_AREG<AR> since those would lay in a new
register window and thus don't require an explicit save/restore.
Regards,
Gwenole Beauchesne
<http://svn.mandriva.com/cgi-bin/viewvc.cgi/packages/cooker/qemu/
current/SOURCES/qemu-0.8.3-preserve-host-registers.patch>
[-- Attachment #2: qemu-0.8.3-preserve-host-registers.patch --]
[-- Type: application/octet-stream, Size: 6280 bytes --]
2007-02-03 Gwenole Beauchesne <gbeauchesne@mandriva.com>
* Preserve callee-saved global registers.
--- qemu-0.8.3/cpu-exec.c.preserve-host-registers 2007-02-02 13:45:47.000000000 +0100
+++ qemu-0.8.3/cpu-exec.c 2007-02-03 09:51:55.000000000 +0100
@@ -226,37 +226,9 @@ static inline TranslationBlock *tb_find_
int cpu_exec(CPUState *env1)
{
- target_ulong saved_T0, saved_T1;
-#if defined(reg_T2)
- target_ulong saved_T2;
-#endif
- CPUState *saved_env;
-#if defined(TARGET_I386)
-#ifdef reg_EAX
- int saved_EAX;
-#endif
-#ifdef reg_ECX
- int saved_ECX;
-#endif
-#ifdef reg_EDX
- int saved_EDX;
-#endif
-#ifdef reg_EBX
- int saved_EBX;
-#endif
-#ifdef reg_ESP
- int saved_ESP;
-#endif
-#ifdef reg_EBP
- int saved_EBP;
-#endif
-#ifdef reg_ESI
- int saved_ESI;
-#endif
-#ifdef reg_EDI
- int saved_EDI;
-#endif
-#elif defined(TARGET_SPARC)
+#define DO_HOST_REGS_DECLARE
+#include "hostregs_helper.h"
+#if defined(TARGET_SPARC)
#if defined(reg_REGWPTR)
uint32_t *saved_regwptr;
#endif
@@ -325,44 +297,15 @@ int cpu_exec(CPUState *env1)
cpu_single_env = env1;
/* first we save global registers */
- saved_env = env;
+#define DO_HOST_REGS_SAVE
+#include "hostregs_helper.h"
env = env1;
- saved_T0 = T0;
- saved_T1 = T1;
-#if defined(reg_T2)
- saved_T2 = T2;
-#endif
#if defined(__sparc__) && !defined(HOST_SOLARIS)
/* we also save i7 because longjmp may not restore it */
asm volatile ("mov %%i7, %0" : "=r" (saved_i7));
#endif
#if defined(TARGET_I386)
-#ifdef reg_EAX
- saved_EAX = EAX;
-#endif
-#ifdef reg_ECX
- saved_ECX = ECX;
-#endif
-#ifdef reg_EDX
- saved_EDX = EDX;
-#endif
-#ifdef reg_EBX
- saved_EBX = EBX;
-#endif
-#ifdef reg_ESP
- saved_ESP = ESP;
-#endif
-#ifdef reg_EBP
- saved_EBP = EBP;
-#endif
-#ifdef reg_ESI
- saved_ESI = ESI;
-#endif
-#ifdef reg_EDI
- saved_EDI = EDI;
-#endif
-
env_to_regs();
/* put eflags in CPU temporary format */
CC_SRC = env->eflags & (CC_O | CC_S | CC_Z | CC_A | CC_P | CC_C);
@@ -827,32 +770,6 @@ int cpu_exec(CPUState *env1)
#endif
/* restore flags in standard format */
env->eflags = env->eflags | cc_table[CC_OP].compute_all() | (DF & DF_MASK);
-
- /* restore global registers */
-#ifdef reg_EAX
- EAX = saved_EAX;
-#endif
-#ifdef reg_ECX
- ECX = saved_ECX;
-#endif
-#ifdef reg_EDX
- EDX = saved_EDX;
-#endif
-#ifdef reg_EBX
- EBX = saved_EBX;
-#endif
-#ifdef reg_ESP
- ESP = saved_ESP;
-#endif
-#ifdef reg_EBP
- EBP = saved_EBP;
-#endif
-#ifdef reg_ESI
- ESI = saved_ESI;
-#endif
-#ifdef reg_EDI
- EDI = saved_EDI;
-#endif
#elif defined(TARGET_ARM)
/* XXX: Save/restore host fpu exception state?. */
#elif defined(TARGET_SPARC)
@@ -871,15 +788,14 @@ int cpu_exec(CPUState *env1)
#else
#error unsupported target CPU
#endif
+
+ /* restore global registers */
#if defined(__sparc__) && !defined(HOST_SOLARIS)
asm volatile ("mov %0, %%i7" : : "r" (saved_i7));
#endif
- T0 = saved_T0;
- T1 = saved_T1;
-#if defined(reg_T2)
- T2 = saved_T2;
-#endif
- env = saved_env;
+#define DO_HOST_REGS_RESTORE
+#include "hostregs_helper.h"
+
/* fail safe : never use cpu_single_env outside cpu_exec() */
cpu_single_env = NULL;
return ret;
--- qemu-0.8.3/hostregs_helper.h.preserve-host-registers 2007-02-03 10:02:51.000000000 +0100
+++ qemu-0.8.3/hostregs_helper.h 2007-02-03 10:02:27.000000000 +0100
@@ -0,0 +1,145 @@
+/*
+ * host registers helpers for dyngen execution engine
+ *
+ * Copyright (c) 2003 Fabrice Bellard
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#ifdef AREG0
+#define DO_AREG0(x) x
+#else
+#define DO_AREG0(x)
+#endif
+
+#ifdef AREG1
+#define DO_AREG1(x) x
+#else
+#define DO_AREG1(x)
+#endif
+
+#ifdef AREG2
+#define DO_AREG2(x) x
+#else
+#define DO_AREG2(x)
+#endif
+
+#ifdef AREG3
+#define DO_AREG3(x) x
+#else
+#define DO_AREG3(x)
+#endif
+
+#ifdef AREG4
+#define DO_AREG4(x) x
+#else
+#define DO_AREG4(x)
+#endif
+
+#ifdef AREG5
+#define DO_AREG5(x) x
+#else
+#define DO_AREG5(x)
+#endif
+
+#ifdef AREG6
+#define DO_AREG6(x) x
+#else
+#define DO_AREG6(x)
+#endif
+
+#ifdef AREG7
+#define DO_AREG7(x) x
+#else
+#define DO_AREG7(x)
+#endif
+
+#ifdef AREG8
+#define DO_AREG8(x) x
+#else
+#define DO_AREG8(x)
+#endif
+
+#ifdef AREG9
+#define DO_AREG9(x) x
+#else
+#define DO_AREG9(x)
+#endif
+
+#ifdef AREG10
+#define DO_AREG10(x) x
+#else
+#define DO_AREG10(x)
+#endif
+
+#ifdef AREG11
+#define DO_AREG11(x) x
+#else
+#define DO_AREG11(x)
+#endif
+
+#if defined(DO_HOST_REGS_DECLARE)
+#define DO_EXPR(REG) \
+ register uintptr_t reg_AREG##REG asm(AREG##REG); \
+ volatile uintptr_t saved_AREG##REG;
+#elif defined(DO_HOST_REGS_SAVE)
+#define DO_EXPR(REG) saved_AREG##REG = reg_AREG##REG;
+#elif defined(DO_HOST_REGS_RESTORE)
+#define DO_EXPR(REG) do { \
+ reg_AREG##REG = saved_AREG##REG; \
+ __asm__ __volatile__ ("" : : "r" (reg_AREG##REG)); \
+} while (0);
+#endif
+
+DO_AREG0(DO_EXPR(0))
+#undef DO_AREG0
+
+DO_AREG1(DO_EXPR(1))
+#undef DO_AREG1
+
+DO_AREG2(DO_EXPR(2))
+#undef DO_AREG2
+
+DO_AREG3(DO_EXPR(3))
+#undef DO_AREG3
+
+DO_AREG4(DO_EXPR(4))
+#undef DO_AREG4
+
+DO_AREG5(DO_EXPR(5))
+#undef DO_AREG5
+
+DO_AREG6(DO_EXPR(6))
+#undef DO_AREG6
+
+DO_AREG7(DO_EXPR(7))
+#undef DO_AREG7
+
+DO_AREG8(DO_EXPR(8))
+#undef DO_AREG8
+
+DO_AREG9(DO_EXPR(9))
+#undef DO_AREG9
+
+DO_AREG10(DO_EXPR(10))
+#undef DO_AREG10
+
+DO_AREG11(DO_EXPR(11))
+#undef DO_AREG11
+
+#undef DO_EXPR
+#undef DO_HOST_REGS_DECLARE
+#undef DO_HOST_REGS_SAVE
+#undef DO_HOST_REGS_RESTORE
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [Qemu-devel] Re: strange crash on FreeBSD-current/amd64 (pointer truncation?)
2007-02-03 18:12 ` Gwenole Beauchesne
@ 2007-02-04 13:38 ` Paul Brook
0 siblings, 0 replies; 11+ messages in thread
From: Paul Brook @ 2007-02-04 13:38 UTC (permalink / raw)
To: qemu-devel
On Saturday 03 February 2007 18:12, Gwenole Beauchesne wrote:
> Hi,
>
> > Hmm. All I can say is the upper half of rbx (which holds T0) gets
> > spilled on FreeBSD-current/amd64 hosts unless saving and restoring
> > the full 64 bit of it...
>
> That's also what I got with VirtualBox on x86_64. Here is an update to
> the patch I posted yesterday and that applies to current QEMU CVS
> instead.
I rewrote it a bit and applied to CVS.
Paul
^ permalink raw reply [flat|nested] 11+ messages in thread
end of thread, other threads:[~2007-02-04 13:38 UTC | newest]
Thread overview: 11+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2007-01-24 20:00 [Qemu-devel] strange crash on FreeBSD-current/amd64 (pointer truncation?) Juergen Lock
2007-02-01 20:00 ` [Qemu-devel] " Juergen Lock
2007-02-02 4:02 ` Thiemo Seufer
2007-02-02 21:45 ` Juergen Lock
2007-02-02 22:24 ` Thiemo Seufer
2007-02-03 1:43 ` Juergen Lock
2007-02-03 18:12 ` Gwenole Beauchesne
2007-02-04 13:38 ` Paul Brook
2007-02-02 22:27 ` andrzej zaborowski
2007-02-02 22:40 ` Paul Brook
2007-02-02 22:57 ` andrzej zaborowski
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.