Re: [Qemu-devel] cvttps2dq, movdq2q, movq2dq incorrect behaviour
On Wed, 21 Jun 2006, Julian Seward wrote:
Malc, your sse-movq.patch works for me. Thanks.
soft-float was a red herring, translate.c is at fault here (interpreter
does not use it, hence behaved correctly)
translate.c:3009
if (b1 = 2 ((b = 0x50 b = 0x5f) ||
b == 0xc2)) {
/* specific case for SSE single instructions */
if (b1 == 2) {
/* 32 bit access */
gen_op_ld_T0_A0[OT_LONG + s-mem_index]();
gen_op_movl_env_T0(offsetof(CPUX86State,xmm_t0.XMM_L(0)));
} else {
/* 64 bit access */
gen_ldq_env_A0[s-mem_index
2](offsetof(CPUX86State,xmm_t0.XMM_D(0))); }
} else {
gen_ldo_env_A0[s-mem_index 2](op2_offset);
}
cvttps2dq is 0x5b(b=0x5b) with repn prefix (b1=2) the above code is
optimized a bit more than it should have been, as it loads only 4 bytes
into xmm_t0 instead of 16.
Uh, fine, but I don't understand how/what to fix. Can you advise?
Following will fix the _specific_ case of cvttps2dq, ideally one
should go through all the [0x50..0x5f, 0xc2] with (repnz,repz prefix)
range and check wether the rules imposed by the above snippet apply.
--- /mnt/big/npf/cvs/qemux/qemu/target-i386/translate.c Tue Jun 20 15:19:20 2006
+++ translate.c Tue Jun 20 18:17:19 2006
@@ -3009,7 +3009,9 @@
if (b1 = 2 ((b = 0x50 b = 0x5f) ||
b == 0xc2)) {
/* specific case for SSE single instructions */
-if (b1 == 2) {
+if (b1 == 2 b == 0x5b) {
+gen_ldo_env_A0[s-mem_index
2](offsetof(CPUX86State,xmm_t0.XMM_Q(0)));
+} else if (b1 == 2) {
/* 32 bit access */
gen_op_ld_T0_A0[OT_LONG + s-mem_index]();
gen_op_movl_env_T0(offsetof(CPUX86State,xmm_t0.XMM_L(0)));
--
mailto:[EMAIL PROTECTED]
___
Qemu-devel mailing list
Qemu-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/qemu-devel
Re: [Qemu-devel] cvttps2dq, movdq2q, movq2dq incorrect behaviour
On Wed, 21 Jun 2006, malc wrote:
On Wed, 21 Jun 2006, Julian Seward wrote:
Malc, your sse-movq.patch works for me. Thanks.
soft-float was a red herring, translate.c is at fault here (interpreter
does not use it, hence behaved correctly)
[..snip..]
cvttps2dq is 0x5b(b=0x5b) with repn prefix (b1=2) the above code is
optimized a bit more than it should have been, as it loads only 4 bytes
into xmm_t0 instead of 16.
Uh, fine, but I don't understand how/what to fix. Can you advise?
Following will fix the _specific_ case of cvttps2dq, ideally one
should go through all the [0x50..0x5f, 0xc2] with (repnz,repz prefix)
range and check wether the rules imposed by the above snippet apply.
[..snip..]
It appears that cvttps2dq is indeed the only exception in the range,
combined patch that fixes both movd?q2d?q and cvttps2dq is attached.
I don't have any kind of SSE on this machine so would apprecaite if
someone would run tests/test-i386 with the patch attached.
--
mailto:[EMAIL PROTECTED]Index: target-i386/translate.c
===
RCS file: /cvsroot/qemu/qemu/target-i386/translate.c,v
retrieving revision 1.57
diff -u -u -r1.57 translate.c
--- target-i386/translate.c 14 Jun 2006 14:29:34 - 1.57
+++ target-i386/translate.c 21 Jun 2006 11:01:47 -
@@ -2947,15 +2947,15 @@
case 0x2d6: /* movq2dq */
gen_op_enter_mmx();
rm = (modrm 7) | REX_B(s);
-gen_op_movq(offsetof(CPUX86State,xmm_regs[rm].XMM_Q(0)),
-offsetof(CPUX86State,fpregs[reg 7].mmx));
-gen_op_movq_env_0(offsetof(CPUX86State,xmm_regs[rm].XMM_Q(1)));
+gen_op_movq(offsetof(CPUX86State,xmm_regs[reg 7].XMM_Q(0)),
+offsetof(CPUX86State,fpregs[rm].mmx));
+gen_op_movq_env_0(offsetof(CPUX86State,xmm_regs[reg
7].XMM_Q(1)));
break;
case 0x3d6: /* movdq2q */
gen_op_enter_mmx();
rm = (modrm 7);
-gen_op_movq(offsetof(CPUX86State,fpregs[rm].mmx),
-offsetof(CPUX86State,xmm_regs[reg].XMM_Q(0)));
+gen_op_movq(offsetof(CPUX86State,fpregs[reg].mmx),
+offsetof(CPUX86State,xmm_regs[rm].XMM_Q(0)));
break;
case 0xd7: /* pmovmskb */
case 0x1d7:
@@ -3006,8 +3006,9 @@
if (mod != 3) {
gen_lea_modrm(s, modrm, reg_addr, offset_addr);
op2_offset = offsetof(CPUX86State,xmm_t0);
-if (b1 = 2 ((b = 0x50 b = 0x5f) ||
-b == 0xc2)) {
+if (!(b1 == 2 b == 0x5b)
+(b1 = 2 ((b = 0x50 b = 0x5f) ||
+b == 0xc2))) {
/* specific case for SSE single instructions */
if (b1 == 2) {
/* 32 bit access */
___
Qemu-devel mailing list
Qemu-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/qemu-devel
[Qemu-devel] [PATCH] Soft float for Sparc32/64 (update)
Last time I missed a lot of the functions, this version should be complete. I see no difference in operation with or without the patch. Comments? Description: Change all uses of float/double and related functions in Sparc32/64 to soft float replacements. _ Express yourself instantly with MSN Messenger! Download today it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ sparc-softfloat.diff.bz2 Description: Binary data ___ Qemu-devel mailing list Qemu-devel@nongnu.org http://lists.nongnu.org/mailman/listinfo/qemu-devel
[Qemu-devel] qemu/linux-user main.c syscall.c
CVSROOT:/sources/qemu
Module name:qemu
Changes by: Fabrice Bellard bellard 06/06/21 18:15:50
Modified files:
linux-user : main.c syscall.c
Log message:
sparc user fixes (Blue Swirl)
CVSWeb URLs:
http://cvs.savannah.gnu.org/viewcvs/qemu/linux-user/main.c?cvsroot=qemur1=1.88r2=1.89
http://cvs.savannah.gnu.org/viewcvs/qemu/linux-user/syscall.c?cvsroot=qemur1=1.72r2=1.73
Patches:
Index: main.c
===
RCS file: /sources/qemu/qemu/linux-user/main.c,v
retrieving revision 1.88
retrieving revision 1.89
diff -u -b -r1.88 -r1.89
--- main.c 18 Jun 2006 19:12:54 - 1.88
+++ main.c 21 Jun 2006 18:15:50 - 1.89
@@ -598,7 +598,8 @@
#else
// XXX
#endif
- case 0x100: // XXX, why do we get these?
+case EXCP_INTERRUPT:
+/* just indicate that signals should be handled asap */
break;
case EXCP_DEBUG:
{
Index: syscall.c
===
RCS file: /sources/qemu/qemu/linux-user/syscall.c,v
retrieving revision 1.72
retrieving revision 1.73
diff -u -b -r1.72 -r1.73
--- syscall.c 14 Jun 2006 13:36:59 - 1.72
+++ syscall.c 21 Jun 2006 18:15:50 - 1.73
@@ -1538,6 +1538,11 @@
new_env-regs[13] = newsp;
new_env-regs[0] = 0;
#elif defined(TARGET_SPARC)
+if (!newsp)
+newsp = env-regwptr[22];
+new_env-regwptr[22] = newsp;
+new_env-regwptr[0] = 0;
+ /* X */
printf (HELPME: %s:%d\n, __FILE__, __LINE__);
#elif defined(TARGET_MIPS)
printf (HELPME: %s:%d\n, __FILE__, __LINE__);
@@ -3598,10 +3603,14 @@
case TARGET_NR_get_thread_area:
goto unimplemented_nowarn;
#endif
+#ifdef TARGET_NR_getdomainname
+case TARGET_NR_getdomainname:
+goto unimplemented_nowarn;
+#endif
default:
unimplemented:
gemu_log(qemu: Unsupported syscall: %d\n, num);
-#if defined(TARGET_NR_setxattr) || defined(TARGET_NR_set_thread_area)
+#if defined(TARGET_NR_setxattr) || defined(TARGET_NR_set_thread_area) ||
defined(TARGET_NR_getdomainname)
unimplemented_nowarn:
#endif
ret = -ENOSYS;
___
Qemu-devel mailing list
Qemu-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/qemu-devel
[Qemu-devel] qemu/target-sparc translate.c
CVSROOT:/sources/qemu
Module name:qemu
Changes by: Fabrice Bellard bellard 06/06/21 18:26:15
Modified files:
target-sparc : translate.c
Log message:
sparc branch fix (Blue Swirl)
CVSWeb URLs:
http://cvs.savannah.gnu.org/viewcvs/qemu/target-sparc/translate.c?cvsroot=qemur1=1.27r2=1.28
Patches:
Index: translate.c
===
RCS file: /sources/qemu/qemu/target-sparc/translate.c,v
retrieving revision 1.27
retrieving revision 1.28
diff -u -b -r1.27 -r1.28
--- translate.c 18 Jun 2006 19:36:58 - 1.27
+++ translate.c 21 Jun 2006 18:26:15 - 1.28
@@ -956,8 +956,8 @@
int cc;
target = GET_FIELD_SP(insn, 0, 18);
- target = 2;
target = sign_extend(target, 18);
+ target = 2;
cc = GET_FIELD_SP(insn, 20, 21);
if (cc == 0)
do_branch(dc, target, insn, 0);
@@ -971,8 +971,8 @@
{
target = GET_FIELD_SP(insn, 0, 13) |
(GET_FIELD_SP(insn, 20, 21) 7);
- target = 2;
target = sign_extend(target, 16);
+ target = 2;
rs1 = GET_FIELD(insn, 13, 17);
gen_movl_reg_T0(rs1);
do_branch_reg(dc, target, insn);
@@ -986,8 +986,8 @@
gen_op_trap_ifnofpu();
#endif
target = GET_FIELD_SP(insn, 0, 18);
- target = 2;
target = sign_extend(target, 19);
+ target = 2;
do_fbranch(dc, target, insn, cc);
goto jmp_insn;
}
@@ -995,8 +995,8 @@
case 0x2: /* BN+x */
{
target = GET_FIELD(insn, 10, 31);
- target = 2;
target = sign_extend(target, 22);
+ target = 2;
do_branch(dc, target, insn, 0);
goto jmp_insn;
}
@@ -1007,8 +1007,8 @@
gen_op_trap_ifnofpu();
#endif
target = GET_FIELD(insn, 10, 31);
- target = 2;
target = sign_extend(target, 22);
+ target = 2;
do_fbranch(dc, target, insn, 0);
goto jmp_insn;
}
___
Qemu-devel mailing list
Qemu-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/qemu-devel
[Qemu-devel] qemu configure target-sparc/cpu.h target-sparc/...
CVSROOT:/sources/qemu
Module name:qemu
Changes by: Fabrice Bellard bellard 06/06/21 18:37:05
Modified files:
. : configure
target-sparc : cpu.h op.c op_helper.c
Log message:
soft floats for SPARC (Blue Swirl)
CVSWeb URLs:
http://cvs.savannah.gnu.org/viewcvs/qemu/configure?cvsroot=qemur1=1.108r2=1.109
http://cvs.savannah.gnu.org/viewcvs/qemu/target-sparc/cpu.h?cvsroot=qemur1=1.21r2=1.22
http://cvs.savannah.gnu.org/viewcvs/qemu/target-sparc/op.c?cvsroot=qemur1=1.19r2=1.20
http://cvs.savannah.gnu.org/viewcvs/qemu/target-sparc/op_helper.c?cvsroot=qemur1=1.19r2=1.20
Patches:
Index: configure
===
RCS file: /sources/qemu/qemu/configure,v
retrieving revision 1.108
retrieving revision 1.109
diff -u -b -r1.108 -r1.109
--- configure 18 Jun 2006 19:16:53 - 1.108
+++ configure 21 Jun 2006 18:37:05 - 1.109
@@ -850,7 +850,7 @@
echo #define CONFIG_USER_ONLY 1 $config_h
fi
-if test $target_cpu = arm -o $target_cpu = armeb ; then
+if test $target_cpu = arm -o $target_cpu = armeb -o $target_cpu =
sparc -o $target_cpu = sparc64; then
echo CONFIG_SOFTFLOAT=yes $config_mak
echo #define CONFIG_SOFTFLOAT 1 $config_h
fi
Index: target-sparc/cpu.h
===
RCS file: /sources/qemu/qemu/target-sparc/cpu.h,v
retrieving revision 1.21
retrieving revision 1.22
diff -u -b -r1.21 -r1.22
--- target-sparc/cpu.h 5 Dec 2005 20:31:52 - 1.21
+++ target-sparc/cpu.h 21 Jun 2006 18:37:05 - 1.22
@@ -12,7 +12,7 @@
#define TARGET_FPREGS 64
#define TARGET_PAGE_BITS 12 /* XXX */
#endif
-#define TARGET_FPREG_T float
+#define TARGET_FPREG_T float32
#include cpu-defs.h
@@ -146,7 +146,7 @@
typedef struct CPUSPARCState {
target_ulong gregs[8]; /* general registers */
target_ulong *regwptr; /* pointer to current register window */
-TARGET_FPREG_Tfpr[TARGET_FPREGS]; /* floating point registers */
+float32 fpr[TARGET_FPREGS]; /* floating point registers */
target_ulong pc; /* program counter */
target_ulong npc; /* next program counter */
target_ulong y;/* multiply/divide register */
@@ -187,8 +187,8 @@
uint32_t mmuregs[16];
#endif
/* temporary float registers */
-float ft0, ft1;
-double dt0, dt1;
+float32 ft0, ft1;
+float64 dt0, dt1;
float_status fp_status;
#if defined(TARGET_SPARC64)
#define MAXTL 4
@@ -236,8 +236,6 @@
CPUSPARCState *cpu_sparc_init(void);
int cpu_sparc_exec(CPUSPARCState *s);
int cpu_sparc_close(CPUSPARCState *s);
-void cpu_get_fp64(uint64_t *pmant, uint16_t *pexp, double f);
-double cpu_put_fp64(uint64_t mant, uint16_t exp);
/* Fake impl 0, version 4 */
#define GET_PSR(env) ((0 28) | (4 24) | (env-psr PSR_ICC) | \
Index: target-sparc/op.c
===
RCS file: /sources/qemu/qemu/target-sparc/op.c,v
retrieving revision 1.19
retrieving revision 1.20
diff -u -b -r1.19 -r1.20
--- target-sparc/op.c 23 Apr 2006 21:33:48 - 1.19
+++ target-sparc/op.c 21 Jun 2006 18:37:05 - 1.20
@@ -1339,94 +1339,66 @@
helper_flush(T0);
}
-void OPPROTO op_fnegs(void)
-{
-FT0 = -FT1;
-}
-
-void OPPROTO op_fabss(void)
-{
-do_fabss();
-}
+#define F_OP(name, p) void OPPROTO op_f##name##p(void)
-#ifdef TARGET_SPARC64
-void OPPROTO op_fnegd(void)
-{
-DT0 = -DT1;
-}
-
-void OPPROTO op_fabsd(void)
-{
-do_fabsd();
-}
-#endif
-
-void OPPROTO op_fsqrts(void)
-{
-do_fsqrts();
-}
-
-void OPPROTO op_fsqrtd(void)
-{
-do_fsqrtd();
-}
-
-void OPPROTO op_fmuls(void)
-{
-FT0 *= FT1;
-}
+#define F_BINOP(name) \
+F_OP(name, s) \
+{ \
+FT0 = float32_ ## name (FT0, FT1, env-fp_status); \
+} \
+F_OP(name, d) \
+{ \
+DT0 = float64_ ## name (DT0, DT1, env-fp_status); \
+}
-void OPPROTO op_fmuld(void)
-{
-DT0 *= DT1;
-}
+F_BINOP(add);
+F_BINOP(sub);
+F_BINOP(mul);
+F_BINOP(div);
+#undef F_BINOP
void OPPROTO op_fsmuld(void)
{
-DT0 = FT0 * FT1;
-}
-
-void OPPROTO op_fadds(void)
-{
-FT0 += FT1;
+DT0 = float64_mul(float32_to_float64(FT0, env-fp_status),
+ float32_to_float64(FT1, env-fp_status),
+ env-fp_status);
}
-void OPPROTO op_faddd(void)
-{
-DT0 += DT1;
-}
+#define F_HELPER(name)\
+F_OP(name, s) \
+{ \
+do_f##name##s(); \
+} \
+F_OP(name, d) \
+{ \
+do_f##name##d(); \
+}
-void
[Qemu-devel] QEMU GUI
Hi, Concerning the QEMU GUI, my mind slightly evolved since my last posts on the topic: I think that a wxWidgets GUI would be the best as it is reasonnably portable and because it uses the native GUIs. If someone is interested, I am ready to try to include such a GUI in the QEMU repository even if it is not usable yet. Regards, Fabrice. ___ Qemu-devel mailing list Qemu-devel@nongnu.org http://lists.nongnu.org/mailman/listinfo/qemu-devel
[Qemu-devel] qemu vl.c target-sparc/cpu.h
CVSROOT:/sources/qemu
Module name:qemu
Changes by: Fabrice Bellard bellard 06/06/21 18:48:02
Modified files:
. : vl.c
target-sparc : cpu.h
Log message:
fixed sparc64 cpu fp save/restore
CVSWeb URLs:
http://cvs.savannah.gnu.org/viewcvs/qemu/vl.c?cvsroot=qemur1=1.189r2=1.190
http://cvs.savannah.gnu.org/viewcvs/qemu/target-sparc/cpu.h?cvsroot=qemur1=1.22r2=1.23
Patches:
Index: vl.c
===
RCS file: /sources/qemu/qemu/vl.c,v
retrieving revision 1.189
retrieving revision 1.190
diff -u -b -r1.189 -r1.190
--- vl.c14 Jun 2006 17:32:25 - 1.189
+++ vl.c21 Jun 2006 18:48:00 - 1.190
@@ -4146,11 +4146,11 @@
/* FPU */
for(i = 0; i TARGET_FPREGS; i++) {
union {
-TARGET_FPREG_T f;
-target_ulong i;
+float32 f;
+uint32_t i;
} u;
u.f = env-fpr[i];
-qemu_put_betl(f, u.i);
+qemu_put_be32(f, u.i);
}
qemu_put_betls(f, env-pc);
@@ -4182,10 +4182,10 @@
/* FPU */
for(i = 0; i TARGET_FPREGS; i++) {
union {
-TARGET_FPREG_T f;
-target_ulong i;
+float32 f;
+uint32_t i;
} u;
-u.i = qemu_get_betl(f);
+u.i = qemu_get_be32(f);
env-fpr[i] = u.f;
}
Index: target-sparc/cpu.h
===
RCS file: /sources/qemu/qemu/target-sparc/cpu.h,v
retrieving revision 1.22
retrieving revision 1.23
diff -u -b -r1.22 -r1.23
--- target-sparc/cpu.h 21 Jun 2006 18:37:05 - 1.22
+++ target-sparc/cpu.h 21 Jun 2006 18:48:01 - 1.23
@@ -12,7 +12,6 @@
#define TARGET_FPREGS 64
#define TARGET_PAGE_BITS 12 /* XXX */
#endif
-#define TARGET_FPREG_T float32
#include cpu-defs.h
___
Qemu-devel mailing list
Qemu-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/qemu-devel
Re: [Qemu-devel] QEMU GUI
Great Idea... This would be in c++ then, or do You fancy another wxWidget flavour? (I remember You did not like c++ in QEMU) If people are interested, we could try to port Q as a base, since it's going to be obsolete anyway (either by the new QEMU GUI or leopard)... :) http://www.kju-app.org http://www.kju-app.org/proj (trac) Mike On 21.06.2006, at 20:11, Fabrice Bellard wrote: Hi, Concerning the QEMU GUI, my mind slightly evolved since my last posts on the topic: I think that a wxWidgets GUI would be the best as it is reasonnably portable and because it uses the native GUIs. If someone is interested, I am ready to try to include such a GUI in the QEMU repository even if it is not usable yet. Regards, Fabrice. ___ Qemu-devel mailing list Qemu-devel@nongnu.org http://lists.nongnu.org/mailman/listinfo/qemu-devel ___ Qemu-devel mailing list Qemu-devel@nongnu.org http://lists.nongnu.org/mailman/listinfo/qemu-devel
[Qemu-devel] QEMU FreeBSD 6.1 npxdna: fpcurthread == curthread spewage
Hi- I'm using qemu-0.8.1 under FreeBSD 6.1-RELEASE from ports with kqemu-1.3.0pre7 (also from ports). The guest OS is Windows XP SP2. When I use the -kernel-kqemu flag to qemu, I get the following kernel console messages: Jun 21 14:04:57 jitter kernel: npxdna: fpcurthread == curthread 5173 times Jun 21 14:04:57 jitter kernel: npxdna: fpcurthread == curthread 5174 times ... This happens quite a bit (5-10 a second) when using any block device. I assume this is because kernel I/O calls are being virtualized and are the cause of this. When I do not use -kernel-kqemu, I do not get these messages. In both cases, Windows appears to work fine. This thread: http://tinyurl.com/mbm5l appears to have some information on why this happens. Has a patch been released to fix this bug? --- Zach ___ Qemu-devel mailing list Qemu-devel@nongnu.org http://lists.nongnu.org/mailman/listinfo/qemu-devel
[Qemu-devel] qemu qemu-doc.texi
CVSROOT:/sources/qemu
Module name:qemu
Changes by: Fabrice Bellard bellard 06/06/21 21:19:50
Modified files:
. : qemu-doc.texi
Log message:
more info about -std-vga
CVSWeb URLs:
http://cvs.savannah.gnu.org/viewcvs/qemu/qemu-doc.texi?cvsroot=qemur1=1.95r2=1.96
Patches:
Index: qemu-doc.texi
===
RCS file: /sources/qemu/qemu/qemu-doc.texi,v
retrieving revision 1.95
retrieving revision 1.96
diff -u -b -r1.95 -r1.96
--- qemu-doc.texi 16 Jun 2006 21:48:48 - 1.95
+++ qemu-doc.texi 21 Jun 2006 21:19:50 - 1.96
@@ -556,7 +556,10 @@
@item -std-vga
Simulate a standard VGA card with Bochs VBE extensions (default is
-Cirrus Logic GD5446 PCI VGA)
+Cirrus Logic GD5446 PCI VGA). If your guest OS supports the VESA 2.0
+VBE extensions (e.g. Windows XP) and if you want to use high
+resolution modes (= 1280x1024x16) then you should use this option.
+
@item -loadvm file
Start right away with a saved state (@code{loadvm} in monitor)
@end table
@@ -1260,6 +1263,11 @@
and use this graphic card. For optimal performances, use 16 bit color
depth in the guest and the host OS.
+If you are using Windows XP as guest OS and if you want to use high
+resolution modes which the Cirrus Logic BIOS does not support (i.e. =
+1280x1024x16), then you should use the VESA VBE virtual graphic card
+(option @option{-std-vga}).
+
@subsubsection CPU usage reduction
Windows 9x does not correctly use the CPU HLT
___
Qemu-devel mailing list
Qemu-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/qemu-devel
Re: [Qemu-devel] QEMU FreeBSD 6.1 npxdna: fpcurthread == curthread spewage
There is a bug in the FPU virtualization in kqemu which is only visible in the kernel virtualization case. I am trying a fix. Fabrice. Zach Metzinger wrote: Hi- I'm using qemu-0.8.1 under FreeBSD 6.1-RELEASE from ports with kqemu-1.3.0pre7 (also from ports). The guest OS is Windows XP SP2. When I use the -kernel-kqemu flag to qemu, I get the following kernel console messages: Jun 21 14:04:57 jitter kernel: npxdna: fpcurthread == curthread 5173 times Jun 21 14:04:57 jitter kernel: npxdna: fpcurthread == curthread 5174 times ... This happens quite a bit (5-10 a second) when using any block device. I assume this is because kernel I/O calls are being virtualized and are the cause of this. When I do not use -kernel-kqemu, I do not get these messages. In both cases, Windows appears to work fine. This thread: http://tinyurl.com/mbm5l appears to have some information on why this happens. Has a patch been released to fix this bug? --- Zach ___ Qemu-devel mailing list Qemu-devel@nongnu.org http://lists.nongnu.org/mailman/listinfo/qemu-devel ___ Qemu-devel mailing list Qemu-devel@nongnu.org http://lists.nongnu.org/mailman/listinfo/qemu-devel
Re: [Qemu-devel] cvttps2dq, movdq2q, movq2dq incorrect behaviour
It appears that cvttps2dq is indeed the only exception in the range, combined patch that fixes both movd?q2d?q and cvttps2dq is attached. I don't have any kind of SSE on this machine so would apprecaite if someone would run tests/test-i386 with the patch attached. That works for me. Thanks. Valgrind's integer/x87/MMX/SSE/SSE2 tests now all pass on i386-softmmu. I didn't try tests/test-i386 though. Fabrice, can you commit this? J ___ Qemu-devel mailing list Qemu-devel@nongnu.org http://lists.nongnu.org/mailman/listinfo/qemu-devel
Re: [Qemu-devel] QEMU FreeBSD 6.1 npxdna: fpcurthread == curthread spewage
Let me know if you need any testing. I have both 6.1-RELEASE and 5.3-RELEASE machines on which to test. Thanks, and this is a truly great bit of software! --- Zach On 6/21/06, Fabrice Bellard [EMAIL PROTECTED] wrote: There is a bug in the FPU virtualization in kqemu which is only visible in the kernel virtualization case. I am trying a fix. Fabrice. Zach Metzinger wrote: Hi- I'm using qemu-0.8.1 under FreeBSD 6.1-RELEASE from ports with kqemu-1.3.0pre7 (also from ports). The guest OS is Windows XP SP2. When I use the -kernel-kqemu flag to qemu, I get the following kernel console messages: Jun 21 14:04:57 jitter kernel: npxdna: fpcurthread == curthread 5173 times Jun 21 14:04:57 jitter kernel: npxdna: fpcurthread == curthread 5174 times ... This happens quite a bit (5-10 a second) when using any block device. I assume this is because kernel I/O calls are being virtualized and are the cause of this. When I do not use -kernel-kqemu, I do not get these messages. In both cases, Windows appears to work fine. This thread: http://tinyurl.com/mbm5l appears to have some information on why this happens. Has a patch been released to fix this bug? --- Zach ___ Qemu-devel mailing list Qemu-devel@nongnu.org http://lists.nongnu.org/mailman/listinfo/qemu-devel ___ Qemu-devel mailing list Qemu-devel@nongnu.org http://lists.nongnu.org/mailman/listinfo/qemu-devel ___ Qemu-devel mailing list Qemu-devel@nongnu.org http://lists.nongnu.org/mailman/listinfo/qemu-devel
