[Qemu-devel] [PATCH v2] translate-all: Fix formatting of dump output
The page dump writes a table with 3 abi_ulong values in each row.
These values take 8 or 16 characters (depending on sizeof abi_ulong).
Fix the table headings to be aligned with the table columns.
old:
startend size prot
00012000-00012021e000 0021e000 rwx
0040-00402000 2000 ---
00402000-004000802000 0080 rw-
new:
startend size prot
00012000-00012021e000 0021e000 rwx
0040-00402000 2000 ---
00402000-004000802000 0080 rw-
Signed-off-by: Stefan Weil s...@weilnetz.de
---
v2:
'length' is now an int value instead of a size_t.
Compilers require an int for the '*' flag.
size_t causes a compiler warning in some configurations.
translate-all.c |5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/translate-all.c b/translate-all.c
index 2c923c6..e7aff92 100644
--- a/translate-all.c
+++ b/translate-all.c
@@ -1679,8 +1679,9 @@ static int dump_region(void *priv, abi_ulong start,
/* dump memory mappings */
void page_dump(FILE *f)
{
-(void) fprintf(f, %-8s %-8s %-8s %s\n,
-start, end, size, prot);
+const int length = sizeof(abi_ulong) * 2;
+(void) fprintf(f, %-*s %-*s %-*s %s\n,
+length, start, length, end, length, size, prot);
walk_memory_regions(f, dump_region);
}
--
1.7.10.4
[Qemu-devel] [PATCH v2] block: Fix compiler warning (-Werror=uninitialized)
The patch fixes a warning from gcc (Debian 4.6.3-14+rpi1) 4.6.3:
block/stream.c:141:22: error:
‘copy’ may be used uninitialized in this function [-Werror=uninitialized]
This is not a real bug - a better compiler would not complain.
Now 'copy' has always a defined value, so the check for ret = 0
can be removed.
Signed-off-by: Stefan Weil s...@weilnetz.de
---
v2:
As noted by Andreas Färber, 'copy' must be set to false after the goto label.
Thanks for this hint.
Regards,
Stefan
block/stream.c |5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/block/stream.c b/block/stream.c
index 078ce4a..45837f4 100644
--- a/block/stream.c
+++ b/block/stream.c
@@ -119,11 +119,12 @@ wait:
break;
}
+copy = false;
+
ret = bdrv_is_allocated(bs, sector_num,
STREAM_BUFFER_SIZE / BDRV_SECTOR_SIZE, n);
if (ret == 1) {
/* Allocated in the top, no need to copy. */
-copy = false;
} else if (ret = 0) {
/* Copy if allocated in the intermediate images. Limit to the
* known-unallocated area [sector_num, sector_num+n). */
@@ -138,7 +139,7 @@ wait:
copy = (ret == 1);
}
trace_stream_one_iteration(s, sector_num, n, ret);
-if (ret = 0 copy) {
+if (copy) {
if (s-common.speed) {
delay_ns = ratelimit_calculate_delay(s-limit, n);
if (delay_ns 0) {
--
1.7.10.4
Re: [Qemu-devel] in_asm substitute for accel=kvm:tcg
On Thu, Sep 19, 2013 at 08:49:51PM +0300, Andriy Gapon wrote: on 19/09/2013 19:53 Paolo Bonzini said the following: Il 19/09/2013 16:36, Andriy Gapon ha scritto: Not sure how the code ends up at 0x9315 after that. Events are dropped, probably corresponding to more emulation. I've got a trace without dropped events between the last normal instruction and the loop (and also including a snippet where the same code is executed without a problem): Which kernel version is this? What BSD version? ... qemu-system-x86-12024 [003] 278157.048876: kvm_emulate_insn: 0:9366:b1 10 (prot32) qemu-system-x86-12024 [003] 278157.048877: kvm_entry:vcpu 0 qemu-system-x86-12024 [003] 278157.048878: kvm_emulate_insn: 0:9368:8e d1 (prot32) qemu-system-x86-12024 [003] 278157.048880: kvm_entry:vcpu 0 qemu-system-x86-12024 [003] 278157.048882: kvm_exit: reason CR_ACCESS rip 0x9312 info 0 0 qemu-system-x86-12024 [003] 278157.048883: kvm_cr: cr_write 0 = 0x10 qemu-system-x86-12024 [003] 278157.048885: kvm_entry:vcpu 0 qemu-system-x86-12024 [003] 278157.048886: kvm_emulate_insn: 0:9315:ea 1a 93 00 00 (real) qemu-system-x86-12024 [003] 278157.048887: kvm_entry:vcpu 0 qemu-system-x86-12024 [003] 278157.04: kvm_emulate_insn: 0:931a:31 c0 (real) ... ... qemu-system-x86-12024 [003] 278157.048990: kvm_set_irq: gsi 4 level 0 source 0 qemu-system-x86-12024 [003] 278157.048991: kvm_pic_set_irq: chip 0 pin 4 (edge|masked) qemu-system-x86-12024 [003] 278157.048992: kvm_ioapic_set_irq: pin 4 dst 0 vec=0 (Fixed|physical|edge|masked) qemu-system-x86-12024 [003] 278157.049001: kvm_entry:vcpu 0 qemu-system-x86-12024 [003] 278157.049002: kvm_exit: reason IO_INSTRUCTION rip 0x1e675 info 3fd0008 0 qemu-system-x86-12024 [003] 278157.049005: kvm_emulate_insn: a000:1e675:ec (prot32) qemu-system-x86-12024 [003] 278157.049005: kvm_pio: pio_read at 0x3fd size 1 count 1 qemu-system-x86-12024 [003] 278157.049006: kvm_userspace_exit: reason KVM_EXIT_IO (2) qemu-system-x86-12024 [003] 278157.049024: kvm_entry:vcpu 0 qemu-system-x86-12024 [003] 278157.049027: kvm_exit: reason CR_ACCESS rip 0x9312 info 0 0 qemu-system-x86-12024 [003] 278157.049028: kvm_cr: cr_write 0 = 0x10 qemu-system-x86-12024 [003] 278157.049030: kvm_entry:vcpu 0 qemu-system-x86-12024 [003] 278157.049031: kvm_emulate_insn: 0:9315: (real) qemu-system-x86-12024 [003] 278157.049033: kvm_emulate_insn: 0:9315: (real) qemu-system-x86-12024 [003] 278157.049034: kvm_emulate_insn: 0:9315: (real) ... It's strange that no instruction gets reported in those repeating 0:9315: (real) lines. It's like kvm is somehow losing track of what should be executed and just loops over the same ip without actually doing anything. -- Andriy Gapon -- Gleb.
Re: [Qemu-devel] [PATCH] target-i386: Enable x2apic by default on more recent CPU models
On Fri, Sep 20, 2013 at 04:15:17PM -0300, Eduardo Habkost wrote:
This enables x2apic on the following CPU models: Conroe, Penryn,
Nehalem, Westmere, Opteron_G[12345].
Normally we try to keep the CPU model definitions as close as the real
CPUs as possible, but x2apic can be emulated by KVM without host CPU
support for x2apic, and it improves performance by reducing APIC access
overhead. x2apic emulation is available on KVM since 2009 (Linux
2.6.32-rc1), there's no reason for not enabling x2apic by default when
running KVM.
About testing: Conroe, Penryn, Nehalem, Westemere and Opteron_G[123]
have x2apic enabled on RHEL-6 since RHEL-6.0, so the presence of x2apic
on those CPU models got lots of testing in the last few years. I want to
eventually enable x2apic on all other CPU models as well, but it will
require some testing to ensure it won't confuse guests.
This shouldn't affect TCG at all because features not supported by TCG
are automatically and silently disabled by QEMU when initializing the
CPU.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
Acked-by: Gleb Natapov g...@redhat.com
---
hw/i386/pc_piix.c | 9 +
hw/i386/pc_q35.c | 9 +
target-i386/cpu.c | 37 +++--
3 files changed, 37 insertions(+), 18 deletions(-)
diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c
index 907792b..0af00ef 100644
--- a/hw/i386/pc_piix.c
+++ b/hw/i386/pc_piix.c
@@ -240,6 +240,15 @@ static void pc_compat_1_6(QEMUMachineInitArgs *args)
{
has_pci_info = false;
rom_file_in_ram = false;
+x86_cpu_compat_set_features(Conroe, FEAT_1_ECX, 0, CPUID_EXT_X2APIC);
+x86_cpu_compat_set_features(Penryn, FEAT_1_ECX, 0, CPUID_EXT_X2APIC);
+x86_cpu_compat_set_features(Nehalem, FEAT_1_ECX, 0, CPUID_EXT_X2APIC);
+x86_cpu_compat_set_features(Westmere, FEAT_1_ECX, 0, CPUID_EXT_X2APIC);
+x86_cpu_compat_set_features(Opteron_G1, FEAT_1_ECX, 0,
CPUID_EXT_X2APIC);
+x86_cpu_compat_set_features(Opteron_G2, FEAT_1_ECX, 0,
CPUID_EXT_X2APIC);
+x86_cpu_compat_set_features(Opteron_G3, FEAT_1_ECX, 0,
CPUID_EXT_X2APIC);
+x86_cpu_compat_set_features(Opteron_G4, FEAT_1_ECX, 0,
CPUID_EXT_X2APIC);
+x86_cpu_compat_set_features(Opteron_G5, FEAT_1_ECX, 0,
CPUID_EXT_X2APIC);
}
static void pc_compat_1_5(QEMUMachineInitArgs *args)
diff --git a/hw/i386/pc_q35.c b/hw/i386/pc_q35.c
index ca84e1c..82e7a23 100644
--- a/hw/i386/pc_q35.c
+++ b/hw/i386/pc_q35.c
@@ -224,6 +224,15 @@ static void pc_compat_1_6(QEMUMachineInitArgs *args)
{
has_pci_info = false;
rom_file_in_ram = false;
+x86_cpu_compat_set_features(Conroe, FEAT_1_ECX, 0, CPUID_EXT_X2APIC);
+x86_cpu_compat_set_features(Penryn, FEAT_1_ECX, 0, CPUID_EXT_X2APIC);
+x86_cpu_compat_set_features(Nehalem, FEAT_1_ECX, 0, CPUID_EXT_X2APIC);
+x86_cpu_compat_set_features(Westmere, FEAT_1_ECX, 0, CPUID_EXT_X2APIC);
+x86_cpu_compat_set_features(Opteron_G1, FEAT_1_ECX, 0,
CPUID_EXT_X2APIC);
+x86_cpu_compat_set_features(Opteron_G2, FEAT_1_ECX, 0,
CPUID_EXT_X2APIC);
+x86_cpu_compat_set_features(Opteron_G3, FEAT_1_ECX, 0,
CPUID_EXT_X2APIC);
+x86_cpu_compat_set_features(Opteron_G4, FEAT_1_ECX, 0,
CPUID_EXT_X2APIC);
+x86_cpu_compat_set_features(Opteron_G5, FEAT_1_ECX, 0,
CPUID_EXT_X2APIC);
}
static void pc_compat_1_5(QEMUMachineInitArgs *args)
diff --git a/target-i386/cpu.c b/target-i386/cpu.c
index 9abb73f..23a44c3 100644
--- a/target-i386/cpu.c
+++ b/target-i386/cpu.c
@@ -791,7 +791,7 @@ static x86_def_t builtin_x86_defs[] = {
CPUID_MCE | CPUID_PAE | CPUID_MSR | CPUID_TSC | CPUID_PSE |
CPUID_DE | CPUID_FP87,
.features[FEAT_1_ECX] =
-CPUID_EXT_SSSE3 | CPUID_EXT_SSE3,
+CPUID_EXT_X2APIC | CPUID_EXT_SSSE3 | CPUID_EXT_SSE3,
.features[FEAT_8000_0001_EDX] =
CPUID_EXT2_LM | CPUID_EXT2_NX | CPUID_EXT2_SYSCALL,
.features[FEAT_8000_0001_ECX] =
@@ -813,8 +813,8 @@ static x86_def_t builtin_x86_defs[] = {
CPUID_MCE | CPUID_PAE | CPUID_MSR | CPUID_TSC | CPUID_PSE |
CPUID_DE | CPUID_FP87,
.features[FEAT_1_ECX] =
-CPUID_EXT_SSE41 | CPUID_EXT_CX16 | CPUID_EXT_SSSE3 |
- CPUID_EXT_SSE3,
+CPUID_EXT_X2APIC | CPUID_EXT_SSE41 | CPUID_EXT_CX16 |
+CPUID_EXT_SSSE3 | CPUID_EXT_SSE3,
.features[FEAT_8000_0001_EDX] =
CPUID_EXT2_LM | CPUID_EXT2_NX | CPUID_EXT2_SYSCALL,
.features[FEAT_8000_0001_ECX] =
@@ -836,8 +836,8 @@ static x86_def_t builtin_x86_defs[] = {
CPUID_MCE | CPUID_PAE | CPUID_MSR | CPUID_TSC | CPUID_PSE |
CPUID_DE | CPUID_FP87,
.features[FEAT_1_ECX] =
-CPUID_EXT_POPCNT | CPUID_EXT_SSE42 | CPUID_EXT_SSE41 |
- CPUID_EXT_CX16 | CPUID_EXT_SSSE3 | CPUID_EXT_SSE3,
+CPUID_EXT_POPCNT | CPUID_EXT_X2APIC |
Re: [Qemu-devel] [PATCH V4 0/3] qemu-iotests: add test for fd passing via SCM rights
On 09/20/2013 11:11 PM, Stefan Hajnoczi wrote: On Mon, Sep 09, 2013 at 01:57:20PM +0200, Stefan Hajnoczi wrote: On Fri, Sep 06, 2013 at 11:24:31AM +0800, Wenchao Xia wrote: tests/qemu-iotests/socket_scm_helper.c | 135 After a clean QEMU build ./check -raw fails now. This can be resolved by manually building the helper program but I think it should be part of the main QEMU build process (make install simply won't copy the helper program). Wenchao: Can you send a patch to include the helper into the build process? Sorry for late response, I was on vacation before. Will sent a patch later. 045 1s ... [failed, exit status 1] - output mismatch (see 045.out.bad) --- 045.out 2013-09-20 16:08:02.193074640 +0200 +++ 045.out.bad 2013-09-20 17:02:51.492683466 +0200 @@ -1,5 +1,49 @@ -... +..FF.FF +== +FAIL: test_add_fd (__main__.TestSCMFd) +-- +Traceback (most recent call last): + File ./045, line 147, in test_add_fd +self._send_fd_by_SCM() + File ./045, line 144, in _send_fd_by_SCM +self.assertEqual(ret, 0, 'Failed to send fd with UNIX SCM') +AssertionError: Failed to send fd with UNIX SCM + +== +FAIL: test_closefd (__main__.TestSCMFd) +-- +Traceback (most recent call last): + File ./045, line 164, in test_closefd +self._send_fd_by_SCM() + File ./045, line 144, in _send_fd_by_SCM +self.assertEqual(ret, 0, 'Failed to send fd with UNIX SCM') +AssertionError: Failed to send fd with UNIX SCM + +== +FAIL: test_getfd (__main__.TestSCMFd) +-- +Traceback (most recent call last): + File ./045, line 152, in test_getfd +self._send_fd_by_SCM() + File ./045, line 144, in _send_fd_by_SCM +self.assertEqual(ret, 0, 'Failed to send fd with UNIX SCM') +AssertionError: Failed to send fd with UNIX SCM + +== +FAIL: test_getfd_invalid_fdname (__main__.TestSCMFd) +-- +Traceback (most recent call last): + File ./045, line 157, in test_getfd_invalid_fdname +self._send_fd_by_SCM() + File ./045, line 144, in _send_fd_by_SCM +self.assertEqual(ret, 0, 'Failed to send fd with UNIX SCM') +AssertionError: Failed to send fd with UNIX SCM + -- Ran 11 tests -OK +FAILED (failures=4) +Scm help program does not present, path 'socket_scm_helper'. +Scm help program does not present, path 'socket_scm_helper'. +Scm help program does not present, path 'socket_scm_helper'. +Scm help program does not present, path 'socket_scm_helper'.
[Qemu-devel] [PATCH] Extend qemu-ga's 'guest-info' command to expose flag 'success-response'
Now we have several qemu-ga commands not returning response on success.
It has been documented in qga/qapi-schema.json already. This patch exposes
the 'success-response' flag by extending 'guest-info' command. With this
change, the clients can handle the command response more flexibly.
Changes:
v2: add the notation 'since 1.7' to the option 'success-response'
(per Eric Blake's comments)
Signed-off-by: Mark Wu wu...@linux.vnet.ibm.com
---
include/qapi/qmp/dispatch.h | 1 +
qapi/qmp-registry.c | 13 +
qga/commands.c | 2 ++
qga/qapi-schema.json| 5 -
4 files changed, 20 insertions(+), 1 deletion(-)
diff --git a/include/qapi/qmp/dispatch.h b/include/qapi/qmp/dispatch.h
index 1ce11f5..f08c1dd 100644
--- a/include/qapi/qmp/dispatch.h
+++ b/include/qapi/qmp/dispatch.h
@@ -48,6 +48,7 @@ QObject *qmp_dispatch(QObject *request);
void qmp_disable_command(const char *name);
void qmp_enable_command(const char *name);
bool qmp_command_is_enabled(const char *name);
+bool qmp_command_has_success_response(const char *name);
char **qmp_get_command_list(void);
QObject *qmp_build_error_object(Error *errp);
diff --git a/qapi/qmp-registry.c b/qapi/qmp-registry.c
index 28bbbe8..f86c3a5 100644
--- a/qapi/qmp-registry.c
+++ b/qapi/qmp-registry.c
@@ -79,6 +79,19 @@ bool qmp_command_is_enabled(const char *name)
return false;
}
+bool qmp_command_has_success_response(const char *name)
+{
+QmpCommand *cmd;
+
+QTAILQ_FOREACH(cmd, qmp_commands, node) {
+if (strcmp(cmd-name, name) == 0) {
+return cmd-options != QCO_NO_SUCCESS_RESP;
+}
+}
+
+return false;
+}
+
char **qmp_get_command_list(void)
{
QmpCommand *cmd;
diff --git a/qga/commands.c b/qga/commands.c
index 528b082..0df910b 100644
--- a/qga/commands.c
+++ b/qga/commands.c
@@ -63,6 +63,8 @@ struct GuestAgentInfo *qmp_guest_info(Error **err)
cmd_info = g_malloc0(sizeof(GuestAgentCommandInfo));
cmd_info-name = g_strdup(*cmd_list);
cmd_info-enabled = qmp_command_is_enabled(cmd_info-name);
+cmd_info-success_response =
+qmp_command_has_success_response(cmd_info-name);
cmd_info_list = g_malloc0(sizeof(GuestAgentCommandInfoList));
cmd_info_list-value = cmd_info;
diff --git a/qga/qapi-schema.json b/qga/qapi-schema.json
index 7155b7a..9c6a58f 100644
--- a/qga/qapi-schema.json
+++ b/qga/qapi-schema.json
@@ -141,10 +141,13 @@
#
# @enabled: whether command is currently enabled by guest admin
#
+# @success-response: whether command returns a response on success
+#and only available since 1.7
+#
# Since 1.1.0
##
{ 'type': 'GuestAgentCommandInfo',
- 'data': { 'name': 'str', 'enabled': 'bool' } }
+ 'data': { 'name': 'str', 'enabled': 'bool', 'success-response': 'bool' } }
##
# @GuestAgentInfo
--
1.8.3.1
[Qemu-devel] [PATCH 0/4] qom: add helpers for integer properties
Add helper functions for adding read-only properties, that work in the case where the value is in memory. Michael S. Tsirkin (4): qemu: add Error to typedefs qom: pull in qemu/typedefs qom: cleanup struct Error references qom: add pointer to int property helpers include/qemu/typedefs.h | 1 + include/qom/object.h| 73 +++-- qom/object.c| 56 + 3 files changed, 104 insertions(+), 26 deletions(-) -- MST
[Qemu-devel] [PATCH 2/4] qom: pull in qemu/typedefs
As usual so we can use typedefs without header dependencies. Signed-off-by: Michael S. Tsirkin m...@redhat.com --- include/qom/object.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/qom/object.h b/include/qom/object.h index 1a7b71a..5b3b743 100644 --- a/include/qom/object.h +++ b/include/qom/object.h @@ -17,6 +17,7 @@ #include glib.h #include stdint.h #include stdbool.h +#include qemu/typedefs.h #include qemu/queue.h struct Visitor; -- MST
[Qemu-devel] [PATCH 1/4] qemu: add Error to typedefs
This is so qom headers can use it without pulling in extra headers. Signed-off-by: Michael S. Tsirkin m...@redhat.com --- include/qemu/typedefs.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/qemu/typedefs.h b/include/qemu/typedefs.h index c5c6e36..bb5acf9 100644 --- a/include/qemu/typedefs.h +++ b/include/qemu/typedefs.h @@ -9,6 +9,7 @@ typedef struct QEMUFile QEMUFile; typedef struct QEMUBH QEMUBH; typedef struct QInt QInt; typedef struct QObject QObject; +typedef struct Error Error; typedef struct AioContext AioContext; -- MST
[Qemu-devel] [PATCH 4/4] qom: add pointer to int property helpers
Make it easy to add read-only helpers for simple
integer properties in memory.
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
include/qom/object.h | 21
qom/object.c | 56
2 files changed, 77 insertions(+)
diff --git a/include/qom/object.h b/include/qom/object.h
index e3a16e4..3b75f5a 100644
--- a/include/qom/object.h
+++ b/include/qom/object.h
@@ -795,6 +795,27 @@ void object_property_add(Object *obj, const char *name,
const char *type,
void object_property_del(Object *obj, const char *name, Error **errp);
/**
+ * object_property_add_uint8_ptr:
+ * object_property_add_uint16_ptr:
+ * object_property_add_uint32_ptr:
+ * object_property_add_uint64_ptr:
+ * @obj: the object to add a property to
+ * @name: the name of the property
+ * @v: pointer to value
+ *
+ * Add an integer property in memory. This function will add a
+ * property of the appropriate type.
+ */
+void object_property_add_uint8_ptr(Object *obj, const char *name,
+ const uint8_t *v, Error **errp);
+void object_property_add_uint16_ptr(Object *obj, const char *name,
+const uint16_t *v, Error **errp);
+void object_property_add_uint32_ptr(Object *obj, const char *name,
+const uint32_t *v, Error **errp);
+void object_property_add_uint64_ptr(Object *obj, const char *name,
+const uint64_t *v, Error **Errp);
+
+/**
* object_property_find:
* @obj: the object
* @name: the name of the property
diff --git a/qom/object.c b/qom/object.c
index e90e382..b929dc6 100644
--- a/qom/object.c
+++ b/qom/object.c
@@ -1344,6 +1344,62 @@ static char *qdev_get_type(Object *obj, Error **errp)
return g_strdup(object_get_typename(obj));
}
+static void property_get_uint8_ptr(Object *obj, Visitor *v,
+ void *opaque, const char *name,
+ Error **errp)
+{
+visit_type_uint8(v, opaque, name, errp);
+}
+
+static void property_get_uint16_ptr(Object *obj, Visitor *v,
+ void *opaque, const char *name,
+ Error **errp)
+{
+visit_type_uint16(v, opaque, name, errp);
+}
+
+static void property_get_uint32_ptr(Object *obj, Visitor *v,
+ void *opaque, const char *name,
+ Error **errp)
+{
+visit_type_uint32(v, opaque, name, errp);
+}
+
+static void property_get_uint64_ptr(Object *obj, Visitor *v,
+ void *opaque, const char *name,
+ Error **errp)
+{
+visit_type_uint64(v, opaque, name, errp);
+}
+
+void object_property_add_uint8_ptr(Object *obj, const char *name,
+ const uint8_t *v, Error **errp)
+{
+object_property_add(obj, name, uint8, property_get_uint8_ptr,
+NULL, NULL, (void *)v, errp);
+}
+
+void object_property_add_uint16_ptr(Object *obj, const char *name,
+const uint16_t *v, Error **errp)
+{
+object_property_add(obj, name, uint16, property_get_uint16_ptr,
+NULL, NULL, (void *)v, errp);
+}
+
+void object_property_add_uint32_ptr(Object *obj, const char *name,
+const uint32_t *v, Error **errp)
+{
+object_property_add(obj, name, uint32, property_get_uint32_ptr,
+NULL, NULL, (void *)v, errp);
+}
+
+void object_property_add_uint64_ptr(Object *obj, const char *name,
+const uint64_t *v, Error **errp)
+{
+object_property_add(obj, name, uint64, property_get_uint64_ptr,
+NULL, NULL, (void *)v, errp);
+}
+
static void object_instance_init(Object *obj)
{
object_property_add_str(obj, type, qdev_get_type, NULL, NULL);
--
MST
[Qemu-devel] [PATCH 3/4] qom: cleanup struct Error references
now that a typedef for struct Error is available, use it in qom/object.h to match coding style rules. Signed-off-by: Michael S. Tsirkin m...@redhat.com --- include/qom/object.h | 51 +-- 1 file changed, 25 insertions(+), 26 deletions(-) diff --git a/include/qom/object.h b/include/qom/object.h index 5b3b743..e3a16e4 100644 --- a/include/qom/object.h +++ b/include/qom/object.h @@ -21,7 +21,6 @@ #include qemu/queue.h struct Visitor; -struct Error; struct TypeImpl; typedef struct TypeImpl *Type; @@ -302,7 +301,7 @@ typedef void (ObjectPropertyAccessor)(Object *obj, struct Visitor *v, void *opaque, const char *name, - struct Error **errp); + Error **errp); /** * ObjectPropertyRelease: @@ -791,9 +790,9 @@ void object_property_add(Object *obj, const char *name, const char *type, ObjectPropertyAccessor *get, ObjectPropertyAccessor *set, ObjectPropertyRelease *release, - void *opaque, struct Error **errp); + void *opaque, Error **errp); -void object_property_del(Object *obj, const char *name, struct Error **errp); +void object_property_del(Object *obj, const char *name, Error **errp); /** * object_property_find: @@ -804,7 +803,7 @@ void object_property_del(Object *obj, const char *name, struct Error **errp); * Look up a property for an object and return its #ObjectProperty if found. */ ObjectProperty *object_property_find(Object *obj, const char *name, - struct Error **errp); + Error **errp); void object_unparent(Object *obj); @@ -819,7 +818,7 @@ void object_unparent(Object *obj); * Reads a property from a object. */ void object_property_get(Object *obj, struct Visitor *v, const char *name, - struct Error **errp); + Error **errp); /** * object_property_set_str: @@ -830,7 +829,7 @@ void object_property_get(Object *obj, struct Visitor *v, const char *name, * Writes a string value to a property. */ void object_property_set_str(Object *obj, const char *value, - const char *name, struct Error **errp); + const char *name, Error **errp); /** * object_property_get_str: @@ -843,7 +842,7 @@ void object_property_set_str(Object *obj, const char *value, * The caller should free the string. */ char *object_property_get_str(Object *obj, const char *name, - struct Error **errp); + Error **errp); /** * object_property_set_link: @@ -854,7 +853,7 @@ char *object_property_get_str(Object *obj, const char *name, * Writes an object's canonical path to a property. */ void object_property_set_link(Object *obj, Object *value, - const char *name, struct Error **errp); + const char *name, Error **errp); /** * object_property_get_link: @@ -867,7 +866,7 @@ void object_property_set_link(Object *obj, Object *value, * string or not a valid object path). */ Object *object_property_get_link(Object *obj, const char *name, - struct Error **errp); + Error **errp); /** * object_property_set_bool: @@ -878,7 +877,7 @@ Object *object_property_get_link(Object *obj, const char *name, * Writes a bool value to a property. */ void object_property_set_bool(Object *obj, bool value, - const char *name, struct Error **errp); + const char *name, Error **errp); /** * object_property_get_bool: @@ -890,7 +889,7 @@ void object_property_set_bool(Object *obj, bool value, * an error occurs (including when the property value is not a bool). */ bool object_property_get_bool(Object *obj, const char *name, - struct Error **errp); + Error **errp); /** * object_property_set_int: @@ -901,7 +900,7 @@ bool object_property_get_bool(Object *obj, const char *name, * Writes an integer value to a property. */ void object_property_set_int(Object *obj, int64_t value, - const char *name, struct Error **errp); + const char *name, Error **errp); /** * object_property_get_int: @@ -913,7 +912,7 @@ void object_property_set_int(Object *obj, int64_t value, * an error occurs (including when the property value is not an integer). */ int64_t object_property_get_int(Object *obj, const char *name, -struct Error **errp); +Error **errp); /**
Re: [Qemu-devel] [PATCH v3 0/4] timers thread-safe stuff
On Wed, Sep 18, 2013 at 9:54 PM, Stefan Hajnoczi stefa...@gmail.com wrote: On Tue, Aug 27, 2013 at 11:20:59AM +0800, Liu Ping Fan wrote: Saw the Alex's patches has been merged, rebase mine onto his. v3: 1. rename seqlock_read_check as seqlock_read_retry 2. Document timerlist were protected by BQL, and discard private lock around qemu_event_wait(tl-ev). v2: 1. fix comment in commit and code 2. fix race issue for qemu_clock_enable(foo,disable) Liu Ping Fan (2): timer: protect timers_state's clock with seqlock timer: make qemu_clock_enable sync between disable and timer's cb Paolo Bonzini (2): seqlock: introduce read-write seqlock qemu-thread: add QemuEvent cpus.c | 36 +++--- include/qemu/seqlock.h | 72 +++ include/qemu/thread-posix.h | 8 +++ include/qemu/thread-win32.h | 4 ++ include/qemu/thread.h | 7 +++ include/qemu/timer.h| 4 ++ qemu-timer.c| 20 +++- util/qemu-thread-posix.c| 116 util/qemu-thread-win32.c| 26 ++ 9 files changed, 286 insertions(+), 7 deletions(-) create mode 100644 include/qemu/seqlock.h Ping Fan: Can you send a final version that addresses Alex's request for documentation? Otherwise we're ready to go. Sorry, miss this letter, will update it immediately. Thx, Pingfan
Re: [Qemu-devel] in_asm substitute for accel=kvm:tcg
on 22/09/2013 09:31 Gleb Natapov said the following: Which kernel version is this? What BSD version? $ uname -a Linux kvm 3.8.0-27-generic #40-Ubuntu SMP Tue Jul 9 00:17:05 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux FreeBSD is 9.x. -- Andriy Gapon
Re: [Qemu-devel] [PATCH 00/11] virtio: cleanup and fix hot-unplug
Il 21/09/2013 21:17, Michael S. Tsirkin ha scritto: On Fri, Sep 20, 2013 at 04:57:49PM +0200, Paolo Bonzini wrote: This series fixes hot-unplug of virtio devices, which can crash due to dangling pointer accesses. Could you please describe the sequence of steps that makes qemu crash? See patch 11. I didn't find out why it fails with PCIe but not PCI, probably a difference in how malloc reuses freed blocks. Paolo
[Qemu-devel] [PATCH v4 0/4] timers thread-safe stuff
v4: fix commit log for protect timers_state's clock with seqlock (Thanks for Alex) v3: 1. rename seqlock_read_check as seqlock_read_retry 2. Document timerlist were protected by BQL, and discard private lock around qemu_event_wait(tl-ev). v2: 1. fix comment in commit and code 2. fix race issue for qemu_clock_enable(foo,disable) Liu Ping Fan (2): timer: protect timers_state's clock with seqlock timer: make qemu_clock_enable sync between disable and timer's cb Paolo Bonzini (2): seqlock: introduce read-write seqlock qemu-thread: add QemuEvent cpus.c | 36 +++--- include/qemu/seqlock.h | 72 +++ include/qemu/thread-posix.h | 8 +++ include/qemu/thread-win32.h | 4 ++ include/qemu/thread.h | 7 +++ include/qemu/timer.h| 4 ++ qemu-timer.c| 20 +++- util/qemu-thread-posix.c| 116 util/qemu-thread-win32.c| 26 ++ 9 files changed, 286 insertions(+), 7 deletions(-) create mode 100644 include/qemu/seqlock.h -- 1.8.1.4
[Qemu-devel] [PATCH v4 1/4] seqlock: introduce read-write seqlock
This lets the read-side access run outside the BQL.
Signed-off-by: Paolo Bonzini pbonz...@redhat.com
---
include/qemu/seqlock.h | 72 ++
1 file changed, 72 insertions(+)
create mode 100644 include/qemu/seqlock.h
diff --git a/include/qemu/seqlock.h b/include/qemu/seqlock.h
new file mode 100644
index 000..3ff118a
--- /dev/null
+++ b/include/qemu/seqlock.h
@@ -0,0 +1,72 @@
+/*
+ * Seqlock implementation for QEMU
+ *
+ * Copyright Red Hat, Inc. 2013
+ *
+ * Author:
+ * Paolo Bonzini pbonz...@redhat.com
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ *
+ */
+#ifndef QEMU_SEQLOCK_H
+#define QEMU_SEQLOCK_H 1
+
+#include qemu/atomic.h
+#include qemu/thread.h
+
+typedef struct QemuSeqLock QemuSeqLock;
+
+struct QemuSeqLock {
+QemuMutex *mutex;
+unsigned sequence;
+};
+
+static inline void seqlock_init(QemuSeqLock *sl, QemuMutex *mutex)
+{
+sl-mutex = mutex;
+sl-sequence = 0;
+}
+
+/* Lock out other writers and update the count. */
+static inline void seqlock_write_lock(QemuSeqLock *sl)
+{
+if (sl-mutex) {
+qemu_mutex_lock(sl-mutex);
+}
+++sl-sequence;
+
+/* Write sequence before updating other fields. */
+smp_wmb();
+}
+
+static inline void seqlock_write_unlock(QemuSeqLock *sl)
+{
+/* Write other fields before finalizing sequence. */
+smp_wmb();
+
+++sl-sequence;
+if (sl-mutex) {
+qemu_mutex_unlock(sl-mutex);
+}
+}
+
+static inline unsigned seqlock_read_begin(QemuSeqLock *sl)
+{
+/* Always fail if a write is in progress. */
+unsigned ret = sl-sequence ~1;
+
+/* Read sequence before reading other fields. */
+smp_rmb();
+return ret;
+}
+
+static int seqlock_read_retry(const QemuSeqLock *sl, unsigned start)
+{
+/* Read other fields before reading final sequence. */
+smp_rmb();
+return unlikely(sl-sequence != start);
+}
+
+#endif
--
1.8.1.4
[Qemu-devel] [PATCH v4 2/4] timer: protect timers_state's clock with seqlock
QEMU_CLOCK_VIRTUAL may be read outside BQL. This will make its
foundation, i.e. timers_state exposed to race condition.
Using private lock to protect it.
After this patch, reading QEMU_CLOCK_VIRTUAL is thread safe
unless use_icount is true, in which case the existing callers
still rely on the BQL
Lock rule: private lock innermost, ie BQL-this lock
Signed-off-by: Liu Ping Fan pingf...@linux.vnet.ibm.com
---
cpus.c | 36 ++--
1 file changed, 30 insertions(+), 6 deletions(-)
diff --git a/cpus.c b/cpus.c
index e566297..870a832 100644
--- a/cpus.c
+++ b/cpus.c
@@ -37,6 +37,7 @@
#include sysemu/qtest.h
#include qemu/main-loop.h
#include qemu/bitmap.h
+#include qemu/seqlock.h
#ifndef _WIN32
#include qemu/compatfd.h
@@ -112,6 +113,13 @@ static int64_t qemu_icount;
typedef struct TimersState {
int64_t cpu_ticks_prev;
int64_t cpu_ticks_offset;
+/* cpu_clock_offset will be read out of BQL, so protect it with private
+ * lock. As for cpu_ticks_*, no requirement to read it outside BQL yet.
+ * Lock rule: innermost
+ */
+QemuSeqLock clock_seqlock;
+/* mutex for seqlock */
+QemuMutex mutex;
int64_t cpu_clock_offset;
int32_t cpu_ticks_enabled;
int64_t dummy;
@@ -137,6 +145,7 @@ int64_t cpu_get_icount(void)
}
/* return the host CPU cycle counter and handle stop/restart */
+/* cpu_ticks is safely if holding BQL */
int64_t cpu_get_ticks(void)
{
if (use_icount) {
@@ -161,33 +170,46 @@ int64_t cpu_get_ticks(void)
int64_t cpu_get_clock(void)
{
int64_t ti;
-if (!timers_state.cpu_ticks_enabled) {
-return timers_state.cpu_clock_offset;
-} else {
-ti = get_clock();
-return ti + timers_state.cpu_clock_offset;
-}
+unsigned start;
+
+do {
+start = seqlock_read_begin(timers_state.clock_seqlock);
+if (!timers_state.cpu_ticks_enabled) {
+ti = timers_state.cpu_clock_offset;
+} else {
+ti = get_clock();
+ti += timers_state.cpu_clock_offset;
+}
+} while (seqlock_read_retry(timers_state.clock_seqlock, start));
+
+return ti;
}
/* enable cpu_get_ticks() */
void cpu_enable_ticks(void)
{
+/* Here, the really thing protected by seqlock is cpu_clock_offset. */
+seqlock_write_lock(timers_state.clock_seqlock);
if (!timers_state.cpu_ticks_enabled) {
timers_state.cpu_ticks_offset -= cpu_get_real_ticks();
timers_state.cpu_clock_offset -= get_clock();
timers_state.cpu_ticks_enabled = 1;
}
+seqlock_write_unlock(timers_state.clock_seqlock);
}
/* disable cpu_get_ticks() : the clock is stopped. You must not call
cpu_get_ticks() after that. */
void cpu_disable_ticks(void)
{
+/* Here, the really thing protected by seqlock is cpu_clock_offset. */
+seqlock_write_lock(timers_state.clock_seqlock);
if (timers_state.cpu_ticks_enabled) {
timers_state.cpu_ticks_offset = cpu_get_ticks();
timers_state.cpu_clock_offset = cpu_get_clock();
timers_state.cpu_ticks_enabled = 0;
}
+seqlock_write_unlock(timers_state.clock_seqlock);
}
/* Correlation between real and virtual time is always going to be
@@ -371,6 +393,8 @@ static const VMStateDescription vmstate_timers = {
void configure_icount(const char *option)
{
+qemu_mutex_init(timers_state.mutex);
+seqlock_init(timers_state.clock_seqlock, timers_state.mutex);
vmstate_register(NULL, 0, vmstate_timers, timers_state);
if (!option) {
return;
--
1.8.1.4
[Qemu-devel] [PATCH v4 1/4] seqlock: introduce read-write seqlock
This lets the read-side access run outside the BQL.
Signed-off-by: Paolo Bonzini pbonz...@redhat.com
---
include/qemu/seqlock.h | 72 ++
1 file changed, 72 insertions(+)
create mode 100644 include/qemu/seqlock.h
diff --git a/include/qemu/seqlock.h b/include/qemu/seqlock.h
new file mode 100644
index 000..3ff118a
--- /dev/null
+++ b/include/qemu/seqlock.h
@@ -0,0 +1,72 @@
+/*
+ * Seqlock implementation for QEMU
+ *
+ * Copyright Red Hat, Inc. 2013
+ *
+ * Author:
+ * Paolo Bonzini pbonz...@redhat.com
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ *
+ */
+#ifndef QEMU_SEQLOCK_H
+#define QEMU_SEQLOCK_H 1
+
+#include qemu/atomic.h
+#include qemu/thread.h
+
+typedef struct QemuSeqLock QemuSeqLock;
+
+struct QemuSeqLock {
+QemuMutex *mutex;
+unsigned sequence;
+};
+
+static inline void seqlock_init(QemuSeqLock *sl, QemuMutex *mutex)
+{
+sl-mutex = mutex;
+sl-sequence = 0;
+}
+
+/* Lock out other writers and update the count. */
+static inline void seqlock_write_lock(QemuSeqLock *sl)
+{
+if (sl-mutex) {
+qemu_mutex_lock(sl-mutex);
+}
+++sl-sequence;
+
+/* Write sequence before updating other fields. */
+smp_wmb();
+}
+
+static inline void seqlock_write_unlock(QemuSeqLock *sl)
+{
+/* Write other fields before finalizing sequence. */
+smp_wmb();
+
+++sl-sequence;
+if (sl-mutex) {
+qemu_mutex_unlock(sl-mutex);
+}
+}
+
+static inline unsigned seqlock_read_begin(QemuSeqLock *sl)
+{
+/* Always fail if a write is in progress. */
+unsigned ret = sl-sequence ~1;
+
+/* Read sequence before reading other fields. */
+smp_rmb();
+return ret;
+}
+
+static int seqlock_read_retry(const QemuSeqLock *sl, unsigned start)
+{
+/* Read other fields before reading final sequence. */
+smp_rmb();
+return unlikely(sl-sequence != start);
+}
+
+#endif
--
1.8.1.4
[Qemu-devel] [PATCH v4 4/4] timer: make qemu_clock_enable sync between disable and timer's cb
After disabling the QemuClock, we should make sure that no QemuTimers
are still in flight. To implement that with light overhead, we resort
to QemuEvent. The caller of disabling will wait on QemuEvent of each
timerlist.
Note, qemu_clock_enable(foo,false) can _not_ be called from timer's cb.
And the callers of qemu_clock_enable() should be sync by themselves,
not protected by this patch.
Signed-off-by: Liu Ping Fan pingf...@linux.vnet.ibm.com
---
include/qemu/timer.h | 4
qemu-timer.c | 20 +++-
2 files changed, 23 insertions(+), 1 deletion(-)
diff --git a/include/qemu/timer.h b/include/qemu/timer.h
index e4934dd..b26909a 100644
--- a/include/qemu/timer.h
+++ b/include/qemu/timer.h
@@ -185,6 +185,10 @@ void qemu_clock_notify(QEMUClockType type);
* @enabled: true to enable, false to disable
*
* Enable or disable a clock
+ * Disabling the clock will wait for related timerlists to stop
+ * executing qemu_run_timers. Thus, this functions should not
+ * be used from the callback of a timer that is based on @clock.
+ * Doing so would cause a deadlock.
*/
void qemu_clock_enable(QEMUClockType type, bool enabled);
diff --git a/qemu-timer.c b/qemu-timer.c
index 95ff47f..c500a76 100644
--- a/qemu-timer.c
+++ b/qemu-timer.c
@@ -45,6 +45,7 @@
/* timers */
typedef struct QEMUClock {
+ /* We rely on BQL to protect the timerlists */
QLIST_HEAD(, QEMUTimerList) timerlists;
NotifierList reset_notifiers;
@@ -70,6 +71,8 @@ struct QEMUTimerList {
QLIST_ENTRY(QEMUTimerList) list;
QEMUTimerListNotifyCB *notify_cb;
void *notify_opaque;
+/* light weight method to mark the end of timerlist's running */
+QemuEvent ev;
};
/**
@@ -98,6 +101,7 @@ QEMUTimerList *timerlist_new(QEMUClockType type,
QEMUClock *clock = qemu_clock_ptr(type);
timer_list = g_malloc0(sizeof(QEMUTimerList));
+qemu_event_init(timer_list-ev, false);
timer_list-clock = clock;
timer_list-notify_cb = cb;
timer_list-notify_opaque = opaque;
@@ -140,13 +144,24 @@ void qemu_clock_notify(QEMUClockType type)
}
}
+/* Disabling the clock will wait for related timerlists to stop
+ * executing qemu_run_timers. Thus, this functions should not
+ * be used from the callback of a timer that is based on @clock.
+ * Doing so would cause a deadlock.
+ */
void qemu_clock_enable(QEMUClockType type, bool enabled)
{
QEMUClock *clock = qemu_clock_ptr(type);
+QEMUTimerList *tl;
bool old = clock-enabled;
clock-enabled = enabled;
if (enabled !old) {
qemu_clock_notify(type);
+} else if (!enabled old) {
+/* We rely on BQL to protect the timerlists */
+QLIST_FOREACH(tl, clock-timerlists, list) {
+qemu_event_wait(tl-ev);
+}
}
}
@@ -373,8 +388,10 @@ bool timerlist_run_timers(QEMUTimerList *timer_list)
QEMUTimer *ts;
int64_t current_time;
bool progress = false;
-
+
+qemu_event_reset(timer_list-ev);
if (!timer_list-clock-enabled) {
+qemu_event_set(timer_list-ev);
return progress;
}
@@ -392,6 +409,7 @@ bool timerlist_run_timers(QEMUTimerList *timer_list)
ts-cb(ts-opaque);
progress = true;
}
+qemu_event_set(timer_list-ev);
return progress;
}
--
1.8.1.4
[Qemu-devel] [PATCH v4 3/4] qemu-thread: add QemuEvent
This emulates Win32 manual-reset events using futexes or conditional
variables. Typical ways to use them are with multi-producer,
single-consumer data structures, to test for a complex condition whose
elements come from different threads:
for (;;) {
qemu_event_reset(ev);
... test complex condition ...
if (condition is true) {
break;
}
qemu_event_wait(ev);
}
Or more efficiently (but with some duplication):
... evaluate condition ...
while (!condition) {
qemu_event_reset(ev);
... evaluate condition ...
if (!condition) {
qemu_event_wait(ev);
... evaluate condition ...
}
}
QemuEvent provides a very fast userspace path in the common case when
no other thread is waiting, or the event is not changing state. It
is used to report RCU quiescent states to the thread calling
synchronize_rcu (the latter being the single consumer), and to report
call_rcu invocations to the thread that receives them.
Signed-off-by: Paolo Bonzini pbonz...@redhat.com
---
include/qemu/thread-posix.h | 8 +++
include/qemu/thread-win32.h | 4 ++
include/qemu/thread.h | 7 +++
util/qemu-thread-posix.c| 116
util/qemu-thread-win32.c| 26 ++
5 files changed, 161 insertions(+)
diff --git a/include/qemu/thread-posix.h b/include/qemu/thread-posix.h
index 361566a..eb5c7a1 100644
--- a/include/qemu/thread-posix.h
+++ b/include/qemu/thread-posix.h
@@ -21,6 +21,14 @@ struct QemuSemaphore {
#endif
};
+struct QemuEvent {
+#ifndef __linux__
+pthread_mutex_t lock;
+pthread_cond_t cond;
+#endif
+unsigned value;
+};
+
struct QemuThread {
pthread_t thread;
};
diff --git a/include/qemu/thread-win32.h b/include/qemu/thread-win32.h
index 13adb95..3d58081 100644
--- a/include/qemu/thread-win32.h
+++ b/include/qemu/thread-win32.h
@@ -17,6 +17,10 @@ struct QemuSemaphore {
HANDLE sema;
};
+struct QemuEvent {
+HANDLE event;
+};
+
typedef struct QemuThreadData QemuThreadData;
struct QemuThread {
QemuThreadData *data;
diff --git a/include/qemu/thread.h b/include/qemu/thread.h
index c02404b..3e32c65 100644
--- a/include/qemu/thread.h
+++ b/include/qemu/thread.h
@@ -7,6 +7,7 @@
typedef struct QemuMutex QemuMutex;
typedef struct QemuCond QemuCond;
typedef struct QemuSemaphore QemuSemaphore;
+typedef struct QemuEvent QemuEvent;
typedef struct QemuThread QemuThread;
#ifdef _WIN32
@@ -45,6 +46,12 @@ void qemu_sem_wait(QemuSemaphore *sem);
int qemu_sem_timedwait(QemuSemaphore *sem, int ms);
void qemu_sem_destroy(QemuSemaphore *sem);
+void qemu_event_init(QemuEvent *ev, bool init);
+void qemu_event_set(QemuEvent *ev);
+void qemu_event_reset(QemuEvent *ev);
+void qemu_event_wait(QemuEvent *ev);
+void qemu_event_destroy(QemuEvent *ev);
+
void qemu_thread_create(QemuThread *thread,
void *(*start_routine)(void *),
void *arg, int mode);
diff --git a/util/qemu-thread-posix.c b/util/qemu-thread-posix.c
index 4de133e..37dd298 100644
--- a/util/qemu-thread-posix.c
+++ b/util/qemu-thread-posix.c
@@ -20,7 +20,12 @@
#include limits.h
#include unistd.h
#include sys/time.h
+#ifdef __linux__
+#include sys/syscall.h
+#include linux/futex.h
+#endif
#include qemu/thread.h
+#include qemu/atomic.h
static void error_exit(int err, const char *msg)
{
@@ -272,6 +277,117 @@ void qemu_sem_wait(QemuSemaphore *sem)
#endif
}
+#ifdef __linux__
+#define futex(...) syscall(__NR_futex, __VA_ARGS__)
+
+static inline void futex_wake(QemuEvent *ev, int n)
+{
+futex(ev, FUTEX_WAKE, n, NULL, NULL, 0);
+}
+
+static inline void futex_wait(QemuEvent *ev, unsigned val)
+{
+futex(ev, FUTEX_WAIT, (int) val, NULL, NULL, 0);
+}
+#else
+static inline void futex_wake(QemuEvent *ev, int n)
+{
+if (n == 1) {
+pthread_cond_signal(ev-cond);
+} else {
+pthread_cond_broadcast(ev-cond);
+}
+}
+
+static inline void futex_wait(QemuEvent *ev, unsigned val)
+{
+pthread_mutex_lock(ev-lock);
+if (ev-value == val) {
+pthread_cond_wait(ev-cond, ev-lock);
+}
+pthread_mutex_unlock(ev-lock);
+}
+#endif
+
+/* Valid transitions:
+ * - free-set, when setting the event
+ * - busy-set, when setting the event, followed by futex_wake
+ * - set-free, when resetting the event
+ * - free-busy, when waiting
+ *
+ * set-busy does not happen (it can be observed from the outside but
+ * it really is set-free-busy).
+ *
+ * busy-free provably cannot happen; to enforce it, the set-free transition
+ * is done with an OR, which becomes a no-op if the event has concurrently
+ * transitioned to free or busy.
+ */
+
+#define EV_SET 0
+#define EV_FREE1
+#define EV_BUSY -1
+
+void qemu_event_init(QemuEvent *ev, bool init)
+{
+#ifndef __linux__
+pthread_mutex_init(ev-lock, NULL);
+pthread_cond_init(ev-cond, NULL);
+#endif
+
+ev-value =
Re: [Qemu-devel] in_asm substitute for accel=kvm:tcg
On Sun, Sep 22, 2013 at 11:05:37AM +0300, Andriy Gapon wrote: on 22/09/2013 09:31 Gleb Natapov said the following: Which kernel version is this? What BSD version? $ uname -a Linux kvm 3.8.0-27-generic #40-Ubuntu SMP Tue Jul 9 00:17:05 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux That's pretty old kernel and there were a lot fixes in emulation since. Before we spend more time tracking this can you verify that the bug is reproducible with 3.11? -- Gleb.
[Qemu-devel] [PATCH 1/2] tests: build the helper program in main build process
This is a quick way to update helper program when qemu main code is changed or built, instead of adding new Makefile under test/qemu-iotest. Signed-off-by: Wenchao Xia xiaw...@linux.vnet.ibm.com --- tests/Makefile |8 +++- 1 files changed, 7 insertions(+), 1 deletions(-) diff --git a/tests/Makefile b/tests/Makefile index 994fef1..5f2894f 100644 --- a/tests/Makefile +++ b/tests/Makefile @@ -252,8 +252,10 @@ check-report.html: check-report.xml # Other tests +QEMU_IOTESTS_HELPERS-$(CONFIG_LINUX) = tests/qemu-iotests/socket_scm_helper$(EXESUF) + .PHONY: check-tests/qemu-iotests-quick.sh -check-tests/qemu-iotests-quick.sh: tests/qemu-iotests-quick.sh qemu-img$(EXESUF) qemu-io$(EXESUF) tests/qemu-iotests/socket_scm_helper$(EXESUF) +check-tests/qemu-iotests-quick.sh: tests/qemu-iotests-quick.sh qemu-img$(EXESUF) qemu-io$(EXESUF) $(QEMU_IOTESTS_HELPERS-y) $ .PHONY: check-tests/test-qapi.py @@ -275,5 +277,9 @@ check-unit: $(patsubst %,check-%, $(check-unit-y)) check-block: $(patsubst %,check-%, $(check-block-y)) check: check-qapi-schema check-unit check-qtest +# Build the help program automatically + +all: $(QEMU_IOTESTS_HELPERS-y) + -include $(wildcard tests/*.d) -include $(wildcard tests/libqos/*.d) -- 1.7.1
[Qemu-devel] [PATCH 0/2] build: trivial patches for test
Wenchao Xia (2): tests: build the helper program in main build process build: add command check-clean Makefile |1 - tests/Makefile | 16 ++-- 2 files changed, 14 insertions(+), 3 deletions(-)
[Qemu-devel] [PATCH 2/2] build: add command check-clean
This command will package the clean operations in tests. Now root Makefile simply calls the command and do not care the details of it any more. Original the built binaries for test will not be removed, now they will be deleted in clean operation. Signed-off-by: Wenchao Xia xiaw...@linux.vnet.ibm.com --- Makefile |1 - tests/Makefile |8 +++- 2 files changed, 7 insertions(+), 2 deletions(-) diff --git a/Makefile b/Makefile index 362fe3e..befc23a 100644 --- a/Makefile +++ b/Makefile @@ -245,7 +245,6 @@ clean: rm -f $(foreach f,$(GENERATED_SOURCES),$(f) $(f)-timestamp) rm -rf qapi-generated rm -rf qga/qapi-generated - $(MAKE) -C tests/tcg clean for d in $(ALL_SUBDIRS); do \ if test -d $$d; then $(MAKE) -C $$d $@ || exit 1; fi; \ rm -f $$d/qemu-options.def; \ diff --git a/tests/Makefile b/tests/Makefile index 5f2894f..643c89c 100644 --- a/tests/Makefile +++ b/tests/Makefile @@ -196,6 +196,7 @@ check-help: @echo make check-qapi-schemaRun QAPI schema tests @echo make check-block Run block tests @echo make check-report.htmlGenerates an HTML test report + @echo make check-clean Clean the tests @echo @echo Please note that HTML reports do not regenerate if the unit tests @echo has not changed. @@ -270,12 +271,17 @@ $(patsubst %, check-%, $(check-qapi-schema-y)): check-%.json: $(SRC_PATH)/%.json # Consolidated targets -.PHONY: check-qapi-schema check-qtest check-unit check +.PHONY: check-qapi-schema check-qtest check-unit check check-clean check-qapi-schema: $(patsubst %,check-%, $(check-qapi-schema-y)) check-qtest: $(patsubst %,check-qtest-%, $(QTEST_TARGETS)) check-unit: $(patsubst %,check-%, $(check-unit-y)) check-block: $(patsubst %,check-%, $(check-block-y)) check: check-qapi-schema check-unit check-qtest +check-clean: + $(MAKE) -C tests/tcg clean + rm -rf $(check-unit-y) $(check-qtest-i386-y) $(check-qtest-x86_64-y) $(check-qtest-sparc64-y) $(check-qtest-sparc-y) tests/*.o $(QEMU_IOTESTS_HELPERS-y) + +clean: check-clean # Build the help program automatically -- 1.7.1
[Qemu-devel] [RFC] sync NIC's MAC maintained in NICConf as soon as emualted NIC's MAC changed in guest
Hi, all Do live migration if emulated NIC's MAC has been changed, RARP with wrong MAC address will broadcast via qemu_announce_self in destination, so, long time network disconnection probably happen. I want to do below works to resolve this problem, 1. change NICConf's MAC as soon as emulated NIC's MAC changed in guest 2. sync NIC's (more precisely, queue) MAC to corresponding NICConf in NIC's migration load handler Any better ideas? Thanks, Zhang Haoyu
Re: [Qemu-devel] [RFC] sync NIC's MAC maintained in NICConf as soon as emualted NIC's MAC changed in guest
On Sun, Sep 22, 2013 at 08:35:29AM +, Zhanghaoyu (A) wrote: Hi, all Do live migration if emulated NIC's MAC has been changed, RARP with wrong MAC address will broadcast via qemu_announce_self in destination, so, long time network disconnection probably happen. Good catch. I want to do below works to resolve this problem, 1. change NICConf's MAC as soon as emulated NIC's MAC changed in guest This will make it impossible to revert it correctly on reset, won't it? 2. sync NIC's (more precisely, queue) MAC to corresponding NICConf in NIC's migration load handler Any better ideas? Thanks, Zhang Haoyu I think announce needs to poke at the current MAC instead of the default one in NICConf. We can make it respect link down state while we are at it. Happily recent linux guests aren't affected since they do announcements from guest. -- MST
[Qemu-devel] [PATCH V2 0/4] export internal snapshot by qemu-nbd
This series allow user to read internal snapshot's contents without qemu-img convert. V2: Address Stefan's comments: 02: add 'fall through' comments in the case statement. 03: add doc about the difference of internal snapshot and backing chain snapshot, which is used in previous '--snapshot' parameter. Other: 01,04: rebased on upstream with conflict resolved. Wenchao Xia (4): 1 snapshot: distinguish id and name in load_tmp 2 qemu-nbd: support internal snapshot export 3 qemu-nbd: add doc for internal snapshot export 4 qemu-iotests: add 058 internal snapshot export with qemu-nbd case block/qcow2-snapshot.c | 16 +++- block/qcow2.h |5 ++- block/snapshot.c | 37 ++- include/block/block_int.h |4 ++- include/block/snapshot.h |4 ++- qemu-img.c | 14 ++- qemu-nbd.c | 62 ++- qemu-nbd.texi |8 - tests/qemu-iotests/058 | 87 tests/qemu-iotests/058.out | 26 + tests/qemu-iotests/group |1 + 11 files changed, 252 insertions(+), 12 deletions(-) create mode 100755 tests/qemu-iotests/058 create mode 100644 tests/qemu-iotests/058.out
[Qemu-devel] [PATCH V2 4/4] qemu-iotests: add 058 internal snapshot export with qemu-nbd case
Signed-off-by: Wenchao Xia xiaw...@linux.vnet.ibm.com
---
tests/qemu-iotests/058 | 87
tests/qemu-iotests/058.out | 26 +
tests/qemu-iotests/group |1 +
3 files changed, 114 insertions(+), 0 deletions(-)
create mode 100755 tests/qemu-iotests/058
create mode 100644 tests/qemu-iotests/058.out
diff --git a/tests/qemu-iotests/058 b/tests/qemu-iotests/058
new file mode 100755
index 000..301ef1f
--- /dev/null
+++ b/tests/qemu-iotests/058
@@ -0,0 +1,87 @@
+#!/bin/bash
+#
+# Test export internal snapshot by qemu-nbd.
+#
+# Copyright (C) 2013 IBM, Inc.
+#
+# Based on 029.
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program 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 General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program. If not, see http://www.gnu.org/licenses/.
+#
+
+# creator
+owner=xiaw...@linux.vnet.ibm.com
+
+seq=`basename $0`
+echo QA output created by $seq
+
+here=`pwd`
+tmp=/tmp/$$
+status=1 # failure is the default!
+
+nbd_snapshot_port=10850
+nbd_snapshot_img=nbd:127.0.0.1:$nbd_snapshot_port
+
+_export_nbd_snapshot()
+{
+eval $QEMU_NBD -v -t -b 127.0.0.1 -p $nbd_snapshot_port $TEST_IMG -l
name=$1
+NBD_SNAPSHOT_PID=$!
+sleep 1
+}
+
+_cleanup()
+{
+if [ -n $NBD_SNAPSHOT_PID ]; then
+kill $NBD_SNAPSHOT_PID
+fi
+ _cleanup_test_img
+}
+trap _cleanup; exit \$status 0 1 2 3 15
+
+# get standard environment, filters and checks
+. ./common.rc
+. ./common.filter
+. ./common.pattern
+
+# Any format supporting intenal snapshots
+_supported_fmt qcow2
+_supported_proto generic
+_supported_os Linux
+
+echo
+echo == preparing image ==
+_make_test_img 64M
+$QEMU_IO -c 'write -P 0xa 0x1000 0x1000' $TEST_IMG | _filter_qemu_io
+$QEMU_IO -c 'write -P 0xb 0x2000 0x1000' $TEST_IMG | _filter_qemu_io
+$QEMU_IMG snapshot -c foo $TEST_IMG
+$QEMU_IO -c 'write -P 0xc 0x1000 0x1000' $TEST_IMG | _filter_qemu_io
+$QEMU_IO -c 'write -P 0xd 0x2000 0x1000' $TEST_IMG | _filter_qemu_io
+_check_test_img
+
+echo
+echo == verifying the image file with patterns ==
+$QEMU_IO -c 'read -P 0xc 0x1000 0x1000' $TEST_IMG | _filter_qemu_io
+$QEMU_IO -c 'read -P 0xd 0x2000 0x1000' $TEST_IMG | _filter_qemu_io
+
+_export_nbd_snapshot foo
+
+echo
+echo == verifying the exported snapshot with patterns ==
+$QEMU_IO -c 'read -P 0xa 0x1000 0x1000' $nbd_snapshot_img | _filter_qemu_io
+$QEMU_IO -c 'read -P 0xb 0x2000 0x1000' $nbd_snapshot_img | _filter_qemu_io
+
+# success, all done
+echo *** done
+rm -f $seq.full
+status=0
diff --git a/tests/qemu-iotests/058.out b/tests/qemu-iotests/058.out
new file mode 100644
index 000..b174f3b
--- /dev/null
+++ b/tests/qemu-iotests/058.out
@@ -0,0 +1,26 @@
+QA output created by 058
+
+== preparing image ==
+Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=67108864
+wrote 4096/4096 bytes at offset 4096
+4 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+wrote 4096/4096 bytes at offset 8192
+4 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+wrote 4096/4096 bytes at offset 4096
+4 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+wrote 4096/4096 bytes at offset 8192
+4 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+No errors were found on the image.
+
+== verifying the image file with patterns ==
+read 4096/4096 bytes at offset 4096
+4 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 4096/4096 bytes at offset 8192
+4 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+== verifying the exported snapshot with patterns ==
+read 4096/4096 bytes at offset 4096
+4 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 4096/4096 bytes at offset 8192
+4 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+*** done
diff --git a/tests/qemu-iotests/group b/tests/qemu-iotests/group
index 1ad02e5..2793e1d 100644
--- a/tests/qemu-iotests/group
+++ b/tests/qemu-iotests/group
@@ -64,6 +64,7 @@
055 rw auto
056 rw auto backing
057 rw auto
+058 rw auto
059 rw auto
060 rw auto
061 rw auto
--
1.7.1
[Qemu-devel] [PATCH V2 1/4] snapshot: distinguish id and name in load_tmp
Since later this function will be used so improve it. The only caller of it
now is qemu-img, and it is not impacted by call the function twice to keep
old search logic.
Signed-off-by: Wenchao Xia xiaw...@linux.vnet.ibm.com
---
block/qcow2-snapshot.c| 16 ++--
block/qcow2.h |5 -
block/snapshot.c | 37 +++--
include/block/block_int.h |4 +++-
include/block/snapshot.h |4 +++-
qemu-img.c| 14 --
6 files changed, 71 insertions(+), 9 deletions(-)
diff --git a/block/qcow2-snapshot.c b/block/qcow2-snapshot.c
index 7d14420..fe8f0eb 100644
--- a/block/qcow2-snapshot.c
+++ b/block/qcow2-snapshot.c
@@ -669,7 +669,10 @@ int qcow2_snapshot_list(BlockDriverState *bs,
QEMUSnapshotInfo **psn_tab)
return s-nb_snapshots;
}
-int qcow2_snapshot_load_tmp(BlockDriverState *bs, const char *snapshot_name)
+int qcow2_snapshot_load_tmp(BlockDriverState *bs,
+const char *snapshot_id,
+const char *name,
+Error **errp)
{
int i, snapshot_index;
BDRVQcowState *s = bs-opaque;
@@ -677,12 +680,17 @@ int qcow2_snapshot_load_tmp(BlockDriverState *bs, const
char *snapshot_name)
uint64_t *new_l1_table;
int new_l1_bytes;
int ret;
+const char *device = bdrv_get_device_name(bs);
assert(bs-read_only);
/* Search the snapshot */
-snapshot_index = find_snapshot_by_id_or_name(bs, snapshot_name);
+snapshot_index = find_snapshot_by_id_and_name(bs, snapshot_id, name);
if (snapshot_index 0) {
+error_setg(errp,
+ Can't find a snapshot with ID '%s' and name '%s'
+ on device '%s',
+ STR_OR_NULL(snapshot_id), STR_OR_NULL(name), device);
return -ENOENT;
}
sn = s-snapshots[snapshot_index];
@@ -693,6 +701,10 @@ int qcow2_snapshot_load_tmp(BlockDriverState *bs, const
char *snapshot_name)
ret = bdrv_pread(bs-file, sn-l1_table_offset, new_l1_table,
new_l1_bytes);
if (ret 0) {
+error_setg(errp,
+ Failed to read l1 table for snapshot with ID '%s' and name
+ '%s' on device '%s',
+ sn-id_str, sn-name, device);
g_free(new_l1_table);
return ret;
}
diff --git a/block/qcow2.h b/block/qcow2.h
index c90e5d6..12cfeaf 100644
--- a/block/qcow2.h
+++ b/block/qcow2.h
@@ -472,7 +472,10 @@ int qcow2_snapshot_delete(BlockDriverState *bs,
const char *name,
Error **errp);
int qcow2_snapshot_list(BlockDriverState *bs, QEMUSnapshotInfo **psn_tab);
-int qcow2_snapshot_load_tmp(BlockDriverState *bs, const char *snapshot_name);
+int qcow2_snapshot_load_tmp(BlockDriverState *bs,
+const char *snapshot_id,
+const char *name,
+Error **errp);
void qcow2_free_snapshots(BlockDriverState *bs);
int qcow2_read_snapshots(BlockDriverState *bs);
diff --git a/block/snapshot.c b/block/snapshot.c
index a05c0c0..0dab05d 100644
--- a/block/snapshot.c
+++ b/block/snapshot.c
@@ -265,18 +265,51 @@ int bdrv_snapshot_list(BlockDriverState *bs,
return -ENOTSUP;
}
+/**
+ * Temporarily load an internal snapshot by @snapshot_id and @name.
+ * @bs: block device used in the operation
+ * @snapshot_id: unique snapshot ID, or NULL
+ * @name: snapshot name, or NULL
+ * @errp: location to store error
+ *
+ * If both @snapshot_id and @name are specified, load the first one with
+ * id @snapshot_id and name @name.
+ * If only @snapshot_id is specified, load the first one with id
+ * @snapshot_id.
+ * If only @name is specified, load the first one with name @name.
+ * if none is specified, return -ENINVAL.
+ *
+ * Returns: 0 on success, -errno on fail. If @bs is not inserted, return
+ * -ENOMEDIUM. If @bs is not readonly, return -EINVAL. If @bs did not support
+ * internal snapshot, return -ENOTSUP. If qemu can't find one matching @id and
+ * @name, return -ENOENT. If @bs do not support parameter @snapshot_id or
+ * @name, return -EINVAL. If @errp != NULL, it will always be filled on
+ * failure.
+ */
int bdrv_snapshot_load_tmp(BlockDriverState *bs,
-const char *snapshot_name)
+ const char *snapshot_id,
+ const char *name,
+ Error **errp)
{
BlockDriver *drv = bs-drv;
+const char *device = bdrv_get_device_name(bs);
if (!drv) {
+error_set(errp, QERR_DEVICE_HAS_NO_MEDIUM, device);
return -ENOMEDIUM;
}
+if (!snapshot_id !name) {
+error_setg(errp, snapshot_id and name are both NULL);
+return -EINVAL;
+}
if (!bs-read_only) {
+error_setg(errp, Device '%s' is not readonly, device);
return -EINVAL;
}
if (drv-bdrv_snapshot_load_tmp) {
-return
[Qemu-devel] [PATCH V2 3/4] qemu-nbd: add doc for internal snapshot export
Signed-off-by: Wenchao Xia xiaw...@linux.vnet.ibm.com
---
qemu-nbd.c|8 +++-
qemu-nbd.texi |8 +++-
2 files changed, 14 insertions(+), 2 deletions(-)
diff --git a/qemu-nbd.c b/qemu-nbd.c
index e450d04..8cb4bf1 100644
--- a/qemu-nbd.c
+++ b/qemu-nbd.c
@@ -80,7 +80,13 @@ static void usage(const char *name)
\n
Block device options:\n
-r, --read-only export read-only\n
- -s, --snapshot use snapshot file\n
+ -s, --snapshot use FILE as an external snapshot, create a temporary\n
+ file with backing_file=FILE, redirect the write to\n
+ the temporary one\n
+ -l, --snapshot-load=PARAM\n
+ load an internal snapshot inside FILE and export it\n
+ as an read-only device, PARAM format is\n
+ 'id=[ID],name=[NAME]'\n
-n, --nocachedisable host cache\n
--cache=MODE set cache mode (none, writeback, ...)\n
#ifdef CONFIG_LINUX_AIO
diff --git a/qemu-nbd.texi b/qemu-nbd.texi
index 6055ec6..69e9e9a 100644
--- a/qemu-nbd.texi
+++ b/qemu-nbd.texi
@@ -27,7 +27,13 @@ Export QEMU disk image using NBD protocol.
@item -P, --partition=@var{num}
only expose partition @var{num}
@item -s, --snapshot
- use snapshot file
+ use @var{filename} as an external snapshot, create a temporary
+ file with backing_file=@var{filename}, redirect the write to
+ the temporary one
+@item -l, --snapshot-load=@var{param}
+ load an internal snapshot inside @var{filename} and export it
+ as an read-only device, @var{param} format is
+ 'id=[ID],name=[NAME]'
@item -n, --nocache
@itemx --cache=@var{cache}
set cache mode to be used with the file. See the documentation of
--
1.7.1
[Qemu-devel] [PATCH V2 2/4] qemu-nbd: support internal snapshot export
Now it is possible to directly export an internal snapshot, which
can be used to probe the snapshot's contents without qemu-img
convert.
Signed-off-by: Wenchao Xia xiaw...@linux.vnet.ibm.com
---
qemu-nbd.c | 54 +-
1 files changed, 53 insertions(+), 1 deletions(-)
diff --git a/qemu-nbd.c b/qemu-nbd.c
index c26c98e..e450d04 100644
--- a/qemu-nbd.c
+++ b/qemu-nbd.c
@@ -20,6 +20,7 @@
#include block/block.h
#include block/nbd.h
#include qemu/main-loop.h
+#include block/snapshot.h
#include stdarg.h
#include stdio.h
@@ -304,6 +305,23 @@ static void nbd_accept(void *opaque)
}
}
+#define SNAPSHOT_OPT_ID id
+#define SNAPSHOT_OPT_NAME name
+
+static QEMUOptionParameter snapshot_options[] = {
+{
+.name = SNAPSHOT_OPT_ID,
+.type = OPT_STRING,
+.help = snapshot id
+},
+{
+.name = SNAPSHOT_OPT_NAME,
+.type = OPT_STRING,
+.help = snapshot name
+},
+{ NULL }
+};
+
int main(int argc, char **argv)
{
BlockDriverState *bs;
@@ -315,7 +333,10 @@ int main(int argc, char **argv)
char *device = NULL;
int port = NBD_DEFAULT_PORT;
off_t fd_size;
-const char *sopt = hVb:o:p:rsnP:c:dvk:e:f:t;
+QEMUOptionParameter *sn_param = NULL;
+const QEMUOptionParameter *sn_param_id, *sn_param_name;
+const char *sn_id = NULL, *sn_name = NULL;
+const char *sopt = hVb:o:p:rsnP:c:dvk:e:f:tl:;
struct option lopt[] = {
{ help, 0, NULL, 'h' },
{ version, 0, NULL, 'V' },
@@ -328,6 +349,7 @@ int main(int argc, char **argv)
{ connect, 1, NULL, 'c' },
{ disconnect, 0, NULL, 'd' },
{ snapshot, 0, NULL, 's' },
+{ snapshot-load, 1, NULL, 'l' },
{ nocache, 0, NULL, 'n' },
{ cache, 1, NULL, QEMU_NBD_OPT_CACHE },
#ifdef CONFIG_LINUX_AIO
@@ -428,6 +450,14 @@ int main(int argc, char **argv)
errx(EXIT_FAILURE, Offset must be positive `%s', optarg);
}
break;
+case 'l':
+sn_param = parse_option_parameters(optarg,
+ snapshot_options, sn_param);
+if (!sn_param) {
+errx(EXIT_FAILURE,
+ Invalid snapshot-load options '%s', optarg);
+}
+/* fall through */
case 'r':
nbdflags |= NBD_FLAG_READ_ONLY;
flags = ~BDRV_O_RDWR;
@@ -581,6 +611,24 @@ int main(int argc, char **argv)
error_get_pretty(local_err));
}
+if (sn_param) {
+sn_param_id = get_option_parameter(sn_param, SNAPSHOT_OPT_ID);
+sn_param_name = get_option_parameter(sn_param, SNAPSHOT_OPT_NAME);
+if (sn_param_id) {
+sn_id = sn_param_id-value.s;
+}
+if (sn_param_name) {
+sn_name = sn_param_name-value.s;
+}
+ret = bdrv_snapshot_load_tmp(bs, sn_id, sn_name, local_err);
+if (ret 0) {
+errno = -ret;
+err(EXIT_FAILURE,
+Failed to load snapshot, reason:\n%s,
+error_get_pretty(local_err));
+}
+}
+
fd_size = bdrv_getlength(bs);
if (partition != -1) {
@@ -641,6 +689,10 @@ int main(int argc, char **argv)
unlink(sockpath);
}
+if (sn_param) {
+free_option_parameters(sn_param);
+}
+
if (device) {
void *ret;
pthread_join(client_thread, ret);
--
1.7.1
Re: [Qemu-devel] in_asm substitute for accel=kvm:tcg
on 22/09/2013 11:17 Gleb Natapov said the following: On Sun, Sep 22, 2013 at 11:05:37AM +0300, Andriy Gapon wrote: on 22/09/2013 09:31 Gleb Natapov said the following: Which kernel version is this? What BSD version? $ uname -a Linux kvm 3.8.0-27-generic #40-Ubuntu SMP Tue Jul 9 00:17:05 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux That's pretty old kernel and there were a lot fixes in emulation since. Before we spend more time tracking this can you verify that the bug is reproducible with 3.11? Gleb, thank you very much -- right on the spot! With 3.11 the boot proceeds past that point. -- Andriy Gapon
Re: [Qemu-devel] [SeaBIOS] [PATCH v2 1/5] linker: utility to patch in-memory ROM files
On Thu, Jul 25, 2013 at 08:06:27PM -0400, Kevin O'Connor wrote:
On Thu, Jul 25, 2013 at 03:55:56PM +0300, Michael S. Tsirkin wrote:
On Mon, Jul 15, 2013 at 11:01:02AM +0300, Michael S. Tsirkin wrote:
On Sun, Jul 14, 2013 at 02:24:52PM -0400, Kevin O'Connor wrote:
I'd prefer to see this tracked within the linker code and not in the
generic romfile struct.
A way to associate a romfile instance with a value seems generally
useful, no? Still, that's not too hard - it would only mean an extra
linked list of
struct linker {
char name[56]
void *data;
struct hlist_node node;
}
is this preferable?
Sure, but it's probably easier to do something like:
struct linkfiles { char *name; void *data; };
void linker_loader_execute(const char *name)
{
int size;
struct linker_loader_entry_s *entries = romfile_loadfile(name, size);
int numentries = size/sizeof(entries[0]);
if (! entries)
return;
struct linkfiles *files = malloc_tmp(sizeof(files[0]) * numentries);
and then just populate and use the array of filenames.
OK I'll do this but it's more code as I can't use plain romfile_find
anymore, and have to code up my own lookup.
Also, is there another name besides linker that could be used?
SeaBIOS has code to self-relocate and fixup code relocations. I think
having code in the repo called linker could cause confusion.
romfile_loader?
Shrug. How about tabledeploy?
-Kevin
Re: [Qemu-devel] [SeaBIOS] [PATCH v2 1/5] linker: utility to patch in-memory ROM files
On Sun, Sep 22, 2013 at 01:49:58PM +0300, Michael S. Tsirkin wrote:
On Thu, Jul 25, 2013 at 08:06:27PM -0400, Kevin O'Connor wrote:
On Thu, Jul 25, 2013 at 03:55:56PM +0300, Michael S. Tsirkin wrote:
On Mon, Jul 15, 2013 at 11:01:02AM +0300, Michael S. Tsirkin wrote:
On Sun, Jul 14, 2013 at 02:24:52PM -0400, Kevin O'Connor wrote:
I'd prefer to see this tracked within the linker code and not in the
generic romfile struct.
A way to associate a romfile instance with a value seems generally
useful, no? Still, that's not too hard - it would only mean an extra
linked list of
struct linker {
char name[56]
void *data;
struct hlist_node node;
}
is this preferable?
Sure, but it's probably easier to do something like:
struct linkfiles { char *name; void *data; };
void linker_loader_execute(const char *name)
{
int size;
struct linker_loader_entry_s *entries = romfile_loadfile(name, size);
int numentries = size/sizeof(entries[0]);
if (! entries)
return;
struct linkfiles *files = malloc_tmp(sizeof(files[0]) * numentries);
and then just populate and use the array of filenames.
OK I'll do this but it's more code as I can't use plain romfile_find
anymore, and have to code up my own lookup.
Also, is there another name besides linker that could be used?
SeaBIOS has code to self-relocate and fixup code relocations. I think
having code in the repo called linker could cause confusion.
romfile_loader?
Shrug. How about tabledeploy?
-Kevin
So I tried this out, and I don't like this much:
see below.
Lots of code, a single data pointer in file struct seems much easier.
Further, without a pointer from file to data,
there's no way to find the tables in memory,
so I will have to make the loader interface ACPI-specific,
make it look into tables loaded for the RSDP signature
and return the address of RSDP.
diff --git a/src/util.h b/src/util.h
index 1e883f2..d777521 100644
--- a/src/util.h
+++ b/src/util.h
@@ -441,7 +441,6 @@ struct romfile_s {
char name[128];
u32 size;
int (*copy)(struct romfile_s *file, void *dest, u32 maxlen);
-void *data;
};
void romfile_add(struct romfile_s *file);
struct romfile_s *romfile_findprefix(const char *prefix, struct romfile_s
*prev);
diff --git a/src/acpi.c b/src/acpi.c
index 24cb1fa..c3a3c16 100644
--- a/src/acpi.c
+++ b/src/acpi.c
@@ -611,14 +611,14 @@ acpi_find_rsdp_rom(void)
if (!file)
break;
-if (!file-data || !pmm_test_fseg(file-data) ||
+if (/*!file-data ||*/ !pmm_test_fseg(NULL /*file-data*/) ||
file-size sizeof(rsdp-signature))
continue;
void *data;
-for (data = file-data;
- data + sizeof(*rsdp) = file-data + file-size;
+for (data = NULL /*file-data */;
+ data + sizeof(*rsdp) = /* file-data */ NULL + file-size;
data++) {
rsdp = data;
if (rsdp-signature == cpu_to_le64(RSDP_SIGNATURE))
diff --git a/src/romfile_loader.c b/src/romfile_loader.c
index 6ba03ed..5e98810 100644
--- a/src/romfile_loader.c
+++ b/src/romfile_loader.c
@@ -2,17 +2,33 @@
#include byteorder.h // leXX_to_cpu/cpu_to_leXX
#include util.h // checksum
-static struct romfile_s *romfile_loader_find(const char *name)
+struct romfile_loader_file {
+struct romfile_s *file;
+void *data;
+};
+struct romfile_loader_files {
+int nfiles;
+struct romfile_loader_file files[];
+};
+
+static struct romfile_loader_file *
+romfile_loader_find(const char *name,
+struct romfile_loader_files *files)
{
+int i;
if (name[ROMFILE_LOADER_FILESZ - 1])
return NULL;
-return romfile_find(name);
+for (i = 0; i files-nfiles; ++i)
+if (!strcmp(files-files[i].file-name, name))
+return files-files[i];
+return NULL;
}
-static void romfile_loader_allocate(struct romfile_loader_entry_s *entry)
+static void romfile_loader_allocate(struct romfile_loader_entry_s *entry,
+struct romfile_loader_files *files)
{
struct zone_s *zone;
-struct romfile_s *file;
+struct romfile_loader_file *file = files-files[files-nfiles];
void *data;
int ret;
unsigned alloc_align = le32_to_cpu(entry-alloc_align);
@@ -32,20 +48,21 @@ static void romfile_loader_allocate(struct
romfile_loader_entry_s *entry)
}
if (alloc_align MALLOC_MIN_ALIGN)
alloc_align = MALLOC_MIN_ALIGN;
-file = romfile_loader_find(entry-alloc_file);
-if (!file || file-data)
+if (entry-alloc_file[ROMFILE_LOADER_FILESZ - 1])
goto err;
-if (!file-size)
+file-file = romfile_find(entry-alloc_file);
+if (!file-file || !file-file-size)
return;
-data = pmm_malloc(zone, PMM_DEFAULT_HANDLE, file-size, alloc_align);
+data
[Qemu-devel] [PATCH] sPAPR: implement route_intx_to_irq to get gsi of pci device.
This is useful when pci assignment happens on sPAPR.
Signed-off-by: Liu Ping Fan pingf...@linux.vnet.ibm.com
---
This patch will apply on patches which enable xics in kernel.
---
hw/intc/xics.c| 5 +
hw/ppc/spapr_pci.c| 14 ++
include/hw/ppc/xics.h | 1 +
3 files changed, 20 insertions(+)
diff --git a/hw/intc/xics.c b/hw/intc/xics.c
index bb018d1..02cdab8 100644
--- a/hw/intc/xics.c
+++ b/hw/intc/xics.c
@@ -442,6 +442,11 @@ void xics_set_irq_type(XICSState *icp, int irq, bool lsi)
icp-ics-islsi[irq - icp-ics-offset] = lsi;
}
+int xics_get_irq_offset(XICSState *icp)
+{
+return icp-ics-offset;
+}
+
/*
* Guest interfaces
*/
diff --git a/hw/ppc/spapr_pci.c b/hw/ppc/spapr_pci.c
index 9b6ee32..6d3657a 100644
--- a/hw/ppc/spapr_pci.c
+++ b/hw/ppc/spapr_pci.c
@@ -432,6 +432,19 @@ static void pci_spapr_set_irq(void *opaque, int irq_num,
int level)
qemu_set_irq(spapr_phb_lsi_qirq(phb, irq_num), level);
}
+static PCIINTxRoute spapr_phb_route_intx_to_irq(void *opaque, int pirq_pin)
+{
+int gsi;
+PCIINTxRoute route;
+sPAPRPHBState *phb = opaque;
+
+gsi = phb-lsi_table[pirq_pin].irq;
+gsi += xics_get_irq_offset(spapr-icp);
+route.mode = PCI_INTX_ENABLED;
+route.irq = gsi;
+return route;
+}
+
/*
* MSI/MSIX memory region implementation.
* The handler handles both MSI and MSIX.
@@ -595,6 +608,7 @@ static int spapr_phb_init(SysBusDevice *s)
pci_spapr_set_irq, pci_spapr_map_irq, sphb,
sphb-memspace, sphb-iospace,
PCI_DEVFN(0, 0), PCI_NUM_PINS, TYPE_PCI_BUS);
+pci_bus_set_route_irq_fn(bus, spapr_phb_route_intx_to_irq);
phb-bus = bus;
sphb-dma_window_start = 0;
diff --git a/include/hw/ppc/xics.h b/include/hw/ppc/xics.h
index 66364c5..6ed1f4d 100644
--- a/include/hw/ppc/xics.h
+++ b/include/hw/ppc/xics.h
@@ -97,6 +97,7 @@ struct ICSIRQState {
qemu_irq xics_get_qirq(XICSState *icp, int irq);
void xics_set_irq_type(XICSState *icp, int irq, bool lsi);
+int xics_get_irq_offset(XICSState *icp);
void xics_cpu_setup(XICSState *icp, PowerPCCPU *cpu);
--
1.8.1.4
[Qemu-devel] [PATCH 1/2] block: fix backing file overriding
Providing backing.file.filename doesn't override backing file as expected:
$ x86_64-softmmu/qemu-system-x86_64 -drive \
file=/tmp/child.qcow2,backing.file.filename=/tmp/fake.qcow2
qemu-system-x86_64: -drive \
file=/tmp/child.qcow2,backing.file.filename=/tmp/fake.qcow2: could not
open disk image /tmp/child.qcow2: Can't specify 'file' and 'filename'
options at the same time
With
$ qemu-img info /tmp/child.qcow2
image: /tmp/child.qcow2
file format: qcow2
virtual size: 1.0G (1073741824 bytes)
disk size: 196K
cluster_size: 65536
backing file: /tmp/fake.qcow2
This fixes it by calling bdrv_get_full_backing_filename only if
backing.file.filename is not provided. Also save the backing file name
to bs-backing_file so the information is correct with HMP info block.
Signed-off-by: Fam Zheng f...@redhat.com
---
block.c | 7 +--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/block.c b/block.c
index e176c6f..a2ea39a 100644
--- a/block.c
+++ b/block.c
@@ -978,11 +978,12 @@ int bdrv_open_backing_file(BlockDriverState *bs, QDict
*options, Error **errp)
} else if (bs-backing_file[0] == '\0' qdict_size(options) == 0) {
QDECREF(options);
return 0;
+} else {
+bdrv_get_full_backing_filename(bs, backing_filename,
+ sizeof(backing_filename));
}
bs-backing_hd = bdrv_new();
-bdrv_get_full_backing_filename(bs, backing_filename,
- sizeof(backing_filename));
if (bs-backing_format[0] != '\0') {
back_drv = bdrv_find_format(bs-backing_format);
@@ -994,6 +995,8 @@ int bdrv_open_backing_file(BlockDriverState *bs, QDict
*options, Error **errp)
ret = bdrv_open(bs-backing_hd,
*backing_filename ? backing_filename : NULL, options,
back_flags, back_drv, local_err);
+pstrcpy(bs-backing_file, sizeof(bs-backing_file),
+bs-backing_hd-file-filename);
if (ret 0) {
bdrv_unref(bs-backing_hd);
bs-backing_hd = NULL;
--
1.8.3.1
[Qemu-devel] [PATCH 2/2] qemu-iotests: add test for backing file overriding
Test that backing.file.filename option can be parsed and override the
backing file from image (backing file reflected with info block).
Signed-off-by: Fam Zheng f...@redhat.com
---
tests/qemu-iotests/051 | 17 -
tests/qemu-iotests/051.out | 11 +++
2 files changed, 27 insertions(+), 1 deletion(-)
diff --git a/tests/qemu-iotests/051 b/tests/qemu-iotests/051
index 1f39c6a..78e1182 100755
--- a/tests/qemu-iotests/051
+++ b/tests/qemu-iotests/051
@@ -45,7 +45,14 @@ _supported_os Linux
function do_run_qemu()
{
echo Testing: $@
-echo quit | $QEMU -nographic -monitor stdio -serial none $@
+(
+if ! test -t 0; then
+while read cmd; do
+echo $cmd
+done
+fi
+echo quit
+) | $QEMU -nographic -monitor stdio -serial none $@
echo
}
@@ -57,6 +64,9 @@ function run_qemu()
size=128M
_make_test_img $size
+cp $TEST_IMG $TEST_IMG.orig
+mv $TEST_IMG $TEST_IMG.base
+_make_test_img -b $TEST_IMG.base $size
echo
echo === Unknown option ===
@@ -67,6 +77,11 @@ run_qemu -drive file=$TEST_IMG,format=qcow2,unknown_opt=on
run_qemu -drive file=$TEST_IMG,format=qcow2,unknown_opt=1234
run_qemu -drive file=$TEST_IMG,format=qcow2,unknown_opt=foo
+echo
+echo === Overriding backing file ===
+echo
+
+echo info block | run_qemu -drive
file=$TEST_IMG,driver=qcow2,backing.file.filename=$TEST_IMG.orig -nodefaults
echo
echo === Enable and disable lazy refcounting on the command line, plus some
invalid values ===
diff --git a/tests/qemu-iotests/051.out b/tests/qemu-iotests/051.out
index 88e8fa7..335fbe3 100644
--- a/tests/qemu-iotests/051.out
+++ b/tests/qemu-iotests/051.out
@@ -1,5 +1,6 @@
QA output created by 051
Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
+Formatting 'TEST_DIR/t.IMGFMT', fmt=IMGFMT size=134217728
backing_file='TEST_DIR/t.IMGFMT.base'
=== Unknown option ===
@@ -16,6 +17,16 @@ Testing: -drive
file=TEST_DIR/t.qcow2,format=qcow2,unknown_opt=foo
QEMU_PROG: -drive file=TEST_DIR/t.qcow2,format=qcow2,unknown_opt=foo: could
not open disk image TEST_DIR/t.qcow2: Block format 'qcow2' used by device
'ide0-hd0' doesn't support the option 'unknown_opt'
+=== Overriding backing file ===
+
+Testing: -drive
file=TEST_DIR/t.qcow2,driver=qcow2,backing.file.filename=TEST_DIR/t.qcow2.orig
-nodefaults
+QEMU X.Y.Z monitor - type 'help' for more information
+(qemu) i[K[Din[K[D[Dinf[K[D[D[Dinfo[K[D[D[D[Dinfo
[K[D[D[D[D[Dinfo b[K[D[D[D[D[D[Dinfo
bl[K[D[D[D[D[D[D[Dinfo blo[K[D[D[D[D[D[D[D[Dinfo
bloc[K[D[D[D[D[D[D[D[D[Dinfo block[K
+ide0-hd0: TEST_DIR/t.qcow2 (qcow2)
+Backing file: TEST_DIR/t.qcow2.orig (chain depth: 1)
+ [not inserted](qemu) q[K[Dqu[K[D[Dqui[K[D[D[Dquit[K
+
+
=== Enable and disable lazy refcounting on the command line, plus some invalid
values ===
Testing: -drive file=TEST_DIR/t.qcow2,format=qcow2,lazy-refcounts=on
--
1.8.3.1
[Qemu-devel] [PATCH 0/2] block: fix backing file overriding
The backing.file.filename option is not working as expected: if there's also a backing file name from the format driver, adding this option fails bdrv_open; if there's no backing file name info in the image, info block doesn't show the overrided file name. A test case is updated to catch these issues. Fam Zheng (2): block: fix backing file overriding qemu-iotests: add test for backing file overriding block.c| 7 +-- tests/qemu-iotests/051 | 17 - tests/qemu-iotests/051.out | 11 +++ 3 files changed, 32 insertions(+), 3 deletions(-) -- 1.8.3.1
[Qemu-devel] [PATCH] .gitignore: ignore tests/qemu-iotests/socket_scm_helper
Signed-off-by: Fam Zheng f...@redhat.com --- tests/qemu-iotests/.gitignore | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/qemu-iotests/.gitignore b/tests/qemu-iotests/.gitignore index 62b4002..0541f80 100644 --- a/tests/qemu-iotests/.gitignore +++ b/tests/qemu-iotests/.gitignore @@ -2,6 +2,7 @@ check.log check.time *.out.bad *.notrun +socket_scm_helper # ignore everything in the scratch directory scratch/ -- 1.8.3.1
[Qemu-devel] [PATCH v4 05/23] fw_cfg: interface to trigger callback on read
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
include/hw/nvram/fw_cfg.h | 4
hw/nvram/fw_cfg.c | 33 -
2 files changed, 32 insertions(+), 5 deletions(-)
diff --git a/include/hw/nvram/fw_cfg.h b/include/hw/nvram/fw_cfg.h
index f60dd67..2ab0fc2 100644
--- a/include/hw/nvram/fw_cfg.h
+++ b/include/hw/nvram/fw_cfg.h
@@ -60,6 +60,7 @@ typedef struct FWCfgFiles {
} FWCfgFiles;
typedef void (*FWCfgCallback)(void *opaque, uint8_t *data);
+typedef void (*FWCfgReadCallback)(void *opaque, uint32_t offset);
void fw_cfg_add_bytes(FWCfgState *s, uint16_t key, void *data, size_t len);
void fw_cfg_add_string(FWCfgState *s, uint16_t key, const char *value);
@@ -70,6 +71,9 @@ void fw_cfg_add_callback(FWCfgState *s, uint16_t key,
FWCfgCallback callback,
void *callback_opaque, void *data, size_t len);
void fw_cfg_add_file(FWCfgState *s, const char *filename, void *data,
size_t len);
+void fw_cfg_add_file_callback(FWCfgState *s, const char *filename,
+ FWCfgReadCallback callback, void
*callback_opaque,
+ void *data, size_t len);
FWCfgState *fw_cfg_init(uint32_t ctl_port, uint32_t data_port,
hwaddr crl_addr, hwaddr data_addr);
diff --git a/hw/nvram/fw_cfg.c b/hw/nvram/fw_cfg.c
index d0820e5..f5dc3ea 100644
--- a/hw/nvram/fw_cfg.c
+++ b/hw/nvram/fw_cfg.c
@@ -42,6 +42,7 @@ typedef struct FWCfgEntry {
uint8_t *data;
void *callback_opaque;
FWCfgCallback callback;
+FWCfgReadCallback read_callback;
} FWCfgEntry;
struct FWCfgState {
@@ -249,8 +250,12 @@ static uint8_t fw_cfg_read(FWCfgState *s)
if (s-cur_entry == FW_CFG_INVALID || !e-data || s-cur_offset = e-len)
ret = 0;
-else
+else {
+if (e-read_callback) {
+e-read_callback(e-callback_opaque, s-cur_offset);
+}
ret = e-data[s-cur_offset++];
+}
trace_fw_cfg_read(s, ret);
return ret;
@@ -381,7 +386,10 @@ static const VMStateDescription vmstate_fw_cfg = {
}
};
-void fw_cfg_add_bytes(FWCfgState *s, uint16_t key, void *data, size_t len)
+static void fw_cfg_add_bytes_read_callback(FWCfgState *s, uint16_t key,
+ FWCfgReadCallback callback,
+ void *callback_opaque,
+ void *data, size_t len)
{
int arch = !!(key FW_CFG_ARCH_LOCAL);
@@ -391,6 +399,13 @@ void fw_cfg_add_bytes(FWCfgState *s, uint16_t key, void
*data, size_t len)
s-entries[arch][key].data = data;
s-entries[arch][key].len = (uint32_t)len;
+s-entries[arch][key].read_callback = callback;
+s-entries[arch][key].callback_opaque = callback_opaque;
+}
+
+void fw_cfg_add_bytes(FWCfgState *s, uint16_t key, void *data, size_t len)
+{
+fw_cfg_add_bytes_read_callback(s, key, NULL, NULL, data, len);
}
void fw_cfg_add_string(FWCfgState *s, uint16_t key, const char *value)
@@ -444,8 +459,9 @@ void fw_cfg_add_callback(FWCfgState *s, uint16_t key,
FWCfgCallback callback,
s-entries[arch][key].callback = callback;
}
-void fw_cfg_add_file(FWCfgState *s, const char *filename,
- void *data, size_t len)
+void fw_cfg_add_file_callback(FWCfgState *s, const char *filename,
+ FWCfgReadCallback callback, void
*callback_opaque,
+ void *data, size_t len)
{
int i, index;
size_t dsize;
@@ -459,7 +475,8 @@ void fw_cfg_add_file(FWCfgState *s, const char *filename,
index = be32_to_cpu(s-files-count);
assert(index FW_CFG_FILE_SLOTS);
-fw_cfg_add_bytes(s, FW_CFG_FILE_FIRST + index, data, len);
+fw_cfg_add_bytes_read_callback(s, FW_CFG_FILE_FIRST + index,
+ callback, callback_opaque, data, len);
pstrcpy(s-files-f[index].name, sizeof(s-files-f[index].name),
filename);
@@ -477,6 +494,12 @@ void fw_cfg_add_file(FWCfgState *s, const char *filename,
s-files-count = cpu_to_be32(index+1);
}
+void fw_cfg_add_file(FWCfgState *s, const char *filename,
+ void *data, size_t len)
+{
+fw_cfg_add_file_callback(s, filename, NULL, NULL, data, len);
+}
+
static void fw_cfg_machine_ready(struct Notifier *n, void *data)
{
size_t len;
--
MST
[Qemu-devel] [PATCH v4 00/23] qemu: generate acpi tables for the guest
This code can also be found here: git://git.kernel.org/pub/scm/virt/kvm/mst/qemu.git acpi While this patch still uses info not available in QOM, I think it's reasonable to merge it and then refactor as QOM properties cover more ground. In particular, merging this patchset blocks other projects so I think its preferable to merge now and not wait for all required QOM properties to materialize. I added QOM properties in ich/piix where I knew how to do this. If everything's in order, I intend to merge this through my tree. Please review, and comment. Changes from v3: - reworked code to use QOM properties some info isn't yet available in QOM, use old-style APIs and lookups by type - address comments by Gerd: tables are now updated on guest access after pci configuration Changes from v2 repost: - address comment by Anthony - convert to use APIs implemented using QOM - address comment by Anthony - avoid tricky pointer path, use GArray from glib instead - Address lots of comments by Hu Tao and Laszlo Ersek Changes from v2: - added missing patches to make it actually build Changes from v1 RFC: - added code to address cross version compatibility - rebased to latest bits - updated seabios code to latest bits (added pvpanic device) This patchset moves all generation of ACPI tables from guest BIOS to the hypervisor. Although ACPI tables come from a system BIOS on real hw, it makes sense that the ACPI tables are coupled with the virtual machine, since they have to abstract the x86 machine to the OS's. This is widely desired as a way to avoid the churn and proliferation of QEMU-specific interfaces associated with ACPI tables in bios code. There's a bit of code duplication where we already declare similar acpi structures in qemu. I think it's best to do it in this order: port code directly, and apply cleanups and reduce duplication that results, on top. This way it's much easier to see that we don't introduce regressions. In particular, I booted a guest on qemu with and without the change, and verified that ACPI tables are unchanged except for trivial pointer address changes. Such binary compatibility makes it easier to be confident that this change won't break things. Michael S. Tsirkin (23): qemu: add Error to typedefs qom: pull in qemu/typedefs qom: cleanup struct Error references qom: add pointer to int property helpers fw_cfg: interface to trigger callback on read loader: support for unmapped ROM blobs pcie_host: expose UNMAPPED macro pcie_host: expose address format q35: use macro for MCFG property name q35: expose mmcfg size as a property i386: add ACPI table files from seabios acpi: add rules to compile ASL source acpi: pre-compiled ASL files loader: use file path size from fw_cfg.h i386: add bios linker/loader loader: allow adding ROMs in done callbacks i386: define pc guest info acpi/piix: add macros for acpi property names piix: APIs for pc guest info ich9: APIs for pc guest info pvpanic: add API to access io port hpet: add API to find it i386: ACPI table generation code from seabios configure|9 +- hw/i386/acpi-defs.h | 331 ++ hw/lm32/lm32_hwsetup.h |2 +- include/hw/acpi/ich9.h |2 + include/hw/acpi/piix4.h |8 + include/hw/i386/acpi-build.h |9 + include/hw/i386/bios-linker-loader.h | 26 + include/hw/i386/ich9.h |2 + include/hw/i386/pc.h | 23 + include/hw/loader.h |8 +- include/hw/nvram/fw_cfg.h|8 +- include/hw/pci-host/q35.h|2 + include/hw/pci/pcie_host.h | 27 + include/hw/timer/hpet.h |2 + include/qemu/typedefs.h |1 + include/qom/object.h | 73 +- hw/acpi/ich9.c | 24 + hw/acpi/piix4.c | 50 +- hw/core/loader.c | 31 +- hw/i386/acpi-build.c | 1190 ++ hw/i386/bios-linker-loader.c | 156 + hw/i386/pc.c | 37 + hw/i386/pc_piix.c|5 + hw/i386/pc_q35.c |3 + hw/isa/lpc_ich9.c| 40 + hw/misc/pvpanic.c| 13 +- hw/nvram/fw_cfg.c| 33 +- hw/pci-host/piix.c |8 + hw/pci-host/q35.c| 26 +- hw/pci/pcie_host.c | 24 - hw/timer/hpet.c |5 + qom/object.c | 56 + vl.c |3 + hw/i386/Makefile.objs| 27 + hw/i386/acpi-dsdt-cpu-hotplug.dsl| 93 + hw/i386/acpi-dsdt-dbug.dsl | 41 + hw/i386/acpi-dsdt-hpet.dsl | 51 + hw/i386/acpi-dsdt-isa.dsl| 117 + hw/i386/acpi-dsdt-pci-crs.dsl| 105 + hw/i386/acpi-dsdt.dsl|
[Qemu-devel] [PATCH v4 01/23] qemu: add Error to typedefs
This is so qom headers can use it without pulling in extra headers. Signed-off-by: Michael S. Tsirkin m...@redhat.com --- include/qemu/typedefs.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/qemu/typedefs.h b/include/qemu/typedefs.h index a4c1b84..46c3599 100644 --- a/include/qemu/typedefs.h +++ b/include/qemu/typedefs.h @@ -7,6 +7,7 @@ typedef struct QEMUTimer QEMUTimer; typedef struct QEMUTimerListGroup QEMUTimerListGroup; typedef struct QEMUFile QEMUFile; typedef struct QEMUBH QEMUBH; +typedef struct Error Error; typedef struct AioContext AioContext; -- MST
[Qemu-devel] [PATCH v4 12/23] acpi: add rules to compile ASL source
Detect presence of IASL compiler and use it
to process ASL source. If not there, use pre-compiled
files in-tree. Add script to update the in-tree files.
Note: distros are known to silently update iasl
so detect correct iasl flags for the installed version on each run as
opposed to at configure time.
Signed-off-by: Michael S. Tsirkin m...@redhat.com
Reviewed-by: Laszlo Ersek ler...@redhat.com
---
configure | 9 -
hw/i386/Makefile.objs | 22 ++
scripts/update-acpi.sh | 4
3 files changed, 34 insertions(+), 1 deletion(-)
create mode 100644 scripts/update-acpi.sh
diff --git a/configure b/configure
index 2b83936..15405e1 100755
--- a/configure
+++ b/configure
@@ -119,6 +119,7 @@ path_of() {
# default parameters
source_path=`dirname $0`
cpu=
+iasl=iasl
interp_prefix=/usr/gnemul/qemu-%M
static=no
cross_prefix=
@@ -257,6 +258,8 @@ for opt do
;;
--cxx=*) CXX=$optarg
;;
+ --iasl=*) iasl=$optarg
+ ;;
--source-path=*) source_path=$optarg
;;
--cpu=*) cpu=$optarg
@@ -1055,6 +1058,7 @@ echo Advanced options (experts only):
echo --source-path=PATH path of source code [$source_path]
echo --cross-prefix=PREFIXuse PREFIX for compile tools [$cross_prefix]
echo --cc=CC use C compiler CC [$cc]
+echo --iasl=IASL use ACPI compiler IASL [$iasl]
echo --host-cc=CC use C compiler CC [$host_cc] for code run at
echobuild time
echo --cxx=CXXuse C++ compiler CXX [$cxx]
@@ -4239,6 +4243,9 @@ else
fi
echo PYTHON=$python $config_host_mak
echo CC=$cc $config_host_mak
+if $iasl -h /dev/null 21; then
+ echo IASL=$iasl $config_host_mak
+fi
echo CC_I386=$cc_i386 $config_host_mak
echo HOST_CC=$host_cc $config_host_mak
echo CXX=$cxx $config_host_mak
@@ -4691,7 +4698,7 @@ for rom in seabios vgabios ; do
echo BCC=bcc $config_mak
echo CPP=$cpp $config_mak
echo OBJCOPY=objcopy $config_mak
-echo IASL=iasl $config_mak
+echo IASL=$iasl $config_mak
echo LD=$ld $config_mak
done
diff --git a/hw/i386/Makefile.objs b/hw/i386/Makefile.objs
index 45e6165..f950707 100644
--- a/hw/i386/Makefile.objs
+++ b/hw/i386/Makefile.objs
@@ -5,3 +5,25 @@ obj-y += pc_sysfw.o
obj-$(CONFIG_XEN) += xen_domainbuild.o xen_machine_pv.o
obj-y += kvmvapic.o
+
+iasl-option=$(shell if test -z `$(1) $(2) 21 /dev/null` \
+; then echo $(2); else echo $(3); fi ;)
+
+ifdef IASL
+#IASL Present. Generate hex files from .dsl
+hw/i386/%.hex: $(SRC_PATH)/hw/i386/%.dsl
$(SRC_PATH)/scripts/acpi_extract_preprocess.py
$(SRC_PATH)/scripts/acpi_extract.py
+ $(call quiet-command, cpp -P $ -o $*.dsl.i.orig, CPP
$(TARGET_DIR)$*.dsl.i.orig)
+ $(call quiet-command, $(PYTHON)
$(SRC_PATH)/scripts/acpi_extract_preprocess.py $*.dsl.i.orig $*.dsl.i,
ACPI_PREPROCESS $(TARGET_DIR)$*.dsl.i)
+ $(call quiet-command, $(IASL) $(call iasl-option,$(IASL),-Pn,) -vs -l
-tc -p $* $*.dsl.i $(if $(V), , /dev/null) 21 , IASL
$(TARGET_DIR)$*.dsl.i)
+ $(call quiet-command, $(SRC_PATH)/scripts/acpi_extract.py $*.lst
$*.off, ACPI_EXTRACT $(TARGET_DIR)$*.off)
+ $(call quiet-command, cat $*.off $@, CAT $(TARGET_DIR)$@)
+else
+#IASL Not present. Restore pre-generated hex files.
+hw/i386/%.hex: $(SRC_PATH)/hw/i386/%.hex.generated
+ $(call quiet-command, cp -f $ $@, CP $(TARGET_DIR)$@)
+endif
+
+.PHONY: cleanhex
+cleanhex:
+ rm -f hw/i386/*hex
+clean: cleanhex
diff --git a/scripts/update-acpi.sh b/scripts/update-acpi.sh
new file mode 100644
index 000..b5f05ff
--- /dev/null
+++ b/scripts/update-acpi.sh
@@ -0,0 +1,4 @@
+cd x86_64-softmmu
+for file in hw/i386/*.hex; do
+cp -f $file ../$file.generated
+done
--
MST
[Qemu-devel] [PATCH v4 06/23] loader: support for unmapped ROM blobs
Support ROM blobs not mapped into guest memory:
same as ROM files really but use caller's buffer.
Support incoking callback on access and
return memory pointer making it easier
for caller to update memory if necessary.
Signed-off-by: Michael S. Tsirkin m...@redhat.com
Reviewed-by: Laszlo Ersek ler...@redhat.com
---
hw/lm32/lm32_hwsetup.h | 2 +-
include/hw/loader.h| 7 ---
hw/core/loader.c | 23 ---
3 files changed, 25 insertions(+), 7 deletions(-)
diff --git a/hw/lm32/lm32_hwsetup.h b/hw/lm32/lm32_hwsetup.h
index 3449bd8..9fd5e69 100644
--- a/hw/lm32/lm32_hwsetup.h
+++ b/hw/lm32/lm32_hwsetup.h
@@ -73,7 +73,7 @@ static inline void hwsetup_free(HWSetup *hw)
static inline void hwsetup_create_rom(HWSetup *hw,
hwaddr base)
{
-rom_add_blob(hwsetup, hw-data, TARGET_PAGE_SIZE, base);
+rom_add_blob(hwsetup, hw-data, TARGET_PAGE_SIZE, base, NULL, NULL,
NULL);
}
static inline void hwsetup_add_u8(HWSetup *hw, uint8_t u)
diff --git a/include/hw/loader.h b/include/hw/loader.h
index 6145736..e0c576b 100644
--- a/include/hw/loader.h
+++ b/include/hw/loader.h
@@ -40,8 +40,9 @@ extern bool rom_file_in_ram;
int rom_add_file(const char *file, const char *fw_dir,
hwaddr addr, int32_t bootindex);
-int rom_add_blob(const char *name, const void *blob, size_t len,
- hwaddr addr);
+void *rom_add_blob(const char *name, const void *blob, size_t len,
+ hwaddr addr, const char *fw_file_name,
+ FWCfgReadCallback fw_callback, void *callback_opaque);
int rom_add_elf_program(const char *name, void *data, size_t datasize,
size_t romsize, hwaddr addr);
int rom_load_all(void);
@@ -53,7 +54,7 @@ void do_info_roms(Monitor *mon, const QDict *qdict);
#define rom_add_file_fixed(_f, _a, _i) \
rom_add_file(_f, NULL, _a, _i)
#define rom_add_blob_fixed(_f, _b, _l, _a) \
-rom_add_blob(_f, _b, _l, _a)
+(rom_add_blob(_f, _b, _l, _a, NULL, NULL, NULL) ? 0 : -1)
#define PC_ROM_MIN_VGA 0xc
#define PC_ROM_MIN_OPTION 0xc8000
diff --git a/hw/core/loader.c b/hw/core/loader.c
index 7b3d3ee..449bd4c 100644
--- a/hw/core/loader.c
+++ b/hw/core/loader.c
@@ -700,10 +700,12 @@ err:
return -1;
}
-int rom_add_blob(const char *name, const void *blob, size_t len,
- hwaddr addr)
+void *rom_add_blob(const char *name, const void *blob, size_t len,
+ hwaddr addr, const char *fw_file_name,
+ FWCfgReadCallback fw_callback, void *callback_opaque)
{
Rom *rom;
+void *data = NULL;
rom = g_malloc0(sizeof(*rom));
rom-name = g_strdup(name);
@@ -713,7 +715,22 @@ int rom_add_blob(const char *name, const void *blob,
size_t len,
rom-data = g_malloc0(rom-datasize);
memcpy(rom-data, blob, len);
rom_insert(rom);
-return 0;
+if (fw_file_name fw_cfg) {
+char devpath[100];
+
+snprintf(devpath, sizeof(devpath), /rom@%s, fw_file_name);
+
+if (rom_file_in_ram) {
+data = rom_set_mr(rom, OBJECT(fw_cfg), devpath);
+} else {
+data = rom-data;
+}
+
+fw_cfg_add_file_callback(fw_cfg, fw_file_name,
+ fw_callback, callback_opaque,
+ data, rom-romsize);
+}
+return data;
}
/* This function is specific for elf program because we don't need to allocate
--
MST
[Qemu-devel] [PATCH v4 08/23] pcie_host: expose address format
Callers pass in the address so it's helpful for them to be able to decode it. Signed-off-by: Michael S. Tsirkin m...@redhat.com --- include/hw/pci/pcie_host.h | 21 + hw/pci/pcie_host.c | 21 - 2 files changed, 21 insertions(+), 21 deletions(-) diff --git a/include/hw/pci/pcie_host.h b/include/hw/pci/pcie_host.h index bac3c67..da0f275 100644 --- a/include/hw/pci/pcie_host.h +++ b/include/hw/pci/pcie_host.h @@ -54,4 +54,25 @@ void pcie_host_mmcfg_update(PCIExpressHost *e, hwaddr addr, uint32_t size); +/* + * PCI express ECAM (Enhanced Configuration Address Mapping) format. + * AKA mmcfg address + * bit 20 - 28: bus number + * bit 15 - 19: device number + * bit 12 - 14: function number + * bit 0 - 11: offset in configuration space of a given device + */ +#define PCIE_MMCFG_SIZE_MAX (1ULL 28) +#define PCIE_MMCFG_SIZE_MIN (1ULL 20) +#define PCIE_MMCFG_BUS_BIT 20 +#define PCIE_MMCFG_BUS_MASK 0x1ff +#define PCIE_MMCFG_DEVFN_BIT12 +#define PCIE_MMCFG_DEVFN_MASK 0xff +#define PCIE_MMCFG_CONFOFFSET_MASK 0xfff +#define PCIE_MMCFG_BUS(addr)(((addr) PCIE_MMCFG_BUS_BIT) \ + PCIE_MMCFG_BUS_MASK) +#define PCIE_MMCFG_DEVFN(addr) (((addr) PCIE_MMCFG_DEVFN_BIT) \ + PCIE_MMCFG_DEVFN_MASK) +#define PCIE_MMCFG_CONFOFFSET(addr) ((addr) PCIE_MMCFG_CONFOFFSET_MASK) + #endif /* PCIE_HOST_H */ diff --git a/hw/pci/pcie_host.c b/hw/pci/pcie_host.c index 410ac08..c6e1b57 100644 --- a/hw/pci/pcie_host.c +++ b/hw/pci/pcie_host.c @@ -24,27 +24,6 @@ #include hw/pci/pcie_host.h #include exec/address-spaces.h -/* - * PCI express mmcfig address - * bit 20 - 28: bus number - * bit 15 - 19: device number - * bit 12 - 14: function number - * bit 0 - 11: offset in configuration space of a given device - */ -#define PCIE_MMCFG_SIZE_MAX (1ULL 28) -#define PCIE_MMCFG_SIZE_MIN (1ULL 20) -#define PCIE_MMCFG_BUS_BIT 20 -#define PCIE_MMCFG_BUS_MASK 0x1ff -#define PCIE_MMCFG_DEVFN_BIT12 -#define PCIE_MMCFG_DEVFN_MASK 0xff -#define PCIE_MMCFG_CONFOFFSET_MASK 0xfff -#define PCIE_MMCFG_BUS(addr)(((addr) PCIE_MMCFG_BUS_BIT) \ - PCIE_MMCFG_BUS_MASK) -#define PCIE_MMCFG_DEVFN(addr) (((addr) PCIE_MMCFG_DEVFN_BIT) \ - PCIE_MMCFG_DEVFN_MASK) -#define PCIE_MMCFG_CONFOFFSET(addr) ((addr) PCIE_MMCFG_CONFOFFSET_MASK) - - /* a helper function to get a PCIDevice for a given mmconfig address */ static inline PCIDevice *pcie_dev_find_by_mmcfg_addr(PCIBus *s, uint32_t mmcfg_addr) -- MST
[Qemu-devel] [PATCH v4 15/23] i386: add bios linker/loader
This adds a dynamic bios linker/loader.
This will be used by acpi table generation
code to:
- load each table in the appropriate memory segment
- link tables to each other
- fix up checksums after said linking
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
include/hw/i386/bios-linker-loader.h | 26 ++
hw/i386/bios-linker-loader.c | 156 +++
hw/i386/Makefile.objs| 1 +
3 files changed, 183 insertions(+)
create mode 100644 include/hw/i386/bios-linker-loader.h
create mode 100644 hw/i386/bios-linker-loader.c
diff --git a/include/hw/i386/bios-linker-loader.h
b/include/hw/i386/bios-linker-loader.h
new file mode 100644
index 000..18c3868
--- /dev/null
+++ b/include/hw/i386/bios-linker-loader.h
@@ -0,0 +1,26 @@
+#ifndef BIOS_LINKER_LOADER_H
+#define BIOS_LINKER_LOADER_H
+
+#include glib.h
+#include stdbool.h
+#include inttypes.h
+
+GArray *bios_linker_init(void);
+
+void bios_linker_alloc(GArray *linker,
+ const char *file,
+ uint32_t alloc_align,
+ bool alloc_fseg);
+
+void bios_linker_add_checksum(GArray *linker, const char *file, void *table,
+ void *start, unsigned size, uint8_t *checksum);
+
+
+void bios_linker_add_pointer(GArray *linker,
+ const char *dest_file,
+ const char *src_file,
+ GArray *table, void *pointer,
+ uint8_t pointer_size);
+
+void *bios_linker_cleanup(GArray *linker);
+#endif
diff --git a/hw/i386/bios-linker-loader.c b/hw/i386/bios-linker-loader.c
new file mode 100644
index 000..644016b
--- /dev/null
+++ b/hw/i386/bios-linker-loader.c
@@ -0,0 +1,156 @@
+/* Dynamic linker/loader of ACPI tables
+ *
+ * Copyright (C) 2013 Red Hat Inc
+ *
+ * Author: Michael S. Tsirkin m...@redhat.com
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+
+ * This program 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 General Public License for more details.
+
+ * You should have received a copy of the GNU General Public License along
+ * with this program; if not, see http://www.gnu.org/licenses/.
+ */
+
+#include hw/i386/bios-linker-loader.h
+#include hw/nvram/fw_cfg.h
+
+#include string.h
+#include assert.h
+#include qemu/bswap.h
+
+#define BIOS_LINKER_LOADER_FILESZ FW_CFG_MAX_FILE_PATH
+
+struct BiosLinkerLoaderEntry {
+uint32_t command;
+union {
+/*
+ * COMMAND_ALLOCATE - allocate a table from @alloc_file
+ * subject to @alloc_align alignment (must be power of 2)
+ * and @alloc_zone (can be HIGH or FSEG) requirements.
+ *
+ * Must appear exactly once for each file, and before
+ * this file is referenced by any other command.
+ */
+struct {
+char alloc_file[BIOS_LINKER_LOADER_FILESZ];
+uint32_t alloc_align;
+uint8_t alloc_zone;
+};
+
+/*
+ * COMMAND_ADD_POINTER - patch the table (originating from
+ * @dest_file) at @pointer_offset, by adding a pointer to the table
+ * originating from @src_file. 1,2,4 or 8 byte unsigned
+ * addition is used depending on @pointer_size.
+ */
+struct {
+char pointer_dest_file[BIOS_LINKER_LOADER_FILESZ];
+char pointer_src_file[BIOS_LINKER_LOADER_FILESZ];
+uint32_t pointer_offset;
+uint8_t pointer_size;
+};
+
+/*
+ * COMMAND_ADD_CHECKSUM - calculate checksum of the range specified by
+ * @cksum_start and @cksum_length fields,
+ * and then add the value at @cksum_offset.
+ * Checksum simply sums -X for each byte X in the range
+ * using 8-bit math.
+ */
+struct {
+char cksum_file[BIOS_LINKER_LOADER_FILESZ];
+uint32_t cksum_offset;
+uint32_t cksum_start;
+uint32_t cksum_length;
+};
+
+/* padding */
+char pad[124];
+};
+} QEMU_PACKED;
+typedef struct BiosLinkerLoaderEntry BiosLinkerLoaderEntry;
+
+enum {
+BIOS_LINKER_LOADER_COMMAND_ALLOCATE = 0x1,
+BIOS_LINKER_LOADER_COMMAND_ADD_POINTER = 0x2,
+BIOS_LINKER_LOADER_COMMAND_ADD_CHECKSUM = 0x3,
+};
+
+enum {
+BIOS_LINKER_LOADER_ALLOC_ZONE_HIGH = 0x1,
+BIOS_LINKER_LOADER_ALLOC_ZONE_FSEG = 0x2,
+};
+
+GArray *bios_linker_init(void)
+{
+return g_array_new(false, true /* clear */, sizeof(BiosLinkerLoaderEntry));
+}
+
+/* Free linker wrapper and return the linker array. */
+void
[Qemu-devel] [PATCH v4 22/23] hpet: add API to find it
Add API to find HPET using QOM.
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
include/hw/timer/hpet.h | 2 ++
hw/timer/hpet.c | 5 +
2 files changed, 7 insertions(+)
diff --git a/include/hw/timer/hpet.h b/include/hw/timer/hpet.h
index 757f79f..ab44bd3 100644
--- a/include/hw/timer/hpet.h
+++ b/include/hw/timer/hpet.h
@@ -71,4 +71,6 @@ struct hpet_fw_config
} QEMU_PACKED;
extern struct hpet_fw_config hpet_cfg;
+
+bool hpet_find(void);
#endif
diff --git a/hw/timer/hpet.c b/hw/timer/hpet.c
index fcd22ae..676bd7d 100644
--- a/hw/timer/hpet.c
+++ b/hw/timer/hpet.c
@@ -757,6 +757,11 @@ static void hpet_device_class_init(ObjectClass *klass,
void *data)
dc-props = hpet_device_properties;
}
+bool hpet_find(void)
+{
+return object_resolve_path_type(, hpet, NULL);
+}
+
static const TypeInfo hpet_device_info = {
.name = TYPE_HPET,
.parent= TYPE_SYS_BUS_DEVICE,
--
MST
[Qemu-devel] [PATCH v4 09/23] q35: use macro for MCFG property name
Useful to make it accessible through QOM.
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
include/hw/pci/pcie_host.h | 2 ++
hw/pci-host/q35.c | 2 +-
2 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/include/hw/pci/pcie_host.h b/include/hw/pci/pcie_host.h
index da0f275..33d75bd 100644
--- a/include/hw/pci/pcie_host.h
+++ b/include/hw/pci/pcie_host.h
@@ -28,6 +28,8 @@
#define PCIE_HOST_BRIDGE(obj) \
OBJECT_CHECK(PCIExpressHost, (obj), TYPE_PCIE_HOST_BRIDGE)
+#define PCIE_HOST_MCFG_BASE MCFG
+
/* pcie_host::base_addr == PCIE_BASE_ADDR_UNMAPPED when it isn't mapped. */
#define PCIE_BASE_ADDR_UNMAPPED ((hwaddr)-1ULL)
diff --git a/hw/pci-host/q35.c b/hw/pci-host/q35.c
index 23dbeea..e46f286 100644
--- a/hw/pci-host/q35.c
+++ b/hw/pci-host/q35.c
@@ -110,7 +110,7 @@ static void q35_host_get_pci_hole64_end(Object *obj,
Visitor *v,
}
static Property mch_props[] = {
-DEFINE_PROP_UINT64(MCFG, Q35PCIHost, parent_obj.base_addr,
+DEFINE_PROP_UINT64(PCIE_HOST_MCFG_BASE, Q35PCIHost, parent_obj.base_addr,
MCH_HOST_BRIDGE_PCIEXBAR_DEFAULT),
DEFINE_PROP_SIZE(PCI_HOST_PROP_PCI_HOLE64_SIZE, Q35PCIHost,
mch.pci_hole64_size, DEFAULT_PCI_HOLE64_SIZE),
--
MST
[Qemu-devel] [PATCH v4 07/23] pcie_host: expose UNMAPPED macro
Make it possible to test unmapped status through QMP.
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
include/hw/pci/pcie_host.h | 3 +++
hw/pci/pcie_host.c | 3 ---
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/include/hw/pci/pcie_host.h b/include/hw/pci/pcie_host.h
index 1228e36..bac3c67 100644
--- a/include/hw/pci/pcie_host.h
+++ b/include/hw/pci/pcie_host.h
@@ -28,6 +28,9 @@
#define PCIE_HOST_BRIDGE(obj) \
OBJECT_CHECK(PCIExpressHost, (obj), TYPE_PCIE_HOST_BRIDGE)
+/* pcie_host::base_addr == PCIE_BASE_ADDR_UNMAPPED when it isn't mapped. */
+#define PCIE_BASE_ADDR_UNMAPPED ((hwaddr)-1ULL)
+
struct PCIExpressHost {
PCIHostState pci;
diff --git a/hw/pci/pcie_host.c b/hw/pci/pcie_host.c
index b70e5ad..410ac08 100644
--- a/hw/pci/pcie_host.c
+++ b/hw/pci/pcie_host.c
@@ -104,9 +104,6 @@ static const MemoryRegionOps pcie_mmcfg_ops = {
.endianness = DEVICE_NATIVE_ENDIAN,
};
-/* pcie_host::base_addr == PCIE_BASE_ADDR_UNMAPPED when it isn't mapped. */
-#define PCIE_BASE_ADDR_UNMAPPED ((hwaddr)-1ULL)
-
int pcie_host_init(PCIExpressHost *e)
{
e-base_addr = PCIE_BASE_ADDR_UNMAPPED;
--
MST
[Qemu-devel] [PATCH v4 16/23] loader: allow adding ROMs in done callbacks
Don't abort if machine done callbacks add ROMs.
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
include/hw/loader.h | 1 +
hw/core/loader.c| 6 +-
vl.c| 3 +++
3 files changed, 9 insertions(+), 1 deletion(-)
diff --git a/include/hw/loader.h b/include/hw/loader.h
index e0c576b..58eca98 100644
--- a/include/hw/loader.h
+++ b/include/hw/loader.h
@@ -46,6 +46,7 @@ void *rom_add_blob(const char *name, const void *blob, size_t
len,
int rom_add_elf_program(const char *name, void *data, size_t datasize,
size_t romsize, hwaddr addr);
int rom_load_all(void);
+void rom_load_done(void);
void rom_set_fw(FWCfgState *f);
int rom_copy(uint8_t *dest, hwaddr addr, size_t size);
void *rom_ptr(hwaddr addr);
diff --git a/hw/core/loader.c b/hw/core/loader.c
index 060729f..60d2ebd 100644
--- a/hw/core/loader.c
+++ b/hw/core/loader.c
@@ -812,10 +812,14 @@ int rom_load_all(void)
memory_region_unref(section.mr);
}
qemu_register_reset(rom_reset, NULL);
-roms_loaded = 1;
return 0;
}
+void rom_load_done(void)
+{
+roms_loaded = 1;
+}
+
void rom_set_fw(FWCfgState *f)
{
fw_cfg = f;
diff --git a/vl.c b/vl.c
index 4e709d5..7469145 100644
--- a/vl.c
+++ b/vl.c
@@ -4336,6 +4336,9 @@ int main(int argc, char **argv, char **envp)
qemu_register_reset(qbus_reset_all_fn, sysbus_get_default());
qemu_run_machine_init_done_notifiers();
+/* Done notifiers can load ROMs */
+rom_load_done();
+
qemu_system_reset(VMRESET_SILENT);
if (loadvm) {
if (load_vmstate(loadvm) 0) {
--
MST
[Qemu-devel] [PATCH v4 10/23] q35: expose mmcfg size as a property
Address is already exposed, expose size for symmetry.
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
include/hw/pci/pcie_host.h | 1 +
hw/pci-host/q35.c | 14 ++
2 files changed, 15 insertions(+)
diff --git a/include/hw/pci/pcie_host.h b/include/hw/pci/pcie_host.h
index 33d75bd..acca45e 100644
--- a/include/hw/pci/pcie_host.h
+++ b/include/hw/pci/pcie_host.h
@@ -29,6 +29,7 @@
OBJECT_CHECK(PCIExpressHost, (obj), TYPE_PCIE_HOST_BRIDGE)
#define PCIE_HOST_MCFG_BASE MCFG
+#define PCIE_HOST_MCFG_SIZE mcfg_size
/* pcie_host::base_addr == PCIE_BASE_ADDR_UNMAPPED when it isn't mapped. */
#define PCIE_BASE_ADDR_UNMAPPED ((hwaddr)-1ULL)
diff --git a/hw/pci-host/q35.c b/hw/pci-host/q35.c
index e46f286..a051b58 100644
--- a/hw/pci-host/q35.c
+++ b/hw/pci-host/q35.c
@@ -109,6 +109,16 @@ static void q35_host_get_pci_hole64_end(Object *obj,
Visitor *v,
visit_type_uint64(v, w64.end, name, errp);
}
+static void q35_host_get_mmcfg_size(Object *obj, Visitor *v,
+void *opaque, const char *name,
+Error **errp)
+{
+PCIExpressHost *e = PCIE_HOST_BRIDGE(obj);
+uint32_t value = e-size;
+
+visit_type_uint32(v, value, name, errp);
+}
+
static Property mch_props[] = {
DEFINE_PROP_UINT64(PCIE_HOST_MCFG_BASE, Q35PCIHost, parent_obj.base_addr,
MCH_HOST_BRIDGE_PCIEXBAR_DEFAULT),
@@ -160,6 +170,10 @@ static void q35_host_initfn(Object *obj)
q35_host_get_pci_hole64_end,
NULL, NULL, NULL, NULL);
+object_property_add(obj, PCIE_HOST_MCFG_SIZE, int,
+q35_host_get_mmcfg_size,
+NULL, NULL, NULL, NULL);
+
/* Leave enough space for the biggest MCFG BAR */
/* TODO: this matches current bios behaviour, but
* it's not a power of two, which means an MTRR
--
MST
[Qemu-devel] [PATCH v4 17/23] i386: define pc guest info
This defines a structure that will be used to fill in acpi tables
where relevant properties are not yet available using QOM.
Signed-off-by: Michael S. Tsirkin m...@redhat.com
Reviewed-by: Laszlo Ersek ler...@redhat.com
---
include/hw/i386/pc.h | 10 ++
hw/i386/pc.c | 35 +++
2 files changed, 45 insertions(+)
diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
index 9b2ddc4..7a343f8 100644
--- a/include/hw/i386/pc.h
+++ b/include/hw/i386/pc.h
@@ -9,6 +9,9 @@
#include hw/i386/ioapic.h
#include qemu/range.h
+#include qemu/bitmap.h
+#include sysemu/sysemu.h
+#include hw/pci/pci.h
/* PC-style peripherals (also used by other machines). */
@@ -20,6 +23,13 @@ typedef struct PcPciInfo {
struct PcGuestInfo {
bool has_pci_info;
bool isapc_ram_fw;
+hwaddr ram_size;
+unsigned apic_id_limit;
+bool apic_xrupt_override;
+uint64_t numa_nodes;
+uint64_t *node_mem;
+uint64_t *node_cpu;
+DECLARE_BITMAP(found_cpus, MAX_CPUMASK_BITS + 1);
FWCfgState *fw_cfg;
};
diff --git a/hw/i386/pc.c b/hw/i386/pc.c
index 0c313fe..3aa4088 100644
--- a/hw/i386/pc.c
+++ b/hw/i386/pc.c
@@ -1028,6 +1028,26 @@ static void pc_fw_cfg_guest_info(PcGuestInfo *guest_info)
fw_cfg_add_file(guest_info-fw_cfg, etc/pci-info, info, sizeof *info);
}
+static void pc_set_cpu_guest_info(CPUState *cpu, PcGuestInfo *guest_info)
+{
+CPUClass *klass = CPU_GET_CLASS(cpu);
+uint64_t apic_id = klass-get_arch_id(cpu);
+int j;
+
+assert(apic_id = MAX_CPUMASK_BITS);
+assert(apic_id guest_info-apic_id_limit);
+
+set_bit(apic_id, guest_info-found_cpus);
+
+for (j = 0; j guest_info-numa_nodes; j++) {
+assert(cpu-cpu_index max_cpus);
+if (test_bit(cpu-cpu_index, node_cpumask[j])) {
+guest_info-node_cpu[apic_id] = cpu_to_le64(j);
+break;
+}
+}
+}
+
typedef struct PcGuestInfoState {
PcGuestInfo info;
Notifier machine_done;
@@ -1047,6 +1067,21 @@ PcGuestInfo *pc_guest_info_init(ram_addr_t
below_4g_mem_size,
{
PcGuestInfoState *guest_info_state = g_malloc0(sizeof *guest_info_state);
PcGuestInfo *guest_info = guest_info_state-info;
+CPUState *cpu;
+
+guest_info-ram_size = below_4g_mem_size + above_4g_mem_size;
+guest_info-apic_id_limit = pc_apic_id_limit(max_cpus);
+guest_info-apic_xrupt_override = kvm_allows_irq0_override();
+guest_info-numa_nodes = nb_numa_nodes;
+guest_info-node_mem = g_memdup(node_mem, guest_info-numa_nodes *
+sizeof *guest_info-node_mem);
+guest_info-node_cpu = g_malloc0(guest_info-apic_id_limit *
+ sizeof *guest_info-node_cpu);
+
+memset(guest_info-found_cpus, 0, sizeof guest_info-found_cpus);
+CPU_FOREACH(cpu) {
+pc_set_cpu_guest_info(cpu, guest_info);
+}
guest_info_state-machine_done.notify = pc_guest_info_machine_done;
qemu_add_machine_init_done_notifier(guest_info_state-machine_done);
--
MST
[Qemu-devel] [PATCH v4 21/23] pvpanic: add API to access io port
Add API to find pvpanic device and get its io port.
Will be used to fill in guest info structure.
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
include/hw/i386/pc.h | 1 +
hw/misc/pvpanic.c| 13 -
2 files changed, 13 insertions(+), 1 deletion(-)
diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
index 2a5d996..61d1ee7 100644
--- a/include/hw/i386/pc.h
+++ b/include/hw/i386/pc.h
@@ -236,6 +236,7 @@ void pc_system_firmware_init(MemoryRegion *rom_memory,
/* pvpanic.c */
void pvpanic_init(ISABus *bus);
+uint16_t pvpanic_port(void);
/* e820 types */
#define E820_RAM1
diff --git a/hw/misc/pvpanic.c b/hw/misc/pvpanic.c
index b64e3bb..226e298 100644
--- a/hw/misc/pvpanic.c
+++ b/hw/misc/pvpanic.c
@@ -117,8 +117,19 @@ void pvpanic_init(ISABus *bus)
isa_create_simple(bus, TYPE_ISA_PVPANIC_DEVICE);
}
+#define PVPANIC_IOPORT_PROP ioport
+
+uint16_t pvpanic_port(void)
+{
+Object *o = object_resolve_path_type(, TYPE_ISA_PVPANIC_DEVICE, NULL);
+if (!o) {
+return 0;
+}
+return object_property_get_int(o, PVPANIC_IOPORT_PROP, NULL);
+}
+
static Property pvpanic_isa_properties[] = {
-DEFINE_PROP_UINT16(ioport, PVPanicState, ioport, 0x505),
+DEFINE_PROP_UINT16(PVPANIC_IOPORT_PROP, PVPanicState, ioport, 0x505),
DEFINE_PROP_END_OF_LIST(),
};
--
MST
[Qemu-devel] [PATCH v4 14/23] loader: use file path size from fw_cfg.h
Avoid a bit of code duplication, make
max file path constant reusable.
Suggested-by: Laszlo Ersek ler...@redhat.com
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
include/hw/nvram/fw_cfg.h | 4 +++-
hw/core/loader.c | 2 +-
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/include/hw/nvram/fw_cfg.h b/include/hw/nvram/fw_cfg.h
index 2ab0fc2..72b1549 100644
--- a/include/hw/nvram/fw_cfg.h
+++ b/include/hw/nvram/fw_cfg.h
@@ -46,12 +46,14 @@
#define FW_CFG_INVALID 0x
+#define FW_CFG_MAX_FILE_PATH56
+
#ifndef NO_QEMU_PROTOS
typedef struct FWCfgFile {
uint32_t size;/* file size */
uint16_t select; /* write this to 0x510 to read it */
uint16_t reserved;
-char name[56];
+char name[FW_CFG_MAX_FILE_PATH];
} FWCfgFile;
typedef struct FWCfgFiles {
diff --git a/hw/core/loader.c b/hw/core/loader.c
index 449bd4c..060729f 100644
--- a/hw/core/loader.c
+++ b/hw/core/loader.c
@@ -663,7 +663,7 @@ int rom_add_file(const char *file, const char *fw_dir,
rom_insert(rom);
if (rom-fw_file fw_cfg) {
const char *basename;
-char fw_file_name[56];
+char fw_file_name[FW_CFG_MAX_FILE_PATH];
void *data;
basename = strrchr(rom-fw_file, '/');
--
MST
[Qemu-devel] [PATCH v4 18/23] acpi/piix: add macros for acpi property names
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
include/hw/i386/pc.h | 10 ++
hw/acpi/piix4.c | 6 +++---
2 files changed, 13 insertions(+), 3 deletions(-)
diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
index 7a343f8..f966cef 100644
--- a/include/hw/i386/pc.h
+++ b/include/hw/i386/pc.h
@@ -20,6 +20,16 @@ typedef struct PcPciInfo {
Range w64;
} PcPciInfo;
+#define ACPI_PM_PROP_S3_DISABLED disable_s3
+#define ACPI_PM_PROP_S4_DISABLED disable_s4
+#define ACPI_PM_PROP_S4_VAL s4_val
+#define ACPI_PM_PROP_SCI_INT sci_int
+#define ACPI_PM_PROP_ACPI_ENABLE_CMD acpi_enable_cmd
+#define ACPI_PM_PROP_ACPI_DISABLE_CMD acpi_disable_cmd
+#define ACPI_PM_PROP_PM_IO_BASE pm_io_base
+#define ACPI_PM_PROP_GPE0_BLK gpe0_blk
+#define ACPI_PM_PROP_GPE0_BLK_LEN gpe0_blk_len
+
struct PcGuestInfo {
bool has_pci_info;
bool isapc_ram_fw;
diff --git a/hw/acpi/piix4.c b/hw/acpi/piix4.c
index b46bd5e..4b8c1da 100644
--- a/hw/acpi/piix4.c
+++ b/hw/acpi/piix4.c
@@ -489,9 +489,9 @@ i2c_bus *piix4_pm_init(PCIBus *bus, int devfn, uint32_t
smb_io_base,
static Property piix4_pm_properties[] = {
DEFINE_PROP_UINT32(smb_io_base, PIIX4PMState, smb_io_base, 0),
-DEFINE_PROP_UINT8(disable_s3, PIIX4PMState, disable_s3, 0),
-DEFINE_PROP_UINT8(disable_s4, PIIX4PMState, disable_s4, 0),
-DEFINE_PROP_UINT8(s4_val, PIIX4PMState, s4_val, 2),
+DEFINE_PROP_UINT8(ACPI_PM_PROP_S3_DISABLED, PIIX4PMState, disable_s3, 0),
+DEFINE_PROP_UINT8(ACPI_PM_PROP_S4_DISABLED, PIIX4PMState, disable_s4, 0),
+DEFINE_PROP_UINT8(ACPI_PM_PROP_S4_VAL, PIIX4PMState, s4_val, 2),
DEFINE_PROP_END_OF_LIST(),
};
--
MST
Re: [Qemu-devel] [SeaBIOS] [PATCH v2 1/5] linker: utility to patch in-memory ROM files
On Sun, Sep 22, 2013 at 02:18:45PM +0300, Michael S. Tsirkin wrote:
On Sun, Sep 22, 2013 at 01:49:58PM +0300, Michael S. Tsirkin wrote:
On Thu, Jul 25, 2013 at 08:06:27PM -0400, Kevin O'Connor wrote:
On Thu, Jul 25, 2013 at 03:55:56PM +0300, Michael S. Tsirkin wrote:
On Mon, Jul 15, 2013 at 11:01:02AM +0300, Michael S. Tsirkin wrote:
On Sun, Jul 14, 2013 at 02:24:52PM -0400, Kevin O'Connor wrote:
I'd prefer to see this tracked within the linker code and not in
the
generic romfile struct.
A way to associate a romfile instance with a value seems generally
useful, no? Still, that's not too hard - it would only mean an extra
linked list of
struct linker {
char name[56]
void *data;
struct hlist_node node;
}
is this preferable?
Sure, but it's probably easier to do something like:
struct linkfiles { char *name; void *data; };
void linker_loader_execute(const char *name)
{
int size;
struct linker_loader_entry_s *entries = romfile_loadfile(name, size);
int numentries = size/sizeof(entries[0]);
if (! entries)
return;
struct linkfiles *files = malloc_tmp(sizeof(files[0]) * numentries);
and then just populate and use the array of filenames.
OK I'll do this but it's more code as I can't use plain romfile_find
anymore, and have to code up my own lookup.
Also, is there another name besides linker that could be used?
SeaBIOS has code to self-relocate and fixup code relocations. I
think
having code in the repo called linker could cause confusion.
romfile_loader?
Shrug. How about tabledeploy?
-Kevin
So I tried this out
Latest version that I posted uses the approach suggested by Kevin here.
It adds about 200 bytes to code size.
If we want to cut that out and go back to data pointer,
we can do this as a patch on top.
Pls let me know.
--
MST
[Qemu-devel] [PATCH v4 19/23] piix: APIs for pc guest info
This adds APIs that will be used to fill in guest acpi tables.
Some required information is still lacking in QOM, so we
fall back on lookups by type and returning explicit types.
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
include/hw/acpi/piix4.h | 8
include/hw/i386/pc.h| 1 +
hw/acpi/piix4.c | 44
hw/pci-host/piix.c | 8
4 files changed, 57 insertions(+), 4 deletions(-)
create mode 100644 include/hw/acpi/piix4.h
diff --git a/include/hw/acpi/piix4.h b/include/hw/acpi/piix4.h
new file mode 100644
index 000..65e6fd7
--- /dev/null
+++ b/include/hw/acpi/piix4.h
@@ -0,0 +1,8 @@
+#ifndef HW_ACPI_PIIX4_H
+#define HW_ACPI_PIIX4_H
+
+#include qemu/typedefs.h
+
+Object *piix4_pm_find(void);
+
+#endif
diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
index f966cef..2a5d996 100644
--- a/include/hw/i386/pc.h
+++ b/include/hw/i386/pc.h
@@ -193,6 +193,7 @@ PCIBus *i440fx_init(PCII440FXState **pi440fx_state, int
*piix_devfn,
MemoryRegion *pci_memory,
MemoryRegion *ram_memory);
+PCIBus *find_i440fx(void);
/* piix4.c */
extern PCIDevice *piix4_dev;
int piix4_init(PCIBus *bus, ISABus **isa_bus, int devfn);
diff --git a/hw/acpi/piix4.c b/hw/acpi/piix4.c
index 4b8c1da..3bcd890 100644
--- a/hw/acpi/piix4.c
+++ b/hw/acpi/piix4.c
@@ -29,6 +29,7 @@
#include exec/ioport.h
#include hw/nvram/fw_cfg.h
#include exec/address-spaces.h
+#include hw/acpi/piix4.h
//#define DEBUG
@@ -69,6 +70,8 @@ typedef struct PIIX4PMState {
/* public */
MemoryRegion io;
+uint32_t io_base;
+
MemoryRegion io_gpe;
MemoryRegion io_pci;
MemoryRegion io_cpu;
@@ -152,14 +155,13 @@ static void apm_ctrl_changed(uint32_t val, void *arg)
static void pm_io_space_update(PIIX4PMState *s)
{
PCIDevice *d = PCI_DEVICE(s);
-uint32_t pm_io_base;
-pm_io_base = le32_to_cpu(*(uint32_t *)(d-config + 0x40));
-pm_io_base = 0xffc0;
+s-io_base = le32_to_cpu(*(uint32_t *)(d-config + 0x40));
+s-io_base = 0xffc0;
memory_region_transaction_begin();
memory_region_set_enabled(s-io, d-config[0x80] 1);
-memory_region_set_address(s-io, pm_io_base);
+memory_region_set_address(s-io, s-io_base);
memory_region_transaction_commit();
}
@@ -407,6 +409,28 @@ static void piix4_pm_machine_ready(Notifier *n, void
*opaque)
(memory_region_present(io_as, 0x2f8) ? 0x90 : 0);
}
+static void piix4_pm_add_propeties(PIIX4PMState *s)
+{
+static const uint8_t acpi_enable_cmd = ACPI_ENABLE;
+static const uint8_t acpi_disable_cmd = ACPI_DISABLE;
+static const uint32_t gpe0_blk = GPE_BASE;
+static const uint32_t gpe0_blk_len = GPE_LEN;
+static const uint16_t sci_int = 9;
+
+object_property_add_uint8_ptr(OBJECT(s), ACPI_PM_PROP_ACPI_ENABLE_CMD,
+ acpi_enable_cmd, NULL);
+object_property_add_uint8_ptr(OBJECT(s), ACPI_PM_PROP_ACPI_DISABLE_CMD,
+ acpi_disable_cmd, NULL);
+object_property_add_uint32_ptr(OBJECT(s), ACPI_PM_PROP_GPE0_BLK,
+ gpe0_blk, NULL);
+object_property_add_uint32_ptr(OBJECT(s), ACPI_PM_PROP_GPE0_BLK_LEN,
+ gpe0_blk_len, NULL);
+object_property_add_uint16_ptr(OBJECT(s), ACPI_PM_PROP_SCI_INT,
+ sci_int, NULL);
+object_property_add_uint32_ptr(OBJECT(s), ACPI_PM_PROP_PM_IO_BASE,
+ s-io_base, NULL);
+}
+
static int piix4_pm_initfn(PCIDevice *dev)
{
PIIX4PMState *s = PIIX4_PM(dev);
@@ -456,9 +480,21 @@ static int piix4_pm_initfn(PCIDevice *dev)
piix4_acpi_system_hot_add_init(pci_address_space_io(dev), dev-bus, s);
+piix4_pm_add_propeties(s);
return 0;
}
+Object *piix4_pm_find(void)
+{
+bool ambig;
+Object *o = object_resolve_path_type(, TYPE_PIIX4_PM, ambig);
+
+if (ambig || !o) {
+return NULL;
+}
+return o;
+}
+
i2c_bus *piix4_pm_init(PCIBus *bus, int devfn, uint32_t smb_io_base,
qemu_irq sci_irq, qemu_irq smi_irq,
int kvm_enabled, FWCfgState *fw_cfg)
diff --git a/hw/pci-host/piix.c b/hw/pci-host/piix.c
index c041149..bad3953 100644
--- a/hw/pci-host/piix.c
+++ b/hw/pci-host/piix.c
@@ -416,6 +416,14 @@ PCIBus *i440fx_init(PCII440FXState **pi440fx_state,
return b;
}
+PCIBus *find_i440fx(void)
+{
+PCIHostState *s = OBJECT_CHECK(PCIHostState,
+ object_resolve_path(/machine/i440fx,
NULL),
+ TYPE_PCI_HOST_BRIDGE);
+return s ? s-bus : NULL;
+}
+
/* PIIX3 PCI to ISA bridge */
static void piix3_set_irq_pic(PIIX3State *piix3, int pic_irq)
{
--
MST
[Qemu-devel] [PATCH v4 20/23] ich9: APIs for pc guest info
This adds APIs that will be used to fill in
acpi tables, implemented using QOM,
to various ich9 components.
Some information is still missing in QOM,
so we fall back on lookups by type instead.
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
include/hw/acpi/ich9.h| 2 ++
include/hw/i386/ich9.h| 2 ++
include/hw/pci-host/q35.h | 2 ++
hw/acpi/ich9.c| 24
hw/isa/lpc_ich9.c | 40
hw/pci-host/q35.c | 10 ++
6 files changed, 80 insertions(+)
diff --git a/include/hw/acpi/ich9.h b/include/hw/acpi/ich9.h
index b1fe71f..82fcf9f 100644
--- a/include/hw/acpi/ich9.h
+++ b/include/hw/acpi/ich9.h
@@ -49,4 +49,6 @@ void ich9_pm_init(PCIDevice *lpc_pci, ICH9LPCPMRegs *pm,
void ich9_pm_iospace_update(ICH9LPCPMRegs *pm, uint32_t pm_io_base);
extern const VMStateDescription vmstate_ich9_pm;
+void ich9_pm_add_properties(Object *obj, ICH9LPCPMRegs *pm, Error **errp);
+
#endif /* HW_ACPI_ICH9_H */
diff --git a/include/hw/i386/ich9.h b/include/hw/i386/ich9.h
index c5f637b..4a68b35 100644
--- a/include/hw/i386/ich9.h
+++ b/include/hw/i386/ich9.h
@@ -66,6 +66,8 @@ typedef struct ICH9LPCState {
qemu_irq *ioapic;
} ICH9LPCState;
+Object *ich9_lpc_find(void);
+
#define Q35_MASK(bit, ms_bit, ls_bit) \
((uint##bit##_t)(((1ULL ((ms_bit) + 1)) - 1) ~((1ULL ls_bit) - 1)))
diff --git a/include/hw/pci-host/q35.h b/include/hw/pci-host/q35.h
index 6eb7ab6..f9db770 100644
--- a/include/hw/pci-host/q35.h
+++ b/include/hw/pci-host/q35.h
@@ -156,4 +156,6 @@ typedef struct Q35PCIHost {
#define MCH_PCIE_DEV 1
#define MCH_PCIE_FUNC 0
+uint64_t mch_mcfg_base(void);
+
#endif /* HW_Q35_H */
diff --git a/hw/acpi/ich9.c b/hw/acpi/ich9.c
index 3fb443d..7e0429e 100644
--- a/hw/acpi/ich9.c
+++ b/hw/acpi/ich9.c
@@ -24,6 +24,7 @@
* GNU GPL, version 2 or (at your option) any later version.
*/
#include hw/hw.h
+#include qapi/visitor.h
#include hw/i386/pc.h
#include hw/pci/pci.h
#include qemu/timer.h
@@ -228,3 +229,26 @@ void ich9_pm_init(PCIDevice *lpc_pci, ICH9LPCPMRegs *pm,
pm-powerdown_notifier.notify = pm_powerdown_req;
qemu_register_powerdown_notifier(pm-powerdown_notifier);
}
+
+static void ich9_pm_get_gpe0_blk(Object *obj, Visitor *v,
+ void *opaque, const char *name,
+ Error **errp)
+{
+ICH9LPCPMRegs *pm = opaque;
+uint32_t value = pm-pm_io_base + ICH9_PMIO_GPE0_STS;
+
+visit_type_uint32(v, value, name, errp);
+}
+
+void ich9_pm_add_properties(Object *obj, ICH9LPCPMRegs *pm, Error **errp)
+{
+static const uint32_t gpe0_len = ICH9_PMIO_GPE0_LEN;
+
+object_property_add_uint32_ptr(obj, ACPI_PM_PROP_PM_IO_BASE,
+ pm-pm_io_base, errp);
+object_property_add(obj, ACPI_PM_PROP_GPE0_BLK, uint32,
+ich9_pm_get_gpe0_blk,
+NULL, NULL, pm, NULL);
+object_property_add_uint32_ptr(obj, ACPI_PM_PROP_GPE0_BLK_LEN,
+ gpe0_len, errp);
+}
diff --git a/hw/isa/lpc_ich9.c b/hw/isa/lpc_ich9.c
index 5633d08..19b2198 100644
--- a/hw/isa/lpc_ich9.c
+++ b/hw/isa/lpc_ich9.c
@@ -29,6 +29,7 @@
*/
#include qemu-common.h
#include hw/hw.h
+#include qapi/visitor.h
#include qemu/range.h
#include hw/isa/isa.h
#include hw/sysbus.h
@@ -525,6 +526,43 @@ static const MemoryRegionOps ich9_rst_cnt_ops = {
.endianness = DEVICE_LITTLE_ENDIAN
};
+Object *ich9_lpc_find(void)
+{
+bool ambig;
+Object *o = object_resolve_path_type(, TYPE_ICH9_LPC_DEVICE, ambig);
+
+if (ambig) {
+return NULL;
+}
+return o;
+}
+
+static void ich9_lpc_get_sci_int(Object *obj, Visitor *v,
+ void *opaque, const char *name,
+ Error **errp)
+{
+ICH9LPCState *lpc = ICH9_LPC_DEVICE(obj);
+uint32_t value = ich9_lpc_sci_irq(lpc);
+
+visit_type_uint32(v, value, name, errp);
+}
+
+static void ich9_lpc_add_properties(ICH9LPCState *lpc)
+{
+static const uint8_t acpi_enable_cmd = ICH9_APM_ACPI_ENABLE;
+static const uint8_t acpi_disable_cmd = ICH9_APM_ACPI_DISABLE;
+
+object_property_add(OBJECT(lpc), ACPI_PM_PROP_SCI_INT, uint32,
+ich9_lpc_get_sci_int,
+NULL, NULL, NULL, NULL);
+object_property_add_uint8_ptr(OBJECT(lpc), ACPI_PM_PROP_ACPI_ENABLE_CMD,
+ acpi_enable_cmd, NULL);
+object_property_add_uint8_ptr(OBJECT(lpc), ACPI_PM_PROP_ACPI_DISABLE_CMD,
+ acpi_disable_cmd, NULL);
+
+ich9_pm_add_properties(OBJECT(lpc), lpc-pm, NULL);
+}
+
static int ich9_lpc_initfn(PCIDevice *d)
{
ICH9LPCState *lpc = ICH9_LPC_DEVICE(d);
@@ -552,6 +590,8 @@ static int ich9_lpc_initfn(PCIDevice *d)
ICH9_RST_CNT_IOPORT, lpc-rst_cnt_mem,
[Qemu-devel] [PATCH v4 03/23] qom: cleanup struct Error references
now that a typedef for struct Error is available, use it in qom/object.h to match coding style rules. Signed-off-by: Michael S. Tsirkin m...@redhat.com --- include/qom/object.h | 51 +-- 1 file changed, 25 insertions(+), 26 deletions(-) diff --git a/include/qom/object.h b/include/qom/object.h index 5b3b743..e3a16e4 100644 --- a/include/qom/object.h +++ b/include/qom/object.h @@ -21,7 +21,6 @@ #include qemu/queue.h struct Visitor; -struct Error; struct TypeImpl; typedef struct TypeImpl *Type; @@ -302,7 +301,7 @@ typedef void (ObjectPropertyAccessor)(Object *obj, struct Visitor *v, void *opaque, const char *name, - struct Error **errp); + Error **errp); /** * ObjectPropertyRelease: @@ -791,9 +790,9 @@ void object_property_add(Object *obj, const char *name, const char *type, ObjectPropertyAccessor *get, ObjectPropertyAccessor *set, ObjectPropertyRelease *release, - void *opaque, struct Error **errp); + void *opaque, Error **errp); -void object_property_del(Object *obj, const char *name, struct Error **errp); +void object_property_del(Object *obj, const char *name, Error **errp); /** * object_property_find: @@ -804,7 +803,7 @@ void object_property_del(Object *obj, const char *name, struct Error **errp); * Look up a property for an object and return its #ObjectProperty if found. */ ObjectProperty *object_property_find(Object *obj, const char *name, - struct Error **errp); + Error **errp); void object_unparent(Object *obj); @@ -819,7 +818,7 @@ void object_unparent(Object *obj); * Reads a property from a object. */ void object_property_get(Object *obj, struct Visitor *v, const char *name, - struct Error **errp); + Error **errp); /** * object_property_set_str: @@ -830,7 +829,7 @@ void object_property_get(Object *obj, struct Visitor *v, const char *name, * Writes a string value to a property. */ void object_property_set_str(Object *obj, const char *value, - const char *name, struct Error **errp); + const char *name, Error **errp); /** * object_property_get_str: @@ -843,7 +842,7 @@ void object_property_set_str(Object *obj, const char *value, * The caller should free the string. */ char *object_property_get_str(Object *obj, const char *name, - struct Error **errp); + Error **errp); /** * object_property_set_link: @@ -854,7 +853,7 @@ char *object_property_get_str(Object *obj, const char *name, * Writes an object's canonical path to a property. */ void object_property_set_link(Object *obj, Object *value, - const char *name, struct Error **errp); + const char *name, Error **errp); /** * object_property_get_link: @@ -867,7 +866,7 @@ void object_property_set_link(Object *obj, Object *value, * string or not a valid object path). */ Object *object_property_get_link(Object *obj, const char *name, - struct Error **errp); + Error **errp); /** * object_property_set_bool: @@ -878,7 +877,7 @@ Object *object_property_get_link(Object *obj, const char *name, * Writes a bool value to a property. */ void object_property_set_bool(Object *obj, bool value, - const char *name, struct Error **errp); + const char *name, Error **errp); /** * object_property_get_bool: @@ -890,7 +889,7 @@ void object_property_set_bool(Object *obj, bool value, * an error occurs (including when the property value is not a bool). */ bool object_property_get_bool(Object *obj, const char *name, - struct Error **errp); + Error **errp); /** * object_property_set_int: @@ -901,7 +900,7 @@ bool object_property_get_bool(Object *obj, const char *name, * Writes an integer value to a property. */ void object_property_set_int(Object *obj, int64_t value, - const char *name, struct Error **errp); + const char *name, Error **errp); /** * object_property_get_int: @@ -913,7 +912,7 @@ void object_property_set_int(Object *obj, int64_t value, * an error occurs (including when the property value is not an integer). */ int64_t object_property_get_int(Object *obj, const char *name, -struct Error **errp); +Error **errp); /**
[Qemu-devel] [PATCH v4 02/23] qom: pull in qemu/typedefs
As usual so we can use typedefs without header dependencies. Signed-off-by: Michael S. Tsirkin m...@redhat.com --- include/qom/object.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/qom/object.h b/include/qom/object.h index 1a7b71a..5b3b743 100644 --- a/include/qom/object.h +++ b/include/qom/object.h @@ -17,6 +17,7 @@ #include glib.h #include stdint.h #include stdbool.h +#include qemu/typedefs.h #include qemu/queue.h struct Visitor; -- MST
[Qemu-devel] [PATCH v4 04/23] qom: add pointer to int property helpers
Make it easy to add read-only helpers for simple
integer properties in memory.
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
include/qom/object.h | 21
qom/object.c | 56
2 files changed, 77 insertions(+)
diff --git a/include/qom/object.h b/include/qom/object.h
index e3a16e4..3b75f5a 100644
--- a/include/qom/object.h
+++ b/include/qom/object.h
@@ -795,6 +795,27 @@ void object_property_add(Object *obj, const char *name,
const char *type,
void object_property_del(Object *obj, const char *name, Error **errp);
/**
+ * object_property_add_uint8_ptr:
+ * object_property_add_uint16_ptr:
+ * object_property_add_uint32_ptr:
+ * object_property_add_uint64_ptr:
+ * @obj: the object to add a property to
+ * @name: the name of the property
+ * @v: pointer to value
+ *
+ * Add an integer property in memory. This function will add a
+ * property of the appropriate type.
+ */
+void object_property_add_uint8_ptr(Object *obj, const char *name,
+ const uint8_t *v, Error **errp);
+void object_property_add_uint16_ptr(Object *obj, const char *name,
+const uint16_t *v, Error **errp);
+void object_property_add_uint32_ptr(Object *obj, const char *name,
+const uint32_t *v, Error **errp);
+void object_property_add_uint64_ptr(Object *obj, const char *name,
+const uint64_t *v, Error **Errp);
+
+/**
* object_property_find:
* @obj: the object
* @name: the name of the property
diff --git a/qom/object.c b/qom/object.c
index e90e382..b929dc6 100644
--- a/qom/object.c
+++ b/qom/object.c
@@ -1344,6 +1344,62 @@ static char *qdev_get_type(Object *obj, Error **errp)
return g_strdup(object_get_typename(obj));
}
+static void property_get_uint8_ptr(Object *obj, Visitor *v,
+ void *opaque, const char *name,
+ Error **errp)
+{
+visit_type_uint8(v, opaque, name, errp);
+}
+
+static void property_get_uint16_ptr(Object *obj, Visitor *v,
+ void *opaque, const char *name,
+ Error **errp)
+{
+visit_type_uint16(v, opaque, name, errp);
+}
+
+static void property_get_uint32_ptr(Object *obj, Visitor *v,
+ void *opaque, const char *name,
+ Error **errp)
+{
+visit_type_uint32(v, opaque, name, errp);
+}
+
+static void property_get_uint64_ptr(Object *obj, Visitor *v,
+ void *opaque, const char *name,
+ Error **errp)
+{
+visit_type_uint64(v, opaque, name, errp);
+}
+
+void object_property_add_uint8_ptr(Object *obj, const char *name,
+ const uint8_t *v, Error **errp)
+{
+object_property_add(obj, name, uint8, property_get_uint8_ptr,
+NULL, NULL, (void *)v, errp);
+}
+
+void object_property_add_uint16_ptr(Object *obj, const char *name,
+const uint16_t *v, Error **errp)
+{
+object_property_add(obj, name, uint16, property_get_uint16_ptr,
+NULL, NULL, (void *)v, errp);
+}
+
+void object_property_add_uint32_ptr(Object *obj, const char *name,
+const uint32_t *v, Error **errp)
+{
+object_property_add(obj, name, uint32, property_get_uint32_ptr,
+NULL, NULL, (void *)v, errp);
+}
+
+void object_property_add_uint64_ptr(Object *obj, const char *name,
+const uint64_t *v, Error **errp)
+{
+object_property_add(obj, name, uint64, property_get_uint64_ptr,
+NULL, NULL, (void *)v, errp);
+}
+
static void object_instance_init(Object *obj)
{
object_property_add_str(obj, type, qdev_get_type, NULL, NULL);
--
MST
[Qemu-devel] [PATCH] virtio-net: broken RX filtering logic fixed
From: Dmitry Fleytman dfley...@redhat.com
Upon processing of VIRTIO_NET_CTRL_MAC_TABLE_SET command
multicast list overwrites unicast list in mac_table.
This leads to broken logic for both unicast and multicast RX filtering.
Signed-off-by: Dmitry Fleytman dfley...@redhat.com
---
hw/net/virtio-net.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/hw/net/virtio-net.c b/hw/net/virtio-net.c
index dd41008..e822ab1 100644
--- a/hw/net/virtio-net.c
+++ b/hw/net/virtio-net.c
@@ -656,7 +656,8 @@ static int virtio_net_handle_mac(VirtIONet *n, uint8_t cmd,
}
if (n-mac_table.in_use + mac_data.entries = MAC_TABLE_ENTRIES) {
-s = iov_to_buf(iov, iov_cnt, 0, n-mac_table.macs,
+s = iov_to_buf(iov, iov_cnt, 0,
+ n-mac_table.macs[n-mac_table.in_use * ETH_ALEN],
mac_data.entries * ETH_ALEN);
if (s != mac_data.entries * ETH_ALEN) {
goto error;
--
1.8.3.1
[Qemu-devel] the way i read the registers may wrong
Hi all I am using the following code to read registers, i can read the value
of EIP, but when i dump the value pointed by EIP, all are zero, byte code
should not be zero, so i think my code is wrong. Please point me out thanks?
static void gkd_read_registers(CPUState *cpu, char *buffer) {
printf(gkd_read_registers()\n);
CPUArchState *env = cpu-env_ptr;
sprintf(buffer,
cs=%x,eip=%x,ds=%x,es=%x,fs=%x,gs=%x,ss=%x,eflags=%x,eax=%x,ebx=%x,ecx=%x,edx=%x,esi=%x,edi=%x,ebp=%x,esp=%x,cr0=%x,cr2=%x,cr3=%x,cr4=%x,
env-segs[R_CS].base, env-eip, env-segs[R_DS].base,
env-segs[R_ES].base,
env-segs[R_FS].base, env-segs[R_GS].base,
env-segs[R_SS].base,
env-eflags, env-regs[R_EAX], env-regs[R_EBX],
env-regs[R_ECX], env-regs[R_EDX], env-regs[R_ESI],
env-regs[R_EDI], env-regs[R_EBP], env-regs[R_ESP],
env-cr[0],
env-cr[2], env-cr[3], env-cr[4]);
sprintf(buffer + strlen(buffer),
,dr0=%x,dr1=%x,dr2=%x,dr3=%x,dr6=%x,dr7=%x,
env-dr[0],
env-dr[1], env-dr[2], env-dr[3], env-dr[6],
env-dr[7]);
sprintf(buffer + strlen(buffer), ,gdtr=%x,gdtr_limit=%x,
env-gdt.base,
env-gdt.limit);
sprintf(buffer + strlen(buffer), ,ldtr=%x,ldtr_limit=%x,
env-ldt.base,
env-ldt.limit);
sprintf(buffer + strlen(buffer), ,idtr=%x,idtr_limit=%x,
env-idt.base,
env-idt.limit);
sprintf(buffer + strlen(buffer), ,tr=%x, env-tr);
}
Thanksfrom Peter
Re: [Qemu-devel] [Xen-devel] RESEND [Xen-unstable][Qemu-xen] HVM Guest reading of Expansion ROM from passthroughed PCI device returns data from emulated VGA rom
On Sun, Sep 22, 2013 at 05:00:58PM +0200, Sander Eikelenboom wrote: I'm trying to get secondary vga-passthrough on a HVM guest to work with a AMD HD6570 and the native kernel radeon driver and kernel modesetting. So the guest still gets the emulated stdvga or cirrus device(used in my case here) as primary/boot vga adapter. - When i don't passthrough the radeon card, the linux native radeon driver loads fine. - When i do passtrough the device to a HVM with the same kernel: The driver in the guest tries to read the pci expansion rom from the passthroughed device to get the vbios. The driver reports a successful read, but fails because it can't find the right string at the right offset. Did you try with qemu-dm-traditional aswell? Does it have the same problem? Hi Pasi, Yes i did and yes the same problem. From what i recall i used to have succes with vga passthrough with a secondary vga card, but that was some time ago. I don't know which of the components (xen, dom0 kernel, domU kernel, radeon driver, qemu has changed in such a way that it fails to work now ... But in the mean time i tried to debug it further and from what i can see: - Only the io port en mem of the pci device are mapped through the hypervisor. - The rom is not, (a hypercall to do the memory mapping is never made) i tried several things to get it to do the mapping, but so far failt to do so. - It seems to be a 64bit capable device, some code comments and git commit messages seem to suggest that there were/are some problems with that (in the way the involved components interact) Yeah, there has been some patches related to that, and I think still some open bugs.. Btw did you try with Xen 4.3? Or 4.2? -- Pasi
Re: [Qemu-devel] [RFC 00/16] TCG indirect registers
On Fri, Sep 20, 2013 at 1:24 AM, Richard Henderson r...@twiddle.net wrote:
This is an attempt to improve performance of target-sparc
by exposing the windowed registers as TCG globals, and all
the optimization that we can do there.
This is done via allowing tcg_global_mem_new to be used
with any base pointer, not just off of a fixed register.
Thus the sparc windowed registers are globals off cpu_regwptr.
In the process of working through this, I attempt to remove
as many uses of int as I can throughout the TCG code gen
paths, replacing them with TCGReg when we're talking about
hard registers, and TCGTemp pointers when we're talking about
temporaries. This, IMO, reduces confusion as to what kind of
int we mean at any given time.
By the time we get to patch 14, actually implementing the
indirect temps, it's fairly easy to recurse in order to
load the base pointer when we need to load or store an
indirect temp.
I've not yet tried to measure the performance. As far as
testing, linux-user-0.3 and sparc-test-0.2 works. I've
scanned some of the dumps from those. In the cases where
no real optimization was possible, we generate practically
the same code -- usually with different registers selected.
In the cases where we can optimize, I've seen some TB's
cut in half.
Anyway, I wanted some feedback before I take this any further.
Hi Richard,
I've reimplemented xtensa windowed registers in the same
way as done for sparc on top of this series. Haven't got any
measurable performance change. From op,out_asm output most
TBs got longer by 1-4 instructions and all temp indices got
doubled.
---8---
From 73300be7dd6b3d31cbfa45225714d5e43c52f077 Mon Sep 17 00:00:00 2001
From: Max Filippov jcmvb...@gmail.com
Date: Sun, 22 Sep 2013 18:54:53 +0400
Subject: [PATCH] target-xtensa: reimplement windowed registers
Signed-off-by: Max Filippov jcmvb...@gmail.com
---
target-xtensa/cpu.c | 1 +
target-xtensa/cpu.h | 5 +++--
target-xtensa/op_helper.c | 46 ++
target-xtensa/translate.c | 7 +--
4 files changed, 19 insertions(+), 40 deletions(-)
diff --git a/target-xtensa/cpu.c b/target-xtensa/cpu.c
index c19d17a..a30511d 100644
--- a/target-xtensa/cpu.c
+++ b/target-xtensa/cpu.c
@@ -59,6 +59,7 @@ static void xtensa_cpu_reset(CPUState *s)
env-sregs[CACHEATTR] = 0x;
env-sregs[ATOMCTL] = xtensa_option_enabled(env-config,
XTENSA_OPTION_ATOMCTL) ? 0x28 : 0x15;
+rotate_window_abs(env, env-sregs[WINDOW_BASE]);
env-pending_irq_level = 0;
reset_mmu(env);
diff --git a/target-xtensa/cpu.h b/target-xtensa/cpu.h
index 95103e9..8100f18 100644
--- a/target-xtensa/cpu.h
+++ b/target-xtensa/cpu.h
@@ -334,11 +334,11 @@ typedef struct XtensaConfigList {
typedef struct CPUXtensaState {
const XtensaConfig *config;
-uint32_t regs[16];
+uint32_t *regs;
uint32_t pc;
uint32_t sregs[256];
uint32_t uregs[256];
-uint32_t phys_regs[MAX_NAREG];
+uint32_t phys_regs[MAX_NAREG + 12];
float32 fregs[16];
float_status fp_status;
@@ -396,6 +396,7 @@ void xtensa_timer_irq(CPUXtensaState *env, uint32_t id,
uint32_t active);
void xtensa_rearm_ccompare_timer(CPUXtensaState *env);
int cpu_xtensa_signal_handler(int host_signum, void *pinfo, void *puc);
void xtensa_cpu_list(FILE *f, fprintf_function cpu_fprintf);
+void rotate_window_abs(CPUXtensaState *env, uint32_t position);
void xtensa_sync_window_from_phys(CPUXtensaState *env);
void xtensa_sync_phys_from_window(CPUXtensaState *env);
uint32_t xtensa_tlb_get_addr_mask(const CPUXtensaState *env, bool dtlb,
uint32_t way);
diff --git a/target-xtensa/op_helper.c b/target-xtensa/op_helper.c
index cf97025..ee21550 100644
--- a/target-xtensa/op_helper.c
+++ b/target-xtensa/op_helper.c
@@ -166,39 +166,6 @@ uint32_t HELPER(nsau)(uint32_t v)
return v ? clz32(v) : 32;
}
-static void copy_window_from_phys(CPUXtensaState *env,
-uint32_t window, uint32_t phys, uint32_t n)
-{
-assert(phys env-config-nareg);
-if (phys + n = env-config-nareg) {
-memcpy(env-regs + window, env-phys_regs + phys,
-n * sizeof(uint32_t));
-} else {
-uint32_t n1 = env-config-nareg - phys;
-memcpy(env-regs + window, env-phys_regs + phys,
-n1 * sizeof(uint32_t));
-memcpy(env-regs + window + n1, env-phys_regs,
-(n - n1) * sizeof(uint32_t));
-}
-}
-
-static void copy_phys_from_window(CPUXtensaState *env,
-uint32_t phys, uint32_t window, uint32_t n)
-{
-assert(phys env-config-nareg);
-if (phys + n = env-config-nareg) {
-memcpy(env-phys_regs + phys, env-regs + window,
-n * sizeof(uint32_t));
-} else {
-uint32_t n1 = env-config-nareg - phys;
-memcpy(env-phys_regs + phys, env-regs + window,
-n1 * sizeof(uint32_t));
-memcpy(env-phys_regs, env-regs + window + n1,
-(n -
Re: [Qemu-devel] [PATCH v4 22/23] hpet: add API to find it
Il 22/09/2013 15:38, Michael S. Tsirkin ha scritto:
Add API to find HPET using QOM.
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
include/hw/timer/hpet.h | 2 ++
hw/timer/hpet.c | 5 +
2 files changed, 7 insertions(+)
diff --git a/include/hw/timer/hpet.h b/include/hw/timer/hpet.h
index 757f79f..ab44bd3 100644
--- a/include/hw/timer/hpet.h
+++ b/include/hw/timer/hpet.h
@@ -71,4 +71,6 @@ struct hpet_fw_config
} QEMU_PACKED;
extern struct hpet_fw_config hpet_cfg;
+
+bool hpet_find(void);
#endif
diff --git a/hw/timer/hpet.c b/hw/timer/hpet.c
index fcd22ae..676bd7d 100644
--- a/hw/timer/hpet.c
+++ b/hw/timer/hpet.c
@@ -757,6 +757,11 @@ static void hpet_device_class_init(ObjectClass *klass,
void *data)
dc-props = hpet_device_properties;
}
+bool hpet_find(void)
+{
+return object_resolve_path_type(, hpet, NULL);
s/hpet/TYPE_HPET/
Paolo
+}
+
static const TypeInfo hpet_device_info = {
.name = TYPE_HPET,
.parent= TYPE_SYS_BUS_DEVICE,
Re: [Qemu-devel] [PATCH v4 22/23] hpet: add API to find it
On Sun, Sep 22, 2013 at 09:22:09PM +0200, Paolo Bonzini wrote:
Il 22/09/2013 15:38, Michael S. Tsirkin ha scritto:
Add API to find HPET using QOM.
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
include/hw/timer/hpet.h | 2 ++
hw/timer/hpet.c | 5 +
2 files changed, 7 insertions(+)
diff --git a/include/hw/timer/hpet.h b/include/hw/timer/hpet.h
index 757f79f..ab44bd3 100644
--- a/include/hw/timer/hpet.h
+++ b/include/hw/timer/hpet.h
@@ -71,4 +71,6 @@ struct hpet_fw_config
} QEMU_PACKED;
extern struct hpet_fw_config hpet_cfg;
+
+bool hpet_find(void);
#endif
diff --git a/hw/timer/hpet.c b/hw/timer/hpet.c
index fcd22ae..676bd7d 100644
--- a/hw/timer/hpet.c
+++ b/hw/timer/hpet.c
@@ -757,6 +757,11 @@ static void hpet_device_class_init(ObjectClass *klass,
void *data)
dc-props = hpet_device_properties;
}
+bool hpet_find(void)
+{
+return object_resolve_path_type(, hpet, NULL);
s/hpet/TYPE_HPET/
Paolo
Right. I'll fix it in my tree, won't repost just for this.
Thanks!
+}
+
static const TypeInfo hpet_device_info = {
.name = TYPE_HPET,
.parent= TYPE_SYS_BUS_DEVICE,
Re: [Qemu-devel] [PATCH] sPAPR: implement route_intx_to_irq to get gsi of pci device.
Am 22.09.2013 um 13:47 schrieb Liu Ping Fan qemul...@gmail.com: This is useful when pci assignment happens on sPAPR. This patch doesn't sound useful on its own to me, thus probably belongs in a greater patch set. And without even a clear commit message that explains why exactly this is going to be useful eventually I don't see any reason to apply this patch. Alex
Re: [Qemu-devel] [PATCH 1/2] tests: build the helper program in main build process
22.09.2013 12:30, Wenchao Xia пишет: This is a quick way to update helper program when qemu main code is changed or built, instead of adding new Makefile under test/qemu-iotest. Why? [] +# Build the help program automatically + +all: $(QEMU_IOTESTS_HELPERS-y) This appears to be somewhat awkward way (to do something which I clearly don't see). First, it is not clear why this iotest helper should be built by default when we hasn't been asked to perform the tests. Second, it apparently does not achieve the (unknown) goal anyway. At least the patch does not do what the commit coment says. Commit says that the helper gets updated when qemu main code is changed or built (which is main code?). But when I build, say, subdir-x86_64-softmmu, which is definitely some main code, the helper isn't being built... Count me confused... :) /mjt
Re: [Qemu-devel] [PATCH] .gitignore: ignore tests/qemu-iotests/socket_scm_helper
Reviewed-by: Wenchao Xiaxiaw...@linux.vnet.ibm.com Signed-off-by: Fam Zhengf...@redhat.com --- tests/qemu-iotests/.gitignore | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/qemu-iotests/.gitignore b/tests/qemu-iotests/.gitignore index 62b4002..0541f80 100644 --- a/tests/qemu-iotests/.gitignore +++ b/tests/qemu-iotests/.gitignore @@ -2,6 +2,7 @@ check.log check.time *.out.bad *.notrun +socket_scm_helper # ignore everything in the scratch directory scratch/
Re: [Qemu-devel] [PATCH 1/2] tests: build the helper program in main build process
On 09/23/2013 04:36 AM, Michael Tokarev wrote: 22.09.2013 12:30, Wenchao Xia пишет: This is a quick way to update helper program when qemu main code is changed or built, instead of adding new Makefile under test/qemu-iotest. Why? [] I think it is a bit overkill to have a new Makefile for one single .c file. If more file comes, we can write Makefile at that time. +# Build the help program automatically + +all: $(QEMU_IOTESTS_HELPERS-y) This appears to be somewhat awkward way (to do something which I clearly don't see). First, it is not clear why this iotest helper should be built by default when we hasn't been asked to perform the tests. Usually we may: 1 configure and make. 2 goto tests/qemu-iotests, ./check. It may encounter an error since build of the helper was not triggered. Second, it apparently does not achieve the (unknown) goal anyway. At least the patch does not do what the commit coment says. Commit says that the helper gets updated when qemu main code is changed or built (which is main code?). But when I build, say, subdir-x86_64-softmmu, which is definitely some main code, the helper isn't being built... Main code may refer to the codes other than ./tests, the commit message is not accurate, I think it should be build the helper by default. Count me confused... :) /mjt
Re: [Qemu-devel] [PATCH] sPAPR: implement route_intx_to_irq to get gsi of pci device.
On Mon, Sep 23, 2013 at 4:02 AM, Alexander Graf ag...@suse.de wrote: Am 22.09.2013 um 13:47 schrieb Liu Ping Fan qemul...@gmail.com: This is useful when pci assignment happens on sPAPR. This patch doesn't sound useful on its own to me, thus probably belongs in a greater patch set. Yes, I think, it will be applied after Alexey Kardashevskiy's patch-set, [PATCH v4 00/12] xics: reworks and in-kernel support CCing Alexey Kardashevskiy And without even a clear commit message that explains why exactly this is going to be useful eventually I don't see any reason to apply this patch. Pci assignment adopts irqfd in kernel, which need gsi as input param, right? Thanks, Pingfan
Re: [Qemu-devel] [PATCH] sPAPR: implement route_intx_to_irq to get gsi of pci device.
On 09/22/2013 09:47 PM, Liu Ping Fan wrote:
This is useful when pci assignment happens on sPAPR.
I have almost the same patch in my queue already, it will enable irqfd for
both INTX and MSI, I am just waiting till in-kernel XICS patchset gets in
upstream and then I'll post it.
Signed-off-by: Liu Ping Fan pingf...@linux.vnet.ibm.com
---
This patch will apply on patches which enable xics in kernel.
---
hw/intc/xics.c| 5 +
hw/ppc/spapr_pci.c| 14 ++
include/hw/ppc/xics.h | 1 +
3 files changed, 20 insertions(+)
diff --git a/hw/intc/xics.c b/hw/intc/xics.c
index bb018d1..02cdab8 100644
--- a/hw/intc/xics.c
+++ b/hw/intc/xics.c
@@ -442,6 +442,11 @@ void xics_set_irq_type(XICSState *icp, int irq, bool lsi)
icp-ics-islsi[irq - icp-ics-offset] = lsi;
}
+int xics_get_irq_offset(XICSState *icp)
+{
+return icp-ics-offset;
+}
+
/*
* Guest interfaces
*/
diff --git a/hw/ppc/spapr_pci.c b/hw/ppc/spapr_pci.c
index 9b6ee32..6d3657a 100644
--- a/hw/ppc/spapr_pci.c
+++ b/hw/ppc/spapr_pci.c
@@ -432,6 +432,19 @@ static void pci_spapr_set_irq(void *opaque, int irq_num,
int level)
qemu_set_irq(spapr_phb_lsi_qirq(phb, irq_num), level);
}
+static PCIINTxRoute spapr_phb_route_intx_to_irq(void *opaque, int pirq_pin)
+{
+int gsi;
+PCIINTxRoute route;
+sPAPRPHBState *phb = opaque;
+
+gsi = phb-lsi_table[pirq_pin].irq;
+gsi += xics_get_irq_offset(spapr-icp);
Why do you need this? lsi_table[].irq is received from spapr_allocate_lsi()
which already adds this offset.
+route.mode = PCI_INTX_ENABLED;
+route.irq = gsi;
+return route;
+}
+
/*
* MSI/MSIX memory region implementation.
* The handler handles both MSI and MSIX.
@@ -595,6 +608,7 @@ static int spapr_phb_init(SysBusDevice *s)
pci_spapr_set_irq, pci_spapr_map_irq, sphb,
sphb-memspace, sphb-iospace,
PCI_DEVFN(0, 0), PCI_NUM_PINS, TYPE_PCI_BUS);
+pci_bus_set_route_irq_fn(bus, spapr_phb_route_intx_to_irq);
phb-bus = bus;
sphb-dma_window_start = 0;
diff --git a/include/hw/ppc/xics.h b/include/hw/ppc/xics.h
index 66364c5..6ed1f4d 100644
--- a/include/hw/ppc/xics.h
+++ b/include/hw/ppc/xics.h
@@ -97,6 +97,7 @@ struct ICSIRQState {
qemu_irq xics_get_qirq(XICSState *icp, int irq);
void xics_set_irq_type(XICSState *icp, int irq, bool lsi);
+int xics_get_irq_offset(XICSState *icp);
void xics_cpu_setup(XICSState *icp, PowerPCCPU *cpu);
--
Alexey
Re: [Qemu-devel] [PATCH v4 00/12] xics: reworks and in-kernel support
On 09/16/2013 02:10 PM, Alexey Kardashevskiy wrote: On 09/10/2013 02:26 PM, Alexey Kardashevskiy wrote: On 09/04/2013 12:56 PM, Alexey Kardashevskiy wrote: On 08/30/2013 03:28 PM, Alexey Kardashevskiy wrote: Yet another try with XICS and XICS-KVM. v3-v4: Addressed multiple comments from Alex; Split out many tiny patches to make them easier to review; Fixed xics_cpu_setup not to call the parent; And many, many small changes. v2-v3: Addressed multiple comments from Andreas; Added 2 patches for XICS from Ben - I included them into the series as they are about XICS and they won't rebase automatically if moved before XICS rework so it seemed to me that it would be better to carry them toghether. If it is wrong, please let me know, I'll repost them separately. v1-v2: The main change is this adds xics-common parent for emulated XICS and XICS-KVM. And many, many small changes, mostly to address Andreas comments. Migration from XICS to XICS-KVM and vice versa still works. Alexey Kardashevskiy (8): xics: move reset and cpu_setup spapr: move cpu_setup after kvmppc_set_papr xics: replace fprintf with error_report xics: add pre_save/post_load dispatchers xics: convert init() to realize() xics: add missing const specifiers to TypeInfo xics: split to xics and xics-common xics: add cpu_setup callback Benjamin Herrenschmidt (2): xics: Implement H_IPOLL xics: Implement H_XIRR_X David Gibson (2): target-ppc: Add helper for KVM_PPC_RTAS_DEFINE_TOKEN xics-kvm: Support for in-kernel XICS interrupt controller default-configs/ppc64-softmmu.mak | 1 + hw/intc/Makefile.objs | 1 + hw/intc/xics.c| 331 +- hw/intc/xics_kvm.c| 488 ++ hw/ppc/spapr.c| 27 ++- include/hw/ppc/spapr.h| 1 + include/hw/ppc/xics.h | 57 + target-ppc/kvm.c | 14 ++ target-ppc/kvm_ppc.h | 7 + 9 files changed, 865 insertions(+), 62 deletions(-) create mode 100644 hw/intc/xics_kvm.c Alex, ping? Anyone, ping? Anyone, ping? *sigh*. ping. -- Alexey
Re: [Qemu-devel] [PATCH 5/5] hw: arm_gic_kvm: Add KVM VGIC save/restore logic
On Sat, Sep 21, 2013 at 06:38:19PM +0900, Peter Maydell wrote: On 21 September 2013 06:46, Christoffer Dall christoffer.d...@linaro.org wrote: On Sat, Sep 21, 2013 at 06:22:23AM +0900, Peter Maydell wrote: *) for getting TCG-KVM and KVM-with-non-host-CPU cases right we need to do translation anyway, or at least think about it. Why? Wouldn't we always only support the case where QEMU emulates the same model as KVM in the first case, and the kernel should behave the same and export the same state if you ask for a specific target no matter what the underlying hardware is, no? If the kernel has to be able to do translation of the state, why not make its life easier by having it only need to do one thing (host h/w format - whatever standard format we pick) rather than lots and lots of things (host CPU X h/w format - format for supported guest CPU A, host CPU X h/w format - format for supported guest CPU B, host CPU Y h/w format - format for guest CPU A, host CPU Y h/w format - format for guest CPU B, etc etc etc) ? That's basically a cross product over every CPU we support. Good point. So after reading the GIC specs again, the way I understand it is that the APR regs have a bit set, if that group priority (a.k.a. preemption level) has an active interrupt. Further, multiple set bits would would only happen if software acknowledges an interrupt and before EOIing it, the GIC gets preempted by an interrupt with a higher group priority (lower number). Correct? Further, and again, I don't think the spec is particularly clear on this point, but I think it suggests that if bit[0] is set, then there's an interrupt from interrupt priority group 0 (the group with the highest priority) in the active state, if bit[1] is set, one from group 1 is active, and so on. That would be a perfectly fine format for the APR in the GICstate structure, and the only remaining questions would be: (1) How many preemption levels should be supported, which would be most easily solved by just defining GICC_APR0-GICC_APR3 for all cpu interfaces. (2) How does the arm_gic_kvm.c code detect the underlying host CPU that the kernel read the register from when it returned the value of the register to do the proper translation? I don't even want to think about how this will work on Big.Little... -Christoffer
Re: [Qemu-devel] [PATCH] sPAPR: implement route_intx_to_irq to get gsi of pci device.
On Mon, Sep 23, 2013 at 9:59 AM, Alexey Kardashevskiy a...@ozlabs.ru wrote:
On 09/22/2013 09:47 PM, Liu Ping Fan wrote:
This is useful when pci assignment happens on sPAPR.
I have almost the same patch in my queue already, it will enable irqfd for
both INTX and MSI, I am just waiting till in-kernel XICS patchset gets in
upstream and then I'll post it.
Ok,
Signed-off-by: Liu Ping Fan pingf...@linux.vnet.ibm.com
---
This patch will apply on patches which enable xics in kernel.
---
hw/intc/xics.c| 5 +
hw/ppc/spapr_pci.c| 14 ++
include/hw/ppc/xics.h | 1 +
3 files changed, 20 insertions(+)
diff --git a/hw/intc/xics.c b/hw/intc/xics.c
index bb018d1..02cdab8 100644
--- a/hw/intc/xics.c
+++ b/hw/intc/xics.c
@@ -442,6 +442,11 @@ void xics_set_irq_type(XICSState *icp, int irq, bool
lsi)
icp-ics-islsi[irq - icp-ics-offset] = lsi;
}
+int xics_get_irq_offset(XICSState *icp)
+{
+return icp-ics-offset;
+}
+
/*
* Guest interfaces
*/
diff --git a/hw/ppc/spapr_pci.c b/hw/ppc/spapr_pci.c
index 9b6ee32..6d3657a 100644
--- a/hw/ppc/spapr_pci.c
+++ b/hw/ppc/spapr_pci.c
@@ -432,6 +432,19 @@ static void pci_spapr_set_irq(void *opaque, int
irq_num, int level)
qemu_set_irq(spapr_phb_lsi_qirq(phb, irq_num), level);
}
+static PCIINTxRoute spapr_phb_route_intx_to_irq(void *opaque, int pirq_pin)
+{
+int gsi;
+PCIINTxRoute route;
+sPAPRPHBState *phb = opaque;
+
+gsi = phb-lsi_table[pirq_pin].irq;
+gsi += xics_get_irq_offset(spapr-icp);
Why do you need this? lsi_table[].irq is received from spapr_allocate_lsi()
which already adds this offset.
Oh, you are right, next_irq begin at XICS_IRQ_BASE
Thx,
Pingfan
+route.mode = PCI_INTX_ENABLED;
+route.irq = gsi;
+return route;
+}
+
/*
* MSI/MSIX memory region implementation.
* The handler handles both MSI and MSIX.
@@ -595,6 +608,7 @@ static int spapr_phb_init(SysBusDevice *s)
pci_spapr_set_irq, pci_spapr_map_irq, sphb,
sphb-memspace, sphb-iospace,
PCI_DEVFN(0, 0), PCI_NUM_PINS, TYPE_PCI_BUS);
+pci_bus_set_route_irq_fn(bus, spapr_phb_route_intx_to_irq);
phb-bus = bus;
sphb-dma_window_start = 0;
diff --git a/include/hw/ppc/xics.h b/include/hw/ppc/xics.h
index 66364c5..6ed1f4d 100644
--- a/include/hw/ppc/xics.h
+++ b/include/hw/ppc/xics.h
@@ -97,6 +97,7 @@ struct ICSIRQState {
qemu_irq xics_get_qirq(XICSState *icp, int irq);
void xics_set_irq_type(XICSState *icp, int irq, bool lsi);
+int xics_get_irq_offset(XICSState *icp);
void xics_cpu_setup(XICSState *icp, PowerPCCPU *cpu);
--
Alexey
