Also, add weak symbols that will be used if qdev-system.o is not
compiled in (i.e. on *-user).
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
Changes v1 - v2:
- Add GCC_WEAK_DECL to function declarations
---
hw/qdev-core.c | 16 +++-
hw/qdev-core.h | 3 +++
hw/qdev
On Wed, Oct 17, 2012 at 01:00:55PM -0500, Anthony Liguori wrote:
Igor Mammedov imamm...@redhat.com writes:
From: Eduardo Habkost ehabk...@redhat.com
This change should help on two things:
- Allowing DeviceState to be used by *-user;
- Writing qdev unit tests without pulling too many
On Wed, Oct 17, 2012 at 01:08:23PM -0500, Anthony Liguori wrote:
Eduardo Habkost ehabk...@redhat.com writes:
The reset register/unregister code is specific to qemu-system-*, so
isolate it so it can be moved to qdev-system.c.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
hw
On Wed, Oct 17, 2012 at 09:40:59PM +0200, Michael S. Tsirkin wrote:
Enable KVM PV EOI by default. You can still disable it with
-kvm_pv_eoi cpu flag. To avoid breaking cross-version migration,
enable only for qemu 1.3 or newer machine type.
Signed-off-by: Michael S. Tsirkin m...@redhat.com
On Wed, Jul 15, 2009 at 01:43:31PM +0200, Gerd Hoffmann wrote:
[...]
diff --git a/hw/qdev-properties.c b/hw/qdev-properties.c
new file mode 100644
index 000..8b0d0ff
--- /dev/null
+++ b/hw/qdev-properties.c
@@ -0,0 +1,246 @@
Gerd, could you clarify what's the copyright/license of this
On Wed, Oct 17, 2012 at 10:48:58PM +0200, Michael S. Tsirkin wrote:
On Wed, Oct 17, 2012 at 05:03:02PM -0300, Eduardo Habkost wrote:
On Wed, Oct 17, 2012 at 09:40:59PM +0200, Michael S. Tsirkin wrote:
Enable KVM PV EOI by default. You can still disable it with
-kvm_pv_eoi cpu flag
On Wed, Oct 17, 2012 at 11:05:22PM +0200, Stefan Weil wrote:
On Wed, Jul 15, 2009 at 01:43:31PM +0200, Gerd Hoffmann wrote:
[...]
diff --git a/hw/qdev-properties.c b/hw/qdev-properties.c
new file mode 100644
index 000..8b0d0ff
--- /dev/null
+++ b/hw/qdev-properties.c
@@ -0,0
On Wed, Oct 17, 2012 at 11:17:26PM +0200, Andre Przywara wrote:
Update QEMU's knowledge of CPUID bit names. This allows to
enable/disable those new features on QEMU's command line when
using KVM and prepares future feature enablement in QEMU.
This adds F16C, RDRAND, LWP, TBM, TopoExt,
On Fri, Oct 19, 2012 at 01:34:19AM +0200, Andre Przywara wrote:
On 10/18/12 18:33, Eduardo Habkost wrote:
On Wed, Oct 17, 2012 at 11:17:26PM +0200, Andre Przywara wrote:
Update QEMU's knowledge of CPUID bit names. This allows to
enable/disable those new features on QEMU's command line when
On Fri, Oct 19, 2012 at 09:31:00AM +0200, Markus Armbruster wrote:
Eduardo Habkost ehabk...@redhat.com writes:
On Wed, Oct 17, 2012 at 01:00:55PM -0500, Anthony Liguori wrote:
[...]
I don't really understand the split here and the 'system' suffix really
doesn't explain it for me. Could
On Wed, Oct 17, 2012 at 08:21:17PM +0100, Peter Maydell wrote:
On 17 October 2012 19:11, Anthony Liguori anth...@codemonkey.ws wrote:
It's dangerously close to bike-shedding, but i don't think qdev belongs
in qom/. It's not core infrastructure. It's the device base class and
belongs IMHO
On Mon, Oct 22, 2012 at 03:12:00PM +0200, Michael S. Tsirkin wrote:
On Mon, Oct 22, 2012 at 12:06:42PM +0200, Andreas Färber wrote:
Am 18.10.2012 17:01, schrieb Michael S. Tsirkin:
Enable KVM PV EOI by default. You can still disable it with
-kvm_pv_eoi cpu flag. To avoid breaking
m...@redhat.com
Reviewed-by: Eduardo Habkost ehabk...@redhat.com
---
Changes from v2:
Address comments by Andreas:
whitespace fixes and moving function around
Changes from v1:
Address comments by Eduardo:
use include instead of duplicate definition
On Mon, Oct 22, 2012 at 02:36:01PM +0200, Igor Mammedov wrote:
On Tue, 16 Oct 2012 16:08:42 -0300
Eduardo Habkost ehabk...@redhat.com wrote:
[...]
diff --git a/qom/qdev-properties.c b/qom/qdev-properties.c
new file mode 100644
index 000..2e82cb9
--- /dev/null
+++ b/qom/qdev
On Tue, Oct 23, 2012 at 03:26:40PM +0200, Igor Mammedov wrote:
convert x86_cpu_list() to use QDEV_FIND_PROP_FROM_BIT() for getting
CPUID feature name.
In addition since x86_cpu_list() was the last user of *feature_name
arrays, clean them up.
Signed-off-by: Igor Mammedov imamm...@redhat.com
On Wed, Oct 17, 2012 at 01:08:23PM -0500, Anthony Liguori wrote:
Eduardo Habkost ehabk...@redhat.com writes:
The reset register/unregister code is specific to qemu-system-*, so
isolate it so it can be moved to qdev-system.c.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
hw
On Tue, Oct 23, 2012 at 12:56:18PM -0200, Eduardo Habkost wrote:
On Wed, Oct 17, 2012 at 01:08:23PM -0500, Anthony Liguori wrote:
Eduardo Habkost ehabk...@redhat.com writes:
The reset register/unregister code is specific to qemu-system-*, so
isolate it so it can be moved to qdev
.
This simply adds a comment documenting the rules about qemu-common.h,
hoping that people will see it before including qemu-common.h from other
header files, and before adding more declarations to qemu-common.h.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
qemu-common.h | 11
Include:
- glib.h for g_malloc0()
- string.h for memset()
Some of those headers were probably being included by accident because
some other headers were including qemu-common.h, but those headers
should eventually stop including qemu-common.h.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
module.h is where machine_init() is defined, but qemu-fsdev-dummy.c
doesn't include it.
The header is probably being included by accident because some other
headers are including qemu-common.h, but those headers should eventually
stop including qemu-common.h.
Signed-off-by: Eduardo Habkost ehabk
of the list.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
The only reason I am doing this is because:
- I am going to mvoe some of the qdev-properties.c file to another file
- I need to add a copyright/license header to the new file.
See http://article.gmane.org
It just needs the Monitor and DeviceState typedefs, so it doesn't need
all of qemu-common.h.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
sysemu.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/sysemu.h b/sysemu.h
index 0c39a3a..2b103a5 100644
--- a/sysemu.h
+++ b
.
The copyright/license header for the new file is directly copied from
qdev-properties.c.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
[imammedo: keep qdev_get_child_bus() in hw/qdev.c]
[imammedo: put qdev_set_nic_properties() in hw/qdev-properties-system.c]
Signed-off-by: Igor Mammedov imamm
stdbool.h is needed for the 'bool' type, used in the header.
The header is probably being included by accident because some other
headers are including qemu-common.h, but those headers should eventually
stop including qemu-common.h.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
ui/vnc
Include:
- glib.h for g_malloc0()
- string.h for strcmp()
Some of those headers were probably being included by accident because
some other headers were including qemu-common.h, but those headers
should eventually stop including qemu-common.h.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
qdev_exists() declaration]
[ehabkost: keep using 'QemuOpts' instead of 'struct QemuOpts', as
qdev-core.h includes qemu-option.h]
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
Yes, there is changelog data before the --- mark, but I believe that
in this case they are important to indicate
cpus.o, and depends on the CPU list from exec.o).
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
hw/qdev.c | 8 +---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/hw/qdev.c b/hw/qdev.c
index 7ddcd24..2cc6434 100644
--- a/hw/qdev.c
+++ b/hw/qdev.c
@@ -109,10 +109,12
-by: Eduardo Habkost ehabk...@redhat.com
---
qga/channel-posix.c | 4
1 file changed, 4 insertions(+)
diff --git a/qga/channel-posix.c b/qga/channel-posix.c
index 57eea06..1539ed3 100644
--- a/qga/channel-posix.c
+++ b/qga/channel-posix.c
@@ -1,5 +1,9 @@
#include glib.h
#include termios.h
From: Igor Mammedov imamm...@redhat.com
needed to prevent build breakage when CPU becomes a child of DeviceState
Signed-off-by: Igor Mammedov imamm...@redhat.com
[ehabkost: include stdbool.h too]
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
Yes, there is changelog data before
The header file is specific for *-user, but I plan to introduce a more
generic qemu-types.h file, so I'm renaming it.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
Cc: Blue Swirl blauwir...@gmail.com
Cc: Riku Voipio riku.voi...@iki.fi
---
bsd-user/qemu.h | 2 +-
cpu
files that use qemu_register_reset().
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
hw/hw.h | 6 +-
hw/qdev-reset.h | 11 +++
hw/qdev.c | 41 +
hw/qdev.h | 1 +
sysemu.h| 1 -
vl.c| 40
From: Igor Mammedov imamm...@redhat.com
Signed-off-by: Igor Mammedov imamm...@redhat.com
[ehabkost: change CPU type declaration to hae TYPE_DEVICE as parent]
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
Yes, there is changelog data before the --- mark, but I believe that
in this case
This version will be used on cases where sysbus.c is not compiled in
(e.g. *-user).
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
hw/qdev-core.h | 2 +-
hw/qdev.c | 12
2 files changed, 13 insertions(+), 1 deletion(-)
diff --git a/hw/qdev-core.h b/hw/qdev-core.h
Include:
- stdio.h for FILE
- qemu-option.h for QemuOptsList
Some of those headers were probably being included by accident because
some other headers were including qemu-common.h, but those headers
should eventually stop including qemu-common.h.
Signed-off-by: Eduardo Habkost ehabk
The code depends on some functions from qemu-option.o, so add
qemu-option.o to universal-obj-y to make sure it's included.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
Changes v1 - v2:
- Keep files on the hw/ directory
(it's simply easier to keep them there, as qdev.o depends
to include qdev-core.h only because of
DeviceState and other typedefs.
This will help us remove qemu-common.h dependencies from some headers
later.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
hw/qdev-core.h | 11 +--
qemu-common.h | 53
This will help reduce the qemu-common.h dependency hell.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
--
Changes v1 - v2:
- move qemu_open() qemu_close() to qemu-stdio.h, too
---
qemu-common.h | 59 ++
qemu-stdio.h | 76
it can be used in cpu.h
Eduardo Habkost (20):
user: move *-user/qemu-types.h to main directory
user: rename qemu-types.h to qemu-user-types.h
qemu-common.h: comment about usage rules
move I/O-related definitions from qemu-common.h to a new header
(qemu-stdio.h)
qemu-fsdev-dummy.c: include
I don't know why it was including it, as I don't see any code that
depends on anything from qemu-common.h.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
qlist.h | 1 -
1 file changed, 1 deletion(-)
diff --git a/qlist.h b/qlist.h
index ae776f9..7408947 100644
--- a/qlist.h
+++ b/qlist.h
Add GCC_WEAK vmstate symbols to qdev.c, so that qdev.o can be used
without savevm.o (i.e. by *-user).
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
Originally submitted as:
Subject: qdev-core: isolate vmstate handling into separate functions
Changes v1 - v2:
- Add GCC_WEAK_DECL
The bsd-user/qemu-types.h and linux-user/qemu-types.h files are almost
the same, but linux-user have the additional definitions of tswapal().
This moves the linux-user file to the main directory, so the same file
can be used by linux-user and bsd-user.
Signed-off-by: Eduardo Habkost ehabk
From: Igor Mammedov imamm...@redhat.com
it's necessary for making CPU child of DEVICE without
causing circular header deps.
Signed-off-by: Igor Mammedov imamm...@redhat.com
[ehabkost: re-added the typedef to hw/irq.h after rebasing]
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
Changes
On Wed, Oct 24, 2012 at 10:42:55AM +0200, Markus Armbruster wrote:
Peter Maydell peter.mayd...@linaro.org writes:
A lot of the ARM board models follow the pattern of having a
single common init function which is called with various
parameters from the QEMUMachine init function for several
On Tue, Oct 23, 2012 at 03:56:52PM +0200, Igor Mammedov wrote:
On Tue, 23 Oct 2012 09:32:53 -0400
Don Slutz d...@cloudswitch.com wrote:
On 10/22/12 11:02, Igor Mammedov wrote:
v5:
- Use static properties instead of dynamic ones
- Compile in KVM CPUID features only if
I removed a line by mistake on commit
3b671a40cab2404bc63e57db8cd3afa4ec70bfab, containing the flags lm/i64,
3dnow, and 3dnowext. This patch restores the removed line.
Reported-by: Don Slutz d...@cloudswitch.com
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
target-i386/cpu.c | 1 +
1
On Tue, Oct 23, 2012 at 01:23:43PM -0400, Don Slutz wrote:
Turns out that patch #32 is the one that causes the command line:
./x86_64-softmmu/qemu-system-x86_64 -cpu
This will allow each architecture to define how the VCPU ID is set on
the KVM_CREATE_VCPU ioctl call.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
kvm-all.c | 2 +-
kvm.h | 3 +++
target-i386/kvm.c | 5 +
target-ppc/kvm.c | 5 +
target-s390x/kvm.c | 5
This will help reduce the qemu-common.h dependency hell.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
--
Changes v1 - v2:
- move qemu_open() qemu_close() to qemu-stdio.h, too
---
qemu-common.h | 59 ++
qemu-stdio.h | 76
This changes FW_CFG_MAX_CPUS and FW_CFG_NUMA to use apic_id_for_cpu(),
so the NUMA table can be based on the APIC IDs, instead of CPU index
(SeaBIOS knows nothing about CPU indexes, just APIC IDs).
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
Changes v1 - v2:
- Get PC object
This will make the code simpler when the CPU initialization code start using
additional initialization parameters.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
hw/pc.c | 8
hw/pc.h | 2 +-
hw/pc_piix.c | 3 +--
3 files changed, 6 insertions(+), 7 deletions(-)
diff
PC will not use max_cpus for that field, so move it outside the common
code so it can use a different value on PC.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
hw/fw_cfg.c | 1 -
hw/pc.c | 2 +-
hw/ppc_newworld.c | 1 +
hw/ppc_oldworld.c | 1 +
hw/sun4m.c| 3
This keeps compatibility on machine-types pc-1.2 and older, and prints a
warning in case the requested configuration won't get the correct
topology.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
Changes v1 - v2:
- Move code to cpu.c
- keep using cpu_index on *-user
- Use
Has anybody looked at this series?
Avi, Marcelo?
On Thu, Oct 04, 2012 at 05:48:52PM -0300, Eduardo Habkost wrote:
Most of this series are just cleanups that will help when making -cpu
check/enforce work properly, with some fixes.
In addition to code movements, the main changes
This code is not specific to the Bochs BIOS, so rename it.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
hw/pc.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/hw/pc.c b/hw/pc.c
index 4e971c8..c38d8b2 100644
--- a/hw/pc.c
+++ b/hw/pc.c
@@ -570,7 +570,7 @@ int
The code will get more complex, so put it outside the for loop, into a
separate function.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
hw/pc.c | 13 +
1 file changed, 9 insertions(+), 4 deletions(-)
diff --git a/hw/pc.c b/hw/pc.c
index 0e01003..85eab04 100644
--- a/hw/pc.c
When pci_enabled is false, rom_memory is already pointing to
system_memory, so there's no need to check pci_enabled again when
calling pc_memory_init().
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
hw/pc_piix.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw
Instead of changing pc_init1() arguments every time some additional
machine-type-specific behavior has to be introduced, create a struct that will
carry all the information needed by the PC initialization functions.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
hw/pc.h | 8
If x86_cpu_realize() set any errors, print an error message and return
NULL.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
target-i386/helper.c | 8 +++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/target-i386/helper.c b/target-i386/helper.c
index c5d42c5..1e5f61f
creation and the x86_cpu_realize() call.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
hw/xen_machine_pv.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/hw/xen_machine_pv.c b/hw/xen_machine_pv.c
index 4264703..95833f4 100644
--- a/hw/xen_machine_pv.c
+++ b/hw
Instead of having a large list of arguments, just pass the PCInitArgs
struct to pc_memory_init().
This is being done mainly to facilitate the addition of an argument to
be used by bochs_bios_init() (enabling compatibility mode for APIC ID
generation).
Signed-off-by: Eduardo Habkost ehabk
Instead of passing each QEMUMachineInitArgs field individually, just
pass the whole QEMUMachineInitArgs object to pc_init1().
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
hw/pc_piix.c | 47 ---
1 file changed, 12 insertions(+), 35 deletions
From: Igor Mammedov imamm...@redhat.com
(L)APIC is a part of cpu [1] so move APIC initialization inside of
x86_cpu object. Since cpu_model and override flags currently specify
whether APIC should be created or not, APIC creationinitialization is
moved into x86_cpu_apic_init() which is called from
To make unit tests that depend on target-specific files, use
check-unit-arch-y and test-obj-arch-y.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
tests/Makefile | 16 +---
1 file changed, 13 insertions(+), 3 deletions(-)
diff --git a/tests/Makefile b/tests/Makefile
index
- kvm_arch_init_vcpu(), that's called by kvm_init_vcpu()
So it's safe to change it before x86_cpu_realize() is called.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
This is based on the patch that I have originally suybmitted as:
Subject: pc: create apic_id_for_cpu() function (v3)
---
hw/pc.c | 22
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
Changes v1 - v2:
- Coding style change: break too-long line
---
hw/apic.c | 35 ++-
1 file changed, 18 insertions(+), 17 deletions(-)
diff --git a/hw/apic.c b/hw/apic.c
index 49f0015..1772f2c 100644
--- a/hw
I don't know why the code uses void* if the FWCfgState typedef is declared at
the same header where fw_cfg_init() and other fw_cfg_*() functions are
declared. This changes the code to use FWCfgState* instead of void* in the PC
initialization code.
Signed-off-by: Eduardo Habkost ehabk
advanced PC CPU
initialization steps
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
hw/pc.c | 12 +++-
target-i386/cpu.h| 14 ++
target-i386/helper.c | 11 ---
3 files changed, 29 insertions(+), 8 deletions(-)
diff --git a/hw/pc.c b/hw/pc.c
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
Cc: Blue Swirl blauwir...@gmail.com
Changes v1 - v2:
- Support 32-bit APIC IDs (in case x2APIC is going to be used)
- Coding style changes
- Use TARGET_I386_TOPOLOGY_H instead of __QEMU_X86_TOPOLOGY_H__
- Rename topo_make_apic_id
This:
- Renames the init function for pc-1.2 and lower to pc_init_pci_v1_2;
- Creates a pc_init_pci_v1_3 function for pc-1.3.
Right now both functions have exactly the same code, but the following patch
will change pc_init_v1_2 to set compatibility PCInitArgs fields.
Signed-off-by: Eduardo
Make the pc_cpus_init() code more clear, by removing an #ifdef from
the middle of the code.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
hw/pc.c | 12 +++-
1 file changed, 7 insertions(+), 5 deletions(-)
diff --git a/hw/pc.c b/hw/pc.c
index 408823b..67fdbe2 100644
--- a/hw
All calls to pc_init1() pass the same expressions as arguments, so move the
expressions inside pc_init1().
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
hw/pc_piix.c | 18 ++
1 file changed, 6 insertions(+), 12 deletions(-)
diff --git a/hw/pc_piix.c b/hw/pc_piix.c
index
The argument will be used later, when enabling compatibility mode for
the APIC ID generation code.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
hw/pc.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/hw/pc.c b/hw/pc.c
index c38d8b2..0e01003 100644
--- a/hw/pc.c
The CPU ID in KVM is supposed to be the APIC ID, so change the
KVM_CREATE_VCPU call to match it. The current behavior didn't break
anything yet because today the APIC ID is assumed to be == the CPU
index, but this won't be true in the future.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
Needed for the definition of fprint_function.
This is not necessary right now, but it will be necessary if code that
doesn't include cpu-common.h includes cpus.h.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
cpus.h | 2 ++
1 file changed, 2 insertions(+)
diff --git a/cpus.h b/cpus.h
://software.intel.com/en-us/articles/intel-64-architecture-processor-topolog
y-enumeration/
This series is available on git, at:
https://github.com/ehabkost/qemu-hacks/commits/apicid-topology-v3
Eduardo Habkost (26):
move I/O-related definitions from qemu-common.h to a new header
(qemu-stdio.h
This depends on a previous series I have submitted:
Subject: [QEMU PATCH 00/15] QEMU KVM_GET_SUPPORTED_CPUID cleanups and fixes
Message-Id: 1349383747-19383-1-git-send-email-ehabk...@redhat.com
http://article.gmane.org/gmane.comp.emulators.kvm.devel/99375
Eduardo Habkost (3):
target-i385
The return value of that function is always 0, and is always ignored.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
target-i386/cpu.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/target-i386/cpu.c b/target-i386/cpu.c
index 38fca92..42171c9 100644
--- a/target
as such, and rename them
to kvm_check_features_against_host() and kvm_cpu_fill_host().
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
target-i386/cpu.c | 25 ++---
1 file changed, 18 insertions(+), 7 deletions(-)
diff --git a/target-i386/cpu.c b/target-i386/cpu.c
index 42171c9
on
the host CPU (e.g. x2apic);
* check and enforce options were not detecting problems when the
host CPU did support a feature, but the KVM kernel code didn't
support it.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
target-i386/cpu.c | 25 +++--
1 file changed, 15
On Wed, Apr 03, 2013 at 07:58:00PM +0200, Igor Mammedov wrote:
snip
+void do_cpu_hot_add(const int64_t id, Error **errp)
+{
+pc_new_cpu(saved_cpu_model, id, errp);
+}
+
Missing x86_cpu_apic_id_from_index(id)?
There was(is?) opposition to using cpu_index to identify x86
On Wed, Apr 03, 2013 at 08:59:07PM +0200, Igor Mammedov wrote:
On Wed, 3 Apr 2013 15:10:05 -0300
Eduardo Habkost ehabk...@redhat.com wrote:
On Wed, Apr 03, 2013 at 07:58:00PM +0200, Igor Mammedov wrote:
snip
+void do_cpu_hot_add(const int64_t id, Error **errp
On Wed, Apr 03, 2013 at 10:09:25PM +0200, Igor Mammedov wrote:
On Wed, 3 Apr 2013 16:27:11 -0300
Eduardo Habkost ehabk...@redhat.com wrote:
On Wed, Apr 03, 2013 at 08:59:07PM +0200, Igor Mammedov wrote:
On Wed, 3 Apr 2013 15:10:05 -0300
Eduardo Habkost ehabk...@redhat.com wrote
Instead of open-coding the filtering code for each feature word, change
the existing code to use the feature_word_info array, that have exactly
the same CPUID eax/ecx/register values for each feature word.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
target-i386/cpu.c | 24
item[7].cpuid-register: EDX
item[7].cpuid-input-eax: 1
item[7].features: 0
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
target-i386/cpu.c | 11 +++
1 file changed, 7 insertions(+), 4 deletions(-)
diff --git a/target-i386/cpu.c b/target-i386/cpu.c
index 64e5570..0dceaf2
This field will contain the feature bits that were filtered out because
of missing host support.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
target-i386/cpu-qom.h | 3 +++
target-i386/cpu.c | 9 ++---
2 files changed, 9 insertions(+), 3 deletions(-)
diff --git a/target-i386
].cpuid-register: EAX
item[6].cpuid-input-eax: 1073741825
item[6].features: 0
item[7].cpuid-register: EDX
item[7].cpuid-input-eax: 2147483658
item[7].features: 0
Eduardo Habkost (5):
target-i386: Add ECX information to FeatureWordInfo
target-i386: Add feature-words property
target
FEAT_7_0_EBX uses ECX as input, so we have to take that into account
when reporting feature word values.
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
target-i386/cpu.c | 10 +++---
1 file changed, 7 insertions(+), 3 deletions(-)
diff --git a/target-i386/cpu.c b/target-i386/cpu.c
item[7].cpuid-register: EDX
item[7].cpuid-input-eax: 1
item[7].features: 126614521
Signed-off-by: Eduardo Habkost ehabk...@redhat.com
---
Changes v1 - v2:
* Merge the non-qapi and qapi patches, to keep series simpler
* Use the feature word array series as base, so we don't have
to set
On Fri, Apr 05, 2013 at 04:37:10PM +0200, Igor Mammedov wrote:
[...]
void pc_cpus_init(const char *cpu_model)
{
int i;
+X86CPU *cpu;
Error *error = NULL;
+SysBusDevice *ib;
/* init CPUs */
if (cpu_model == NULL) {
@@ -902,14 +907,22 @@ void
On Fri, Apr 05, 2013 at 04:37:16PM +0200, Igor Mammedov wrote:
[...]
diff --git a/qapi-schema.json b/qapi-schema.json
index db542f6..a760ed5 100644
--- a/qapi-schema.json
+++ b/qapi-schema.json
@@ -1387,6 +1387,17 @@
{ 'command': 'cpu', 'data': {'index': 'int'} }
##
+# @cpu-add
+#
On Fri, Apr 05, 2013 at 02:10:54PM -0300, Eduardo Habkost wrote:
On Fri, Apr 05, 2013 at 04:37:16PM +0200, Igor Mammedov wrote:
[...]
diff --git a/qapi-schema.json b/qapi-schema.json
index db542f6..a760ed5 100644
--- a/qapi-schema.json
+++ b/qapi-schema.json
@@ -1387,6 +1387,17
env-mp_state.
So split APIC device creation from its initialization and realize APIC
after CPU is created, when it's safe to call APIC's reset method.
Signed-off-by: Igor Mammedov imamm...@redhat.com
Reviewed-by: Eduardo Habkost ehabk...@redhat.com
---
v2:
* s/x86_cpu_apic_init
On Fri, Apr 05, 2013 at 04:36:55PM +0200, Igor Mammedov wrote:
Move CPU creation and features parsing into a separate cpu_x86_create()
function, so that board would be able to set board specific CPU
properties before CPU is realized.
Keep cpu_x86_init() for compatibility with the code that
On Fri, Apr 05, 2013 at 04:36:56PM +0200, Igor Mammedov wrote:
... so it could be called from without requiring CPUArchState
Signed-off-by: Igor Mammedov imamm...@redhat.com
Reviewed-by: Eduardo Habkost ehabk...@redhat.com
---
cpus.c | 4 ++--
include/sysemu/kvm.h | 12
On Fri, Apr 05, 2013 at 04:36:57PM +0200, Igor Mammedov wrote:
... to synchronize CPU state to KVM
* in addition link kvm-stub.o to *-user target and fix related compiling
issues.
Signed-off-by: Igor Mammedov imamm...@redhat.com
---
Makefile.target | 6 ++
On Fri, Apr 05, 2013 at 04:36:58PM +0200, Igor Mammedov wrote:
... and call it if defined from CPUClass.realize() if CPU was hotplugged
by default leave .resume() unset (i.e. NULL) and override it for softmmu
in qemu_init_vcpu() if it's still unset.
Signed-off-by: Igor Mammedov
On Tue, Apr 09, 2013 at 10:19:11PM +0200, Igor Mammedov wrote:
On Fri, 5 Apr 2013 14:10:54 -0300
Eduardo Habkost ehabk...@redhat.com wrote:
On Fri, Apr 05, 2013 at 04:37:16PM +0200, Igor Mammedov wrote:
[...]
diff --git a/qapi-schema.json b/qapi-schema.json
index db542f6..a760ed5
On Tue, Apr 09, 2013 at 11:05:26PM +0200, Igor Mammedov wrote:
On Tue, 9 Apr 2013 17:46:21 -0300
Eduardo Habkost ehabk...@redhat.com wrote:
On Tue, Apr 09, 2013 at 10:19:11PM +0200, Igor Mammedov wrote:
On Fri, 5 Apr 2013 14:10:54 -0300
Eduardo Habkost ehabk...@redhat.com wrote
On Thu, Apr 11, 2013 at 05:17:29PM +0200, Igor Mammedov wrote:
[...]
Now, my question is: suppose a caller starts QEMU as:
$ qemu -smp 18,cores=3,threads=3,maxcpus=36 \
-numa node,cpus=0-8 -numa node,cpus=9-17 \
-numa node,cpus=18-26 -numa node,cpus=27-35
and
that uses cpu_init()
and doesn't need to modify CPU properties.
Signed-off-by: Igor Mammedov imamm...@redhat.com
Nice, much simpler to review than v1. :-)
Reviewed-by: Eduardo Habkost ehabk...@redhat.com
---
v2:
* use error_propagate() to rise error up stack
---
target-i386/cpu.c | 18
801 - 900 of 11343 matches
Mail list logo
