* [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support
@ 2011-12-08 11:52 Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 01/15] msi: Generalize msix_supported to msi_supported Jan Kiszka
` (16 more replies)
0 siblings, 17 replies; 32+ messages in thread
From: Jan Kiszka @ 2011-12-08 11:52 UTC (permalink / raw)
To: Avi Kivity, Marcelo Tosatti
Cc: Anthony Liguori, Lai Jiangshan, kvm, Michael S. Tsirkin,
qemu-devel, Blue Swirl
Changes in v4:
- rebased of current uq/master
- fixed stupid bugs that broke bisectability and user space irqchip mode
- integrated NMI-over-LINT1 injection logic
CC: Lai Jiangshan <laijs@cn.fujitsu.com>
Jan Kiszka (15):
msi: Generalize msix_supported to msi_supported
kvm: Move kvmclock into hw/kvm folder
apic: Stop timer on reset
apic: Inject external NMI events via LINT1
apic: Introduce backend/frontend infrastructure for KVM reuse
apic: Open-code timer save/restore
i8259: Introduce backend/frontend infrastructure for KVM reuse
ioapic: Introduce backend/frontend infrastructure for KVM reuse
memory: Introduce memory_region_init_reservation
kvm: Introduce core services for in-kernel irqchip support
kvm: x86: Establish IRQ0 override control
kvm: x86: Add user space part for in-kernel APIC
kvm: x86: Add user space part for in-kernel i8259
kvm: x86: Add user space part for in-kernel IOAPIC
kvm: Arm in-kernel irqchip support
Makefile.objs | 2 +-
Makefile.target | 6 +-
configure | 1 +
hw/apic.c | 309 ++++-----------------------------------
hw/apic.h | 1 +
hw/apic_common.c | 312 ++++++++++++++++++++++++++++++++++++++++
hw/apic_internal.h | 122 ++++++++++++++++
hw/i8259.c | 127 ++--------------
hw/i8259_common.c | 173 ++++++++++++++++++++++
hw/i8259_internal.h | 82 +++++++++++
hw/ioapic.c | 130 ++---------------
hw/ioapic_common.c | 138 ++++++++++++++++++
hw/ioapic_internal.h | 106 ++++++++++++++
hw/kvm/apic.c | 154 ++++++++++++++++++++
hw/{kvmclock.c => kvm/clock.c} | 4 +-
hw/{kvmclock.h => kvm/clock.h} | 0
hw/kvm/i8259.c | 126 ++++++++++++++++
hw/kvm/ioapic.c | 101 +++++++++++++
hw/msi.c | 8 +
hw/msi.h | 2 +
hw/msix.c | 9 +-
hw/msix.h | 2 -
hw/pc.c | 19 ++-
hw/pc.h | 1 +
hw/pc_piix.c | 66 ++++++++-
kvm-all.c | 154 ++++++++++++++++++++
kvm-stub.c | 5 +
kvm.h | 13 ++
memory.c | 36 +++++
memory.h | 16 ++
monitor.c | 6 +-
qemu-config.c | 4 +
qemu-options.hx | 5 +-
sysemu.h | 1 -
target-i386/kvm.c | 19 +++
trace-events | 2 +-
vl.c | 1 -
37 files changed, 1724 insertions(+), 539 deletions(-)
create mode 100644 hw/apic_common.c
create mode 100644 hw/apic_internal.h
create mode 100644 hw/i8259_common.c
create mode 100644 hw/i8259_internal.h
create mode 100644 hw/ioapic_common.c
create mode 100644 hw/ioapic_internal.h
create mode 100644 hw/kvm/apic.c
rename hw/{kvmclock.c => kvm/clock.c} (98%)
rename hw/{kvmclock.h => kvm/clock.h} (100%)
create mode 100644 hw/kvm/i8259.c
create mode 100644 hw/kvm/ioapic.c
--
1.7.3.4
^ permalink raw reply [flat|nested] 32+ messages in thread
* [Qemu-devel] [PATCH v4 01/15] msi: Generalize msix_supported to msi_supported
2011-12-08 11:52 [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support Jan Kiszka
@ 2011-12-08 11:52 ` Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 02/15] kvm: Move kvmclock into hw/kvm folder Jan Kiszka
` (15 subsequent siblings)
16 siblings, 0 replies; 32+ messages in thread
From: Jan Kiszka @ 2011-12-08 11:52 UTC (permalink / raw)
To: Avi Kivity, Marcelo Tosatti
Cc: Blue Swirl, Anthony Liguori, qemu-devel, kvm, Michael S. Tsirkin
Rename msix_supported to msi_supported and control MSI and MSI-X
activation this way. That was likely to original intention for this
flag, but MSI support came after MSI-X.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
---
hw/msi.c | 8 ++++++++
hw/msi.h | 2 ++
hw/msix.c | 9 ++++-----
hw/msix.h | 2 --
hw/pc.c | 4 ++--
5 files changed, 16 insertions(+), 9 deletions(-)
diff --git a/hw/msi.c b/hw/msi.c
index f214fcf..5d6ceb6 100644
--- a/hw/msi.c
+++ b/hw/msi.c
@@ -36,6 +36,9 @@
#define PCI_MSI_VECTORS_MAX 32
+/* Flag for interrupt controller to declare MSI/MSI-X support */
+bool msi_supported;
+
/* If we get rid of cap allocator, we won't need this. */
static inline uint8_t msi_cap_sizeof(uint16_t flags)
{
@@ -116,6 +119,11 @@ int msi_init(struct PCIDevice *dev, uint8_t offset,
uint16_t flags;
uint8_t cap_size;
int config_offset;
+
+ if (!msi_supported) {
+ return -ENOTSUP;
+ }
+
MSI_DEV_PRINTF(dev,
"init offset: 0x%"PRIx8" vector: %"PRId8
" 64bit %d mask %d\n",
diff --git a/hw/msi.h b/hw/msi.h
index 5766018..3040bb0 100644
--- a/hw/msi.h
+++ b/hw/msi.h
@@ -24,6 +24,8 @@
#include "qemu-common.h"
#include "pci.h"
+extern bool msi_supported;
+
bool msi_enabled(const PCIDevice *dev);
int msi_init(struct PCIDevice *dev, uint8_t offset,
unsigned int nr_vectors, bool msi64bit, bool msi_per_vector_mask);
diff --git a/hw/msix.c b/hw/msix.c
index 149eed2..107d4e5 100644
--- a/hw/msix.c
+++ b/hw/msix.c
@@ -12,6 +12,7 @@
*/
#include "hw.h"
+#include "msi.h"
#include "msix.h"
#include "pci.h"
#include "range.h"
@@ -32,9 +33,6 @@
#define MSIX_MAX_ENTRIES 32
-/* Flag for interrupt controller to declare MSI-X support */
-int msix_supported;
-
/* Add MSI-X capability to the config space for the device. */
/* Given a bar and its size, add MSI-X table on top of it
* and fill MSI-X capability in the config space.
@@ -235,10 +233,11 @@ int msix_init(struct PCIDevice *dev, unsigned short nentries,
unsigned bar_nr, unsigned bar_size)
{
int ret;
+
/* Nothing to do if MSI is not supported by interrupt controller */
- if (!msix_supported)
+ if (!msi_supported) {
return -ENOTSUP;
-
+ }
if (nentries > MSIX_MAX_ENTRIES)
return -EINVAL;
diff --git a/hw/msix.h b/hw/msix.h
index 7e04336..5aba22b 100644
--- a/hw/msix.h
+++ b/hw/msix.h
@@ -29,6 +29,4 @@ void msix_notify(PCIDevice *dev, unsigned vector);
void msix_reset(PCIDevice *dev);
-extern int msix_supported;
-
#endif
diff --git a/hw/pc.c b/hw/pc.c
index 7c4bfa8..240aaae 100644
--- a/hw/pc.c
+++ b/hw/pc.c
@@ -36,7 +36,7 @@
#include "elf.h"
#include "multiboot.h"
#include "mc146818rtc.h"
-#include "msix.h"
+#include "msi.h"
#include "sysbus.h"
#include "sysemu.h"
#include "blockdev.h"
@@ -896,7 +896,7 @@ static DeviceState *apic_init(void *env, uint8_t apic_id)
apic_mapped = 1;
}
- msix_supported = 1;
+ msi_supported = true;
return dev;
}
--
1.7.3.4
^ permalink raw reply related [flat|nested] 32+ messages in thread
* [Qemu-devel] [PATCH v4 02/15] kvm: Move kvmclock into hw/kvm folder
2011-12-08 11:52 [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 01/15] msi: Generalize msix_supported to msi_supported Jan Kiszka
@ 2011-12-08 11:52 ` Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 03/15] apic: Stop timer on reset Jan Kiszka
` (14 subsequent siblings)
16 siblings, 0 replies; 32+ messages in thread
From: Jan Kiszka @ 2011-12-08 11:52 UTC (permalink / raw)
To: Avi Kivity, Marcelo Tosatti
Cc: Blue Swirl, Anthony Liguori, qemu-devel, kvm, Michael S. Tsirkin
More KVM-specific devices will come, so let's start with moving the
kvmclock into a dedicated folder.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
---
Makefile.target | 4 ++--
configure | 1 +
hw/{kvmclock.c => kvm/clock.c} | 4 ++--
hw/{kvmclock.h => kvm/clock.h} | 0
hw/pc_piix.c | 2 +-
5 files changed, 6 insertions(+), 5 deletions(-)
rename hw/{kvmclock.c => kvm/clock.c} (98%)
rename hw/{kvmclock.h => kvm/clock.h} (100%)
diff --git a/Makefile.target b/Makefile.target
index a111521..1d24a30 100644
--- a/Makefile.target
+++ b/Makefile.target
@@ -236,7 +236,7 @@ obj-i386-y += vmport.o
obj-i386-y += device-hotplug.o pci-hotplug.o smbios.o wdt_ib700.o
obj-i386-y += debugcon.o multiboot.o
obj-i386-y += pc_piix.o
-obj-i386-$(CONFIG_KVM) += kvmclock.o
+obj-i386-$(CONFIG_KVM) += kvm/clock.o
obj-i386-$(CONFIG_SPICE) += qxl.o qxl-logger.o qxl-render.o
# shared objects
@@ -428,7 +428,7 @@ qmp-commands-old.h: $(SRC_PATH)/qmp-commands.hx
clean:
rm -f *.o *.a *~ $(PROGS) nwfpe/*.o fpu/*.o
- rm -f *.d */*.d tcg/*.o ide/*.o 9pfs/*.o
+ rm -f *.d */*.d tcg/*.o ide/*.o 9pfs/*.o kvm/*.o
rm -f hmp-commands.h qmp-commands-old.h gdbstub-xml.c
ifdef CONFIG_TRACE_SYSTEMTAP
rm -f *.stp
diff --git a/configure b/configure
index ac4840d..12cd9d1 100755
--- a/configure
+++ b/configure
@@ -3338,6 +3338,7 @@ mkdir -p $target_dir/fpu
mkdir -p $target_dir/tcg
mkdir -p $target_dir/ide
mkdir -p $target_dir/9pfs
+mkdir -p $target_dir/kvm
if test "$target" = "arm-linux-user" -o "$target" = "armeb-linux-user" -o "$target" = "arm-bsd-user" -o "$target" = "armeb-bsd-user" ; then
mkdir -p $target_dir/nwfpe
fi
diff --git a/hw/kvmclock.c b/hw/kvm/clock.c
similarity index 98%
rename from hw/kvmclock.c
rename to hw/kvm/clock.c
index 5388bc4..5983271 100644
--- a/hw/kvmclock.c
+++ b/hw/kvm/clock.c
@@ -13,9 +13,9 @@
#include "qemu-common.h"
#include "sysemu.h"
-#include "sysbus.h"
#include "kvm.h"
-#include "kvmclock.h"
+#include "hw/sysbus.h"
+#include "hw/kvm/clock.h"
#include <linux/kvm.h>
#include <linux/kvm_para.h>
diff --git a/hw/kvmclock.h b/hw/kvm/clock.h
similarity index 100%
rename from hw/kvmclock.h
rename to hw/kvm/clock.h
diff --git a/hw/pc_piix.c b/hw/pc_piix.c
index 970f43c..530fe9c 100644
--- a/hw/pc_piix.c
+++ b/hw/pc_piix.c
@@ -34,7 +34,7 @@
#include "boards.h"
#include "ide.h"
#include "kvm.h"
-#include "kvmclock.h"
+#include "kvm/clock.h"
#include "sysemu.h"
#include "sysbus.h"
#include "arch_init.h"
--
1.7.3.4
^ permalink raw reply related [flat|nested] 32+ messages in thread
* [Qemu-devel] [PATCH v4 03/15] apic: Stop timer on reset
2011-12-08 11:52 [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 01/15] msi: Generalize msix_supported to msi_supported Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 02/15] kvm: Move kvmclock into hw/kvm folder Jan Kiszka
@ 2011-12-08 11:52 ` Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 04/15] apic: Inject external NMI events via LINT1 Jan Kiszka
` (13 subsequent siblings)
16 siblings, 0 replies; 32+ messages in thread
From: Jan Kiszka @ 2011-12-08 11:52 UTC (permalink / raw)
To: Avi Kivity, Marcelo Tosatti
Cc: Blue Swirl, Anthony Liguori, qemu-devel, kvm, Michael S. Tsirkin
All LVTs are masked on reset, so the timer becomes ineffective. Letting
it tick nevertheless is harmless, but will at least create a spurious
trace event.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
---
hw/apic.c | 2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/hw/apic.c b/hw/apic.c
index 9d0f460..4b97b17 100644
--- a/hw/apic.c
+++ b/hw/apic.c
@@ -528,6 +528,8 @@ void apic_init_reset(DeviceState *d)
s->initial_count_load_time = 0;
s->next_time = 0;
s->wait_for_sipi = 1;
+
+ qemu_del_timer(s->timer);
}
static void apic_startup(APICState *s, int vector_num)
--
1.7.3.4
^ permalink raw reply related [flat|nested] 32+ messages in thread
* [Qemu-devel] [PATCH v4 04/15] apic: Inject external NMI events via LINT1
2011-12-08 11:52 [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support Jan Kiszka
` (2 preceding siblings ...)
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 03/15] apic: Stop timer on reset Jan Kiszka
@ 2011-12-08 11:52 ` Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 05/15] apic: Introduce backend/frontend infrastructure for KVM reuse Jan Kiszka
` (12 subsequent siblings)
16 siblings, 0 replies; 32+ messages in thread
From: Jan Kiszka @ 2011-12-08 11:52 UTC (permalink / raw)
To: Avi Kivity, Marcelo Tosatti
Cc: Anthony Liguori, Lai Jiangshan, kvm, Michael S. Tsirkin,
qemu-devel, Blue Swirl
On real hardware, NMI button events are injected via the LINT1 line of
the APICs. E.g. kdump expect this wiring and gets upset if the per-APIC
LINT1 mask is not respected, i.e. if NMIs are injected to VCPUs that
should not receive them. Change the APIC emulation code to reflect this.
Based on qemu-kvm patch by Lai Jiangshan.
CC: Lai Jiangshan <laijs@cn.fujitsu.com>
Reported-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
---
hw/apic.c | 7 +++++++
hw/apic.h | 1 +
monitor.c | 6 +++++-
3 files changed, 13 insertions(+), 1 deletions(-)
diff --git a/hw/apic.c b/hw/apic.c
index 4b97b17..b9d733c 100644
--- a/hw/apic.c
+++ b/hw/apic.c
@@ -205,6 +205,13 @@ void apic_deliver_pic_intr(DeviceState *d, int level)
}
}
+void apic_deliver_nmi(DeviceState *d)
+{
+ APICState *s = DO_UPCAST(APICState, busdev.qdev, d);
+
+ apic_local_deliver(s, APIC_LVT_LINT1);
+}
+
#define foreach_apic(apic, deliver_bitmask, code) \
{\
int __i, __j, __mask;\
diff --git a/hw/apic.h b/hw/apic.h
index a5c910f..a62d83b 100644
--- a/hw/apic.h
+++ b/hw/apic.h
@@ -8,6 +8,7 @@ void apic_deliver_irq(uint8_t dest, uint8_t dest_mode, uint8_t delivery_mode,
uint8_t vector_num, uint8_t trigger_mode);
int apic_accept_pic_intr(DeviceState *s);
void apic_deliver_pic_intr(DeviceState *s, int level);
+void apic_deliver_nmi(DeviceState *d);
int apic_get_interrupt(DeviceState *s);
void apic_reset_irq_delivered(void);
int apic_get_irq_delivered(void);
diff --git a/monitor.c b/monitor.c
index 1be222e..6bd0fb1 100644
--- a/monitor.c
+++ b/monitor.c
@@ -2354,7 +2354,11 @@ static int do_inject_nmi(Monitor *mon, const QDict *qdict, QObject **ret_data)
CPUState *env;
for (env = first_cpu; env != NULL; env = env->next_cpu) {
- cpu_interrupt(env, CPU_INTERRUPT_NMI);
+ if (!env->apic_state) {
+ cpu_interrupt(env, CPU_INTERRUPT_NMI);
+ } else {
+ apic_deliver_nmi(env->apic_state);
+ }
}
return 0;
--
1.7.3.4
^ permalink raw reply related [flat|nested] 32+ messages in thread
* [Qemu-devel] [PATCH v4 05/15] apic: Introduce backend/frontend infrastructure for KVM reuse
2011-12-08 11:52 [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support Jan Kiszka
` (3 preceding siblings ...)
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 04/15] apic: Inject external NMI events via LINT1 Jan Kiszka
@ 2011-12-08 11:52 ` Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 06/15] apic: Open-code timer save/restore Jan Kiszka
` (11 subsequent siblings)
16 siblings, 0 replies; 32+ messages in thread
From: Jan Kiszka @ 2011-12-08 11:52 UTC (permalink / raw)
To: Avi Kivity, Marcelo Tosatti
Cc: Blue Swirl, Anthony Liguori, qemu-devel, kvm, Michael S. Tsirkin
The KVM in-kernel APIC model will reuse parts of the user space model
while providing the same frontend view to guest and most management
interfaces. Introduce an APIC backend concept to encapsulate those
parts that will tell user space and KVM model apart. The backend offers
callback hooks for init, base/tpr setting, and the external NMI delivery
that will be implemented accordingly.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
---
Makefile.target | 2 +-
hw/apic.c | 282 +++-------------------------------------------------
hw/apic_common.c | 265 ++++++++++++++++++++++++++++++++++++++++++++++++
hw/apic_internal.h | 119 ++++++++++++++++++++++
hw/pc.c | 1 +
trace-events | 2 +-
6 files changed, 403 insertions(+), 268 deletions(-)
create mode 100644 hw/apic_common.c
create mode 100644 hw/apic_internal.h
diff --git a/Makefile.target b/Makefile.target
index 1d24a30..c46f062 100644
--- a/Makefile.target
+++ b/Makefile.target
@@ -231,7 +231,7 @@ obj-$(CONFIG_IVSHMEM) += ivshmem.o
# Hardware support
obj-i386-y += vga.o
obj-i386-y += mc146818rtc.o pc.o
-obj-i386-y += cirrus_vga.o sga.o apic.o ioapic.o piix_pci.o
+obj-i386-y += cirrus_vga.o sga.o apic_common.o apic.o ioapic.o piix_pci.o
obj-i386-y += vmport.o
obj-i386-y += device-hotplug.o pci-hotplug.o smbios.o wdt_ib700.o
obj-i386-y += debugcon.o multiboot.o
diff --git a/hw/apic.c b/hw/apic.c
index b9d733c..f25be80 100644
--- a/hw/apic.c
+++ b/hw/apic.c
@@ -16,53 +16,13 @@
* You should have received a copy of the GNU Lesser General Public
* License along with this library; if not, see <http://www.gnu.org/licenses/>
*/
-#include "hw.h"
+#include "apic_internal.h"
#include "apic.h"
#include "ioapic.h"
-#include "qemu-timer.h"
#include "host-utils.h"
-#include "sysbus.h"
#include "trace.h"
#include "pc.h"
-/* APIC Local Vector Table */
-#define APIC_LVT_TIMER 0
-#define APIC_LVT_THERMAL 1
-#define APIC_LVT_PERFORM 2
-#define APIC_LVT_LINT0 3
-#define APIC_LVT_LINT1 4
-#define APIC_LVT_ERROR 5
-#define APIC_LVT_NB 6
-
-/* APIC delivery modes */
-#define APIC_DM_FIXED 0
-#define APIC_DM_LOWPRI 1
-#define APIC_DM_SMI 2
-#define APIC_DM_NMI 4
-#define APIC_DM_INIT 5
-#define APIC_DM_SIPI 6
-#define APIC_DM_EXTINT 7
-
-/* APIC destination mode */
-#define APIC_DESTMODE_FLAT 0xf
-#define APIC_DESTMODE_CLUSTER 1
-
-#define APIC_TRIGGER_EDGE 0
-#define APIC_TRIGGER_LEVEL 1
-
-#define APIC_LVT_TIMER_PERIODIC (1<<17)
-#define APIC_LVT_MASKED (1<<16)
-#define APIC_LVT_LEVEL_TRIGGER (1<<15)
-#define APIC_LVT_REMOTE_IRR (1<<14)
-#define APIC_INPUT_POLARITY (1<<13)
-#define APIC_SEND_PENDING (1<<12)
-
-#define ESR_ILLEGAL_ADDRESS (1 << 7)
-
-#define APIC_SV_DIRECTED_IO (1<<12)
-#define APIC_SV_ENABLE (1<<8)
-
-#define MAX_APICS 255
#define MAX_APIC_WORDS 8
/* Intel APIC constants: from include/asm/msidef.h */
@@ -75,40 +35,7 @@
#define MSI_ADDR_DEST_ID_SHIFT 12
#define MSI_ADDR_DEST_ID_MASK 0x00ffff0
-#define MSI_ADDR_SIZE 0x100000
-
-typedef struct APICState APICState;
-
-struct APICState {
- SysBusDevice busdev;
- MemoryRegion io_memory;
- void *cpu_env;
- uint32_t apicbase;
- uint8_t id;
- uint8_t arb_id;
- uint8_t tpr;
- uint32_t spurious_vec;
- uint8_t log_dest;
- uint8_t dest_mode;
- uint32_t isr[8]; /* in service register */
- uint32_t tmr[8]; /* trigger mode register */
- uint32_t irr[8]; /* interrupt request register */
- uint32_t lvt[APIC_LVT_NB];
- uint32_t esr; /* error register */
- uint32_t icr[2];
-
- uint32_t divide_conf;
- int count_shift;
- uint32_t initial_count;
- int64_t initial_count_load_time, next_time;
- uint32_t idx;
- QEMUTimer *timer;
- int sipi_vector;
- int wait_for_sipi;
-};
-
static APICState *local_apics[MAX_APICS + 1];
-static int apic_irq_delivered;
static void apic_set_irq(APICState *s, int vector_num, int trigger_mode);
static void apic_update_irq(APICState *s);
@@ -205,10 +132,8 @@ void apic_deliver_pic_intr(DeviceState *d, int level)
}
}
-void apic_deliver_nmi(DeviceState *d)
+static void apic_external_nmi(APICState *s)
{
- APICState *s = DO_UPCAST(APICState, busdev.qdev, d);
-
apic_local_deliver(s, APIC_LVT_LINT1);
}
@@ -300,14 +225,8 @@ void apic_deliver_irq(uint8_t dest, uint8_t dest_mode, uint8_t delivery_mode,
apic_bus_deliver(deliver_bitmask, delivery_mode, vector_num, trigger_mode);
}
-void cpu_set_apic_base(DeviceState *d, uint64_t val)
+static void apic_set_base(APICState *s, uint64_t val)
{
- APICState *s = DO_UPCAST(APICState, busdev.qdev, d);
-
- trace_cpu_set_apic_base(val);
-
- if (!s)
- return;
s->apicbase = (val & 0xfffff000) |
(s->apicbase & (MSR_IA32_APICBASE_BSP | MSR_IA32_APICBASE_ENABLE));
/* if disabled, cannot be enabled again */
@@ -318,32 +237,12 @@ void cpu_set_apic_base(DeviceState *d, uint64_t val)
}
}
-uint64_t cpu_get_apic_base(DeviceState *d)
+static void apic_set_tpr(APICState *s, uint8_t val)
{
- APICState *s = DO_UPCAST(APICState, busdev.qdev, d);
-
- trace_cpu_get_apic_base(s ? (uint64_t)s->apicbase: 0);
-
- return s ? s->apicbase : 0;
-}
-
-void cpu_set_apic_tpr(DeviceState *d, uint8_t val)
-{
- APICState *s = DO_UPCAST(APICState, busdev.qdev, d);
-
- if (!s)
- return;
s->tpr = (val & 0x0f) << 4;
apic_update_irq(s);
}
-uint8_t cpu_get_apic_tpr(DeviceState *d)
-{
- APICState *s = DO_UPCAST(APICState, busdev.qdev, d);
-
- return s ? s->tpr >> 4 : 0;
-}
-
/* return -1 if no bit is set */
static int get_highest_priority_int(uint32_t *tab)
{
@@ -413,25 +312,9 @@ static void apic_update_irq(APICState *s)
}
}
-void apic_reset_irq_delivered(void)
-{
- trace_apic_reset_irq_delivered(apic_irq_delivered);
-
- apic_irq_delivered = 0;
-}
-
-int apic_get_irq_delivered(void)
-{
- trace_apic_get_irq_delivered(apic_irq_delivered);
-
- return apic_irq_delivered;
-}
-
static void apic_set_irq(APICState *s, int vector_num, int trigger_mode)
{
- apic_irq_delivered += !get_bit(s->irr, vector_num);
-
- trace_apic_set_irq(apic_irq_delivered);
+ apic_set_irq_delivered(!get_bit(s->irr, vector_num));
set_bit(s->irr, vector_num);
if (trigger_mode)
@@ -510,35 +393,6 @@ static void apic_get_delivery_bitmask(uint32_t *deliver_bitmask,
}
}
-void apic_init_reset(DeviceState *d)
-{
- APICState *s = DO_UPCAST(APICState, busdev.qdev, d);
- int i;
-
- if (!s)
- return;
-
- s->tpr = 0;
- s->spurious_vec = 0xff;
- s->log_dest = 0;
- s->dest_mode = 0xf;
- memset(s->isr, 0, sizeof(s->isr));
- memset(s->tmr, 0, sizeof(s->tmr));
- memset(s->irr, 0, sizeof(s->irr));
- for(i = 0; i < APIC_LVT_NB; i++)
- s->lvt[i] = 1 << 16; /* mask LVT */
- s->esr = 0;
- memset(s->icr, 0, sizeof(s->icr));
- s->divide_conf = 0;
- s->count_shift = 0;
- s->initial_count = 0;
- s->initial_count_load_time = 0;
- s->next_time = 0;
- s->wait_for_sipi = 1;
-
- qemu_del_timer(s->timer);
-}
-
static void apic_startup(APICState *s, int vector_num)
{
s->sipi_vector = vector_num;
@@ -899,96 +753,6 @@ static void apic_mem_writel(void *opaque, target_phys_addr_t addr, uint32_t val)
}
}
-/* This function is only used for old state version 1 and 2 */
-static int apic_load_old(QEMUFile *f, void *opaque, int version_id)
-{
- APICState *s = opaque;
- int i;
-
- if (version_id > 2)
- return -EINVAL;
-
- /* XXX: what if the base changes? (registered memory regions) */
- qemu_get_be32s(f, &s->apicbase);
- qemu_get_8s(f, &s->id);
- qemu_get_8s(f, &s->arb_id);
- qemu_get_8s(f, &s->tpr);
- qemu_get_be32s(f, &s->spurious_vec);
- qemu_get_8s(f, &s->log_dest);
- qemu_get_8s(f, &s->dest_mode);
- for (i = 0; i < 8; i++) {
- qemu_get_be32s(f, &s->isr[i]);
- qemu_get_be32s(f, &s->tmr[i]);
- qemu_get_be32s(f, &s->irr[i]);
- }
- for (i = 0; i < APIC_LVT_NB; i++) {
- qemu_get_be32s(f, &s->lvt[i]);
- }
- qemu_get_be32s(f, &s->esr);
- qemu_get_be32s(f, &s->icr[0]);
- qemu_get_be32s(f, &s->icr[1]);
- qemu_get_be32s(f, &s->divide_conf);
- s->count_shift=qemu_get_be32(f);
- qemu_get_be32s(f, &s->initial_count);
- s->initial_count_load_time=qemu_get_be64(f);
- s->next_time=qemu_get_be64(f);
-
- if (version_id >= 2)
- qemu_get_timer(f, s->timer);
- return 0;
-}
-
-static const VMStateDescription vmstate_apic = {
- .name = "apic",
- .version_id = 3,
- .minimum_version_id = 3,
- .minimum_version_id_old = 1,
- .load_state_old = apic_load_old,
- .fields = (VMStateField []) {
- VMSTATE_UINT32(apicbase, APICState),
- VMSTATE_UINT8(id, APICState),
- VMSTATE_UINT8(arb_id, APICState),
- VMSTATE_UINT8(tpr, APICState),
- VMSTATE_UINT32(spurious_vec, APICState),
- VMSTATE_UINT8(log_dest, APICState),
- VMSTATE_UINT8(dest_mode, APICState),
- VMSTATE_UINT32_ARRAY(isr, APICState, 8),
- VMSTATE_UINT32_ARRAY(tmr, APICState, 8),
- VMSTATE_UINT32_ARRAY(irr, APICState, 8),
- VMSTATE_UINT32_ARRAY(lvt, APICState, APIC_LVT_NB),
- VMSTATE_UINT32(esr, APICState),
- VMSTATE_UINT32_ARRAY(icr, APICState, 2),
- VMSTATE_UINT32(divide_conf, APICState),
- VMSTATE_INT32(count_shift, APICState),
- VMSTATE_UINT32(initial_count, APICState),
- VMSTATE_INT64(initial_count_load_time, APICState),
- VMSTATE_INT64(next_time, APICState),
- VMSTATE_TIMER(timer, APICState),
- VMSTATE_END_OF_LIST()
- }
-};
-
-static void apic_reset(DeviceState *d)
-{
- APICState *s = DO_UPCAST(APICState, busdev.qdev, d);
- int bsp;
-
- bsp = cpu_is_bsp(s->cpu_env);
- s->apicbase = 0xfee00000 |
- (bsp ? MSR_IA32_APICBASE_BSP : 0) | MSR_IA32_APICBASE_ENABLE;
-
- apic_init_reset(d);
-
- if (bsp) {
- /*
- * LINT0 delivery mode on CPU #0 is set to ExtInt at initialization
- * time typically by BIOS, so PIC interrupt can be delivered to the
- * processor when local APIC is enabled.
- */
- s->lvt[APIC_LVT_LINT0] = 0x700;
- }
-}
-
static const MemoryRegionOps apic_io_ops = {
.old_mmio = {
.read = { apic_mem_readb, apic_mem_readw, apic_mem_readl, },
@@ -997,41 +761,27 @@ static const MemoryRegionOps apic_io_ops = {
.endianness = DEVICE_NATIVE_ENDIAN,
};
-static int apic_init1(SysBusDevice *dev)
+static void apic_backend_init(APICState *s)
{
- APICState *s = FROM_SYSBUS(APICState, dev);
- static int last_apic_idx;
-
- if (last_apic_idx >= MAX_APICS) {
- return -1;
- }
- memory_region_init_io(&s->io_memory, &apic_io_ops, s, "apic",
- MSI_ADDR_SIZE);
- sysbus_init_mmio(dev, &s->io_memory);
+ memory_region_init_io(&s->io_memory, &apic_io_ops, s, "apic-msi",
+ MSI_SPACE_SIZE);
s->timer = qemu_new_timer_ns(vm_clock, apic_timer, s);
- s->idx = last_apic_idx++;
local_apics[s->idx] = s;
- return 0;
}
-static SysBusDeviceInfo apic_info = {
- .init = apic_init1,
- .qdev.name = "apic",
- .qdev.size = sizeof(APICState),
- .qdev.vmsd = &vmstate_apic,
- .qdev.reset = apic_reset,
- .qdev.no_user = 1,
- .qdev.props = (Property[]) {
- DEFINE_PROP_UINT8("id", APICState, id, -1),
- DEFINE_PROP_PTR("cpu_env", APICState, cpu_env),
- DEFINE_PROP_END_OF_LIST(),
- }
+static APICBackend apic_backend = {
+ .name = "QEMU",
+ .init = apic_backend_init,
+ .set_base = apic_set_base,
+ .set_tpr = apic_set_tpr,
+ .external_nmi = apic_external_nmi,
};
static void apic_register_devices(void)
{
- sysbus_register_withprop(&apic_info);
+ apic_register_device();
+ apic_register_backend(&apic_backend);
}
device_init(apic_register_devices)
diff --git a/hw/apic_common.c b/hw/apic_common.c
new file mode 100644
index 0000000..73241e4
--- /dev/null
+++ b/hw/apic_common.c
@@ -0,0 +1,265 @@
+/*
+ * APIC support - common bits of emulated and KVM kernel model
+ *
+ * Copyright (c) 2004-2005 Fabrice Bellard
+ * Copyright (c) 2011 Jan Kiszka, Siemens AG
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see <http://www.gnu.org/licenses/>
+ */
+#include "apic.h"
+#include "apic_internal.h"
+#include "trace.h"
+
+static QSIMPLEQ_HEAD(, APICBackend) backends =
+ QSIMPLEQ_HEAD_INITIALIZER(backends);
+static int apic_irq_delivered;
+
+void cpu_set_apic_base(DeviceState *d, uint64_t val)
+{
+ APICState *s = DO_UPCAST(APICState, busdev.qdev, d);
+
+ trace_cpu_set_apic_base(val);
+
+ if (s) {
+ s->backend->set_base(s, val);
+ }
+}
+
+uint64_t cpu_get_apic_base(DeviceState *d)
+{
+ APICState *s = DO_UPCAST(APICState, busdev.qdev, d);
+
+ trace_cpu_get_apic_base(s ? (uint64_t)s->apicbase : 0);
+
+ return s ? s->apicbase : 0;
+}
+
+void cpu_set_apic_tpr(DeviceState *d, uint8_t val)
+{
+ APICState *s = DO_UPCAST(APICState, busdev.qdev, d);
+
+ if (s) {
+ s->backend->set_tpr(s, val);
+ }
+}
+
+uint8_t cpu_get_apic_tpr(DeviceState *d)
+{
+ APICState *s = DO_UPCAST(APICState, busdev.qdev, d);
+
+ return s ? s->tpr >> 4 : 0;
+}
+
+void apic_set_irq_delivered(int delivered)
+{
+ apic_irq_delivered += delivered;
+
+ trace_apic_set_irq_delivered(apic_irq_delivered);
+}
+
+void apic_reset_irq_delivered(void)
+{
+ trace_apic_reset_irq_delivered(apic_irq_delivered);
+
+ apic_irq_delivered = false;
+}
+
+int apic_get_irq_delivered(void)
+{
+ trace_apic_get_irq_delivered(apic_irq_delivered);
+
+ return apic_irq_delivered;
+}
+
+void apic_deliver_nmi(DeviceState *d)
+{
+ APICState *s = DO_UPCAST(APICState, busdev.qdev, d);
+
+ s->backend->external_nmi(s);
+}
+
+void apic_init_reset(DeviceState *d)
+{
+ APICState *s = DO_UPCAST(APICState, busdev.qdev, d);
+ int i;
+
+ if (!s) {
+ return;
+ }
+ s->tpr = 0;
+ s->spurious_vec = 0xff;
+ s->log_dest = 0;
+ s->dest_mode = 0xf;
+ memset(s->isr, 0, sizeof(s->isr));
+ memset(s->tmr, 0, sizeof(s->tmr));
+ memset(s->irr, 0, sizeof(s->irr));
+ for (i = 0; i < APIC_LVT_NB; i++) {
+ s->lvt[i] = APIC_LVT_MASKED;
+ }
+ s->esr = 0;
+ memset(s->icr, 0, sizeof(s->icr));
+ s->divide_conf = 0;
+ s->count_shift = 0;
+ s->initial_count = 0;
+ s->initial_count_load_time = 0;
+ s->next_time = 0;
+ s->wait_for_sipi = 1;
+
+ qemu_del_timer(s->timer);
+}
+
+static void apic_reset(DeviceState *d)
+{
+ APICState *s = DO_UPCAST(APICState, busdev.qdev, d);
+ bool bsp;
+
+ bsp = cpu_is_bsp(s->cpu_env);
+ s->apicbase = 0xfee00000 |
+ (bsp ? MSR_IA32_APICBASE_BSP : 0) | MSR_IA32_APICBASE_ENABLE;
+
+ apic_init_reset(d);
+
+ if (bsp) {
+ /*
+ * LINT0 delivery mode on CPU #0 is set to ExtInt at initialization
+ * time typically by BIOS, so PIC interrupt can be delivered to the
+ * processor when local APIC is enabled.
+ */
+ s->lvt[APIC_LVT_LINT0] = 0x700;
+ }
+}
+
+/* This function is only used for old state version 1 and 2 */
+static int apic_load_old(QEMUFile *f, void *opaque, int version_id)
+{
+ APICState *s = opaque;
+ int i;
+
+ if (version_id > 2) {
+ return -EINVAL;
+ }
+
+ /* XXX: what if the base changes? (registered memory regions) */
+ qemu_get_be32s(f, &s->apicbase);
+ qemu_get_8s(f, &s->id);
+ qemu_get_8s(f, &s->arb_id);
+ qemu_get_8s(f, &s->tpr);
+ qemu_get_be32s(f, &s->spurious_vec);
+ qemu_get_8s(f, &s->log_dest);
+ qemu_get_8s(f, &s->dest_mode);
+ for (i = 0; i < 8; i++) {
+ qemu_get_be32s(f, &s->isr[i]);
+ qemu_get_be32s(f, &s->tmr[i]);
+ qemu_get_be32s(f, &s->irr[i]);
+ }
+ for (i = 0; i < APIC_LVT_NB; i++) {
+ qemu_get_be32s(f, &s->lvt[i]);
+ }
+ qemu_get_be32s(f, &s->esr);
+ qemu_get_be32s(f, &s->icr[0]);
+ qemu_get_be32s(f, &s->icr[1]);
+ qemu_get_be32s(f, &s->divide_conf);
+ s->count_shift = qemu_get_be32(f);
+ qemu_get_be32s(f, &s->initial_count);
+ s->initial_count_load_time = qemu_get_be64(f);
+ s->next_time = qemu_get_be64(f);
+
+ if (version_id >= 2) {
+ qemu_get_timer(f, s->timer);
+ }
+ return 0;
+}
+
+static const VMStateDescription vmstate_apic = {
+ .name = "apic",
+ .version_id = 3,
+ .minimum_version_id = 3,
+ .minimum_version_id_old = 1,
+ .load_state_old = apic_load_old,
+ .fields = (VMStateField[]) {
+ VMSTATE_UINT32(apicbase, APICState),
+ VMSTATE_UINT8(id, APICState),
+ VMSTATE_UINT8(arb_id, APICState),
+ VMSTATE_UINT8(tpr, APICState),
+ VMSTATE_UINT32(spurious_vec, APICState),
+ VMSTATE_UINT8(log_dest, APICState),
+ VMSTATE_UINT8(dest_mode, APICState),
+ VMSTATE_UINT32_ARRAY(isr, APICState, 8),
+ VMSTATE_UINT32_ARRAY(tmr, APICState, 8),
+ VMSTATE_UINT32_ARRAY(irr, APICState, 8),
+ VMSTATE_UINT32_ARRAY(lvt, APICState, APIC_LVT_NB),
+ VMSTATE_UINT32(esr, APICState),
+ VMSTATE_UINT32_ARRAY(icr, APICState, 2),
+ VMSTATE_UINT32(divide_conf, APICState),
+ VMSTATE_INT32(count_shift, APICState),
+ VMSTATE_UINT32(initial_count, APICState),
+ VMSTATE_INT64(initial_count_load_time, APICState),
+ VMSTATE_INT64(next_time, APICState),
+ VMSTATE_TIMER(timer, APICState),
+ VMSTATE_END_OF_LIST()
+ }
+};
+
+static int apic_init(SysBusDevice *dev)
+{
+ APICState *s = FROM_SYSBUS(APICState, dev);
+ static int apic_no;
+ APICBackend *b;
+
+ if (apic_no >= MAX_APICS) {
+ return -1;
+ }
+ s->idx = apic_no++;
+
+ QSIMPLEQ_FOREACH(b, &backends, entry) {
+ if (strcmp(b->name, s->backend_name) == 0) {
+ s->backend = b;
+ break;
+ }
+ }
+ if (!s->backend) {
+ hw_error("APIC backend '%s' not found!", s->backend_name);
+ exit(1);
+ }
+
+ b->init(s);
+
+ sysbus_init_mmio(&s->busdev, &s->io_memory);
+ return 0;
+}
+
+static SysBusDeviceInfo apic_info = {
+ .init = apic_init,
+ .qdev.name = "apic",
+ .qdev.size = sizeof(APICState),
+ .qdev.vmsd = &vmstate_apic,
+ .qdev.reset = apic_reset,
+ .qdev.no_user = 1,
+ .qdev.props = (Property[]) {
+ DEFINE_PROP_UINT8("id", APICState, id, -1),
+ DEFINE_PROP_PTR("cpu_env", APICState, cpu_env),
+ DEFINE_PROP_STRING("backend", APICState, backend_name),
+ DEFINE_PROP_END_OF_LIST(),
+ }
+};
+
+void apic_register_backend(APICBackend *backend)
+{
+ QSIMPLEQ_INSERT_TAIL(&backends, backend, entry);
+}
+
+void apic_register_device(void)
+{
+ sysbus_register_withprop(&apic_info);
+}
diff --git a/hw/apic_internal.h b/hw/apic_internal.h
new file mode 100644
index 0000000..80d6e62
--- /dev/null
+++ b/hw/apic_internal.h
@@ -0,0 +1,119 @@
+/*
+ * APIC support - internal interfaces
+ *
+ * Copyright (c) 2004-2005 Fabrice Bellard
+ * Copyright (c) 2011 Jan Kiszka, Siemens AG
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see <http://www.gnu.org/licenses/>
+ */
+#ifndef QEMU_APIC_INTERNAL_H
+#define QEMU_APIC_INTERNAL_H
+
+#include "memory.h"
+#include "sysbus.h"
+#include "qemu-timer.h"
+#include "qemu-queue.h"
+
+/* APIC Local Vector Table */
+#define APIC_LVT_TIMER 0
+#define APIC_LVT_THERMAL 1
+#define APIC_LVT_PERFORM 2
+#define APIC_LVT_LINT0 3
+#define APIC_LVT_LINT1 4
+#define APIC_LVT_ERROR 5
+#define APIC_LVT_NB 6
+
+/* APIC delivery modes */
+#define APIC_DM_FIXED 0
+#define APIC_DM_LOWPRI 1
+#define APIC_DM_SMI 2
+#define APIC_DM_NMI 4
+#define APIC_DM_INIT 5
+#define APIC_DM_SIPI 6
+#define APIC_DM_EXTINT 7
+
+/* APIC destination mode */
+#define APIC_DESTMODE_FLAT 0xf
+#define APIC_DESTMODE_CLUSTER 1
+
+#define APIC_TRIGGER_EDGE 0
+#define APIC_TRIGGER_LEVEL 1
+
+#define APIC_LVT_TIMER_PERIODIC (1<<17)
+#define APIC_LVT_MASKED (1<<16)
+#define APIC_LVT_LEVEL_TRIGGER (1<<15)
+#define APIC_LVT_REMOTE_IRR (1<<14)
+#define APIC_INPUT_POLARITY (1<<13)
+#define APIC_SEND_PENDING (1<<12)
+
+#define ESR_ILLEGAL_ADDRESS (1 << 7)
+
+#define APIC_SV_DIRECTED_IO (1<<12)
+#define APIC_SV_ENABLE (1<<8)
+
+#define MAX_APICS 255
+
+#define MSI_SPACE_SIZE 0x100000
+
+typedef struct APICBackend APICBackend;
+typedef struct APICState APICState;
+
+struct APICBackend {
+ const char *name;
+ void (*init)(APICState *s);
+ void (*set_base)(APICState *s, uint64_t val);
+ void (*set_tpr)(APICState *s, uint8_t val);
+ void (*external_nmi)(APICState *s);
+
+ QSIMPLEQ_ENTRY(APICBackend) entry;
+};
+
+struct APICState {
+ SysBusDevice busdev;
+ MemoryRegion io_memory;
+ void *cpu_env;
+ uint32_t apicbase;
+ uint8_t id;
+ uint8_t arb_id;
+ uint8_t tpr;
+ uint32_t spurious_vec;
+ uint8_t log_dest;
+ uint8_t dest_mode;
+ uint32_t isr[8]; /* in service register */
+ uint32_t tmr[8]; /* trigger mode register */
+ uint32_t irr[8]; /* interrupt request register */
+ uint32_t lvt[APIC_LVT_NB];
+ uint32_t esr; /* error register */
+ uint32_t icr[2];
+
+ uint32_t divide_conf;
+ int count_shift;
+ uint32_t initial_count;
+ int64_t initial_count_load_time;
+ int64_t next_time;
+ int idx;
+ QEMUTimer *timer;
+ int sipi_vector;
+ int wait_for_sipi;
+
+ char *backend_name;
+ APICBackend *backend;
+};
+
+void apic_register_device(void);
+void apic_register_backend(APICBackend *backend);
+
+void apic_set_irq_delivered(int delivered);
+
+#endif /* !QEMU_APIC_INTERNAL_H */
diff --git a/hw/pc.c b/hw/pc.c
index 240aaae..ee6e59b 100644
--- a/hw/pc.c
+++ b/hw/pc.c
@@ -884,6 +884,7 @@ static DeviceState *apic_init(void *env, uint8_t apic_id)
dev = qdev_create(NULL, "apic");
qdev_prop_set_uint8(dev, "id", apic_id);
qdev_prop_set_ptr(dev, "cpu_env", env);
+ qdev_prop_set_string(dev, "backend", g_strdup("QEMU"));
qdev_init_nofail(dev);
d = sysbus_from_qdev(dev);
diff --git a/trace-events b/trace-events
index 962caca..210ee66 100644
--- a/trace-events
+++ b/trace-events
@@ -98,7 +98,7 @@ apic_mem_writel(uint64_t addr, uint32_t val) "%"PRIx64" = %08x"
# coalescing
apic_reset_irq_delivered(int apic_irq_delivered) "old coalescing %d"
apic_get_irq_delivered(int apic_irq_delivered) "returning coalescing %d"
-apic_set_irq(int apic_irq_delivered) "coalescing %d"
+apic_set_irq_delivered(int apic_irq_delivered) "coalescing %d"
# hw/cs4231.c
cs4231_mem_readl_dreg(uint32_t reg, uint32_t ret) "read dreg %d: 0x%02x"
--
1.7.3.4
^ permalink raw reply related [flat|nested] 32+ messages in thread
* [Qemu-devel] [PATCH v4 06/15] apic: Open-code timer save/restore
2011-12-08 11:52 [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support Jan Kiszka
` (4 preceding siblings ...)
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 05/15] apic: Introduce backend/frontend infrastructure for KVM reuse Jan Kiszka
@ 2011-12-08 11:52 ` Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 07/15] i8259: Introduce backend/frontend infrastructure for KVM reuse Jan Kiszka
` (10 subsequent siblings)
16 siblings, 0 replies; 32+ messages in thread
From: Jan Kiszka @ 2011-12-08 11:52 UTC (permalink / raw)
To: Avi Kivity, Marcelo Tosatti
Cc: Blue Swirl, Anthony Liguori, qemu-devel, kvm, Michael S. Tsirkin
To enable migration between accelerated and non-accelerated APIC models,
we will need to handle the timer saving and restoring specially and can
no longer rely on the automatics of VMSTATE_TIMER. Specifically,
accelerated model will not start any QEMUTimer.
This patch therefore factors out the generic bits into apic_next_timer
and introduces a post-load callback that can be implemented differently
by both models.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
---
hw/apic.c | 30 ++++++++++++------------------
hw/apic_common.c | 51 +++++++++++++++++++++++++++++++++++++++++++++++++--
hw/apic_internal.h | 3 +++
3 files changed, 64 insertions(+), 20 deletions(-)
diff --git a/hw/apic.c b/hw/apic.c
index f25be80..ed6411d 100644
--- a/hw/apic.c
+++ b/hw/apic.c
@@ -521,25 +521,9 @@ static uint32_t apic_get_current_count(APICState *s)
static void apic_timer_update(APICState *s, int64_t current_time)
{
- int64_t next_time, d;
-
- if (!(s->lvt[APIC_LVT_TIMER] & APIC_LVT_MASKED)) {
- d = (current_time - s->initial_count_load_time) >>
- s->count_shift;
- if (s->lvt[APIC_LVT_TIMER] & APIC_LVT_TIMER_PERIODIC) {
- if (!s->initial_count)
- goto no_timer;
- d = ((d / ((uint64_t)s->initial_count + 1)) + 1) * ((uint64_t)s->initial_count + 1);
- } else {
- if (d >= s->initial_count)
- goto no_timer;
- d = (uint64_t)s->initial_count + 1;
- }
- next_time = s->initial_count_load_time + (d << s->count_shift);
- qemu_mod_timer(s->timer, next_time);
- s->next_time = next_time;
+ if (apic_next_timer(s, current_time)) {
+ qemu_mod_timer(s->timer, s->next_time);
} else {
- no_timer:
qemu_del_timer(s->timer);
}
}
@@ -770,12 +754,22 @@ static void apic_backend_init(APICState *s)
local_apics[s->idx] = s;
}
+static void apic_post_load(APICState *s)
+{
+ if (s->timer_expiry != -1) {
+ qemu_mod_timer(s->timer, s->timer_expiry);
+ } else {
+ qemu_del_timer(s->timer);
+ }
+}
+
static APICBackend apic_backend = {
.name = "QEMU",
.init = apic_backend_init,
.set_base = apic_set_base,
.set_tpr = apic_set_tpr,
.external_nmi = apic_external_nmi,
+ .post_load = apic_post_load,
};
static void apic_register_devices(void)
diff --git a/hw/apic_common.c b/hw/apic_common.c
index 73241e4..f38ffc1 100644
--- a/hw/apic_common.c
+++ b/hw/apic_common.c
@@ -89,6 +89,39 @@ void apic_deliver_nmi(DeviceState *d)
s->backend->external_nmi(s);
}
+bool apic_next_timer(APICState *s, int64_t current_time)
+{
+ int64_t d;
+
+ /* We need to store the timer state separately to support APIC
+ * implementations that maintain a non-QEMU timer, e.g. inside the
+ * host kernel. This open-coded state allows us to migrate between
+ * both models. */
+ s->timer_expiry = -1;
+
+ if (s->lvt[APIC_LVT_TIMER] & APIC_LVT_MASKED) {
+ return false;
+ }
+
+ d = (current_time - s->initial_count_load_time) >> s->count_shift;
+
+ if (s->lvt[APIC_LVT_TIMER] & APIC_LVT_TIMER_PERIODIC) {
+ if (!s->initial_count) {
+ return false;
+ }
+ d = ((d / ((uint64_t)s->initial_count + 1)) + 1) *
+ ((uint64_t)s->initial_count + 1);
+ } else {
+ if (d >= s->initial_count) {
+ return false;
+ }
+ d = (uint64_t)s->initial_count + 1;
+ }
+ s->next_time = s->initial_count_load_time + (d << s->count_shift);
+ s->timer_expiry = s->next_time;
+ return true;
+}
+
void apic_init_reset(DeviceState *d)
{
APICState *s = DO_UPCAST(APICState, busdev.qdev, d);
@@ -116,7 +149,10 @@ void apic_init_reset(DeviceState *d)
s->next_time = 0;
s->wait_for_sipi = 1;
- qemu_del_timer(s->timer);
+ if (s->timer) {
+ qemu_del_timer(s->timer);
+ }
+ s->timer_expiry = -1;
}
static void apic_reset(DeviceState *d)
@@ -181,12 +217,23 @@ static int apic_load_old(QEMUFile *f, void *opaque, int version_id)
return 0;
}
+static int apic_dispatch_post_load(void *opaque, int version_id)
+{
+ APICState *s = opaque;
+
+ if (s->backend->post_load) {
+ s->backend->post_load(s);
+ }
+ return 0;
+}
+
static const VMStateDescription vmstate_apic = {
.name = "apic",
.version_id = 3,
.minimum_version_id = 3,
.minimum_version_id_old = 1,
.load_state_old = apic_load_old,
+ .post_load = apic_dispatch_post_load,
.fields = (VMStateField[]) {
VMSTATE_UINT32(apicbase, APICState),
VMSTATE_UINT8(id, APICState),
@@ -206,7 +253,7 @@ static const VMStateDescription vmstate_apic = {
VMSTATE_UINT32(initial_count, APICState),
VMSTATE_INT64(initial_count_load_time, APICState),
VMSTATE_INT64(next_time, APICState),
- VMSTATE_TIMER(timer, APICState),
+ VMSTATE_INT64(timer_expiry, APICState), /* open-coded timer state */
VMSTATE_END_OF_LIST()
}
};
diff --git a/hw/apic_internal.h b/hw/apic_internal.h
index 80d6e62..8665bbb 100644
--- a/hw/apic_internal.h
+++ b/hw/apic_internal.h
@@ -75,6 +75,7 @@ struct APICBackend {
void (*set_base)(APICState *s, uint64_t val);
void (*set_tpr)(APICState *s, uint8_t val);
void (*external_nmi)(APICState *s);
+ void (*post_load)(APICState *s);
QSIMPLEQ_ENTRY(APICBackend) entry;
};
@@ -104,6 +105,7 @@ struct APICState {
int64_t next_time;
int idx;
QEMUTimer *timer;
+ int64_t timer_expiry;
int sipi_vector;
int wait_for_sipi;
@@ -114,6 +116,7 @@ struct APICState {
void apic_register_device(void);
void apic_register_backend(APICBackend *backend);
+bool apic_next_timer(APICState *s, int64_t current_time);
void apic_set_irq_delivered(int delivered);
#endif /* !QEMU_APIC_INTERNAL_H */
--
1.7.3.4
^ permalink raw reply related [flat|nested] 32+ messages in thread
* [Qemu-devel] [PATCH v4 07/15] i8259: Introduce backend/frontend infrastructure for KVM reuse
2011-12-08 11:52 [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support Jan Kiszka
` (5 preceding siblings ...)
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 06/15] apic: Open-code timer save/restore Jan Kiszka
@ 2011-12-08 11:52 ` Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 08/15] ioapic: " Jan Kiszka
` (9 subsequent siblings)
16 siblings, 0 replies; 32+ messages in thread
From: Jan Kiszka @ 2011-12-08 11:52 UTC (permalink / raw)
To: Avi Kivity, Marcelo Tosatti
Cc: Blue Swirl, Anthony Liguori, qemu-devel, kvm, Michael S. Tsirkin
Analogously to the APIC, we will reuse some parts of the user space
i8259 model for KVM. Again, we create a PIC backend infrastructure and
provide hooks for init, reset, and vmload/save. This also introduces a
common helper to instantiate a single i8259 chip from the cascade-
creating i8259_init function.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
---
Makefile.objs | 2 +-
hw/i8259.c | 127 +++++---------------------------------
hw/i8259_common.c | 173 +++++++++++++++++++++++++++++++++++++++++++++++++++
hw/i8259_internal.h | 82 ++++++++++++++++++++++++
4 files changed, 271 insertions(+), 113 deletions(-)
create mode 100644 hw/i8259_common.c
create mode 100644 hw/i8259_internal.h
diff --git a/Makefile.objs b/Makefile.objs
index d7a6539..72d8ee7 100644
--- a/Makefile.objs
+++ b/Makefile.objs
@@ -221,7 +221,7 @@ hw-obj-$(CONFIG_APPLESMC) += applesmc.o
hw-obj-$(CONFIG_SMARTCARD) += usb-ccid.o ccid-card-passthru.o
hw-obj-$(CONFIG_SMARTCARD_NSS) += ccid-card-emulated.o
hw-obj-$(CONFIG_USB_REDIR) += usb-redir.o
-hw-obj-$(CONFIG_I8259) += i8259.o
+hw-obj-$(CONFIG_I8259) += i8259_common.o i8259.o
# PPC devices
hw-obj-$(CONFIG_PREP_PCI) += prep_pci.o
diff --git a/hw/i8259.c b/hw/i8259.c
index ab519de..413802c 100644
--- a/hw/i8259.c
+++ b/hw/i8259.c
@@ -26,6 +26,7 @@
#include "isa.h"
#include "monitor.h"
#include "qemu-timer.h"
+#include "i8259_internal.h"
/* debug PIC */
//#define DEBUG_PIC
@@ -40,33 +41,6 @@
//#define DEBUG_IRQ_LATENCY
//#define DEBUG_IRQ_COUNT
-struct PicState {
- ISADevice dev;
- uint8_t last_irr; /* edge detection */
- uint8_t irr; /* interrupt request register */
- uint8_t imr; /* interrupt mask register */
- uint8_t isr; /* interrupt service register */
- uint8_t priority_add; /* highest irq priority */
- uint8_t irq_base;
- uint8_t read_reg_select;
- uint8_t poll;
- uint8_t special_mask;
- uint8_t init_state;
- uint8_t auto_eoi;
- uint8_t rotate_on_auto_eoi;
- uint8_t special_fully_nested_mode;
- uint8_t init4; /* true if 4 byte init */
- uint8_t single_mode; /* true if slave pic is not initialized */
- uint8_t elcr; /* PIIX edge/trigger selection*/
- uint8_t elcr_mask;
- qemu_irq int_out[1];
- uint32_t master; /* reflects /SP input pin */
- uint32_t iobase;
- uint32_t elcr_addr;
- MemoryRegion base_io;
- MemoryRegion elcr_io;
-};
-
#if defined(DEBUG_PIC) || defined(DEBUG_IRQ_COUNT)
static int irq_level[16];
#endif
@@ -248,29 +222,12 @@ int pic_read_irq(PicState *s)
static void pic_init_reset(PicState *s)
{
- s->last_irr = 0;
- s->irr = 0;
- s->imr = 0;
- s->isr = 0;
- s->priority_add = 0;
- s->irq_base = 0;
- s->read_reg_select = 0;
- s->poll = 0;
- s->special_mask = 0;
- s->init_state = 0;
- s->auto_eoi = 0;
- s->rotate_on_auto_eoi = 0;
- s->special_fully_nested_mode = 0;
- s->init4 = 0;
- s->single_mode = 0;
- /* Note: ELCR is not reset */
+ pic_reset_internal(s);
pic_update_irq(s);
}
-static void pic_reset(DeviceState *dev)
+static void pic_reset(PicState *s)
{
- PicState *s = container_of(dev, PicState, dev.qdev);
-
pic_init_reset(s);
s->elcr = 0;
}
@@ -418,32 +375,6 @@ static uint64_t elcr_ioport_read(void *opaque, target_phys_addr_t addr,
return s->elcr;
}
-static const VMStateDescription vmstate_pic = {
- .name = "i8259",
- .version_id = 1,
- .minimum_version_id = 1,
- .minimum_version_id_old = 1,
- .fields = (VMStateField[]) {
- VMSTATE_UINT8(last_irr, PicState),
- VMSTATE_UINT8(irr, PicState),
- VMSTATE_UINT8(imr, PicState),
- VMSTATE_UINT8(isr, PicState),
- VMSTATE_UINT8(priority_add, PicState),
- VMSTATE_UINT8(irq_base, PicState),
- VMSTATE_UINT8(read_reg_select, PicState),
- VMSTATE_UINT8(poll, PicState),
- VMSTATE_UINT8(special_mask, PicState),
- VMSTATE_UINT8(init_state, PicState),
- VMSTATE_UINT8(auto_eoi, PicState),
- VMSTATE_UINT8(rotate_on_auto_eoi, PicState),
- VMSTATE_UINT8(special_fully_nested_mode, PicState),
- VMSTATE_UINT8(init4, PicState),
- VMSTATE_UINT8(single_mode, PicState),
- VMSTATE_UINT8(elcr, PicState),
- VMSTATE_END_OF_LIST()
- }
-};
-
static const MemoryRegionOps pic_base_ioport_ops = {
.read = pic_ioport_read,
.write = pic_ioport_write,
@@ -462,24 +393,13 @@ static const MemoryRegionOps pic_elcr_ioport_ops = {
},
};
-static int pic_initfn(ISADevice *dev)
+static void pic_backend_init(PicState *s)
{
- PicState *s = DO_UPCAST(PicState, dev, dev);
-
memory_region_init_io(&s->base_io, &pic_base_ioport_ops, s, "pic", 2);
memory_region_init_io(&s->elcr_io, &pic_elcr_ioport_ops, s, "elcr", 1);
- isa_register_ioport(NULL, &s->base_io, s->iobase);
- if (s->elcr_addr != -1) {
- isa_register_ioport(NULL, &s->elcr_io, s->elcr_addr);
- }
-
- qdev_init_gpio_out(&dev->qdev, s->int_out, ARRAY_SIZE(s->int_out));
- qdev_init_gpio_in(&dev->qdev, pic_set_irq, 8);
-
- qdev_set_legacy_instance_id(&dev->qdev, s->iobase, 1);
-
- return 0;
+ qdev_init_gpio_out(&s->dev.qdev, s->int_out, ARRAY_SIZE(s->int_out));
+ qdev_init_gpio_in(&s->dev.qdev, pic_set_irq, 8);
}
void pic_info(Monitor *mon)
@@ -526,12 +446,7 @@ qemu_irq *i8259_init(qemu_irq parent_irq)
irq_set = g_malloc(ISA_NUM_IRQS * sizeof(qemu_irq));
- dev = isa_create("isa-i8259");
- qdev_prop_set_uint32(&dev->qdev, "iobase", 0x20);
- qdev_prop_set_uint32(&dev->qdev, "elcr_addr", 0x4d0);
- qdev_prop_set_uint8(&dev->qdev, "elcr_mask", 0xf8);
- qdev_prop_set_bit(&dev->qdev, "master", true);
- qdev_init_nofail(&dev->qdev);
+ dev = i8259_init_chip(true, "QEMU");
qdev_connect_gpio_out(&dev->qdev, 0, parent_irq);
for (i = 0 ; i < 8; i++) {
@@ -540,11 +455,7 @@ qemu_irq *i8259_init(qemu_irq parent_irq)
isa_pic = DO_UPCAST(PicState, dev, dev);
- dev = isa_create("isa-i8259");
- qdev_prop_set_uint32(&dev->qdev, "iobase", 0xa0);
- qdev_prop_set_uint32(&dev->qdev, "elcr_addr", 0x4d1);
- qdev_prop_set_uint8(&dev->qdev, "elcr_mask", 0xde);
- qdev_init_nofail(&dev->qdev);
+ dev = i8259_init_chip(false, "QEMU");
qdev_connect_gpio_out(&dev->qdev, 0, irq_set[2]);
for (i = 0 ; i < 8; i++) {
@@ -556,24 +467,16 @@ qemu_irq *i8259_init(qemu_irq parent_irq)
return irq_set;
}
-static ISADeviceInfo i8259_info = {
- .qdev.name = "isa-i8259",
- .qdev.size = sizeof(PicState),
- .qdev.vmsd = &vmstate_pic,
- .qdev.reset = pic_reset,
- .qdev.no_user = 1,
- .init = pic_initfn,
- .qdev.props = (Property[]) {
- DEFINE_PROP_HEX32("iobase", PicState, iobase, -1),
- DEFINE_PROP_HEX32("elcr_addr", PicState, elcr_addr, -1),
- DEFINE_PROP_HEX8("elcr_mask", PicState, elcr_mask, -1),
- DEFINE_PROP_BIT("master", PicState, master, 0, false),
- DEFINE_PROP_END_OF_LIST(),
- },
+static PICBackend pic_backend = {
+ .name = "QEMU",
+ .init = pic_backend_init,
+ .reset = pic_reset,
};
static void pic_register(void)
{
- isa_qdev_register(&i8259_info);
+ pic_register_device();
+ pic_register_backend(&pic_backend);
}
+
device_init(pic_register)
diff --git a/hw/i8259_common.c b/hw/i8259_common.c
new file mode 100644
index 0000000..403077d
--- /dev/null
+++ b/hw/i8259_common.c
@@ -0,0 +1,173 @@
+/*
+ * QEMU 8259 - common bits of emulated and KVM kernel model
+ *
+ * Copyright (c) 2003-2004 Fabrice Bellard
+ * Copyright (c) 2011 Jan Kiszka, Siemens AG
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ */
+#include "pc.h"
+#include "i8259_internal.h"
+
+static QSIMPLEQ_HEAD(, PICBackend) backends =
+ QSIMPLEQ_HEAD_INITIALIZER(backends);
+
+void pic_reset_internal(PicState *s)
+{
+ s->last_irr = 0;
+ s->irr = 0;
+ s->imr = 0;
+ s->isr = 0;
+ s->priority_add = 0;
+ s->irq_base = 0;
+ s->read_reg_select = 0;
+ s->poll = 0;
+ s->special_mask = 0;
+ s->init_state = 0;
+ s->auto_eoi = 0;
+ s->rotate_on_auto_eoi = 0;
+ s->special_fully_nested_mode = 0;
+ s->init4 = 0;
+ s->single_mode = 0;
+ /* Note: ELCR is not reset */
+}
+
+static void pic_dispatch_pre_save(void *opaque)
+{
+ PicState *s = opaque;
+
+ if (s->backend->pre_save) {
+ s->backend->pre_save(s);
+ }
+}
+
+static int pic_dispatch_post_load(void *opaque, int version_id)
+{
+ PicState *s = opaque;
+
+ if (s->backend->post_load) {
+ s->backend->post_load(s);
+ }
+ return 0;
+}
+
+static const VMStateDescription vmstate_pic = {
+ .name = "i8259",
+ .version_id = 1,
+ .minimum_version_id = 1,
+ .minimum_version_id_old = 1,
+ .pre_save = pic_dispatch_pre_save,
+ .post_load = pic_dispatch_post_load,
+ .fields = (VMStateField[]) {
+ VMSTATE_UINT8(last_irr, PicState),
+ VMSTATE_UINT8(irr, PicState),
+ VMSTATE_UINT8(imr, PicState),
+ VMSTATE_UINT8(isr, PicState),
+ VMSTATE_UINT8(priority_add, PicState),
+ VMSTATE_UINT8(irq_base, PicState),
+ VMSTATE_UINT8(read_reg_select, PicState),
+ VMSTATE_UINT8(poll, PicState),
+ VMSTATE_UINT8(special_mask, PicState),
+ VMSTATE_UINT8(init_state, PicState),
+ VMSTATE_UINT8(auto_eoi, PicState),
+ VMSTATE_UINT8(rotate_on_auto_eoi, PicState),
+ VMSTATE_UINT8(special_fully_nested_mode, PicState),
+ VMSTATE_UINT8(init4, PicState),
+ VMSTATE_UINT8(single_mode, PicState),
+ VMSTATE_UINT8(elcr, PicState),
+ VMSTATE_END_OF_LIST()
+ }
+};
+
+static int pic_initfn(ISADevice *dev)
+{
+ PicState *s = DO_UPCAST(PicState, dev, dev);
+ PICBackend *b;
+
+ QSIMPLEQ_FOREACH(b, &backends, entry) {
+ if (strcmp(b->name, s->backend_name) == 0) {
+ s->backend = b;
+ break;
+ }
+ }
+ if (!s->backend) {
+ hw_error("PIC backend '%s' not found!", s->backend_name);
+ exit(1);
+ }
+
+ b->init(s);
+
+ isa_register_ioport(NULL, &s->base_io, s->iobase);
+ if (s->elcr_addr != -1) {
+ isa_register_ioport(NULL, &s->elcr_io, s->elcr_addr);
+ }
+
+ qdev_set_legacy_instance_id(&s->dev.qdev, s->iobase, 1);
+
+ return 0;
+}
+
+static void pic_reset(DeviceState *dev)
+{
+ PicState *s = container_of(dev, PicState, dev.qdev);
+
+ s->backend->reset(s);
+}
+
+static ISADeviceInfo i8259_info = {
+ .qdev.name = "isa-i8259",
+ .qdev.size = sizeof(PicState),
+ .qdev.vmsd = &vmstate_pic,
+ .qdev.reset = pic_reset,
+ .qdev.no_user = 1,
+ .init = pic_initfn,
+ .qdev.props = (Property[]) {
+ DEFINE_PROP_HEX32("iobase", PicState, iobase, -1),
+ DEFINE_PROP_HEX32("elcr_addr", PicState, elcr_addr, -1),
+ DEFINE_PROP_HEX8("elcr_mask", PicState, elcr_mask, -1),
+ DEFINE_PROP_BIT("master", PicState, master, 0, false),
+ DEFINE_PROP_STRING("backend", PicState, backend_name),
+ DEFINE_PROP_END_OF_LIST(),
+ },
+};
+
+void pic_register_backend(PICBackend *backend)
+{
+ QSIMPLEQ_INSERT_TAIL(&backends, backend, entry);
+}
+
+void pic_register_device(void)
+{
+ isa_qdev_register(&i8259_info);
+}
+
+ISADevice *i8259_init_chip(bool master, const char *backend)
+{
+ ISADevice *dev;
+
+ dev = isa_create("isa-i8259");
+ qdev_prop_set_uint32(&dev->qdev, "iobase", master ? 0x20 : 0xa0);
+ qdev_prop_set_uint32(&dev->qdev, "elcr_addr", master ? 0x4d0 : 0x4d1);
+ qdev_prop_set_uint8(&dev->qdev, "elcr_mask", master ? 0xf8 : 0xde);
+ qdev_prop_set_bit(&dev->qdev, "master", master);
+ qdev_prop_set_string(&dev->qdev, "backend", g_strdup(backend));
+ qdev_init_nofail(&dev->qdev);
+
+ return dev;
+}
diff --git a/hw/i8259_internal.h b/hw/i8259_internal.h
new file mode 100644
index 0000000..e11c312
--- /dev/null
+++ b/hw/i8259_internal.h
@@ -0,0 +1,82 @@
+/*
+ * QEMU 8259 - internal interfaces
+ *
+ * Copyright (c) 2011 Jan Kiszka, Siemens AG
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ */
+
+#ifndef QEMU_I8259_INTERNAL_H
+#define QEMU_I8259_INTERNAL_H
+
+#include "hw.h"
+#include "pc.h"
+#include "isa.h"
+#include "qemu-queue.h"
+
+typedef struct PICBackend PICBackend;
+
+struct PICBackend {
+ const char *name;
+ void (*init)(PicState *s);
+ void (*reset)(PicState *s);
+ void (*pre_save)(PicState *s);
+ void (*post_load)(PicState *s);
+
+ QSIMPLEQ_ENTRY(PICBackend) entry;
+};
+
+struct PicState {
+ ISADevice dev;
+ uint8_t last_irr; /* edge detection */
+ uint8_t irr; /* interrupt request register */
+ uint8_t imr; /* interrupt mask register */
+ uint8_t isr; /* interrupt service register */
+ uint8_t priority_add; /* highest irq priority */
+ uint8_t irq_base;
+ uint8_t read_reg_select;
+ uint8_t poll;
+ uint8_t special_mask;
+ uint8_t init_state;
+ uint8_t auto_eoi;
+ uint8_t rotate_on_auto_eoi;
+ uint8_t special_fully_nested_mode;
+ uint8_t init4; /* true if 4 byte init */
+ uint8_t single_mode; /* true if slave pic is not initialized */
+ uint8_t elcr; /* PIIX edge/trigger selection*/
+ uint8_t elcr_mask;
+ qemu_irq int_out[1];
+ uint32_t master; /* reflects /SP input pin */
+ uint32_t iobase;
+ uint32_t elcr_addr;
+ MemoryRegion base_io;
+ MemoryRegion elcr_io;
+
+ char *backend_name;
+ PICBackend *backend;
+};
+
+void pic_register_device(void);
+void pic_register_backend(PICBackend *backend);
+
+void pic_reset_internal(PicState *s);
+
+ISADevice *i8259_init_chip(bool master, const char *backend);
+
+#endif /* !QEMU_I8259_INTERNAL_H */
--
1.7.3.4
^ permalink raw reply related [flat|nested] 32+ messages in thread
* [Qemu-devel] [PATCH v4 08/15] ioapic: Introduce backend/frontend infrastructure for KVM reuse
2011-12-08 11:52 [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support Jan Kiszka
` (6 preceding siblings ...)
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 07/15] i8259: Introduce backend/frontend infrastructure for KVM reuse Jan Kiszka
@ 2011-12-08 11:52 ` Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 09/15] memory: Introduce memory_region_init_reservation Jan Kiszka
` (8 subsequent siblings)
16 siblings, 0 replies; 32+ messages in thread
From: Jan Kiszka @ 2011-12-08 11:52 UTC (permalink / raw)
To: Avi Kivity, Marcelo Tosatti
Cc: Blue Swirl, Anthony Liguori, qemu-devel, kvm, Michael S. Tsirkin
Split up the IOAPIC analogously to APIC and i8259. KVM will share the
device description, reset logic and certain init parts with the user
space model.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
---
Makefile.target | 2 +-
hw/ioapic.c | 130 ++++-------------------------------------------
hw/ioapic_common.c | 137 ++++++++++++++++++++++++++++++++++++++++++++++++++
hw/ioapic_internal.h | 105 ++++++++++++++++++++++++++++++++++++++
hw/pc_piix.c | 1 +
5 files changed, 254 insertions(+), 121 deletions(-)
create mode 100644 hw/ioapic_common.c
create mode 100644 hw/ioapic_internal.h
diff --git a/Makefile.target b/Makefile.target
index c46f062..b549988 100644
--- a/Makefile.target
+++ b/Makefile.target
@@ -231,7 +231,7 @@ obj-$(CONFIG_IVSHMEM) += ivshmem.o
# Hardware support
obj-i386-y += vga.o
obj-i386-y += mc146818rtc.o pc.o
-obj-i386-y += cirrus_vga.o sga.o apic_common.o apic.o ioapic.o piix_pci.o
+obj-i386-y += cirrus_vga.o sga.o apic_common.o apic.o ioapic_common.o ioapic.o piix_pci.o
obj-i386-y += vmport.o
obj-i386-y += device-hotplug.o pci-hotplug.o smbios.o wdt_ib700.o
obj-i386-y += debugcon.o multiboot.o
diff --git a/hw/ioapic.c b/hw/ioapic.c
index 27b07c6..2db72e0 100644
--- a/hw/ioapic.c
+++ b/hw/ioapic.c
@@ -24,9 +24,7 @@
#include "pc.h"
#include "apic.h"
#include "ioapic.h"
-#include "qemu-timer.h"
-#include "host-utils.h"
-#include "sysbus.h"
+#include "ioapic_internal.h"
//#define DEBUG_IOAPIC
@@ -37,62 +35,6 @@
#define DPRINTF(fmt, ...)
#endif
-#define MAX_IOAPICS 1
-
-#define IOAPIC_VERSION 0x11
-
-#define IOAPIC_LVT_DEST_SHIFT 56
-#define IOAPIC_LVT_MASKED_SHIFT 16
-#define IOAPIC_LVT_TRIGGER_MODE_SHIFT 15
-#define IOAPIC_LVT_REMOTE_IRR_SHIFT 14
-#define IOAPIC_LVT_POLARITY_SHIFT 13
-#define IOAPIC_LVT_DELIV_STATUS_SHIFT 12
-#define IOAPIC_LVT_DEST_MODE_SHIFT 11
-#define IOAPIC_LVT_DELIV_MODE_SHIFT 8
-
-#define IOAPIC_LVT_MASKED (1 << IOAPIC_LVT_MASKED_SHIFT)
-#define IOAPIC_LVT_REMOTE_IRR (1 << IOAPIC_LVT_REMOTE_IRR_SHIFT)
-
-#define IOAPIC_TRIGGER_EDGE 0
-#define IOAPIC_TRIGGER_LEVEL 1
-
-/*io{apic,sapic} delivery mode*/
-#define IOAPIC_DM_FIXED 0x0
-#define IOAPIC_DM_LOWEST_PRIORITY 0x1
-#define IOAPIC_DM_PMI 0x2
-#define IOAPIC_DM_NMI 0x4
-#define IOAPIC_DM_INIT 0x5
-#define IOAPIC_DM_SIPI 0x6
-#define IOAPIC_DM_EXTINT 0x7
-#define IOAPIC_DM_MASK 0x7
-
-#define IOAPIC_VECTOR_MASK 0xff
-
-#define IOAPIC_IOREGSEL 0x00
-#define IOAPIC_IOWIN 0x10
-
-#define IOAPIC_REG_ID 0x00
-#define IOAPIC_REG_VER 0x01
-#define IOAPIC_REG_ARB 0x02
-#define IOAPIC_REG_REDTBL_BASE 0x10
-#define IOAPIC_ID 0x00
-
-#define IOAPIC_ID_SHIFT 24
-#define IOAPIC_ID_MASK 0xf
-
-#define IOAPIC_VER_ENTRIES_SHIFT 16
-
-typedef struct IOAPICState IOAPICState;
-
-struct IOAPICState {
- SysBusDevice busdev;
- MemoryRegion io_memory;
- uint8_t id;
- uint8_t ioregsel;
- uint32_t irr;
- uint64_t ioredtbl[IOAPIC_NUM_PINS];
-};
-
static IOAPICState *ioapics[MAX_IOAPICS];
static void ioapic_service(IOAPICState *s)
@@ -278,83 +220,31 @@ ioapic_mem_write(void *opaque, target_phys_addr_t addr, uint64_t val,
}
}
-static int ioapic_post_load(void *opaque, int version_id)
-{
- IOAPICState *s = opaque;
-
- if (version_id == 1) {
- /* set sane value */
- s->irr = 0;
- }
- return 0;
-}
-
-static const VMStateDescription vmstate_ioapic = {
- .name = "ioapic",
- .version_id = 3,
- .post_load = ioapic_post_load,
- .minimum_version_id = 1,
- .minimum_version_id_old = 1,
- .fields = (VMStateField[]) {
- VMSTATE_UINT8(id, IOAPICState),
- VMSTATE_UINT8(ioregsel, IOAPICState),
- VMSTATE_UNUSED_V(2, 8), /* to account for qemu-kvm's v2 format */
- VMSTATE_UINT32_V(irr, IOAPICState, 2),
- VMSTATE_UINT64_ARRAY(ioredtbl, IOAPICState, IOAPIC_NUM_PINS),
- VMSTATE_END_OF_LIST()
- }
-};
-
-static void ioapic_reset(DeviceState *d)
-{
- IOAPICState *s = DO_UPCAST(IOAPICState, busdev.qdev, d);
- int i;
-
- s->id = 0;
- s->ioregsel = 0;
- s->irr = 0;
- for (i = 0; i < IOAPIC_NUM_PINS; i++) {
- s->ioredtbl[i] = 1 << IOAPIC_LVT_MASKED_SHIFT;
- }
-}
-
static const MemoryRegionOps ioapic_io_ops = {
.read = ioapic_mem_read,
.write = ioapic_mem_write,
.endianness = DEVICE_NATIVE_ENDIAN,
};
-static int ioapic_init1(SysBusDevice *dev)
+static void ioapic_backend_init(IOAPICState *s, int index)
{
- IOAPICState *s = FROM_SYSBUS(IOAPICState, dev);
- static int ioapic_no;
-
- if (ioapic_no >= MAX_IOAPICS) {
- return -1;
- }
-
memory_region_init_io(&s->io_memory, &ioapic_io_ops, s, "ioapic", 0x1000);
- sysbus_init_mmio(dev, &s->io_memory);
-
- qdev_init_gpio_in(&dev->qdev, ioapic_set_irq, IOAPIC_NUM_PINS);
- ioapics[ioapic_no++] = s;
+ qdev_init_gpio_in(&s->busdev.qdev, ioapic_set_irq, IOAPIC_NUM_PINS);
- return 0;
+ ioapics[index] = s;
}
-static SysBusDeviceInfo ioapic_info = {
- .init = ioapic_init1,
- .qdev.name = "ioapic",
- .qdev.size = sizeof(IOAPICState),
- .qdev.vmsd = &vmstate_ioapic,
- .qdev.reset = ioapic_reset,
- .qdev.no_user = 1,
+static IOAPICBackend ioapic_backend = {
+ .name = "QEMU",
+ .init = ioapic_backend_init,
+ .reset = ioapic_reset_internal,
};
static void ioapic_register_devices(void)
{
- sysbus_register_withprop(&ioapic_info);
+ ioapic_register_device();
+ ioapic_register_backend(&ioapic_backend);
}
device_init(ioapic_register_devices)
diff --git a/hw/ioapic_common.c b/hw/ioapic_common.c
new file mode 100644
index 0000000..094551c
--- /dev/null
+++ b/hw/ioapic_common.c
@@ -0,0 +1,137 @@
+/*
+ * IOAPIC emulation logic - common bits of emulated and KVM kernel model
+ *
+ * Copyright (c) 2004-2005 Fabrice Bellard
+ * Copyright (c) 2009 Xiantao Zhang, Intel
+ * Copyright (c) 2011 Jan Kiszka, Siemens AG
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "ioapic.h"
+#include "ioapic_internal.h"
+#include "sysbus.h"
+
+static QSIMPLEQ_HEAD(, IOAPICBackend) backends =
+ QSIMPLEQ_HEAD_INITIALIZER(backends);
+
+void ioapic_reset_internal(IOAPICState *s)
+{
+ int i;
+
+ s->id = 0;
+ s->ioregsel = 0;
+ s->irr = 0;
+ for (i = 0; i < IOAPIC_NUM_PINS; i++) {
+ s->ioredtbl[i] = 1 << IOAPIC_LVT_MASKED_SHIFT;
+ }
+}
+
+static void ioapic_dispatch_pre_save(void *opaque)
+{
+ IOAPICState *s = opaque;
+
+ if (s->backend->pre_save) {
+ s->backend->pre_save(s);
+ }
+}
+
+static int ioapic_post_load(void *opaque, int version_id)
+{
+ IOAPICState *s = opaque;
+
+ if (version_id == 1) {
+ /* set sane value */
+ s->irr = 0;
+ }
+ if (s->backend->post_load) {
+ s->backend->post_load(s);
+ }
+ return 0;
+}
+
+const VMStateDescription vmstate_ioapic = {
+ .name = "ioapic",
+ .version_id = 3,
+ .minimum_version_id = 1,
+ .minimum_version_id_old = 1,
+ .pre_save = ioapic_dispatch_pre_save,
+ .post_load = ioapic_post_load,
+ .fields = (VMStateField[]) {
+ VMSTATE_UINT8(id, IOAPICState),
+ VMSTATE_UINT8(ioregsel, IOAPICState),
+ VMSTATE_UNUSED_V(2, 8), /* to account for qemu-kvm's v2 format */
+ VMSTATE_UINT32_V(irr, IOAPICState, 2),
+ VMSTATE_UINT64_ARRAY(ioredtbl, IOAPICState, IOAPIC_NUM_PINS),
+ VMSTATE_END_OF_LIST()
+ }
+};
+
+static int ioapic_init(SysBusDevice *dev)
+{
+ IOAPICState *s = FROM_SYSBUS(IOAPICState, dev);
+ static int ioapic_no;
+ IOAPICBackend *b;
+
+ if (ioapic_no >= MAX_IOAPICS) {
+ return -1;
+ }
+
+ QSIMPLEQ_FOREACH(b, &backends, entry) {
+ if (strcmp(b->name, s->backend_name) == 0) {
+ s->backend = b;
+ break;
+ }
+ }
+ if (!s->backend) {
+ hw_error("IOAPIC backend '%s' not found!", s->backend_name);
+ exit(1);
+ }
+
+ b->init(s, ioapic_no++);
+
+ sysbus_init_mmio(&s->busdev, &s->io_memory);
+
+ return 0;
+}
+
+static void ioapic_reset(DeviceState *dev)
+{
+ IOAPICState *s = DO_UPCAST(IOAPICState, busdev.qdev, dev);
+
+ s->backend->reset(s);
+}
+
+static SysBusDeviceInfo ioapic_info = {
+ .init = ioapic_init,
+ .qdev.name = "ioapic",
+ .qdev.size = sizeof(IOAPICState),
+ .qdev.vmsd = &vmstate_ioapic,
+ .qdev.reset = ioapic_reset,
+ .qdev.no_user = 1,
+ .qdev.props = (Property[]) {
+ DEFINE_PROP_STRING("backend", IOAPICState, backend_name),
+ DEFINE_PROP_END_OF_LIST(),
+ },
+};
+
+void ioapic_register_backend(IOAPICBackend *backend)
+{
+ QSIMPLEQ_INSERT_TAIL(&backends, backend, entry);
+}
+
+void ioapic_register_device(void)
+{
+ sysbus_register_withprop(&ioapic_info);
+}
diff --git a/hw/ioapic_internal.h b/hw/ioapic_internal.h
new file mode 100644
index 0000000..c5fab8b
--- /dev/null
+++ b/hw/ioapic_internal.h
@@ -0,0 +1,105 @@
+/*
+ * IOAPIC emulation logic - internal interfaces
+ *
+ * Copyright (c) 2004-2005 Fabrice Bellard
+ * Copyright (c) 2009 Xiantao Zhang, Intel
+ * Copyright (c) 2011 Jan Kiszka, Siemens AG
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef QEMU_IOAPIC_INTERNAL_H
+#define QEMU_IOAPIC_INTERNAL_H
+
+#include "hw.h"
+#include "memory.h"
+#include "sysbus.h"
+#include "qemu-queue.h"
+
+#define MAX_IOAPICS 1
+
+#define IOAPIC_VERSION 0x11
+
+#define IOAPIC_LVT_DEST_SHIFT 56
+#define IOAPIC_LVT_MASKED_SHIFT 16
+#define IOAPIC_LVT_TRIGGER_MODE_SHIFT 15
+#define IOAPIC_LVT_REMOTE_IRR_SHIFT 14
+#define IOAPIC_LVT_POLARITY_SHIFT 13
+#define IOAPIC_LVT_DELIV_STATUS_SHIFT 12
+#define IOAPIC_LVT_DEST_MODE_SHIFT 11
+#define IOAPIC_LVT_DELIV_MODE_SHIFT 8
+
+#define IOAPIC_LVT_MASKED (1 << IOAPIC_LVT_MASKED_SHIFT)
+#define IOAPIC_LVT_REMOTE_IRR (1 << IOAPIC_LVT_REMOTE_IRR_SHIFT)
+
+#define IOAPIC_TRIGGER_EDGE 0
+#define IOAPIC_TRIGGER_LEVEL 1
+
+/*io{apic,sapic} delivery mode*/
+#define IOAPIC_DM_FIXED 0x0
+#define IOAPIC_DM_LOWEST_PRIORITY 0x1
+#define IOAPIC_DM_PMI 0x2
+#define IOAPIC_DM_NMI 0x4
+#define IOAPIC_DM_INIT 0x5
+#define IOAPIC_DM_SIPI 0x6
+#define IOAPIC_DM_EXTINT 0x7
+#define IOAPIC_DM_MASK 0x7
+
+#define IOAPIC_VECTOR_MASK 0xff
+
+#define IOAPIC_IOREGSEL 0x00
+#define IOAPIC_IOWIN 0x10
+
+#define IOAPIC_REG_ID 0x00
+#define IOAPIC_REG_VER 0x01
+#define IOAPIC_REG_ARB 0x02
+#define IOAPIC_REG_REDTBL_BASE 0x10
+#define IOAPIC_ID 0x00
+
+#define IOAPIC_ID_SHIFT 24
+#define IOAPIC_ID_MASK 0xf
+
+#define IOAPIC_VER_ENTRIES_SHIFT 16
+
+typedef struct IOAPICBackend IOAPICBackend;
+typedef struct IOAPICState IOAPICState;
+
+struct IOAPICBackend {
+ const char *name;
+ void (*init)(IOAPICState *s, int index);
+ void (*reset)(IOAPICState *s);
+ void (*pre_save)(IOAPICState *s);
+ void (*post_load)(IOAPICState *s);
+
+ QSIMPLEQ_ENTRY(IOAPICBackend) entry;
+};
+
+struct IOAPICState {
+ SysBusDevice busdev;
+ MemoryRegion io_memory;
+ uint8_t id;
+ uint8_t ioregsel;
+ uint32_t irr;
+ uint64_t ioredtbl[IOAPIC_NUM_PINS];
+
+ char *backend_name;
+ IOAPICBackend *backend;
+};
+
+void ioapic_register_device(void);
+void ioapic_register_backend(IOAPICBackend *backend);
+
+void ioapic_reset_internal(IOAPICState *s);
+
+#endif /* !QEMU_IOAPIC_INTERNAL_H */
diff --git a/hw/pc_piix.c b/hw/pc_piix.c
index 530fe9c..98f2822 100644
--- a/hw/pc_piix.c
+++ b/hw/pc_piix.c
@@ -60,6 +60,7 @@ static void ioapic_init(GSIState *gsi_state)
unsigned int i;
dev = qdev_create(NULL, "ioapic");
+ qdev_prop_set_string(dev, "backend", g_strdup("QEMU"));
qdev_init_nofail(dev);
d = sysbus_from_qdev(dev);
sysbus_mmio_map(d, 0, 0xfec00000);
--
1.7.3.4
^ permalink raw reply related [flat|nested] 32+ messages in thread
* [Qemu-devel] [PATCH v4 09/15] memory: Introduce memory_region_init_reservation
2011-12-08 11:52 [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support Jan Kiszka
` (7 preceding siblings ...)
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 08/15] ioapic: " Jan Kiszka
@ 2011-12-08 11:52 ` Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 10/15] kvm: Introduce core services for in-kernel irqchip support Jan Kiszka
` (7 subsequent siblings)
16 siblings, 0 replies; 32+ messages in thread
From: Jan Kiszka @ 2011-12-08 11:52 UTC (permalink / raw)
To: Avi Kivity, Marcelo Tosatti
Cc: Blue Swirl, Anthony Liguori, qemu-devel, kvm, Michael S. Tsirkin
Introduce a memory region type that can reserve I/O space. Such regions
are useful for modeling I/O that is only handled outside of QEMU, i.e.
in the context of an accelerator like KVM.
Any access to such a region from QEMU is a bug, but could theoretically
be triggered by guest code (DMA to reserved region). So only warning
about such events once, then ignore them.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
---
memory.c | 36 ++++++++++++++++++++++++++++++++++++
memory.h | 16 ++++++++++++++++
2 files changed, 52 insertions(+), 0 deletions(-)
diff --git a/memory.c b/memory.c
index adfdf14..71a252a 100644
--- a/memory.c
+++ b/memory.c
@@ -1031,6 +1031,42 @@ void memory_region_init_rom_device(MemoryRegion *mr,
mr->backend_registered = true;
}
+static uint64_t invalid_read(void *opaque, target_phys_addr_t addr,
+ unsigned size)
+{
+ MemoryRegion *mr = opaque;
+
+ if (!mr->warning_printed) {
+ fprintf(stderr, "Invalid read from memory region %s\n", mr->name);
+ mr->warning_printed = true;
+ }
+ return -1U;
+}
+
+static void invalid_write(void *opaque, target_phys_addr_t addr, uint64_t data,
+ unsigned size)
+{
+ MemoryRegion *mr = opaque;
+
+ if (!mr->warning_printed) {
+ fprintf(stderr, "Invalid write to memory region %s\n", mr->name);
+ mr->warning_printed = true;
+ }
+}
+
+static const MemoryRegionOps reservation_ops = {
+ .read = invalid_read,
+ .write = invalid_write,
+ .endianness = DEVICE_NATIVE_ENDIAN,
+};
+
+void memory_region_init_reservation(MemoryRegion *mr,
+ const char *name,
+ uint64_t size)
+{
+ memory_region_init_io(mr, &reservation_ops, mr, name, size);
+}
+
void memory_region_destroy(MemoryRegion *mr)
{
assert(QTAILQ_EMPTY(&mr->subregions));
diff --git a/memory.h b/memory.h
index 53bf261..1097eac 100644
--- a/memory.h
+++ b/memory.h
@@ -123,6 +123,7 @@ struct MemoryRegion {
bool terminates;
bool readable;
bool readonly; /* For RAM regions */
+ bool warning_printed; /* For reservations */
MemoryRegion *alias;
target_phys_addr_t alias_offset;
unsigned priority;
@@ -250,6 +251,21 @@ void memory_region_init_rom_device(MemoryRegion *mr,
uint64_t size);
/**
+ * memory_region_init_reservation: Initialize a memory region that reserves
+ * I/O space.
+ *
+ * A reservation region primariy serves debugging purposes. It claims I/O
+ * space that is not supposed to be handled by QEMU itself. Any access via
+ * the memory API will cause an abort().
+ *
+ * @mr: the #MemoryRegion to be initialized
+ * @name: used for debugging; not visible to the user or ABI
+ * @size: size of the region.
+ */
+void memory_region_init_reservation(MemoryRegion *mr,
+ const char *name,
+ uint64_t size);
+/**
* memory_region_destroy: Destroy a memory region and relaim all resources.
*
* @mr: the region to be destroyed. May not currently be a subregion
--
1.7.3.4
^ permalink raw reply related [flat|nested] 32+ messages in thread
* [Qemu-devel] [PATCH v4 10/15] kvm: Introduce core services for in-kernel irqchip support
2011-12-08 11:52 [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support Jan Kiszka
` (8 preceding siblings ...)
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 09/15] memory: Introduce memory_region_init_reservation Jan Kiszka
@ 2011-12-08 11:52 ` Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 11/15] kvm: x86: Establish IRQ0 override control Jan Kiszka
` (6 subsequent siblings)
16 siblings, 0 replies; 32+ messages in thread
From: Jan Kiszka @ 2011-12-08 11:52 UTC (permalink / raw)
To: Avi Kivity, Marcelo Tosatti
Cc: Blue Swirl, Anthony Liguori, qemu-devel, kvm, Michael S. Tsirkin
Add the basic infrastructure to active in-kernel irqchip support, inject
interrupts into these models, and maintain IRQ routes.
Routing is optional and depends on the host arch supporting
KVM_CAP_IRQ_ROUTING. When it's not available on x86, we looe the HPET as
we can't route GSI0 to IOAPIC pin 2.
In-kernel irqchip support will once be controlled by the machine
property 'kernel_irqchip', but this is not yet wired up.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
---
kvm-all.c | 149 +++++++++++++++++++++++++++++++++++++++++++++++++++++
kvm.h | 8 +++
target-i386/kvm.c | 11 ++++
3 files changed, 168 insertions(+), 0 deletions(-)
diff --git a/kvm-all.c b/kvm-all.c
index 4c466d6..8958abd 100644
--- a/kvm-all.c
+++ b/kvm-all.c
@@ -77,6 +77,13 @@ struct KVMState
int pit_in_kernel;
int xsave, xcrs;
int many_ioeventfds;
+ int irqchip_inject_ioctl;
+#ifdef KVM_CAP_IRQ_ROUTING
+ struct kvm_irq_routing *irq_routes;
+ int nr_allocated_irq_routes;
+ uint32_t *used_gsi_bitmap;
+ unsigned int max_gsi;
+#endif
};
KVMState *kvm_state;
@@ -693,6 +700,138 @@ static void kvm_handle_interrupt(CPUState *env, int mask)
}
}
+int kvm_irqchip_set_irq(KVMState *s, int irq, int level)
+{
+ struct kvm_irq_level event;
+ int ret;
+
+ assert(s->irqchip_in_kernel);
+
+ event.level = level;
+ event.irq = irq;
+ ret = kvm_vm_ioctl(s, s->irqchip_inject_ioctl, &event);
+ if (ret < 0) {
+ perror("kvm_set_irqchip_line");
+ abort();
+ }
+
+ return (s->irqchip_inject_ioctl == KVM_IRQ_LINE) ? 1 : event.status;
+}
+
+#ifdef KVM_CAP_IRQ_ROUTING
+static void set_gsi(KVMState *s, unsigned int gsi)
+{
+ assert(gsi < s->max_gsi);
+
+ s->used_gsi_bitmap[gsi / 32] |= 1U << (gsi % 32);
+}
+
+static void kvm_init_irq_routing(KVMState *s)
+{
+ int gsi_count;
+
+ gsi_count = kvm_check_extension(s, KVM_CAP_IRQ_ROUTING);
+ if (gsi_count > 0) {
+ unsigned int gsi_bits, i;
+
+ /* Round up so we can search ints using ffs */
+ gsi_bits = (gsi_count + 31) / 32;
+ s->used_gsi_bitmap = g_malloc0(gsi_bits / 8);
+ s->max_gsi = gsi_bits;
+
+ /* Mark any over-allocated bits as already in use */
+ for (i = gsi_count; i < gsi_bits; i++) {
+ set_gsi(s, i);
+ }
+ }
+
+ s->irq_routes = g_malloc0(sizeof(*s->irq_routes));
+ s->nr_allocated_irq_routes = 0;
+
+ kvm_arch_init_irq_routing(s);
+}
+
+static void kvm_add_routing_entry(KVMState *s,
+ struct kvm_irq_routing_entry *entry)
+{
+ struct kvm_irq_routing_entry *new;
+ int n, size;
+
+ if (s->irq_routes->nr == s->nr_allocated_irq_routes) {
+ n = s->nr_allocated_irq_routes * 2;
+ if (n < 64) {
+ n = 64;
+ }
+ size = sizeof(struct kvm_irq_routing);
+ size += n * sizeof(*new);
+ s->irq_routes = g_realloc(s->irq_routes, size);
+ s->nr_allocated_irq_routes = n;
+ }
+ n = s->irq_routes->nr++;
+ new = &s->irq_routes->entries[n];
+ memset(new, 0, sizeof(*new));
+ new->gsi = entry->gsi;
+ new->type = entry->type;
+ new->flags = entry->flags;
+ new->u = entry->u;
+
+ set_gsi(s, entry->gsi);
+}
+
+void kvm_irqchip_add_route(KVMState *s, int irq, int irqchip, int pin)
+{
+ struct kvm_irq_routing_entry e;
+
+ e.gsi = irq;
+ e.type = KVM_IRQ_ROUTING_IRQCHIP;
+ e.flags = 0;
+ e.u.irqchip.irqchip = irqchip;
+ e.u.irqchip.pin = pin;
+ kvm_add_routing_entry(s, &e);
+}
+
+int kvm_irqchip_commit_routes(KVMState *s)
+{
+ s->irq_routes->flags = 0;
+ return kvm_vm_ioctl(s, KVM_SET_GSI_ROUTING, s->irq_routes);
+}
+
+#else /* !KVM_CAP_IRQ_ROUTING */
+
+static void kvm_init_irq_routing(KVMState *s)
+{
+}
+#endif /* !KVM_CAP_IRQ_ROUTING */
+
+static int kvm_irqchip_create(KVMState *s)
+{
+ QemuOptsList *list = qemu_find_opts("machine");
+ int ret;
+
+ if (QTAILQ_EMPTY(&list->head) ||
+ !qemu_opt_get_bool(QTAILQ_FIRST(&list->head),
+ "kernel_irqchip", false) ||
+ !kvm_check_extension(s, KVM_CAP_IRQCHIP)) {
+ return 0;
+ }
+
+ ret = kvm_vm_ioctl(s, KVM_CREATE_IRQCHIP);
+ if (ret < 0) {
+ fprintf(stderr, "Create kernel irqchip failed\n");
+ return ret;
+ }
+
+ s->irqchip_inject_ioctl = KVM_IRQ_LINE;
+ if (kvm_check_extension(s, KVM_CAP_IRQ_INJECT_STATUS)) {
+ s->irqchip_inject_ioctl = KVM_IRQ_LINE_STATUS;
+ }
+ s->irqchip_in_kernel = 1;
+
+ kvm_init_irq_routing(s);
+
+ return 0;
+}
+
int kvm_init(void)
{
static const char upgrade_note[] =
@@ -788,6 +927,11 @@ int kvm_init(void)
goto err;
}
+ ret = kvm_irqchip_create(s);
+ if (ret < 0) {
+ goto err;
+ }
+
kvm_state = s;
cpu_register_phys_memory_client(&kvm_cpu_phys_memory_client);
@@ -1122,6 +1266,11 @@ int kvm_has_many_ioeventfds(void)
return kvm_state->many_ioeventfds;
}
+int kvm_has_gsi_routing(void)
+{
+ return kvm_check_extension(kvm_state, KVM_CAP_IRQ_ROUTING);
+}
+
void kvm_setup_guest_memory(void *start, size_t size)
{
if (!kvm_has_sync_mmu()) {
diff --git a/kvm.h b/kvm.h
index 243b063..0d6c453 100644
--- a/kvm.h
+++ b/kvm.h
@@ -51,6 +51,7 @@ int kvm_has_debugregs(void);
int kvm_has_xsave(void);
int kvm_has_xcrs(void);
int kvm_has_many_ioeventfds(void);
+int kvm_has_gsi_routing(void);
#ifdef NEED_CPU_H
int kvm_init_vcpu(CPUState *env);
@@ -124,6 +125,13 @@ void kvm_arch_reset_vcpu(CPUState *env);
int kvm_arch_on_sigbus_vcpu(CPUState *env, int code, void *addr);
int kvm_arch_on_sigbus(int code, void *addr);
+void kvm_arch_init_irq_routing(KVMState *s);
+
+int kvm_irqchip_set_irq(KVMState *s, int irq, int level);
+
+void kvm_irqchip_add_route(KVMState *s, int gsi, int irqchip, int pin);
+int kvm_irqchip_commit_routes(KVMState *s);
+
struct kvm_guest_debug;
struct kvm_debug_exit_arch;
diff --git a/target-i386/kvm.c b/target-i386/kvm.c
index d206852..9d1191f 100644
--- a/target-i386/kvm.c
+++ b/target-i386/kvm.c
@@ -1879,3 +1879,14 @@ bool kvm_arch_stop_on_emulation_error(CPUState *env)
return !(env->cr[0] & CR0_PE_MASK) ||
((env->segs[R_CS].selector & 3) != 3);
}
+
+void kvm_arch_init_irq_routing(KVMState *s)
+{
+ if (!kvm_check_extension(s, KVM_CAP_IRQ_ROUTING)) {
+ /* If kernel can't do irq routing, interrupt source
+ * override 0->2 cannot be set up as required by HPET.
+ * So we have to disable it.
+ */
+ no_hpet = 1;
+ }
+}
--
1.7.3.4
^ permalink raw reply related [flat|nested] 32+ messages in thread
* [Qemu-devel] [PATCH v4 11/15] kvm: x86: Establish IRQ0 override control
2011-12-08 11:52 [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support Jan Kiszka
` (9 preceding siblings ...)
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 10/15] kvm: Introduce core services for in-kernel irqchip support Jan Kiszka
@ 2011-12-08 11:52 ` Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 12/15] kvm: x86: Add user space part for in-kernel APIC Jan Kiszka
` (5 subsequent siblings)
16 siblings, 0 replies; 32+ messages in thread
From: Jan Kiszka @ 2011-12-08 11:52 UTC (permalink / raw)
To: Avi Kivity, Marcelo Tosatti
Cc: Blue Swirl, Anthony Liguori, qemu-devel, kvm, Michael S. Tsirkin
KVM is forced to disable the IRQ0 override when we run with in-kernel
irqchip but without IRQ routing support of the kernel. Set the fwcfg
value correspondingly. This aligns us with qemu-kvm.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
---
hw/pc.c | 3 ++-
kvm-all.c | 5 +++++
kvm-stub.c | 5 +++++
kvm.h | 2 ++
sysemu.h | 1 -
vl.c | 1 -
6 files changed, 14 insertions(+), 3 deletions(-)
diff --git a/hw/pc.c b/hw/pc.c
index ee6e59b..066edc4 100644
--- a/hw/pc.c
+++ b/hw/pc.c
@@ -39,6 +39,7 @@
#include "msi.h"
#include "sysbus.h"
#include "sysemu.h"
+#include "kvm.h"
#include "blockdev.h"
#include "ui/qemu-spice.h"
#include "memory.h"
@@ -609,7 +610,7 @@ static void *bochs_bios_init(void)
fw_cfg_add_i64(fw_cfg, FW_CFG_RAM_SIZE, (uint64_t)ram_size);
fw_cfg_add_bytes(fw_cfg, FW_CFG_ACPI_TABLES, (uint8_t *)acpi_tables,
acpi_tables_len);
- fw_cfg_add_bytes(fw_cfg, FW_CFG_IRQ0_OVERRIDE, &irq0override, 1);
+ fw_cfg_add_i32(fw_cfg, FW_CFG_IRQ0_OVERRIDE, kvm_allows_irq0_override());
smbios_table = smbios_get_table(&smbios_len);
if (smbios_table)
diff --git a/kvm-all.c b/kvm-all.c
index 8958abd..7387dd3 100644
--- a/kvm-all.c
+++ b/kvm-all.c
@@ -1271,6 +1271,11 @@ int kvm_has_gsi_routing(void)
return kvm_check_extension(kvm_state, KVM_CAP_IRQ_ROUTING);
}
+int kvm_allows_irq0_override(void)
+{
+ return !kvm_enabled() || !kvm_irqchip_in_kernel() || kvm_has_gsi_routing();
+}
+
void kvm_setup_guest_memory(void *start, size_t size)
{
if (!kvm_has_sync_mmu()) {
diff --git a/kvm-stub.c b/kvm-stub.c
index 06064b9..6c2b06b 100644
--- a/kvm-stub.c
+++ b/kvm-stub.c
@@ -78,6 +78,11 @@ int kvm_has_many_ioeventfds(void)
return 0;
}
+int kvm_allows_irq0_override(void)
+{
+ return 1;
+}
+
void kvm_setup_guest_memory(void *start, size_t size)
{
}
diff --git a/kvm.h b/kvm.h
index 0d6c453..a3c87af 100644
--- a/kvm.h
+++ b/kvm.h
@@ -53,6 +53,8 @@ int kvm_has_xcrs(void);
int kvm_has_many_ioeventfds(void);
int kvm_has_gsi_routing(void);
+int kvm_allows_irq0_override(void);
+
#ifdef NEED_CPU_H
int kvm_init_vcpu(CPUState *env);
diff --git a/sysemu.h b/sysemu.h
index 22cd720..3bd896e 100644
--- a/sysemu.h
+++ b/sysemu.h
@@ -102,7 +102,6 @@ extern int vga_interface_type;
extern int graphic_width;
extern int graphic_height;
extern int graphic_depth;
-extern uint8_t irq0override;
extern DisplayType display_type;
extern const char *keyboard_layout;
extern int win2k_install_hack;
diff --git a/vl.c b/vl.c
index de5ecef..f9a8caf 100644
--- a/vl.c
+++ b/vl.c
@@ -218,7 +218,6 @@ int no_reboot = 0;
int no_shutdown = 0;
int cursor_hide = 1;
int graphic_rotate = 0;
-uint8_t irq0override = 1;
const char *watchdog;
QEMUOptionRom option_rom[MAX_OPTION_ROMS];
int nb_option_roms;
--
1.7.3.4
^ permalink raw reply related [flat|nested] 32+ messages in thread
* [Qemu-devel] [PATCH v4 12/15] kvm: x86: Add user space part for in-kernel APIC
2011-12-08 11:52 [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support Jan Kiszka
` (10 preceding siblings ...)
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 11/15] kvm: x86: Establish IRQ0 override control Jan Kiszka
@ 2011-12-08 11:52 ` Jan Kiszka
2011-12-08 21:16 ` Blue Swirl
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 13/15] kvm: x86: Add user space part for in-kernel i8259 Jan Kiszka
` (4 subsequent siblings)
16 siblings, 1 reply; 32+ messages in thread
From: Jan Kiszka @ 2011-12-08 11:52 UTC (permalink / raw)
To: Avi Kivity, Marcelo Tosatti
Cc: Anthony Liguori, Lai Jiangshan, kvm, Michael S. Tsirkin,
qemu-devel, Blue Swirl
This introduces the alternative APIC backend which makes use of KVM's
in-kernel device model. External NMI injection via LINT1 is emulated by
checking the current state of the in-kernel APIC, only injecting a NMI
into the VCPU if LINT1 is unmasked and configured to DM_NMI.
MSI is not yet supported, so we disable this when the in-kernel model is
in use.
CC: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
---
Makefile.target | 2 +-
hw/kvm/apic.c | 154 +++++++++++++++++++++++++++++++++++++++++++++++++++++
hw/pc.c | 15 ++++--
kvm.h | 3 +
target-i386/kvm.c | 8 +++
5 files changed, 176 insertions(+), 6 deletions(-)
create mode 100644 hw/kvm/apic.c
diff --git a/Makefile.target b/Makefile.target
index b549988..76de485 100644
--- a/Makefile.target
+++ b/Makefile.target
@@ -236,7 +236,7 @@ obj-i386-y += vmport.o
obj-i386-y += device-hotplug.o pci-hotplug.o smbios.o wdt_ib700.o
obj-i386-y += debugcon.o multiboot.o
obj-i386-y += pc_piix.o
-obj-i386-$(CONFIG_KVM) += kvm/clock.o
+obj-i386-$(CONFIG_KVM) += kvm/clock.o kvm/apic.o
obj-i386-$(CONFIG_SPICE) += qxl.o qxl-logger.o qxl-render.o
# shared objects
diff --git a/hw/kvm/apic.c b/hw/kvm/apic.c
new file mode 100644
index 0000000..3924f9e
--- /dev/null
+++ b/hw/kvm/apic.c
@@ -0,0 +1,154 @@
+/*
+ * KVM in-kernel APIC support
+ *
+ * Copyright (c) 2011 Siemens AG
+ *
+ * Authors:
+ * Jan Kiszka <jan.kiszka@siemens.com>
+ *
+ * This work is licensed under the terms of the GNU GPL version 2.
+ * See the COPYING file in the top-level directory.
+ */
+#include "hw/apic_internal.h"
+#include "kvm.h"
+
+static inline void kvm_apic_set_reg(struct kvm_lapic_state *kapic,
+ int reg_id, uint32_t val)
+{
+ *((uint32_t *)(kapic->regs + (reg_id << 4))) = val;
+}
+
+static inline uint32_t kvm_apic_get_reg(struct kvm_lapic_state *kapic,
+ int reg_id)
+{
+ return *((uint32_t *)(kapic->regs + (reg_id << 4)));
+}
+
+int kvm_put_apic(CPUState *env)
+{
+ APICState *s = DO_UPCAST(APICState, busdev.qdev, env->apic_state);
+ struct kvm_lapic_state kapic;
+ int i;
+
+ if (s && kvm_enabled() && kvm_irqchip_in_kernel()) {
+ memset(&kapic, 0, sizeof(kapic));
+ kvm_apic_set_reg(&kapic, 0x2, s->id << 24);
+ kvm_apic_set_reg(&kapic, 0x8, s->tpr);
+ kvm_apic_set_reg(&kapic, 0xd, s->log_dest << 24);
+ kvm_apic_set_reg(&kapic, 0xe, s->dest_mode << 28 | 0x0fffffff);
+ kvm_apic_set_reg(&kapic, 0xf, s->spurious_vec);
+ for (i = 0; i < 8; i++) {
+ kvm_apic_set_reg(&kapic, 0x10 + i, s->isr[i]);
+ kvm_apic_set_reg(&kapic, 0x18 + i, s->tmr[i]);
+ kvm_apic_set_reg(&kapic, 0x20 + i, s->irr[i]);
+ }
+ kvm_apic_set_reg(&kapic, 0x28, s->esr);
+ kvm_apic_set_reg(&kapic, 0x30, s->icr[0]);
+ kvm_apic_set_reg(&kapic, 0x31, s->icr[1]);
+ for (i = 0; i < APIC_LVT_NB; i++) {
+ kvm_apic_set_reg(&kapic, 0x32 + i, s->lvt[i]);
+ }
+ kvm_apic_set_reg(&kapic, 0x38, s->initial_count);
+ kvm_apic_set_reg(&kapic, 0x3e, s->divide_conf);
+
+ return kvm_vcpu_ioctl(env, KVM_SET_LAPIC, &kapic);
+ }
+
+ return 0;
+}
+
+int kvm_get_apic(CPUState *env)
+{
+ APICState *s = DO_UPCAST(APICState, busdev.qdev, env->apic_state);
+ struct kvm_lapic_state kapic;
+ int ret, i, v;
+
+ if (s && kvm_enabled() && kvm_irqchip_in_kernel()) {
+ ret = kvm_vcpu_ioctl(env, KVM_GET_LAPIC, &kapic);
+ if (ret < 0) {
+ return ret;
+ }
+
+ s->id = kvm_apic_get_reg(&kapic, 0x2) >> 24;
+ s->tpr = kvm_apic_get_reg(&kapic, 0x8);
+ s->arb_id = kvm_apic_get_reg(&kapic, 0x9);
+ s->log_dest = kvm_apic_get_reg(&kapic, 0xd) >> 24;
+ s->dest_mode = kvm_apic_get_reg(&kapic, 0xe) >> 28;
+ s->spurious_vec = kvm_apic_get_reg(&kapic, 0xf);
+ for (i = 0; i < 8; i++) {
+ s->isr[i] = kvm_apic_get_reg(&kapic, 0x10 + i);
+ s->tmr[i] = kvm_apic_get_reg(&kapic, 0x18 + i);
+ s->irr[i] = kvm_apic_get_reg(&kapic, 0x20 + i);
+ }
+ s->esr = kvm_apic_get_reg(&kapic, 0x28);
+ s->icr[0] = kvm_apic_get_reg(&kapic, 0x30);
+ s->icr[1] = kvm_apic_get_reg(&kapic, 0x31);
+ for (i = 0; i < APIC_LVT_NB; i++) {
+ s->lvt[i] = kvm_apic_get_reg(&kapic, 0x32 + i);
+ }
+ s->initial_count = kvm_apic_get_reg(&kapic, 0x38);
+ s->divide_conf = kvm_apic_get_reg(&kapic, 0x3e);
+
+ v = (s->divide_conf & 3) | ((s->divide_conf >> 1) & 4);
+ s->count_shift = (v + 1) & 7;
+
+ s->initial_count_load_time = qemu_get_clock_ns(vm_clock);
+ apic_next_timer(s, s->initial_count_load_time);
+ }
+ return 0;
+}
+
+static void kvm_apic_set_base(APICState *s, uint64_t val)
+{
+ s->apicbase = val;
+}
+
+static void kvm_apic_set_tpr(APICState *s, uint8_t val)
+{
+ s->tpr = (val & 0x0f) << 4;
+}
+
+static void do_inject_external_nmi(void *data)
+{
+ APICState *s = data;
+ CPUState *env = s->cpu_env;
+ uint32_t lvt;
+ int ret;
+
+ cpu_synchronize_state(env);
+
+ lvt = s->lvt[APIC_LVT_LINT1];
+ if (!(lvt & APIC_LVT_MASKED) && ((lvt >> 8) & 7) == APIC_DM_NMI) {
+ ret = kvm_vcpu_ioctl(env, KVM_NMI);
+ if (ret < 0) {
+ fprintf(stderr, "KVM: injection failed, NMI lost (%s)\n",
+ strerror(-ret));
+ }
+ }
+}
+
+static void kvm_apic_external_nmi(APICState *s)
+{
+ run_on_cpu(s->cpu_env, do_inject_external_nmi, s);
+}
+
+static void kvm_apic_backend_init(APICState *s)
+{
+ memory_region_init_reservation(&s->io_memory, "kvm-apic-msi",
+ MSI_SPACE_SIZE);
+}
+
+static APICBackend kvm_apic_backend = {
+ .name = "KVM",
+ .init = kvm_apic_backend_init,
+ .set_base = kvm_apic_set_base,
+ .set_tpr = kvm_apic_set_tpr,
+ .external_nmi = kvm_apic_external_nmi,
+};
+
+static void kvm_apic_register_backend(void)
+{
+ apic_register_backend(&kvm_apic_backend);
+}
+
+device_init(kvm_apic_register_backend)
diff --git a/hw/pc.c b/hw/pc.c
index 066edc4..8c8aa49 100644
--- a/hw/pc.c
+++ b/hw/pc.c
@@ -878,27 +878,32 @@ DeviceState *cpu_get_current_apic(void)
static DeviceState *apic_init(void *env, uint8_t apic_id)
{
+ const char *backend = "QEMU";
DeviceState *dev;
- SysBusDevice *d;
static int apic_mapped;
dev = qdev_create(NULL, "apic");
qdev_prop_set_uint8(dev, "id", apic_id);
qdev_prop_set_ptr(dev, "cpu_env", env);
- qdev_prop_set_string(dev, "backend", g_strdup("QEMU"));
+ if (kvm_enabled() && kvm_irqchip_in_kernel()) {
+ backend = "KVM";
+ }
+ qdev_prop_set_string(dev, "backend", g_strdup(backend));
qdev_init_nofail(dev);
- d = sysbus_from_qdev(dev);
/* XXX: mapping more APICs at the same memory location */
if (apic_mapped == 0) {
/* NOTE: the APIC is directly connected to the CPU - it is not
on the global memory bus. */
/* XXX: what if the base changes? */
- sysbus_mmio_map(d, 0, MSI_ADDR_BASE);
+ sysbus_mmio_map(sysbus_from_qdev(dev), 0, MSI_ADDR_BASE);
apic_mapped = 1;
}
- msi_supported = true;
+ /* KVM does not support MSI yet. */
+ if (!kvm_enabled() || !kvm_irqchip_in_kernel()) {
+ msi_supported = true;
+ }
return dev;
}
diff --git a/kvm.h b/kvm.h
index a3c87af..446360a 100644
--- a/kvm.h
+++ b/kvm.h
@@ -134,6 +134,9 @@ int kvm_irqchip_set_irq(KVMState *s, int irq, int level);
void kvm_irqchip_add_route(KVMState *s, int gsi, int irqchip, int pin);
int kvm_irqchip_commit_routes(KVMState *s);
+int kvm_put_apic(CPUState *env);
+int kvm_get_apic(CPUState *env);
+
struct kvm_guest_debug;
struct kvm_debug_exit_arch;
diff --git a/target-i386/kvm.c b/target-i386/kvm.c
index 9d1191f..956f436 100644
--- a/target-i386/kvm.c
+++ b/target-i386/kvm.c
@@ -1450,6 +1450,10 @@ int kvm_arch_put_registers(CPUState *env, int level)
if (ret < 0) {
return ret;
}
+ ret = kvm_put_apic(env);
+ if (ret < 0) {
+ return ret;
+ }
}
ret = kvm_put_vcpu_events(env, level);
if (ret < 0) {
@@ -1497,6 +1501,10 @@ int kvm_arch_get_registers(CPUState *env)
if (ret < 0) {
return ret;
}
+ ret = kvm_get_apic(env);
+ if (ret < 0) {
+ return ret;
+ }
ret = kvm_get_vcpu_events(env);
if (ret < 0) {
return ret;
--
1.7.3.4
^ permalink raw reply related [flat|nested] 32+ messages in thread
* [Qemu-devel] [PATCH v4 13/15] kvm: x86: Add user space part for in-kernel i8259
2011-12-08 11:52 [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support Jan Kiszka
` (11 preceding siblings ...)
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 12/15] kvm: x86: Add user space part for in-kernel APIC Jan Kiszka
@ 2011-12-08 11:52 ` Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 14/15] kvm: x86: Add user space part for in-kernel IOAPIC Jan Kiszka
` (3 subsequent siblings)
16 siblings, 0 replies; 32+ messages in thread
From: Jan Kiszka @ 2011-12-08 11:52 UTC (permalink / raw)
To: Avi Kivity, Marcelo Tosatti
Cc: Blue Swirl, Anthony Liguori, qemu-devel, kvm, Michael S. Tsirkin
Introduce the alternative i8259 backend that exploits KVM in-kernel
acceleration.
The PIIX3 initialization code is furthermore extended by KVM specific
IRQ route setup. GSI injection differs in KVM mode from the user space
model. As we can dispatch ISA-range IRQs to both IOAPIC and PIC inside
the kernel, we do not need to inject them separately. This is reflected
by a KVM-specific GSI handler.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
---
Makefile.target | 2 +-
hw/kvm/i8259.c | 126 +++++++++++++++++++++++++++++++++++++++++++++++++++++++
hw/pc.h | 1 +
hw/pc_piix.c | 50 ++++++++++++++++++++--
4 files changed, 174 insertions(+), 5 deletions(-)
create mode 100644 hw/kvm/i8259.c
diff --git a/Makefile.target b/Makefile.target
index 76de485..fb10143 100644
--- a/Makefile.target
+++ b/Makefile.target
@@ -236,7 +236,7 @@ obj-i386-y += vmport.o
obj-i386-y += device-hotplug.o pci-hotplug.o smbios.o wdt_ib700.o
obj-i386-y += debugcon.o multiboot.o
obj-i386-y += pc_piix.o
-obj-i386-$(CONFIG_KVM) += kvm/clock.o kvm/apic.o
+obj-i386-$(CONFIG_KVM) += kvm/clock.o kvm/apic.o kvm/i8259.o
obj-i386-$(CONFIG_SPICE) += qxl.o qxl-logger.o qxl-render.o
# shared objects
diff --git a/hw/kvm/i8259.c b/hw/kvm/i8259.c
new file mode 100644
index 0000000..98d7141
--- /dev/null
+++ b/hw/kvm/i8259.c
@@ -0,0 +1,126 @@
+/*
+ * KVM in-kernel PIC (i8259) support
+ *
+ * Copyright (c) 2011 Siemens AG
+ *
+ * Authors:
+ * Jan Kiszka <jan.kiszka@siemens.com>
+ *
+ * This work is licensed under the terms of the GNU GPL version 2.
+ * See the COPYING file in the top-level directory.
+ */
+#include "hw/i8259_internal.h"
+#include "hw/apic_internal.h"
+#include "kvm.h"
+
+static void kvm_pic_get(PicState *s)
+{
+ struct kvm_irqchip chip;
+ struct kvm_pic_state *kpic;
+ int ret;
+
+ chip.chip_id = s->master ? KVM_IRQCHIP_PIC_MASTER : KVM_IRQCHIP_PIC_SLAVE;
+ ret = kvm_vm_ioctl(kvm_state, KVM_GET_IRQCHIP, &chip);
+ if (ret < 0) {
+ fprintf(stderr, "KVM_GET_IRQCHIP failed: %s\n", strerror(ret));
+ abort();
+ }
+
+ kpic = &chip.chip.pic;
+
+ s->last_irr = kpic->last_irr;
+ s->irr = kpic->irr;
+ s->imr = kpic->imr;
+ s->isr = kpic->isr;
+ s->priority_add = kpic->priority_add;
+ s->irq_base = kpic->irq_base;
+ s->read_reg_select = kpic->read_reg_select;
+ s->poll = kpic->poll;
+ s->special_mask = kpic->special_mask;
+ s->init_state = kpic->init_state;
+ s->auto_eoi = kpic->auto_eoi;
+ s->rotate_on_auto_eoi = kpic->rotate_on_auto_eoi;
+ s->special_fully_nested_mode = kpic->special_fully_nested_mode;
+ s->init4 = kpic->init4;
+ s->elcr = kpic->elcr;
+ s->elcr_mask = kpic->elcr_mask;
+}
+
+static void kvm_pic_put(PicState *s)
+{
+ struct kvm_irqchip chip;
+ struct kvm_pic_state *kpic;
+ int ret;
+
+ chip.chip_id = s->master ? KVM_IRQCHIP_PIC_MASTER : KVM_IRQCHIP_PIC_SLAVE;
+
+ kpic = &chip.chip.pic;
+
+ kpic->last_irr = s->last_irr;
+ kpic->irr = s->irr;
+ kpic->imr = s->imr;
+ kpic->isr = s->isr;
+ kpic->priority_add = s->priority_add;
+ kpic->irq_base = s->irq_base;
+ kpic->read_reg_select = s->read_reg_select;
+ kpic->poll = s->poll;
+ kpic->special_mask = s->special_mask;
+ kpic->init_state = s->init_state;
+ kpic->auto_eoi = s->auto_eoi;
+ kpic->rotate_on_auto_eoi = s->rotate_on_auto_eoi;
+ kpic->special_fully_nested_mode = s->special_fully_nested_mode;
+ kpic->init4 = s->init4;
+ kpic->elcr = s->elcr;
+ kpic->elcr_mask = s->elcr_mask;
+
+ ret = kvm_vm_ioctl(kvm_state, KVM_SET_IRQCHIP, &chip);
+ if (ret < 0) {
+ fprintf(stderr, "KVM_GET_IRQCHIP failed: %s\n", strerror(ret));
+ abort();
+ }
+}
+
+static void kvm_pic_reset(PicState *s)
+{
+ pic_reset_internal(s);
+ s->elcr = 0;
+
+ kvm_pic_put(s);
+}
+
+static void kvm_pic_set_irq(void *opaque, int irq, int level)
+{
+ int delivered;
+
+ delivered = kvm_irqchip_set_irq(kvm_state, irq, level);
+ apic_set_irq_delivered(delivered);
+}
+
+static void kvm_pic_backend_init(PicState *s)
+{
+ memory_region_init_reservation(&s->base_io, "kvm-pic", 2);
+ memory_region_init_reservation(&s->elcr_io, "kvm-elcr", 1);
+}
+
+qemu_irq *kvm_i8259_init(void)
+{
+ i8259_init_chip(true, "KVM");
+ i8259_init_chip(false, "KVM");
+
+ return qemu_allocate_irqs(kvm_pic_set_irq, NULL, ISA_NUM_IRQS);
+}
+
+static PICBackend kvm_pic_backend = {
+ .name = "KVM",
+ .init = kvm_pic_backend_init,
+ .reset = kvm_pic_reset,
+ .pre_save = kvm_pic_get,
+ .post_load = kvm_pic_put,
+};
+
+static void kvm_pic_register(void)
+{
+ pic_register_backend(&kvm_pic_backend);
+}
+
+device_init(kvm_pic_register)
diff --git a/hw/pc.h b/hw/pc.h
index b7b7e40..fc6f446 100644
--- a/hw/pc.h
+++ b/hw/pc.h
@@ -64,6 +64,7 @@ bool parallel_mm_init(MemoryRegion *address_space,
typedef struct PicState PicState;
extern PicState *isa_pic;
qemu_irq *i8259_init(qemu_irq parent_irq);
+qemu_irq *kvm_i8259_init(void);
int pic_read_irq(PicState *s);
int pic_get_output(PicState *s);
void pic_info(Monitor *mon);
diff --git a/hw/pc_piix.c b/hw/pc_piix.c
index 98f2822..8650319 100644
--- a/hw/pc_piix.c
+++ b/hw/pc_piix.c
@@ -53,6 +53,40 @@ static const int ide_iobase[MAX_IDE_BUS] = { 0x1f0, 0x170 };
static const int ide_iobase2[MAX_IDE_BUS] = { 0x3f6, 0x376 };
static const int ide_irq[MAX_IDE_BUS] = { 14, 15 };
+static void kvm_piix3_setup_irq_routing(bool pci_enabled)
+{
+ KVMState *s = kvm_state;
+ int ret, i;
+
+ if (kvm_check_extension(s, KVM_CAP_IRQ_ROUTING)) {
+ for (i = 0; i < 8; ++i) {
+ if (i == 2) {
+ continue;
+ }
+ kvm_irqchip_add_route(s, i, KVM_IRQCHIP_PIC_MASTER, i);
+ }
+ for (i = 8; i < 16; ++i) {
+ kvm_irqchip_add_route(s, i, KVM_IRQCHIP_PIC_SLAVE, i - 8);
+ }
+ ret = kvm_irqchip_commit_routes(s);
+ if (ret < 0) {
+ hw_error("KVM IRQ routing setup failed");
+ }
+ }
+}
+
+static void kvm_piix3_gsi_handler(void *opaque, int n, int level)
+{
+ GSIState *s = opaque;
+
+ if (n < ISA_NUM_IRQS) {
+ /* Kernel will forward to both PIC and IOAPIC */
+ qemu_set_irq(s->i8259_irq[n], level);
+ } else {
+ qemu_set_irq(s->ioapic_irq[n], level);
+ }
+}
+
static void ioapic_init(GSIState *gsi_state)
{
DeviceState *dev;
@@ -133,7 +167,13 @@ static void pc_init1(MemoryRegion *system_memory,
}
gsi_state = g_malloc0(sizeof(*gsi_state));
- gsi = qemu_allocate_irqs(gsi_handler, gsi_state, GSI_NUM_PINS);
+ if (kvm_enabled() && kvm_irqchip_in_kernel()) {
+ kvm_piix3_setup_irq_routing(pci_enabled);
+ gsi = qemu_allocate_irqs(kvm_piix3_gsi_handler, gsi_state,
+ GSI_NUM_PINS);
+ } else {
+ gsi = qemu_allocate_irqs(gsi_handler, gsi_state, GSI_NUM_PINS);
+ }
if (pci_enabled) {
pci_bus = i440fx_init(&i440fx_state, &piix3_devfn, gsi,
@@ -153,11 +193,13 @@ static void pc_init1(MemoryRegion *system_memory,
}
isa_bus_irqs(gsi);
- if (!xen_enabled()) {
+ if (kvm_enabled() && kvm_irqchip_in_kernel()) {
+ i8259 = kvm_i8259_init();
+ } else if (xen_enabled()) {
+ i8259 = xen_interrupt_controller_init();
+ } else {
cpu_irq = pc_allocate_cpu_irq();
i8259 = i8259_init(cpu_irq[0]);
- } else {
- i8259 = xen_interrupt_controller_init();
}
for (i = 0; i < ISA_NUM_IRQS; i++) {
--
1.7.3.4
^ permalink raw reply related [flat|nested] 32+ messages in thread
* [Qemu-devel] [PATCH v4 14/15] kvm: x86: Add user space part for in-kernel IOAPIC
2011-12-08 11:52 [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support Jan Kiszka
` (12 preceding siblings ...)
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 13/15] kvm: x86: Add user space part for in-kernel i8259 Jan Kiszka
@ 2011-12-08 11:52 ` Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 15/15] kvm: Arm in-kernel irqchip support Jan Kiszka
` (2 subsequent siblings)
16 siblings, 0 replies; 32+ messages in thread
From: Jan Kiszka @ 2011-12-08 11:52 UTC (permalink / raw)
To: Avi Kivity, Marcelo Tosatti
Cc: Blue Swirl, Anthony Liguori, qemu-devel, kvm, Michael S. Tsirkin
This introduces the KVM-accelerated IOAPIC backend and extends the IRQ
routing setup by the 0->2 redirection when needed.
The IOAPIC gains a KVM-specific property that allows to define the GSI
base for injecting interrupts into the kernel model. This will allow to
disentangle PIC and IOAPIC pins for chipsets that support more
sophisticated IRQ routes than the PIIX3. So far the base is kept at 0,
i.e. PIC and IOAPIC share pins 0..15.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
---
Makefile.target | 2 +-
hw/ioapic_common.c | 1 +
hw/ioapic_internal.h | 1 +
hw/kvm/ioapic.c | 101 ++++++++++++++++++++++++++++++++++++++++++++++++++
hw/pc_piix.c | 15 +++++++-
5 files changed, 118 insertions(+), 2 deletions(-)
create mode 100644 hw/kvm/ioapic.c
diff --git a/Makefile.target b/Makefile.target
index fb10143..b48bb57 100644
--- a/Makefile.target
+++ b/Makefile.target
@@ -236,7 +236,7 @@ obj-i386-y += vmport.o
obj-i386-y += device-hotplug.o pci-hotplug.o smbios.o wdt_ib700.o
obj-i386-y += debugcon.o multiboot.o
obj-i386-y += pc_piix.o
-obj-i386-$(CONFIG_KVM) += kvm/clock.o kvm/apic.o kvm/i8259.o
+obj-i386-$(CONFIG_KVM) += kvm/clock.o kvm/apic.o kvm/i8259.o kvm/ioapic.o
obj-i386-$(CONFIG_SPICE) += qxl.o qxl-logger.o qxl-render.o
# shared objects
diff --git a/hw/ioapic_common.c b/hw/ioapic_common.c
index 094551c..efc1d44 100644
--- a/hw/ioapic_common.c
+++ b/hw/ioapic_common.c
@@ -122,6 +122,7 @@ static SysBusDeviceInfo ioapic_info = {
.qdev.no_user = 1,
.qdev.props = (Property[]) {
DEFINE_PROP_STRING("backend", IOAPICState, backend_name),
+ DEFINE_PROP_UINT32("kvm_gsi_base", IOAPICState, kvm_gsi_base, 0),
DEFINE_PROP_END_OF_LIST(),
},
};
diff --git a/hw/ioapic_internal.h b/hw/ioapic_internal.h
index c5fab8b..bf63115 100644
--- a/hw/ioapic_internal.h
+++ b/hw/ioapic_internal.h
@@ -95,6 +95,7 @@ struct IOAPICState {
char *backend_name;
IOAPICBackend *backend;
+ uint32_t kvm_gsi_base;
};
void ioapic_register_device(void);
diff --git a/hw/kvm/ioapic.c b/hw/kvm/ioapic.c
new file mode 100644
index 0000000..0e66240
--- /dev/null
+++ b/hw/kvm/ioapic.c
@@ -0,0 +1,101 @@
+/*
+ * KVM in-kernel IOPIC support
+ *
+ * Copyright (c) 2011 Siemens AG
+ *
+ * Authors:
+ * Jan Kiszka <jan.kiszka@siemens.com>
+ *
+ * This work is licensed under the terms of the GNU GPL version 2.
+ * See the COPYING file in the top-level directory.
+ */
+
+#include "hw/pc.h"
+#include "hw/ioapic_internal.h"
+#include "hw/apic_internal.h"
+#include "kvm.h"
+
+static void kvm_ioapic_get(IOAPICState *s)
+{
+ struct kvm_irqchip chip;
+ struct kvm_ioapic_state *kioapic;
+ int ret, i;
+
+ chip.chip_id = KVM_IRQCHIP_IOAPIC;
+ ret = kvm_vm_ioctl(kvm_state, KVM_GET_IRQCHIP, &chip);
+ if (ret < 0) {
+ fprintf(stderr, "KVM_GET_IRQCHIP failed: %s\n", strerror(ret));
+ abort();
+ }
+
+ kioapic = &chip.chip.ioapic;
+
+ s->id = kioapic->id;
+ s->ioregsel = kioapic->ioregsel;
+ s->irr = kioapic->irr;
+ for (i = 0; i < IOAPIC_NUM_PINS; i++) {
+ s->ioredtbl[i] = kioapic->redirtbl[i].bits;
+ }
+}
+
+static void kvm_ioapic_put(IOAPICState *s)
+{
+ struct kvm_irqchip chip;
+ struct kvm_ioapic_state *kioapic;
+ int ret, i;
+
+ chip.chip_id = KVM_IRQCHIP_IOAPIC;
+ kioapic = &chip.chip.ioapic;
+
+ kioapic->id = s->id;
+ kioapic->ioregsel = s->ioregsel;
+ kioapic->base_address = s->busdev.mmio[0].addr;
+ kioapic->irr = s->irr;
+ for (i = 0; i < IOAPIC_NUM_PINS; i++) {
+ kioapic->redirtbl[i].bits = s->ioredtbl[i];
+ }
+
+ ret = kvm_vm_ioctl(kvm_state, KVM_SET_IRQCHIP, &chip);
+ if (ret < 0) {
+ fprintf(stderr, "KVM_GET_IRQCHIP failed: %s\n", strerror(ret));
+ abort();
+ }
+}
+
+static void kvm_ioapic_reset(IOAPICState *s)
+{
+ ioapic_reset_internal(s);
+
+ kvm_ioapic_put(s);
+}
+
+static void kvm_ioapic_set_irq(void *opaque, int irq, int level)
+{
+ IOAPICState *s = opaque;
+ int delivered;
+
+ delivered = kvm_irqchip_set_irq(kvm_state, s->kvm_gsi_base + irq, level);
+ apic_set_irq_delivered(delivered);
+}
+
+static void kvm_ioapic_backend_init(IOAPICState *s, int index)
+{
+ memory_region_init_reservation(&s->io_memory, "kvm-ioapic", 0x1000);
+
+ qdev_init_gpio_in(&s->busdev.qdev, kvm_ioapic_set_irq, IOAPIC_NUM_PINS);
+}
+
+static IOAPICBackend kvm_ioapic_backend = {
+ .name = "KVM",
+ .init = kvm_ioapic_backend_init,
+ .reset = kvm_ioapic_reset,
+ .pre_save = kvm_ioapic_get,
+ .post_load = kvm_ioapic_put,
+};
+
+static void kvm_ioapic_register(void)
+{
+ ioapic_register_backend(&kvm_ioapic_backend);
+}
+
+device_init(kvm_ioapic_register)
diff --git a/hw/pc_piix.c b/hw/pc_piix.c
index 8650319..93d0eba 100644
--- a/hw/pc_piix.c
+++ b/hw/pc_piix.c
@@ -68,6 +68,15 @@ static void kvm_piix3_setup_irq_routing(bool pci_enabled)
for (i = 8; i < 16; ++i) {
kvm_irqchip_add_route(s, i, KVM_IRQCHIP_PIC_SLAVE, i - 8);
}
+ if (pci_enabled) {
+ for (i = 0; i < 24; ++i) {
+ if (i == 0) {
+ kvm_irqchip_add_route(s, i, KVM_IRQCHIP_IOAPIC, 2);
+ } else if (i != 2) {
+ kvm_irqchip_add_route(s, i, KVM_IRQCHIP_IOAPIC, i);
+ }
+ }
+ }
ret = kvm_irqchip_commit_routes(s);
if (ret < 0) {
hw_error("KVM IRQ routing setup failed");
@@ -89,12 +98,16 @@ static void kvm_piix3_gsi_handler(void *opaque, int n, int level)
static void ioapic_init(GSIState *gsi_state)
{
+ const char *backend = "QEMU";
DeviceState *dev;
SysBusDevice *d;
unsigned int i;
dev = qdev_create(NULL, "ioapic");
- qdev_prop_set_string(dev, "backend", g_strdup("QEMU"));
+ if (kvm_enabled() && kvm_irqchip_in_kernel()) {
+ backend = "KVM";
+ }
+ qdev_prop_set_string(dev, "backend", g_strdup(backend));
qdev_init_nofail(dev);
d = sysbus_from_qdev(dev);
sysbus_mmio_map(d, 0, 0xfec00000);
--
1.7.3.4
^ permalink raw reply related [flat|nested] 32+ messages in thread
* [Qemu-devel] [PATCH v4 15/15] kvm: Arm in-kernel irqchip support
2011-12-08 11:52 [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support Jan Kiszka
` (13 preceding siblings ...)
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 14/15] kvm: x86: Add user space part for in-kernel IOAPIC Jan Kiszka
@ 2011-12-08 11:52 ` Jan Kiszka
2011-12-08 21:25 ` [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic " Blue Swirl
2011-12-12 16:37 ` Marcelo Tosatti
16 siblings, 0 replies; 32+ messages in thread
From: Jan Kiszka @ 2011-12-08 11:52 UTC (permalink / raw)
To: Avi Kivity, Marcelo Tosatti
Cc: Blue Swirl, Anthony Liguori, qemu-devel, kvm, Michael S. Tsirkin
Make the basic in-kernel irqchip support selectable via
-machine ...,kernel_irqchip=on. Leave it off by default until it can
fully replace user space models.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
---
qemu-config.c | 4 ++++
qemu-options.hx | 5 ++++-
2 files changed, 8 insertions(+), 1 deletions(-)
diff --git a/qemu-config.c b/qemu-config.c
index 597d7e1..a761bea 100644
--- a/qemu-config.c
+++ b/qemu-config.c
@@ -490,6 +490,10 @@ static QemuOptsList qemu_machine_opts = {
.name = "accel",
.type = QEMU_OPT_STRING,
.help = "accelerator list",
+ }, {
+ .name = "kernel_irqchip",
+ .type = QEMU_OPT_BOOL,
+ .help = "use KVM in-kernel irqchip",
},
{ /* End of list */ }
},
diff --git a/qemu-options.hx b/qemu-options.hx
index 681eaf1..60b7dc0 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -31,7 +31,8 @@ DEF("machine", HAS_ARG, QEMU_OPTION_machine, \
"-machine [type=]name[,prop[=value][,...]]\n"
" selects emulated machine (-machine ? for list)\n"
" property accel=accel1[:accel2[:...]] selects accelerator\n"
- " supported accelerators are kvm, xen, tcg (default: tcg)\n",
+ " supported accelerators are kvm, xen, tcg (default: tcg)\n"
+ " kernel_irqchip=on|off controls accelerated irqchip support\n",
QEMU_ARCH_ALL)
STEXI
@item -machine [type=]@var{name}[,prop=@var{value}[,...]]
@@ -44,6 +45,8 @@ This is used to enable an accelerator. Depending on the target architecture,
kvm, xen, or tcg can be available. By default, tcg is used. If there is more
than one accelerator specified, the next one is used if the previous one fails
to initialize.
+@item kernel_irqchip=on|off
+Enables in-kernel irqchip support for the chosen accelerator when available.
@end table
ETEXI
--
1.7.3.4
^ permalink raw reply related [flat|nested] 32+ messages in thread
* Re: [Qemu-devel] [PATCH v4 12/15] kvm: x86: Add user space part for in-kernel APIC
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 12/15] kvm: x86: Add user space part for in-kernel APIC Jan Kiszka
@ 2011-12-08 21:16 ` Blue Swirl
2011-12-09 7:45 ` Jan Kiszka
0 siblings, 1 reply; 32+ messages in thread
From: Blue Swirl @ 2011-12-08 21:16 UTC (permalink / raw)
To: Jan Kiszka
Cc: Anthony Liguori, Lai Jiangshan, kvm, Michael S. Tsirkin,
Marcelo Tosatti, qemu-devel, Avi Kivity
On Thu, Dec 8, 2011 at 11:52, Jan Kiszka <jan.kiszka@siemens.com> wrote:
> This introduces the alternative APIC backend which makes use of KVM's
> in-kernel device model. External NMI injection via LINT1 is emulated by
> checking the current state of the in-kernel APIC, only injecting a NMI
> into the VCPU if LINT1 is unmasked and configured to DM_NMI.
>
> MSI is not yet supported, so we disable this when the in-kernel model is
> in use.
>
> CC: Lai Jiangshan <laijs@cn.fujitsu.com>
> Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
> ---
> Makefile.target | 2 +-
> hw/kvm/apic.c | 154 +++++++++++++++++++++++++++++++++++++++++++++++++++++
> hw/pc.c | 15 ++++--
> kvm.h | 3 +
> target-i386/kvm.c | 8 +++
> 5 files changed, 176 insertions(+), 6 deletions(-)
> create mode 100644 hw/kvm/apic.c
>
> diff --git a/Makefile.target b/Makefile.target
> index b549988..76de485 100644
> --- a/Makefile.target
> +++ b/Makefile.target
> @@ -236,7 +236,7 @@ obj-i386-y += vmport.o
> obj-i386-y += device-hotplug.o pci-hotplug.o smbios.o wdt_ib700.o
> obj-i386-y += debugcon.o multiboot.o
> obj-i386-y += pc_piix.o
> -obj-i386-$(CONFIG_KVM) += kvm/clock.o
> +obj-i386-$(CONFIG_KVM) += kvm/clock.o kvm/apic.o
> obj-i386-$(CONFIG_SPICE) += qxl.o qxl-logger.o qxl-render.o
>
> # shared objects
> diff --git a/hw/kvm/apic.c b/hw/kvm/apic.c
> new file mode 100644
> index 0000000..3924f9e
> --- /dev/null
> +++ b/hw/kvm/apic.c
> @@ -0,0 +1,154 @@
> +/*
> + * KVM in-kernel APIC support
> + *
> + * Copyright (c) 2011 Siemens AG
> + *
> + * Authors:
> + * Jan Kiszka <jan.kiszka@siemens.com>
> + *
> + * This work is licensed under the terms of the GNU GPL version 2.
> + * See the COPYING file in the top-level directory.
> + */
> +#include "hw/apic_internal.h"
> +#include "kvm.h"
> +
> +static inline void kvm_apic_set_reg(struct kvm_lapic_state *kapic,
> + int reg_id, uint32_t val)
> +{
> + *((uint32_t *)(kapic->regs + (reg_id << 4))) = val;
> +}
> +
> +static inline uint32_t kvm_apic_get_reg(struct kvm_lapic_state *kapic,
> + int reg_id)
> +{
> + return *((uint32_t *)(kapic->regs + (reg_id << 4)));
> +}
> +
> +int kvm_put_apic(CPUState *env)
> +{
> + APICState *s = DO_UPCAST(APICState, busdev.qdev, env->apic_state);
Please pass APICState instead of CPUState.
> + struct kvm_lapic_state kapic;
> + int i;
> +
> + if (s && kvm_enabled() && kvm_irqchip_in_kernel()) {
> + memset(&kapic, 0, sizeof(kapic));
> + kvm_apic_set_reg(&kapic, 0x2, s->id << 24);
> + kvm_apic_set_reg(&kapic, 0x8, s->tpr);
> + kvm_apic_set_reg(&kapic, 0xd, s->log_dest << 24);
> + kvm_apic_set_reg(&kapic, 0xe, s->dest_mode << 28 | 0x0fffffff);
> + kvm_apic_set_reg(&kapic, 0xf, s->spurious_vec);
> + for (i = 0; i < 8; i++) {
> + kvm_apic_set_reg(&kapic, 0x10 + i, s->isr[i]);
> + kvm_apic_set_reg(&kapic, 0x18 + i, s->tmr[i]);
> + kvm_apic_set_reg(&kapic, 0x20 + i, s->irr[i]);
> + }
> + kvm_apic_set_reg(&kapic, 0x28, s->esr);
> + kvm_apic_set_reg(&kapic, 0x30, s->icr[0]);
> + kvm_apic_set_reg(&kapic, 0x31, s->icr[1]);
> + for (i = 0; i < APIC_LVT_NB; i++) {
> + kvm_apic_set_reg(&kapic, 0x32 + i, s->lvt[i]);
> + }
> + kvm_apic_set_reg(&kapic, 0x38, s->initial_count);
> + kvm_apic_set_reg(&kapic, 0x3e, s->divide_conf);
> +
> + return kvm_vcpu_ioctl(env, KVM_SET_LAPIC, &kapic);
> + }
> +
> + return 0;
> +}
> +
> +int kvm_get_apic(CPUState *env)
Same here.
> +{
> + APICState *s = DO_UPCAST(APICState, busdev.qdev, env->apic_state);
> + struct kvm_lapic_state kapic;
> + int ret, i, v;
> +
> + if (s && kvm_enabled() && kvm_irqchip_in_kernel()) {
> + ret = kvm_vcpu_ioctl(env, KVM_GET_LAPIC, &kapic);
> + if (ret < 0) {
> + return ret;
> + }
> +
> + s->id = kvm_apic_get_reg(&kapic, 0x2) >> 24;
> + s->tpr = kvm_apic_get_reg(&kapic, 0x8);
> + s->arb_id = kvm_apic_get_reg(&kapic, 0x9);
> + s->log_dest = kvm_apic_get_reg(&kapic, 0xd) >> 24;
> + s->dest_mode = kvm_apic_get_reg(&kapic, 0xe) >> 28;
> + s->spurious_vec = kvm_apic_get_reg(&kapic, 0xf);
> + for (i = 0; i < 8; i++) {
> + s->isr[i] = kvm_apic_get_reg(&kapic, 0x10 + i);
> + s->tmr[i] = kvm_apic_get_reg(&kapic, 0x18 + i);
> + s->irr[i] = kvm_apic_get_reg(&kapic, 0x20 + i);
> + }
> + s->esr = kvm_apic_get_reg(&kapic, 0x28);
> + s->icr[0] = kvm_apic_get_reg(&kapic, 0x30);
> + s->icr[1] = kvm_apic_get_reg(&kapic, 0x31);
> + for (i = 0; i < APIC_LVT_NB; i++) {
> + s->lvt[i] = kvm_apic_get_reg(&kapic, 0x32 + i);
> + }
> + s->initial_count = kvm_apic_get_reg(&kapic, 0x38);
> + s->divide_conf = kvm_apic_get_reg(&kapic, 0x3e);
> +
> + v = (s->divide_conf & 3) | ((s->divide_conf >> 1) & 4);
> + s->count_shift = (v + 1) & 7;
> +
> + s->initial_count_load_time = qemu_get_clock_ns(vm_clock);
> + apic_next_timer(s, s->initial_count_load_time);
> + }
> + return 0;
> +}
> +
> +static void kvm_apic_set_base(APICState *s, uint64_t val)
> +{
> + s->apicbase = val;
> +}
> +
> +static void kvm_apic_set_tpr(APICState *s, uint8_t val)
> +{
> + s->tpr = (val & 0x0f) << 4;
> +}
> +
> +static void do_inject_external_nmi(void *data)
> +{
> + APICState *s = data;
> + CPUState *env = s->cpu_env;
> + uint32_t lvt;
> + int ret;
> +
> + cpu_synchronize_state(env);
> +
> + lvt = s->lvt[APIC_LVT_LINT1];
> + if (!(lvt & APIC_LVT_MASKED) && ((lvt >> 8) & 7) == APIC_DM_NMI) {
> + ret = kvm_vcpu_ioctl(env, KVM_NMI);
> + if (ret < 0) {
> + fprintf(stderr, "KVM: injection failed, NMI lost (%s)\n",
> + strerror(-ret));
> + }
> + }
> +}
> +
> +static void kvm_apic_external_nmi(APICState *s)
> +{
> + run_on_cpu(s->cpu_env, do_inject_external_nmi, s);
Here probably CPUState would make more sense.
> +}
> +
> +static void kvm_apic_backend_init(APICState *s)
> +{
> + memory_region_init_reservation(&s->io_memory, "kvm-apic-msi",
> + MSI_SPACE_SIZE);
> +}
> +
> +static APICBackend kvm_apic_backend = {
> + .name = "KVM",
> + .init = kvm_apic_backend_init,
> + .set_base = kvm_apic_set_base,
> + .set_tpr = kvm_apic_set_tpr,
> + .external_nmi = kvm_apic_external_nmi,
> +};
> +
> +static void kvm_apic_register_backend(void)
> +{
> + apic_register_backend(&kvm_apic_backend);
> +}
> +
> +device_init(kvm_apic_register_backend)
> diff --git a/hw/pc.c b/hw/pc.c
> index 066edc4..8c8aa49 100644
> --- a/hw/pc.c
> +++ b/hw/pc.c
> @@ -878,27 +878,32 @@ DeviceState *cpu_get_current_apic(void)
>
> static DeviceState *apic_init(void *env, uint8_t apic_id)
> {
> + const char *backend = "QEMU";
> DeviceState *dev;
> - SysBusDevice *d;
> static int apic_mapped;
>
> dev = qdev_create(NULL, "apic");
> qdev_prop_set_uint8(dev, "id", apic_id);
> qdev_prop_set_ptr(dev, "cpu_env", env);
> - qdev_prop_set_string(dev, "backend", g_strdup("QEMU"));
> + if (kvm_enabled() && kvm_irqchip_in_kernel()) {
> + backend = "KVM";
> + }
> + qdev_prop_set_string(dev, "backend", g_strdup(backend));
> qdev_init_nofail(dev);
> - d = sysbus_from_qdev(dev);
>
> /* XXX: mapping more APICs at the same memory location */
> if (apic_mapped == 0) {
> /* NOTE: the APIC is directly connected to the CPU - it is not
> on the global memory bus. */
> /* XXX: what if the base changes? */
> - sysbus_mmio_map(d, 0, MSI_ADDR_BASE);
> + sysbus_mmio_map(sysbus_from_qdev(dev), 0, MSI_ADDR_BASE);
> apic_mapped = 1;
> }
>
> - msi_supported = true;
> + /* KVM does not support MSI yet. */
> + if (!kvm_enabled() || !kvm_irqchip_in_kernel()) {
> + msi_supported = true;
> + }
>
> return dev;
> }
> diff --git a/kvm.h b/kvm.h
> index a3c87af..446360a 100644
> --- a/kvm.h
> +++ b/kvm.h
> @@ -134,6 +134,9 @@ int kvm_irqchip_set_irq(KVMState *s, int irq, int level);
> void kvm_irqchip_add_route(KVMState *s, int gsi, int irqchip, int pin);
> int kvm_irqchip_commit_routes(KVMState *s);
>
> +int kvm_put_apic(CPUState *env);
> +int kvm_get_apic(CPUState *env);
> +
> struct kvm_guest_debug;
> struct kvm_debug_exit_arch;
>
> diff --git a/target-i386/kvm.c b/target-i386/kvm.c
> index 9d1191f..956f436 100644
> --- a/target-i386/kvm.c
> +++ b/target-i386/kvm.c
> @@ -1450,6 +1450,10 @@ int kvm_arch_put_registers(CPUState *env, int level)
> if (ret < 0) {
> return ret;
> }
> + ret = kvm_put_apic(env);
> + if (ret < 0) {
> + return ret;
> + }
> }
> ret = kvm_put_vcpu_events(env, level);
> if (ret < 0) {
> @@ -1497,6 +1501,10 @@ int kvm_arch_get_registers(CPUState *env)
> if (ret < 0) {
> return ret;
> }
> + ret = kvm_get_apic(env);
> + if (ret < 0) {
> + return ret;
> + }
> ret = kvm_get_vcpu_events(env);
> if (ret < 0) {
> return ret;
> --
> 1.7.3.4
>
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support
2011-12-08 11:52 [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support Jan Kiszka
` (14 preceding siblings ...)
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 15/15] kvm: Arm in-kernel irqchip support Jan Kiszka
@ 2011-12-08 21:25 ` Blue Swirl
2011-12-09 7:40 ` Jan Kiszka
2011-12-12 16:37 ` Marcelo Tosatti
16 siblings, 1 reply; 32+ messages in thread
From: Blue Swirl @ 2011-12-08 21:25 UTC (permalink / raw)
To: Jan Kiszka
Cc: Anthony Liguori, Lai Jiangshan, kvm, Michael S. Tsirkin,
Marcelo Tosatti, qemu-devel, Avi Kivity
On Thu, Dec 8, 2011 at 11:52, Jan Kiszka <jan.kiszka@siemens.com> wrote:
> Changes in v4:
> - rebased of current uq/master
> - fixed stupid bugs that broke bisectability and user space irqchip mode
> - integrated NMI-over-LINT1 injection logic
I had comments to one patch, others look fine.
Overall, string based subtype selection does not somehow seem to be a
hot idea, but this could be used as a starting point which should be
cleaned up later when we have proper device composition. APIC and x86
interrupt handling need more cleanup anyway.
> CC: Lai Jiangshan <laijs@cn.fujitsu.com>
>
> Jan Kiszka (15):
> msi: Generalize msix_supported to msi_supported
> kvm: Move kvmclock into hw/kvm folder
> apic: Stop timer on reset
> apic: Inject external NMI events via LINT1
> apic: Introduce backend/frontend infrastructure for KVM reuse
> apic: Open-code timer save/restore
> i8259: Introduce backend/frontend infrastructure for KVM reuse
> ioapic: Introduce backend/frontend infrastructure for KVM reuse
> memory: Introduce memory_region_init_reservation
> kvm: Introduce core services for in-kernel irqchip support
> kvm: x86: Establish IRQ0 override control
> kvm: x86: Add user space part for in-kernel APIC
> kvm: x86: Add user space part for in-kernel i8259
> kvm: x86: Add user space part for in-kernel IOAPIC
> kvm: Arm in-kernel irqchip support
>
> Makefile.objs | 2 +-
> Makefile.target | 6 +-
> configure | 1 +
> hw/apic.c | 309 ++++-----------------------------------
> hw/apic.h | 1 +
> hw/apic_common.c | 312 ++++++++++++++++++++++++++++++++++++++++
> hw/apic_internal.h | 122 ++++++++++++++++
> hw/i8259.c | 127 ++--------------
> hw/i8259_common.c | 173 ++++++++++++++++++++++
> hw/i8259_internal.h | 82 +++++++++++
> hw/ioapic.c | 130 ++---------------
> hw/ioapic_common.c | 138 ++++++++++++++++++
> hw/ioapic_internal.h | 106 ++++++++++++++
> hw/kvm/apic.c | 154 ++++++++++++++++++++
> hw/{kvmclock.c => kvm/clock.c} | 4 +-
> hw/{kvmclock.h => kvm/clock.h} | 0
> hw/kvm/i8259.c | 126 ++++++++++++++++
> hw/kvm/ioapic.c | 101 +++++++++++++
> hw/msi.c | 8 +
> hw/msi.h | 2 +
> hw/msix.c | 9 +-
> hw/msix.h | 2 -
> hw/pc.c | 19 ++-
> hw/pc.h | 1 +
> hw/pc_piix.c | 66 ++++++++-
> kvm-all.c | 154 ++++++++++++++++++++
> kvm-stub.c | 5 +
> kvm.h | 13 ++
> memory.c | 36 +++++
> memory.h | 16 ++
> monitor.c | 6 +-
> qemu-config.c | 4 +
> qemu-options.hx | 5 +-
> sysemu.h | 1 -
> target-i386/kvm.c | 19 +++
> trace-events | 2 +-
> vl.c | 1 -
> 37 files changed, 1724 insertions(+), 539 deletions(-)
> create mode 100644 hw/apic_common.c
> create mode 100644 hw/apic_internal.h
> create mode 100644 hw/i8259_common.c
> create mode 100644 hw/i8259_internal.h
> create mode 100644 hw/ioapic_common.c
> create mode 100644 hw/ioapic_internal.h
> create mode 100644 hw/kvm/apic.c
> rename hw/{kvmclock.c => kvm/clock.c} (98%)
> rename hw/{kvmclock.h => kvm/clock.h} (100%)
> create mode 100644 hw/kvm/i8259.c
> create mode 100644 hw/kvm/ioapic.c
>
> --
> 1.7.3.4
>
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support
2011-12-08 21:25 ` [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic " Blue Swirl
@ 2011-12-09 7:40 ` Jan Kiszka
0 siblings, 0 replies; 32+ messages in thread
From: Jan Kiszka @ 2011-12-09 7:40 UTC (permalink / raw)
To: Blue Swirl
Cc: Anthony Liguori, Lai Jiangshan, kvm, Michael S. Tsirkin,
Marcelo Tosatti, qemu-devel, Avi Kivity
On 2011-12-08 22:25, Blue Swirl wrote:
> On Thu, Dec 8, 2011 at 11:52, Jan Kiszka <jan.kiszka@siemens.com> wrote:
>> Changes in v4:
>> - rebased of current uq/master
>> - fixed stupid bugs that broke bisectability and user space irqchip mode
>> - integrated NMI-over-LINT1 injection logic
>
> I had comments to one patch, others look fine.
>
> Overall, string based subtype selection does not somehow seem to be a
> hot idea, but this could be used as a starting point which should be
> cleaned up later when we have proper device composition. APIC and x86
> interrupt handling need more cleanup anyway.
Yes, more than likely. However, I think the string-based selection is
the best that can be done right now. It gives us, e.g., pretty-printing
of the backend property so that you can easily check what mode is in use.
The next area of ugliness will be the PIT and its relation to the HPET.
I started looking into this, found some bugs, but do not really have a
nice model for all that yet.
Looking into the remaining remarks now.
Thanks,
Jan
--
Siemens AG, Corporate Technology, CT T DE IT 1
Corporate Competence Center Embedded Linux
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [Qemu-devel] [PATCH v4 12/15] kvm: x86: Add user space part for in-kernel APIC
2011-12-08 21:16 ` Blue Swirl
@ 2011-12-09 7:45 ` Jan Kiszka
2011-12-09 7:52 ` Jan Kiszka
0 siblings, 1 reply; 32+ messages in thread
From: Jan Kiszka @ 2011-12-09 7:45 UTC (permalink / raw)
To: Blue Swirl
Cc: Anthony Liguori, Lai Jiangshan, kvm@vger.kernel.org,
Michael S. Tsirkin, Marcelo Tosatti, qemu-devel, Avi Kivity
On 2011-12-08 22:16, Blue Swirl wrote:
> On Thu, Dec 8, 2011 at 11:52, Jan Kiszka <jan.kiszka@siemens.com> wrote:
>> This introduces the alternative APIC backend which makes use of KVM's
>> in-kernel device model. External NMI injection via LINT1 is emulated by
>> checking the current state of the in-kernel APIC, only injecting a NMI
>> into the VCPU if LINT1 is unmasked and configured to DM_NMI.
>>
>> MSI is not yet supported, so we disable this when the in-kernel model is
>> in use.
>>
>> CC: Lai Jiangshan <laijs@cn.fujitsu.com>
>> Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
>> ---
>> Makefile.target | 2 +-
>> hw/kvm/apic.c | 154 +++++++++++++++++++++++++++++++++++++++++++++++++++++
>> hw/pc.c | 15 ++++--
>> kvm.h | 3 +
>> target-i386/kvm.c | 8 +++
>> 5 files changed, 176 insertions(+), 6 deletions(-)
>> create mode 100644 hw/kvm/apic.c
>>
>> diff --git a/Makefile.target b/Makefile.target
>> index b549988..76de485 100644
>> --- a/Makefile.target
>> +++ b/Makefile.target
>> @@ -236,7 +236,7 @@ obj-i386-y += vmport.o
>> obj-i386-y += device-hotplug.o pci-hotplug.o smbios.o wdt_ib700.o
>> obj-i386-y += debugcon.o multiboot.o
>> obj-i386-y += pc_piix.o
>> -obj-i386-$(CONFIG_KVM) += kvm/clock.o
>> +obj-i386-$(CONFIG_KVM) += kvm/clock.o kvm/apic.o
>> obj-i386-$(CONFIG_SPICE) += qxl.o qxl-logger.o qxl-render.o
>>
>> # shared objects
>> diff --git a/hw/kvm/apic.c b/hw/kvm/apic.c
>> new file mode 100644
>> index 0000000..3924f9e
>> --- /dev/null
>> +++ b/hw/kvm/apic.c
>> @@ -0,0 +1,154 @@
>> +/*
>> + * KVM in-kernel APIC support
>> + *
>> + * Copyright (c) 2011 Siemens AG
>> + *
>> + * Authors:
>> + * Jan Kiszka <jan.kiszka@siemens.com>
>> + *
>> + * This work is licensed under the terms of the GNU GPL version 2.
>> + * See the COPYING file in the top-level directory.
>> + */
>> +#include "hw/apic_internal.h"
>> +#include "kvm.h"
>> +
>> +static inline void kvm_apic_set_reg(struct kvm_lapic_state *kapic,
>> + int reg_id, uint32_t val)
>> +{
>> + *((uint32_t *)(kapic->regs + (reg_id << 4))) = val;
>> +}
>> +
>> +static inline uint32_t kvm_apic_get_reg(struct kvm_lapic_state *kapic,
>> + int reg_id)
>> +{
>> + return *((uint32_t *)(kapic->regs + (reg_id << 4)));
>> +}
>> +
>> +int kvm_put_apic(CPUState *env)
>> +{
>> + APICState *s = DO_UPCAST(APICState, busdev.qdev, env->apic_state);
>
> Please pass APICState instead of CPUState.
DeviceState, I suppose. Yes, makes more sense, update will follow.
>
>> + struct kvm_lapic_state kapic;
>> + int i;
>> +
>> + if (s && kvm_enabled() && kvm_irqchip_in_kernel()) {
>> + memset(&kapic, 0, sizeof(kapic));
>> + kvm_apic_set_reg(&kapic, 0x2, s->id << 24);
>> + kvm_apic_set_reg(&kapic, 0x8, s->tpr);
>> + kvm_apic_set_reg(&kapic, 0xd, s->log_dest << 24);
>> + kvm_apic_set_reg(&kapic, 0xe, s->dest_mode << 28 | 0x0fffffff);
>> + kvm_apic_set_reg(&kapic, 0xf, s->spurious_vec);
>> + for (i = 0; i < 8; i++) {
>> + kvm_apic_set_reg(&kapic, 0x10 + i, s->isr[i]);
>> + kvm_apic_set_reg(&kapic, 0x18 + i, s->tmr[i]);
>> + kvm_apic_set_reg(&kapic, 0x20 + i, s->irr[i]);
>> + }
>> + kvm_apic_set_reg(&kapic, 0x28, s->esr);
>> + kvm_apic_set_reg(&kapic, 0x30, s->icr[0]);
>> + kvm_apic_set_reg(&kapic, 0x31, s->icr[1]);
>> + for (i = 0; i < APIC_LVT_NB; i++) {
>> + kvm_apic_set_reg(&kapic, 0x32 + i, s->lvt[i]);
>> + }
>> + kvm_apic_set_reg(&kapic, 0x38, s->initial_count);
>> + kvm_apic_set_reg(&kapic, 0x3e, s->divide_conf);
>> +
>> + return kvm_vcpu_ioctl(env, KVM_SET_LAPIC, &kapic);
>> + }
>> +
>> + return 0;
>> +}
>> +
>> +int kvm_get_apic(CPUState *env)
>
> Same here.
>
>> +{
>> + APICState *s = DO_UPCAST(APICState, busdev.qdev, env->apic_state);
>> + struct kvm_lapic_state kapic;
>> + int ret, i, v;
>> +
>> + if (s && kvm_enabled() && kvm_irqchip_in_kernel()) {
>> + ret = kvm_vcpu_ioctl(env, KVM_GET_LAPIC, &kapic);
>> + if (ret < 0) {
>> + return ret;
>> + }
>> +
>> + s->id = kvm_apic_get_reg(&kapic, 0x2) >> 24;
>> + s->tpr = kvm_apic_get_reg(&kapic, 0x8);
>> + s->arb_id = kvm_apic_get_reg(&kapic, 0x9);
>> + s->log_dest = kvm_apic_get_reg(&kapic, 0xd) >> 24;
>> + s->dest_mode = kvm_apic_get_reg(&kapic, 0xe) >> 28;
>> + s->spurious_vec = kvm_apic_get_reg(&kapic, 0xf);
>> + for (i = 0; i < 8; i++) {
>> + s->isr[i] = kvm_apic_get_reg(&kapic, 0x10 + i);
>> + s->tmr[i] = kvm_apic_get_reg(&kapic, 0x18 + i);
>> + s->irr[i] = kvm_apic_get_reg(&kapic, 0x20 + i);
>> + }
>> + s->esr = kvm_apic_get_reg(&kapic, 0x28);
>> + s->icr[0] = kvm_apic_get_reg(&kapic, 0x30);
>> + s->icr[1] = kvm_apic_get_reg(&kapic, 0x31);
>> + for (i = 0; i < APIC_LVT_NB; i++) {
>> + s->lvt[i] = kvm_apic_get_reg(&kapic, 0x32 + i);
>> + }
>> + s->initial_count = kvm_apic_get_reg(&kapic, 0x38);
>> + s->divide_conf = kvm_apic_get_reg(&kapic, 0x3e);
>> +
>> + v = (s->divide_conf & 3) | ((s->divide_conf >> 1) & 4);
>> + s->count_shift = (v + 1) & 7;
>> +
>> + s->initial_count_load_time = qemu_get_clock_ns(vm_clock);
>> + apic_next_timer(s, s->initial_count_load_time);
>> + }
>> + return 0;
>> +}
>> +
>> +static void kvm_apic_set_base(APICState *s, uint64_t val)
>> +{
>> + s->apicbase = val;
>> +}
>> +
>> +static void kvm_apic_set_tpr(APICState *s, uint8_t val)
>> +{
>> + s->tpr = (val & 0x0f) << 4;
>> +}
>> +
>> +static void do_inject_external_nmi(void *data)
>> +{
>> + APICState *s = data;
>> + CPUState *env = s->cpu_env;
>> + uint32_t lvt;
>> + int ret;
>> +
>> + cpu_synchronize_state(env);
>> +
>> + lvt = s->lvt[APIC_LVT_LINT1];
>> + if (!(lvt & APIC_LVT_MASKED) && ((lvt >> 8) & 7) == APIC_DM_NMI) {
>> + ret = kvm_vcpu_ioctl(env, KVM_NMI);
>> + if (ret < 0) {
>> + fprintf(stderr, "KVM: injection failed, NMI lost (%s)\n",
>> + strerror(-ret));
>> + }
>> + }
>> +}
>> +
>> +static void kvm_apic_external_nmi(APICState *s)
>> +{
>> + run_on_cpu(s->cpu_env, do_inject_external_nmi, s);
>
> Here probably CPUState would make more sense.
But here I think it's more consistent to stay with APICState (for the
hook) and DeviceState (for apic_deliver_nmi).
Jan
--
Siemens AG, Corporate Technology, CT T DE IT 1
Corporate Competence Center Embedded Linux
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [Qemu-devel] [PATCH v4 12/15] kvm: x86: Add user space part for in-kernel APIC
2011-12-09 7:45 ` Jan Kiszka
@ 2011-12-09 7:52 ` Jan Kiszka
2011-12-10 15:40 ` Blue Swirl
0 siblings, 1 reply; 32+ messages in thread
From: Jan Kiszka @ 2011-12-09 7:52 UTC (permalink / raw)
To: Blue Swirl
Cc: Anthony Liguori, Lai Jiangshan, kvm@vger.kernel.org,
Michael S. Tsirkin, Marcelo Tosatti, qemu-devel, Avi Kivity
On 2011-12-09 08:45, Jan Kiszka wrote:
> On 2011-12-08 22:16, Blue Swirl wrote:
>> On Thu, Dec 8, 2011 at 11:52, Jan Kiszka <jan.kiszka@siemens.com> wrote:
>>> This introduces the alternative APIC backend which makes use of KVM's
>>> in-kernel device model. External NMI injection via LINT1 is emulated by
>>> checking the current state of the in-kernel APIC, only injecting a NMI
>>> into the VCPU if LINT1 is unmasked and configured to DM_NMI.
>>>
>>> MSI is not yet supported, so we disable this when the in-kernel model is
>>> in use.
>>>
>>> CC: Lai Jiangshan <laijs@cn.fujitsu.com>
>>> Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
>>> ---
>>> Makefile.target | 2 +-
>>> hw/kvm/apic.c | 154 +++++++++++++++++++++++++++++++++++++++++++++++++++++
>>> hw/pc.c | 15 ++++--
>>> kvm.h | 3 +
>>> target-i386/kvm.c | 8 +++
>>> 5 files changed, 176 insertions(+), 6 deletions(-)
>>> create mode 100644 hw/kvm/apic.c
>>>
>>> diff --git a/Makefile.target b/Makefile.target
>>> index b549988..76de485 100644
>>> --- a/Makefile.target
>>> +++ b/Makefile.target
>>> @@ -236,7 +236,7 @@ obj-i386-y += vmport.o
>>> obj-i386-y += device-hotplug.o pci-hotplug.o smbios.o wdt_ib700.o
>>> obj-i386-y += debugcon.o multiboot.o
>>> obj-i386-y += pc_piix.o
>>> -obj-i386-$(CONFIG_KVM) += kvm/clock.o
>>> +obj-i386-$(CONFIG_KVM) += kvm/clock.o kvm/apic.o
>>> obj-i386-$(CONFIG_SPICE) += qxl.o qxl-logger.o qxl-render.o
>>>
>>> # shared objects
>>> diff --git a/hw/kvm/apic.c b/hw/kvm/apic.c
>>> new file mode 100644
>>> index 0000000..3924f9e
>>> --- /dev/null
>>> +++ b/hw/kvm/apic.c
>>> @@ -0,0 +1,154 @@
>>> +/*
>>> + * KVM in-kernel APIC support
>>> + *
>>> + * Copyright (c) 2011 Siemens AG
>>> + *
>>> + * Authors:
>>> + * Jan Kiszka <jan.kiszka@siemens.com>
>>> + *
>>> + * This work is licensed under the terms of the GNU GPL version 2.
>>> + * See the COPYING file in the top-level directory.
>>> + */
>>> +#include "hw/apic_internal.h"
>>> +#include "kvm.h"
>>> +
>>> +static inline void kvm_apic_set_reg(struct kvm_lapic_state *kapic,
>>> + int reg_id, uint32_t val)
>>> +{
>>> + *((uint32_t *)(kapic->regs + (reg_id << 4))) = val;
>>> +}
>>> +
>>> +static inline uint32_t kvm_apic_get_reg(struct kvm_lapic_state *kapic,
>>> + int reg_id)
>>> +{
>>> + return *((uint32_t *)(kapic->regs + (reg_id << 4)));
>>> +}
>>> +
>>> +int kvm_put_apic(CPUState *env)
>>> +{
>>> + APICState *s = DO_UPCAST(APICState, busdev.qdev, env->apic_state);
>>
>> Please pass APICState instead of CPUState.
>
> DeviceState, I suppose. Yes, makes more sense, update will follow.
On second look: no, I'll keep it as is. All kvm_get/put_* helpers have
this kind of signature, i.e. are working against env. kvm_get/put_apic
just happens to be implemented outside of target-i386/kvm.c. And they
require both APIC and CPUState anyway, so it makes no difference.
Jan
--
Siemens AG, Corporate Technology, CT T DE IT 1
Corporate Competence Center Embedded Linux
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [Qemu-devel] [PATCH v4 12/15] kvm: x86: Add user space part for in-kernel APIC
2011-12-09 7:52 ` Jan Kiszka
@ 2011-12-10 15:40 ` Blue Swirl
2011-12-10 15:58 ` Jan Kiszka
0 siblings, 1 reply; 32+ messages in thread
From: Blue Swirl @ 2011-12-10 15:40 UTC (permalink / raw)
To: Jan Kiszka
Cc: Anthony Liguori, Lai Jiangshan, kvm@vger.kernel.org,
Michael S. Tsirkin, Marcelo Tosatti, qemu-devel, Avi Kivity
On Fri, Dec 9, 2011 at 07:52, Jan Kiszka <jan.kiszka@siemens.com> wrote:
> On 2011-12-09 08:45, Jan Kiszka wrote:
>> On 2011-12-08 22:16, Blue Swirl wrote:
>>> On Thu, Dec 8, 2011 at 11:52, Jan Kiszka <jan.kiszka@siemens.com> wrote:
>>>> This introduces the alternative APIC backend which makes use of KVM's
>>>> in-kernel device model. External NMI injection via LINT1 is emulated by
>>>> checking the current state of the in-kernel APIC, only injecting a NMI
>>>> into the VCPU if LINT1 is unmasked and configured to DM_NMI.
>>>>
>>>> MSI is not yet supported, so we disable this when the in-kernel model is
>>>> in use.
>>>>
>>>> CC: Lai Jiangshan <laijs@cn.fujitsu.com>
>>>> Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
>>>> ---
>>>> Makefile.target | 2 +-
>>>> hw/kvm/apic.c | 154 +++++++++++++++++++++++++++++++++++++++++++++++++++++
>>>> hw/pc.c | 15 ++++--
>>>> kvm.h | 3 +
>>>> target-i386/kvm.c | 8 +++
>>>> 5 files changed, 176 insertions(+), 6 deletions(-)
>>>> create mode 100644 hw/kvm/apic.c
>>>>
>>>> diff --git a/Makefile.target b/Makefile.target
>>>> index b549988..76de485 100644
>>>> --- a/Makefile.target
>>>> +++ b/Makefile.target
>>>> @@ -236,7 +236,7 @@ obj-i386-y += vmport.o
>>>> obj-i386-y += device-hotplug.o pci-hotplug.o smbios.o wdt_ib700.o
>>>> obj-i386-y += debugcon.o multiboot.o
>>>> obj-i386-y += pc_piix.o
>>>> -obj-i386-$(CONFIG_KVM) += kvm/clock.o
>>>> +obj-i386-$(CONFIG_KVM) += kvm/clock.o kvm/apic.o
>>>> obj-i386-$(CONFIG_SPICE) += qxl.o qxl-logger.o qxl-render.o
>>>>
>>>> # shared objects
>>>> diff --git a/hw/kvm/apic.c b/hw/kvm/apic.c
>>>> new file mode 100644
>>>> index 0000000..3924f9e
>>>> --- /dev/null
>>>> +++ b/hw/kvm/apic.c
>>>> @@ -0,0 +1,154 @@
>>>> +/*
>>>> + * KVM in-kernel APIC support
>>>> + *
>>>> + * Copyright (c) 2011 Siemens AG
>>>> + *
>>>> + * Authors:
>>>> + * Jan Kiszka <jan.kiszka@siemens.com>
>>>> + *
>>>> + * This work is licensed under the terms of the GNU GPL version 2.
>>>> + * See the COPYING file in the top-level directory.
>>>> + */
>>>> +#include "hw/apic_internal.h"
>>>> +#include "kvm.h"
>>>> +
>>>> +static inline void kvm_apic_set_reg(struct kvm_lapic_state *kapic,
>>>> + int reg_id, uint32_t val)
>>>> +{
>>>> + *((uint32_t *)(kapic->regs + (reg_id << 4))) = val;
>>>> +}
>>>> +
>>>> +static inline uint32_t kvm_apic_get_reg(struct kvm_lapic_state *kapic,
>>>> + int reg_id)
>>>> +{
>>>> + return *((uint32_t *)(kapic->regs + (reg_id << 4)));
>>>> +}
>>>> +
>>>> +int kvm_put_apic(CPUState *env)
>>>> +{
>>>> + APICState *s = DO_UPCAST(APICState, busdev.qdev, env->apic_state);
>>>
>>> Please pass APICState instead of CPUState.
>>
>> DeviceState, I suppose. Yes, makes more sense, update will follow.
>
> On second look: no, I'll keep it as is. All kvm_get/put_* helpers have
> this kind of signature, i.e. are working against env.
There's kvm_get_supported_msrs for example.
> kvm_get/put_apic
> just happens to be implemented outside of target-i386/kvm.c. And they
> require both APIC and CPUState anyway, so it makes no difference.
It does, passing CPUState violates layering. Please split the
functions so that the ioctl calls which need CPUState go to kvm.c. For
example, the functions in kvm/apic.c could just perform copying from
kvm_lapic_state fields to APICstate fields and vice versa.
The KVM interface by the way does not look so clever. Why isn't there
just an array of 32 bit fields so the casts can be avoided? Perhaps
APICState should be (later) changed to match KVM version so that the
structure can be passed directly without copying.
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [Qemu-devel] [PATCH v4 12/15] kvm: x86: Add user space part for in-kernel APIC
2011-12-10 15:40 ` Blue Swirl
@ 2011-12-10 15:58 ` Jan Kiszka
2011-12-10 16:11 ` Blue Swirl
0 siblings, 1 reply; 32+ messages in thread
From: Jan Kiszka @ 2011-12-10 15:58 UTC (permalink / raw)
To: Blue Swirl
Cc: Anthony Liguori, Lai Jiangshan, kvm@vger.kernel.org,
Michael S. Tsirkin, Marcelo Tosatti, qemu-devel, Avi Kivity
[-- Attachment #1: Type: text/plain, Size: 4101 bytes --]
On 2011-12-10 16:40, Blue Swirl wrote:
> On Fri, Dec 9, 2011 at 07:52, Jan Kiszka <jan.kiszka@siemens.com> wrote:
>> On 2011-12-09 08:45, Jan Kiszka wrote:
>>> On 2011-12-08 22:16, Blue Swirl wrote:
>>>> On Thu, Dec 8, 2011 at 11:52, Jan Kiszka <jan.kiszka@siemens.com> wrote:
>>>>> This introduces the alternative APIC backend which makes use of KVM's
>>>>> in-kernel device model. External NMI injection via LINT1 is emulated by
>>>>> checking the current state of the in-kernel APIC, only injecting a NMI
>>>>> into the VCPU if LINT1 is unmasked and configured to DM_NMI.
>>>>>
>>>>> MSI is not yet supported, so we disable this when the in-kernel model is
>>>>> in use.
>>>>>
>>>>> CC: Lai Jiangshan <laijs@cn.fujitsu.com>
>>>>> Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
>>>>> ---
>>>>> Makefile.target | 2 +-
>>>>> hw/kvm/apic.c | 154 +++++++++++++++++++++++++++++++++++++++++++++++++++++
>>>>> hw/pc.c | 15 ++++--
>>>>> kvm.h | 3 +
>>>>> target-i386/kvm.c | 8 +++
>>>>> 5 files changed, 176 insertions(+), 6 deletions(-)
>>>>> create mode 100644 hw/kvm/apic.c
>>>>>
>>>>> diff --git a/Makefile.target b/Makefile.target
>>>>> index b549988..76de485 100644
>>>>> --- a/Makefile.target
>>>>> +++ b/Makefile.target
>>>>> @@ -236,7 +236,7 @@ obj-i386-y += vmport.o
>>>>> obj-i386-y += device-hotplug.o pci-hotplug.o smbios.o wdt_ib700.o
>>>>> obj-i386-y += debugcon.o multiboot.o
>>>>> obj-i386-y += pc_piix.o
>>>>> -obj-i386-$(CONFIG_KVM) += kvm/clock.o
>>>>> +obj-i386-$(CONFIG_KVM) += kvm/clock.o kvm/apic.o
>>>>> obj-i386-$(CONFIG_SPICE) += qxl.o qxl-logger.o qxl-render.o
>>>>>
>>>>> # shared objects
>>>>> diff --git a/hw/kvm/apic.c b/hw/kvm/apic.c
>>>>> new file mode 100644
>>>>> index 0000000..3924f9e
>>>>> --- /dev/null
>>>>> +++ b/hw/kvm/apic.c
>>>>> @@ -0,0 +1,154 @@
>>>>> +/*
>>>>> + * KVM in-kernel APIC support
>>>>> + *
>>>>> + * Copyright (c) 2011 Siemens AG
>>>>> + *
>>>>> + * Authors:
>>>>> + * Jan Kiszka <jan.kiszka@siemens.com>
>>>>> + *
>>>>> + * This work is licensed under the terms of the GNU GPL version 2.
>>>>> + * See the COPYING file in the top-level directory.
>>>>> + */
>>>>> +#include "hw/apic_internal.h"
>>>>> +#include "kvm.h"
>>>>> +
>>>>> +static inline void kvm_apic_set_reg(struct kvm_lapic_state *kapic,
>>>>> + int reg_id, uint32_t val)
>>>>> +{
>>>>> + *((uint32_t *)(kapic->regs + (reg_id << 4))) = val;
>>>>> +}
>>>>> +
>>>>> +static inline uint32_t kvm_apic_get_reg(struct kvm_lapic_state *kapic,
>>>>> + int reg_id)
>>>>> +{
>>>>> + return *((uint32_t *)(kapic->regs + (reg_id << 4)));
>>>>> +}
>>>>> +
>>>>> +int kvm_put_apic(CPUState *env)
>>>>> +{
>>>>> + APICState *s = DO_UPCAST(APICState, busdev.qdev, env->apic_state);
>>>>
>>>> Please pass APICState instead of CPUState.
>>>
>>> DeviceState, I suppose. Yes, makes more sense, update will follow.
>>
>> On second look: no, I'll keep it as is. All kvm_get/put_* helpers have
>> this kind of signature, i.e. are working against env.
>
> There's kvm_get_supported_msrs for example.
>
>> kvm_get/put_apic
>> just happens to be implemented outside of target-i386/kvm.c. And they
>> require both APIC and CPUState anyway, so it makes no difference.
>
> It does, passing CPUState violates layering. Please split the
> functions so that the ioctl calls which need CPUState go to kvm.c. For
> example, the functions in kvm/apic.c could just perform copying from
> kvm_lapic_state fields to APICstate fields and vice versa.
That's a good idea.
>
> The KVM interface by the way does not look so clever. Why isn't there
> just an array of 32 bit fields so the casts can be avoided? Perhaps
> APICState should be (later) changed to match KVM version so that the
> structure can be passed directly without copying.
Wouldn't that complicate the use in the user space model again? At least
for registers that are used with both backends.
Jan
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 262 bytes --]
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [Qemu-devel] [PATCH v4 12/15] kvm: x86: Add user space part for in-kernel APIC
2011-12-10 15:58 ` Jan Kiszka
@ 2011-12-10 16:11 ` Blue Swirl
0 siblings, 0 replies; 32+ messages in thread
From: Blue Swirl @ 2011-12-10 16:11 UTC (permalink / raw)
To: Jan Kiszka
Cc: Anthony Liguori, Lai Jiangshan, kvm@vger.kernel.org,
Michael S. Tsirkin, Marcelo Tosatti, qemu-devel, Avi Kivity
On Sat, Dec 10, 2011 at 15:58, Jan Kiszka <jan.kiszka@web.de> wrote:
> On 2011-12-10 16:40, Blue Swirl wrote:
>> On Fri, Dec 9, 2011 at 07:52, Jan Kiszka <jan.kiszka@siemens.com> wrote:
>>> On 2011-12-09 08:45, Jan Kiszka wrote:
>>>> On 2011-12-08 22:16, Blue Swirl wrote:
>>>>> On Thu, Dec 8, 2011 at 11:52, Jan Kiszka <jan.kiszka@siemens.com> wrote:
>>>>>> This introduces the alternative APIC backend which makes use of KVM's
>>>>>> in-kernel device model. External NMI injection via LINT1 is emulated by
>>>>>> checking the current state of the in-kernel APIC, only injecting a NMI
>>>>>> into the VCPU if LINT1 is unmasked and configured to DM_NMI.
>>>>>>
>>>>>> MSI is not yet supported, so we disable this when the in-kernel model is
>>>>>> in use.
>>>>>>
>>>>>> CC: Lai Jiangshan <laijs@cn.fujitsu.com>
>>>>>> Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
>>>>>> ---
>>>>>> Makefile.target | 2 +-
>>>>>> hw/kvm/apic.c | 154 +++++++++++++++++++++++++++++++++++++++++++++++++++++
>>>>>> hw/pc.c | 15 ++++--
>>>>>> kvm.h | 3 +
>>>>>> target-i386/kvm.c | 8 +++
>>>>>> 5 files changed, 176 insertions(+), 6 deletions(-)
>>>>>> create mode 100644 hw/kvm/apic.c
>>>>>>
>>>>>> diff --git a/Makefile.target b/Makefile.target
>>>>>> index b549988..76de485 100644
>>>>>> --- a/Makefile.target
>>>>>> +++ b/Makefile.target
>>>>>> @@ -236,7 +236,7 @@ obj-i386-y += vmport.o
>>>>>> obj-i386-y += device-hotplug.o pci-hotplug.o smbios.o wdt_ib700.o
>>>>>> obj-i386-y += debugcon.o multiboot.o
>>>>>> obj-i386-y += pc_piix.o
>>>>>> -obj-i386-$(CONFIG_KVM) += kvm/clock.o
>>>>>> +obj-i386-$(CONFIG_KVM) += kvm/clock.o kvm/apic.o
>>>>>> obj-i386-$(CONFIG_SPICE) += qxl.o qxl-logger.o qxl-render.o
>>>>>>
>>>>>> # shared objects
>>>>>> diff --git a/hw/kvm/apic.c b/hw/kvm/apic.c
>>>>>> new file mode 100644
>>>>>> index 0000000..3924f9e
>>>>>> --- /dev/null
>>>>>> +++ b/hw/kvm/apic.c
>>>>>> @@ -0,0 +1,154 @@
>>>>>> +/*
>>>>>> + * KVM in-kernel APIC support
>>>>>> + *
>>>>>> + * Copyright (c) 2011 Siemens AG
>>>>>> + *
>>>>>> + * Authors:
>>>>>> + * Jan Kiszka <jan.kiszka@siemens.com>
>>>>>> + *
>>>>>> + * This work is licensed under the terms of the GNU GPL version 2.
>>>>>> + * See the COPYING file in the top-level directory.
>>>>>> + */
>>>>>> +#include "hw/apic_internal.h"
>>>>>> +#include "kvm.h"
>>>>>> +
>>>>>> +static inline void kvm_apic_set_reg(struct kvm_lapic_state *kapic,
>>>>>> + int reg_id, uint32_t val)
>>>>>> +{
>>>>>> + *((uint32_t *)(kapic->regs + (reg_id << 4))) = val;
>>>>>> +}
>>>>>> +
>>>>>> +static inline uint32_t kvm_apic_get_reg(struct kvm_lapic_state *kapic,
>>>>>> + int reg_id)
>>>>>> +{
>>>>>> + return *((uint32_t *)(kapic->regs + (reg_id << 4)));
>>>>>> +}
>>>>>> +
>>>>>> +int kvm_put_apic(CPUState *env)
>>>>>> +{
>>>>>> + APICState *s = DO_UPCAST(APICState, busdev.qdev, env->apic_state);
>>>>>
>>>>> Please pass APICState instead of CPUState.
>>>>
>>>> DeviceState, I suppose. Yes, makes more sense, update will follow.
>>>
>>> On second look: no, I'll keep it as is. All kvm_get/put_* helpers have
>>> this kind of signature, i.e. are working against env.
>>
>> There's kvm_get_supported_msrs for example.
>>
>>> kvm_get/put_apic
>>> just happens to be implemented outside of target-i386/kvm.c. And they
>>> require both APIC and CPUState anyway, so it makes no difference.
>>
>> It does, passing CPUState violates layering. Please split the
>> functions so that the ioctl calls which need CPUState go to kvm.c. For
>> example, the functions in kvm/apic.c could just perform copying from
>> kvm_lapic_state fields to APICstate fields and vice versa.
>
> That's a good idea.
>
>>
>> The KVM interface by the way does not look so clever. Why isn't there
>> just an array of 32 bit fields so the casts can be avoided? Perhaps
>> APICState should be (later) changed to match KVM version so that the
>> structure can be passed directly without copying.
>
> Wouldn't that complicate the use in the user space model again? At least
> for registers that are used with both backends.
Well, we have (at least) two styles how to model devices.
In the first one, the device state structure contains an array of
registers, so the functions which use them may need for example to
perform some bit field extraction to get what they need.
In the model used by APIC and others, the structure contains cooked
values, for example divide_count and count_shift in APICState. This
means that the CPU accesses get slightly slower since the fields need
to be packed and unpacked but the other functions may be faster.
Which one is better depends on frequency and importance of register
accesses by CPU vs. other accesses. But it shouldn't complicate that
much either way. Actually design choices like this may have been taken
without too much consideration.
Alternatively, KVM interface could be changed to take QEMU structure
directly, but I don't suppose that would be a good idea. It would be
easier for everyone if QEMU changed instead.
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support
2011-12-08 11:52 [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support Jan Kiszka
` (15 preceding siblings ...)
2011-12-08 21:25 ` [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic " Blue Swirl
@ 2011-12-12 16:37 ` Marcelo Tosatti
2011-12-12 16:51 ` Jan Kiszka
16 siblings, 1 reply; 32+ messages in thread
From: Marcelo Tosatti @ 2011-12-12 16:37 UTC (permalink / raw)
To: Jan Kiszka
Cc: Anthony Liguori, Lai Jiangshan, kvm, Michael S. Tsirkin,
qemu-devel, Blue Swirl, Avi Kivity
On Thu, Dec 08, 2011 at 12:52:19PM +0100, Jan Kiszka wrote:
> Changes in v4:
> - rebased of current uq/master
> - fixed stupid bugs that broke bisectability and user space irqchip mode
> - integrated NMI-over-LINT1 injection logic
>
> CC: Lai Jiangshan <laijs@cn.fujitsu.com>
>
> Jan Kiszka (15):
> msi: Generalize msix_supported to msi_supported
> kvm: Move kvmclock into hw/kvm folder
> apic: Stop timer on reset
> apic: Inject external NMI events via LINT1
> apic: Introduce backend/frontend infrastructure for KVM reuse
> apic: Open-code timer save/restore
> i8259: Introduce backend/frontend infrastructure for KVM reuse
> ioapic: Introduce backend/frontend infrastructure for KVM reuse
> memory: Introduce memory_region_init_reservation
> kvm: Introduce core services for in-kernel irqchip support
> kvm: x86: Establish IRQ0 override control
> kvm: x86: Add user space part for in-kernel APIC
> kvm: x86: Add user space part for in-kernel i8259
> kvm: x86: Add user space part for in-kernel IOAPIC
> kvm: Arm in-kernel irqchip support
Looks good to me.
Any thoughts on the qemu-kvm merge plan? Sounds painful.
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support
2011-12-12 16:37 ` Marcelo Tosatti
@ 2011-12-12 16:51 ` Jan Kiszka
2011-12-12 17:37 ` Avi Kivity
2011-12-15 10:33 ` Jan Kiszka
0 siblings, 2 replies; 32+ messages in thread
From: Jan Kiszka @ 2011-12-12 16:51 UTC (permalink / raw)
To: Marcelo Tosatti
Cc: Anthony Liguori, Lai Jiangshan, kvm@vger.kernel.org,
Michael S. Tsirkin, qemu-devel, Blue Swirl, Avi Kivity
On 2011-12-12 17:37, Marcelo Tosatti wrote:
> On Thu, Dec 08, 2011 at 12:52:19PM +0100, Jan Kiszka wrote:
>> Changes in v4:
>> - rebased of current uq/master
>> - fixed stupid bugs that broke bisectability and user space irqchip mode
>> - integrated NMI-over-LINT1 injection logic
>>
>> CC: Lai Jiangshan <laijs@cn.fujitsu.com>
>>
>> Jan Kiszka (15):
>> msi: Generalize msix_supported to msi_supported
>> kvm: Move kvmclock into hw/kvm folder
>> apic: Stop timer on reset
>> apic: Inject external NMI events via LINT1
>> apic: Introduce backend/frontend infrastructure for KVM reuse
>> apic: Open-code timer save/restore
>> i8259: Introduce backend/frontend infrastructure for KVM reuse
>> ioapic: Introduce backend/frontend infrastructure for KVM reuse
>> memory: Introduce memory_region_init_reservation
>> kvm: Introduce core services for in-kernel irqchip support
>> kvm: x86: Establish IRQ0 override control
>> kvm: x86: Add user space part for in-kernel APIC
>> kvm: x86: Add user space part for in-kernel i8259
>> kvm: x86: Add user space part for in-kernel IOAPIC
>> kvm: Arm in-kernel irqchip support
>
> Looks good to me.
>
> Any thoughts on the qemu-kvm merge plan? Sounds painful.
Pain will be where the existing qemu-kvm extensions collide with these
refactored upstream devices (backend/frontend split specifically).
That's where we have to merge very carefully. Haven't tried this yet,
will give it a spin tomorrow or so.
>From that point on, disabling the new stuff for now and at some point
switching over from the old one should be simple again.
BTW, PIT+HPET+speaker will cause similar issues for the same reasons.
Jan
--
Siemens AG, Corporate Technology, CT T DE IT 1
Corporate Competence Center Embedded Linux
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support
2011-12-12 16:51 ` Jan Kiszka
@ 2011-12-12 17:37 ` Avi Kivity
2011-12-12 17:42 ` Jan Kiszka
2011-12-15 10:33 ` Jan Kiszka
1 sibling, 1 reply; 32+ messages in thread
From: Avi Kivity @ 2011-12-12 17:37 UTC (permalink / raw)
To: Jan Kiszka
Cc: Anthony Liguori, Lai Jiangshan, kvm@vger.kernel.org,
Michael S. Tsirkin, Marcelo Tosatti, qemu-devel, Blue Swirl
On 12/12/2011 06:51 PM, Jan Kiszka wrote:
> >
> > Any thoughts on the qemu-kvm merge plan? Sounds painful.
>
> Pain will be where the existing qemu-kvm extensions collide with these
> refactored upstream devices (backend/frontend split specifically).
> That's where we have to merge very carefully. Haven't tried this yet,
> will give it a spin tomorrow or so.
>
> From that point on, disabling the new stuff for now and at some point
> switching over from the old one should be simple again.
>
> BTW, PIT+HPET+speaker will cause similar issues for the same reasons.
>
It's a little late for this, but refactoring qemu-kvm in-tree and then
splitting it into patches would have been easier. Let's try it this way
for the next batch.
--
error compiling committee.c: too many arguments to function
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support
2011-12-12 17:37 ` Avi Kivity
@ 2011-12-12 17:42 ` Jan Kiszka
2011-12-12 17:59 ` Avi Kivity
0 siblings, 1 reply; 32+ messages in thread
From: Jan Kiszka @ 2011-12-12 17:42 UTC (permalink / raw)
To: Avi Kivity
Cc: Anthony Liguori, Lai Jiangshan, kvm@vger.kernel.org,
Michael S. Tsirkin, Marcelo Tosatti, qemu-devel, Blue Swirl
On 2011-12-12 18:37, Avi Kivity wrote:
> On 12/12/2011 06:51 PM, Jan Kiszka wrote:
>>>
>>> Any thoughts on the qemu-kvm merge plan? Sounds painful.
>>
>> Pain will be where the existing qemu-kvm extensions collide with these
>> refactored upstream devices (backend/frontend split specifically).
>> That's where we have to merge very carefully. Haven't tried this yet,
>> will give it a spin tomorrow or so.
>>
>> From that point on, disabling the new stuff for now and at some point
>> switching over from the old one should be simple again.
>>
>> BTW, PIT+HPET+speaker will cause similar issues for the same reasons.
>>
>
> It's a little late for this, but refactoring qemu-kvm in-tree and then
> splitting it into patches would have been easier. Let's try it this way
> for the next batch.
I thought about this, but it definitely takes a clean, qemu-kvm free
base as start. The point is to design something free of all the legacy,
only looking at the other code base to extract the logic.
Moreover, there was and still is quite some upstream cleanup necessary,
and that never goes well with the delta of qemu-kvm.
Jan
--
Siemens AG, Corporate Technology, CT T DE IT 1
Corporate Competence Center Embedded Linux
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support
2011-12-12 17:42 ` Jan Kiszka
@ 2011-12-12 17:59 ` Avi Kivity
0 siblings, 0 replies; 32+ messages in thread
From: Avi Kivity @ 2011-12-12 17:59 UTC (permalink / raw)
To: Jan Kiszka
Cc: Anthony Liguori, Lai Jiangshan, kvm@vger.kernel.org,
Michael S. Tsirkin, Marcelo Tosatti, qemu-devel, Blue Swirl
On 12/12/2011 07:42 PM, Jan Kiszka wrote:
> > It's a little late for this, but refactoring qemu-kvm in-tree and then
> > splitting it into patches would have been easier. Let's try it this way
> > for the next batch.
>
> I thought about this, but it definitely takes a clean, qemu-kvm free
> base as start. The point is to design something free of all the legacy,
> only looking at the other code base to extract the logic.
If qemu-kvm was merged into qemu as is, then I'm sure you'd be able to
refactor it into good shape (though there might be a little less
motivation).
> Moreover, there was and still is quite some upstream cleanup necessary,
> and that never goes well with the delta of qemu-kvm.
That's a good point.
--
error compiling committee.c: too many arguments to function
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support
2011-12-12 16:51 ` Jan Kiszka
2011-12-12 17:37 ` Avi Kivity
@ 2011-12-15 10:33 ` Jan Kiszka
2011-12-15 11:54 ` Avi Kivity
1 sibling, 1 reply; 32+ messages in thread
From: Jan Kiszka @ 2011-12-15 10:33 UTC (permalink / raw)
To: Marcelo Tosatti
Cc: Anthony Liguori, Lai Jiangshan, kvm@vger.kernel.org,
Michael S. Tsirkin, qemu-devel, Blue Swirl, Avi Kivity
On 2011-12-12 17:51, Jan Kiszka wrote:
> On 2011-12-12 17:37, Marcelo Tosatti wrote:
>> On Thu, Dec 08, 2011 at 12:52:19PM +0100, Jan Kiszka wrote:
>>> Changes in v4:
>>> - rebased of current uq/master
>>> - fixed stupid bugs that broke bisectability and user space irqchip mode
>>> - integrated NMI-over-LINT1 injection logic
>>>
>>> CC: Lai Jiangshan <laijs@cn.fujitsu.com>
>>>
>>> Jan Kiszka (15):
>>> msi: Generalize msix_supported to msi_supported
>>> kvm: Move kvmclock into hw/kvm folder
>>> apic: Stop timer on reset
>>> apic: Inject external NMI events via LINT1
>>> apic: Introduce backend/frontend infrastructure for KVM reuse
>>> apic: Open-code timer save/restore
>>> i8259: Introduce backend/frontend infrastructure for KVM reuse
>>> ioapic: Introduce backend/frontend infrastructure for KVM reuse
>>> memory: Introduce memory_region_init_reservation
>>> kvm: Introduce core services for in-kernel irqchip support
>>> kvm: x86: Establish IRQ0 override control
>>> kvm: x86: Add user space part for in-kernel APIC
>>> kvm: x86: Add user space part for in-kernel i8259
>>> kvm: x86: Add user space part for in-kernel IOAPIC
>>> kvm: Arm in-kernel irqchip support
>>
>> Looks good to me.
>>
>> Any thoughts on the qemu-kvm merge plan? Sounds painful.
>
> Pain will be where the existing qemu-kvm extensions collide with these
> refactored upstream devices (backend/frontend split specifically).
> That's where we have to merge very carefully. Haven't tried this yet,
> will give it a spin tomorrow or so.
Done yesterday, still seems to work fine. The result can be found at
git://git.kiszka.org/qemu-kvm.git kvm-irqchip-merge
The integration of the upstream irqchip patches was, as expected, not
that hard. But the merge of my earlier refactorings, the
backend/frontend split-up caused some efforts.
I'm not sure what to do with that branch. We could either try to merge
it before pulling in an upstream version that includes the new irqchips.
But that won't work without manual conflict resolution as well. Or the
branch can serve as a reference for re-doing a merge later on.
Jan
--
Siemens AG, Corporate Technology, CT T DE IT 1
Corporate Competence Center Embedded Linux
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support
2011-12-15 10:33 ` Jan Kiszka
@ 2011-12-15 11:54 ` Avi Kivity
2011-12-15 11:59 ` Jan Kiszka
0 siblings, 1 reply; 32+ messages in thread
From: Avi Kivity @ 2011-12-15 11:54 UTC (permalink / raw)
To: Jan Kiszka
Cc: Anthony Liguori, Lai Jiangshan, kvm@vger.kernel.org,
Michael S. Tsirkin, Marcelo Tosatti, qemu-devel, Blue Swirl
On 12/15/2011 12:33 PM, Jan Kiszka wrote:
> >>
> >> Any thoughts on the qemu-kvm merge plan? Sounds painful.
> >
> > Pain will be where the existing qemu-kvm extensions collide with these
> > refactored upstream devices (backend/frontend split specifically).
> > That's where we have to merge very carefully. Haven't tried this yet,
> > will give it a spin tomorrow or so.
>
> Done yesterday, still seems to work fine. The result can be found at
>
> git://git.kiszka.org/qemu-kvm.git kvm-irqchip-merge
>
> The integration of the upstream irqchip patches was, as expected, not
> that hard. But the merge of my earlier refactorings, the
> backend/frontend split-up caused some efforts.
>
> I'm not sure what to do with that branch. We could either try to merge
> it before pulling in an upstream version that includes the new irqchips.
> But that won't work without manual conflict resolution as well. Or the
> branch can serve as a reference for re-doing a merge later on.
If we merge this before upstream, will the two sides end up equivalent?
Sounds like it'll be pretty easy to resolve the conflicts if so.
--
error compiling committee.c: too many arguments to function
^ permalink raw reply [flat|nested] 32+ messages in thread
* Re: [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support
2011-12-15 11:54 ` Avi Kivity
@ 2011-12-15 11:59 ` Jan Kiszka
0 siblings, 0 replies; 32+ messages in thread
From: Jan Kiszka @ 2011-12-15 11:59 UTC (permalink / raw)
To: Avi Kivity
Cc: Anthony Liguori, Lai Jiangshan, kvm@vger.kernel.org,
Michael S. Tsirkin, Marcelo Tosatti, qemu-devel, Blue Swirl
On 2011-12-15 12:54, Avi Kivity wrote:
> On 12/15/2011 12:33 PM, Jan Kiszka wrote:
>>>>
>>>> Any thoughts on the qemu-kvm merge plan? Sounds painful.
>>>
>>> Pain will be where the existing qemu-kvm extensions collide with these
>>> refactored upstream devices (backend/frontend split specifically).
>>> That's where we have to merge very carefully. Haven't tried this yet,
>>> will give it a spin tomorrow or so.
>>
>> Done yesterday, still seems to work fine. The result can be found at
>>
>> git://git.kiszka.org/qemu-kvm.git kvm-irqchip-merge
>>
>> The integration of the upstream irqchip patches was, as expected, not
>> that hard. But the merge of my earlier refactorings, the
>> backend/frontend split-up caused some efforts.
>>
>> I'm not sure what to do with that branch. We could either try to merge
>> it before pulling in an upstream version that includes the new irqchips.
>> But that won't work without manual conflict resolution as well. Or the
>> branch can serve as a reference for re-doing a merge later on.
>
> If we merge this before upstream, will the two sides end up equivalent?
qemu-kvm will carry the new upstream bits but will keep them widely
disabled until upstream is feature equivalent. Of course, merging the
kvm-irqchip-merge only makes sense once the corresponding upstream queue
was actually merged.
> Sounds like it'll be pretty easy to resolve the conflicts if so.
>
Jan
--
Siemens AG, Corporate Technology, CT T DE IT 1
Corporate Competence Center Embedded Linux
^ permalink raw reply [flat|nested] 32+ messages in thread
end of thread, other threads:[~2011-12-15 12:00 UTC | newest]
Thread overview: 32+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-12-08 11:52 [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic irqchip support Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 01/15] msi: Generalize msix_supported to msi_supported Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 02/15] kvm: Move kvmclock into hw/kvm folder Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 03/15] apic: Stop timer on reset Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 04/15] apic: Inject external NMI events via LINT1 Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 05/15] apic: Introduce backend/frontend infrastructure for KVM reuse Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 06/15] apic: Open-code timer save/restore Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 07/15] i8259: Introduce backend/frontend infrastructure for KVM reuse Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 08/15] ioapic: " Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 09/15] memory: Introduce memory_region_init_reservation Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 10/15] kvm: Introduce core services for in-kernel irqchip support Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 11/15] kvm: x86: Establish IRQ0 override control Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 12/15] kvm: x86: Add user space part for in-kernel APIC Jan Kiszka
2011-12-08 21:16 ` Blue Swirl
2011-12-09 7:45 ` Jan Kiszka
2011-12-09 7:52 ` Jan Kiszka
2011-12-10 15:40 ` Blue Swirl
2011-12-10 15:58 ` Jan Kiszka
2011-12-10 16:11 ` Blue Swirl
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 13/15] kvm: x86: Add user space part for in-kernel i8259 Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 14/15] kvm: x86: Add user space part for in-kernel IOAPIC Jan Kiszka
2011-12-08 11:52 ` [Qemu-devel] [PATCH v4 15/15] kvm: Arm in-kernel irqchip support Jan Kiszka
2011-12-08 21:25 ` [Qemu-devel] [PATCH v4 00/15] uq/master: Introduce basic " Blue Swirl
2011-12-09 7:40 ` Jan Kiszka
2011-12-12 16:37 ` Marcelo Tosatti
2011-12-12 16:51 ` Jan Kiszka
2011-12-12 17:37 ` Avi Kivity
2011-12-12 17:42 ` Jan Kiszka
2011-12-12 17:59 ` Avi Kivity
2011-12-15 10:33 ` Jan Kiszka
2011-12-15 11:54 ` Avi Kivity
2011-12-15 11:59 ` Jan Kiszka
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).