* [PATCH 1/9] msix: avoid leaking kvm data on init failure
2010-09-20 15:06 [PATCH 0/9] msix/kvm integration cleanups Avi Kivity
@ 2010-09-20 15:06 ` Avi Kivity
2010-09-20 15:06 ` [PATCH 2/9] msix: make kvm specific initialization a function Avi Kivity
` (8 subsequent siblings)
9 siblings, 0 replies; 19+ messages in thread
From: Avi Kivity @ 2010-09-20 15:06 UTC (permalink / raw)
To: Marcelo Tosatti, kvm, Michael S. Tsirkin
Move initialization after we're certain to succeed, so we don't leak
memory on failure.
Signed-off-by: Avi Kivity <avi@redhat.com>
---
hw/msix.c | 13 +++++++------
1 files changed, 7 insertions(+), 6 deletions(-)
diff --git a/hw/msix.c b/hw/msix.c
index 3dd0456..312439a 100644
--- a/hw/msix.c
+++ b/hw/msix.c
@@ -373,12 +373,6 @@ int msix_init(struct PCIDevice *dev, unsigned short nentries,
if (nentries > MSIX_MAX_ENTRIES)
return -EINVAL;
-#ifdef KVM_CAP_IRQCHIP
- if (kvm_enabled() && kvm_irqchip_in_kernel()) {
- dev->msix_irq_entries = qemu_malloc(nentries *
- sizeof *dev->msix_irq_entries);
- }
-#endif
dev->msix_mask_notifier_opaque =
qemu_mallocz(nentries * sizeof *dev->msix_mask_notifier_opaque);
dev->msix_mask_notifier = NULL;
@@ -400,6 +394,13 @@ int msix_init(struct PCIDevice *dev, unsigned short nentries,
if (ret)
goto err_config;
+#ifdef KVM_CAP_IRQCHIP
+ if (kvm_enabled() && kvm_irqchip_in_kernel()) {
+ dev->msix_irq_entries = qemu_malloc(nentries *
+ sizeof *dev->msix_irq_entries);
+ }
+#endif
+
dev->cap_present |= QEMU_PCI_CAP_MSIX;
return 0;
--
1.7.2.3
^ permalink raw reply related [flat|nested] 19+ messages in thread* [PATCH 2/9] msix: make kvm specific initialization a function
2010-09-20 15:06 [PATCH 0/9] msix/kvm integration cleanups Avi Kivity
2010-09-20 15:06 ` [PATCH 1/9] msix: avoid leaking kvm data on init failure Avi Kivity
@ 2010-09-20 15:06 ` Avi Kivity
2010-09-20 15:06 ` [PATCH 3/9] msix: move kvm specific msix notify into " Avi Kivity
` (7 subsequent siblings)
9 siblings, 0 replies; 19+ messages in thread
From: Avi Kivity @ 2010-09-20 15:06 UTC (permalink / raw)
To: Marcelo Tosatti, kvm, Michael S. Tsirkin
Signed-off-by: Avi Kivity <avi@redhat.com>
---
hw/msix.c | 19 +++++++++++++------
1 files changed, 13 insertions(+), 6 deletions(-)
diff --git a/hw/msix.c b/hw/msix.c
index 312439a..2e2ce5a 100644
--- a/hw/msix.c
+++ b/hw/msix.c
@@ -48,6 +48,17 @@ int msix_supported;
#ifdef CONFIG_KVM
/* KVM specific MSIX helpers */
+
+static void kvm_msix_init(PCIDevice *dev)
+{
+#ifdef KVM_CAP_IRQCHIP
+ if (kvm_enabled() && kvm_irqchip_in_kernel()) {
+ dev->msix_irq_entries = qemu_malloc(dev->msix_entries_nr *
+ sizeof *dev->msix_irq_entries);
+ }
+#endif
+}
+
static void kvm_msix_free(PCIDevice *dev)
{
int vector, changed = 0;
@@ -150,6 +161,7 @@ static void kvm_msix_del(PCIDevice *dev, unsigned vector)
}
#else
+static void kvm_msix_init(PCIDevice *dev) {}
static void kvm_msix_free(PCIDevice *dev) {}
static void kvm_msix_update(PCIDevice *dev, int vector,
int was_masked, int is_masked) {}
@@ -394,12 +406,7 @@ int msix_init(struct PCIDevice *dev, unsigned short nentries,
if (ret)
goto err_config;
-#ifdef KVM_CAP_IRQCHIP
- if (kvm_enabled() && kvm_irqchip_in_kernel()) {
- dev->msix_irq_entries = qemu_malloc(nentries *
- sizeof *dev->msix_irq_entries);
- }
-#endif
+ kvm_msix_init(dev);
dev->cap_present |= QEMU_PCI_CAP_MSIX;
return 0;
--
1.7.2.3
^ permalink raw reply related [flat|nested] 19+ messages in thread* [PATCH 3/9] msix: move kvm specific msix notify into a function
2010-09-20 15:06 [PATCH 0/9] msix/kvm integration cleanups Avi Kivity
2010-09-20 15:06 ` [PATCH 1/9] msix: avoid leaking kvm data on init failure Avi Kivity
2010-09-20 15:06 ` [PATCH 2/9] msix: make kvm specific initialization a function Avi Kivity
@ 2010-09-20 15:06 ` Avi Kivity
2010-09-20 15:06 ` [PATCH 4/9] msix: move definitions from msix.c to msix.h Avi Kivity
` (6 subsequent siblings)
9 siblings, 0 replies; 19+ messages in thread
From: Avi Kivity @ 2010-09-20 15:06 UTC (permalink / raw)
To: Marcelo Tosatti, kvm, Michael S. Tsirkin
Signed-off-by: Avi Kivity <avi@redhat.com>
---
hw/msix.c | 19 +++++++++++++++----
1 files changed, 15 insertions(+), 4 deletions(-)
diff --git a/hw/msix.c b/hw/msix.c
index 2e2ce5a..d762870 100644
--- a/hw/msix.c
+++ b/hw/msix.c
@@ -159,6 +159,18 @@ static void kvm_msix_del(PCIDevice *dev, unsigned vector)
kvm_del_routing_entry(kvm_context, &dev->msix_irq_entries[vector]);
kvm_commit_irq_routes(kvm_context);
}
+
+static bool kvm_msix_notify(PCIDevice *dev, unsigned vector)
+{
+#ifdef KVM_CAP_IRQCHIP
+ if (kvm_enabled() && kvm_irqchip_in_kernel()) {
+ kvm_set_irq(dev->msix_irq_entries[vector].gsi, 1, NULL);
+ return true;
+ }
+#endif
+ return false;
+}
+
#else
static void kvm_msix_init(PCIDevice *dev) {}
@@ -167,6 +179,8 @@ static void kvm_msix_update(PCIDevice *dev, int vector,
int was_masked, int is_masked) {}
static int kvm_msix_add(PCIDevice *dev, unsigned vector) { return -1; }
static void kvm_msix_del(PCIDevice *dev, unsigned vector) {}
+static bool kvm_msix_notify(PCIDevice *dev, unsigned vector) { return false; }
+
#endif
/* Add MSI-X capability to the config space for the device. */
@@ -525,12 +539,9 @@ void msix_notify(PCIDevice *dev, unsigned vector)
return;
}
-#ifdef KVM_CAP_IRQCHIP
- if (kvm_enabled() && kvm_irqchip_in_kernel()) {
- kvm_set_irq(dev->msix_irq_entries[vector].gsi, 1, NULL);
+ if (kvm_msix_notify(dev, vector)) {
return;
}
-#endif
address = pci_get_long(table_entry + MSIX_MSG_UPPER_ADDR);
address = (address << 32) | pci_get_long(table_entry + MSIX_MSG_ADDR);
--
1.7.2.3
^ permalink raw reply related [flat|nested] 19+ messages in thread* [PATCH 4/9] msix: move definitions from msix.c to msix.h
2010-09-20 15:06 [PATCH 0/9] msix/kvm integration cleanups Avi Kivity
` (2 preceding siblings ...)
2010-09-20 15:06 ` [PATCH 3/9] msix: move kvm specific msix notify into " Avi Kivity
@ 2010-09-20 15:06 ` Avi Kivity
2010-09-20 16:56 ` Michael S. Tsirkin
2010-09-20 15:06 ` [PATCH 5/9] Add missing #include to hw/irq.h Avi Kivity
` (5 subsequent siblings)
9 siblings, 1 reply; 19+ messages in thread
From: Avi Kivity @ 2010-09-20 15:06 UTC (permalink / raw)
To: Marcelo Tosatti, kvm, Michael S. Tsirkin
This allows us to reuse them from the kvm support code.
Signed-off-by: Avi Kivity <avi@redhat.com>
---
hw/msix.c | 27 ---------------------------
hw/msix.h | 26 ++++++++++++++++++++++++++
2 files changed, 26 insertions(+), 27 deletions(-)
diff --git a/hw/msix.c b/hw/msix.c
index d762870..dda1a24 100644
--- a/hw/msix.c
+++ b/hw/msix.c
@@ -16,33 +16,6 @@
#include "pci.h"
#include "kvm.h"
-/* MSI-X capability structure */
-#define MSIX_TABLE_OFFSET 4
-#define MSIX_PBA_OFFSET 8
-#define MSIX_CAP_LENGTH 12
-
-/* MSI enable bit and maskall bit are in byte 1 in FLAGS register */
-#define MSIX_CONTROL_OFFSET (PCI_MSIX_FLAGS + 1)
-#define MSIX_ENABLE_MASK (PCI_MSIX_FLAGS_ENABLE >> 8)
-#define MSIX_MASKALL_MASK (PCI_MSIX_FLAGS_MASKALL >> 8)
-
-/* MSI-X table format */
-#define MSIX_MSG_ADDR 0
-#define MSIX_MSG_UPPER_ADDR 4
-#define MSIX_MSG_DATA 8
-#define MSIX_VECTOR_CTRL 12
-#define MSIX_ENTRY_SIZE 16
-#define MSIX_VECTOR_MASK 0x1
-
-/* How much space does an MSIX table need. */
-/* The spec requires giving the table structure
- * a 4K aligned region all by itself. */
-#define MSIX_PAGE_SIZE 0x1000
-/* Reserve second half of the page for pending bits */
-#define MSIX_PAGE_PENDING (MSIX_PAGE_SIZE / 2)
-#define MSIX_MAX_ENTRIES 32
-
-
/* Flag for interrupt controller to declare MSI-X support */
int msix_supported;
diff --git a/hw/msix.h b/hw/msix.h
index 6b21ffb..5c84b3e 100644
--- a/hw/msix.h
+++ b/hw/msix.h
@@ -4,6 +4,32 @@
#include "qemu-common.h"
#include "pci.h"
+/* MSI-X capability structure */
+#define MSIX_TABLE_OFFSET 4
+#define MSIX_PBA_OFFSET 8
+#define MSIX_CAP_LENGTH 12
+
+/* MSI enable bit and maskall bit are in byte 1 in FLAGS register */
+#define MSIX_CONTROL_OFFSET (PCI_MSIX_FLAGS + 1)
+#define MSIX_ENABLE_MASK (PCI_MSIX_FLAGS_ENABLE >> 8)
+#define MSIX_MASKALL_MASK (PCI_MSIX_FLAGS_MASKALL >> 8)
+
+/* MSI-X table format */
+#define MSIX_MSG_ADDR 0
+#define MSIX_MSG_UPPER_ADDR 4
+#define MSIX_MSG_DATA 8
+#define MSIX_VECTOR_CTRL 12
+#define MSIX_ENTRY_SIZE 16
+#define MSIX_VECTOR_MASK 0x1
+
+/* How much space does an MSIX table need. */
+/* The spec requires giving the table structure
+ * a 4K aligned region all by itself. */
+#define MSIX_PAGE_SIZE 0x1000
+/* Reserve second half of the page for pending bits */
+#define MSIX_PAGE_PENDING (MSIX_PAGE_SIZE / 2)
+#define MSIX_MAX_ENTRIES 32
+
int msix_init(PCIDevice *pdev, unsigned short nentries,
unsigned bar_nr, unsigned bar_size);
--
1.7.2.3
^ permalink raw reply related [flat|nested] 19+ messages in thread* Re: [PATCH 4/9] msix: move definitions from msix.c to msix.h
2010-09-20 15:06 ` [PATCH 4/9] msix: move definitions from msix.c to msix.h Avi Kivity
@ 2010-09-20 16:56 ` Michael S. Tsirkin
2010-09-21 16:03 ` Avi Kivity
0 siblings, 1 reply; 19+ messages in thread
From: Michael S. Tsirkin @ 2010-09-20 16:56 UTC (permalink / raw)
To: Avi Kivity; +Cc: Marcelo Tosatti, kvm
On Mon, Sep 20, 2010 at 05:06:45PM +0200, Avi Kivity wrote:
> This allows us to reuse them from the kvm support code.
>
> Signed-off-by: Avi Kivity <avi@redhat.com>
I would rather all dealings with MSI-X table stayed in one place. All we
need is just the entry, so let's add APIs to retrieve MSIX address and
data:
uint64_t msix_get_address(dev, vector)
uint32_t msix_get_data(dev, vector)
and that will be enough for KVM.
> ---
> hw/msix.c | 27 ---------------------------
> hw/msix.h | 26 ++++++++++++++++++++++++++
> 2 files changed, 26 insertions(+), 27 deletions(-)
>
> diff --git a/hw/msix.c b/hw/msix.c
> index d762870..dda1a24 100644
> --- a/hw/msix.c
> +++ b/hw/msix.c
> @@ -16,33 +16,6 @@
> #include "pci.h"
> #include "kvm.h"
>
> -/* MSI-X capability structure */
> -#define MSIX_TABLE_OFFSET 4
> -#define MSIX_PBA_OFFSET 8
> -#define MSIX_CAP_LENGTH 12
> -
> -/* MSI enable bit and maskall bit are in byte 1 in FLAGS register */
> -#define MSIX_CONTROL_OFFSET (PCI_MSIX_FLAGS + 1)
> -#define MSIX_ENABLE_MASK (PCI_MSIX_FLAGS_ENABLE >> 8)
> -#define MSIX_MASKALL_MASK (PCI_MSIX_FLAGS_MASKALL >> 8)
> -
> -/* MSI-X table format */
> -#define MSIX_MSG_ADDR 0
> -#define MSIX_MSG_UPPER_ADDR 4
> -#define MSIX_MSG_DATA 8
> -#define MSIX_VECTOR_CTRL 12
> -#define MSIX_ENTRY_SIZE 16
> -#define MSIX_VECTOR_MASK 0x1
> -
> -/* How much space does an MSIX table need. */
> -/* The spec requires giving the table structure
> - * a 4K aligned region all by itself. */
> -#define MSIX_PAGE_SIZE 0x1000
> -/* Reserve second half of the page for pending bits */
> -#define MSIX_PAGE_PENDING (MSIX_PAGE_SIZE / 2)
> -#define MSIX_MAX_ENTRIES 32
> -
> -
> /* Flag for interrupt controller to declare MSI-X support */
> int msix_supported;
>
> diff --git a/hw/msix.h b/hw/msix.h
> index 6b21ffb..5c84b3e 100644
> --- a/hw/msix.h
> +++ b/hw/msix.h
> @@ -4,6 +4,32 @@
> #include "qemu-common.h"
> #include "pci.h"
>
> +/* MSI-X capability structure */
> +#define MSIX_TABLE_OFFSET 4
> +#define MSIX_PBA_OFFSET 8
> +#define MSIX_CAP_LENGTH 12
> +
> +/* MSI enable bit and maskall bit are in byte 1 in FLAGS register */
> +#define MSIX_CONTROL_OFFSET (PCI_MSIX_FLAGS + 1)
> +#define MSIX_ENABLE_MASK (PCI_MSIX_FLAGS_ENABLE >> 8)
> +#define MSIX_MASKALL_MASK (PCI_MSIX_FLAGS_MASKALL >> 8)
> +
> +/* MSI-X table format */
> +#define MSIX_MSG_ADDR 0
> +#define MSIX_MSG_UPPER_ADDR 4
> +#define MSIX_MSG_DATA 8
> +#define MSIX_VECTOR_CTRL 12
> +#define MSIX_ENTRY_SIZE 16
> +#define MSIX_VECTOR_MASK 0x1
> +
> +/* How much space does an MSIX table need. */
> +/* The spec requires giving the table structure
> + * a 4K aligned region all by itself. */
> +#define MSIX_PAGE_SIZE 0x1000
> +/* Reserve second half of the page for pending bits */
> +#define MSIX_PAGE_PENDING (MSIX_PAGE_SIZE / 2)
> +#define MSIX_MAX_ENTRIES 32
> +
> int msix_init(PCIDevice *pdev, unsigned short nentries,
> unsigned bar_nr, unsigned bar_size);
>
> --
> 1.7.2.3
^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH 4/9] msix: move definitions from msix.c to msix.h
2010-09-20 16:56 ` Michael S. Tsirkin
@ 2010-09-21 16:03 ` Avi Kivity
0 siblings, 0 replies; 19+ messages in thread
From: Avi Kivity @ 2010-09-21 16:03 UTC (permalink / raw)
To: Michael S. Tsirkin; +Cc: Marcelo Tosatti, kvm
On 09/20/2010 06:56 PM, Michael S. Tsirkin wrote:
> On Mon, Sep 20, 2010 at 05:06:45PM +0200, Avi Kivity wrote:
> > This allows us to reuse them from the kvm support code.
> >
> > Signed-off-by: Avi Kivity<avi@redhat.com>
>
> I would rather all dealings with MSI-X table stayed in one place. All we
> need is just the entry, so let's add APIs to retrieve MSIX address and
> data:
>
> uint64_t msix_get_address(dev, vector)
> uint32_t msix_get_data(dev, vector)
>
> and that will be enough for KVM.
>
Ok, will do.
--
error compiling committee.c: too many arguments to function
^ permalink raw reply [flat|nested] 19+ messages in thread
* [PATCH 5/9] Add missing #include to hw/irq.h
2010-09-20 15:06 [PATCH 0/9] msix/kvm integration cleanups Avi Kivity
` (3 preceding siblings ...)
2010-09-20 15:06 ` [PATCH 4/9] msix: move definitions from msix.c to msix.h Avi Kivity
@ 2010-09-20 15:06 ` Avi Kivity
2010-09-20 15:06 ` [PATCH 6/9] msix: move kvm support functions to kvm-all.c and kvm-stub.c Avi Kivity
` (4 subsequent siblings)
9 siblings, 0 replies; 19+ messages in thread
From: Avi Kivity @ 2010-09-20 15:06 UTC (permalink / raw)
To: Marcelo Tosatti, kvm, Michael S. Tsirkin
Needs qemu-common.h for qemu_irq.
Signed-off-by: Avi Kivity <avi@redhat.com>
---
hw/irq.h | 2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/hw/irq.h b/hw/irq.h
index 5daae44..e9f0542 100644
--- a/hw/irq.h
+++ b/hw/irq.h
@@ -1,6 +1,8 @@
#ifndef QEMU_IRQ_H
#define QEMU_IRQ_H
+#include "qemu-common.h"
+
/* Generic IRQ/GPIO pin infrastructure. */
/* FIXME: Rmove one of these. */
--
1.7.2.3
^ permalink raw reply related [flat|nested] 19+ messages in thread* [PATCH 6/9] msix: move kvm support functions to kvm-all.c and kvm-stub.c
2010-09-20 15:06 [PATCH 0/9] msix/kvm integration cleanups Avi Kivity
` (4 preceding siblings ...)
2010-09-20 15:06 ` [PATCH 5/9] Add missing #include to hw/irq.h Avi Kivity
@ 2010-09-20 15:06 ` Avi Kivity
2010-09-20 15:06 ` [PATCH 7/9] msix: Move kvm_enabled() guards to kvm-all.c functions Avi Kivity
` (3 subsequent siblings)
9 siblings, 0 replies; 19+ messages in thread
From: Avi Kivity @ 2010-09-20 15:06 UTC (permalink / raw)
To: Marcelo Tosatti, kvm, Michael S. Tsirkin
This removes the dependency on CONFIG_KVM.
Signed-off-by: Avi Kivity <avi@redhat.com>
---
hw/msix.c | 137 ------------------------------------------------------------
kvm-all.c | 124 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
kvm-stub.c | 28 ++++++++++++
kvm.h | 16 +++++++
4 files changed, 168 insertions(+), 137 deletions(-)
diff --git a/hw/msix.c b/hw/msix.c
index dda1a24..bdec0b3 100644
--- a/hw/msix.c
+++ b/hw/msix.c
@@ -19,143 +19,6 @@
/* Flag for interrupt controller to declare MSI-X support */
int msix_supported;
-#ifdef CONFIG_KVM
-/* KVM specific MSIX helpers */
-
-static void kvm_msix_init(PCIDevice *dev)
-{
-#ifdef KVM_CAP_IRQCHIP
- if (kvm_enabled() && kvm_irqchip_in_kernel()) {
- dev->msix_irq_entries = qemu_malloc(dev->msix_entries_nr *
- sizeof *dev->msix_irq_entries);
- }
-#endif
-}
-
-static void kvm_msix_free(PCIDevice *dev)
-{
- int vector, changed = 0;
- for (vector = 0; vector < dev->msix_entries_nr; ++vector) {
- if (dev->msix_entry_used[vector]) {
- kvm_del_routing_entry(kvm_context, &dev->msix_irq_entries[vector]);
- changed = 1;
- }
- }
- if (changed) {
- kvm_commit_irq_routes(kvm_context);
- }
-}
-
-static void kvm_msix_routing_entry(PCIDevice *dev, unsigned vector,
- struct kvm_irq_routing_entry *entry)
-{
- uint8_t *table_entry = dev->msix_table_page + vector * MSIX_ENTRY_SIZE;
- entry->type = KVM_IRQ_ROUTING_MSI;
- entry->flags = 0;
- entry->u.msi.address_lo = pci_get_long(table_entry + MSIX_MSG_ADDR);
- entry->u.msi.address_hi = pci_get_long(table_entry + MSIX_MSG_UPPER_ADDR);
- entry->u.msi.data = pci_get_long(table_entry + MSIX_MSG_DATA);
-}
-
-static void kvm_msix_update(PCIDevice *dev, int vector,
- int was_masked, int is_masked)
-{
- struct kvm_irq_routing_entry e = {}, *entry;
- int mask_cleared = was_masked && !is_masked;
- /* It is only legal to change an entry when it is masked. Therefore, it is
- * enough to update the routing in kernel when mask is being cleared. */
- if (!mask_cleared) {
- return;
- }
- if (!dev->msix_entry_used[vector]) {
- return;
- }
- entry = dev->msix_irq_entries + vector;
- e.gsi = entry->gsi;
- kvm_msix_routing_entry(dev, vector, &e);
- if (memcmp(&entry->u.msi, &e.u.msi, sizeof entry->u.msi)) {
- int r;
- r = kvm_update_routing_entry(kvm_context, entry, &e);
- if (r) {
- fprintf(stderr, "%s: kvm_update_routing_entry failed: %s\n", __func__,
- strerror(-r));
- exit(1);
- }
- memcpy(&entry->u.msi, &e.u.msi, sizeof entry->u.msi);
- r = kvm_commit_irq_routes(kvm_context);
- if (r) {
- fprintf(stderr, "%s: kvm_commit_irq_routes failed: %s\n", __func__,
- strerror(-r));
- exit(1);
- }
- }
-}
-
-static int kvm_msix_add(PCIDevice *dev, unsigned vector)
-{
- struct kvm_irq_routing_entry *entry = dev->msix_irq_entries + vector;
- int r;
-
- if (!kvm_has_gsi_routing(kvm_context)) {
- fprintf(stderr, "Warning: no MSI-X support found. "
- "At least kernel 2.6.30 is required for MSI-X support.\n"
- );
- return -EOPNOTSUPP;
- }
-
- r = kvm_get_irq_route_gsi(kvm_context);
- if (r < 0) {
- fprintf(stderr, "%s: kvm_get_irq_route_gsi failed: %s\n", __func__, strerror(-r));
- return r;
- }
- entry->gsi = r;
- kvm_msix_routing_entry(dev, vector, entry);
- r = kvm_add_routing_entry(kvm_context, entry);
- if (r < 0) {
- fprintf(stderr, "%s: kvm_add_routing_entry failed: %s\n", __func__, strerror(-r));
- return r;
- }
-
- r = kvm_commit_irq_routes(kvm_context);
- if (r < 0) {
- fprintf(stderr, "%s: kvm_commit_irq_routes failed: %s\n", __func__, strerror(-r));
- return r;
- }
- return 0;
-}
-
-static void kvm_msix_del(PCIDevice *dev, unsigned vector)
-{
- if (dev->msix_entry_used[vector]) {
- return;
- }
- kvm_del_routing_entry(kvm_context, &dev->msix_irq_entries[vector]);
- kvm_commit_irq_routes(kvm_context);
-}
-
-static bool kvm_msix_notify(PCIDevice *dev, unsigned vector)
-{
-#ifdef KVM_CAP_IRQCHIP
- if (kvm_enabled() && kvm_irqchip_in_kernel()) {
- kvm_set_irq(dev->msix_irq_entries[vector].gsi, 1, NULL);
- return true;
- }
-#endif
- return false;
-}
-
-#else
-
-static void kvm_msix_init(PCIDevice *dev) {}
-static void kvm_msix_free(PCIDevice *dev) {}
-static void kvm_msix_update(PCIDevice *dev, int vector,
- int was_masked, int is_masked) {}
-static int kvm_msix_add(PCIDevice *dev, unsigned vector) { return -1; }
-static void kvm_msix_del(PCIDevice *dev, unsigned vector) {}
-static bool kvm_msix_notify(PCIDevice *dev, unsigned vector) { return false; }
-
-#endif
-
/* 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.
diff --git a/kvm-all.c b/kvm-all.c
index d4b0861..0860a16 100644
--- a/kvm-all.c
+++ b/kvm-all.c
@@ -24,6 +24,8 @@
#include "qemu-barrier.h"
#include "sysemu.h"
#include "hw/hw.h"
+#include "hw/pci.h"
+#include "hw/msix.h"
#include "gdbstub.h"
#include "kvm.h"
#include "bswap.h"
@@ -1338,5 +1340,127 @@ int kvm_set_irqfd(int gsi, int fd, bool assigned)
}
#endif
+void kvm_msix_init(PCIDevice *dev)
+{
+#ifdef KVM_CAP_IRQCHIP
+ if (kvm_enabled() && kvm_irqchip_in_kernel()) {
+ dev->msix_irq_entries = qemu_malloc(dev->msix_entries_nr *
+ sizeof *dev->msix_irq_entries);
+ }
+#endif
+}
+
+void kvm_msix_free(PCIDevice *dev)
+{
+ int vector, changed = 0;
+ for (vector = 0; vector < dev->msix_entries_nr; ++vector) {
+ if (dev->msix_entry_used[vector]) {
+ kvm_del_routing_entry(kvm_context, &dev->msix_irq_entries[vector]);
+ changed = 1;
+ }
+ }
+ if (changed) {
+ kvm_commit_irq_routes(kvm_context);
+ }
+}
+
+static void kvm_msix_routing_entry(PCIDevice *dev, unsigned vector,
+ struct kvm_irq_routing_entry *entry)
+{
+ uint8_t *table_entry = dev->msix_table_page + vector * MSIX_ENTRY_SIZE;
+ entry->type = KVM_IRQ_ROUTING_MSI;
+ entry->flags = 0;
+ entry->u.msi.address_lo = pci_get_long(table_entry + MSIX_MSG_ADDR);
+ entry->u.msi.address_hi = pci_get_long(table_entry + MSIX_MSG_UPPER_ADDR);
+ entry->u.msi.data = pci_get_long(table_entry + MSIX_MSG_DATA);
+}
+
+void kvm_msix_update(PCIDevice *dev, int vector,
+ int was_masked, int is_masked)
+{
+ struct kvm_irq_routing_entry e = {}, *entry;
+ int mask_cleared = was_masked && !is_masked;
+ /* It is only legal to change an entry when it is masked. Therefore, it is
+ * enough to update the routing in kernel when mask is being cleared. */
+ if (!mask_cleared) {
+ return;
+ }
+ if (!dev->msix_entry_used[vector]) {
+ return;
+ }
+ entry = dev->msix_irq_entries + vector;
+ e.gsi = entry->gsi;
+ kvm_msix_routing_entry(dev, vector, &e);
+ if (memcmp(&entry->u.msi, &e.u.msi, sizeof entry->u.msi)) {
+ int r;
+ r = kvm_update_routing_entry(kvm_context, entry, &e);
+ if (r) {
+ fprintf(stderr, "%s: kvm_update_routing_entry failed: %s\n", __func__,
+ strerror(-r));
+ exit(1);
+ }
+ memcpy(&entry->u.msi, &e.u.msi, sizeof entry->u.msi);
+ r = kvm_commit_irq_routes(kvm_context);
+ if (r) {
+ fprintf(stderr, "%s: kvm_commit_irq_routes failed: %s\n", __func__,
+ strerror(-r));
+ exit(1);
+ }
+ }
+}
+
+int kvm_msix_add(PCIDevice *dev, unsigned vector)
+{
+ struct kvm_irq_routing_entry *entry = dev->msix_irq_entries + vector;
+ int r;
+
+ if (!kvm_has_gsi_routing(kvm_context)) {
+ fprintf(stderr, "Warning: no MSI-X support found. "
+ "At least kernel 2.6.30 is required for MSI-X support.\n"
+ );
+ return -EOPNOTSUPP;
+ }
+
+ r = kvm_get_irq_route_gsi(kvm_context);
+ if (r < 0) {
+ fprintf(stderr, "%s: kvm_get_irq_route_gsi failed: %s\n", __func__, strerror(-r));
+ return r;
+ }
+ entry->gsi = r;
+ kvm_msix_routing_entry(dev, vector, entry);
+ r = kvm_add_routing_entry(kvm_context, entry);
+ if (r < 0) {
+ fprintf(stderr, "%s: kvm_add_routing_entry failed: %s\n", __func__, strerror(-r));
+ return r;
+ }
+
+ r = kvm_commit_irq_routes(kvm_context);
+ if (r < 0) {
+ fprintf(stderr, "%s: kvm_commit_irq_routes failed: %s\n", __func__, strerror(-r));
+ return r;
+ }
+ return 0;
+}
+
+void kvm_msix_del(PCIDevice *dev, unsigned vector)
+{
+ if (dev->msix_entry_used[vector]) {
+ return;
+ }
+ kvm_del_routing_entry(kvm_context, &dev->msix_irq_entries[vector]);
+ kvm_commit_irq_routes(kvm_context);
+}
+
+bool kvm_msix_notify(PCIDevice *dev, unsigned vector)
+{
+#ifdef KVM_CAP_IRQCHIP
+ if (kvm_enabled() && kvm_irqchip_in_kernel()) {
+ kvm_set_irq(dev->msix_irq_entries[vector].gsi, 1, NULL);
+ return true;
+ }
+#endif
+ return false;
+}
+
#undef PAGE_SIZE
#include "qemu-kvm.c"
diff --git a/kvm-stub.c b/kvm-stub.c
index d45f9fa..1c54020 100644
--- a/kvm-stub.c
+++ b/kvm-stub.c
@@ -13,6 +13,7 @@
#include "qemu-common.h"
#include "sysemu.h"
#include "hw/hw.h"
+#include "hw/pci.h"
#include "exec-all.h"
#include "gdbstub.h"
#include "kvm.h"
@@ -141,3 +142,30 @@ int kvm_set_ioeventfd_mmio_long(int fd, uint32_t adr, uint32_t val, bool assign)
{
return -ENOSYS;
}
+
+void kvm_msix_init(PCIDevice *dev)
+{
+}
+
+void kvm_msix_free(PCIDevice *dev)
+{
+}
+
+void kvm_msix_update(PCIDevice *dev, int vector,
+ int was_masked, int is_masked)
+{
+}
+
+int kvm_msix_add(PCIDevice *dev, unsigned vector)
+{
+ return -1;
+}
+
+void kvm_msix_del(PCIDevice *dev, unsigned vector)
+{
+}
+
+bool kvm_msix_notify(PCIDevice *dev, unsigned vector)
+{
+ return false;
+}
diff --git a/kvm.h b/kvm.h
index 56236ae..62dc039 100644
--- a/kvm.h
+++ b/kvm.h
@@ -18,6 +18,9 @@
#include "config-host.h"
#include "qemu-queue.h"
#include "qemu-kvm.h"
+#ifndef __DYNGEN_EXEC_H__
+#include "hw/pci.h"
+#endif
#ifdef CONFIG_KVM
#include <linux/kvm.h>
@@ -194,5 +197,18 @@ int kvm_set_irqfd(int gsi, int fd, bool assigned)
}
#endif
+#ifndef __DYNGEN_EXEC_H__
+
int kvm_set_ioeventfd_pio_word(int fd, uint16_t adr, uint16_t val, bool assign);
+
+void kvm_msix_init(PCIDevice *dev);
+void kvm_msix_free(PCIDevice *dev);
+void kvm_msix_update(PCIDevice *dev, int vector,
+ int was_masked, int is_masked);
+int kvm_msix_add(PCIDevice *dev, unsigned vector);
+void kvm_msix_del(PCIDevice *dev, unsigned vector);
+bool kvm_msix_notify(PCIDevice *dev, unsigned vector);
+
+#endif
+
#endif
--
1.7.2.3
^ permalink raw reply related [flat|nested] 19+ messages in thread* [PATCH 7/9] msix: Move kvm_enabled() guards to kvm-all.c functions
2010-09-20 15:06 [PATCH 0/9] msix/kvm integration cleanups Avi Kivity
` (5 preceding siblings ...)
2010-09-20 15:06 ` [PATCH 6/9] msix: move kvm support functions to kvm-all.c and kvm-stub.c Avi Kivity
@ 2010-09-20 15:06 ` Avi Kivity
2010-09-20 15:06 ` [PATCH 8/9] Protect qemu-kvm.h declarations with NEED_CPU_H Avi Kivity
` (2 subsequent siblings)
9 siblings, 0 replies; 19+ messages in thread
From: Avi Kivity @ 2010-09-20 15:06 UTC (permalink / raw)
To: Marcelo Tosatti, kvm, Michael S. Tsirkin
Simplifies the msix code.
Signed-off-by: Avi Kivity <avi@redhat.com>
---
hw/msix.c | 22 ++++++++--------------
kvm-all.c | 18 ++++++++++++++++++
kvm-stub.c | 2 +-
3 files changed, 27 insertions(+), 15 deletions(-)
diff --git a/hw/msix.c b/hw/msix.c
index bdec0b3..fc4cbb1 100644
--- a/hw/msix.c
+++ b/hw/msix.c
@@ -159,9 +159,7 @@ static void msix_mmio_writel(void *opaque, target_phys_addr_t addr,
int vector = offset / MSIX_ENTRY_SIZE;
int was_masked = msix_is_masked(dev, vector);
pci_set_long(dev->msix_table_page + offset, val);
- if (kvm_enabled() && kvm_irqchip_in_kernel()) {
- kvm_msix_update(dev, vector, was_masked, msix_is_masked(dev, vector));
- }
+ kvm_msix_update(dev, vector, was_masked, msix_is_masked(dev, vector));
if (was_masked != msix_is_masked(dev, vector) &&
dev->msix_mask_notifier && dev->msix_mask_notifier_opaque[vector]) {
int r = dev->msix_mask_notifier(dev, vector,
@@ -276,9 +274,7 @@ static void msix_free_irq_entries(PCIDevice *dev)
{
int vector;
- if (kvm_enabled() && kvm_irqchip_in_kernel()) {
- kvm_msix_free(dev);
- }
+ kvm_msix_free(dev);
for (vector = 0; vector < dev->msix_entries_nr; ++vector) {
dev->msix_entry_used[vector] = 0;
@@ -413,12 +409,12 @@ int msix_vector_use(PCIDevice *dev, unsigned vector)
if (dev->msix_entry_used[vector]) {
return 0;
}
- if (kvm_enabled() && kvm_irqchip_in_kernel()) {
- ret = kvm_msix_add(dev, vector);
- if (ret) {
- return ret;
- }
+
+ ret = kvm_msix_add(dev, vector);
+ if (ret) {
+ return ret;
}
+
++dev->msix_entry_used[vector];
return 0;
}
@@ -432,9 +428,7 @@ void msix_vector_unuse(PCIDevice *dev, unsigned vector)
if (--dev->msix_entry_used[vector]) {
return;
}
- if (kvm_enabled() && kvm_irqchip_in_kernel()) {
- kvm_msix_del(dev, vector);
- }
+ kvm_msix_del(dev, vector);
msix_clr_pending(dev, vector);
}
diff --git a/kvm-all.c b/kvm-all.c
index 0860a16..a5e7545 100644
--- a/kvm-all.c
+++ b/kvm-all.c
@@ -1353,6 +1353,11 @@ void kvm_msix_init(PCIDevice *dev)
void kvm_msix_free(PCIDevice *dev)
{
int vector, changed = 0;
+
+ if (!(kvm_enabled() && kvm_irqchip_in_kernel())) {
+ return;
+ }
+
for (vector = 0; vector < dev->msix_entries_nr; ++vector) {
if (dev->msix_entry_used[vector]) {
kvm_del_routing_entry(kvm_context, &dev->msix_irq_entries[vector]);
@@ -1380,6 +1385,11 @@ void kvm_msix_update(PCIDevice *dev, int vector,
{
struct kvm_irq_routing_entry e = {}, *entry;
int mask_cleared = was_masked && !is_masked;
+
+ if (!(kvm_enabled() && kvm_irqchip_in_kernel())) {
+ return;
+ }
+
/* It is only legal to change an entry when it is masked. Therefore, it is
* enough to update the routing in kernel when mask is being cleared. */
if (!mask_cleared) {
@@ -1414,6 +1424,10 @@ int kvm_msix_add(PCIDevice *dev, unsigned vector)
struct kvm_irq_routing_entry *entry = dev->msix_irq_entries + vector;
int r;
+ if (!(kvm_enabled() && kvm_irqchip_in_kernel())) {
+ return 0;
+ }
+
if (!kvm_has_gsi_routing(kvm_context)) {
fprintf(stderr, "Warning: no MSI-X support found. "
"At least kernel 2.6.30 is required for MSI-X support.\n"
@@ -1444,6 +1458,10 @@ int kvm_msix_add(PCIDevice *dev, unsigned vector)
void kvm_msix_del(PCIDevice *dev, unsigned vector)
{
+ if (!(kvm_enabled() && kvm_irqchip_in_kernel())) {
+ return;
+ }
+
if (dev->msix_entry_used[vector]) {
return;
}
diff --git a/kvm-stub.c b/kvm-stub.c
index 1c54020..37d2b7a 100644
--- a/kvm-stub.c
+++ b/kvm-stub.c
@@ -158,7 +158,7 @@ void kvm_msix_update(PCIDevice *dev, int vector,
int kvm_msix_add(PCIDevice *dev, unsigned vector)
{
- return -1;
+ return 0;
}
void kvm_msix_del(PCIDevice *dev, unsigned vector)
--
1.7.2.3
^ permalink raw reply related [flat|nested] 19+ messages in thread* [PATCH 8/9] Protect qemu-kvm.h declarations with NEED_CPU_H
2010-09-20 15:06 [PATCH 0/9] msix/kvm integration cleanups Avi Kivity
` (6 preceding siblings ...)
2010-09-20 15:06 ` [PATCH 7/9] msix: Move kvm_enabled() guards to kvm-all.c functions Avi Kivity
@ 2010-09-20 15:06 ` Avi Kivity
2010-09-20 17:05 ` Michael S. Tsirkin
2010-09-20 15:06 ` [PATCH 9/9] Move msix.o build back to Makefile.objs Avi Kivity
2010-09-20 17:02 ` [PATCH 0/9] msix/kvm integration cleanups Michael S. Tsirkin
9 siblings, 1 reply; 19+ messages in thread
From: Avi Kivity @ 2010-09-20 15:06 UTC (permalink / raw)
To: Marcelo Tosatti, kvm, Michael S. Tsirkin
Target-specific definitions need to be qualified with NEED_CPU_H so kvm.h
can be included from non-target-specific files.
Signed-off-by: Avi Kivity <avi@redhat.com>
---
kvm-stub.c | 1 +
qemu-kvm.h | 21 ++++++++++++++++++++-
2 files changed, 21 insertions(+), 1 deletions(-)
diff --git a/kvm-stub.c b/kvm-stub.c
index 37d2b7a..2e4bf00 100644
--- a/kvm-stub.c
+++ b/kvm-stub.c
@@ -169,3 +169,4 @@ bool kvm_msix_notify(PCIDevice *dev, unsigned vector)
{
return false;
}
+
diff --git a/qemu-kvm.h b/qemu-kvm.h
index 9809574..66ae18f 100644
--- a/qemu-kvm.h
+++ b/qemu-kvm.h
@@ -8,7 +8,9 @@
#ifndef THE_ORIGINAL_AND_TRUE_QEMU_KVM_H
#define THE_ORIGINAL_AND_TRUE_QEMU_KVM_H
+#ifdef NEED_CPU_H
#include "cpu.h"
+#endif
#include <signal.h>
#include <stdlib.h>
@@ -794,6 +796,8 @@ typedef struct kvm_vcpu_context *kvm_vcpu_context_t;
struct kvm_pit_state {
};
+#ifdef NEED_CPU_H
+
static inline void kvm_inject_x86_mce(CPUState *cenv, int bank,
uint64_t status, uint64_t mcg_status,
uint64_t addr, uint64_t misc,
@@ -803,8 +807,9 @@ static inline void kvm_inject_x86_mce(CPUState *cenv, int bank,
abort();
}
-#endif /* !CONFIG_KVM */
+#endif
+#endif /* !CONFIG_KVM */
/*!
* \brief Create new KVM context
@@ -820,9 +825,11 @@ int kvm_init(int smp_cpus);
int kvm_main_loop(void);
int kvm_init_ap(void);
+#ifdef NEED_CPU_H
int kvm_vcpu_inited(CPUState *env);
void kvm_save_lapic(CPUState *env);
void kvm_load_lapic(CPUState *env);
+#endif
void kvm_hpet_enable_kpit(void);
void kvm_hpet_disable_kpit(void);
@@ -830,11 +837,13 @@ int kvm_set_irq(int irq, int level, int *status);
int kvm_physical_memory_set_dirty_tracking(int enable);
+#ifdef NEED_CPU_H
void qemu_kvm_call_with_env(void (*func)(void *), void *data, CPUState *env);
void qemu_kvm_cpuid_on_env(CPUState *env);
void kvm_inject_interrupt(CPUState *env, int mask);
void kvm_update_after_sipi(CPUState *env);
void kvm_update_interrupt_request(CPUState *env);
+#endif
#ifndef CONFIG_USER_ONLY
void *kvm_cpu_create_phys_mem(target_phys_addr_t start_addr, unsigned long size,
int log, int writable);
@@ -850,6 +859,7 @@ int kvm_qemu_destroy_memory_alias(uint64_t phys_start);
int kvm_arch_qemu_create_context(void);
+#ifdef NEED_CPU_H
void kvm_arch_save_regs(CPUState *env);
void kvm_arch_load_regs(CPUState *env, int level);
int kvm_arch_has_work(CPUState *env);
@@ -858,12 +868,15 @@ int kvm_arch_try_push_interrupts(void *opaque);
void kvm_arch_push_nmi(void *opaque);
void kvm_arch_cpu_reset(CPUState *env);
int kvm_set_boot_cpu_id(uint32_t id);
+#endif
void qemu_kvm_aio_wait_start(void);
void qemu_kvm_aio_wait(void);
void qemu_kvm_aio_wait_end(void);
+#ifdef NEED_CPU_H
void kvm_tpr_access_report(CPUState *env, uint64_t rip, int is_write);
+#endif
int kvm_arch_init_irq_routing(void);
@@ -873,15 +886,20 @@ int kvm_mmio_write(void *opaque, uint64_t addr, uint8_t * data, int len);
#ifdef CONFIG_KVM_DEVICE_ASSIGNMENT
struct ioperm_data;
+#ifdef NEED_CPU_H
void kvm_ioperm(CPUState *env, void *data);
void kvm_add_ioperm_data(struct ioperm_data *data);
void kvm_remove_ioperm_data(unsigned long start_port, unsigned long num);
void kvm_arch_do_ioperm(void *_data);
#endif
+#endif
#define ALIGN(x, y) (((x)+(y)-1) & ~((y)-1))
+#ifdef NEED_CPU_H
#define BITMAP_SIZE(m) (ALIGN(((m)>>TARGET_PAGE_BITS), HOST_LONG_BITS) / 8)
+#endif
+#ifdef NEED_CPU_H
#ifdef CONFIG_KVM
#include "qemu-queue.h"
@@ -915,6 +933,7 @@ int handle_tpr_access(void *opaque, CPUState *env, uint64_t rip,
#endif
#define qemu_kvm_cpu_stop(env) do {} while(0)
#endif
+#endif
#ifdef CONFIG_KVM
--
1.7.2.3
^ permalink raw reply related [flat|nested] 19+ messages in thread* Re: [PATCH 8/9] Protect qemu-kvm.h declarations with NEED_CPU_H
2010-09-20 15:06 ` [PATCH 8/9] Protect qemu-kvm.h declarations with NEED_CPU_H Avi Kivity
@ 2010-09-20 17:05 ` Michael S. Tsirkin
2010-09-21 16:03 ` Avi Kivity
0 siblings, 1 reply; 19+ messages in thread
From: Michael S. Tsirkin @ 2010-09-20 17:05 UTC (permalink / raw)
To: Avi Kivity; +Cc: Marcelo Tosatti, kvm
On Mon, Sep 20, 2010 at 05:06:49PM +0200, Avi Kivity wrote:
> Target-specific definitions need to be qualified with NEED_CPU_H so kvm.h
> can be included from non-target-specific files.
>
> Signed-off-by: Avi Kivity <avi@redhat.com>
Long term, would be cleaner to split this into two files ...
> ---
> kvm-stub.c | 1 +
> qemu-kvm.h | 21 ++++++++++++++++++++-
> 2 files changed, 21 insertions(+), 1 deletions(-)
>
> diff --git a/kvm-stub.c b/kvm-stub.c
> index 37d2b7a..2e4bf00 100644
> --- a/kvm-stub.c
> +++ b/kvm-stub.c
> @@ -169,3 +169,4 @@ bool kvm_msix_notify(PCIDevice *dev, unsigned vector)
> {
> return false;
> }
> +
intentional?
> diff --git a/qemu-kvm.h b/qemu-kvm.h
> index 9809574..66ae18f 100644
> --- a/qemu-kvm.h
> +++ b/qemu-kvm.h
> @@ -8,7 +8,9 @@
> #ifndef THE_ORIGINAL_AND_TRUE_QEMU_KVM_H
> #define THE_ORIGINAL_AND_TRUE_QEMU_KVM_H
>
> +#ifdef NEED_CPU_H
> #include "cpu.h"
> +#endif
>
> #include <signal.h>
> #include <stdlib.h>
> @@ -794,6 +796,8 @@ typedef struct kvm_vcpu_context *kvm_vcpu_context_t;
> struct kvm_pit_state {
> };
>
> +#ifdef NEED_CPU_H
> +
> static inline void kvm_inject_x86_mce(CPUState *cenv, int bank,
> uint64_t status, uint64_t mcg_status,
> uint64_t addr, uint64_t misc,
> @@ -803,8 +807,9 @@ static inline void kvm_inject_x86_mce(CPUState *cenv, int bank,
> abort();
> }
>
> -#endif /* !CONFIG_KVM */
> +#endif
>
> +#endif /* !CONFIG_KVM */
>
> /*!
> * \brief Create new KVM context
> @@ -820,9 +825,11 @@ int kvm_init(int smp_cpus);
>
> int kvm_main_loop(void);
> int kvm_init_ap(void);
> +#ifdef NEED_CPU_H
> int kvm_vcpu_inited(CPUState *env);
> void kvm_save_lapic(CPUState *env);
> void kvm_load_lapic(CPUState *env);
> +#endif
>
> void kvm_hpet_enable_kpit(void);
> void kvm_hpet_disable_kpit(void);
> @@ -830,11 +837,13 @@ int kvm_set_irq(int irq, int level, int *status);
>
> int kvm_physical_memory_set_dirty_tracking(int enable);
>
> +#ifdef NEED_CPU_H
> void qemu_kvm_call_with_env(void (*func)(void *), void *data, CPUState *env);
> void qemu_kvm_cpuid_on_env(CPUState *env);
> void kvm_inject_interrupt(CPUState *env, int mask);
> void kvm_update_after_sipi(CPUState *env);
> void kvm_update_interrupt_request(CPUState *env);
> +#endif
> #ifndef CONFIG_USER_ONLY
> void *kvm_cpu_create_phys_mem(target_phys_addr_t start_addr, unsigned long size,
> int log, int writable);
> @@ -850,6 +859,7 @@ int kvm_qemu_destroy_memory_alias(uint64_t phys_start);
>
> int kvm_arch_qemu_create_context(void);
>
> +#ifdef NEED_CPU_H
> void kvm_arch_save_regs(CPUState *env);
> void kvm_arch_load_regs(CPUState *env, int level);
> int kvm_arch_has_work(CPUState *env);
> @@ -858,12 +868,15 @@ int kvm_arch_try_push_interrupts(void *opaque);
> void kvm_arch_push_nmi(void *opaque);
> void kvm_arch_cpu_reset(CPUState *env);
> int kvm_set_boot_cpu_id(uint32_t id);
> +#endif
>
> void qemu_kvm_aio_wait_start(void);
> void qemu_kvm_aio_wait(void);
> void qemu_kvm_aio_wait_end(void);
>
> +#ifdef NEED_CPU_H
> void kvm_tpr_access_report(CPUState *env, uint64_t rip, int is_write);
> +#endif
>
> int kvm_arch_init_irq_routing(void);
>
> @@ -873,15 +886,20 @@ int kvm_mmio_write(void *opaque, uint64_t addr, uint8_t * data, int len);
> #ifdef CONFIG_KVM_DEVICE_ASSIGNMENT
> struct ioperm_data;
>
> +#ifdef NEED_CPU_H
> void kvm_ioperm(CPUState *env, void *data);
> void kvm_add_ioperm_data(struct ioperm_data *data);
> void kvm_remove_ioperm_data(unsigned long start_port, unsigned long num);
> void kvm_arch_do_ioperm(void *_data);
> #endif
> +#endif
>
> #define ALIGN(x, y) (((x)+(y)-1) & ~((y)-1))
> +#ifdef NEED_CPU_H
> #define BITMAP_SIZE(m) (ALIGN(((m)>>TARGET_PAGE_BITS), HOST_LONG_BITS) / 8)
> +#endif
>
> +#ifdef NEED_CPU_H
> #ifdef CONFIG_KVM
> #include "qemu-queue.h"
>
> @@ -915,6 +933,7 @@ int handle_tpr_access(void *opaque, CPUState *env, uint64_t rip,
> #endif
> #define qemu_kvm_cpu_stop(env) do {} while(0)
> #endif
> +#endif
>
> #ifdef CONFIG_KVM
>
> --
> 1.7.2.3
^ permalink raw reply [flat|nested] 19+ messages in thread* Re: [PATCH 8/9] Protect qemu-kvm.h declarations with NEED_CPU_H
2010-09-20 17:05 ` Michael S. Tsirkin
@ 2010-09-21 16:03 ` Avi Kivity
0 siblings, 0 replies; 19+ messages in thread
From: Avi Kivity @ 2010-09-21 16:03 UTC (permalink / raw)
To: Michael S. Tsirkin; +Cc: Marcelo Tosatti, kvm
On 09/20/2010 07:05 PM, Michael S. Tsirkin wrote:
> On Mon, Sep 20, 2010 at 05:06:49PM +0200, Avi Kivity wrote:
> > Target-specific definitions need to be qualified with NEED_CPU_H so kvm.h
> > can be included from non-target-specific files.
> >
> > Signed-off-by: Avi Kivity<avi@redhat.com>
>
> Long term, would be cleaner to split this into two files ...
Yes, this is a pain to deal with.
> > ---
> > kvm-stub.c | 1 +
> > qemu-kvm.h | 21 ++++++++++++++++++++-
> > 2 files changed, 21 insertions(+), 1 deletions(-)
> >
> > diff --git a/kvm-stub.c b/kvm-stub.c
> > index 37d2b7a..2e4bf00 100644
> > --- a/kvm-stub.c
> > +++ b/kvm-stub.c
> > @@ -169,3 +169,4 @@ bool kvm_msix_notify(PCIDevice *dev, unsigned vector)
> > {
> > return false;
> > }
> > +
>
> intentional?
Nope.
--
error compiling committee.c: too many arguments to function
^ permalink raw reply [flat|nested] 19+ messages in thread
* [PATCH 9/9] Move msix.o build back to Makefile.objs
2010-09-20 15:06 [PATCH 0/9] msix/kvm integration cleanups Avi Kivity
` (7 preceding siblings ...)
2010-09-20 15:06 ` [PATCH 8/9] Protect qemu-kvm.h declarations with NEED_CPU_H Avi Kivity
@ 2010-09-20 15:06 ` Avi Kivity
2010-09-20 17:02 ` [PATCH 0/9] msix/kvm integration cleanups Michael S. Tsirkin
9 siblings, 0 replies; 19+ messages in thread
From: Avi Kivity @ 2010-09-20 15:06 UTC (permalink / raw)
To: Marcelo Tosatti, kvm, Michael S. Tsirkin
Now that hw/msix.c does not depend on CONFIG_KVM, remove it from
Makefile.target and put it back in Makefile.objs.
Signed-off-by: Avi Kivity <avi@redhat.com>
---
Makefile.objs | 4 +---
Makefile.target | 4 ----
2 files changed, 1 insertions(+), 7 deletions(-)
diff --git a/Makefile.objs b/Makefile.objs
index dbee210..44b2399 100644
--- a/Makefile.objs
+++ b/Makefile.objs
@@ -196,9 +196,7 @@ hw-obj-$(CONFIG_PIIX4) += piix4.o
# PCI watchdog devices
hw-obj-y += wdt_i6300esb.o
-# MSI-X depends on kvm for interrupt injection,
-# so moved it from Makefile.objs to Makefile.target for now
-# hw-obj-y += msix.o
+hw-obj-y += msix.o
# PCI network cards
hw-obj-y += ne2000.o
diff --git a/Makefile.target b/Makefile.target
index 5c1eec4..7c8f19c 100644
--- a/Makefile.target
+++ b/Makefile.target
@@ -183,10 +183,6 @@ obj-y += rwhandler.o
obj-$(CONFIG_KVM) += kvm.o kvm-all.o
obj-$(CONFIG_NO_KVM) += kvm-stub.o
-# MSI-X depends on kvm for interrupt injection,
-# so moved it from Makefile.objs to Makefile.target for now
-obj-y += msix.o
-
LIBS+=-lz
QEMU_CFLAGS += $(VNC_TLS_CFLAGS)
--
1.7.2.3
^ permalink raw reply related [flat|nested] 19+ messages in thread* Re: [PATCH 0/9] msix/kvm integration cleanups
2010-09-20 15:06 [PATCH 0/9] msix/kvm integration cleanups Avi Kivity
` (8 preceding siblings ...)
2010-09-20 15:06 ` [PATCH 9/9] Move msix.o build back to Makefile.objs Avi Kivity
@ 2010-09-20 17:02 ` Michael S. Tsirkin
2010-09-21 16:05 ` Avi Kivity
9 siblings, 1 reply; 19+ messages in thread
From: Michael S. Tsirkin @ 2010-09-20 17:02 UTC (permalink / raw)
To: Avi Kivity; +Cc: Marcelo Tosatti, kvm
On Mon, Sep 20, 2010 at 05:06:41PM +0200, Avi Kivity wrote:
> This cleans up msix/kvm integration a bit. The really important patch is the
> last one, which allows msix.o to be part of non-target-specific build.
I actually thoought this later move should be done in a different way:
- add all functions msix uses to kvm-stub.c
- kvm_irq_routing_entry should also have a stub
I sent some minor comments in case you have a reason
to prefer this way.
> Avi Kivity (9):
> msix: avoid leaking kvm data on init failure
> msix: make kvm specific initialization a function
> msix: move kvm specific msix notify into a function
> msix: move definitions from msix.c to msix.h
> Add missing #include to hw/irq.h
> msix: move kvm support functions to kvm-all.c and kvm-stub.c
> msix: Move kvm_enabled() guards to kvm-all.c functions
> Protect qemu-kvm.h declarations with NEED_CPU_H
> Move msix.o build back to Makefile.objs
>
> Makefile.objs | 4 +-
> Makefile.target | 4 -
> hw/irq.h | 2 +
> hw/msix.c | 173 ++++---------------------------------------------------
> hw/msix.h | 26 ++++++++
> kvm-all.c | 142 +++++++++++++++++++++++++++++++++++++++++++++
> kvm-stub.c | 29 +++++++++
> kvm.h | 16 +++++
> qemu-kvm.h | 21 ++++++-
> 9 files changed, 247 insertions(+), 170 deletions(-)
>
> --
> 1.7.2.3
^ permalink raw reply [flat|nested] 19+ messages in thread* Re: [PATCH 0/9] msix/kvm integration cleanups
2010-09-20 17:02 ` [PATCH 0/9] msix/kvm integration cleanups Michael S. Tsirkin
@ 2010-09-21 16:05 ` Avi Kivity
2010-09-21 16:14 ` Michael S. Tsirkin
2010-10-06 9:39 ` Michael S. Tsirkin
0 siblings, 2 replies; 19+ messages in thread
From: Avi Kivity @ 2010-09-21 16:05 UTC (permalink / raw)
To: Michael S. Tsirkin; +Cc: Marcelo Tosatti, kvm
On 09/20/2010 07:02 PM, Michael S. Tsirkin wrote:
> On Mon, Sep 20, 2010 at 05:06:41PM +0200, Avi Kivity wrote:
> > This cleans up msix/kvm integration a bit. The really important patch is the
> > last one, which allows msix.o to be part of non-target-specific build.
>
> I actually thoought this later move should be done in a different way:
> - add all functions msix uses to kvm-stub.c
Isn't that what I did?
> - kvm_irq_routing_entry should also have a stub
>
> I sent some minor comments in case you have a reason
> to prefer this way.
My motivation is really the last patch. If you explain what you'd like
to see I'll try to do it.
--
error compiling committee.c: too many arguments to function
^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH 0/9] msix/kvm integration cleanups
2010-09-21 16:05 ` Avi Kivity
@ 2010-09-21 16:14 ` Michael S. Tsirkin
2010-10-06 9:39 ` Michael S. Tsirkin
1 sibling, 0 replies; 19+ messages in thread
From: Michael S. Tsirkin @ 2010-09-21 16:14 UTC (permalink / raw)
To: Avi Kivity; +Cc: Marcelo Tosatti, kvm
On Tue, Sep 21, 2010 at 06:05:10PM +0200, Avi Kivity wrote:
> On 09/20/2010 07:02 PM, Michael S. Tsirkin wrote:
> >On Mon, Sep 20, 2010 at 05:06:41PM +0200, Avi Kivity wrote:
> >> This cleans up msix/kvm integration a bit. The really important patch is the
> >> last one, which allows msix.o to be part of non-target-specific build.
> >
> >I actually thoought this later move should be done in a different way:
> >- add all functions msix uses to kvm-stub.c
>
> Isn't that what I did?
>
> >- kvm_irq_routing_entry should also have a stub
> >
> >I sent some minor comments in case you have a reason
> >to prefer this way.
>
> My motivation is really the last patch. If you explain what you'd
> like to see I'll try to do it.
Basically my idea was to avoid all ifdefs in msix.c
*without changing it*, by stubbing out kvm APIs
and structures we use there.
--
MST
^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH 0/9] msix/kvm integration cleanups
2010-09-21 16:05 ` Avi Kivity
2010-09-21 16:14 ` Michael S. Tsirkin
@ 2010-10-06 9:39 ` Michael S. Tsirkin
2010-10-06 9:50 ` Avi Kivity
1 sibling, 1 reply; 19+ messages in thread
From: Michael S. Tsirkin @ 2010-10-06 9:39 UTC (permalink / raw)
To: Avi Kivity; +Cc: Marcelo Tosatti, kvm
On Tue, Sep 21, 2010 at 06:05:10PM +0200, Avi Kivity wrote:
> On 09/20/2010 07:02 PM, Michael S. Tsirkin wrote:
> >On Mon, Sep 20, 2010 at 05:06:41PM +0200, Avi Kivity wrote:
> >> This cleans up msix/kvm integration a bit. The really important patch is the
> >> last one, which allows msix.o to be part of non-target-specific build.
> >
> >I actually thoought this later move should be done in a different way:
> >- add all functions msix uses to kvm-stub.c
>
> Isn't that what I did?
>
> >- kvm_irq_routing_entry should also have a stub
> >
> >I sent some minor comments in case you have a reason
> >to prefer this way.
>
> My motivation is really the last patch. If you explain what you'd
> like to see I'll try to do it.
Still looking at this?
> --
> error compiling committee.c: too many arguments to function
^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH 0/9] msix/kvm integration cleanups
2010-10-06 9:39 ` Michael S. Tsirkin
@ 2010-10-06 9:50 ` Avi Kivity
0 siblings, 0 replies; 19+ messages in thread
From: Avi Kivity @ 2010-10-06 9:50 UTC (permalink / raw)
To: Michael S. Tsirkin; +Cc: Marcelo Tosatti, kvm
On 10/06/2010 11:39 AM, Michael S. Tsirkin wrote:
> On Tue, Sep 21, 2010 at 06:05:10PM +0200, Avi Kivity wrote:
> > On 09/20/2010 07:02 PM, Michael S. Tsirkin wrote:
> > >On Mon, Sep 20, 2010 at 05:06:41PM +0200, Avi Kivity wrote:
> > >> This cleans up msix/kvm integration a bit. The really important patch is the
> > >> last one, which allows msix.o to be part of non-target-specific build.
> > >
> > >I actually thoought this later move should be done in a different way:
> > >- add all functions msix uses to kvm-stub.c
> >
> > Isn't that what I did?
> >
> > >- kvm_irq_routing_entry should also have a stub
> > >
> > >I sent some minor comments in case you have a reason
> > >to prefer this way.
> >
> > My motivation is really the last patch. If you explain what you'd
> > like to see I'll try to do it.
>
> Still looking at this?
>
I plan to do this yes, when I get a bit of time.
--
error compiling committee.c: too many arguments to function
^ permalink raw reply [flat|nested] 19+ messages in thread