Patchwork kvm-ppc: halt secondary cpus when guest reset

login
register
mail settings
Submitter Liu Yu-B13201
Date Nov. 29, 2011, 6:41 a.m.
Message ID <1322548878-14000-1-git-send-email-yu.liu@freescale.com>
Download mbox | patch
Permalink /patch/128231/
State New
Headers show

Comments

Liu Yu-B13201 - Nov. 29, 2011, 6:41 a.m.
When guest reset, we need to halt secondary cpus until guest kick them.
This already works for tcg. The patch add the support for kvm.

For kvm, if use in-kernel mpic, it should not be halted.
In this case, we pause the sencondaries instead.
This has impact to handling of halt exit,
but fortunately, kvmppc doesn't use halt exit.

Signed-off-by: Liu Yu <yu.liu@freescale.com>
---
 hw/ppce500_spin.c |    1 +
 target-ppc/kvm.c  |   11 ++++++++++-
 2 files changed, 11 insertions(+), 1 deletions(-)
Alexander Graf - Dec. 19, 2011, 1:14 p.m.
On 29.11.2011, at 07:41, Liu Yu wrote:

> When guest reset, we need to halt secondary cpus until guest kick them.
> This already works for tcg. The patch add the support for kvm.
> 
> For kvm, if use in-kernel mpic, it should not be halted.
> In this case, we pause the sencondaries instead.
> This has impact to handling of halt exit,
> but fortunately, kvmppc doesn't use halt exit.
> 
> Signed-off-by: Liu Yu <yu.liu@freescale.com>
> ---
> hw/ppce500_spin.c |    1 +
> target-ppc/kvm.c  |   11 ++++++++++-
> 2 files changed, 11 insertions(+), 1 deletions(-)
> 
> diff --git a/hw/ppce500_spin.c b/hw/ppce500_spin.c
> index cccd940..2b52728 100644
> --- a/hw/ppce500_spin.c
> +++ b/hw/ppce500_spin.c
> @@ -112,6 +112,7 @@ static void spin_kick(void *data)
> 
>     env->halted = 0;
>     env->exception_index = -1;
> +    env->stopped = 0;
>     qemu_cpu_kick(env);
> }
> 
> diff --git a/target-ppc/kvm.c b/target-ppc/kvm.c
> index 429349f..ff4cfdb 100644
> --- a/target-ppc/kvm.c
> +++ b/target-ppc/kvm.c
> @@ -504,7 +504,16 @@ void kvm_arch_post_run(CPUState *env, struct kvm_run *run)
> 
> int kvm_arch_process_async_events(CPUState *env)
> {
> -    return 0;
> +    if (kvm_irqchip_in_kernel()) {

There is no in-kernel irqchip in upstream KVM for BookE. I'll drop this branch. The rest is applied to ppc-next.


Alex

Patch

diff --git a/hw/ppce500_spin.c b/hw/ppce500_spin.c
index cccd940..2b52728 100644
--- a/hw/ppce500_spin.c
+++ b/hw/ppce500_spin.c
@@ -112,6 +112,7 @@  static void spin_kick(void *data)
 
     env->halted = 0;
     env->exception_index = -1;
+    env->stopped = 0;
     qemu_cpu_kick(env);
 }
 
diff --git a/target-ppc/kvm.c b/target-ppc/kvm.c
index 429349f..ff4cfdb 100644
--- a/target-ppc/kvm.c
+++ b/target-ppc/kvm.c
@@ -504,7 +504,16 @@  void kvm_arch_post_run(CPUState *env, struct kvm_run *run)
 
 int kvm_arch_process_async_events(CPUState *env)
 {
-    return 0;
+    if (kvm_irqchip_in_kernel()) {
+        if (env->halted == 1 && env->exception_index == EXCP_HLT) {
+            env->stopped = 1;
+            return 1;
+        } else {
+            return 0;
+        }
+    }
+
+    return env->halted;
 }
 
 static int kvmppc_handle_halt(CPUState *env)