2019-11-28 03:10:55

by Miaohe Lin

[permalink] [raw]
Subject: [PATCH] KVM: vgic: Use warpper function to lock/unlock all vcpus in kvm_vgic_create()

From: Miaohe Lin <[email protected]>

Use warpper function lock_all_vcpus()/unlock_all_vcpus()
in kvm_vgic_create() to remove duplicated code dealing
with locking and unlocking all vcpus in a vm.

Signed-off-by: Miaohe Lin <[email protected]>
---
virt/kvm/arm/vgic/vgic-init.c | 14 ++++----------
1 file changed, 4 insertions(+), 10 deletions(-)

diff --git a/virt/kvm/arm/vgic/vgic-init.c b/virt/kvm/arm/vgic/vgic-init.c
index b3c5de48064c..53e3969dfb52 100644
--- a/virt/kvm/arm/vgic/vgic-init.c
+++ b/virt/kvm/arm/vgic/vgic-init.c
@@ -70,7 +70,7 @@ void kvm_vgic_early_init(struct kvm *kvm)
*/
int kvm_vgic_create(struct kvm *kvm, u32 type)
{
- int i, vcpu_lock_idx = -1, ret;
+ int i, ret;
struct kvm_vcpu *vcpu;

if (irqchip_in_kernel(kvm))
@@ -92,11 +92,8 @@ int kvm_vgic_create(struct kvm *kvm, u32 type)
* that no other VCPUs are run while we create the vgic.
*/
ret = -EBUSY;
- kvm_for_each_vcpu(i, vcpu, kvm) {
- if (!mutex_trylock(&vcpu->mutex))
- goto out_unlock;
- vcpu_lock_idx = i;
- }
+ if (!lock_all_vcpus(kvm))
+ return ret;

kvm_for_each_vcpu(i, vcpu, kvm) {
if (vcpu->arch.has_run_once)
@@ -125,10 +122,7 @@ int kvm_vgic_create(struct kvm *kvm, u32 type)
INIT_LIST_HEAD(&kvm->arch.vgic.rd_regions);

out_unlock:
- for (; vcpu_lock_idx >= 0; vcpu_lock_idx--) {
- vcpu = kvm_get_vcpu(kvm, vcpu_lock_idx);
- mutex_unlock(&vcpu->mutex);
- }
+ unlock_all_vcpus(kvm);
return ret;
}

--
2.19.1


2019-11-28 09:10:12

by Eric Auger

[permalink] [raw]
Subject: Re: [PATCH] KVM: vgic: Use warpper function to lock/unlock all vcpus in kvm_vgic_create()

Hi,

On 11/28/19 4:09 AM, linmiaohe wrote:
> From: Miaohe Lin <[email protected]>
>
> Use warpper function lock_all_vcpus()/unlock_all_vcpus()
s/warpper/wrapper and also in the title.
> in kvm_vgic_create() to remove duplicated code dealing
> with locking and unlocking all vcpus in a vm.
>
> Signed-off-by: Miaohe Lin <[email protected]>
> ---
> virt/kvm/arm/vgic/vgic-init.c | 14 ++++----------
> 1 file changed, 4 insertions(+), 10 deletions(-)
>
> diff --git a/virt/kvm/arm/vgic/vgic-init.c b/virt/kvm/arm/vgic/vgic-init.c
> index b3c5de48064c..53e3969dfb52 100644
> --- a/virt/kvm/arm/vgic/vgic-init.c
> +++ b/virt/kvm/arm/vgic/vgic-init.c
> @@ -70,7 +70,7 @@ void kvm_vgic_early_init(struct kvm *kvm)
> */
> int kvm_vgic_create(struct kvm *kvm, u32 type)
> {
> - int i, vcpu_lock_idx = -1, ret;
> + int i, ret;
> struct kvm_vcpu *vcpu;
>
> if (irqchip_in_kernel(kvm))
> @@ -92,11 +92,8 @@ int kvm_vgic_create(struct kvm *kvm, u32 type)
> * that no other VCPUs are run while we create the vgic.
> */
> ret = -EBUSY;
> - kvm_for_each_vcpu(i, vcpu, kvm) {
> - if (!mutex_trylock(&vcpu->mutex))
> - goto out_unlock;
> - vcpu_lock_idx = i;
> - }
> + if (!lock_all_vcpus(kvm))
> + return ret;
>
> kvm_for_each_vcpu(i, vcpu, kvm) {
> if (vcpu->arch.has_run_once)
> @@ -125,10 +122,7 @@ int kvm_vgic_create(struct kvm *kvm, u32 type)
> INIT_LIST_HEAD(&kvm->arch.vgic.rd_regions);
>
> out_unlock:
> - for (; vcpu_lock_idx >= 0; vcpu_lock_idx--) {
> - vcpu = kvm_get_vcpu(kvm, vcpu_lock_idx);
> - mutex_unlock(&vcpu->mutex);
> - }
> + unlock_all_vcpus(kvm);
> return ret;
> }
>
>
Besides, looks good to me
Reviewed-by: Eric Auger <[email protected]>

Thanks

Eric

2019-11-28 09:22:03

by Miaohe Lin

[permalink] [raw]
Subject: Re: [PATCH] KVM: vgic: Use warpper function to lock/unlock all vcpus in kvm_vgic_create()

Eric wrote:
>> From: Miaohe Lin <[email protected]>
>>
>> Use warpper function lock_all_vcpus()/unlock_all_vcpus()
> s/warpper/wrapper and also in the title.

Hi, Eric:
Many thanks for your review. I would fix this and send a patch v2.
Thanks again.

2019-11-28 09:48:50

by Marc Gonzalez

[permalink] [raw]
Subject: Re: [PATCH] KVM: vgic: Use warpper function to lock/unlock all vcpus in kvm_vgic_create()

On 28/11/2019 10:07, Auger Eric wrote:

> s/warpper/wrapper and also in the title.

Warp functions sounded very cool ;-)

"Beam me up, Scotty. There's no intelligent life down here."

Regards.