2021-10-22 11:30:54

by Janis Schoetterl-Glausch

Subject: [PATCH] KVM: s390: Fix handle_sske page fault handling

Retry if fixup_user_fault succeeds.
The same issue in handle_pfmf was fixed by
a11bdb1a6b78 (KVM: s390: Fix pfmf and conditional skey emulation).

Fixes: bd096f644319 ("KVM: s390: Add skey emulation fault handling")
Signed-off-by: Janis Schoetterl-Glausch <[email protected]>
---
arch/s390/kvm/priv.c | 2 ++
1 file changed, 2 insertions(+)

diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c
index 53da4ceb16a3..417154b314a6 100644
--- a/arch/s390/kvm/priv.c
+++ b/arch/s390/kvm/priv.c
@@ -397,6 +397,8 @@ static int handle_sske(struct kvm_vcpu *vcpu)
mmap_read_unlock(current->mm);
if (rc == -EFAULT)
return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
+ if (rc == -EAGAIN)
+ continue;
if (rc < 0)
return rc;
start += PAGE_SIZE;
--
2.25.1

2021-10-22 12:04:10

by Christian Borntraeger

Subject: Re: [PATCH] KVM: s390: Fix handle_sske page fault handling

Am 22.10.21 um 13:29 schrieb Janis Schoetterl-Glausch:
> Retry if fixup_user_fault succeeds.

Maybe rephrase that with a more verbose description (e.g. if fixup_user_fault succeeds
we return EAGAIN and thus we ust retry the loop and blabla....)

> The same issue in handle_pfmf was fixed by
> a11bdb1a6b78 (KVM: s390: Fix pfmf and conditional skey emulation).
>
> Fixes: bd096f644319 ("KVM: s390: Add skey emulation fault handling")
> Signed-off-by: Janis Schoetterl-Glausch <[email protected]>

Patch itself looks good:

Reviewed-by: Christian Borntraeger <[email protected]>
> ---
> arch/s390/kvm/priv.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c
> index 53da4ceb16a3..417154b314a6 100644
> --- a/arch/s390/kvm/priv.c
> +++ b/arch/s390/kvm/priv.c
> @@ -397,6 +397,8 @@ static int handle_sske(struct kvm_vcpu *vcpu)
> mmap_read_unlock(current->mm);
> if (rc == -EFAULT)
> return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
> + if (rc == -EAGAIN)
> + continue;
> if (rc < 0)
> return rc;
> start += PAGE_SIZE;
>

2021-10-22 12:16:10

by Claudio Imbrenda

Subject: Re: [PATCH] KVM: s390: Fix handle_sske page fault handling

On Fri, 22 Oct 2021 13:29:13 +0200
Janis Schoetterl-Glausch <[email protected]> wrote:

> Retry if fixup_user_fault succeeds.
> The same issue in handle_pfmf was fixed by
> a11bdb1a6b78 (KVM: s390: Fix pfmf and conditional skey emulation).
>
> Fixes: bd096f644319 ("KVM: s390: Add skey emulation fault handling")
> Signed-off-by: Janis Schoetterl-Glausch <[email protected]>

with the description fixed as indicated by Christian:

Reviewed-by: Claudio Imbrenda <[email protected]>

> ---
> arch/s390/kvm/priv.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c
> index 53da4ceb16a3..417154b314a6 100644
> --- a/arch/s390/kvm/priv.c
> +++ b/arch/s390/kvm/priv.c
> @@ -397,6 +397,8 @@ static int handle_sske(struct kvm_vcpu *vcpu)
> mmap_read_unlock(current->mm);
> if (rc == -EFAULT)
> return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
> + if (rc == -EAGAIN)
> + continue;
> if (rc < 0)
> return rc;
> start += PAGE_SIZE;

2021-10-25 07:19:28

by Christian Borntraeger

Subject: Re: [PATCH] KVM: s390: Fix handle_sske page fault handling

Am 22.10.21 um 17:26 schrieb Janis Schoetterl-Glausch:
> On Fri, Oct 22, 2021 at 02:00:13PM +0200, Christian Borntraeger wrote:
>> Am 22.10.21 um 13:29 schrieb Janis Schoetterl-Glausch:
>>> Retry if fixup_user_fault succeeds.
>>
>> Maybe rephrase that with a more verbose description (e.g. if fixup_user_fault succeeds
>> we return EAGAIN and thus we ust retry the loop and blabla....)
>>
> Done
>
> [...]
> -- >8 --
> Subject: [PATCH v2] KVM: s390: Fix handle_sske page fault handling
>
> If handle_sske cannot set the storage key, because there is no
> page table entry or no present large page entry, it calls
> fixup_user_fault.
> However, currently, if the call succeeds, handle_sske returns
> -EAGAIN, without having set the storage key.
> Instead, retry by continue'ing the loop without incrementing the
> address.
> The same issue in handle_pfmf was fixed by
> a11bdb1a6b78 (KVM: s390: Fix pfmf and conditional skey emulation).
>
> Fixes: bd096f644319 ("KVM: s390: Add skey emulation fault handling")
> Signed-off-by: Janis Schoetterl-Glausch <[email protected]>
> Reviewed-by: Christian Borntraeger <[email protected]>
> Reviewed-by: Claudio Imbrenda <[email protected]>
> ---
> arch/s390/kvm/priv.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c
> index 53da4ceb16a3..417154b314a6 100644
> --- a/arch/s390/kvm/priv.c
> +++ b/arch/s390/kvm/priv.c
> @@ -397,6 +397,8 @@ static int handle_sske(struct kvm_vcpu *vcpu)
> mmap_read_unlock(current->mm);
> if (rc == -EFAULT)
> return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
> + if (rc == -EAGAIN)
> + continue;
> if (rc < 0)
> return rc;
> start += PAGE_SIZE;
>

thanks applied.