mbox series

[v3,0/2] KVM: x86: Check hypercall's exit to userspace generically

Message ID 20240826022255.361406-1-binbin.wu@linux.intel.com (mailing list archive)
Headers show
Series KVM: x86: Check hypercall's exit to userspace generically | expand

Message

Binbin Wu Aug. 26, 2024, 2:22 a.m. UTC 
Currently in kvm_emulate_hypercall, KVM_HC_MAP_GPA_RANGE is checked
specifically to decide whether a KVM hypercall needs to exit to userspace
or not.  Do the check based on the hypercall_exit_enabled field of
struct kvm_arch.

Also use the API user_exit_on_hypercall() to replace the opencode.

---
v3:
- Rename is_kvm_hc_exit_enabled() to user_exit_on_hypercall(). (Sean)
- Remove the WARN_ON_ONCE(). (Isaku, Sean)
- Use BIT(hc_nr) instead of (1 << nr) (Yuan)
- Added a comment to explain why check the !ret first. (Kai)
- Add Kai and Isaku's Reviewed-by.

v2:
- Check the return value of __kvm_emulate_hypercall() before checking
  hypercall_exit_enabled to avoid an invalid KVM hypercall nr.
  https://lore.kernel.org/kvm/184d90a8-14a0-494a-9112-365417245911@linux.intel.com/
- Add a warning if a hypercall nr out of the range of hypercall_exit_enabled
  can express.

Binbin Wu (2):
  KVM: x86: Check hypercall's exit to userspace generically
  KVM: x86: Use user_exit_on_hypercall() instead of opencode

 arch/x86/kvm/svm/sev.c | 4 ++--
 arch/x86/kvm/x86.c     | 7 ++++---
 arch/x86/kvm/x86.h     | 4 ++++
 3 files changed, 10 insertions(+), 5 deletions(-)


base-commit: a1206bc992c3cd3f758a9b46117dfc7e59e8c10f

Comments

Binbin Wu Oct. 9, 2024, 5:37 a.m. UTC | #1 
Gentle ping...


On 8/26/2024 10:22 AM, Binbin Wu wrote:
> Currently in kvm_emulate_hypercall, KVM_HC_MAP_GPA_RANGE is checked
> specifically to decide whether a KVM hypercall needs to exit to userspace
> or not.  Do the check based on the hypercall_exit_enabled field of
> struct kvm_arch.
>
> Also use the API user_exit_on_hypercall() to replace the opencode.
>
> ---
> v3:
> - Rename is_kvm_hc_exit_enabled() to user_exit_on_hypercall(). (Sean)
> - Remove the WARN_ON_ONCE(). (Isaku, Sean)
> - Use BIT(hc_nr) instead of (1 << nr) (Yuan)
> - Added a comment to explain why check the !ret first. (Kai)
> - Add Kai and Isaku's Reviewed-by.
>
> v2:
> - Check the return value of __kvm_emulate_hypercall() before checking
>    hypercall_exit_enabled to avoid an invalid KVM hypercall nr.
>    https://lore.kernel.org/kvm/184d90a8-14a0-494a-9112-365417245911@linux.intel.com/
> - Add a warning if a hypercall nr out of the range of hypercall_exit_enabled
>    can express.
>
> Binbin Wu (2):
>    KVM: x86: Check hypercall's exit to userspace generically
>    KVM: x86: Use user_exit_on_hypercall() instead of opencode
>
>   arch/x86/kvm/svm/sev.c | 4 ++--
>   arch/x86/kvm/x86.c     | 7 ++++---
>   arch/x86/kvm/x86.h     | 4 ++++
>   3 files changed, 10 insertions(+), 5 deletions(-)
>
>
> base-commit: a1206bc992c3cd3f758a9b46117dfc7e59e8c10f