diff mbox

[Xen-devel,v3,1/3] x86/hvm: Add check when register io handler

Message ID 1463874151-6950-2-git-send-email-suravee.suthikulpanit@amd.com
State Superseded
Headers show

Commit Message

Suthikulpanit, Suravee May 21, 2016, 11:42 p.m. UTC 
From: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com>

At the time of registering HVM I/O handler, the HVM domain might
not have been initialized, which means the hvm_domain.io_handler
would be NULL. In the hvm_next_io_handler(), this should be asserted.

Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
---
 xen/arch/x86/hvm/intercept.c | 2 ++
 1 file changed, 2 insertions(+)

Comments

Suthikulpanit, Suravee May 25, 2016, 7:04 p.m. UTC | #1 
On 5/23/2016 6:46 AM, Jan Beulich wrote:
>>>> On 22.05.16 at 01:42, <suravee.suthikulpanit@amd.com> wrote:
>> --- a/xen/arch/x86/hvm/intercept.c
>> +++ b/xen/arch/x86/hvm/intercept.c
>> @@ -258,6 +258,8 @@ struct hvm_io_handler *hvm_next_io_handler(struct domain *d)
>>  {
>>      unsigned int i = d->arch.hvm_domain.io_handler_count++;
>>
>> +    ASSERT( d->arch.hvm_domain.io_handler );
>
> Am I wrong in understanding that without patch 2 this ASSERT() will
> actually trigger? In which case the patch order would be wrong (but
> that could be taken care of during commit).
>
> Jan
>

Right, I can fix this in V4 if we decide to change the 
iommu_domain_initialise() ordering.

Suravee
diff mbox

Patch

diff --git a/xen/arch/x86/hvm/intercept.c b/xen/arch/x86/hvm/intercept.c
index fc757d0..2f8d57f 100644
--- a/xen/arch/x86/hvm/intercept.c
+++ b/xen/arch/x86/hvm/intercept.c
@@ -258,6 +258,8 @@  struct hvm_io_handler *hvm_next_io_handler(struct domain *d)
 {
     unsigned int i = d->arch.hvm_domain.io_handler_count++;
 
+    ASSERT( d->arch.hvm_domain.io_handler );
+
     if ( i == NR_IO_HANDLERS )
     {
         domain_crash(d);