[PATCH v2 0/2] Misc cleanup for iommu/vt-d

[PATCH v2 0/2] Misc cleanup for iommu/vt-d

[Yanfei Xu]

No functional changes expected.

v1->v2:
1.Change to use check and return for dmar_reenable_qi()
2.Remove WARN_ON of 'table' instead of using BUG_ON
3.Also remove useless WARN_ON of domain id

Thanks for Baolu's suggestion!

Yanfei Xu (2):
  iommu/vt-d: Handle the failure case of dmar_reenable_qi()
  iommu/vt-d: Remove two WARN_ON in domain_context_mapping_one()

 drivers/iommu/intel/iommu.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

-- 
2.34.1

[PATCH v2 1/2] iommu/vt-d: Handle the failure case of dmar_reenable_qi()

[Yanfei Xu]

dmar_reenable_qi() may not succeed. Check and return when it fails.

Signed-off-by: Yanfei Xu <yanfei.xu@intel.com>
---
 drivers/iommu/intel/iommu.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/iommu/intel/iommu.c b/drivers/iommu/intel/iommu.c
index 8096273b034c..e9188d045609 100644
--- a/drivers/iommu/intel/iommu.c
+++ b/drivers/iommu/intel/iommu.c
@@ -2967,10 +2967,14 @@ static int init_iommu_hw(void)
 {
 	struct dmar_drhd_unit *drhd;
 	struct intel_iommu *iommu = NULL;
+	int ret;
 
-	for_each_active_iommu(iommu, drhd)
+	for_each_active_iommu(iommu, drhd) {
 		if (iommu->qi)
-			dmar_reenable_qi(iommu);
+			ret = dmar_reenable_qi(iommu);
+		if (ret)
+			return ret;
+	}
 
 	for_each_iommu(iommu, drhd) {
 		if (drhd->ignored) {
-- 
2.34.1

Re: [PATCH v2 1/2] iommu/vt-d: Handle the failure case of dmar_reenable_qi()

[Robin Murphy]

On 2023-06-02 03:05, Yanfei Xu wrote:
> dmar_reenable_qi() may not succeed. Check and return when it fails.
> 
> Signed-off-by: Yanfei Xu <yanfei.xu@intel.com>
> ---
>   drivers/iommu/intel/iommu.c | 8 ++++++--
>   1 file changed, 6 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/iommu/intel/iommu.c b/drivers/iommu/intel/iommu.c
> index 8096273b034c..e9188d045609 100644
> --- a/drivers/iommu/intel/iommu.c
> +++ b/drivers/iommu/intel/iommu.c
> @@ -2967,10 +2967,14 @@ static int init_iommu_hw(void)
>   {
>   	struct dmar_drhd_unit *drhd;
>   	struct intel_iommu *iommu = NULL;
> +	int ret;
>   
> -	for_each_active_iommu(iommu, drhd)
> +	for_each_active_iommu(iommu, drhd) {
>   		if (iommu->qi)
> -			dmar_reenable_qi(iommu);
> +			ret = dmar_reenable_qi(iommu);
> +		if (ret)

Nit: either this should be inside the previous condition, or you need to 
initialise ret to 0.

Thanks,
Robin.

> +			return ret;
> +	}
>   
>   	for_each_iommu(iommu, drhd) {
>   		if (drhd->ignored) {

Re: [PATCH v2 1/2] iommu/vt-d: Handle the failure case of dmar_reenable_qi()

[Yanfei Xu]

On 6/2/2023 5:27 PM, Robin Murphy wrote:
> On 2023-06-02 03:05, Yanfei Xu wrote:
>> dmar_reenable_qi() may not succeed. Check and return when it fails.
>>
>> Signed-off-by: Yanfei Xu <yanfei.xu@intel.com>
>> ---
>>   drivers/iommu/intel/iommu.c | 8 ++++++--
>>   1 file changed, 6 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/iommu/intel/iommu.c b/drivers/iommu/intel/iommu.c
>> index 8096273b034c..e9188d045609 100644
>> --- a/drivers/iommu/intel/iommu.c
>> +++ b/drivers/iommu/intel/iommu.c
>> @@ -2967,10 +2967,14 @@ static int init_iommu_hw(void)
>>   {
>>       struct dmar_drhd_unit *drhd;
>>       struct intel_iommu *iommu = NULL;
>> +    int ret;
>>   -    for_each_active_iommu(iommu, drhd)
>> +    for_each_active_iommu(iommu, drhd) {
>>           if (iommu->qi)
>> -            dmar_reenable_qi(iommu);
>> +            ret = dmar_reenable_qi(iommu);
>> +        if (ret)
>
> Nit: either this should be inside the previous condition, or you need 
> to initialise ret to 0.
>
Oh, you are right. Will correct it.

Thanks,
Yanfei

> Thanks,
> Robin.
>
>> +            return ret;
>> +    }
>>         for_each_iommu(iommu, drhd) {
>>           if (drhd->ignored) {

[PATCH v2 2/2] iommu/vt-d: Remove two WARN_ON in domain_context_mapping_one()

[Yanfei Xu]

Remove the WARN_ON(did == 0) as the domain id 0 is reserved and
set once the domain_ids is allocated. So iommu_init_domains will
never return 0.

Remvoe the WARN_ON(!table) as this pointer will be accessed in
the following code, if empty "table" really happens, the kernel
will report a NULL pointer reference warning at the first place.

Signed-off-by: Yanfei Xu <yanfei.xu@intel.com>
---
 drivers/iommu/intel/iommu.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/iommu/intel/iommu.c b/drivers/iommu/intel/iommu.c
index e9188d045609..3953dd69200d 100644
--- a/drivers/iommu/intel/iommu.c
+++ b/drivers/iommu/intel/iommu.c
@@ -1897,8 +1897,6 @@ static int domain_context_mapping_one(struct dmar_domain *domain,
 	struct context_entry *context;
 	int ret;
 
-	WARN_ON(did == 0);
-
 	if (hw_pass_through && domain_type_is_si(domain))
 		translation = CONTEXT_TT_PASS_THROUGH;
 
@@ -1944,8 +1942,6 @@ static int domain_context_mapping_one(struct dmar_domain *domain,
 	if (sm_supported(iommu)) {
 		unsigned long pds;
 
-		WARN_ON(!table);
-
 		/* Setup the PASID DIR pointer: */
 		pds = context_get_sm_pds(table);
 		context->lo = (u64)virt_to_phys(table->table) |
-- 
2.34.1