Re: [PATCH v1 02/15] intel_iommu: Optimize context entry cache utilization
On 6/18/25 4:10 AM, Duan, Zhenzhong wrote:
> Hi Eric,
>
>> -Original Message-
>> From: Eric Auger
>> Subject: Re: [PATCH v1 02/15] intel_iommu: Optimize context entry cache
>> utilization
>>
>> Hi Zhenzhong,
>>
>> On 6/6/25 12:04 PM, Zhenzhong Duan wrote:
>>> There are many call sites referencing context entry by calling
>>> vtd_dev_to_context_entry() which will traverse the DMAR table.
>>>
>>> In most cases we can use cached context entry in vtd_as->context_cache_entry
>>> except when its entry is stale. Currently only global and domain context
>>> invalidation stale it.
>>>
>>> So introduce a helper function vtd_as_to_context_entry() to fetch from cache
>>> before trying with vtd_dev_to_context_entry().
>>>
>>> Signed-off-by: Zhenzhong Duan
>>> ---
>>> hw/i386/intel_iommu.c | 36 +++-
>>> 1 file changed, 23 insertions(+), 13 deletions(-)
>>>
>>> diff --git a/hw/i386/intel_iommu.c b/hw/i386/intel_iommu.c
>>> index f0b1f90eff..a2f3250724 100644
>>> --- a/hw/i386/intel_iommu.c
>>> +++ b/hw/i386/intel_iommu.c
>>> @@ -1597,6 +1597,22 @@ static int
>> vtd_dev_to_context_entry(IntelIOMMUState *s, uint8_t bus_num,
>>> return 0;
>>> }
>>>
>>> +static int vtd_as_to_context_entry(VTDAddressSpace *vtd_as,
>> VTDContextEntry *ce)
>>> +{
>>> +IntelIOMMUState *s = vtd_as->iommu_state;
>>> +uint8_t bus_num = pci_bus_num(vtd_as->bus);
>>> +uint8_t devfn = vtd_as->devfn;
>>> +VTDContextCacheEntry *cc_entry = &vtd_as->context_cache_entry;
>>> +
>>> +/* Try to fetch context-entry from cache first */
>>> +if (cc_entry->context_cache_gen == s->context_cache_gen) {
>>> +*ce = cc_entry->context_entry;
>>> +return 0;
>>> +} else {
>>> +return vtd_dev_to_context_entry(s, bus_num, devfn, ce);
>>> +}
>>> +}
>>> +
>> While the patch looks good to me can't you use the helper also in
>> vtd_do_iommu_translate()?
>> See " /* Try to fetch context-entry from cache first */"
> It can, but it finally calls into vtd_dev_to_context_entry() so we can call
> vtd_dev_to_context_entry() directly.
> I will drop this patch following Yi's suggestion.
OK
Cheers
Eric
>
> Thanks
> Zhenzhong
RE: [PATCH v1 02/15] intel_iommu: Optimize context entry cache utilization
Hi Eric,
>-Original Message-
>From: Eric Auger
>Subject: Re: [PATCH v1 02/15] intel_iommu: Optimize context entry cache
>utilization
>
>Hi Zhenzhong,
>
>On 6/6/25 12:04 PM, Zhenzhong Duan wrote:
>> There are many call sites referencing context entry by calling
>> vtd_dev_to_context_entry() which will traverse the DMAR table.
>>
>> In most cases we can use cached context entry in vtd_as->context_cache_entry
>> except when its entry is stale. Currently only global and domain context
>> invalidation stale it.
>>
>> So introduce a helper function vtd_as_to_context_entry() to fetch from cache
>> before trying with vtd_dev_to_context_entry().
>>
>> Signed-off-by: Zhenzhong Duan
>> ---
>> hw/i386/intel_iommu.c | 36 +++-
>> 1 file changed, 23 insertions(+), 13 deletions(-)
>>
>> diff --git a/hw/i386/intel_iommu.c b/hw/i386/intel_iommu.c
>> index f0b1f90eff..a2f3250724 100644
>> --- a/hw/i386/intel_iommu.c
>> +++ b/hw/i386/intel_iommu.c
>> @@ -1597,6 +1597,22 @@ static int
>vtd_dev_to_context_entry(IntelIOMMUState *s, uint8_t bus_num,
>> return 0;
>> }
>>
>> +static int vtd_as_to_context_entry(VTDAddressSpace *vtd_as,
>VTDContextEntry *ce)
>> +{
>> +IntelIOMMUState *s = vtd_as->iommu_state;
>> +uint8_t bus_num = pci_bus_num(vtd_as->bus);
>> +uint8_t devfn = vtd_as->devfn;
>> +VTDContextCacheEntry *cc_entry = &vtd_as->context_cache_entry;
>> +
>> +/* Try to fetch context-entry from cache first */
>> +if (cc_entry->context_cache_gen == s->context_cache_gen) {
>> +*ce = cc_entry->context_entry;
>> +return 0;
>> +} else {
>> +return vtd_dev_to_context_entry(s, bus_num, devfn, ce);
>> +}
>> +}
>> +
>While the patch looks good to me can't you use the helper also in
>vtd_do_iommu_translate()?
>See " /* Try to fetch context-entry from cache first */"
It can, but it finally calls into vtd_dev_to_context_entry() so we can call
vtd_dev_to_context_entry() directly.
I will drop this patch following Yi's suggestion.
Thanks
Zhenzhong
RE: [PATCH v1 02/15] intel_iommu: Optimize context entry cache utilization
>-Original Message-
>From: Liu, Yi L
>Subject: Re: [PATCH v1 02/15] intel_iommu: Optimize context entry cache
>utilization
>
>On 2025/6/11 18:06, Duan, Zhenzhong wrote:
>>
>>
>>> -Original Message-----
>>> From: Liu, Yi L
>>> Subject: Re: [PATCH v1 02/15] intel_iommu: Optimize context entry cache
>>> utilization
>>>
>>> On 2025/6/6 18:04, Zhenzhong Duan wrote:
>>>> There are many call sites referencing context entry by calling
>>>> vtd_dev_to_context_entry() which will traverse the DMAR table.
>>>>
>>>> In most cases we can use cached context entry in vtd_as-
>>context_cache_entry
>>>> except when its entry is stale. Currently only global and domain context
>>>> invalidation stale it.
>>>>
>>>> So introduce a helper function vtd_as_to_context_entry() to fetch from
>cache
>>>> before trying with vtd_dev_to_context_entry().
>>>
>>> The cached context entry is now protected by vtd_iommu_lock(). While not
>>> all caller of vtd_dev_to_context_entry() are under this lock.
>>>
>>> Also, the cached context entry is created in the translate path. IMHO,
>>> this path is not supposed to be triggered for passthrough devices.
>>> While this may need double check and may change in the future. But let's
>>> see if any locking issue with the current code.
>>
>> Good finding, yes.
>> Previously I thought translation path updates cc_entry->context_entry after
>cc_entry->context_cache_gen.
>> In vtd_as_to_context_entry() cc_entry->context_cache_gen is checked first, so
>there was no real race.
>> But I still missed a memory barrier like below:
>
>yeah, testing context_cache_gen is necessary. But without lock, this
>cannot guarantee the cc_entry is valid after the test.
>
>> @@ -2277,6 +2286,7 @@ static bool
>vtd_do_iommu_translate(VTDAddressSpace *vtd_as, PCIBus *bus,
>> cc_entry->context_cache_gen,
>> s->context_cache_gen);
>> cc_entry->context_entry = ce;
>> +smp_wmb();
>> cc_entry->context_cache_gen = s->context_cache_gen;
>> }
>>
>> Another option I can think of is adding lock to cache reading like below:
>
>this is in-enough as well since the cc_entry->context_entry can be modified
>after lock is released.
>
>> @@ -1659,11 +1659,15 @@ static int
>vtd_as_to_context_entry(VTDAddressSpace *vtd_as, VTDContextEntry *ce)
>> uint8_t devfn = vtd_as->devfn;
>> VTDContextCacheEntry *cc_entry = &vtd_as->context_cache_entry;
>>
>> +vtd_iommu_lock(s);
>> +
>> /* Try to fetch context-entry from cache first */
>> if (cc_entry->context_cache_gen == s->context_cache_gen) {
>> *ce = cc_entry->context_entry;
>> +vtd_iommu_unlock(s);
>> return 0;
>> } else {
>> +vtd_iommu_unlock(s);
>> return vtd_dev_to_context_entry(s, bus_num, devfn, ce);
>> }
>> }
>>
>> Which one do you prefer?
>
>If it's just optimization, perhaps just drop it. :)
Fine for me, will do.
Thanks
Zhenzhong
Re: [PATCH v1 02/15] intel_iommu: Optimize context entry cache utilization
Hi Zhenzhong,
On 6/6/25 12:04 PM, Zhenzhong Duan wrote:
> There are many call sites referencing context entry by calling
> vtd_dev_to_context_entry() which will traverse the DMAR table.
>
> In most cases we can use cached context entry in vtd_as->context_cache_entry
> except when its entry is stale. Currently only global and domain context
> invalidation stale it.
>
> So introduce a helper function vtd_as_to_context_entry() to fetch from cache
> before trying with vtd_dev_to_context_entry().
>
> Signed-off-by: Zhenzhong Duan
> ---
> hw/i386/intel_iommu.c | 36 +++-
> 1 file changed, 23 insertions(+), 13 deletions(-)
>
> diff --git a/hw/i386/intel_iommu.c b/hw/i386/intel_iommu.c
> index f0b1f90eff..a2f3250724 100644
> --- a/hw/i386/intel_iommu.c
> +++ b/hw/i386/intel_iommu.c
> @@ -1597,6 +1597,22 @@ static int vtd_dev_to_context_entry(IntelIOMMUState
> *s, uint8_t bus_num,
> return 0;
> }
>
> +static int vtd_as_to_context_entry(VTDAddressSpace *vtd_as, VTDContextEntry
> *ce)
> +{
> +IntelIOMMUState *s = vtd_as->iommu_state;
> +uint8_t bus_num = pci_bus_num(vtd_as->bus);
> +uint8_t devfn = vtd_as->devfn;
> +VTDContextCacheEntry *cc_entry = &vtd_as->context_cache_entry;
> +
> +/* Try to fetch context-entry from cache first */
> +if (cc_entry->context_cache_gen == s->context_cache_gen) {
> +*ce = cc_entry->context_entry;
> +return 0;
> +} else {
> +return vtd_dev_to_context_entry(s, bus_num, devfn, ce);
> +}
> +}
> +
While the patch looks good to me can't you use the helper also in
vtd_do_iommu_translate()?
See " /* Try to fetch context-entry from cache first */"
If not you may add a comment in the commit desc while it can't be
applied there.
Thanks
Eric
> static int vtd_sync_shadow_page_hook(const IOMMUTLBEvent *event,
> void *private)
> {
> @@ -1649,9 +1665,7 @@ static int vtd_address_space_sync(VTDAddressSpace
> *vtd_as)
> return 0;
> }
>
> -ret = vtd_dev_to_context_entry(vtd_as->iommu_state,
> - pci_bus_num(vtd_as->bus),
> - vtd_as->devfn, &ce);
> +ret = vtd_as_to_context_entry(vtd_as, &ce);
> if (ret) {
> if (ret == -VTD_FR_CONTEXT_ENTRY_P) {
> /*
> @@ -1710,8 +1724,7 @@ static bool vtd_as_pt_enabled(VTDAddressSpace *as)
> assert(as);
>
> s = as->iommu_state;
> -if (vtd_dev_to_context_entry(s, pci_bus_num(as->bus), as->devfn,
> - &ce)) {
> +if (vtd_as_to_context_entry(as, &ce)) {
> /*
> * Possibly failed to parse the context entry for some reason
> * (e.g., during init, or any guest configuration errors on
> @@ -2435,8 +2448,7 @@ static void vtd_iotlb_domain_invalidate(IntelIOMMUState
> *s, uint16_t domain_id)
> vtd_iommu_unlock(s);
>
> QLIST_FOREACH(vtd_as, &s->vtd_as_with_notifiers, next) {
> -if (!vtd_dev_to_context_entry(s, pci_bus_num(vtd_as->bus),
> - vtd_as->devfn, &ce) &&
> +if (!vtd_as_to_context_entry(vtd_as, &ce) &&
> domain_id == vtd_get_domain_id(s, &ce, vtd_as->pasid)) {
> vtd_address_space_sync(vtd_as);
> }
> @@ -2458,8 +2470,7 @@ static void
> vtd_iotlb_page_invalidate_notify(IntelIOMMUState *s,
> hwaddr size = (1 << am) * VTD_PAGE_SIZE;
>
> QLIST_FOREACH(vtd_as, &(s->vtd_as_with_notifiers), next) {
> -ret = vtd_dev_to_context_entry(s, pci_bus_num(vtd_as->bus),
> - vtd_as->devfn, &ce);
> +ret = vtd_as_to_context_entry(vtd_as, &ce);
> if (!ret && domain_id == vtd_get_domain_id(s, &ce, vtd_as->pasid)) {
> uint32_t rid2pasid = PCI_NO_PASID;
>
> @@ -2966,8 +2977,7 @@ static void vtd_piotlb_pasid_invalidate(IntelIOMMUState
> *s,
> vtd_iommu_unlock(s);
>
> QLIST_FOREACH(vtd_as, &s->vtd_as_with_notifiers, next) {
> -if (!vtd_dev_to_context_entry(s, pci_bus_num(vtd_as->bus),
> - vtd_as->devfn, &ce) &&
> +if (!vtd_as_to_context_entry(vtd_as, &ce) &&
> domain_id == vtd_get_domain_id(s, &ce, vtd_as->pasid)) {
> uint32_t rid2pasid = VTD_CE_GET_RID2PASID(&ce);
>
> @@ -4146,7 +4156,7 @@ static void
> vtd_report_ir_illegal_access(VTDAddressSpace *vtd_as,
> assert(vtd_as->pasid != PCI_NO_PASID);
>
> /* Try out best to fetch FPD, we can't do anything more */
> -if (vtd_dev_to_context_entry(s, bus_n, vtd_as->devfn, &ce) == 0) {
> +if (vtd_as_to_context_entry(vtd_as, &ce) == 0) {
> is_fpd_set = ce.lo & VTD_CONTEXT_ENTRY_FPD;
> if (!is_fpd_set && s->root_scalable) {
> vtd_ce_get_pasid_fpd(s, &ce, &is_fpd_set, vtd_as->pasid);
> @@ -4506,7 +4516,7 @@ static void vtd_iommu_replay(IOMMUMemoryRegion
> *iommu_mr, IOMMUNotifier *n)
> /* rep
Re: [PATCH v1 02/15] intel_iommu: Optimize context entry cache utilization
On 2025/6/11 18:06, Duan, Zhenzhong wrote:
-Original Message-
From: Liu, Yi L
Subject: Re: [PATCH v1 02/15] intel_iommu: Optimize context entry cache
utilization
On 2025/6/6 18:04, Zhenzhong Duan wrote:
There are many call sites referencing context entry by calling
vtd_dev_to_context_entry() which will traverse the DMAR table.
In most cases we can use cached context entry in vtd_as->context_cache_entry
except when its entry is stale. Currently only global and domain context
invalidation stale it.
So introduce a helper function vtd_as_to_context_entry() to fetch from cache
before trying with vtd_dev_to_context_entry().
The cached context entry is now protected by vtd_iommu_lock(). While not
all caller of vtd_dev_to_context_entry() are under this lock.
Also, the cached context entry is created in the translate path. IMHO,
this path is not supposed to be triggered for passthrough devices.
While this may need double check and may change in the future. But let's
see if any locking issue with the current code.
Good finding, yes.
Previously I thought translation path updates cc_entry->context_entry after
cc_entry->context_cache_gen.
In vtd_as_to_context_entry() cc_entry->context_cache_gen is checked first, so
there was no real race.
But I still missed a memory barrier like below:
yeah, testing context_cache_gen is necessary. But without lock, this
cannot guarantee the cc_entry is valid after the test.
@@ -2277,6 +2286,7 @@ static bool vtd_do_iommu_translate(VTDAddressSpace
*vtd_as, PCIBus *bus,
cc_entry->context_cache_gen,
s->context_cache_gen);
cc_entry->context_entry = ce;
+smp_wmb();
cc_entry->context_cache_gen = s->context_cache_gen;
}
Another option I can think of is adding lock to cache reading like below:
this is in-enough as well since the cc_entry->context_entry can be modified
after lock is released.
@@ -1659,11 +1659,15 @@ static int vtd_as_to_context_entry(VTDAddressSpace
*vtd_as, VTDContextEntry *ce)
uint8_t devfn = vtd_as->devfn;
VTDContextCacheEntry *cc_entry = &vtd_as->context_cache_entry;
+vtd_iommu_lock(s);
+
/* Try to fetch context-entry from cache first */
if (cc_entry->context_cache_gen == s->context_cache_gen) {
*ce = cc_entry->context_entry;
+vtd_iommu_unlock(s);
return 0;
} else {
+vtd_iommu_unlock(s);
return vtd_dev_to_context_entry(s, bus_num, devfn, ce);
}
}
Which one do you prefer?
If it's just optimization, perhaps just drop it. :)
--
Regards,
Yi Liu
RE: [PATCH v1 02/15] intel_iommu: Optimize context entry cache utilization
>-Original Message-
>From: Liu, Yi L
>Subject: Re: [PATCH v1 02/15] intel_iommu: Optimize context entry cache
>utilization
>
>On 2025/6/6 18:04, Zhenzhong Duan wrote:
>> There are many call sites referencing context entry by calling
>> vtd_dev_to_context_entry() which will traverse the DMAR table.
>>
>> In most cases we can use cached context entry in vtd_as->context_cache_entry
>> except when its entry is stale. Currently only global and domain context
>> invalidation stale it.
>>
>> So introduce a helper function vtd_as_to_context_entry() to fetch from cache
>> before trying with vtd_dev_to_context_entry().
>
>The cached context entry is now protected by vtd_iommu_lock(). While not
>all caller of vtd_dev_to_context_entry() are under this lock.
>
>Also, the cached context entry is created in the translate path. IMHO,
>this path is not supposed to be triggered for passthrough devices.
>While this may need double check and may change in the future. But let's
>see if any locking issue with the current code.
Good finding, yes.
Previously I thought translation path updates cc_entry->context_entry after
cc_entry->context_cache_gen.
In vtd_as_to_context_entry() cc_entry->context_cache_gen is checked first, so
there was no real race.
But I still missed a memory barrier like below:
@@ -2277,6 +2286,7 @@ static bool vtd_do_iommu_translate(VTDAddressSpace
*vtd_as, PCIBus *bus,
cc_entry->context_cache_gen,
s->context_cache_gen);
cc_entry->context_entry = ce;
+smp_wmb();
cc_entry->context_cache_gen = s->context_cache_gen;
}
Another option I can think of is adding lock to cache reading like below:
@@ -1659,11 +1659,15 @@ static int vtd_as_to_context_entry(VTDAddressSpace
*vtd_as, VTDContextEntry *ce)
uint8_t devfn = vtd_as->devfn;
VTDContextCacheEntry *cc_entry = &vtd_as->context_cache_entry;
+vtd_iommu_lock(s);
+
/* Try to fetch context-entry from cache first */
if (cc_entry->context_cache_gen == s->context_cache_gen) {
*ce = cc_entry->context_entry;
+vtd_iommu_unlock(s);
return 0;
} else {
+vtd_iommu_unlock(s);
return vtd_dev_to_context_entry(s, bus_num, devfn, ce);
}
}
Which one do you prefer?
Thanks
Zhenzhong
Re: [PATCH v1 02/15] intel_iommu: Optimize context entry cache utilization
On 2025/6/6 18:04, Zhenzhong Duan wrote:
There are many call sites referencing context entry by calling
vtd_dev_to_context_entry() which will traverse the DMAR table.
In most cases we can use cached context entry in vtd_as->context_cache_entry
except when its entry is stale. Currently only global and domain context
invalidation stale it.
So introduce a helper function vtd_as_to_context_entry() to fetch from cache
before trying with vtd_dev_to_context_entry().
The cached context entry is now protected by vtd_iommu_lock(). While not
all caller of vtd_dev_to_context_entry() are under this lock.
Also, the cached context entry is created in the translate path. IMHO,
this path is not supposed to be triggered for passthrough devices.
While this may need double check and may change in the future. But let's
see if any locking issue with the current code.
Signed-off-by: Zhenzhong Duan
---
hw/i386/intel_iommu.c | 36 +++-
1 file changed, 23 insertions(+), 13 deletions(-)
diff --git a/hw/i386/intel_iommu.c b/hw/i386/intel_iommu.c
index f0b1f90eff..a2f3250724 100644
--- a/hw/i386/intel_iommu.c
+++ b/hw/i386/intel_iommu.c
@@ -1597,6 +1597,22 @@ static int vtd_dev_to_context_entry(IntelIOMMUState *s,
uint8_t bus_num,
return 0;
}
+static int vtd_as_to_context_entry(VTDAddressSpace *vtd_as, VTDContextEntry *ce)
+{
+IntelIOMMUState *s = vtd_as->iommu_state;
+uint8_t bus_num = pci_bus_num(vtd_as->bus);
+uint8_t devfn = vtd_as->devfn;
+VTDContextCacheEntry *cc_entry = &vtd_as->context_cache_entry;
+
+/* Try to fetch context-entry from cache first */
+if (cc_entry->context_cache_gen == s->context_cache_gen) {
+*ce = cc_entry->context_entry;
+return 0;
+} else {
+return vtd_dev_to_context_entry(s, bus_num, devfn, ce);
+}
+}
+
static int vtd_sync_shadow_page_hook(const IOMMUTLBEvent *event,
void *private)
{
@@ -1649,9 +1665,7 @@ static int vtd_address_space_sync(VTDAddressSpace *vtd_as)
return 0;
}
-ret = vtd_dev_to_context_entry(vtd_as->iommu_state,
- pci_bus_num(vtd_as->bus),
- vtd_as->devfn, &ce);
+ret = vtd_as_to_context_entry(vtd_as, &ce);
if (ret) {
if (ret == -VTD_FR_CONTEXT_ENTRY_P) {
/*
@@ -1710,8 +1724,7 @@ static bool vtd_as_pt_enabled(VTDAddressSpace *as)
assert(as);
s = as->iommu_state;
-if (vtd_dev_to_context_entry(s, pci_bus_num(as->bus), as->devfn,
- &ce)) {
+if (vtd_as_to_context_entry(as, &ce)) {
/*
* Possibly failed to parse the context entry for some reason
* (e.g., during init, or any guest configuration errors on
@@ -2435,8 +2448,7 @@ static void vtd_iotlb_domain_invalidate(IntelIOMMUState
*s, uint16_t domain_id)
vtd_iommu_unlock(s);
QLIST_FOREACH(vtd_as, &s->vtd_as_with_notifiers, next) {
-if (!vtd_dev_to_context_entry(s, pci_bus_num(vtd_as->bus),
- vtd_as->devfn, &ce) &&
+if (!vtd_as_to_context_entry(vtd_as, &ce) &&
this one apparently is called out of lock.
domain_id == vtd_get_domain_id(s, &ce, vtd_as->pasid)) {
vtd_address_space_sync(vtd_as);
}
@@ -2458,8 +2470,7 @@ static void
vtd_iotlb_page_invalidate_notify(IntelIOMMUState *s,
hwaddr size = (1 << am) * VTD_PAGE_SIZE;
QLIST_FOREACH(vtd_as, &(s->vtd_as_with_notifiers), next) {
-ret = vtd_dev_to_context_entry(s, pci_bus_num(vtd_as->bus),
- vtd_as->devfn, &ce);
+ret = vtd_as_to_context_entry(vtd_as, &ce);
if (!ret && domain_id == vtd_get_domain_id(s, &ce, vtd_as->pasid)) {
uint32_t rid2pasid = PCI_NO_PASID;
@@ -2966,8 +2977,7 @@ static void vtd_piotlb_pasid_invalidate(IntelIOMMUState *s,
vtd_iommu_unlock(s);
QLIST_FOREACH(vtd_as, &s->vtd_as_with_notifiers, next) {
-if (!vtd_dev_to_context_entry(s, pci_bus_num(vtd_as->bus),
- vtd_as->devfn, &ce) &&
+if (!vtd_as_to_context_entry(vtd_as, &ce) &&
domain_id == vtd_get_domain_id(s, &ce, vtd_as->pasid)) {
uint32_t rid2pasid = VTD_CE_GET_RID2PASID(&ce);
@@ -4146,7 +4156,7 @@ static void vtd_report_ir_illegal_access(VTDAddressSpace *vtd_as,
assert(vtd_as->pasid != PCI_NO_PASID);
/* Try out best to fetch FPD, we can't do anything more */
-if (vtd_dev_to_context_entry(s, bus_n, vtd_as->devfn, &ce) == 0) {
+if (vtd_as_to_context_entry(vtd_as, &ce) == 0) {
is_fpd_set = ce.lo & VTD_CONTEXT_ENTRY_FPD;
if (!is_fpd_set && s->root_scalable) {
vtd_ce_get_pasid_fpd(s, &ce, &is_fpd_set, vtd_as->pasid);
@@ -4506,7 +4516,7 @@ static void vtd_iommu_replay(IOMMUMemoryRegion *iommu_mr,
IOMMUNotifier
