* [PATCH] iommu/vt-d: Fix a bug in intel_iommu_iova_to_phys() for huge page
@ 2020-02-18 22:23 Yonghyun Hwang
2020-02-19 3:15 ` Lu Baolu
0 siblings, 1 reply; 7+ messages in thread
From: Yonghyun Hwang @ 2020-02-18 22:23 UTC (permalink / raw)
To: David Woodhouse, Lu Baolu, Joerg Roedel
Cc: iommu, linux-kernel, Havard Skinnemoen, Deepa Dinamani,
Moritz Fischer, Yonghyun Hwang
intel_iommu_iova_to_phys() has a bug when it translates an IOVA for a huge
page onto its corresponding physical address. This commit fixes the bug by
accomodating the level of page entry for the IOVA and adds IOVA's lower
address to the physical address.
Signed-off-by: Yonghyun Hwang <yonghyun@google.com>
---
drivers/iommu/intel-iommu.c | 11 +++++++++--
1 file changed, 9 insertions(+), 2 deletions(-)
diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index 0c8d81f56a30..ed6e69adb578 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -5555,13 +5555,20 @@ static phys_addr_t intel_iommu_iova_to_phys(struct iommu_domain *domain,
struct dma_pte *pte;
int level = 0;
u64 phys = 0;
+ const unsigned long pfn = iova >> VTD_PAGE_SHIFT;
if (dmar_domain->flags & DOMAIN_FLAG_LOSE_CHILDREN)
return 0;
- pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT, &level);
- if (pte)
+ pte = pfn_to_dma_pte(dmar_domain, pfn, &level);
+ if (pte) {
phys = dma_pte_addr(pte);
+ if (level > 1)
+ phys += (pfn &
+ ((1UL << level_to_offset_bits(level)) - 1))
+ << VTD_PAGE_SHIFT;
+ phys += iova & (VTD_PAGE_SIZE - 1);
+ }
return phys;
}
--
2.25.0.265.gbab2e86ba0-goog
^ permalink raw reply related [flat|nested] 7+ messages in thread
* Re: [PATCH] iommu/vt-d: Fix a bug in intel_iommu_iova_to_phys() for huge page
2020-02-18 22:23 [PATCH] iommu/vt-d: Fix a bug in intel_iommu_iova_to_phys() for huge page Yonghyun Hwang
@ 2020-02-19 3:15 ` Lu Baolu
2020-02-19 4:42 ` Moritz Fischer
0 siblings, 1 reply; 7+ messages in thread
From: Lu Baolu @ 2020-02-19 3:15 UTC (permalink / raw)
To: Yonghyun Hwang, David Woodhouse, Joerg Roedel
Cc: iommu, linux-kernel, Havard Skinnemoen, Deepa Dinamani, Moritz Fischer
Hi Yonghyun,
Thanks for the patch.
On 2020/2/19 6:23, Yonghyun Hwang wrote:
> intel_iommu_iova_to_phys() has a bug when it translates an IOVA for a huge
> page onto its corresponding physical address. This commit fixes the bug by
> accomodating the level of page entry for the IOVA and adds IOVA's lower
> address to the physical address. >
> Signed-off-by: Yonghyun Hwang <yonghyun@google.com>
> ---
> drivers/iommu/intel-iommu.c | 11 +++++++++--
> 1 file changed, 9 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
> index 0c8d81f56a30..ed6e69adb578 100644
> --- a/drivers/iommu/intel-iommu.c
> +++ b/drivers/iommu/intel-iommu.c
> @@ -5555,13 +5555,20 @@ static phys_addr_t intel_iommu_iova_to_phys(struct iommu_domain *domain,
> struct dma_pte *pte;
> int level = 0;
> u64 phys = 0;
> + const unsigned long pfn = iova >> VTD_PAGE_SHIFT;
Why do you need a "const unsigned long" here?
>
> if (dmar_domain->flags & DOMAIN_FLAG_LOSE_CHILDREN)
> return 0;
>
> - pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT, &level);
> - if (pte)
> + pte = pfn_to_dma_pte(dmar_domain, pfn, &level);
> + if (pte) {
> phys = dma_pte_addr(pte);
> + if (level > 1)
> + phys += (pfn &
> + ((1UL << level_to_offset_bits(level)) - 1))
> + << VTD_PAGE_SHIFT;
> + phys += iova & (VTD_PAGE_SIZE - 1);
How about
diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index 9dc37672bf89..bd17c2510bb2 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -5693,8 +5693,14 @@ static phys_addr_t
intel_iommu_iova_to_phys(struct iommu_domain *domain,
u64 phys = 0;
pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT, &level);
- if (pte)
+ if (pte) {
+ unsigned long offset_mask;
+
+ offset_mask = BIT_MASK(level_to_offset_bits(level) +
+ VTD_PAGE_SHIFT) - 1;
phys = dma_pte_addr(pte);
+ phys += iova & (bitmask - 1);
+ }
return phys;
}
Best regards,
baolu
^ permalink raw reply related [flat|nested] 7+ messages in thread
* Re: [PATCH] iommu/vt-d: Fix a bug in intel_iommu_iova_to_phys() for huge page
2020-02-19 3:15 ` Lu Baolu
@ 2020-02-19 4:42 ` Moritz Fischer
[not found] ` <CAEauFbw_FGWuQJMpr7PgyyCAefKChXoszKAz-ykBKgxFc-aa4g@mail.gmail.com>
0 siblings, 1 reply; 7+ messages in thread
From: Moritz Fischer @ 2020-02-19 4:42 UTC (permalink / raw)
To: Lu Baolu
Cc: Yonghyun Hwang, David Woodhouse, Joerg Roedel, iommu,
linux-kernel, Havard Skinnemoen, Deepa Dinamani, Moritz Fischer
Hi Baolu, Yonghyun
On Wed, Feb 19, 2020 at 11:15:36AM +0800, Lu Baolu wrote:
> Hi Yonghyun,
>
> Thanks for the patch.
>
> On 2020/2/19 6:23, Yonghyun Hwang wrote:
> > intel_iommu_iova_to_phys() has a bug when it translates an IOVA for a huge
> > page onto its corresponding physical address. This commit fixes the bug by
> > accomodating the level of page entry for the IOVA and adds IOVA's lower
> > address to the physical address. >
> > Signed-off-by: Yonghyun Hwang <yonghyun@google.com>
> > ---
> > drivers/iommu/intel-iommu.c | 11 +++++++++--
> > 1 file changed, 9 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
> > index 0c8d81f56a30..ed6e69adb578 100644
> > --- a/drivers/iommu/intel-iommu.c
> > +++ b/drivers/iommu/intel-iommu.c
> > @@ -5555,13 +5555,20 @@ static phys_addr_t intel_iommu_iova_to_phys(struct iommu_domain *domain,
> > struct dma_pte *pte;
> > int level = 0;
> > u64 phys = 0;
> > + const unsigned long pfn = iova >> VTD_PAGE_SHIFT;
>
> Why do you need a "const unsigned long" here?
>
> > if (dmar_domain->flags & DOMAIN_FLAG_LOSE_CHILDREN)
> > return 0;
> > - pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT, &level);
> > - if (pte)
> > + pte = pfn_to_dma_pte(dmar_domain, pfn, &level);
> > + if (pte) {
> > phys = dma_pte_addr(pte);
> > + if (level > 1)
> > + phys += (pfn &
> > + ((1UL << level_to_offset_bits(level)) - 1))
> > + << VTD_PAGE_SHIFT;
> > + phys += iova & (VTD_PAGE_SIZE - 1);
>
> How about
>
> diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
> index 9dc37672bf89..bd17c2510bb2 100644
> --- a/drivers/iommu/intel-iommu.c
> +++ b/drivers/iommu/intel-iommu.c
> @@ -5693,8 +5693,14 @@ static phys_addr_t intel_iommu_iova_to_phys(struct
> iommu_domain *domain,
> u64 phys = 0;
>
> pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT, &level);
> - if (pte)
> + if (pte) {
> + unsigned long offset_mask;
> +
> + offset_mask = BIT_MASK(level_to_offset_bits(level) +
> + VTD_PAGE_SHIFT) - 1;
> phys = dma_pte_addr(pte);
> + phys += iova & (bitmask - 1);
Did you mean:
phys += iova & (offset_mask - 1);
> + }
At that point why not define a helper:
static inline unsigned long offset_mask(int level)
{
return BIT_MASK(level_to_offset_bits(level) + VTD_PAGE_SHIFT) - 1;
}
At which point the above would reduce to:
if (pte)
phys = dma_pte_addr(pte) + iova & offset_mask(level) - 1;
I might've fatfingered something here, but you get the idea :)
Cheers,
Moritz
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] iommu/vt-d: Fix a bug in intel_iommu_iova_to_phys() for huge page
[not found] ` <CAEauFbw_FGWuQJMpr7PgyyCAefKChXoszKAz-ykBKgxFc-aa4g@mail.gmail.com>
@ 2020-02-19 23:32 ` Lu Baolu
2020-02-20 0:12 ` Yonghyun Hwang
0 siblings, 1 reply; 7+ messages in thread
From: Lu Baolu @ 2020-02-19 23:32 UTC (permalink / raw)
To: Yonghyun Hwang, Moritz Fischer
Cc: David Woodhouse, Joerg Roedel, iommu, linux-kernel,
Havard Skinnemoen, Deepa Dinamani, Moritz Fischer
Hi,
On 2020/2/20 2:51, Yonghyun Hwang wrote:
> Hello Lu and Moritz,
>
> I think it's better to keep "if (level > 1)" because
> level_to_offset_bits() returns (unsigned int) -1 * LEVEL_STRIDE if level
> becomes 0.
@level will never be 0 in this case.
Best regards,
baolu
>
> static inline unsigned int level_to_offset_bits(int level)
> {
> return (level - 1) * LEVEL_STRIDE;
> }
>
> On Tue, Feb 18, 2020 at 8:42 PM Moritz Fischer <mdf@kernel.org
> <mailto:mdf@kernel.org>> wrote:
>
> Hi Baolu, Yonghyun
>
> On Wed, Feb 19, 2020 at 11:15:36AM +0800, Lu Baolu wrote:
> > Hi Yonghyun,
> >
> > Thanks for the patch.
> >
> > On 2020/2/19 6:23, Yonghyun Hwang wrote:
> > > intel_iommu_iova_to_phys() has a bug when it translates an IOVA
> for a huge
> > > page onto its corresponding physical address. This commit fixes
> the bug by
> > > accomodating the level of page entry for the IOVA and adds
> IOVA's lower
> > > address to the physical address. >
> > > Signed-off-by: Yonghyun Hwang <yonghyun@google.com
> <mailto:yonghyun@google.com>>
> > > ---
> > > drivers/iommu/intel-iommu.c | 11 +++++++++--
> > > 1 file changed, 9 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/drivers/iommu/intel-iommu.c
> b/drivers/iommu/intel-iommu.c
> > > index 0c8d81f56a30..ed6e69adb578 100644
> > > --- a/drivers/iommu/intel-iommu.c
> > > +++ b/drivers/iommu/intel-iommu.c
> > > @@ -5555,13 +5555,20 @@ static phys_addr_t
> intel_iommu_iova_to_phys(struct iommu_domain *domain,
> > > struct dma_pte *pte;
> > > int level = 0;
> > > u64 phys = 0;
> > > + const unsigned long pfn = iova >> VTD_PAGE_SHIFT;
> >
> > Why do you need a "const unsigned long" here?
>
>
>
> It's because pfn_to_dma_pte() takes pfn for its second parameter as
> "unsigned int" type and pfn is not changed in intel_iommu_iova_to_phys().
>
> >
> > > if (dmar_domain->flags & DOMAIN_FLAG_LOSE_CHILDREN)
> > > return 0;
> > > - pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT,
> &level);
> > > - if (pte)
> > > + pte = pfn_to_dma_pte(dmar_domain, pfn, &level);
> > > + if (pte) {
> > > phys = dma_pte_addr(pte);
> > > + if (level > 1)
> > > + phys += (pfn &
> > > + ((1UL <<
> level_to_offset_bits(level)) - 1))
> > > + << VTD_PAGE_SHIFT;
> > > + phys += iova & (VTD_PAGE_SIZE - 1);
> >
> > How about
>
> >
> > diff --git a/drivers/iommu/intel-iommu.c
> b/drivers/iommu/intel-iommu.c
> > index 9dc37672bf89..bd17c2510bb2 100644
> > --- a/drivers/iommu/intel-iommu.c
> > +++ b/drivers/iommu/intel-iommu.c
> > @@ -5693,8 +5693,14 @@ static phys_addr_t
> intel_iommu_iova_to_phys(struct
> > iommu_domain *domain,
> > u64 phys = 0;
> >
> > pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT,
> &level);
> > - if (pte)
> > + if (pte) {
> > + unsigned long offset_mask;
> > +
> > + offset_mask = BIT_MASK(level_to_offset_bits(level) +
> > + VTD_PAGE_SHIFT) - 1;
> > phys = dma_pte_addr(pte);
> > + phys += iova & (bitmask - 1);
> Did you mean:
>
> phys += iova & (offset_mask - 1);
>
> > + }
>
> At that point why not define a helper:
>
> static inline unsigned long offset_mask(int level)
> {
> return BIT_MASK(level_to_offset_bits(level) +
> VTD_PAGE_SHIFT) - 1;
> }
>
> At which point the above would reduce to:
>
> if (pte)
> phys = dma_pte_addr(pte) + iova & offset_mask(level) - 1;
>
>
>
> Let me introduce the helper function and upstream the patch again. Thank
> you Lu and Moritz.
>
>
> I might've fatfingered something here, but you get the idea :)
>
> Cheers,
> Moritz
>
>
> On Tue, Feb 18, 2020 at 8:42 PM Moritz Fischer <mdf@kernel.org
> <mailto:mdf@kernel.org>> wrote:
>
> Hi Baolu, Yonghyun
>
> On Wed, Feb 19, 2020 at 11:15:36AM +0800, Lu Baolu wrote:
> > Hi Yonghyun,
> >
> > Thanks for the patch.
> >
> > On 2020/2/19 6:23, Yonghyun Hwang wrote:
> > > intel_iommu_iova_to_phys() has a bug when it translates an IOVA
> for a huge
> > > page onto its corresponding physical address. This commit fixes
> the bug by
> > > accomodating the level of page entry for the IOVA and adds
> IOVA's lower
> > > address to the physical address. >
> > > Signed-off-by: Yonghyun Hwang <yonghyun@google.com
> <mailto:yonghyun@google.com>>
> > > ---
> > > drivers/iommu/intel-iommu.c | 11 +++++++++--
> > > 1 file changed, 9 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/drivers/iommu/intel-iommu.c
> b/drivers/iommu/intel-iommu.c
> > > index 0c8d81f56a30..ed6e69adb578 100644
> > > --- a/drivers/iommu/intel-iommu.c
> > > +++ b/drivers/iommu/intel-iommu.c
> > > @@ -5555,13 +5555,20 @@ static phys_addr_t
> intel_iommu_iova_to_phys(struct iommu_domain *domain,
> > > struct dma_pte *pte;
> > > int level = 0;
> > > u64 phys = 0;
> > > + const unsigned long pfn = iova >> VTD_PAGE_SHIFT;
> >
> > Why do you need a "const unsigned long" here?
> >
> > > if (dmar_domain->flags & DOMAIN_FLAG_LOSE_CHILDREN)
> > > return 0;
> > > - pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT,
> &level);
> > > - if (pte)
> > > + pte = pfn_to_dma_pte(dmar_domain, pfn, &level);
> > > + if (pte) {
> > > phys = dma_pte_addr(pte);
> > > + if (level > 1)
> > > + phys += (pfn &
> > > + ((1UL <<
> level_to_offset_bits(level)) - 1))
> > > + << VTD_PAGE_SHIFT;
> > > + phys += iova & (VTD_PAGE_SIZE - 1);
> >
> > How about
>
> >
> > diff --git a/drivers/iommu/intel-iommu.c
> b/drivers/iommu/intel-iommu.c
> > index 9dc37672bf89..bd17c2510bb2 100644
> > --- a/drivers/iommu/intel-iommu.c
> > +++ b/drivers/iommu/intel-iommu.c
> > @@ -5693,8 +5693,14 @@ static phys_addr_t
> intel_iommu_iova_to_phys(struct
> > iommu_domain *domain,
> > u64 phys = 0;
> >
> > pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT,
> &level);
> > - if (pte)
> > + if (pte) {
> > + unsigned long offset_mask;
> > +
> > + offset_mask = BIT_MASK(level_to_offset_bits(level) +
> > + VTD_PAGE_SHIFT) - 1;
> > phys = dma_pte_addr(pte);
> > + phys += iova & (bitmask - 1);
> Did you mean:
>
> phys += iova & (offset_mask - 1);
>
> > + }
>
> At that point why not define a helper:
>
> static inline unsigned long offset_mask(int level)
> {
> return BIT_MASK(level_to_offset_bits(level) +
> VTD_PAGE_SHIFT) - 1;
> }
>
> At which point the above would reduce to:
>
> if (pte)
> phys = dma_pte_addr(pte) + iova & offset_mask(level) - 1;
>
> I might've fatfingered something here, but you get the idea :)
>
> Cheers,
> Moritz
>
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] iommu/vt-d: Fix a bug in intel_iommu_iova_to_phys() for huge page
2020-02-19 23:32 ` Lu Baolu
@ 2020-02-20 0:12 ` Yonghyun Hwang
2020-02-20 2:35 ` Lu Baolu
0 siblings, 1 reply; 7+ messages in thread
From: Yonghyun Hwang @ 2020-02-20 0:12 UTC (permalink / raw)
To: Lu Baolu
Cc: Moritz Fischer, David Woodhouse, Joerg Roedel, iommu,
linux-kernel, Havard Skinnemoen, Deepa Dinamani, Moritz Fischer
Hello Baolu,
Yes, pfn_to_dma_pte() doesn't put 0 for level. However, it's hard for
me to get that without comment, which is the reason why I put a dummy
"if" there. Anyway, what about the following code?
static inline unsigned long offset_mask(int level)
{
return BIT_MASK(level_to_offset_bits(level) + VTD_PAGE_SHIFT) - 1;
}
static phys_addr_t intel_iommu_iova_to_phys(struct iommu_domain *domain,
dma_addr_t iova)
{
struct dmar_domain *dmar_domain = to_dmar_domain(domain);
struct dma_pte *pte;
int level = 0;
u64 phys = 0;
if (dmar_domain->flags & DOMAIN_FLAG_LOSE_CHILDREN)
return 0;
pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT, &level);
if (pte)
phys = dma_pte_addr(pte) + (iova & offset_mask(level));
return phys;
}
On Wed, Feb 19, 2020 at 3:33 PM Lu Baolu <baolu.lu@linux.intel.com> wrote:
>
> Hi,
>
> On 2020/2/20 2:51, Yonghyun Hwang wrote:
> > Hello Lu and Moritz,
> >
> > I think it's better to keep "if (level > 1)" because
> > level_to_offset_bits() returns (unsigned int) -1 * LEVEL_STRIDE if level
> > becomes 0.
>
> @level will never be 0 in this case.
>
> Best regards,
> baolu
>
> >
> > static inline unsigned int level_to_offset_bits(int level)
> > {
> > return (level - 1) * LEVEL_STRIDE;
> > }
> >
> > On Tue, Feb 18, 2020 at 8:42 PM Moritz Fischer <mdf@kernel.org
> > <mailto:mdf@kernel.org>> wrote:
> >
> > Hi Baolu, Yonghyun
> >
> > On Wed, Feb 19, 2020 at 11:15:36AM +0800, Lu Baolu wrote:
> > > Hi Yonghyun,
> > >
> > > Thanks for the patch.
> > >
> > > On 2020/2/19 6:23, Yonghyun Hwang wrote:
> > > > intel_iommu_iova_to_phys() has a bug when it translates an IOVA
> > for a huge
> > > > page onto its corresponding physical address. This commit fixes
> > the bug by
> > > > accomodating the level of page entry for the IOVA and adds
> > IOVA's lower
> > > > address to the physical address. >
> > > > Signed-off-by: Yonghyun Hwang <yonghyun@google.com
> > <mailto:yonghyun@google.com>>
> > > > ---
> > > > drivers/iommu/intel-iommu.c | 11 +++++++++--
> > > > 1 file changed, 9 insertions(+), 2 deletions(-)
> > > >
> > > > diff --git a/drivers/iommu/intel-iommu.c
> > b/drivers/iommu/intel-iommu.c
> > > > index 0c8d81f56a30..ed6e69adb578 100644
> > > > --- a/drivers/iommu/intel-iommu.c
> > > > +++ b/drivers/iommu/intel-iommu.c
> > > > @@ -5555,13 +5555,20 @@ static phys_addr_t
> > intel_iommu_iova_to_phys(struct iommu_domain *domain,
> > > > struct dma_pte *pte;
> > > > int level = 0;
> > > > u64 phys = 0;
> > > > + const unsigned long pfn = iova >> VTD_PAGE_SHIFT;
> > >
> > > Why do you need a "const unsigned long" here?
> >
> >
> >
> > It's because pfn_to_dma_pte() takes pfn for its second parameter as
> > "unsigned int" type and pfn is not changed in intel_iommu_iova_to_phys().
> >
> > >
> > > > if (dmar_domain->flags & DOMAIN_FLAG_LOSE_CHILDREN)
> > > > return 0;
> > > > - pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT,
> > &level);
> > > > - if (pte)
> > > > + pte = pfn_to_dma_pte(dmar_domain, pfn, &level);
> > > > + if (pte) {
> > > > phys = dma_pte_addr(pte);
> > > > + if (level > 1)
> > > > + phys += (pfn &
> > > > + ((1UL <<
> > level_to_offset_bits(level)) - 1))
> > > > + << VTD_PAGE_SHIFT;
> > > > + phys += iova & (VTD_PAGE_SIZE - 1);
> > >
> > > How about
> >
> > >
> > > diff --git a/drivers/iommu/intel-iommu.c
> > b/drivers/iommu/intel-iommu.c
> > > index 9dc37672bf89..bd17c2510bb2 100644
> > > --- a/drivers/iommu/intel-iommu.c
> > > +++ b/drivers/iommu/intel-iommu.c
> > > @@ -5693,8 +5693,14 @@ static phys_addr_t
> > intel_iommu_iova_to_phys(struct
> > > iommu_domain *domain,
> > > u64 phys = 0;
> > >
> > > pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT,
> > &level);
> > > - if (pte)
> > > + if (pte) {
> > > + unsigned long offset_mask;
> > > +
> > > + offset_mask = BIT_MASK(level_to_offset_bits(level) +
> > > + VTD_PAGE_SHIFT) - 1;
> > > phys = dma_pte_addr(pte);
> > > + phys += iova & (bitmask - 1);
> > Did you mean:
> >
> > phys += iova & (offset_mask - 1);
> >
> > > + }
> >
> > At that point why not define a helper:
> >
> > static inline unsigned long offset_mask(int level)
> > {
> > return BIT_MASK(level_to_offset_bits(level) +
> > VTD_PAGE_SHIFT) - 1;
> > }
> >
> > At which point the above would reduce to:
> >
> > if (pte)
> > phys = dma_pte_addr(pte) + iova & offset_mask(level) - 1;
> >
> >
> >
> > Let me introduce the helper function and upstream the patch again. Thank
> > you Lu and Moritz.
> >
> >
> > I might've fatfingered something here, but you get the idea :)
> >
> > Cheers,
> > Moritz
> >
> >
> > On Tue, Feb 18, 2020 at 8:42 PM Moritz Fischer <mdf@kernel.org
> > <mailto:mdf@kernel.org>> wrote:
> >
> > Hi Baolu, Yonghyun
> >
> > On Wed, Feb 19, 2020 at 11:15:36AM +0800, Lu Baolu wrote:
> > > Hi Yonghyun,
> > >
> > > Thanks for the patch.
> > >
> > > On 2020/2/19 6:23, Yonghyun Hwang wrote:
> > > > intel_iommu_iova_to_phys() has a bug when it translates an IOVA
> > for a huge
> > > > page onto its corresponding physical address. This commit fixes
> > the bug by
> > > > accomodating the level of page entry for the IOVA and adds
> > IOVA's lower
> > > > address to the physical address. >
> > > > Signed-off-by: Yonghyun Hwang <yonghyun@google.com
> > <mailto:yonghyun@google.com>>
> > > > ---
> > > > drivers/iommu/intel-iommu.c | 11 +++++++++--
> > > > 1 file changed, 9 insertions(+), 2 deletions(-)
> > > >
> > > > diff --git a/drivers/iommu/intel-iommu.c
> > b/drivers/iommu/intel-iommu.c
> > > > index 0c8d81f56a30..ed6e69adb578 100644
> > > > --- a/drivers/iommu/intel-iommu.c
> > > > +++ b/drivers/iommu/intel-iommu.c
> > > > @@ -5555,13 +5555,20 @@ static phys_addr_t
> > intel_iommu_iova_to_phys(struct iommu_domain *domain,
> > > > struct dma_pte *pte;
> > > > int level = 0;
> > > > u64 phys = 0;
> > > > + const unsigned long pfn = iova >> VTD_PAGE_SHIFT;
> > >
> > > Why do you need a "const unsigned long" here?
> > >
> > > > if (dmar_domain->flags & DOMAIN_FLAG_LOSE_CHILDREN)
> > > > return 0;
> > > > - pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT,
> > &level);
> > > > - if (pte)
> > > > + pte = pfn_to_dma_pte(dmar_domain, pfn, &level);
> > > > + if (pte) {
> > > > phys = dma_pte_addr(pte);
> > > > + if (level > 1)
> > > > + phys += (pfn &
> > > > + ((1UL <<
> > level_to_offset_bits(level)) - 1))
> > > > + << VTD_PAGE_SHIFT;
> > > > + phys += iova & (VTD_PAGE_SIZE - 1);
> > >
> > > How about
> >
> > >
> > > diff --git a/drivers/iommu/intel-iommu.c
> > b/drivers/iommu/intel-iommu.c
> > > index 9dc37672bf89..bd17c2510bb2 100644
> > > --- a/drivers/iommu/intel-iommu.c
> > > +++ b/drivers/iommu/intel-iommu.c
> > > @@ -5693,8 +5693,14 @@ static phys_addr_t
> > intel_iommu_iova_to_phys(struct
> > > iommu_domain *domain,
> > > u64 phys = 0;
> > >
> > > pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT,
> > &level);
> > > - if (pte)
> > > + if (pte) {
> > > + unsigned long offset_mask;
> > > +
> > > + offset_mask = BIT_MASK(level_to_offset_bits(level) +
> > > + VTD_PAGE_SHIFT) - 1;
> > > phys = dma_pte_addr(pte);
> > > + phys += iova & (bitmask - 1);
> > Did you mean:
> >
> > phys += iova & (offset_mask - 1);
> >
> > > + }
> >
> > At that point why not define a helper:
> >
> > static inline unsigned long offset_mask(int level)
> > {
> > return BIT_MASK(level_to_offset_bits(level) +
> > VTD_PAGE_SHIFT) - 1;
> > }
> >
> > At which point the above would reduce to:
> >
> > if (pte)
> > phys = dma_pte_addr(pte) + iova & offset_mask(level) - 1;
> >
> > I might've fatfingered something here, but you get the idea :)
> >
> > Cheers,
> > Moritz
> >
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] iommu/vt-d: Fix a bug in intel_iommu_iova_to_phys() for huge page
2020-02-20 0:12 ` Yonghyun Hwang
@ 2020-02-20 2:35 ` Lu Baolu
0 siblings, 0 replies; 7+ messages in thread
From: Lu Baolu @ 2020-02-20 2:35 UTC (permalink / raw)
To: Yonghyun Hwang
Cc: Moritz Fischer, David Woodhouse, Joerg Roedel, iommu,
linux-kernel, Havard Skinnemoen, Deepa Dinamani, Moritz Fischer
Hi Yonghyun,
On 2020/2/20 8:12, Yonghyun Hwang wrote:
> Hello Baolu,
>
> Yes, pfn_to_dma_pte() doesn't put 0 for level. However, it's hard for
> me to get that without comment, which is the reason why I put a dummy
> "if" there. Anyway, what about the following code?
>
> static inline unsigned long offset_mask(int level)
> {
> return BIT_MASK(level_to_offset_bits(level) + VTD_PAGE_SHIFT) - 1;
> }
This helper is unnecessary since it's only for the iova_to_phys(). I'd
like to put it in the code directly. Other looks good to me.
By the way, can you please rebase your patch on the top of the mainline
kernel tree and add a patch version if you are willing to send a new
version of this patch?
Best regards,
baolu
>
> static phys_addr_t intel_iommu_iova_to_phys(struct iommu_domain *domain,
> dma_addr_t iova)
> {
> struct dmar_domain *dmar_domain = to_dmar_domain(domain);
> struct dma_pte *pte;
> int level = 0;
> u64 phys = 0;
>
> if (dmar_domain->flags & DOMAIN_FLAG_LOSE_CHILDREN)
> return 0;
>
> pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT, &level);
> if (pte)
> phys = dma_pte_addr(pte) + (iova & offset_mask(level));
>
> return phys;
> }
>
> On Wed, Feb 19, 2020 at 3:33 PM Lu Baolu <baolu.lu@linux.intel.com> wrote:
>>
>> Hi,
>>
>> On 2020/2/20 2:51, Yonghyun Hwang wrote:
>>> Hello Lu and Moritz,
>>>
>>> I think it's better to keep "if (level > 1)" because
>>> level_to_offset_bits() returns (unsigned int) -1 * LEVEL_STRIDE if level
>>> becomes 0.
>>
>> @level will never be 0 in this case.
>>
>> Best regards,
>> baolu
>>
>>>
>>> static inline unsigned int level_to_offset_bits(int level)
>>> {
>>> return (level - 1) * LEVEL_STRIDE;
>>> }
>>>
>>> On Tue, Feb 18, 2020 at 8:42 PM Moritz Fischer <mdf@kernel.org
>>> <mailto:mdf@kernel.org>> wrote:
>>>
>>> Hi Baolu, Yonghyun
>>>
>>> On Wed, Feb 19, 2020 at 11:15:36AM +0800, Lu Baolu wrote:
>>> > Hi Yonghyun,
>>> >
>>> > Thanks for the patch.
>>> >
>>> > On 2020/2/19 6:23, Yonghyun Hwang wrote:
>>> > > intel_iommu_iova_to_phys() has a bug when it translates an IOVA
>>> for a huge
>>> > > page onto its corresponding physical address. This commit fixes
>>> the bug by
>>> > > accomodating the level of page entry for the IOVA and adds
>>> IOVA's lower
>>> > > address to the physical address. >
>>> > > Signed-off-by: Yonghyun Hwang <yonghyun@google.com
>>> <mailto:yonghyun@google.com>>
>>> > > ---
>>> > > drivers/iommu/intel-iommu.c | 11 +++++++++--
>>> > > 1 file changed, 9 insertions(+), 2 deletions(-)
>>> > >
>>> > > diff --git a/drivers/iommu/intel-iommu.c
>>> b/drivers/iommu/intel-iommu.c
>>> > > index 0c8d81f56a30..ed6e69adb578 100644
>>> > > --- a/drivers/iommu/intel-iommu.c
>>> > > +++ b/drivers/iommu/intel-iommu.c
>>> > > @@ -5555,13 +5555,20 @@ static phys_addr_t
>>> intel_iommu_iova_to_phys(struct iommu_domain *domain,
>>> > > struct dma_pte *pte;
>>> > > int level = 0;
>>> > > u64 phys = 0;
>>> > > + const unsigned long pfn = iova >> VTD_PAGE_SHIFT;
>>> >
>>> > Why do you need a "const unsigned long" here?
>>>
>>>
>>>
>>> It's because pfn_to_dma_pte() takes pfn for its second parameter as
>>> "unsigned int" type and pfn is not changed in intel_iommu_iova_to_phys().
>>>
>>> >
>>> > > if (dmar_domain->flags & DOMAIN_FLAG_LOSE_CHILDREN)
>>> > > return 0;
>>> > > - pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT,
>>> &level);
>>> > > - if (pte)
>>> > > + pte = pfn_to_dma_pte(dmar_domain, pfn, &level);
>>> > > + if (pte) {
>>> > > phys = dma_pte_addr(pte);
>>> > > + if (level > 1)
>>> > > + phys += (pfn &
>>> > > + ((1UL <<
>>> level_to_offset_bits(level)) - 1))
>>> > > + << VTD_PAGE_SHIFT;
>>> > > + phys += iova & (VTD_PAGE_SIZE - 1);
>>> >
>>> > How about
>>>
>>> >
>>> > diff --git a/drivers/iommu/intel-iommu.c
>>> b/drivers/iommu/intel-iommu.c
>>> > index 9dc37672bf89..bd17c2510bb2 100644
>>> > --- a/drivers/iommu/intel-iommu.c
>>> > +++ b/drivers/iommu/intel-iommu.c
>>> > @@ -5693,8 +5693,14 @@ static phys_addr_t
>>> intel_iommu_iova_to_phys(struct
>>> > iommu_domain *domain,
>>> > u64 phys = 0;
>>> >
>>> > pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT,
>>> &level);
>>> > - if (pte)
>>> > + if (pte) {
>>> > + unsigned long offset_mask;
>>> > +
>>> > + offset_mask = BIT_MASK(level_to_offset_bits(level) +
>>> > + VTD_PAGE_SHIFT) - 1;
>>> > phys = dma_pte_addr(pte);
>>> > + phys += iova & (bitmask - 1);
>>> Did you mean:
>>>
>>> phys += iova & (offset_mask - 1);
>>>
>>> > + }
>>>
>>> At that point why not define a helper:
>>>
>>> static inline unsigned long offset_mask(int level)
>>> {
>>> return BIT_MASK(level_to_offset_bits(level) +
>>> VTD_PAGE_SHIFT) - 1;
>>> }
>>>
>>> At which point the above would reduce to:
>>>
>>> if (pte)
>>> phys = dma_pte_addr(pte) + iova & offset_mask(level) - 1;
>>>
>>>
>>>
>>> Let me introduce the helper function and upstream the patch again. Thank
>>> you Lu and Moritz.
>>>
>>>
>>> I might've fatfingered something here, but you get the idea :)
>>>
>>> Cheers,
>>> Moritz
>>>
>>>
>>> On Tue, Feb 18, 2020 at 8:42 PM Moritz Fischer <mdf@kernel.org
>>> <mailto:mdf@kernel.org>> wrote:
>>>
>>> Hi Baolu, Yonghyun
>>>
>>> On Wed, Feb 19, 2020 at 11:15:36AM +0800, Lu Baolu wrote:
>>> > Hi Yonghyun,
>>> >
>>> > Thanks for the patch.
>>> >
>>> > On 2020/2/19 6:23, Yonghyun Hwang wrote:
>>> > > intel_iommu_iova_to_phys() has a bug when it translates an IOVA
>>> for a huge
>>> > > page onto its corresponding physical address. This commit fixes
>>> the bug by
>>> > > accomodating the level of page entry for the IOVA and adds
>>> IOVA's lower
>>> > > address to the physical address. >
>>> > > Signed-off-by: Yonghyun Hwang <yonghyun@google.com
>>> <mailto:yonghyun@google.com>>
>>> > > ---
>>> > > drivers/iommu/intel-iommu.c | 11 +++++++++--
>>> > > 1 file changed, 9 insertions(+), 2 deletions(-)
>>> > >
>>> > > diff --git a/drivers/iommu/intel-iommu.c
>>> b/drivers/iommu/intel-iommu.c
>>> > > index 0c8d81f56a30..ed6e69adb578 100644
>>> > > --- a/drivers/iommu/intel-iommu.c
>>> > > +++ b/drivers/iommu/intel-iommu.c
>>> > > @@ -5555,13 +5555,20 @@ static phys_addr_t
>>> intel_iommu_iova_to_phys(struct iommu_domain *domain,
>>> > > struct dma_pte *pte;
>>> > > int level = 0;
>>> > > u64 phys = 0;
>>> > > + const unsigned long pfn = iova >> VTD_PAGE_SHIFT;
>>> >
>>> > Why do you need a "const unsigned long" here?
>>> >
>>> > > if (dmar_domain->flags & DOMAIN_FLAG_LOSE_CHILDREN)
>>> > > return 0;
>>> > > - pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT,
>>> &level);
>>> > > - if (pte)
>>> > > + pte = pfn_to_dma_pte(dmar_domain, pfn, &level);
>>> > > + if (pte) {
>>> > > phys = dma_pte_addr(pte);
>>> > > + if (level > 1)
>>> > > + phys += (pfn &
>>> > > + ((1UL <<
>>> level_to_offset_bits(level)) - 1))
>>> > > + << VTD_PAGE_SHIFT;
>>> > > + phys += iova & (VTD_PAGE_SIZE - 1);
>>> >
>>> > How about
>>>
>>> >
>>> > diff --git a/drivers/iommu/intel-iommu.c
>>> b/drivers/iommu/intel-iommu.c
>>> > index 9dc37672bf89..bd17c2510bb2 100644
>>> > --- a/drivers/iommu/intel-iommu.c
>>> > +++ b/drivers/iommu/intel-iommu.c
>>> > @@ -5693,8 +5693,14 @@ static phys_addr_t
>>> intel_iommu_iova_to_phys(struct
>>> > iommu_domain *domain,
>>> > u64 phys = 0;
>>> >
>>> > pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT,
>>> &level);
>>> > - if (pte)
>>> > + if (pte) {
>>> > + unsigned long offset_mask;
>>> > +
>>> > + offset_mask = BIT_MASK(level_to_offset_bits(level) +
>>> > + VTD_PAGE_SHIFT) - 1;
>>> > phys = dma_pte_addr(pte);
>>> > + phys += iova & (bitmask - 1);
>>> Did you mean:
>>>
>>> phys += iova & (offset_mask - 1);
>>>
>>> > + }
>>>
>>> At that point why not define a helper:
>>>
>>> static inline unsigned long offset_mask(int level)
>>> {
>>> return BIT_MASK(level_to_offset_bits(level) +
>>> VTD_PAGE_SHIFT) - 1;
>>> }
>>>
>>> At which point the above would reduce to:
>>>
>>> if (pte)
>>> phys = dma_pte_addr(pte) + iova & offset_mask(level) - 1;
>>>
>>> I might've fatfingered something here, but you get the idea :)
>>>
>>> Cheers,
>>> Moritz
>>>
^ permalink raw reply [flat|nested] 7+ messages in thread
* [PATCH] iommu/vt-d: Fix a bug in intel_iommu_iova_to_phys() for huge page
@ 2020-02-19 19:32 Yonghyun Hwang
0 siblings, 0 replies; 7+ messages in thread
From: Yonghyun Hwang @ 2020-02-19 19:32 UTC (permalink / raw)
To: David Woodhouse, Lu Baolu, Joerg Roedel
Cc: iommu, linux-kernel, Havard Skinnemoen, Deepa Dinamani,
Moritz Fischer, Yonghyun Hwang
intel_iommu_iova_to_phys() has a bug when it translates an IOVA for a huge
page onto its corresponding physical address. This commit fixes the bug by
accomodating the level of page entry for the IOVA and adds IOVA's lower
address to the physical address.
Signed-off-by: Yonghyun Hwang <yonghyun@google.com>
---
drivers/iommu/intel-iommu.c | 15 +++++++++++++--
1 file changed, 13 insertions(+), 2 deletions(-)
diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index 0c8d81f56a30..beff7ede41f4 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -5548,6 +5548,15 @@ static size_t intel_iommu_unmap(struct iommu_domain *domain,
return size;
}
+static inline unsigned long pfn_level_mask(unsigned long pfn, int level)
+{
+ if (level > 1)
+ return pfn & (BIT_MASK(level_to_offset_bits(level)) - 1)
+ << VTD_PAGE_SHIFT;
+ else
+ return 0;
+}
+
static phys_addr_t intel_iommu_iova_to_phys(struct iommu_domain *domain,
dma_addr_t iova)
{
@@ -5555,13 +5564,15 @@ static phys_addr_t intel_iommu_iova_to_phys(struct iommu_domain *domain,
struct dma_pte *pte;
int level = 0;
u64 phys = 0;
+ const unsigned long pfn = iova >> VTD_PAGE_SHIFT;
if (dmar_domain->flags & DOMAIN_FLAG_LOSE_CHILDREN)
return 0;
- pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT, &level);
+ pte = pfn_to_dma_pte(dmar_domain, pfn, &level);
if (pte)
- phys = dma_pte_addr(pte);
+ phys = dma_pte_addr(pte) + pfn_level_mask(pfn, level) +
+ (iova & ~VTD_PAGE_MASK);
return phys;
}
--
2.25.0.265.gbab2e86ba0-goog
^ permalink raw reply related [flat|nested] 7+ messages in thread
end of thread, other threads:[~2020-02-20 2:35 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-02-18 22:23 [PATCH] iommu/vt-d: Fix a bug in intel_iommu_iova_to_phys() for huge page Yonghyun Hwang
2020-02-19 3:15 ` Lu Baolu
2020-02-19 4:42 ` Moritz Fischer
[not found] ` <CAEauFbw_FGWuQJMpr7PgyyCAefKChXoszKAz-ykBKgxFc-aa4g@mail.gmail.com>
2020-02-19 23:32 ` Lu Baolu
2020-02-20 0:12 ` Yonghyun Hwang
2020-02-20 2:35 ` Lu Baolu
2020-02-19 19:32 Yonghyun Hwang
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).