[PATCH] iommu: Flow ERR_PTR out from __iommu_domain_alloc()

[PATCH] iommu: Flow ERR_PTR out from __iommu_domain_alloc()

[Jason Gunthorpe]

Most of the calling code now has error handling that can carry an error
code further up the call chain. Keep the exported interface
iommu_domain_alloc() returning NULL and reflow the internal code to use
ERR_PTR not NULL for domain allocation failure.

Optionally allow drivers to return ERR_PTR from any of the alloc ops. Many
of the new ops (user, sva, etc) already return ERR_PTR, so having two
rules is confusing and hard on drivers. This fixes a bug in DART that was
returning ERR_PTR.

This also fixes a bug in iommu_group_alloc_default_domain() where it
returned both NULL and ERR_PTR, now it always returns ERR_PTR.

Fixes: 482feb5c6492 ("iommu/dart: Call apple_dart_finalize_domain() as part of alloc_paging()")
Fixes: 1c68cbc64fe6 ("iommu: Add IOMMU_DOMAIN_PLATFORM")
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/linux-iommu/b85e0715-3224-4f45-ad6b-ebb9f08c015d@moroto.mountain/
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
---
 drivers/iommu/iommu.c | 53 +++++++++++++++++++++++++++++--------------
 1 file changed, 36 insertions(+), 17 deletions(-)

diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
index f17a1113f3d6a3..6bc044d2310487 100644
--- a/drivers/iommu/iommu.c
+++ b/drivers/iommu/iommu.c
@@ -1802,10 +1802,10 @@ iommu_group_alloc_default_domain(struct iommu_group *group, int req_type)
 
 	/* Otherwise IDENTITY and DMA_FQ defaults will try DMA */
 	if (iommu_def_domain_type == IOMMU_DOMAIN_DMA)
-		return NULL;
+		return ERR_PTR(-ENODEV);
 	dom = __iommu_group_alloc_default_domain(group, IOMMU_DOMAIN_DMA);
 	if (!dom)
-		return NULL;
+		return dom;
 
 	pr_warn("Failed to allocate default IOMMU domain of type %u for group %s - Falling back to IOMMU_DOMAIN_DMA",
 		iommu_def_domain_type, group->name);
@@ -2094,10 +2094,17 @@ static struct iommu_domain *__iommu_domain_alloc(const struct iommu_ops *ops,
 	else if (ops->domain_alloc)
 		domain = ops->domain_alloc(alloc_type);
 	else
-		return NULL;
+		return ERR_PTR(-EOPNOTSUPP);
 
+	/*
+	 * Many domain_alloc ops now return ERR_PTR, make things easier for the
+	 * driver by accepting ERR_PTR from all domain_alloc ops instead of
+	 * having two rules.
+	 */
+	if (IS_ERR(domain))
+		return domain;
 	if (!domain)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 	domain->type = type;
 	/*
@@ -2110,9 +2117,14 @@ static struct iommu_domain *__iommu_domain_alloc(const struct iommu_ops *ops,
 	if (!domain->ops)
 		domain->ops = ops->default_domain_ops;
 
-	if (iommu_is_dma_domain(domain) && iommu_get_dma_cookie(domain)) {
-		iommu_domain_free(domain);
-		domain = NULL;
+	if (iommu_is_dma_domain(domain)) {
+		int rc;
+
+		rc = iommu_get_dma_cookie(domain);
+		if (rc) {
+			iommu_domain_free(domain);
+			return ERR_PTR(rc);
+		}
 	}
 	return domain;
 }
@@ -2129,10 +2141,15 @@ __iommu_group_domain_alloc(struct iommu_group *group, unsigned int type)
 
 struct iommu_domain *iommu_domain_alloc(const struct bus_type *bus)
 {
+	struct iommu_domain *domain;
+
 	if (bus == NULL || bus->iommu_ops == NULL)
 		return NULL;
-	return __iommu_domain_alloc(bus->iommu_ops, NULL,
+	domain = __iommu_domain_alloc(bus->iommu_ops, NULL,
 				    IOMMU_DOMAIN_UNMANAGED);
+	if (IS_ERR(domain))
+		return NULL;
+	return domain;
 }
 EXPORT_SYMBOL_GPL(iommu_domain_alloc);
 
@@ -3041,8 +3058,8 @@ static int iommu_setup_default_domain(struct iommu_group *group,
 		return -EINVAL;
 
 	dom = iommu_group_alloc_default_domain(group, req_type);
-	if (!dom)
-		return -ENODEV;
+	if (IS_ERR(dom))
+		return PTR_ERR(dom);
 
 	if (group->default_domain == dom)
 		return 0;
@@ -3243,21 +3260,23 @@ void iommu_device_unuse_default_domain(struct device *dev)
 
 static int __iommu_group_alloc_blocking_domain(struct iommu_group *group)
 {
+	struct iommu_domain *domain;
+
 	if (group->blocking_domain)
 		return 0;
 
-	group->blocking_domain =
-		__iommu_group_domain_alloc(group, IOMMU_DOMAIN_BLOCKED);
-	if (!group->blocking_domain) {
+	domain = __iommu_group_domain_alloc(group, IOMMU_DOMAIN_BLOCKED);
+	if (IS_ERR(domain)) {
 		/*
 		 * For drivers that do not yet understand IOMMU_DOMAIN_BLOCKED
 		 * create an empty domain instead.
 		 */
-		group->blocking_domain = __iommu_group_domain_alloc(
-			group, IOMMU_DOMAIN_UNMANAGED);
-		if (!group->blocking_domain)
-			return -EINVAL;
+		domain = __iommu_group_domain_alloc(group,
+						    IOMMU_DOMAIN_UNMANAGED);
+		if (IS_ERR(domain))
+			return PTR_ERR(domain);
 	}
+	group->blocking_domain = domain;
 	return 0;
 }
 

base-commit: e8cca466a84a75f8ff2a7a31173c99ee6d1c59d2
-- 
2.42.0

Re: [PATCH] iommu: Flow ERR_PTR out from __iommu_domain_alloc()

[Jerry Snitselaar]

On Wed, Nov 01, 2023 at 02:51:02PM -0300, Jason Gunthorpe wrote:
> Most of the calling code now has error handling that can carry an error
> code further up the call chain. Keep the exported interface
> iommu_domain_alloc() returning NULL and reflow the internal code to use
> ERR_PTR not NULL for domain allocation failure.
> 
> Optionally allow drivers to return ERR_PTR from any of the alloc ops. Many
> of the new ops (user, sva, etc) already return ERR_PTR, so having two
> rules is confusing and hard on drivers. This fixes a bug in DART that was
> returning ERR_PTR.
> 
> This also fixes a bug in iommu_group_alloc_default_domain() where it
> returned both NULL and ERR_PTR, now it always returns ERR_PTR.
> 
> Fixes: 482feb5c6492 ("iommu/dart: Call apple_dart_finalize_domain() as part of alloc_paging()")
> Fixes: 1c68cbc64fe6 ("iommu: Add IOMMU_DOMAIN_PLATFORM")
> Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
> Link: https://lore.kernel.org/linux-iommu/b85e0715-3224-4f45-ad6b-ebb9f08c015d@moroto.mountain/
> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
> ---
>  drivers/iommu/iommu.c | 53 +++++++++++++++++++++++++++++--------------
>  1 file changed, 36 insertions(+), 17 deletions(-)
> 
> diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
> index f17a1113f3d6a3..6bc044d2310487 100644
> --- a/drivers/iommu/iommu.c
> +++ b/drivers/iommu/iommu.c
> @@ -1802,10 +1802,10 @@ iommu_group_alloc_default_domain(struct iommu_group *group, int req_type)
>  
>  	/* Otherwise IDENTITY and DMA_FQ defaults will try DMA */
>  	if (iommu_def_domain_type == IOMMU_DOMAIN_DMA)
> -		return NULL;
> +		return ERR_PTR(-ENODEV);
>  	dom = __iommu_group_alloc_default_domain(group, IOMMU_DOMAIN_DMA);
>  	if (!dom)

Should this be if(IS_ERR(dom)) ?

> -		return NULL;
> +		return dom;
>  
>  	pr_warn("Failed to allocate default IOMMU domain of type %u for group %s - Falling back to IOMMU_DOMAIN_DMA",
>  		iommu_def_domain_type, group->name);

Re: [PATCH] iommu: Flow ERR_PTR out from __iommu_domain_alloc()

[Jason Gunthorpe]

On Wed, Nov 01, 2023 at 11:52:08AM -0700, Jerry Snitselaar wrote:
> On Wed, Nov 01, 2023 at 02:51:02PM -0300, Jason Gunthorpe wrote:
> > Most of the calling code now has error handling that can carry an error
> > code further up the call chain. Keep the exported interface
> > iommu_domain_alloc() returning NULL and reflow the internal code to use
> > ERR_PTR not NULL for domain allocation failure.
> > 
> > Optionally allow drivers to return ERR_PTR from any of the alloc ops. Many
> > of the new ops (user, sva, etc) already return ERR_PTR, so having two
> > rules is confusing and hard on drivers. This fixes a bug in DART that was
> > returning ERR_PTR.
> > 
> > This also fixes a bug in iommu_group_alloc_default_domain() where it
> > returned both NULL and ERR_PTR, now it always returns ERR_PTR.
> > 
> > Fixes: 482feb5c6492 ("iommu/dart: Call apple_dart_finalize_domain() as part of alloc_paging()")
> > Fixes: 1c68cbc64fe6 ("iommu: Add IOMMU_DOMAIN_PLATFORM")
> > Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
> > Link: https://lore.kernel.org/linux-iommu/b85e0715-3224-4f45-ad6b-ebb9f08c015d@moroto.mountain/
> > Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
> > ---
> >  drivers/iommu/iommu.c | 53 +++++++++++++++++++++++++++++--------------
> >  1 file changed, 36 insertions(+), 17 deletions(-)
> > 
> > diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
> > index f17a1113f3d6a3..6bc044d2310487 100644
> > --- a/drivers/iommu/iommu.c
> > +++ b/drivers/iommu/iommu.c
> > @@ -1802,10 +1802,10 @@ iommu_group_alloc_default_domain(struct iommu_group *group, int req_type)
> >  
> >  	/* Otherwise IDENTITY and DMA_FQ defaults will try DMA */
> >  	if (iommu_def_domain_type == IOMMU_DOMAIN_DMA)
> > -		return NULL;
> > +		return ERR_PTR(-ENODEV);
> >  	dom = __iommu_group_alloc_default_domain(group, IOMMU_DOMAIN_DMA);
> >  	if (!dom)
> 
> Should this be if(IS_ERR(dom)) ?

Blah, I think I somehow missed this function:

@@ -1788,7 +1788,7 @@ iommu_group_alloc_default_domain(struct iommu_group *group, int req_type)
         */
        if (ops->default_domain) {
                if (req_type)
-                       return NULL;
+                       return ERR_PTR(-EINVAL);
                return ops->default_domain;
        }
 
@@ -1797,14 +1797,14 @@ iommu_group_alloc_default_domain(struct iommu_group *group, int req_type)
 
        /* The driver gave no guidance on what type to use, try the default */
        dom = __iommu_group_alloc_default_domain(group, iommu_def_domain_type);
-       if (dom)
+       if (!IS_ERR(dom))
                return dom;
 
        /* Otherwise IDENTITY and DMA_FQ defaults will try DMA */
        if (iommu_def_domain_type == IOMMU_DOMAIN_DMA)
-               return ERR_PTR(-ENODEV);
+               return ERR_PTR(-EINVAL);
        dom = __iommu_group_alloc_default_domain(group, IOMMU_DOMAIN_DMA);
-       if (!dom)
+       if (IS_ERR(dom))
                return dom;
 
        pr_warn("Failed to allocate default IOMMU domain of type %u for group %s - Falling back to IOMMU_DOMAIN_DMA",


Thanks! 
Jason

Re: [PATCH] iommu: Flow ERR_PTR out from __iommu_domain_alloc()

[Dan Carpenter]

On Wed, Nov 01, 2023 at 08:17:27PM -0300, Jason Gunthorpe wrote:
>         /* The driver gave no guidance on what type to use, try the default */
>         dom = __iommu_group_alloc_default_domain(group, iommu_def_domain_type);
> -       if (dom)
> +       if (!IS_ERR(dom))
>                 return dom;
>  
>         /* Otherwise IDENTITY and DMA_FQ defaults will try DMA */
>         if (iommu_def_domain_type == IOMMU_DOMAIN_DMA)
> -               return ERR_PTR(-ENODEV);
> +               return ERR_PTR(-EINVAL);
>         dom = __iommu_group_alloc_default_domain(group, IOMMU_DOMAIN_DMA);
> -       if (!dom)
> +       if (IS_ERR(dom))
>                 return dom;
>  
>         pr_warn("Failed to allocate default IOMMU domain of type %u for group %s - Falling back to IOMMU_DOMAIN_DMA",

Not related to your patch, but we should delete/move this warning as
well.  It's on a success path.

regards,
dan carpenter

Re: [PATCH] iommu: Flow ERR_PTR out from __iommu_domain_alloc()

[Jerry Snitselaar]

On Thu, Nov 02, 2023 at 07:10:57AM +0300, Dan Carpenter wrote:
> On Wed, Nov 01, 2023 at 08:17:27PM -0300, Jason Gunthorpe wrote:
> >         /* The driver gave no guidance on what type to use, try the default */
> >         dom = __iommu_group_alloc_default_domain(group, iommu_def_domain_type);
> > -       if (dom)
> > +       if (!IS_ERR(dom))
> >                 return dom;
> >  
> >         /* Otherwise IDENTITY and DMA_FQ defaults will try DMA */
> >         if (iommu_def_domain_type == IOMMU_DOMAIN_DMA)
> > -               return ERR_PTR(-ENODEV);
> > +               return ERR_PTR(-EINVAL);
> >         dom = __iommu_group_alloc_default_domain(group, IOMMU_DOMAIN_DMA);
> > -       if (!dom)
> > +       if (IS_ERR(dom))
> >                 return dom;
> >  
> >         pr_warn("Failed to allocate default IOMMU domain of type %u for group %s - Falling back to IOMMU_DOMAIN_DMA",
> 
> Not related to your patch, but we should delete/move this warning as
> well.  It's on a success path.
> 
> regards,
> dan carpenter
> 

It seems worthwhile to note due to the performance implications, if
nothing else, of having to fall back to the dma domain type from the
two types mentioned in the comment. I'm not sure where would be a
better place since this is first spot where it knows that is going to
be the case.

Re: [PATCH] iommu: Flow ERR_PTR out from __iommu_domain_alloc()

[Dan Carpenter]

On Wed, Nov 01, 2023 at 10:09:18PM -0700, Jerry Snitselaar wrote:
> On Thu, Nov 02, 2023 at 07:10:57AM +0300, Dan Carpenter wrote:
> > On Wed, Nov 01, 2023 at 08:17:27PM -0300, Jason Gunthorpe wrote:
> > >         /* The driver gave no guidance on what type to use, try the default */
> > >         dom = __iommu_group_alloc_default_domain(group, iommu_def_domain_type);
> > > -       if (dom)
> > > +       if (!IS_ERR(dom))
> > >                 return dom;
> > >  
> > >         /* Otherwise IDENTITY and DMA_FQ defaults will try DMA */
> > >         if (iommu_def_domain_type == IOMMU_DOMAIN_DMA)
> > > -               return ERR_PTR(-ENODEV);
> > > +               return ERR_PTR(-EINVAL);
> > >         dom = __iommu_group_alloc_default_domain(group, IOMMU_DOMAIN_DMA);
> > > -       if (!dom)
> > > +       if (IS_ERR(dom))
> > >                 return dom;
> > >  
> > >         pr_warn("Failed to allocate default IOMMU domain of type %u for group %s - Falling back to IOMMU_DOMAIN_DMA",
> > 
> > Not related to your patch, but we should delete/move this warning as
> > well.  It's on a success path.
> > 
> > regards,
> > dan carpenter
> > 
> 
> It seems worthwhile to note due to the performance implications, if
> nothing else, of having to fall back to the dma domain type from the
> two types mentioned in the comment. I'm not sure where would be a
> better place since this is first spot where it knows that is going to
> be the case.

Maybe it should be a pr_info() instead of a pr_warn()?  Maybe I'm just
wrong.

regards,
dan carpenter

Re: [PATCH] iommu: Flow ERR_PTR out from __iommu_domain_alloc()

[Jason Gunthorpe]

On Thu, Nov 02, 2023 at 08:22:13AM +0300, Dan Carpenter wrote:
> On Wed, Nov 01, 2023 at 10:09:18PM -0700, Jerry Snitselaar wrote:
> > On Thu, Nov 02, 2023 at 07:10:57AM +0300, Dan Carpenter wrote:
> > > On Wed, Nov 01, 2023 at 08:17:27PM -0300, Jason Gunthorpe wrote:
> > > >         /* The driver gave no guidance on what type to use, try the default */
> > > >         dom = __iommu_group_alloc_default_domain(group, iommu_def_domain_type);
> > > > -       if (dom)
> > > > +       if (!IS_ERR(dom))
> > > >                 return dom;
> > > >  
> > > >         /* Otherwise IDENTITY and DMA_FQ defaults will try DMA */
> > > >         if (iommu_def_domain_type == IOMMU_DOMAIN_DMA)
> > > > -               return ERR_PTR(-ENODEV);
> > > > +               return ERR_PTR(-EINVAL);
> > > >         dom = __iommu_group_alloc_default_domain(group, IOMMU_DOMAIN_DMA);
> > > > -       if (!dom)
> > > > +       if (IS_ERR(dom))
> > > >                 return dom;
> > > >  
> > > >         pr_warn("Failed to allocate default IOMMU domain of type %u for group %s - Falling back to IOMMU_DOMAIN_DMA",
> > > 
> > > Not related to your patch, but we should delete/move this warning as
> > > well.  It's on a success path.
> > > 
> > > regards,
> > > dan carpenter
> > > 
> > 
> > It seems worthwhile to note due to the performance implications, if
> > nothing else, of having to fall back to the dma domain type from the
> > two types mentioned in the comment. I'm not sure where would be a
> > better place since this is first spot where it knows that is going to
> > be the case.
> 
> Maybe it should be a pr_info() instead of a pr_warn()?  Maybe I'm just
> wrong.

TBH, I'm not entirely sure what it is for, it has been there a long
time.. I think it means that kernel config and command line options
are incompatible with this iommu driver

But at this moment there are very few drivers that don't support
IDENTITY (s390 notably, I suppose) and the DMA_FQ stuff is no longer
driver visible, so it probably never triggers..

Jason

Re: [PATCH] iommu: Flow ERR_PTR out from __iommu_domain_alloc()

[Robin Murphy]

On 2023-11-02 12:12 pm, Jason Gunthorpe wrote:
> On Thu, Nov 02, 2023 at 08:22:13AM +0300, Dan Carpenter wrote:
>> On Wed, Nov 01, 2023 at 10:09:18PM -0700, Jerry Snitselaar wrote:
>>> On Thu, Nov 02, 2023 at 07:10:57AM +0300, Dan Carpenter wrote:
>>>> On Wed, Nov 01, 2023 at 08:17:27PM -0300, Jason Gunthorpe wrote:
>>>>>          /* The driver gave no guidance on what type to use, try the default */
>>>>>          dom = __iommu_group_alloc_default_domain(group, iommu_def_domain_type);
>>>>> -       if (dom)
>>>>> +       if (!IS_ERR(dom))
>>>>>                  return dom;
>>>>>   
>>>>>          /* Otherwise IDENTITY and DMA_FQ defaults will try DMA */
>>>>>          if (iommu_def_domain_type == IOMMU_DOMAIN_DMA)
>>>>> -               return ERR_PTR(-ENODEV);
>>>>> +               return ERR_PTR(-EINVAL);
>>>>>          dom = __iommu_group_alloc_default_domain(group, IOMMU_DOMAIN_DMA);
>>>>> -       if (!dom)
>>>>> +       if (IS_ERR(dom))
>>>>>                  return dom;
>>>>>   
>>>>>          pr_warn("Failed to allocate default IOMMU domain of type %u for group %s - Falling back to IOMMU_DOMAIN_DMA",
>>>>
>>>> Not related to your patch, but we should delete/move this warning as
>>>> well.  It's on a success path.
>>>>
>>>> regards,
>>>> dan carpenter
>>>>
>>>
>>> It seems worthwhile to note due to the performance implications, if
>>> nothing else, of having to fall back to the dma domain type from the
>>> two types mentioned in the comment. I'm not sure where would be a
>>> better place since this is first spot where it knows that is going to
>>> be the case.
>>
>> Maybe it should be a pr_info() instead of a pr_warn()?  Maybe I'm just
>> wrong.
> 
> TBH, I'm not entirely sure what it is for, it has been there a long
> time.. I think it means that kernel config and command line options
> are incompatible with this iommu driver

Yes, it's a failure path in the sense that the user has requested 
something (via config or command line) that the IOMMU driver cannot 
provide. Falling back to allocating something different is more "damage 
limitation" than "success".

> But at this moment there are very few drivers that don't support
> IDENTITY (s390 notably, I suppose) and the DMA_FQ stuff is no longer
> driver visible, so it probably never triggers..

Indeed if we get to the point where any platform/driver combination that 
would support IOMMU_DOMAIN_DMA also supports IOMMU_DOMAIN_IDENTITY 
anyway, then this fallback is no longer useful (and we could 
consequently also simplify some stuff further out), however as things 
stand I see at least apple-dart remaining a thorn in the side of that plan.

Thanks,
Robin.

Re: [PATCH] iommu: Flow ERR_PTR out from __iommu_domain_alloc()

[Jerry Snitselaar]

On Thu, Nov 02, 2023 at 01:09:50PM +0000, Robin Murphy wrote:
> On 2023-11-02 12:12 pm, Jason Gunthorpe wrote:
> > On Thu, Nov 02, 2023 at 08:22:13AM +0300, Dan Carpenter wrote:
> > > On Wed, Nov 01, 2023 at 10:09:18PM -0700, Jerry Snitselaar wrote:
> > > > On Thu, Nov 02, 2023 at 07:10:57AM +0300, Dan Carpenter wrote:
> > > > > On Wed, Nov 01, 2023 at 08:17:27PM -0300, Jason Gunthorpe wrote:
> > > > > >          /* The driver gave no guidance on what type to use, try the default */
> > > > > >          dom = __iommu_group_alloc_default_domain(group, iommu_def_domain_type);
> > > > > > -       if (dom)
> > > > > > +       if (!IS_ERR(dom))
> > > > > >                  return dom;
> > > > > >          /* Otherwise IDENTITY and DMA_FQ defaults will try DMA */
> > > > > >          if (iommu_def_domain_type == IOMMU_DOMAIN_DMA)
> > > > > > -               return ERR_PTR(-ENODEV);
> > > > > > +               return ERR_PTR(-EINVAL);
> > > > > >          dom = __iommu_group_alloc_default_domain(group, IOMMU_DOMAIN_DMA);
> > > > > > -       if (!dom)
> > > > > > +       if (IS_ERR(dom))
> > > > > >                  return dom;
> > > > > >          pr_warn("Failed to allocate default IOMMU domain of type %u for group %s - Falling back to IOMMU_DOMAIN_DMA",
> > > > > 
> > > > > Not related to your patch, but we should delete/move this warning as
> > > > > well.  It's on a success path.
> > > > > 
> > > > > regards,
> > > > > dan carpenter
> > > > > 
> > > > 
> > > > It seems worthwhile to note due to the performance implications, if
> > > > nothing else, of having to fall back to the dma domain type from the
> > > > two types mentioned in the comment. I'm not sure where would be a
> > > > better place since this is first spot where it knows that is going to
> > > > be the case.
> > > 
> > > Maybe it should be a pr_info() instead of a pr_warn()?  Maybe I'm just
> > > wrong.
> > 
> > TBH, I'm not entirely sure what it is for, it has been there a long
> > time.. I think it means that kernel config and command line options
> > are incompatible with this iommu driver
> 
> Yes, it's a failure path in the sense that the user has requested something
> (via config or command line) that the IOMMU driver cannot provide. Falling
> back to allocating something different is more "damage limitation" than
> "success".
> 
> > But at this moment there are very few drivers that don't support
> > IDENTITY (s390 notably, I suppose) and the DMA_FQ stuff is no longer
> > driver visible, so it probably never triggers..
> 
> Indeed if we get to the point where any platform/driver combination that
> would support IOMMU_DOMAIN_DMA also supports IOMMU_DOMAIN_IDENTITY anyway,
> then this fallback is no longer useful (and we could consequently also
> simplify some stuff further out), however as things stand I see at least
> apple-dart remaining a thorn in the side of that plan.
> 
> Thanks,
> Robin.

Btw Robin, I was able to verify the PCI SAC change works well in a
case recently.  Went from a drop off of 40Gbps to 1Gbps, to no
noticeable drop off.

Regards,
Jerry