Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1734035 > unrolled thread

[PATCH] iommu/ipmmu-vmsa: Do not replace bus IOMMU ops on driver init.

Back to article view | Back to linux.kernel

Contents

  [PATCH] iommu/ipmmu-vmsa: Do not replace bus IOMMU ops on driver init. Liviu Dudau <Liviu.Dudau@arm.com> - 2017-09-18 13:10 +0200
    Re: [PATCH] iommu/ipmmu-vmsa: Do not replace bus IOMMU ops on driver init. Laurent Pinchart <laurent.pinchart@ideasonboard.com> - 2017-09-19 09:10 +0200
      Re: [PATCH] iommu/ipmmu-vmsa: Do not replace bus IOMMU ops on driver  init. Liviu Dudau <Liviu.Dudau@arm.com> - 2017-09-19 11:50 +0200
      Re: [PATCH] iommu/ipmmu-vmsa: Do not replace bus IOMMU ops on driver  init. Robin Murphy <robin.murphy@arm.com> - 2017-09-19 12:50 +0200
    [PATCH v2] iommu/ipmmu-vmsa: Do not replace bus IOMMU ops on driver init. Liviu Dudau <Liviu.Dudau@arm.com> - 2017-09-20 16:20 +0200

#1734035 — [PATCH] iommu/ipmmu-vmsa: Do not replace bus IOMMU ops on driver init.

If the IPMMU driver is compiled in the kernel it will replace the
platform bus IOMMU ops on running the ipmmu_init() function, regardless
if there is any IPMMU hardware present or not. This screws up systems
that just want to build a generic kernel that runs on multiple platforms
and use a different IOMMU implementation.

Move the bus_set_iommu() call at the end of the ipmmu_probe() function
when we know that hardware is present. With current IOMMU framework it
should be safe (at least for OF case).

Now that the ipmmu_init() and ipmmu_exit() functions are simple calls to
platform_driver_register() and platform_driver_unregister(), replace
them with the module_platform_driver() macro call.

Signed-off-by: Liviu Dudau <Liviu.Dudau@arm.com>
Cc: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
---
 drivers/iommu/ipmmu-vmsa.c | 29 +++++------------------------
 1 file changed, 5 insertions(+), 24 deletions(-)

diff --git a/drivers/iommu/ipmmu-vmsa.c b/drivers/iommu/ipmmu-vmsa.c
index 2a38aa15be17d..c60569c74678d 100644
--- a/drivers/iommu/ipmmu-vmsa.c
+++ b/drivers/iommu/ipmmu-vmsa.c
@@ -1055,10 +1055,11 @@ static int ipmmu_probe(struct platform_device *pdev)
 	ipmmu_device_reset(mmu);
 
 	/*
-	 * We can't create the ARM mapping here as it requires the bus to have
-	 * an IOMMU, which only happens when bus_set_iommu() is called in
-	 * ipmmu_init() after the probe function returns.
+	 * Now that we have validated the presence of the hardware, set
+	 * the bus IOMMU ops to enable future domain and device setup.
 	 */
+	if (!iommu_present(&platform_bus_type))
+		bus_set_iommu(&platform_bus_type, &ipmmu_ops);
 
 	spin_lock(&ipmmu_devices_lock);
 	list_add(&mmu->list, &ipmmu_devices);
@@ -1100,27 +1101,7 @@ static struct platform_driver ipmmu_driver = {
 	.remove	= ipmmu_remove,
 };
 
-static int __init ipmmu_init(void)
-{
-	int ret;
-
-	ret = platform_driver_register(&ipmmu_driver);
-	if (ret < 0)
-		return ret;
-
-	if (!iommu_present(&platform_bus_type))
-		bus_set_iommu(&platform_bus_type, &ipmmu_ops);
-
-	return 0;
-}
-
-static void __exit ipmmu_exit(void)
-{
-	return platform_driver_unregister(&ipmmu_driver);
-}
-
-subsys_initcall(ipmmu_init);
-module_exit(ipmmu_exit);
+module_platform_driver(ipmmu_driver);
 
 MODULE_DESCRIPTION("IOMMU API for Renesas VMSA-compatible IPMMU");
 MODULE_AUTHOR("Laurent Pinchart <laurent.pinchart@ideasonboard.com>");
-- 
2.14.1

[toc] | [next] | [standalone]

#1734697

Hi Liviu,

Thank you for the patch.

On Monday, 18 September 2017 13:04:44 EEST Liviu Dudau wrote:
> If the IPMMU driver is compiled in the kernel it will replace the
> platform bus IOMMU ops on running the ipmmu_init() function, regardless
> if there is any IPMMU hardware present or not. This screws up systems
> that just want to build a generic kernel that runs on multiple platforms
> and use a different IOMMU implementation.
> 
> Move the bus_set_iommu() call at the end of the ipmmu_probe() function
> when we know that hardware is present. With current IOMMU framework it
> should be safe (at least for OF case).

If I recall correctly the issue is that the IPMMU might be probed after bus 
master devices when using the legacy IOMMU DT integration, which the ipmmu-
vmsa driver does on ARM32. Calling bus_set_iommu() from the probe function 
will then result in some devices losing IOMMU support.

> Now that the ipmmu_init() and ipmmu_exit() functions are simple calls to
> platform_driver_register() and platform_driver_unregister(), replace
> them with the module_platform_driver() macro call.
> 
> Signed-off-by: Liviu Dudau <Liviu.Dudau@arm.com>
> Cc: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
> ---
>  drivers/iommu/ipmmu-vmsa.c | 29 +++++------------------------
>  1 file changed, 5 insertions(+), 24 deletions(-)
> 
> diff --git a/drivers/iommu/ipmmu-vmsa.c b/drivers/iommu/ipmmu-vmsa.c
> index 2a38aa15be17d..c60569c74678d 100644
> --- a/drivers/iommu/ipmmu-vmsa.c
> +++ b/drivers/iommu/ipmmu-vmsa.c
> @@ -1055,10 +1055,11 @@ static int ipmmu_probe(struct platform_device *pdev)
> ipmmu_device_reset(mmu);
> 
>  	/*
> -	 * We can't create the ARM mapping here as it requires the bus to have
> -	 * an IOMMU, which only happens when bus_set_iommu() is called in
> -	 * ipmmu_init() after the probe function returns.
> +	 * Now that we have validated the presence of the hardware, set
> +	 * the bus IOMMU ops to enable future domain and device setup.
>  	 */
> +	if (!iommu_present(&platform_bus_type))
> +		bus_set_iommu(&platform_bus_type, &ipmmu_ops);
> 
>  	spin_lock(&ipmmu_devices_lock);
>  	list_add(&mmu->list, &ipmmu_devices);

What branch is this patch based on ? ipmmu_devices_lock isn't in mainline.

> @@ -1100,27 +1101,7 @@ static struct platform_driver ipmmu_driver = {
>  	.remove	= ipmmu_remove,
>  };
> 
> -static int __init ipmmu_init(void)
> -{
> -	int ret;
> -
> -	ret = platform_driver_register(&ipmmu_driver);
> -	if (ret < 0)
> -		return ret;
> -
> -	if (!iommu_present(&platform_bus_type))
> -		bus_set_iommu(&platform_bus_type, &ipmmu_ops);
> -
> -	return 0;
> -}
> -
> -static void __exit ipmmu_exit(void)
> -{
> -	return platform_driver_unregister(&ipmmu_driver);
> -}
> -
> -subsys_initcall(ipmmu_init);
> -module_exit(ipmmu_exit);
> +module_platform_driver(ipmmu_driver);
> 
>  MODULE_DESCRIPTION("IOMMU API for Renesas VMSA-compatible IPMMU");
>  MODULE_AUTHOR("Laurent Pinchart <laurent.pinchart@ideasonboard.com>");

-- 
Regards,

Laurent Pinchart

[toc] | [prev] | [next] | [standalone]

#1734774 — Re: [PATCH] iommu/ipmmu-vmsa: Do not replace bus IOMMU ops on driver init.

On Tue, Sep 19, 2017 at 10:07:58AM +0300, Laurent Pinchart wrote:
> Hi Liviu,
> 
> Thank you for the patch.
> 
> On Monday, 18 September 2017 13:04:44 EEST Liviu Dudau wrote:
> > If the IPMMU driver is compiled in the kernel it will replace the
> > platform bus IOMMU ops on running the ipmmu_init() function, regardless
> > if there is any IPMMU hardware present or not. This screws up systems
> > that just want to build a generic kernel that runs on multiple platforms
> > and use a different IOMMU implementation.
> > 
> > Move the bus_set_iommu() call at the end of the ipmmu_probe() function
> > when we know that hardware is present. With current IOMMU framework it
> > should be safe (at least for OF case).
> 
> If I recall correctly the issue is that the IPMMU might be probed after bus 
> master devices when using the legacy IOMMU DT integration, which the ipmmu-
> vmsa driver does on ARM32. Calling bus_set_iommu() from the probe function 
> will then result in some devices losing IOMMU support.

This is on arm64, on the Arm Juno board.

> 
> > Now that the ipmmu_init() and ipmmu_exit() functions are simple calls to
> > platform_driver_register() and platform_driver_unregister(), replace
> > them with the module_platform_driver() macro call.
> > 
> > Signed-off-by: Liviu Dudau <Liviu.Dudau@arm.com>
> > Cc: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
> > ---
> >  drivers/iommu/ipmmu-vmsa.c | 29 +++++------------------------
> >  1 file changed, 5 insertions(+), 24 deletions(-)
> > 
> > diff --git a/drivers/iommu/ipmmu-vmsa.c b/drivers/iommu/ipmmu-vmsa.c
> > index 2a38aa15be17d..c60569c74678d 100644
> > --- a/drivers/iommu/ipmmu-vmsa.c
> > +++ b/drivers/iommu/ipmmu-vmsa.c
> > @@ -1055,10 +1055,11 @@ static int ipmmu_probe(struct platform_device *pdev)
> > ipmmu_device_reset(mmu);
> > 
> >  	/*
> > -	 * We can't create the ARM mapping here as it requires the bus to have
> > -	 * an IOMMU, which only happens when bus_set_iommu() is called in
> > -	 * ipmmu_init() after the probe function returns.
> > +	 * Now that we have validated the presence of the hardware, set
> > +	 * the bus IOMMU ops to enable future domain and device setup.
> >  	 */
> > +	if (!iommu_present(&platform_bus_type))
> > +		bus_set_iommu(&platform_bus_type, &ipmmu_ops);
> > 
> >  	spin_lock(&ipmmu_devices_lock);
> >  	list_add(&mmu->list, &ipmmu_devices);
> 
> What branch is this patch based on ? ipmmu_devices_lock isn't in mainline.

drm-next. :)
Looks like things have move forward since I've made the patch before
LPC, will update and resend.

Best regards,
Liviu

> 
> > @@ -1100,27 +1101,7 @@ static struct platform_driver ipmmu_driver = {
> >  	.remove	= ipmmu_remove,
> >  };
> > 
> > -static int __init ipmmu_init(void)
> > -{
> > -	int ret;
> > -
> > -	ret = platform_driver_register(&ipmmu_driver);
> > -	if (ret < 0)
> > -		return ret;
> > -
> > -	if (!iommu_present(&platform_bus_type))
> > -		bus_set_iommu(&platform_bus_type, &ipmmu_ops);
> > -
> > -	return 0;
> > -}
> > -
> > -static void __exit ipmmu_exit(void)
> > -{
> > -	return platform_driver_unregister(&ipmmu_driver);
> > -}
> > -
> > -subsys_initcall(ipmmu_init);
> > -module_exit(ipmmu_exit);
> > +module_platform_driver(ipmmu_driver);
> > 
> >  MODULE_DESCRIPTION("IOMMU API for Renesas VMSA-compatible IPMMU");
> >  MODULE_AUTHOR("Laurent Pinchart <laurent.pinchart@ideasonboard.com>");
> 
> -- 
> Regards,
> 
> Laurent Pinchart
> 

-- 
====================
| I would like to |
| fix the world,  |
| but they're not |
| giving me the   |
 \ source code!  /
  ---------------
    ¯\_(ツ)_/¯

[toc] | [prev] | [next] | [standalone]

#1734808 — Re: [PATCH] iommu/ipmmu-vmsa: Do not replace bus IOMMU ops on driver init.

Hi Laurent,

On 19/09/17 08:07, Laurent Pinchart wrote:
> Hi Liviu,
> 
> Thank you for the patch.
> 
> On Monday, 18 September 2017 13:04:44 EEST Liviu Dudau wrote:
>> If the IPMMU driver is compiled in the kernel it will replace the
>> platform bus IOMMU ops on running the ipmmu_init() function, regardless
>> if there is any IPMMU hardware present or not. This screws up systems
>> that just want to build a generic kernel that runs on multiple platforms
>> and use a different IOMMU implementation.
>>
>> Move the bus_set_iommu() call at the end of the ipmmu_probe() function
>> when we know that hardware is present. With current IOMMU framework it
>> should be safe (at least for OF case).
> 
> If I recall correctly the issue is that the IPMMU might be probed after bus 
> master devices when using the legacy IOMMU DT integration, which the ipmmu-
> vmsa driver does on ARM32. Calling bus_set_iommu() from the probe function 
> will then result in some devices losing IOMMU support.

Yes, that used to be the problem, but since IPMMU uses the generic
bindings and iommu_device_register(), it should now benefit from client
probe-deferral even when it doesn't support of_xlate().

Robin.

>> Now that the ipmmu_init() and ipmmu_exit() functions are simple calls to
>> platform_driver_register() and platform_driver_unregister(), replace
>> them with the module_platform_driver() macro call.
>>
>> Signed-off-by: Liviu Dudau <Liviu.Dudau@arm.com>
>> Cc: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
>> ---
>>  drivers/iommu/ipmmu-vmsa.c | 29 +++++------------------------
>>  1 file changed, 5 insertions(+), 24 deletions(-)
>>
>> diff --git a/drivers/iommu/ipmmu-vmsa.c b/drivers/iommu/ipmmu-vmsa.c
>> index 2a38aa15be17d..c60569c74678d 100644
>> --- a/drivers/iommu/ipmmu-vmsa.c
>> +++ b/drivers/iommu/ipmmu-vmsa.c
>> @@ -1055,10 +1055,11 @@ static int ipmmu_probe(struct platform_device *pdev)
>> ipmmu_device_reset(mmu);
>>
>>  	/*
>> -	 * We can't create the ARM mapping here as it requires the bus to have
>> -	 * an IOMMU, which only happens when bus_set_iommu() is called in
>> -	 * ipmmu_init() after the probe function returns.
>> +	 * Now that we have validated the presence of the hardware, set
>> +	 * the bus IOMMU ops to enable future domain and device setup.
>>  	 */
>> +	if (!iommu_present(&platform_bus_type))
>> +		bus_set_iommu(&platform_bus_type, &ipmmu_ops);
>>
>>  	spin_lock(&ipmmu_devices_lock);
>>  	list_add(&mmu->list, &ipmmu_devices);
> 
> What branch is this patch based on ? ipmmu_devices_lock isn't in mainline.
> 
>> @@ -1100,27 +1101,7 @@ static struct platform_driver ipmmu_driver = {
>>  	.remove	= ipmmu_remove,
>>  };
>>
>> -static int __init ipmmu_init(void)
>> -{
>> -	int ret;
>> -
>> -	ret = platform_driver_register(&ipmmu_driver);
>> -	if (ret < 0)
>> -		return ret;
>> -
>> -	if (!iommu_present(&platform_bus_type))
>> -		bus_set_iommu(&platform_bus_type, &ipmmu_ops);
>> -
>> -	return 0;
>> -}
>> -
>> -static void __exit ipmmu_exit(void)
>> -{
>> -	return platform_driver_unregister(&ipmmu_driver);
>> -}
>> -
>> -subsys_initcall(ipmmu_init);
>> -module_exit(ipmmu_exit);
>> +module_platform_driver(ipmmu_driver);
>>
>>  MODULE_DESCRIPTION("IOMMU API for Renesas VMSA-compatible IPMMU");
>>  MODULE_AUTHOR("Laurent Pinchart <laurent.pinchart@ideasonboard.com>");
> 

[toc] | [prev] | [next] | [standalone]

#1735791 — [PATCH v2] iommu/ipmmu-vmsa: Do not replace bus IOMMU ops on driver init.

If the IPMMU driver is compiled in the kernel it will replace the
platform bus IOMMU ops on running the ipmmu_init() function, regardless
if there is any IPMMU hardware present or not. This screws up systems
that just want to build a generic kernel that runs on multiple platforms
and use a different IOMMU implementation.

Move the bus_set_iommu() call at the end of the ipmmu_probe() function
when we know that hardware is present. With current IOMMU framework it
should be safe (at least for OF case).

Now that the ipmmu_init() and ipmmu_exit() functions are simple calls to
platform_driver_register() and platform_driver_unregister(), replace
them with the module_platform_driver() macro call.

Signed-off-by: Liviu Dudau <Liviu.Dudau@arm.com>
Cc: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
---
 drivers/iommu/ipmmu-vmsa.c | 29 +++++------------------------
 1 file changed, 5 insertions(+), 24 deletions(-)

diff --git a/drivers/iommu/ipmmu-vmsa.c b/drivers/iommu/ipmmu-vmsa.c
index 195d6e93ac718..31912997bffdf 100644
--- a/drivers/iommu/ipmmu-vmsa.c
+++ b/drivers/iommu/ipmmu-vmsa.c
@@ -966,10 +966,11 @@ static int ipmmu_probe(struct platform_device *pdev)
 		return ret;
 
 	/*
-	 * We can't create the ARM mapping here as it requires the bus to have
-	 * an IOMMU, which only happens when bus_set_iommu() is called in
-	 * ipmmu_init() after the probe function returns.
+	 * Now that we have validated the presence of the hardware, set
+	 * the bus IOMMU ops to enable future domain and device setup.
 	 */
+	if (!iommu_present(&platform_bus_type))
+		bus_set_iommu(&platform_bus_type, &ipmmu_ops);
 
 	platform_set_drvdata(pdev, mmu);
 
@@ -1006,27 +1007,7 @@ static struct platform_driver ipmmu_driver = {
 	.remove	= ipmmu_remove,
 };
 
-static int __init ipmmu_init(void)
-{
-	int ret;
-
-	ret = platform_driver_register(&ipmmu_driver);
-	if (ret < 0)
-		return ret;
-
-	if (!iommu_present(&platform_bus_type))
-		bus_set_iommu(&platform_bus_type, &ipmmu_ops);
-
-	return 0;
-}
-
-static void __exit ipmmu_exit(void)
-{
-	return platform_driver_unregister(&ipmmu_driver);
-}
-
-subsys_initcall(ipmmu_init);
-module_exit(ipmmu_exit);
+module_platform_driver(ipmmu_driver);
 
 MODULE_DESCRIPTION("IOMMU API for Renesas VMSA-compatible IPMMU");
 MODULE_AUTHOR("Laurent Pinchart <laurent.pinchart@ideasonboard.com>");
-- 
2.14.1

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web