On 09/19/2012 08:52:27 AM, Kumar Gala wrote:
=20
On Sep 19, 2012, at 8:17 AM, [off-list ref] =20
[off-list ref] wrote:
=20
quoted
From: Varun Sethi <redacted>
Added the following domain attributes required by FSL PAMU driver:
1. Subwindows field added to the iommu domain geometry attribute.
2. Added new iommu stash attribute, which allows setting of the
LIODN specific stash id parameter through IOMMU API.
3. Added an attribute for enabling/disabling DMA to a particular
memory window.
Signed-off-by: Varun Sethi <redacted>
---
include/linux/iommu.h | 30 ++++++++++++++++++++++++++++++
1 files changed, 30 insertions(+), 0 deletions(-)
diff --git a/include/linux/iommu.h b/include/linux/iommu.h
index 7e83370..eaa40c6 100644
--- a/include/linux/iommu.h
+++ b/include/linux/iommu.h
@@ -44,6 +44,28 @@ struct iommu_domain_geometry { dma_addr_t aperture_start; /* First address that can be =20
mapped */
quoted
dma_addr_t aperture_end; /* Last address that can be =20
mapped */
quoted
bool force_aperture; /* DMA only allowed in mappable =20
range? */
quoted
+
+ /* The subwindows field indicates number of DMA subwindows =20
supported
quoted
+ * by the geometry. Following is the interpretation of
+ * values for this field:
+ * 0 : This implies that the supported geometry size is 1 MB
+ * with each subwindow size being 4KB. Thus number of =20
subwindows
quoted
+ * being =3D 1MB/4KB =3D 256.
+ * 1 : Only one DMA window i.e. no subwindows.
+ * value other than 0 or 1 would indicate actual number of =20
subwindows.
quoted
+ */
+ u32 subwindows;
+};
+
+/* This attribute corresponds to IOMMUs capable of generating
+ * a stash transaction. A stash transaction is typically a
+ * hardware initiated prefetch of data from memory to cache.
+ * This attribute allows configuring stashig specific parameters
+ * in the IOMMU hardware.
+ */
+struct iommu_stash_attribute {
+ u32 cpu; /* cpu number */
+ u32 cache; /* cache to stash to: L1,L2,L3 */
=20
seems like this should be enum instead of u32 for cache
=20
With enum being something like:
=20
enum iommu_attr_stash_cache {
IOMMU_ATTR_CACHE_L1,
IOMMU_ATTR_CACHE_L2,
IOMMU_ATTR_CACHE_L3,
};
Don't we want these structs to be usable via some VFIO ioctl? In that =20
case they need to use fixed size types.
-Scott=
-----Original Message-----
From: Wood Scott-B07421
Sent: Thursday, September 20, 2012 5:42 AM
To: Kumar Gala
Cc: Sethi Varun-B16395; joerg.roedel@amd.com; iommu@lists.linux-
foundation.org; linuxppc-dev@lists.ozlabs.org; linux-
kernel@vger.kernel.org; Sethi Varun-B16395
Subject: Re: [RFC][PATCH 2/3] iommu/fsl: Add iommu domain attributes
required by fsl PAMU driver.
=20
On 09/19/2012 08:52:27 AM, Kumar Gala wrote:
quoted
On Sep 19, 2012, at 8:17 AM, [off-list ref]
[off-list ref] wrote:
quoted
From: Varun Sethi <redacted>
Added the following domain attributes required by FSL PAMU driver:
1. Subwindows field added to the iommu domain geometry attribute.
2. Added new iommu stash attribute, which allows setting of the
LIODN specific stash id parameter through IOMMU API.
3. Added an attribute for enabling/disabling DMA to a particular
memory window.
Signed-off-by: Varun Sethi <redacted>
---
include/linux/iommu.h | 30 ++++++++++++++++++++++++++++++
1 files changed, 30 insertions(+), 0 deletions(-)
diff --git a/include/linux/iommu.h b/include/linux/iommu.h index
7e83370..eaa40c6 100644
--- a/include/linux/iommu.h
+++ b/include/linux/iommu.h
@@ -44,6 +44,28 @@ struct iommu_domain_geometry { dma_addr_t aperture_start; /* First address that can be
mapped */
quoted
dma_addr_t aperture_end; /* Last address that can be
mapped */
quoted
bool force_aperture; /* DMA only allowed in mappable
range? */
quoted
+
+ /* The subwindows field indicates number of DMA subwindows
supported
quoted
+ * by the geometry. Following is the interpretation of
+ * values for this field:
+ * 0 : This implies that the supported geometry size is 1 MB
+ * with each subwindow size being 4KB. Thus number of
subwindows
quoted
+ * being =3D 1MB/4KB =3D 256.
+ * 1 : Only one DMA window i.e. no subwindows.
+ * value other than 0 or 1 would indicate actual number of
subwindows.
quoted
+ */
+ u32 subwindows;
+};
+
+/* This attribute corresponds to IOMMUs capable of generating
+ * a stash transaction. A stash transaction is typically a
+ * hardware initiated prefetch of data from memory to cache.
+ * This attribute allows configuring stashig specific parameters
+ * in the IOMMU hardware.
+ */
+struct iommu_stash_attribute {
+ u32 cpu; /* cpu number */
+ u32 cache; /* cache to stash to: L1,L2,L3 */
seems like this should be enum instead of u32 for cache
With enum being something like:
enum iommu_attr_stash_cache {
IOMMU_ATTR_CACHE_L1,
IOMMU_ATTR_CACHE_L2,
IOMMU_ATTR_CACHE_L3,
};
=20
Don't we want these structs to be usable via some VFIO ioctl? In that
case they need to use fixed size types.
=20
Yes, this would be usable via vfio ioctl. But, then the caller should be
aware of supported stash targets. May be I should add an interface for the =
caller,
to query supported stash targets.
-Varun
On Sep 20, 2012, at 4:46 AM, Sethi Varun-B16395 wrote:
=20
=20
quoted
-----Original Message-----
From: Wood Scott-B07421
Sent: Thursday, September 20, 2012 5:42 AM
To: Kumar Gala
Cc: Sethi Varun-B16395; joerg.roedel@amd.com; iommu@lists.linux-
foundation.org; linuxppc-dev@lists.ozlabs.org; linux-
kernel@vger.kernel.org; Sethi Varun-B16395
Subject: Re: [RFC][PATCH 2/3] iommu/fsl: Add iommu domain attributes
required by fsl PAMU driver.
=20
On 09/19/2012 08:52:27 AM, Kumar Gala wrote:
quoted
=20
On Sep 19, 2012, at 8:17 AM, [off-list ref]
[off-list ref] wrote:
=20
quoted
From: Varun Sethi <redacted>
=20
Added the following domain attributes required by FSL PAMU driver:
1. Subwindows field added to the iommu domain geometry attribute.
2. Added new iommu stash attribute, which allows setting of the
LIODN specific stash id parameter through IOMMU API.
3. Added an attribute for enabling/disabling DMA to a particular
memory window.
=20
Signed-off-by: Varun Sethi <redacted>
---
include/linux/iommu.h | 30 ++++++++++++++++++++++++++++++
1 files changed, 30 insertions(+), 0 deletions(-)
=20
diff --git a/include/linux/iommu.h b/include/linux/iommu.h index
7e83370..eaa40c6 100644
--- a/include/linux/iommu.h
+++ b/include/linux/iommu.h
@@ -44,6 +44,28 @@ struct iommu_domain_geometry { dma_addr_t aperture_start; /* First address that can be
mapped */
quoted
dma_addr_t aperture_end; /* Last address that can be
mapped */
quoted
bool force_aperture; /* DMA only allowed in mappable
range? */
quoted
+
+ /* The subwindows field indicates number of DMA subwindows
supported
quoted
+ * by the geometry. Following is the interpretation of
+ * values for this field:
+ * 0 : This implies that the supported geometry size is 1 MB
+ * with each subwindow size being 4KB. Thus number of
subwindows
quoted
+ * being =3D 1MB/4KB =3D 256.
+ * 1 : Only one DMA window i.e. no subwindows.
+ * value other than 0 or 1 would indicate actual number of
subwindows.
quoted
+ */
+ u32 subwindows;
+};
+
+/* This attribute corresponds to IOMMUs capable of generating
+ * a stash transaction. A stash transaction is typically a
+ * hardware initiated prefetch of data from memory to cache.
+ * This attribute allows configuring stashig specific parameters
+ * in the IOMMU hardware.
+ */
+struct iommu_stash_attribute {
+ u32 cpu; /* cpu number */
+ u32 cache; /* cache to stash to: L1,L2,L3 */
=20
seems like this should be enum instead of u32 for cache
=20
With enum being something like:
=20
enum iommu_attr_stash_cache {
IOMMU_ATTR_CACHE_L1,
IOMMU_ATTR_CACHE_L2,
IOMMU_ATTR_CACHE_L3,
};
=20
Don't we want these structs to be usable via some VFIO ioctl? In =
that
quoted
case they need to use fixed size types.
=20
Yes, this would be usable via vfio ioctl. But, then the caller should =
be
aware of supported stash targets. May be I should add an interface for =
the caller,
to query supported stash targets.
Guess the caller probably knows, but thinking we should move the =
#defines for valid values into this file out of pamu specific files.
- k=