[PATCH char-misc-next v2 00/22] misc: mic: Enable COSM and remaining SCIF functionality

[PATCH char-misc-next v2 00/22] misc: mic: Enable COSM and remaining SCIF functionality

[Ashutosh Dixit]

Changelog:
=========
v1 => v2:
* Use existing patches for IOMMU IOVA separation by Sakari Alius.
  These patches are being sent via char-misc since this patch
  series has a dependency on them. The IOMMU patches can be
  dropped once they are merged in via the iommu tree
* Add new dmaengine API instead of overloading the memcpy API
* Bug fixes and checkpatch cleanups since last post

v1: Initial post @ https://lkml.org/lkml/2015/7/27/958

The Symmetric Communication Interface (SCIF) API was accepted upstream
during the v4.2 merge window @ https://lkml.org/lkml/2015/3/30/865.
This patch series completes the implementation of the SCIF API and
also introduces the MIC Coprocessor State Management (COSM) drivers.

The first three patches break the IOMMU IOVA out of the IOMMU and make
it into a separate module. These patches were submitted in the IOMMU
tree but are being resent via char-misc since they are not available
upstream and the SCIF patches depend on them. The first three patches
can be dropped if they have been merged via the IOMMU tree.

Patch 4 solves a build issue with IOMMU IOVA and patch 5 extends the
DMA engine API to support MIC X100 status descriptor writes.

Patch 6 introduces SCIF poll which allows user and kernel mode clients
to block for events on a SCIF endpoint. Patch 7 introduces support for
SCIF kernel mode clients. SCIF clients are probed when new SCIF nodes
come online. Similarly the client remove method is called as SCIF
nodes go offline. The SCIF client framework makes use of the SCIF peer
bus infrastructure submitted previously.

Patches 8 through 14 are a re-factoring of functionality for boot,
shutdown and reset of MIC cards. This functionality, called COSM, was
previously present in the MIC host driver but has now been re-factored
into a separate driver so that it can be shared across multiple
generations of MIC products. The MIC sysfs interface documented
earlier has been updated. Further, the COSM host driver communicates
with a COSM client on the card using SCIF. COSM is therefore the first
kernel mode SCIF client and demonstrates the use of a subset of the
SCIF API.

Patch 15 onward completes the SCIF implementation by enabling the
following SCIF Remote Memory Access (RMA) functionality:
a. Memory registration to pin and unpin pages
b. Remote Memory mapping for low latency CPU accesses
c. Remote DMA (RDMA) for high bandwidth DMA transfers
d. Fence mechanisms to synchronize RDMAs

Documentation/mic/scif_overview.txt contains more information about
SCIF. This patch series is divided into 22 patches as follows:

1) iommu: iova: Move iova cache management to the iova library

2) iommu: iova: Export symbols

3) iommu: Make the iova library a module

4) Allow iova to be used without requiring IOMMU_SUPPORT

5) Add support to program MIC x100 status descriptiors

6) SCIF poll

7) Support for kernel mode SCIF clients

8) MIC COSM bus

9) Coprocessor State Management (COSM) driver

10) COSM SCIF server

11) COSM client driver

12) Remove COSM functionality from the MIC host driver

13) Remove COSM functionality from the MIC card driver

14) Update MIC host daemon with COSM changes

15) SCIF RMA header file and IOCTL changes

16) SCIF RMA header file

17) SCIF memory registration and unregistration

18) SCIF RMA list operations

19) SCIF remote memory map/unmap interface

20) SCIF DMA and CPU copy interface

21) SCIF fence

22) SCIF RMA nodeqp and minor miscellaneous changes

These patches have also been scanned by Fengguang Wu's 0-day
infrastructure and no issues have been reported.

Ashutosh Dixit (9):
  misc: mic: SCIF poll
  misc: mic: Add support for kernel mode SCIF clients
  misc: mic: MIC COSM bus
  misc: mic: Coprocessor State Management (COSM) driver
  misc: mic: COSM SCIF server
  misc: mic: COSM client driver
  misc: mic: Remove COSM functionality from the MIC host driver
  misc: mic: Remove COSM functionality from the MIC card driver
  misc: mic: Update MIC host daemon with COSM changes

Sakari Ailus (3):
  iommu: iova: Move iova cache management to the iova library
  iommu: iova: Export symbols
  iommu: Make the iova library a module

Siva Yerramreddy (1):
  dma: Add support to program MIC x100 status descriptiors

Sudeep Dutt (9):
  iommu: Allow iova to be used without requiring IOMMU_SUPPORT
  misc: mic: SCIF RMA header file and IOCTL changes
  misc: mic: SCIF RMA header file
  misc: mic: SCIF memory registration and unregistration
  misc: mic: SCIF RMA list operations
  misc: mic: SCIF remote memory map/unmap interface
  misc: mic: SCIF DMA and CPU copy interface
  misc: mic: SCIF fence
  misc: mic: SCIF RMA nodeqp and minor miscellaneous changes

 Documentation/ABI/testing/sysfs-class-mic.txt   |   29 +-
 Documentation/mic/mic_overview.txt              |   31 +-
 Documentation/mic/mpssd/mpss                    |    4 +-
 Documentation/mic/mpssd/mpssd.c                 |  362 +++--
 Documentation/mic/mpssd/mpssd.h                 |    1 +
 drivers/dma/mic_x100_dma.c                      |   39 +-
 drivers/iommu/Kconfig                           |    6 +-
 drivers/iommu/intel-iommu.c                     |    6 +-
 drivers/iommu/iova.c                            |   97 +-
 drivers/misc/mic/Kconfig                        |   23 +-
 drivers/misc/mic/Makefile                       |    2 +
 drivers/misc/mic/bus/Makefile                   |    1 +
 drivers/misc/mic/bus/cosm_bus.c                 |  141 ++
 drivers/misc/mic/bus/cosm_bus.h                 |  134 ++
 drivers/misc/mic/bus/mic_bus.c                  |   22 +-
 drivers/misc/mic/bus/scif_bus.c                 |    7 +-
 drivers/misc/mic/bus/scif_bus.h                 |    6 +-
 drivers/misc/mic/card/mic_device.c              |   88 +-
 drivers/misc/mic/card/mic_x100.c                |    2 +-
 drivers/misc/mic/common/mic_dev.h               |   13 +
 drivers/misc/mic/cosm/Makefile                  |   10 +
 drivers/misc/mic/cosm/cosm_debugfs.c            |  155 ++
 drivers/misc/mic/cosm/cosm_main.c               |  388 +++++
 drivers/misc/mic/cosm/cosm_main.h               |   70 +
 drivers/misc/mic/cosm/cosm_scif_server.c        |  405 +++++
 drivers/misc/mic/cosm/cosm_sysfs.c              |  461 ++++++
 drivers/misc/mic/cosm_client/Makefile           |    7 +
 drivers/misc/mic/cosm_client/cosm_scif_client.c |  275 ++++
 drivers/misc/mic/host/Makefile                  |    1 -
 drivers/misc/mic/host/mic_boot.c                |  317 +---
 drivers/misc/mic/host/mic_debugfs.c             |  114 +-
 drivers/misc/mic/host/mic_device.h              |   88 +-
 drivers/misc/mic/host/mic_fops.c                |    4 +-
 drivers/misc/mic/host/mic_intr.c                |   46 +-
 drivers/misc/mic/host/mic_main.c                |  223 +--
 drivers/misc/mic/host/mic_smpt.c                |   30 +-
 drivers/misc/mic/host/mic_sysfs.c               |  459 ------
 drivers/misc/mic/host/mic_virtio.c              |   17 +-
 drivers/misc/mic/host/mic_virtio.h              |    2 +-
 drivers/misc/mic/host/mic_x100.c                |   46 +-
 drivers/misc/mic/scif/Makefile                  |    5 +
 drivers/misc/mic/scif/scif_api.c                |  234 ++-
 drivers/misc/mic/scif/scif_debugfs.c            |   85 +-
 drivers/misc/mic/scif/scif_dma.c                | 1979 +++++++++++++++++++++++
 drivers/misc/mic/scif/scif_epd.c                |   26 +-
 drivers/misc/mic/scif/scif_epd.h                |   50 +
 drivers/misc/mic/scif/scif_fd.c                 |  178 +-
 drivers/misc/mic/scif/scif_fence.c              |  771 +++++++++
 drivers/misc/mic/scif/scif_main.c               |  111 +-
 drivers/misc/mic/scif/scif_main.h               |   37 +-
 drivers/misc/mic/scif/scif_map.h                |   25 +-
 drivers/misc/mic/scif/scif_mmap.c               |  699 ++++++++
 drivers/misc/mic/scif/scif_nm.c                 |   20 +-
 drivers/misc/mic/scif/scif_nodeqp.c             |  130 +-
 drivers/misc/mic/scif/scif_nodeqp.h             |   42 +-
 drivers/misc/mic/scif/scif_peer_bus.c           |  179 +-
 drivers/misc/mic/scif/scif_peer_bus.h           |   42 +-
 drivers/misc/mic/scif/scif_rma.c                | 1770 ++++++++++++++++++++
 drivers/misc/mic/scif/scif_rma.h                |  464 ++++++
 drivers/misc/mic/scif/scif_rma_list.c           |  291 ++++
 drivers/misc/mic/scif/scif_rma_list.h           |   57 +
 include/linux/dmaengine.h                       |    4 +
 include/linux/iova.h                            |    4 +-
 include/linux/mic_bus.h                         |    3 +-
 include/linux/scif.h                            |  366 ++++-
 include/uapi/linux/mic_common.h                 |   16 +-
 include/uapi/linux/scif_ioctl.h                 |   85 +
 67 files changed, 10065 insertions(+), 1740 deletions(-)
 create mode 100644 drivers/misc/mic/bus/cosm_bus.c
 create mode 100644 drivers/misc/mic/bus/cosm_bus.h
 create mode 100644 drivers/misc/mic/cosm/Makefile
 create mode 100644 drivers/misc/mic/cosm/cosm_debugfs.c
 create mode 100644 drivers/misc/mic/cosm/cosm_main.c
 create mode 100644 drivers/misc/mic/cosm/cosm_main.h
 create mode 100644 drivers/misc/mic/cosm/cosm_scif_server.c
 create mode 100644 drivers/misc/mic/cosm/cosm_sysfs.c
 create mode 100644 drivers/misc/mic/cosm_client/Makefile
 create mode 100644 drivers/misc/mic/cosm_client/cosm_scif_client.c
 delete mode 100644 drivers/misc/mic/host/mic_sysfs.c
 create mode 100644 drivers/misc/mic/scif/scif_dma.c
 create mode 100644 drivers/misc/mic/scif/scif_fence.c
 create mode 100644 drivers/misc/mic/scif/scif_mmap.c
 create mode 100644 drivers/misc/mic/scif/scif_rma.c
 create mode 100644 drivers/misc/mic/scif/scif_rma.h
 create mode 100644 drivers/misc/mic/scif/scif_rma_list.c
 create mode 100644 drivers/misc/mic/scif/scif_rma_list.h

-- 
2.0.0.rc3.2.g998f840

[PATCH char-misc-next v2 01/22] iommu: iova: Move iova cache management to the iova library

[Ashutosh Dixit]

From: Sakari Ailus <sakari.ailus-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>

This is necessary to separate intel-iommu from the iova library.

Signed-off-by: Sakari Ailus <sakari.ailus-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
Signed-off-by: David Woodhouse <David.Woodhouse-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
---
 drivers/iommu/intel-iommu.c |  6 ++--
 drivers/iommu/iova.c        | 83 ++++++++++++++++++++++++++-------------------
 include/linux/iova.h        |  4 +--
 3 files changed, 54 insertions(+), 39 deletions(-)

diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index 2d7349a..7b27bcd 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -3711,7 +3711,7 @@ static inline int iommu_devinfo_cache_init(void)
 static int __init iommu_init_mempool(void)
 {
 	int ret;
-	ret = iommu_iova_cache_init();
+	ret = iova_cache_get();
 	if (ret)
 		return ret;
 
@@ -3725,7 +3725,7 @@ static int __init iommu_init_mempool(void)
 
 	kmem_cache_destroy(iommu_domain_cache);
 domain_error:
-	iommu_iova_cache_destroy();
+	iova_cache_put();
 
 	return -ENOMEM;
 }
@@ -3734,7 +3734,7 @@ static void __init iommu_exit_mempool(void)
 {
 	kmem_cache_destroy(iommu_devinfo_cache);
 	kmem_cache_destroy(iommu_domain_cache);
-	iommu_iova_cache_destroy();
+	iova_cache_put();
 }
 
 static void quirk_ioat_snb_local_iommu(struct pci_dev *pdev)
diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c
index b7c3d92..400f4d1 100644
--- a/drivers/iommu/iova.c
+++ b/drivers/iommu/iova.c
@@ -20,40 +20,6 @@
 #include <linux/iova.h>
 #include <linux/slab.h>
 
-static struct kmem_cache *iommu_iova_cache;
-
-int iommu_iova_cache_init(void)
-{
-	int ret = 0;
-
-	iommu_iova_cache = kmem_cache_create("iommu_iova",
-					 sizeof(struct iova),
-					 0,
-					 SLAB_HWCACHE_ALIGN,
-					 NULL);
-	if (!iommu_iova_cache) {
-		pr_err("Couldn't create iova cache\n");
-		ret = -ENOMEM;
-	}
-
-	return ret;
-}
-
-void iommu_iova_cache_destroy(void)
-{
-	kmem_cache_destroy(iommu_iova_cache);
-}
-
-struct iova *alloc_iova_mem(void)
-{
-	return kmem_cache_alloc(iommu_iova_cache, GFP_ATOMIC);
-}
-
-void free_iova_mem(struct iova *iova)
-{
-	kmem_cache_free(iommu_iova_cache, iova);
-}
-
 void
 init_iova_domain(struct iova_domain *iovad, unsigned long granule,
 	unsigned long start_pfn, unsigned long pfn_32bit)
@@ -242,6 +208,55 @@ iova_insert_rbtree(struct rb_root *root, struct iova *iova)
 	rb_insert_color(&iova->node, root);
 }
 
+static struct kmem_cache *iova_cache;
+static unsigned int iova_cache_users;
+static DEFINE_MUTEX(iova_cache_mutex);
+
+struct iova *alloc_iova_mem(void)
+{
+	return kmem_cache_alloc(iova_cache, GFP_ATOMIC);
+}
+EXPORT_SYMBOL(alloc_iova_mem);
+
+void free_iova_mem(struct iova *iova)
+{
+	kmem_cache_free(iova_cache, iova);
+}
+EXPORT_SYMBOL(free_iova_mem);
+
+int iova_cache_get(void)
+{
+	mutex_lock(&iova_cache_mutex);
+	if (!iova_cache_users) {
+		iova_cache = kmem_cache_create(
+			"iommu_iova", sizeof(struct iova), 0,
+			SLAB_HWCACHE_ALIGN, NULL);
+		if (!iova_cache) {
+			mutex_unlock(&iova_cache_mutex);
+			printk(KERN_ERR "Couldn't create iova cache\n");
+			return -ENOMEM;
+		}
+	}
+
+	iova_cache_users++;
+	mutex_unlock(&iova_cache_mutex);
+
+	return 0;
+}
+
+void iova_cache_put(void)
+{
+	mutex_lock(&iova_cache_mutex);
+	if (WARN_ON(!iova_cache_users)) {
+		mutex_unlock(&iova_cache_mutex);
+		return;
+	}
+	iova_cache_users--;
+	if (!iova_cache_users)
+		kmem_cache_destroy(iova_cache);
+	mutex_unlock(&iova_cache_mutex);
+}
+
 /**
  * alloc_iova - allocates an iova
  * @iovad: - iova domain in question
diff --git a/include/linux/iova.h b/include/linux/iova.h
index 3920a19..92f7177 100644
--- a/include/linux/iova.h
+++ b/include/linux/iova.h
@@ -68,8 +68,8 @@ static inline unsigned long iova_pfn(struct iova_domain *iovad, dma_addr_t iova)
 	return iova >> iova_shift(iovad);
 }
 
-int iommu_iova_cache_init(void);
-void iommu_iova_cache_destroy(void);
+int iova_cache_get(void);
+void iova_cache_put(void);
 
 struct iova *alloc_iova_mem(void);
 void free_iova_mem(struct iova *iova);
-- 
2.0.0.rc3.2.g998f840

[PATCH char-misc-next v2 02/22] iommu: iova: Export symbols

[Ashutosh Dixit]

From: Sakari Ailus <sakari.ailus@linux.intel.com>

Use EXPORT_SYMBOL_GPL() to export the iova library symbols. The symbols
include:

	init_iova_domain();
	iova_cache_get();
	iova_cache_put();
	iova_cache_init();
	alloc_iova();
	find_iova();
	__free_iova();
	free_iova();
	put_iova_domain();
	reserve_iova();
	copy_reserved_iova();

Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/iommu/iova.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c
index 400f4d1..dd87123 100644
--- a/drivers/iommu/iova.c
+++ b/drivers/iommu/iova.c
@@ -38,6 +38,7 @@ init_iova_domain(struct iova_domain *iovad, unsigned long granule,
 	iovad->start_pfn = start_pfn;
 	iovad->dma_32bit_pfn = pfn_32bit;
 }
+EXPORT_SYMBOL_GPL(init_iova_domain);
 
 static struct rb_node *
 __get_cached_rbnode(struct iova_domain *iovad, unsigned long *limit_pfn)
@@ -243,6 +244,7 @@ int iova_cache_get(void)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(iova_cache_get);
 
 void iova_cache_put(void)
 {
@@ -256,6 +258,7 @@ void iova_cache_put(void)
 		kmem_cache_destroy(iova_cache);
 	mutex_unlock(&iova_cache_mutex);
 }
+EXPORT_SYMBOL_GPL(iova_cache_put);
 
 /**
  * alloc_iova - allocates an iova
@@ -296,6 +299,7 @@ alloc_iova(struct iova_domain *iovad, unsigned long size,
 
 	return new_iova;
 }
+EXPORT_SYMBOL_GPL(alloc_iova);
 
 /**
  * find_iova - find's an iova for a given pfn
@@ -336,6 +340,7 @@ struct iova *find_iova(struct iova_domain *iovad, unsigned long pfn)
 	spin_unlock_irqrestore(&iovad->iova_rbtree_lock, flags);
 	return NULL;
 }
+EXPORT_SYMBOL_GPL(find_iova);
 
 /**
  * __free_iova - frees the given iova
@@ -354,6 +359,7 @@ __free_iova(struct iova_domain *iovad, struct iova *iova)
 	spin_unlock_irqrestore(&iovad->iova_rbtree_lock, flags);
 	free_iova_mem(iova);
 }
+EXPORT_SYMBOL_GPL(__free_iova);
 
 /**
  * free_iova - finds and frees the iova for a given pfn
@@ -371,6 +377,7 @@ free_iova(struct iova_domain *iovad, unsigned long pfn)
 		__free_iova(iovad, iova);
 
 }
+EXPORT_SYMBOL_GPL(free_iova);
 
 /**
  * put_iova_domain - destroys the iova doamin
@@ -393,6 +400,7 @@ void put_iova_domain(struct iova_domain *iovad)
 	}
 	spin_unlock_irqrestore(&iovad->iova_rbtree_lock, flags);
 }
+EXPORT_SYMBOL_GPL(put_iova_domain);
 
 static int
 __is_range_overlap(struct rb_node *node,
@@ -482,6 +490,7 @@ finish:
 	spin_unlock_irqrestore(&iovad->iova_rbtree_lock, flags);
 	return iova;
 }
+EXPORT_SYMBOL_GPL(reserve_iova);
 
 /**
  * copy_reserved_iova - copies the reserved between domains
@@ -508,6 +517,7 @@ copy_reserved_iova(struct iova_domain *from, struct iova_domain *to)
 	}
 	spin_unlock_irqrestore(&from->iova_rbtree_lock, flags);
 }
+EXPORT_SYMBOL_GPL(copy_reserved_iova);
 
 struct iova *
 split_and_remove_iova(struct iova_domain *iovad, struct iova *iova,
-- 
2.0.0.rc3.2.g998f840

[PATCH char-misc-next v2 03/22] iommu: Make the iova library a module

[Ashutosh Dixit]

From: Sakari Ailus <sakari.ailus@linux.intel.com>

The iova library has use outside the intel-iommu driver, thus make it a
module.

Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/iommu/Kconfig | 2 +-
 drivers/iommu/iova.c  | 4 ++++
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig
index 4664c2a..d9da766 100644
--- a/drivers/iommu/Kconfig
+++ b/drivers/iommu/Kconfig
@@ -43,7 +43,7 @@ config IOMMU_IO_PGTABLE_LPAE_SELFTEST
 endmenu
 
 config IOMMU_IOVA
-	bool
+	tristate
 
 config OF_IOMMU
        def_bool y
diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c
index dd87123..dd64d33 100644
--- a/drivers/iommu/iova.c
+++ b/drivers/iommu/iova.c
@@ -18,6 +18,7 @@
  */
 
 #include <linux/iova.h>
+#include <linux/module.h>
 #include <linux/slab.h>
 
 void
@@ -559,3 +560,6 @@ error:
 		free_iova_mem(prev);
 	return NULL;
 }
+
+MODULE_AUTHOR("Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>");
+MODULE_LICENSE("GPL");
-- 
2.0.0.rc3.2.g998f840

[PATCH char-misc-next v2 04/22] iommu: Allow iova to be used without requiring IOMMU_SUPPORT

[Ashutosh Dixit]

From: Sudeep Dutt <sudeep.dutt-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>

iova is a library which can be built without IOMMU_SUPPORT

Signed-off-by: Sudeep Dutt <sudeep.dutt-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
---
 drivers/iommu/Kconfig | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig
index d9da766..71d1c46 100644
--- a/drivers/iommu/Kconfig
+++ b/drivers/iommu/Kconfig
@@ -12,6 +12,9 @@ menuconfig IOMMU_SUPPORT
 	  remap DMA requests and/or remap interrupts from other devices on the
 	  system.
 
+config IOMMU_IOVA
+	tristate
+
 if IOMMU_SUPPORT
 
 menu "Generic IOMMU Pagetable Support"
@@ -42,9 +45,6 @@ config IOMMU_IO_PGTABLE_LPAE_SELFTEST
 
 endmenu
 
-config IOMMU_IOVA
-	tristate
-
 config OF_IOMMU
        def_bool y
        depends on OF && IOMMU_API
-- 
2.0.0.rc3.2.g998f840

Re: [PATCH char-misc-next v2 04/22] iommu: Allow iova to be used without requiring IOMMU_SUPPORT

[Woodhouse, David]

[-- Attachment #1: Type: text/plain, Size: 620 bytes --]

On Tue, 2015-09-29 at 18:09 -0700, Ashutosh Dixit wrote:
> From: Sudeep Dutt <sudeep.dutt@intel.com>
> 
> iova is a library which can be built without IOMMU_SUPPORT
> 
> Signed-off-by: Sudeep Dutt <sudeep.dutt@intel.com>

The first three of these patches are in 4.3-rc4 already. Apologies for
the delay in pushing them out.

This one looks sane enough too, but perhaps in that case we should move
the code *out* of drivers/iommu/ and into lib/iova/ ?

-- 
David Woodhouse                            Open Source Technology Centre
David.Woodhouse@intel.com                              Intel Corporation


[-- Attachment #2: smime.p7s --]
[-- Type: application/x-pkcs7-signature, Size: 3437 bytes --]

Re: [PATCH char-misc-next v2 04/22] iommu: Allow iova to be used without requiring IOMMU_SUPPORT

[Sudeep Dutt]

On Mon, 2015-10-05 at 03:50 -0700, Woodhouse, David wrote:
> On Tue, 2015-09-29 at 18:09 -0700, Ashutosh Dixit wrote:
> > From: Sudeep Dutt <sudeep.dutt-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
> > 
> > iova is a library which can be built without IOMMU_SUPPORT
> > 
> > Signed-off-by: Sudeep Dutt <sudeep.dutt-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
> 
> The first three of these patches are in 4.3-rc4 already. Apologies for
> the delay in pushing them out.
> 
> This one looks sane enough too, but perhaps in that case we should move
> the code *out* of drivers/iommu/ and into lib/iova/ ?
> 

Yes, moving the code into lib/iova is the correct long term solution. I
have sent Greg a patch which reverts this commit since it is no longer
required and will create a merge conflict for him unnecessarily as well
with 4.3-rc4.

Thanks,
Sudeep Dutt

Re: [PATCH char-misc-next v2 04/22] iommu: Allow iova to be used without requiring IOMMU_SUPPORT

[gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r]

On Mon, Oct 05, 2015 at 10:38:43AM -0700, Sudeep Dutt wrote:
> On Mon, 2015-10-05 at 03:50 -0700, Woodhouse, David wrote:
> > On Tue, 2015-09-29 at 18:09 -0700, Ashutosh Dixit wrote:
> > > From: Sudeep Dutt <sudeep.dutt-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
> > > 
> > > iova is a library which can be built without IOMMU_SUPPORT
> > > 
> > > Signed-off-by: Sudeep Dutt <sudeep.dutt-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
> > 
> > The first three of these patches are in 4.3-rc4 already. Apologies for
> > the delay in pushing them out.
> > 
> > This one looks sane enough too, but perhaps in that case we should move
> > the code *out* of drivers/iommu/ and into lib/iova/ ?
> > 
> 
> Yes, moving the code into lib/iova is the correct long term solution. I
> have sent Greg a patch which reverts this commit since it is no longer
> required and will create a merge conflict for him unnecessarily as well
> with 4.3-rc4.

I can handle merge issues, that's trivial.  Reverting the patch
shoulnd't really be needed, right?  Let me see what happens when I merge
to see if your patch is necessary...

Re: [PATCH char-misc-next v2 04/22] iommu: Allow iova to be used without requiring IOMMU_SUPPORT

[gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r]

On Tue, Oct 06, 2015 at 06:12:40AM +0100, gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org wrote:
> On Mon, Oct 05, 2015 at 10:38:43AM -0700, Sudeep Dutt wrote:
> > On Mon, 2015-10-05 at 03:50 -0700, Woodhouse, David wrote:
> > > On Tue, 2015-09-29 at 18:09 -0700, Ashutosh Dixit wrote:
> > > > From: Sudeep Dutt <sudeep.dutt-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
> > > > 
> > > > iova is a library which can be built without IOMMU_SUPPORT
> > > > 
> > > > Signed-off-by: Sudeep Dutt <sudeep.dutt-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
> > > 
> > > The first three of these patches are in 4.3-rc4 already. Apologies for
> > > the delay in pushing them out.
> > > 
> > > This one looks sane enough too, but perhaps in that case we should move
> > > the code *out* of drivers/iommu/ and into lib/iova/ ?
> > > 
> > 
> > Yes, moving the code into lib/iova is the correct long term solution. I
> > have sent Greg a patch which reverts this commit since it is no longer
> > required and will create a merge conflict for him unnecessarily as well
> > with 4.3-rc4.
> 
> I can handle merge issues, that's trivial.  Reverting the patch
> shoulnd't really be needed, right?  Let me see what happens when I merge
> to see if your patch is necessary...

Ok, I don't think it is needed, the merge was pretty trivial.

Can you test out my char-misc-testing branch right now to see if it's
all ok with the merge?  If so, I'll move it all over to the "real" place
for it to start showing up in linux-next, i.e. my char-misc-next branch.

thanks,

greg k-h

Re: [PATCH char-misc-next v2 04/22] iommu: Allow iova to be used without requiring IOMMU_SUPPORT

[Sudeep Dutt]

On Tue, 2015-10-06 at 06:20 +0100, gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org wrote:
> On Tue, Oct 06, 2015 at 06:12:40AM +0100, gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org wrote:
> > On Mon, Oct 05, 2015 at 10:38:43AM -0700, Sudeep Dutt wrote:
> > > On Mon, 2015-10-05 at 03:50 -0700, Woodhouse, David wrote:
> > > > On Tue, 2015-09-29 at 18:09 -0700, Ashutosh Dixit wrote:
> > > > > From: Sudeep Dutt <sudeep.dutt-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
> > > > > 
> > > > > iova is a library which can be built without IOMMU_SUPPORT
> > > > > 
> > > > > Signed-off-by: Sudeep Dutt <sudeep.dutt-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
> > > > 
> > > > The first three of these patches are in 4.3-rc4 already. Apologies for
> > > > the delay in pushing them out.
> > > > 
> > > > This one looks sane enough too, but perhaps in that case we should move
> > > > the code *out* of drivers/iommu/ and into lib/iova/ ?
> > > > 
> > > 
> > > Yes, moving the code into lib/iova is the correct long term solution. I
> > > have sent Greg a patch which reverts this commit since it is no longer
> > > required and will create a merge conflict for him unnecessarily as well
> > > with 4.3-rc4.
> > 
> > I can handle merge issues, that's trivial.  Reverting the patch
> > shoulnd't really be needed, right?  Let me see what happens when I merge
> > to see if your patch is necessary...
> 
> Ok, I don't think it is needed, the merge was pretty trivial.
> 
> Can you test out my char-misc-testing branch right now to see if it's
> all ok with the merge?  If so, I'll move it all over to the "real" place
> for it to start showing up in linux-next, i.e. my char-misc-next branch.
> 

Hi Greg,

I think it is best to revert this patch as it is incorrect. The iommu
folder gets compiled only if IOMMU_SUPPORT is enabled so IOMMU_IOVA
should indeed be included only when IOMMU_SUPPORT is enabled.

Sincere apologies for the mess here but I believe it will all get fixed
up if you accept the revert of 353649e5da I sent across earlier today.

Thanks,
Sudeep Dutt

Re: [PATCH char-misc-next v2 04/22] iommu: Allow iova to be used without requiring IOMMU_SUPPORT

[gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r]

On Mon, Oct 05, 2015 at 10:23:38PM -0700, Sudeep Dutt wrote:
> On Tue, 2015-10-06 at 06:20 +0100, gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org wrote:
> > On Tue, Oct 06, 2015 at 06:12:40AM +0100, gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org wrote:
> > > On Mon, Oct 05, 2015 at 10:38:43AM -0700, Sudeep Dutt wrote:
> > > > On Mon, 2015-10-05 at 03:50 -0700, Woodhouse, David wrote:
> > > > > On Tue, 2015-09-29 at 18:09 -0700, Ashutosh Dixit wrote:
> > > > > > From: Sudeep Dutt <sudeep.dutt-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
> > > > > > 
> > > > > > iova is a library which can be built without IOMMU_SUPPORT
> > > > > > 
> > > > > > Signed-off-by: Sudeep Dutt <sudeep.dutt-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
> > > > > 
> > > > > The first three of these patches are in 4.3-rc4 already. Apologies for
> > > > > the delay in pushing them out.
> > > > > 
> > > > > This one looks sane enough too, but perhaps in that case we should move
> > > > > the code *out* of drivers/iommu/ and into lib/iova/ ?
> > > > > 
> > > > 
> > > > Yes, moving the code into lib/iova is the correct long term solution. I
> > > > have sent Greg a patch which reverts this commit since it is no longer
> > > > required and will create a merge conflict for him unnecessarily as well
> > > > with 4.3-rc4.
> > > 
> > > I can handle merge issues, that's trivial.  Reverting the patch
> > > shoulnd't really be needed, right?  Let me see what happens when I merge
> > > to see if your patch is necessary...
> > 
> > Ok, I don't think it is needed, the merge was pretty trivial.
> > 
> > Can you test out my char-misc-testing branch right now to see if it's
> > all ok with the merge?  If so, I'll move it all over to the "real" place
> > for it to start showing up in linux-next, i.e. my char-misc-next branch.
> > 
> 
> Hi Greg,
> 
> I think it is best to revert this patch as it is incorrect. The iommu
> folder gets compiled only if IOMMU_SUPPORT is enabled so IOMMU_IOVA
> should indeed be included only when IOMMU_SUPPORT is enabled.
> 
> Sincere apologies for the mess here but I believe it will all get fixed
> up if you accept the revert of 353649e5da I sent across earlier today.

Again, look at the merge, I think I already handled this in that manner.
If not, let me know.

thanks,

greg k-h

Re: [PATCH char-misc-next v2 04/22] iommu: Allow iova to be used without requiring IOMMU_SUPPORT

[Sudeep Dutt]

On Tue, 2015-10-06 at 08:56 +0100, gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org wrote:
> On Mon, Oct 05, 2015 at 10:23:38PM -0700, Sudeep Dutt wrote:
> > On Tue, 2015-10-06 at 06:20 +0100, gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org wrote:
> > > On Tue, Oct 06, 2015 at 06:12:40AM +0100, gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org wrote:
> > > > On Mon, Oct 05, 2015 at 10:38:43AM -0700, Sudeep Dutt wrote:
> > > > > On Mon, 2015-10-05 at 03:50 -0700, Woodhouse, David wrote:
> > > > > > On Tue, 2015-09-29 at 18:09 -0700, Ashutosh Dixit wrote:
> > > > > > > From: Sudeep Dutt <sudeep.dutt-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
> > > > > > > 
> > > > > > > iova is a library which can be built without IOMMU_SUPPORT
> > > > > > > 
> > > > > > > Signed-off-by: Sudeep Dutt <sudeep.dutt-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
> > > > > > 
> > > > > > The first three of these patches are in 4.3-rc4 already. Apologies for
> > > > > > the delay in pushing them out.
> > > > > > 
> > > > > > This one looks sane enough too, but perhaps in that case we should move
> > > > > > the code *out* of drivers/iommu/ and into lib/iova/ ?
> > > > > > 
> > > > > 
> > > > > Yes, moving the code into lib/iova is the correct long term solution. I
> > > > > have sent Greg a patch which reverts this commit since it is no longer
> > > > > required and will create a merge conflict for him unnecessarily as well
> > > > > with 4.3-rc4.
> > > > 
> > > > I can handle merge issues, that's trivial.  Reverting the patch
> > > > shoulnd't really be needed, right?  Let me see what happens when I merge
> > > > to see if your patch is necessary...
> > > 
> > > Ok, I don't think it is needed, the merge was pretty trivial.
> > > 
> > > Can you test out my char-misc-testing branch right now to see if it's
> > > all ok with the merge?  If so, I'll move it all over to the "real" place
> > > for it to start showing up in linux-next, i.e. my char-misc-next branch.
> > > 
> > 
> > Hi Greg,
> > 
> > I think it is best to revert this patch as it is incorrect. The iommu
> > folder gets compiled only if IOMMU_SUPPORT is enabled so IOMMU_IOVA
> > should indeed be included only when IOMMU_SUPPORT is enabled.
> > 
> > Sincere apologies for the mess here but I believe it will all get fixed
> > up if you accept the revert of 353649e5da I sent across earlier today.
> 
> Again, look at the merge, I think I already handled this in that manner.
> If not, let me know.
> 

Hi Greg,

I took a look at your latest char-misc-testing tree and it needs to be
fixed up. IOMMU_IOVA should be inside the "if IOMMU_SUPPORT" block
instead of above it.

git revert 353649e5da in the char-misc-testing tree will fix everything
up or you could also apply the patch I sent earlier today which has the
same revert.

Thanks,
Sudeep Dutt

Re: [PATCH char-misc-next v2 04/22] iommu: Allow iova to be used without requiring IOMMU_SUPPORT

[gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r]

On Tue, Oct 06, 2015 at 01:05:27AM -0700, Sudeep Dutt wrote:
> On Tue, 2015-10-06 at 08:56 +0100, gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org wrote:
> > On Mon, Oct 05, 2015 at 10:23:38PM -0700, Sudeep Dutt wrote:
> > > On Tue, 2015-10-06 at 06:20 +0100, gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org wrote:
> > > > On Tue, Oct 06, 2015 at 06:12:40AM +0100, gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org wrote:
> > > > > On Mon, Oct 05, 2015 at 10:38:43AM -0700, Sudeep Dutt wrote:
> > > > > > On Mon, 2015-10-05 at 03:50 -0700, Woodhouse, David wrote:
> > > > > > > On Tue, 2015-09-29 at 18:09 -0700, Ashutosh Dixit wrote:
> > > > > > > > From: Sudeep Dutt <sudeep.dutt-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
> > > > > > > > 
> > > > > > > > iova is a library which can be built without IOMMU_SUPPORT
> > > > > > > > 
> > > > > > > > Signed-off-by: Sudeep Dutt <sudeep.dutt-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
> > > > > > > 
> > > > > > > The first three of these patches are in 4.3-rc4 already. Apologies for
> > > > > > > the delay in pushing them out.
> > > > > > > 
> > > > > > > This one looks sane enough too, but perhaps in that case we should move
> > > > > > > the code *out* of drivers/iommu/ and into lib/iova/ ?
> > > > > > > 
> > > > > > 
> > > > > > Yes, moving the code into lib/iova is the correct long term solution. I
> > > > > > have sent Greg a patch which reverts this commit since it is no longer
> > > > > > required and will create a merge conflict for him unnecessarily as well
> > > > > > with 4.3-rc4.
> > > > > 
> > > > > I can handle merge issues, that's trivial.  Reverting the patch
> > > > > shoulnd't really be needed, right?  Let me see what happens when I merge
> > > > > to see if your patch is necessary...
> > > > 
> > > > Ok, I don't think it is needed, the merge was pretty trivial.
> > > > 
> > > > Can you test out my char-misc-testing branch right now to see if it's
> > > > all ok with the merge?  If so, I'll move it all over to the "real" place
> > > > for it to start showing up in linux-next, i.e. my char-misc-next branch.
> > > > 
> > > 
> > > Hi Greg,
> > > 
> > > I think it is best to revert this patch as it is incorrect. The iommu
> > > folder gets compiled only if IOMMU_SUPPORT is enabled so IOMMU_IOVA
> > > should indeed be included only when IOMMU_SUPPORT is enabled.
> > > 
> > > Sincere apologies for the mess here but I believe it will all get fixed
> > > up if you accept the revert of 353649e5da I sent across earlier today.
> > 
> > Again, look at the merge, I think I already handled this in that manner.
> > If not, let me know.
> > 
> 
> Hi Greg,
> 
> I took a look at your latest char-misc-testing tree and it needs to be
> fixed up. IOMMU_IOVA should be inside the "if IOMMU_SUPPORT" block
> instead of above it.
> 
> git revert 353649e5da in the char-misc-testing tree will fix everything
> up or you could also apply the patch I sent earlier today which has the
> same revert.

Ok, I've applied your patch, rolled back the merge, and that should be
good, right?  If so, I'll push this branch out to char-misc-next and
then merge in 4.3-rc4 just to keep everything up to date.

thanks,

greg k-h

Re: [PATCH char-misc-next v2 04/22] iommu: Allow iova to be used without requiring IOMMU_SUPPORT

[Sudeep Dutt]

On Tue, 2015-10-06 at 09:33 +0100, gregkh@linuxfoundation.org wrote:
> On Tue, Oct 06, 2015 at 01:05:27AM -0700, Sudeep Dutt wrote:
> > On Tue, 2015-10-06 at 08:56 +0100, gregkh@linuxfoundation.org wrote:
> > > On Mon, Oct 05, 2015 at 10:23:38PM -0700, Sudeep Dutt wrote:
> > > > On Tue, 2015-10-06 at 06:20 +0100, gregkh@linuxfoundation.org wrote:
> > > > > On Tue, Oct 06, 2015 at 06:12:40AM +0100, gregkh@linuxfoundation.org wrote:
> > > > > > On Mon, Oct 05, 2015 at 10:38:43AM -0700, Sudeep Dutt wrote:
> > > > > > > On Mon, 2015-10-05 at 03:50 -0700, Woodhouse, David wrote:
> > > > > > > > On Tue, 2015-09-29 at 18:09 -0700, Ashutosh Dixit wrote:
> > > > > > > > > From: Sudeep Dutt <sudeep.dutt@intel.com>
> > > > > > > > > 
> > > > > > > > > iova is a library which can be built without IOMMU_SUPPORT
> > > > > > > > > 
> > > > > > > > > Signed-off-by: Sudeep Dutt <sudeep.dutt@intel.com>
> > > > > > > > 
> > > > > > > > The first three of these patches are in 4.3-rc4 already. Apologies for
> > > > > > > > the delay in pushing them out.
> > > > > > > > 
> > > > > > > > This one looks sane enough too, but perhaps in that case we should move
> > > > > > > > the code *out* of drivers/iommu/ and into lib/iova/ ?
> > > > > > > > 
> > > > > > > 
> > > > > > > Yes, moving the code into lib/iova is the correct long term solution. I
> > > > > > > have sent Greg a patch which reverts this commit since it is no longer
> > > > > > > required and will create a merge conflict for him unnecessarily as well
> > > > > > > 0000d3bd-0010
> > > > > > 
> > > > > > I can handle merge issues, that's trivial.  Reverting the patch
> > > > > > shoulnd't really be needed, right?  Let me see what happens when I merge
> > > > > > to see if your patch is necessary...
> > > > > 
> > > > > Ok, I don't think it is needed, the merge was pretty trivial.
> > > > > 
> > > > > Can you test out my char-misc-testing branch right now to see if it's
> > > > > all ok with the merge?  If so, I'll move it all over to the "real" place
> > > > > for it to start showing up in linux-next, i.e. my char-misc-next branch.
> > > > > 
> > > > 
> > > > Hi Greg,
> > > > 
> > > > I think it is best to revert this patch as it is incorrect. The iommu
> > > > folder gets compiled only if IOMMU_SUPPORT is enabled so IOMMU_IOVA
> > > > should indeed be included only when IOMMU_SUPPORT is enabled.
> > > > 
> > > > Sincere apologies for the mess here but I believe it will all get fixed
> > > > up if you accept the revert of 353649e5da I sent across earlier today.
> > > 
> > > Again, look at the merge, I think I already handled this in that manner.
> > > If not, let me know.
> > > 
> > 
> > Hi Greg,
> > 
> > I took a look at your latest char-misc-testing tree and it needs to be
> > fixed up. IOMMU_IOVA should be inside the "if IOMMU_SUPPORT" block
> > instead of above it.
> > 
> > git revert 353649e5da in the char-misc-testing tree will fix everything
> > up or you could also apply the patch I sent earlier today which has the
> > same revert.
> 
> Ok, I've applied your patch, rolled back the merge, and that should be
> good, right?  If so, I'll push this branch out to char-misc-next and
> then merge in 4.3-rc4 just to keep everything up to date.

Yes, it looks good now.

Thanks,
Sudeep Dutt