7 years agoiommu/amd: Convert dev_table_entry to u64
Joerg Roedel [Wed, 9 Nov 2011 11:06:03 +0000]
iommu/amd: Convert dev_table_entry to u64

Convert the contents of 'struct dev_table_entry' to u64 to
allow updating the DTE wit 64bit writes as required by the
spec.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/omap: be verbose when omap_iommu_iova_to_phys fails
Ohad Ben-Cohen [Tue, 6 Dec 2011 13:22:10 +0000]
iommu/omap: be verbose when omap_iommu_iova_to_phys fails

An omap_iommu_iova_to_phys failure usually means that iova wasn't mapped.

When that happens, it's helpful to know the value of iova, so add it
to the error message.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agointr_remapping: Fix section mismatch in ir_dev_scope_init()
Sergey Senozhatsky [Wed, 26 Oct 2011 16:15:07 +0000]
intr_remapping: Fix section mismatch in ir_dev_scope_init()

Fix:

 Section mismatch in reference from the function
 ir_dev_scope_init() to the function
 .init.text:dmar_dev_scope_init() The function
 ir_dev_scope_init() references the function __init dmar_dev_scope_init().

Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Youquan Song <youquan.song@intel.com>
Cc: Ohad Ben-Cohen <ohad@wizery.com>
Link: http://lkml.kernel.org/r/20111026161507.GB10103@swordfish
Signed-off-by: Ingo Molnar <mingo@elte.hu>

7 years agointel-iommu: Fix section mismatch in dmar_parse_rmrr_atsr_dev()
Sergey Senozhatsky [Wed, 26 Oct 2011 15:45:39 +0000]
intel-iommu: Fix section mismatch in dmar_parse_rmrr_atsr_dev()

dmar_parse_rmrr_atsr_dev() calls rmrr_parse_dev() and
atsr_parse_dev() which are both marked as __init.

Section mismatch in reference from the function
dmar_parse_rmrr_atsr_dev() to the function
.init.text:dmar_parse_dev_scope() The function
dmar_parse_rmrr_atsr_dev() references the function __init
dmar_parse_dev_scope().

Section mismatch in reference from the function
dmar_parse_rmrr_atsr_dev() to the function
.init.text:dmar_parse_dev_scope() The function
dmar_parse_rmrr_atsr_dev() references the function __init
dmar_parse_dev_scope().

Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: iommu@lists.linux-foundation.org
Cc: Joerg Roedel <joerg.roedel@amd.com>
Cc: Ohad Ben-Cohen <ohad@wizery.com>
Link: http://lkml.kernel.org/r/20111026154539.GA10103@swordfish
Signed-off-by: Ingo Molnar <mingo@elte.hu>

7 years agoiommu/omap: eliminate the public omap_find_iommu_device() method
Ohad Ben-Cohen [Mon, 10 Oct 2011 22:18:33 +0000]
iommu/omap: eliminate the public omap_find_iommu_device() method

Eliminate the public omap_find_iommu_device() method, and don't
expect clients to provide the omap_iommu handle anymore.

Instead, OMAP's iommu driver now utilizes dev_archdata's private iommu
extension to be able to access the required iommu information.

This way OMAP IOMMU users are now able to use the generic IOMMU API without
having to call any omap-specific binding method.

Update omap3isp appropriately.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Cc: Hiroshi Doyu <hdoyu@nvidia.com>

7 years agoiommu/omap: eliminate the public omap_find_iommu_device() method
Ohad Ben-Cohen [Mon, 10 Oct 2011 22:18:33 +0000]
iommu/omap: eliminate the public omap_find_iommu_device() method

Eliminate the public omap_find_iommu_device() method, and don't
expect clients to provide the omap_iommu handle anymore.

Instead, OMAP's iommu driver now utilizes dev_archdata's private iommu
extension to be able to access the required iommu information.

This way OMAP IOMMU users are now able to use the generic IOMMU API without
having to call any omap-specific binding method.

Update omap3isp appropriately.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Cc: Hiroshi Doyu <hdoyu@nvidia.com>

Conflicts:

drivers/media/video/omap3isp/ispccdc.c

Change-Id: Ia9b3421662fab7bb4bee298693ed6da716599ee5

7 years agoiommu: Fix compile error with !IOMMU_API
Joerg Roedel [Tue, 15 Nov 2011 11:48:29 +0000]
iommu: Fix compile error with !IOMMU_API

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu: Add option to group multi-function devices
Alex Williamson [Fri, 21 Oct 2011 19:56:24 +0000]
iommu: Add option to group multi-function devices

The option iommu=group_mf indicates the that the iommu driver should
expose all functions of a multi-function PCI device as the same
iommu_device_group.  This is useful for disallowing individual functions
being exposed as independent devices to userspace as there are often
hidden dependencies.  Virtual functions are not affected by this option.

Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/amd: Implement iommu_device_group
Alex Williamson [Fri, 21 Oct 2011 19:56:18 +0000]
iommu/amd: Implement iommu_device_group

Just use the amd_iommu_alias_table directly.

Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

Conflicts:

drivers/iommu/amd_iommu.c

Change-Id: I3aaeb78a1359e1ad8f4a37ba07fb1d404daa8d83

7 years agoiommu/intel: Implement iommu_device_group
Alex Williamson [Fri, 21 Oct 2011 19:56:11 +0000]
iommu/intel: Implement iommu_device_group

We generally have BDF granularity for devices, so we just need
to make sure devices aren't hidden behind PCIe-to-PCI bridges.
We can then make up a group number that's simply the concatenated
seg|bus|dev|fn so we don't have to track them (not that users
should depend on that).

Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Acked-By: David Woodhouse <David.Woodhouse@intel.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

Conflicts:

drivers/iommu/intel-iommu.c

Change-Id: Ifc8108a42383f317c2db42003eb96713a98a6899

7 years agoiommu: Add iommu_device_group callback and iommu_group sysfs entry
Alex Williamson [Fri, 21 Oct 2011 19:56:05 +0000]
iommu: Add iommu_device_group callback and iommu_group sysfs entry

An IOMMU group is a set of devices for which the IOMMU cannot
distinguish transactions.  For PCI devices, a group often occurs
when a PCI bridge is involved.  Transactions from any device
behind the bridge appear to be sourced from the bridge itself.
We leave it to the IOMMU driver to define the grouping restraints
for their platform.

Using this new interface, the group for a device can be retrieved
using the iommu_device_group() callback.  Users will compare the
value returned against the value returned for other devices to
determine whether they are part of the same group.  Devices with
no group are not translated by the IOMMU.  There should be no
expectations about the group numbers as they may be arbitrarily
assigned by the IOMMU driver and may not be persistent across boots.

We also provide a sysfs interface to the group numbers here so
that userspace can understand IOMMU dependencies between devices
for managing safe, userspace drivers.

[Some code changes by Joerg Roedel <joerg.roedel@amd.com>]

Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

Conflicts:

include/linux/iommu.h

Change-Id: I524607da569e42730bdc706759c49c4e0783555c

7 years agoiommu/core: remove the temporary pgsize settings
Ohad Ben-Cohen [Thu, 10 Nov 2011 09:32:31 +0000]
iommu/core: remove the temporary pgsize settings

Now that all IOMMU drivers are exporting their supported pgsizes,
we can remove the default pgsize settings in register_iommu().

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/intel: announce supported page sizes
Ohad Ben-Cohen [Thu, 10 Nov 2011 09:32:30 +0000]
iommu/intel: announce supported page sizes

Let the IOMMU core know we support arbitrary page sizes (as long as
they're an order of 4KiB).

This way the IOMMU core will retain the existing behavior we're used to;
it will let us map regions that:
- their size is an order of 4KiB
- they are naturally aligned

Note: Intel IOMMU hardware doesn't support arbitrary page sizes,
but the driver does (it splits arbitrary-sized mappings into
the pages supported by the hardware).

To make everything simpler for now, though, this patch effectively tells
the IOMMU core to keep giving this driver the same memory regions it did
before, so nothing is changed as far as it's concerned.

At this point, the page sizes announced remain static within the IOMMU
core. To correctly utilize the pgsize-splitting of the IOMMU core by
this driver, it seems that some core changes should still be done,
because Intel's IOMMU page size capabilities seem to have the potential
to be different between different DMA remapping devices.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/amd: announce supported page sizes
Ohad Ben-Cohen [Thu, 10 Nov 2011 09:32:29 +0000]
iommu/amd: announce supported page sizes

Let the IOMMU core know we support arbitrary page sizes (as long as
they're an order of 4KiB).

This way the IOMMU core will retain the existing behavior we're used to;
it will let us map regions that:
- their size is an order of 4KiB
- they are naturally aligned

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Cc: Joerg Roedel <Joerg.Roedel@amd.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/msm: announce supported page sizes
Ohad Ben-Cohen [Thu, 10 Nov 2011 09:32:28 +0000]
iommu/msm: announce supported page sizes

Let the IOMMU core know we support 4KiB, 64KiB, 1MiB and 16MiB page sizes.

This way the IOMMU core can split any arbitrary-sized physically
contiguous regions (that it needs to map) as needed.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Acked-by: David Brown <davidb@codeaurora.org>
Cc: Stepan Moskovchenko <stepanm@codeaurora.org>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/omap: announce supported page sizes
Ohad Ben-Cohen [Thu, 10 Nov 2011 09:32:27 +0000]
iommu/omap: announce supported page sizes

Let the IOMMU core know we support 4KiB, 64KiB, 1MiB and 16MiB page sizes.

This way the IOMMU core can split any arbitrary-sized physically
contiguous regions (that it needs to map) as needed.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Cc: Hiroshi DOYU <hdoyu@nvidia.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/core: split mapping to page sizes as supported by the hardware
Ohad Ben-Cohen [Thu, 10 Nov 2011 09:32:26 +0000]
iommu/core: split mapping to page sizes as supported by the hardware

When mapping a memory region, split it to page sizes as supported
by the iommu hardware. Always prefer bigger pages, when possible,
in order to reduce the TLB pressure.

The logic to do that is now added to the IOMMU core, so neither the iommu
drivers themselves nor users of the IOMMU API have to duplicate it.

This allows a more lenient granularity of mappings; traditionally the
IOMMU API took 'order' (of a page) as a mapping size, and directly let
the low level iommu drivers handle the mapping, but now that the IOMMU
core can split arbitrary memory regions into pages, we can remove this
limitation, so users don't have to split those regions by themselves.

Currently the supported page sizes are advertised once and they then
remain static. That works well for OMAP and MSM but it would probably
not fly well with intel's hardware, where the page size capabilities
seem to have the potential to be different between several DMA
remapping devices.

register_iommu() currently sets a default pgsize behavior, so we can convert
the IOMMU drivers in subsequent patches. After all the drivers
are converted, the temporary default settings will be removed.

Mainline users of the IOMMU API (kvm and omap-iovmm) are adopted
to deal with bytes instead of page order.

Many thanks to Joerg Roedel <Joerg.Roedel@amd.com> for significant review!

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Cc: David Brown <davidb@codeaurora.org>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Joerg Roedel <Joerg.Roedel@amd.com>
Cc: Stepan Moskovchenko <stepanm@codeaurora.org>
Cc: KyongHo Cho <pullip.cho@samsung.com>
Cc: Hiroshi DOYU <hdoyu@nvidia.com>
Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Cc: kvm@vger.kernel.org
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/core: stop converting bytes to page order back and forth
Ohad Ben-Cohen [Thu, 10 Nov 2011 09:32:25 +0000]
iommu/core: stop converting bytes to page order back and forth

Express sizes in bytes rather than in page order, to eliminate the
size->order->size conversions we have whenever the IOMMU API is calling
the low level drivers' map/unmap methods.

Adopt all existing drivers.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Cc: David Brown <davidb@codeaurora.org>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Joerg Roedel <Joerg.Roedel@amd.com>
Cc: Stepan Moskovchenko <stepanm@codeaurora.org>
Cc: KyongHo Cho <pullip.cho@samsung.com>
Cc: Hiroshi DOYU <hdoyu@nvidia.com>
Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu: omap: Fix compile failure
Ming Lei [Tue, 8 Nov 2011 10:29:15 +0000]
iommu: omap: Fix compile failure

Fix compile failure in drivers/iommu/omap-iommu-debug.c
because of missing module.h include.

Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agointel-iommu: now needs export.h for EXPORT_SYMBOL_GPL
Paul Gortmaker [Sat, 29 Oct 2011 14:26:25 +0000]
intel-iommu: now needs export.h for EXPORT_SYMBOL_GPL

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>

7 years agoiommu/core: Remove global iommu_ops and register_iommu
Joerg Roedel [Tue, 6 Sep 2011 16:58:54 +0000]
iommu/core: Remove global iommu_ops and register_iommu

With all IOMMU drivers being converted to bus_set_iommu the
global iommu_ops are no longer required. The same is true
for the deprecated register_iommu function.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/msm: Use bus_set_iommu instead of register_iommu
Joerg Roedel [Tue, 6 Sep 2011 15:56:07 +0000]
iommu/msm: Use bus_set_iommu instead of register_iommu

Convert the MSM IOMMU driver for ARM to use the new
interface for publishing the iommu_ops.

Acked-by: David Brown <davidb@codeaurora.org>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/omap: Use bus_set_iommu instead of register_iommu
Joerg Roedel [Tue, 6 Sep 2011 15:56:07 +0000]
iommu/omap: Use bus_set_iommu instead of register_iommu

Convert the OMAP IOMMU driver on ARM to use the new
interface for publishing the iommu_ops.

Cc: Ohad Ben-Cohen <ohad@wizery.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/vt-d: Use bus_set_iommu instead of register_iommu
Joerg Roedel [Tue, 6 Sep 2011 15:56:07 +0000]
iommu/vt-d: Use bus_set_iommu instead of register_iommu

Convert the Intel IOMMU driver to use the new interface for
publishing the iommu_ops.

Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/amd: Use bus_set_iommu instead of register_iommu
Joerg Roedel [Tue, 6 Sep 2011 15:56:07 +0000]
iommu/amd: Use bus_set_iommu instead of register_iommu

Convert the AMD IOMMU driver to use the new interface for
publishing the iommu_ops.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/core: Use bus->iommu_ops in the iommu-api
Joerg Roedel [Tue, 6 Sep 2011 14:44:29 +0000]
iommu/core: Use bus->iommu_ops in the iommu-api

Use the per-bus iommu-ops in the functions of the iommu-api
instead of the global iommu_ops.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/core: Convert iommu_found to iommu_present
Joerg Roedel [Tue, 6 Sep 2011 16:46:34 +0000]
iommu/core: Convert iommu_found to iommu_present

With per-bus iommu_ops the iommu_found function needs to
work on a bus_type too. This patch adds a bus_type parameter
to that function and converts all call-places.
The function is also renamed to iommu_present because the
function now checks if an iommu is present for a given bus
and does not check for a global iommu anymore.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/core: Add bus_type parameter to iommu_domain_alloc
Joerg Roedel [Tue, 6 Sep 2011 14:03:26 +0000]
iommu/core: Add bus_type parameter to iommu_domain_alloc

This is necessary to store a pointer to the bus-specific
iommu_ops in the iommu-domain structure. It will be used
later to call into bus-specific iommu-ops.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

Conflicts:

drivers/iommu/iommu.c

Change-Id: Iddbd561739552b663a4be293f1992314eb0f775a

7 years agoDriver core: Add iommu_ops to bus_type
Joerg Roedel [Fri, 26 Aug 2011 14:48:26 +0000]
Driver core: Add iommu_ops to bus_type

This is the starting point to make the iommu_ops used for
the iommu-api a per-bus-type structure. It is required to
easily implement bus-specific setup in the iommu-layer.
The first user will be the iommu-group attribute in sysfs.

Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/core: Define iommu_ops and register_iommu only with CONFIG_IOMMU_API
Joerg Roedel [Tue, 6 Sep 2011 14:48:40 +0000]
iommu/core: Define iommu_ops and register_iommu only with CONFIG_IOMMU_API

This makes it impossible to compile an iommu driver into the
kernel without selecting CONFIG_IOMMU_API.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/omap: always provide iommu debug code
Arnd Bergmann [Sun, 2 Oct 2011 18:34:05 +0000]
iommu/omap: always provide iommu debug code

The iommu module on omap contains a few functions that are
only used by the debug module. These are however only there
when the debug code is built as a module. Since it is possible
to build the debug code into the kernel, the functions should
also be provided for the built-in case.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/core: let drivers know if an iommu fault handler isn't installed
Ohad Ben-Cohen [Tue, 27 Sep 2011 11:36:40 +0000]
iommu/core: let drivers know if an iommu fault handler isn't installed

Make report_iommu_fault() return -ENOSYS whenever an iommu fault
handler isn't installed, so IOMMU drivers can then do their own
platform-specific default behavior if they wanted.

Fault handlers can still return -ENOSYS in case they want to elicit the
default behavior of the IOMMU drivers.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/core: export iommu_set_fault_handler()
Ohad Ben-Cohen [Mon, 26 Sep 2011 13:11:46 +0000]
iommu/core: export iommu_set_fault_handler()

commit 4f3f8d9 "iommu/core: Add fault reporting mechanism" added
the public iommu_set_fault_handler() symbol but forgot to export it.

Fix that.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu: Rename the DMAR and INTR_REMAP config options
Suresh Siddha [Wed, 24 Aug 2011 00:05:25 +0000]
iommu: Rename the DMAR and INTR_REMAP config options

Change the CONFIG_DMAR to CONFIG_INTEL_IOMMU to be consistent
with the other IOMMU options.

Rename the CONFIG_INTR_REMAP to CONFIG_IRQ_REMAP to match the
irq subsystem name.

And define the CONFIG_DMAR_TABLE for the common ACPI DMAR
routines shared by both CONFIG_INTEL_IOMMU and CONFIG_IRQ_REMAP.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: yinghai@kernel.org
Cc: youquan.song@intel.com
Cc: joerg.roedel@amd.com
Cc: tony.luck@intel.com
Cc: dwmw2@infradead.org
Link: http://lkml.kernel.org/r/20110824001456.558630224@sbsiddha-desk.sc.intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>

Conflicts:

arch/x86/include/asm/irq_remapping.h
arch/x86/kernel/apic/io_apic.c
drivers/iommu/Makefile

Change-Id: I3f09baf327e59d94f38801d107268f4e20c1da8e

7 years agoiommu: Cleanup ifdefs in detect_intel_iommu()
Suresh Siddha [Wed, 24 Aug 2011 00:05:22 +0000]
iommu: Cleanup ifdefs in detect_intel_iommu()

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: yinghai@kernel.org
Cc: youquan.song@intel.com
Cc: joerg.roedel@amd.com
Cc: tony.luck@intel.com
Cc: dwmw2@infradead.org
Link: http://lkml.kernel.org/r/20110824001456.386003047@sbsiddha-desk.sc.intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>

7 years agoiommu: No need to set dmar_disabled in check_zero_address()
Suresh Siddha [Wed, 24 Aug 2011 00:05:21 +0000]
iommu: No need to set dmar_disabled in check_zero_address()

Before the restructruing of the x86 IOMMU code,
intel_iommu_init() was getting called directly from
pci_iommu_init() and hence needed to explicitly set
dmar_disabled to 1 for the failure conditions of
check_zero_address().

Recent changes don't call intel_iommu_init() if the intel iommu
detection fails as a result of failure in check_zero_address().

So no need for this ifdef and the code inside it.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: yinghai@kernel.org
Cc: youquan.song@intel.com
Cc: joerg.roedel@amd.com
Cc: tony.luck@intel.com
Cc: dwmw2@infradead.org
Link: http://lkml.kernel.org/r/20110824001456.334878686@sbsiddha-desk.sc.intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>

7 years agoiommu: Move IOMMU specific code to intel-iommu.c
Suresh Siddha [Wed, 24 Aug 2011 00:05:20 +0000]
iommu: Move IOMMU specific code to intel-iommu.c

Move the IOMMU specific routines to intel-iommu.c leaving the
dmar.c to the common ACPI dmar code shared between DMA-remapping
and Interrupt-remapping.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: yinghai@kernel.org
Cc: youquan.song@intel.com
Cc: joerg.roedel@amd.com
Cc: tony.luck@intel.com
Cc: dwmw2@infradead.org
Link: http://lkml.kernel.org/r/20110824001456.282401285@sbsiddha-desk.sc.intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>

7 years agointr_remap: Call dmar_dev_scope_init() explicitly
Suresh Siddha [Wed, 24 Aug 2011 00:05:19 +0000]
intr_remap: Call dmar_dev_scope_init() explicitly

Both DMA-remapping aswell as Interrupt-remapping depend on the
dmar dev scope to be initialized. When both DMA and
IRQ-remapping are enabled, we depend on DMA-remapping init code
to call dmar_dev_scope_init(). This resulted in not doing this
init when DMA-remapping was turned off but interrupt-remapping
turned on in the kernel config.

This caused interrupt routing to break with CONFIG_INTR_REMAP=y
and CONFIG_DMAR=n.

This issue was introduced by this commit:

 | commit 9d5ce73a64be2be8112147a3e0b551ad9cd1247b
 | Author: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
 | Date:   Tue Nov 10 19:46:16 2009 +0900
 |
 |    x86: intel-iommu: Convert detect_intel_iommu to use iommu_init hook

Fix this by calling dmar_dev_scope_init() explicitly from the
interrupt remapping code too.

Reported-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: yinghai@kernel.org
Cc: youquan.song@intel.com
Cc: joerg.roedel@amd.com
Cc: tony.luck@intel.com
Cc: dwmw2@infradead.org
Link: http://lkml.kernel.org/r/20110824001456.229207526@sbsiddha-desk.sc.intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>

7 years agox86, x2apic: Enable the bios request for x2apic optout
Suresh Siddha [Wed, 24 Aug 2011 00:05:18 +0000]
x86, x2apic: Enable the bios request for x2apic optout

On the platforms which are x2apic and interrupt-remapping
capable, Linux kernel is enabling x2apic even if the BIOS
doesn't. This is to take advantage of the features that x2apic
brings in.

Some of the OEM platforms are running into issues because of
this, as their bios is not x2apic aware. For example, this was
resulting in interrupt migration issues on one of the platforms.
Also if the BIOS SMI handling uses APIC interface to send SMI's,
then the BIOS need to be aware of x2apic mode that OS has
enabled.

On some of these platforms, BIOS doesn't have a HW mechanism to
turnoff the x2apic feature to prevent OS from enabling it.

To resolve this mess, recent changes to the VT-d2 specification:

 http://download.intel.com/technology/computing/vptech/Intel(r)_VT_for_Direct_IO.pdf

includes a mechanism that provides BIOS a way to request system
software to opt out of enabling x2apic mode.

Look at the x2apic optout flag in the DMAR tables before
enabling the x2apic mode in the platform. Also print a warning
that we have disabled x2apic based on the BIOS request.

Kernel boot parameter "intremap=no_x2apic_optout" can be used to
override the BIOS x2apic optout request.

Signed-off-by: Youquan Song <youquan.song@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: yinghai@kernel.org
Cc: joerg.roedel@amd.com
Cc: tony.luck@intel.com
Cc: dwmw2@infradead.org
Link: http://lkml.kernel.org/r/20110824001456.171766616@sbsiddha-desk.sc.intel.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>

7 years agoiommu/omap: Fix build error with !IOMMU_SUPPORT
Joerg Roedel [Wed, 14 Sep 2011 14:03:45 +0000]
iommu/omap: Fix build error with !IOMMU_SUPPORT

Without this patch it is possible to select the VIDEO_OMAP3
driver which then selects OMAP_IOVMM. But the omap iommu
driver is not compiled without IOMMU_SUPPORT enabled. Fix
that by forcing OMAP_IOMMU and OMAP_IOVMM are enabled before
VIDEO_OMAP3 can be selected.

Cc: Ohad Ben-Cohen <ohad@wizery.com>
Cc: iommu@lists.linux-foundation.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/omap: Migrate to the generic fault report mechanism
Ohad Ben-Cohen [Tue, 13 Sep 2011 19:26:29 +0000]
iommu/omap: Migrate to the generic fault report mechanism

Start using the generic fault report mechanism, as provided by
the IOMMU core, and remove its now-redundant omap_iommu_set_isr API.

Currently we're only interested in letting upper layers know about the
fault, so in case the faulting device is a remote processor, they could
restart it.

Dynamic PTE/TLB loading is not supported.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/core: Add fault reporting mechanism
Ohad Ben-Cohen [Tue, 13 Sep 2011 19:25:23 +0000]
iommu/core: Add fault reporting mechanism

Add iommu fault report mechanism to the IOMMU API, so implementations
could report about mmu faults (translation errors, hardware errors,
etc..).

Fault reports can be used in several ways:
- mere logging
- reset the device that accessed the faulting address (may be necessary
  in case the device is a remote processor for example)
- implement dynamic PTE/TLB loading

A dedicated iommu_set_fault_handler() API has been added to allow
users, who are interested to receive such reports, to provide
their handler.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agolocking, x86, iommu: Annotate qi->q_lock as raw
Thomas Gleixner [Tue, 19 Jul 2011 15:02:07 +0000]
locking, x86, iommu: Annotate qi->q_lock as raw

The qi->q_lock lock can be taken in atomic context and therefore
cannot be preempted on -rt - annotate it.

In mainline this change documents the low level nature of
the lock - otherwise there's no functional difference. Lockdep
and Sparse checking will work as usual.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

7 years agolocking, x86, iommu: Annotate irq_2_ir_lock as raw
Thomas Gleixner [Tue, 19 Jul 2011 14:28:19 +0000]
locking, x86, iommu: Annotate irq_2_ir_lock as raw

The irq_2_ir_lock can be taken in atomic context and therefore
cannot be preempted on -rt - annotate it.

In mainline this change documents the low level nature of
the lock - otherwise there's no functional difference. Lockdep
and Sparse checking will work as usual.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

7 years agolocking, x86, iommu: Annotate iommu->register_lock as raw
Thomas Gleixner [Tue, 19 Jul 2011 14:19:51 +0000]
locking, x86, iommu: Annotate iommu->register_lock as raw

The iommu->register_lock can be taken in atomic context and therefore
must not be preempted on -rt - annotate it.

In mainline this change documents the low level nature of
the lock - otherwise there's no functional difference. Lockdep
and Sparse checking will work as usual.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

7 years agoiommu/core: Use PAGE_SIZE instead of hard-coded value
Joerg Roedel [Tue, 6 Sep 2011 12:36:17 +0000]
iommu/core: Use PAGE_SIZE instead of hard-coded value

Replace the hard-coded 4kb by PAGE_SIZE to make iommu-api
implementations possible on architectures where
PAGE_SIZE != 4kb.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/core: use the existing IS_ALIGNED macro
Ohad Ben-Cohen [Fri, 2 Sep 2011 17:32:32 +0000]
iommu/core: use the existing IS_ALIGNED macro

Replace iommu's alignment checks with the existing IS_ALIGNED macro,
to drop a few lines of code and utilize IS_ALIGNED's type safety.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/msm: ->unmap() should return order of unmapped page
Ohad Ben-Cohen [Fri, 2 Sep 2011 17:32:34 +0000]
iommu/msm: ->unmap() should return order of unmapped page

Users of the IOMMU API (kvm specifically) assume that iommu_unmap()
returns the order of the unmapped page (on success).

Fix msm_iommu_unmap() accordingly.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Cc: Stepan Moskovchenko <stepanm@codeaurora.org>
Cc: David Brown <davidb@codeaurora.org>
Acked-by: David Brown <davidb@codeaurora.org>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/omap: ->unmap() should return order of unmapped page
Ohad Ben-Cohen [Fri, 2 Sep 2011 17:32:33 +0000]
iommu/omap: ->unmap() should return order of unmapped page

Users of the IOMMU API (kvm specifically) assume that iommu_unmap()
returns the order of the unmapped page.

Fix omap_iommu_unmap() to do so and adopt omap-iovmm accordingly.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/omap: cleanup: remove a redundant statement
Ohad Ben-Cohen [Fri, 2 Sep 2011 17:32:31 +0000]
iommu/omap: cleanup: remove a redundant statement

Tiny cleanup that removes a redundant 'return' statement.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu/omap-iovmm: support non page-aligned buffers in iommu_vmap
Laurent Pinchart [Fri, 2 Sep 2011 17:32:30 +0000]
iommu/omap-iovmm: support non page-aligned buffers in iommu_vmap

omap_iovmm requires page-aligned buffers, and that sometimes causes
omap3isp failures (i.e. whenever the buffer passed from userspace is not
page-aligned).

Remove this limitation by rounding the address of the first page entry
down, and adding the offset back to the device address.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: Hiroshi DOYU <Hiroshi.DOYU@nokia.com>
[ohad@wizery.com: rebased, but tested only with aligned buffers]
[ohad@wizery.com: slightly edited the commit log]
Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoiommu: omap: add Kconfig OMAP dependency
Ohad Ben-Cohen [Mon, 29 Aug 2011 04:57:44 +0000]
iommu: omap: add Kconfig OMAP dependency

Make CONFIG_OMAP_IOMMU depend on CONFIG_ARCH_OMAP so other
allmodconfig builds won't fail.

Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoomap: iommu: Fix up mutex->spin_lock conversion of iommu_lock
Joerg Roedel [Fri, 26 Aug 2011 11:20:06 +0000]
omap: iommu: Fix up mutex->spin_lock conversion of iommu_lock

The omap_iommu_set_isr() was still using the mutex functions
but the iommu_lock was converted to a spin_lock. Fix that
up.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoomap: iommu: omapify 'struct iommu' and exposed API
Ohad Ben-Cohen [Wed, 17 Aug 2011 19:57:56 +0000]
omap: iommu: omapify 'struct iommu' and exposed API

Prepend 'omap_' to OMAP's 'struct iommu' and exposed API, to prevent
namespace pollution and generally to improve readability of the code
that still uses the driver directly.

Update the users as needed as well.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: Hiroshi DOYU <Hiroshi.DOYU@nokia.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

Conflicts:

arch/arm/plat-omap/include/plat/iommu.h
arch/arm/plat-omap/include/plat/iovmm.h
drivers/iommu/omap-iommu.c
drivers/iommu/omap-iovmm.c
drivers/media/video/omap3isp/isp.c
drivers/media/video/omap3isp/isp.h
drivers/media/video/omap3isp/ispccdc.c
drivers/media/video/omap3isp/ispstat.c
drivers/media/video/omap3isp/ispvideo.c

Change-Id: I6e4825d93115596184ed3e2274fa759d13a5d6dc

7 years agoomap: iovmm: remove unused functionality
Ohad Ben-Cohen [Tue, 16 Aug 2011 12:31:16 +0000]
omap: iovmm: remove unused functionality

Remove unused functionality from OMAP's iovmm module.

The intention is to eventually completely replace iovmm with the
generic DMA-API, so new code that'd need this iovmm functionality
will have to extend the DMA-API instead.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Acked-by: Hiroshi DOYU <Hiroshi.DOYU@nokia.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

Conflicts:

arch/arm/plat-omap/include/plat/iovmm.h
drivers/iommu/omap-iovmm.c

Change-Id: I9ebfc8aaeeb28d4fc7d733f0f7fe146c0af7431b

7 years agoomap: iommu: PREFETCH_IOTLB cleanup
Ohad Ben-Cohen [Tue, 16 Aug 2011 12:19:10 +0000]
omap: iommu: PREFETCH_IOTLB cleanup

Use PREFETCH_IOTLB to control the content of the called function,
instead of inlining it in the code.

This improves readability of the code, and also prevents an "unused
function" warning to show up when PREFETCH_IOTLB isn't set.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Acked-by: Hiroshi DOYU <Hiroshi.DOYU@nokia.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoomap: iommu: stop exporting local functions
Ohad Ben-Cohen [Tue, 16 Aug 2011 11:58:14 +0000]
omap: iommu: stop exporting local functions

Stop exporting functions that are used only within the iommu
driver itself.

Eventually OMAP's iommu driver should only expose API via the generic
IOMMU framework.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Acked-by: Hiroshi DOYU <Hiroshi.DOYU@nokia.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

7 years agoomap: iommu/iovmm: move to dedicated iommu folder
Ohad Ben-Cohen [Mon, 15 Aug 2011 20:21:41 +0000]
omap: iommu/iovmm: move to dedicated iommu folder

Move OMAP's iommu drivers to the dedicated iommu drivers folder.

While OMAP's iovmm (virtual memory manager) driver does not strictly
belong to the iommu drivers folder, move it there as well, because
it's by no means OMAP-specific (in concept. technically it is still
coupled with OMAP's iommu).

Eventually, iovmm will be completely replaced with the generic,
iommu-based, dma-mapping API.

Signed-off-by: Ohad Ben-Cohen <ohad@wizery.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: Hiroshi DOYU <Hiroshi.DOYU@nokia.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>

Conflicts:

arch/arm/plat-omap/Kconfig
drivers/iommu/Kconfig
drivers/iommu/Makefile

Change-Id: Ie437869b8cbb92ea8fafc3d18333331a3b51ec06

7 years agoARM: tegra: usb: check for valid phy clock
Krishna Yarlagadda [Fri, 18 Nov 2011 09:32:17 +0000]
ARM: tegra: usb: check for valid phy clock

Check for valid clock after setting up phy
in UTMIP reset

Bug 853708
Bug 886080

(reviewed on http://git-master/r/51555)
(cherry picked from commit 25b340b64ca27971f6d2d555299a749f0394e77e)

Change-Id: I98c79fb1e8961b9af19f3001fe6fbde2b28509a0
Reviewed-on: http://git-master/r/75502
Reviewed-by: Simone Willett <swillett@nvidia.com>
Signed-off-by: Ken Chang <kenc@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/76468
Reviewed-by: Automatic_Commit_Validation_User

7 years agoASoC: WM8903: Use 2x of miniumum bitclk in master mode
Sumit Bhattacharya [Tue, 17 Jan 2012 13:56:18 +0000]
ASoC: WM8903: Use 2x of miniumum bitclk in master mode

Instead of using tight bitclk use 2x of minimum bitclk requirement
in codec master mode. This change is needed to have clean audio
in codec master mode on Tegra30 based cardhu platform.

Bug 915107

Change-Id: Ia59e1a906415ad8fd1ed324ab4dabef61746b62f
Reviewed-on: http://git-master/r/75706
Reviewed-by: Ravindra Lokhande <rlokhande@nvidia.com>
Reviewed-by: Scott Peterson <speterson@nvidia.com>
Signed-off-by: Sumit Bhattacharya <sumitb@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/76467
Reviewed-by: Automatic_Commit_Validation_User

7 years agoALSA: HDA: Fix MODPOST errors
Sumit Bhattacharya [Tue, 17 Jan 2012 12:48:57 +0000]
ALSA: HDA: Fix MODPOST errors

Bug 923312

Change-Id: I7c68f261989d2ca1ef7b273b7d41a519bc91c868
Reviewed-on: http://git-master/r/75700
Reviewed-by: Ravindra Lokhande <rlokhande@nvidia.com>
Reviewed-by: Scott Peterson <speterson@nvidia.com>
Reviewed-by: Scott Williams <scwilliams@nvidia.com>
Signed-off-by: Sumit Bhattacharya <sumitb@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/76466
Reviewed-by: Automatic_Commit_Validation_User

7 years agoarm: tegra: cardhu: Use fixed regulator
Laxman Dewangan [Wed, 11 Jan 2012 10:25:27 +0000]
arm: tegra: cardhu: Use fixed regulator

Using fixed regulator in place of gpio-switch regulator.
Fixed regulator is very similar to gpio-switch regulator
and it is available by default in linux kernel.

gpio-switch-regulator will be used only for the open
collector load switches.

bug 923713

Change-Id: I42480cb314135d475b4d4b56bb5e8e73eba2f77e
Reviewed-on: http://git-master/r/74536
Reviewed-by: Simone Willett <swillett@nvidia.com>
Signed-off-by: Laxman Dewangan <ldewangan@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/76465
Reviewed-by: Automatic_Commit_Validation_User

7 years agotegra: usb: phy: support for minus calib offset
Ken Chang [Mon, 17 Oct 2011 05:26:15 +0000]
tegra: usb: phy: support for minus calib offset

need to support negative values for usb calibration.
change xcvr_setup_offset from unsigned to signed.

bug 872648

(cherry picked from commit 06258b46589436b5579c8265405b1cb286c406aa)
(reviewed on http://git-master/r/66101)

Change-Id: I1ab6a63184fe48bc734152546a541085ac7c6efc
Reviewed-on: http://git-master/r/74503
Reviewed-by: Simone Willett <swillett@nvidia.com>
Signed-off-by: Ken Chang <kenc@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/76464
Reviewed-by: Automatic_Commit_Validation_User

7 years agoARM: tegra: clock: Add Tegra3 0.95V core voltage step
Alex Frid [Tue, 10 Jan 2012 03:46:25 +0000]
ARM: tegra: clock: Add Tegra3 0.95V core voltage step

- Expanded Tegra3 DVFS tables with 0.95V core voltage step
- Updated cbus minimum rate calculation, since cbus can not
  run at 0.95V
- Updated PLLM dvfs initialization, since PLLM can no longer
  be voltage independent, even when its usage is restricted.

Bug 817679
Bug 841336

Change-Id: I4973dc19d351ce237f2b249ebf75a79abf3afef4
Reviewed-on: http://git-master/r/74141
Reviewed-by: Yu-Huan Hsu <yhsu@nvidia.com>
Reviewed-by: Diwakar Tundlam <dtundlam@nvidia.com>
Signed-off-by: Alex Frid <afrid@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/76463
Reviewed-by: Automatic_Commit_Validation_User

7 years agoregulator: Fix possible nullpointer dereference in regulator_enable()
Heiko Stübner [Fri, 14 Oct 2011 16:00:29 +0000]
regulator: Fix possible nullpointer dereference in regulator_enable()

In the case where _regulator_enable returns an error it was not checked
if a supplying regulator exists before trying to disable it, leading
to a null pointer-dereference if no supplying regulator existed.

Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Cherry-pick from main stream:
 commit d1685e4e2c3854782272f32b71f2f3eff5c6e0d0

Change-Id: Ie4d8db9184a65e6a98964259299bfc7daa0f4048
Reviewed-on: http://git-master/r/75908
Reviewed-by: Pavan Kunapuli <pkunapuli@nvidia.com>
Signed-off-by: Laxman Dewangan <ldewangan@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/76462
Reviewed-by: Automatic_Commit_Validation_User

7 years agoDTV: dma: fixed dma burst size issue DTV xmit
Adam Jiang [Fri, 13 Jan 2012 16:16:02 +0000]
DTV: dma: fixed dma burst size issue DTV xmit

Since dtv interface was designed based on SPI bus, it shares the same
dma configuration with SPI bus. However, it is not proper because DTV
interface has to transmit data in 4 words long constantly. The patch
added an option in tegra_dma_req to set burst size to be fixed 4 words.

fixed Bug 910227

Change-Id: I1436f0c8d108dd39edc57ae4c9cb750d9574b62c
Reviewed-on: http://git-master/r/75509
Reviewed-by: David Schalig <dschalig@nvidia.com>
Reviewed-by: Laxman Dewangan <ldewangan@nvidia.com>
Signed-off-by: Adam Jiang <chaoj@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/76461
Reviewed-by: Automatic_Commit_Validation_User

7 years agocdc_ncm: Modify NCM network interface.
BH Hsieh [Wed, 18 Jan 2012 05:10:14 +0000]
cdc_ncm: Modify NCM network interface.

Adding FLAG_RMNET to change ncm net device name
for Ericsson modem.

Bug 901367

Change-Id: I4a029cc10a02176035319ed4ac1e7b9af349cbd1
Reviewed-on: http://git-master/r/74912
Reviewed-by: WK Tsai <wtsai@nvidia.com>
Reviewed-by: Steve Lin <stlin@nvidia.com>
Signed-off-by: BH Hsieh <bhsieh@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/76459
Reviewed-by: Automatic_Commit_Validation_User

7 years agoarm: config: tegra: Enable NCT1008
Wei Ni [Wed, 11 Jan 2012 11:20:34 +0000]
arm: config: tegra: Enable NCT1008

Enable CONFIG_NCT1008

Change-Id: I700557b8900401b8f9ab36faab5903ba0afde542
Reviewed-on: http://git-master/r/74564
Reviewed-by: Allen Martin <amartin@nvidia.com>
Signed-off-by: Wei Ni <wni@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/76458
Reviewed-by: Automatic_Commit_Validation_User

7 years agoARM: tegra: cardhu: update elpida dqs
Ray Poudrier [Sat, 7 Jan 2012 00:59:58 +0000]
ARM: tegra: cardhu: update elpida dqs

Bug 852560

Change-Id: I335bed2bb4a35e8c740ee54829c1c2e3203d97aa
Reviewed-on: http://git-master/r/73838
Reviewed-by: Aleksandr Frid <afrid@nvidia.com>
Reviewed-by: Yu-Huan Hsu <yhsu@nvidia.com>
Signed-off-by: Ray Poudrier <rapoudrier@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/76457
Reviewed-by: Automatic_Commit_Validation_User

7 years agocrypto: tegra-aes: get hardware ownership before copying key
Varun Wadekar [Thu, 19 Jan 2012 05:45:17 +0000]
crypto: tegra-aes: get hardware ownership before copying key

Bug 917607

Change-Id: I049ca03efe7953dc6a6c03eaa4acce85e15662d1
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/76154
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Sanjay Singh Rawat <srawat@nvidia.com>

7 years agocrypto: tegra-aes: remove aes_lock usage
Varun Wadekar [Wed, 18 Jan 2012 13:45:18 +0000]
crypto: tegra-aes: remove aes_lock usage

Bug 917607

Change-Id: I6036b12456d3b5fb22f479a9e0eefd500cb6c059
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/76153
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Sanjay Singh Rawat <srawat@nvidia.com>

7 years agocrypto: tegra-aes: rng: check if bsea is busy before using it
Varun Wadekar [Mon, 12 Dec 2011 07:02:36 +0000]
crypto: tegra-aes: rng: check if bsea is busy before using it

Bug 917607

Change-Id: I7497411932b6ba8c155026f1662063e87c21eb40
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/74012
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Sanjay Singh Rawat <srawat@nvidia.com>

7 years agousb: gadget: fsl_udc: use can_pullup function in vbus_session.
Jubeom Kim [Wed, 18 Jan 2012 13:52:43 +0000]
usb: gadget: fsl_udc: use can_pullup function in vbus_session.

Even if the value of is_active is true, vbus_session is needed to check
whether udc can do pullup or not.
If the return value of can_pullup function is false
because of udc->softconnect value, udc has to make the STOP status by pullup.

Bug 925453

Change-Id: Iaefbc44cc3b3aefa2bb8e2a42d7c042ad535b308
Signed-off-by: Jubeom Kim <jubeomk@nvidia.com>
Reviewed-on: http://git-master/r/75954
Reviewed-by: Rakesh Bodla <rbodla@nvidia.com>
Tested-by: Rakesh Bodla <rbodla@nvidia.com>
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Kurt Yi <kyi@nvidia.com>
Tested-by: Kurt Yi <kyi@nvidia.com>
Reviewed-by: Varun Wadekar <vwadekar@nvidia.com>

7 years agotegra: nor: always build partitioning support
Preetham Chandru [Wed, 18 Jan 2012 09:43:38 +0000]
tegra: nor: always build partitioning support

Remove the CONFIG_MTD_PARTITIONS preprocessor conditionals as
partitioning is always available.
Moreover as none of the drivers use CONFIG_MTD_PARTITIONS, this has
been undefined in Kconfig from 2.6.39 onwords.
The following commit 6a8a98b22b10f1560d5f90aded4a54234b9b2724 has
removed the CONFIG_MTD_PARTITIONS in Kconfig.

Bug: 923135
Change-Id: I8acb6ac2df26d9983b4cfe5869bc97a5861da573
Signed-off-by: Preetham Chandru R <pchandru@nvidia.com>
Reviewed-on: http://git-master/r/75895
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Shashank Sharma <shashanks@nvidia.com>
Reviewed-by: Kiran Adduri <kadduri@nvidia.com>
Reviewed-by: Varun Wadekar <vwadekar@nvidia.com>

7 years agocrypto: tegra-se: Fix SG buffer length mismatch
Mallikarjun Kasoju [Tue, 17 Jan 2012 14:22:05 +0000]
crypto: tegra-se: Fix SG buffer length mismatch

Fixed the case where SG buffer length is more than actual number
of bytes to process.

Bug 922857

Change-Id: I8445d8ae74b8fc6c964c19523fcd731f7ba4bd37
Reviewed-on: http://git-master/r/75701
Signed-off-by: Mallikarjun Kasoju <mkasoju@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/75891
Reviewed-by: Automatic_Commit_Validation_User

7 years agomfd: max77663: Support Low-Power Mode during sleep mode
Jin Park [Tue, 17 Jan 2012 05:48:47 +0000]
mfd: max77663: Support Low-Power Mode during sleep mode

Put the power rails into Low-Power Mode during sleep mode,
if SLP_LPM is enabled and the power rail's power mode is GLPM.

Bug 924686

Change-Id: I460f8bfa6be73a440a4924160d8e271695fc4973
Reviewed-on: http://git-master/r/75626
Signed-off-by: Jin Park <jinyoungp@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/75890
Reviewed-by: Automatic_Commit_Validation_User

7 years agoarm: tegra: defconfig: build cfg80211 as module
Om Prakash Singh [Thu, 12 Jan 2012 13:09:47 +0000]
arm: tegra: defconfig: build cfg80211 as module

Bug 926128

Change-Id: I326e3806a429e031bea78a3e051c22518e705f86
Reviewed-on: http://git-master/r/74973
Signed-off-by: Om Prakash Singh <omp@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/75889
Reviewed-by: Automatic_Commit_Validation_User

7 years agousb: host: tegra: replace CONFIG_USB_HOTPLUG to tegra_ehci_platform_data
Kurt Yi [Wed, 11 Jan 2012 12:11:07 +0000]
usb: host: tegra: replace CONFIG_USB_HOTPLUG to tegra_ehci_platform_data

remove CONFIG_USB_HOTPLUG and add default_enable into tegra_ehci_platform_data
We can't use different def_config because whistler & ventana source code
is merged in same tree.

BUG 923705

Change-Id: I3dfdf237d80865831264f02104c89c8088101905
Reviewed-on: http://git-master/r/74574
Signed-off-by: Kurt Yi <kyi@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/75888
Reviewed-by: Varun Colbert <vcolbert@nvidia.com>
Tested-by: Varun Colbert <vcolbert@nvidia.com>

7 years agoarm: tegra2: defconfig: enable nct1008
Artiste Hsu [Tue, 10 Jan 2012 09:37:47 +0000]
arm: tegra2: defconfig: enable nct1008

bug 924363

Change-Id: Ic5e88e0887d912234e01a759674945524a3350a1
Reviewed-on: http://git-master/r/74258
Signed-off-by: Artiste Hsu <chhsu@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/75887
Reviewed-by: Automatic_Commit_Validation_User

7 years agocpufreq: Preserve sysfs min/max request
Antti P Miettinen [Thu, 5 Jan 2012 00:13:23 +0000]
cpufreq: Preserve sysfs min/max request

Store the value received via sysfs as the user_policy
min/max value instead of the currently enforced min/max.
This allows restoring the user min/max values when
constraints on enforced min/max change.

Bug 888312

Change-Id: I8a5adf022585e5bd91fa3b1d8d448065efa6827a
Reviewed-on: http://git-master/r/73982
Signed-off-by: Antti P Miettinen <amiettinen@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/75886
Reviewed-by: Automatic_Commit_Validation_User

7 years agovideo: tegra: host: Fix includes
Juha Tukkinen [Mon, 2 Jan 2012 15:18:38 +0000]
video: tegra: host: Fix includes

Remove one duplicate include and relative paths in includes. Replace one
include with forward declaration. Make dependencies to t20 and t30
explicit.

Change-Id: I195020bedbb46277dbd6a92b65be2cde5f6276b8
Reviewed-on: http://git-master/r/72903
Signed-off-by: Juha Tukkinen <jtukkinen@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/75885
Reviewed-by: Automatic_Commit_Validation_User

7 years agocpufreq: Enforce PM QoS min/max limits
Antti P Miettinen [Tue, 27 Dec 2011 10:41:39 +0000]
cpufreq: Enforce PM QoS min/max limits

Observe PM QoS CPU frequency minimum and maximum in addition
to policy settings.

Bug 888312

Change-Id: Ia4f60a1649a9952e02f6847c8add3b2ea5d47524
Reviewed-on: http://git-master/r/72207
Signed-off-by: Antti P Miettinen <amiettinen@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/75884
Reviewed-by: Automatic_Commit_Validation_User

7 years agoPM QoS: Add CPU frequency min/max as PM QoS params
Antti P Miettinen [Tue, 27 Dec 2011 10:28:21 +0000]
PM QoS: Add CPU frequency min/max as PM QoS params

Add minimum and maximum CPU frequency as PM QoS parameters.

Bug 888312

Change-Id: I18abddded35a044a6ad8365035e31d1a2213a329
Reviewed-on: http://git-master/r/72206
Signed-off-by: Antti P Miettinen <amiettinen@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/75883
Reviewed-by: Automatic_Commit_Validation_User

7 years agocpufreq: Export user_policy min/max
Antti P Miettinen [Tue, 27 Dec 2011 10:09:56 +0000]
cpufreq: Export user_policy min/max

Add sysfs nodes for user_policy min and max settings.

Bug 888312

Change-Id: Ife8ec07f8faebc0c04e7ecf6e24976a311bad73b
Reviewed-on: http://git-master/r/72201
Signed-off-by: Antti P Miettinen <amiettinen@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/75882
Reviewed-by: Automatic_Commit_Validation_User

7 years agovideo: tegra: nvmap: Remove high mem page cache flush.
Krishna Reddy [Fri, 2 Dec 2011 20:42:20 +0000]
video: tegra: nvmap: Remove high mem page cache flush.

Remove cache flush for high mem pages from nvmap as CPA is taking
care of cache flush for high mem pages.
Bug 865816

Change-Id: Ia7b06cdd80a16de0996c119dd950e68c5fd24b5b
Reviewed-on: http://git-master/r/71726
Signed-off-by: Krishna Reddy <vdumpa@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/75881
Reviewed-by: Automatic_Commit_Validation_User

7 years agoarm: mm: Optimze cache flush in CPA.
Krishna Reddy [Fri, 2 Dec 2011 20:40:21 +0000]
arm: mm: Optimze cache flush in CPA.

Optimze cache flush time and enable cache flush for high mem pages in CPA.
Bug 865816

Change-Id: I15736010bd26c18ea0d3350c15769675f07ac055
Reviewed-on: http://git-master/r/71725
Signed-off-by: Krishna Reddy <vdumpa@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/75880
Reviewed-by: Automatic_Commit_Validation_User

7 years agoarm: tegra: usb: Use T30 specific hsic bus reset
Vinayak Pane [Fri, 9 Dec 2011 05:38:51 +0000]
arm: tegra: usb: Use T30 specific hsic bus reset

HSIC bus reset apparently needs a WAR. The ehci generic
port reset is not enough for modem to connect in stress
tests.
Instead use pull-down pull-up on Strobe signal of
USB HSIC bus. This is AP30 specific workaround.

Bug 898008
Bug 912850
Bug 922444

Change-Id: I4f542d0398b05c388f6fa7e08477b65877dc705c
Reviewed-on: http://git-master/r/73266
Signed-off-by: Vinayak Pane <vpane@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/75879
Reviewed-by: Automatic_Commit_Validation_User

7 years agoarm: tegra3: dvfs: update tegra3 speedo thresholds
Ken Chang [Fri, 13 Jan 2012 02:31:58 +0000]
arm: tegra3: dvfs: update tegra3 speedo thresholds

bug 921903

Change-Id: I8932dbde45cb0b025ba8a537b528ebec82c3424e
Signed-off-by: Ken Chang <kenc@nvidia.com>
Reviewed-on: http://git-master/r/75130
Reviewed-by: Diwakar Tundlam <dtundlam@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/75878
Reviewed-by: Automatic_Commit_Validation_User

7 years agodrivers: misc: nct: Show lo/hi alerts in nct debug
Diwakar Tundlam [Thu, 12 Jan 2012 20:24:02 +0000]
drivers: misc: nct: Show lo/hi alerts in nct debug

Also fixed some error strings

Reviewed-on: http://git-master/r/75035

Change-Id: Ia071d25f4e0402c55bdc9830cca948759e4aea2c
Signed-off-by: Diwakar Tundlam <dtundlam@nvidia.com>
Reviewed-by: Joshua Primero <jprimero@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/75877
Reviewed-by: Automatic_Commit_Validation_User
Tested-by: Joshua Primero <jprimero@nvidia.com>

7 years agoarm: tegra: baseband: cleanup code for modem power driver
Seshendra Gadagottu [Tue, 27 Dec 2011 04:08:24 +0000]
arm: tegra: baseband: cleanup code for modem power driver

Disable wakeup irq when removing the modem power driver.
BUG 920108

Change-Id: I371644421fc7d986af2292f736b459b141651ed1
Signed-off-by: Steve Lin <stlin@nvidia.com>
Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/75876
Reviewed-by: Automatic_Commit_Validation_User

7 years agotegra: nand: always build partitioning support
Preetham Chandru [Wed, 18 Jan 2012 09:31:40 +0000]
tegra: nand: always build partitioning support

Remove the CONFIG_MTD_PARTITIONS preprocessor conditionals as
partitioning is always available.
Moreover as none of the drivers use CONFIG_MTD_PARTITIONS, this has
been undefined in Kconfig from 2.6.39 onwords.
The following commit 6a8a98b22b10f1560d5f90aded4a54234b9b2724 has
removed the CONFIG_MTD_PARTITIONS in Kconfig.

Bug: 923135
Change-Id: Id75da462c4ec58baf4da72a0f210b02517374de6
Signed-off-by: Preetham Chandru R <pchandru@nvidia.com>
Reviewed-on: http://git-master/r/75871
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Venu Byravarasu <vbyravarasu@nvidia.com>
Reviewed-by: Shashank Sharma <shashanks@nvidia.com>
Reviewed-by: Kiran Adduri <kadduri@nvidia.com>
Reviewed-by: Varun Wadekar <vwadekar@nvidia.com>

7 years agoarm: tegra: ventana: combine all camera gpios init stuffs
Pritesh Raithatha [Tue, 17 Jan 2012 10:13:55 +0000]
arm: tegra: ventana: combine all camera gpios init stuffs

Bug 925547

Change-Id: Iee05a2dda861b4a630045caaeb341f0365a74e12
Signed-off-by: Pritesh Raithatha <praithatha@nvidia.com>
Reviewed-on: http://git-master/r/75690
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-by: Mayuresh Kulkarni <mkulkarni@nvidia.com>

7 years agoarm: tegra: ventana: add defines and cleanup header file
Pritesh Raithatha [Tue, 17 Jan 2012 10:05:40 +0000]
arm: tegra: ventana: add defines and cleanup header file

-move sensors defines to board header file.
-add defines for pca954x i2c mux instead of using direct i2c
bus number.
-cleanup board-ventana.h.

Bug 925547

Change-Id: I76a42b66f3a3fa0adbc8c06b2edb9fd006c7ae44
Signed-off-by: Pritesh Raithatha <praithatha@nvidia.com>
Reviewed-on: http://git-master/r/75689
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-by: Mayuresh Kulkarni <mkulkarni@nvidia.com>

7 years agousb: gadget: android: Don't allow changing the functions list if enabled
Benoit Goby [Sat, 10 Dec 2011 02:05:00 +0000]
usb: gadget: android: Don't allow changing the functions list if enabled

cherry-pick I3ad39b420ce79a8602a7eca1daac1f56b30bad5c from
https://android.googlesource.com/kernel/common.git

Change-Id: Icb16df8b5227e471a202c8207ecfbb1b8414a7c6
Signed-off-by: Benoit Goby <benoit@android.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/75464
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Rakesh Bodla <rbodla@nvidia.com>

7 years agommc: Set suspend/resume bus operations if CONFIG_PM_RUNTIME is used
Dmitry Shmidt [Sat, 17 Dec 2011 01:52:18 +0000]
mmc: Set suspend/resume bus operations if CONFIG_PM_RUNTIME is used

cherry-pick ea693bf7f87603b072f4ceea6684221fa0b8e863 from
https://android.googlesource.com/kernel/common.git

Change-Id: I9575b542af664973a03ad35d7bc48da130ce5a89
Signed-off-by: Dmitry Shmidt <dimitrysh@google.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/75463
Reviewed-by: Automatic_Commit_Validation_User

7 years agowireless: Protect regdomain change by mutex
Dmitry Shmidt [Mon, 19 Dec 2011 20:32:21 +0000]
wireless: Protect regdomain change by mutex

cherry-pick 35047200c401cdd1bb43b5574270d1ac86a47937 from
https://android.googlesource.com/kernel/common.git

Change-Id: I23efd4bab367bf49f7aca34daa19904baf5b9170
Signed-off-by: Dmitry Shmidt <dimitrysh@google.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/75462
Reviewed-by: Automatic_Commit_Validation_User

7 years agoRevert "usb: gadget: rndis: don't use dev_get_stats"
Varun Wadekar [Sat, 14 Jan 2012 07:12:19 +0000]
Revert "usb: gadget: rndis: don't use dev_get_stats"

This reverts commit cf1684b901a1fc239512760beea224090e56aa2f.

Not needed anymore in 2.6.39 and 3.0, dev_get_stats has been fixed
and may be called from atomic context. See:
1ac9ad1 net: remove dev_txq_stats_fold()

cherry-pick f16e6fb3e3bbbc279579b3c3def7951787c88a0a from
https://android.googlesource.com/kernel/common.git

Change-Id: If51b9959add812a057ab4cfd6c6ddf7acf0e4c02
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/75461
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Rakesh Bodla <rbodla@nvidia.com>

7 years agoram_console: set CON_ANYTIME console flag
Varun Wadekar [Sat, 14 Jan 2012 06:53:26 +0000]
ram_console: set CON_ANYTIME console flag

We want to ensure that we get all the console messages, even ones
that occur while the printing CPU is not yet online.

cherry-pick a65e28a0149261776678977962cfa0f90973e1d4 from
https://android.googlesource.com/kernel/common.git

Change-Id: I8b83014a95fade7200e5201c29fb6624e1140aaf
Signed-off-by: Dima Zavin <dima@android.com>
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Reviewed-on: http://git-master/r/75460
Reviewed-by: Automatic_Commit_Validation_User

7 years agoBluetooth: Drop rfcomm session reference only once for incoming session
Ville Tervo [Fri, 26 Feb 2010 10:21:01 +0000]
Bluetooth: Drop rfcomm session reference only once for incoming session

Move decision to drop reference for incoming session to
rfcomm_session_close to get more clear
rfcomm_session_hold()/rfcomm_session_put() pairs.

Rebase by: Jaikumar Ganesh <jaikumarg@android.com>

Signed-off-by: Ville Tervo <ville.tervo@nokia.com>
Signed-off-by: Jaikumar Ganesh <jaikumarg@android.com>
(cherry picked from commit dacd12646bbe15c98ab2b74b266092271a41d278)

Change-Id: I6b9374bb732d4f9706a65be19f8dbfddbe1de2df
Signed-off-by: Nagarjuna Kristam <nkristam@nvidia.com>
Reviewed-on: http://git-master/r/75236
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>
Reviewed-by: Rakesh Goyal <rgoyal@nvidia.com>

7 years agolib: genalloc: Add API to allocate at specified addr.
Krishna Reddy [Wed, 11 Jan 2012 00:33:33 +0000]
lib: genalloc: Add API to allocate at specified addr.

Add API to allocate at specified alloc address.

Change-Id: I188e5430220c050026c6a3e17a586012d9a9fa04
Signed-off-by: Krishna Reddy <vdumpa@nvidia.com>
Reviewed-on: http://git-master/r/74468
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Stephen Warren <swarren@nvidia.com>