platform/kernel/linux-starfive.git
6 years agomei: define dma ring buffer sizes for PCH12 HW and newer
Alexander Usyskin [Tue, 31 Jul 2018 06:35:37 +0000 (09:35 +0300)]
mei: define dma ring buffer sizes for PCH12 HW and newer

Define dma ring buffer sizes for PCH12 (CLN HW and newer)

Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agomei: restrict dma ring support to hbm version 2.1
Tomas Winkler [Tue, 31 Jul 2018 06:35:36 +0000 (09:35 +0300)]
mei: restrict dma ring support to hbm version 2.1

Only a firmware with version 2.1 and above supports dma ring feature.

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agomei: hbm: introduce dma bit in the message header
Tomas Winkler [Tue, 31 Jul 2018 06:35:35 +0000 (09:35 +0300)]
mei: hbm: introduce dma bit in the message header

Add dma_ring bit in the mei message header for conveying
that the message data itself are on the dma ring.

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agomei: hbm: define dma ring setup protocol
Tomas Winkler [Tue, 31 Jul 2018 06:35:34 +0000 (09:35 +0300)]
mei: hbm: define dma ring setup protocol

The protocol defines how to setup an I/O ring on top of host
memory to utilize the device DMA engine for faster transport.

Three memory buffers are allocated.
A Host circular buffer for from the Host to Device communication.
A Device circular buffer for from Device to the Host communication.
And finally a Control block where the pointers for the both
circular buffers are managed.

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agomei: add support for variable length mei headers.
Tomas Winkler [Tue, 31 Jul 2018 06:35:33 +0000 (09:35 +0300)]
mei: add support for variable length mei headers.

Remove header size knowledge from me and txe hw layers,
this requires to change the write handler to accept
header and its length as well as data and its length.

HBM messages are fixed to use basic header, hence we add mei_hbm2slots()
that converts HBM message length and mei message header,
while mei_data2slots() converts data length directly to the slots.

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agouio: fix possible circular locking dependency
Xiubo Li [Mon, 30 Jul 2018 07:11:48 +0000 (03:11 -0400)]
uio: fix possible circular locking dependency

The call trace:
XXX/1910 is trying to acquire lock:
 (&mm->mmap_sem){++++++}, at: [<ffffffff97008c87>] might_fault+0x57/0xb0

but task is already holding lock:
 (&idev->info_lock){+.+...}, at: [<ffffffffc0638a06>] uio_write+0x46/0x130 [uio]

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:

-> #1 (&idev->info_lock){+.+...}:
       [<ffffffff96f31fc9>] lock_acquire+0x99/0x1e0
       [<ffffffff975edad3>] mutex_lock_nested+0x93/0x410
       [<ffffffffc063873d>] uio_mmap+0x2d/0x170 [uio]
       [<ffffffff97016b58>] mmap_region+0x428/0x650
       [<ffffffff97017138>] do_mmap+0x3b8/0x4e0
       [<ffffffff96ffaba3>] vm_mmap_pgoff+0xd3/0x120
       [<ffffffff97015261>] SyS_mmap_pgoff+0x1f1/0x270
       [<ffffffff96e387c2>] SyS_mmap+0x22/0x30
       [<ffffffff975ff315>] system_call_fastpath+0x1c/0x21

-> #0 (&mm->mmap_sem){++++++}:
       [<ffffffff96f30e9c>] __lock_acquire+0xdac/0x15f0
       [<ffffffff96f31fc9>] lock_acquire+0x99/0x1e0
       [<ffffffff97008cb4>] might_fault+0x84/0xb0
       [<ffffffffc0638a74>] uio_write+0xb4/0x130 [uio]
       [<ffffffff9706ffa3>] vfs_write+0xc3/0x1f0
       [<ffffffff97070e2a>] SyS_write+0x8a/0x100
       [<ffffffff975ff315>] system_call_fastpath+0x1c/0x21

other info that might help us debug this:
 Possible unsafe locking scenario:
       CPU0                    CPU1
       ----                    ----
  lock(&idev->info_lock);
                               lock(&mm->mmap_sem);
                               lock(&idev->info_lock);
  lock(&mm->mmap_sem);

 *** DEADLOCK ***
1 lock held by XXX/1910:
 #0:  (&idev->info_lock){+.+...}, at: [<ffffffffc0638a06>] uio_write+0x46/0x130 [uio]

stack backtrace:
CPU: 0 PID: 1910 Comm: XXX Kdump: loaded Not tainted #1
Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 05/19/2017
Call Trace:
 [<ffffffff975e9211>] dump_stack+0x19/0x1b
 [<ffffffff975e260a>] print_circular_bug+0x1f9/0x207
 [<ffffffff96f2f6a7>] check_prevs_add+0x957/0x960
 [<ffffffff96f30e9c>] __lock_acquire+0xdac/0x15f0
 [<ffffffff96f2fb19>] ? mark_held_locks+0xb9/0x140
 [<ffffffff96f31fc9>] lock_acquire+0x99/0x1e0
 [<ffffffff97008c87>] ? might_fault+0x57/0xb0
 [<ffffffff97008cb4>] might_fault+0x84/0xb0
 [<ffffffff97008c87>] ? might_fault+0x57/0xb0
 [<ffffffffc0638a74>] uio_write+0xb4/0x130 [uio]
 [<ffffffff9706ffa3>] vfs_write+0xc3/0x1f0
 [<ffffffff9709349c>] ? fget_light+0xfc/0x510
 [<ffffffff97070e2a>] SyS_write+0x8a/0x100
 [<ffffffff975ff315>] system_call_fastpath+0x1c/0x21

Signed-off-by: Xiubo Li <xiubli@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agouio: pruss: fix error handling in probe
Dan Carpenter [Thu, 2 Aug 2018 07:15:58 +0000 (10:15 +0300)]
uio: pruss: fix error handling in probe

There are two bugs here.  First the error codes weren't set on several
paths.  And second, if the call to request_threaded_irq() inside
uio_register_device() fails then it would lead to a double free when
we call uio_unregister_device() inside pruss_cleanup().

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agoDrivers: hv: vmbus: add numa_node to sysfs
Stephen Hemminger [Sat, 28 Jul 2018 21:58:48 +0000 (21:58 +0000)]
Drivers: hv: vmbus: add numa_node to sysfs

Being able to find the numa_node for a device is useful for userspace
drivers (DPDK) and also for diagnosing performance issues.  This makes
vmbus similar to pci.

Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agoDrivers: hv: vmbus: Get rid of MSR access from vmbus_drv.c
Sunil Muthuswamy [Sat, 28 Jul 2018 21:58:47 +0000 (21:58 +0000)]
Drivers: hv: vmbus: Get rid of MSR access from vmbus_drv.c

Get rid of ISA specific code from vmus_drv.c which is common code.

Fixes: 81b18bce48af ("Drivers: HV: Send one page worth of kmsg dump over Hyper-V during panic")

Signed-off-by: Sunil Muthuswamy <sunilmut@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agoDrivers: hv: vmbus: Fix the issue with freeing up hv_ctl_table_hdr
Sunil Muthuswamy [Sat, 28 Jul 2018 21:58:46 +0000 (21:58 +0000)]
Drivers: hv: vmbus: Fix the issue with freeing up hv_ctl_table_hdr

The check to free the Hyper-V control table header was reversed. This
fixes it.

Fixes: 81b18bce48af ("Drivers: HV: Send one page worth of kmsg dump over Hyper-V during panic")

Signed-off-by: Sunil Muthuswamy <sunilmut@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agoDrivers: hv: vmus: Fix the check for return value from kmsg get dump buffer
Sunil Muthuswamy [Sat, 28 Jul 2018 21:58:45 +0000 (21:58 +0000)]
Drivers: hv: vmus: Fix the check for return value from kmsg get dump buffer

The code to support panic control message was checking the return was
checking the return value from kmsg_dump_get_buffer as error value, which
is not what the routine returns. This fixes it.

Fixes: 81b18bce48af ("Drivers: HV: Send one page worth of kmsg dump over Hyper-V during panic")

Signed-off-by: Sunil Muthuswamy <sunilmut@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agoMerge tag 'fsi-updates-2018-07-27' of git://git.kernel.org/pub/scm/linux/kernel/git...
Greg Kroah-Hartman [Fri, 27 Jul 2018 08:37:08 +0000 (10:37 +0200)]
Merge tag 'fsi-updates-2018-07-27' of git://git./linux/kernel/git/benh/linux-fsi into char-misc-next

Ben writes:

Last round of FSI updates for 4.19

This adds a few fixes for things reported since the last merge,
and the latch batch of changes pending for FSI for 4.19.

That batch is a rather mechanical conversion of the misc devices
into proper char devices.

The misc devices were ill suited, the minor space for them is
limited and we can have a lot of chips in a system creating FSI
devices.

This also allows us to better control (and fix) object lifetime
getting rid of the bad devm_kzalloc() of the structures containing
the devices etc...

Finally, we add a chardev to the core FSI that provides raw CFAM
access to FSI slaves as a replacement for the current "raw" binary
sysfs file which will be ultimately deprecated and removed.

6 years agofsi: Prevent multiple concurrent rescans
Benjamin Herrenschmidt [Thu, 21 Jun 2018 08:00:05 +0000 (18:00 +1000)]
fsi: Prevent multiple concurrent rescans

The bus scanning process isn't terribly good at parallel attempts
at rescanning the same bus. Let's have a per-master mutex protecting
the scanning process.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
6 years agofsi: Add cfam char devices
Benjamin Herrenschmidt [Thu, 21 Jun 2018 02:34:22 +0000 (12:34 +1000)]
fsi: Add cfam char devices

This aims to deprecate the "raw" sysfs file used for directly
accessing the CFAM and instead use a char device like the
other sub drivers.

Since it reworks the slave creation code and adds a cfam device
type, we also use the opportunity to convert the attributes
to attribute groups and add a couple more.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
6 years agofsi: scom: Convert to use the new chardev
Benjamin Herrenschmidt [Wed, 20 Jun 2018 05:33:03 +0000 (15:33 +1000)]
fsi: scom: Convert to use the new chardev

This converts FSI scom to use the new fsi-core controlled
chardev allocator and use a real cdev instead of a miscdev.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
6 years agofsi: sbefifo: Convert to use the new chardev
Benjamin Herrenschmidt [Tue, 24 Jul 2018 04:39:55 +0000 (14:39 +1000)]
fsi: sbefifo: Convert to use the new chardev

This converts FSI sbefifo to use the new fsi-core controlled
chardev allocator and use a real cdev instead of a miscdev.

One side effect is to fix the object lifetime by removing
the use of devm_kzalloc() for something that contains kobjects,
and using proper reference counting.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
6 years agofsi: Add new central chardev support
Benjamin Herrenschmidt [Wed, 20 Jun 2018 05:22:52 +0000 (15:22 +1000)]
fsi: Add new central chardev support

The various FSI devices (sbefifo, occ, scom, more to come)
currently use misc devices.

This is problematic as the minor device space for misc is
limited and there can be a lot of them. Also it limits our
ability to move them to a dedicated /dev/fsi directory or
to be smart about device naming and numbering.

It also means we have IDAs on every single of these drivers

This creates a common fsi "device_type" for the optional
/dev/fsi grouping and a dev_t allocator for all FSI devices.

"Legacy" devices get to use a backward compatible numbering
scheme (as long as chip id <16 and there's only one copy
of a given unit type per chip).

A single major number and a single IDA are shared for all
FSI devices.

This doesn't convert the FSI device drivers to use the new
scheme yet, they will be converted individually.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
6 years agofsi: master-ast-cf: Rename dump_trace() to avoid name collision
Benjamin Herrenschmidt [Thu, 26 Jul 2018 04:49:50 +0000 (14:49 +1000)]
fsi: master-ast-cf: Rename dump_trace() to avoid name collision

s390 defines a global dump_trace() symbol. Rename ours to
dump_ucode_trace() to avoid a collision in build tests.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
6 years agofsi: master-ast-cf: Fix memory leak
Gustavo A. R. Silva [Wed, 25 Jul 2018 13:38:32 +0000 (08:38 -0500)]
fsi: master-ast-cf: Fix memory leak

In case memory resources for *fw* were allocated, release them
before return.

Addresses-Coverity-ID: 1472044 ("Resource leak")
Fixes: 6a794a27daca ("fsi: master-ast-cf: Add new FSI master using Aspeed ColdFire")
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
6 years agothunderbolt: Add support for runtime PM
Mika Westerberg [Wed, 25 Jul 2018 08:48:39 +0000 (11:48 +0300)]
thunderbolt: Add support for runtime PM

When Thunderbolt host controller is set to RTD3 mode (Runtime D3) it is
present all the time. Because of this it is important to runtime suspend
the controller whenever possible. In case of ICM we have following rules
which all needs to be true before the host controller can be put to D3:

  - The controller firmware reports to support RTD3
  - All the connected devices announce support for RTD3
  - There is no active XDomain connection

Implement this using standard Linux runtime PM APIs so that when all the
children devices are runtime suspended, the Thunderbolt host controller
PCI device is runtime suspended as well. The ICM firmware then starts
powering down power domains towards RTD3 but it can prevent this if it
detects that there is an active Display Port stream (this is not visible
to the software, though).

The Thunderbolt host controller will be runtime resumed either when
there is a remote wake event (device is connected or disconnected), or
when there is access from userspace that requires hardware access.

Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agothunderbolt: Remove redundant variable 'approved'
Colin Ian King [Wed, 25 Jul 2018 08:03:21 +0000 (11:03 +0300)]
thunderbolt: Remove redundant variable 'approved'

Variable 'approved' is being assigned but is never used hence it is
redundant and can be removed.

Cleans up clang warning:
warning: variable 'approved' set but not used [-Wunused-but-set-variable]

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agothunderbolt: Use correct ICM commands in system suspend
Mika Westerberg [Wed, 25 Jul 2018 08:03:19 +0000 (11:03 +0300)]
thunderbolt: Use correct ICM commands in system suspend

The correct way to put the ICM into suspend state is to send it
NHI_MAILBOX_DRV_UNLOADS mailbox command. NHI_MAILBOX_SAVE_DEVS is not
needed on Intel Titan Ridge so we can skip it.

Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agothunderbolt: No need to take tb->lock in domain suspend/complete
Mika Westerberg [Wed, 25 Jul 2018 08:03:18 +0000 (11:03 +0300)]
thunderbolt: No need to take tb->lock in domain suspend/complete

If the connection manager implementation needs to touch the domain
structures it ought to take the lock itself. Currently only ICM
implements these hooks and it does not need the lock because we there
will be no notifications before driver ready message is sent to it.

Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agothunderbolt: Do not unnecessarily call ICM get route
Mika Westerberg [Wed, 25 Jul 2018 08:03:17 +0000 (11:03 +0300)]
thunderbolt: Do not unnecessarily call ICM get route

This command is not really fast and can make resume time slower. We only
need to get route again if the link was changed and during initial
device connected message.

Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agothunderbolt: Use 64-bit DMA mask if supported by the platform
Mika Westerberg [Wed, 25 Jul 2018 08:03:16 +0000 (11:03 +0300)]
thunderbolt: Use 64-bit DMA mask if supported by the platform

PCI defaults to 32-bit DMA mask but this device is capable of full
64-bit addressing, so make sure we first try 64-bit DMA mask before
falling back to the default 32-bit.

Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agothunderbolt: Fix small typo in variable name
Nathan Ciobanu [Wed, 25 Jul 2018 08:03:15 +0000 (11:03 +0300)]
thunderbolt: Fix small typo in variable name

Fixes small variable name typo and the associated
checkpatch spelling warning.

Signed-off-by: Nathan Ciobanu <nathan.d.ciobanu@linux.intel.com>
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofsi: master-ast-cf: Mask unused bits in RTAG/RCRC
Benjamin Herrenschmidt [Wed, 25 Jul 2018 04:57:21 +0000 (14:57 +1000)]
fsi: master-ast-cf: Mask unused bits in RTAG/RCRC

Then reading the RTAG/RCRC "registers" from the coprocessor after
a command is complete, mask out the top bits, only keep the relevant
bits. Microcode v5 will leave garbage in those top bits as a
result of a performance optimization.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---

6 years agoandroid: binder: Include asm/cacheflush.h after linux/ include files
Guenter Roeck [Mon, 23 Jul 2018 21:41:38 +0000 (14:41 -0700)]
android: binder: Include asm/cacheflush.h after linux/ include files

If asm/cacheflush.h is included first, the following build warnings are
seen with sparc32 builds.

In file included from arch/sparc/include/asm/cacheflush.h:11:0,
        from drivers/android/binder.c:54:
arch/sparc/include/asm/cacheflush_32.h:40:37: warning:
'struct page' declared inside parameter list will not be visible
outside of this definition or declaration

Moving the asm/ include after linux/ includes solves the problem.

Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agoandroid: binder_alloc: Include asm/cacheflush.h after linux/ include files
Guenter Roeck [Mon, 23 Jul 2018 21:47:23 +0000 (14:47 -0700)]
android: binder_alloc: Include asm/cacheflush.h after linux/ include files

If asm/cacheflush.h is included first, the following build warnings are
seen with sparc32 builds.

In file included from ./arch/sparc/include/asm/cacheflush.h:11:0,
from drivers/android/binder_alloc.c:20:
./arch/sparc/include/asm/cacheflush_32.h:40:37: warning:
'struct page' declared inside parameter list

Moving the asm/ include after linux/ includes fixes the problem.

Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agomei: add mei_msg_hdr_init wrapper.
Tomas Winkler [Mon, 23 Jul 2018 10:21:24 +0000 (13:21 +0300)]
mei: add mei_msg_hdr_init wrapper.

Wrap the mei header boilerplate initialization code in
mei_msg_hdr_init function. On the way remove 'completed'
field from mei_cl_cb structure as this information
is already included in the header and is local to particular
fragment.

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agomei: move hbuf_depth from the mei device to the hw modules
Tomas Winkler [Mon, 23 Jul 2018 10:21:23 +0000 (13:21 +0300)]
mei: move hbuf_depth from the mei device to the hw modules

The host buffer depth is hardware specific so it's better to
handle it inside the me and txe hw modules. In me the depth
is read from register in txe it's a constant number.
The value is now retrieved via mei_hbuf_depth accessor,
while it replaces mei_hbuf_max_len.

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agomei: cleanup slots to data conversions
Tomas Winkler [Mon, 23 Jul 2018 10:21:22 +0000 (13:21 +0300)]
mei: cleanup slots to data conversions

Cleanup conversions between slots and data.
Define MEI_SLOT_SIZE instead of using 4 or sizeof(u32) across
the source code.

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofsi: master-ast-cf: Fix build warnings on 64-bit platforms
Benjamin Herrenschmidt [Tue, 24 Jul 2018 06:35:03 +0000 (16:35 +1000)]
fsi: master-ast-cf: Fix build warnings on 64-bit platforms

A couple of places forgot the 'z' qualifier for dev_dbg
when printing a size_t

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
6 years agoMerge tag 'fsi-updates-2018-07-24' of git://git.kernel.org/pub/scm/linux/kernel/git...
Greg Kroah-Hartman [Tue, 24 Jul 2018 06:00:13 +0000 (08:00 +0200)]
Merge tag 'fsi-updates-2018-07-24' of git://git./linux/kernel/git/benh/linux-fsi into char-misc-testing

Ben writes:

This adds support for offloading the FSI low level bitbanging to the
ColdFire coprocessor of the Aspeed SoCs. All the pre-requisites have
already been merged, this is the final piece in the puzzle.

This branch also pull gpio/ib-aspeed which is a topic branch already
in gpio/for-next (and thus in next) whic contains pre-requisites.

Finally, there's also a bug fix to the sbefifo driver for some
inconsistent use of a mutex in the error handling code.

6 years agofsi: Add support for device-tree provided chip IDs
Benjamin Herrenschmidt [Wed, 20 Jun 2018 05:16:31 +0000 (15:16 +1000)]
fsi: Add support for device-tree provided chip IDs

They get retrieved from the device-tree and exposed
as an attribute in sysfs

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
6 years agodt-bindings: fsi: Add optional chip-id to CFAMs
Benjamin Herrenschmidt [Wed, 20 Jun 2018 05:15:00 +0000 (15:15 +1000)]
dt-bindings: fsi: Add optional chip-id to CFAMs

This represents a physical chip in the system and allows
a stable numbering scheme to be passed to udev for userspace
to recognize which chip is which.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: Rob Herring <robh@kernel.org>
6 years agofsi: sbefifo: Fix inconsistent use of ffdc mutex
Benjamin Herrenschmidt [Mon, 23 Jul 2018 06:27:21 +0000 (16:27 +1000)]
fsi: sbefifo: Fix inconsistent use of ffdc mutex

Some of the exit path missed the unlock. Move the mutex to
an outer function to avoid the problem completely

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
6 years agofsi: master-ast-cf: Add new FSI master using Aspeed ColdFire
Benjamin Herrenschmidt [Mon, 11 Jun 2018 23:55:04 +0000 (09:55 +1000)]
fsi: master-ast-cf: Add new FSI master using Aspeed ColdFire

The Aspeed AST2x00 can contain a ColdFire v1 coprocessor which
is currently unused on OpenPower systems.

This adds an alternative to the fsi-master-gpio driver that
uses that coprocessor instead of bit banging from the ARM
core itself. The end result is about 4 times faster.

The firmware for the coprocessor and its source code can be
found at https://github.com/ozbenh/cf-fsi and is system specific.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
6 years agodt-bindings: fsi: Document binding for the fsi-master-ast-cf "device"
Benjamin Herrenschmidt [Wed, 13 Jun 2018 11:28:38 +0000 (21:28 +1000)]
dt-bindings: fsi: Document binding for the fsi-master-ast-cf "device"

This isn't per-se a real device, it's a pseudo-device that
represents the use of the Aspeed built-in ColdFire to
implement the FSI protocol by bitbanging the GPIOs instead
of doing it from the ARM core.

Thus it's a drop-in replacement for the existing
fsi-master-gpio pseudo-device for use on systems based
on the Aspeed chips. It has most of the same properties,
plus some more needed to operate the coprocessor.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Reviewed-by: Rob Herring <robh@kernel.org>
6 years agodevres: Add devm_of_iomap()
Benjamin Herrenschmidt [Tue, 5 Jun 2018 03:21:26 +0000 (13:21 +1000)]
devres: Add devm_of_iomap()

There are still quite a few cases where a device might want
to get to a different node of the device-tree, obtain the
resources and map them.

We have of_iomap() and of_io_request_and_map() but they both
have shortcomings, such as not returning the size of the
resource found (which can be useful) and not being "managed".

This adds a devm_of_iomap() that provides all of these and
should probably replace uses of the above in most drivers.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Joel Stanley <joel@jms.id.au>
6 years agoMerge remote-tracking branch 'gpio/ib-aspeed' into upstream-ready
Benjamin Herrenschmidt [Mon, 23 Jul 2018 05:21:39 +0000 (15:21 +1000)]
Merge remote-tracking branch 'gpio/ib-aspeed' into upstream-ready

Merge the GPIO tree "ib-aspeed" topic branch which contains pre-requisites
for subsequent changes. This branch is also in gpio "next".

6 years agouio: add SPDX license tags
Stephen Hemminger [Sat, 21 Jul 2018 13:31:39 +0000 (06:31 -0700)]
uio: add SPDX license tags

For those without any license text present or short reference
to GPL, add SPDX tag.

Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agouio: fix wrong return value from uio_mmap()
Hailong Liu [Fri, 20 Jul 2018 00:31:56 +0000 (08:31 +0800)]
uio: fix wrong return value from uio_mmap()

uio_mmap has multiple fail paths to set return value to nonzero then
goto out. However, it always returns *0* from the *out* at end, and
this will mislead callers who check the return value of this function.

Fixes: 57c5f4df0a5a0ee ("uio: fix crash after the device is unregistered")
CC: Xiubo Li <xiubli@redhat.com>
Signed-off-by: Hailong Liu <liu.hailong6@zte.com.cn>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Jiang Biao <jiang.biao2@zte.com.cn>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agouio: add SPDX license tag
Stephen Hemminger [Fri, 20 Jul 2018 22:11:26 +0000 (15:11 -0700)]
uio: add SPDX license tag

Replace short statement in comment with proper SPDX
license tag.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agoMerge tag 'extcon-next-for-4.19' of git://git.kernel.org/pub/scm/linux/kernel/git...
Greg Kroah-Hartman [Sat, 21 Jul 2018 06:13:02 +0000 (08:13 +0200)]
Merge tag 'extcon-next-for-4.19' of git://git./linux/kernel/git/chanwoo/extcon into char-misc-next

Chanwoo writes:

Update extcon for 4.19

Detailed description for this pull request:
1. Release locking when sending the connector state in extcon_sync()
-  Previously, extcon used the spinlock before calling the notifier_call_chain
to prevent the scheduled out of task and to prevent the notification delay.
When spinlock is locked for sending the notification, deadlock issue
occured on the side of extcon consumer device. To fix this issue on extcon
consumer device, release locking when sending the connector state.

2. Fix minor issues of extcon provider driver
- extcon-intel-int3496.c uses 'linux/gpio/consumer.h' instead of 'linux/gpio.h'
- extcon-usbc-cors-ec.c adds SPDX license and fix the wrong license information

6 years agomisc: vexpress/syscfg: Use devm_ioremap_resource() to map memory
Sudeep Holla [Wed, 11 Jul 2018 15:41:10 +0000 (16:41 +0100)]
misc: vexpress/syscfg: Use devm_ioremap_resource() to map memory

Instead of checking the return value of platform_get_resource(), we can
use devm_ioremap_resource() which has the NULL pointer check and the
memory region requesting. devm_ioremap_resource is designed to replace
calls to devm_request_mem_region followed by devm_ioremap, so let's use
the same.

Cc: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Signed-off-by: Sudeep Holla <sudeep.holla@arm.com>
Acked-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agodrivers/misc: Aspeed LPC snoop output using misc chardev
Robert Lippert [Fri, 6 Jul 2018 18:25:32 +0000 (11:25 -0700)]
drivers/misc: Aspeed LPC snoop output using misc chardev

Provides the data bytes snooped over the LPC snoop bus to userspace
as a (blocking) misc character device.

Bytes output from the host using LPC I/O transactions to the snooped port
can be watched or retrieved from the character device using a simple
command like this:
~#  od -w1 -A n -t x1 /dev/aspeed-lpc-snoop0
 10
 de
 ad
 c0
 ff
 ee

Signed-off-by: Robert Lippert <rlippert@google.com>
Signed-off-by: Patrick Venture <venture@google.com>
Signed-off-by: Benjamin Fair <benjaminfair@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agomei: bus: suppress sign-compare warnings
Tomas Winkler [Sun, 15 Jul 2018 11:41:18 +0000 (14:41 +0300)]
mei: bus: suppress sign-compare warnings

Comparison between signed and unsigned warnings
and associated type promotion may cause error
condition not be detected.
The type promotion issue in mei bus was addressed by two patches:
commit b40b3e9358fb ("mei: bus: type promotion bug in mei_nfc_if_version()")
commit cf1ed2c59b98 ("mei: bus: type promotion bug in mei_fwver()")
Now it is possible to suppress the warning, by adding proper
casting to move out of radar.

Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agognss: fix potential error pointer dereference
Dan Carpenter [Mon, 16 Jul 2018 10:42:03 +0000 (12:42 +0200)]
gnss: fix potential error pointer dereference

The gnss_allocate_device() function returns a mix of NULL and error
pointers on error.  It should only return one or the other.  Since the
callers both check for NULL, I've modified it to return NULL on error.

Fixes: 2b6a44035143 ("gnss: add GNSS receiver subsystem")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agognss: ubx: make struct ubx_gserial_ops static
Colin Ian King [Mon, 16 Jul 2018 10:42:02 +0000 (12:42 +0200)]
gnss: ubx: make struct ubx_gserial_ops static

The structure ubx_gserial_ops is local to the source and does not need
to be in global scope, so make it static.

Cleans up sparse warning:
symbol 'ubx_gserial_ops' was not declared. Should it be static?

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agovirtio: virtconsole: Use seq_file for debugfs operations
Tuomas Tynkkynen [Thu, 12 Jul 2018 21:39:56 +0000 (00:39 +0300)]
virtio: virtconsole: Use seq_file for debugfs operations

Simplifies the code and is more conventional to what's used in the rest
of the kernel for debugfs ops.

Signed-off-by: Tuomas Tynkkynen <tuomas@tuxera.com>
Reviewed-by: Amit Shah <amit@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agovme: ca91cx42: remove redundant variable i
Colin Ian King [Sat, 14 Jul 2018 16:33:32 +0000 (17:33 +0100)]
vme: ca91cx42: remove redundant variable i

Variable i is being assigned but is never used hence it is redundant
and can be removed.

Cleans up clang warning:
warning: variable 'i' set but not used [-Wunused-but-set-variable]

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Martyn Welch <martyn@welchs.me.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agoMerge 4.18-rc5 into char-misc-next
Greg Kroah-Hartman [Mon, 16 Jul 2018 07:04:54 +0000 (09:04 +0200)]
Merge 4.18-rc5 into char-misc-next

We want the char-misc fixes in here as well.

Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agoLinux 4.18-rc5
Linus Torvalds [Sun, 15 Jul 2018 19:49:31 +0000 (12:49 -0700)]
Linux 4.18-rc5

6 years agoMerge tag 'armsoc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc
Linus Torvalds [Sun, 15 Jul 2018 16:49:21 +0000 (09:49 -0700)]
Merge tag 'armsoc-fixes' of git://git./linux/kernel/git/arm/arm-soc

Pull ARM SoC fixes from Olof Johansson:

 - A fix for OMAP5 and DRA7 to make the branch predictor hardening
   settings take proper effect on secondary cores

 - Disable USB OTG on am3517 since current driver isn't working

 - Fix thermal sensor register settings on Armada 38x

 - Fix suspend/resume IRQs on pxa3xx

* tag 'armsoc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc:
  ARM: dts: am3517.dtsi:  Disable reference to OMAP3 OTG controller
  ARM: DRA7/OMAP5: Enable ACTLR[0] (Enable invalidates of BTB) for secondary cores
  ARM: pxa: irq: fix handling of ICMR registers in suspend/resume
  ARM: dts: armada-38x: use the new thermal binding

6 years agoparport: ieee1284: mark expected switch fall-throughs
Gustavo A. R. Silva [Thu, 12 Jul 2018 21:29:56 +0000 (22:29 +0100)]
parport: ieee1284: mark expected switch fall-throughs

In preparation to enabling -Wimplicit-fallthrough, mark switch cases
where we are expecting to fall through.

Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: Sudip Mukherjee <sudipm.mukherjee@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agoparport: sunbpp: fix error return code
Julia Lawall [Thu, 12 Jul 2018 21:29:55 +0000 (22:29 +0100)]
parport: sunbpp: fix error return code

Return an error code on failure.  Change leading spaces to tab on the
first if.

Problem found using Coccinelle.

Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Sudip Mukherjee <sudipm.mukherjee@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agoeventpoll.h: wrap casts in () properly
Greg Kroah-Hartman [Tue, 10 Jul 2018 15:15:38 +0000 (17:15 +0200)]
eventpoll.h: wrap casts in () properly

When importing the latest copy of the kernel headers into Bionic,
Christpher and Elliott noticed that the eventpoll.h casts were not
wrapped in ().  As it is, clang complains about macros without
surrounding (), so this makes it a pain for userspace tools.

So fix it up by adding another () pair, and make them line up purty by
using tabs.

Fixes: 65aaf87b3aa2 ("add EPOLLNVAL, annotate EPOLL... and event_poll->event")
Reported-by: Christopher Ferris <cferris@google.com>
Reported-by: Elliott Hughes <enh@google.com>
Cc: stable <stable@vger.kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agonvmem: imx-ocotp: add support for imx6sll
Anson Huang [Wed, 11 Jul 2018 10:20:43 +0000 (11:20 +0100)]
nvmem: imx-ocotp: add support for imx6sll

i.MX6SLL is a new SoC of i.MX6 family, enable ocotp
driver support for this SoC.

Signed-off-by: Anson Huang <Anson.Huang@nxp.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agodt-bindings: nvmem: imx-ocotp: add compatible string for imx6sll
Anson Huang [Wed, 11 Jul 2018 10:20:42 +0000 (11:20 +0100)]
dt-bindings: nvmem: imx-ocotp: add compatible string for imx6sll

Add new compatible string for i.MX6SLL SOC.

Signed-off-by: Anson Huang <Anson.Huang@nxp.com>
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agonvmem: Add Spreadtrum SC27XX efuse support
Freeman Liu [Wed, 11 Jul 2018 10:20:41 +0000 (11:20 +0100)]
nvmem: Add Spreadtrum SC27XX efuse support

This patch add the efuse driver which is embeded in Spreadtrum SC27XX
series PMICs. The sc27xx efuse contains 32 blocks and each block's
data width is 16 bits.

Signed-off-by: Freeman Liu <freeman.liu@spreadtrum.com>
Signed-off-by: Baolin Wang <baolin.wang@linaro.org>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agodt-bindings: nvmem: Add Spreadtrum SC27XX efuse controller documentation
Baolin Wang [Wed, 11 Jul 2018 10:20:40 +0000 (11:20 +0100)]
dt-bindings: nvmem: Add Spreadtrum SC27XX efuse controller documentation

This patch adds the binding documentation for Spreadtrum SC27XX series
PMICs efuse controller device.

Signed-off-by: Baolin Wang <baolin.wang@linaro.org>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agoMAINTAINERS: add entry for FPGA DFL drivers
Wu Hao [Sat, 30 Jun 2018 00:53:36 +0000 (08:53 +0800)]
MAINTAINERS: add entry for FPGA DFL drivers

Add entry for FPGA Device Feature List (DFL) drivers.

Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Acked-by: Moritz Fischer <mdf@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: dfl: afu: add DFL_FPGA_PORT_DMA_MAP/UNMAP ioctls support
Wu Hao [Sat, 30 Jun 2018 00:53:35 +0000 (08:53 +0800)]
fpga: dfl: afu: add DFL_FPGA_PORT_DMA_MAP/UNMAP ioctls support

DMA memory regions are required for Accelerated Function Unit (AFU) usage.
These two ioctls allow user space applications to map user memory regions
for dma, and unmap them after use. Iova is returned from driver to user
space application via DFL_FPGA_PORT_DMA_MAP ioctl. Application needs to
unmap it after use, otherwise, driver will unmap them in device file
release operation.

Each AFU has its own rb tree to keep track of its mapped DMA regions.

Ioctl interfaces:
* DFL_FPGA_PORT_DMA_MAP
  Do the dma mapping per user_addr and length provided by user.
  Return iova in provided struct dfl_fpga_port_dma_map.

* DFL_FPGA_PORT_DMA_UNMAP
  Unmap the dma region per iova provided by user.

Signed-off-by: Tim Whisonant <tim.whisonant@intel.com>
Signed-off-by: Enno Luebbers <enno.luebbers@intel.com>
Signed-off-by: Shiva Rao <shiva.rao@intel.com>
Signed-off-by: Christopher Rauer <christopher.rauer@intel.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: dfl: afu: add afu sub feature support
Xiao Guangrong [Sat, 30 Jun 2018 00:53:34 +0000 (08:53 +0800)]
fpga: dfl: afu: add afu sub feature support

User Accelerated Function Unit sub feature exposes the MMIO region of
the AFU. After valid PR bitstream is programmed and the port is enabled,
then this MMIO region could be accessed.

This patch adds support to enumerate the AFU MMIO region and expose it
to userspace via mmap file operation. Below interfaces are exposed to user:

Sysfs interface:
* /sys/class/fpga_region/<regionX>/<dfl-port.x>/afu_id
  Read-only. Indicate which PR bitstream is programmed to this AFU.

Ioctl interfaces:
* DFL_FPGA_PORT_GET_INFO
  Provide info to userspace on the number of supported region.
  Only UAFU region is supported now.

* DFL_FPGA_PORT_GET_REGION_INFO
  Provide region information, including access permission, region size,
  offset from the start of device fd.

Signed-off-by: Tim Whisonant <tim.whisonant@intel.com>
Signed-off-by: Enno Luebbers <enno.luebbers@intel.com>
Signed-off-by: Shiva Rao <shiva.rao@intel.com>
Signed-off-by: Christopher Rauer <christopher.rauer@intel.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: dfl: afu: add DFL_FPGA_GET_API_VERSION/CHECK_EXTENSION ioctls support
Wu Hao [Sat, 30 Jun 2018 00:53:33 +0000 (08:53 +0800)]
fpga: dfl: afu: add DFL_FPGA_GET_API_VERSION/CHECK_EXTENSION ioctls support

DFL_FPGA_GET_API_VERSION and DFL_FPGA_CHECK_EXTENSION ioctls are common
ones which need to be supported by all feature devices drivers including
FME and AFU. This patch implements above 2 ioctls in FPGA Accelerated
Function Unit (AFU) driver.

Signed-off-by: Tim Whisonant <tim.whisonant@intel.com>
Signed-off-by: Enno Luebbers <enno.luebbers@intel.com>
Signed-off-by: Shiva Rao <shiva.rao@intel.com>
Signed-off-by: Christopher Rauer <christopher.rauer@intel.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Acked-by: Moritz Fischer <mdf@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: dfl: afu: add header sub feature support
Wu Hao [Sat, 30 Jun 2018 00:53:32 +0000 (08:53 +0800)]
fpga: dfl: afu: add header sub feature support

The port header register set is always present for port, it is mainly
for capability, control and status of the ports that AFU connected to.

This patch implements header sub feature support. Below user interfaces
are created by this patch.

Sysfs interface:
* /sys/class/fpga_region/<regionX>/<dfl-port.x>/id
  Read-only. Port ID.

Ioctl interface:
* DFL_FPGA_PORT_RESET
  Reset the FPGA Port and its AFU.

Signed-off-by: Tim Whisonant <tim.whisonant@intel.com>
Signed-off-by: Enno Luebbers <enno.luebbers@intel.com>
Signed-off-by: Shiva Rao <shiva.rao@intel.com>
Signed-off-by: Christopher Rauer <christopher.rauer@intel.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: dfl: afu: add port ops support
Wu Hao [Sat, 30 Jun 2018 00:53:31 +0000 (08:53 +0800)]
fpga: dfl: afu: add port ops support

This patch registers the port ops into the global list in the DFL
framework, and it allows other modules to use the port ops. And
This patch includes the implementation of the get_id and enable_set
ops too.

Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: dfl: add FPGA Accelerated Function Unit driver basic framework
Wu Hao [Sat, 30 Jun 2018 00:53:30 +0000 (08:53 +0800)]
fpga: dfl: add FPGA Accelerated Function Unit driver basic framework

On DFL FPGA devices, the Accelerated Function Unit (AFU), can be
reprogrammed for different functions. It connects to the FPGA
infrastructure (static FPGA region) via a Port. Port CSRs are
implemented separately from the AFU CSRs to provide control and
status of the Port. Once valid PR bitstream is programmed into
the AFU, it allows access to the AFU CSRs in the AFU MMIO space.

This patch only implements basic driver framework for AFU, including
device file operation framework.

Signed-off-by: Tim Whisonant <tim.whisonant@intel.com>
Signed-off-by: Enno Luebbers <enno.luebbers@intel.com>
Signed-off-by: Shiva Rao <shiva.rao@intel.com>
Signed-off-by: Christopher Rauer <christopher.rauer@intel.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: dfl: fme-region: add support for compat_id
Wu Hao [Sat, 30 Jun 2018 00:53:29 +0000 (08:53 +0800)]
fpga: dfl: fme-region: add support for compat_id

This patch adds compat_id support, it reuses fme manager's
compat id, as the per region compat id is actually from the
fme manager's register.

Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Acked-by: Moritz Fischer <mdf@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: dfl: add fpga region platform driver for FME
Wu Hao [Sat, 30 Jun 2018 00:53:28 +0000 (08:53 +0800)]
fpga: dfl: add fpga region platform driver for FME

This patch adds fpga region platform driver for FPGA Management Engine.
It register an fpga region with given fpga manager / bridge device.

Signed-off-by: Tim Whisonant <tim.whisonant@intel.com>
Signed-off-by: Enno Luebbers <enno.luebbers@intel.com>
Signed-off-by: Shiva Rao <shiva.rao@intel.com>
Signed-off-by: Christopher Rauer <christopher.rauer@intel.com>
Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Acked-by: Moritz Fischer <mdf@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: dfl: add fpga bridge platform driver for FME
Wu Hao [Sat, 30 Jun 2018 00:53:27 +0000 (08:53 +0800)]
fpga: dfl: add fpga bridge platform driver for FME

This patch adds fpga bridge platform driver for FPGA Management Engine.
It implements the enable_set callback for fpga bridge.

Signed-off-by: Tim Whisonant <tim.whisonant@intel.com>
Signed-off-by: Enno Luebbers <enno.luebbers@intel.com>
Signed-off-by: Shiva Rao <shiva.rao@intel.com>
Signed-off-by: Christopher Rauer <christopher.rauer@intel.com>
Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Acked-by: Moritz Fischer <mdf@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: dfl: fme-mgr: add compat_id support
Wu Hao [Sat, 30 Jun 2018 00:53:26 +0000 (08:53 +0800)]
fpga: dfl: fme-mgr: add compat_id support

This patch adds compat_id support to fme manager driver, it
reads the ID from the hardware register. And it could be used
for compatibility check before partial reconfiguration.

Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Acked-by: Moritz Fischer <mdf@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: dfl: add fpga manager platform driver for FME
Wu Hao [Sat, 30 Jun 2018 00:53:25 +0000 (08:53 +0800)]
fpga: dfl: add fpga manager platform driver for FME

This patch adds fpga manager driver for FPGA Management Engine (FME). It
implements fpga_manager_ops for FPGA Partial Reconfiguration function.

Signed-off-by: Tim Whisonant <tim.whisonant@intel.com>
Signed-off-by: Enno Luebbers <enno.luebbers@intel.com>
Signed-off-by: Shiva Rao <shiva.rao@intel.com>
Signed-off-by: Christopher Rauer <christopher.rauer@intel.com>
Signed-off-by: Kang Luwei <luwei.kang@intel.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: dfl: fme: add partial reconfiguration sub feature support
Kang Luwei [Sat, 30 Jun 2018 00:53:24 +0000 (08:53 +0800)]
fpga: dfl: fme: add partial reconfiguration sub feature support

Partial Reconfiguration (PR) is the most important function for FME. It
allows reconfiguration for given Port/Accelerated Function Unit (AFU).

It creates platform devices for fpga-mgr, fpga-regions and fpga-bridges,
and invokes fpga-region's interface (fpga_region_program_fpga) for PR
operation once PR request received via ioctl. Below user space interface
is exposed by this sub feature.

Ioctl interface:
* DFL_FPGA_FME_PORT_PR
  Do partial reconfiguration per information from userspace, including
  target port(AFU), buffer size and address info. It returns error code
  to userspace if failed. For detailed PR error information, user needs
  to read fpga-mgr's status sysfs interface.

Signed-off-by: Tim Whisonant <tim.whisonant@intel.com>
Signed-off-by: Enno Luebbers <enno.luebbers@intel.com>
Signed-off-by: Shiva Rao <shiva.rao@intel.com>
Signed-off-by: Christopher Rauer <christopher.rauer@intel.com>
Signed-off-by: Kang Luwei <luwei.kang@intel.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: dfl: fme: add DFL_FPGA_GET_API_VERSION/CHECK_EXTENSION ioctls support
Wu Hao [Sat, 30 Jun 2018 00:53:23 +0000 (08:53 +0800)]
fpga: dfl: fme: add DFL_FPGA_GET_API_VERSION/CHECK_EXTENSION ioctls support

DFL_FPGA_GET_API_VERSION and DFL_FPGA_CHECK_EXTENSION ioctls are common
ones which need to be supported by all feature devices drivers including
FME and AFU. Userspace application can use these ioctl interfaces to get
the API info and check if specific extension is supported or not in
current driver.

This patch implements above 2 ioctls in FPGA Management Engine (FME)
driver.

Signed-off-by: Tim Whisonant <tim.whisonant@intel.com>
Signed-off-by: Enno Luebbers <enno.luebbers@intel.com>
Signed-off-by: Shiva Rao <shiva.rao@intel.com>
Signed-off-by: Christopher Rauer <christopher.rauer@intel.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Acked-by: Moritz Fischer <mdf@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: dfl: fme: add header sub feature support
Kang Luwei [Sat, 30 Jun 2018 00:53:22 +0000 (08:53 +0800)]
fpga: dfl: fme: add header sub feature support

The Header Register set is always present for FPGA Management Engine (FME),
this patch implements init and uinit function for header sub feature and
introduces several read-only sysfs interfaces for the capability and
status.

Sysfs interfaces:
* /sys/class/fpga_region/<regionX>/<dfl-fme.x>/ports_num
  Read-only. Number of ports implemented

* /sys/class/fpga_region/<regionX>/<dfl-fme.x>/bitstream_id
  Read-only. Bitstream (static FPGA region) identifier number. It contains
  the detailed version and other information of this static FPGA region.

* /sys/class/fpga_region/<regionX>/<dfl-fme.x>/bitstream_metadata
  Read-only. Bitstream (static FPGA region) meta data. It contains the
  synthesis date, seed and other information of this static FPGA region.

Signed-off-by: Tim Whisonant <tim.whisonant@intel.com>
Signed-off-by: Enno Luebbers <enno.luebbers@intel.com>
Signed-off-by: Shiva Rao <shiva.rao@intel.com>
Signed-off-by: Christopher Rauer <christopher.rauer@intel.com>
Signed-off-by: Kang Luwei <luwei.kang@intel.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: dfl: add FPGA Management Engine driver basic framework
Kang Luwei [Sat, 30 Jun 2018 00:53:21 +0000 (08:53 +0800)]
fpga: dfl: add FPGA Management Engine driver basic framework

The FPGA Management Engine (FME) provides power, thermal management,
performance counters, partial reconfiguration and other functions. For each
function, it is packaged into a private feature linked to the FME feature
device in the 'Device Feature List'. It's a platform device created by
DFL framework.

This patch adds the basic framework of FME platform driver. It defines
sub feature drivers to handle the different sub features, including init,
uinit and ioctl. It also registers the file operations for the device file.

Signed-off-by: Tim Whisonant <tim.whisonant@intel.com>
Signed-off-by: Enno Luebbers <enno.luebbers@intel.com>
Signed-off-by: Shiva Rao <shiva.rao@intel.com>
Signed-off-by: Christopher Rauer <christopher.rauer@intel.com>
Signed-off-by: Kang Luwei <luwei.kang@intel.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Acked-by: Moritz Fischer <mdf@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: dfl-pci: add enumeration for feature devices
Wu Hao [Sat, 30 Jun 2018 00:53:20 +0000 (08:53 +0800)]
fpga: dfl-pci: add enumeration for feature devices

The Device Feature List (DFL) is implemented in MMIO and features
are linked via the DFLs. This patch enables pcie driver to prepare
enumeration information (e.g. locations of all device feature lists
in MMIO) and use common APIs provided by the Device Feature List
framework to enumerate each feature device linked.

Signed-off-by: Tim Whisonant <tim.whisonant@intel.com>
Signed-off-by: Enno Luebbers <enno.luebbers@intel.com>
Signed-off-by: Shiva Rao <shiva.rao@intel.com>
Signed-off-by: Christopher Rauer <christopher.rauer@intel.com>
Signed-off-by: Zhang Yi <yi.z.zhang@intel.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: add FPGA DFL PCIe device driver
Zhang Yi [Sat, 30 Jun 2018 00:53:19 +0000 (08:53 +0800)]
fpga: add FPGA DFL PCIe device driver

This patch implements the basic framework of the driver for FPGA PCIe
device which implements the Device Feature List (DFL) in its MMIO space.
This driver is verified on Intel(R) PCIe-based FPGA DFL devices, including
both integrated (e.g. Intel Server Platform with In-package FPGA) and
discrete (e.g. Intel FPGA PCIe Acceleration Cards) solutions.

Signed-off-by: Tim Whisonant <tim.whisonant@intel.com>
Signed-off-by: Enno Luebbers <enno.luebbers@intel.com>
Signed-off-by: Shiva Rao <shiva.rao@intel.com>
Signed-off-by: Christopher Rauer <christopher.rauer@intel.com>
Signed-off-by: Zhang Yi <yi.z.zhang@intel.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Acked-by: Moritz Fischer <mdf@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: dfl: add dfl_fpga_check_port_id function.
Wu Hao [Sat, 30 Jun 2018 00:53:18 +0000 (08:53 +0800)]
fpga: dfl: add dfl_fpga_check_port_id function.

This patch adds one common function in DFL framework. It uses
port_ops get_id callback to get port id and compare it with given
value. This function could be used as match function of the
dfl_fpga_cdev_find_port function.

Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: dfl: add dfl_fpga_port_ops support.
Wu Hao [Sat, 30 Jun 2018 00:53:17 +0000 (08:53 +0800)]
fpga: dfl: add dfl_fpga_port_ops support.

In some cases, other DFL driver modules may need to access some port
operations, e.g. disable / enable port for partial reconfiguration in
FME module. In order to avoid dependency between port and FME modules,
this patch introduces the dfl_fpga_port_ops support in DFL framework.
A global dfl_fpga_port_ops list is added in the DFL framework, and
it allows other DFL modules to use these port operations registered
to this list, even in virtualization case, the port platform device
is turned into VF / guest VM and hidden in host, the registered
port_ops is still usable. It resolves the dependency issues between
modules, but once get port ops API returns a valid port ops, that
means related port driver module has been module_get to prevent from
unexpected unload, and put port ops API must be invoked after use.

These APIs introduced by this patch is listed below:
 * dfl_fpga_port_ops_add
   add one port ops to the global list.

 * dfl_fpga_port_ops_del
   del one port ops from the global list.

 * dfl_fpga_port_ops_get / dfl_fpga_port_ops_put
   get/put the port ops before/after use.

Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: dfl: add feature device infrastructure
Xiao Guangrong [Sat, 30 Jun 2018 00:53:16 +0000 (08:53 +0800)]
fpga: dfl: add feature device infrastructure

This patch abstracts the common operations of the sub features and defines
the feature_ops data structure, including init, uinit and ioctl function
pointers. And this patch adds some common helper functions for FME and AFU
drivers, e.g. dfl_feature_dev_use_begin/end which are used to ensure
exclusive usage of the feature device file.

Signed-off-by: Tim Whisonant <tim.whisonant@intel.com>
Signed-off-by: Enno Luebbers <enno.luebbers@intel.com>
Signed-off-by: Shiva Rao <shiva.rao@intel.com>
Signed-off-by: Christopher Rauer <christopher.rauer@intel.com>
Signed-off-by: Kang Luwei <luwei.kang@intel.com>
Signed-off-by: Zhang Yi <yi.z.zhang@intel.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: dfl: add dfl_fpga_cdev_find_port
Wu Hao [Sat, 30 Jun 2018 00:53:15 +0000 (08:53 +0800)]
fpga: dfl: add dfl_fpga_cdev_find_port

For feature devices, we need a method to find the port dedicated
to the device. This patch adds a function dfl_fpga_cdev_find_port
for this purpose. e.g. FPGA Management Engine (FME) Partial
Reconfiguration sub feature, it uses this function to find
dedicated port on the device for PR function implementation.

Signed-off-by: Tim Whisonant <tim.whisonant@intel.com>
Signed-off-by: Enno Luebbers <enno.luebbers@intel.com>
Signed-off-by: Shiva Rao <shiva.rao@intel.com>
Signed-off-by: Christopher Rauer <christopher.rauer@intel.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Acked-by: Moritz Fischer <mdf@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: dfl: add chardev support for feature devices
Wu Hao [Sat, 30 Jun 2018 00:53:14 +0000 (08:53 +0800)]
fpga: dfl: add chardev support for feature devices

For feature devices drivers, both the FPGA Management Engine (FME) and
Accelerated Function Unit (AFU) driver need to expose user interfaces via
the device file, for example, mmap and ioctls.

This patch adds chardev support in the dfl driver for feature devices,
FME and AFU. It reserves the chardev regions for FME and AFU and provide
interfaces for FME and AFU driver to register their device file operations.

Signed-off-by: Tim Whisonant <tim.whisonant@intel.com>
Signed-off-by: Enno Luebbers <enno.luebbers@intel.com>
Signed-off-by: Shiva Rao <shiva.rao@intel.com>
Signed-off-by: Christopher Rauer <christopher.rauer@intel.com>
Signed-off-by: Zhang Yi <yi.z.zhang@intel.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: add device feature list support
Wu Hao [Sat, 30 Jun 2018 00:53:13 +0000 (08:53 +0800)]
fpga: add device feature list support

Device Feature List (DFL) defines a feature list structure that creates
a linked list of feature headers within the MMIO space to provide an
extensible way of adding features. This patch introduces a kernel module
to provide basic infrastructure to support FPGA devices which implement
the Device Feature List.

Usually there will be different features and their sub features linked into
the DFL. This code provides common APIs for feature enumeration, it creates
a container device (FPGA base region), walks through the DFLs and creates
platform devices for feature devices (Currently it only supports two
different feature devices, FPGA Management Engine (FME) and Port which
the Accelerator Function Unit (AFU) connected to). In order to enumerate
the DFLs, the common APIs required low level driver to provide necessary
enumeration information (e.g. address for each device feature list for
given device) and fill it to the dfl_fpga_enum_info data structure. Please
refer to below description for APIs added for enumeration.

Functions for enumeration information preparation:
 *dfl_fpga_enum_info_alloc
   allocate enumeration information data structure.

 *dfl_fpga_enum_info_add_dfl
   add a device feature list to dfl_fpga_enum_info data structure.

 *dfl_fpga_enum_info_free
   free dfl_fpga_enum_info data structure and related resources.

Functions for feature device enumeration:
 *dfl_fpga_feature_devs_enumerate
   enumerate feature devices and return container device.

 *dfl_fpga_feature_devs_remove
   remove feature devices under given container device.

Signed-off-by: Tim Whisonant <tim.whisonant@intel.com>
Signed-off-by: Enno Luebbers <enno.luebbers@intel.com>
Signed-off-by: Shiva Rao <shiva.rao@intel.com>
Signed-off-by: Christopher Rauer <christopher.rauer@intel.com>
Signed-off-by: Zhang Yi <yi.z.zhang@intel.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: region: add compat_id support
Wu Hao [Sat, 30 Jun 2018 00:53:12 +0000 (08:53 +0800)]
fpga: region: add compat_id support

This patch introduces a compat_id pointer member and sysfs interface
for each fpga region, similar as compat_id for fpga manager, it allows
applications to read the per region compat_id for compatibility
checking before other actions on this fpga-region (e.g. PR).

Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Acked-by: Moritz Fischer <mdf@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: mgr: add compat_id support
Wu Hao [Sat, 30 Jun 2018 00:53:11 +0000 (08:53 +0800)]
fpga: mgr: add compat_id support

This patch introduces compat_id support to fpga manager, it adds
a fpga_compat_id pointer to fpga manager data structure to allow
fpga manager drivers to save the compatibility id. This compat_id
could be used for compatibility checking before doing partial
reconfiguration to associated fpga regions.

Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: mgr: add status for fpga-manager
Wu Hao [Sat, 30 Jun 2018 00:53:10 +0000 (08:53 +0800)]
fpga: mgr: add status for fpga-manager

This patch adds status sysfs interface for fpga manager, it's a
read only interface which allows user to get fpga manager status,
including full/partial reconfiguration error and other status
information. It adds a status callback to fpga_manager_ops too,
allows each fpga_manager driver to define its own method to
collect latest status from hardware.

The following sysfs file is created:
* /sys/class/fpga_manager/<fpga>/status
  Return status of fpga manager, including reconfiguration errors.

Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Acked-by: Moritz Fischer <mdf@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agofpga: mgr: add region_id to fpga_image_info
Wu Hao [Sat, 30 Jun 2018 00:53:09 +0000 (08:53 +0800)]
fpga: mgr: add region_id to fpga_image_info

This patch adds region_id to fpga_image_info data structure, it
allows driver to pass region id information to fpga-mgr via
fpga_image_info for fpga reconfiguration function.

Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Moritz Fischer <mdf@kernel.org>
Acked-by: Alan Tull <atull@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agodocs: fpga: add a document for FPGA Device Feature List (DFL) Framework Overview
Wu Hao [Sat, 30 Jun 2018 00:53:08 +0000 (08:53 +0800)]
docs: fpga: add a document for FPGA Device Feature List (DFL) Framework Overview

Add a document for FPGA Device Feature List (DFL) Framework Overview.

Signed-off-by: Enno Luebbers <enno.luebbers@intel.com>
Signed-off-by: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Signed-off-by: Wu Hao <hao.wu@intel.com>
Acked-by: Alan Tull <atull@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agocoresight: tpiu: Fix disabling timeouts
Robin Murphy [Wed, 11 Jul 2018 19:40:35 +0000 (13:40 -0600)]
coresight: tpiu: Fix disabling timeouts

Probing the TPIU driver under UBSan triggers an out-of-bounds shift
warning in coresight_timeout():

...
[    5.677530] UBSAN: Undefined behaviour in drivers/hwtracing/coresight/coresight.c:929:16
[    5.685542] shift exponent 64 is too large for 64-bit type 'long unsigned int'
...

On closer inspection things are exponentially out of whack because we're
passing a bitmask where a bit number should be. Amusingly, it seems that
both calls will find their expected values by sheer luck and appear to
succeed: 1 << FFCR_FON_MAN ends up at bit 64 which whilst undefined
evaluates as zero in practice, while 1 << FFSR_FT_STOPPED finds bit 2
(TCPresent) which apparently is usually tied high.

Following the examples of other drivers, define separate FOO and FOO_BIT
macros for masks vs. indices, and put things right.

CC: Robert Walker <robert.walker@arm.com>
CC: Mike Leach <mike.leach@linaro.org>
CC: Mathieu Poirier <mathieu.poirier@linaro.org>
Fixes: 11595db8e17f ("coresight: Fix disabling of CoreSight TPIU")
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agocoresight: catu: Plug in CATU as a backend for ETR buffer
Suzuki K Poulose [Wed, 11 Jul 2018 19:40:34 +0000 (13:40 -0600)]
coresight: catu: Plug in CATU as a backend for ETR buffer

Now that we can use a CATU with a scatter gather table, add support
for the TMC ETR to make use of the connected CATU in translate mode.
This is done by adding CATU as new buffer mode.

Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agocoresight: catu: Add support for scatter gather tables
Suzuki K Poulose [Wed, 11 Jul 2018 19:40:33 +0000 (13:40 -0600)]
coresight: catu: Add support for scatter gather tables

This patch adds the support for setting up a SG table for use
by the CATU. We reuse the tmc_sg_table to represent the table/data
pages, even though the table format is different.

Similar to ETR SG table, CATU uses a 4KB page size for data buffers
as well as page tables. All table entries are 64bit wide and have
the following format:

        63                      12      1  0
        x-----------------------------------x
        |        Address [63-12] | SBZ  | V |
        x-----------------------------------x

Where [V] ->  0 - Pointer is invalid
 1 - Pointer is Valid

CATU uses only first half of the page for data page pointers.
i.e, single table page will only have 256 page pointers, addressing
upto 1MB of data. The second half of a table page contains only two
pointers at the end of the page (i.e, pointers at index 510 and 511),
which are used as links to the "Previous" and "Next" page tables
respectively.

The first table page has an "Invalid" previous pointer and the
next pointer entry points to the second page table if there is one.
Similarly the last table page has an "Invalid" next pointer to
indicate the end of the table chain.

Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agodts: bindings: Document device tree binding for CATU
Suzuki K Poulose [Wed, 11 Jul 2018 19:40:32 +0000 (13:40 -0600)]
dts: bindings: Document device tree binding for CATU

Document CATU device-tree bindings. CATU augments the TMC-ETR
by providing an improved Scatter Gather mechanism for streaming
trace data to non-contiguous system RAM pages.

Cc: devicetree@vger.kernel.org
Cc: frowand.list@gmail.com
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agocoresight: Introduce support for Coresight Address Translation Unit
Suzuki K Poulose [Wed, 11 Jul 2018 19:40:31 +0000 (13:40 -0600)]
coresight: Introduce support for Coresight Address Translation Unit

Add the initial support for Coresight Address Translation Unit, which
augments the TMC in Coresight SoC-600 by providing an improved Scatter
Gather mechanism. CATU is always connected to a single TMC-ETR and
converts the AXI address with a translated address (from a given SG
table with specific format). The CATU should be programmed in pass
through mode and enabled even if the ETR doesn't use the translation
by CATU.

This patch provides mechanism to enable/disable the CATU always in the
pass through mode.

We reuse the existing ports mechanism to link the TMC-ETR to the
connected CATU.

i.e, TMC-ETR:output_port0 -> CATU:input_port0

Reference manual for CATU component is avilable in version r2p0 of :
"Arm Coresight System-on-Chip SoC-600 Technical Reference Manual".

Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agocoresight: Add helper device type
Suzuki K Poulose [Wed, 11 Jul 2018 19:40:30 +0000 (13:40 -0600)]
coresight: Add helper device type

Add a new coresight device type, which do not belong to any
of the existing types, i.e, source, sink, link etc. A helper
device could be connected to a coresight device, which could
augment the functionality of the coresight device.

This is intended to cover Coresight Address Translation Unit (CATU)
devices, which provide improved Scatter Gather mechanism for TMC
ETR. The idea is that the helper device could be controlled by
the driver of the device it is attached to (in this case ETR),
transparent to the generic coresight driver (and paths).

The operations include enable(), disable(), both of which could
accept a device specific "data" which the driving device and
the helper device could share. Since they don't appear in the
coresight "path" tracked by software, we have to ensure that
they are powered up/down whenever the master device is turned
on.

Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agocoresight: Cleanup device subtype struct
Suzuki K Poulose [Wed, 11 Jul 2018 19:40:29 +0000 (13:40 -0600)]
coresight: Cleanup device subtype struct

Clean up our struct a little bit by using a union instead of
a struct for tracking the subtype of a device.

Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agocoresight: Handle errors in finding input/output ports
Suzuki K Poulose [Wed, 11 Jul 2018 19:40:28 +0000 (13:40 -0600)]
coresight: Handle errors in finding input/output ports

If we fail to find the input / output port for a LINK component
while enabling a path, we should fail gracefully rather than
assuming port "0".

Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
6 years agocoresight: Cleanup platform description data
Suzuki K Poulose [Wed, 11 Jul 2018 19:40:27 +0000 (13:40 -0600)]
coresight: Cleanup platform description data

Nobody uses the "clk" field in struct coresight_platform_data.
Remove it.

Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>