This hooks up the endo, and modules, into the device tree.  All modules
for a specific endo are created when the host device is initialized.
When an interface is registered, the correct module for it is found and
that module is used for the sysfs tree.  When the interface is removed,
the reference on the module is dropped.

When the host device goes away, the whole endo and modules are removed
at once.
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Reviewed-by: Alex Elder <elder@linaro.org>

This adds endo.c and endo.h and provides functions to create an endo and
the initial 0x0555 set of modules.

But, it doesn't hook this logic up into the running code yet, that comes
next.
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Reviewed-by: Alex Elder <elder@linaro.org>

This adds the attributes power_control and present to a module.  It also
removes the unneeded module_id attribute, as that comes from the name of
the module itself.
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Reviewed-by: Alex Elder <elder@linaro.org>

Drop the host-driver buffer headroom that was used to transfer the cport
id on ES1 and ES2.

Rather than transferring additional bytes on the wire and having to deal
with buffer-alignment issues (e.g. requiring the headroom to be a
multiple of 8 bytes) simply drop the headroom functionality.

Host drivers are expected set up their transfer descriptors separately
from the data buffers and any intermediate drivers (e.g. for Greybus
over USB) can (ab)use the operation message pad bytes for now.
Signed-off-by: Johan Hovold <johan@hovoldconsulting.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

Fix transfer-buffer alignment of es2 as well.
Signed-off-by: Johan Hovold <johan@hovoldconsulting.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

Fix transfer-buffer alignment of outgoing transfers which are currently
byte aligned.

Some USB host drivers cannot handle byte-aligned buffers and will
allocate temporary buffers, which the data is copied to or from on every
transfer. This affects for example musb (e.g. Beaglebone Black) and
ehci-tegra (e.g. Jetson).

Instead of transferring pad bytes on the wire, let's (ab)use the pad
bytes of the operation message header to transfer the cport id. This
gives us properly aligned buffers and more efficient transfers in both
directions.

By using both pad bytes, we can also remove the arbitrary limitation of
256 cports.

Note that the protocol between the host driver and the UniPro bridge is
not necessarily Greybus. As long as the firmware clears the pad bytes
before forwarding the data, and the host driver does the same before
passing received data up the stack, this should be considered "legal"
use.
Signed-off-by: Johan Hovold <johan@hovoldconsulting.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

Add explicit pad bytes to the message header.
Signed-off-by: Johan Hovold <johan@hovoldconsulting.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

Make sure to allocate the message transfer-buffer separately from the
containing message structure to avoid data corruption on systems without
DMA-coherent caches.

The message structure contains state that is updated while the buffer
may be used for DMA, something which could lead to data corruption due
to cache-line sharing on some architectures.

Use the (renamed) message cache for the message structure itself and
allocate the buffer separately.

If the additional allocation is a concern, the message structures
could eventually be allocated as part of the operation structure.
Signed-off-by: Johan Hovold <johan@hovoldconsulting.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

Pass structured greybus messages rather than buffers to the host
drivers.

This will allow us to separate the transfer buffers from the message
structures.

Rename the related functions to reflect the new interface.
Signed-off-by: Johan Hovold <johan@hovoldconsulting.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

Move operation message-header to operation.h so that it can be used
by host drivers.
Signed-off-by: Johan Hovold <johan@hovoldconsulting.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

Remove unused and unnecessary buffer-alignment define that host driver
were supposed to use.

We can handle unaligned incoming buffers just fine by accessing the
operation-message header via a copy in the receive path, rather than
requiring host drivers to make sure the alignment is correct.
Signed-off-by: Johan Hovold <johan@hovoldconsulting.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

The buffer received from our current host driver is 1-byte aligned and
will therefore cause unaligned memory accesses if simply cast to an
operation-message header.

Fix this by making a properly aligned copy of the header in
gb_connection_recv_response before accessing its fields.

Note that this does not affect protocol drivers as the whole buffer is
copied when creating the corresponding request or response before being
forwarded.
Signed-off-by: Johan Hovold <johan@hovoldconsulting.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

Fix two bugs in es2 and do some minor clean up.
Signed-off-by: Johan Hovold <johan@hovoldconsulting.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

The maximum buffer size does not include the headroom, so subtract the
headroom size from the actual buffer size.
Signed-off-by: Johan Hovold <johan@hovoldconsulting.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

A stack-allocated buffer is not generally DMA-able and must not be used
for USB control transfers.

Note that the memset and extra buffer byte were redundant as no more
than the bytes actually transferred was ever added to the fifo.
Signed-off-by: Johan Hovold <johan@hovoldconsulting.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

Drop unnecessary explicit casts.
Signed-off-by: Johan Hovold <johan@hovoldconsulting.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

Alex suggested to name it class instead of class type.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

A Greybus driver will bind to a bundle, not an interface. Lets follow
this rule in code.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

A module can have more than one interfaces and we get hotplug events or
manifests for interfaces, not modules. Details like version, vendor,
product id, etc. can be different for different interfaces within the
same module and so shall be fetched from interface descriptor instead of
module descriptor.

So what we have been doing for module descriptors until now must be done
for interface descriptors. There can only be one interface descriptor in
the manifest. Module descriptor isn't used anymore and probably most of
its fields can be removed now.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

An interface can have 1 or more bundles. On link-up event, we must initialize
all the bundles associated with the interface.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

Currently we are creating bundles based on interface descriptors. An interface
can have one or more bundles associated with it and so a bundle must be created
based on a bundle descriptor.

Also get class_type from bundle descriptor.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

Report size isn't passed as first two bytes of the report according to
USB-HID spec. Get it from payload-size.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Reviewed-by: Johan Hovold <johan@hovoldconsulting.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

A bundle corresponds to a device and a greybus driver binds to it. This patch
adds a type and descriptor for bundle.

This also shuffles the values of 'enum greybus_descriptor_type' to align
them with Greybus Specifications.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

There can be more than one interface on a module and we need to know the
interface for which the event has occurred.

But at the same time we may not need the module id at all. During initial phase
when AP is probed, the AP will receive the unique Endo id which shall be enough
to draw relationships between interface and module ids.

Code for that isn't available today and so lets create another routine to get
module id (which needs to be fixed separately), which will simply return
interface id passed to it.

Now that we have interface id, update rest of the code to use it.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

Devices registered with the device-core needs to be freed by calling
device_unregister(). For module we are calling just put_device() and for
bundle, connection and interface we are calling device_del().

All of these are incomplete and so none of them get freed, i.e. the
.release() routine is never called for their devices.

Module being a special case that it needs to maintain a refcount or a
list of interfaces to trace its usage count. I have chosen refcount.

And so once the refcount is zero, we can Unregister the device and
module will get free as well.

Because of this bug in freeing devices, their sysfs directories were not
getting removed properly and after a manifest is parsed with the help of
gbsim, removing modules was creating problems. The sysfs directory
'greybus' wasn't getting removed. And inserting the modules again
resulted in warnings and insmod failure.

WARNING: CPU: 3 PID: 4277 at
/build/buildd/linux-3.13.0/fs/sysfs/dir.c:486
sysfs_warn_dup+0x86/0xa0()
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

These are definitions from Mark that I've consolidated into
one header file. I'd like to get these merged at some point
soon, so the audio driver and gbsim work can avoid having
out-of-tree dependencies.
Signed-off-by: Mark A. Greer <mgreer@animalcreek.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

This adds a proposed sysfs layout for greybus to Documentation to make
it easier for people to discuss / test things.  It includes a module, an
interface, a bundle, and a gpbridge binding to that bundle.

This was discussed on the projectara software mailing list.
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

Cut out some comments that are no longer operative.
Signed-off-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

These functions showed up in 3.12 or so, and we are stuck on 3.10 for
various reasons, so provide backports in kernel_ver.h so that we can
rely on these functions.
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

We should use the attribute groups, not group, for the device, so
add and remove it.  No one should ever be updating a sysfs group for a
device, as that can be pretty dangerous if you don't duplicate _all_
existing attribute for that device, and I don't think we were doing that
here.
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

x86 doesn't include SZ_4K somehow so explicitly include <linux/sizes.h>
to fix the build breakage.
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

Add a simple Greybus protocol in order to stress USB and Greybus.
This protocol currently support 2 requests: ping and transfer.

ping request is useful to measure latency.
Kernel send a ping request and firmware should respond with a ping.

The transfer request request is useful to stress Greybus and USB.
Kernel can send data from 0 to 4k and the firmware must send back the data to kernel.

This behaviour of gb-loopback module is controlled via sysfs.
Curently, connection sysfs folder is updated with new entries:
- type: Type of loopback message to send
  * 0 => Don't send message
  * 1 => Send ping message continuously (message without payload)
  * 2 => Send transer message continuously (message with payload)
- size: Size of transfer message payload: 0-4096 bytes
- ms_wait: Time to wait between two messages: 0-1024 ms

Module also export some statistics about connection:
- latency: Time to send and receive one message
- frequency: Number of packet sent per second on this cport
- throughput: Quantity of data sent and received on this cport
- error
All this statistics are cleared everytime type, size or ms_wait entries are updated.
Signed-off-by: Alexandre Bailon <abailon@baylibre.com>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

When usb_log_enable() is called, the global apb1_log_task is used to
hold the result of kthread_run().  It is possible for kthread_run()
to return an error pointer, so tests of apb_log_task against NULL
are insufficient to determine its validity.

Note that kthread_run() never returns NULL so we don't have to check
for that.  But apb1_log_task is initially NULL, so that global must
be both non-null and not an error in order to be considered valid.
Signed-off-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

When usb_log_enable() is called, the global apb1_log_task is used to
hold the result of kthread_run().  It is possible for kthread_run()
to return an error pointer, so tests of apb_log_task against NULL
are insufficient to determine its validity.

Note that kthread_run() never returns NULL so we don't have to check
for that.  But apb1_log_task is initially NULL, so that global must
be both non-null and not an error in order to be considered valid.
Signed-off-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

In identify_descriptor(), the variable desc_size represents the size
of a memory object.  So change its type from int to size_t.

The return value for this function can be desc_size cast to int.
One can verify by inspection this will never exceed INT_MAX.
Signed-off-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

Send response to incoming requests from the operation request handler
rather than in every protocol request_recv callback.

This simplifies request_recv error handling and allows for further code
reuse.

Note that if we ever get protocols that need to hold off sending
responses we could implement this by letting them return a special
value (after acquiring the necessary operation references) to suppress
the response from being sent by greybus core.
Signed-off-by: Johan Hovold <johan@hovoldconsulting.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

Send response also to incoming requests that cannot be fulfilled.
Signed-off-by: Johan Hovold <johan@hovoldconsulting.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

Add minimal verification of incoming report size, before using it to
determine what buffer and size to pass on to HID core.

Add comment about protocol needing to be revisited. If we are going to
be parsing the report data received, then those fields have to be
defined in the Greybus specification at least.
Signed-off-by: Johan Hovold <johan@hovoldconsulting.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>

Make sure to verify the length of incoming requests before trying to
parse the request buffer, which can even be NULL on empty requests.
Signed-off-by: Johan Hovold <johan@hovoldconsulting.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>