Add MEI BUS and NFC Device
Subsystems (3)
cover Add MEI BUS and NFC Device
This patch set adds implementation of MEI BUS abstraction over MEI device, this allows standard Linux device drivers to access functionality exposed by MEI device that was previously available only to the user space through /dev/mei
The first exercises is to export the NFC radio
More information can be found underDocumentation/misc-devices/mei/mei-bus.txt
Samuel Ortiz (11):
mei: bus: Initial MEI bus type implementation
mei: bus: Implement driver registration
mei: bus: Initial implementation for I/O routines
mei: bus: Add bus related structures to mei_cl
mei: bus: Call bus routines from the core code
mei: bus: Synchronous API for the data transmission
mei: bus: Implement bus driver data setter/getter
mei: nfc: Initial nfc implementation
mei: nfc: Connect also the regular ME client
mei: nfc: Add NFC device to the MEI bus
mei: nfc: Implement MEI bus IO ops
Documentation/misc-devices/mei/mei-bus.txt | 137 ++++++++ drivers/misc/mei/Kconfig | 7 + drivers/misc/mei/Makefile | 2 + drivers/misc/mei/bus.c | 483 ++++++++++++++++++++++++++++ drivers/misc/mei/bus.h | 32 ++ drivers/misc/mei/client.c | 4 + drivers/misc/mei/init.c | 1 + drivers/misc/mei/interrupt.c | 2 + drivers/misc/mei/mei_dev.h | 78 +++++ drivers/misc/mei/nfc.c | 460 ++++++++++++++++++++++++++ drivers/misc/mei/nfc.h | 141 ++++++++ drivers/misc/mei/pci-me.c | 8 + include/linux/mei_bus.h | 108 ++++++ 13 files changed, 1463 insertions(+), 0 deletions(-) create mode 100644 Documentation/misc-devices/mei/mei-bus.txt create mode 100644 drivers/misc/mei/bus.c create mode 100644 drivers/misc/mei/bus.h create mode 100644 drivers/misc/mei/nfc.c create mode 100644 drivers/misc/mei/nfc.h create mode 100644 include/linux/mei_bus.h
Signature · 1 line
1.7.4.4
PATCH 1/11 mei: bus: Initial MEI bus type implementation
From: Samuel Ortiz <[email protected]>
mei bus will present some of the me clients
as devices for other standard subsystems
Implement the probe, remove, match and the device addtion routines. A mei-bus.txt document describing the rationale and the API usage is also added.
--- Documentation/misc-devices/mei/mei-bus.txt | 137 ++++++++++++++++++++++++ drivers/misc/mei/Makefile | 1 + drivers/misc/mei/bus.c | 155 ++++++++++++++++++++++++++++ drivers/misc/mei/bus.h | 27 +++++ drivers/misc/mei/mei_dev.h | 24 +++++ include/linux/mei_bus.h | 91 ++++++++++++++++ 6 files changed, 435 insertions(+), 0 deletions(-) create mode 100644 Documentation/misc-devices/mei/mei-bus.txt create mode 100644 drivers/misc/mei/bus.c create mode 100644 drivers/misc/mei/bus.h create mode 100644 include/linux/mei_bus.h
Documentation/misc-devices/mei/mei-bus.txt new file +137 -0
@@ -0,0 +1,137 @@ +Intel(R) Management Engine (ME) bus API +======================================= + + +Rationale +========= +While a misc character device is useful for applications to send and receive +data to the many IP blocks found in Intel's ME, kernel drivers rely on the +device model to be probed. +By adding a kernel virtual bus abstraction on top of the MEI driver we can +implement drivers for the various MEI features as standalone ones, found in +their respective subsystem. Existing drivers can even potentially be re-used +by adding an MEI bus layer to the existing code. + + +MEI bus API +=========== +A driver implementation for an MEI IP block is very similar to existing bus +based device drivers. The driver registers itself as an MEI bus driver through +the mei_bus_driver structure: + +struct mei_bus_driver { + struct device_driver driver; + + struct mei_id id; + + int (*probe)(struct mei_bus_client *client); + int (*remove)(struct mei_bus_client *client); +}; + +struct mei_id { + char name[MEI_NAME_SIZE]; + uuid_le uuid; +}; + +The mei_id structure allows the driver to bind itself against an ME UUID and a +device name. There typically is one ME UUID per technology and the mei_id name +field matches a specific device name within that technology. As an example, +the ME supports several NFC devices: All of them have the same ME UUID but the +ME bus code will assign each of them a different name. + +To actually register a driver on the ME bus one must call the mei_add_driver() +API. This is typically called at module init time. + +Once registered on the ME bus, a driver will typically try to do some I/O on +this bus and this should be done through the mei_bus_send() and mei_bus_recv() +routines. The latter is synchronous (blocks and sleeps until data shows up). +In order for drivers to be notified of pending events waiting for them (e.g. +an Rx event) they can register an event handler through the +mei_bus_register_event_cb() routine. Currently only the MEI_BUS_EVENT_RX event +will trigger an event handler call and the driver implementation is supposed +to call mei_bus_recv() from the event handler in order to fetch the pending +received buffers. + + +Example +======= +As a theoretical example let's pretend the ME comes with a "contact" NFC IP. +The driver init and exit routines for this device would look like: + +#define CONTACT_DRIVER_NAME "contact" + +#define NFC_UUID UUID_LE(0x0bb17a78, 0x2a8e, 0x4c50, 0x94, \ + 0xd4, 0x50, 0x26, 0x67, 0x23, 0x77, 0x5c) + +static struct mei_bus_driver contact_driver = { + .driver = { + .name = CONTAC_DRIVER_NAME, + }, + .id = { + .name = CONTACT_DRIVER_NAME, + .uuid = NFC_UUID, + }, + + .probe = contact_probe, + .remove = contact_remove, +}; + +static int contact_init(void) +{ + int r; + + pr_debug(DRIVER_DESC ": %s\n", __func__); + + r = mei_add_driver(&contact_driver); + if (r) { + pr_err(CONTACT_DRIVER_NAME ": driver registration failed\n"); + return r; + } + + return 0; +} + +static void __exit contact_exit(void) +{ + mei_del_driver(&contact_driver); +} + +module_init(contact_init); +module_exit(contact_exit); + +And the driver's simplified probe routine would look like that: + +int contact_probe(struct mei_bus_client *client) +{ + struct contact_driver *contact; + + [...] + mei_bus_register_event_cb(client, contact_event_cb, contact); + + return 0; + } + +In the probe routine the driver basically registers an ME bus event handler +which is as close as it can get to registering a threaded IRQ handler. +The handler implementation will typically call some I/O routine depending on +the pending events: + +#define MAX_NFC_PAYLOAD + +static void contact_event_cb(struct mei_bus_client *client, u32 events, + void *context) +{ + struct contact_driver *contact = context; + + if (events & BIT(MEI_BUS_EVENT_RX)) { + u8 payload[MAX_NFC_PAYLOAD]; + int payload_size; + + payload_size = mei_bus_recv(client, payload, MAX_NFC_PAYLOAD); + if (payload_size <= 0) + return; + + /* Hook to the NFC subsystem */ + nfc_hci_recv_frame(contact->hdev, payload, payload_size); + } +}
drivers/misc/mei/Makefile +1 -0
@@ -10,5 +10,6 @@ mei-objs += client.o mei-objs += main.o mei-objs += amthif.o mei-objs += wd.o +mei-objs += bus.o mei-$(CONFIG_INTEL_MEI_ME) += pci-me.o mei-$(CONFIG_INTEL_MEI_ME) += hw-me.o
drivers/misc/mei/bus.c new file +155 -0
155 lines not on this page · raw message
drivers/misc/mei/bus.h new file +27 -0
@@ -0,0 +1,27 @@ +/* +* +* Intel Management Engine Interface (Intel MEI) Linux driver +* Copyright (c) 2003-2012, Intel Corporation. +* +* This program is free software; you can redistribute it and/or modify it +* under the terms and conditions of the GNU General Public License, +* version 2, as published by the Free Software Foundation. +* +* This program is distributed in the hope it will be useful, but WITHOUT +* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or +* FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for +* more details. +* +*/ + +#ifndef _MEI_BUS_H_ +#define _MEI_BUS_H_ + +#define to_mei_driver (d) container_of(d, struct mei_bus_driver, driver) +#define to_mei_client (d) container_of(d, struct mei_bus_client, dev) + +struct mei_bus_client *mei_add_device (struct mei_device *mei_dev, +uuid_le uuid,char *name); +void mei_remove_device (struct mei_bus_client *client); + +#endif /* _MEI_BUS_H_ */
drivers/misc/mei/mei_dev.h +24 -0
@@ -21,6 +21,7 @@#include <linux/watchdog.h> #include <linux/poll.h> #include <linux/mei.h> +#include <linux/mei_bus.h> #include "hw.h" #include "hw-me-regs.h"
@@ -264,6 +265,29 @@ struct mei_hw_ops { };/** +* mei_bus_client +* +* @uuid: me client uuid +* @name: client symbolic name +* @me_dev: mei device +* @cl: mei client +* @driver: mei bus driver for this client +* @dev: linux driver model device pointer +* @priv_data: client private data +*/ +struct mei_bus_client { +uuid_le uuid; +char name[MEI_NAME_SIZE ]; + +struct mei_device *mei_dev; +struct mei_cl *cl; +struct mei_bus_driver *driver; +struct device dev; + +void *priv_data; +}; + +/** * struct mei_device - MEI private device struct * @mem_addr - mem mapped base register address
include/linux/mei_bus.h new file +91 -0
91 lines not on this page · raw message
Signature · 1 line
1.7.4.4
Arnd Bergmann
maintainerRe: [char-misc-next 01/11] mei: bus: Initial MEI bus type implementation
Quoted 8 lines · Tomas Winkler
From: Samuel Ortiz <[email protected]>
mei bus will present some of the me clients
as devices for other standard subsystemsImplement the probe, remove, match and the device addtion routines. A mei-bus.txt document describing the rationale and the API usage is also added.
This looks really nice, but I also think you have reached the point where you are outgrowing the scope of drivers/misc. How about turning mei into a top-level subsystem along with this new bus_type?
Another option would be moving it to drivers/bus/mei/ if you don't want to be quite at the top.
Arnd
Arnd Bergmann
maintainerRe: [char-misc-next 01/11] mei: bus: Initial MEI bus type implementation
In reply to Tomas Winklerdrivers/misc/mei/bus.c line 155
Quoted 8 lines · Tomas Winkler
drivers/misc/mei/bus.c new file +8
-0+} +EXPORT_SYMBOL (mei_add_device); + +void mei_remove_device (struct mei_bus_client *client) +{ +device_unregister (&client->dev); +} +EXPORT_SYMBOL (mei_remove_device);
One more point: did you intentionally pick EXPORT_SYMBOL over EXPORT_SYMBOL_GPL here? It is your decision as the copyright holder, but the default is often EXPORT_SYMBOL_GPL these days, to make it clear that you don't expect closed source drivers to plug in there.
Arnd
Re: [char-misc-next 01/11] mei: bus: Initial MEI bus type implementation
Quoted 12 lines · Arnd Bergmann
drivers/misc/mei/bus.c line 155
Quoted 8 lines · Tomas Winkler
drivers/misc/mei/bus.c new file +8
-0+} +EXPORT_SYMBOL (mei_add_device); + +void mei_remove_device (struct mei_bus_client *client) +{ +device_unregister (&client->dev); +} +EXPORT_SYMBOL (mei_remove_device);One more point: did you intentionally pick
EXPORT_SYMBOLoverEXPORT_SYMBOL_GPLhere?
No, I didn't.
Quoted 4 lines
It is your decision as the copyright
holder, but the default is oftenEXPORT_SYMBOL_GPLthese days, to make it clear that you don't expect closed source drivers to plug in there.
I'll fix that if Tomas is ok with it.
Cheers,
Samuel.
Signature · 2 lines
Intel Open Source Technology Centre http://oss.intel.com/
RE: [char-misc-next 01/11] mei: bus: Initial MEI bus type implementation
Quoted 22 lines
-----Original Message-----
From: Samuel Ortiz [mailto:[email protected]]
Sent: Friday, February 08, 2013 00:59
To: Arnd Bergmann
Cc: Winkler, Tomas; [email protected]; linux- [email protected]
Subject: Re: [char-misc-next 01/11] mei: bus: Initial MEI bus type implementationQuoted 10 lines · Arnd Bergmann
drivers/misc/mei/bus.c line 155
Quoted 6 lines · Tomas Winkler
drivers/misc/mei/bus.c new file +6
-0+} +EXPORT_SYMBOL (mei_add_device); + +void mei_remove_device (struct mei_bus_client *client) { +device_unregister (&client->dev); } +EXPORT_SYMBOL (mei_remove_device);One more point: did you intentionally pick
EXPORT_SYMBOLoverEXPORT_SYMBOL_GPLhere?
Quoted 6 lines
No, I didn't.
Quoted 3 lines
It is your decision as the copyright
holder, but the default is oftenEXPORT_SYMBOL_GPLthese days, to make it clear that you don't expect closed source drivers to plug in there.I'll fix that if Tomas is ok with it.
So far I'm happy with not forcing GPL, but I will check that issue.
Thanks
Tomas
PATCH 2/11 mei: bus: Implement driver registration
In reply to Tomas WinklerFrom: Samuel Ortiz <[email protected]>
--- drivers/misc/mei/bus.c | 29 +++++++++++++++++++++++++++++ include/linux/mei_bus.h | 3 +++ 2 files changed, 32 insertions(+), 0 deletions(-)
drivers/misc/mei/bus.c +29 -0
@@ -153,3 +153,32 @@ void mei_remove_device(struct mei_bus_client *client)device_unregister (&client->dev); }EXPORT_SYMBOL (mei_remove_device); + +int mei_add_driver (struct mei_bus_driver *driver) +{ +int err; + +/* Can't register until after driver model init */ +if (unlikely (WARN_ON (!mei_bus_type.p))) +return -EAGAIN ; + + driver->driver.owner =THIS_MODULE ; + driver->driver.bus = &mei_bus_type; + + err =driver_register (&driver->driver); +if (err) +return err; + +pr_debug ("mei: driver [%s] registered\n" , driver->driver.name); + +return 0 ; +} +EXPORT_SYMBOL (mei_add_driver); + +void mei_del_driver (struct mei_bus_driver *driver) +{ +driver_unregister (&driver->driver); + +pr_debug ("mei: driver [%s] unregistered\n" , driver->driver.name); +} +EXPORT_SYMBOL (mei_del_driver);
include/linux/mei_bus.h +3 -0
@@ -88,4 +88,7 @@ struct mei_bus_driver { int (*remove)(struct mei_bus_client *client); }; +int mei_add_driver (struct mei_bus_driver *driver); +void mei_del_driver (struct mei_bus_driver *driver); +#endif /* _LINUX_MEI_BUS_H */
Signature · 1 line
1.7.4.4
Arnd Bergmann
maintainerRe: [char-misc-next 02/11] mei: bus: Implement driver registration
Quoted 1 line · Tomas Winkler
+int
mei_add_driver(struct mei_bus_driver *driver)
By convention, I think this should be called mei_driver_register(), matching what we do on most other subsystem. Similarly for mei_driver_unregister().
Arnd
PATCH 3/11 mei: bus: Initial implementation for I/O routines
In reply to Tomas WinklerFrom: Samuel Ortiz <[email protected]>
--- drivers/misc/mei/bus.c | 226 ++++++++++++++++++++++++++++++++++++++++++++ drivers/misc/mei/bus.h | 3 + drivers/misc/mei/mei_dev.h | 14 +++ include/linux/mei_bus.h | 11 ++ 4 files changed, 254 insertions(+), 0 deletions(-)
drivers/misc/mei/bus.c +226 -0
@@ -16,6 +16,7 @@#include <linux/module.h> #include <linux/device.h> #include <linux/kernel.h> +#include <linux/sched.h> #include <linux/init.h> #include <linux/errno.h> #include <linux/slab.h>
@@ -25,6 +26,8 @@#include <linux/mei_bus.h> #include "mei_dev.h" +#include "hw-me.h" +#include "client.h" #include "bus.h" static int mei_device_match (struct device *dev,struct device_driver *drv)
@@ -73,6 +76,11 @@ static int mei_device_remove(struct device *dev)if (!client || !dev->driver)return 0 ; +if (client->event_cb) { + client->event_cb =NULL ; +cancel_work_sync (&client->event_work); + } + driver =to_mei_driver (dev->driver);if (!driver->remove) { dev->driver =NULL ;
@@ -182,3 +190,221 @@ void mei_del_driver(struct mei_bus_driver *driver)pr_debug ("mei: driver [%s] unregistered\n" , driver->driver.name); }EXPORT_SYMBOL (mei_del_driver); + +int mei_send (struct mei_cl *cl,u8 *buf,size_t length) +{ +struct mei_device *dev; +struct mei_msg_hdr mei_hdr; +struct mei_cl_cb *cb; +int me_cl_id, err; + +if (WARN_ON (!cl || !cl->dev)) +return -ENODEV ; + +if (cl->state !=MEI_FILE_CONNECTED ) +return -ENODEV ; + + cb =mei_io_cb_init (cl,NULL ); +if (!cb) +return -ENOMEM ; + + err =mei_io_cb_alloc_req_buf (cb, length); +if (err <0 ) { +mei_io_cb_free (cb); +return err; + } + +memcpy (cb->request_buffer.data, buf, length); + cb->fop_type =MEI_FOP_WRITE ; + + dev = cl->dev; + +mutex_lock (&dev->device_lock); + +/* Check if we have an ME client device */ + me_cl_id =mei_me_cl_by_id (dev, cl->me_client_id); +if (me_cl_id == dev->me_clients_num) { + err = -ENODEV ; + goto out_err; + } + +if (length > dev->me_clients[me_cl_id].props.max_msg_length) { + err = -EINVAL ; + goto out_err; + } + + err =mei_cl_flow_ctrl_creds (cl); +if (err <0 ) + goto out_err; + +/* Host buffer is not ready, we queue the request */ +if (err ==0 || !dev->hbuf_is_ready) { + cb->buf_idx =0 ; + mei_hdr.msg_complete =0 ; + cl->writing_state =MEI_WRITING ; +list_add_tail (&cb->list, &dev->write_list.list); + +mutex_unlock (&dev->device_lock); + +return length; + } + + dev->hbuf_is_ready = false; + +/* Check for a maximum length */ +if (length >mei_hbuf_max_len (dev)) { + mei_hdr.length =mei_hbuf_max_len (dev); + mei_hdr.msg_complete =0 ; + }else { + mei_hdr.length = length; + mei_hdr.msg_complete =1 ; + } + + mei_hdr.host_addr = cl->host_client_id; + mei_hdr.me_addr = cl->me_client_id; + mei_hdr.reserved =0 ; + +if (mei_write_message (dev, &mei_hdr, buf)) { + err = -EIO ; + goto out_err; + } + + cl->writing_state =MEI_WRITING ; + cb->buf_idx = mei_hdr.length; + +if (!mei_hdr.msg_complete) { +list_add_tail (&cb->list, &dev->write_list.list); + }else { +if (mei_cl_flow_ctrl_reduce (cl)) { + err = -EIO ; + goto out_err; + } + +list_add_tail (&cb->list, &dev->write_waiting_list.list); + } + +mutex_unlock (&dev->device_lock); + +return mei_hdr.length; + +out_err: +mutex_unlock (&dev->device_lock); +mei_io_cb_free (cb); + +return err; +} + +int mei_recv (struct mei_cl *cl,u8 *buf,size_t length) +{ +struct mei_device *dev; +struct mei_cl_cb *cb; +size_t r_length; +int err; + +if (WARN_ON (!cl || !cl->dev)) +return -ENODEV ; + + dev = cl->dev; + +mutex_lock (&dev->device_lock); + +if (!cl->read_cb) { + err =mei_cl_read_start (cl); +if (err <0 ) { +mutex_unlock (&dev->device_lock); +return err; + } + } + +if (cl->reading_state !=MEI_READ_COMPLETE && + !waitqueue_active (&cl->rx_wait)) { +mutex_unlock (&dev->device_lock); + +if (wait_event_interruptible (cl->rx_wait, + (MEI_READ_COMPLETE == cl->reading_state))) { +if (signal_pending (current)) +return -EINTR ; +return -ERESTARTSYS ; + } + +mutex_lock (&dev->device_lock); + } + + cb = cl->read_cb; + +if (cl->reading_state !=MEI_READ_COMPLETE ) { + r_length =0 ; + goto out; + } + + r_length =min_t (size_t, length, cb->buf_idx); + +memcpy (buf, cb->response_buffer.data, r_length); + +mei_io_cb_free (cb); + cl->reading_state =MEI_IDLE ; + cl->read_cb =NULL ; + +out: +mutex_unlock (&dev->device_lock); + +return r_length; +} + +int mei_bus_send (struct mei_bus_client *client,u8 *buf,size_t length) +{ +struct mei_cl *cl =NULL ; + +/* TODO: hook between mei_bus_client and mei_cl */ + +if (client->ops && client->ops->send) +return client->ops->send (client, buf, length); + +return mei_send (cl, buf, length); +} +EXPORT_SYMBOL (mei_bus_send); + +int mei_bus_recv (struct mei_bus_client *client,u8 *buf,size_t length) +{ +struct mei_cl *cl =NULL ; + +/* TODO: hook between mei_bus_client and mei_cl */ + +if (client->ops && client->ops->recv) +return client->ops->recv (client, buf, length); + +return mei_recv (cl, buf, length); +} +EXPORT_SYMBOL (mei_bus_recv); + +static void mei_bus_event_work (struct work_struct *work) +{ +struct mei_bus_client *client; + + client = container_of(work, struct mei_bus_client, event_work); + +if (client->event_cb) + client->event_cb (client, client->events, client->event_context); + + client->events =0 ; + +/* Prepare for the next read */ +mei_cl_read_start (client->cl); +} + +int mei_bus_register_event_cb (struct mei_bus_client *client, +mei_bus_event_cb_t event_cb,void *context) +{ +if (client->event_cb) +return -EALREADY ; + + client->events =0 ; + client->event_cb = event_cb; + client->event_context = context; +INIT_WORK (&client->event_work, mei_bus_event_work); + +mei_cl_read_start (client->cl); + +return 0 ; +} +EXPORT_SYMBOL (mei_bus_register_event_cb);
drivers/misc/mei/bus.h +3 -0
@@ -24,4 +24,7 @@ struct mei_bus_client *mei_add_device(struct mei_device *mei_dev, uuid_le uuid, char *name);void mei_remove_device (struct mei_bus_client *client); +int mei_send (struct mei_cl *cl,u8 *buf,size_t length); +int mei_recv (struct mei_cl *cl,u8 *buf,size_t length); +#endif /* _MEI_BUS_H_ */
drivers/misc/mei/mei_dev.h +14 -0
@@ -264,6 +264,13 @@ struct mei_hw_ops { unsigned char *buf, unsigned long len); }; +struct mei_bus_client ; + +struct mei_bus_ops { +int (*send)(struct mei_bus_client *client,u8 *buf,size_t length); +int (*recv)(struct mei_bus_client *client,u8 *buf,size_t length); +}; +/** * mei_bus_client *
@@ -284,6 +291,13 @@ struct mei_bus_client {struct mei_bus_driver *driver;struct device dev; +struct mei_bus_ops *ops; + +struct work_struct event_work; +mei_bus_event_cb_t event_cb; +void *event_context; +unsigned long events; +void *priv_data; };
include/linux/mei_bus.h +11 -0
@@ -88,7 +88,18 @@ struct mei_bus_driver { int (*remove)(struct mei_bus_client *client); }; +#define MEI_BUS_EVENT_RX 0 +#define MEI_BUS_EVENT_TX 1 +int mei_add_driver (struct mei_bus_driver *driver);void mei_del_driver (struct mei_bus_driver *driver); +int mei_bus_send (struct mei_bus_client *client,u8 *buf,size_t length); +int mei_bus_recv (struct mei_bus_client *client,u8 *buf,size_t length); + +typedef void (*mei_bus_event_cb_t )(struct mei_bus_client *client, +u32 events,void *context); +int mei_bus_register_event_cb (struct mei_bus_client *client, +mei_bus_event_cb_t read_cb,void *context); +#endif /* _LINUX_MEI_BUS_H */
Signature · 1 line
1.7.4.4
Arnd Bergmann
maintainerRe: [char-misc-next 03/11] mei: bus: Initial implementation for I/O routines
drivers/misc/mei/mei_dev.h struct mei_hw_ops { · line 273
Quoted 6 lines · Tomas Winkler
drivers/misc/mei/mei_dev.h +6
-0+ +struct mei_bus_ops { +int (*send)(struct mei_bus_client *client,u8 *buf,size_t length); +int (*recv)(struct mei_bus_client *client,u8 *buf,size_t length); +}; +
Can you have more than one set of mei_bus_ops in a driver? If not, how about adding the callbacks to the mei_bus_driver structure directly as a simplification? Never mind if that doesn't work for some obvious reason I missed.
Arnd
Re: [char-misc-next 03/11] mei: bus: Initial implementation for I/O routines
Hi Arnd,
Quoted 9 lines · Arnd Bergmann
drivers/misc/mei/mei_dev.h struct mei_hw_ops { · line 273
Quoted 6 lines · Tomas Winkler
drivers/misc/mei/mei_dev.h +6
-0+ +struct mei_bus_ops { +int (*send)(struct mei_bus_client *client,u8 *buf,size_t length); +int (*recv)(struct mei_bus_client *client,u8 *buf,size_t length); +}; +Can you have more than one set of
mei_bus_opsin a driver?
You can have at most one mei_bus_ops per mei_bus_client.
Quoted 2 lines
If not, how about adding the callbacks to the
mei_bus_driverstructure directly as a simplification?
I can add the ops directly to the mei_bus_client structure, yes.
Cheers,
Samuel.
Signature · 2 lines
Intel Open Source Technology Centre http://oss.intel.com/
Arnd Bergmann
maintainerRe: [char-misc-next 03/11] mei: bus: Initial implementation for I/O routines
Quoted 10 lines · Samuel Ortiz
Quoted 9 lines · Arnd Bergmann
drivers/misc/mei/mei_dev.h struct mei_hw_ops { · line 273
Quoted 6 lines · Tomas Winkler
drivers/misc/mei/mei_dev.h +6
-0+ +struct mei_bus_ops { +int (*send)(struct mei_bus_client *client,u8 *buf,size_t length); +int (*recv)(struct mei_bus_client *client,u8 *buf,size_t length); +}; +Can you have more than one set of
mei_bus_opsin a driver?
Quoted 5 lines
You can have at most one
mei_bus_opspermei_bus_client.Quoted 2 lines
If not, how about adding the callbacks to the
mei_bus_driverstructure directly as a simplification?I can add the ops directly to the
mei_bus_clientstructure, yes.
I looked at the new version, and it's not what I assumed it would be. I thought the operations were specific to a client driver and should be part of the /mei_bus_driver/ structure, not the /mei_bus_client/.
Did I misunderstand what these functions do, or did you misunderstand what I was asking for?
Arnd
Re: [char-misc-next 03/11] mei: bus: Initial implementation for I/O routines
Hi Arnd,
Quoted 13 lines · Arnd Bergmann
Quoted 12 lines · Samuel Ortiz
Quoted 9 lines · Arnd Bergmann
drivers/misc/mei/mei_dev.h struct mei_hw_ops { · line 273
Quoted 6 lines · Tomas Winkler
drivers/misc/mei/mei_dev.h +6
-0+ +struct mei_bus_ops { +int (*send)(struct mei_bus_client *client,u8 *buf,size_t length); +int (*recv)(struct mei_bus_client *client,u8 *buf,size_t length); +}; +Can you have more than one set of
mei_bus_opsin a driver?You can have at most one
mei_bus_opspermei_bus_client.
Quoted 7 lines
Quoted 3 lines
Quoted 2 lines
If not, how about adding the callbacks to the
mei_bus_driverstructure directly as a simplification?I can add the ops directly to the
mei_bus_clientstructure, yes.I looked at the new version, and it's not what I assumed it would be. I thought the operations were specific to a client driver and should be part of the /mei_bus_driver/ structure, not the /mei_bus_client/.
The ops should be part of mei_bus_client as they're specific to the MEI protocol for a given IP block on the ME. You need to have MEI and HECI knowledge to implement those ops and drivers (defining their mei_bus_driver structure) should not have that kind of knowledge but rather focus on the technology they're driving.
If we take the NFC example again, the drivers/nfc/ code will send NFC payloads to the bus I/O routines and this is where the mei_bus_client ops will add the ME specific protocol (command and request id for the NFC block) on top of it. In practice, this is an additional header which handles a transport layer that's specific not only to the ME but to the NFC block of it. So each ME block can have its own protocol to send and receive technology specific payloads, that's what those ops implement.
That's why I think that those ops should not be defined by the drivers/nfc/ code and in fact should be opaque to it.
Quoted 2 lines
Did I misunderstand what these functions do, or did you misunderstand what I was asking for?
Probably both. I really thought you were looking for a structure cleanup by directly putting the ops into the mei_bus_client structure.
Cheers,
Samuel.
Signature · 2 lines
Intel Open Source Technology Centre http://oss.intel.com/
Arnd Bergmann
maintainerRe: [char-misc-next 03/11] mei: bus: Initial implementation for I/O routines
Quoted 21 lines · Samuel Ortiz
Quoted 13 lines · Arnd Bergmann
Quoted 12 lines · Samuel Ortiz
Quoted 9 lines · Arnd Bergmann
drivers/misc/mei/mei_dev.h struct mei_hw_ops { · line 273
Quoted 6 lines · Tomas Winkler
drivers/misc/mei/mei_dev.h +6
-0+ +struct mei_bus_ops { +int (*send)(struct mei_bus_client *client,u8 *buf,size_t length); +int (*recv)(struct mei_bus_client *client,u8 *buf,size_t length); +}; +Can you have more than one set of
mei_bus_opsin a driver?You can have at most one
mei_bus_opspermei_bus_client.Quoted 7 lines
Quoted 3 lines
Quoted 2 lines
If not, how about adding the callbacks to the
mei_bus_driverstructure directly as a simplification?I can add the ops directly to the
mei_bus_clientstructure, yes.I looked at the new version, and it's not what I assumed it would be. I thought the operations were specific to a client driver and should be part of the /mei_bus_driver/ structure, not the /mei_bus_client/.
Quoted 14 lines
The ops should be part of
mei_bus_clientas they're specific to the MEI protocol for a given IP block on the ME. You need to have MEI and HECI knowledge to implement those ops and drivers (defining theirmei_bus_driverstructure) should not have that kind of knowledge but rather focus on the technology they're driving.
If we take the NFC example again, thedrivers/nfc/code will send NFC payloads to the bus I/O routines and this is where themei_bus_clientops will add the ME specific protocol (command and request id for the NFC block) on top of it. In practice, this is an additional header which handles a transport layer that's specific not only to the ME but to the NFC block of it. So each ME block can have its own protocol to send and receive technology specific payloads, that's what those ops implement.
That's why I think that those ops should not be defined by thedrivers/nfc/code and in fact should be opaque to it.
I think I'm still confused. What I read in your description is that unlike other subsystems that have a common bus implementation that is hooked into by two kinds of drivers (bus drivers that provide devices like a USB host controller, and device drivers that use those devices), you have three components that can be mixed independently: the bus (based on which PCI device you have), the transport and the endpoint device. Is that correct?
If so, how do you know which transport to use?
Arnd
Re: [char-misc-next 03/11] mei: bus: Initial implementation for I/O routines
Hi Arnd,
Quoted 37 lines · Arnd Bergmann
Quoted 21 lines · Samuel Ortiz
Quoted 13 lines · Arnd Bergmann
Quoted 12 lines · Samuel Ortiz
Quoted 9 lines · Arnd Bergmann
drivers/misc/mei/mei_dev.h struct mei_hw_ops { · line 273
Quoted 6 lines · Tomas Winkler
drivers/misc/mei/mei_dev.h +6
-0+ +struct mei_bus_ops { +int (*send)(struct mei_bus_client *client,u8 *buf,size_t length); +int (*recv)(struct mei_bus_client *client,u8 *buf,size_t length); +}; +Can you have more than one set of
mei_bus_opsin a driver?You can have at most one
mei_bus_opspermei_bus_client.Quoted 7 lines
Quoted 3 lines
Quoted 2 lines
If not, how about adding the callbacks to the
mei_bus_driverstructure directly as a simplification?I can add the ops directly to the
mei_bus_clientstructure, yes.I looked at the new version, and it's not what I assumed it would be. I thought the operations were specific to a client driver and should be part of the /mei_bus_driver/ structure, not the /mei_bus_client/.
Quoted 14 lines
The ops should be part of
mei_bus_clientas they're specific to the MEI protocol for a given IP block on the ME. You need to have MEI and HECI knowledge to implement those ops and drivers (defining theirmei_bus_driverstructure) should not have that kind of knowledge but rather focus on the technology they're driving.
If we take the NFC example again, thedrivers/nfc/code will send NFC payloads to the bus I/O routines and this is where themei_bus_clientops will add the ME specific protocol (command and request id for the NFC block) on top of it. In practice, this is an additional header which handles a transport layer that's specific not only to the ME but to the NFC block of it. So each ME block can have its own protocol to send and receive technology specific payloads, that's what those ops implement.
That's why I think that those ops should not be defined by thedrivers/nfc/code and in fact should be opaque to it.
Quoted 1 line
I think I'm still confused.
It is confusing, and my explanations are probably not as good as they should be.
Quoted 7 lines
What I read in your description is that unlike
other subsystems that have a common bus implementation that is hooked into by two kinds of drivers (bus drivers that provide devices like a USB host controller, and device drivers that use those devices), you have three components that can be mixed independently: the bus (based on which PCI device you have), the transport and the endpoint device. Is that correct?
That's quite accurate, yes.
Quoted 1 line
If so, how do you know which transport to use?
Through the mei_bus_client ops. Device drivers get a mei_bus_client pointer from their probe routine and the ops pointers there (If any) are set by whoever creates the device. In the NFC case mei/nfc.c does that and implements the NFC specific transport code for this technology. mei/nfc.c is also the part of the code that actually adds the device to the bus.
So when a device driver wants e.g. to send its payload through the MEI bus, it calls mei_bus_send() which takes the device driver mei_bus client pointer as its first argument.
Then the payload may go through mei_bus_client->send() first which will eventually physically sent the newly built frame through mei_send(). Some ME blocks don't require any additional transport layer and in that case the device driver payload will go straight to mei_send() since the mei_bus_client ops will be NULL.
Does that make more sense now ?
Cheers,
Samuel.
Signature · 2 lines
Intel Open Source Technology Centre http://oss.intel.com/
Arnd Bergmann
maintainerRe: [char-misc-next 03/11] mei: bus: Initial implementation for I/O routines
Quoted 7 lines · On Monday 11 February 2013 16:48:12 Samuel Ortiz wrote:
Quoted 2 lines
If so, how do you know which transport to use?
Through the
mei_bus_clientops. Device drivers get amei_bus_clientpointer from their probe routine and the ops pointers there (If any) are set by whoever creates the device. In the NFC casemei/nfc.cdoes that and implements the NFC specific transport code for this technology.mei/nfc.cis also the part of the code that actually adds the device to the bus.
Ok.
Quoted 9 lines
So when a device driver wants e.g. to send its payload through the MEI bus, it calls
mei_bus_send()which takes the device drivermei_busclient pointer as its first argument.
Then the payload may go throughmei_bus_client->send()first which will eventually physically sent the newly built frame throughmei_send(). Some ME blocks don't require any additional transport layer and in that case the device driver payload will go straight tomei_send()since themei_bus_clientops will be NULL.Does that make more sense now ?
Yes, so it's not actually as complicated as I thought, because the operations are already known by the bus driver (in the normal sense of that word, meaning the driver that creates the device not the one driving it) at the time when the device is created.
In that case, please ignore my original comment and put the struct mei_bus_ops (mei_transport_ops?) back, ideally as a "const" member of the struct mei_bus_device (to be called mei_device then), so you can declare the structure statically and constant in the driver that has the functions.
Arnd
PATCH 4/11 mei: bus: Add bus related structures to mei_cl
In reply to Tomas WinklerFrom: Samuel Ortiz <[email protected]>
We keep track of all MEI bus clients through a specific linked list. We also have a mei_bus_client instance in the mei_cl structure.
--- drivers/misc/mei/bus.c | 47 +++++++++++++++++++++++++++++++------------ drivers/misc/mei/client.c | 1 + drivers/misc/mei/init.c | 1 + drivers/misc/mei/mei_dev.h | 8 +++++++ 4 files changed, 44 insertions(+), 13 deletions(-)
drivers/misc/mei/bus.c +34 -13
@@ -119,16 +119,37 @@ static struct device_type mei_client_type = { .release = mei_client_dev_release, }; +static struct mei_cl *mei_bus_find_mei_cl_by_uuid (struct mei_device *mei_dev, +uuid_le uuid) +{ +struct mei_cl *cl, *next; + +list_for_each_entry_safe (cl, next, + &mei_dev->bus_client_list, bus_client_link) { +if (!uuid_le_cmp (uuid, cl->bus_client_uuid)) +return cl; + } + +return NULL ; +} +struct mei_bus_client *mei_add_device (struct mei_device *mei_dev,uuid_le uuid,char *name) {struct mei_bus_client *client; +struct mei_cl *cl;int status; + cl =mei_bus_find_mei_cl_by_uuid (mei_dev, uuid); +if (cl ==NULL ) +return NULL ; + client =kzalloc (sizeof (struct mei_bus_client ),GFP_KERNEL );if (!client)return NULL ; + client->cl = cl; + client->mei_dev = mei_dev; client->uuid = uuid;strlcpy (client->name, name,sizeof (client->name));
@@ -140,19 +161,17 @@ struct mei_bus_client *mei_add_device(struct mei_device *mei_dev,dev_set_name (&client->dev,"%s" , client->name); status =device_register (&client->dev); -if (status) - goto out_err; +if (status) { +kfree (client); +dev_err (client->dev.parent,"Failed to register MEI client\n" ); +return NULL ; + } + + cl->client = client;dev_dbg (&client->dev,"client %s registered\n" , client->name);return client; - -out_err: -dev_err (client->dev.parent,"Failed to register MEI client\n" ); - -kfree (client); - -return NULL ; }EXPORT_SYMBOL (mei_add_device);
@@ -353,9 +372,10 @@ out:int mei_bus_send (struct mei_bus_client *client,u8 *buf,size_t length) { -struct mei_cl *cl =NULL ; +struct mei_cl *cl = client->cl; -/* TODO: hook between mei_bus_client and mei_cl */ +if (cl ==NULL ) +return -ENODEV ;if (client->ops && client->ops->send)return client->ops->send (client, buf, length);
@@ -366,9 +386,10 @@ EXPORT_SYMBOL(mei_bus_send);int mei_bus_recv (struct mei_bus_client *client,u8 *buf,size_t length) { -struct mei_cl *cl =NULL ; +struct mei_cl *cl = client->cl; -/* TODO: hook between mei_bus_client and mei_cl */ +if (cl ==NULL ) +return -ENODEV ;if (client->ops && client->ops->recv)return client->ops->recv (client, buf, length);
drivers/misc/mei/client.c +1 -0
@@ -216,6 +216,7 @@ void mei_cl_init(struct mei_cl *cl, struct mei_device *dev)init_waitqueue_head (&cl->rx_wait);init_waitqueue_head (&cl->tx_wait);INIT_LIST_HEAD (&cl->link); +INIT_LIST_HEAD (&cl->bus_client_link); cl->reading_state =MEI_IDLE ; cl->writing_state =MEI_IDLE ; cl->dev = dev;
drivers/misc/mei/init.c +1 -0
@@ -46,6 +46,7 @@ void mei_device_init(struct mei_device *dev) {/* setup our list array */ INIT_LIST_HEAD (&dev->file_list); +INIT_LIST_HEAD (&dev->bus_client_list);mutex_init (&dev->device_lock);init_waitqueue_head (&dev->wait_recvd_msg);init_waitqueue_head (&dev->wait_stop_wd);
drivers/misc/mei/mei_dev.h +8 -0
@@ -210,6 +210,11 @@ struct mei_cl {enum mei_file_transaction_states writing_state;int sm_state;struct mei_cl_cb *read_cb; + +/* MEI bus data */ +struct mei_bus_client *client; +struct list_head bus_client_link; +uuid_le bus_client_uuid; };/** struct mei_hw_ops
@@ -403,6 +408,9 @@ struct mei_device {struct work_struct init_work; +/* List of bus clients */ +struct list_head bus_client_list; +const struct mei_hw_ops *ops;char hw[0 ]__aligned(sizeof(void *)) ; };
Signature · 1 line
1.7.4.4
PATCH 5/11 mei: bus: Call bus routines from the core code
In reply to Tomas WinklerFrom: Samuel Ortiz <[email protected]>
Register the MEI bus type against the kernel core bus APIs and call the bus Rx handler from interrupt.c
--- drivers/misc/mei/bus.c | 22 ++++++++++++++++++++++ drivers/misc/mei/interrupt.c | 2 ++ drivers/misc/mei/mei_dev.h | 4 ++++ drivers/misc/mei/pci-me.c | 6 ++++++ 4 files changed, 34 insertions(+), 0 deletions(-)
drivers/misc/mei/bus.c +22 -0
@@ -429,3 +429,25 @@ int mei_bus_register_event_cb(struct mei_bus_client *client,return 0 ; }EXPORT_SYMBOL (mei_bus_register_event_cb); + +void mei_bus_rx_event (struct mei_cl *cl) +{ +struct mei_bus_client *client = cl->client; + +if (!client || !client->event_cb) +return ; + +set_bit (MEI_BUS_EVENT_RX , &client->events); + +schedule_work (&client->event_work); +} + +int mei_bus_init (struct pci_dev *pdev) +{ +return bus_register (&mei_bus_type); +} + +void mei_bus_exit (void ) +{ +bus_unregister (&mei_bus_type); +}
drivers/misc/mei/interrupt.c +2 -0
@@ -49,6 +49,8 @@ void mei_irq_complete_handler(struct mei_cl *cl, struct mei_cl_cb *cb_pos) cl->reading_state =MEI_READ_COMPLETE ;if (waitqueue_active (&cl->rx_wait))wake_up_interruptible (&cl->rx_wait); +else +mei_bus_rx_event (cl); } }
drivers/misc/mei/mei_dev.h +4 -0
@@ -420,6 +420,10 @@ static inline unsigned long mei_secs_to_jiffies(unsigned long sec)return msecs_to_jiffies (sec *MSEC_PER_SEC ); } +void mei_bus_rx_event (struct mei_cl *cl); +int mei_bus_init (struct pci_dev *pdev); +void mei_bus_exit (void ); +/* * mei init function prototypes
drivers/misc/mei/pci-me.c +6 -0
@@ -197,6 +197,9 @@ static int mei_probe(struct pci_dev *pdev, const struct pci_device_id *ent) mei_pdev = pdev;pci_set_drvdata (pdev, dev); + err =mei_bus_init (mei_pdev); +if (err) + goto deregister_mei;schedule_delayed_work (&dev->timer_work,HZ );
@@ -206,6 +209,8 @@ static int mei_probe(struct pci_dev *pdev, const struct pci_device_id *ent)return 0 ; +deregister_mei: +mei_deregister (); release_irq:mei_disable_interrupts (dev);flush_scheduled_work ();
@@ -300,6 +305,7 @@ static void mei_remove(struct pci_dev *pdev)pci_release_regions (pdev);pci_disable_device (pdev); +mei_bus_exit ();mei_deregister (); }
Signature · 1 line
1.7.4.4
Arnd Bergmann
maintainerRe: [char-misc-next 05/11] mei: bus: Call bus routines from the core code
drivers/misc/mei/pci-me.c static int mei_probe(struct pci_dev *pdev, const struct pci_device_id *ent) · line 205
Quoted 10 lines · Tomas Winkler
drivers/misc/mei/pci-me.c +3
-0@@ -197,6 +197,9 @@ static int mei_probe(struct pci_dev *pdev, const struct pci_device_id *ent) mei_pdev = pdev;pci_set_drvdata (pdev, dev); + err =mei_bus_init (mei_pdev); +if (err) + goto deregister_mei;schedule_delayed_work (&dev->timer_work,HZ );
This is fairly unusual, and will break if you ever have multiple mei devices in one system, because you end up registering the bus type for each device. I think it would be more logical to register/unregister the bus_type from the module_init/exit functions of the module that contains the bus_type object.
Arnd
RE: [char-misc-next 05/11] mei: bus: Call bus routines from the core code
Quoted 21 lines
-----Original Message-----
From: Arnd Bergmann [mailto:[email protected]]
Sent: Friday, February 08, 2013 00:38
To: Winkler, Tomas
Cc: [email protected]; [email protected]; linux- [email protected]
Subject: Re: [char-misc-next 05/11] mei: bus: Call bus routines from the core codeReview static int mei_probe(struct pci_dev *pdev, const
Quoted 1 line · Tomas Winkler
(quoted diff) +0
-0@@ -197,6 +197,9 @@ static int mei_probe(struct pci_dev *pdev, conststruct
pci_device_id*ent)Quoted 7 lines
drivers/misc/mei/pci-me.c +3
-0mei_pdev = pdev;pci_set_drvdata (pdev, dev); + err =mei_bus_init (mei_pdev); +if (err) + goto deregister_mei;
Quoted 7 lines
drivers/misc/mei/pci-me.c static int mei_probe(struct pci_dev *pdev, const struct pci_device_id *ent) · line 205
Quoted 2 lines
drivers/misc/mei/pci-me.c +0
-0schedule_delayed_work (&dev->timer_work,HZ );This is fairly unusual, and will break if you ever have multiple mei devices in one system, because you end up registering the bus type for each device. I think it would be more logical to register/unregister the
bus_typefrom the module_init/exit functions of the module that contains thebus_typeobject.
MEI is a singleton and it is enforced also in software in the probe
Second why to register anything if the MEI device is not present on the system.
Thanks
Tomas
Arnd Bergmann
maintainerRe: [char-misc-next 05/11] mei: bus: Call bus routines from the core code
Quoted 1 line · Tomas
Second why to register anything if the MEI device is not present on the system.
Mostly to match the expectations of readers of that code. Note that no memory is wasted if you do this, because it's a static data structure. You can actually save a few bytes because the registration call moves into an __init section that is discarded after boot.
Arnd
Re: [char-misc-next 05/11] mei: bus: Call bus routines from the core code
In reply to Arnd BergmannHi Arnd,
Quoted 9 lines · Arnd Bergmann
Quoted 8 lines · Tomas Winkler
drivers/misc/mei/pci-me.c +3
-0@@ -197,6 +197,9 @@ static int mei_probe(struct pci_dev *pdev, const struct pci_device_id *ent) mei_pdev = pdev;pci_set_drvdata (pdev, dev); + err =mei_bus_init (mei_pdev); +if (err) + goto deregister_mei;
Quoted 8 lines
drivers/misc/mei/pci-me.c static int mei_probe(struct pci_dev *pdev, const struct pci_device_id *ent) · line 205
Quoted 2 lines
drivers/misc/mei/pci-me.c +0
-0schedule_delayed_work (&dev->timer_work,HZ );This is fairly unusual, and will break if you ever have multiple mei devices in one system, because you end up registering the bus type for each device. I think it would be more logical to register/unregister the
bus_typefrom the module_init/exit functions of the module that contains thebus_typeobject.
That makes sense, I'll fix that.
Cheers,
Samuel.
Signature · 2 lines
Intel Open Source Technology Centre http://oss.intel.com/
PATCH 6/11 mei: bus: Synchronous API for the data transmission
In reply to Tomas WinklerFrom: Samuel Ortiz <[email protected]>
Define a truly synchronous API for the bus Tx path by putting all pending request to the write list and wait for the interrupt tx handler to wake us up.
The __mei_send() out path is also slightly reworked to make it look more like main.c:mei_write().
--- drivers/misc/mei/bus.c | 38 ++++++++++++++++++++++++++++---------- drivers/misc/mei/bus.h | 2 ++ 2 files changed, 30 insertions(+), 10 deletions(-)
drivers/misc/mei/bus.c +28 -10
@@ -210,7 +210,7 @@ void mei_del_driver(struct mei_bus_driver *driver) }EXPORT_SYMBOL (mei_del_driver); -int mei_send (struct mei_cl *cl,u8 *buf,size_t length) +static int __mei_send (struct mei_cl *cl,u8 *buf,size_t length,bool blocking) {struct mei_device *dev;struct mei_msg_hdr mei_hdr;
@@ -261,11 +261,8 @@ int mei_send(struct mei_cl *cl, u8 *buf, size_t length) cb->buf_idx =0 ; mei_hdr.msg_complete =0 ; cl->writing_state =MEI_WRITING ; -list_add_tail (&cb->list, &dev->write_list.list); - -mutex_unlock (&dev->device_lock); -return length; + goto out; } dev->hbuf_is_ready = false;
@@ -291,19 +288,30 @@ int mei_send(struct mei_cl *cl, u8 *buf, size_t length) cl->writing_state =MEI_WRITING ; cb->buf_idx = mei_hdr.length; -if (!mei_hdr.msg_complete) { -list_add_tail (&cb->list, &dev->write_list.list); - }else { +out: +if (mei_hdr.msg_complete) {if (mei_cl_flow_ctrl_reduce (cl)) { - err = -EIO ; + err = -ENODEV ; goto out_err; } -list_add_tail (&cb->list, &dev->write_waiting_list.list); + }else { +list_add_tail (&cb->list, &dev->write_list.list); }mutex_unlock (&dev->device_lock); +if (blocking && cl->writing_state !=MEI_WRITE_COMPLETE ) { +if (wait_event_interruptible (cl->tx_wait, + cl->writing_state ==MEI_WRITE_COMPLETE )) { +if (signal_pending (current)) + err = -EINTR ; + err = -ERESTARTSYS ; +mutex_lock (&dev->device_lock); + goto out_err; + } + } +return mei_hdr.length; out_err:
@@ -370,6 +378,16 @@ out:return r_length; } +inline int mei_async_send (struct mei_cl *cl,u8 *buf,size_t length) +{ +return __mei_send (cl, buf, length,0 ); +} + +inline int mei_send (struct mei_cl *cl,u8 *buf,size_t length) +{ +return __mei_send (cl, buf, length,1 ); +} +int mei_bus_send (struct mei_bus_client *client,u8 *buf,size_t length) {struct mei_cl *cl = client->cl;
drivers/misc/mei/bus.h +2 -0
@@ -24,7 +24,9 @@ struct mei_bus_client *mei_add_device(struct mei_device *mei_dev, uuid_le uuid, char *name);void mei_remove_device (struct mei_bus_client *client); +int mei_async_send (struct mei_cl *cl,u8 *buf,size_t length);int mei_send (struct mei_cl *cl,u8 *buf,size_t length);int mei_recv (struct mei_cl *cl,u8 *buf,size_t length); +#endif /* _MEI_BUS_H_ */
Signature · 1 line
1.7.4.4
PATCH 7/11 mei: bus: Implement bus driver data setter/getter
In reply to Tomas WinklerFrom: Samuel Ortiz <[email protected]>
MEI bus drivers should be able to carry their private data around.
--- drivers/misc/mei/bus.c | 12 ++++++++++++ include/linux/mei_bus.h | 3 +++ 2 files changed, 15 insertions(+), 0 deletions(-)
drivers/misc/mei/bus.c +12 -0
@@ -448,6 +448,18 @@ int mei_bus_register_event_cb(struct mei_bus_client *client, }EXPORT_SYMBOL (mei_bus_register_event_cb); +inline void *mei_bus_get_clientdata (const struct mei_bus_client *client) +{ +return dev_get_drvdata (&client->dev); +} +EXPORT_SYMBOL (mei_bus_get_clientdata); + +inline void mei_bus_set_clientdata (struct mei_bus_client *client,void *data) +{ +dev_set_drvdata (&client->dev, data); +} +EXPORT_SYMBOL (mei_bus_set_clientdata); +void mei_bus_rx_event (struct mei_cl *cl) {struct mei_bus_client *client = cl->client;
include/linux/mei_bus.h +3 -0
@@ -102,4 +102,7 @@ typedef void (*mei_bus_event_cb_t)(struct mei_bus_client *client,int mei_bus_register_event_cb (struct mei_bus_client *client,mei_bus_event_cb_t read_cb,void *context); +void *mei_bus_get_clientdata (const struct mei_bus_client *client); +void mei_bus_set_clientdata (struct mei_bus_client *client,void *data); +#endif /* _LINUX_MEI_BUS_H */
Signature · 1 line
1.7.4.4
Arnd Bergmann
maintainerRe: [char-misc-next 07/11] mei: bus: Implement bus driver data setter/getter
drivers/misc/mei/bus.c int mei_bus_register_event_cb(struct mei_bus_client *client, · line 456
Quoted 7 lines · Tomas Winkler
drivers/misc/mei/bus.c +6
-0+inline void *mei_bus_get_clientdata (const struct mei_bus_client *client) +{ +return dev_get_drvdata (&client->dev); +} +EXPORT_SYMBOL (mei_bus_get_clientdata); +
Did you really mean to export an inline function?
Can you make this a static inline function in a header file instead?
Arnd
Re: [char-misc-next 07/11] mei: bus: Implement bus driver data setter/getter
Quoted 11 lines · Arnd Bergmann
drivers/misc/mei/bus.c int mei_bus_register_event_cb(struct mei_bus_client *client, · line 456
Quoted 7 lines · Tomas Winkler
drivers/misc/mei/bus.c +6
-0+inline void *mei_bus_get_clientdata (const struct mei_bus_client *client) +{ +return dev_get_drvdata (&client->dev); +} +EXPORT_SYMBOL (mei_bus_get_clientdata); +Did you really mean to export an inline function?
Quoted 1 line
Can you make this a static inline function in a header file instead?
Sure, will do.
Cheers,
Samuel.
Signature · 2 lines
Intel Open Source Technology Centre http://oss.intel.com/
Re: [char-misc-next 07/11] mei: bus: Implement bus driver data setter/getter
Quoted 12 lines · Samuel Ortiz
Quoted 11 lines · Arnd Bergmann
drivers/misc/mei/bus.c int mei_bus_register_event_cb(struct mei_bus_client *client, · line 456
Quoted 7 lines · Tomas Winkler
drivers/misc/mei/bus.c +6
-0+inline void *mei_bus_get_clientdata (const struct mei_bus_client *client) +{ +return dev_get_drvdata (&client->dev); +} +EXPORT_SYMBOL (mei_bus_get_clientdata); +Did you really mean to export an inline function?
Quoted 2 lines
Quoted 1 line
Can you make this a static inline function in a header file instead?
Sure, will do.
Actually, I'd like to keep the mei_bus_client structure opaque to the MEI bus drivers. So I'll still export those symbols without inlining them.
Cheers,
Samuel.
Signature · 2 lines
Intel Open Source Technology Centre http://oss.intel.com/
Arnd Bergmann
maintainerRe: [char-misc-next 07/11] mei: bus: Implement bus driver data setter/getter
Quoted 14 lines · Samuel Ortiz
Quoted 12 lines · Samuel Ortiz
Quoted 11 lines · Arnd Bergmann
drivers/misc/mei/bus.c int mei_bus_register_event_cb(struct mei_bus_client *client, · line 456
Quoted 7 lines · Tomas Winkler
drivers/misc/mei/bus.c +6
-0+inline void *mei_bus_get_clientdata (const struct mei_bus_client *client) +{ +return dev_get_drvdata (&client->dev); +} +EXPORT_SYMBOL (mei_bus_get_clientdata); +Did you really mean to export an inline function?
Quoted 4 lines
Quoted 2 lines
Quoted 1 line
Can you make this a static inline function in a header file instead?
Sure, will do.
Actually, I'd like to keep the
mei_bus_clientstructure opaque to the MEI bus drivers. So I'll still export those symbols without inlining them.
Ok, makes sense. I assume when you say "bus driver" you mean what other subsystems call a "device driver".
Arnd
Re: [char-misc-next 07/11] mei: bus: Implement bus driver data setter/getter
Hi Arnd,
Quoted 15 lines · Arnd Bergmann
Quoted 14 lines · Samuel Ortiz
Quoted 12 lines · Samuel Ortiz
Quoted 11 lines · Arnd Bergmann
drivers/misc/mei/bus.c int mei_bus_register_event_cb(struct mei_bus_client *client, · line 456
Quoted 7 lines · Tomas Winkler
drivers/misc/mei/bus.c +6
-0+inline void *mei_bus_get_clientdata (const struct mei_bus_client *client) +{ +return dev_get_drvdata (&client->dev); +} +EXPORT_SYMBOL (mei_bus_get_clientdata); +Did you really mean to export an inline function?
Quoted 7 lines
Quoted 4 lines
Quoted 2 lines
Quoted 1 line
Can you make this a static inline function in a header file instead?
Sure, will do.
Actually, I'd like to keep the
mei_bus_clientstructure opaque to the MEI bus drivers. So I'll still export those symbols without inlining them.Ok, makes sense. I assume when you say "bus driver" you mean what other subsystems call a "device driver".
That's correct, yes.
Cheers,
Samuel.
Signature · 2 lines
Intel Open Source Technology Centre http://oss.intel.com/
Greg KH
maintainerRe: [char-misc-next 07/11] mei: bus: Implement bus driver data setter/getter
Quoted 18 lines · Samuel Ortiz
Hi Arnd,
Quoted 15 lines · Arnd Bergmann
Quoted 14 lines · Samuel Ortiz
Quoted 12 lines · Samuel Ortiz
Quoted 11 lines · Arnd Bergmann
drivers/misc/mei/bus.c int mei_bus_register_event_cb(struct mei_bus_client *client, · line 456
Quoted 7 lines · Tomas Winkler
drivers/misc/mei/bus.c +6
-0+inline void *mei_bus_get_clientdata (const struct mei_bus_client *client) +{ +return dev_get_drvdata (&client->dev); +} +EXPORT_SYMBOL (mei_bus_get_clientdata); +Did you really mean to export an inline function?
Quoted 8 lines
Quoted 7 lines
Quoted 4 lines
Quoted 2 lines
Quoted 1 line
Can you make this a static inline function in a header file instead?
Sure, will do.
Actually, I'd like to keep the
mei_bus_clientstructure opaque to the MEI bus drivers. So I'll still export those symbols without inlining them.Ok, makes sense. I assume when you say "bus driver" you mean what other subsystems call a "device driver".
That's correct, yes.
Thanks for clearing that up. So, in your next revision, can you use the "proper" names for everything so that we all can understand the code easier as it will follow the normal standards for kernel driver model interaction?
thanks,
greg k-h
Re: [char-misc-next 07/11] mei: bus: Implement bus driver data setter/getter
Quoted 24 lines · Greg KH
Quoted 23 lines · Samuel Ortiz
Hi Arnd,
Quoted 20 lines · Arnd Bergmann
Quoted 18 lines · Samuel Ortiz
Quoted 12 lines · Samuel Ortiz
Quoted 11 lines · Arnd Bergmann
Review
Quoted 7 lines · Tomas Winkler
(quoted diff) +6
-0+inline void *mei_bus_get_clientdata(const struct mei_bus_client *client) +{ + return dev_get_drvdata(&client->dev); +} +EXPORT_SYMBOL(mei_bus_get_clientdata); +Did you really mean to export an inline function?
Quoted 2 lines
Quoted 1 line
Can you make this a static inline function in a header file instead?
Sure, will do.
Actually, I'd like to keep the
mei_bus_clientstructure opaque to the MEI bus drivers. So I'll still export those symbols without inlining them.
Quoted 8 lines
Quoted 3 lines
Quoted 2 lines
Ok, makes sense. I assume when you say "bus driver" you mean what other subsystems call a "device driver".
That's correct, yes.
Thanks for clearing that up. So, in your next revision, can you use the "proper" names for everything so that we all can understand the code easier as it will follow the normal standards for kernel driver model interaction?
I will certainly do that, yes.
Cheers,
Samuel.
Signature · 2 lines
Intel Open Source Technology Centre http://oss.intel.com/
PATCH 8/11 mei: nfc: Initial nfc implementation
In reply to Tomas WinklerFrom: Samuel Ortiz <[email protected]>
NFC ME client is exported through mei bus to be consumed by the NFC subsystem.
NFC is represented by two mei clients: An info one and the actual NFC one. In order for correct connection we first need to retrieve the firmware information from the info client.
--- drivers/misc/mei/Kconfig | 7 ++ drivers/misc/mei/Makefile | 1 + drivers/misc/mei/client.c | 3 + drivers/misc/mei/mei_dev.h | 28 ++++++ drivers/misc/mei/nfc.c | 210 ++++++++++++++++++++++++++++++++++++++++++++ drivers/misc/mei/nfc.h | 122 +++++++++++++++++++++++++ drivers/misc/mei/pci-me.c | 2 + 7 files changed, 373 insertions(+), 0 deletions(-) create mode 100644 drivers/misc/mei/nfc.c create mode 100644 drivers/misc/mei/nfc.h
drivers/misc/mei/Kconfig +7 -0
@@ -35,3 +35,10 @@ config INTEL_MEI_ME 82Q33 Express 82X38/X48 Express +config INTEL_MEI_BUS_NFC + bool "MEI bus NFC support" + depends on INTEL_MEI + help + When selecting this option the ME NFC device will be added to the + MEI bus. This is needed by the NFC kernel subsystem for sending and + receiving HCI frames to and from the NFC device.
drivers/misc/mei/Makefile +1 -0
@@ -11,5 +11,6 @@ mei-objs += main.o mei-objs += amthif.o mei-objs += wd.o mei-objs += bus.o +mei-$(CONFIG_INTEL_MEI_BUS_NFC) += nfc.o mei-$(CONFIG_INTEL_MEI_ME) += pci-me.o mei-$(CONFIG_INTEL_MEI_ME) += hw-me.o
drivers/misc/mei/client.c +3 -0
@@ -358,6 +358,9 @@ void mei_host_client_init(struct work_struct *work)mei_amthif_host_init (dev); else if (!uuid_le_cmp(client_props->protocol_name, mei_wd_guid))mei_wd_host_init (dev); + else if (!uuid_le_cmp(client_props->protocol_name, mei_nfc_guid)) +mei_nfc_host_init (dev); + } dev->dev_state =MEI_DEV_ENABLED ;
drivers/misc/mei/mei_dev.h +28 -0
@@ -475,6 +475,34 @@ int mei_amthif_irq_read_message(struct mei_cl_cb *complete_list, struct mei_device *dev, struct mei_msg_hdr *mei_hdr);int mei_amthif_irq_read (struct mei_device *dev,s32 *slots); +#ifdef CONFIG_INTEL_MEI_BUS_NFC +/* +* NFC functions +*/ +int mei_nfc_host_init (struct mei_device *dev); +void mei_nfc_host_exit (void ); + +/* +* NFC Client UUID +*/ +extern const uuid_le mei_nfc_guid; + +#else + +static inline int mei_nfc_host_init (struct mei_device *dev) +{ +return 0 ; +} + +static inline void mei_nfc_host_exit (void ) +{ +return ; +} + +static const uuid_le mei_nfc_guid =UUID_LE (0x0bb17a78 ,0x2a8e ,0x4c50 , +0x94 ,0xd4 ,0x50 ,0x26 , +0x67 ,0x23 ,0x77 ,0x5c ); +#endif int mei_wd_send (struct mei_device *dev);int mei_wd_stop (struct mei_device *dev);
drivers/misc/mei/nfc.c new file +210 -0
210 lines not on this page · raw message
drivers/misc/mei/nfc.h new file +122 -0
122 lines not on this page · raw message
drivers/misc/mei/pci-me.c +2 -0
@@ -283,6 +283,8 @@ static void mei_remove(struct pci_dev *pdev) dev->open_handle_count--;mei_cl_unlink (&dev->iamthif_cl); +mei_nfc_host_exit (); + dev->iamthif_current_cb =NULL ; dev->me_clients_num =0 ;
Signature · 1 line
1.7.4.4
Arnd Bergmann
maintainerRe: [char-misc-next 08/11] mei: nfc: Initial nfc implementation
Quoted 6 lines · Tomas Winkler
From: Samuel Ortiz <[email protected]>
NFC ME client is exported through mei bus to be consumed by the NFC subsystem.
Quoted 6 lines
NFC is represented by two mei clients: An info one and the actual NFC one. In order for correct connection we first need to retrieve the firmware information from the info client.
Signed-off-by Samuel Ortiz <[email protected]>Signed-off-by Tomas Winkler <[email protected]>
Shouldn't this be moved to the drivers/nfc directory? Generally speaking, all drivers nowadays tend to live in the directories of the subsystems they are implementing support for, not the subsystems that they are implemented with.
Arnd
Re: [char-misc-next 08/11] mei: nfc: Initial nfc implementation
Hi Arnd,
Quoted 11 lines · Arnd Bergmann
Quoted 10 lines · Tomas Winkler
From: Samuel Ortiz <[email protected]>
NFC ME client is exported through mei bus to be consumed by the NFC subsystem.
NFC is represented by two mei clients: An info one and the actual NFC one. In order for correct connection we first need to retrieve the firmware information from the info client.
Quoted 7 lines
Quoted 2 lines
Signed-off-by Samuel Ortiz <[email protected]>Signed-off-by Tomas Winkler <[email protected]>Shouldn't this be moved to the
drivers/nfcdirectory? Generally speaking, all drivers nowadays tend to live in the directories of the subsystems they are implementing support for, not the subsystems that they are implemented with.
This is a bit on the edge: This part of the MEI bus code doesn't implement support for any NFC chipset in particular but for the MEI specific commands to send and receive NFC HCI payloads.
So the drivers under driver/nfc call into the MEI bus I/O API which then call this code to encapsulate the HCI payload into MEI (HECI) commands. The microread driver in my nfc-next git.kernel.org tree uses the MEI bus API for example.
Moreover this code actually talks to the ME at init time to understand which exact NFC chipset is living behind the ME and add the correct device string to the bus. So it really needs to access the MEI internal API to do so.
Cheers,
Samuel.
Signature · 2 lines
Intel Open Source Technology Centre http://oss.intel.com/
PATCH 9/11 mei: nfc: Connect also the regular ME client
In reply to Tomas WinklerFrom: Samuel Ortiz <[email protected]>
After receiving the NFC interface version, IVN and radio type, we can connect to the the actual nfc me client and send the initialization (nfc connect) message.
--- drivers/misc/mei/nfc.c | 128 ++++++++++++++++++++++++++++++++++++++++++++++-- 1 files changed, 123 insertions(+), 5 deletions(-)
drivers/misc/mei/nfc.c +123 -5
@@ -29,6 +29,7 @@/** mei_nfc_bdev - nfc mei bus device * +* @cl: nfc info host client * @cl_info: nfc info host client * @init_work: perform connection to the info client * @fw_ivn: NFC Intervace Version Number
@@ -36,6 +37,7 @@* @radio_type: NFC radio type */ struct mei_bus_dev_nfc { +struct mei_cl *cl;struct mei_cl *cl_info;struct work_struct init_work;u8 fw_ivn;
@@ -56,6 +58,12 @@ const uuid_le mei_nfc_info_guid = UUID_LE(0xd2de1625, 0x382d, 0x417d,static void mei_nfc_free (struct mei_bus_dev_nfc *bdev) { +if (bdev->cl) { +list_del (&bdev->cl->bus_client_link); +mei_cl_unlink (bdev->cl); +kfree (bdev->cl); + } +if (bdev->cl_info) {list_del (&bdev->cl_info->bus_client_link);mei_cl_unlink (bdev->cl_info);
@@ -63,6 +71,73 @@ static void mei_nfc_free(struct mei_bus_dev_nfc *bdev) } } +static int mei_nfc_connect (struct mei_bus_dev_nfc *bdev) +{ +struct mei_device *dev; +struct mei_cl *cl; +struct mei_nfc_cmd *cmd, *reply; +struct mei_nfc_connect *connect; +struct mei_nfc_connect_resp *connect_resp; +size_t connect_length, connect_resp_length; +int bytes_recv, ret; + + cl = bdev->cl; + dev = cl->dev; + + connect_length =sizeof (struct mei_nfc_cmd ) + +sizeof (struct mei_nfc_connect ); + + connect_resp_length =sizeof (struct mei_nfc_cmd ) + +sizeof (struct mei_nfc_connect_resp ); + + cmd =kzalloc (connect_length,GFP_KERNEL ); +if (!cmd) +return -ENOMEM ; + connect = (struct mei_nfc_connect *)cmd->data; + + reply =kzalloc (connect_resp_length,GFP_KERNEL ); +if (!reply) { +kfree (cmd); +return -ENOMEM ; + } + + connect_resp = (struct mei_nfc_connect_resp *)reply->data; + + cmd->command =MEI_NFC_CMD_MAINTENANCE ; + cmd->data_size =3 ; + cmd->sub_command =MEI_NFC_SUBCMD_CONNECT ; + connect->fw_ivn = bdev->fw_ivn; + connect->vendor_id = bdev->vendor_id; + + ret =mei_send (cl, (u8 *)cmd, connect_length); +if (ret <0 ) { +dev_err (&dev->pdev->dev,"Could not send connect cmd\n" ); + goto err; + } + + bytes_recv =mei_recv (cl, (u8 *)reply, connect_resp_length); +if (bytes_recv <0 ) { +dev_err (&dev->pdev->dev,"Could not read connect response\n" ); + ret = bytes_recv; + goto err; + } + +dev_info (&dev->pdev->dev,"IVN 0x%x Vendor ID 0x%x\n" , + connect_resp->fw_ivn, connect_resp->vendor_id); + +dev_info (&dev->pdev->dev,"ME FW %d.%d.%d.%d\n" , + connect_resp->me_major, connect_resp->me_minor, + connect_resp->me_hotfix, connect_resp->me_build); + + ret =0 ; + +err: +kfree (reply); +kfree (cmd); + +return ret; +} +static int mei_nfc_if_version (struct mei_bus_dev_nfc *bdev) {struct mei_device *dev;
@@ -118,12 +193,13 @@ static void mei_nfc_init(struct work_struct *work) {struct mei_device *dev;struct mei_bus_dev_nfc *bdev; -struct mei_cl *cl_info; +struct mei_cl *cl_info, *cl;int ret; bdev = container_of(work, struct mei_bus_dev_nfc, init_work); cl_info = bdev->cl_info; + cl = bdev->cl; dev = cl_info->dev;mutex_lock (&dev->device_lock);
@@ -149,6 +225,24 @@ static void mei_nfc_init(struct work_struct *work)"NFC MEI VERSION: IVN 0x%x Vendor ID 0x%x Type 0x%x\n" , bdev->fw_ivn, bdev->vendor_id, bdev->radio_type); +mutex_lock (&dev->device_lock); + +if (mei_cl_connect (cl,NULL ) <0 ) { +mutex_unlock (&dev->device_lock); +dev_err (&dev->pdev->dev, +"Could not connect to the NFC ME client" ); + + goto err; + } + +mutex_unlock (&dev->device_lock); + + ret =mei_nfc_connect (bdev); +if (ret <0 ) { +dev_err (&dev->pdev->dev,"Could not connect to NFC" ); +return ; + } +return ; err:
@@ -161,7 +255,7 @@ err:int mei_nfc_host_init (struct mei_device *dev) {struct mei_bus_dev_nfc *bdev = &nfc_bdev; -struct mei_cl *cl_info; +struct mei_cl *cl_info, *cl =NULL ;int i, ret;/* already initialzed */
@@ -169,14 +263,19 @@ int mei_nfc_host_init(struct mei_device *dev)return 0 ; cl_info =mei_cl_allocate (dev); -if (!cl_info) -return -ENOMEM ; + cl =mei_cl_allocate (dev); + +if (!cl || !cl_info) { + ret = -ENOMEM ; + goto err; + }/* check for valid client id */ i =mei_me_cl_by_uuid (dev, &mei_nfc_info_guid);if (i <0 ) {dev_info (&dev->pdev->dev,"nfc: failed to find the client\n" ); -return -ENOENT ; + ret = -ENOENT ; + goto err; } cl_info->me_client_id = dev->me_clients[i].client_id;
@@ -189,7 +288,26 @@ int mei_nfc_host_init(struct mei_device *dev)list_add_tail (&cl_info->bus_client_link, &dev->bus_client_list); +/* check for valid client id */ + i =mei_me_cl_by_uuid (dev, &mei_nfc_guid); +if (i <0 ) { +dev_info (&dev->pdev->dev,"nfc: failed to find the client\n" ); + ret = -ENOENT ; + goto err; + } + + cl->me_client_id = dev->me_clients[i].client_id; + + ret =mei_cl_link (cl,MEI_HOST_CLIENT_ID_ANY ); +if (ret) + goto err; + + cl->bus_client_uuid = mei_nfc_guid; + +list_add_tail (&cl->bus_client_link, &dev->bus_client_list); + bdev->cl_info = cl_info; + bdev->cl = cl;INIT_WORK (&bdev->init_work, mei_nfc_init);schedule_work (&bdev->init_work);
Signature · 1 line
1.7.4.4
PATCH 10/11 mei: nfc: Add NFC device to the MEI bus
In reply to Tomas WinklerFrom: Samuel Ortiz <[email protected]>
After building its bus name as a string based on its vendor id and radio type, we can add it to the bus.
--- drivers/misc/mei/nfc.c | 56 ++++++++++++++++++++++++++++++++++++++++++++++++ drivers/misc/mei/nfc.h | 6 +++++ 2 files changed, 62 insertions(+), 0 deletions(-)
drivers/misc/mei/nfc.c +56 -0
@@ -43,6 +43,8 @@ struct mei_bus_dev_nfc {u8 fw_ivn;u8 vendor_id;u8 radio_type; + +char *bus_name; };struct mei_bus_dev_nfc nfc_bdev;
@@ -71,6 +73,39 @@ static void mei_nfc_free(struct mei_bus_dev_nfc *bdev) } } +static int mei_nfc_build_bus_name (struct mei_bus_dev_nfc *bdev) +{ +struct mei_device *dev; + +if (!bdev->cl) +return -ENODEV ; + + dev = bdev->cl->dev; + +switch (bdev->vendor_id) { +case MEI_NFC_VENDOR_INSIDE : +switch (bdev->radio_type) { +case MEI_NFC_VENDOR_INSIDE_UREAD : + bdev->bus_name ="microread" ; +return 0 ; + +default : +dev_err (&dev->pdev->dev,"Unknow radio type 0x%x\n" , + bdev->radio_type); + +return -EINVAL ; + } + +default : +dev_err (&dev->pdev->dev,"Unknow vendor ID 0x%x\n" , + bdev->vendor_id); + +return -EINVAL ; + } + +return 0 ; +} +static int mei_nfc_connect (struct mei_bus_dev_nfc *bdev) {struct mei_device *dev;
@@ -192,6 +227,7 @@ err:static void mei_nfc_init (struct work_struct *work) {struct mei_device *dev; +struct mei_bus_client *bus_client;struct mei_bus_dev_nfc *bdev;struct mei_cl *cl_info, *cl;int ret;
@@ -243,6 +279,23 @@ static void mei_nfc_init(struct work_struct *work)return ; } +if (mei_nfc_build_bus_name (bdev) <0 ) { +dev_err (&dev->pdev->dev, +"Could not build the bus ID name\n" ); +return ; + } + + bus_client =mei_add_device (dev, mei_nfc_guid, + bdev->bus_name); +if (!bus_client) { +dev_err (&dev->pdev->dev, +"Could not add the NFC device to the MEI bus\n" ); + + goto err; + } + + bus_client->priv_data = bdev; +return ; err:
@@ -324,5 +377,8 @@ void mei_nfc_host_exit(void) {struct mei_bus_dev_nfc *bdev = &nfc_bdev; +if (bdev->cl && bdev->cl->client) +mei_remove_device (bdev->cl->client); +mei_nfc_free (bdev); }
drivers/misc/mei/nfc.h +6 -0
@@ -119,4 +119,10 @@ struct mei_nfc_connect_resp {#define MEI_NFC_SUBCMD_CONNECT 0x00#define MEI_NFC_SUBCMD_IF_VERSION 0x01 +/* Vendors */ +#define MEI_NFC_VENDOR_INSIDE 0x00 + +/* Radio types */ +#define MEI_NFC_VENDOR_INSIDE_UREAD 0x00 +#endif /* _MEI_NFC_H */
Signature · 1 line
1.7.4.4
PATCH 11/11 mei: nfc: Implement MEI bus IO ops
In reply to Tomas WinklerFrom: Samuel Ortiz <[email protected]>
The send ops for NFC builds the command header, updates the request id and then waits for an ACK.
The recv ops check if it receives data or an ACK and in the latter case wakes the send ops up.
--- drivers/misc/mei/nfc.c | 76 ++++++++++++++++++++++++++++++++++++++++++++++++ drivers/misc/mei/nfc.h | 13 ++++++++ 2 files changed, 89 insertions(+), 0 deletions(-)
drivers/misc/mei/nfc.c +76 -0
@@ -15,6 +15,7 @@*/ #include <linux/kernel.h> +#include <linux/sched.h> #include <linux/module.h> #include <linux/moduleparam.h> #include <linux/device.h>
@@ -40,11 +41,15 @@ struct mei_bus_dev_nfc {struct mei_cl *cl;struct mei_cl *cl_info;struct work_struct init_work; +wait_queue_head_t send_wq;u8 fw_ivn;u8 vendor_id;u8 radio_type;char *bus_name; + +u16 req_id; +u16 recv_req_id; };struct mei_bus_dev_nfc nfc_bdev;
@@ -224,6 +229,74 @@ err:return ret; } +static int mei_nfc_send (struct mei_bus_client *client,u8 *buf,size_t length) +{ +struct mei_device *dev; +struct mei_bus_dev_nfc *bdev; +struct mei_nfc_hci_hdr *hdr; +u8 *mei_buf; +int err; + + bdev = (struct mei_bus_dev_nfc *) client->priv_data; + dev = bdev->cl->dev; + + mei_buf =kzalloc (length +MEI_NFC_HEADER_SIZE ,GFP_KERNEL ); +if (!mei_buf) +return -ENOMEM ; + + hdr = (struct mei_nfc_hci_hdr *) mei_buf; + hdr->cmd =MEI_NFC_CMD_HCI_SEND ; + hdr->status =0 ; + hdr->req_id = bdev->req_id; + hdr->reserved =0 ; + hdr->data_size = length; + +memcpy (mei_buf +MEI_NFC_HEADER_SIZE , buf, length); + + err =mei_send (bdev->cl, mei_buf, length +MEI_NFC_HEADER_SIZE ); + +kfree (mei_buf); + +if (!wait_event_interruptible_timeout (bdev->send_wq, + bdev->recv_req_id == bdev->req_id,HZ )) { +dev_err (&dev->pdev->dev,"NFC MEI command timeout\n" ); + err = -ETIMEDOUT ; + }else { + bdev->req_id++; + } + +return err; +} + +static int mei_nfc_recv (struct mei_bus_client *client,u8 *buf,size_t length) +{ +struct mei_bus_dev_nfc *bdev; +struct mei_nfc_hci_hdr *hci_hdr; +int received_length; + + bdev = (struct mei_bus_dev_nfc *) client->priv_data; + + received_length =mei_recv (bdev->cl, buf, length); +if (received_length <0 ) +return received_length; + + hci_hdr = (struct mei_nfc_hci_hdr *) buf; + +if (hci_hdr->cmd ==MEI_NFC_CMD_HCI_SEND ) { + bdev->recv_req_id = hci_hdr->req_id; +wake_up (&bdev->send_wq); + +return 0 ; + } + +return received_length; +} + +static struct mei_bus_ops nfc_ops = { + .send = mei_nfc_send, + .recv = mei_nfc_recv, +}; +static void mei_nfc_init (struct work_struct *work) {struct mei_device *dev;
@@ -295,6 +368,7 @@ static void mei_nfc_init(struct work_struct *work) } bus_client->priv_data = bdev; + bus_client->ops = &nfc_ops;return ;
@@ -361,8 +435,10 @@ int mei_nfc_host_init(struct mei_device *dev) bdev->cl_info = cl_info; bdev->cl = cl; + bdev->req_id =1 ;INIT_WORK (&bdev->init_work, mei_nfc_init); +init_waitqueue_head (&bdev->send_wq);schedule_work (&bdev->init_work);return 0 ;
drivers/misc/mei/nfc.h +13 -0
@@ -114,11 +114,24 @@ struct mei_nfc_connect_resp {uint16_t me_build; } __packed; +struct mei_nfc_hci_hdr { +u8 cmd; +u8 status; +u16 req_id; +u32 reserved; +u16 data_size; +} __packed; +#define MEI_NFC_CMD_MAINTENANCE 0x00 +#define MEI_NFC_CMD_HCI_SEND 0x01 +#define MEI_NFC_CMD_HCI_RECV 0x02#define MEI_NFC_SUBCMD_CONNECT 0x00#define MEI_NFC_SUBCMD_IF_VERSION 0x01 +#define MEI_NFC_HEADER_SIZE 10 +#define MEI_NFC_MAX_HCI_PAYLOAD 300 +/* Vendors */ #define MEI_NFC_VENDOR_INSIDE 0x00
Signature · 1 line
1.7.4.4