i2c: Get rid of the legacy binding model
Subsystems (3)
cover i2c: Get rid of the legacy binding model
As the last few remaining legacy i2c drivers are being converted, we can finally get rid of the legacy binding model supporting code in i2c-core. Here come 6 patches which do exactly that:
[1/6] i2c: Kill client_register and client_unregister methods [2/6] i2c: Get rid of the legacy binding model [3/6] i2c: Drop i2c_probe function
[4/6] i2c: Merge i2c_attach_client into i2c_new_device [5/6] i2c: Kill is_newstyle_driver
[6/6] i2c: Kill the redundant client list
As one would expect, the size reduction is very significant:
Documentation/feature-removal-schedule.txt | 10 - Documentation/i2c/writing-clients | 16 - drivers/i2c/i2c-core.c | 366 +++-------------------------- include/linux/i2c.h | 57 +---- 4 files changed, 48 insertions(+), 401 deletions(-)
Quoted 1 line
From a binary perspective, "size" announces an 18% shrink.
There are two things which should still be investigated.
1* In i2c_device_uevent() there is the following piece of code:
/* by definition, legacy drivers can't hotplug */ if (dev->driver) return 0;
David, I presume this can go away?
2* Core locking should be checked. I am not completely sure what core_lock was supposed to protect, nor what it should protect now. I think it should protect at least i2c_adapter_idr, driver->detect() calls (which may add a device to the driver->clients list) and driver->attach_adapter() calls (for the same reason), although for the latter two, a per-driver lock may be more appropriate.
At the moment, we also hold the lock for all of i2c_register_adapter(), i2c_del_adapter() and i2c_del_driver(), but not i2c_register_driver(), which seems a little inconsistent. David, do you know if we need to serialize the calls to device_register() and device_unregister(), or if the driver core takes care of this for us?
Anything else needs to be protected?
Signature · 1 line
Jean Delvare
A message that is not in the archive (sent off-list or not yet received): [email protected]
PATCH 1/6 i2c: Kill client_register and client_unregister methods
These methods were useful in the legacy binding model but no longer in the new (standard) binding model. There are no users left so we can drop them.
--- Documentation/feature-removal-schedule.txt | 3 -- drivers/i2c/i2c-core.c | 30 ---------------------------- include/linux/i2c.h | 4 --- 3 files changed, 2 insertions(+), 35 deletions(-)
Documentation/feature-removal-schedule.txt +1 -2
@@ -354,8 +354,7 @@ Who: Krzysztof Piotr Oledzki <[email protected] --------------------------- -What: i2c_attach_client(), i2c_detach_client(), i2c_driver->detach_client(), - i2c_adapter->client_register(), i2c_adapter->client_unregister +What: i2c_attach_client(), i2c_detach_client(), i2c_driver->detach_client() When: 2.6.30 Check: i2c_attach_client i2c_detach_client Why: Deprecated by the new (standard) device driver binding model. Use
drivers/i2c/i2c-core.c +1 -29
@@ -309,14 +309,6 @@ void i2c_unregister_device(struct i2c_clreturn ; } -if (adapter->client_unregister) { -if (adapter->client_unregister (client)) { -dev_warn (&client->dev, -"client_unregister [%s] failed\n" , - client->name); - } - } -mutex_lock (&adapter->clist_lock);list_del (&client->list);mutex_unlock (&adapter->clist_lock);
@@ -867,14 +859,6 @@ int i2c_attach_client(struct i2c_clientdev_dbg (&adapter->dev,"client [%s] registered with bus id %s\n" , client->name,dev_name (&client->dev)); -if (adapter->client_register) { -if (adapter->client_register (client)) { -dev_dbg (&adapter->dev,"client_register " -"failed for client [%s] at 0x%02x\n" , - client->name, client->addr); - } - } -return 0 ; out_err:
@@ -887,17 +871,6 @@ EXPORT_SYMBOL(i2c_attach_client);int i2c_detach_client (struct i2c_client *client) {struct i2c_adapter *adapter = client->adapter; -int res =0 ; - -if (adapter->client_unregister) { - res = adapter->client_unregister (client); -if (res) { -dev_err (&client->dev, -"client_unregister [%s] failed, " -"client not detached\n" , client->name); - goto out; - } - }mutex_lock (&adapter->clist_lock);list_del (&client->list);
@@ -907,8 +880,7 @@ int i2c_detach_client(struct i2c_clientdevice_unregister (&client->dev);wait_for_completion (&client->released); - out: -return res; +return 0 ; }EXPORT_SYMBOL (i2c_detach_client);
include/linux/i2c.h +0 -4
@@ -352,10 +352,6 @@ struct i2c_adapter {const struct i2c_algorithm *algo;/* the algorithm to access the bus */ void *algo_data; -/* --- administration stuff. */ -int (*client_register)(struct i2c_client *)__deprecated ; -int (*client_unregister)(struct i2c_client *)__deprecated ; -/* data fields that are valid for all devices */ u8 level;/* nesting level for lockdep */ struct mutex bus_lock;
Signature · 1 line
Jean Delvare
PATCH 2/6 i2c: Get rid of the legacy binding model
In reply to a message not in the archiveWe converted all the legacy i2c drivers so we can finally get rid of the legacy binding model. Hooray!
--- Documentation/feature-removal-schedule.txt | 9 --- Documentation/i2c/writing-clients | 16 +---- drivers/i2c/i2c-core.c | 81 +--------------------------- include/linux/i2c.h | 39 +++---------- 4 files changed, 18 insertions(+), 127 deletions(-)
Documentation/feature-removal-schedule.txt +0 -9
@@ -354,15 +354,6 @@ Who: Krzysztof Piotr Oledzki <[email protected] --------------------------- -What: i2c_attach_client(), i2c_detach_client(), i2c_driver->detach_client() -When: 2.6.30 -Check: i2c_attach_client i2c_detach_client -Why: Deprecated by the new (standard) device driver binding model. Use - i2c_driver->probe() and ->remove() instead. -Who: Jean Delvare <[email protected]> - ---------------------------- - What: fscher and fscpos drivers When: June 2009 Why: Deprecated by the new fschmd driver.
drivers/i2c/i2c-core.c +5 -76
144 lines not on this page · raw message
include/linux/i2c.h +10 -29
@@ -100,9 +100,8 @@ extern s32 i2c_smbus_write_i2c_block_dat* @class: What kind of i2c device we instantiate (for detect) * @attach_adapter: Callback for bus addition (for legacy drivers) * @detach_adapter: Callback for bus removal (for legacy drivers) -* @detach_client: Callback for device removal (for legacy drivers) -* @probe: Callback for device binding (new-style drivers) -* @remove: Callback for device unbinding (new-style drivers) +* @probe: Callback for device binding +* @remove: Callback for device unbinding * @shutdown: Callback for device shutdown * @suspend: Callback for device suspend * @resume: Callback for device resume
@@ -137,26 +136,14 @@ struct i2c_driver {int id;unsigned int class; -/* Notifies the driver that a new bus has appeared. This routine -* can be used by the driver to test if the bus meets its conditions -* & seek for the presence of the chip(s) it supports. If found, it -* registers the client(s) that are on the bus to the i2c admin. via -* i2c_attach_client. (LEGACY I2C DRIVERS ONLY) +/* Notifies the driver that a new bus has appeared or is about to be +* removed. You should avoid using this if you can, it will probably +* be removed in a near future. */ int (*attach_adapter)(struct i2c_adapter *);int (*detach_adapter)(struct i2c_adapter *); -/* tells the driver that a client is about to be deleted & gives it -* the chance to remove its private data. Also, if the client struct -* has been dynamically allocated by the driver in the function above, -* it must be freed here. (LEGACY I2C DRIVERS ONLY) -*/ -int (*detach_client)(struct i2c_client *)__deprecated ; - -/* Standard driver model interfaces, for "new style" i2c drivers. -* With the driver model, device enumeration is NEVER done by drivers; -* it's done by infrastructure. (NEW STYLE DRIVERS ONLY) -*/ +/* Standard driver model interfaces */ int (*probe)(struct i2c_client *,const struct i2c_device_id *);int (*remove)(struct i2c_client *);
@@ -248,11 +235,10 @@ static inline void i2c_set_clientdata(st* that, such as chip type, configuration, associated IRQ, and so on. * * i2c_board_info is used to build tables of information listing I2C devices -* that are present. This information is used to grow the driver model tree -* for "new style" I2C drivers. For mainboards this is done statically using -* i2c_register_board_info(); bus numbers identify adapters that aren't -* yet available. For add-on boards, i2c_new_device() does this dynamically -* with the adapter already known. +* that are present. This information is used to grow the driver model tree. +* For mainboards this is done statically using i2c_register_board_info(); +* bus numbers identify adapters that aren't yet available. For add-on boards, +* i2c_new_device() does this dynamically with the adapter already known. */ struct i2c_board_info {char type[I2C_NAME_SIZE ];
@@ -425,11 +411,6 @@ static inline int i2c_add_driver(structreturn i2c_register_driver (THIS_MODULE , driver); } -/* These are deprecated, your driver should use the standard .probe() -* and .remove() methods instead. */ -extern int __deprecated i2c_attach_client (struct i2c_client *); -extern int __deprecated i2c_detach_client (struct i2c_client *); -extern struct i2c_client *i2c_use_client (struct i2c_client *client);extern void i2c_release_client (struct i2c_client *client);
Documentation/i2c/writing-clients +3 -13
@@ -126,19 +126,9 @@ different) configuration information, as that can't be distinguished by protocol probing, or which need some board specific information to operate correctly. -Accordingly, the I2C stack now has two models for associating I2C devices -with their drivers: the original "legacy" model, and a newer one that's -fully compatible with the Linux 2.6 driver model. These models do not mix, -since the "legacy" model requires drivers to create "i2c_client" device -objects after SMBus style probing, while the Linux driver model expects -drivers to be given such device objects in their probe() routines. -The legacy model is deprecated now and will soon be removed, so we no -longer document it here. - - -Standard Driver Model Binding ("New Style") -------------------------------------------- +Device/Driver Binding +--------------------- System infrastructure, typically board-specific initialization code or boot firmware, reports what I2C devices exist. For example, there may be
@@ -201,7 +191,7 @@ a given I2C bus. This is for example th devices on a PC's SMBus. In that case, you may want to let your driver detect supported devices automatically. This is how the legacy model was working, and is now available as an extension to the standard -driver model (so that we can finally get rid of the legacy model.) +driver model. You simply have to define a detect callback which will attempt to identify supported devices (returning 0 for supported ones and -ENODEV
Signature · 1 line
Jean Delvare
A message that is not in the archive (sent off-list or not yet received): [email protected]
include/linux/i2c.h extern s32 i2c_smbus_write_i2c_block_dat · line 102
Quoted 4 lines · Jean Delvare
include/linux/i2c.h +0
-0@@ -100,9 +100,8 @@ extern s32 i2c_smbus_write_i2c_block_dat * @class: What kind of i2c device we instantiate (for detect) * @attach_adapter: Callback for bus addition (for legacy drivers) * @detach_adapter: Callback for bus removal (for legacy drivers)
Remnants of the legacy model still remain ...
Quoted 8 lines
include/linux/i2c.h +2
-3-* @detach_client: Callback for device removal (for legacy drivers) -* @probe: Callback for device binding (new-style drivers) -* @remove: Callback for device unbinding (new-style drivers) +* @probe: Callback for device binding +* @remove: Callback for device unbinding * @shutdown: Callback for device shutdown * @suspend: Callback for device suspend * @resume: Callback for device resume
A message that is not in the archive (sent off-list or not yet received): 200905021203.43974.david-b-yBeKhBN/[email protected]
Quoted 7 lines · David Brownell
include/linux/i2c.h extern s32 i2c_smbus_write_i2c_block_dat · line 102
Quoted 4 lines · Jean Delvare
include/linux/i2c.h +0
-0@@ -100,9 +100,8 @@ extern s32 i2c_smbus_write_i2c_block_dat * @class: What kind of i2c device we instantiate (for detect) * @attach_adapter: Callback for bus addition (for legacy drivers) * @detach_adapter: Callback for bus removal (for legacy drivers)Remnants of the legacy model still remain ...
Depends on what exactly you call "the legacy model". As long as the lifetime of the devices is those of the standard model, I don't much care much about the rest.
At this point there are still a few drivers using attach_adapter, so we can't get rid of it: 9 macintosh drivers, and i2c-dev. The fix for the macintosh drivers would be to convert the powermac to fixed bus numbers. This is probably not very difficult, but I'd rather let powerpc people take care of that.
For i2c-dev, I just don't know. We could merge it into i2c-core. Or we could split the notification mechanism out of i2c_driver. Or maybe we can leverage the driver core notification mechanism and get rid of ours.
Either way this is probably not the highest priority change at this point in time. I'd rather work on the sysfs replacement for I2C_CLIENT_INSMOD* macros, the multiplexing support and the conversion of i2c-adapter to bus devices first.
Quoted 8 lines
Quoted 8 lines
include/linux/i2c.h +2
-3-* @detach_client: Callback for device removal (for legacy drivers) -* @probe: Callback for device binding (new-style drivers) -* @remove: Callback for device unbinding (new-style drivers) +* @probe: Callback for device binding +* @remove: Callback for device unbinding * @shutdown: Callback for device shutdown * @suspend: Callback for device suspend * @resume: Callback for device resume
Signature · 1 line
Jean Delvare
PATCH 3/6 i2c: Drop i2c_probe function
In reply to a message not in the archiveThe legacy i2c_probe() function has no users left, get rid of it.
--- drivers/i2c/i2c-core.c | 137 ------------------------------------------------ include/linux/i2c.h | 8 -- 2 files changed, 145 deletions(-)
drivers/i2c/i2c-core.c +0 -137
@@ -1042,144 +1042,7 @@ EXPORT_SYMBOL(i2c_master_recv);* Will not work for 10-bit addresses! * ---------------------------------------------------- */ -static int i2c_probe_address (struct i2c_adapter *adapter,int addr,int kind, -int (*found_proc) (struct i2c_adapter *,int ,int )) -{ -int err; - -/* Make sure the address is valid */ -if (addr <0x03 || addr >0x77 ) { -dev_warn (&adapter->dev,"Invalid probe address 0x%02x\n" , - addr); -return -EINVAL ; - } - -/* Skip if already in use */ -if (i2c_check_addr (adapter, addr)) -return 0 ; - -/* Make sure there is something at this address, unless forced */ -if (kind <0 ) { -if (i2c_smbus_xfer (adapter, addr,0 ,0 ,0 , -I2C_SMBUS_QUICK ,NULL ) <0 ) -return 0 ; - -/* prevent 24RF08 corruption */ -if ((addr & ~0x0f ) ==0x50 ) -i2c_smbus_xfer (adapter, addr,0 ,0 ,0 , -I2C_SMBUS_QUICK ,NULL ); - } - -/* Finally call the custom detection function */ - err =found_proc (adapter, addr, kind); -/* -ENODEV can be returned if there is a chip at the given address -but it isn't supported by this chip driver. We catch it here as -this isn't an error. */ -if (err == -ENODEV ) - err =0 ; - -if (err) -dev_warn (&adapter->dev,"Client creation failed at 0x%x (%d)\n" , - addr, err); -return err; -} - -int i2c_probe (struct i2c_adapter *adapter, -const struct i2c_client_address_data *address_data, -int (*found_proc) (struct i2c_adapter *,int ,int )) -{ -int i, err; -int adap_id =i2c_adapter_id (adapter); - -/* Force entries are done first, and are not affected by ignore -entries */ -if (address_data->forces) { -const unsigned short *const *forces = address_data->forces; -int kind; - -for (kind =0 ; forces[kind]; kind++) { -for (i =0 ; forces[kind][i] !=I2C_CLIENT_END ; - i +=2 ) { -if (forces[kind][i] == adap_id - || forces[kind][i] ==ANY_I2C_BUS ) { -dev_dbg (&adapter->dev,"found force " -"parameter for adapter %d, " -"addr 0x%02x, kind %d\n" , - adap_id, forces[kind][i +1 ], - kind); - err =i2c_probe_address (adapter, - forces[kind][i +1 ], - kind, found_proc); -if (err) -return err; - } - } - } - } - -/* Stop here if we can't use SMBUS_QUICK */ -if (!i2c_check_functionality (adapter,I2C_FUNC_SMBUS_QUICK )) { -if (address_data->probe[0 ] ==I2C_CLIENT_END - && address_data->normal_i2c[0 ] ==I2C_CLIENT_END ) -return 0 ; - -dev_dbg (&adapter->dev,"SMBus Quick command not supported, " -"can't probe for chips\n" ); -return -EOPNOTSUPP ; - } - -/* Probe entries are done second, and are not affected by ignore -entries either */ -for (i =0 ; address_data->probe[i] !=I2C_CLIENT_END ; i +=2 ) { -if (address_data->probe[i] == adap_id - || address_data->probe[i] ==ANY_I2C_BUS ) { -dev_dbg (&adapter->dev,"found probe parameter for " -"adapter %d, addr 0x%02x\n" , adap_id, - address_data->probe[i +1 ]); - err =i2c_probe_address (adapter, - address_data->probe[i +1 ], --1 , found_proc); -if (err) -return err; - } - } - -/* Normal entries are done last, unless shadowed by an ignore entry */ -for (i =0 ; address_data->normal_i2c[i] !=I2C_CLIENT_END ; i +=1 ) { -int j, ignore; - - ignore =0 ; -for (j =0 ; address_data->ignore[j] !=I2C_CLIENT_END ; - j +=2 ) { -if ((address_data->ignore[j] == adap_id || - address_data->ignore[j] ==ANY_I2C_BUS ) - && address_data->ignore[j +1 ] - == address_data->normal_i2c[i]) { -dev_dbg (&adapter->dev,"found ignore " -"parameter for adapter %d, " -"addr 0x%02x\n" , adap_id, - address_data->ignore[j +1 ]); - ignore =1 ; -break ; - } - } -if (ignore) -continue ; - -dev_dbg (&adapter->dev,"found normal entry for adapter %d, " -"addr 0x%02x\n" , adap_id, - address_data->normal_i2c[i]); - err =i2c_probe_address (adapter, address_data->normal_i2c[i], --1 , found_proc); -if (err) -return err; - } - -return 0 ; -} -EXPORT_SYMBOL (i2c_probe); -/* Separate detection function for new-style drivers */ static int i2c_detect_address (struct i2c_client *temp_client,int kind,struct i2c_driver *driver) {
include/linux/i2c.h +0 -8
@@ -419,14 +419,6 @@ extern void i2c_release_client(struct i2extern void i2c_clients_command (struct i2c_adapter *adap,unsigned int cmd,void *arg); -/* Detect function. It iterates over all possible addresses itself. -* It will only call found_proc if some client is connected at the -* specific address (unless a 'force' matched); -*/ -extern int i2c_probe (struct i2c_adapter *adapter, -const struct i2c_client_address_data *address_data, -int (*found_proc) (struct i2c_adapter *,int ,int )); -extern struct i2c_adapter *i2c_get_adapter (int id);extern void i2c_put_adapter (struct i2c_adapter *adap);
Signature · 1 line
Jean Delvare
PATCH 4/6 i2c: Merge i2c_attach_client into i2c_new_device
In reply to a message not in the archiveNow that i2c_attach_client is no longer exported, it doesn't need to be a separate function. Merge it into its only user, i2c_new_device.
--- drivers/i2c/i2c-core.c | 100 +++++++++++++++++++----------------------------- 1 file changed, 41 insertions(+), 59 deletions(-)
drivers/i2c/i2c-core.c +41 -59
@@ -44,7 +44,7 @@ static DEFINE_IDR(i2c_adapter_idr);#define is_newstyle_driver (d) ((d)->probe || (d)->remove || (d)->detect) -static int i2c_attach_client (struct i2c_client *client); +static int i2c_check_addr (struct i2c_adapter *adapter,int addr);static int i2c_detect (struct i2c_adapter *adapter,struct i2c_driver *driver);/* ------------------------------------------------------------------------- */
@@ -242,15 +242,17 @@ EXPORT_SYMBOL(i2c_verify_client);/** -* i2c_new_device - instantiate an i2c device for use with a new style driver +* i2c_new_device - instantiate an i2c device * @adap: the adapter managing the device * @info: describes one I2C device; bus_num is ignored * Context: can sleep * -* Create a device to work with a new style i2c driver, where binding is -* handled through driver model probe()/remove() methods. This call is not -* appropriate for use by mainboad initialization logic, which usually runs -* during an arch_initcall() long before any i2c_adapter could exist. +* Create an i2c device. Binding is handled through driver model +* probe()/remove() methods. A driver may be bound to this device when we +* return from this function, or any later moment (e.g. maybe hotplugging will +* load the driver module). This call is not appropriate for use by mainboard +* initialization logic, which usually runs during an arch_initcall() long +* before any i2c_adapter could exist. * * This returns the new i2c client, which may be saved for later use with * i2c_unregister_device(); or NULL to indicate an error.
@@ -278,17 +280,40 @@ i2c_new_device(struct i2c_adapter *adap,strlcpy (client->name, info->type,sizeof (client->name)); -/* a new style driver may be bound to this device when we -* return from this function, or any later moment (e.g. maybe -* hotplugging will load the driver module). and the device -* refcount model is the standard driver model one. -*/ - status =i2c_attach_client (client); -if (status <0 ) { -kfree (client); - client =NULL ; - } +/* Check for address business */ + status =i2c_check_addr (adap, client->addr); +if (status) + goto out_err; + + client->dev.parent = &client->adapter->dev; + client->dev.bus = &i2c_bus_type; + +if (client->driver && !is_newstyle_driver (client->driver)) { + client->dev.release = i2c_client_release; +dev_set_uevent_suppress (&client->dev,1 ); + }else + client->dev.release = i2c_client_dev_release; + +dev_set_name (&client->dev,"%d-%04x" ,i2c_adapter_id (adap), + client->addr); + status =device_register (&client->dev); +if (status) + goto out_err; + +mutex_lock (&adap->clist_lock); +list_add_tail (&client->list, &adap->clients); +mutex_unlock (&adap->clist_lock); + +dev_dbg (&adap->dev,"client [%s] registered with bus id %s\n" , + client->name,dev_name (&client->dev)); +return client; + +out_err: +dev_err (&adap->dev,"Failed to register i2c client %s at 0x%02x " +"(%d)\n" , client->name, client->addr, status); +kfree (client); +return NULL ; }EXPORT_SYMBOL_GPL (i2c_new_device);
@@ -770,49 +795,6 @@ static int i2c_check_addr(struct i2c_adareturn device_for_each_child (&adapter->dev, &addr, __i2c_check_addr); } -static int i2c_attach_client (struct i2c_client *client) -{ -struct i2c_adapter *adapter = client->adapter; -int res; - -/* Check for address business */ - res =i2c_check_addr (adapter, client->addr); -if (res) -return res; - - client->dev.parent = &client->adapter->dev; - client->dev.bus = &i2c_bus_type; - -if (client->driver) - client->dev.driver = &client->driver->driver; - -if (client->driver && !is_newstyle_driver (client->driver)) { - client->dev.release = i2c_client_release; -dev_set_uevent_suppress (&client->dev,1 ); - }else - client->dev.release = i2c_client_dev_release; - -dev_set_name (&client->dev,"%d-%04x" ,i2c_adapter_id (adapter), - client->addr); - res =device_register (&client->dev); -if (res) - goto out_err; - -mutex_lock (&adapter->clist_lock); -list_add_tail (&client->list, &adapter->clients); -mutex_unlock (&adapter->clist_lock); - -dev_dbg (&adapter->dev,"client [%s] registered with bus id %s\n" , - client->name,dev_name (&client->dev)); - -return 0 ; - -out_err: -dev_err (&adapter->dev,"Failed to attach i2c client %s at 0x%02x " -"(%d)\n" , client->name, client->addr, res); -return res; -} -/** * i2c_use_client - increments the reference count of the i2c client structure * @client: the client being referenced
Signature · 1 line
Jean Delvare
PATCH 5/6 i2c: Kill is_newstyle_driver
In reply to a message not in the archiveLegacy i2c drivers are gone, all drivers are new-style now, so there is no point to check.
--- drivers/i2c/i2c-core.c | 32 +------------------------------- include/linux/i2c.h | 2 -- 2 files changed, 1 insertion(+), 33 deletions(-)
drivers/i2c/i2c-core.c +1 -31
@@ -42,8 +42,6 @@static DEFINE_MUTEX (core_lock);static DEFINE_IDR (i2c_adapter_idr); -#define is_newstyle_driver (d) ((d)->probe || (d)->remove || (d)->detect) -static int i2c_check_addr (struct i2c_adapter *adapter,int addr);static int i2c_detect (struct i2c_adapter *adapter,struct i2c_driver *driver);
@@ -65,12 +63,6 @@ static int i2c_device_match(struct devicstruct i2c_client *client =to_i2c_client (dev);struct i2c_driver *driver =to_i2c_driver (drv); -/* make legacy i2c drivers bypass driver model probing entirely; -* such drivers scan each i2c adapter/bus themselves. -*/ -if (!is_newstyle_driver (driver)) -return 0 ; -/* match on an id table if there is one */ if (driver->id_table)return i2c_match_id (driver->id_table, client) !=NULL ;
@@ -177,12 +169,6 @@ static int i2c_device_resume(struct devireturn driver->resume (to_i2c_client (dev)); } -static void i2c_client_release (struct device *dev) -{ -struct i2c_client *client =to_i2c_client (dev); -complete (&client->released); -} -static void i2c_client_dev_release (struct device *dev) {kfree (to_i2c_client (dev));
@@ -287,12 +273,7 @@ i2c_new_device(struct i2c_adapter *adap, client->dev.parent = &client->adapter->dev; client->dev.bus = &i2c_bus_type; - -if (client->driver && !is_newstyle_driver (client->driver)) { - client->dev.release = i2c_client_release; -dev_set_uevent_suppress (&client->dev,1 ); - }else - client->dev.release = i2c_client_dev_release; + client->dev.release = i2c_client_dev_release;dev_set_name (&client->dev,"%d-%04x" ,i2c_adapter_id (adap), client->addr);
@@ -326,14 +307,6 @@ EXPORT_SYMBOL_GPL(i2c_new_device);void i2c_unregister_device (struct i2c_client *client) {struct i2c_adapter *adapter = client->adapter; -struct i2c_driver *driver = client->driver; - -if (driver && !is_newstyle_driver (driver)) { -dev_err (&client->dev,"can't unregister devices " -"with legacy drivers\n" ); -WARN_ON (1 ); -return ; - }mutex_lock (&adapter->clist_lock);list_del (&client->list);
@@ -746,9 +719,6 @@ static int __detach_adapter(struct devici2c_unregister_device (client); } -if (is_newstyle_driver (driver)) -return 0 ; -if (driver->detach_adapter) {if (driver->detach_adapter (adapter)) dev_err(&adapter->dev,
include/linux/i2c.h +0 -2
@@ -180,7 +180,6 @@ struct i2c_driver {* @irq: indicates the IRQ generated by this device (if any) * @list: list of active/busy clients (DEPRECATED) * @detected: member of an i2c_driver.clients list -* @released: used to synchronize client releases & detaches and references * * An i2c_client identifies a single device (i.e. chip) connected to an * i2c bus. The behaviour exposed to Linux is defined by the driver
@@ -198,7 +197,6 @@ struct i2c_client {int irq;/* irq issued by device */ struct list_head list;/* DEPRECATED */ struct list_head detected; -struct completion released; };#define to_i2c_client (d) container_of(d, struct i2c_client, dev)
Signature · 1 line
Jean Delvare
PATCH 6/6 i2c: Kill the redundant client list
In reply to a message not in the archiveWe used to maintain our own per-adapter list of i2c clients, but this is redundant with what the driver core does, and no longer needed. Just drop the redundant list.
--- drivers/i2c/i2c-core.c | 12 ------------ include/linux/i2c.h | 4 ---- 2 files changed, 16 deletions(-)
drivers/i2c/i2c-core.c +0 -12
@@ -281,10 +281,6 @@ i2c_new_device(struct i2c_adapter *adap,if (status) goto out_err; -mutex_lock (&adap->clist_lock); -list_add_tail (&client->list, &adap->clients); -mutex_unlock (&adap->clist_lock); -dev_dbg (&adap->dev,"client [%s] registered with bus id %s\n" , client->name,dev_name (&client->dev));
@@ -306,12 +302,6 @@ EXPORT_SYMBOL_GPL(i2c_new_device);*/ void i2c_unregister_device (struct i2c_client *client) { -struct i2c_adapter *adapter = client->adapter; - -mutex_lock (&adapter->clist_lock); -list_del (&client->list); -mutex_unlock (&adapter->clist_lock); -device_unregister (&client->dev); }EXPORT_SYMBOL_GPL (i2c_unregister_device);
@@ -437,8 +427,6 @@ static int i2c_register_adapter(struct ireturn -EAGAIN ;mutex_init (&adap->bus_lock); -mutex_init (&adap->clist_lock); -INIT_LIST_HEAD (&adap->clients);mutex_lock (&core_lock);
include/linux/i2c.h +0 -4
@@ -178,7 +178,6 @@ struct i2c_driver {* @driver: device's driver, hence pointer to access routines * @dev: Driver model device node for the slave. * @irq: indicates the IRQ generated by this device (if any) -* @list: list of active/busy clients (DEPRECATED) * @detected: member of an i2c_driver.clients list * * An i2c_client identifies a single device (i.e. chip) connected to an
@@ -195,7 +194,6 @@ struct i2c_client {struct i2c_driver *driver;/* and our access routines */ struct device dev;/* the device structure */ int irq;/* irq issued by device */ -struct list_head list;/* DEPRECATED */ struct list_head detected; };#define to_i2c_client (d) container_of(d, struct i2c_client, dev)
@@ -339,14 +337,12 @@ struct i2c_adapter {/* data fields that are valid for all devices */ u8 level;/* nesting level for lockdep */ struct mutex bus_lock; -struct mutex clist_lock;int timeout;/* in jiffies */ int retries;struct device dev;/* the adapter device */ int nr; -struct list_head clients;/* DEPRECATED */ char name[48 ];struct completion dev_released; };
Signature · 1 line
Jean Delvare
Quoted 7 lines · Jean Delvare
1* In
i2c_device_uevent()there is the following piece of code:/* by definition, legacy drivers can't hotplug */ if (
dev->driver)
return 0;David, I presume this can go away?
Legacy drivers are going away, so ... yes. :)
Quoted 6 lines
2* Core locking should be checked. I am not completely sure what
core_lockwas supposed to protect, nor what it should protect now. I think it should protect at leasti2c_adapter_idr,driver->detect()calls (which may add a device to thedriver->clientslist) anddriver->attach_adapter()calls (for the same reason), although for the latter two, a per-driver lock may be more appropriate.
Protecting the IDR seems right to me. I'd have to look at the post-patches code to say more ... but for now I'll just say that locking driver->*() calls seems less desirable than just making sure the updates to clients_list are protected. You should work to minimize lock scope, so in general *drop* locks before calling out from one module to another. (Lots of nastiness comes from the other module making calls that involve more locking...)
Quoted 5 lines
At the moment, we also hold the lock for all of
i2c_register_adapter(),i2c_del_adapter()andi2c_del_driver(), but noti2c_register_driver(), which seems a little inconsistent. David, do you know if we need to serialize the calls todevice_register()anddevice_unregister(), or if the driver core takes care of this for us?
Driver core locking should eventually suffice for everything except data structures directly managed by the I2C framework.
- Dave
Quoted 4 lines · Jean Delvare
Documentation/feature-removal-schedule.txt | 10 -
Documentation/i2c/writing-clients| 16 -drivers/i2c/i2c-core.c| 366 +++--------------------------include/linux/i2c.h| 57 +----
Which makes all but patch #4 *very* easy to review. All they do is remove code. Unless that breaks a build, those removals have a hard time making any trouble. ;)
Quoted 3 lines
4 files changed, 48 insertions(+), 401 deletions(-)From a binary perspective, "size" announces an 18% shrink.
And Linux now builds without all those warnings, yes?
Plus from a code complexity perspective, it's even better ... it removes 100% of the high level confusion associated with that legacy model, and the bugs and fragility caused by that. (Or maybe I should say that it *finishes* removing that.)
Glad to see this. I presume this will sit in -next for a while and merge in 2.6.31-early?
- Dave
A message that is not in the archive (sent off-list or not yet received): 200905021214.02937.david-b-yBeKhBN/[email protected]
Quoted 10 lines · David Brownell
Quoted 4 lines · Jean Delvare
Documentation/feature-removal-schedule.txt | 10 -
Documentation/i2c/writing-clients| 16 -drivers/i2c/i2c-core.c| 366 +++--------------------------include/linux/i2c.h| 57 +----Which makes all but patch #4 *very* easy to review. All they do is remove code. Unless that breaks a build, those removals have a hard time making any trouble. ;)
Well, removing code can always break things. For example, say I remove all the locking code in the kernel... ;)
Quoted 5 lines
Quoted 3 lines
4 files changed, 48 insertions(+), 401 deletions(-)From a binary perspective, "size" announces an 18% shrink.
And Linux now builds without all those warnings, yes?
Yes it does :)
Quoted 5 lines
Plus from a code complexity perspective, it's even better ... it removes 100% of the high level confusion associated with that legacy model, and the bugs and fragility caused by that. (Or maybe I should say that it *finishes* removing that.)
Thanks for doing all the hard work :)
Quoted 2 lines
Glad to see this. I presume this will sit in -next for a while and merge in 2.6.31-early?
Exactly. Where "early" is unfortunately bound to the few remaining legacy drivers, the conversion of which is supposed to be merged by their respective maintainers. But anyway, the hard deadline is 2.6.31-rc1.
Signature · 1 line
Jean Delvare