)]}' { "log": [ { "commit": "3bece55aa5356af0171aaa64fd9c4f7601c47f1c", "tree": "8dd08d51859062defc424ff234f1032b89e9410e", "parents": [ "43699dea1ea21a0d5786317a794cb2ba27a6f4fe" ], "author": { "name": "Linus Walleij", "email": "linus.walleij@linaro.org", "time": "Sun Dec 18 23:44:26 2011 +0100" }, "committer": { "name": "Linus Walleij", "email": "linus.walleij@linaro.org", "time": "Tue Jan 03 09:10:08 2012 +0100" }, "message": "pinctrl: rename U300 and SIRF pin controllers\n\nFor stringent order, rename the pinmux-* pin controllers to\npinctrl-* and also rename the Kconfig symbols and in-kernel\nusers.\n\nCc: Rongjun Ying \u003cRongjun.Ying@csr.com\u003e\nCc: Jean-Christophe PLAGNIOL-VILLARD \u003cplagnioj@jcrosoft.com\u003e\nAcked-by: Stephen Warren \u003cswarren@nvidia.com\u003e\nSigned-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n" }, { "commit": "ae6b4d8588f4fc95520b0e62c4b1f474c82191a9", "tree": "3da8e553a6374f02e89b5a6ba52b83f34c3abea2", "parents": [ "b4e3ac74d5cd4152f2ec6b3280b1ff3428952f7f" ], "author": { "name": "Linus Walleij", "email": "linus.walleij@linaro.org", "time": "Wed Oct 19 18:14:33 2011 +0200" }, "committer": { "name": "Linus Walleij", "email": "linus.walleij@linaro.org", "time": "Tue Jan 03 09:10:04 2012 +0100" }, "message": "pinctrl: add a pin config interface\n\nThis add per-pin and per-group pin config interfaces for biasing,\ndriving and other such electronic properties. The details of passed\nconfigurations are passed in an opaque unsigned long which may be\ndereferences to integer types, structs or lists on either side\nof the configuration interface.\n\nChangeLog v1-\u003ev2:\n- Clear split of terminology: we now have pin controllers, and\n those may support two interfaces using vtables: pin\n multiplexing and pin configuration.\n- Break out pin configuration to its own C file, controllers may\n implement only config without mux, and vice versa, so keep each\n sub-functionality of pin controllers separate. Introduce\n CONFIG_PINCONF in Kconfig.\n- Implement some core logic around pin configuration in the\n pinconf.c file.\n- Remove UNKNOWN config states, these were just surplus baggage.\n- Remove FLOAT config state - HIGH_IMPEDANCE should be enough for\n everyone.\n- PIN_CONFIG_POWER_SOURCE added to handle switching the power\n supply for the pin logic between different sources\n- Explicit DISABLE config enums to turn schmitt-trigger,\n wakeup etc OFF.\n- Update documentation to reflect all the recent reasoning.\nChangeLog v2-\u003ev3:\n- Twist API around to pass around arrays of config tuples instead\n of (param, value) pairs everywhere.\n- Explicit drive strength semantics for push/pull and similar\n drive modes, this shall be the number of drive stages vs\n nominal load impedance, which should match the actual\n electronics used in push/pull CMOS or TTY totempoles.\n- Drop load capacitance configuration - I probably don\u0027t know\n what I\u0027m doing here so leave it out.\n- Drop PIN_CONFIG_INPUT_SCHMITT_OFF, instead the argument zero to\n PIN_CONFIG_INPUT_SCHMITT turns schmitt trigger off.\n- Drop PIN_CONFIG_NORMAL_POWER_MODE and have a well defined\n argument to PIN_CONFIG_LOW_POWER_MODE to get out of it instead.\n- Drop PIN_CONFIG_WAKEUP_ENABLE/DISABLE and just use\n PIN_CONFIG_WAKEUP with defined value zero to turn wakeup off.\n- Add PIN_CONFIG_INPUT_DEBOUNCE for configuring debounce time\n on input lines.\n- Fix a bug when we tried to configure pins for pin controllers\n without pinconf support.\n- Initialized debugfs properly so it works.\n- Initialize the mutex properly and lock around config tampering\n sections.\n- Check the return value from get_initial_config() properly.\nChangeLog v3-\u003ev4:\n- Export the pin_config_get(), pin_config_set() and\n pin_config_group() functions.\n- Drop the entire concept of just getting initial config and\n keeping track of pin states internally, instead ask the pins\n what state they are in. Previous idea was plain wrong, if the\n device cannot keep track of its state, the driver should do\n it.\n- Drop the generic configuration layout, it seems this impose\n too much restriction on some pin controllers, so let them do\n things the way they want and split off support for generic\n config as an optional add-on.\nChangeLog v4-\u003ev5:\n- Introduce two symmetric driver calls for group configuration,\n .pin_config_group_[get|set] and corresponding external calls.\n- Remove generic semantic meanings of return values from config\n calls, these belong in the generic config patch. Just pass the\n return value through instead.\n- Add a debugfs entry \"pinconf-groups\" to read status from group\n configuration only, also slam in a per-group debug callback in\n the pinconf_ops so custom drivers can display something\n meaningful for their pins.\n- Fix some dangling newline.\n- Drop dangling #else clause.\n- Update documentation to match the above.\nChangeLog v5-\u003ev6:\n- Change to using a pin name as parameter for the\n [get|set]_config() functions, as suggested by Stephen Warren.\n This is more natural as names will be what a developer has\n access to in written documentation etc.\nChangeLog v6-\u003ev7:\n- Refactor out by-pin and by-name get/set functions, only expose\n the by-name functions externally, expose the by-pin functions\n internally.\n- Show supported pin control functionality in the debugfs\n pinctrl-devices file.\n\nAcked-by: Stephen Warren \u003cswarren@nvidia.com\u003e\nSigned-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n" }, { "commit": "b4e3ac74d5cd4152f2ec6b3280b1ff3428952f7f", "tree": "b1dc770b28ac3dcdf2fd2b770827e4112804155e", "parents": [ "f812f0f53e5643c06b929ce3299cbaffb307c488" ], "author": { "name": "Linus Walleij", "email": "linus.walleij@linaro.org", "time": "Wed Nov 16 10:24:39 2011 +0100" }, "committer": { "name": "Linus Walleij", "email": "linus.walleij@linaro.org", "time": "Tue Jan 03 09:10:04 2012 +0100" }, "message": "pinctrl/coh901: driver to request its pins\n\nThis makes the COH 901 driver request muxing of its GPIO pins\nfrom the pinmux-u300 driver using the standard API calls.\n\nSigned-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n" }, { "commit": "ca402d37dccf2b797440c5f03bd0db16f977acc9", "tree": "7dee0cb62c951122dea275a680da837ee81f69d0", "parents": [ "dd68acc7cc256c928256eb5f53e163233605de5d" ], "author": { "name": "Linus Walleij", "email": "linus.walleij@linaro.org", "time": "Wed Nov 16 09:22:59 2011 +0100" }, "committer": { "name": "Linus Walleij", "email": "linus.walleij@linaro.org", "time": "Tue Jan 03 09:10:03 2012 +0100" }, "message": "pinctrl: move the U300 GPIO driver to pinctrl\n\nThis driver will be converted to a dual GPIO + pinctrl driver\nsince it supports biasing and driving control options. Hopefully\nit can serve as an example.\n\nSigned-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n" }, { "commit": "45f034ef205e5439a50d6f7e5f89add93131c0cc", "tree": "aec6b5d06227b16f8f9a243ab1599cfab5da6c39", "parents": [ "e0e20753c15fc418d94fee826af394907df856d8" ], "author": { "name": "Linus Walleij", "email": "linus.walleij@linaro.org", "time": "Sat Nov 05 21:28:46 2011 +0100" }, "committer": { "name": "Linus Walleij", "email": "linus.walleij@linaro.org", "time": "Thu Nov 10 09:02:12 2011 +0100" }, "message": "pinctrl: hide subsystem from the populace\n\nMachines that have embedded pin controllers need to select them\nexplicitly, so why broadcast their config options to menuconfig.\nWe provide a helpful submenu for those machines that do select\nit, making it possible to enable debugging for example.\n\nReported-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nAcked-by: Stephen Warren \u003cswarren@nvidia.com\u003e\nSigned-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n" }, { "commit": "393daa814f4bbc6f5c099178c073fae9f7ef6177", "tree": "6bf1296eb65da30fe71a82d5bc0fd43dfe2c3d1d", "parents": [ "98da3529536ed3c78ae493f4cc3d7ac8d43fc72c" ], "author": { "name": "Rongjun Ying", "email": "Rongjun.Ying@csr.com", "time": "Sun Oct 09 03:11:13 2011 -0700" }, "committer": { "name": "Linus Walleij", "email": "linus.walleij@stericsson.com", "time": "Thu Oct 13 12:57:46 2011 +0200" }, "message": "pinctrl: add a driver for the CSR SiRFprimaII pinmux\n\nThis creates a pin controller driver for the SiRFprinaII\npin mux portions.\n\nSigned-off-by: Rongjun Ying \u003cRongjun.Ying@csr.com\u003e\nSigned-off-by: Barry Song \u003cBaohua.Song@csr.com\u003e\n[Fixup for changed function names and semantics in the v10 patch]\nSigned-off-by: Linus Walleij \u003clinus.walleij@stericsson.com\u003e\n" }, { "commit": "98da3529536ed3c78ae493f4cc3d7ac8d43fc72c", "tree": "4d10ff569e1cbea384c7e0e162b0481be4766e30", "parents": [ "2744e8afb3b76343e7eb8197e8b3e085036010a5" ], "author": { "name": "Linus Walleij", "email": "linus.walleij@linaro.org", "time": "Mon May 02 20:54:38 2011 +0200" }, "committer": { "name": "Linus Walleij", "email": "linus.walleij@stericsson.com", "time": "Thu Oct 13 12:57:45 2011 +0200" }, "message": "pinctrl: add a driver for the U300 pinmux\n\nThis adds a driver for the U300 pinmux portions of the system\ncontroller \"SYSCON\". It also serves as an example of how to use\nthe pinmux subsystem. This driver also houses the platform data\nfor the only supported platform.\n\nThis deletes the old U300 driver in arch/arm/mach-u300 and\nreplace it with a driver using the new subsystem.\n\nThe new driver is considerably fatter than the old one, but it\nalso registers all 467 pins of the system and adds the power\nand EMIF pin groups and corresponding functions. The idea\nis to use this driver as a a reference for other\nimplementation so it needs to be as complete and verbose\nas possible.\n\nReviewed-by: Barry Song \u003c21cnbao@gmail.com\u003e\n[Fixup for changed function names and semantics in the v10 patch]\nSigned-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n" }, { "commit": "2744e8afb3b76343e7eb8197e8b3e085036010a5", "tree": "2a96122ef4a6114483bda0fe9696d61a6e6f1bb5", "parents": [ "a102a9ece5489e1718cd7543aa079082450ac3a2" ], "author": { "name": "Linus Walleij", "email": "linus.walleij@linaro.org", "time": "Mon May 02 20:50:54 2011 +0200" }, "committer": { "name": "Linus Walleij", "email": "linus.walleij@stericsson.com", "time": "Thu Oct 13 12:49:17 2011 +0200" }, "message": "drivers: create a pin control subsystem\n\nThis creates a subsystem for handling of pin control devices.\nThese are devices that control different aspects of package\npins.\n\nCurrently it handles pinmuxing, i.e. assigning electronic\nfunctions to groups of pins on primarily PGA and BGA type of\nchip packages which are common in embedded systems.\n\nThe plan is to also handle other I/O pin control aspects\nsuch as biasing, driving, input properties such as\nschmitt-triggering, load capacitance etc within this\nsubsystem, to remove a lot of ARM arch code as well as\nfeature-creepy GPIO drivers which are implementing the same\nthing over and over again.\n\nThis is being done to depopulate the arch/arm/* directory\nof such custom drivers and try to abstract the infrastructure\nthey all need. See the Documentation/pinctrl.txt file that is\npart of this patch for more details.\n\nChangeLog v1-\u003ev2:\n\n- Various minor fixes from Joe\u0027s and Stephens review comments\n- Added a pinmux_config() that can invoke custom configuration\n with arbitrary data passed in or out to/from the pinmux driver\n\nChangeLog v2-\u003ev3:\n\n- Renamed subsystem folder to \"pinctrl\" since we will likely\n want to keep other pin control such as biasing in this\n subsystem too, so let us keep to something generic even though\n we\u0027re mainly doing pinmux now.\n- As a consequence, register pins as an abstract entity separate\n from the pinmux. The muxing functions will claim pins out of the\n pin pool and make sure they do not collide. Pins can now be\n named by the pinctrl core.\n- Converted the pin lookup from a static array into a radix tree,\n I agreed with Grant Likely to try to avoid any static allocation\n (which is crap for device tree stuff) so I just rewrote this\n to be dynamic, just like irq number descriptors. The\n platform-wide definition of number of pins goes away - this is\n now just the sum total of the pins registered to the subsystem.\n- Make sure mappings with only a function name and no device\n works properly.\n\nChangeLog v3-\u003ev4:\n\n- Define a number space per controller instead of globally,\n Stephen and Grant requested the same thing so now maps need to\n define target controller, and the radix tree of pin descriptors\n is a property on each pin controller device.\n- Add a compulsory pinctrl device entry to the pinctrl mapping\n table. This must match the pinctrl device, like \"pinctrl.0\"\n- Split the file core.c in two: core.c and pinmux.c where the\n latter carry all pinmux stuff, the core is for generic pin\n control, and use local headers to access functionality between\n files. It is now possible to implement a \"blank\" pin controller\n without pinmux capabilities. This split will make new additions\n like pindrive.c, pinbias.c etc possible for combined drivers\n and chunks of functionality which is a GoodThing(TM).\n- Rewrite the interaction with the GPIO subsystem - the pin\n controller descriptor now handles this by defining an offset\n into the GPIO numberspace for its handled pin range. This is\n used to look up the apropriate pin controller for a GPIO pin.\n Then that specific GPIO range is matched 1-1 for the target\n controller instance.\n- Fixed a number of review comments from Joe Perches.\n- Broke out a header file pinctrl.h for the core pin handling\n stuff that will be reused by other stuff than pinmux.\n- Fixed some erroneous EXPORT() stuff.\n- Remove mispatched U300 Kconfig and Makefile entries\n- Fixed a number of review comments from Stephen Warren, not all\n of them - still WIP. But I think the new mapping that will\n specify which function goes to which pin mux controller address\n 50% of your concerns (else beat me up).\n\nChangeLog v4-\u003ev5:\n\n- Defined a \"position\" for each function, so the pin controller now\n tracks a function in a certain position, and the pinmux maps define\n what position you want the function in. (Feedback from Stephen\n Warren and Sascha Hauer).\n- Since we now need to request a combined function+position from\n the machine mapping table that connect mux settings to drivers,\n it was extended with a position field and a name field. The\n name field is now used if you e.g. need to switch between two\n mux map settings at runtime.\n- Switched from a class device to using struct bus_type for this\n subsystem. Verified sysfs functionality: seems to work fine.\n (Feedback from Arnd Bergmann and Greg Kroah-Hartman)\n- Define a per pincontroller list of GPIO ranges from the GPIO\n pin space that can be handled by the pin controller. These can\n be added one by one at runtime. (Feedback from Barry Song)\n- Expanded documentation of regulator_[get|enable|disable|put]\n semantics.\n- Fixed a number of review comments from Barry Song. (Thanks!)\n\nChangeLog v5-\u003ev6:\n\n- Create an abstract pin group concept that can sort pins into\n named and enumerated groups no matter what the use of these\n groups may be, one possible usecase is a group of pins being\n muxed in or so. The intention is however to also use these\n groups for other pin control activities.\n- Make it compulsory for pinmux functions to associate with\n at least one group, so the abstract pin group concept is used\n to define the groups of pins affected by a pinmux function.\n The pinmux driver interface has been altered so as to enforce\n a function to list applicable groups per function.\n- Provide an optional .group entry in the pinmux machine map\n so the map can select beteween different available groups\n to be used with a certain function.\n- Consequent changes all over the place so that e.g. debugfs\n present reasonable information about the world.\n- Drop the per-pin mux (*config) function in the pinmux_ops\n struct - I was afraid that this would start to be used for\n things totally unrelated to muxing, we can introduce that to\n the generic struct pinctrl_ops if needed. I want to keep\n muxing orthogonal to other pin control subjects and not mix\n these things up.\n\nChangeLog v6-\u003ev7:\n\n- Make it possible to have several map entries matching the\n same device, pin controller and function, but using\n a different group, and alter the semantics so that\n pinmux_get() will pick all matching map entries, and\n store the associated groups in a list. The list will\n then be iterated over at pinmux_enable()/pinmux_disable()\n and corresponding driver functions called for each\n defined group. Notice that you\u0027re only allowed to map\n multiple *groups* to the same\n { device, pin controller, function } triplet, attempts\n to map the same device to multiple pin controllers will\n for example fail. This is hopefully the crucial feature\n requested by Stephen Warren.\n- Add a pinmux hogging field to the pinmux mapping entries,\n and enable the pinmux core to hog pinmux map entries.\n This currently only works for pinmuxes without assigned\n devices as it looks now, but with device trees we can\n look up the corresponding struct device * entries when\n we register the pinmux driver, and have it hog each\n pinmux map in turn, for a simple approach to\n non-dynamic pin muxing. This addresses an issue from\n Grant Likely that the machine should take care of as\n much of the pinmux setup as possible, not the devices.\n By supplying a list of hogs, it can now instruct the\n core to take care of any static mappings.\n- Switch pinmux group retrieveal function to grab an\n array of strings representing the groups rather than an\n array of unsigned and rewrite accordingly.\n- Alter debugfs to show the grouplist handled by each\n pinmux. Also add a list of hogs.\n- Dynamically allocate a struct pinmux at pinmux_get() and\n free it at pinmux_put(), then add these to the global\n list of pinmuxes active as we go along.\n- Go over the list of pinmux maps at pinmux_get() time\n and repeatedly apply matches.\n- Retrieve applicable groups per function from the driver\n as a string array rather than a unsigned array, then\n lookup the enumerators.\n- Make the device to pinmux map a singleton - only allow the\n mapping table to be registered once and even tag the\n registration function with __init so it surely won\u0027t be\n abused.\n- Create a separate debugfs file to view the pinmux map at\n runtime.\n- Introduce a spin lock to the pin descriptor struct, lock it\n when modifying pin status entries. Reported by Stijn Devriendt.\n- Fix up the documentation after review from Stephen Warren.\n- Let the GPIO ranges give names as const char * instead of some\n fixed-length string.\n- add a function to unregister GPIO ranges to mirror the\n registration function.\n- Privatized the struct pinctrl_device and removed it from the\n \u003clinux/pinctrl/pinctrl.h\u003e API, the drivers do not need to know\n the members of this struct. It is now in the local header\n \"core.h\".\n- Rename the concept of \"anonymous\" mux maps to \"system\" muxes\n and add convenience macros and documentation.\n\nChangeLog v7-\u003ev8:\n\n- Delete the leftover pinmux_config() function from the\n \u003clinux/pinctrl/pinmux.h\u003e header.\n- Fix a race condition found by Stijn Devriendt in pin_request()\n\nChangeLog v8-\u003ev9:\n\n- Drop the bus_type and the sysfs attributes and all, we\u0027re not on\n the clear about how this should be used for e.g. userspace\n interfaces so let us save this for the future.\n- Use the right name in MAINTAINERS, PIN CONTROL rather than\n PINMUX\n- Don\u0027t kfree() the device state holder, let the .remove() callback\n handle this.\n- Fix up numerous kerneldoc headers to have one line for the function\n description and more verbose documentation below the parameters\n\nChangeLog v9-\u003ev10:\n- pinctrl: EXPORT_SYMBOL needs export.h, folded in a patch\n from Steven Rothwell\n- fix pinctrl_register error handling, folded in a patch from\n Axel Lin\n- Various fixes to documentation text so that it\u0027s consistent.\n- Removed pointless comment from drivers/Kconfig\n- Removed dependency on SYSFS since we removed the bus in\n v9.\n- Renamed hopelessly abbreviated pctldev_* functions to the\n more verbose pinctrl_dev_*\n- Drop mutex properly when looking up GPIO ranges\n- Return NULL instead of ERR_PTR() errors on registration of\n pin controllers, using cast pointers is fragile. We can\n live without the detailed error codes for sure.\n\nCc: Stijn Devriendt \u003chighguy@gmail.com\u003e\nCc: Joe Perches \u003cjoe@perches.com\u003e\nCc: Russell King \u003clinux@arm.linux.org.uk\u003e\nAcked-by: Grant Likely \u003cgrant.likely@secretlab.ca\u003e\nAcked-by: Stephen Warren \u003cswarren@nvidia.com\u003e\nTested-by: Barry Song \u003c21cnbao@gmail.com\u003e\nSigned-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\n" } ] }