/* * Copyright (C) 2009 Francisco Jerez. * All Rights Reserved. * * Permission is hereby granted, free of charge, to any person obtaining * a copy of this software and associated documentation files (the * "Software"), to deal in the Software without restriction, including * without limitation the rights to use, copy, modify, merge, publish, * distribute, sublicense, and/or sell copies of the Software, and to * permit persons to whom the Software is furnished to do so, subject to * the following conditions: * * The above copyright notice and this permission notice (including the * next paragraph) shall be included in all copies or substantial * portions of the Software. * * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. * IN NO EVENT SHALL THE COPYRIGHT OWNER(S) AND/OR ITS SUPPLIERS BE * LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION * OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION * WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. * */ #ifndef __DRM_ENCODER_SLAVE_H__ #define __DRM_ENCODER_SLAVE_H__ #include #include /** * struct drm_encoder_slave_funcs - Entry points exposed by a slave encoder driver * @set_config: Initialize any encoder-specific modesetting parameters. * The meaning of the @params parameter is implementation * dependent. It will usually be a structure with DVO port * data format settings or timings. It's not required for * the new parameters to take effect until the next mode * is set. * * Most of its members are analogous to the function pointers in * &drm_encoder_helper_funcs and they can optionally be used to * initialize the latter. Connector-like methods (e.g. @get_modes and * @set_property) will typically be wrapped around and only be called * if the encoder is the currently selected one for the connector. */ struct drm_encoder_slave_funcs { void (*set_config)(struct drm_encoder *encoder, void *params); void (*destroy)(struct drm_encoder *encoder); void (*dpms)(struct drm_encoder *encoder, int mode); void (*save)(struct drm_encoder *encoder); void (*restore)(struct drm_encoder *encoder); bool (*mode_fixup)(struct drm_encoder *encoder, const struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode); int (*mode_valid)(struct drm_encoder *encoder, struct drm_display_mode *mode); void (*mode_set)(struct drm_encoder *encoder, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode); enum drm_connector_status (*detect)(struct drm_encoder *encoder, struct drm_connector *connector); int (*get_modes)(struct drm_encoder *encoder, struct drm_connector *connector); int (*create_resources)(struct drm_encoder *encoder, struct drm_connector *connector); int (*set_property)(struct drm_encoder *encoder, struct drm_connector *connector, struct drm_property *property, uint64_t val); }; /** * struct drm_encoder_slave - Slave encoder struct * @base: DRM encoder object. * @slave_funcs: Slave encoder callbacks. * @slave_priv: Slave encoder private data. * @bus_priv: Bus specific data. * * A &drm_encoder_slave has two sets of callbacks, @slave_funcs and the * ones in @base. The former are never actually called by the common * CRTC code, it's just a convenience for splitting the encoder * functions in an upper, GPU-specific layer and a (hopefully) * GPU-agnostic lower layer: It's the GPU driver responsibility to * call the slave methods when appropriate. * * drm_i2c_encoder_init() provides a way to get an implementation of * this. */ struct drm_encoder_slave { struct drm_encoder base; const struct drm_encoder_slave_funcs *slave_funcs; void *slave_priv; void *bus_priv; }; #define to_encoder_slave(x) container_of((x), struct drm_encoder_slave, base) int drm_i2c_encoder_init(struct drm_device *dev, struct drm_encoder_slave *encoder, struct i2c_adapter *adap, const struct i2c_board_info *info); /** * struct drm_i2c_encoder_driver * * Describes a device driver for an encoder connected to the GPU * through an I2C bus. In addition to the entry points in @i2c_driver * an @encoder_init function should be provided. It will be called to * give the driver an opportunity to allocate any per-encoder data * structures and to initialize the @slave_funcs and (optionally) * @slave_priv members of @encoder. */ struct drm_i2c_encoder_driver { struct i2c_driver i2c_driver; int (*encoder_init)(struct i2c_client *client, struct drm_device *dev, struct drm_encoder_slave *encoder); }; #define to_drm_i2c_encoder_driver(x) container_of((x), \ struct drm_i2c_encoder_driver, \ i2c_driver) /** * drm_i2c_encoder_get_client - Get the I2C client corresponding to an encoder */ static inline struct i2c_client *drm_i2c_encoder_get_client(struct drm_encoder *encoder) { return (struct i2c_client *)to_encoder_slave(encoder)->bus_priv; } /** * drm_i2c_encoder_register - Register an I2C encoder driver * @owner: Module containing the driver. * @driver: Driver to be registered. */ static inline int drm_i2c_encoder_register(struct module *owner, struct drm_i2c_encoder_driver *driver) { return i2c_register_driver(owner, &driver->i2c_driver); } /** * drm_i2c_encoder_unregister - Unregister an I2C encoder driver * @driver: Driver to be unregistered. */ static inline void drm_i2c_encoder_unregister(struct drm_i2c_encoder_driver *driver) { i2c_del_driver(&driver->i2c_driver); } void drm_i2c_encoder_destroy(struct drm_encoder *encoder); /* * Wrapper fxns which can be plugged in to drm_encoder_helper_funcs: */ void drm_i2c_encoder_dpms(struct drm_encoder *encoder, int mode); bool drm_i2c_encoder_mode_fixup(struct drm_encoder *encoder, const struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode); void drm_i2c_encoder_prepare(struct drm_encoder *encoder); void drm_i2c_encoder_commit(struct drm_encoder *encoder); void drm_i2c_encoder_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode); enum drm_connector_status drm_i2c_encoder_detect(struct drm_encoder *encoder, struct drm_connector *connector); void drm_i2c_encoder_save(struct drm_encoder *encoder); void drm_i2c_encoder_restore(struct drm_encoder *encoder); #endif e atomic bit operations return the old value of the same bit we changed, not the value of an unrelated bit. On x86, we could put the lock bit in the high bit of the byte, and use "xadd" with that bit (where the overflow ends up not touching other bits), and look at the other bits of the result. However, an even simpler model is to just use a regular atomic "and" to clear the lock bit, and then the sign bit in eflags will indicate the resulting state of the unrelated bit #7. So by moving the PageWaiters bit up to bit #7, we can atomically clear the lock bit and test the waiters bit on x86 too. And architectures with LL/SC (which is all the usual RISC suspects), the particular bit doesn't matter, so they are fine with this approach too. This avoids the extra access to the same atomic word, and thus avoids the costly stall at page unlock time. The only downside is that the interface ends up being a bit odd and specialized: clear a bit in a byte, and test the sign bit. Nick doesn't love the resulting name of the new primitive, but I'd rather make the name be descriptive and very clear about the limitation imposed by trying to work across all relevant architectures than make it be some generic thing that doesn't make the odd semantics explicit. So this introduces the new architecture primitive clear_bit_unlock_is_negative_byte(); and adds the trivial implementation for x86. We have a generic non-optimized fallback (that just does a "clear_bit()"+"test_bit(7)" combination) which can be overridden by any architecture that can do better. According to Nick, Power has the same hickup x86 has, for example, but some other architectures may not even care. All these optimizations mean that my page locking stress-test (which is just executing a lot of small short-lived shell scripts: "make test" in the git source tree) no longer makes our page locking look horribly bad. Before all these optimizations, just the unlock_page() costs were just over 3% of all CPU overhead on "make test". After this, it's down to 0.66%, so just a quarter of the cost it used to be. (The difference on NUMA is bigger, but there this micro-optimization is likely less noticeable, since the big issue on NUMA was not the accesses to 'struct page', but the waitqueue accesses that were already removed by Nick's earlier commit). Acked-by: Nick Piggin <npiggin@gmail.com> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: Bob Peterson <rpeterso@redhat.com> Cc: Steven Whitehouse <swhiteho@redhat.com> Cc: Andrew Lutomirski <luto@kernel.org> Cc: Andreas Gruenbacher <agruenba@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Mel Gorman <mgorman@techsingularity.net> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'include/sound')