1 /****************************************************************************** 2 * Filename: interrupt_doc.h 3 * 4 * Copyright (c) 2015 - 2022, Texas Instruments Incorporated 5 * All rights reserved. 6 * 7 * Redistribution and use in source and binary forms, with or without 8 * modification, are permitted provided that the following conditions are met: 9 * 10 * 1) Redistributions of source code must retain the above copyright notice, 11 * this list of conditions and the following disclaimer. 12 * 13 * 2) Redistributions in binary form must reproduce the above copyright notice, 14 * this list of conditions and the following disclaimer in the documentation 15 * and/or other materials provided with the distribution. 16 * 17 * 3) Neither the name of the ORGANIZATION nor the names of its contributors may 18 * be used to endorse or promote products derived from this software without 19 * specific prior written permission. 20 * 21 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" 22 * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 23 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 24 * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE 25 * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 26 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 27 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 28 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 29 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 30 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 31 * POSSIBILITY OF SUCH DAMAGE. 32 * 33 ******************************************************************************/ 34 //! \addtogroup interrupt_api 35 //! @{ 36 //! \section sec_interrupt Introduction 37 //! 38 //! The interrupt controller API provides a set of functions for dealing with the 39 //! Nested Vectored Interrupt Controller (NVIC). Functions are provided to enable 40 //! and disable interrupts, register interrupt handlers, and set the priority of 41 //! interrupts. 42 //! 43 //! The event sources that trigger the interrupt lines in the NVIC are controlled by 44 //! the MCU event fabric. All event sources are statically connected to the NVIC interrupt lines 45 //! except one which is programmable. For more information about the MCU event fabric, see the 46 //! [MCU event fabric API](\ref event_api). 47 //! 48 //! \section sec_interrupt_api API 49 //! 50 //! Interrupts and system exceptions must be individually enabled and disabled through: 51 //! - \ref IntEnable() 52 //! - \ref IntDisable() 53 //! 54 //! The global CPU interrupt can be enabled and disabled with the following functions: 55 //! - \ref IntMasterEnable() 56 //! - \ref IntMasterDisable() 57 //! 58 //! This does not affect the individual interrupt enable states. Masking of the CPU 59 //! interrupt can be used as a simple critical section (only an NMI can interrupt the 60 //! CPU while the CPU interrupt is disabled), although masking the CPU 61 //! interrupt can increase the interrupt response time. 62 //! 63 //! It is possible to access the NVIC to see if any interrupts are pending and manually 64 //! clear pending interrupts which have not yet been serviced or set a specific interrupt as 65 //! pending to be handled based on its priority. Pending interrupts are cleared automatically 66 //! when the interrupt is accepted and executed. However, the event source which caused the 67 //! interrupt might need to be cleared manually to avoid re-triggering the corresponding interrupt. 68 //! The functions to read, clear, and set pending interrupts are: 69 //! - \ref IntPendGet() 70 //! - \ref IntPendClear() 71 //! - \ref IntPendSet() 72 //! 73 //! The interrupt prioritization in the NVIC allows handling of higher priority interrupts 74 //! before lower priority interrupts, as well as allowing preemption of lower priority interrupt 75 //! handlers by higher priority interrupts. 76 //! The device supports eight priority levels from 0 to 7 with 0 being the highest priority. 77 //! The priority of each interrupt source can be set and examined using: 78 //! - \ref IntPrioritySet() 79 //! - \ref IntPriorityGet() 80 //! 81 //! Interrupts can be masked based on their priority such that interrupts with the same or lower 82 //! priority than the mask are effectively disabled. This can be configured with: 83 //! - \ref IntPriorityMaskSet() 84 //! - \ref IntPriorityMaskGet() 85 //! 86 //! Subprioritization is also possible. Instead of having three bits of preemptable 87 //! prioritization (eight levels), the NVIC can be configured for 3 - M bits of 88 //! preemptable prioritization and M bits of subpriority. In this scheme, two 89 //! interrupts with the same preemptable prioritization but different subpriorities 90 //! do not cause a preemption. Instead, tail chaining is used to process 91 //! the two interrupts back-to-back. 92 //! If two interrupts with the same priority (and subpriority if so configured) are 93 //! asserted at the same time, the one with the lower interrupt number is 94 //! processed first. 95 //! Subprioritization is handled by: 96 //! - \ref IntPriorityGroupingSet() 97 //! - \ref IntPriorityGroupingGet() 98 //! 99 //! \section sec_interrupt_table Interrupt Vector Table 100 //! 101 //! The interrupt vector table can be configured in one of two ways: 102 //! - Statically (at compile time): Vector table is placed in Flash and each entry has a fixed 103 //! pointer to an interrupt handler (ISR). 104 //! - Dynamically (at runtime): Vector table is placed in SRAM and each entry can be changed 105 //! (registered or unregistered) at runtime. This allows a single interrupt to trigger different 106 //! interrupt handlers (ISRs) depending on which interrupt handler is registered at the time the 107 //! System CPU responds to the interrupt. 108 //! 109 //! When configured, the interrupts must be explicitly enabled in the NVIC through \ref IntEnable() 110 //! before the CPU can respond to the interrupt (in addition to any interrupt enabling required 111 //! within the peripheral). 112 //! 113 //! \subsection sec_interrupt_table_static Static Vector Table 114 //! 115 //! Static registration of interrupt handlers is accomplished by editing the interrupt handler 116 //! table in the startup code of the application. Texas Instruments provides startup files for 117 //! each supported compiler ( \ti_code{startup_<compiler>.c} ) and these startup files include 118 //! a default static interrupt vector table. 119 //! All entries, except ResetISR, are declared as \c extern with weak assignment to a default 120 //! interrupt handler. This allows the user to declare and define a function (in the user's code) 121 //! with the same name as an entry in the vector table. At compile time, the linker then replaces 122 //! the pointer to the default interrupt handler in the vector table with the pointer to the 123 //! interrupt handler defined by the user. 124 //! 125 //! Statically configuring the interrupt table provides the fastest interrupt response time 126 //! because the stacking operation (a write to SRAM on the data bus) is performed in parallel 127 //! with the interrupt handler table fetch (a read from Flash on the instruction bus), as well 128 //! as the prefetch of the interrupt handler (assuming it is also in Flash). 129 //! 130 //! \subsection sec_interrupt_table_dynamic Dynamic Vector Table 131 //! 132 //! Alternatively, interrupts can be registered in the vector table at runtime, thus dynamically. 133 //! The dynamic vector table is placed in SRAM and the code can then modify the pointers to 134 //! interrupt handlers throughout the application. 135 //! 136 //! DriverLib uses these two functions to modify the dynamic vector table: 137 //! - \ref IntRegister() : Write a pointer to an interrupt handler into the vector table. 138 //! - \ref IntUnregister() : Write pointer to default interrupt handler into the vector table. 139 //! 140 //! \note First call to \ref IntRegister() initializes the vector table in SRAM by copying the 141 //! static vector table from Flash and forcing the NVIC to use the dynamic vector table from 142 //! this point forward. If using the dynamic vector table it is highly recommended to 143 //! initialize it during the setup phase of the application. The NVIC uses the static vector 144 //! table in Flash until the application initializes the dynamic vector table in SRAM. 145 //! 146 //! Runtime configuration of interrupts adds a small latency to the interrupt response time 147 //! because the stacking operation (a write to SRAM on the data bus) and the interrupt handler 148 //! fetch from the vector table (a read from SRAM on the instruction bus) must be performed 149 //! sequentially. 150 //! 151 //! The dynamic vector table, \ref g_pfnRAMVectors, is placed in SRAM in the section called 152 //! \c vtable_ram which is a section defined in the linker file. By default the linker file 153 //! places this section at the start of the SRAM but this is configurable by the user. 154 //! 155 //! \warning Runtime configuration of interrupt handlers requires that the interrupt 156 //! handler table is placed on a 256-byte boundary in SRAM (typically, this is 157 //! at the beginning of SRAM). Failure to do so results in an incorrect vector 158 //! address being fetched in response to an interrupt. 159 //! 160 //! @} 161