Lines Matching full:the
8 * this software and associated documentation files (the "Software"), to deal in
9 * the Software without restriction, including without limitation the rights to
11 * the Software, and to permit persons to whom the Software is furnished to do so,
12 * subject to the following conditions:
14 * The above copyright notice and this permission notice shall be included in all
15 * copies or substantial portions of the Software.
17 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
18 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS
19 * FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR
22 * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
30 * Implementation of functions defined in portable.h for the ARM CM3 port.
41 /* Constants required to manipulate the core. Registers first... */
46 /* ...then bits in the registers. */
59 /* Constants required to check the validity of an interrupt priority. */
69 /* Masks off all bits but the VECTACTIVE bits in the ICSR register. */
72 /* Constants required to set up the initial stack. */
75 /* The systick is a 24-bit counter. */
78 /* A fiddle factor to estimate the number of SysTick counts that would have
79 * occurred while the SysTick counter is stopped during tickless idle
83 /* For strict compliance with the Cortex-M spec the task start address should
84 * have bit-0 clear, as it is loaded into the PC on exit from an ISR. */
87 /* Let the user override the default SysTick clock rate. If defined by the
88 * user, this symbol must equal the SysTick clock rate when the CLK bit is 0 in the
92 /* Ensure the SysTick is clocked at the same frequency as the core. */
95 /* Select the option to clock SysTick not at the same frequency as the core. */
100 * Setup the timer to generate the tick interrupts. The implementation in this
101 * file is weak to allow application writers to change the timer used to
102 * generate the tick interrupt.
123 /* Required to allow portasm.asm access the configMAX_SYSCALL_INTERRUPT_PRIORITY
127 /* Each task maintains its own interrupt status in the critical nesting
132 * The number of SysTick increments that make up one tick period.
139 * The maximum number of tick periods that can be suppressed is limited by the
140 * 24 bit resolution of the SysTick timer.
147 * Compensate for the CPU cycles that pass while the SysTick is stopped (low
155 * Used by the portASSERT_IF_INTERRUPT_PRIORITY_INVALID() macro to ensure
174 /* Simulate the stack frame as it would be created by a context switch in pxPortInitialiseStack()
177 /* Offset added to account for the way the MCU uses the stack on entry/exit in pxPortInitialiseStack()
204 * defined, then stop here so application writers can catch the error. */ in prvTaskExitError()
226 /* Determine the maximum priority from which ISR safe FreeRTOS API in xPortStartScheduler()
231 * Save the interrupt priority value that is about to be clobbered. */ in xPortStartScheduler()
234 /* Determine the number of priority bits available. First write to all in xPortStartScheduler()
238 /* Read the value back to see how many bits stuck. */ in xPortStartScheduler()
241 /* Use the same mask on the maximum system call priority. */ in xPortStartScheduler()
244 /* Check that the maximum system call priority is nonzero after in xPortStartScheduler()
245 * accounting for the number of priority bits supported by the in xPortStartScheduler()
246 * hardware. A priority of 0 is invalid because setting the BASEPRI in xPortStartScheduler()
252 /* Check that the bits not implemented in hardware are zero in in xPortStartScheduler()
256 /* Calculate the maximum acceptable priority group value for the number in xPortStartScheduler()
267 /* When the hardware implements 8 priority bits, there is no way for in xPortStartScheduler()
268 * the software to configure PRIGROUP to not have sub-priorities. As in xPortStartScheduler()
269 * a result, the least significant bit is always used for sub-priority in xPortStartScheduler()
275 * are at the same preemption priority. This may appear confusing as in xPortStartScheduler()
279 * to 4, this confusion does not happen and the behaviour remains the same. in xPortStartScheduler()
281 * The following assert ensures that the sub-priority bit in the in xPortStartScheduler()
282 * configMAX_SYSCALL_INTERRUPT_PRIORITY is clear to avoid the above mentioned in xPortStartScheduler()
292 /* Shift the priority group value back to its position within the AIRCR in xPortStartScheduler()
297 /* Restore the clobbered interrupt priority register to its original in xPortStartScheduler()
303 /* Make PendSV and SysTick the lowest priority interrupts. */ in xPortStartScheduler()
307 /* Start the timer that generates the tick ISR. Interrupts are disabled in xPortStartScheduler()
311 /* Initialise the critical nesting count ready for the first task. */ in xPortStartScheduler()
314 /* Start the first task. */ in xPortStartScheduler()
335 /* This is not the interrupt safe version of the enter critical function so in vPortEnterCritical()
338 * the critical nesting count is 1 to protect against recursive calls if the in vPortEnterCritical()
361 /* The SysTick runs at the lowest interrupt priority, so when this interrupt in xPortSysTickHandler()
363 * save and then restore the interrupt mask value as its value is already in xPortSysTickHandler()
368 /* Increment the RTOS tick. */ in xPortSysTickHandler()
374 * the PendSV interrupt. Pend the PendSV interrupt. */ in xPortSysTickHandler()
394 /* Make sure the SysTick reload value does not overflow the counter. */ in vPortSuppressTicksAndSleep()
400 /* Enter a critical section but don't use the taskENTER_CRITICAL() in vPortSuppressTicksAndSleep()
406 /* If a context switch is pending or a task is waiting for the scheduler in vPortSuppressTicksAndSleep()
407 * to be unsuspended then abandon the low power entry. */ in vPortSuppressTicksAndSleep()
410 /* Re-enable interrupts - see comments above the cpsid instruction in vPortSuppressTicksAndSleep()
416 /* Stop the SysTick momentarily. The time the SysTick is stopped for in vPortSuppressTicksAndSleep()
417 * is accounted for as best it can be, but using the tickless mode will in vPortSuppressTicksAndSleep()
418 * inevitably result in some tiny drift of the time maintained by the in vPortSuppressTicksAndSleep()
422 /* Use the SysTick current-value register to determine the number of in vPortSuppressTicksAndSleep()
423 * SysTick decrements remaining until the next tick interrupt. If the in vPortSuppressTicksAndSleep()
425 * ulTimerCountsForOneTick decrements remaining, not zero, because the in vPortSuppressTicksAndSleep()
426 * SysTick requests the interrupt when decrementing from 1 to 0. */ in vPortSuppressTicksAndSleep()
434 /* Calculate the reload value required to wait xExpectedIdleTime in vPortSuppressTicksAndSleep()
436 * way through the first tick period. But if the SysTick IRQ is now in vPortSuppressTicksAndSleep()
437 * pending, then clear the IRQ, suppressing the first tick, and correct in vPortSuppressTicksAndSleep()
438 * the reload value to reflect that the second tick period is already in vPortSuppressTicksAndSleep()
439 * underway. The expected idle time is always at least two ticks. */ in vPortSuppressTicksAndSleep()
453 /* Set the new reload value. */ in vPortSuppressTicksAndSleep()
456 /* Clear the SysTick count flag and set the count value back to in vPortSuppressTicksAndSleep()
466 * should not be executed again. However, the original expected idle in vPortSuppressTicksAndSleep()
480 /* Re-enable interrupts to allow the interrupt that brought the MCU in vPortSuppressTicksAndSleep()
482 * the cpsid instruction above. */ in vPortSuppressTicksAndSleep()
487 /* Disable interrupts again because the clock is about to be stopped in vPortSuppressTicksAndSleep()
488 * and interrupts that execute while the clock is stopped will increase in vPortSuppressTicksAndSleep()
489 * any slippage between the time maintained by the RTOS and calendar in vPortSuppressTicksAndSleep()
495 /* Disable the SysTick clock without reading the in vPortSuppressTicksAndSleep()
496 * portNVIC_SYSTICK_CTRL_REG register to ensure the in vPortSuppressTicksAndSleep()
498 * the time the SysTick is stopped for is accounted for as best it can in vPortSuppressTicksAndSleep()
499 * be, but using the tickless mode will inevitably result in some tiny in vPortSuppressTicksAndSleep()
500 * drift of the time maintained by the kernel with respect to calendar in vPortSuppressTicksAndSleep()
504 /* Determine whether the SysTick has already counted to zero. */ in vPortSuppressTicksAndSleep()
509 /* The tick interrupt ended the sleep (or is now pending), and in vPortSuppressTicksAndSleep()
511 * with whatever remains of the new tick period. */ in vPortSuppressTicksAndSleep()
515 * underflowed because the post sleep hook did something in vPortSuppressTicksAndSleep()
516 * that took too long or because the SysTick current-value register in vPortSuppressTicksAndSleep()
525 /* As the pending tick will be processed as soon as this in vPortSuppressTicksAndSleep()
526 * function exits, the tick value maintained by the tick is stepped in vPortSuppressTicksAndSleep()
527 * forward by one less than the time spent waiting. */ in vPortSuppressTicksAndSleep()
532 /* Something other than the tick interrupt ended the sleep. */ in vPortSuppressTicksAndSleep()
534 /* Use the SysTick current-value register to determine the in vPortSuppressTicksAndSleep()
535 * number of SysTick decrements remaining until the expected idle in vPortSuppressTicksAndSleep()
540 /* If the SysTick is not using the core clock, the current- in vPortSuppressTicksAndSleep()
541 * value register might still be zero here. In that case, the in vPortSuppressTicksAndSleep()
542 * SysTick didn't load from the reload register, and there are in vPortSuppressTicksAndSleep()
543 * ulReloadValue decrements remaining in the expected idle in vPortSuppressTicksAndSleep()
552 /* Work out how long the sleep lasted rounded to complete tick in vPortSuppressTicksAndSleep()
553 * periods (not the ulReload value which accounted for part in vPortSuppressTicksAndSleep()
557 /* How many complete tick periods passed while the processor in vPortSuppressTicksAndSleep()
561 /* The reload value is set to whatever fraction of a single tick in vPortSuppressTicksAndSleep()
568 * the SysTick is not using the core clock, temporarily configure it to in vPortSuppressTicksAndSleep()
569 * use the core clock. This configuration forces the SysTick to load in vPortSuppressTicksAndSleep()
570 * from portNVIC_SYSTICK_LOAD_REG immediately instead of at the next in vPortSuppressTicksAndSleep()
571 * cycle of the other clock. Then portNVIC_SYSTICK_LOAD_REG is ready in vPortSuppressTicksAndSleep()
572 * to receive the standard value immediately. */ in vPortSuppressTicksAndSleep()
581 /* The temporary usage of the core clock has served its purpose, in vPortSuppressTicksAndSleep()
582 * as described above. Resume usage of the other clock. */ in vPortSuppressTicksAndSleep()
587 /* The partial tick period already ended. Be sure the SysTick in vPortSuppressTicksAndSleep()
597 /* Step the tick to account for any tick periods that elapsed. */ in vPortSuppressTicksAndSleep()
609 * Setup the systick timer to generate the tick interrupts at the required
615 /* Calculate the constants required to configure the tick interrupt. */ in vPortSetupTimerInterrupt()
624 /* Stop and clear the SysTick. */ in vPortSetupTimerInterrupt()
628 /* Configure SysTick to interrupt at the requested rate. */ in vPortSetupTimerInterrupt()
644 /* Is the interrupt number a user defined interrupt? */ in vPortValidateInterruptPriority()
647 /* Look up the interrupt's priority. */ in vPortValidateInterruptPriority()
650 /* The following assertion will fail if a service routine (ISR) for in vPortValidateInterruptPriority()
658 * interrupt priorities, therefore the priority of the interrupt must in vPortValidateInterruptPriority()
662 * Interrupts that use the FreeRTOS API must not be left at their in vPortValidateInterruptPriority()
663 * default priority of zero as that is the highest possible priority, in vPortValidateInterruptPriority()
670 * The following links provide detailed information: in vPortValidateInterruptPriority()
676 /* Priority grouping: The interrupt controller (NVIC) allows the bits in vPortValidateInterruptPriority()
678 * define the interrupt's pre-emption priority bits and bits that define in vPortValidateInterruptPriority()
679 * the interrupt's sub-priority. For simplicity all bits must be defined in vPortValidateInterruptPriority()
680 * to be pre-emption priority bits. The following assertion will fail if in vPortValidateInterruptPriority()
681 * this is not the case (if some bits represent a sub-priority). in vPortValidateInterruptPriority()
683 * If the application only uses CMSIS libraries for interrupt in vPortValidateInterruptPriority()
684 * configuration then the correct setting can be achieved on all Cortex-M in vPortValidateInterruptPriority()
685 * devices by calling NVIC_SetPriorityGrouping( 0 ); before starting the in vPortValidateInterruptPriority()