| поискавой системы для электроныых деталей |
|
AVR32AP датащи(PDF) 46 Page - ATMEL Corporation |
|
|
|||||||||||||||||||||||||||||
AVR32AP датащи(HTML) 46 Page - ATMEL Corporation |
|
46 / 181 page ![]() 46 32001A–AVR32–06/06 AVR32 3.12 Interrupt latencies The following features in AVR32 AP ensure low and deterministic interrupt latency: • Four different interrupt levels and an NMI ensures that the user can efficiently prioritize the interrupt sources. • Interrupts are autovectored, allowing the CPU to jump directly to the interrupt handler. • A shadowed interrupt context for INT3 is provided so that critical interrupt handlers can start directly without having to stack registers. • Interrupt handler code can be locked in the icache, and the corresponding page table information can be locked in the TLB. The following calculations makes the following assumptions: • The interrupt handler code is present in the icache and fetching handler instructions does not cause any MMU exceptions. • The pending interrupt is of higher priority than any executing interrupts, so that it can be handled immediately. • Any instructions in DA or D do not cause a cache miss. Any interrupts will wait until instructions in DA or D have left these pipeline stages. If the instruction in DA or D cause a cache miss, the time for the cache line to be loaded so that the instruction can complete will depend on the timing of the memory the data will be loaded from. Any time spent reloading a cache line must be added to the maximum interrupt latency calculated below. 3.12.1 Maximum interrupt latency The maximum interrupt latency occurs when a long-running instruction is present in DA. Any instruction must have left DA and D before interrupt handling will commence. The latency can be calculated as follows: 3.12.2 Minimum interrupt latency The maximum interrupt latency can be calculated as follows: Table 3-4. Maximum interrupt latency Source Delay Wait for the slowest instruction (ldm/stm) to leave DA and D 10 Wait for autovector target instruction to be fetched 4 TOTAL 14 Table 3-5. Maximum interrupt latency Source Delay DA and D are empty 0 Wait for autovector target instruction to be fetched 4 TOTAL 4 |
|
ссылки URL |
| Вашему бизинису помогли Аллдатащит? [ DONATE ] |
Что такое Аллдатащит | реклама | контакт | Конфиденциальность | Ссылка на техническое описание | обмен ссыками | поиск по производителю All Rights Reserved©Alldatasheet.com |
| Russian : Alldatasheetru.com | Korean : Alldatasheet.co.kr | Spanish : Alldatasheet.es | French : Alldatasheet.fr | Italian : Alldatasheetit.com Portuguese : Alldatasheetpt.com | Polish : Alldatasheet.pl | Vietnamese : Alldatasheet.vn Indian : Alldatasheet.in | Mexican : Alldatasheet.com.mx | British : Alldatasheet.co.uk | New Zealand : Alldatasheet.co.nz |
|
Family Site : ic2ic.com |
icmetro.com |