MPUs and MCUs
How to accelerate peripheral monitoring in low power wearables with DMA
By Brandon Hurst, hardware and embedded firmware engineer, Analog Devices T
his article explains the use cases, advantages, and disadvantages of utilising direct memory access (DMA) in embedded systems programming. The article describes how DMA interacts with peripheral and memory modules for more efficient operation of CPUs. The article will also introduce the reader to different DMA bus access architectures, and the advantages of each. One task that is common for embedded systems is managing external input. Managing input can put a lot of unnecessary computational strain on the processor, causing longer periods in active power modes and slow response times. For optimising power, preserving quick responses to events, and managing large continuous data transfers, a microcontroller with direct memory access (DMA) may offer the best solution.
Direct memory access (DMA) In system applications involving peripherals, there are many points at which a microprocessor can become bottlenecked. For instance, when managing an ADC that is constantly sending data, a processor can be interrupted so often that it struggles to accomplish other tasks. DMA is a method of moving data and minimising processor involvement in large or fast data transactions. You can think of the DMA controller as a coprocessor whose sole purpose is to interact with memory and peripherals. This allows the main processor to successfully manage a greedy peripheral, focus on another task, or even go to sleep and conserve power while data transactions happen in the background. For example, on Arm architectures, a DMA
38 April 2023
Advantages of using DMA CPU Time
Power Consumption Parallel Operation
DMA minimises the need for processor execution and interrupts, decreasing the required CPU time for data transactions.
Using DMA can yield opportunities to minimise power consumption if it allows the processor to sleep during DMA transfers.
Depending on the architectural details of the system bus, the processor may be able to execute other operations while peripheral transactions are taking place.
Disdvantages of using DMA Cost Complexity Platform dependence
Incorporating a system with DMA requires a DMA controller, and this can make a system more expensive.
While DMA can reduce the frequency of interrupts, it can increase the size and complexity of application firmware.
DMA controllers have differing internal architectures between and within manufacturers and can have different behaviour depending on their native bus access schemes.
Cache Incoherency
DMA transactions can cause logical errors to occur by writing to a cached layer of the memory hierarchy. This can be solved by using cache-coherent system architectures or by invalidating cache storage upon DMA completion.
module can operate during LP2 (sleep) or LP3 (run) modes. This can give a distinct advantage in applications that require extended battery life, such as wearable sensor hubs and smart watches.
Advantages and drawbacks DMA is useful in many digital systems, and sometimes it is even required to manage large amounts of bus traffic. It has been used in network cards, graphics cards, and even some of the original IBM PCs. That being said, incorporating DMA into a design does have some trade-offs.
Bus access and CPU cycles While DMA controllers can be incredibly
Components in Electronics
effective at conserving power or speeding up embedded systems, their implementation is not heavily standardised. There are multiple schemes for making sure that internal bus access
is not granted simultaneously with the CPU. The goal of the bus access scheme is primarily to avoid concurrent access to the same memory locations, which can lead to cache incoherency and logical errors. A single DMA controller will usually be configured to employ one of these schemes, since different hardware or firmware control may be required to use each of them. The bus access schemes used by most DMA controllers are burst, cycle-stealing, and transparent DMA. Transparent DMA can only execute a single operation at a time, but it must also wait for the processor to execute instructions in which it yields access to the desired data or address buses. Extra logic is required to verify this access restriction, and this type of DMA is generally the slowest. Transparent DMA may be advantageous in applications where one has extra processing to do that does not require access to the memory buses. The advantage in this case would be the elimination of throttling the CPU, since the processor does not have to stop operating completely.
Burst DMA occurs through infrequent large bursts, where the DMA controller sends as much data to the destination buffer as the buffer can hold. The DMA
Summary of DMA types and their pros/cons Type of DMA
Pros Burst DMA Cycle-Stealing DMA Transparent DMA Fastest type of DMA
DCPU is not idle for long contiguous periods
No throttling of CPU use needed
Cons
Relatively long periods of CPU idle time
Slower than burst DMA Slowest form of DMA
www.cieonline.co.uk
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62
