RM0090 Rev 18 303/1749
RM0090 DMA controller (DMA)
338
FIFO to ensure an immediate data transfer as soon as a DMA request is triggered
by a peripheral.
• Each stream can be configured by hardware to be:
– a regular channel that supports peripheral-to-memory, memory-to-peripheral and
memory-to-memory transfers
– a double buffer channel that also supports double buffering on the memory side
• Each of the 8 streams are connected to dedicated hardware DMA channels (requests)
• Priorities between DMA stream requests are software-programmable (4 levels
consisting of very high, high, medium, low) or hardware in case of equality (request 0
has priority over request 1, etc.)
• Each stream also supports software trigger for memory-to-memory transfers (only
available for the DMA2 controller)
• Each stream request can be selected among up to 8 possible channel requests. This
selection is software-configurable and allows several peripherals to initiate DMA
requests
• The number of data items to be transferred can be managed either by the DMA
controller or by the peripheral:
– DMA flow controller: the number of data items to be transferred is software-
programmable from 1 to 65535
– Peripheral flow controller: the number of data items to be transferred is unknown
and controlled by the source or the destination peripheral that signals the end of
the transfer by hardware
• Independent source and destination transfer width (byte, half-word, word): when the
data widths of the source and destination are not equal, the DMA automatically
packs/unpacks the necessary transfers to optimize the bandwidth. This feature is only
available in FIFO mode
• Incrementing or non-incrementing addressing for source and destination
• Supports incremental burst transfers of 4, 8 or 16 beats. The size of the burst is
software-configurable, usually equal to half the FIFO size of the peripheral
• Each stream supports circular buffer management
• 5 event flags (DMA Half Transfer, DMA Transfer complete, DMA Transfer Error, DMA
FIFO Error, Direct Mode Error) logically ORed together in a single interrupt request for
each stream
To Next Page
Loading...
Need help?
General
