Types of EDMA Transfers
6-22
Figure 6–10. Non-2D EDMA Transfer With Frame Sync
Frame 0
Frame 1
Frame 2
EC=1
E0
E1
E2
En
E0
E1
E2
En
EC=1
FC=0
E0
E1
E2
En
+EIX
+EIX
FS=1
FS=1
FS=1
+FIX
+FIX
+EIX
EC=1
6.8.2
2-Dimensional Transfers
2-dimensional transfers are useful for imaging applications where contiguous
set of elements (referred to as array) has to be transferred on receiving a sync
event. This means there is no spacing or indexing between elements in an
array and hence, EIX is not used in 2D transfers. The number of elements in
an array makes up for the first dimension of the transfer. A group of arrays
forms the second dimension and is called a block.
6.8.2.1
R/WSYNC 2D Transfer (FS = 0)
A conceptual diagram in Figure 6–11 shows a 2-dimensional, read/write syn-
chronized transfer without frame synchronization. Since this 2-D transfer is not
frame synchronized, the R/WSYNC is the sync event on which every array or
contiguous group of elements is transferred. The example shows ‘n’ elements
in an array and number of arrays to be transferred as 3 (frame count = 2).
Frame count (FC) decrements after each array is transferred. Frame index is
added to an array’s start address to derive the next array’s start address
depending on the address update mode chosen (SUM/DUM).
When FC reaches zero and if linking is enabled (LINK = 1, see section 6.9),
the complete transfer parameters get reloaded (from the parameter reload
space in EDMA parameter RAM) after sending the last transfer request to the
address generation hardware.