
Altera Corporation 9.1 3–47
November 2009 Nios II C2H Compiler User Guide
C-to-Hardware Mapping Reference
that performs Avalon-MM pipelined read transfers. Inside the
accelerator, the master port connects to a FIFO, which guarantees the
accelerator can receive data for all pending read transfers, regardless of
whether the state machine stalls.
Figure 3–24 shows the dependency graph for Example 3–37, which
demonstrates a loop that pipelines memory accesses with latency. This
example uses the connection pragma to connect the master port for
variable list to a slave memory named
my_mem_with_two_cycles_read_latency, as described in section
“Master-Slave Connections” on page 3–23.
Example 3–37. Accessing Memory with Latency
#pragma altera_accelerate connect_variable \
sum_elements/list to \
my_mem_with_two_cycles_read_latency
int sum_elements (int *list, int len)
{
int i;
int sum = 0;
for (i=0; i<len; i++)
sum += *list++;
}
Figure 3–24. Accessing Memory with Latency
Kommentare zu diesen Handbüchern