EPXA4 ALTERA [Altera Corporation], EPXA4 Datasheet - Page 25

no-image

EPXA4

Manufacturer Part Number
EPXA4
Description
Excalibur Devices
Manufacturer
ALTERA [Altera Corporation]
Datasheet

Available stocks

Company
Part Number
Manufacturer
Quantity
Price
Part Number:
EPXA4F1020C1
Manufacturer:
ALTERA
Quantity:
3 000
Part Number:
EPXA4F1020C1
Manufacturer:
ALTERA
0
Part Number:
EPXA4F1020C1ES
Manufacturer:
ALTERA
Quantity:
3 000
Part Number:
EPXA4F1020C1ES
Manufacturer:
ALTERA
0
Part Number:
EPXA4F1020C2
Manufacturer:
ALTERA
0
Part Number:
EPXA4F1020C2ES
Manufacturer:
ALTERA
Quantity:
3 000
Part Number:
EPXA4F1020C2ES
Manufacturer:
ALTERA
0
Part Number:
EPXA4F1020C3
Manufacturer:
ALTERA
Quantity:
3 000
Part Number:
EPXA4F1020C3
Manufacturer:
ALTERA
0
Part Number:
EPXA4F672C1
Manufacturer:
ALTERA
Quantity:
784
Part Number:
EPXA4F672C1
Manufacturer:
ALTERA
Quantity:
30
A.2.7
Altera Corporation
618
619
620
621
Table 7. Example of how to calculate the correct location of a stall (Part 2 of 2)
FIFOFULL LOW for one cycle during overflow—Category 3
Note: If 2018 is also an LDR, rather than an ADD, then it would not be
possible to determine which of the two instructions 2018 or 201C caused
the stall.
Description
Although FIFOFULL correctly becomes asserted when the space available
in the FIFO is less than the minimum value specified in the FIFOFULL
Level register of the ETM (register 0x0b ), it incorrectly becomes de-
asserted for a single cycle when the FIFO overflows.
Conditions
This occurs for one cycle, on the same cycle in which the FIFO overflows.
Implications
One extra cycle of trace can be lost during FIFO overflow. Since this case
only occurs when overflow is already about to occur, the ability of
FIFOFULL to prevent overflow is unaffected.
Workaround
None required. The erratum causes the loss of an extra cycle of trace when
there is an overflow, so the effect is not significant.
Implications of workaround
No implications.
IE for 2018
IE for 201C
IE for 2020
IE for 2024
WT
IE for 201C
IE for 2020
IE for 2024
The stall is far more likely to
correspond to 201C, as this causes a
LDR to on-chip memory. We can
therefore deduce that this LDR
caused precisely 1 stall cycle.
The stall could not correspond to this
load, as it is not 6 cycles ahead of a
detected external stall in the trace.
Note that the total number of cycles
taken for the sequence of instructions
is reported correctly.
25

Related parts for EPXA4