Filter
Top Products
System Control Coprocessor
ARM DDI 0363E Copyright © 2009 ARM Limited. All rights reserved. 4-24
ID013010 Non-Confidential, Unrestricted Access
c0, Memory Model Feature Register 2, MMFR2
The Memory Model Feature Register 2 provides information about the memory model, memory
management, and cache support operations of the processor.
The Memory Model Feature Register 2 is:
• a read-only register
• accessible in Privileged mode only.
Figure 4-17 shows the bit arrangement for Memory Model Feature Register 2.
Figure 4-17 Memory Model Feature Register 2 format
Table 4-12 shows how the bit values correspond with the Memory Model Feature Register 2
functions.
31 28 27 24 23 20 19 16 15 12 11 8 7 4 3 0
Hardware
access flag
WFI
Memory
barrier
TLB maintenance operations (unified)
TLB maintenance operations (Harward)
L1 cache maintenance range operations (Harward)
L1 background prefetch cache operations
L1 foreground prefetch cache operations
Table 4-12 Memory Model Feature Register 2 bit functions
Bits Field Function
[31:28] Hardware access flag Indicates support for Hardware Access Flag.
0x0
, no support.
[27:24] WFI Indicates support for Wait-For-Interrupt stalling.
0x1
, the processor supports Wait-For-Interrupt.
[23:20] Memory barrier Indicates support for memory barrier operations.
0x2
, the processor supports:
• DSB (formerly DWB)
• ISB (formerly Prefetch Flush)
•DMB.
[19:16] TLB maintenance
operations (unified)
Indicates support for TLB maintenance operations, unified architecture.
0x0
, no support.
[15:12] TLB maintenance
operations (Harvard)
Indicates support for TLB maintenance operations, Harvard architecture.
0x0
, no support.
[11:8] L1 cache
maintenance range
operations (Harvard)
Indicates support for cache maintenance range operations, Harvard architecture.
0x0
, no support.
[7:4] L1 background
prefetch cache
operations
Indicates support for background prefetch cache range operations, Harvard
architecture.
0x0
, no support.
[3:0] L1 foreground
prefetch cache
operations
Indicates support for foreground prefetch cache range operations, Harvard
architecture.
0x0
, no support.