ARM R4 Manuale Utente
Level One Memory System
ARM DDI 0363E
Copyright © 2009 ARM Limited. All rights reserved.
8-18
ID013010
Non-Confidential, Unrestricted Access
8.5
About the caches
The L1 memory system can be configured to include instruction and data caches of varying
sizes. You can configure whether the cache controller is included and, if it is, configure the size
of each cache independently. The cached instructions or data are fetched from external memory
using the L2 memory interface. The cache controllers use RAMs that are integrated into the
Cortex-R4 macrocell during implementation.
sizes. You can configure whether the cache controller is included and, if it is, configure the size
of each cache independently. The cached instructions or data are fetched from external memory
using the L2 memory interface. The cache controllers use RAMs that are integrated into the
Cortex-R4 macrocell during implementation.
Any access that is not for a TCM is handled by the appropriate cache controller. If the access is
to Cacheable memory, and the cache is enabled, a lookup is performed in the cache and, if found
in the cache, that is, a cache hit, the data is fetched from or written into the cache. When the
cache is not enabled and for Non-cacheable memory, the accesses are performed using the L2
memory interface.
to Cacheable memory, and the cache is enabled, a lookup is performed in the cache and, if found
in the cache, that is, a cache hit, the data is fetched from or written into the cache. When the
cache is not enabled and for Non-cacheable memory, the accesses are performed using the L2
memory interface.
Both caches allocate a memory location to a cache line on a cache miss because of a read, that
is, all Cacheable locations are Read-Allocate (RA). In addition, the data cache can allocate on a
write access if the memory location is marked as Write-Allocate (WA). When a cache line is
allocated, the appropriate memory is fetched into a linefill buffer by the L2 memory interface
before being written to the cache. See Linefill buffers and the AXI master interface on page 9-4.
The linefill buffers always fetch the requested data first, and then the rest of the cache line. This
enables the data read to be used by the pipeline without waiting for the linefill to complete and
is known as critical word first and non-blocking behavior. If an error is reported to the L2
memory interface for a linefill, the linefill does not update the cache RAMs, but an abort is only
generated if the error was reported on the critical word.
is, all Cacheable locations are Read-Allocate (RA). In addition, the data cache can allocate on a
write access if the memory location is marked as Write-Allocate (WA). When a cache line is
allocated, the appropriate memory is fetched into a linefill buffer by the L2 memory interface
before being written to the cache. See Linefill buffers and the AXI master interface on page 9-4.
The linefill buffers always fetch the requested data first, and then the rest of the cache line. This
enables the data read to be used by the pipeline without waiting for the linefill to complete and
is known as critical word first and non-blocking behavior. If an error is reported to the L2
memory interface for a linefill, the linefill does not update the cache RAMs, but an abort is only
generated if the error was reported on the critical word.
If all the cache lines in a set are valid, to allocate a different address to the cache, the cache
controller must evict a line from the cache.
controller must evict a line from the cache.
Writes accesses that hit in the cache are written into the cache RAMs. If the memory location is
marked as Write-Through (WT), the write is also performed on the L2 memory interface, so that
the data stored in the RAM remains coherent with the external memory system. If the memory
is Write-Back (WB), the cache line is marked as dirty, and the write is only performed on the L2
memory interface when the line is evicted. When a dirty cache line is evicted, the data is passed
to the Eviction Buffer in the L2 memory interface to be written to the external memory system.
See Eviction buffer on page 9-5 for more information.
marked as Write-Through (WT), the write is also performed on the L2 memory interface, so that
the data stored in the RAM remains coherent with the external memory system. If the memory
is Write-Back (WB), the cache line is marked as dirty, and the write is only performed on the L2
memory interface when the line is evicted. When a dirty cache line is evicted, the data is passed
to the Eviction Buffer in the L2 memory interface to be written to the external memory system.
See Eviction buffer on page 9-5 for more information.
The cache controllers also manage the cache maintenance operations described in Cache
maintenance operations on page 8-19.
maintenance operations on page 8-19.
Each cache can also be configured with either parity or ECC error checking schemes. If an error
checking scheme is implemented and enabled, then the tags associated with each line, and data
read from the cache are checked whenever a lookup is performed in the cache. See Cache error
detection and correction on page 8-20 for more information.
checking scheme is implemented and enabled, then the tags associated with each line, and data
read from the cache are checked whenever a lookup is performed in the cache. See Cache error
detection and correction on page 8-20 for more information.
For more information on the general rules about memory attributes and behavior, see the ARM
Architecture Reference Manual.
Architecture Reference Manual.
8.5.1
Store buffer
The cache controller includes a store buffer to hold data before it is written to the cache RAMs
or passed to the AXI master interface. The store buffer has four entries. Each entry can contain
up to 64 bits of data and a 32-bit address. All write requests from the data-side that are not to a
TCM interface are stored in the store buffer.
or passed to the AXI master interface. The store buffer has four entries. Each entry can contain
up to 64 bits of data and a 32-bit address. All write requests from the data-side that are not to a
TCM interface are stored in the store buffer.