mirror of
https://github.com/espressif/esp-idf.git
synced 2024-10-05 20:47:46 -04:00
32 lines
1.5 KiB
Plaintext
32 lines
1.5 KiB
Plaintext
|
menu "ESP-MM: Memory Management Configurations"
|
||
|
|
||
|
config ESP_MM_CACHE_MSYNC_C2M_CHUNKED_OPS
|
||
|
depends on SOC_CACHE_WRITEBACK_SUPPORTED
|
||
|
bool "Enable esp_cache_msync C2M chunked operation"
|
||
|
help
|
||
|
`esp_cache_msync` C2M direction takes critical sections, which means during
|
||
|
the operation, the interrupts are disabled. Whereas Cache writebacks for
|
||
|
large buffers could be especially time intensive, and might cause interrupts
|
||
|
to be disabled for a significant amount of time.
|
||
|
|
||
|
Sometimes you want other ISRs to be responded during this C2M process.
|
||
|
This option is to slice one C2M operation into multiple chunks,
|
||
|
with CONFIG_ESP_MM_CACHE_MSYNC_C2M_CHUNKED_OPS_MAX_LEN max len. This will give you
|
||
|
a breath during the C2M process as sometimes the C2M process is quite long.
|
||
|
|
||
|
Note if the buffer processed by the `esp_cache_msync` (C2M sliced) is interrupted by an ISR,
|
||
|
and this ISR also accesses this buffer, this may lead to data coherence issue.
|
||
|
|
||
|
config ESP_MM_CACHE_MSYNC_C2M_CHUNKED_OPS_MAX_LEN
|
||
|
hex "Max len in bytes per C2M chunk"
|
||
|
depends on ESP_MM_CACHE_MSYNC_C2M_CHUNKED_OPS
|
||
|
range 0 0x80000
|
||
|
default 0x20000 if IDF_TARGET_ESP32P4
|
||
|
default 0x2000 if IDF_TARGET_ESP32S2
|
||
|
default 0x8000 if IDF_TARGET_ESP32S3
|
||
|
help
|
||
|
Max len in bytes per C2M chunk, operations with size over the max len will be
|
||
|
sliced into multiple chunks.
|
||
|
|
||
|
endmenu
|