armando
9dfbc9a59c
fix(cache): fixed l2 cache suspended/frozen but l1cache triggers l2 writeback issue
2025-04-02 18:11:02 +08:00
armando
950e54fc96
Revert "fix(cache): fixed l2 cache suspended/frozen but l1cache triggers l2 writeback issue"
...
This reverts commit df1f0f51f4
2025-04-02 18:07:49 +08:00
armando
7948c53dde
Revert "fix(cache): disable branch predictor when cache freeze"
...
This reverts commit 6bb78e4573
2025-04-02 18:07:40 +08:00
armando
6bb78e4573
fix(cache): disable branch predictor when cache freeze
2025-03-26 18:59:37 +08:00
armando
df1f0f51f4
fix(cache): fixed l2 cache suspended/frozen but l1cache triggers l2 writeback issue
2025-03-26 18:59:33 +08:00
Jeroen Domburg
df4195062d
change(system): heap_caps_alloc returns aligned memory if caps indicate a need for it
...
The implicit promise of heap_alloc_caps() and friends is that the memory it
returns is fit for the purpose as requested in the caps field. Before
this commit, that did not happen; e.g. DMA-capable memory wass returned
from a correct region, but not aligned/sized to something the DMA subsystem
can handle.
This commit adds an API to the esp_mm component that is then used by the
heap component to adjust allocation alignment, caps and size dependent on
the hardware requirement of the requested allocation caps.
2024-05-30 16:02:03 +08:00
Armando
0df418facd
change(cache): update to use heap cap malloc flags
2024-04-15 15:34:51 +08:00
Armando
b658c712e7
change(cache): change esp_cache_aligned_alloc_log_e_to_w
2024-04-15 09:49:25 +08:00
Armando
f0518b3c16
feat(dma): advanced dma malloc helper
2024-04-02 14:30:14 +08:00
Armando
de70ed1c84
feat(cache): added cache_prefer_m/calloc
2024-03-26 18:03:15 +08:00
Armando
9e36994a7b
bugfix(cache): don't allow M2C direction ESP_CACHE_MSYNC_FLAG_UNALIGNED
2024-03-22 15:38:50 +08:00
Konstantin Kondrashov
3f89072af1
feat(all): Use PRIx macro in all logs
2024-03-12 11:15:53 +02:00
Armando
1f8f0a5285
feat(cache): added private API to get cache alignment requirements
2023-10-10 18:09:01 +08:00
Armando
3d4b60afc0
refactor(esp_mm): reformat code with astyle_py
2023-10-08 10:36:04 +08:00
Armando
ec27891af6
change(cache): swap cache hal arg 'type' and 'level'
2023-09-22 14:19:41 +08:00
Armando
0a1503897c
feat(cache): support esp_cache_msync for instructions
2023-09-22 14:19:41 +08:00
Armando
e8813ddd1d
feat(cache): added an helper API for cache-align-malloc
2023-09-22 14:19:41 +08:00
Armando
ea38a2e9a4
feat(cache): support cache driver on esp32p4
2023-09-22 14:19:41 +08:00
Armando
ae6d9e2b93
fix(cache): added alignment check for M2C direction
2023-07-20 12:20:42 +08:00
Armando
eb1831f8d7
fix(cache): no longer use freeze in esp_cache_msync
...
Writeback and invalidation don't need cache to be frozen first
2023-07-20 10:45:57 +08:00
Armando
da8afe7c78
feat(cache): added direction selection to esp_cache_msync API
...
ESP_CACHE_MSYNC_FLAG_DIR_C2M: From cache to memory
ESP_CACHE_MSYNC_FLAG_DIR_M2C: From memory to cache
By default, if no direction flag is set, it will fallback C2M direction.
For M2C direction, now this API will do an invalidation.
2023-07-11 11:19:10 +08:00
Armando
fda9746bb8
esp_mm: cache_msync API
2023-02-28 10:42:22 +08:00
