This is an automated email from the ASF dual-hosted git repository.

xiaoxiang pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/nuttx.git

commit 5b3ef206299c801b0451848b0c5f50de3963be5a
Author: Jani Paalijarvi <[email protected]>
AuthorDate: Tue May 7 15:14:13 2024 +0300

    arm64/imx9: Allocate 64B granules instead of 256B
    
    Optimal size of granule is 64B (the dcache line size).
    We can use it now as we don't have max. 32 granules limitation anymore.
    
    Signed-off-by: Jani Paalijarvi <[email protected]>
---
 arch/arm64/src/imx9/imx9_dma_alloc.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/arch/arm64/src/imx9/imx9_dma_alloc.c 
b/arch/arm64/src/imx9/imx9_dma_alloc.c
index 172e2276db..1708e1ae69 100644
--- a/arch/arm64/src/imx9/imx9_dma_alloc.c
+++ b/arch/arm64/src/imx9/imx9_dma_alloc.c
@@ -89,11 +89,7 @@ int imx9_dma_alloc_init(void)
 {
   /* Allocate 64B granules with 64B alignment */
 
-  /* REVISIT: Use 256B granule size to get 8K maximum allocation. This is a
-   * limitation in the granule allocator itself.
-   */
-
-  dma_allocator = gran_initialize(g_dma_heap, sizeof(g_dma_heap), 8, 6);
+  dma_allocator = gran_initialize(g_dma_heap, sizeof(g_dma_heap), 6, 6);
 
   if (dma_allocator == NULL)
     {

Reply via email to