dlmalloc seems to work more efficient with this chunk size.
Copying a bigger file (3MB) takes 271ms vs. 125ms on a i.MX27
board. Even bigger chunk sizes do not further improve performance.
Signed-off-by: Sascha Hauer <s.hauer@xxxxxxxxxxxxxx>
---
fs/ramfs.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/fs/ramfs.c b/fs/ramfs.c
index db417e1..5e352f2 100644
--- a/fs/ramfs.c
+++ b/fs/ramfs.c
@@ -30,7 +30,7 @@
#include <linux/stat.h>
#include <xfuncs.h>
-#define CHUNK_SIZE 4096
+#define CHUNK_SIZE (4096 * 2)
struct ramfs_chunk {
char *data;
--
1.7.7.3
_______________________________________________
barebox mailing list
barebox@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/barebox
