If CONFIG_USE_ARCH_MEMCPY is selected, let's use the assembly optimized memcpy implementation for the 'cp' command.
Currently only Blackfin uses memcpy for the 'cp' command, so extend this to the CONFIG_USE_ARCH_MEMCPY users.
Tested on a mx6qsabreauto board where a 5x gain in performance is seen when reading 10MB from the parallel NOR memory.
Signed-off-by: Fabio Estevam fabio.estevam@nxp.com --- Sending as RFC as I don't know if any extra checks would be required here, so I appreciate any feedback.
cmd/mem.c | 5 +++++ 1 file changed, 5 insertions(+)
diff --git a/cmd/mem.c b/cmd/mem.c index a690957..c1d9a7c 100644 --- a/cmd/mem.c +++ b/cmd/mem.c @@ -465,6 +465,11 @@ static int do_mem_cp(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[]) } #endif
+#ifdef CONFIG_USE_ARCH_MEMCPY + memcpy((void *)dest, (void *)addr, count * size); + return 0; +#endif + bytes = size * count; buf = map_sysmem(dest, bytes); src = map_sysmem(addr, bytes);