On Sun, Sep 18, 2022, at 2:42 PM, Giulio Benetti wrote:
> Signed-off-by: Giulio Benetti <giulio.benetti@xxxxxxxxxxxxxxxxxxxxxx>
> ---
> V1->V2:
> * prefix subjet with 'mmc-utils:' as pointed by Avri Altman
> ---
> mmc_cmds.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/mmc_cmds.c b/mmc_cmds.c
> index 12b7802..777d649 100644
> --- a/mmc_cmds.c
> +++ b/mmc_cmds.c
> @@ -2436,7 +2436,7 @@ int do_rpmb_write_block(int nargs, char **argv)
> int ret, dev_fd, key_fd, data_fd;
> unsigned char key[32];
> uint16_t addr;
> - unsigned int cnt;
> + unsigned int cnt = 0;
> struct rpmb_frame frame_in = {
> .req_resp = htobe16(MMC_RPMB_WRITE),
> .block_count = htobe16(1)
I think this is the wrong place to initialize it, I would
suggest doing it inside of rpmb_read_counter() when returning
an error, to make it clear that this is not the intended
counter but something that is not actually meant to be used.
If you initialize local variables to bogus values, you remove
the possibility for the compiler to warn a function actually uses
something without initialization.
Arnd
