On Thu, 2017-04-13 at 10:48 -0400, Hal Rosenstock wrote:
> %d is not good as LIDs above 32K print as negative numbers.
> %#x seems to be the right choice here; %u was used by IB
> management tools for unicast LID printing.
Hello Hal,
Although the patch itself looks fine to me: LIDs are converted from network
to host format using be16toh(). That function returns an unsigned 16-bit
number. These numbers are converted to type "int" before pr_err() is called.
The type "int" is at least 32 bits on all architectures supported by Linux.
So how could using %d result in a negative number being printed?
> diff --git a/srp_daemon/srp_daemon.c b/srp_daemon/srp_daemon.c
> index 59a6137..4d8476e 100644
> --- a/srp_daemon/srp_daemon.c
> +++ b/srp_daemon/srp_daemon.c
> @@ -632,7 +632,7 @@ recv:
> ret = umad_status(in_mad);
> if (ret) {
> pr_err(
> - "bad MAD status (%u) from lid %d\n",
> + "bad MAD status (%u) from lid %#x\n",
> ret, (uint16_t) be16toh(out_mad->hdr.addr.lid));
> return -ret;
If you have to repost this patch, please remove the (uint16_t) cast since it's
superfluous.
Thanks,
Bart.--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
