On Sat, Jul 03, 2021 at 04:00:49PM +0100, Sean Young wrote:
> Signed-off-by: Sean Young <sean@xxxxxxxx>
> ---
> utils/ir-ctl/ir-ctl.c | 9 ++++++---
> 1 file changed, 6 insertions(+), 3 deletions(-)
>
> diff --git a/utils/ir-ctl/ir-ctl.c b/utils/ir-ctl/ir-ctl.c
> index 3c3bcca1..66a257a7 100644
> --- a/utils/ir-ctl/ir-ctl.c
> +++ b/utils/ir-ctl/ir-ctl.c
> @@ -933,7 +933,7 @@ static int lirc_options(struct arguments *args, int fd, unsigned features)
> if (features & LIRC_CAN_SET_TRANSMITTER_MASK) {
> rc = ioctl(fd, LIRC_SET_TRANSMITTER_MASK, &args->emitters);
> if (rc > 0)
> - fprintf(stderr, _("warning: %s: failed to set send transmitters: only %d available\n"), dev, rc);
> + fprintf(stderr, _("warning: %s: failed to set send transmitters: only %d available\n"), dev, ffs(~rc) - 1);
I was mistaken; the ioctl does return the tx emitter count, not the valid
mask.
All the existing drivers except rc-loopback return the count, not the mask.
The documentation also says so.
I'll have to revert.
Sean
> else if (rc < 0)
> fprintf(stderr, _("warning: %s: failed to set send transmitters: %m\n"), dev);
> } else
> @@ -1026,8 +1026,11 @@ static void lirc_features(struct arguments *args, int fd, unsigned features)
> fprintf(stderr, _("warning: %s: device supports setting transmitter mask but returns 0 as number of transmitters\n"), dev);
> else if (rc < 0)
> fprintf(stderr, _("warning: %s: device supports setting transmitter mask but returns: %m\n"), dev);
> - else
> - printf(_(" - Set transmitter (%d available)\n"), rc);
> + else {
> + unsigned count = ffs(~rc) - 1;
> +
> + printf(_(" - Set transmitter (%d available)\n"), count);
> + }
> }
> } else if (features & LIRC_CAN_SEND_LIRCCODE) {
> printf(_(" - Device can send using device dependent LIRCCODE mode (not supported)\n"));
> --
> 2.31.1
