[Click] [Patch] fix endianess in radiotap encap

[Roberto Riggio]

  is this patch suitable for merging?

Il 11/05/2010 20:12, Roberto Riggio ha scritto:
> This patch fixes the following issues
> 1) len and flag must always be LE.
> 2) the radiotapencap element specifies that the data_retries entry
> is present, however it does not check the parameter is really different
> from zero. this is a problem on the ath9k driver that in this case does not
> transmitt the frame
>
> --
> diff --git a/elements/wifi/radiotapencap.cc b/elements/wifi/radiotapencap.cc
> index 37d2d69..d481f1c 100644
> --- a/elements/wifi/radiotapencap.cc
> +++ b/elements/wifi/radiotapencap.cc
> @@ -88,14 +88,16 @@ RadiotapEncap::simple_action(Packet *p)
>          memset(crh, 0, sizeof(struct click_radiotap_header));
>
>          crh->wt_ihdr.it_version = 0;
> -      crh->wt_ihdr.it_len = sizeof(struct click_radiotap_header);
> -      crh->wt_ihdr.it_present = CLICK_RADIOTAP_PRESENT;
> +      crh->wt_ihdr.it_len = cpu_to_le16(sizeof(struct
> click_radiotap_header));
> +      crh->wt_ihdr.it_present = cpu_to_le32(CLICK_RADIOTAP_PRESENT);
>
>          crh->wt_rate = ceh->rate;
>          crh->wt_txpower = ceh->power;
>          crh->wt_rts_retries = 0;
>          if (ceh->max_tries>  0) {
>              crh->wt_data_retries = ceh->max_tries - 1;
> +      } else {
> +          crh->wt_data_retries = WIFI_MAX_RETRIES+1;
>          }
>      }
> _______________________________________________
> click mailing list
> click at amsterdam.lcs.mit.edu
> https://amsterdam.lcs.mit.edu/mailman/listinfo/click
>