On Sun, Feb 27, 2011 at 11:28:18PM +0300, Stanislav Fomichev wrote:
> --- a/drivers/staging/brcm80211/brcmsmac/wlc_ampdu.c
> +++ b/drivers/staging/brcm80211/brcmsmac/wlc_ampdu.c
> @@ -637,11 +637,12 @@ wlc_sendampdu(struct ampdu_info *ampdu, wlc_txq_info_t *qi,
> mcl |= TXC_STARTMSDU;
> rts = (struct ieee80211_rts *)&txh->rts_frame;
> fc = le16_to_cpu(rts->frame_control);
> - if ((fc & FC_KIND_MASK) == FC_RTS) {
> +
> + if (ieee80211_is_rts(rts->frame_control)) {
fc isn't used anymore so it can be removed.
> @@ -6051,7 +6044,8 @@ wlc_d11hdrs_mac80211(struct wlc_info *wlc, struct ieee80211_hw *hw,
> txrate[1]->count = 0;
>
> /* (2) PROTECTION, may change rspec */
> - if ((ieee80211_is_data(fc) || ieee80211_is_mgmt(fc)) &&
> + if ((ieee80211_is_data(h->frame_control) ||
> + ieee80211_is_mgmt(h->frame_control)) &&
This is a bugfix because fc is cpu endian and h->frame_control is big
endian. Thank you for fixing the bug, :), but please not behavior
changes in the commit message.
> @@ -6684,7 +6680,7 @@ wlc_dotxstatus(struct wlc_info *wlc, tx_status_t *txs, u32 frm_tx2)
> tx_rts_count =
> (txs->status & TX_STATUS_RTS_RTX_MASK) >> TX_STATUS_RTS_RTX_SHIFT;
>
> - lastframe = (fc & IEEE80211_FCTL_MOREFRAGS) == 0;
> + lastframe = ieee80211_has_morefrags(h->frame_control);
This is reversed. It should be:
if (ieee80211_has_morefrags(h->frame_control))
lastframe = false;
else
lastframe = true;
regards,
dan carpenter
_______________________________________________
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxx
http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
