Hey,
I wanted to modify iw to support NL80211_CMD_SET_TX_BITRATE_MASK and thus
wanted to be clear about the expected input from the user-space. I see that
the function rateset_to_mask converts an array of u8 received from userspace
via netlink to a u32 mask for legacy rate.
It uses following lines of code to achieve that:
*for (i = 0; i < rates_len; i++) {
int rate = (rates[i] & 0x7f) * 5;
int ridx;
for (ridx = 0; ridx < sband->n_bitrates; ridx++) {
struct ieee80211_rate *srate =
&sband->bitrates[ridx];
if (rate == srate->bitrate) {
mask |= 1 << ridx;
break;
}
}
if (ridx == sband->n_bitrates)
return 0; /* rate not found */
}*
What I don't understand is the line * int rate = (rates[i] &
0x7f) * 5;*
*
*
I am working with ath9k driver and the sband->bitrates array contains
following values in that case:
*2 RATE(10, 0x1b, 0),*
*4 RATE(20, 0x1a, IEEE80211_RATE_SHORT_PREAMBLE),*
*11 RATE(55, 0x19, IEEE80211_RATE_SHORT_PREAMBLE),*
*22 RATE(110, 0x18, IEEE80211_RATE_SHORT_PREAMBLE),*
*12 RATE(60, 0x0b, 0),*
*18 RATE(90, 0x0f, 0),*
*24 RATE(120, 0x0a, 0),*
*36 RATE(180, 0x0e, 0),*
*48 RATE(240, 0x09, 0),*
*72 RATE(360, 0x0d, 0),*
*96 RATE(480, 0x08, 0),*
*108 RATE(540, 0x0c, 0),*
But what beats me is that *why are we multiplying the incoming rate numbers
by 5*?
A prompt reply shall be highly appreciated.
--
Rakesh
_______________________________________________
ath9k-devel mailing list
[email protected]
https://lists.ath9k.org/mailman/listinfo/ath9k-devel
