Re: [PATCH v2] mac80211: Fix output of minstrels rc_stats

[Arnd Hannemann <hannemann@nets.rwth-aachen.de>]

Julian Calaby wrote:
> On Tue, Aug 25, 2009 at 04:51, Arnd
> Hannemann<hannemann@nets.rwth-aachen.de> wrote:
>> An integer overflow in the minstrel debug code prevented the
>> throughput to be displayed correctly. This patch fixes that,
>> by permutating operations like proposed by Pavel Roskin.
>>
>> Signed-off-by: Arnd Hannemann <hannemann@nets.rwth-aachen.de>
>> ---
>>  net/mac80211/rc80211_minstrel_debugfs.c |    2 +-
>>  1 files changed, 1 insertions(+), 1 deletions(-)
>>
>> diff --git a/net/mac80211/rc80211_minstrel_debugfs.c b/net/mac80211/rc80211_minstrel_debugfs.c
>> index 98f4807..3d72ec5 100644
>> --- a/net/mac80211/rc80211_minstrel_debugfs.c
>> +++ b/net/mac80211/rc80211_minstrel_debugfs.c
>> @@ -83,7 +83,7 @@ minstrel_stats_open(struct inode *inode, struct file *file)
>>                p += sprintf(p, "%3u%s", mr->bitrate / 2,
>>                                (mr->bitrate & 1 ? ".5" : "  "));
>>
>> -               tp = ((mr->cur_tp * 96) / 18000) >> 10;
>> +               tp = mr->cur_tp / ((18000 << 10) / 96);
> 
> Sorry about being so late, but wouldn't:
> 
> tp = ((mr->cur_tp * 2) / 375) >> 10;
> 
> also work? (Assuming that the numbers in the constant aren't important)

I needed a while to figure out why those constants are used, but finally
they made some sense, so I think its best to preserve them.

mr->cur_tp is the time to send one 1200 byte packet (9600 bits),
the loss probelity is scaled between 0-18000 to reduce rounding error.

> 
> or even:
> 
> tp = (mr->cur_tp / 375) >> 9;

See above.

Best regards,
Arnd