On Thu, 2015-05-21 at 01:53 +0200, Thomas Graf wrote:
> On 05/15/15 at 08:06am, Herbert Xu wrote:
> > On Thu, May 14, 2015 at 07:37:56AM -0700, Eric Dumazet wrote:
> > >
> > > This solves the corruption thanks Herbert.
> >
> > Great.
> >
> > > But wasn't rhashtable meant to be faster ? ;)
> >
> > Is it, that's news to me :)
>
> Eric, can you share the scripts you used to test this?
I believe I posted the getaddrinfo C prog already.
Here is the shell script :
# cat getaddrinfo_many.sh
#!/bin/bash
TRANSACTIONS_PER_JOB=500
JOBS=200
run_netperf() {
for ((i=0; i<$JOBS; i++)); do
./getaddrinfo $TRANSACTIONS_PER_JOB &> $i.out && rm $i.out &
done
wait
return 0
}
run_once() {
seconds=$(TIMEFORMAT="%R";(time run_netperf) 2>&1)
total_transactions=$(($TRANSACTIONS_PER_JOB * $JOBS))
transactions_per_second=$(echo "print $total_transactions / $seconds" |
python)
echo $transactions_per_second
return 0
}
start_time=$(date +%s)
while true; do
result=$(run_once)
current_time=$(date +%s)
elapsed=$(($current_time - $start_time))
echo -n "TPS: $result Seconds Elapsed: $elapsed Current time: "
date
done
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
