[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <e1ad29bb-7766-7c9d-3191-47a5e866e07e@gmail.com>
Date: Mon, 9 Mar 2020 10:54:45 -0700
From: Eric Dumazet <eric.dumazet@...il.com>
To: David Ahern <dsahern@...il.com>,
Leslie Monis <lesliemonis@...il.com>,
Linux NetDev <netdev@...r.kernel.org>
Cc: Stephen Hemminger <stephen@...workplumber.org>,
"Mohit P . Tahiliani" <tahiliani@...k.edu.in>,
Gautam Ramakrishnan <gautamramk@...il.com>
Subject: Re: [PATCH iproute2-next] tc: pie: change maximum integer value of
tc_pie_xstats->prob
On 3/9/20 10:48 AM, Eric Dumazet wrote:
>
>
> On 3/8/20 7:49 PM, David Ahern wrote:
>> On 3/5/20 9:25 AM, Leslie Monis wrote:
>>> Kernel commit 105e808c1da2 ("pie: remove pie_vars->accu_prob_overflows"),
>>> changes the maximum value of tc_pie_xstats->prob from (2^64 - 1) to
>>> (2^56 - 1).
>>>
>>> Signed-off-by: Mohit P. Tahiliani <tahiliani@...k.edu.in>
>>> Signed-off-by: Gautam Ramakrishnan <gautamramk@...il.com>
>>> Signed-off-by: Leslie Monis <lesliemonis@...il.com>
>>> ---
>>> tc/q_pie.c | 4 ++--
>>> 1 file changed, 2 insertions(+), 2 deletions(-)
>>>
>>
>> applied to iproute2-next. Thanks
>>
>>
>
> This means that iproute2 is incompatible with old kernels.
>
> commit 105e808c1da2 ("pie: remove pie_vars->accu_prob_overflows") was wrong,
> it should not have changed user ABI.
>
> The rule is : iproute2 v-X should work with linux-<whatever-version>
>
> Since pie MAX_PROB was implicitly in the user ABI, it can not be changed,
> at least from user point of view.
>
So this kernel patch might be needed :
diff --git a/net/sched/sch_pie.c b/net/sched/sch_pie.c
index f52442d39bf57a7cf7af2595638a277e9c1ecf60..c65077f0c0f39832ee97f4e89f25639306b19281 100644
--- a/net/sched/sch_pie.c
+++ b/net/sched/sch_pie.c
@@ -493,7 +493,7 @@ static int pie_dump_stats(struct Qdisc *sch, struct gnet_dump *d)
{
struct pie_sched_data *q = qdisc_priv(sch);
struct tc_pie_xstats st = {
- .prob = q->vars.prob,
+ .prob = q->vars.prob << BITS_PER_BYTE,
.delay = ((u32)PSCHED_TICKS2NS(q->vars.qdelay)) /
NSEC_PER_USEC,
.packets_in = q->stats.packets_in,
Powered by blists - more mailing lists