[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20170729071958.GA1858@nanopsycho>
Date: Sat, 29 Jul 2017 09:19:58 +0200
From: Jiri Pirko <jiri@...nulli.us>
To: Jamal Hadi Salim <jhs@...atatu.com>
Cc: davem@...emloft.net, netdev@...r.kernel.org,
xiyou.wangcong@...il.com, dsahern@...il.com,
eric.dumazet@...il.com, mrv@...atatu.com,
simon.horman@...ronome.com, alex.aring@...il.com
Subject: Re: [PATCH net-next v11 3/4] net sched actions: dump more than
TCA_ACT_MAX_PRIO actions per batch
Sat, Jul 29, 2017 at 12:10:20AM CEST, jhs@...atatu.com wrote:
>On 17-07-28 11:45 AM, Jiri Pirko wrote:
>> Fri, Jul 28, 2017 at 05:08:10PM CEST, jhs@...atatu.com wrote:
>> > On 17-07-28 10:52 AM, Jamal Hadi Salim wrote:
>> > > On 17-07-28 10:12 AM, Jiri Pirko wrote:
>
>
>> > /*disallow invalid selector */
>> > if ((bf->selector & *valid_flags_allowed) > *valid_flags_allowed)
>>
>> I don't get the ">"....
>> Just (bf->selector & ~*valid_flags_allowed) should be enought, no?
>>
>
>It may be enough - I will try it out tommorow.
>I was worried about the selector having more bits than the allowed flags.
That is what the code I wrote above do. And I think that your code does
not (if you don't assume that allowed flags are always starting from the
least significant bit)
>
>cheers,
>jamal
Powered by blists - more mailing lists