netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Eric Dumazet <dada1@cosmosbay.com>
To: Neil Horman <nhorman@tuxdriver.com>
Cc: David Miller <davem@davemloft.net>,
	billfink@mindspring.com, netdev@vger.kernel.org,
	kuznet@ms2.inr.ac.ru, pekkas@netcore.fi, jmorris@namei.org,
	yoshfuji@linux-ipv6.org, kaber@trash.net, johnpol@2ka.mipt.ru
Subject: Re: [PATCH] net: implement emergency route cache rebulds when	gc_elasticity is exceeded
Date: Thu, 16 Oct 2008 14:25:47 +0200	[thread overview]
Message-ID: <48F732CB.8030704@cosmosbay.com> (raw)
In-Reply-To: <20081016114155.GA15877@hmsreliant.think-freely.org>

Neil Horman a écrit :
> On Wed, Oct 15, 2008 at 11:55:56PM -0700, David Miller wrote:
>> From: Neil Horman <nhorman@tuxdriver.com>
>> Date: Mon, 13 Oct 2008 14:26:55 -0400
>>
>>> If this meets everyones approval I think we can follow up with a
>>> patch to remove the secret interval code entirely.
>> This patch looks pretty good to me.
>>
>> Just some minor coding style nits:
>>
>>> +static void rt_secret_rebuild_oneshot(struct net *net) {
>> Openning brace on new line please.
>>
>>> +static void rt_emergency_hash_rebuild(struct net *net) {
>> Likewise.
>>
> 
> Thanks Dave, new patch, with those nits fixed up.  I also cleaned up a few
> checkpatch errors (all trailing whitespace and 80 col errors)
> 
> Best
> Neil
> 
> Signed-off-by: Neil Horman <nhorman@tuxdriver.com>

>  
> +/*
> + * While freeing expired entries, we compute average chain length
> + * and standard deviation, using fixed-point arithmetic.
> + * This to have an estimation of rt_chain_length_max
> + *  rt_chain_length_max = max(elasticity, AVG + 4*SD)
> + * We use 3 bits for frational part, and 29 (or 61) for magnitude.
> + */
> +
> +#define FRACT_BITS 3
> +#define ONE (1UL << FRACT_BITS)
> +
>  static void rt_check_expire(void)
>  {
>  	static unsigned int rover;
>  	unsigned int i = rover, goal;
>  	struct rtable *rth, **rthp;
> +	unsigned long length;
>  	u64 mult;
>  
>  	mult = ((u64)ip_rt_gc_interval) << rt_hash_log;
> @@ -784,11 +812,29 @@ static void rt_check_expire(void)
>  				if (time_before_eq(jiffies, rth->u.dst.expires)) {
>  					tmo >>= 1;
>  					rthp = &rth->u.dst.rt_next;
> +					/*
> +					 * Only bump our length if the hash
> +					 * inputs on entries n and n+1 are not
> +					 * the same, we only count entries on
> +					 * a chain with equal hash inputs once
> +					 * so that entries for different QOS
> +					 * levels, and other non-hash input
> +					 * attributes don't unfairly skew
> +					 * the length computation
> +					 */
> +					if (*rthp &&
> +					    !compare_hash_inputs(&(*rthp)->fl,
> +								 &rth->fl))
> +						length += ONE;
>  					continue;
>  				}
>  			} else if (!rt_may_expire(rth, tmo, ip_rt_gc_timeout)) {
>  				tmo >>= 1;
>  				rthp = &rth->u.dst.rt_next;
> +				if (*rthp &&
> +				    !compare_hash_inputs(&(*rthp)->fl,
> +							 &rth->fl))
> +					length += ONE;
>  				continue;
>  			}

Incomplete patch ?

You added a 'length' variable, and update it but nowhere initialize and/or read it ?

Some way to change rt_chain_length_max is needed, sysctl or dynamically...




  reply	other threads:[~2008-10-16 12:29 UTC|newest]

Thread overview: 64+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-09-29 19:12 [PATCH] net: implement emergency route cache rebulds when gc_elasticity is exceeded Neil Horman
2008-09-29 20:22 ` Eric Dumazet
2008-09-29 20:27   ` Neil Horman
2008-09-29 21:00     ` Eric Dumazet
2008-09-29 22:38       ` Neil Horman
2008-09-30  6:02         ` Eric Dumazet
2008-09-30 11:23           ` Neil Horman
2008-09-30 14:10           ` David Miller
2008-09-30 17:16             ` Eric Dumazet
2008-09-30 18:42               ` Neil Horman
2008-10-02  7:16                 ` Evgeniy Polyakov
2008-10-02 13:14                   ` Neil Horman
2008-10-01 18:08               ` Neil Horman
2008-10-02  5:01                 ` Bill Fink
2008-10-02  6:56                   ` Eric Dumazet
2008-10-02  8:15                     ` Eric Dumazet
2008-10-02 14:20                       ` Eric Dumazet
2008-10-03  0:31                       ` Neil Horman
2008-10-03 20:36                         ` Neil Horman
2008-10-06 10:49                           ` Eric Dumazet
2008-10-06 13:14                             ` Neil Horman
2008-10-06 20:54                             ` Neil Horman
2008-10-06 21:21                               ` Eric Dumazet
2008-10-06 22:52                                 ` Neil Horman
2008-10-07  5:13                                   ` Eric Dumazet
2008-10-07 10:54                                     ` Neil Horman
2008-10-13 18:26                                     ` Neil Horman
2008-10-16  6:55                                       ` David Miller
2008-10-16  9:19                                         ` Eric Dumazet
2008-10-16 21:18                                           ` David Miller
2008-10-16 11:41                                         ` Neil Horman
2008-10-16 12:25                                           ` Eric Dumazet [this message]
2008-10-16 16:36                                             ` Neil Horman
2008-10-16 23:35                                               ` Neil Horman
2008-10-17  4:53                                                 ` Eric Dumazet
2008-10-17  5:23                                                   ` David Miller
2008-10-17  5:03                                                 ` Stephen Hemminger
2008-10-17  5:06                                                 ` Stephen Hemminger
2008-10-17 10:39                                                   ` Neil Horman
     [not found]                                                     ` <48F8806A.6090306@cosmosbay.com>
     [not found]                                                       ` <20081017152328.GB23591@hmsreliant.think-freely.org>
     [not found]                                                         ` <48F8AFBE.5080503@cosmosbay.com>
2008-10-17 20:44                                                           ` Neil Horman
2008-10-18  0:54                                                             ` Neil Horman
2008-10-18  4:36                                                               ` Eric Dumazet
2008-10-18 13:30                                                                 ` Neil Horman
2008-10-20  0:07                                                                 ` Neil Horman
2008-10-20  8:12                                                                   ` Eric Dumazet
2008-10-27 19:28                                                                     ` David Miller
2008-10-02  7:13               ` Evgeniy Polyakov
2008-09-30 14:08   ` David Miller
2008-09-30 14:08 ` David Miller
2008-09-30 17:47   ` Eric Dumazet
2008-10-05  3:26   ` Herbert Xu
2008-10-05  4:45     ` Andrew Dickinson
2008-10-05 17:34       ` David Miller
2008-10-05 18:06         ` Andrew Dickinson
2008-10-06  4:21         ` Herbert Xu
2008-10-06 10:50           ` Neil Horman
2008-10-06 11:02             ` Herbert Xu
2008-10-06 12:43               ` Neil Horman
2008-09-30 14:17 ` Denis V. Lunev
2008-09-30 14:35   ` Neil Horman
2008-09-30 14:49     ` Denis V. Lunev
2008-10-05  3:17 ` Herbert Xu
2008-10-05  3:20   ` Herbert Xu
2008-10-06  0:52     ` Neil Horman

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=48F732CB.8030704@cosmosbay.com \
    --to=dada1@cosmosbay.com \
    --cc=billfink@mindspring.com \
    --cc=davem@davemloft.net \
    --cc=jmorris@namei.org \
    --cc=johnpol@2ka.mipt.ru \
    --cc=kaber@trash.net \
    --cc=kuznet@ms2.inr.ac.ru \
    --cc=netdev@vger.kernel.org \
    --cc=nhorman@tuxdriver.com \
    --cc=pekkas@netcore.fi \
    --cc=yoshfuji@linux-ipv6.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).