fuzzy.patch and fuzzy.patch.ipv6 - again.
Hi Harald !
Have you applied that patch ?
As I said before , the new behavior is , in practice , equivalent to the old one.
Regards.
----- Original Message -----
From: "Wojciech \"Sas\" Cieciwa" <[email protected]>
Date: Fri, 6 Jun 2003 15:31:42 +0200 (CEST)
To: Harald Welte <[email protected]>
Subject: fuzzy.patch and fuzzy.patch.ipv6 - again.
Quoted 74 lines
Hi,
I try to build p-o-m [2003 06 05] on PPC and I take some unresolved symbols. I don't know, that I send patch which resole this problem.
If yes, please appaly this.
If not here is the patch.diff -Nur
netfilter.org/patch-o-matic/base/fuzzy.patchnetfilter/patch-o-matic/base/fuzzy.patchpatch-o-matic/base/fuzzy.patch +3
-3@@ -59,7 +59,7 @@ + Expressed in percentage +*/ + -+#define PAR_LOW 0.01 ++#define PAR_LOW 1/100 +#define PAR_HIGH 1 + +static spinlock_t fuzzy_lock = SPIN_LOCK_UNLOCKED ;@@ -138,7 +138,7 @@ + howlow = mf_low(info->mean_rate,info->minimum_rate,info->maximum_rate); + + info->acceptance_rate = (u_int8_t) \ -+ ( PAR_LOW*howhigh + PAR_HIGH*howlow ) ; ++ ( howhigh*PAR_LOW + PAR_HIGH*howlow ) ; + + /* In fact , the above defuzzification would require a denominator + proportional to (howhigh+howlow) but , in this particular case ,@@ -158,7 +158,7 @@ + get_random_bytes((void *)(&random_number), 1); + + /* If within the acceptance , it can pass => don't match */ -+ if ( random_number <= 2.55 * info->acceptance_rate ) ++ if ( random_number <= (255 * info->acceptance_rate) / 100 ) + return 0 ; + else + return 1; /* It can't pass ( It matches ) */diff -Nur netfilter.org/patch-o-matic/base/fuzzy6.patch.ipv6 netfilter/patch-o-matic/base/fuzzy6.patch.ipv6
patch-o-matic/base/fuzzy6.patch.ipv6 +3
-3@@ -60,7 +60,7 @@ + Expressed in percentage +*/ + -+#define PAR_LOW 0.01 ++#define PAR_LOW 1/100 +#define PAR_HIGH 1 + +static spinlock_t fuzzy_lock = SPIN_LOCK_UNLOCKED;@@ -138,7 +138,7 @@ + howlow = mf_low(info->mean_rate,info->minimum_rate,info->maximum_rate); + + info->acceptance_rate = (u_int8_t) \ -+ (PAR_LOW * howhigh + PAR_HIGH * howlow); ++ (howhigh * PAR_LOW + PAR_HIGH * howlow); + + /* In fact, the above defuzzification would require a denominator + * proportional to (howhigh+howlow) but, in this particular case,@@ -158,7 +158,7 @@ + get_random_bytes((void *)(&random_number), 1); + + /* If within the acceptance , it can pass => don't match */ -+ if (random_number <= 2.55 * info->acceptance_rate) ++ if (random_number <= (255 * info->acceptance_rate) / 100) + return 0; + else + return 1; /* It can't pass (It matches) */Signature · 3 lines
-- {Wojciech 'Sas' Cieciwa} {Member of PLD Team } {e-mail: [email protected], http://www2.zarz.agh.edu.pl/~cieciwa}
Hime Junior
___________________________
Electronic and Control Engineer
Signature · 3 lines
__________________________________________________________ Sign-up for your own FREE Personalized E-mail at Mail.com http://www.mail.com/?sr=signup
Hi,
Quoted 3 lines
Hi Harald !
Have you applied that patch ?
The patch has not been applied yet.
You can subscribe to netfilet-cvslog to follow the changes to CVS
https://lists.netfilter.org/mailman/listinfo/netfilter-cvslog
Regards,
Maciej