DSCP match/target patches
PATCH 1/1 DSCP match/target patches
Hi,
Here is a simple patch to the DSCP match to allow the use of DiffServ class names in the matches, in a similar manner to the target.
Also attached is a one line fix for a typo in the DSCP target where CS6 was defined twice in the match list.
Iain
libipt_DSCP.c renamed from libipt_DSCP.c.orig +1 -1
@@ -35,7 +35,7 @@ { "CS4", 0x20 }, { "CS5", 0x28 }, { "CS6", 0x30 }, - { "CS6", 0x38 }, + { "CS7", 0x38 }, { "BE", 0 }, { "AF11", 0x0a }, { "AF12", 0x0c },
libipt_dscp.c renamed from libipt_dscp.c.orig +67 -1
103 lines not on this page · raw message
Quoted 7 lines · On Mon, Jun 10, 2002 at 01:49:13PM +0000, [email protected] wrote:
Hi,
Here is a simple patch to the DSCP match to allow the use of DiffServ class names in the matches, in a similar manner to the target.
Also attached is a one line fix for a typo in the DSCP target where CS6 was defined twice in the match list.
thanks for your contribution. What about putting all the class name <-> numerical resolving in a seperate .c file, which is then compiled once and linked into both libipt_dscp.so and libipt_DSCP.so ?
I think this would be the ideal case, without any code duplication.
Quoted 1 line
Iain
Signature · 5 lines
Live long and prosper - Harald Welte / [email protected] http://www.gnumonks.org/ ============================================================================ GCS/E/IT d- s-: a-- C+++ UL++++$ P+++ L++++$ E--- W- N++ o? K- w--- O- M+ V-- PS++ PE-- Y++ PGP++ t+ 5-- !X !R tv-- b+++ !DI !D G+ e* h--- r++ y+(*)
Harald
Ok, here is the first cut of a set of patches.
The only thing missing is the Makefile modifications. My knowledge of the fine art of Makefile writing is sorely lacking, so any help on that front, from anybody, would be greatly appreciated.
The helper functions have been placed in a libipt_dscp_helper.c file in the extensions directory.
Both the DSCP match and target have been modified to use the new functions.
Iain
Quoted 23 lines · Harald Welte
Quoted 7 lines · On Mon, Jun 10, 2002 at 01:49:13PM +0000, [email protected] wrote:
Hi,
Here is a simple patch to the DSCP match to allow the use of DiffServ class names in the matches, in a similar manner to the target.
Also attached is a one line fix for a typo in the DSCP target where CS6 was defined twice in the match list.
thanks for your contribution. What about putting all the class name <-> numerical resolving in a seperate .c file, which is then compiled once and linked into both libipt_dscp.so and libipt_DSCP.so ?
I think this would be the ideal case, without any code duplication.
Quoted 1 line
Iain
Signature · 5 lines
Live long and prosper - Harald Welte / [email protected] http://www.gnumonks.org/ ============================================================================ GCS/E/IT d- s-: a-- C+++ UL++++$ P+++ L++++$ E--- W- N++ o? K- w--- O- M+ V-- PS++ PE-- Y++ PGP++ t+ 5-- !X !R tv-- b+++ !DI !D G+ e* h--- r++ y+(*)
Signature · 1 line
Iain Barnes <[email protected]>
extensions/libipt_DSCP.c +4 -42
@@ -18,39 +18,8 @@#include <linux/netfilter_ipv4/ip_tables.h> #include <linux/netfilter_ipv4/ipt_DSCP.h> +extern unsigned int class_to_dscp (const char *name); -/* see http://www.iana.org/assignments/dscp-registry */ - -static struct ds_class -{ -char *class; -unsigned int dscp; -} ds_classes[] = -{ - {"CS0" ,0 }, - {"CS1" ,0x08 }, - {"CS2" ,0x10 }, - {"CS3" ,0x18 }, - {"CS3" ,0x18 }, - {"CS4" ,0x20 }, - {"CS5" ,0x28 }, - {"CS6" ,0x30 }, - {"CS6" ,0x38 }, - {"BE" ,0 }, - {"AF11" ,0x0a }, - {"AF12" ,0x0c }, - {"AF13" ,0x0e }, - {"AF21" ,0x12 }, - {"AF22" ,0x14 }, - {"AF23" ,0x16 }, - {"AF31" ,0x1a }, - {"AF32" ,0x1c }, - {"AF33" ,0x1e }, - {"AF41" ,0x22 }, - {"AF42" ,0x24 }, - {"AF43" ,0x26 }, - {"EF" ,0x2e } -};static void init (struct ipt_entry_target *t,unsigned int *nfcache) {
@@ -99,17 +68,10 @@static void parse_class (const unsigned char *s,struct ipt_DSCP_info *dinfo) { -int i; - -for (i =0 ; i <sizeof (ds_classes) /sizeof (struct ds_class ); i++) { -if (!strncasecmp (s, ds_classes[i].class, -strlen (ds_classes[i].class))) { - dinfo->dscp = (u_int8_t )ds_classes[i].dscp; -return ; - } - } +unsigned int dscp =class_to_dscp (s); -exit_error (PARAMETER_PROBLEM ,"Invalid DSCP class value '%s'" , s); +/* Assign the value */ + dinfo->dscp = (u_int8_t )dscp; }
extensions/libipt_dscp.c +34 -1
@@ -6,6 +6,8 @@* * libipt_dscp.c borrowed heavily from libipt_tos.c * +* --class support added by Iain Barnes +* * For a list of DSCP codepoints see * http://www.iana.org/assignments/dscp-registry *
@@ -19,6 +21,9 @@#include <linux/netfilter_ipv4/ip_tables.h> #include <linux/netfilter_ipv4/ipt_dscp.h> + +extern unsigned int class_to_dscp (const char *name); +static void init (struct ipt_entry_match *m,unsigned int *nfcache) { *nfcache |=NFC_IP_TOS ;
@@ -30,12 +35,18 @@"DSCP match v%s options\n" "[!] --dscp value Match DSCP codepoint with numerical value\n" " This value can be in decimal (ex: 32)\n" -" or in hex (ex: 0x20)\n" ,IPTABLES_VERSION +" or in hex (ex: 0x20)\n" +"[!] --class name Match the DiffServ class. This value may\n" +" be any of the BE,EF, AFxx or CSx classes\n" +"\n" +" These two options are mutually exclusive !\n" + ,IPTABLES_VERSION ); }static struct option opts[] = { {"dscp" ,1 ,0 ,'F' }, + {"class" ,1 ,0 ,'G' }, {0 } };
@@ -56,6 +67,17 @@return ; } + +static void +parse_class (const char *s,struct ipt_dscp_info *dinfo) +{ +unsigned int dscp =class_to_dscp (s); + +/* Assign the value */ + dinfo->dscp = (u_int8_t )dscp; +} + + static int parse(int c, char **argv, int invert, unsigned int *flags, const struct ipt_entry *entry,
@@ -72,6 +94,17 @@ "DSCP match: Only use --dscp ONCE!");check_inverse (optarg, &invert, &optind,0 );parse_dscp (argv[optind-1 ], dinfo); +if (invert) + dinfo->invert =1 ; + *flags =1 ; +break ; + +case 'G' : +if (*flags) +exit_error (PARAMETER_PROBLEM , +"DSCP match: Only use --class ONCE!" ); +check_inverse (optarg, &invert, &optind,0 ); +parse_class (argv[optind -1 ], dinfo);if (invert) dinfo->invert =1 ; *flags =1 ;
extensions/libipt_dscp_helper.c +82 -0
82 lines not on this page · raw message
Quoted 5 lines · Iain Barnes
Harald
The only thing missing is the
Makefilemodifications. My knowledge of the fine art ofMakefilewriting is sorely lacking, so any help on that front, from anybody, would be greatly appreciated.
I've now managed to solve the problem without Makefile modifications. Feel free to look at the current CVS code :)
Quoted 1 line
Iain
Signature · 5 lines
Live long and prosper - Harald Welte / [email protected] http://www.gnumonks.org/ ============================================================================ GCS/E/IT d- s-: a-- C+++ UL++++$ P+++ L++++$ E--- W- N++ o? K- w--- O- M+ V-- PS++ PE-- Y++ PGP++ t+ 5-- !X !R tv-- b+++ !DI !D G+ e* h--- r++ y+(*)