| Message ID | 1279892621.2481.53.camel@edumazet-laptop |
|---|---|
| State | Not Applicable, archived |
| Delegated to: | David Miller |
| Headers | show |
On 23.07.2010 15:43, Eric Dumazet wrote: > extension: add xt_cpu match > > Kernel 2.6.36 supports xt_cpu match > > In some situations a CPU match permits a better spreading of > connections, or select targets only for a given cpu. > > With Remote Packet Steering or multiqueue NIC and appropriate IRQ > affinities, we can distribute trafic on available cpus, per session. > (all RX packets for a given flow are handled by a given cpu) > > Some legacy applications being not SMP friendly, one way to scale a > server is to run multiple copies of them. > > Instead of randomly choosing an instance, we can use the cpu number as a > key so that softirq handler for a whole instance is running on a single > cpu, maximizing cache effects in TCP/UDP stacks. > > Using NAT for example, a four ways machine might run four copies of > server application, using a separate listening port for each instance, > but still presenting an unique external port : > > iptables -t nat -A PREROUTING -p tcp --dport 80 -m cpu --cpu 0 \ > -j REDIRECT --to-port 8080 > > iptables -t nat -A PREROUTING -p tcp --dport 80 -m cpu --cpu 1 \ > -j REDIRECT --to-port 8081 > > iptables -t nat -A PREROUTING -p tcp --dport 80 -m cpu --cpu 2 \ > -j REDIRECT --to-port 8082 > > iptables -t nat -A PREROUTING -p tcp --dport 80 -m cpu --cpu 3 \ > -j REDIRECT --to-port 8083 > Applied to the iptables-next branch, thanks Eric. -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Friday 2010-07-23 15:43, Eric Dumazet wrote: >+ >+static const struct option cpu_opts[] = { >+ { "cpu", 1, NULL, '1' }, >+ { .name = NULL } >+}; I will never understand that sort of style mix logic. Why the C99 initializer only on the sentinel? { {.name = "cpu", .has_arg = true, .val = '1'}, {NULL}, }; >+cpu_print(const void *ip, const struct xt_entry_match *match, int numeric) >+{ >+ const struct xt_cpu_info *info = (void *)match->data; >+ >+ printf("cpu %s%u ", info->invert ? "! ":"", info->cpu); >+} >+ >+static void cpu_save(const void *ip, const struct xt_entry_match *match) >+{ >+ const struct xt_cpu_info *info = (void *)match->data; >+ >+ printf("%s--cpu %u ", info->invert ? "! ":"", info->cpu); >+} Using if (info->invert) would save the empty string. >diff --git a/extensions/libxt_cpu.man b/extensions/libxt_cpu.man >index e69de29..f42ac7a 100644 >--- a/extensions/libxt_cpu.man >+++ b/extensions/libxt_cpu.man >@@ -0,0 +1,16 @@ >+.TP >+[\fB!\fP] \fB\-\-cpu\fP \fInumber\fP >+ >+Match cpu handling this packet. cpus are numbered from 0 to NR_CPUS-1 Unwanted blank line. >+Can be used in combination with RPS (Remote Packet Steering) or >+multiqueue NICS to spread network traffic on different queues. >+.PP >+Example: >+.PP >+iptables \-t nat \-A PREROUTING \-p tcp \-\-dport 80 \-m cpu \-\-cpu 0 >+ \-j REDIRECT \-\-to\-port 8080 Unwanted indent. >+.PP >+iptables \-t nat \-A PREROUTING \-p tcp \-\-dport 80 \-m cpu \-\-cpu 1 >+ \-j REDIRECT \-\-to\-port 8081 >+.PP >+Available since linux 2.6.36 Linux. -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Le vendredi 23 juillet 2010 à 18:46 +0200, Jan Engelhardt a écrit : > On Friday 2010-07-23 15:43, Eric Dumazet wrote: > >+ > >+static const struct option cpu_opts[] = { > >+ { "cpu", 1, NULL, '1' }, > >+ { .name = NULL } > >+}; > > I will never understand that sort of style mix logic. Why the > C99 initializer only on the sentinel? > > { > {.name = "cpu", .has_arg = true, .val = '1'}, > {NULL}, > }; > copy/paste from another module ? > >+cpu_print(const void *ip, const struct xt_entry_match *match, int numeric) > >+{ > >+ const struct xt_cpu_info *info = (void *)match->data; > >+ > >+ printf("cpu %s%u ", info->invert ? "! ":"", info->cpu); > >+} > >+ > >+static void cpu_save(const void *ip, const struct xt_entry_match *match) > >+{ > >+ const struct xt_cpu_info *info = (void *)match->data; > >+ > >+ printf("%s--cpu %u ", info->invert ? "! ":"", info->cpu); > >+} > > Using if (info->invert) would save the empty string. > Not sure what you mean. You want to save an empty string (1 byte long), and add multiple printf() calls ? > >diff --git a/extensions/libxt_cpu.man b/extensions/libxt_cpu.man > >index e69de29..f42ac7a 100644 > >--- a/extensions/libxt_cpu.man > >+++ b/extensions/libxt_cpu.man > >@@ -0,0 +1,16 @@ > >+.TP > >+[\fB!\fP] \fB\-\-cpu\fP \fInumber\fP > >+ > >+Match cpu handling this packet. cpus are numbered from 0 to NR_CPUS-1 > > Unwanted blank line. > > >+Can be used in combination with RPS (Remote Packet Steering) or > >+multiqueue NICS to spread network traffic on different queues. > >+.PP > >+Example: > >+.PP > >+iptables \-t nat \-A PREROUTING \-p tcp \-\-dport 80 \-m cpu \-\-cpu 0 > >+ \-j REDIRECT \-\-to\-port 8080 > > Unwanted indent. > > >+.PP > >+iptables \-t nat \-A PREROUTING \-p tcp \-\-dport 80 \-m cpu \-\-cpu 1 > >+ \-j REDIRECT \-\-to\-port 8081 > >+.PP > >+Available since linux 2.6.36 > > Linux. OK ;) I'll provide a cleanup patch, not only to xt_cpu but all other iptables modules that dont meet your coding style requirements ;) Thanks -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Friday 2010-07-23 19:30, Eric Dumazet wrote: >> >+ >> >+static const struct option cpu_opts[] = { >> >+ { "cpu", 1, NULL, '1' }, >> >+ { .name = NULL } >> >+}; >> >> I will never understand that sort of style mix logic. Why the >> C99 initializer only on the sentinel? >> >> { >> {.name = "cpu", .has_arg = true, .val = '1'}, >> {NULL}, >> }; >> > >copy/paste from another module ? > > >> >diff --git a/extensions/libxt_cpu.man b/extensions/libxt_cpu.man >> >index e69de29..f42ac7a 100644 >> >--- a/extensions/libxt_cpu.man >> >+++ b/extensions/libxt_cpu.man >> >@@ -0,0 +1,16 @@ >> >+.TP >> >+[\fB!\fP] \fB\-\-cpu\fP \fInumber\fP >> >+ >> >+Match cpu handling this packet. cpus are numbered from 0 to NR_CPUS-1 >> >> Unwanted blank line. >> >> >+Can be used in combination with RPS (Remote Packet Steering) or >> >+multiqueue NICS to spread network traffic on different queues. >> >+.PP >> >+Example: >> >+.PP >> >+iptables \-t nat \-A PREROUTING \-p tcp \-\-dport 80 \-m cpu \-\-cpu 0 >> >+ \-j REDIRECT \-\-to\-port 8080 >> >> Unwanted indent. >> >> >+.PP >> >+iptables \-t nat \-A PREROUTING \-p tcp \-\-dport 80 \-m cpu \-\-cpu 1 >> >+ \-j REDIRECT \-\-to\-port 8081 >> >+.PP >> >+Available since linux 2.6.36 >> >> Linux. > > >OK ;) > >I'll provide a cleanup patch, not only to xt_cpu but all other iptables >modules that dont meet your coding style requirements ;) Well nah I'm already on it myself, given Patrick has already imported the patches. -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/extensions/libxt_cpu.c b/extensions/libxt_cpu.c index e69de29..869998d 100644 --- a/extensions/libxt_cpu.c +++ b/extensions/libxt_cpu.c @@ -0,0 +1,98 @@ +/* Shared library add-on to iptables to add CPU match support. */ +#include <stdio.h> +#include <netdb.h> +#include <string.h> +#include <stdlib.h> +#include <getopt.h> +#include <xtables.h> +#include <linux/netfilter/xt_cpu.h> + +static void cpu_help(void) +{ + printf( +"cpu match options:\n" +"[!] --cpu number Match CPU number\n"); +} + +static const struct option cpu_opts[] = { + { "cpu", 1, NULL, '1' }, + { .name = NULL } +}; + +static void +parse_cpu(const char *s, struct xt_cpu_info *info) +{ + unsigned int cpu; + char *end; + + if (!xtables_strtoui(s, &end, &cpu, 0, UINT32_MAX)) + xtables_param_act(XTF_BAD_VALUE, "cpu", "--cpu", s); + + if (*end != '\0') + xtables_param_act(XTF_BAD_VALUE, "cpu", "--cpu", s); + + info->cpu = cpu; +} + +static int +cpu_parse(int c, char **argv, int invert, unsigned int *flags, + const void *entry, struct xt_entry_match **match) +{ + struct xt_cpu_info *cpuinfo = (struct xt_cpu_info *)(*match)->data; + + switch (c) { + case '1': + xtables_check_inverse(optarg, &invert, &optind, 0, argv); + parse_cpu(optarg, cpuinfo); + if (invert) + cpuinfo->invert = 1; + *flags = 1; + break; + + default: + return 0; + } + + return 1; +} + +static void cpu_check(unsigned int flags) +{ + if (!flags) + xtables_error(PARAMETER_PROBLEM, + "You must specify `--cpu'"); +} + +static void +cpu_print(const void *ip, const struct xt_entry_match *match, int numeric) +{ + const struct xt_cpu_info *info = (void *)match->data; + + printf("cpu %s%u ", info->invert ? "! ":"", info->cpu); +} + +static void cpu_save(const void *ip, const struct xt_entry_match *match) +{ + const struct xt_cpu_info *info = (void *)match->data; + + printf("%s--cpu %u ", info->invert ? "! ":"", info->cpu); +} + +static struct xtables_match cpu_match = { + .family = NFPROTO_UNSPEC, + .name = "cpu", + .version = XTABLES_VERSION, + .size = XT_ALIGN(sizeof(struct xt_cpu_info)), + .userspacesize = XT_ALIGN(sizeof(struct xt_cpu_info)), + .help = cpu_help, + .parse = cpu_parse, + .final_check = cpu_check, + .print = cpu_print, + .save = cpu_save, + .extra_opts = cpu_opts, +}; + +void _init(void) +{ + xtables_register_match(&cpu_match); +} diff --git a/extensions/libxt_cpu.man b/extensions/libxt_cpu.man index e69de29..f42ac7a 100644 --- a/extensions/libxt_cpu.man +++ b/extensions/libxt_cpu.man @@ -0,0 +1,16 @@ +.TP +[\fB!\fP] \fB\-\-cpu\fP \fInumber\fP + +Match cpu handling this packet. cpus are numbered from 0 to NR_CPUS-1 +Can be used in combination with RPS (Remote Packet Steering) or +multiqueue NICS to spread network traffic on different queues. +.PP +Example: +.PP +iptables \-t nat \-A PREROUTING \-p tcp \-\-dport 80 \-m cpu \-\-cpu 0 + \-j REDIRECT \-\-to\-port 8080 +.PP +iptables \-t nat \-A PREROUTING \-p tcp \-\-dport 80 \-m cpu \-\-cpu 1 + \-j REDIRECT \-\-to\-port 8081 +.PP +Available since linux 2.6.36 diff --git a/include/linux/netfilter/xt_cpu.h b/include/linux/netfilter/xt_cpu.h index e69de29..93c7f11 100644 --- a/include/linux/netfilter/xt_cpu.h +++ b/include/linux/netfilter/xt_cpu.h @@ -0,0 +1,11 @@ +#ifndef _XT_CPU_H +#define _XT_CPU_H + +#include <linux/types.h> + +struct xt_cpu_info { + __u32 cpu; + __u32 invert; +}; + +#endif /*_XT_CPU_H*/
Patrick, Here is iptables extension for xt_cpu match. I put same changelog than kernel one, tell me if its ok or not ;) Thanks [PATCH iptables] extension: add xt_cpu match Kernel 2.6.36 supports xt_cpu match In some situations a CPU match permits a better spreading of connections, or select targets only for a given cpu. With Remote Packet Steering or multiqueue NIC and appropriate IRQ affinities, we can distribute trafic on available cpus, per session. (all RX packets for a given flow are handled by a given cpu) Some legacy applications being not SMP friendly, one way to scale a server is to run multiple copies of them. Instead of randomly choosing an instance, we can use the cpu number as a key so that softirq handler for a whole instance is running on a single cpu, maximizing cache effects in TCP/UDP stacks. Using NAT for example, a four ways machine might run four copies of server application, using a separate listening port for each instance, but still presenting an unique external port : iptables -t nat -A PREROUTING -p tcp --dport 80 -m cpu --cpu 0 \ -j REDIRECT --to-port 8080 iptables -t nat -A PREROUTING -p tcp --dport 80 -m cpu --cpu 1 \ -j REDIRECT --to-port 8081 iptables -t nat -A PREROUTING -p tcp --dport 80 -m cpu --cpu 2 \ -j REDIRECT --to-port 8082 iptables -t nat -A PREROUTING -p tcp --dport 80 -m cpu --cpu 3 \ -j REDIRECT --to-port 8083 Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> --- extensions/libxt_cpu.c | 98 +++++++++++++++++++++++++++++ extensions/libxt_cpu.man | 16 ++++ include/linux/netfilter/xt_cpu.h | 11 +++ 3 files changed, 125 insertions(+) -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html