Upon starting netem with the trace enhancement (e.g. netem trace) a new
process (called flowseed) is created which sends the "delay-values" to
the netem kernel module with the help of the configfs. The values are
written in chunks of 1000, thus avoiding to many context switches.
When the corresponding qdisc gets deleted the flowseed process
terminates himself (because he receives a negative return value from
the write call).
Since we have inserted a subdirectory iproute2/netem/distributions the
patch became quite huge. For ease of discussion I inline the relevant
part of the patch in this email and the whole patch can be found on
http://www.tcn.hypert.net/tcn_iproute2_2_6_23
Signed-off-by: Ariane Keller <[EMAIL PROTECTED]>
---
diff -uprN originIPRoute/include/linux/pkt_sched.h
iproute2-2.6.23/include/linux/pkt_sched.h
--- originIPRoute/include/linux/pkt_sched.h 2007-10-16
23:27:42.000000000 +0200
+++ iproute2-2.6.23/include/linux/pkt_sched.h 2007-11-19
18:42:48.000000000 +0100
@@ -439,6 +439,8 @@ enum
TCA_NETEM_DELAY_DIST,
TCA_NETEM_REORDER,
TCA_NETEM_CORRUPT,
+ TCA_NETEM_TRACE,
+ TCA_NETEM_STATS,
__TCA_NETEM_MAX,
};
@@ -454,6 +456,37 @@ struct tc_netem_qopt
__u32 jitter; /* random jitter in latency (us) */
};
+struct tc_netem_stats
+{
+ int packetcount;
+ int packetok;
+ int normaldelay;
+ int drops;
+ int dupl;
+ int corrupt;
+ int novaliddata;
+ int uninitialized;
+ int bufferunderrun;
+ int bufferinuseempty;
+ int noemptybuffer;
+ int readbehindbuffer;
+ int buffer1_reloads;
+ int buffer2_reloads;
+ int tobuffer1_switch;
+ int tobuffer2_switch;
+ int switch_to_emptybuffer1;
+ int switch_to_emptybuffer2;
+};
+
+
+struct tc_netem_trace
+{
+ __u32 fid;
+ __u32 def;
+ __u32 ticks;
+};
+
+
struct tc_netem_corr
{
__u32 delay_corr; /* delay correlation */
diff -uprN originIPRoute/netem/trace/flowseed.c
iproute2-2.6.23/netem/trace/flowseed.c
--- originIPRoute/netem/trace/flowseed.c 1970-01-01 01:00:00.000000000
+0100
+++ iproute2-2.6.23/netem/trace/flowseed.c 2007-11-20 14:32:54.000000000
+0100
@@ -0,0 +1,117 @@
+/* flowseed.c flowseedprocess to deliver values for packet delay,
+ * duplication, loss and curruption form userspace to netem
+ *
+ * This program is free software; you can redistribute
it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * Authors: Ariane Keller <[EMAIL PROTECTED]> ETH Zurich
+ * Rainer Baumann <[EMAIL PROTECTED]> ETH Zurich
+ */
+
+#include <ctype.h>
+#include <stdio.h>
+#include <fcntl.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <unistd.h>
+
+#define DATA_PACKAGE 4000
+#define DATA_PACKAGE_ID DATA_PACKAGE + sizeof(unsigned int) + sizeof(int)
+
+/* maximal amount of parallel flows */
+#define MAX_FLOWS 4
+
+int main(int argc, char *argv[])
+{
+ if (argc < 3) {
+ printf("usage: <tracefilename> <loop>");
+ return 0;
+ }
+ char *sendpkg;
+ sendpkg = malloc(DATA_PACKAGE_ID);
+
+ if (sendpkg == NULL) {
+ printf("out of memory\n");
+ return 0;
+ }
+
+ int fid = getpid();
+ char dirname[20];
+ char path[30];
+ int fdflowseed, fdtcn, dirsux;
+ unsigned int loop;
+ int infinity = 0;
+ int moreData = 1, r = 0, rold = 0;
+ loop = strtoul(argv[2], NULL, 10);
+
+ snprintf(dirname, sizeof(path), "/config/tcn/%d", fid);
+ dirsux = mkdir(dirname, S_IRWXO);
+ snprintf(path, sizeof(path), "%s/delayvalue", dirname);
+
+ if ((fdtcn = open(path, O_WRONLY, 0)) < 0) {
+ perror("fdtcn: ");
+ return 0;
+ }
+
+ if ((fdflowseed = open(argv[1], O_RDONLY, 0)) < 0) {
+ perror("cannot open tracefile");
+ return 0;
+ }
+
+ if (loop == 0)
+ infinity = 1;
+
+ while (loop > 0 || infinity) {
+ loop--;
+ int w;
+ /*read action values from tracefile */
+ while ((r = read(fdflowseed, sendpkg + rold, DATA_PACKAGE - rold))
>= 0) {
+ if (r + rold < DATA_PACKAGE) {
+ /* Tail of input file reached,
+ set rest at start from next iteration */
+ rold = r;
+ if (lseek(fdflowseed, 0L, SEEK_SET) < 0)
+ perror("lseek reset");
+ break;
+ }
+ r = 0;
+ rold = 0;
+ memcpy(sendpkg + DATA_PACKAGE, &fid, sizeof(int));
+ memcpy(sendpkg + DATA_PACKAGE + sizeof(int), &moreData,
+ sizeof(int));
+ w = write(fdtcn, sendpkg, DATA_PACKAGE_ID);
+ if (w < 0) {
+ perror("write");
+ close(fdflowseed);
+ close(fdtcn);
+ rmdir(dirname);
+ free(sendpkg);
+ return 0;
+ }
+
+ }
+ }
+ /* last packet: say kernel that no more data is available
+ enables to distinguish between bufferunderrun and no
+ more data available.
+ */
+ moreData = 0;
+ memcpy(sendpkg + DATA_PACKAGE, &fid, sizeof(int));
+ memcpy(sendpkg + DATA_PACKAGE + sizeof(int), &moreData, sizeof(int));
+ int w = write(fdtcn, sendpkg, DATA_PACKAGE_ID);
+ if (w < 0)
+ perror("Failure writing last msg to kernel");
+
+ printf("Tail of input file reached. Exit.\n");
+
+ close(fdflowseed);
+ close(fdtcn);
+ rmdir(dirname);
+ free(sendpkg);
+
+ return 0;
+}
diff -uprN originIPRoute/tc/q_netem.c iproute2-2.6.23/tc/q_netem.c
--- originIPRoute/tc/q_netem.c 2007-10-16 23:27:42.000000000 +0200
+++ iproute2-2.6.23/tc/q_netem.c 2007-11-20 14:46:24.000000000 +0100
@@ -6,7 +6,12 @@
* as published by the Free Software Foundation; either version
* 2 of the License, or (at your option) any later version.
*
+ * README files: iproute2/netem/distribution
+ * iproute2/netem/trace
+ *
* Authors: Stephen Hemminger <[EMAIL PROTECTED]>
+ * netem trace: Ariane Keller <[EMAIL PROTECTED]> ETH Zurich
+ * Rainer Baumann <[EMAIL PROTECTED]> ETH Zurich
*
*/
@@ -20,6 +25,10 @@
#include <arpa/inet.h>
#include <string.h>
#include <errno.h>
+#include <sys/mount.h>
+#include <ctype.h>
+#include <string.h>
+#include <sys/types.h>
#include "utils.h"
#include "tc_util.h"
@@ -42,6 +51,7 @@ static void explain1(const char *arg)
fprintf(stderr, "Illegal \"%s\"\n", arg);
}
+#define FLOWPATH "/usr/local/bin/flowseed"
#define usage() return(-1)
/*
@@ -129,6 +139,7 @@ static int netem_parse_opt(struct qdisc_
struct tc_netem_corr cor;
struct tc_netem_reorder reorder;
struct tc_netem_corrupt corrupt;
+ struct tc_netem_trace traceopt;
__s16 *dist_data = NULL;
int present[__TCA_NETEM_MAX];
@@ -137,6 +148,7 @@ static int netem_parse_opt(struct qdisc_
memset(&cor, 0, sizeof(cor));
memset(&reorder, 0, sizeof(reorder));
memset(&corrupt, 0, sizeof(corrupt));
+ memset(&traceopt, 0, sizeof(traceopt));
memset(present, 0, sizeof(present));
while (argc > 0) {
@@ -243,6 +255,73 @@ static int netem_parse_opt(struct qdisc_
} else if (strcmp(*argv, "help") == 0) {
explain();
return -1;
+ } else if (strcmp(*argv, "trace") == 0) {
+ int pid_tc = getpid();
+ int fd;
+ int execvl;
+ char *filename;
+ int pid;
+ /* configfs for data transfer user <-> kernel space */
+ int b = mkdir("/config", S_IRWXO);
+ if (b && errno != EEXIST) {
+ perror("mkdir");
+ return -1;
+ }
+ int a = mount("", "/config", "configfs", 0, "");
+ if (a && errno != EBUSY) {
+ perror("mounting configfs");
+ return -1;
+ }
+
+ /*get ticks correct since tracefile is in us,
+ *and ticks may not be equal to us
+ */
+ get_ticks(&traceopt.ticks, "1000us");
+ NEXT_ARG();
+ filename = *argv;
+ if ((fd = open(filename, O_RDONLY, 0)) < 0) {
+ fprintf(stderr, "Cannot open trace file \n");
+ return -1;
+ }
+ close(fd);
+ if (NEXT_IS_NUMBER()) {
+ NEXT_ARG();
+ /*child will load tracefile to kernel */
+ switch (pid = fork()) {
+ case -1:{
+ fprintf(stderr,
+ "Cannot fork\n");
+ return -1;
+ }
+ case 0:{
+ /* child wait for parent to die to be
+ * sure that the kernel is ready for
+ * for the tracefiledata
+ */
+ while (pid_tc == getppid())
+ sleep(0);
+ execvl = execl(FLOWPATH,
+ "flowseed",
+ filename,
+ *argv, 0);
+ if (execvl < 0) {
+ fprintf(stderr,
+ "starting child failed\n");
+ return -1;
+ }
+ }
+ }
+ }
+ else {
+ explain();
+ return -1;
+ }
+ traceopt.def = 0;
+ if (NEXT_IS_NUMBER()) {
+ NEXT_ARG();
+ traceopt.def = atoi(*argv);
+ }
+ traceopt.fid = pid;
} else {
fprintf(stderr, "What is \"%s\"?\n", *argv);
explain();
@@ -291,7 +370,13 @@ static int netem_parse_opt(struct qdisc_
dist_data, dist_size*sizeof(dist_data[0])) < 0)
return -1;
}
- tail->rta_len = (void *) NLMSG_TAIL(n) - (void *) tail;
+ if (traceopt.fid) {
+ if (addattr_l(n, TCA_BUF_MAX, TCA_NETEM_TRACE, &traceopt,
+ sizeof(traceopt)) < 0)
+ return -1;
+ }
+
+ tail->rta_len = (void *)NLMSG_TAIL(n) - (void *)tail;
return 0;
}
@@ -300,6 +385,8 @@ static int netem_print_opt(struct qdisc_
const struct tc_netem_corr *cor = NULL;
const struct tc_netem_reorder *reorder = NULL;
const struct tc_netem_corrupt *corrupt = NULL;
+ const struct tc_netem_trace *traceopt = NULL;
+ const struct tc_netem_stats *tracestats = NULL;
struct tc_netem_qopt qopt;
int len = RTA_PAYLOAD(opt) - sizeof(qopt);
SPRINT_BUF(b1);
@@ -333,9 +420,49 @@ static int netem_print_opt(struct qdisc_
return -1;
corrupt = RTA_DATA(tb[TCA_NETEM_CORRUPT]);
}
+ if (tb[TCA_NETEM_TRACE]) {
+ if (RTA_PAYLOAD(tb[TCA_NETEM_TRACE]) <
sizeof(*traceopt))
+ return -1;
+ traceopt = RTA_DATA(tb[TCA_NETEM_TRACE]);
+ }
+ if (tb[TCA_NETEM_STATS]) {
+ if (RTA_PAYLOAD(tb[TCA_NETEM_STATS]) <
sizeof(*tracestats))
+ return -1;
+ tracestats = RTA_DATA(tb[TCA_NETEM_STATS]);
+ }
}
fprintf(f, "limit %d", qopt.limit);
+ if (traceopt->fid) {
+ fprintf(f, " trace\n");
+
+ fprintf(f, "packetcount= %d\n", tracestats->packetcount);
+ fprintf(f, "packetok= %d\n", tracestats->packetok);
+ fprintf(f, "normaldelay= %d\n", tracestats->normaldelay);
+ fprintf(f, "drops= %d\n", tracestats->drops);
+ fprintf(f, "dupl= %d\n", tracestats->dupl);
+ fprintf(f, "corrupt= %d\n", tracestats->corrupt);
+ fprintf(f, "novaliddata= %d\n", tracestats->novaliddata);
+ fprintf(f, "uninitialized= %d\n", tracestats->uninitialized);
+ fprintf(f, "bufferunderrun= %d\n", tracestats->bufferunderrun);
+ fprintf(f, "bufferinuseempty= %d\n",
+ tracestats->bufferinuseempty);
+ fprintf(f, "noemptybuffer= %d\n", tracestats->noemptybuffer);
+ fprintf(f, "readbehindbuffer= %d\n",
+ tracestats->readbehindbuffer);
+ fprintf(f, "buffer1_reloads= %d\n",
+ tracestats->buffer1_reloads);
+ fprintf(f, "buffer2_reloads= %d\n",
+ tracestats->buffer2_reloads);
+ fprintf(f, "tobuffer1_switch= %d\n",
+ tracestats->tobuffer1_switch);
+ fprintf(f, "tobuffer2_switch= %d\n",
+ tracestats->tobuffer2_switch);
+ fprintf(f, "switch_to_emptybuffer1= %d\n",
+ tracestats->switch_to_emptybuffer1);
+ fprintf(f, "switch_to_emptybuffer2= %d\n",
+ tracestats->switch_to_emptybuffer2);
+ }
if (qopt.latency) {
fprintf(f, " delay %s", sprint_ticks(qopt.latency, b1));
-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
