[netperf-dev] netperf2 commit notice r88 - in trunk: . src
src/NetPerfDir src/NetServerDir
raj at netperf.org
raj at netperf.org
Wed Feb 7 16:37:04 PST 2007
Author: raj
Date: 2007-02-07 16:37:01 -0800 (Wed, 07 Feb 2007)
New Revision: 88
Modified:
trunk/AUTHORS
trunk/README
trunk/README.hpux
trunk/README.osx
trunk/README.solaris
trunk/Release_Notes
trunk/src/Makefile.am
trunk/src/Makefile.in
trunk/src/NetPerfDir/sources
trunk/src/NetServerDir/sources
trunk/src/netlib.c
trunk/src/nettest_bsd.c
trunk/src/nettest_sctp.c
Log:
remove nettest_ipv6.* from source distibutions and fix some bogosity when both intervals and demo mode were active at the same time
Modified: trunk/AUTHORS
===================================================================
--- trunk/AUTHORS 2007-02-07 23:16:58 UTC (rev 87)
+++ trunk/AUTHORS 2007-02-08 00:37:01 UTC (rev 88)
@@ -115,7 +115,7 @@
Fixes for -DHISTOGRAM and -DUNIX on Mac OS X, updates to usage strings
Spencer Frink
-Fixes and Cleanup for WIN32
+Fixes and Cleanup for WIN32. Many over many years.
Nicholas Thomas
Fixes for DLPI on SVR4 Streams under Linux
Modified: trunk/README
===================================================================
--- trunk/README 2007-02-07 23:16:58 UTC (rev 87)
+++ trunk/README 2007-02-08 00:37:01 UTC (rev 88)
@@ -15,14 +15,24 @@
Eagles gave me. I am not sure if the legalese is clear, but the intent
is to say "Here is a benchmark. Use it in good health. Pass it along,
port it, enhance it. You didn't pay for this tool, so don't expect it
-to be perfect ;-)" The rest of it is there to keep the layers happy...
+to be perfect ;-)" The rest of it is there to keep the layers happy...
-Feel free to report results in public forums, but please be
+While the copyright is pretty much in spirit an "open source" one, it
+is not in letter - I never took the time to try to get it approved.
+If you feel strongly about the license of the code you use and want
+something under the GPL, consider netperf4:
+
+http://www.netperf.org/svn/netperf4/trunk
+
+Feel free to report netperf results in public forums, but please be
excruciatingly complete in your description of the test envorinment.
-Better still, report results to the netperf database at
-http://www.netperf.org/netperf/NetperfPage.html. However, the netperf
-database is rather long in the tooth these days.
+The old netperf database at:
+ http://www.netperf.org/netperf/NetperfPage.html
+
+is no more - or rather the utilities for accessing it no longer run.
+The data is still present in the tree, albeit _VERY_ old now.
+
There is an Internet mailing list devoted to netperf. It is called
netperf-talk and it is hosted on netperf.org. Subscription requests
should go to netperf-talk-request at netperf.org.
Modified: trunk/README.hpux
===================================================================
--- trunk/README.hpux 2007-02-07 23:16:58 UTC (rev 87)
+++ trunk/README.hpux 2007-02-08 00:37:01 UTC (rev 88)
@@ -20,3 +20,12 @@
We will still calculate fractions for user, kernel and interrupt, and
report them in debug (-d) output, but with a warning for interrupt
time. See src/netcpu_pstatnew.c for all the details.
+
+Up through HP-UX 11.23 (aka 11iV2) if you enable burst mode, and
+happen to send sub-MSS requests and/or responses you _cannot_ assume
+that the packet per second rate on the wire will match the transaction
+per second rate reported by netperf, even if you set TCP_NODELAY with
+the test-specific -D option. The HP-UX 11.X TCP stack likely will be
+generating some immediate 'standalone' ACKnowledgements which may not
+be generated by other stacks. This has been reported to the HP-UX TCP
+folks, and an announcement will be made when that issue is resolved.
Modified: trunk/README.osx
===================================================================
--- trunk/README.osx 2007-02-07 23:16:58 UTC (rev 87)
+++ trunk/README.osx 2007-02-08 00:37:01 UTC (rev 88)
@@ -1,16 +1,4 @@
-When compiling under MacOS X you may encounter problems with the
-NetPerf and NetServer directories in src/. Seems that filenames can
-be case insensitive in OSX and so there are conflicts with the netperf
-and netserver binaries the makefile will try to create. Brilliant!
-
-As NetPerf/ and NetServer/ are only for Windows compilation, you can
-safely rename them to NetPerf.dir and NetServer.dir and the make will
-be fat dumb and happy.
-
-However, if you are using subversion to grab bits, it will become
-rather cranky at commit or update time - it will miss NetPerf and
-NetServer and will complain bitterly until you move them back again.
-
-At some point it is hoped that those two directories can be renamed so
-there is no case-insensitive conflict between them and the names of
-the binaries.
+If you are reading this, it suggests you are using a version of
+netperf in which the issue of file/directory name case insensitivity
+in OSX has been worked-around by renaming the NetPerf and NetServer
+directories to NetPerfDir and NetServerDir respectively.
Modified: trunk/README.solaris
===================================================================
--- trunk/README.solaris 2007-02-07 23:16:58 UTC (rev 87)
+++ trunk/README.solaris 2007-02-08 00:37:01 UTC (rev 88)
@@ -15,6 +15,15 @@
automagic (although for SCTP you still must --enable-sctp at configure
time)
-beware CPU util figures on anything before Solaris 10 that does not
+Beware CPU util figures on anything before Solaris 10 that does not
say 100%, and still be a triffle cautious with Solaris 10 CPU util
-reports.
+reports. The CPU time accounting mechanisms either do not track time
+spent servicing interrupts, or do so in parallel with time spent in
+user/kernel/idle which means that some idle time isn't _really_ idle
+time.
+
+And beyond that, it is still not clear if the CPU utilization reported
+on systems with hardware threading support (eg UltraSPARC-T1) is
+really accurate even ignoring the issue with interrupt time. It is
+likely that to be truely accurate, it is necessary to know how much
+"real work" any one strand performed.
Modified: trunk/Release_Notes
===================================================================
--- trunk/Release_Notes 2007-02-07 23:16:58 UTC (rev 87)
+++ trunk/Release_Notes 2007-02-08 00:37:01 UTC (rev 88)
@@ -1,7 +1,18 @@
-These are the Release Notes leading-up to Revision 2.4.3 of netperf:
+These are the Release Notes for Revision 2.4.3 of netperf:
Things changed in this release:
+*) Eliminate a bogus fprintf from the signal catching routine which
+ was being executed when both intervals and demo mode were active at
+ the same time.
+
+*) The nettest_ipv6.* files are no longer included in the source
+ tar/zip file. IPv6 functionality has been subsumed into the
+ nettest_bsd.* files for some time now. Paranoia keeps the Netperf
+ Contributing Editor from removing it from the top of trunk of the
+ source repository until after he can do a bit more reading about
+ subversion :)
+
*) Use a higher resolution "time" source for HISTOGRAM support under
Windows, courtesy of Spencer Frink. Prior to this it had no better
than 10ms granularity which could lead to some rather strange
Modified: trunk/src/Makefile.am
===================================================================
--- trunk/src/Makefile.am 2007-02-07 23:16:58 UTC (rev 87)
+++ trunk/src/Makefile.am 2007-02-08 00:37:01 UTC (rev 88)
@@ -6,7 +6,7 @@
USE_CPU_SOURCE=netcpu_ at NETCPU_SOURCE@.c
-EXTRA_DIST = netcpu_none.c netcpu_looper.c netcpu_pstat.c netcpu_pstatnew.c netcpu_perfstat.c netcpu_procstat.c netcpu_kstat.c netcpu_kstat10.c netcpu_sysctl.c netcpu_ntperf.c netcpu_osx.c nettest_ipv6.c nettest_ipv6.h missing/* dirs NetPerfDir/* NetServerDir/*
+EXTRA_DIST = netcpu_none.c netcpu_looper.c netcpu_pstat.c netcpu_pstatnew.c netcpu_perfstat.c netcpu_procstat.c netcpu_kstat.c netcpu_kstat10.c netcpu_sysctl.c netcpu_ntperf.c netcpu_osx.c missing/* dirs NetPerfDir/* NetServerDir/*
COMMON_SRC = hist.h netlib.c netlib.h netcpu.h netsh.c netsh.h nettest_bsd.c nettest_bsd.h nettest_dlpi.c nettest_dlpi.h nettest_dns.c nettest_dns.h nettest_unix.c nettest_unix.h nettest_xti.c nettest_xti.h nettest_sctp.c nettest_sctp.h
Modified: trunk/src/Makefile.in
===================================================================
--- trunk/src/Makefile.in 2007-02-07 23:16:58 UTC (rev 87)
+++ trunk/src/Makefile.in 2007-02-08 00:37:01 UTC (rev 88)
@@ -134,7 +134,7 @@
USE_CPU_SOURCE = netcpu_ at NETCPU_SOURCE@.c
-EXTRA_DIST = netcpu_none.c netcpu_looper.c netcpu_pstat.c netcpu_pstatnew.c netcpu_perfstat.c netcpu_procstat.c netcpu_kstat.c netcpu_kstat10.c netcpu_sysctl.c netcpu_ntperf.c netcpu_osx.c nettest_ipv6.c nettest_ipv6.h missing/* dirs NetPerfDir/* NetServerDir/*
+EXTRA_DIST = netcpu_none.c netcpu_looper.c netcpu_pstat.c netcpu_pstatnew.c netcpu_perfstat.c netcpu_procstat.c netcpu_kstat.c netcpu_kstat10.c netcpu_sysctl.c netcpu_ntperf.c netcpu_osx.c missing/* dirs NetPerfDir/* NetServerDir/*
COMMON_SRC = hist.h netlib.c netlib.h netcpu.h netsh.c netsh.h nettest_bsd.c nettest_bsd.h nettest_dlpi.c nettest_dlpi.h nettest_dns.c nettest_dns.h nettest_unix.c nettest_unix.h nettest_xti.c nettest_xti.h nettest_sctp.c nettest_sctp.h
Modified: trunk/src/NetPerfDir/sources
===================================================================
--- trunk/src/NetPerfDir/sources 2007-02-07 23:16:58 UTC (rev 87)
+++ trunk/src/NetPerfDir/sources 2007-02-08 00:37:01 UTC (rev 88)
@@ -21,7 +21,6 @@
..\netlib.c \
..\netsh.c \
..\nettest_bsd.c \
- ..\nettest_ipv6.c \
inet_ntop.c \
..\netperf.c
Modified: trunk/src/NetServerDir/sources
===================================================================
--- trunk/src/NetServerDir/sources 2007-02-07 23:16:58 UTC (rev 87)
+++ trunk/src/NetServerDir/sources 2007-02-08 00:37:01 UTC (rev 88)
@@ -21,7 +21,6 @@
..\netlib.c \
..\netsh.c \
..\nettest_bsd.c \
- ..\nettest_ipv6.c \
inet_ntop.c \
..\netserver.c
Modified: trunk/src/netlib.c
===================================================================
--- trunk/src/netlib.c 2007-02-07 23:16:58 UTC (rev 87)
+++ trunk/src/netlib.c 2007-02-08 00:37:01 UTC (rev 88)
@@ -322,7 +322,7 @@
new_interval.it_value.tv_usec = 0;
if (setitimer(ITIMER_REAL,&new_interval,&old_interval) != 0) {
/* there was a problem arming the interval timer */
- perror("clusterperf: cluster_root: setitimer");
+ perror("netperf: setitimer");
exit(1);
}
return;
@@ -785,16 +785,6 @@
scp->sc_syscall_action = SIG_RESTART;
}
#endif /* __hpux */
- if (demo_mode) {
- /* spit-out what the performance was in units/s. based on our */
- /* knowledge of the interval length we do not need to call */
- /* gettimeofday() raj 2/95 */
- fprintf(where,"%g\n",(units_this_tick *
- (double) 1000000 /
- (double) usec_per_itvl));
- fflush(where);
- units_this_tick = (double) 0.0;
- }
#else /* WANT_INTERVALS */
fprintf(where,
"catcher: interval timer running unexpectedly!\n");
@@ -1046,7 +1036,7 @@
new_interval.it_value.tv_usec = usec_per_itvl % 1000000;
if (setitimer(ITIMER_REAL,&new_interval,&old_interval) != 0) {
/* there was a problem arming the interval timer */
- perror("clusterperf: cluster_root: setitimer");
+ perror("netperf: setitimer");
exit(1);
}
}
Modified: trunk/src/nettest_bsd.c
===================================================================
--- trunk/src/nettest_bsd.c 2007-02-07 23:16:58 UTC (rev 87)
+++ trunk/src/nettest_bsd.c 2007-02-08 00:37:01 UTC (rev 88)
@@ -233,7 +233,7 @@
if ((interval_burst) && (--interval_count == 0)) { \
/* call sigsuspend and wait for the interval timer to get us \
out */ \
- if (debug) { \
+ if (debug > 1) { \
fprintf(where,"about to suspend\n"); \
fflush(where); \
} \
@@ -280,7 +280,7 @@
if ((interval_burst) && (--interval_count == 0)) { \
/* call sigsuspend and wait for the interval timer to get us \
out */ \
- if (debug) { \
+ if (debug > 1) { \
fprintf(where,"about to spin suspend\n"); \
fflush(where); \
} \
Modified: trunk/src/nettest_sctp.c
===================================================================
--- trunk/src/nettest_sctp.c 2007-02-07 23:16:58 UTC (rev 87)
+++ trunk/src/nettest_sctp.c 2007-02-08 00:37:01 UTC (rev 88)
@@ -748,7 +748,7 @@
if ((interval_burst) && (--interval_count == 0)) {
/* call sigsuspend and wait for the interval timer to get us */
/* out */
- if (debug) {
+ if (debug > 1) {
fprintf(where,"about to suspend\n");
fflush(where);
}
@@ -1898,7 +1898,7 @@
if ((interval_burst) && (--interval_count == 0)) {
/* call sigsuspend and wait for the interval timer to get us */
/* out */
- if (debug) {
+ if (debug > 1) {
fprintf(where,"about to suspend\n");
fflush(where);
}
@@ -2981,7 +2981,7 @@
if ((interval_burst) && (--interval_count == 0)) {
/* call sigsuspend and wait for the interval timer to get us */
/* out */
- if (debug) {
+ if (debug > ) {
fprintf(where,"about to suspend\n");
fflush(where);
}
More information about the netperf-dev
mailing list