*) Initial pass at support for --enable-intervals (WANT_INTERVALS) for
Windows, courtesy of Jonathan Cook.
*) When in demo mode (./configure --enable-demo and a global -D
<interval> option netperf will make sure it emits "one last
interval result" when the test is terminated. This should assist
when post-processing results through the likes of rrdtool when
there is a slow-down in the performance just at the end that would
have stretched the interval to beyond the test termination.
*) A fix to have the AF_UNIX tests realize that the value for "take
the system default" socket buffer size became -1 years ago. Bug
found by Eric Dumazet.
*) Include a patch from Dave Taht to enable symbolic manipulation of
IP_TOS values.
*) Include a patch from Sachar Raindel to enable the omni tests to get
ENOBUFS under Linux when the socket buffer is larger than the tx
queue of the egress interface. This will help preclude netperf's
reporting a larger than link-rate send-side figure.
*) Fix a problem with late checking of the return from select() in
src/netserver.c. Reported by Waqar Sheikh.
*) A new global -Z option has been added to netperf and netserver.
This takes as an argument a passphrase. In the case of netserver
it will expect a control message with the passphrase as the first
thing it receives on the control connection. If netserver does not
receive a control message with the passphrase it will close the
control connection and move-on. If the netserver receives a
control message with a passprhase when it is not lookign for one,
it will be ignored. There is at present a 20 second timeout on the
attempted receipt of the request message. In the case of netperf,
the passphrase will be the first thing sent on the control
connection. There is no response to a passphrase control message.
*) Demo mode output format will now track the omni output format. So,
if the omni ouput format is CSV then the interim results will be
emitted in csv. Likewise for keyval. If the mode is human (default
and test-specific -O) then the output remains unchanged. Keyval
output includes the count of interval, with a mind towards being
able to source it in shells and whatnot. Subject to change without
notice.
*) A patch to correctly handle IPv6 addresses in the control messages,
courtesy of Bjoern Zeeb.
*) The global -F option can now be used specify a local and/or remote
fill file.
*) It is now possible to set/get the TCP congestion control algorithm
being used by either end of the test connection when using the omni
code. The output selectors are LOCAL_CONG_CONTROL and
REMOTE_CONG_CONTROL and setting is via the test-specific -K option.
*) Stop leaking file descriptors when looking-up probable egress
interface names and I/O slot numbers.
*) The global -Y option can be used to set IP_TOS on those platforms
which support it. Since this is specific to IP (v4 or v6) it may
move to a test-specific otion in the future. It is presently
global for foolish consistency with the -y option to set
SO_PRIORITY.
*) The global -y option can be used to set SO_PRIORITY on those
platforms which support it. Based on patches from Amir Vidai.
*) The control message size has been increased from 256 bytes to 512
bytes. THIS WILL BREAK COMPATABILITY WITH PREVIOUS VERSIONS OF
NETPERF. However, we need more room on the pinhead on which the
angels dance.
*) Make the "sum" field of the histogram structure a 64 bit int to
avoid having it wrap-around on tests where the sum of all the
measured latencies was larger than 31 bits. This was causing
statistics like stddev to go negative in some cases.
*) If the time delta between two events is negative, do not bother
doing any math with it in the histogram/statistics code, just
increment the ridiculous count and move-on.
*) Fixed a bug which caused local transport retransmissions to be
reported as -1 even though the getsockopt() call was
successful. (Linux). Later included remote transport
retransmissions.
*) The src/nettest_omni.c and re-written src/netserver.c code are now
known to have compiled under Windows 7 x64 with the Microsoft
WDK. There remains a timing issue with confidence intervals which
is yet to be addressed, and may have been there for ages. Netserver
has been run as a non-spawning (-f) server, netperf has been run,
both have run "classic" and "omni" tests.
1.) Comment out "MASTER_SITES". The URL no longer works and all copies
that Google can find are "pkgsrc" distfile mirrors.
2.) Fix various "pkglint" errors.
DNSPerf, ResPerf, and DHCPerf are free tools developed by Nominum that
make it simple to gather accurate latency and throughput metrics for
Domain Name Service (DNS) and Dynamic Host Configuration Protocol
(DHCP). These tools are easy-to-use and simulate typical Internet so
network operators can benchmark their naming and addressing
infrastructure and plan for upgrades. The latest version of the DNS
test tools (DNSPerf and ResPerf) can be used with new test files that
include IPv6 queries.
DNSPerf "self-paces" the DNS query load to simulate network
conditions. New features in DNSPerf improve the precision of latency
measurements and allow for per packet per-query latency reporting is
possible. DNSPerf is now multithreaded, multiple DNSPerf clients can
be supported in multicore systems (each client requires two
cores). The output of DNSPerf has also been improved so it is more
concise and useful. Latency data can be used to make detailed graphs
so it is simple for network operators to take advantage of the data.
The point of paranoia is to assess a compiler/machine environment.
Therefore, this package should be built with the default CFLAGS, and
not add options to try to get the tests to pass. This will cause
errors to be reported on i386.
Take MAINTAINERSHIP.
to address issues with NetBSD-6(and earlier)'s fontconfig not being
new enough for pango.
While doing that, also bump freetype2 dependency to current pkgsrc
version.
Suggested by tron in PR 47882
a) refer 'perl' in their Makefile, or
b) have a directory name of p5-*, or
c) have any dependency on any p5-* package
Like last time, where this caused no complaints.
File too long (should be no more than 24 lines).
Line too long (should be no more than 80 characters).
Trailing empty lines.
Trailing white-space.
Trucated the long files as best as possible while preserving the most info
contained in them.
RAMspeed, a cache and memory benchmarking tool
(for uniprocessor machines running UNIX-like operating systems).
This command line utility measures effective bandwidth of both cache and memory
subsystems. It has been written entirely in C for portability purposes, though
benchmark routines are also available in several assembly languages for
performance reasons.
The "-r" option is highly discouraged in BSD although it's reluctantly
supported. The problem is that it doesn't behave the same on all
platforms. For example, "cp -r pts-core/ destdir/" won't always give
the same results:
On NetBSD, pts-core files are put: /destdir/pts-core/<files>
On DragonFly pts-core files are put: /destdir/<files>
That messes the PLIST up.
This changes "cp -r <dir>/" to "cp -R <dir>" which has the same
behavior on different platforms. Tested on NetBSD and DragonFly
Filebench is a file system and storage benchmark that allows
to generate a high variety of workloads. It employs extensive
Workload Model Language (WML) for detailed workload specification.
Features
* Extensive Workload Model Language (WML) for workload specification
* Includes popular pre-defined workloads: webserver, fileserver,
varmail, etc.
* Reports throughput and latency histograms
Changes since version 3.397:
Revision 3.408
Bug fix for clobbering of buffer. "tfile" needed to be bigger, as its
address was loaded into filearray[] and then that was overwritten
by the mfflag causing parsed names to get loaded.
Revision 3.407
Work around for cache_line_size suddenly becoming zero and breaking fetchit()
Revision 3.406
Increase buffer size for configuration lines. They might be 700 chars long.
Revision 3.405
Reduce CPU consumption in the op_rate control mechanism.
Revision 3.404
Fix type-oh in usage message.
Revision 3.403
Add -+W to permit chid_skew. This permits adding files, and continuing
the proper sharing/dedup within each quadrant, within same seed group
and with previously existing files that were created with a different
number of threads.
Revision 3.402
Mods for DragonFly support.
Revision 3.401
bug fix for re-write rec.
Revision 3.398
Adding thread_read_test and thread_write_test.
The update from version 3.308 to 3.397 broke DragonFly 11 days after I
fixed it due to missing a macro change. An integer redeclaration was
also removed.
DESCR:
The Phoronix Test Suite is the most comprehensive testing and benchmarking
platform available for the *nix operating system. This software is designed
to effectively carry out both qualitative and quantitative benchmarks in a
clean, reproducible, and easy-to-use manner.
It appears that the authors of Iozone tried to add DragonFly support without
actually building it. The configuration was broken, and the iozone.c
was uncompilable. With the two modified patches, it will compile now, and a
quick check using "iozone -a" seemed to work as expected.
No PKGREVISION bump necessary.
Despite several additional patches, postal 0.62 wouldn't compile on
DragonFly (broke on stl vectors), so upgrading the package was the
most effective approach to restore this package.
License updated in Makefile.
Summary of changes since version 0.62
=====================================
0.65 Significant improvement, many new features and many bugs fixed!
0.66 Made GNUTLS work in BHM and added Messageld to Postal
0.67 Changed license to GPLv3
0.68 Fixed compiler warnings on GCC 4.3, make it work for OpenSolaris
0.69 Fixed compiler warnings and more GCC 4.3 compatibility work
0.70 Updated man pages and allowed build without SSL support
This commit fixes the run-libmicro command which was broken for all
platforms. It was pointing at the destdir version instead of the
installed version.
It also provides a DragonFly-specific makefile which enables it to
build and run on DragonFly.
part of, a new upstream version, but I need to talk to some people first.
(I'm keeping this separate from the previous update in case it breaks stuff.)
Changes:
roll in parts of patch-aa and patch-aq;
roll in all of patch-ar and patch-src_memsize_c;
make the ARM-only adjustment from patch-ar universal;
add enough volatile to make 'mhz' work more or less properly again;
fix the loop calibration to bail out instead of spinning forever
if things go too fast (e.g. because the compiler's optimized out
the benchmark loop);
various hacking and modernization to get sort of a clean compile;
fix some non-stdc stringifying macros;
hopefully advance the clang build a bit further.
Based on PR#44354 by Tony Young.
set LICENSE=public-domain, unit-test support by me.
Version 0.2 2002-09-11
Added unit test framework.
Read from a list of files, rather than just one, seeking evenly within
all of them based on the number of blocks in each.
Allow reading several contiguous blocks for each seek, so we can do
sequential/random and sequential read testing.
Print more and clearer information.
Added disktest shell scripts.
