summaryrefslogtreecommitdiff
path: root/benchmarks
AgeCommit message (Collapse)AuthorFilesLines
2005-11-11Rename LATEX_ACCEPTED and LATEX_DEFAULT to TEX_ACCEPTED and TEX_DEFAULTtonio1-2/+2
It is more consistent with the tex.buildlink3.mk name. Also, if a package really needs latex, it just has to set TEX_ACCEPTED to latex distributions altough today, all TEX_ACCEPTED possibilities are latex distributions
2005-11-08convert to use tex.buildlink3.mktonio1-2/+3
Accepts teTeX1, teTeX2 and teTeX3
2005-11-07Don't declare errno manually, depend on errno.h for it. While here,joerg3-13/+43
fix the signal handling to use a volatile variable. Compilers can optimise the access otherwise. Bump revision, since this is a bugfix. Reviewed by jlam and the maintainer.
2005-11-07Depend on errno.h to provide errno.joerg2-12/+20
2005-11-06Use O_FSYNC instead of O_SYNC on DragonFly. Disable extattr_get_filejoerg3-7/+25
on DragonFly, the semantic is different and it isn't very useful in the current form.
2005-11-05Do not intermingle build phases, CONFIGURE_DIRS can specify a list ofjoerg7-689/+147
directories to process. The removal of pre-configure fixes the config.* handling at the same time. Also specify the config.status overrides. Patch configure to not try to detect -lrfftw, just link the static version explicitly. Do the same for -lfftw. This ensures that we actually do get the static versions without having to worry about libtool being too clever. Retire manual config.sub hacks, this should be handled by config.guess already. <jlam> just go ahead and commit that benchfft patch and let's <jlam> never speak of it again :)
2005-11-05Regenerate hand-edited patches so that the Mac OS X patch programkristerw4-22/+22
can apply them.
2005-10-31NOT_FOR_PLATFORM Interix.tv1-1/+3
2005-10-31NOT_FOR_PLATFORM Interix (only 32-bit off_t).tv1-1/+3
2005-10-31This needs libgetopt on some systems.tv1-1/+2
2005-10-314.1.0nb1: Interix has no sync(2). Use fsync(2) there instead.tv3-3/+55
While there, use fsync(2) on all hosts. Without this, the fstime benchmark will be disproportionately biased toward OS's and hardware that buffer larger amounts of data in memory before [background] syncing.
2005-10-31NOT_FOR_PLATFORM Interix (only 32-bit off_t).tv1-1/+3
2005-10-20Add RMD160 checksum.wiz1-1/+2
2005-10-19Changes 3.248:adam3-15/+16
* Improvements
2005-10-13Added benchmarks/forkbombxtraeme1-1/+2
2005-10-13Initial import of forkbomb-1.4.xtraeme4-0/+35
Classic Unix fork() bomber. Includes CPU hanger, memory allocator, memory toucher and zombie dance team. You can test how will your computer behave under heavy CPU, process, memory load. Forkbomb is also useful as realloc() benchmark.
2005-10-13Correct execl() so that the argument list is NULL terminated.ben2-9/+14
Terminating with 0 worked under NetBSD 2.0.2, but it fails with EFAULT under NetBSD/amd64 3.0_BETA.
2005-10-10Explicitly specify to Makefile.in the directory on which the manpagesxtraeme3-4/+25
will be installed, otherwise as --mandir now defaults to ${PREFIX}/${PKGMANDIR}, they won't be installed into the correct location. Bump PKGREVISION.
2005-09-28Removed unneeded patch-aa.rillig1-122/+0
2005-09-28Fixed pkglint warnings. Converted patch-ab to unified diff format. Movedrillig4-24/+133
hint.1 from inside patch-aa to the files/ directory.
2005-09-25Added a chunk to patch-aa that comments out unnecessary code (statvfs). Nowrillig3-6/+22
the package builds on NetBSD-1.6.2. Bumped PKGREVISION.
2005-09-03Removed the $Id$ line from the patch. Restored -kkv mode.rillig2-5/+4
2005-09-03Re-add the patch with -ko, as suggests pkglint.xtraeme2-2/+16
2005-09-03Remove the patch, contains $Id$ expandedxtraeme1-15/+0
2005-09-03Remove the line $Id$ in the patch.xtraeme2-6/+7
2005-09-03eek, fix the patch.xtraeme2-6/+6
2005-09-03Fix HOMEPAGExtraeme1-2/+2
2005-09-03Added benchmarks/pipebenchxtraeme1-1/+2
2005-09-03Initial import of pipebench-0.40.xtraeme5-0/+41
Measures the speed of a pipe, by sitting in the middle passing the data along to the next process.
2005-09-03Added benchmarks/nbenchxtraeme1-1/+2
2005-09-03Initial import of nbench-2.2.2.xtraeme5-0/+51
This benchmark utility is used to determine how the processor, its caches and coprocessors influence overall system performance. Its measurements can indicate problems with the processor subsystem and (since the processor is a major influence on overall system system performance) give us an idea of how well a given system will perform.
2005-09-03Added benchmarks/ubenchxtraeme1-1/+2
2005-09-03Initial import of ubench-0.32.xtraeme5-0/+59
The Unix Benchmark Utility "ubench" provides a single mesure of performance for machines running various flavors of Unix operating systems.
2005-08-23Use Dimension instead of int for the width and the height.hira3-8/+17
This makes the package work properly on NetBSD/macppc. Bump PKGREVISION.
2005-08-10Simplify the check for SunOS, LIBS.SunOS is enough.xtraeme1-4/+2
2005-08-10Use "PTHREAD_LIBS" in build phase to make sure that this builds ontron1-2/+2
platforms without a "-pthread" option supported by the compiler.
2005-08-09Update "netio" package to version 1.23. Changes since version 1.13 include:tron3-47/+12
- UDP support - sets socket buffer size by default for decent performance under *BSD
2005-08-06Removed PKG_MISC_TARGETS, as it doesn't work anyway. Many packages in thisrillig1-2/+1
category don't even implement these targets. ok'ed by jlam.
2005-08-02explicitely include <sys/types.h> before <machine/ieee.h>drochner2-7/+8
might fix build error on older NetBSD releases (seen in 1.6.2 bulk build)
2005-07-27+zelibmdrochner1-1/+2
2005-07-27import "zelibm", a tool to check the accuracy of some libm functions,drochner9-0/+218
by Ulrich Drepper
2005-07-16Remove some unnecessarily strong dependencies on perl that resultedjlam2-5/+4
from including perl5/buildlink3.mk. These packages just need the Perl interpreter, and can just add "perl" to USE_TOOLS instead.
2005-07-16Get rid of USE_PERL5. The new way to express needing the Perl executablejlam2-5/+4
around at either build-time or at run-time is: USE_TOOLS+= perl # build-time USE_TOOLS+= perl:run # run-time Also remove some places where perl5/buildlink3.mk was being included by a package Makefile, but all that the package wanted was the Perl executable.
2005-07-15Drop distinction between PKGSRC_USE_TOOLS and USE_TOOLS by makingjlam2-4/+4
PKGSRC_USE_TOOLS go away. There is now only a single USE_TOOLS variable that specifies all of the tools we need to build/run the package.
2005-07-12Changes 3.03:adam4-57/+14
* Many improvements
2005-07-06Import thrulay-0.6, TCP and UDP network capacity tester.reed1-1/+2
Packaged by Rui Paulo.
2005-07-06Import thrulay-0.6, a TCP and UDP network capacity tester.reed4-0/+50
This package is from rpaulo@, who asked me to import. (My only change was to use INSTALLATION_DIRS.) I only tested build. DESCR: thrulay is used to measure the capacity of a network by sending a bulk TCP stream over it. Like other tools (such as iperf, netperf, nettest, nuttcp, ttcp, etc.), thrulay can report TCP throughput periodically so that TCP performance plots can be produced. Unlike other tools, thrulay not only reports goodput, but round-trip delay time as well. The output of thrulay is easy to parse by machine (in fact, it's ready to be used as a data file for gnuplot). Starting from version 0.5, thrulay supports UDP tests. Unique feature of thrulay is that it can send a Poisson stream of very precisely positioned packets; the TSC register is used for timing rather than very coarse (20ms on most systems) system sleep functionality.
2005-06-17Fix inappropriate uses of ${LOCALBASE} or ${X11PREFIX} instead ofjlam1-3/+3
${PREFIX}.
2005-06-16Create directories before installing files into them.jlam1-1/+3
2005-06-12regen patch-aa (as standard unidiff). Addresses part of PR 29877.wiz2-29/+21
generated by cgit v1.2.3 (git 2.39.1) at 2024-08-31 13:25:54 +0000