Increase array test sizes and reps to wash out sampling error on modern
hardware, where run times were becoming too short and inconsistent.
A few other minor enhancements
Update the CONFLICTS definitions of ports in the following categories:
- accessibility
- archivers
- audio
- benchmarks
- biology
- cad
- chinese
- comms
- converters
An attempt has been made to use generic conflicts patterns that do not
have to be updated whenever a new version of a conflicting port is
added to the ports system.
There is a misunderstanding that the port being built/installed has to
be omitted from the conflicts pattern. This is not true - the port
being built is implicitly non-conflicting due to logic in bsd.port.mk.
Approved by: portmgr (implicit)
There have been lots of missing CONFLICTS_INSTALL entries, either
because conflicting ports were added without updating existing ports,
due to name changes of generated packages, due to mis-understanding
the format and semantics of the conflicts entries, or just due to
typoes in package names.
This patch is the result of a comparison of all files contained in
the official packages with each other. This comparison was based on
packages built with default options and may therefore have missed
further conflicts with optionally installed files.
Where possible, version numbers in conflicts entries have been
generalized, some times taking advantage of the fact that a port
cannot conflict with itself (due to logic in bsd.port.mk that
supresses the pattern match result in that case).
A few ports that set the conflicts variables depending on complex
conditions (e.g. port options), have been left unmodified, despite
probably containing outdated package names.
These changes should only affect the installation of locally built
ports, not the package building with poudriere. They should give an
early indication of the install conflict in cases where currently
the pkg command aborts an installation when it detects that an
existing file would be overwritten,
Approved by: portmgr (implicit)
The CPU_SET macros in -CURRENT have been made compatible with GLIBC.
A simple test for the new signature of CPU_AND, CPU_OR, ... is the
existence of a CPU_ALLOC macro.
Approved by: portmgr (implicit)
MFH: 2022Q1
In CURRENT, the initial configuration detects now support for the
scheduler affinity, enabling the build of an additional stressor.
In this stressor, the CPU_OR macro has 3 arguments (Linux implementation),
while on FreeBSD it has only 2 arguments.
This patch provides the fix
No bumprevision is provided, because it's only fixing CURRENT where the
build was failing.
- Report CPU frequency and temperature on the HUD
- For gathering system information, use sysctlbyname() if
possible instead of capturing external programs' output
Now comes with CPU load monitoring enabled by default and better
system information report in debug mode. Note that those things
currently require mounted linprocfs(5) to work.
Reported by: portscout
- Add missing dependency for playing music: USE_LINUX+=openal-soft
- In the `do-extract', rather than changing current directory, pass it to
tar(1) as a -C argument and don't needlessly use semi-private _DISTDIR
variable: this makes the command shorter and avoids bogus double slash
in the build log
- Provide more user-friendly wrapper script which anticipates potential
problems that typically affect running this program and advises on how
to solve them
- Spell GNU/Linux correctly in the COMMENT and port description; reformat
the latter with `fmt -w 75' for better looks (less ragged right edge)
error[E0432]: unresolved imports `core::sync::atomic::AtomicI64`, `core::sync::atomic::AtomicU64`
--> /wrkdirs/usr/ports/benchmarks/inferno/work/inferno-0.10.7/cargo-crates/crossbeam-utils-0.8.5/src/lib.rs:79:49
|
79 | pub(crate) use core::sync::atomic::{AtomicI64, AtomicU64};
| ^^^^^^^^^ ^^^^^^^^^ no `AtomicU64` in `sync::atomic`
| |
| no `AtomicI64` in `sync::atomic`
|
help: a similar name exists in the module
|
79 | pub(crate) use core::sync::atomic::{AtomicI8, AtomicU64};
| ~~~~~~~~
help: a similar name exists in the module
|
79 | pub(crate) use core::sync::atomic::{AtomicI64, AtomicU8};
| ~~~~~~~~
error[E0412]: cannot find type `AtomicU64` in module `core::sync::atomic`
--> /wrkdirs/usr/ports/benchmarks/inferno/work/inferno-0.10.7/cargo-crates/crossbeam-utils-0.8.5/src/atomic/consume.rs:78:14
|
78 | impl_atomic!(AtomicU64, u64);
| ^^^^^^^^^ help: a struct with a similar name exists: `AtomicU16`
error[E0412]: cannot find type `AtomicI64` in module `core::sync::atomic`
--> /wrkdirs/usr/ports/benchmarks/inferno/work/inferno-0.10.7/cargo-crates/crossbeam-utils-0.8.5/src/atomic/consume.rs:80:14
|
80 | impl_atomic!(AtomicI64, i64);
| ^^^^^^^^^ help: a struct with a similar name exists: `AtomicI16`
Some errors have detailed explanations: E0412, E0432.
For more information about an error, try `rustc --explain E0412`.
error: could not compile `crossbeam-utils` due to 3 previous errors
Unigine benchmarks are extreme performance and stability test for your
hardware: video card, power supply, cooling system. Check your rig in
stock and overclocked modes with real-life load!
These two benchmarks depict a Gothic chapel and a tropical paradise.
WWW: https://benchmark.unigine.com/
This program allows to compare the performance of different POSIX
shells (e.g. sh, bash, dash, ksh, zsh, ...).
A number of sample scripts are included and can be easily extended to
cover more areas.
vkpeak is a synthetic benchmarking tool to measure peak capabilities
of Vulkan devices. It only measures the peak metrics that can be
achieved using vector operations and does not represent a real-world
use case.
https://github.com/nihui/vkpeak
- Update KDiskMark to version 2.2.1
- Make cache clearing and KDE dependency optional
- Amend port description and add pkg-help file
Reported by: portscout
- Belatedly update to version 1.2.1 and sync port description
- Builds with CMake now and uses modern libraries instead of the
Apache Portable Runtime (devel/apr1)
- While upstream asks for C++11, force C++17 because that's how
Abseil C++ libraries are currently built, otherwise it won't
link due to ABI incompatibilities
- Hook two provided tests to our framework (USES+=cmake:testing)
Per discussion with bapt on helping pkg handle the changing of these
deps and avoiding impossible upgrade senarios.
PR: 246767
Reviewed by: manu, bapt
Approved by: x11
Differential Revision: https://reviews.freebsd.org/D30824
MST-bench is a simple program to measure optimal sustained memory and disk
performance.
Unlike many benchmarks, it does not attempt to simulate naturally occurring
loads. Rather, it indicates what is the best performance you can expect out of
your hardware, providing a reference to which software can be compared.
From [1]
* What is new in gsl-2.7:
* fixed doc bug for gsl_histogram_min_bin (lhcsky at 163.com)
* fixed bug #60335 (spmatrix test failure, J. Lamb)
* fixed bug #36577
* clarified documentation on interpolation accelerators (V. Krishnan)
* fixed bug #45521 (erroneous GSL_ERROR_NULL in ode-initval2, thanks to M. Sitte)
* fixed doc bug #59758
* fixed bug #58202 (rstat median for n=5)
* added support for native C complex number types in gsl_complex
when using a C11 compiler
* upgraded to autoconf 2.71, automake 1.16.3, libtool 2.4.6
* updated exponential fitting example for nonlinear least squares
* added banded LU decomposition and solver (gsl_linalg_LU_band)
* New functions added to the library:
- gsl_matrix_norm1
- gsl_spmatrix_norm1
- gsl_matrix_complex_conjtrans_memcpy
- gsl_linalg_QL: decomp, unpack
- gsl_linalg_complex_QR_* (thanks to Christian Krueger)
- gsl_vector_sum
- gsl_matrix_scale_rows
- gsl_matrix_scale_columns
- gsl_multilarge_linear_matrix_ptr
- gsl_multilarge_linear_rhs_ptr
- gsl_spmatrix_dense_add (renamed from gsl_spmatrix_add_to_dense)
- gsl_spmatrix_dense_sub
- gsl_linalg_cholesky_band: solvem, svxm, scale, scale_apply
- gsl_linalg_QR_UD: decomp, lssolve
- gsl_linalg_QR_UU: decomp, lssolve, QTvec
- gsl_linalg_QR_UZ: decomp
- gsl_multifit_linear_lcurvature
- gsl_spline2d_eval_extrap
* bug fix in checking vector lengths in gsl_vector_memcpy (dieggsy@pm.me)
* made gsl_sf_legendre_array_index() inline and documented
- gsl_sf_legendre_nlm()
[1] http://git.savannah.gnu.org/cgit/gsl.git/tree/NEWS
PR: 256423
Exp-run by: antoine
USE_GCC=any has been equivalent to USE_GCC=yes in most cases (such
as i386 and amd64 since 12.x and depending on configuration 11.x,
most newer installations on other platforms, and 13.x across the
board).
Since commit 96c17633d9 Mk/bsd.gcc.mk is treating them as
different spellings of the same, so continue the deorbiting of the
USE_GCC=any form and simply replace it with USE_GCC=yes.
This should not make any functional difference at all.
Discussed with: mat, linimon, pkubaj
For ports that already use the licenses framwork, merge the content of
RESTRICTED/NO_CDROM/LEGAL* entries into LICENSEs.
Approved by: rene
Differential Revision: https://reviews.freebsd.org/D30010
- Bump portrevision of dependent ports.
- Add patches to dependent ports math/octave-forge-* if I can figure them out.
- Mark broken dependent ports I couldn't figure out.
Reason: Many bug reports (including bug#253256) related to jlaffaye's ports timed out in the period from April 2020 to now.
PR: 253256
Approved by: jlaffaye (maintainer's timeout 14 days)
- Gargage-collect patches integrated upstream
- Now really requires C++14-capable compiler, so
stop trying to force C++11
- Disable Wayland flavor and mark it as BROKEN
Reported by: portscout
The option LIBINOTIFY, on by default, can be used to not build inotify
stressor and install its dependency.
No portrevision bump needed, inotify support was already enable.
stress-ng supports libinotify. A kevent based implementation can be used
to provide support, fixing a build issue reported in PR 252661
PR: 252661
Reported by: dewayne@heuristicsystems.com.au
Windowless can mean a window without border, so mention console.
GBM needs KMS-backed console as EFI and text mode won't work.
Simply loading DRM drivers after boot is enough.
Version 0.4.2
- Fixed globals in master and slave (had a few non-static globals with same
name)
Version 0.4.1
- Fixed bug that included random bytes in the hash input used to generate the
packet id in cases where the captured bytes did not include the TCP option
bytes or the first 12 bytes for TCP/UDP payload. Now for the TCP options, the
UDP/TCP payloads and data for non UDP/TCP, the capture length (snap size) will
limit the hash input.
Changes: https://bitbucket.org/caia-swin/spp/src/v0.4.2/Changelog.txt
PR: 251002
