Add a partial fix for what was reported in PR 231804, the part that's
actually relevant upstream and that I made there, cf.
https://gcc.gnu.org/ml/gcc-patches/2018-10/msg00076.html as a new
local patch files/patch-libgfortran until I have backported this to the
upstream release branch and hence we get it via the weekly snapshots.
PR: 231804
ports r393217 via bug 200622 [1] originally set MAKE_JOBS_UNSAFE=yes due to
incorrect uses of recursive make [2], causing intermittent build failures when
run with multiple jobs (-jN).
Upstream committed a fix for the issue in default (3.6, at the time), 3.5 and
2.7 which are now contained in all released lang/python?? port versions. 3.4 did
not receieve a backport merge.
lang/python3.5+ ports inadvertently inherited MAKE_JOBS_UNSAFE=yes, via
repocopies from lang/python34 on their creation, when they were infact safe to
use with -j.
Remove MAKE_JOBS_UNSAFE in all lang/python?? ports except python34 accordingly.
[1] https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=200622
[2] https://bugs.python.org/issue22359
PR: 232308
Reported by: cem
Reviewed by: cem
Approved by: koobs (python)
MFH: 2018Q4
Differential Revision: D17579
Add a partial fix for what was reported in PR 231804, specifically the
part that's actually relevant upstream and that I made there, cf.
https://gcc.gnu.org/ml/gcc-patches/2018-10/msg00076.html as a new local
patch files/patch-libgfortran until I have backported this to the upstream
release branch and hence we get it via the weekly snapshots.
PR: 231804
20181014:
- Simple function profiling has been added to the compiler. Profiling
is enabled, when programs are compiled with the option -p. When
programs with profiling are executed they write profiling information
to the file profile_out, when they terminate.
- An explanation of the compiler option -p (simple profiling) has been
added to the FAQ.
- The function str_dd_mm_yy has been added to the library time.s7i.
- In chkflt.sd7 the tests for the operator ** (with an exponent of NaN)
have been improved.
- Documentation comments have been improved in integer.s7i, intlib.c,
int_rtl.c, tim_unx.c and tim_win.c.
- In hsh_rtl.c the function hshIdxEnterDefault has been simplified to
use fewer parameters. Usages of hshIdxEnterDefault in gkb_win.c,
gkb_x11.c, pol_sel.c, pol_unx.c, ref_data.c and typ_data.c have been
adjusted to the reduced number of parameters.
- In int_rtl.c the performance of the function intParse has been
improved (when two's complement integers are used).
- In pcs_unx.c the function pcsStart has been improved to work correct
when stdin, stdout or stderr are not redirected.
- In flt_rtl.c the function fltPow() has been improved to work correct,
when the base is zero and the exponent is NaN.
- The program chkccomp.c has been improved to check the function pow()
with more test cases.
- In striutl.c the function heapAllocOsStri() has been refactored to
return the allocated string instead of using a pointer to a string
variable. This opens the possibility to optimize usages of the macro
os_stri_alloc().
- The definition of WITH_SQL has been removed from the makefiles and
added to the file config.h instead.
- The file config.h has been changed to define the macros WITH_FLOAT,
WITH_REFERENCE, WITH_DRAW, WITH_STATISTIC, WITH_COMPILATION_INFO and
WITH_TYPE_CHECK always.
- The function timMicroSec in tim_unx.c and tim_win.c has been changed
to be useable for the simple profiling.
Notable changes:
- Switch from PCRE to PCRE2
- Many modules now require PCRE2 for building
- graphics/php73-gd: X11 option is no longer default
Changes to Mk/Uses/php.mk approved by ale
Add a partial fix for what was reported in PR 231804, the part that's
actually relevant upstream and that I made there, cf.
https://gcc.gnu.org/ml/gcc-patches/2018-10/msg00076.html as a new
local patch files/patch-libgfortran until I have backported this to the
upstream release branch and hence we get it via the weekly snapshots.
PR: 231804
the default version of GCC for the Ports Collection.
And backport r478722 | gerald | 2018-09-01 from lang/gcc7: [1]
Disable the build/use of libssp/gets-chk since FreeBSD 12 and later
do not feature gets() any longer.
PR: 222796, 231066 [1]
Differential Revision: https://reviews.freebsd.org/D12298 [1]
- While here, do not force ${SHAREMODE} (444) on docfiles as well: default
permissions (644) are perfectly fine
Reported by: pkg-fallout
PR: 231802
Submitted by: maintainer
These ports link some non-PIC code, which fails with lld as it defaults
to disallowing relocations against read-only segments. For i386 we can
just add -znotext unconditionally: for GNU BFD ld it just affirms BFD's
existing default.
PR: 214864
Approved by: portmgr (lld blanket)
Sponsored by: The FreeBSD Foundation
This resolves a ninja-related issues, corrects passing private args to linker
while building libruby itself (without external LDFLAGS), exports the full
symbols table, and pet 'pkg check -d'. Issues with weechat, vim, facter,
rubygem-hpricot and rubygem-rdiscount should be fixed. See [1] for long story
of hunting and fixing these issues.
Bump PORTREVISIONS accordingly.
PR: 230238 [1], 231361
Submitted by: fuffy
Approved by: fuffy
mv: rename /wrkdirs/usr/ports/lang/rust-nightly/work/rustc-nightly-src/src/vendor/backtrace-sys/src/libbacktrace/config/libtool.m4 to /wrkdirs/usr/ports/lang/rust-nightly/work/rustc-nightly-src/src/vendor/backtrace-sys/src/libbacktrace/config/libtool.m4.dont-fix: No such file or directory
*** Error code 1
As the port appears to be a staging ground for future updates don't
deprecate yet.
PR: 229826
Reported by: pkg-fallout
- libgit2 in ports is nowadays newer than bundled version
- libssh2 was already unbundled until 1.29.0 update
- Unbundle libssh2 for USES=cargo (unused atm)
PR: 231411
Approved by: rust (dumbbell, tobik)
It was initially got from ruby23, but recent ruby releases have some
differences in runtime linking, which was not hit by exp-run, and caused
trouble with some ports (e.g. devel/leatherman, sysutils/facter).
PR: 230238
Submitted by: fluffy
Reported by: freebsd@ptty.com
Approved by: fluffy
This port links some non-PIC code, which fails with lld as it defaults
to disallowing relocations against read-only segments. For i386 we can
just add -znotext unconditionally: for GNU BFD ld it just affirms BFD's
existing default.
PR: 214864, 230207
Approved by: bapt
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D17193
Example error:
/usr/bin/ld: error: can't create dynamic relocation R_386_32 against
symbol: caml_last_return_address in readonly segment; recompile
object files with -fPIC
This port links some non-PIC code, which fails with lld as it defaults
to disallowing relocations against read-only segments. For i386 we can
just add -znotext unconditionally: for GNU BFD ld it just affirms BFD's
existing default.
PR: 214864, 230209
Approved by: bapt
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D17200
Release notes:
* https://blog.rust-lang.org/2018/07/20/Rust-1.27.2.html
* https://blog.rust-lang.org/2018/08/02/Rust-1.28.html
* https://blog.rust-lang.org/2018/09/13/Rust-1.29.html
Up to and including Rust 1.27.x, the Rust build system shelled out to
a configure script to detect the presence and usability of libunwind.
Since Rust 1.28.0, it's using a static result in a `build.rs` file and
expects libunwind to be used. It was not the case on FreeBSD so far, so
we need a patch to this `build.rs` to disable that. We still need to
study if the FreeBSD port should use libunwind and what to do with this
patch. But this problem prevented the update to Rust 1.28.0 already, so
enough delay.
The update also comes with a patch to a few `USE_GECKO`-based ports such
as Firefox [1]. Their configure script has some asumptions on the output
of `rustc --print target-list` which are not true anymore. The patch was
already committed upstream.
The aarch64 version is still marked as BROKEN because I didn't find the
time to work on it. As a consequence, there is also no aarch64 bootstrap
for Rust 1.29.0.
PR: 229826
Approved by: jbeich [1]
Obtained from: https://bugzilla.mozilla.org/show_bug.cgi?id=1479540 [1]
Differential Revision: https://reviews.freebsd.org/D17178
This change merges the following upstream pull request into the Ruby
interpreter ports:
https://github.com/ruby/ruby/pull/1954
Adding this patch to these ports will ensure that once we patch up
qsort_r() to be compatible with glibc, Ruby will automatically pick them
up. Ruby should also build fine without this patch, but this will cause
it to use its own implementation, which blows up the binary size
slightly.
Poudriere runs seem to pass for 11.x amd64. Logs indicate that this
doesn't negatively affect the existing qsort_r() detection:
checking whether qsort_r is GNU version... no
checking whether qsort_r is BSD version... yes
PR: 231256
Approved by: sunpoet
Differential Revision: https://reviews.freebsd.org/D17157
Change x11/xorgproto to become a build time dependency when added to
USE_XORG. Change the dependency to be on the port, rather than a file the
port installs.
Fix fallout.
Bump portrevision on depending ports.
PR: 230909
Reviewed by: eadler
Approved by: portmgr (antoine)
Obtained from: https://github.com/FreeBSDDesktop/freebsd-ports/tree/feature/xorgproto
exp-run: antoine
Differential Revision: https://reviews.freebsd.org/D16906
20180821:
- In fil_dos.c in the function setupFiles the handling of ctrl-c has
been improved such that ctrl-c is sent immediate (without the need to
press return afterwards). Additionally redirected stdin, stdout and
stderr files are switched to binary mode now.
- In the compiler (in comp/const.s7i) the actions BIN_XOR, ENU_ICONV2,
ENU_ORD2, INT_ICONV1 and INT_ICONV3 have been added to the list of
special actions.
- The batch files bin/call_gcc.bat and bin/call_ar.bat have been
improved to set environment variables with the command 'set'. This
is necessary to execute them successfully under Wine.
- The description of log, log10 and log2 in the manual has been
improved.
- In flt_rtl.c the functions fltLog, fltLog10 and fltLog2 have been
added. This functions are based on the original functions log(),
log10() and log2() and work according to IEEE 754.
- The configuration values LOG_FUNCTION_OKAY, LOG10_FUNCTION_OKAY and
LOG2_FUNCTION_OKAY have been added to cc_conf.s7i.
- The configuration value HAS_LOG2 has been removed from cc_conf.s7i.
- The compiler (s7c) has been improved to use the functions fltLog,
fltLog10 and fltLog2.
- In comp/inline.s7i the functions push_inline_func_param,
pop_inline_func_param, push_inline_proc_param and
pop_inline_proc_param have been improved to support inline functions
with 'attr' parameters.
- The example program diff7.sd7 has been improved.
- Documentation comments have been improved in math.s7i and tar.s7i.
- In chkflt.sd7 the tests for the operators 'parse' and 'sci' and for
the function str() have been improved.
- The program chkccomp.c has been improved to define the macros
POW_OF_NEGATIVE_OKAY, LOG_OF_NAN_OKAY, LOG_OF_ZERO_OKAY,
LOG_OF_NEGATIVE_OKAY, LOG10_OF_NAN_OKAY, LOG10_OF_ZERO_OKAY,
LOG10_OF_NEGATIVE_OKAY, LOG2_OF_NAN_OKAY, LOG2_OF_ZERO_OKAY,
LOG2_OF_NEGATIVE_OKAY, MEMCMP_RETURNS_SIGNUM and
WMEMCMP_RETURNS_SIGNUM.
- In chkccomp.c the function determineFseekFunctions has been improved
to make tests that actually use fseek() and ftell().
- Calls of logFunction have been added in cmd_rtl.c, str_rtl.c,
strlib.c and striutl.c.
- In bstlib.c the function bst_cmp() has been improved to be based on
bstCmp().
- The functions strCompare(), str_cmp() and bstCmp() have been
improved to use different code depending on MEMCMP_RETURNS_SIGNUM
and WMEMCMP_RETURNS_SIGNUM.
- In s7c.sd7 the function canTakeAddress has been improved to return
FALSE for a CONSTENUMOBJECT.
Disable the build/use of libssp/gets-chk since FreeBSD 12 and later
do not feature gets() any longer.
(I was planning to make this part of a routine update, alas last week's
snapshot was broken on i386-unknown-freebsd10.x and so will this week's
still.)
PR: 222796, 231066
Differential Revision: https://reviews.freebsd.org/D12298
I was too optimistic about the fix from r478932 and it quickly falls
to pieces with a higher number (> 16) of jobs. Upstream has never
supported building in parallel either.
