little bit and allows to proceed to a more recent linux_base from
a stable (read as: the major bugs should be ironed out or identified
and most linux ports build just fine) source.
It also allows to ship 4.11 with a working linuxolator (the EOLed
linux_base is marked forbidden because of a security hole).
This is a major update, please read UPDATING (and CHANGES if you
develop linux ports).
Changes:
- change the default linux_base from v7 to v8
- add a newer freetype to linux_base-8 for nicer fonts display [1]
- don't let cpio use hardlinks in the linux_base-8 port to quiet some
warnings in some cases [2]
- fix a cut&past error in the linux_base-8 pkg-install script [3]
- convert the binary knob "USE_LINUX" to a version specifier, e.g.
USE_LINUX=<value> specifies a dependency upon
emulators/linux_base-<value>, exceptions are a value of "7" (which
does what you want and adds a dependency to linux_base) and any
value without a corresponding port in
PORTSDIR/emulators/linux_base-<value> (which adds a dependency to
the default linux_base)
- don't implicitly add USE_LINUX with the USE_LINUX_PREFIX knob,
this allows us to use the USE_LINUX_PREFIX knob for linux_base and
paves the way for splitting up future linux base ports into
individual pieces
- remove RESTRICTED from some GPL licensed ports, even when we only
distribute binaries, we get them from official linux sites, so
anyone can grab them there if he needs to
- add a dependency upon the linux X11 bits where necessary (based upon
guesswork)
- don't use USE_X_PREFIX in some linux ports since it adds a dependency
to the FreeBSD X11 libs, as a workaround use PREFIX?= (the clean
solution would be to remove the implicit USE_XLIB from USE_X_PREFIX)
- bump the portrevision of the linux ports ("better safe than sorry"
algorithm)
- pass maintainership of the important linux infrastructure to a
mailinglist, hijack freebsd-emulation@ for this purpose (if somebody
doesn't like this: tell us your bikeshed color at freebsd-emulation@,
my color would be "linuxolator@" in case someone cares...)
- add a pkg-install script for linux-fontconfig, but don't use it;
everything should work without it (the FreeBSD fc-cache program should
do all the work), but in case we need it we just need to decomment the
pkg-install part in the Makefile
- fix some dependencies
- fix some bugs
- add some static plists
- unbreak the ports with dependecies to more than one linux_base
This also fixes some ports which are marked BROKEN because of dependencies
to v7 and v8 of linux_base at the same time.
Known bugs:
- the linux-mesa and linux-devtools ports install libGL*.so symlinks
- some "minor" plist bugs (e.g. ld.so.{conf,cache} are modified by
the linux X11 port, so linx_Base-8 moans at deinstall time)
Future work (interested souls should coordinate with freebsd-emulation@):
- add some kind of USE_LINUX_X11 knob to streamline the X11 dependencies,
or modify the behavior of USE_XLIB in the USE_LINUX case
AFAIK trevor has some patches.
- make USE_XLIB and USE_X_PREFIX orthogonal to be able to get rid of
the PREFIX?= workaround in some linux ports
Should be discussed/coordinated on/with x11@.
- move the RPM bits from x11-toolkits/linux-gtk/Makefile to PORTSDIR/Mk/
- update to a more recent linux base
PR: 69997, 70539 (and maybe others)
Discussed with/on: java@, x11@, trevor, portmgr
Tested by: mezz, portmgr, pointyhat
RPM hunted down by: Joseph Gelinas <scirocco@tasam.com> [1]
Requested by: portmgr [2]
Submitted by: kris [3]
Approved by: portmgr
- Incorporate the C++ DSO Object Destruction API for < 502101 systems
(icc v8 already does this). [1]
If you use icc7 on such a system, you have to recompile C++ programs
(including stldport-icc).
- Simplify the post-extract now that we only extract one rpm. [1]
Submitted by: Marius Strobl <marius@alchemy.franken.de> [1]
where appropriate [1]
- make portlint happy [1]
- sync icc7 and icc [1]
- add linux_base as a patch depends for icc v8
Submitted by: Marius Strobl <marius@alchemy.franken.de> [1]
Requested by: maintainer [1]
- correct the use of ECHO_CMD and ECHO (swap them) [1]
icc:
- fix the DISTFILE handling, it's automatically available after
bsd.port.post.mk, not after bsd.port.pre.mk, so set it explicitly
to be able to use it in the check for the IGNORE message [1]
icc7:
- don't extract the Intel debugger, it's not usable without a
threads debugging lib
- USE_SIZE
Noticed after: reading the commit log/diff of the
ifc port [1]
Submitted indirectly by: maho, hrs [1]
There slipped in a hardcoded /usr/local, it was supposed to be a
placeholder for LOCALBASE (to find the stlport lib in the ld wrapper).
Thanks to: Marius Strobl <marius@alchemy.franken.de>
for reviewing the commits.
NOTE: you need to rebuild stlport-icc and maybe some other C++
programs/libs.
- rework ld.c to fix the build of stlport-icc on 4.x (first part
of the build fix, the second part follows shortly in a stlport
commit) [1]
Submitted by: Marius Strobl <marius@alchemy.franken.de> [1]
script was renamed to solve a conflict with archivers/rpm) to fix possible
build problems.
I've tested this with lang/icc. Any new errors because of this commit in
one of the modified ports may be because the ports previously may have used
rpm2cpio from archivers/rpm instead of the used {EXTRACT,BUILD}_DEPENDS
archivers/rpm2cpio.
- Transform some warnings into errors as suggested by some included
docs (some kind of MSVC compatibility which isn't reverted in icc
for linux).
ld.c:
- add possibility to use a different threads lib via PTHREAD_LIBS
variable (e.g. PTHREAD_LIBS=-lthr) [1]
this may be subject to change when gcc learns how to handle our
different threads libs
- refactor some code [1][2]
- remove mailwrapper license, there's no code from mailwrapper
anymore [2]
- correct the order of libc and libc_r [1][2]
Submitted by: mi [1]
Submitted by: Marius Strobl <marius@alchemy.franken.de> [2]
Reviewed by: Marius Strobl <marius@alchemy.franken.de> [1]
icc -shared -o libfoo.so foo.o -lbaz
the ld wrapper gets confused and thinks that a static link is intended
and the link fails. This patch appears to fix things.
Submitted by: dfr
(this may break ports which depend upon OpenSSL from ports which was
compiled as a base system replacement because it includes a system
header directory again)
- ignore "-pipe" in CFLAGS, this should unbreak some ports with hardcoded
"-pipe"
Noticed by: Krzysztof Parzyszek <kristof@swissmail.org> [1]
Tested by: Krzysztof Parzyszek <kristof@swissmail.org> [1]
path, thus adding a system path with -I results in not respecting the
sunstitute headers. This results in problems because we have some important
changes there.
Parts of this commit where
Submitted by: marius@alchemy.franken.de
- fix [dfi]vec.h with stlport-iostreams
- do not install a Windows header (mathf.h)
- do not install libompstub (depends on pthread_atfork(), see PR 17437)
Submitted by: marius@alchemy.franken.de
- point to the icc errata after make install
Note:
Object files built with version 6.0 of Intel(R) Compilers may not link
properly with objects built with version 7.0 of Intel(R) Compilers.
Rebuilding of the object files is suggested.
Changes between the submitted version and the committed version:
- update to 078 instead of 075
- treat excessive parameters in function calls as an error instead
of a warning (removed compatibility with a not named MS product, as
suggested by Intel)
- allow to override the INTEL_FLEXLM_LICENSE variable
From the submitter:
- Unbreak usage of signal.h on -current and RELENG_5_0.
(committers note: see below)
- Fix ownership of installed files when built as non-root.
- Fix the patched versions of the icc- and the icpc-script to work with options
containing spaces (e.g. '-DFOOBAR="foo bar"').
- Install 'clicense' and 'csupport' in the docs-dir even if NOPORTDOCS is
defined, the later is not optional but required by 'icid'.
- Move the ld-wrapper to a sub-dir and thus out of what is normally set in
PATH to prevent footshooting (this isn't perfect but simple until someone
reworks this port to provide FreeBSD-compatibility in another way... or
Intel releases a native version).
- Patch the headers to better fit for FreeBSD (hopefully...).
Notes: - Icc7 is more picky about unknow options than the previous versions,
i.e. the gcc-options normally supplied in CFLAGS unless altered in
/etc/make.conf. Therefore compilation of ports with icc likely fail
if the standard CFLAGS are not unset (e.g. `make CC=icc CFLAGS=""`)
or replaced by options valid for icc.
- When compiling multi-threaded C-code make sure to link with icc and
option '-mt' (for ports e.g. via PTHREAD_CFLAGS) otherwise libc_r
won't be linked in correctly.
Submitted by: marius@alchemy.franken.de
Additional information:
- rev 1.20 of src/sys/i386/include/signal.h introduced __aligned which
isn't handled in sys/cdefs.h for non GCC compilers (at least not in
publically available sources)
- the FreeBSD stdarg.h is missing a macro for va_copy() in the non GCC case,
we handle it in the port, but be prepared to get problems when the base
system gets fixed
- we don't use the icc-buildin of alignof()
- Intel provides a float.h which has different floatingpoint values
(e.g. MAX_FLOAT) than we have in the base system, in the port we use the
FreeBSD header (Warner knows about the issue... at least he got a mail
from me and Marius)
- we replace __wchar_t with __ct_rune_t because icc has a build-in type
for it, this affects "typedef __ct_rune_t __wchar_t;" in sys/_types.h.
It isn't known if this may be evil...
- icc doesn't need the option "-mt" anymore to generate thread safe code,
but our FreeBSD version still needs it to be able to choose libc_r
instead of libc
If a src-committer wants to address any of the above mentioned issues he
should first contact me, I may have already some fixes in my local tree.
