The LLVM Project is a collection of modular and reusable compiler and
toolchain technologies.
This port includes Clang (a C/C++/Objective-C compiler), LLD (a linker),
LLDB (a debugger), an OpenMP runtime library, and the LLVM infrastructure
these are built on.
===> The following configuration options are available for llvm80-8.0.1_7:
BE_AMDGPU=on: AMD GPU backend (required by mesa)
CLANG=on: Build clang
COMPILER_RT=on: Sanitizer libraries
DOCS=on: Build and/or install documentation
EXTRAS=on: Extra clang tools
GOLD=on: Build the LLVM Gold plugin for LTO
LIT=on: Install lit and FileCheck test tools
LLD=on: Install lld, the LLVM linker
LLDB=on: Install lldb, the LLVM debugger
OPENMP=on: Install libomp, the LLVM OpenMP runtime library
PYCLANG=on: Install python bindings to libclang
====> Options available for the single BACKENDS: you have to select exactly one of them
BE_FREEBSD=off: Backends for FreeBSD architectures
BE_NATIVE=off: Backend(s) for this architecture (X86)
BE_STANDARD=on: All non-experimental backends
===> Use 'make config' to modify these settings
Cleanup: Remove devel/llvm[789]0
* remove old devel/llvm70
* remove old devel/llvm80
* remove old devel/llvm90
* remove ports explicitely depending on either of the three
Differential Revision: https://reviews.freebsd.org/D37936
devel/llvm*: Fix CONFLICT on cmark
If the user has the textproc/cmark package installed, the py*-CommonMark
package will fail to install. This in turn leads to build failures down
the line when sphinx tries to load commonmark. The result is confusing
and potentially wastes a lot of build time so just conflict upfront.
The previous code attempted to do this, but I'd listed the conflict
incorrectly in 2019. :(
PR: 239636
Reported by: Stephen Roznowski <sjrbulk@verizon.net>
Remove WWW entries moved into port Makefiles
Commit b7f05445c00f has added WWW entries to port Makefiles based on
WWW: lines in pkg-descr files.
This commit removes the WWW: lines of moved-over URLs from these
pkg-descr files.
Approved by: portmgr (tcberner)
Add WWW entries to port Makefiles
It has been common practice to have one or more URLs at the end of the
ports' pkg-descr files, one per line and prefixed with "WWW:". These
URLs should point at a project website or other relevant resources.
Access to these URLs required processing of the pkg-descr files, and
they have often become stale over time. If more than one such URL was
present in a pkg-descr file, only the first one was tarnsfered into
the port INDEX, but for many ports only the last line did contain the
port specific URL to further information.
There have been several proposals to make a project URL available as
a macro in the ports' Makefiles, over time.
(Only the first 15 lines of the commit message are shown above )
textproc/libxml2: bump all LIB_DEPENDS consumers
This is a separate commit to facilitate easier cherry-picking for
quarterly.
PR: 262853, 262940, 262877, 263126
Approved by: fluffy (mentor)
textproc/libxml2: bump all dependencies
This should make sure that all dependent ports will pick
up the new version commited with a13ec21cd733f67a9fc0dc00ab45268bdc236246
*/*: Remove redundant '-*' from CONFLICTS definitions
The conflict checks compare the patterns first against the package
names without version (as reported by "pkg query "%n"), then - if
there was no match - agsinst the full package names including the
version (as reported by "pkg query "%n-%v").
Approved by: portmgr (blanket)
devel/llvm*: Fix BE_NATIVE on aarch64
Fix two bugs in native-only backend support (BE_NATIVE). First, ARCH
is aarch64 not arm64. Second, consistently spell AArch64 not AAarch64
(extra 'a').
Because this only effects aarch64 in a non-default configuration, skip
the PORTREVISION bump that is technically required to reduce excess
heating in common cases.
Reported by: Mark Millard
Also mark BROKEN on riscv64 (the ports framework is picky about such
things). Add a note shooing people away from trying it anyway.
While here, pet portlint (trailing whitespace).
Package is not affected, so no need for revision bump.
Approved by: portmgr (tier-2 blanket)
Obtained from: lonesome.com build testing
devel/llvm*: Avoid extraneous escapes
All of these single quotes are safe on their own, as they're enclosed in
double quotes. Removes the escapes, because they're currently passed through
to the underlying regex engine as \' to be interpreted as '.
A future change will forbid this escape, then an even further future change
will actually imbue \' with special meaning based on GNU interpretation
("end of subject").
No PORTREVISION bump or rebuild is needed, as they're functionally
equivalent.
PR: 247058
Approved by: koobs (mentor), brooks (maintainer)
MFH: 2020Q2 (blanket: trivial build fix)
devel/llvm*: Upgrade Sphinx dependency to 3.X
This avoids conflicts between Sphinx ports when not using Poudriere.
PR: 246500
Approved by: brooks (maintainer)
Update Sphinx
- Repocopy textproc/py-sphinx to textproc/py-sphinx18
Update it to 1.8.5 (latest version from 1.8.X).
This version supports Python 2 and 3.
Add test target.
- textproc/py-sphinx: Update to 3.0.2
Python 3 only (3.5+).
Add test target.
- Mk/Uses/python.mk: Add PY_SPHINX
(Only the first 15 lines of the commit message are shown above )
Drop C++11 workaround for GCC < 6 after r449590
lang/gcc5 will expire in ~1 month. If someone still uses lang/gcc48
they should backport C++11 fix instead. QA on GCC architecuters is
currently limited to powerpc64 which uses GCC_DEFAULT.
PR: 193528
Assorted build improvements:
- Add a build conflict for commonmark-cmark-* when DOCS are enabled.
This prevents a failure later on in the build. [0]
- Add a new option BE_AMDGPU which can be used to enable the AMDGPU
backed used by mesa when BE_NATIVE or BE_FREEBSD is set. Enable this
option by default to limit later surprises. [1]
- New option PYCLANG to add python binding for clang. [2]
PR: 239636 [0], 230789 [1], 239990 [2]
Submitted by: chardon.frederic@gmail.com [2]
Sponsored by: DARPA, AFRL
Allow builds with a default python version on 3.7.
Further refinement and testing is required, but this version achieves
the stated goal and is tested.
Submitted by: leres
Address a code generation bug that could allow the ARM stack protector
to be bypassed.
This change merges upstream r366369, r366371, and r267068 (minus some
test improvements).
Also:
- Address bugs breaking the build with all options disabled. [0]
- Pin the python version to 3.6 rather than 2.7.
PR: 239503 [0]
Security: https://kb.cert.org/vuls/id/129209
Bump PORTREVISION for ports depending on the canonical version of GCC
as defined in Mk/bsd.default-versions.mk which has moved from GCC 8.3
to GCC 9.1 under most circumstances now after revision 507371.
This includes ports
- with USE_GCC=yes or USE_GCC=any,
- with USES=fortran,
- using Mk/bsd.octave.mk which in turn features USES=fortran, and
- with USES=compiler specifying openmp, nestedfct, c11, c++0x, c++11-lang,
c++11-lib, c++14-lang, c++17-lang, or gcc-c++11-lib
plus, everything INDEX-11 shows with a dependency on lang/gcc9 now.
PR: 238330
Fix a bug when ${PREFIX} != /usr/local.[0]
As of LLVM 8.0.1, pre-releases have moved to github. Chase this change.
PR: 238603 [0]
Submitted by: mi [0]
MFH: 2019Q2
Fix the installation of llvm-lit's "cat" utility.
On llvm80, also disable discovery of libz3 to avoid an untracked
dependency when math/z3 is installed during a manual build.
PR: 237284
Reported by: arrowd
Sponsored by: DARPA, AFRL
Move the list of llvm commands to seperate Makefile.COMMANDS to be used
by another port to come later.
Reorder Makefiles a bit to group similar things together.
Remove support for generating patch files from upstream SVN revisions.
The required interfaces have been disabled for some time and LLVM is in
the process if migrating to Git.
No functional changes (pkg-plist changes are simply ordering).
Fix libclangDoc.a installation, install clang-doc80
Bug 230791 mistakenly put libclangDoc.a under %%DOCS%%. It actually
belongs to clang extras.
Plus, on 8.0.0, the clang-doc binary does work, so we can install that
too.
PORTREVISION bumped for llvm80 due to new file, not bumped for llvm70 as
this plist change only fixes previously broken OPTIONS configurations.
PR: 230791
Submitted by: Greg V
Differential Revision: https://reviews.freebsd.org/D19068
Add support for reducing the number of LLVM backends built.
There are three options:
BE_FREEBSD: Backends for FreeBSD architectures
BE_NATIVE: Backend(s) the native architecture
BE_STANDARD: All non-experimental backends (as configured by LLVM)
BE_STANDARD is the default and leaves the plist unchanged.
This feature is experimental and will be migrated to other devel/llvm ports
after we gain some experince with it.
Requested by: many
Pointer in the right direction: jhibbits
Sponsored by: DARPA, AFRL