This breaks the riscv64 build; clearly the version we have in-tree is
not new enough to work. That or our build system needs tweaking. Either
way, we're not ready for this, and enabling it without testing was
misguided.
This reverts commit c59a47dc6c.
Previously it was enabled for all architectures except but RISC-V.
Upstream made significant progress on RISC-V support over the last
year[1], and although there's still some FreeBSD-specific work to do we
can start building it now.
[1] https://jonasdevlieghere.com/post/lldb-2025/
Sponsored by: The FreeBSD Foundation
Build fails with BHYVE_SNAPSHOT enabled on non-amd64,
so add it to BROKEN_OPTIONS for these arches.
PR: 292686
Reviewed by: emaste, markj
MFC after: 3 days
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D54873
NO_FP_LIBC was added in 2004 to save space by disabling FP support in
*printf()/*scanf(). The size benefit is negligible on modern systems
and conflicts with assumptions made by current base utilities.
Remove the option and always build libc with floating-point support.
Reported by: Oskar Holmlund <eovholmlund at gmail com>
MFC after: 2 weeks
Similar to the build epoch check, cache a list of source options in the
object tree, and force a clean build if the cached list does not match
the current list, after filtering out options which are known not to
affect the build (e.g. CLEAN, TESTS, WARNS).
This also adds a DEPEND_CLEANUP option (which defaults to yes unless
the CLEAN option is set) which can be used to skip depend-cleanup for
faster incremental builds.
MFC after: 1 week
Reviewed by: imp
Differential Revision: https://reviews.freebsd.org/D52011
In commit faeaa25f56 MK_REPRODUCIBLE_PATHS was introduced, but some
ports that include bsd.debug.mk then started to fail with:
make: /usr/share/mk/bsd.debug.mk:19: Variable "MK_REPRODUCIBLE_PATHS" is undefined
in /usr/share/mk/bsd.lib.mk:525
in /wrkdirs/usr/ports/devel/libsysinfo/work/libsysinfo-0.0.3/Makefile:16
make: Fatal errors encountered -- cannot continue
This is because bsd.debug.mk includes bsd.opts.mk, not src.opts.mk. Move
the MK_REPRODUCIBLE_PATHS option to bsd.opts.mk to fix this.
Reviewed by: imp, emaste
Fixes: faeaa25f56
There are many environments that you do not want the paths to be
canonicalized. For example, if I always build a product from a specific
path in a jail that's dependent on the product version, I don't want
that to be sanitized to /usr/src because when I have profiling tools run
against multiple kernels, I can't do A/B testing and results processing
if everything points to /usr/src (sadly, the profiling tools cannot
remap back to the original like gdb can). Since we need both active at
the same time, symbloic link fallbacks don't work.
We do want the rest of the build to be reproducible, however. We'll get
the same binaries if we rebuild later since we always rebuild with the
same path.
Turn them both on for maximum reproduction for environments where that
cannot be guaranteed.
Sponsored by: Netflix
Input from: emaste, ivy
Differential Revision: https://reviews.freebsd.org/D52959
In commit 2e47f35be5 libllvm, libclang and liblldb were converted into
private shared libraries. This allowed clang, lld, lldb, and other llvm
tools to be linked against these shared libraries, which makes them
smaller and avoids duplication.
However, this also comes at the cost of some performance, since the
dynamic libraries are quite large, and contain lots of long symbols
(mangled C++ identifiers).
Add a WITH_LLVM_LINK_STATIC_LIBRARIES build knob that can be used to go
back to the previous behavior: libllvm, libclang and liblldb are built
as internal static libraries, i.e. only available during buildworld, and
fully linked into the various executables such as clang, lld, etc.
PR: 287447
Reviewed by: emaste
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D50956
At the moment ipfs(8) is a tool that can be easily abused. Though the
concept is sound the implementation needs some work.
ipfs(8) should be considered experimental at the moment.
This commit also makes ipfs support in the kernel optional.
Reviewed by: emaste, glebius
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D53787
It was previously disabled when we built for older 32-bit arm targets
which lack 64-bit atomics. As we now support armv7 at a minimum there
is no need to disable LLDB.
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D52668
Follow up upstream rename from blacklist to blocklist.
- Old names and rc scripts are still valid, but emitting an ugly warning
- Old firewall rules and anchor names should work, but emitting an ugly
warning
- Old MK_BLACKLIST* knobs are wired to the new ones
Although care has been taken not to break current configurations, this
is a large patch containing mostly duplicated code. If issues arise, it
will be swiftly reverted.
Reviewed by: ivy (pkgbase)
Approved by: emaste (mentor)
MFC after: 2 days
Relnotes: yes
It was added to the list in bsd.opts.mk in commit 4f81c42fbd, so
should have been removed here.
Reported by: cperciva
Fixes: 4f81c42fbd ("share/mk: Substitute reproducible prefixes in dwarf info")
MFC after: 3 days
Sponsored by: The FreeBSD Foundation
Sponsored by: Klara, Inc.
For MIT Kerberos, MK_GSSAPI has no meaning: GSSAPI is a required part of
Kerberos and is always built if MK_KERBEROS is enabled. Backport this
behaviour to Heimdal so it works the same way.
While here, change Heimdal's libcom_err and compile_et to be selected by
MK_KERBEROS, not MK_KERBEROS_SUPPORT, since these are part of Kerberos
and third-party users might need it even if Kerberos support is disabled
in the base system. This means MK_KERBEROS_SUPPORT installs the same
files with both MIT and Heimdal.
Reviewed by: cy
Differential Revision: https://reviews.freebsd.org/D51859
Now that we have the notion of a build epoch, we have sufficient tooling
to handle changes that might need a larger hammer to unbreak the build.
Future changes may extend our cleanup to avoid scenarios like with the
MITKRB5 knob where flipping it will still require a cleanup, but on the
whole most users aren't regularly flipping breaking knobs enough for
this to cause the same level of problems. Those users are advised to do
a CLEAN build when they flip it until we start recording build options
and doing another clean for important knob switches (see D52011).
When knobs of that sort are newly introduced or switch defaults, we can
explicitly record an epoch for them in the interim until
option-switching is detected and handled properly.
This reverts commit 6dab48b9de.
Reviewed by: jrtc27, markj
Differential Revision: https://reviews.freebsd.org/D51923
There have been too many issues with non-META_MODE incremental builds
recently, and it is clear that most users, even developers, were not
aware that the default had been switched.
This will be revisited once more work has been done to help prevent
future breakage.
This reverts commit ba373fca78.
Reviewed by: markj, jhb
Differential Revision: https://reviews.freebsd.org/D51828
Enabling and disabling LLVM assertions changes the ABI. There have
been multiple reports on the freebsd-current@ mailing list of broken
'cc'.
Out of caution, you may wish to do a clean build while crossing this
commit after building the reverted commit if you do not set
LLVM_ASSERTIONS in your src.conf.
Discussed with: kevans, jrtc27
This reverts commit 9b3055d0d4.
This noticeably slows down the compiler execution and with the move to
pkgbase is one of the remaining differences that cannot be
administratively controlled; pkg kernels and malloc.conf(5) will allow
quickly testing a snapshot in a configuration closer to stable and
release branches.
dim@ notes that you need to set WITH_LLVM_ASSERTIONS in src.conf(5) if
you want to report an issue with LLVM.
Enabling WITH_LLVM_ASSERTIONS is also a consideration for CI and pkg
builders especially tracking main.
Approved by: dim, kevans
Differential Revision: https://reviews.freebsd.org/D50388
This is the first in a series of commits to replace Heimdal with
MIT KRB5. This first commit adds the WITH_MITKRB5/MK_MITKRB5 knob
to src.opts.mk and its corresponding documentation in
tools/build/options. The default is off.
This change of and by itself is of no consequence as MIT KRB5 has
yet to be imported in to HEAD. But it does insulate the build from
inremental change until the last patch is committed when WITH_MITKRB5
will build MIT KRB5 1.21.3 instead of Heimdal.
The affords us the opportunity to review smaller commits.
This is the first of many commits.
Reviewed by: imp, jhb, brooks, markj
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D50684
Starting in 2014 FreeBSD migrated from GNU binutils to ELF Tool Chain
tools. At that time there were no usable LLVM versions of those tools,
but they have been developing rapidly since then. Migrate to LLVML's
tools for both functionality and maintainability reasons.
This will eventually support the use of link-time optimization (LTO) in
the FreeBSD base system. LTO runs optimization passes over the entire
executable (or library) at link time and thus allows for more effective
optimization than when performed on individual compilation units.
When using LTO object files (.o) including those contained in static
library archives (.a) contain LLVM IR bitcode rather than target
object code. This means that utilities that operate on object files
need to support LLVM IR.
As with ELF Tool Chain the LLVM tools aim for command line and output
format compatibility with GNU binutils, although there are a few minor
differences. Where these cause a material issue (breaking a port or
eliminating required functionality) we can submit LLVM bugs and work
on patches.
PR: 258872 (exp-run)
Reviewed by: bapt
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D49361
In 2020, an improved implementation of the bc and dc commands
developed by Gavin D. Howard has been imported into FreeBSD.
It has replaced the OpenBSD-derived versions of these commands
in all currently supported FreeBSD releases.
The OpenBSD versions could still be built using the WITHOUT_GH_BC
option. There have been no reports of problems or unexpected
deviations from the OpenBSD version for some time, therefore
keeping the OpenBSD version is no longer required in FreeBSD.
This commit removes the option to build the OpenBSD version and
corresponding source files from -CURRENT. No MFC is planned, all
currently released FreeBSD versions should retain the build option.
The WITHOUT_GH_BC option is no longer accepted and will cause
make buildworld to fail.
Reviewed by: des, emaste
Approved by: des
Relnotes: yes
Differential Revision: https://reviews.freebsd.org/D46876
After talking with a number of people about the removal of some things
to make the loader fit, readjust things a little.
Add back GZIP and BZIP2 compression support. Many of the downstream MFC
packaging systems depend on this. This adds back 20k to the size of the
loader.
Make the boot loader text-only by default. This saves 40k in size. Net,
we're 20k smaller. The graphics loader for BIOS is less useful than the
zip functionality: You can still boot w/a text only one it and you can
build a custom one if you really want it. It's also the default we use
for dual console.
This should be merged back into stable/14 and stable/13 so it's in the
next release for each of these. That way we have only one release (13.4)
with the other defaults.
MFC After: 3 days
Sponsored by: Netflix
Reviewed by: olce, rgrimes, emaste
Differential Revision: https://reviews.freebsd.org/D47203
and allow to revert it back to 48 with WITHOUT_JEMALLOC_LG_VADDR_WIDE build
option.
Reviewed by: andrew, emaste
Sponsored by: Advanced Micro Devices (AMD)
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D46686
In preparation for supporting 64-bit machines with 32-bit UEFI firmware,
add a build option for compiling the ia32 loader. Currently unused.
Reviewed by: imp
Pull Request: https://github.com/freebsd/freebsd-src/pull/1098
The in-tree ZFS test suite is somewhat outdated and I see a number of
failures there. I tend to think that we want to integrate the OpenZFS
test suite somehow, replacing the legacy one, though it's also possible
to run that as a separate test suite.
In any case, if one wants to run the OpenZFS test suite separately, it's
useful to be able to disable installation of the legacy ZFS test suite,
so let's provide a src.conf option to do that.
Reviewed by: asomers
MFC after: 2 weeks
Differential Revision: https://reviews.freebsd.org/D46476
Make it possible to disable pxeboot. This loader will fail to build when
it's too large. When /boot/loader needs to be larger like that, this
options will disable a component whose build will fail. It is an explicit
option rather than implicit when things are too large to force the user to
make the explicit tradeoffs rather than wonder why they have a stale pxeboot
or other odd failure mode.
MFC After: 3 days
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D46212
As discussed on the freebsd-arch mailing list[1]. For historical
reasons FreeBSD's buildworld and buildkernel targets started by cleaning
the object tree, for traditional (non-metamode) builds.
Cleaning is not necessary when dependencies are properly tracked, and we
have a somewhat kludgey script[2] to handle some known cases where deps
were mishandled by traditional builds. Be consistent with the vast
majority of open source build systems by default, and do not clean at
the beginning of buildworld or buildkernel.
Users may set WITH_CLEAN in src.conf(5) to restore the previous
behaviour, or run `make cleanworld` and/or `make cleankernel` before
starting a build.
[1] https://lists.freebsd.org/archives/freebsd-arch/2024-July/000727.html
[2] tools/build/depend-cleanup.sh
Reviewed by: jhb, imp
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D46172
gcc doesn't have -m32 support on aarch64 so mark LIB32 broken there.
We have to check both COMPILER_TYPE and X_COMPILER_TYPE becuase
X_COMPILER_TYPE is only conditionally set and COMPILER_TYPE is the host
compiler in Makefile.inc1.
Reviewed by: emaste
Differential Revision: https://reviews.freebsd.org/D45089
This commit broke "make makeman" checks in github CI due to a lack of
option description files. The split between VIMAGE and VIMAGE_SUPPORT
is not clearly justified and the code is broken because there is no
opt_vimage.h (it's in opt_global.h).
This reverts commit 22ca6db50f.
The drivers and utilities are now built and installed unconditionally.
Reviewed by: imp, emaste
Sponsored by: Chelsio Communications
Differential Revision: https://reviews.freebsd.org/D44843
Capsicum is non-optional as of c24c117b96 ("Remove
WITHOUT_{CAPSICUM,CASPER} options").
`#ifndef WITHOUT_CAPSICUM` is left in the source for the benefit of
downstream consumers, but is never defined in FreeBSD.
Reviewed by: oshogbo
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D42077
Default to VIMAGE as yes.
Add VIMAGE to __DEFAULT_DEPENDENT_OPTIONS (to define VIMAGE_SUPPORT)
Only output VIMAGE to opt_global.h when VIMAGE support is wanted.
Obtained from: Juniper Networks, Inc.
Differential Revision: https://reviews.freebsd.org/D39636
this is a very early script to support cloudinit, it does not intend to
be a full featured cloudinit client, but will support a good enough
subset to be viable in most case.
It support nocloud and openstack config-2 config drive mode (iso9660 or
msdosfs)
The following features are currently supported:
- adding users (including a default user named 'freebsd' with password
'freebsd'
- adding groups
- adding ssh keys
- static ipv4, static ipv6, dynamic ipv4
With this one is able to use the 'bring your own image feature" out of
box.
It is expected that the script grows the support of other clouds
supporting cloud-init, contributions are welcomed.
It is designed to be only run once via the firstboot mecanism.
Sponsored by: OVHCloud
MFC After: 3 weeks
Differential Revision: https://reviews.freebsd.org/D44141
When BEARSSL is enabled, we pull in libsecureboot, which has EFI
dependencies which don't exist on powerpc. This needs to be detangled,
but until then mark it as broken.
Sponsored by: Netflix
This option will omit all the graphics support, the teken terminal
library, video mode support, etc and support a simple, basic, text-only
video console for the x86 BIOS boot loader. It uses the FreeBSD 12
version of vidconsole.c. It defaults to NO.
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D43912
Commit de6feefdb7 limited the amount of debuginfo generated for clang
and other llvm-related executables. This was done to save disk space and
memory during building, but it makes debugging any of these executables
much harder.
Add a new src.conf(5) setting, WITH_LLVM_FULL_DEBUGINFO, to generate
full debuginfo instead. This is off by default, but could for example be
enabled for release builds or snapshots, so llvm executables are easier
to debug.
Reviewed by: emaste
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D43839
The option was added to parallel the CLANG_IS_CC which was removed in
commit 20a66ab4bf.
Reviewed by: imp, dim, emaste
Differential Revision: https://reviews.freebsd.org/D42575
Allow building 32bit libs for host.
Move CFLAGS additions from local.sys.dirdeps.mk (which is too early
and impacts CFLAGS defaults) to local.sys.mk
Reviewed by: stevek
Sponsored by: Juniper Networks, Inc.
Differential Revision: https://reviews.freebsd.org/D41946
This will enable alternative mallocs to be included in the tree and
selected by setting LIBC_MALLOC. As there is only one today (jemalloc)
this option does nothing, but we expect to add other implementations
in the future. This will also reduce diffs to CheriBSD.
Reviewed by: emaste
Differential Revision: https://reviews.freebsd.org/D41660
Baptiste Daroussin