Commit Graph

228 Commits

Author SHA1 Message Date
Andrew Turner
2422138952 Fix the KCSAN_ENABLED check when building modules
The KCSAN_ENABLED variable is non-empty when the kernel is being built
with KCSAN. This allows us to disable modules that are known to be
broken.

There was a bug where we would check if it was defined. As this is
always the case the KCSAN_ENABLED variable would be set when building
modules so we would never build such a module. Fix this by checking
if the value is empty before passing it on to the module stage.

This doesn't affect how modules are built as the CFLAGS passed to
modules has the correct check.

Reported by:	rstone
Sponsored by:	Innovate UK
2021-06-02 10:59:19 +00:00
Konstantin Belousov
aa3ea612be x86: remove gcov kernel support
Reviewed by:	jhb
Sponsored by:	The FreeBSD Foundation
Differential revision:	https://reviews.freebsd.org/D29529
2021-04-02 15:41:51 +03:00
John Baldwin
f54c6ef100 Use a template assembly file to generate the embedded MFS.
This uses the .incbin directive to pull in the MFS image contents.
Using assembly directly ensures that symbols can be defined with the
name and properties (such as .size) desired without having to rename
symbols, etc. via a second objcopy invocation.  Since it is compiled
by the C compiler driver, it also avoids the need for all of the
EMBEDFS* make variables.

Suggested by:	jrtc27
Reviewed by:	kib, markj
Obtained from:	CheriBSD
MFC after:	2 weeks
Sponsored by:	DARPA
Differential Revision:	https://reviews.freebsd.org/D26781
2020-10-20 16:48:45 +00:00
Alex Richardson
2b6ee34cf6 Pass -fuse-ld=/path/to/ld if ${LD} != "ld"
This is needed so that setting LD/XLD is not ignored when linking with $CC
instead of directly using $LD. Currently only clang accepts an absolute
path for -fuse-ld= (Clang 12+ will add a new --ld-path flag), so we now
warn when building with GCC and $LD != "ld" since that might result in the
wrong linker being used.

We have been setting XLD=/path/to/cheri/ld.lld in CheriBSD for a long time and
used a similar version of this patch to avoid linking with /usr/bin/ld.
This change is also required when building FreeBSD on an Ubuntu with Clang:
In that case we set XCC=/usr/lib/llvm-10/bin/clang and since
/usr/lib/llvm-10/bin/ does not contain a "ld" binary the build fails with
`clang: error: unable to execute command: Executable "ld" doesn't exist!`
unless we pass -fuse-ld=/usr/lib/llvm-10/bin/ld.lld.

This change passes -fuse-ld instead of copying ${XLD} to WOLRDTMP/bin/ld
since then we would have to ensure that this file does not exist while
building the bootstrap tools. The cross-linker might not be compatible with
the host linker (e.g. when building on macos: host-linker= Mach-O /usr/bin/ld,
cross-linker=LLVM ld.lld).

Reviewed By:	brooks, emaste
Differential Revision: https://reviews.freebsd.org/D26055
2020-08-25 13:30:03 +00:00
Alex Richardson
5bb9250e0a Add necessary Makefile.inc1 infrastructure for building on non-FreeBSD
The most awkward bit in this patch is the bootstrapping of m4:
We can't simply use the host version of m4 since that is not compatible
with the flags passed by lex (at least on macOS, possibly also on Linux).
Therefore we need to bootstrap m4, but lex needs m4 to build and m4 also
depends on lex (which needs m4 to generate any files). To work around this
cyclic dependency we can build a bootstrap version of m4 (with pre-generated
files) then use that to build the real m4.

This patch also changes the xz/unxz/dd tools to always use the host version
since the version in the source tree cannot easily be bootstrapped on macOS
or Linux.

Reviewed By:	brooks, imp (earlier version)
Differential Revision: https://reviews.freebsd.org/D25992
2020-08-25 13:29:57 +00:00
Eric van Gyzen
fac6dee9eb Remove tests for obsolete compilers in the build system
Assume gcc is at least 6.4, the oldest xtoolchain in the ports tree.
Assume clang is at least 6, which was in 11.2-RELEASE.  Drop conditions
for older compilers.

Reviewed by:	imp (earlier version), emaste, jhb
MFC after:	2 weeks
Sponsored by:	Dell EMC Isilon
Differential Revision:	https://reviews.freebsd.org/D24802
2020-05-12 15:22:40 +00:00
Brooks Davis
ab2b8d671b Allow the kernel to build with a compiler that sets -fno-common.
The mechanism that generates assym.inc and offset.inc depends on the
symbols in question being common. For now, simply force the object files
to be created with -fcommon.

-fno-common will be the default in GCC10/LLVM11.

Submitted by:	arichardson
Reviewed by:	kevans
Sponsored by:	DARPA
Differential Revision:	https://reviews.freebsd.org/D24322
2020-04-07 15:32:08 +00:00
Bryan Drewery
82952c4065 make all is needed to generate .depend.*
PR:		241746
X-MFC-With:	r357043
MFC after:	1 week
2020-01-31 21:08:33 +00:00
John Baldwin
1207cda961 Compile hack.c with normal CFLAGS + -shared -nostdlib.
Originally, hack.c was compiled into a shard object with just -shared
-nostdlib.  This assumed that ${CC} did not require any additional
flags for ABIs, cross-building, etc.

When kern.post.mk was created in r89509 by reducing duplication in
kernel Makefile.<arch> files, the -shared flag was moved into a
HACK_EXTRA_FLAGS variable so that sparc64 could override it with
-Wl,-shared.  The sparc64 hack was removed in r111650, but
HACK_EXTRA_FLAGS was left in place.  Over time, we have started
support toolchains that require flags to support alternate ABIs on
MIPS and PowerPC and started (ab)using HACK_EXTRA_FLAGS to set only
those flags.

I need to fix risc-v to pass -mno-relax to the hack.c build for lld in
llvm 10, and the patches to support cross-build from non-FreeBSD hosts
need to include -target for clang in CFLAGS for hack.c.  Rather than
adding more hacks into HACK_EXTRA_FLAGS, just use the full set of
CFLAGS with hack.c.

Reviewed by:	kib, arichardson
MFC after:	1 month
Sponsored by:	DARPA
Differential Revision:	https://reviews.freebsd.org/D23362
2020-01-26 14:19:08 +00:00
Yoshihiro Takahashi
10cd2843a5 Fix kernel-tags target.
- A depend-file is broken up into .depend.*.o files. [1]
  - Fix an assembly file support.

PR:		241746
Submitted by:	leres [1]
MFC after:	1 week
2020-01-23 13:56:12 +00:00
Andrew Turner
849aef496d Port the NetBSD KCSAN runtime to FreeBSD.
Update the NetBSD Kernel Concurrency Sanitizer (KCSAN) runtime to work in
the FreeBSD kernel. It is a useful tool for finding data races between
threads executing on different CPUs.

This can be enabled by enabling KCSAN in the kernel config, or by using the
GENERIC-KCSAN amd64 kernel. It works on amd64 and arm64, however the later
needs a compiler change to allow -fsanitize=thread that KCSAN uses.

Sponsored by:	DARPA, AFRL
Differential Revision:	https://reviews.freebsd.org/D22315
2019-11-21 11:22:08 +00:00
Kyle Evans
1f6453b126 Back out r351799
empty does not appear to work like I thought it did and it actively breaks
real LOCAL_MODULES usage, of which I have none at the moment...
2019-09-04 14:32:04 +00:00
Kyle Evans
a3f59fe262 LOCAL_MODULES: Allow LOCAL_MODULES="" in src.conf to work
Currently LOCAL_MODULES= works, but LOCAL_MODULES="" causes build errors as
.for still has the empty string to loop over. An .if empty prior to the loop
was considered, but LOCAL_MODULES has empty quotes at that point and thus,
isn't empty. A better solution likely exists, but this floats us by for
now...
2019-09-03 22:01:12 +00:00
John Baldwin
9b82c66474 Fix LOCAL_MODULES and improve the make output.
The exists() check guarding the invocation of ls was not working
correctly as it was expanding '$L' to determine the path of the local
modules directory.  Fix by using {} around the variable name.

Inline some of the logic from bsd.subdir.mk when invoking local module
builds.  This gives output in 'make buildkernel' the same as if there
was a Makefile in /usr/local/sys/modules with SUBDIR =
${LOCAL_MODULES}.

Reviewed by:	imp
Differential Revision:	https://reviews.freebsd.org/D20991
2019-08-06 23:22:25 +00:00
Bryan Drewery
29317e6a0e Fix build race with machine links and genoffset.o.
Generate the ilinks for all dependency objects not just the ones
in the CLEAN list.

Possibly related to r345351

Reported by:	kmoore
MFC after:	2 weeks
X-MFC-with:	r345351
Sponsored by:	Dell EMC Isilon
2019-05-10 18:09:27 +00:00
Kyle Evans
af44a26351 fdt: stop installing FDT_DTS_FILE
r346307 inadvertently started installing FDT_DTS_FILE along with the kernel.
While this isn't necessarily bad, it was not intended or discussed and it
actively breaks some current setups that don't anticipate any .dtb being
installed when it's using static fdt. This change could be reconsidered down
the line, but it needs to be done with prior discussion.

Fix it by pushing FDT_DTS_FILE build down into the raw dtb.build.mk bits.
This technically allows modules building DTS to accidentally specify an
FDT_DTS_FILE that gets built but isn't otherwise useful (since it's not
installed), but I suspect this isn't a big deal and would get caught with
any kind of testing -- and perhaps this might end up useful in some other
way, for example by some module wanting to embed fdt in some other way than
our current/normal mechanism.

Reported by:	Mori Hiroki <yamori813@yahoo.co.jp>
MFC after:	3 days
X-MFC-With:	r346307
2019-04-24 01:11:50 +00:00
Kyle Evans
924f8189b0 fdt: further consolidate DTB building and revise manpage
FDT_DTS_FILE was built separately with a rule in sys/conf/files and
recreated the rules we used in dtb.mk. Now that we have other infrastructure
to build a DTB along with the kernel, fold FDT_DTS_FILE into that since it
doesn't have any special requirements.

fdt(4) never got revised to mention the DTS/DTSO make options, so do that
now.

Reviewed by:	imp
MFC after:	1 week
Differential Revision:	https://reviews.freebsd.org/D19736
2019-04-17 03:29:16 +00:00
Mark Johnston
963ae7a63e Stop using -fdebug-prefix-map to map the object directory.
We were doing so as a workaround for the problem addressed by r345593, so
it's no longer necessary.

Reviewed by:	jhb
Discussed with:	emaste
MFC after:	2 weeks
Sponsored by:	The FreeBSD Foundation
Differential Revision:	https://reviews.freebsd.org/D19705
2019-03-27 19:34:19 +00:00
Kyle Evans
dc4636133f Allow kernel config to specify DTS/DTSO to build, and out-of-tree support
This allows for directives such as

makeoptions DTS+=/out/of/tree/myboard.dts
# in tree! Same rules applied as if this were in a dtb/ module
makeoptions DTS+=otherboard.dts

to be specified in config(5) and have these built/installed alongside th
kernel. The assumption that overlays live in an overlays/ directory is only
made for in-tree DTSO, but we still make the assumption that out-of-tree
arm64 DTS will be in vendored directories (for now).

This lowers the cost to hacking on an overlay or dts by being able to
quickly throw it in a custom config, especially if it doesn't fit one of the
current dtb/modules quite appropriately or it's not intended for commit
there.

The build/install targets were split out of dtb.mk to centralize the build
logic and leave out the all/realinstall/CLEANFILES additions... it was
believed that we didn't want to pollute the kernel build with these.

The build rules were converted to suffix rules at the suggestion of Ian to
clean things up a little bit in a world where we can have mixed
in-tree/out-of-tree DTS/DTSO specified.

Reviewed by:	ian
MFC after:	1 week
Differential Revision:	https://reviews.freebsd.org/D19351
2019-03-26 02:45:23 +00:00
Bryan Drewery
0a0369387e Build common kernel dependencies before modules.
This ensures files like genassym.o and awk/mfiles are generated before
descending into the modules build.  It may also allow some module builds
to not recreate files that are already present in the KERNBUILDDIR.

This fixes a rare build race where genassym.o is missing and assym.inc
is empty.

More work is planned around this to reduce some redundant dependency
generation in modules.

PR:		233339
MFC after:	2 weeks
Reported by:	markj
2019-03-20 22:49:41 +00:00
Mark Johnston
7114b1763c Use -fdebug-prefix-map to map auto-generated kernel build paths.
The kernel build uses symlinks to make MD #includes like <machine/pcpu.h>
work.  Debug info ends up referencing these symlinks in a relative path,
so debuggers generally don't know how to find the corresponding headers.
Address this by using -fdebug-prefix-map to map relative paths through
the symlinks to their absolute paths in the source tree.  This is
consistent with how regular source file paths are defined in the
kernel's debug info.

Also map the current directory to an absolute path to the object
directory.  This gives debuggers a chance to find auto-generated files
like vnode_if.c if the object directory is available.

Reviewed by:	emaste, jhb (previous version)
MFC after:	1 month
Sponsored by:	The FreeBSD Foundation
Differential Revision:	https://reviews.freebsd.org/D19633
2019-03-20 20:42:44 +00:00
Matt Macy
5426539c71 gcov support
add gcov support and export results as files in debugfs

Reviewed by:	hps@
MFC after:	1 week
Sponsored by:	iX Systems
Differential Revision:	https://reviews.freebsd.org/D19260
2019-02-23 21:14:00 +00:00
Andrew Turner
4ea56599e8 Port the NetBSD ubsan runtime to the FreeBSD kernel.
This allows us to build the ubsan code added in r340189 into the kernel
with the KUBSAN option. This will report when undefined behaviour is
detected in the currently running kernel.

As it can be large, the kernel is 65MB on arm64, loader may not be able to
load the kernel on all architectures so is disabled by default for now.

Sponsored by:	DARPA, AFRL
2018-11-06 17:32:07 +00:00
John Baldwin
58b6812de1 Only invoke 'ls' if the local modules directory exists.
This avoids a spurious make warning if /usr/local/sys/modules doesn't
exist.

Submitted by:	rgrimes
Reported by:	markj
2018-10-30 18:20:34 +00:00
John Baldwin
cd785c1b34 Permit local kernel modules to be built as part of a kernel build.
Add support for "local" modules.  By default, these modules are
located in LOCALBASE/sys/modules (where LOCALBASE defaults to
/usr/local).  Individual modules can be built along with a kernel by
defining LOCAL_MODULES to the list of modules.  Each is assumed to be
a subdirectory containing a valid Makefile.  If LOCAL_MODULES is not
specified, all of the modules present in LOCALBASE/sys/modules are
built and installed along with the kernel.

This means that a port that installs a kernel module can choose to
install its source along with a suitable Makefile to
/usr/local/sys/modules/<foo>.  Future kernel builds will then include
that kernel module using the kernel configuration's opt_*.h headers
and install it into /boot/kernel along with other kernel-specific
modules.

This is not trying to solve the issue of folks running GENERIC release
kernels, but is instead aimed at folks who build their own kernels.
For those folks this ensures that kernel modules from ports will
always be using the right KBI, etc.  This includes folks running any
KBI-breaking kernel configs (such as PAE).

There are still some kinks to be worked out with cross-building (we
probably shouldn't include local modules in cross-built kernels by
default), but this is a sufficient starting point.

Reviewed by:	imp
MFC after:	3 months
Relnotes:	yes
Differential Revision:	https://reviews.freebsd.org/D16966
2018-10-30 00:23:37 +00:00
Ed Maste
08d0704d74 Switch reproducible builds to unmodified src tree mode
newvers.sh supports two modes for reproducible builds:

 -r    Reproducible build.  Do not embed directory names, user
       names, time stamps or other dynamic information into
       the output file.  This is intended to allow two builds
       done at different times and even by different people on
       different hosts to produce identical output.

 -R    Reproducible build if the tree represents an unmodified
       checkout from a version control system.  Metadata is
       included if the tree is modified.

Switch to the second mode when reproducible builds are enabled.
The value of a reproducible build is much less when building from an
uncontrolled, modified src tree, and -R likely provides the best
compromise in allowing the REPRODUCIBLE_BUILD knob to be enabled by
default for the release.

Approved by:	re (kib)
Sponsored by:	The FreeBSD Foundation
2018-09-11 19:19:07 +00:00
Bryan Drewery
817f7baa74 Use OBJS_DEPEND_GUESS for forced opt_global.h dependency.
MFC after:	2 weeks
Sponsored by:	Dell EMC
2018-07-04 00:18:36 +00:00
Matt Macy
9f7b089e95 Enable MODULE_TIED by default for modules compiled with the kernel
Requested by: jhb
2018-07-03 23:05:42 +00:00
Bryan Drewery
cd1acc77e6 Fix .depend.foo.o tracking for sys/conf/files defined compilations.
Some example files:
  ia32_genassym.o
  acpi_wakecode.o

The old mkdep method also lacked tracking these files.

Objects defined in sys/conf/files with no-obj and no-implicit-rule get their
own targets defined in the kernel Makefile but lack having their objects added
to DEPENDOBJS so never get a .depend file generated.  Normally if an object is
in OBJS it will get a .depend file.

Fix this by looking for .o files in CLEAN and ensuring they are part of
the -MD filtering and .depend loading.  This is a hack.  Other solutions
could exist involving sys/conf/files or config(8) to auto add these to
DEPENDFILES/DEPENDOBJS but this method seems reliable enough without being
intrusive or error-prone for new files.

Reported by:	bde
MFC after:	3 weeks
Sponsored by:	Dell EMC
2018-07-03 21:29:04 +00:00
Bryan Drewery
44e4164832 Fix r335906+r335907 to not pass multiple source files to genoffset_test.o target.
This is working around a bug with the FAST_DEPEND logic with custom targets
that use .ALLSRC.

Reported by:	mmacy
2018-07-03 20:03:19 +00:00
Bryan Drewery
a46134b641 Handle .depend for genoffset_test.o 2018-07-03 19:24:52 +00:00
Bryan Drewery
4cee7eae0d No need to make a copy to build genoffset_test.o 2018-07-03 19:24:44 +00:00
Matt Macy
018617b8ba explain the purpose of genoffset_test 2018-07-03 18:39:43 +00:00
Matt Macy
87c259d436 ensure that offset.inc is generated first 2018-07-03 18:20:42 +00:00
Matt Macy
6443773dab make critical_{enter, exit} inline
Avoid pulling in all of the <sys/proc.h> dependencies by
automatically generating a stripped down thread_lite exporting
only the fields of interest. The field declarations are type checked
against the original and the offsets of the generated result is
automatically checked.

kib has expressed disagreement and would have preferred to simply
use genassym style offsets (which loses type check enforcement).
jhb has expressed dislike of it due to header pollution and a
duplicate structure. He would have preferred to just have defined
thread in _thread.h. Nonetheless, he admits that this is the only
viable solution at the moment.

The impetus for this came from mjg's D15331:
"Inline critical_enter/exit for amd64"

Reviewed by: jeff
Differential Revision: https://reviews.freebsd.org/D16078
2018-07-03 01:55:09 +00:00
Jason A. Harmening
28da5c6d7a Remove MK_AUTO_OBJ from env passed to PORTS_MODULES
This fixes a failure to resolve object file paths seen when buildkernel
(which sets MK_AUTO_OBJ=yes) and installkernel (which sets MK_AUTO_OBJ=no)
are run as separate steps.  r329232 partially fixed this scenario by removing
MAKEOBJDIR, but it seems the AUTO_OBJ setting also needs to be on the same
page for the build and install steps.

Reviewed by:	bdrewery
MFC after:	2 weeks
Differential Revision:	https://reviews.freebsd.org/D14143
2018-03-31 05:17:12 +00:00
Ed Maste
fc2a8776a2 Rename assym.s to assym.inc
assym is only to be included by other .s files, and should never
actually be assembled by itself.

Reviewed by:	imp, bdrewery (earlier)
Sponsored by:	The FreeBSD Foundation
Differential Revision:	https://reviews.freebsd.org/D14180
2018-03-20 17:58:51 +00:00
Bryan Drewery
71b126549d Fix rebase mismerge in r330724.
X-MFC-With:	r330724
MFC after:	2 weeks
Sponsored by:	Dell EMC
2018-03-10 02:13:48 +00:00
Bryan Drewery
906ce865c2 Don't skip reading depend for 'make obj' unless it is alone.
This was effectively done in bsd.dep.mk quite some time ago.

MFC after:	2 weeks
Sponsored by:	Dell EMC
2018-03-10 02:10:26 +00:00
Bryan Drewery
003a0576c9 Skip reading depend files with -V unless looking up a depend variable.
This speeds up some simple -V lookups significantly.

Reported by:	bde
MFC after:	2 weeks
Sponsored by:	Dell EMC
2018-03-10 02:10:19 +00:00
Bryan Drewery
501608626d ports modules: Don't leak AUTO_OBJ changes into the port builds.
This came about when r328489 made ports modules builds no longer use the
in-tree share/mk files, but didn't cleanup MAKEOBJDIR from the
environment.

This fixes "Variable OBJTOP is recursive".

Sponsored by:	Dell EMC
2018-02-13 17:51:16 +00:00
Jason A. Harmening
283f8a5c33 Remove system makefile path directives from env passed to PORTS_MODULES step
Previously, MAKESYSPATH as well as '-m' directives in MAKEFLAGS would cause
any port rebuilt during the PORTS_MODULES stage to consume system makefiles
from $(SRCROOT)/share/mk instead of those installed under /usr/share/mk.
For kernel modules that need to build against an updated src tree this
makes sense; less so for <bsd.port.mk> or  any userspace library or utility
the port may also happen to install.

Before 11.0, this probably didn't matter much in practice.  But the addition
of src.libnames.mk under $(SRCROOT)/share/mk in 11.0 breaks any consumer of
bsd.prog.mk and DPADD/LDADD during PORTS_MODULES.

Address the build breakage by removing MAKESYSPATH and any occurrence of
'-m' from MAKEFLAGS in the environment created for the port build.
Instead set SYSDIR so that any kmod built by the port will still consume
conf/kmod.mk from the updated src tree, assuming it uses <bsd.kmod.mk>

Reviewed by:	bdrewery
MFC after:	2 weeks
Differential Revision: 	https://reviews.freebsd.org/D13053
2018-01-27 20:13:36 +00:00
Alex Richardson
454ce38321 Use ln -n instead of -h to allow building the kernel on Linux
Both flags do the same thing but -n is more widely supported.

Reviewed By:	jhb, emaste
Approved By:	jhb (mentor)
Differential Revision: https://reviews.freebsd.org/D13936
2018-01-16 21:43:57 +00:00
Bryan Drewery
7cc780d771 DEPENDSRCS not used here.
Sponsored by:	Dell EMC
2017-12-05 02:23:27 +00:00
Bryan Drewery
b4585565d5 Rename DEPENDFILES_OBJS to DEPENDFILES.
This is to be consistent with bsd.dep.mk using DEPENDFILES after
r325677.

Sponsored by:	Dell EMC
2017-12-05 02:23:23 +00:00
Bryan Drewery
b00ea69b51 Fix PORTS_MODULES+'make reinstallkernel' trying to run bogus 'make redeinstall'.
Also fix 'make installkernel' running 'make deinstall' twice.

PR:		201779
MFC after:	2 weeks
Sponsored by:	Dell
2017-11-18 20:01:05 +00:00
Warner Losh
1cbb58886a Remove build system support for lint.
Differential Revision: https://reviews.freebsd.org/D13124
2017-11-17 18:16:46 +00:00
Bryan Drewery
96dd05dd7d Quote ${MAKE} when passing in env in case it contains spaces.
Downstream we are wrapping MAKE with a limits(1) call which
interferes with these non-quoted cases.

Sponsored by:	Dell EMC Isilon
2017-08-16 17:54:24 +00:00
Bryan Drewery
ecc39238d3 Allow -DNO_SKIP_DEPEND to override the _SKIP_DEPEND logic.
MFC after:	2 weeks
Sponsored by:	Dell EMC Isilon
2017-07-25 16:46:12 +00:00
Bryan Drewery
c2fd1d01e8 Allow disabling dependency tracking if DEPEND_CFLAGS is empty.
This falls back on using the guesssed dependencies if so.
Also remove a pre-bmake check while here.

MFC after:	2 weeks
Sponsored by:	Dell EMC Isilon
2017-07-24 23:32:40 +00:00