Commit Graph

2857 Commits

Author SHA1 Message Date
uwe b453cdfc13 Switch powerpc to GCC 8 again now that alloca() is fixed.
PR port-macppc/54827
2020-01-22 21:46:06 +00:00
uwe 10180bec9c Switch powerpc back to GCC 7 until we fix alloca() PR port-macppc/54827
ok mrg@
2020-01-22 08:32:33 +00:00
jmcneill 8f8fe2de40 Install nouveau and radeon firmware on aarch64 2020-01-19 00:57:18 +00:00
christos f58ad7d37c Back to libarchive's bsdtar 2020-01-17 16:26:12 +00:00
christos 53517b7626 Back to pax as tar until I figure out how I broke libarchive (hard link
extraction creates 0 length files for existing hard-links).
2020-01-14 16:48:18 +00:00
christos 852bfb5491 re-enable BSDTAR now that the two issues (symlinks/atomicity) have been fixed. 2020-01-12 16:11:51 +00:00
christos 0fb78fffb4 Back out previous. This has no chance to work unless the make variable parsing
is not changed do that instead of scanning for a single character delim ':',
it scans for "?:". This is because !empty(COMPILE.c:M*-pg*) contains a ':'.
2020-01-08 22:04:24 +00:00
christos 42506d4e49 tweak the DBG rule that -fomit-frame-pointer for aarch64 when profiling
(because the compiler complains), to use a match with the compile flags
and *pg*, instead of using a match to a target suffix (which is NetBSD
build-specific). Pointed out by phone@.
2020-01-08 04:15:45 +00:00
christos e33a0ba00d move MV to sys.mk because it is used there. Pointed out by joerg@ 2019-12-20 04:04:25 +00:00
mrg 74b9b609f1 switch powerpc to GCC 8. 2019-12-14 10:59:25 +00:00
jmcneill 2aeb9a97d8 Enable DTrace on aarch64 2019-12-04 11:47:52 +00:00
jmcneill 84d4c1fb40 Enable ZFS support on aarch64 2019-12-01 20:28:25 +00:00
jmcneill 2e3c4047ee Build aarch64 modules without fp or simd instructions. 2019-12-01 20:24:47 +00:00
kamil 018b416e9d Disable KCOV instrumentation in x86_machdep.c
This allows to use cpu_intr_p() directly inside KCOV.
2019-12-01 17:25:47 +00:00
martin 9d445e0761 Document NETBSD_OFFICIAL_RELEASE and add it to /etc/release - it is
important information when trying to reproduce a build.
2019-11-22 13:30:19 +00:00
mrg 540580a78d switch riscv32/64 to GCC 8. 2019-11-21 07:56:58 +00:00
christos 249752ac15 Compare libraries that the bsd.prog.mk "knows about" vs the ones we actually
install... Yes, pretty bad right now.
2019-11-18 16:23:28 +00:00
christos 4d3961b8b8 Improve the script to check for installed libraries, and fix the wrong
names and obsolete libraries it found.
2019-11-18 16:09:21 +00:00
christos e2cf0109bf switch arm to gcc-8 2019-11-18 15:38:07 +00:00
mrg bc98eb4506 switch powerpc64 to GCC 8. 2019-11-17 02:59:25 +00:00
mrg 1cf39b370c switch ia64 to GCC 8. boots just as far as GCC 7 in ski. 2019-11-17 02:52:18 +00:00
mrg a7e410a84d sparc & sparc64 -> GCC 8. 2019-11-16 10:23:36 +00:00
mrg 0d35e2c2c3 i386 -> GCC 8. 2019-11-16 10:22:20 +00:00
maxv 65ee926c5a Make kMSan compatible with KCOV. With kMSan we are forced to stay with the
fsanitize flag on subr_kcov.c, which means that kMSan will instrument KCOV.
We add a bunch of __nomsan attributes to reduce this instrumentation, but
it does not remove it completely. That's fine.
2019-11-15 09:44:44 +00:00
joerg 3cdbfa93d9 Restore pax-as-tar. 2019-11-13 22:32:16 +00:00
maxv b7edd3d132 Add Kernel Concurrency Sanitizer (kCSan) support. This sanitizer allows us
to detect race conditions at runtime. It is a variation of TSan that is
easy to implement and more suited to kernel internals, albeit theoretically
less precise than TSan's happens-before.

We do basically two things:

 - On every KCSAN_NACCESSES (=2000) memory accesses, we create a cell
   describing the access, and delay the calling CPU (10ms).

 - On all memory accesses, we verify if the memory we're reading/writing
   is referenced in a cell already.

The combination of the two means that, if for example cpu0 does a read that
is selected and cpu1 does a write at the same address, kCSan will fire,
because cpu1's write collides with cpu0's read cell.

The coverage of the instrumentation is the same as that of kASan. Also, the
code is organized in a way similar to kASan, so it is easy to add support
for more architectures than amd64. kCSan is compatible with KCOV.

Reviewed by Kamil.
2019-11-05 20:19:17 +00:00
martin 66ef6a30f0 PR misc/54657: let RELEASEMACHINEDIR default to ${MACHINE}-${MACHINE_ARCH}
for evb{arm,mips,sh3}*.
2019-10-31 06:43:19 +00:00
maxv a8c6c0bf54 Add nvmmctl, with two commands for now. 2019-10-28 13:04:18 +00:00
christos 20f77b55bf 1. All Makefiles that use ARM_APCS_FLAGS add -marm unconditionally and
bsd.own.mk also adds it for gcc only (so we end up with 2 -marm for gcc).
   Just add -marm unconditionally in bsd.own.mk in ARM_APCS_FLAGS and remove
   it from everywhere else.
2. gcc-8 now warns about thumb internetwork in APCS mode. This does not make
   sense because the documentation says that this flag does not make a
   difference in apcs mode, but might produce larger code in non-APCS mode.
   So we just add it to silence the warnings.
2019-10-24 18:46:20 +00:00
christos 78be1c9f56 move a{md,arch}64 to gcc 8 2019-10-22 13:49:30 +00:00
christos 0c7a7c13c0 Allow individual lex and yacc sources to override/extend LFLAGS and YFLAGS
respectively.
2019-10-15 15:05:00 +00:00
mrg 69f74cd544 make GCC_NO_STRINGOP_OVERFLOW apply to gcc 7 too. 2019-10-13 10:12:13 +00:00
mrg de11d87641 introduce some common variables for use in GCC warning disables:
GCC_NO_FORMAT_TRUNCATION    -Wno-format-truncation (GCC 7/8)
GCC_NO_STRINGOP_TRUNCATION  -Wno-stringop-truncation (GCC 8)
GCC_NO_STRINGOP_OVERFLOW    -Wno-stringop-overflow (GCC 8)
GCC_NO_CAST_FUNCTION_TYPE   -Wno-cast-function-type (GCC 8)

use these to turn off warnings for most GCC-8 complaints.  many
of these are false positives, most of the real bugs are already
commited, or are yet to come.


we plan to introduce versions of (some?) of these that use the
"-Wno-error=" form, which still displays the warnings but does
not make it an error, and all of the above will be re-considered
as either being "fix me" (warning still displayed) or "warning
is wrong."
2019-10-13 07:28:04 +00:00
jhigh 0090dad153 added backend support for argon2 password scheme 2019-10-09 13:13:08 +00:00
uwe 0892295b90 Don't add -Wabi to CXXFLAGS for gcc 8. It warns about differences
from the most up-to-date ABI, which in gcc 8 is used by default.
XXX: This needs a better condition.
2019-10-05 18:37:39 +00:00
mrg 4be7da17b5 gcc 7 can now be found in gcc.old. 2019-09-29 22:10:55 +00:00
bad c296e4670d Also don't add a sysroot at all if a rumpkernel build. 2019-09-15 21:17:08 +00:00
maya 4bf438acf0 Allow overriding the default man substitutions with X11EXTRAMANTRANSFORMS 2019-09-13 10:23:07 +00:00
kamil 1a5f018b01 Enhance the support of LLVM sanitizers
Define _REENTRANT for MKSANITIZER build. This is needed for at least stdio
code. This caused new build issued with duplicated symbols in few places
and rump kernel code picking different code paths borrowed from libc.
Handle all this in one go.

Add bsd.sanitizer.mk to share common code used by programs and libraries.

Switch from realall to beforeinstall target in .syms files. This is more
reliable in MKSANITIZER.
2019-08-27 22:48:53 +00:00
mrg 1db21b1df3 fix MKCXX=no:
- don't build GDB
- don't build sanitizers
- don't build GCC commands
- fix some set lists
2019-08-23 08:17:26 +00:00
kamil 92253c7761 Add PAXCTL_FLAG rules for MKSANITIZER
Add per-program rules to disable ASLR for ASan, TSan and MSan in all
sanitized programs. This flag is not needed for other supported sanitizers.

Without this change, sanitized init(8) dies early on startup.

Approach originally suggested by <joerg>
2019-08-17 00:14:48 +00:00
kamil 0e6f677375 Add SANITIZER_RENAME_CLASSES in bsd.prog.mk
This adds a more flexible version of SANITIZER_RENAME_SYMBOL.

New symbol allows to specify MKSANITIZER specific renames on per-file and
per-symbol basis.
2019-08-15 08:42:59 +00:00
christos a3a94da6c9 Add a script to compare the library lists so we can keep them in sync easier. 2019-08-10 12:50:36 +00:00
christos 9b5a96cd27 Make the X11 library lists consistent again. 2019-08-10 12:46:38 +00:00
christos 806cddf3f2 Don't install PIC libraries by default because they are too big since they
contain debug symbols. I supplied a patch in PR/54449 to remove the debugging
symbols but folks preferred to not install them at all.
2019-08-10 12:20:17 +00:00
roy cb24add28c MKBSDTAR now defaults to yes
Starting to see some modern archives not opening, it's time to make them
work!

OK joerg@ agc@
2019-07-24 09:28:29 +00:00
mrg 9d9d771525 add util-macros src dir. 2019-07-20 20:25:35 +00:00
wiz c4e128da50 Install the nouveau gm206 firmware as used by my graphics card.
Also add the other gm20x nouveau firmwares to the source tree
to make it easier to add them for someone who can test them.

Installed if MKNOUVEAUFIRMWARE is set to 'yes'.
This defaults to no except on amd64 and i386 (like for radeon).
2019-07-20 19:09:21 +00:00
mrg f791ec1b3a break the build if a .pc generation leaves something matching "@.*@".
this is an error in the build that doesn't trigger issues until you
try to integrate with eg, pkgsrc.
2019-07-10 21:53:35 +00:00
hannken ce2428b9c3 When building a module keep symbols if built with "MKCTF" and "COPTS=-g". 2019-07-05 08:28:16 +00:00