bsd.own.mk: rename to CC_WNO_ADDRESS_OF_PACKED_MEMBER

Provide a single variable
CC_WNO_ADDRESS_OF_PACKED_MEMBER
with options for both clang and gcc, to replace
CLANG_NO_ADDR_OF_PACKED_MEMBER
CC_NO_ADDR_OF_PACKED_MEMBER
GCC_NO_ADDR_OF_PACKED_MEMBER

Using the convention CC_compilerflag, where compilerflag
is based on the full compiler flag name.

add support for new GCC 9 warnings that may be too much to fix
right now. new address-of-packed-member and format-overflow
warnings have new GCC_NO_ADDR_OF_PACKED_MEMBER amd
GCC_NO_FORMAT_OVERFLOW variables to remove these warnings.

apply to a bunch of the tree. mostly, these are real bugs that
should be fixed, but in many cases, only by removing the 'packed'
attribute from some structure that doesn't really need it. (i
looked at many different ones, and while perhaps 60-80% were
already properly aligned, it wasn't clear to me that the uses
were always coming from sane data vs network alignment, so it
doesn't seem safe to remove packed without careful research for
each affect struct.) clang already warned (and was not erroring)
for many of these cases, but gcc picked up dozens more.

Restore the MKGROFF=bo MKCXX=yes build

Mark the documentation in dc(1), gprof(1), rogue(6) and fsck_ffs(8) with
the .roff flag in SUBDIR.

Sponsored by <The NetBSD Foundation>

branches: 1.48.10;
Add smaller versions of fsck_ffs(8) and newfs(8) for install media, where
support for Endian-Independent FFS and Apple UFS is disabled unless FFS_EI=1
and APPLE_UFS=1 are added to CRUNCHENV, respectively.

This reduces the size of ramdisk image for atari by over 15KB.

Thanks tsutsui and christos for their useful comments.

Do not compile in progress.c when SMALLPROG is defined.

branches: 1.46.2; 1.46.4;
-O0 for pass1.c and vax

Rework /usr/share/doc.

Update the <bsd.doc.mk> infrastructure, and update the docs to match
the new infrastructure.

- Build and install text, ps, pdf, and/or html, not roff sources.

- Don't wire the chapter numbers into the build system, or use them in
the installed pathnames. This didn't matter much when the docs were a
museum, but now that we're theoretically going to start maintaining
them again, we're going to add and remove documents periodically and
having the chapter numbers baked in creates a lot of thrashing for no
purpose.

- Specify the document name explicitly, rather than implicitly in a
path. Use this name (instead of other random strings) as the name
of the installed files.

- Specify the document section, which is the subdirectory of
/usr/share/doc to install into.

- Allow multiple subdocuments. (That is, multiple documents in one
output directory.)

- Enumerate the .png files groff emits along with html so they can be
installed.

- Remove assorted hand-rolled rules for running roff and roff widgetry
and add enough variable settings to make these unnecessary. This
includes support for
- explicit use of soelim
- refer
- tbl
- pic
- eqn

- Forcibly apply at least minimal amounts of sanity to certain
autogenerated roff files.

- Don't exclude USD.doc, SMM.doc, and PSD.doc directories from the
build, as they now actually do stuff.

Note: currently we can't generate pdf. This turns out to be a
nontrivial problem with no immediate solution forthcoming. So for now,
as a workaround, install compressed .ps as the printable form.

branches: 1.44.2; 1.44.8;
Don't depend on HAVE_GCC being always defined.

branches: 1.43.2;
WARNS=4

remove most of the remaining HAVE_GCC tests that are always true in
the modern world.

branches: 1.41.2;
merge the bouyer-quota2 branch. This adds a new on-disk format
to store disk quota usage and limits, integrated with ffs
metadata. Usage is checked by fsck_ffs (no more quotacheck)
and is covered by the WAPBL journal. Enabled with kernel
option QUOTA2 (added where QUOTA was enabled in kernel config files),
turned on with tunefs(8) on a per-filesystem
basis. mount_mfs(8) can also turn quotas on.

See http://mail-index.netbsd.org/tech-kern/2011/02/19/msg010025.html
for details.

branches: 1.40.2;
Add support for DIOCGDISKINFO to disk like device drivers. Change
partutil.c::getdiskinfo to use it to get disk geometry info.
Use DIOCGWEDGEINFO ioctl to get information about partition size, if disk
driver doesn't support it use old DIOCGDINFO. This patch adds support for
wedge like devices(lvm logical volumes, ZFS zvol partitions) to newfs and
other tools.

No objections on tech-userlevel@.

Enable WARNS=4 by default except for:
dump dump_lfs fsck_ffs fsck_lfs fsdb mount_smbfs
newfs_ext2fs newfs_lfs resize_lfs setkey

branches: 1.38.4;
Add fss(4) snapshot support to fsck_ffs(8) (via -x or -X options, like
dump(8)). This allows fsck_ffs -n to work on a snapshot of a R/W mounted
filesystem, and avoid errors related to filesystem activity.

Wrap compiler-specific flags with HAVE_GCC and HAVE_PCC as necessary. Add a few flags for PCC.

Merge the simonb-wapbl branch. From the original branch commit:

Add Wasabi System's WAPBL (Write Ahead Physical Block Logging)
journaling code. Originally written by Darrin B. Jewell while
at Wasabi and updated to -current by Antti Kantee, Andy Doran,
Greg Oster and Simon Burge.

OK'd by core@, releng@.

branches: 1.35.2; 1.35.4;
There is only one more file which requires HACKS for m68000,
so handle it in each Makefile rather than sys.mk.

These ICEs might be related with GCC Bugzilla Bug 32424
which is not resolved yet even in the upstream.

branches: 1.34.4; 1.34.6;
make all sun2 use -O0 and move most of the hacks out into just 3 files.

branches: 1.33.8; 1.33.10;
- Deal with wedges and the new disk geometry structures, instead of using
struct disklabel.

Functionality lost:
1. struct disklabel used to be updated to contain bsize, fsize, cpg.
This information was used to locate the alternative superblock in
the filesystem if the primary superblock was corrupted. We need
to find a new place to store this information if we need this
functionality.
2. On vax SMD drives that contained bad sector lists, the newfs program
knew how to get the offset and skip to the correct location in order
to place the label.

Revert previous. It is not ready for public consumption.

Don't use || if you need &&.

work around some GCC4 internal problems on m68000 platform.
document the hack, and update another GCC4 list.

sprinkle some -fno-strict-aliasing and -Wno-pointer-sign with GCC4.

WARNS=2 is the default defined in sbin/Makefile.inc. (thanks wiz)

Kill __P(), ANSIfy and WARNS=2

Use -DPROGRESS to enable the progress bar, don't depend on SMALL.

Add a progress meter to fsck_ffs based on the work by thorpej presented
to the mailing lists last January. This is optional.

Add support for the Apple UFS variation on ffs
This is the bulk of PR #17345

The general approach is to use a run time deteriminable value
for DIRBLKSIZ. Additional allowances are included for using
MAXSYMLINKLEN with FS_42INODEFMT and a shift in the cylinder group
cluster summary count array. Support is added for managing
the Apple UFS volume label.

Use ${NETBSDSRCDIR}/some/path instead of ${.CURDIR}/../../some/path

branches: 1.22.4;
#include machine/bswap.h and remove -lutil.

Add support for non-native byteorder FFS, and converting byteorder.
Also, be a bit more conservative with the clean flag: don't mark the FS
clean when we know there may still be errors (user anserwed 'n' to
a question, or fsck says "you must rerun fsck").

Use bsd.subdir.mk as appropriate.

CFLAGS->CPPFLAGS

remove -g from CFLAGS=

resolve conflicts from lite-2 merge.

Back out the .PATH.c changes. The .depend problem (and others)
will be fixed using the new .NOPATH make feature instead.

Use .PATH.c: ...

use += instead of = for CFLAGS

- util.h -> fsutil.h

- fixed all printf formats [there were a lot of %l? <-> %? mistakes]
- added missing prototypes, and made local functions static
- removed parallel preening code; this is part of fsck(8)
- use printing utilities from fsck(8)
- Makefile does not make links to fsck and fsck.8
- removed -l maxparallel option. It has no meaning anymore.

check in changes proposed in PR 2006 (approved by J.T.), to rename fsck
to fsck_ffs, so that in the future 'fsck' can be a wrapper than invokes
appropriate filesystem-specific checker programs. For now, the only
user-visible change is that the names have changed in the manual page
and in error messages; fsck and fsck.8 are now links to fsck_ffs and
fsck_ffs.8, until the rest of the transition is complete.

convert to new RCS Id conventions; reduce my headache

specify man pages the new way.

deal with new share/doc strategy

update for new doc-make/install strategy

Update from 4.4-Lite, with local changes.

Add RCS identifiers.

i don't think we're in sys.386bsd any more...

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

Drop -O[01] hacks for vax; no longer necessary for patched GCC 10

PR port-vax/57646: Import major vax toolchain fix by Kalvis Duckmanton

XXXRO: GCC 12 may require hacks for a while, but it turned out that
GCC 12 should be treated differently from older versions; a lot of new
files need -O[01] flags, while some may no longer require it.

I will take a look later.

Note that these hacks are not required if Kalvis's patches for GCC 12
are applied.

Keep doc/HACKS entries for a while (will be sync with update for GCC 12).

diff --git a/crypto/external/bsd/openssh/lib/Makefile b/crypto/external/bsd/openssh/lib/Makefile
index 96a93f5e396..4e38be1b642 100644
--- a/crypto/external/bsd/openssh/lib/Makefile
+++ b/crypto/external/bsd/openssh/lib/Makefile
@@ -117,12 +117,6 @@ COPTS.${f}.c+= -Wno-pointer-sign
# XXX
COPTS.channels.c+= -fno-strict-aliasing

-# XXX
-.if ${MACHINE} == "vax"
-COPTS.poly1305.c+= -O0
-COPTS.umac.c+= -O0
-.endif
-
COPTS.hostfile.c+= ${CC_WNO_FORMAT_TRUNCATION}
COPTS.sshkey.c+= ${CC_WNO_FORMAT_TRUNCATION}
COPTS.cipher.c+= -Wno-error=deprecated-declarations
diff --git a/distrib/utils/x_ping/Makefile b/distrib/utils/x_ping/Makefile
index f1c452a0d60..8ee77580a3f 100644
--- a/distrib/utils/x_ping/Makefile
+++ b/distrib/utils/x_ping/Makefile
@@ -12,10 +12,6 @@ CPPFLAGS+= -I${SRCDIR} -DCRUNCHOPS
DPADD= ${LIBM}
LDADD= -lm

-.if ${MACHINE_ARCH} == "vax"
-COPTS.ping.c=-O0
-.endif
-
.include <bsd.prog.mk>

.PATH: ${SRCDIR}
diff --git a/external/apache2/argon2/lib/libargon2/Makefile.inc b/external/apache2/argon2/lib/libargon2/Makefile.inc
index d3f4470c6c7..6a1dd146d7d 100644
--- a/external/apache2/argon2/lib/libargon2/Makefile.inc
+++ b/external/apache2/argon2/lib/libargon2/Makefile.inc
@@ -16,7 +16,3 @@ CFLAGS+= -pthread
LDADD+=-lpthread
DPADD+=${LIBPTHREAD}
.endif
-
-.if ${MACHINE} == "vax"
-COPTS.blake2b.c+= -O0
-.endif
diff --git a/external/bsd/jemalloc/lib/Makefile.inc b/external/bsd/jemalloc/lib/Makefile.inc
index 2f524358f46..d5da774f293 100644
--- a/external/bsd/jemalloc/lib/Makefile.inc
+++ b/external/bsd/jemalloc/lib/Makefile.inc
@@ -53,12 +53,6 @@ COPTS.ctl.c+=-Wno-error=stack-protector
COPTS.stats.c+=-Wno-error=stack-protector
COPTS.tcache.c+=-Wno-error=stack-protector

-.if ${MACHINE_ARCH} == "vax"
-# in merge_overlapping_regs, at regrename.c
-COPTS.arena.c+=-O0
-COPTS.extent.c+=-O0
-.endif
-
SRCS+=${JEMALLOC_SRCS}

jemalloc.d jemalloc.pico jemalloc.o jemalloc.ln jemalloc.po jemalloc.go: \
diff --git a/external/bsd/mdocml/lib/libmandoc/Makefile b/external/bsd/mdocml/lib/libmandoc/Makefile
index 40d7e29ae88..dfd3077a2c7 100644
--- a/external/bsd/mdocml/lib/libmandoc/Makefile
+++ b/external/bsd/mdocml/lib/libmandoc/Makefile
@@ -39,11 +39,6 @@ tbl_opts.c

MAN= mandoc.3

-# XXX
-.if ${MACHINE} == "vax"
-COPTS.mdoc_macro.c+=-O0
-.endif
-
COPTS.man_validate.c+=-Wno-error=array-bounds

.include <bsd.lib.mk>
diff --git a/external/gpl3/binutils.old/lib/libbfd/Makefile b/external/gpl3/binutils.old/lib/libbfd/Makefile
index 03477232b55..74a354532fa 100644
--- a/external/gpl3/binutils.old/lib/libbfd/Makefile
+++ b/external/gpl3/binutils.old/lib/libbfd/Makefile
@@ -41,10 +41,6 @@ CPPFLAGS+= -I${.CURDIR}/arch/${BFD_MACHINE_ARCH} -I${DIST}/include -I. \
-DDEBUGDIR=\"${DEBUGDIR}\" -DLIBDIR=\"${LIBDIR}\" \
-DBINDIR=\"${BINDIR}\"

-.if (${BFD_MACHINE_ARCH} == "vax")
-CPPFLAGS.elf.c += -O0
-.endif
-
COPTS.pei-x86_64.c+= -Wno-stack-protector
COPTS.elfxx-mips.c+= -Wno-stack-protector
COPTS.elf.c+= ${${ACTIVE_CC} == "gcc" && ${HAVE_GCC:U0} >= 10:? -Wno-overflow :}
diff --git a/external/gpl3/binutils/lib/libbfd/Makefile b/external/gpl3/binutils/lib/libbfd/Makefile
index 00cb2b20081..a3f0c25fd35 100644
--- a/external/gpl3/binutils/lib/libbfd/Makefile
+++ b/external/gpl3/binutils/lib/libbfd/Makefile
@@ -41,10 +41,6 @@ CPPFLAGS+= -I${.CURDIR}/arch/${BFD_MACHINE_ARCH} -I${DIST}/include -I. \
-DDEBUGDIR=\"${DEBUGDIR}\" -DLIBDIR=\"${LIBDIR}\" \
-DBINDIR=\"${BINDIR}\"

-.if (${BFD_MACHINE_ARCH} == "vax")
-CPPFLAGS.elf.c += -O0
-.endif
-
COPTS.pei-x86_64.c+= -Wno-stack-protector
COPTS.elfxx-mips.c+= -Wno-stack-protector
COPTS.elf.c+= ${${ACTIVE_CC} == "gcc" && ${HAVE_GCC:U0} >= 10:? -Wno-overflow :}
diff --git a/external/gpl3/gcc.old/lib/Makefile.sanitizer b/external/gpl3/gcc.old/lib/Makefile.sanitizer
index efd36ae4bee..5173a5109a0 100644
--- a/external/gpl3/gcc.old/lib/Makefile.sanitizer
+++ b/external/gpl3/gcc.old/lib/Makefile.sanitizer
@@ -88,22 +88,3 @@ COPTS+=-fno-builtin -fno-exceptions -fno-rtti -funwind-tables

# Can't profile without it`
#-fomit-frame-pointer
-
-.if ${MACHINE_ARCH} == "vax"
-COPTS.sanitizer_allocator.cc += -O1
-COPTS.sanitizer_common.cc += -O1
-COPTS.sanitizer_common_libcdep.cc += -O1
-COPTS.sanitizer_coverage_libcdep.cc += -O1
-COPTS.sanitizer_coverage_mapping_libcdep.cc += -O1
-COPTS.sanitizer_deadlock_detector1.cc += -O1
-COPTS.sanitizer_mac.cc += -O1
-COPTS.sanitizer_netbsd.cc += -O1
-COPTS.sanitizer_posix.cc += -O1
-COPTS.sanitizer_printf.cc += -O1
-COPTS.sanitizer_procmaps_common.cc += -O1
-COPTS.sanitizer_stackdepot.cc += -O1
-COPTS.sanitizer_symbolizer_libcdep.cc += -O1
-COPTS.sanitizer_symbolizer_report.cc += -O1
-COPTS.ubsan_diag.cc += -O1
-COPTS.ubsan_init.cc += -O1
-.endif
diff --git a/external/gpl3/gcc.old/lib/libasan/Makefile b/external/gpl3/gcc.old/lib/libasan/Makefile
index 1e6f9608abe..3da7bbf2f58 100644
--- a/external/gpl3/gcc.old/lib/libasan/Makefile
+++ b/external/gpl3/gcc.old/lib/libasan/Makefile
@@ -49,14 +49,6 @@ LIBDPLIBS+= m ${NETBSDSRCDIR}/lib/libm
LIBDPLIBS+= pthread ${NETBSDSRCDIR}/lib/libpthread
CPPFLAGS+=-DCAN_SANITIZE_UB=0

-.if ${MACHINE_ARCH} == "vax"
-COPTS.asan_allocator.cc += -O1
-COPTS.asan_report.cc += -O1
-COPTS.sanitizer_file.cc += -O1
-COPTS.ubsan_diag.cc += -O1
-COPTS.ubsan_init.cc += -O1
-.endif
-
.if ${MACHINE_ARCH} == "mipsel" || ${MACHINE_ARCH} == "mipseb"
COPTS.asan_interceptors.cc += -O1
.endif
diff --git a/external/gpl3/gcc.old/lib/liblsan/Makefile b/external/gpl3/gcc.old/lib/liblsan/Makefile
index 4578544a9a8..ddb7981077c 100644
--- a/external/gpl3/gcc.old/lib/liblsan/Makefile
+++ b/external/gpl3/gcc.old/lib/liblsan/Makefile
@@ -16,11 +16,6 @@ LSAN_SRCS+= \
lsan_linux.cc \
lsan_thread.cc

-.if ${MACHINE_ARCH} == "vax"
-COPTS.lsan_allocator.cc += -O1
-COPTS.sanitizer_file.cc += -O1
-.endif
-
LIB= lsan
SRCS+= ${LSAN_SRCS}
LIBDPLIBS+= m ${NETBSDSRCDIR}/lib/libm
diff --git a/external/gpl3/gcc.old/lib/libubsan/Makefile b/external/gpl3/gcc.old/lib/libubsan/Makefile
index 2e2faca7b1c..9de82550d14 100644
--- a/external/gpl3/gcc.old/lib/libubsan/Makefile
+++ b/external/gpl3/gcc.old/lib/libubsan/Makefile
@@ -26,10 +26,6 @@ UBSAN_SRCS= \
COPTS.${_s}.cc+=-frtti
.endfor

-.if ${MACHINE_ARCH} == "vax"
-COPTS.sanitizer_file.cc += -O1
-.endif
-
LIB= ubsan
SRCS+= ${UBSAN_SRCS}
LIBDPLIBS+= m ${NETBSDSRCDIR}/lib/libm
diff --git a/external/gpl3/gcc.old/usr.bin/cc1/Makefile b/external/gpl3/gcc.old/usr.bin/cc1/Makefile
index c55e292ee8e..ff12a5be524 100644
--- a/external/gpl3/gcc.old/usr.bin/cc1/Makefile
+++ b/external/gpl3/gcc.old/usr.bin/cc1/Makefile
@@ -46,17 +46,6 @@ CPPFLAGS.default-c.c+= -I${BACKENDOBJ}
BUILDSYMLINKS+= ${GNUHOSTDIST}/gcc/config/sh/sh-c.cc sh-c.c
.endif

-.if ${MACHINE_CPU} == "vax"
-COPTS.c-decl.c+=-O0
-COPTS.c-typeck.c+=-O0
-COPTS.c-array-notation.c+=-O0
-COPTS.c-common.c+=-O0
-COPTS.c-ada-spec.c+=-O0
-COPTS.cilk.c+=-O0
-COPTS.c-ubsan.c+=-O0
-COPTS.cc1-checksum.c+=-O0
-.endif
-
.if ${MACHINE_ARCH} == "mipseb" || ${MACHINE_ARCH} == "mipsel"
COPTS.c-common.c+=-O3
.endif
diff --git a/external/gpl3/gcc.old/usr.bin/cc1obj/Makefile b/external/gpl3/gcc.old/usr.bin/cc1obj/Makefile
index 3123ee0d60b..aab82ad3448 100644
--- a/external/gpl3/gcc.old/usr.bin/cc1obj/Makefile
+++ b/external/gpl3/gcc.old/usr.bin/cc1obj/Makefile
@@ -44,16 +44,8 @@ COPTS.objc-act.c+= -Wno-stack-protector
COPTS.objc-gnu-runtime-abi-01.c+= -Wno-stack-protector
COPTS.objc-next-runtime-abi-01.c+= -Wno-stack-protector

-.if ${MACHINE_ARCH} == "vax"
-COPTS.c-decl.c+=-O0
-COPTS.c-typeck.c+=-O0
-COPTS.c-array-notation.c+=-O0
-COPTS.c-common.c+=-O0
-COPTS.c-ada-spec.c+=-O0
-COPTS.cilk.c+=-O0
-COPTS.c-ubsan.c+=-O0
-COPTS.cc1obj-checksum.c+=-O0
-.endif
+COPTS.c-cppbuiltin.c+= ${${ACTIVE_CC} == "gcc" && ${HAVE_GCC:U0} == 8:? -Wno-error=format-overflow :}
+COPTS.c-typeck.c+= ${${ACTIVE_CC} == "gcc" && ${HAVE_GCC:U0} == 8:? -Wno-error=format-overflow :}

.if ${MACHINE_ARCH} == "mipseb" || ${MACHINE_ARCH} == "mipsel"
COPTS.c-common.c+=-O3
diff --git a/external/gpl3/gcc.old/usr.bin/cc1objplus/Makefile b/external/gpl3/gcc.old/usr.bin/cc1objplus/Makefile
index b9c108251c5..e04e8fc7f53 100644
--- a/external/gpl3/gcc.old/usr.bin/cc1objplus/Makefile
+++ b/external/gpl3/gcc.old/usr.bin/cc1objplus/Makefile
@@ -69,18 +69,6 @@ COPTS.objc-act.c+= -Wno-stack-protector
COPTS.objc-gnu-runtime-abi-01.c+= -Wno-stack-protector
COPTS.objc-next-runtime-abi-01.c+= -Wno-stack-protector

-.if ${MACHINE_ARCH} == "vax"
-COPTS.c-decl.c+=-O0
-COPTS.c-typeck.c+=-O0
-COPTS.c-array-notation.c+=-O0
-COPTS.c-common.c+=-O0
-COPTS.c-ada-spec.c+=-O0
-COPTS.cilk.c+=-O0
-COPTS.c-ubsan.c+=-O0
-COPTS.cc1obj-checksum.c+=-O0
-COPTS.decl.c+=-O0
-.endif
-
.if ${MACHINE_ARCH} == "mipseb" || ${MACHINE_ARCH} == "mipsel"
COPTS.c-common.c+=-O3
.endif
diff --git a/external/gpl3/gcc.old/usr.bin/cc1plus/Makefile b/external/gpl3/gcc.old/usr.bin/cc1plus/Makefile
index 7936efd3a5e..7c54559e21a 100644
--- a/external/gpl3/gcc.old/usr.bin/cc1plus/Makefile
+++ b/external/gpl3/gcc.old/usr.bin/cc1plus/Makefile
@@ -31,23 +31,6 @@ CHECKSUM_OBJS= ${LIBBACKTRACEOBJ}/libbacktrace.a \

CFLAGS+= -Wno-error=stack-protector

-.if ${MACHINE_ARCH} == "vax"
-COPTS.call.c+=-O0
-COPTS.decl.c+=-O0
-COPTS.typeck2.c+=-O0
-COPTS.class.c+=-O0
-COPTS.typeck.c+=-O0
-COPTS.init.c+=-O0
-COPTS.semantics.c+=-O0
-COPTS.mangle.c+=-O0
-COPTS.constexpr.c+=-O0
-COPTS.c-common.c+=-O0
-COPTS.c-ada-spec.c+=-O0
-COPTS.cilk.c+=-O0
-COPTS.c-ubsan.c+=-O0
-COPTS.cc1plus-checksum.c+=-O0
-.endif
-
.if ${MACHINE_ARCH} == "mipseb" || ${MACHINE_ARCH} == "mipsel"
COPTS.c-common.c+=-O3
.endif
diff --git a/external/gpl3/gcc.old/usr.bin/gcc/Makefile b/external/gpl3/gcc.old/usr.bin/gcc/Makefile
index 7a66d77ffc7..a5a4b1c3adf 100644
--- a/external/gpl3/gcc.old/usr.bin/gcc/Makefile
+++ b/external/gpl3/gcc.old/usr.bin/gcc/Makefile
@@ -25,9 +25,6 @@ ${SRCS}: ${GCCARCH}/defs.mk
.include <bsd.info.mk>

COPTS.gcc.c= -Wno-stack-protector
-.if ${MACHINE_ARCH} == "vax"
-COPTS.dse.c= -O1
-.endif

.PATH: ${DIST}/gcc ${DIST}/gcc/doc ${DIST}/gcc/c

diff --git a/external/gpl3/gcc.old/usr.bin/libdecnumber/Makefile b/external/gpl3/gcc.old/usr.bin/libdecnumber/Makefile
index 751d1513eec..33aeb1c9898 100644
--- a/external/gpl3/gcc.old/usr.bin/libdecnumber/Makefile
+++ b/external/gpl3/gcc.old/usr.bin/libdecnumber/Makefile
@@ -14,10 +14,6 @@ SRCS= decNumber.c decContext.c decimal32.c decimal64.c decimal128.c
CPPFLAGS+= -I${.CURDIR}/arch/${GCC_MACHINE_ARCH} -I${DIST}/libdecnumber
CPPFLAGS+= -I${DIST}/libgcc

-.if ${MACHINE_ARCH} == "vax"
-COPTS.decNumber.c=-O0
-.endif
-
.include <bsd.lib.mk>

# Force using C++ for this
diff --git a/external/gpl3/gcc.old/usr.bin/lto-dump/Makefile b/external/gpl3/gcc.old/usr.bin/lto-dump/Makefile
index 90f44a9fcf6..fd251464fb7 100644
--- a/external/gpl3/gcc.old/usr.bin/lto-dump/Makefile
+++ b/external/gpl3/gcc.old/usr.bin/lto-dump/Makefile
@@ -24,11 +24,6 @@ COPTS.lto-common.c+= -Wno-stack-protector
.include "../Makefile.backtrace"
.include "../Makefile.libdecnumber"

-.if ${MACHINE_ARCH} == "vax"
-COPTS.lto-lang.c+=-O0
-COPTS.lto-symtab.c+=-O0
-.endif
-
LDADD+= ${LIBIBERTY} ${LIBMPC} ${LIBMPFR} ${LIBGMP} -lintl -lz -lm
DPADD+= ${LIBIBERTY} ${LIBMPC} ${LIBMPFR} ${LIBGMP} ${LIBINTL} ${LIBZ} ${LIBM}

diff --git a/external/gpl3/gcc.old/usr.bin/lto1/Makefile b/external/gpl3/gcc.old/usr.bin/lto1/Makefile
index 4c49161a64c..340d3c6124f 100644
--- a/external/gpl3/gcc.old/usr.bin/lto1/Makefile
+++ b/external/gpl3/gcc.old/usr.bin/lto1/Makefile
@@ -21,11 +21,6 @@ COPTS.lto-common.c+= -Wno-stack-protector
.include "../Makefile.libcpp"
.include "../Makefile.libdecnumber"

-.if ${MACHINE_ARCH} == "vax"
-COPTS.lto-lang.c+=-O0
-COPTS.lto-symtab.c+=-O0
-.endif
-
LDADD+= ${LIBIBERTYOBJ}/libiberty.a
DPADD+= ${LIBIBERTYOBJ}/libiberty.a
LDADD+= ${LIBIBERTY} ${LIBMPC} ${LIBMPFR} ${LIBGMP} -lintl -lz -lm
diff --git a/external/gpl3/gdb.old/lib/libdecnumber/Makefile b/external/gpl3/gdb.old/lib/libdecnumber/Makefile
index bcd4ae5b222..ce28811526b 100644
--- a/external/gpl3/gdb.old/lib/libdecnumber/Makefile
+++ b/external/gpl3/gdb.old/lib/libdecnumber/Makefile
@@ -18,10 +18,6 @@ CPPFLAGS+= -I${.CURDIR}/arch/${GDB_MACHINE_ARCH} \

SRCS= ${G_OBJS:.o=.c} ${G_SOURCES}

-.if ${MACHINE_ARCH} == "vax"
-COPTS.decNumber.c=-O0
-.endif
-
.PATH: ${DIST}/libdecnumber ${DIST}/libdecnumber/dpd

.include <bsd.lib.mk>
diff --git a/external/gpl3/gdb/lib/libdecnumber/Makefile b/external/gpl3/gdb/lib/libdecnumber/Makefile
index c95d36e3dfa..e3169865304 100644
--- a/external/gpl3/gdb/lib/libdecnumber/Makefile
+++ b/external/gpl3/gdb/lib/libdecnumber/Makefile
@@ -18,10 +18,6 @@ CPPFLAGS+= -I${.CURDIR}/arch/${GDB_MACHINE_ARCH} \

SRCS= ${G_OBJS:.o=.c} ${G_SOURCES}

-.if ${MACHINE_ARCH} == "vax"
-COPTS.decNumber.c=-O0
-.endif
-
.PATH: ${DIST}/libdecnumber ${DIST}/libdecnumber/dpd ${DIST}/libdecnumber/bid

.include <bsd.lib.mk>
diff --git a/external/gpl3/gdb/lib/libgdb/Makefile b/external/gpl3/gdb/lib/libgdb/Makefile
index b87a515d13a..c118c363975 100644
--- a/external/gpl3/gdb/lib/libgdb/Makefile
+++ b/external/gpl3/gdb/lib/libgdb/Makefile
@@ -66,12 +66,6 @@ CFLAGS:= ${CXXFLAGS} -std=gnu++17 -Wno-error=stack-protector

ada-exp.c: ada-lex.c

-.if ${MACHINE} == "vax"
-. if ${ACTIVE_CC} == "gcc" && ${HAVE_GCC:U0} >= 8
-COPTS.read.c+= -O0
-. endif
-.endif
-
# These are generated by implicit rules and are not easy to generate
CLEANDIRFILES+= \
ada-exp.c ada-lex.c \
diff --git a/external/mit/xorg/lib/gallium.old/Makefile b/external/mit/xorg/lib/gallium.old/Makefile
index 55b2fe3d1c1..bfa40c8974f 100644
--- a/external/mit/xorg/lib/gallium.old/Makefile
+++ b/external/mit/xorg/lib/gallium.old/Makefile
@@ -1211,10 +1211,6 @@ SYMLINKS+= gallium_dri.so.${SHLIB_MAJOR}.debug ${DRIDEBUGDIR}/${_d}_dri.so.${SHL
COPTS+= ${${ACTIVE_CC} == "clang":? -Wa,-Av8plus :}
.endif

-.if ${MACHINE_ARCH} == "vax"
-COPTS.nir.c += -O1
-.endif
-
# XXXGCC12
.if ${MACHINE_ARCH} == "m68k"
COPTS.st_glsl_to_tgsi.cpp += -O1
diff --git a/external/mit/xorg/lib/gallium/Makefile b/external/mit/xorg/lib/gallium/Makefile
index 6778a8c11c8..9f1422080fe 100644
--- a/external/mit/xorg/lib/gallium/Makefile
+++ b/external/mit/xorg/lib/gallium/Makefile
@@ -1428,10 +1428,6 @@ SYMLINKS+= gallium_dri.so.${SHLIB_MAJOR}.debug ${DRIDEBUGDIR}/${_d}_dri.so.${SHL
COPTS+= ${${ACTIVE_CC} == "clang":? -Wa,-Av8plus :}
.endif

-.if ${MACHINE_ARCH} == "vax"
-COPTS.nir.c += -O1
-.endif
-
COPTS.u_atomic.c+= ${${ACTIVE_CC} == "gcc" && ${HAVE_GCC:U0} >= 10:? -Wno-builtin-declaration-mismatch :}

.include <bsd.lib.mk>
diff --git a/external/mit/xorg/lib/libX11/Makefile.libx11 b/external/mit/xorg/lib/libX11/Makefile.libx11
index e103d9ace12..0e55c2a4732 100644
--- a/external/mit/xorg/lib/libX11/Makefile.libx11
+++ b/external/mit/xorg/lib/libX11/Makefile.libx11
@@ -471,11 +471,6 @@ COPTS.OpenDis.c+= -Wno-error # XXX xf86bigfstr.h
COPTS.XlibInt.c+= -Wno-error # XXX xcmiscstr.h
COPTS.XKBBind.c+= -Wno-deprecated-declarations # uses XKeycodeToKeysym

-# XXX
-.if ${MACHINE} == "vax"
-COPTS.lcWrap.c+= -O0
-.endif
-
CWARNFLAGS.clang+= -Wno-string-plus-int

.include "${NETBSDSRCDIR}/external/mit/xorg/tools/makekeys/Makefile.makekeys"
diff --git a/games/gomoku/Makefile b/games/gomoku/Makefile
index e86a63aaea7..678537066ed 100644
--- a/games/gomoku/Makefile
+++ b/games/gomoku/Makefile
@@ -9,10 +9,6 @@ LDADD= -lcurses -lterminfo
HIDEGAME=hidegame
CPPFLAGS+= ${DEBUG:D-DDEBUG}

-.if ${MACHINE} == "vax"
-COPTS.pickmove.c += -O0
-.endif
-
#WARNS= 6 # would produce warnings about small integer types
LINTFLAGS+= -w # treat warnings as errors
LINTFLAGS+= -T # strict bool mode
diff --git a/games/phantasia/Makefile b/games/phantasia/Makefile
index e9c53f23ae5..b919dac17c6 100644
--- a/games/phantasia/Makefile
+++ b/games/phantasia/Makefile
@@ -52,7 +52,3 @@ map: map.c
./map | plot > /dev/tty

.include <bsd.prog.mk>
-
-.if defined(HAVE_GCC) && ${MACHINE_ARCH} == "vax"
-COPTS.misc.c+= -O0
-.endif
diff --git a/lib/i18n_module/UTF7/Makefile b/lib/i18n_module/UTF7/Makefile
index 712c1515d42..7136f7f0cc3 100644
--- a/lib/i18n_module/UTF7/Makefile
+++ b/lib/i18n_module/UTF7/Makefile
@@ -3,6 +3,3 @@
SRCPRE=citrus_
.include <bsd.lib.mk>

-.if ${MACHINE_ARCH} == "vax" && defined(HAVE_GCC)
-COPTS.citrus_utf7.c+= -O0
-.endif
diff --git a/lib/libbz2/Makefile b/lib/libbz2/Makefile
index b2aea1e04b0..e60a2862d48 100644
--- a/lib/libbz2/Makefile
+++ b/lib/libbz2/Makefile
@@ -15,11 +15,6 @@ SRCS= blocksort.c huffman.c crctable.c randtable.c compress.c \
INCS= bzlib.h
INCSDIR= /usr/include

-# XXX huffman.c gets mis-compiled with 2.95.3
-.if ${MACHINE_ARCH} == "vax"
-COPTS+= -O0
-.endif
-
COPTS+= ${CC_WNO_IMPLICIT_FALLTHROUGH}

# XXX blocksort.c gets mis-compiled with 4.1
diff --git a/lib/libc/gdtoa/Makefile.inc b/lib/libc/gdtoa/Makefile.inc
index 132686fc33d..609da919803 100644
--- a/lib/libc/gdtoa/Makefile.inc
+++ b/lib/libc/gdtoa/Makefile.inc
@@ -44,10 +44,3 @@ SRCS+= dmisc.c \
.if ${MACHINE_ARCH} != "vax"
SRCS+= strtord.c
.endif
-
-# XXX revisit with newer GCC.
-# Ensure numbers like 0xffff319f5fa95963 print correctly
-# ("999999999999999.98", not garbage like "?A>C>@>C:BA;A><.:<")
-.if ${MACHINE_ARCH} == "vax" && defined(HAVE_GCC)
-COPTS.misc.c+= -O0
-.endif
diff --git a/lib/libcrypt/Makefile b/lib/libcrypt/Makefile
index cb3f89d6d09..e9c8be820da 100644
--- a/lib/libcrypt/Makefile
+++ b/lib/libcrypt/Makefile
@@ -30,9 +30,6 @@ SRCS+= crypt-argon2.c
SRCS+= ${src}
COPTS.${src}+= -fvisibility=hidden
. endfor
-. if ${MACHINE} == "vax"
-COPTS.blake2b.c+= -O0
-. endif
.endif

WARNS?= 5
diff --git a/libexec/ld.elf_so/Makefile b/libexec/ld.elf_so/Makefile
index f5baae1e883..24f7e1121e5 100644
--- a/libexec/ld.elf_so/Makefile
+++ b/libexec/ld.elf_so/Makefile
@@ -141,10 +141,6 @@ CPPFLAGS+= -DRTLD_DEFAULT_LIBRARY_PATH=\"${SHLIBDIR}:${LIBDIR}\"
COPTS.rtld.c+= -Wno-stack-protector
COPTS.symbol.c+=-Wno-stack-protector

-.if ${MACHINE_CPU} == "vax"
-COPTS.rtld.c+= -O0
-.endif
-
LDADD+= -Wl,--version-script=${.CURDIR}/symbols.map
LDADD+= -L${CLIBOBJ} -L${DESTDIR}${LIBDIR}
.if ${MKPICLIB} != "no"
diff --git a/sbin/fsck_ffs/Makefile.common b/sbin/fsck_ffs/Makefile.common
index 765638b2801..824e3bf2628 100644
--- a/sbin/fsck_ffs/Makefile.common
+++ b/sbin/fsck_ffs/Makefile.common
@@ -36,7 +36,3 @@ COPTS.ffs_appleufs.c+= -Wno-pointer-sign
.if ${MACHINE_ARCH} == "m68000"
COPTS.pass1.c+= -fno-tree-fre -fno-tree-lrs
.endif
-.if ${MACHINE_ARCH} == "vax"
-COPTS.pass1.c+= -O0
-COPTS.inode.c+= -O0
-.endif
diff --git a/sbin/fsdb/Makefile b/sbin/fsdb/Makefile
index c039f73a1eb..3b0331e5880 100644
--- a/sbin/fsdb/Makefile
+++ b/sbin/fsdb/Makefile
@@ -39,12 +39,6 @@ COPTS.${f}.c+= -Wno-pointer-sign
COPTS.pass1.c+= -fno-tree-fre -fno-tree-lrs
.endif

-.if ${MACHINE_ARCH} == "vax"
-COPTS.pass1.c+= -O0
-COPTS.inode.c+= -O0
-COPTS.fsdb.c+= -O0
-.endif
-
CWARNFLAGS.gcc+= ${CC_WNO_ADDRESS_OF_PACKED_MEMBER}

.include <bsd.prog.mk>
diff --git a/sbin/newfs_ext2fs/Makefile b/sbin/newfs_ext2fs/Makefile
index 40a44b9a038..ecfdbff4d3a 100644
--- a/sbin/newfs_ext2fs/Makefile
+++ b/sbin/newfs_ext2fs/Makefile
@@ -20,8 +20,4 @@ DPADD+=${LIBPROP}

.PATH: ${NETBSDSRCDIR}/sys/ufs/ext2fs ${FSCK}

-.if ${MACHINE_ARCH} == "vax"
-COPTS.mke2fs.c=-O0
-.endif
-
.include <bsd.prog.mk>
diff --git a/sbin/ping/Makefile b/sbin/ping/Makefile
index 4f33501ea7a..1f7f56e3ea5 100644
--- a/sbin/ping/Makefile
+++ b/sbin/ping/Makefile
@@ -12,8 +12,4 @@ CPPFLAGS+= -DIPSEC
LDADD+= -lipsec
DPADD+= ${LIBIPSEC}

-.if ${MACHINE_ARCH} == "vax"
-COPTS.ping.c=-O0
-.endif
-
.include <bsd.prog.mk>
diff --git a/sys/arch/vax/conf/Makefile.vax b/sys/arch/vax/conf/Makefile.vax
index cf095d54a86..c483eaa644a 100644
--- a/sys/arch/vax/conf/Makefile.vax
+++ b/sys/arch/vax/conf/Makefile.vax
@@ -34,7 +34,6 @@ GENASSYM_CONF= ${VAX}/vax/genassym.cf
CPPFLAGS+= -D_VAX_INLINE_
AFLAGS+= -x assembler-with-cpp -fno-pic
CFLAGS+= -fno-pic
-COPTS.wsmux.c+= ${${ACTIVE_CC} == "gcc" && ${HAVE_GCC:U0} >= 9:? -O1 :}


##
diff --git a/sys/lib/libsa/Makefile b/sys/lib/libsa/Makefile
index 718c97ce857..4ef3840c753 100644
--- a/sys/lib/libsa/Makefile
+++ b/sys/lib/libsa/Makefile
@@ -96,7 +96,3 @@ SRCS+= ufs.c
lib${LIB}.o:: ${OBJS:O} __buildstdlib

CPPFLAGS+= -Wno-pointer-sign
-
-.if defined(HAVE_GCC) && ${MACHINE_ARCH} == "vax"
-COPTS.bootp.c+= -O0
-.endif
diff --git a/sys/lib/libz/Makefile b/sys/lib/libz/Makefile
index 6945c0da6c6..61fb7d4f652 100644
--- a/sys/lib/libz/Makefile
+++ b/sys/lib/libz/Makefile
@@ -25,7 +25,3 @@ CLEANFILES+= lib${LIB}.o
.include <bsd.lib.mk>

lib${LIB}.o:: ${OBJS:O} __buildstdlib
-
-.if defined(HAVE_GCC) && ${MACHINE_ARCH} == "vax"
-COPTS.inftrees.c+= -O0
-.endif
diff --git a/sys/modules/lfs/Makefile b/sys/modules/lfs/Makefile
index 962538821a5..ea8f187e160 100644
--- a/sys/modules/lfs/Makefile
+++ b/sys/modules/lfs/Makefile
@@ -18,9 +18,4 @@ SRCS+= ulfs_bmap.c ulfs_dirhash.c ulfs_inode.c ulfs_lookup.c \

WARNS= 3

-.if ${MACHINE} == "vax"
-# GCC 6.5 and 7.4 cannot compile this with -DDIAGNOSTIC and -O2/-O1
-COPTS.lfs_inode.c+= -O0
-.endif
-
.include <bsd.kmodule.mk>
diff --git a/sys/rump/fs/lib/liblfs/Makefile b/sys/rump/fs/lib/liblfs/Makefile
index abade3767d9..68b74c62502 100644
--- a/sys/rump/fs/lib/liblfs/Makefile
+++ b/sys/rump/fs/lib/liblfs/Makefile
@@ -17,9 +17,6 @@ SRCS+= ulfs_bmap.c ulfs_dirhash.c ulfs_extattr.c \


CFLAGS+= -DLFS_KERNEL_RFW
-.if ${MACHINE_ARCH} == "vax"
-COPTS.lfs_inode.c+=-O0
-.endif

.include <bsd.lib.mk>
.include <bsd.klinks.mk>
diff --git a/usr.sbin/mtrace/Makefile b/usr.sbin/mtrace/Makefile
index aef09db6198..9c125c15265 100644
--- a/usr.sbin/mtrace/Makefile
+++ b/usr.sbin/mtrace/Makefile
@@ -12,10 +12,6 @@ PROG= mtrace
SRCS= igmp.c inet.c kern.c mtrace.c
MAN= mtrace.8

-.if ${MACHINE_ARCH} == "vax"
-COPTS.mtrace.c=-O0
-.endif
-
BINMODE=4555
BINOWN= root

diff --git a/external/gpl3/gcc.old/usr.bin/backend/Makefile b/external/gpl3/gcc.old/usr.bin/backend/Makefile
index d69a26fe619..99e881c783d 100644
--- a/external/gpl3/gcc.old/usr.bin/backend/Makefile
+++ b/external/gpl3/gcc.old/usr.bin/backend/Makefile
@@ -533,68 +533,9 @@ COPTS.insn-recog.c+=-Wno-error

.if ${GCC_MACHINE_ARCH} == "vax"
CPPFLAGS+=-I${.CURDIR}/../../lib/libgcc/libgcov/arch/${GCC_MACHINE_ARCH}
-COPTS.builtins.c+=-O0
-COPTS.calls.c+=-O0
-COPTS.convert.c+=-O0
-COPTS.data-streamer-out.c+=-O0
-COPTS.dse.c+=-O0 # XXX port-vax/51967
-COPTS.dwarf2out.c+=-O0
-COPTS.expmed.c+=-O0
-COPTS.expr.c+=-O0 -Wno-error=tautological-compare
-COPTS.fixed-value.c+=-O0
-COPTS.fold-const.c+=-O0
-COPTS.generic-match.c+=-O0
-COPTS.gimple-fold.c+=-O0
-COPTS.gimple-match.c+=-O0
-COPTS.gimple-ssa-strength-reduction.c+=-O0
-COPTS.gimple-ssa-warn-restrict.c+=-O0
-COPTS.gimple.c+=-O0
-COPTS.internal-fn.c+=-O0
-COPTS.lto-streamer-out.c+=-O0
-COPTS.omp-low.c+=-O0
-COPTS.predict.c+=-O0
-COPTS.range-op.cc+=-O0
-COPTS.recog.c+=-O0
-COPTS.sanopt.c+=-O0
-COPTS.stmt.c+=-O0
-COPTS.stor-layout.c+=-O0
-COPTS.targhooks.c+=-O0
-COPTS.tree-affine.c+=-O0
-COPTS.tree-cfg.c+=-O0
-COPTS.tree-data-ref.c+=-O0
-COPTS.tree-eh.c+=-O0
-COPTS.tree-if-conv.c+=-O0
-COPTS.tree-object-size.c+=-O0
-COPTS.tree-parloops.c+=-O0
-COPTS.tree-predcom.c+=-O0
-COPTS.tree-pretty-print.c+=-O0
-COPTS.tree-ssa-alias.c+=-O0
-COPTS.tree-ssa-ccp.c+=-O0
-COPTS.tree-ssa-forwprop.c+=-O0
-COPTS.tree-ssa-loop-ivopts.c+=-O0
-COPTS.tree-ssa-loop-manip.c+=-O0
-COPTS.tree-ssa-loop-niter.c+=-O0
-COPTS.tree-ssa-math-opts.c+=-O0
-COPTS.tree-ssa-phiopt.c+= -O0
-COPTS.tree-ssa-pre.c+=-O0
-COPTS.tree-ssa-reassoc.c+=-O0
-COPTS.tree-ssa-strlen.c+=-O0
-COPTS.tree-ssa-uninit.c+=-O0
-COPTS.tree-ssa.c+=-O0
-COPTS.tree-switch-conversion.c+=-O0
-COPTS.tree-vect-data-refs.c+=-O0
-COPTS.tree-vect-loop-manip.c+=-O0
-COPTS.tree-vect-loop.c+=-O0
-COPTS.tree-vect-patterns.c+=-O0
-COPTS.tree-vect-stmts.c+=-O0
-COPTS.tree-vrp.c+=-O0
-COPTS.tree.c+=-O0
-COPTS.ubsan.c+=-O0
-COPTS.varasm.c+=-O0
-COPTS.vr-values.c+=-O0
-COPTS.web.c+=-O0
-COPTS.wide-int-range.cc+=-O0
-COPTS.wide-int.cc+=-O0
+
+COPTS.expmed.c+=-Wno-error=tautological-compare
+COPTS.expr.c+=-Wno-error=tautological-compare
.else
COPTS.tree.c= ${${ACTIVE_CC} == "clang" :? -O0 :}
.endif

branches: 1.2.8;
inode.c needs -O0 for vax

branches: 1.1.2; 1.1.4; 1.1.16;
Add smaller versions of fsck_ffs(8) and newfs(8) for install media, where
support for Endian-Independent FFS and Apple UFS is disabled unless FFS_EI=1
and APPLE_UFS=1 are added to CRUNCHENV, respectively.

This reduces the size of ramdisk image for atari by over 15KB.

Thanks tsutsui and christos for their useful comments.

fsck_ffs(8): Fix whitespace issues.

- Nix trailing whitespace.
- Omit excessive blank lines.
- Insert missing blank lines between $NetBSD$ and copyright.

No functional change intended.

Add a -z flag to zero out the up to 4 bytes of padding in directory entry
names (including the terminating NUL), as well as directory entries with
extra free space (d->d_reclen > UFS_DIRSIZ(d)).

Inspired from FreeBSD:
https://svnweb.freebsd.org/base?view=revision&revision=347066

While the kernel has been fixed to deal with the padding bytes (new
kernels will correctly zero out all the padding after the name), it
appears that there is still an issue with directory entries with extra
free space, since a newly created and populated filesystem gets modified
with "fsck_ffs -z".

simplify the endian byte-swapping code.

"s/ are are / are /" in comment. No functional change.

branches: 1.58.10; 1.58.12;
Add smaller versions of fsck_ffs(8) and newfs(8) for install media, where
support for Endian-Independent FFS and Apple UFS is disabled unless FFS_EI=1
and APPLE_UFS=1 are added to CRUNCHENV, respectively.

This reduces the size of ramdisk image for atari by over 15KB.

Thanks tsutsui and christos for their useful comments.

branches: 1.57.10; 1.57.14;
Stick ffs_, ext2_, chfs_, filecore_, cd9660_, or mfs_ in front of
the following symbols so as to disambiguate fully. (Christos already
did the lfs ones.)

lblkno
lblktosize
lfragtosize
numfrags
blkroundup
fragroundup

Rename ambiguous macros:
MAXDIRSIZE -> UFS_MAXDIRSIZE or LFS_MAXDIRSIZE
NINDIR -> FFS_NINDIR, EXT2_NINDIR, LFS_NINDIR, or MFS_NINDIR
INOPB -> FFS_INOPB, LFS_INOPB
INOPF -> FFS_INOPF, LFS_INOPF
blksize -> ffs_blksize, ext2_blksize, or lfs_blksize
sblksize -> ffs_blksize

These are not the only ambiguously defined filesystem macros, of
course, there's a pile more. I may not have found all the ambiguous
definitions of blksize(), too, as there are a lot of other things
called 'blksize' in the system.

Stick UFS_ in front of these symbols:
DIRBLKSIZ
DIRECTSIZ
DIRSIZ
OLDDIRFMT
NEWDIRFMT

Part of PR 47909.

Stuff UFS_ in front of a few of ufs's symbols to reduce namespace
pollution. Specifically:
ROOTINO -> UFS_ROOTINO
WINO -> UFS_WINO
NXADDR -> UFS_NXADDR
NDADDR -> UFS_NDADDR
NIADDR -> UFS_NIADDR
MAXSYMLINKLEN -> UFS_MAXSYMLINKLEN
MAXSYMLINKLEN_UFS[12] -> UFS[12]_MAXSYMLINKLEN (for consistency)

Sort out ext2fs's misuse of NDADDR and NIADDR; fortunately, these have
the same values in ext2fs and ffs.

No functional change intended.

branches: 1.53.4; 1.53.10;
merge the bouyer-quota2 branch. This adds a new on-disk format
to store disk quota usage and limits, integrated with ffs
metadata. Usage is checked by fsck_ffs (no more quotacheck)
and is covered by the WAPBL journal. Enabled with kernel
option QUOTA2 (added where QUOTA was enabled in kernel config files),
turned on with tunefs(8) on a per-filesystem
basis. mount_mfs(8) can also turn quotas on.

See http://mail-index.netbsd.org/tech-kern/2011/02/19/msg010025.html
for details.

branches: 1.52.2;
fix sign-compare issues

branches: 1.51.4;
Fix funny whitespace in a variable assignment.

branches: 1.50.4; 1.50.6;
Make sure that the exit values are always sane, and use symbolic instead
of magic constants. Reviewed by go@

branches: 1.49.8; 1.49.16; 1.49.18;
c99 initializers

Don't use cast expressions as lvalues as newer versions of gcc warn.

linkup: reparent() only when lost+found is created successfully.

64 bit inode changes

sprinkle const.

appease gcc -Wuninitialized

whitespace (pointed out by wiz@)

Kill __P(), ANSIfy and WARNS=2

have allocdir fail cleanly when allocino fails

Reduce memory usage slightly.

- some KNF (80 cols)
- fix a printf format issue

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22308, verified by myself.

use bounded string op

Add support for UFS2. UFS2 is an enhanced FFS, adding support for
64 bit block pointers, extended attribute storage, and a few
other things.

This commit does not yet include the code to manipulate the extended
storage (for e.g. ACLs), this will be done later.

Originally written by Kirk McKusick and Network Associates Laboratories for
FreeBSD.

Bump daddr_t to 64 bits. Replace it with int32_t in all places where
it was used on-disk, so that on-disk formats remain the same.
Remove ufs_daddr_t and ufs_lbn_t for the time being.

Add support for the Apple UFS variation on ffs
This is the bulk of PR #17345

The general approach is to use a run time deteriminable value
for DIRBLKSIZ. Additional allowances are included for using
MAXSYMLINKLEN with FS_42INODEFMT and a shift in the cylinder group
cluster summary count array. Support is added for managing
the Apple UFS volume label.

Don't bother testing if a uint8_t is > 256 -- that test is always false.

Fix a rather glaring byte-swapping bug: di_size is 64 bits, not 16.

Whoops; call propagate() with the right child inode number. (Doesn't break
anything, but it would have made reconnect less efficient.)

Abstract the code to attach a directory to its parent's child list into a
separate function, and call it from multiple places in linkup() to handle
reconnects and creation of /lost+found.

The reconnect algorithm was historically O(n^4).
Some years ago I made it O(n^2).
Someone helpfully made it O(n^4) again.
Today I'm making it O(n).
If that's not good enough, I don't know what else to do. B-)

Technical details:
* The graph traversal in propagate() is modified to be able to start from any
point in the tree. To handle certain exceptional cases, it is also modified
to work in two passes, marking the tree with a special tag and then changing
it to DFOUND.
* The reconnect case now modifies the child/sibling pointers and calls
propagate() to propagate the connection state starting with the reconnected
directory.

Pray that you never encounter a file system trashed enough for this to matter.

Need an lfdir global variable now.

Fix an evil ugly bug which causes files placed into lost+found to
be inconsistent, and unremovable. From Ethan Solomita <ethan@geocast.com>.
Reviewed by fvdl.

branches: 1.26.4;
Changes for softdep code.

branches: 1.25.2; 1.25.4; 1.25.8;
Add support for non-native byteorder FFS, and converting byteorder.
Also, be a bit more conservative with the clean flag: don't mark the FS
clean when we know there may still be errors (user anserwed 'n' to
a question, or fsck says "you must rerun fsck").

- don't indiscriminately include <stdlib.h> and <unistd.h> in "fsck.h"
- explicitly pull in <stdio.h>, <stdlib.h> and <unistd.h> in *.c as necessary

resolve conflicts from lite-2 merge.

* cleanup for WARNS=1
* deprecate register
* cleanup manpage
* remove unused docheck() func
* prefix hex numbers with '0x'
* getopt returns -1 not EOF

lost+found directories are mode -1700

- util.h -> fsutil.h

- fixed all printf formats [there were a lot of %l? <-> %? mistakes]
- added missing prototypes, and made local functions static
- removed parallel preening code; this is part of fsck(8)
- use printing utilities from fsck(8)
- Makefile does not make links to fsck and fsck.8
- removed -l maxparallel option. It has no meaning anymore.

Use a different tree-walking algorithm in propagate().

branches: 1.17.4;
convert to new RCS Id conventions; reduce my headache

Mostly sync with CSRG.

more cleanups from Jim Jegers, passed over by me.

On a recursive call to fsck_readdir() through getpathname(), don't attempt to
fix a problem that's already being fixed.

dircheck() shouldn't be looking at d_type or d_namlen in blank entries
*at all*. Not only is it wrong, but it causes a serious problem on
little-endian machines, since after -c2 conversion, d_type will often
be > 15.

Remove some more uses of obsolete functions.

Eliminate uses of some obsolete functions.

A correct fix for the byte order problem when reconnecting.

Fix oversight in last patch.

Fix byte-order problem with directory entry creation.

branches: 1.7.2;
Update from 4.4-Lite, with local changes.

need <sys/time.h>

Patch from Bruce Evans to deal with file names with NULs in them.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

fsck_ffs(8): Fix whitespace issues.

- Nix trailing whitespace.
- Omit excessive blank lines.
- Insert missing blank lines between $NetBSD$ and copyright.

No functional change intended.

Restore backward compatibility of UFS2 with previous NetBSD releases by
disabling support in UFS2 for extended attributes (including ACLs).
Add a new variant of UFS2 called "UFS2ea" that does support extended attributes.
Add new fsck_ffs operations "-c ea" and "-c no-ea" to convert file systems
from UFS2 to UFS2ea and vice-versa (both of which delete all existing extended
attributes in the process).

share more code.

branches: 1.26.2;
merge the bouyer-quota2 branch. This adds a new on-disk format
to store disk quota usage and limits, integrated with ffs
metadata. Usage is checked by fsck_ffs (no more quotacheck)
and is covered by the WAPBL journal. Enabled with kernel
option QUOTA2 (added where QUOTA was enabled in kernel config files),
turned on with tunefs(8) on a per-filesystem
basis. mount_mfs(8) can also turn quotas on.

See http://mail-index.netbsd.org/tech-kern/2011/02/19/msg010025.html
for details.

branches: 1.25.2;
Do some basic checks of the WAPBL journal, to abort the boot before the
kernel refuse to mount a filesystem read-write (booting a system
multiuser with critical filesystems read-only is bad):
Add a check_wapbl() which will check some WAPBL values in the superblock,
and try to read the journal via wapbl_replay_start() if there is one.
pfatal() if one of these fail (abort boot if in preen mode,
as "CONTINUE" otherwise). In non-preen mode the bogus journal will
be cleared.
check_wapbl() is always called if the superblock supports WAPBL.
Even if FS_DOWAPBL is not there, there could be flags asking the
kernel to clear or create a log with bogus values which would cause the
kernel refuse to mount the filesystem.
Discussed in
http://mail-index.netbsd.org/tech-kern/2009/08/17/msg005896.html
and followups.

branches: 1.24.2; 1.24.8;
Add fss(4) snapshot support to fsck_ffs(8) (via -x or -X options, like
dump(8)). This allows fsck_ffs -n to work on a snapshot of a R/W mounted
filesystem, and avoid errors related to filesystem activity.

Merge the simonb-wapbl branch. From the original branch commit:

Add Wasabi System's WAPBL (Write Ahead Physical Block Logging)
journaling code. Originally written by Darrin B. Jewell while
at Wasabi and updated to -current by Antti Kantee, Andy Doran,
Greg Oster and Simon Burge.

OK'd by core@, releng@.

branches: 1.22.20; 1.22.24; 1.22.26;
sprinkle const.

whitespace (pointed out by wiz@)

Kill __P(), ANSIfy and WARNS=2

do not upgrade superblock or set FS_FLAGS_UPDATED unless -c 4 option
is provided.
add compatibility for filesystems before FFSv2 integration
these patches are from pr port-macppc/23925 and should also
fix problems discussed in pr kern/21404 and pr kern/21283

use bounded string op

Write update some old fields when writing the superblock, similar to
ffs_oldfscompat_write() in the kernel. Use the old totals when
time < old_time (i.e. an old kernel or fsck wrote the filesystem last).
When setting the date back on a new kernel, that works out ok, since
new kernels always update both fields.

Add support for UFS2. UFS2 is an enhanced FFS, adding support for
64 bit block pointers, extended attribute storage, and a few
other things.

This commit does not yet include the code to manipulate the extended
storage (for e.g. ACLs), this will be done later.

Originally written by Kirk McKusick and Network Associates Laboratories for
FreeBSD.

Bump daddr_t to 64 bits. Replace it with int32_t in all places where
it was used on-disk, so that on-disk formats remain the same.
Remove ufs_daddr_t and ufs_lbn_t for the time being.

If SIGINFO is received, display summary information to stderr.
Based on changes in FreeBSD, via Chuck Cranor <chuck@research.att.com>

remove redundant declarations

In pass 5, check alternate superblocks for consistency with
the current in-core master superblock, and fix them up if
they're incorrect. Move the code that writes the alternate
superblocks if (cvtlevel || doswap) into pass 5 for efficiency.

Reviewd by Charles Hannum, and used by me to fix up a curdled
file system.

The reconnect algorithm was historically O(n^4).
Some years ago I made it O(n^2).
Someone helpfully made it O(n^4) again.
Today I'm making it O(n).
If that's not good enough, I don't know what else to do. B-)

Technical details:
* The graph traversal in propagate() is modified to be able to start from any
point in the tree. To handle certain exceptional cases, it is also modified
to work in two passes, marking the tree with a special tag and then changing
it to DFOUND.
* The reconnect case now modifies the child/sibling pointers and calls
propagate() to propagate the connection state starting with the reconnected
directory.

Pray that you never encounter a file system trashed enough for this to matter.

branches: 1.10.12;
const poisoning.

Remove redundant declaration.

Add support for non-native byteorder FFS, and converting byteorder.
Also, be a bit more conservative with the clean flag: don't mark the FS
clean when we know there may still be errors (user anserwed 'n' to
a question, or fsck says "you must rerun fsck").

resolve conflicts from lite-2 merge.

- util.h -> fsutil.h

- fixed all printf formats [there were a lot of %l? <-> %? mistakes]
- added missing prototypes, and made local functions static
- removed parallel preening code; this is part of fsck(8)
- use printing utilities from fsck(8)
- Makefile does not make links to fsck and fsck.8
- removed -l maxparallel option. It has no meaning anymore.

Set the clean flag if necessary. If preening, don't check `clean' file
systems.

convert to new RCS Id conventions; reduce my headache

adjust; from jimj.

more cleanups from Jim Jegers, passed over by me.

unnecessary file from lite-2 merge

branches: 1.1.1;
Initial revision

s/ownerchip/ownership/ in comment.

branches: 1.58.2;
fsck_ffs(8): Fix whitespace issues.

- Nix trailing whitespace.
- Omit excessive blank lines.
- Insert missing blank lines between $NetBSD$ and copyright.

No functional change intended.

catch up with sign changes in the fs.h

branches: 1.56.2;
Restore backward compatibility of UFS2 with previous NetBSD releases by
disabling support in UFS2 for extended attributes (including ACLs).
Add a new variant of UFS2 called "UFS2ea" that does support extended attributes.
Add new fsck_ffs operations "-c ea" and "-c no-ea" to convert file systems
from UFS2 to UFS2ea and vice-versa (both of which delete all existing extended
attributes in the process).

add NO_IOBUF_ALIGNED to not pull aligned_alloc() for really constrained
boot media

Fix depenency on common symbols in sbin.

Add a -z flag to zero out the up to 4 bytes of padding in directory entry
names (including the terminating NUL), as well as directory entries with
extra free space (d->d_reclen > UFS_DIRSIZ(d)).

Inspired from FreeBSD:
https://svnweb.freebsd.org/base?view=revision&revision=347066

While the kernel has been fixed to deal with the padding bytes (new
kernels will correctly zero out all the padding after the name), it
appears that there is still an issue with directory entries with extra
free space, since a newly created and populated filesystem gets modified
with "fsck_ffs -z".

branches: 1.52.12;
__empty -> __nothing

use __empty

Add smaller versions of fsck_ffs(8) and newfs(8) for install media, where
support for Endian-Independent FFS and Apple UFS is disabled unless FFS_EI=1
and APPLE_UFS=1 are added to CRUNCHENV, respectively.

This reduces the size of ramdisk image for atari by over 15KB.

Thanks tsutsui and christos for their useful comments.

branches: 1.49.26; 1.49.30;
merge the bouyer-quota2 branch. This adds a new on-disk format
to store disk quota usage and limits, integrated with ffs
metadata. Usage is checked by fsck_ffs (no more quotacheck)
and is covered by the WAPBL journal. Enabled with kernel
option QUOTA2 (added where QUOTA was enabled in kernel config files),
turned on with tunefs(8) on a per-filesystem
basis. mount_mfs(8) can also turn quotas on.

See http://mail-index.netbsd.org/tech-kern/2011/02/19/msg010025.html
for details.

branches: 1.48.2;
minimal ansification/constification

Disable userid to username lookups by default. Add a -U flag to perform them.
In single user mode lookups that involve the network might not work and they
slow down fsck.

Make sure that the exit values are always sane, and use symbolic instead
of magic constants. Reviewed by go@

branches: 1.45.4; 1.45.10; 1.45.18; 1.45.20;
Don't use cast expressions as lvalues as newer versions of gcc warn.

Remove leading __ from __(const|inline|signed|volatile) -- it is obsolete.

sprinkle const.

Kill __P(), ANSIfy and WARNS=2

Use -DPROGRESS to enable the progress bar, don't depend on SMALL.

Add a progress meter to fsck_ffs based on the work by thorpej presented
to the mailing lists last January. This is optional.

Reduce memory usage slightly.

Add ffs internal snapshots. Written by Marshall Kirk McKusick for FreeBSD.

- Not enabled by default. Needs kernel option FFS_SNAPSHOT.
- Change parameters of ffs_blkfree.
- Let the copy-on-write functions return an error so spec_strategy
may fail if the copy-on-write fails.
- Change genfs_*lock*() to use vp->v_vnlock instead of &vp->v_lock.
- Add flag B_METAONLY to VOP_BALLOC to return indirect block buffer.
- Add a function ffs_checkfreefile needed for snapshot creation.
- Add special handling of snapshot files:
Snapshots may not be opened for writing and the attributes are read-only.
Use the mtime as the time this snapshot was taken.
Deny mtime updates for snapshot files.
- Add function transferlockers to transfer any waiting processes from
one lock to another.
- Add vfsop VFS_SNAPSHOT to take a snapshot and make it accessible through
a vnode.
- Add snapshot support to ls, fsck_ffs and dump.

Welcome to 2.0F.

Approved by: Jason R. Thorpe <thorpej@netbsd.org>

do not upgrade superblock or set FS_FLAGS_UPDATED unless -c 4 option
is provided.
add compatibility for filesystems before FFSv2 integration
these patches are from pr port-macppc/23925 and should also
fix problems discussed in pr kern/21404 and pr kern/21283

Add a -q (quiet) option to print nothing for clean filesystems.
Support in fsck_ffs and stub in fsck_xxx.
Push a few more messages through pwarn() instead of printf() to ensure
disk name is shown.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22308, verified by myself.

Oops. The clearinode macro had the UFS2 case reversed, causing it to
zero out two inodes in the plain FFS case, since UFS2 dinodes are
twice as big.

Write update some old fields when writing the superblock, similar to
ffs_oldfscompat_write() in the kernel. Use the old totals when
time < old_time (i.e. an old kernel or fsck wrote the filesystem last).
When setting the date back on a new kernel, that works out ok, since
new kernels always update both fields.

Add support for UFS2. UFS2 is an enhanced FFS, adding support for
64 bit block pointers, extended attribute storage, and a few
other things.

This commit does not yet include the code to manipulate the extended
storage (for e.g. ACLs), this will be done later.

Originally written by Kirk McKusick and Network Associates Laboratories for
FreeBSD.

Consistently spell occurrence with two rs.

Bump daddr_t to 64 bits. Replace it with int32_t in all places where
it was used on-disk, so that on-disk formats remain the same.
Remove ufs_daddr_t and ufs_lbn_t for the time being.

Add support for the Apple UFS variation on ffs
This is the bulk of PR #17345

The general approach is to use a run time deteriminable value
for DIRBLKSIZ. Additional allowances are included for using
MAXSYMLINKLEN with FS_42INODEFMT and a shift in the cylinder group
cluster summary count array. Support is added for managing
the Apple UFS volume label.

If SIGINFO is received, display summary information to stderr.
Based on changes in FreeBSD, via Chuck Cranor <chuck@research.att.com>

remove third argument (`int ns') from ffs_sb_swap(), and let ffs_sb_swap()
determine the endianness of the `struct fs *o' superblock from o->fs_magic
and set needswap as necessary, rather than trusting the caller to get
it right. invariably, almost every caller of ffs_sb_swap() was calling it
with ns set to the wrong value for ns anyway!
ansi KNF ffs_bswap.c declarations whilst here.

this fixes all sorts of problems when trying to use other-endian file systems,
notably the kernel trying to access memory *way* off, possibly corrupting or
panicing, and userland programs SEGVing and/or corrupting things (e.g,
"fsck_ffs -B" to swap a file system endianness).

whilst the previous rev of ffs_bswap.c (1.10, 2000/12/23) made this problem
worse, i suspect that the problem was always there and previous versions
just happened not to trash things at the wrong time.

FFS_EI should now be a lot more stable.

- implement -F; treat provided filesystems as images in regular files
- replace "filesystem" with "file system" as appropriate
- grammar fixes

minor whitespace cleanup

remove redundant declarations

In pass 5, check alternate superblocks for consistency with
the current in-core master superblock, and fix them up if
they're incorrect. Move the code that writes the alternate
superblocks if (cvtlevel || doswap) into pass 5 for efficiency.

Reviewd by Charles Hannum, and used by me to fix up a curdled
file system.

The reconnect algorithm was historically O(n^4).
Some years ago I made it O(n^2).
Someone helpfully made it O(n^4) again.
Today I'm making it O(n).
If that's not good enough, I don't know what else to do. B-)

Technical details:
* The graph traversal in propagate() is modified to be able to start from any
point in the tree. To handle certain exceptional cases, it is also modified
to work in two passes, marking the tree with a special tag and then changing
it to DFOUND.
* The reconnect case now modifies the child/sibling pointers and calls
propagate() to propagate the connection state starting with the reconnected
directory.

Pray that you never encounter a file system trashed enough for this to matter.

Some more `extern's for initialised globals.

branches: 1.20.4;
Update for softdep code.

branches: 1.19.4; 1.19.8;
#include machine/bswap.h and remove -lutil.

Use DINODE_SIZE rather than sizeof(struct dinode) and/or pointer arithmetic.

Add support for non-native byteorder FFS, and converting byteorder.
Also, be a bit more conservative with the clean flag: don't mark the FS
clean when we know there may still be errors (user anserwed 'n' to
a question, or fsck says "you must rerun fsck").

a quad_t is an int64_t, not a u_int64_t...

- don't indiscriminately include <stdlib.h> and <unistd.h> in "fsck.h"
- explicitly pull in <stdio.h>, <stdlib.h> and <unistd.h> in *.c as necessary

resolve conflicts from lite-2 merge.

Bug fix from Kirk McKusick <mckusick@McKusick.COM>:

Fix a bug in fsck_ffs where if a directory somehow develops a hole
(that is a block pointer that has a value of zero), fsck would give the
filesystem a clean bill of health, but the kernel would panic when
accessing the directory with the hole. Fsck now checks for holes
in directories. If found in preen mode, fsck fails. In manual
mode, it can be directed to shorten the directory to the beginning of
the hole. A more complete solution would be to allocate a block to fill
the hole. However, this is a lot more work for a `cannot happen' error,
so the extra effort seems unwarranted.

- fixed all printf formats [there were a lot of %l? <-> %? mistakes]
- added missing prototypes, and made local functions static
- removed parallel preening code; this is part of fsck(8)
- use printing utilities from fsck(8)
- Makefile does not make links to fsck and fsck.8
- removed -l maxparallel option. It has no meaning anymore.

Use a different tree-walking algorithm in propagate().

branches: 1.10.4;
Set the clean flag if necessary. If preening, don't check `clean' file
systems.

type sizes

convert to new RCS Id conventions; reduce my headache

more cleanups from Jim Jegers, passed over by me.

Update from 4.4-Lite, with local changes.

Prototype some things.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

Restore backward compatibility of UFS2 with previous NetBSD releases by
disabling support in UFS2 for extended attributes (including ACLs).
Add a new variant of UFS2 called "UFS2ea" that does support extended attributes.
Add new fsck_ffs operations "-c ea" and "-c no-ea" to convert file systems
from UFS2 to UFS2ea and vice-versa (both of which delete all existing extended
attributes in the process).

Add a -z flag to zero out the up to 4 bytes of padding in directory entry
names (including the terminating NUL), as well as directory entries with
extra free space (d->d_reclen > UFS_DIRSIZ(d)).

Inspired from FreeBSD:
https://svnweb.freebsd.org/base?view=revision&revision=347066

While the kernel has been fixed to deal with the padding bytes (new
kernels will correctly zero out all the padding after the name), it
appears that there is still an issue with directory entries with extra
free space, since a newly created and populated filesystem gets modified
with "fsck_ffs -z".

branches: 1.50.14;
Document the version fsck_ffs first appeared.
Bump date.

Mention scan_ffs -b.
Based on a patch by Matthew Mondor <mm_lists@pulsar-zone.net> on
tech-userlevel.
Bump date.

branches: 1.48.4;
Improve documentation of FFS formats and format levels from PR 32100.
Prompted also by recent discussion on tech-kern. Bump date.

Re-add -q description that got lost when -x was added.
Sort descriptions.
Bump date.

New sentence, new line.

Use "FFSv2" instead of "UFS2".
Expand the description of -c.
Xref dumpfs.

branches: 1.44.4;
Disable userid to username lookups by default. Add a -U flag to perform them.
In single user mode lookups that involve the network might not work and they
slow down fsck.

Add an explanation of why one would want to use -x/-X, after
discussion with bouyer@.

Add fss(4) snapshot support to fsck_ffs(8) (via -x or -X options, like
dump(8)). This allows fsck_ffs -n to work on a snapshot of a R/W mounted
filesystem, and avoid errors related to filesystem activity.

Merge the simonb-wapbl branch. From the original branch commit:

Add Wasabi System's WAPBL (Write Ahead Physical Block Logging)
journaling code. Originally written by Darrin B. Jewell while
at Wasabi and updated to -current by Antti Kantee, Andy Doran,
Greg Oster and Simon Burge.

OK'd by core@, releng@.

branches: 1.40.10; 1.40.26; 1.40.30; 1.40.32;
Sort options. Bump date for -P. Remove superfluous -.

Add a progress meter to fsck_ffs based on the work by thorpej presented
to the mailing lists last January. This is optional.

Correct a misnumbering of options, as noted in PR 25784.

Various improvements; mostly mention arguments of flags by name to
simplify the description and improve lists.

branches: 1.36.2;
Update Dd for previous and fix a case.

do not upgrade superblock or set FS_FLAGS_UPDATED unless -c 4 option
is provided.
add compatibility for filesystems before FFSv2 integration
these patches are from pr port-macppc/23925 and should also
fix problems discussed in pr kern/21404 and pr kern/21283

Add a -q (quiet) option to print nothing for clean filesystems.
Support in fsck_ffs and stub in fsck_xxx.
Push a few more messages through pwarn() instead of printf() to ensure
disk name is shown.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22308, verified by myself.

.Nm does not need a dummy argument ("") before punctuation or
for correct formatting of the SYNOPSIS any longer.

New sentence, new line. By Robert Elz with minimal fixes.

Add support for the Apple UFS variation on ffs
This is the bulk of PR #17345

The general approach is to use a run time deteriminable value
for DIRBLKSIZ. Additional allowances are included for using
MAXSYMLINKLEN with FS_42INODEFMT and a shift in the cylinder group
cluster summary count array. Support is added for managing
the Apple UFS volume label.

If SIGINFO is received, display summary information to stderr.
Based on changes in FreeBSD, via Chuck Cranor <chuck@research.att.com>

Edit -mdoc usage.

* There is no -indent option to .Bd or .Bl, although you would
never know that from its frequent use in this tree. There is a
"-offset indent" combination that makes sense, and you can certainly
say "-width indent".

* Also, you can't markup the -width option argument, tho you CAN
use a callable macro. So "-width Ar filename" doesn't make sense,
but either "-width Ar" or "-width filename" does, as might something
like "-width xxfilename" for a little extra space.

* There are a lot of needlessly complex hanging tag macros in man4 used
to create simple item lists. Those should be simplified one of these
days before someone copies and edits yet another man4 page.

Sort SEE ALSO, whitespace nits.

- changes to -F semantics:
- remove the restriction that filesystem must be a regular file
- don't try and read a disklabel
- use `p' (instead of `h') as the index of the last partition

- implement -F; treat provided filesystems as images in regular files
- replace "filesystem" with "file system" as appropriate
- grammar fixes

Drop arguments of .Os.

Clean up SYNOPSIS formatting.

fix typo.

Add support for non-native byteorder FFS, and converting byteorder.
Also, be a bit more conservative with the clean flag: don't mark the FS
clean when we know there may still be errors (user anserwed 'n' to
a question, or fsck says "you must rerun fsck").

use .Nm correctly

* cleanup for WARNS=1
* deprecate register
* cleanup manpage
* remove unused docheck() func
* prefix hex numbers with '0x'
* getopt returns -1 not EOF

lost+found directories are mode -1700

The fs level (ref: fsck -c ...) can be determined from the second
line of the output from dumpfs(8), not the first line. part of [bin/1353]

alternate -> alternative, per PR 2643

fix reboot(8) xref

eliminate obsolete references to mkfs(8);
from Klaus Klein <kleink@layla.inka.de>

Update this to reflect reality after the fsck/fsck_ffs split.

- fixed all printf formats [there were a lot of %l? <-> %? mistakes]
- added missing prototypes, and made local functions static
- removed parallel preening code; this is part of fsck(8)
- use printing utilities from fsck(8)
- Makefile does not make links to fsck and fsck.8
- removed -l maxparallel option. It has no meaning anymore.

check in changes proposed in PR 2006 (approved by J.T.), to rename fsck
to fsck_ffs, so that in the future 'fsck' can be a wrapper than invokes
appropriate filesystem-specific checker programs. For now, the only
user-visible change is that the names have changed in the manual page
and in error messages; fsck and fsck.8 are now links to fsck_ffs and
fsck_ffs.8, until the rest of the transition is complete.

implement a 'force check' flag, '-f'. I used the SunOS name, but the Digital
semantics. now:
(1) dirty file systems will always be checked; nothing new there.
(2) if not '-f' clean file systems will _NEVER_ be checked,
i.e. they won't be checked even if -p isn't specified. This
allows one to 'fsck -p ; fsck' to preen, then clean up
anything that 'fsck -p' barfs on, without waiting for the
clean file systems to be checked again.
(3) if '-f' clean file systems will ALWAYS be checked. This
allows people to put 'fsck -fp' into /etc/rc on systems
where they're leery of the FS clean flag state, need
the extra reliability, and can afford time 'wasted'
in checks.
The assumption made here is that if a file system is marked clean, it
_IS CLEAN_, really, and shouldn't be checked unless fsck is explicitly
told to (with -f). This should be a valid assumption, but may not be in
the presence of file system bugs. Documentation updated to note '-f'.

convert to new RCS Id conventions; reduce my headache

Mostly sync with CSRG.

Update from 4.4-Lite, with local changes.

typo

Updated to -mandoc macros.

Add RCS indentifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

Fix a few typos in the word 'offset' in comments and error message.

branches: 1.78.2;
Revert "fsck_ffs(8): Ensure A divides S before aligned_alloc(A, S)."

C17 lifted this restriction.

fsck_ffs(8): Fix whitespace issues.

- Nix trailing whitespace.
- Omit excessive blank lines.
- Insert missing blank lines between $NetBSD$ and copyright.

No functional change intended.

fsck_ffs(8): Ensure A divides S before aligned_alloc(A, S).

Required by C11 Sec. 7.22.3.1 The aligned_alloc function, para. 2,
p. 348:

The value of alignment shall be a valid alignment supported by the
implementation and the value of size shall be an integral multiple
of alignment.

XXX pullup-10

Use %zu rather than %lu to print a size_t (should fix i386 build).

But, philosophical question, shouldn't the product of two size_t
variables really be an area_t, or something like that?

catch up with sign changes in the fs.h

branches: 1.73.6;
align buffers used for I/O to DEV_BSIZE so it's executed more optimally
when run for xbd(4) raw (character) device

branches: 1.72.12;
Add smaller versions of fsck_ffs(8) and newfs(8) for install media, where
support for Endian-Independent FFS and Apple UFS is disabled unless FFS_EI=1
and APPLE_UFS=1 are added to CRUNCHENV, respectively.

This reduces the size of ramdisk image for atari by over 15KB.

Thanks tsutsui and christos for their useful comments.

branches: 1.71.8; 1.71.12;
Iterate over fields of struct seperately to avoid warnings from pedantic compilers

Fix pasto in the !ufs2 case (use dp->dp1 and not of dp->dp2).
This would be a problem only when allocating a new data block and the
indir block is already allocated, which explains why automated tests didn't
find it.
Problem reported on tech-kern@ and fix tested by manu@.

Stick ffs_ in front of the following macros:
fragstoblks()
blkstofrags()
fragnum()
blknum()

to finish the job of distinguishing them from the lfs versions, which
Christos renamed the other day.

I believe this is the last of the overtly ambiguous exported symbols
from ffs... or at least, the last of the ones that conflicted with lfs.
ffs still pollutes the C namespace very broadly (as does ufs) and this
needs quite a bit more cleanup.

XXX: boo on macros with lowercase names. But I'm not tackling that just yet.

Stick ffs_, ext2_, chfs_, filecore_, cd9660_, or mfs_ in front of
the following symbols so as to disambiguate fully. (Christos already
did the lfs ones.)

lblkno
lblktosize
lfragtosize
numfrags
blkroundup
fragroundup

fsbtodb() -> FFS_FSBTODB(), EXT2_FSBTODB(), or MFS_FSBTODB()
dbtofsb() -> FFS_DBTOFSB() or EXT2_DBTOFSB()

(Christos already did the lfs ones a few days back)

Rename ambiguous macros:
MAXDIRSIZE -> UFS_MAXDIRSIZE or LFS_MAXDIRSIZE
NINDIR -> FFS_NINDIR, EXT2_NINDIR, LFS_NINDIR, or MFS_NINDIR
INOPB -> FFS_INOPB, LFS_INOPB
INOPF -> FFS_INOPF, LFS_INOPF
blksize -> ffs_blksize, ext2_blksize, or lfs_blksize
sblksize -> ffs_blksize

These are not the only ambiguously defined filesystem macros, of
course, there's a pile more. I may not have found all the ambiguous
definitions of blksize(), too, as there are a lot of other things
called 'blksize' in the system.

Stuff UFS_ in front of a few of ufs's symbols to reduce namespace
pollution. Specifically:
ROOTINO -> UFS_ROOTINO
WINO -> UFS_WINO
NXADDR -> UFS_NXADDR
NDADDR -> UFS_NDADDR
NIADDR -> UFS_NIADDR
MAXSYMLINKLEN -> UFS_MAXSYMLINKLEN
MAXSYMLINKLEN_UFS[12] -> UFS[12]_MAXSYMLINKLEN (for consistency)

Sort out ext2fs's misuse of NDADDR and NIADDR; fortunately, these have
the same values in ext2fs and ffs.

No functional change intended.

branches: 1.64.4; 1.64.6; 1.64.10; 1.64.12; 1.64.14;
merge the bouyer-quota2 branch. This adds a new on-disk format
to store disk quota usage and limits, integrated with ffs
metadata. Usage is checked by fsck_ffs (no more quotacheck)
and is covered by the WAPBL journal. Enabled with kernel
option QUOTA2 (added where QUOTA was enabled in kernel config files),
turned on with tunefs(8) on a per-filesystem
basis. mount_mfs(8) can also turn quotas on.

See http://mail-index.netbsd.org/tech-kern/2011/02/19/msg010025.html
for details.

branches: 1.63.2;
Centralize time printing and deal with ctime possibly returning NULL.

fix sign-compare issues

branches: 1.61.4;
Disable userid to username lookups by default. Add a -U flag to perform them.
In single user mode lookups that involve the network might not work and they
slow down fsck.

Fix funny whitespace in a variable assignment.

branches: 1.59.4; 1.59.6;
Make sure that the exit values are always sane, and use symbolic instead
of magic constants. Reviewed by go@

branches: 1.58.4; 1.58.10; 1.58.12;
apply revision 1.31 yet again:
"make sure that we don't try to allocate negative memory when blks == 0."
or on amd64, "make sure that we don't allocate 32 GB when blks == 0."

branches: 1.57.4;
Don't use cast expressions as lvalues as newer versions of gcc warn.

64 bit inode changes

sprinkle const.

appease gcc -Wuninitialized

Remove stray " I" from message.

Dammit, get rid of the debugging abort() calls. It is no fun to have a broken
fsck program on a busted filesystem.

branches: 1.51.2;
whitespace (pointed out by wiz@)

Kill __P(), ANSIfy and WARNS=2

when allocating inodes, such as for the lost+found directory,
extend the inostat array if needed. Otherwise, inoinfo() will
return the static "unallocated" inode template, which was
getting improperly modified.
Before this fix, any time the lost+found directory got created,
fsck would set all of the inode/directory counts wrong since
suddenly unallocated inodes would turn into directory inodes

Reduce memory usage slightly.

Add ffs internal snapshots. Written by Marshall Kirk McKusick for FreeBSD.

- Not enabled by default. Needs kernel option FFS_SNAPSHOT.
- Change parameters of ffs_blkfree.
- Let the copy-on-write functions return an error so spec_strategy
may fail if the copy-on-write fails.
- Change genfs_*lock*() to use vp->v_vnlock instead of &vp->v_lock.
- Add flag B_METAONLY to VOP_BALLOC to return indirect block buffer.
- Add a function ffs_checkfreefile needed for snapshot creation.
- Add special handling of snapshot files:
Snapshots may not be opened for writing and the attributes are read-only.
Use the mtime as the time this snapshot was taken.
Deny mtime updates for snapshot files.
- Add function transferlockers to transfer any waiting processes from
one lock to another.
- Add vfsop VFS_SNAPSHOT to take a snapshot and make it accessible through
a vnode.
- Add snapshot support to ls, fsck_ffs and dump.

Welcome to 2.0F.

Approved by: Jason R. Thorpe <thorpej@netbsd.org>

fix 64bit bug in chkrange()
problem noticed by nathanw
fix from freebsd

branches: 1.45.2;
fix logic for handling symlinks in inodes when isappleufs

when doing both -c 2 and -B, swap the blocks listed in the inode
of symlinks, since the swapping happens before the symlinks
are moved into the inode.

fix paste-o in previous commit

fix bugs with unsigned comparison ofs fs_maxsymlinklen

realloc pedant

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22308, verified by myself.

use bounded string op

Swap the right blocks in an inode in the byteswapping case.

Add support for UFS2. UFS2 is an enhanced FFS, adding support for
64 bit block pointers, extended attribute storage, and a few
other things.

This commit does not yet include the code to manipulate the extended
storage (for e.g. ACLs), this will be done later.

Originally written by Kirk McKusick and Network Associates Laboratories for
FreeBSD.

Bump daddr_t to 64 bits. Replace it with int32_t in all places where
it was used on-disk, so that on-disk formats remain the same.
Remove ufs_daddr_t and ufs_lbn_t for the time being.

Add support for the Apple UFS variation on ffs
This is the bulk of PR #17345

The general approach is to use a run time deteriminable value
for DIRBLKSIZ. Additional allowances are included for using
MAXSYMLINKLEN with FS_42INODEFMT and a shift in the cylinder group
cluster summary count array. Support is added for managing
the Apple UFS volume label.

use %ll_ instead of the less standard %q_

- ARGH /brick fvdl
- Put back the change from revision 1.31

Changes for softdep code.

branches: 1.31.4;
make sure that we don't try to allocate negative memory when blks == 0.

branches: 1.30.4;
Use DINODE_SIZE rather than sizeof(struct dinode) and/or pointer arithmetic.

__AUDIT__ cleanup.

Need <time.h> for ctime() and time() prototypes.

Add support for non-native byteorder FFS, and converting byteorder.
Also, be a bit more conservative with the clean flag: don't mark the FS
clean when we know there may still be errors (user anserwed 'n' to
a question, or fsck says "you must rerun fsck").

resolve conflicts from lite-2 merge.

make these compile on the alpha after WARNS=1.

* cleanup for WARNS=1
* deprecate register
* cleanup manpage
* remove unused docheck() func
* prefix hex numbers with '0x'
* getopt returns -1 not EOF

Bug fix from Kirk McKusick <mckusick@McKusick.COM>:

Fix a bug in fsck_ffs where if a directory somehow develops a hole
(that is a block pointer that has a value of zero), fsck would give the
filesystem a clean bill of health, but the kernel would panic when
accessing the directory with the hole. Fsck now checks for holes
in directories. If found in preen mode, fsck fails. In manual
mode, it can be directed to shorten the directory to the beginning of
the hole. A more complete solution would be to allocate a block to fill
the hole. However, this is a lot more work for a `cannot happen' error,
so the extra effort seems unwarranted.

- util.h -> fsutil.h

- fixed all printf formats [there were a lot of %l? <-> %? mistakes]
- added missing prototypes, and made local functions static
- removed parallel preening code; this is part of fsck(8)
- use printing utilities from fsck(8)
- Makefile does not make links to fsck and fsck.8
- removed -l maxparallel option. It has no meaning anymore.

Use a different tree-walking algorithm in propagate().

File sizes in FFS are u_int64_t.

branches: 1.18.2;
Adjust the logic a bit so we `nif' can't overflow.

Squish some type bugs pointed out by Jonathan Stone.

Make this compile with -Werror.

typeof(timeval.tv_sec) != time_t

Make sure to recreate the `..' entry in the root directory if missing.

convert to new RCS Id conventions; reduce my headache

more cleanups from Jim Jegers, passed over by me.

Eliminate uses of some obsolete functions.

Fix compatibility with old fastlinks.

Update from 4.4-Lite, with local changes.

Prototype some things.

need <sys/time.h>

kill some code if SMALL is defined...

Add RCS identifiers.

Add support for fast symlinks.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

fsck_ffs(8): Don't assume <endian.h> is transcluded via <sys/types.h>

Prompted by PR pkg/59839

fsck_ffs(8): Fix whitespace issues.

- Nix trailing whitespace.
- Omit excessive blank lines.
- Insert missing blank lines between $NetBSD$ and copyright.

No functional change intended.

ufs: fixed signed/unsigned bugs affecting large file systems

Apply these commits from FreeBSD:

commit e870d1e6f97cc73308c11c40684b775bcfa906a2
Author: Kirk McKusick <mckusick@FreeBSD.org>
Date: Wed Feb 10 20:10:35 2010 +0000

This fix corrects a problem in the file system that treats large
inode numbers as negative rather than unsigned. For a default
(16K block) file system, this bug began to show up at a file system
size above about 16Tb.

To fully handle this problem, newfs must be updated to ensure that
it will never create a filesystem with more than 2^32 inodes. That
patch will be forthcoming soon.

Reported by: Scott Burns, John Kilburg, Bruce Evans
Followup by: Jeff Roberson
PR: 133980
MFC after: 2 weeks

commit 81479e688b0f643ffacd3f335b4b4bba460b769d
Author: Kirk McKusick <mckusick@FreeBSD.org>
Date: Thu Feb 11 18:14:53 2010 +0000

One last pass to get all the unsigned comparisons correct.


In additional to the changes from FreeBSD, this commit includes quite a few
related changes to appease -Wsign-compare.

branches: 1.90.2;
Restore backward compatibility of UFS2 with previous NetBSD releases by
disabling support in UFS2 for extended attributes (including ACLs).
Add a new variant of UFS2 called "UFS2ea" that does support extended attributes.
Add new fsck_ffs operations "-c ea" and "-c no-ea" to convert file systems
from UFS2 to UFS2ea and vice-versa (both of which delete all existing extended
attributes in the process).

Sync "common" declarations with ifdefs in header.

dion't define endian as well as try to declare it as a variable.

Fix depenency on common symbols in sbin.

fsck: Stop defining the same variable concurrently in bss and data

returntosingle was defined in multiple places:

- fsck_lfs/main.c
- fsck_ffs/main.c
- fsck_ext2fs/main.c
- fsck/fsutil.c

Keep the fsutil.c definition as the only one.

Detected during the build of telned with Address Sanitizer (MKSANITIZER).

branches: 1.85.2;
Add a -z flag to zero out the up to 4 bytes of padding in directory entry
names (including the terminating NUL), as well as directory entries with
extra free space (d->d_reclen > UFS_DIRSIZ(d)).

Inspired from FreeBSD:
https://svnweb.freebsd.org/base?view=revision&revision=347066

While the kernel has been fixed to deal with the padding bytes (new
kernels will correctly zero out all the padding after the name), it
appears that there is still an issue with directory entries with extra
free space, since a newly created and populated filesystem gets modified
with "fsck_ffs -z".

branches: 1.84.12;
Add smaller versions of fsck_ffs(8) and newfs(8) for install media, where
support for Endian-Independent FFS and Apple UFS is disabled unless FFS_EI=1
and APPLE_UFS=1 are added to CRUNCHENV, respectively.

This reduces the size of ramdisk image for atari by over 15KB.

Thanks tsutsui and christos for their useful comments.

branches: 1.83.2; 1.83.4;
a few more \n's in errors

fix error messages containing \n

Stuff UFS_ in front of a few of ufs's symbols to reduce namespace
pollution. Specifically:
ROOTINO -> UFS_ROOTINO
WINO -> UFS_WINO
NXADDR -> UFS_NXADDR
NDADDR -> UFS_NDADDR
NIADDR -> UFS_NIADDR
MAXSYMLINKLEN -> UFS_MAXSYMLINKLEN
MAXSYMLINKLEN_UFS[12] -> UFS[12]_MAXSYMLINKLEN (for consistency)

Sort out ext2fs's misuse of NDADDR and NIADDR; fortunately, these have
the same values in ext2fs and ffs.

No functional change intended.

branches: 1.80.2; 1.80.8;
Use __dead

share more code.

branches: 1.78.2;
merge the bouyer-quota2 branch. This adds a new on-disk format
to store disk quota usage and limits, integrated with ffs
metadata. Usage is checked by fsck_ffs (no more quotacheck)
and is covered by the WAPBL journal. Enabled with kernel
option QUOTA2 (added where QUOTA was enabled in kernel config files),
turned on with tunefs(8) on a per-filesystem
basis. mount_mfs(8) can also turn quotas on.

See http://mail-index.netbsd.org/tech-kern/2011/02/19/msg010025.html
for details.

Do Skip device checks, which ends up parsing fstab, when working on a
filesystem image with -F option set.

branches: 1.76.2;
Add -x option which allows to run `fsck_msdos -n' on a snapshot of
a live file system.

While here modify snap_open() to accept a character device as its
first arg and remove now unneeded get_snap_device().

Reviewed by: Manuel Bouyer <bouyer@netbsd.org>

make this compile again.

PR/42568: Pedro F. Giffuni: Better signal handling from OpenBSD, but simplified.

Don't use unicode in usage.
Noted by Anon Ymous.

Sync usage with man page.

Disable userid to username lookups by default. Add a -U flag to perform them.
In single user mode lookups that involve the network might not work and they
slow down fsck.

Make it compile. (HI BOUYER!)

Add fss(4) snapshot support to fsck_ffs(8) (via -x or -X options, like
dump(8)). This allows fsck_ffs -n to work on a snapshot of a R/W mounted
filesystem, and avoid errors related to filesystem activity.

Remove the \n and tabs from the __COPYRIGHT() strings.
(Tweak some to use a consistent format.)

branches: 1.67.4; 1.67.6;
Make sure that the exit values are always sane, and use symbolic instead
of magic constants. Reviewed by go@

branches: 1.66.4; 1.66.10; 1.66.12;
Make all mount(2) return value error checks against -1. Some file
systems just checked != 0, breaking MNT_GETARGS. Others worked with < 0,
but make them check against -1 too for consistency. And sprinkle some
stylish line wrapping where appropriate.

Add additional 'sizeof args' parameter to mount(2).

include <signal.h> where signal(3) is used

Remove more duplicate #includes,
from Slava Semushin <slava.semushin@gmail.com>

branches: 1.62.2;
Fix error in previous. Pass 1 needs to use progress_setrange(),
not progress_sethighlim().

Improve the progress bar displayed by fsck_ffs -p -P.

The progress bar will now travel from 0 to 20% during pass 1, from 20%
to 95% during pass 2, and from 95% to 100% during passes 3, 4, and 5.
Previously, the progress bar was not displayed at all during pass 1,
slowly traveled from 0 to about 50% during pass 2, and then very rapidly
traveled from 50% to 100% during passes 3, 4, and 5.

Note that fsck_ffs is the only user of fsck/progress.{c,h}.

Apply the NFS exports list rototill patch:

- Remove all NFS related stuff from file system specific code.
- Drop the vfs_checkexp hook and generalize it in the new nfs_check_export
function, thus removing redundancy from all file systems.
- Move all NFS export-related stuff from kern/vfs_subr.c to the new
file sys/nfs/nfs_export.c. The former was becoming large and its code
is always compiled, regardless of the build options. Using the latter,
the code is only compiled in when NFSSERVER is enabled. While doing this,
also make some functions in nfs_subs.c conditional to NFSSERVER.
- Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a
path and a set of export entries. At the moment it can only clear the
exports list or append entries, one by one, but it is done in a way that
allows setting the whole set of entries atomically in the future (see the
comment in mountd_set_exports_list or in doc/TODO).
- Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so
that it becomes file system agnostic. In fact, all this whole thing was
done to remove a 'XXX' block from this utility!
- Change the mount*, newfs and fsck* userland utilities to not deal with NFS
exports initialization; done internally by the kernel when initializing
the NFS support for each file system.
- Implement an interface for VFS (called VFS hooks) so that several kernel
subsystems can run arbitrary code upon receipt of specific VFS events.
At the moment, this only provides support for unmount and is used to
destroy NFS exports lists from the file systems being unmounted, though it
has room for extension.

Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments
and advice in the development of this patch.

64 bit inode changes

sprinkle const.

branches: 1.57.2;
whitespace (pointed out by wiz@)

Kill __P(), ANSIfy and WARNS=2

Add -P to usage.

Use -DPROGRESS to enable the progress bar, don't depend on SMALL.

Add a progress meter to fsck_ffs based on the work by thorpej presented
to the mailing lists last January. This is optional.

always print warning and return non-zero exit when there are unresolved inconsistencies.

Add -a to usage. Closes PR 25916 by Kouichirou Hiratsuka.

Replace the statfs() family of system calls with statvfs().
Retain binary compatibility.

print warning if the user specifies a conversion level greater than implemented

Homogenize usage messages: make the 'usage' word all lowercase, as this seems
to be the most common practice in our tree.

Add a -q (quiet) option to print nothing for clean filesystems.
Support in fsck_ffs and stub in fsck_xxx.
Push a few more messages through pwarn() instead of printf() to ensure
disk name is shown.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22308, verified by myself.

Add support for UFS2. UFS2 is an enhanced FFS, adding support for
64 bit block pointers, extended attribute storage, and a few
other things.

This commit does not yet include the code to manipulate the extended
storage (for e.g. ACLs), this will be done later.

Originally written by Kirk McKusick and Network Associates Laboratories for
FreeBSD.

make this build on alpha after daddr_t->64bit

Bump daddr_t to 64 bits. Replace it with int32_t in all places where
it was used on-disk, so that on-disk formats remain the same.
Remove ufs_daddr_t and ufs_lbn_t for the time being.

Add support for the Apple UFS variation on ffs
This is the bulk of PR #17345

The general approach is to use a run time deteriminable value
for DIRBLKSIZ. Additional allowances are included for using
MAXSYMLINKLEN with FS_42INODEFMT and a shift in the cylinder group
cluster summary count array. Support is added for managing
the Apple UFS volume label.

If SIGINFO is received, display summary information to stderr.
Based on changes in FreeBSD, via Chuck Cranor <chuck@research.att.com>

- implement -F; treat provided filesystems as images in regular files
- replace "filesystem" with "file system" as appropriate
- grammar fixes

fix compile errors.

convert to use getprogname()

remove redundant declarations

In pass 5, check alternate superblocks for consistency with
the current in-core master superblock, and fix them up if
they're incorrect. Move the code that writes the alternate
superblocks if (cvtlevel || doswap) into pass 5 for efficiency.

Reviewd by Charles Hannum, and used by me to fix up a curdled
file system.

Fix an annoyingly incorrect message.

branches: 1.34.4;
Changes for softdep code.

branches: 1.33.4; 1.33.8;
const poisoning.

Need <time.h> for time() prototype.

Add support for non-native byteorder FFS, and converting byteorder.
Also, be a bit more conservative with the clean flag: don't mark the FS
clean when we know there may still be errors (user anserwed 'n' to
a question, or fsck says "you must rerun fsck").

Merge with Lite2 + local changes

Hmmm..fsck_ffs has to really be able to alwasy work, and if you've got
a *huge* (like 600GB) filesystem, you need to not be limited by resource
limits.

branches: 1.28.2;
Don't declare optind and optarg.

for now, #ifdef out a couple of chunks that were added in the lite2 merge

- don't indiscriminately include <stdlib.h> and <unistd.h> in "fsck.h"
- explicitly pull in <stdio.h>, <stdlib.h> and <unistd.h> in *.c as necessary

resolve conflicts from lite-2 merge.

* cleanup for WARNS=1
* deprecate register
* cleanup manpage
* remove unused docheck() func
* prefix hex numbers with '0x'
* getopt returns -1 not EOF

- sort and remove unused options.
- If no filesystem is specified or an incorrect option, print usage information

Bug fix from Kirk McKusick <mckusick@McKusick.COM>:

Fix a bug in fsck_ffs where if a directory somehow develops a hole
(that is a block pointer that has a value of zero), fsck would give the
filesystem a clean bill of health, but the kernel would panic when
accessing the directory with the hole. Fsck now checks for holes
in directories. If found in preen mode, fsck fails. In manual
mode, it can be directed to shorten the directory to the beginning of
the hole. A more complete solution would be to allocate a block to fill
the hole. However, this is a lot more work for a `cannot happen' error,
so the extra effort seems unwarranted.

- util.h -> fsutil.h

- fixed all printf formats [there were a lot of %l? <-> %? mistakes]
- added missing prototypes, and made local functions static
- removed parallel preening code; this is part of fsck(8)
- use printing utilities from fsck(8)
- Makefile does not make links to fsck and fsck.8
- removed -l maxparallel option. It has no meaning anymore.

branches: 1.19.4;
merge in changes from 1.1 release branch

branches: 1.18.2;
recognize 'ffs' and 'ufs' as meaning the same thing.

implement a 'force check' flag, '-f'. I used the SunOS name, but the Digital
semantics. now:
(1) dirty file systems will always be checked; nothing new there.
(2) if not '-f' clean file systems will _NEVER_ be checked,
i.e. they won't be checked even if -p isn't specified. This
allows one to 'fsck -p ; fsck' to preen, then clean up
anything that 'fsck -p' barfs on, without waiting for the
clean file systems to be checked again.
(3) if '-f' clean file systems will ALWAYS be checked. This
allows people to put 'fsck -fp' into /etc/rc on systems
where they're leery of the FS clean flag state, need
the extra reliability, and can afford time 'wasted'
in checks.
The assumption made here is that if a file system is marked clean, it
_IS CLEAN_, really, and shouldn't be checked unless fsck is explicitly
told to (with -f). This should be a valid assumption, but may not be in
the presence of file system bugs. Documentation updated to note '-f'.

Set the clean flag if necessary. If preening, don't check `clean' file
systems.

convert to new RCS Id conventions; reduce my headache

more cleanups from Jim Jegers, passed over by me.

Update from 4.4-Lite, with local changes.

need <sys/time.h>

kill the evil U word! from Mike Long <mike.long@analog.com>

try again to keep it from rebooting endlessly

set the 'clean' flag sometimes, but ignore it when deciding to clean
or not. this allows you to use this fsck and... reboot your system;
otherwise it would inf-loop rebooting and marking the FS clean,
which would then mark the fs modified, cause a reboot, etc.

#if 0 the fs_clean check for now.

Don't set clean flag if -n.

Initial count for fs_state is `FS_CLEANFREQ'.

Skip check if filesystem is marked clean and isn't too dusty, only with -p.
Set clean flag after checking a filesystem.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

fsck_ffs(8): Fix whitespace issues.

- Nix trailing whitespace.
- Omit excessive blank lines.
- Insert missing blank lines between $NetBSD$ and copyright.

No functional change intended.

ufs: fixed signed/unsigned bugs affecting large file systems

Apply these commits from FreeBSD:

commit e870d1e6f97cc73308c11c40684b775bcfa906a2
Author: Kirk McKusick <mckusick@FreeBSD.org>
Date: Wed Feb 10 20:10:35 2010 +0000

This fix corrects a problem in the file system that treats large
inode numbers as negative rather than unsigned. For a default
(16K block) file system, this bug began to show up at a file system
size above about 16Tb.

To fully handle this problem, newfs must be updated to ensure that
it will never create a filesystem with more than 2^32 inodes. That
patch will be forthcoming soon.

Reported by: Scott Burns, John Kilburg, Bruce Evans
Followup by: Jeff Roberson
PR: 133980
MFC after: 2 weeks

commit 81479e688b0f643ffacd3f335b4b4bba460b769d
Author: Kirk McKusick <mckusick@FreeBSD.org>
Date: Thu Feb 11 18:14:53 2010 +0000

One last pass to get all the unsigned comparisons correct.


In additional to the changes from FreeBSD, this commit includes quite a few
related changes to appease -Wsign-compare.

branches: 1.62.2;
Fix clearing of permissions when finding unexpected extended attributs
for swapped endian file systems. Ok: chs

Make the "non-zero exattr fields" message show the inode number.

Restore backward compatibility of UFS2 with previous NetBSD releases by
disabling support in UFS2 for extended attributes (including ACLs).
Add a new variant of UFS2 called "UFS2ea" that does support extended attributes.
Add new fsck_ffs operations "-c ea" and "-c no-ea" to convert file systems
from UFS2 to UFS2ea and vice-versa (both of which delete all existing extended
attributes in the process).

Enable the code to clean the extattr blocks

branches: 1.58.4;
Treat an inode with "mode == 0" and "blocks != 0" as partially allocated
and clear it as ffs_newvnode() tests for "blocks == 0".

branches: 1.57.4;
Add smaller versions of fsck_ffs(8) and newfs(8) for install media, where
support for Endian-Independent FFS and Apple UFS is disabled unless FFS_EI=1
and APPLE_UFS=1 are added to CRUNCHENV, respectively.

This reduces the size of ramdisk image for atari by over 15KB.

Thanks tsutsui and christos for their useful comments.

branches: 1.56.8; 1.56.12;
fix unused variable warnings.

Stick ffs_, ext2_, chfs_, filecore_, cd9660_, or mfs_ in front of
the following symbols so as to disambiguate fully. (Christos already
did the lfs ones.)

lblkno
lblktosize
lfragtosize
numfrags
blkroundup
fragroundup

fsbtodb() -> FFS_FSBTODB(), EXT2_FSBTODB(), or MFS_FSBTODB()
dbtofsb() -> FFS_DBTOFSB() or EXT2_DBTOFSB()

(Christos already did the lfs ones a few days back)

blkoff() -> ffs_blkoff() stragglers

Rename ambiguous macros:
MAXDIRSIZE -> UFS_MAXDIRSIZE or LFS_MAXDIRSIZE
NINDIR -> FFS_NINDIR, EXT2_NINDIR, LFS_NINDIR, or MFS_NINDIR
INOPB -> FFS_INOPB, LFS_INOPB
INOPF -> FFS_INOPF, LFS_INOPF
blksize -> ffs_blksize, ext2_blksize, or lfs_blksize
sblksize -> ffs_blksize

These are not the only ambiguously defined filesystem macros, of
course, there's a pile more. I may not have found all the ambiguous
definitions of blksize(), too, as there are a lot of other things
called 'blksize' in the system.

Stuff UFS_ in front of a few of ufs's symbols to reduce namespace
pollution. Specifically:
ROOTINO -> UFS_ROOTINO
WINO -> UFS_WINO
NXADDR -> UFS_NXADDR
NDADDR -> UFS_NDADDR
NIADDR -> UFS_NIADDR
MAXSYMLINKLEN -> UFS_MAXSYMLINKLEN
MAXSYMLINKLEN_UFS[12] -> UFS[12]_MAXSYMLINKLEN (for consistency)

Sort out ext2fs's misuse of NDADDR and NIADDR; fortunately, these have
the same values in ext2fs and ffs.

No functional change intended.

Show pass1 SIGINFO output on stderr like other passes, not on stdout.

branches: 1.49.2; 1.49.4; 1.49.8; 1.49.10;
WARNS=4

share more code.

branches: 1.47.2;
merge the bouyer-quota2 branch. This adds a new on-disk format
to store disk quota usage and limits, integrated with ffs
metadata. Usage is checked by fsck_ffs (no more quotacheck)
and is covered by the WAPBL journal. Enabled with kernel
option QUOTA2 (added where QUOTA was enabled in kernel config files),
turned on with tunefs(8) on a per-filesystem
basis. mount_mfs(8) can also turn quotas on.

See http://mail-index.netbsd.org/tech-kern/2011/02/19/msg010025.html
for details.

branches: 1.46.2; 1.46.6; 1.46.10; 1.46.14; 1.46.16;
reallocate the correct amount. From Anon Ymous

Avoid allocating 2 x inospace and use realloc instead. This also saves a
large memcpy.

Make sure that the exit values are always sane, and use symbolic instead
of magic constants. Reviewed by go@

branches: 1.43.8; 1.43.16; 1.43.18;
Improve the progress bar displayed by fsck_ffs -p -P.

The progress bar will now travel from 0 to 20% during pass 1, from 20%
to 95% during pass 2, and from 95% to 100% during passes 3, 4, and 5.
Previously, the progress bar was not displayed at all during pass 1,
slowly traveled from 0 to about 50% during pass 2, and then very rapidly
traveled from 50% to 100% during passes 3, 4, and 5.

Note that fsck_ffs is the only user of fsck/progress.{c,h}.

comment out impossible code.

Don't use cast expressions as lvalues as newer versions of gcc warn.

ndb is daddr_t which is 64 bits, by truncating it to j which is an int, it
can get negative causing segmentation faults. Check against this.

64 bit inode changes

add code to handle inodes with extended attribute blocks
this is currently turned off with "#ifdef notyet" since the
kernel will not correctly clean up extended attribute blocks on truncation

Originally written by Kirk McKusick and Network Associates Laboratories for
FreeBSD.

Dammit, get rid of the debugging abort() calls. It is no fun to have a broken
fsck program on a busted filesystem.

branches: 1.36.2;
whitespace (pointed out by wiz@)

Kill __P(), ANSIfy and WARNS=2

Use -DPROGRESS to enable the progress bar, don't depend on SMALL.

Add a progress meter to fsck_ffs based on the work by thorpej presented
to the mailing lists last January. This is optional.

Add ffs internal snapshots. Written by Marshall Kirk McKusick for FreeBSD.

- Not enabled by default. Needs kernel option FFS_SNAPSHOT.
- Change parameters of ffs_blkfree.
- Let the copy-on-write functions return an error so spec_strategy
may fail if the copy-on-write fails.
- Change genfs_*lock*() to use vp->v_vnlock instead of &vp->v_lock.
- Add flag B_METAONLY to VOP_BALLOC to return indirect block buffer.
- Add a function ffs_checkfreefile needed for snapshot creation.
- Add special handling of snapshot files:
Snapshots may not be opened for writing and the attributes are read-only.
Use the mtime as the time this snapshot was taken.
Deny mtime updates for snapshot files.
- Add function transferlockers to transfer any waiting processes from
one lock to another.
- Add vfsop VFS_SNAPSHOT to take a snapshot and make it accessible through
a vnode.
- Add snapshot support to ls, fsck_ffs and dump.

Welcome to 2.0F.

Approved by: Jason R. Thorpe <thorpej@netbsd.org>

branches: 1.31.4;
fix bugs with unsigned comparison ofs fs_maxsymlinklen

increase size of buffer used for updating symlinks with -c 1
otherwise, the block read will blow the stack

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22308, verified by myself.

On LP64 hosts, iswap64() result is "long int", so cast result to (long long)
before printing with %lld.

Add support for UFS2. UFS2 is an enhanced FFS, adding support for
64 bit block pointers, extended attribute storage, and a few
other things.

This commit does not yet include the code to manipulate the extended
storage (for e.g. ACLs), this will be done later.

Originally written by Kirk McKusick and Network Associates Laboratories for
FreeBSD.

Bump daddr_t to 64 bits. Replace it with int32_t in all places where
it was used on-disk, so that on-disk formats remain the same.
Remove ufs_daddr_t and ufs_lbn_t for the time being.

Add support for the Apple UFS variation on ffs
This is the bulk of PR #17345

The general approach is to use a run time deteriminable value
for DIRBLKSIZ. Additional allowances are included for using
MAXSYMLINKLEN with FS_42INODEFMT and a shift in the cylinder group
cluster summary count array. Support is added for managing
the Apple UFS volume label.

If SIGINFO is received, display summary information to stderr.
Based on changes in FreeBSD, via Chuck Cranor <chuck@research.att.com>

use %ll_ instead of the less standard %q_

Changes for softdep code.

branches: 1.21.4; 1.21.8;
Add support for non-native byteorder FFS, and converting byteorder.
Also, be a bit more conservative with the clean flag: don't mark the FS
clean when we know there may still be errors (user anserwed 'n' to
a question, or fsck says "you must rerun fsck").

- don't indiscriminately include <stdlib.h> and <unistd.h> in "fsck.h"
- explicitly pull in <stdio.h>, <stdlib.h> and <unistd.h> in *.c as necessary

resolve conflicts from lite-2 merge.

make these compile on the alpha after WARNS=1.

* cleanup for WARNS=1
* deprecate register
* cleanup manpage
* remove unused docheck() func
* prefix hex numbers with '0x'
* getopt returns -1 not EOF

- util.h -> fsutil.h

- fixed all printf formats [there were a lot of %l? <-> %? mistakes]
- added missing prototypes, and made local functions static
- removed parallel preening code; this is part of fsck(8)
- use printing utilities from fsck(8)
- Makefile does not make links to fsck and fsck.8
- removed -l maxparallel option. It has no meaning anymore.

Use fs_csaddr to find the cg summary area. From der Mouse, PR 1366.

convert to new RCS Id conventions; reduce my headache

more cleanups from Jim Jegers, passed over by me.

Remove some more uses of obsolete functions.

Eliminate uses of some obsolete functions.

Reads on raw disks are only guarranteed in multiples of the block size

Fix compatibility with old fastlinks.

Update from 4.4-Lite, with local changes.

need <sys/time.h>

Add RCS identifiers.

Add support for fast symlinks.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

ufs: fixed signed/unsigned bugs affecting large file systems

Apply these commits from FreeBSD:

commit e870d1e6f97cc73308c11c40684b775bcfa906a2
Author: Kirk McKusick <mckusick@FreeBSD.org>
Date: Wed Feb 10 20:10:35 2010 +0000

This fix corrects a problem in the file system that treats large
inode numbers as negative rather than unsigned. For a default
(16K block) file system, this bug began to show up at a file system
size above about 16Tb.

To fully handle this problem, newfs must be updated to ensure that
it will never create a filesystem with more than 2^32 inodes. That
patch will be forthcoming soon.

Reported by: Scott Burns, John Kilburg, Bruce Evans
Followup by: Jeff Roberson
PR: 133980
MFC after: 2 weeks

commit 81479e688b0f643ffacd3f335b4b4bba460b769d
Author: Kirk McKusick <mckusick@FreeBSD.org>
Date: Thu Feb 11 18:14:53 2010 +0000

One last pass to get all the unsigned comparisons correct.


In additional to the changes from FreeBSD, this commit includes quite a few
related changes to appease -Wsign-compare.

branches: 1.23.40;
Stuff UFS_ in front of a few of ufs's symbols to reduce namespace
pollution. Specifically:
ROOTINO -> UFS_ROOTINO
WINO -> UFS_WINO
NXADDR -> UFS_NXADDR
NDADDR -> UFS_NDADDR
NIADDR -> UFS_NIADDR
MAXSYMLINKLEN -> UFS_MAXSYMLINKLEN
MAXSYMLINKLEN_UFS[12] -> UFS[12]_MAXSYMLINKLEN (for consistency)

Sort out ext2fs's misuse of NDADDR and NIADDR; fortunately, these have
the same values in ext2fs and ffs.

No functional change intended.

branches: 1.22.4; 1.22.10;
merge the bouyer-quota2 branch. This adds a new on-disk format
to store disk quota usage and limits, integrated with ffs
metadata. Usage is checked by fsck_ffs (no more quotacheck)
and is covered by the WAPBL journal. Enabled with kernel
option QUOTA2 (added where QUOTA was enabled in kernel config files),
turned on with tunefs(8) on a per-filesystem
basis. mount_mfs(8) can also turn quotas on.

See http://mail-index.netbsd.org/tech-kern/2011/02/19/msg010025.html
for details.

branches: 1.21.48;
whitespace (pointed out by wiz@)

Kill __P(), ANSIfy and WARNS=2

Use -DPROGRESS to enable the progress bar, don't depend on SMALL.

Add a progress meter to fsck_ffs based on the work by thorpej presented
to the mailing lists last January. This is optional.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22308, verified by myself.

Add support for UFS2. UFS2 is an enhanced FFS, adding support for
64 bit block pointers, extended attribute storage, and a few
other things.

This commit does not yet include the code to manipulate the extended
storage (for e.g. ACLs), this will be done later.

Originally written by Kirk McKusick and Network Associates Laboratories for
FreeBSD.

Bump daddr_t to 64 bits. Replace it with int32_t in all places where
it was used on-disk, so that on-disk formats remain the same.
Remove ufs_daddr_t and ufs_lbn_t for the time being.

If SIGINFO is received, display summary information to stderr.
Based on changes in FreeBSD, via Chuck Cranor <chuck@research.att.com>

Add support for non-native byteorder FFS, and converting byteorder.
Also, be a bit more conservative with the clean flag: don't mark the FS
clean when we know there may still be errors (user anserwed 'n' to
a question, or fsck says "you must rerun fsck").

resolve conflicts from lite-2 merge.

* cleanup for WARNS=1
* deprecate register
* cleanup manpage
* remove unused docheck() func
* prefix hex numbers with '0x'
* getopt returns -1 not EOF

- fixed all printf formats [there were a lot of %l? <-> %? mistakes]
- added missing prototypes, and made local functions static
- removed parallel preening code; this is part of fsck(8)
- use printing utilities from fsck(8)
- Makefile does not make links to fsck and fsck.8
- removed -l maxparallel option. It has no meaning anymore.

convert to new RCS Id conventions; reduce my headache

more cleanups from Jim Jegers, passed over by me.