Permission bits for symlinks are taken straight from `lstat()`. However, the
actual bits presented to the user are filesystem/kernel specific. For example,
Linux with ext2/3/4 will use 0777 for symlinks, whereas NetBSD/FFS will
show 0755. As far as `makefs` is in the loop, the target filesystem will likely
be FFS, so use 0755.
`makefs` inserts nodes into its internal data structures in the order as
returned by `readdir()` calls. As this is unpredictable, sort entries by
name before creating the target filesystem.
This is done by first converting the (per-directory) linked list into
a plain array, sort it, finally re-link the list. Special case for the
sorting function: The "." directory entry seems to be ment to be always
at the front, so always check that first.
While here, re-enable lint in those cases where lint was skipped due to
a bug in interpreting abstract types, which was fixed in cgram.y 1.469
from 2023-08-02.
- base32 contains (when MKCOMPAT=yes) shared libraries for 32-bit
compatibility, previously included in base
- debug32 contains (when MKCOMPAT=yes) debug symbols and static libraries
containing debug symbols for 32-bit compatiblity, previously included
in debug
- manhtml contains (when MKHTML=yes) the HTML files previously included
in 'man', which are of limited utility without third-party software.
The motivation for this change is to be able to easily exclude sets
from CD-ROM images that go over the size limit without xz compression
(which many NetBSD platforms struggle to extract at acceptable speeds).
1. For tools that use elftoolchain: always use elftoolchain's
elfdefinitions.h. Don't even think about looking at the host's
sys/exec_elf.h, which makes no sense and should never happen.
(ELF tools that don't use elftoolchain, like m68k-elf2coff,
continue to use nbincludes/sys/exec_elf.h. But no more nbincludes
hacks in elftoolchain.)
2. For kernel components (solaris, zfs, dtrace): always use
sys/exec_elf.h, even in Solaris components via sys/elf.h.
elfdefinitions.h is not wired up in the kernel build at all.
3. For most userland components that involve libelf: use
elfdefinitions.h via libelf header files (libelf.h, gelf.h).
libdtrace in particular requires _all_ R_* reloc type definitions,
but sys/exec_elf.h brings in only the _current machine's_ R_*
reloc type definitions. (While here: Use uintptr_t instead of
Elf_Addr for pointer-to-integer cast, since Elf_Addr is MD and
provided only by sys/exec_elf.h, not by elfdefinitions.h.)
And most userland components using libelf don't rely on any
properties of the current machine from sys/exec_elf.h, so they can
use libelf's elfdefinition.h.
Exceptions:
- dtrace drti.c relies on link.h -> link_elf.h -> sys/exec_elf.h,
but it also relies on sys/dtrace.h -> sys/elf.h ->
elfdefinitions.h like other userland components using sys/elf.h.
- kdump-ioctl.c uses sys/exec_elf.h directly and sys/dtrace.h ->
sys/elf.h -> elfdefinitions like other userland components using
sys/elf.h.
- t_ptrace_wait.c (via t_ptrace_core_wait.h) uses libelf to parse
core files, but relies on sys/exec_elf.h for struct
netbsd_elfcore_procinfo.
None of these exceptions needs all R_* reloc type definitions, so
as a workaround, we can just suppress libelf's elfdefinitions.h by
defining _SYS_ELFDEFINITIONS_H_ and use sys/exec_elf.h in these
exceptions.
And undo the whole BUILTIN_ELF_HEADERS mistake. This was:
- half bogus workarounds for missing build_install dependencies in
tools/Makefile, which are no longer missing now, and
- half futile attempt to use src/sys/sys/exec_elf.h via nbincludes in
tools involving libelf instead of libelf's elfdefinitions.h, which
collides.
Longer-term, we may wish to unify sys/exec_elf.h and libelf's
elfdefinitions.h, so we don't have to play these games.
But at least now the games are limited to three .c files (one of
which is generated by Makefile.ioctl-c), rather than haphazardly
applied tree-wide by monstrous kludges in widely used .h files with
broken hackarounds to get the tools build lurching to completion.
Looks like the nfs server does enforce that after all, in spite of
the rather oblique commentary in the BUGS section about export
options being tied to local mount point options with which they must
be noncontradictory.
And there's no reason in principle it shouldn't enforce this -- it
just need to block various file system _operations_, rather than the
subtree issue where the criteria for evaluating whether operations
are allowed on particular _file handles_ are too painful to
contemplate.
PR misc/58063
The exports(5) man page is full of walls of turgid prose that should
be itemized lists with syntax templates, and I'm itching to rewrite
it, but let's get the security-relevant warnings out of the way
first.
PR misc/58063
keep track of the highest usbN value, not the count of busses.
fixes a problem where you do "drvctl -d usb0" (or a parent), and then
"usbdevs" won't print info on the last bus.
At least one user reported thinking that the previous wording implied
that usernames required padding to 8 characters.
This syncs the english translation with the german.
- use proper local variables instead of adding _ or other prefixes.
- centralize rm use
- use grep -q instead of > /dev/null
- reduce constant duplication
According to POSIX 2018, the syntax between `then' and `elif' and
`fi' must be a _non-empty_ list of commands:
compound_list : linebreak term
| linebreak term separator
;
...
if_clause : If compound_list Then compound_list else_part Fi
| If compound_list Then compound_list Fi
;
else_part : Elif compound_list Then compound_list
| Elif compound_list Then compound_list else_part
| Else compound_list
;
https://pubs.opengroup.org/onlinepubs/9699919799/utilities/V3_chap02.html#tag_18_10_02
NetBSD's sh(1) currently doesn't enforce this and allows an empty
sequence of commands, but let's not rely on that nonstandard quirk.
Noted in PR 57997.
Saves some time running subprocesses. Since this is only used for
non-directories (i.e., there's never trailing / on the inputs), it
suffices to delete the longest prefix matching glob `*/' with shell
parameter expansion -- much cheaper than spawning a subprocess.
Shaves off about 1/3 of the time spent in `certctl list' on an
aarch64 VM in qemu.
PR bin/57993
"path + cur->name" is not same as "cur->root + cur->path + cur->name"
for extra-directory files, as extra-directory files are
in different location. Do what makefs ffs code does.
reproducible.
When we specify a label, the volume label is used to create the root dir
and it gets stamped by the timestamp. We were trying to set the timestamp
in the fs-specific part of the options parsing which did not work since
the method was called only for fs_specific options. Move setting of the
timestamp just before we create the fs, where we reconcile the rest of
the options.
ignore differences in the install target flag - the backend might have
flipped it off already to ensure only a single partition is marked
as install target.
christos