needs (looking at you sgi, but others required this before). This is for
the circumstances we need pagesize known at compile time, not getpagesize()
runtime. Use it for malloc storage sizes, for shm, and to set pthread stack
default sizes. The stack sizes were a mess, and pushing them towards
page-aligned is healthy move (which will also be needed by the coming
stack register checker)
ok guenther kettenis, discussion with stefan
are signalled to shutdown cleanly. Wait for each to finish to avoid too much
busy work at once; this may need revising if it turns out to be too slow with
a larger number of VMs (e.g. signal/delay/signal/delay/... then wait for
shutdowns), but let's avoid making it more complex unless we know it's needed.
Based on a diff from abieber@, discussed with mlarkin@ aja@ rpe@, ok rpe
to skip backup or temp files.
- test if the patterns matched actual files
- warn if ifcreate() fails on an interface and continue with the
subsequent interfaces in the list instead of return'ing
OK dlg sthen tb
These PATH helpers failed to quote their input properly leading to shell
code execution. Noone noticed since import (over 21 years ago), so wipe it.
OK tb rpe
- In ifcreate() use the exit code of the {} block directly
- In vifscreate(), use the ifconfig -C output directly in the for _vif loop
- Remove superfluous and somewhat confusing comment
OK dlg kn sthen
hw.disknames. This can only happen due to a failure or user error.
In either case, silent failure makes it hard to discover and debug.
Now it will be easy to spot in the daily mail.
ok rob, schwarze
- Start with a full page of struct region_info's
- Save an mprotect in the init code: allocate 3 pages with none and
make the middle page r/w instead of a r/w allocation and two calls to make the
guard pages none
The existing example was written over a decade ago, when carp(4) was
not supporting carpdemote and did not track it's interfaces link state itself.
OK jmc@, rob@
but at that time ffs function calls were generated instead of the
compiler inlining the code. Now that ffs is marked protected in
libc this is handled better. Thanks to kshe who prompted me to
look at this again.
dance, mark it protected. This works better for both gcc and clang: gcc
blocks overriding of internal calls, while clang permits inlining again.
ok otto@
- change -a, -o to &&, || inside [[]]
- remove unecessary quoting inside [[]]
- remove X"" constructs inside [[]]
- remove \ (line continuation) in case of &&, || and pipes
- replace backticks with $()
discussed with and OK aja@
OK tb
pledge for a new execve image immediately upon start. Also introduces
"error" which makes violations return -1 ENOSYS instead of killing the
program ("error" may not be handed to a setuid/setgid program, which
may be missing/ignoring syscall return values and would continue with
inconsistant state)
Discussion with many
florian has used this to improve the strictness of a daemon
PF ruleset. This is not a complete block on _pbuild being able to communicate
(e.g. non-TCP/UDP protocols don't have a PCB with userid, so PF can't restrict
in those cases) but avoids some cases, and in particular makes it more obvious
when a port does things like download extra distfiles or dependencies
as part of the build process. Slight tweak from a diff by espie@.
to builtins like mem{set,cpy,move} and __stack_smash_handler. So, when
building with clang, instead mark those as protected visibility to get rid
of the PLT relocations. We can't take the address of them then, but that's
ok: it's a build-time error not a run-time error.
ok kettenis@
deraadt