Add clockintr_advance_random(). Its sole purpose is to simplify the
implementation of statclock's pseudorandom period.
Ideally, nothing else will use it and we will get rid of it and the
pseudorandom statclock period at some point in the near future.
Suggested by mpi@.
Thread: https://marc.info/?l=openbsd-tech&m=169392340028978&w=2
ok mpi@ mlarkin@
The api changed to handle cookies and extended error codes.
Passing in NULL for config_file disables cookie handling so we do not
need to pass a valid comm_reply, NULL will do.
There is no need for a helper function to obfuscate lh_ADDED_OBJ_new().
Just call the real thing directly.
Adding an object with a NID of NID_undef basically amounts to disabling
a built-in OID. It does so in an incoherent fashion and the caller can't
easily tell success from failure of the operation. Arguably the result is
a corrupted objects table.
Let's not allow adding such an object in an attempt at keeping things
slightly more coherent.
Issue noted and initial diff by schwarze while writing documentation
ok schwarze
* Integrate the leftovers of the former NOTES section into the main text,
resulting in a more logical order of information.
* Make many descriptions more precise and tweak many wordings.
For example, the description of OBJ_cmp(3) was totally misleading.
Add a CAVEATS section explaining the scary ownership contracts
of the functions returning ASN1_OBJECT pointers.
Move the discussion of NID_undef to the BUGS section because the
statement "objects which are not in the table have the NID value
NID_undef" was misleading in more than one way.
Considering that an API as fundamental as this one contains such a
gigantic amount of quirks and traps and gaps makes me shudder.
Hash them all into the Multicast Table Array. This matches what
FreeBSD does.
This fixes a problem where the I217-LM would fail to receive packets
for some of the programmed addresses.
Approach agreed by jmatthew@ dlg@
OK claudio@
a page undergoing copy-on-write faulting. We fixed the case 1b
version in rev 1.125 (2022-02-01), but missed this other path.
jsg@ noted that in NetBSD Chuck Silvers had a relevant commit, their
rev 1.234 (2023-08-13), which looks like it fixed both cases due
to their refactoring of common code into a uvmfault_promote()
function.
ok mpi@ jca@
TCP syn cache variable scs_use is basically counting packet insertions
into syn cache. Prefer type long to exclude overflow on fast
machines. Due to counting downwards from a limit, it can become
negative. Copy it out as tcps_sc_uses_left via sysctl, and print
it as signed long long integer.
OK mvs@
This disables the helpers that recognize compressed archives when the
user failed to use the proper flag/command. Those are not terribly
useful on the ramdisks and the fallback behavior is sane.
Went through a make release Just In Case(tm).
Spotted by caspar@, ok millert@ sthen@ caspar@
-DSMALL has never been used to build the ramdisks, thus the support for
reading pax format archives has always been there. This is misleading,
so just zap the ifdef since we want to keep read support.
Went through a make release Just In Case(tm).
Spotted by caspar@, ok millert@ sthen@ caspar@
The single thread API needs to lock the process to enter single thread
mode and does not need to stop the scheduler.
This code changes ps_singlecount from a count down to zero to ps_singlecnt
which counts up until equal to ps_threadcnt (in which case all threads
are properly asleep).
Tested by phessler@, OK mpi@ cheloha@
in the standard man(1) mode that formats a single resulting page
if the respective manpath contained digits, like X11R6 does.
Fortunately, this bug did not trigger for any Xenocara manual page.
cheloha