Lock as writer on establish/disestablish and as reader on list traverse.
For exechook ride "exec_lock" as it is already take as reader when
traversing the list. Add local locks for exithook and forkhook.
Move exec_init before signal_init as signal_init calls exechook_establish()
that needs "exec_lock".
PR kern/39913 "exec, fork, exit hooks need locking"
netinet6/nd6 from the kernel. Now, the minimal compat code can
be successfully loaded and unloaded along with the rest of the
COMPAT_90 code.
XXX pullup-10 - hopefully before RC2
be used VERY early in boot; such consumers statically allocate the vmem
arena and boundary tags, and then explicitly add those static, private
boundary tags to the arena tag free list using the new function vmem_add_bts().
Vmem arenas that use private boundary tags will NOT consume the statically
allocated bootstrap tags used by the vmem system itself; the assumption is
that the consumer of such an arena knows what they're doing, and is responsible
for all necessary resource management. A macro, VMEM_EST_BTCOUNT(), is
provided to help such consumers size the static boundary tag store based
on the expected number of spans and early allocations. Once the private
tags are exhausted, the arena will dynamically allocate tags as usual.
from an arena. This is just a convenience wrapper around vmem_xalloc(),
that's just a bit more obvious how to use and performs some additional
sanity checks.
From OpenBSD:
commit 7b4d35e0a60ba1dd4daf4b1c2932020a22463a89
Author: bluhm <bluhm@openbsd.org>
Date: Fri Oct 20 16:25:15 2023 +0000
Avoid assertion failure when splitting mbuf cluster.
m_split() calls m_align() to initialize the data pointer of newly
allocated mbuf. If the new mbuf will be converted to a cluster,
this is not necessary. If additionally the new mbuf is larger than
MLEN, this can lead to a panic.
Only call m_align() when a valid m_data is needed. This is the
case if we do not refecence the existing cluster, but memcpy() the
data into the new mbuf.
Reported-by: syzbot+0e6817f5877926f0e96a@syzkaller.appspotmail.com
OK claudio@ deraadt@
The issue is harmless if DIAGNOSTIC is not enabled.
XXX pullup-10
XXX pullup-9
kpause(9) forbids zero.
Local workaround for wider problem in PR kern/57718, to address
immediate symptom of crash on any system with hz=50, e.g. alpha in
qemu:
panic: kernel diagnostic assertion "timo != 0 || intr" failed: file "/usr/src/sys/kern/kern_synch.c", line 249
XXX pullup-10
XXX pullup-9
XXX pullup-8
Unfortunately, owing to PR kern/57705 and PR misc/57706, it isn't
convenient to flip the xfail switch on a test for this bug. So we'll
do that separately. (But I did verify that a rumpified version of
the test postd to PR kern/57703 failed without this change, and
passed with this change.)
PR kern/57703
XXX pullup-10
thmap_del can't fail, and it is used in places in npf where sleeping
is forbidden, so it can't rely on allocating memory either.
Instead of having thmap_del allocate memory on the fly for each
object to defer freeing until thmap_gc, arrange to have thmap(9)
preallocate the same storage when allocating all the objects in the
first place, with a GC header.
This is suboptimal for memory usage, especially on insertion- and
lookup-heavy but deletion-light workloads, but it's not clear rmind's
alternative (https://github.com/rmind/thmap/tree/thmap_del_mem_fail)
is ready to use yet, so we'll go with this for correctness.
PR kern/57208
https://github.com/rmind/npf/issues/129
XXX pullup-10
XXX pullup-9
THMAP_GETPTR may return nonnull even though alloc returned zero.
Note that this failure branch is not actually appropriate;
thmap_create should not fail. We really need to pass KM_SLEEP
through in this call site even though there are other call sites for
which KM_NOSLEEP is appropriate.
Adapted from: https://github.com/rmind/thmap/pull/14
PR kern/57666
https://github.com/rmind/thmap/issues/13
XXX pullup-10
XXX pullup-9
Remove it in ddb/db_syncobj.h too.
New sys/wchan.h defines wchan_t so that users need not pull in
sys/syncobj.h to get it.
Sprinkle #include <sys/syncobj.h> in .c files where it is now needed.
- Allocate only one struct pipe not two (no need to be bidirectional here).
- Then use f_flag (FREAD/FWRITE) to figure out what to do in the fileops.
- Never wake the other side or acquire long-term (I/O) lock unless needed.
- Whenever possible, defer wakeups until after locks have been released.
- Do some things locklessly in pipe_ioctl() and pipe_poll().
Some notable results:
- -30% latency on a 486DX2/66 doing 1 byte ping-pong within a single process.
- 2.5x less lock contention during "make cleandir" of src on a 48 CPU machine.
- 1.5x bandwith with 1kB messages on the same 48 CPU machine (8kB: same b/w).
Like cv_broadcast(), but make any LWPs that share the same file descriptor
table as the caller return ERESTART when resuming. Used to dislodge LWPs
waiting for I/O that prevent a file descriptor from being closed, without
upsetting access to the file (not descriptor) made from another direction.
christos