===================================================================
RCS file: /cvsrepo/anoncvs/cvs/www/73.html,v
retrieving revision 1.50
retrieving revision 1.51
diff -u -r1.50 -r1.51
--- www/73.html 2023/04/08 01:10:55 1.50
+++ www/73.html 2023/04/08 01:52:39 1.51
@@ -95,21 +95,17 @@
Introduced clockintr(9), a
machine-independent clock interrupt scheduler. Switched all
- architectures to use this new kernel subsystem.
- Added a priority queue to clockintr(9).
+ architectures to use the new subsystem.
Introduced a new kern.autoconf_serial sysctl(8) that can be used
by userland to monitor state changes of the kernel device tree.
Fixed pmap(9) bugs
involving entering an executable mapping for a page before
synchronizing the data and instruction cache on arm64 and riscv64.
- Removed copystr(9) from public API.
- Add getnsecruntime(9)
- to the kernel timecounting API. Together with getbinruntime(), it
- provides a fast, monotonic clock that only advances while the system
- is not suspended.
+ Removed copystr(9) from the public API.
+ Added getnsecruntime(9).
+ Offers fast access to the system runtime clock at the cost of precision.
Add detection for Spectre-BHB Branch History Injection
vulnerability related CLRBHB, ECBHB and CSV2_3/HCXT feature bits.
@@ -144,9 +140,10 @@
href="https://man.openbsd.org/munmap.2">munmap(2), and mprotect(2).
Unlocked sched_yield(2).
- Added support for per-cpu event counters, to be used for clock and
- IPI counters where the event counted occurs across all CPUs in the
- system.
+ Added support for per-CPU counters to
+ evcount(9).
+ Useful for counting events that are prone to occur simultaneously
+ across multiple CPUs, like clock interrupts and IPIs.
Moved pf(4) purge
tasks out from under the kernel lock.
Unlocked ioctl(2)
@@ -197,8 +194,10 @@
VMM/VMD improvements
- Updated vmm(4) to
- allow guests to read MSR_HWCR and MSR_PSTATEDEF, which is necessary to
- determine the TSC frequency on AMD families 17h and 19h.
+ permit SVM guests read access to MSR_HWCR and MSR_PSTATEDEF.
+ Guests can use these registers on AMD 17h and 19h hosts to
+ determine the TSC frequency without calibrating against a
+ second clock.
- Allocated reference for vm and vcpu SLISTs in vmm(4), keeping vmm from
triggering excessive wakeup calls while iterating through the list of
@@ -288,15 +287,15 @@
that contain a single member.
- Fixed ed(1) to print
bytes read/written and the ? prompt to stdout, not stderr.
-
- Modified the vmstat view in Changed the vmstat view in systat(1) to measure
- elapsed time using clock_gettime(2).
-
- Implemented periodic display in clock_gettime(2)
+ instead of statclock ticks.
+
- Improved the periodic display in iostat(8).
-
- Corrected top(1)
- display of online CPUs which can change based on the sysctl(2) sysctl setting.
+
- Fixed an edge case in top(1)
+ where summary statistics for offline CPUs were displayed.
- Added support for a personal units(1) library by passing
-f multiple times.
@@ -317,9 +316,9 @@
name of each library before relinking as a signal to the operator that
boot has not stalled.
-
- Added the audioctl(8) -w option to
- display variables periodically.
+
- Added a -w flag to audioctl(8) for
+ displaying variables periodically.
- Added short options for timeout(1) --foreground
and --preserve-status.
@@ -644,9 +643,9 @@
- Added or improved wireless network drivers:
- - Bumped tsleep timeout for bwfm(4) PCI devices to help
- prevent failures loading firmware, particularly on Apple M2 laptops.
+
- Increased the timeout for bwfm(4) PCI devices to
+ avoid spurious firmware load failures, particularly on Apple M2 laptops.
- Implemented alternative mailbox handling mechanism required by
newer bwfm(4) firmware.
- Fixed bwfm(4)
@@ -823,11 +822,11 @@
configuring IPv6. This allows non-multicast interfaces such as
point-to-point interfaces and the NBMA / point-to-multipoint
interfaces like mpe(4), mgre(4) and wg(4) to work with IPv6.
-
- Use the new Measure the TCP_KEEPALIVE timeout with getnsecruntime(9)
- timer to check the TCP_KEEPALIVE timer only against the system
- runtime, not the uptime. Prevents TCP connections to fail after
- waking up from suspend.
+ instead of the the system uptime.
+ Prevents TCP connections from needlessly failing en masse after
+ waking a system from suspend.
- Used stoeplitz (symmetric Toeplitz hash algorithm) to generate a
hash/flowid for pf(4) state
keys. With this change, pf will hash traffic the same way that