OpenBSD cvs log

created 2019-04-30T03:26:05Z
begin 2019-03-07T00:00:00Z
end 2019-03-14T00:00:00Z
path src/sys
commits 30

date 2019-03-07T07:58:26Z
author patrick
files src/sys/arch/arm64/dev/simplebus.c log diff annotate
message When accessing the child address cells make sure to use the size for
those instead of the size of the parent address cells.

ok kettenis@

date 2019-03-07T10:46:37Z
author jsg
files src/sys/arch/i386/stand/boot/conf.c log diff annotate
src/sys/arch/i386/stand/cdboot/conf.c log diff annotate
src/sys/arch/i386/stand/libsa/exec_i386.c log diff annotate
src/sys/arch/i386/stand/pxeboot/conf.c log diff annotate
message Return early in ucode loading if cpuid is not available. Should fix
booting on 486s without cpuid. Reported by Falk Richter and diagnosed
by guenther@

date 2019-03-08T10:53:59Z
author fcambus
files src/sys/dev/wsfont/spleen12x24.h log diff annotate
src/sys/dev/wsfont/spleen16x32.h log diff annotate
src/sys/dev/wsfont/spleen32x64.h log diff annotate
src/sys/dev/wsfont/spleen8x16.h log diff annotate
message Update Spleen kernel fonts to the latest released version, bringing
the following improvements:

- Remove strain pixels on the inner upper part of parentheses for the
16x32 version
- Harmonize the 'v' and 'V' characters across all sizes
- Remove artefacts from the 'c' in the copyright sign across all sizes
- Make the upper left corner of the 'R' sharp in the registered sign
for the 16x32 and 32x64 versions

"Go for it" kettenis@, deraadt@

date 2019-03-08T16:33:23Z
author jcs
files src/sys/dev/acpi/acpithinkpad.c log diff annotate
message the previous hkey version check was not accurate enough, so try the
ACPI method for screen backlight changes everywhere and if the ACPI
method fails, fallback on the CMOS method.

also unmask the mic mute event needed on the x260.

required for the x260, no regression on the x40, x230, x250, x280,
t460s, t470p, and x1c6.

date 2019-03-08T17:48:35Z
author mpi
files src/sys/net/bridgectl.c log diff annotate
src/sys/net/if_bridge.c log diff annotate
src/sys/net/if_bridge.h log diff annotate
message Move the tag mechanism outside of net/if_bridge.c.

This will help for future (un)locking.

ok visa@

date 2019-03-08T17:49:36Z
author mpi
files src/sys/net/if_bridge.c log diff annotate
message Do not grab a `bif' pointer again, we already have it.

ok visa@

date 2019-03-09T06:14:21Z
author visa
files src/sys/sys/atomic.h log diff annotate
message Add macros READ_ONCE() and WRITE_ONCE(). They constrain optimizations
done by the compiler.

OK guenther@

date 2019-03-09T06:15:48Z
author visa
files src/sys/sys/smr.h log diff annotate
message Use READ_ONCE() and WRITE_ONCE() in .

OK guenther@

date 2019-03-09T23:04:56Z
author cheloha
files src/sys/kern/kern_tc.c log diff annotate
message tc_windup: read active timecounter once at function start.

tc_windup() is not necessarily called with KERNEL_LOCK, so it is possible
for the timecounter pointer to change in the midst of the call via the
kern.timecounter.hardware sysctl(2). Reading it once and using that local
copy ensures we're referring to the same timecounter consistently.

Apparently the compiler can optimize this out... somehow... so there may
be room for improvement.

Idea from visa@. With input from visa@, mpi@, cjeker@, and guenther@.

ok visa@ mpi@

date 2019-03-10T07:35:33Z
author mlarkin
files src/sys/arch/amd64/amd64/vmm.c log diff annotate
message vmm(4): add a debug printf if the host does not have writeback memory
type for EPT

Noticed by Guillaume Pagnoux and myself while cleaning up some old bugs.

date 2019-03-10T07:44:17Z
author mlarkin
files src/sys/arch/amd64/amd64/vmm.c log diff annotate
message missing newline in a printf

date 2019-03-10T08:27:28Z
author mlarkin
files src/sys/arch/amd64/amd64/vmm.c log diff annotate
message fix two wrong comments

date 2019-03-10T14:20:44Z
author visa
files src/sys/arch/octeon/dev/octcrypto.c log diff annotate
message Use SMR instead of reference counting for session reclamation.
This reduces the need of atomic operations. The lookup tree still
requires serialization, though.

date 2019-03-10T21:16:15Z
author cheloha
files src/sys/kern/kern_tc.c log diff annotate
src/sys/kern/kern_time.c log diff annotate
src/sys/sys/kernel.h log diff annotate
src/sys/sys/timetc.h log diff annotate
message Move adjtimedelta from kern_time.c to kern_tc.c.

This will simplify upcoming MP-safety diffs for the timecounting layer.

adjtimedelta is now accessed nowhere outside of kern_tc.c, so we can
remove its extern declaration from kernel.h. Zeroing adjtimedelta
within timecounter_mtx before we jump the real-time clock is also a
bit safer than what we do now, as we are not racing a simultaneous
tc_windup() call from hardclock(), which itself can modify adjtimedelta
via ntp_update_second().

Discussed with visa@ and mpi@.

ok visa@

date 2019-03-11T06:19:33Z
author kevlo
files src/sys/dev/ic/r92creg.h log diff annotate
src/sys/dev/ic/rtwn.c log diff annotate
src/sys/dev/ic/rtwnvar.h log diff annotate
src/sys/dev/usb/if_urtwn.c log diff annotate
message - in rtwn_r88e_fw_reset function, use the correct bit R88E_RSV_CTRL_MCU_RST
to reset/enable MCU
- there's no need to enable 1.2V power for rtl8192eu

ok jmatthew@

date 2019-03-11T17:13:31Z
author anton
files src/sys/kern/tty_pty.c log diff annotate
message Bring back revision 1.91 now that vmd has been fixed, repeating the
previous commit message:

When closing the slave end of a pty, generate an EOF event to any kqueue
consumer of the master end. This behavior is equivalent to how pipes
already behave with kqueue. Also, FreeBSD and NetBSD behaves the same
way.

ok deraadt@ millert@ visa@

date 2019-03-11T17:50:08Z
author mpi
files src/sys/dev/usb/ehci.c log diff annotate
src/sys/dev/usb/ohci.c log diff annotate
message Add missing bus powered bit, from aalm@

date 2019-03-11T17:50:09Z
author mpi
files src/sys/dev/usb/uhci.c log diff annotate
src/sys/dev/usb/xhci.c log diff annotate
src/sys/dev/usb/dwc2/dwc2.c log diff annotate
message Add missing bus powered bit, from aalm@

date 2019-03-12T01:07:37Z
author jmatthew
files src/sys/dev/pci/if_ixl.c log diff annotate
message Until we can figure out why it causes NMIs on some machines, skip the
pre-reset steps described in Intel's datasheet and also their driver code.

ok dlg@

date 2019-03-12T08:13:50Z
author ratchov
files src/sys/dev/usb/uhci.c log diff annotate
message uhci: Fix delayed completions for isochronous transfers.

When an isochronous transfer of n frames is scheduled, the last frame
i.e. frame number (n - 1) must be set to generate an interrupt.

ok mpi

date 2019-03-12T08:16:29Z
author ratchov
files src/sys/dev/audio.c log diff annotate
src/sys/dev/audio_if.h log diff annotate
message Add new copy_output() and underrun() methods to support drivers
using bounce buffers, like the uaudio(4) driver.

ok mpi

date 2019-03-12T08:32:06Z
author ratchov
files src/sys/dev/usb/uaudio.c log diff annotate
message Add a new driver for USB Audio Class v2.0 devices. It replaces the
current one for UAC v1.0 devices.

The main difference with the old driver is that now we map audio
blocks to USB transfers, which allows precise synchronization and
reliability, including during low-latency operation.

with help from many, ok mpi

date 2019-03-12T08:33:25Z
author ratchov
files src/sys/dev/usb/Attic/uaudioreg.h log diff annotate
message Remove unused uaudioreg.h file.

ok mpi

date 2019-03-12T11:01:25Z
author nicm
files src/sys/sys/ttydefaults.h log diff annotate
message Almost all terminals now support hardware tabs so default to OXTABS
off.

This makes three changes: adds the ht capability to the standard lines
in gettytab(5); removes OXTABS from TTYDEF_OFLAG in ttydefaults.h (the
defaults used by pty(4) - diff from martijn); and only sets OXTABS on
terminals which lack hts and tbc in tset(1) (from Thomas Dickey
upstream).

Addresses problems reported by tedu.

ok millert

date 2019-03-12T11:45:00Z
author mpi
files src/sys/net/if_bridge.c log diff annotate
message Merge copy/pasted code to export STP states via ioctl into a function.

date 2019-03-12T18:13:40Z
author patrick
files src/sys/dev/usb/xhci.c log diff annotate
message Fix typo in debug print: wih -> with

ok mpi@

date 2019-03-12T22:14:50Z
author patrick
files src/sys/arch/arm64/include/fenv.h log diff annotate
message Setting and getting the rounding mode on our arm64 FPU has not worked
in libm since the rounding mode is in fpcr, not fpsr. Since both FPU
registers are 32-bit we can store them in the 64-bit fenv_t to make
handling the bits easier.

While there add FE_DENORMAL, which also exists on x86. Also make sure
that whenever we are being passed an exception mask, we only allow the
bits that are supported by hardware.

Found by regression tests
Debugged with Moritz Buhl
ok kettenis@

date 2019-03-13T09:28:21Z
author patrick
files src/sys/arch/arm/arm/db_interface.c log diff annotate
src/sys/arch/arm/arm/undefined.c log diff annotate
src/sys/arch/arm/arm/vfp.c log diff annotate
src/sys/arch/arm/include/undefined.h log diff annotate
src/sys/arch/arm/include/vfp.h log diff annotate
message Store whether or not the VFP was active on entering the unhandled
instruction trap and pass it to the VFP fault handler, so that it
knows if we faulted with the VFP active or disabled. Reading the
status in the VFP fault handler does not work since at that point
the VFP already got disabled.

ok kettenis@

date 2019-03-13T09:41:12Z
author patrick
files src/sys/dev/fdt/imxccm.c log diff annotate
message Since new bindings might use the same clock index for divs, gates and
muxes, quietly return if the clock has no gate, but it's a div/mux.
With this clocks that are defined in both gates and divs/muxes can
be enabled.

date 2019-03-13T09:49:46Z
author patrick
files src/sys/dev/fdt/imxccm.c log diff annotate
src/sys/dev/fdt/imxccm_clocks.h log diff annotate
message Add support for i.MX8M PWM clocks.