These seem harmless to address. Not sure why the original author did not
just assign the appropriate names if he knew they were wrong.
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D56044
It is a legacy script which is no longer used. Its utility is also
unknown.
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D56043
It is always defined.
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D56042
They introduce an extra level of abstraction for no reason at all.
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D56041
chn_getmute_matrix() does what CHN_GETMUTE() does, but with a few
additional checks.
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D55993
CHN_SETVOLUME() is just a wrapped around chn_setvolume_matrix() anyway,
so use it directly.
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D55992
CHN_GETVOLUME() is just a wrapped around chn_getvolume_matrix() anyway,
so use it directly.
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D55991
There is no reason not to use it. We do it already with CHN_SETVOLUME().
chn_getvolume_matrix() is the same as the non-INVARIANTS
CHN_GETVOLUME(), just without the additional KASSERT
chn_getvolume_matrix() provides.
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D55990
The purpose of this has not been documented, but it seems like it makes
it possible to view the maximum number of bytes that passed to
sndbuf_feed(), as well as the maximum number of cycles taken inside its
main loop. These do not seem particularly useful anymore.
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D55989
The feeder chain gets destroyed when feeder_chain() is called, which is
after the chn_reset() call in chn_init() further down for primary chans,
or vchan_create() for vchans. This makes the root feeder creation in
chn_init() essentially a no-op. Remove it altogether and let
feeder_chain() after chn_reset() take care of creating the feeder
properly. It creates the root one as well.
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D55941
This a legacy thing that is not needed anymore. We can support all of
them just fine.
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D55939
There is no reason to have these legacy controls anymore, all these
formats can be handled just fine nowadays.
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D55937
SND_MULTICHANNEL is always defined, so SND_CHN_MAX will also always be
8. Apart from the fact that there is no other place in the code that
touches SND_MULTICHANNEL, there is also no good reason nowadays to set
SND_CHN_MAX to 2.
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D55934
This is a legacy option and does not serve a good purpose anymore.
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D55933
This performs an attach, not probe.
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D55900
Prune unused code hidden behind 'notdef', bringing us in sync with
the changes in OpenBSD. Despite the `__default_hash` function pointer
having external linkage, no ABI change is expected since it was
never exported.
Sponsored by: Klara, Inc.
Differential Revision: https://reviews.freebsd.org/D55842
Reviewed by: allanjude, des
MFC after: 2 weeks
This change adds support for three new Intel fields to jevents.c:
- MetricThreshold
- MetricgroupNoGroup
- DefaultMetricgroupName
Reviewed by: ali_mashtizadeh.com, mhorne
Differential Revision: https://reviews.freebsd.org/D56017
MFC after: 1 week
Previously was or'ing the indices of the DSMs directly, not their
corresponding bits. This commit rectifies this.
Reviewed by: olce
Approved by: olce
Fixes: c5daa5a4c3 ("acpi_spmc: Add system power management controller driver")
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/https://reviews.freebsd.org/D56062
hgame_dpad_cb() previously exited early whenever conflicting directions
were input (e.g., UP + DOWN) without saving said input to the
hgame_softc state. This led to a desync between the driver and the
gamepad.
This patch instead handles conflicting inputs by cancelling them out
with each other.
Remove early return. Calculate axis value by subtraction.
Reviewed by: obiwac
Approved by: obiwac
Differential Revision: https://reviews.freebsd.org/D55849
There's been a problem where rules which differed only in address ranges
were considered duplicates and not added. Test for this.
MFC after: 1 week
Sponsored by: Rubicon Communications, LLC ("Netgate")
The default values documented for kern.hwpmc.logbuffersize (4KB) and
kern.hwpmc.nbuffers_pcpu (64) have been incorrect since 2981a3420cb1
(2018), which updated the compiled defaults but did not update the
man page.
- Correct logbuffersize default from 4KB to 256KB, add 16MB maximum
- Correct nbuffers_pcpu default from 64 to 32, document 32MB per-CPU
product limit with kern.hwpmc.logbuffersize
- Update DIAGNOSTICS section to reflect current warning messages
Reviewed by: mhorne
MFC after: 1 week
Sponsored by: NLINK (nlink.com.br)
Differential Revision: https://reviews.freebsd.org/D56050
Replace printf() with log(LOG_WARNING, ...) in pmclog_initialize()
so that tunable validation failures are visible in dmesg and
/var/log/messages rather than only on the early console.
Also improve the messages to report both the invalid value and the
default it resets to, making it easier for users to understand why
their tunable was ignored.
While here, adjust some whitespacing/style.
Reviewed by: Ali Mashtizadeh <ali@mashtizadeh.com>, mhorne
MFC after: 1 week
Sponsored by: NLINK (nlink.com.br)
Differential Revision: https://reviews.freebsd.org/D56029
Add support to libpmc for parsing the IBS qualifiers and computing the
ctl register value as a function of the qualifiers and the sample rate.
This includes all of the flags available up to AMD Zen 5. Along side
these user facing changes I included the documentation for AMD IBS.
Reviewed by: mhorne
Sponsored by: Netflix
Pull Request: https://github.com/freebsd/freebsd-src/pull/2081
No need for a barrier here, we are inside an NMI handler and executing a
number of serializing instructions with stronger semantics. Reducing
this overhead will increase our maximum safe sampling rate.
Tested by: Paulo Fragoso <paulo@nlink.com.br>
Reviewed by: mhorne
MFC after: 1 week
Sponsored by: Netflix
Pull Request: https://github.com/freebsd/freebsd-src/pull/2076
We could patch the tzcode config to not use it, but it's simple to
provide an implementation of it and avoid spreading cross-build
bootstrapping special cases.
Fixes: ff2c98b30b ("tzcode: Update to 2026a")
MFC after: 1 week
because cancel_delayed_work_sync() might need to sleep, which cannot be
done in the interrupt thread where the completion runs.
Sponsored by: Nvidia networking
MFC after: 1 week
OpenBSD fixed this separately (in their 6a338f3f70). We appear to not
have had this bug, but test for it anyway.
Sponsored by: Rubicon Communications, LLC ("Netgate")
Christos Margiolis