My initial rate control code was .. suboptimal. I wanted to at least get MCS
rates sent, but it didn't do anywhere near enough to handle low signal level links
or remotely keep accurate statistics.
So, 8 years later, here's what I should've done back then.
* Firstly, I wasn't at all tracking packet sizes other than the two buckets
(250 and 1600 bytes.) So, extend it to include 4096, 8192, 16384, 32768 and
65536. I may go add 2048 at some point if I find it's useful.
This is important for a few reasons. First, when forming A-MPDU or AMSDU
aggregates the frame sizes are larger, and thus the TX time calculation
is woefully, increasingly wrong. Secondly, the behaviour of 802.11 channels
isn't some fixed thing, both due to channel conditions and radios themselves.
Notably, there was some observations done a few years ago on 11n chipsets
which noticed longer aggregates showed an increase in failed A-MPDU sub-frame
reception as you got further along in the transmit time. It could be due to
a variety of things - transmitter linearity, channel conditions changing,
frequency/phase drift, etc - but the observation was to potentially form
shorter aggregates to improve BER.
* .. and then modify the ath TX path to report the length of the aggregate sent,
so as the statistics kept would line up with the correct bucket.
* Then on the rate control look-up side - i was also only using the first frame
length for an A-MPDU rate control lookup which isn't good enough here.
So, add a new method that walks the TID software queue for that node to
find out what the likely length of data available is. It isn't ALL of the
data in the queue because we'll only ever send enough data to fit inside the
block-ack window, so limit how many bytes we return to roughly what ath_tx_form_aggr()
would do.
* .. and cache that in the first ath_buf in the aggregate so it and the eventual
AMPDU length can be returned to the rate control code.
* THEN, modify the rate control code to look at them both when deciding which bucket
to attribute the sent frame on. I'm erring on the side of caution and using the
size bucket that the lookup is based on.
Ok, so now the rate lookups and statistics are "more correct". However, MCS rates
are not the same as 11abg rates in that they're not a monotonically incrementing
set of faster rates and you can't assume that just because a given MCS rate fails,
the next higher one wouldn't work better or be a lower average tx time.
So, I had to do a bunch of surgery to the best rate and sample rate math.
This is the bit that's a WIP.
* First, simplify the statistics updates (update_stats()) to do a single pass on
all rates.
* Next, make sure that each rate average tx time is updated based on /its/ failure/success.
Eg if you sent a frame with { MCS15, MCS12, MCS8 } and MCS8 succeeded, MCS15 and MCS
12 would have their average tx time updated for /their/ part of the transmission,
not the whole transmission.
* Next, EWMA wasn't being fully calculated based on the /failures/ in each of the
rate attempts. So, if MCS15, MCS12 failed above but MCS8 didn't, then ensure
that the statistics noted that /all/ subframes failed at those rates, rather than
the eventual set of transmitted/sent frames. This ensures the EWMA /and/ average
TX time are updated correctly.
* When picking a sample rate and initial rate, probe rates aroud the current MCS
but limit it to MCS0..7 /for all spatial streams/, rather than doing crazy things
like hitting MCS7 and then probing MCS8 - MCS8 is basically MCS0 but two spatial
streams. It's a /lot/ slower than MCS7. Also, the reverse is true - if we're at
MCS8 then don't probe MCS7 as part of it, it's not likely to succeed.
* Fix bugs in pick_best_rate() where I was /immediately/ choosing the highest MCS
rate if there weren't any frames yet transmitted. I was defaulting to 25% EWMA and
.. then each comparison would accept the higher rate. Just skip those; sampling
will fill in the details.
So, this seems to work a lot better. It's not perfect; I'm still seeing a lot of
instability around higher MCS rates because there are bursts of loss/retransmissions
that aren't /too/ bad. But i'll keep iterating over this and tidying up my hacks.
Ok, so why this still something I'm poking at? rather than porting minstrel_ht?
ath_rate_sample tries to minimise airtime, not maximise throughput. I have
extended it with an EWMA based on sub-frame success/failures - high MCS rates
that have partially successful receptions still show super short average frame
times, but a /lot/ of retransmits have to happen for that to work.
So for MCS rates I also track this EWMA and ensure that the rates I'm choosing
don't have super crappy packet failures. I don't mind not getting lower
peak throughput versus minstrel_ht; instead I want to see if I can make "minimise
airtime" work well.
Tested:
* AR9380, STA mode
* AR9344, STA mode
* AR9580, STA/AP mode
Using UI units allows to fit larger numbers in columns.
Stop calling v_page_size - this is a value that doesn't change at runtime.
Renamed WINDOW *wnd to *wd to avoid conflict with global *wnd variable.
Use bit-shift to convert page size to byte.
PR: 246458
Submitted by: ota@j.email.ne.jp
MFC after: 2 weeks
Differential Revision: D24834
Expose the special kernel LAPIC, IOAPIC, and HPET devices to userspace
for use in, e.g., fallback instruction emulation (when userspace has a
newer instruction decode/emulation layer than the kernel vmm(4)).
Plumb the ioctl through libvmmapi and register the memory ranges in
bhyve(8).
Reviewed by: grehan
Differential Revision: https://reviews.freebsd.org/D24525
This problem was found by looking at syzkaller reproducers for some other
problems.
Reviewed by: rrs
Sponsored by: Netflix, Inc.
Differential Revision: https://reviews.freebsd.org/D24831
help of Michael Tuexen. There was some accounting
errors with TCPFO for bbr and also for both rack
and bbr there was a FO case where we should be
jumping to the just_return_nolock label to
exit instead of returning 0. This of course
caused no timer to be running and thus the
stuck sessions.
Reported by: Michael Tuexen and Skyzaller
Sponsored by: Netflix Inc.
Differential Revision: https://reviews.freebsd.org/D24852
This fixes a race where concurrent calls to doenterpgrp() and
leavepgrp() while TIOCSCTTY is executing may result in tp->t_pgrp
changing value so that tty_rel_pgrp() misses clearing it to NULL. For
more details refer to the use of pgdelete() in the kernel.
No functional change intended.
Panic backtrace:
__mtx_lock_sleep() # page fault due to using destroyed mutex
tty_signal_pgrp()
tty_ioctl()
ptsdev_ioctl()
kern_ioctl()
sys_ioctl()
amd64_syscall()
MFC after: 1 week
Sponsored by: Mellanox Technologies
netgraph(3) points to #include <netgraph/netgraph.h>, which is kernel only.
The man page refers to the user-space part of the netgraph module, which is
located in <netgraph.h>.
Submitted by: lutz_donnerhacke.de
Approved by: bcr
Differential Revision: https://reviews.freebsd.org/D23814
After r360820, additional parameters are passed through the argument 'opts', and the name of the backend through the argument 'devname'. So, there is no need to skip the backend name from the 'opts' argument.
This change introduces new jail command hooks that run before and after any
other actions.
The exec.prepare hook can be used for example to invoke a script that checks
if the jail's root exists, creating it if it does not. Since arbitrary
variables in jail.conf can be passed to the command, it can be pretty useful
for templating jails.
An example use case for exec.release would be to remove the filesystem of an
ephemeral jail.
The names "prepare" and "release" are borrowed from the names of similar hooks
in libvirt.
Reviewed by: jamie, manpages, mmacy
Approved by: mmacy (mentor)
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D24829
common_init_tbl is only used within this single CU, so it should be marked
static.
WARNS=6 also complained about the var defined by
`ATF_TC_WITH_CLEANUP(getastats);` being unused, which turns out to be
because it's not been hooked up in ATF_TP_ADD_TCS. kp@ did not immediately
recall any reason for this, and the case passes on my local system, so hook
it up.
Note that I've not yet set WARNS= 6 here. Investigation is underway to see
if we can feasibly default WARNS to 6 for src builds to catch directories
too deep to inherit a WARNS from the top-level subdirectories' Makefile.inc.
Those particular WARNS settings will be subsequently removed as they become
redundant with a more-global default.
MFC after: 1 week
In recent Linux (5.3+) and OpenBSD (6.6+) kernels, and with hosts that
support CPUID 0x15, the local APIC frequency is determined directly
from the reported crystal clock to avoid calibration against the 8254
timer.
However, the local APIC frequency implemented by bhyve is 128MHz, where
most h/w systems report frequencies around 25MHz. This shows up on
OpenBSD guests as repeated keystrokes on the emulated PS2 keyboard
when using VNC, since the kernel's timers are now much shorter.
Fix by reporting all-zeroes for CPUID 0x15. This allows guests to fall
back to using the 8254 to calibrate the local APIC frequency.
Future work could be to compute values returned for 0x15 that would
match the host TSC and bhyve local APIC frequency, though all dependencies
on this would need to be examined (for example, Linux will start using
0x16 for some hosts).
PR: 246321
Reported by: Jason Tubnor (and tested)
Reviewed by: jhb
Approved by: jhb, bz (mentor)
MFC after: 3 days
Differential Revision: https://reviews.freebsd.org/D24837
Reorder flag manipulations and use barrier to ensure that the program
order is followed by compiler and CPU, for unlocked reader of so_state.
In collaboration with: markj
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D24842
Sometimes, when doing read(2) over unix domain socket, for which the
other side socket was closed, read(2) returns -1/ENOTCONN instead of
EOF AKA zero-size read. This is because soreceive_generic() does not
lock socket when testing the so_state SS_ISCONNECTED|SS_ISCONNECTING
flags. It could end up that we do not observe so->so_rcv.sb_state bit
SBS_CANTRCVMORE, and then miss SS_ flags.
Change the test to check that the socket was never connected before
returning ENOTCONN, by adding all state bits for connected.
Reported and tested by: pho
In collaboration with: markj
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D24819
Submitted by: debdrup (with some minor changes by kevans)
Reviewed by: bcr (manpages)
MFC after: 3 days
Differential Revision: https://reviews.freebsd.org/D24818
This function is responsible for setting pc_domain in each pcpu
structure. Call it from the main function that starts APs, rather than
a separate SYSINIT. This makes it easier to close the window where
UMA's per-CPU slab allocator may be called while pc_domain is
uninitialized. In particular, the allocator uses pc_domain to allocate
domain-local pages, so allocations before this point end up using domain
0 for everything.
Reviewed by: kib
MFC after: 1 week
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D24757
Otherwise anything counted before SI_SUB_VM_CONF is discarded. However,
it is useful to be able to see stats from allocations done early during
boot.
Reviewed by: kib
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D24756
This man page contains stat utilities that are available in
the base system. This is a better approach than looking them
up via "apropos stat" or similar commands.
Thanks to Daniel Ebdrup Jensen for writing the original page
and incorporating the feedback given.
Submitted by: Daniel Ebdrup Jensen
Reviewed by: 0mp, allanjude, brueffer, bcr
Approved by: bcr
MFC after: 3 days
Relnotes: yes (new stats(7) man page)
Differential Revision: https://reviews.freebsd.org/D24417
We were accidentally using stfd instead of stw in our SAVEGPR macro.
This has almost certainly been causing crashes when compiling with -Os.
Reviewed by: jhibbits (in irc)
MFC after: 3 days
Sponsored by: Tag1 Consulting, Inc.
Otherwise, removals from the blacklist may not get processed as they should.
While we're here, restructure these to not bother with mkdir(1) if we've
already tested them to exist.
MFC after: 3 days
It's been reported/noted that a well-timed `certctl rehash` will completely
obliterate $CERTDESTDIR, which may get used by ports or system
administrators. While we can't guarantee the certctl semantics when other
non-certctl-controlled bits live here, we should make some amount of effort
to play nice.
Pruning all existing links, which we'll subsequently rebuild as needed, is
sufficient for our needs. This can still be destructive, but it's perhaps
less likely to cause issues.
I also note that we should probably be pruning /etc/ssl/blacklisted upon
rehash as well.
Reported by: cem's dovecot server
MFC after: 3 days
Extattr names are allowed to be 255 bytes -- not 254 bytes plus trailing
NUL. Provide a 256 buffer so that copyinstr() has room for the trailing
NUL.
Re-enable test for maximal name lengths.
PR: 208965
Reported by: asomers
Reviewed by: asomers
Differential Revision: https://reviews.freebsd.org/D24584
The alias needs to be part of the provider instead of the geom to work
properly. To bind the DEV geom, we need to look at the provider's names and
aliases and create the dev entries from there. If this lives in the GEOM, then
it won't propigate down the tree properly. Remove it from geom, add it provider.
Update geli, gmountver, gnop, gpart, and guzip to use it, which handles the bulk
of the uses in FreeBSD. I think this is all the providers that create a new name
based on their parent's name.
__builtin_unreachable doesn't raise any compile-time warnings/errors on its
own, so problems with its usage can't be easily detected. While it would be
nice for this situation to change and compilers to at least add a warning
for trivial cases where local state means the instruction can't be reached,
this isn't the case at the moment and likely will not happen.
This commit adds an __assert_unreachable, whose intent is incredibly clear:
it asserts that this instruction is unreachable. On INVARIANTS builds, it's
a panic(), and on non-INVARIANTS it expands to __unreachable().
Existing users of __unreachable() are converted to __assert_unreachable,
to improve debuggability if this assumption is violated.
Reviewed by: mjg
Differential Revision: https://reviews.freebsd.org/D23793
When protecting a superpage, we would previously fall through to the
non-superpage case and read the contents of the superpage as PTEs,
potentially modifying them and trying to look up underlying VM pages that
don't exist if they happen to look like PTEs we would care about. This led
to nginx causing an unexpected page fault in pmap_protect that panic'ed the
kernel. Instead, if we see a superpage, we are done for this range and
should continue to the next.
Reviewed by: markj, jhb (mentor)
Approved by: markj, jhb (mentor)
Differential Revision: https://reviews.freebsd.org/D24827
This is in preparation for me bumping how many size buckets are used
for ath_rate_sample statistics.
* Bump buffer size to 64k
* Don't waste 4 lines per bucket size, condense it to two
* Alternate colours; my logic made everything after the first two just
be black. Oops.
