♟ imp

cam/iosched: Document latency buckets correctly.

nvmecontrol: Allow optional /dev/ for device names

In D45042#1027404, @andrew wrote:

I've been thinking about adding PAGE_SIZE_MAX/PAGE_SHIFT_MAX or similar to arm64 to define the largest page size the kernel could support. We could then use that here if it's defined.

Thanks for the rework!
I'm just finishing a nvme-cli port :)

In D40676#1027003, @theraven wrote:

After this change, ktrace output is littered with 'CAP system call not allowed: $SYSCALL' on systems w/o capsicum enabled

Are systems without Capsicum still supported? I thought that option was removed in 14.

I've run this patch from you and noticed no world build time regressions with a 4k kernel

ctl: Fix CTL GET EVENT STATUS NOTIFICATION valid bits

package: fix dependency generation

package: FreeBSD-periodic should depend on FreeBSD-cron

services.5: describe better

Add libxo support to du

rights.4: various corrections on capability rights

glabel: Add support for Linux swap

package: move authpf into the FreeBSD-pf package

package: move OpenBSM auditing into its own package

UPDATING: add entry for recent pkgbase changes

setkey: Use nitems(foo) instead of sizeof(foo)/sizeof(foo[0])

ipf: Use nitems(foo) instead of sizeof(foo)/sizeof(foo[0])

camcontrol: Use nitems(foo) instead of sizeof(foo)/sizeof(foo[0])

ipfw: Use nitems(foo) instead of sizeof(foo)/sizeof(foo[0])

zonectl: Use nitems(foo) instead of sizeof(foo)/sizeof(foo[0])

usbconfig: Use nitems(foo) instead of sizeof(foo)/sizeof(foo[0])

wlandebug: Use nitems(foo) instead of sizeof(foo)/sizeof(foo[0])

rwhod: Use nitems(foo) instead of sizeof(foo)/sizeof(foo[0])

pstat: Use nitems(foo) instead of sizeof(foo)/sizeof(foo[0])

ppp: Use nitems(foo) instead of sizeof(foo)/sizeof(foo[0])

ndp: Use nitems(foo) instead of sizeof(foo)/sizeof(foo[0])

moused: Use nitems(foo) instead of sizeof(foo)/sizeof(foo[0])

kbdcontrol: Use nitems(foo) instead of sizeof(foo)/sizeof(foo[0])

ctld: Use nitems(foo) instead of sizeof(foo)/sizeof(foo[0])

ctladm: Use nitems(foo) instead of sizeof(foo)/sizeof(foo[0])

btxld: Use nitems(foo) instead of sizeof(foo)/sizeof(foo[0])

rtadvctl: Use nitems(foo) instead of sizeof(foo)/sizeof(foo[0])

rtadvd: Use nitems(foo) instead of sizeof(foo)/sizeof(foo[0])

bluetooth: Use nitems(foo) instead of sizeof(foo)/sizeof(foo[0])

axgbe: Various stability improvements

ada: Another NCQ Trim instability drive

In D33233#1025516, @olce wrote:

In D33233#1025445, @imp wrote:

There's almost no difference between sync; shutdown and just shutdow other than more typing. The latter does a sync first thing anyway (in the system call) so you may start the io a smidgen earlier with the former... but sync doesn't guarantee the io will be done before shutdown starts... so even doing them on sepatate lines might only gain you the time it takes you to type shutdown at the expense of typing 5 extra characters.

The main difference is that you control the time you leave to the system to actually write to disk. Typing sync and then shutdown almost immediately indeed doesn't make much sense. Waiting for some seconds or much more does, if you know statistically that all writes will have finished by that chosen time.

Yes. I agree. It will be defined for both kernel and loader builds.

linux: ignore setsockopt(IPV6_RECVERR)

There's almost no difference between sync; shutdown and just shutdow other than more typing. The latter does a sync first thing anyway (in the system call) so you may start the io a smidgen earlier with the former... but sync doesn't guarantee the io will be done before shutdown starts... so even doing them on sepatate lines might only gain you the time it takes you to type shutdown at the expense of typing 5 extra characters.

Since reboot and shutdown system calls were added just after v7, the sync 3 times dance is a placebo. It was marginally useful on v7 before powering off. It's existed in lore as being necessary when it hasn't been. And since all it does is schedule IO you can't use it to know data is on the disk.

CTL: READ(6) should be allowed on CD devices

nvmecontrol: Flesh out nvmecontrol format information

update per Chuck's idea of making it clear it's an index. Hopefully this makes
it clearer. Thanks Chuck!

So I'm not saying we need it now, but we likely need to know if we can... get some sleep

I don't suppose that yhese could be generated automatically? It's a huge table to maintain by hand...

Any reason to not always set cn_unmute = 0 unconditionally?

arm_kernel_bothdr.awk: Update to latest ota

Add support for Intel Atom S1200 UART

In D43463#1024283, @delphij wrote:

Can we merge this?

Revert "bsdinstall/distfetch.c: check environment variables before downloading…

Revert "swapon: Do not overwrite Linux swap header"

bsdinstall/distfetch.c: check environment variables before downloading and…

swapon: Do not overwrite Linux swap header

release: Change vmimage EFI GPT label

nanobsd: Set a GPT label for EFI and cfg partitions

sys/*/conf: do not use "../../conf/" when including std.*

sys: add conf/std.debug, generic debugging options

sys/netpfil/pf: fix non-INET module build

sys/net/if_bridge: support non-INET kernels

ahc(4)/ahd(4): target mode: cancel outstanding AIOs and INOTs

ahc(4) clean up old Linux defines

checkstyle9.pl: Warn if there's no SOB line

pow,powf(3),__ieee754_rem_pio2(f): Avoid negative integer left shift UB

I'd thought we didn't need to propogate the locale, since both strerror and strerror_r use the same locale()... but I'd forgotten printf can have a different one, so I proposed something different in the bug.
We could share buf and errormsg, but maybe it's not worth the hassle.

zfsbootcfg(8), zpool_influxdb(8): move to the zfs package

swab.c(libc): use a simplified version of byte swapping

ifconfig tunnelfib is implemented in wg(4)

alc(4): disable MSI-X by default on Killer cards

umount: Support partitions & slices with -d option

share/examples: move examples into appropriate packages

package: add DESC and COMMENT for cron package

package: move cron into its own package

package: move hv_{vss,kvp}_daemon into hyperv-tools

linux: ignore setsockopt(IPV6_RECVERR)

libipsec: make const-correct

this looks good to me now (forget about my other comments... I clicked some something weird maybe?)

I think a rebase is in order... this diff is picking up a lot of "noise" that have nothing to do with fortify....

imp (Warner Losh)
User

Projects

User Details

Recent Activity
View All

Yesterday

Wed, May 1

Tue, Apr 30

Mon, Apr 29

Sun, Apr 28

Sat, Apr 27

Fri, Apr 26

Thu, Apr 25

Wed, Apr 24

Tue, Apr 23

Mon, Apr 22

Sun, Apr 21

Sat, Apr 20

imp (Warner Losh)User