Page MenuHomeFreeBSD

Recent Activity

Today

imp committed rG42b7e9c4a9c5: newbus: Remove redundant check for 0 (authored by imp).
newbus: Remove redundant check for 0
Wed, Oct 16, 3:19 AM
imp closed D44269: newbus: Remove redundant check for 0.
Wed, Oct 16, 3:19 AM
david_crossfamilyweb.com requested review of D47149: nscd step7.
Wed, Oct 16, 3:13 AM
david_crossfamilyweb.com requested review of D47148: nscd step6.
Wed, Oct 16, 3:10 AM
jamie accepted D46896: MAC: mac.h: Fix missing includes/typedefs.
Wed, Oct 16, 3:01 AM
jamie accepted D46899: MAC: Define a common 'mac' node for MAC's jail parameters.

Currently, it seems to suggest (being a jailsys parameter) that there's some sort of valid "new" or "deleted" state for MAC inside the jail.

Not sure what you mean here. SYSCTL_JAIL_PARAM_NODE() just declares the common MAC sub-node. mac_do(4) will then indeed use the new SYSCTL_JAIL_PARAM_SYS_SUBNODE() for the mac.do jail parameter "node", effectively intended to be a jailsys one.

Wed, Oct 16, 3:00 AM
jamie accepted D46898: MAC: 'kernel_mac_support' module: Make an outdated comment more generic.
Wed, Oct 16, 2:58 AM
jamie accepted D46897: MAC: mac.h: Separately test inclusion from userspace.
Wed, Oct 16, 2:57 AM
jamie accepted D46903: MAC: mac_policy.h: Declare common MAC sysctl and jail parameters' nodes.
Wed, Oct 16, 2:57 AM
jamie added a comment to D46284: Add the ability have executable jail.conf.

It would make sense for the exec.clean parameter to apply to the config execution. Bit of a chicken and egg problem there, but there's still the "-l" flag.

Wed, Oct 16, 2:56 AM · Jails
david_crossfamilyweb.com requested review of D47147: nscd step5.
Wed, Oct 16, 2:51 AM
imp abandoned D35040: ada: create ada_reprobe for common probe starting logic.

I'm torn on this, but I'm going to punt to declutter my life

Wed, Oct 16, 2:50 AM
imp abandoned D34357: newbus: Create a preferred list of drivers.

There's problems with this approach, since it relies on ties.

Wed, Oct 16, 2:46 AM
imp abandoned D30352: cam: introduce xpt_setup_stack_ccb.

I still like this idea, but I think that I'm just going to punt on this detail

Wed, Oct 16, 2:44 AM
gallatin added a comment to D30155: ixgbe: Bring back accounting for tx in AIM.

Yeah, my ideal irq rate/queue is < 1000 . We mostly use Chelsio and Mellanox NICs that can do super aggressive irq coalescing without freaking out TCP due to using RX timestamps. Super aggressive coalescing like this lets us build packet trains in excess of 1000 packets to feed to lro via RSS assisted LRO, and we actually have useful LRO on internet workloads with tens of thousands of TCP connections per-queue. That reminds me that I should port RSS assisted LRO to iflib (eg, lro_queue_mbuf()).

Wed, Oct 16, 2:41 AM
imp closed D42400: mpi3mr: Enqueue request from mpi3mr_prepare_sgls.

A different variation of this landed as part of 3208a189c1e2c4ef35daa432fe45629a043d7047

Wed, Oct 16, 2:38 AM
imp closed D35918: stand: In geli, use devformat instead of disk_fmtdev.

This was folded in with D35917 and landed as ad759c73522ef

Wed, Oct 16, 2:34 AM
david_crossfamilyweb.com requested review of D47146: nscd step4.
Wed, Oct 16, 2:26 AM
david_crossfamilyweb.com requested review of D47145: nscd step3.
Wed, Oct 16, 2:14 AM
david_crossfamilyweb.com requested review of D47144: nscd step2.
Wed, Oct 16, 2:06 AM
kbowling added a comment to D30155: ixgbe: Bring back accounting for tx in AIM.

@imp @gallatin if you are able to test your workload, setting this to 1 and 2 would be new behavior versus where you are currently:

I can pull this into our tree and make an image for @dhw to run on the A/B cluster. However, we're not using this hardware very much any more, and there is only 1 pair of machines using it in the A/B cluster. Lmk if you're still interested, and I'll try to build the image tomorrow so that David can test it at his leisure.

Sure, it sounds like that is only enough for one experiment so I would focus on the default algorithm the patch will boot with sysctl dev.ix.<N>.enable_aim=1

Its running now. Eyeballing command-line utilities, the CPU is about 5% higher (27% -> 32%) and we have 2x the irq rate (110k vs 55k irq/sec).
When applying this, I wanted to give it a fair shake, and disabled this tunable: hw.ix.max_interrupt_rate=4000. Perhaps that was a mistake? Is there a runtime way to tweak the algorithm so it doesn't interrupt so fast under this level of load?

Wed, Oct 16, 1:59 AM
david_crossfamilyweb.com requested review of D47143: nscd -- step1.
Wed, Oct 16, 1:48 AM
gallatin added a comment to D30155: ixgbe: Bring back accounting for tx in AIM.

@imp @gallatin if you are able to test your workload, setting this to 1 and 2 would be new behavior versus where you are currently:

I can pull this into our tree and make an image for @dhw to run on the A/B cluster. However, we're not using this hardware very much any more, and there is only 1 pair of machines using it in the A/B cluster. Lmk if you're still interested, and I'll try to build the image tomorrow so that David can test it at his leisure.

Sure, it sounds like that is only enough for one experiment so I would focus on the default algorithm the patch will boot with sysctl dev.ix.<N>.enable_aim=1

Wed, Oct 16, 1:39 AM
bz requested review of D47142: net/wifi-firmware-*kmod: update firmware for iwlwifi, rtw88, rtw89.
Wed, Oct 16, 1:15 AM
david_crossfamilyweb.com updated the diff for D38047: Multiple fixes to the NSS caching system.

rebasing with minor additional changes before I begin the work to split this up

Wed, Oct 16, 1:13 AM · Src Committers
bz requested review of D47141: net/wifi-firmware-kmod: add a release flavor.
Wed, Oct 16, 1:11 AM
pfg committed rG80a5b26871e9: sbin/ping: allow normal users to specify larger packets (authored by pfg).
sbin/ping: allow normal users to specify larger packets
Wed, Oct 16, 1:08 AM
pfg closed D45774: sbin/ping: allow normal users to specify larger packets.
Wed, Oct 16, 1:07 AM
osa committed rGbdbf50660c51: bsd-family-tree: add macOS 15 (authored by osa).
bsd-family-tree: add macOS 15
Wed, Oct 16, 12:54 AM
imp closed D37864: stand: Add eh symbols to linker scripts.

85f2095d4d7e2e981d5535448d9b874307a2dae5 landed this.

Wed, Oct 16, 12:49 AM
dvl committed R11:2ec05f9c7fb9: net-mgmt/p5-OSLV-Monitor: Update to 1.0.2 (authored by dvl).
net-mgmt/p5-OSLV-Monitor: Update to 1.0.2
Wed, Oct 16, 12:46 AM
brooks committed R11:de16aa64b8a8: devel/llvm-devel: new snapshot (authored by brooks).
devel/llvm-devel: new snapshot
Wed, Oct 16, 12:37 AM
brooks committed R11:bb4c04f85eec: devel/llvm19: 19.1.2 release (authored by brooks).
devel/llvm19: 19.1.2 release
Wed, Oct 16, 12:37 AM
brooks committed R11:ffb78cd00420: devel/llvm-{cheri,cheriot,morello}: update (authored by brooks).
devel/llvm-{cheri,cheriot,morello}: update
Wed, Oct 16, 12:37 AM
emaste added a comment to D42247: gpart: Add warning when the start sector is too low..

Sorry I didn't spot this / comment until after commit -- shouldn't this go to stderr? perhaps warnx?

Wed, Oct 16, 12:28 AM
imp committed rG2cbda736cea8: gpart: Add warning when the start sector is too low. (authored by imp).
gpart: Add warning when the start sector is too low.
Wed, Oct 16, 12:15 AM
imp closed D42247: gpart: Add warning when the start sector is too low..
Wed, Oct 16, 12:15 AM

Yesterday

imp abandoned D28585: nvme: Add optimizations for device removal.

OBE. Even when I had this in, it only helped a little and failure is more nuanced than this patch contemplates.

Tue, Oct 15, 11:57 PM
imp abandoned D40525: camcontrol: Add support for -r XX to set reserved bits in defects.

This turns out not to be needed for the thing the vendor was telling me to do. Their firmware didn't behave like they thought it would, and the solution is along a different path. So, rather than add a useless feature, I'm punting.

Tue, Oct 15, 11:55 PM
imp abandoned D34884: cam: debug for lost and found.

OBE long agi. While this still applies, it's kinda useless. The bugs it was chasing have been retired and what's left is in no shape to commit.

Tue, Oct 15, 11:52 PM