Re: "Oddness" in head since around r343678 or so

2019-02-08 Thread Alexander Leidinger 

Hi,

I recently noticed some generic slowness myself. I experienced this during 
replacing disks in a raidz by bigger ones. Long story short, check top -s 
if you have vnlru running for a long period at high CPU... If yes increase 
kern.maxvnodes (I increased to 10 times). Note, we should improve the admin 
page in the FAQ, the vnlru entry could need a little bit more hints and 
explanations.


If you encounter the same issue we have probably introduced a change 
somewhere with an unintended side effect.


Bye,
Alexander.

--
Send from a mobile device, please forgive brevity and misspellings.

Am 7. Februar 2019 4:43:43 nachm. schrieb "Rodney W. Grimes" 
:



I run and track stable/11, stable/12, and head (from separate slices --
no VMs involved) on my laptop; among other things, this permits some
degree of comparison among them.

The laptop is configured to run xdm; while running stable/11 or
stable/12, there's a period of about 5 seconds after xdm's login banner
shows up before either the mouse or keyboard responds.

Up to a few days ago, head was the same, but as of about last weekend (I
think), that period is about 40 seconds while running head.

I realize that's a little ... lacking in specificity, and I apologize
for that.  And while I noticed it a few days ago, I hadn't actually
timed it until this morning.


My addition is going to lack even worse, I have been seeing very
long periods of lag, on this 40 second order, when a bhyve instance
is started.  The interactive response for this time is 0, not
even ^t produces any output.  The prompt comes back almost immediatly
after typing the bhyve command, but that prompt is dead for a
very long time.

I am running 12.0-RELEASE, and see this issue on multiple machines.



Information, including verbose dmesg.boot copies and a log of uname
outputs, may be found from the page at
.

Here's a list of the uname outputs for head over the last several days:

FreeBSD 13.0-CURRENT #245 r343551M/343551: Tue Jan 29 03:54:46 PST 2019 
r...@g1-49.catwhisker.org:/common/S4/obj/usr/src/amd64.amd64/sys/CANARY  
amd64 1300010
FreeBSD 13.0-CURRENT #246 r343572M/343572: Wed Jan 30 05:14:44 PST 2019 
r...@g1-49.catwhisker.org:/common/S4/obj/usr/src/amd64.amd64/sys/CANARY  
amd64 1300010
FreeBSD 13.0-CURRENT #247 r343604M/343604: Thu Jan 31 03:53:37 PST 2019 
r...@g1-49.catwhisker.org:/common/S4/obj/usr/src/amd64.amd64/sys/CANARY  
amd64 1300010
FreeBSD 13.0-CURRENT #248 r343646M/343658: Fri Feb  1 04:05:45 PST 2019 
r...@g1-49.catwhisker.org:/common/S4/obj/usr/src/amd64.amd64/sys/CANARY  
amd64 1300010
FreeBSD 13.0-CURRENT #249 r343678M/343678: Sat Feb  2 03:55:23 PST 2019 
r...@g1-49.catwhisker.org:/common/S4/obj/usr/src/amd64.amd64/sys/CANARY  
amd64 1300010
FreeBSD 13.0-CURRENT #250 r343707M/343711: Sun Feb  3 04:06:08 PST 2019 
r...@g1-49.catwhisker.org:/common/S4/obj/usr/src/amd64.amd64/sys/CANARY  
amd64 1300010
FreeBSD 13.0-CURRENT #251 r343732M/343741: Mon Feb  4 04:24:07 PST 2019 
r...@g1-49.catwhisker.org:/common/S4/obj/usr/src/amd64.amd64/sys/CANARY  
amd64 1300010
FreeBSD 13.0-CURRENT #252 r343770M/343771: Tue Feb  5 04:27:38 PST 2019 
r...@g1-49.catwhisker.org:/common/S4/obj/usr/src/amd64.amd64/sys/CANARY  
amd64 1300010
FreeBSD 13.0-CURRENT #253 r343828M/343834: Wed Feb  6 03:59:36 PST 2019 
r...@g1-49.catwhisker.org:/common/S4/obj/usr/src/amd64.amd64/sys/CANARY  
amd64 1300010
FreeBSD 13.0-CURRENT #254 r343867M/343867: Thu Feb  7 04:52:38 PST 2019 
r...@g1-49.catwhisker.org:/common/S4/obj/usr/src/amd64.amd64/sys/CANARY  
amd64 1300010


Peace,
david
--
David H. Wolfskill  da...@catwhisker.org
"It was like reading a Donald Trump tweet.  It didn't make any sense."

See http://www.catwhisker.org/~david/publickey.gpg for my public key.


--
Rod Grimes rgri...@freebsd.org
___
freebsd-current@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"




___
freebsd-current@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Re: r343843 /usr/ports/ make install error: unable to create target: 'No available targets are compatible with this triple.'

2019-02-08 Thread Graham Perrin 

On 07/02/2019 06:43, I wrote:

> Follow-up to
> 


>
>> r343663 poudriere can not configure pkg
>
> It seems, the problems are not limited to my poudriere environment.
>
> Also, it's not possible to install from /usr/ports
> …

The lost ability to install from /usr/ports was quite show-stopping:

1) I could not build things such as gpu-firmware-kmod or
   drm-legacy-kmod with the updated system; and

2) in the absence of properly built kernel modules, I could
   not start X.

Apparently worked around by removing all LLVM-related lines from
/etc/make.conf
then build and installation of r343883.

Re: 
I still wonder whether these problems arose, somehow, from:

235215 – Bump LLVM_DEFAULT to 80




Now:

root@momh167-gjp4-8570p:~ # grep -v \# /etc/make.conf
PORTS_MODULES=graphics/gpu-firmware-kmod graphics/drm-legacy-kmod
DEFAULT_VERSIONS+=samba=4.8

root@momh167-gjp4-8570p:~ #
___
freebsd-current@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Re: r343843 /usr/ports/ make install error: unable to create target: 'No available targets are compatible with this triple.'

2019-02-08 Thread Mark Linimon 
On Fri, Feb 08, 2019 at 12:18:33PM +, Graham Perrin wrote:
> I still wonder whether these problems arose, somehow, from:
> 
> 235215 – Bump LLVM_DEFAULT to 80
> 

AFAIK that's still being tested (not committed yet).

mcl
___
freebsd-current@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Re: r343843 /usr/ports/ make install error: unable to create target: 'No available targets are compatible with this triple.'

2019-02-08 Thread Graham Perrin 

On 08/02/2019 13:54, Mark Linimon wrote:

On Fri, Feb 08, 2019 at 12:18:33PM +, Graham Perrin wrote:

I still wonder whether these problems arose, somehow, from:

235215 – Bump LLVM_DEFAULT to 80


AFAIK that's still being tested (not committed yet).

mcl


Yep, I was testing. That's me at comments 14 and 15.

Still am, to a degree.

___
freebsd-current@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"