Date: Tue, 28 Feb 2012 17:31:40 GMT From: Eugen Konkov <kes-kes@yandex.ru> To: freebsd-gnats-submit@FreeBSD.org Subject: misc/165536: netisr bad distribution between CPUs Message-ID: <201202281731.q1SHVe8c060456@red.freebsd.org> Resent-Message-ID: <201202281740.q1SHeBQf034693@freefall.freebsd.org>
next in thread | raw e-mail | index | archive | help
>Number: 165536
>Category: misc
>Synopsis: netisr bad distribution between CPUs
>Confidential: no
>Severity: non-critical
>Priority: low
>Responsible: freebsd-bugs
>State: open
>Quarter:
>Keywords:
>Date-Required:
>Class: sw-bug
>Submitter-Id: current-users
>Arrival-Date: Tue Feb 28 17:40:11 UTC 2012
>Closed-Date:
>Last-Modified:
>Originator: Eugen Konkov
>Release: 9.0-CURRENT
>Organization:
ISP FreeLine
>Environment:
FreeBSD flux2 9.0-CURRENT FreeBSD 9.0-CURRENT #4: Fri Jun 10 01:30:12 UTC 2011
>Description:
As you see two of four threads get about 100% cpu usage and other two only 30%
# top -SIHP
last pid: 43373; load averages: 5.14, 5.58, 5.79 up 0+06:25:33 19:15:20
293 processes: 6 running, 268 sleeping, 17 waiting, 2 lock
CPU 0: 3.6% user, 0.0% nice, 24.6% system, 42.0% interrupt, 29.7% idle
CPU 1: 3.6% user, 0.0% nice, 14.5% system, 64.5% interrupt, 17.4% idle
CPU 2: 0.0% user, 0.0% nice, 10.9% system, 77.5% interrupt, 11.6% idle
CPU 3: 0.7% user, 0.0% nice, 10.9% system, 70.3% interrupt, 18.1% idle
Mem: 515M Active, 1309M Inact, 230M Wired, 260K Cache, 112M Buf, 1844M Free
Swap: 32G Total, 32G Free
PID USERNAME PRI NICE SIZE RES STATE C TIME WCPU COMMAND
12 root -72 - 0K 160K *per-i 2 223:06 99.61% {swi1: netisr 0}
12 root -72 - 0K 160K CPU1 1 180:02 88.43% {swi1: netisr 2}
12 root -72 - 0K 160K *per-i 0 65:57 34.72% {swi1: netisr 1}
12 root -72 - 0K 160K WAIT 3 65:10 32.52% {swi1: netisr 3}
11 root 155 ki31 0K 32K CPU0 0 199:44 31.35% {idle: cpu0}
11 root 155 ki31 0K 32K RUN 1 181:36 18.02% {idle: cpu1}
11 root 155 ki31 0K 32K RUN 3 174:51 14.31% {idle: cpu3}
11 root 155 ki31 0K 32K CPU2 2 174:50 13.43% {idle: cpu2}
12 root -92 - 0K 160K WAIT 0 45:35 12.89% {irq256: re0}
13 root -16 - 0K 32K sleep 0 17:41 4.93% {ng_queue2}
13 root -16 - 0K 32K sleep 2 17:40 4.74% {ng_queue1}
13 root -16 - 0K 32K sleep 0 17:44 4.69% {ng_queue3}
13 root -16 - 0K 32K sleep 0 17:41 4.59% {ng_queue0}
# netstat -Q
Configuration:
Setting Current Limit
Thread count 4 4
Default queue limit 256 10240
Direct dispatch disabled n/a
Forced direct dispatch disabled n/a
Threads bound to CPUs disabled n/a
Protocols:
Name Proto QLimit Policy Flags
ip 1 1024 flow ---
igmp 2 256 source ---
rtsock 3 256 source ---
arp 7 256 source ---
ip6 10 256 flow ---
Workstreams:
WSID CPU Name Len WMark Disp'd HDisp'd QDrops Queued Handled
0 0 ip 493 1024 0 0 7217665 377961834 377960765
0 0 igmp 0 0 0 0 0 0 0
0 0 rtsock 0 256 0 0 113 27218 27218
0 0 arp 0 12 0 0 0 36476 36476
0 0 ip6 0 5 0 0 0 191 191
1 1 ip 0 786 0 0 0 103683774 103683774
1 1 igmp 0 0 0 0 0 0 0
1 1 rtsock 0 0 0 0 0 0 0
1 1 arp 0 3 0 0 0 16403 16403
1 1 ip6 0 1 0 0 0 165 165
2 2 ip 33 1024 0 0 20062 406492881 406492832
2 2 igmp 0 0 0 0 0 0 0
2 2 rtsock 0 0 0 0 0 0 0
2 2 arp 0 9 0 0 0 58454 58454
2 2 ip6 0 1 0 0 0 191 191
3 3 ip 1 1024 0 0 8312 267733875 267733873
3 3 igmp 0 0 0 0 0 0 0
3 3 rtsock 0 0 0 0 0 0 0
3 3 arp 0 10 0 0 0 135230 135230
3 3 ip6 0 3 0 0 0 3107 3107
>How-To-Repeat:
>Fix:
>Release-Note:
>Audit-Trail:
>Unformatted:
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201202281731.q1SHVe8c060456>