From: Nicholas Lee <nic-lists@plumtree.co.nz>
To: Ian Pratt <m+Ian.Pratt@cl.cam.ac.uk>,
xen-devel@lists.sourceforge.net, ian.pratt@cl.cam.ac.uk
Subject: Re: Interrupt levels
Date: Wed, 9 Mar 2005 00:45:58 +1300 [thread overview]
Message-ID: <20050308114552.GA366@plumtree.co.nz> (raw)
In-Reply-To: <20050308085053.GB1419@plumtree.co.nz>
Digging into this a little more.
Once I've shut down domUs and a lot of the services (including xend) on
dom0, the interrupt level is unchanged and still high.
nic@stateless:~$ vmstat 3
procs -----------memory---------- ---swap-- -----io---- --system-- ----cpu----
r b swpd free buff cache si so bi bo in cs us sy id wa
0 0 0 98512 104 21328 0 0 53 16 71426 49 0 1 97 1
0 0 0 98520 104 21328 0 0 0 1 183415 16 0 0 100 0
0 0 0 98520 104 21328 0 0 0 0 185419 10 0 0 100 0
0 0 0 98520 104 21328 0 0 0 0 183671 12 0 0 100 0
0 0 0 97872 104 21328 0 0 0 67 170898 47 0 0 100 0
....
So obiovusly the domUs aren't causing the problem.
In fact in a new rebooted machine with hardly anything running the same
is occuring. Seems like some spinlock out of control.
Now if I start compiling xen (make in xen-2.0bk) the following occurs:
nic@stateless:~$ vmstat 3
....
0 1 0 3296 32 102832 0 0 2653 53 73099 658 20 17 11 53
0 1 0 3168 32 101036 0 0 839 5092 71982 558 26 25 5 45
3 1 0 3616 32 97076 0 0 5101 139 36210 811 32 37 2 29
0 1 0 26904 32 95108 0 0 1665 263 67692 832 27 25 3 45
0 1 0 8408 32 99000 0 0 1297 129 55256 399 26 13 8 53
0 1 0 4696 32 98140 0 0 396 83 22446 277 57 19 4 20
1 0 0 3416 32 98476 0 0 795 1169 88617 436 22 12 4 62
1 0 0 19800 32 91668 0 0 364 21 18907 178 65 23 0 12
1 0 0 22552 32 88176 0 0 424 470 14323 204 65 23 1 11
1 0 0 23768 32 86052 0 0 140 283 3218 181 77 19 0 4
1 0 0 35736 32 86260 0 0 28 244 215 119 71 27 0 2
1 0 0 22040 32 86572 0 0 71 475 14472 148 72 17 4 7
3 0 0 21216 32 86272 0 0 85 55 592 239 67 32 0 1
1 0 0 18528 32 86452 0 0 31 1238 19669 131 59 29 10 2
1 0 0 16816 32 86988 0 0 79 4721 9065 106 66 25 1 8
1 0 0 15664 32 81284 0 0 16 43 1786 93 79 20 0 1
1 0 0 26048 32 81452 0 0 25 53 325 81 71 29 0 0
1 0 0 17616 32 81624 0 0 33 21 1337 115 87 13 0 1
1 0 0 41056 32 77984 0 0 73 326 1898 140 79 18 0 2
1 0 0 25248 32 78144 0 0 32 233 503 87 87 13 0 1
1 0 0 33832 32 78304 0 0 19 11 116 58 78 22 0 0
1 0 0 45736 32 78532 0 0 43 114 1477 75 80 18 0 1
1 0 0 40680 32 78760 0 0 43 32 215 63 78 22 0 0
1 0 0 29224 32 78924 0 0 25 162 250 88 79 21 0 1
1 0 0 39976 32 79124 0 0 29 269 5719 90 70 26 3 0
procs -----------memory---------- ---swap-- -----io---- --system-- ----cpu----
r b swpd free buff cache si so bi bo in cs us sy id wa
2 0 0 38816 32 80432 0 0 404 24 9220 119 68 22 1 9
1 0 0 16672 32 80984 0 0 32 90 2226 82 75 23 0 2
1 0 0 23840 32 81788 0 0 213 21 4700 106 75 18 0 7
1 0 0 31136 32 82100 0 0 52 98 3469 88 74 23 0 3
1 0 0 32288 32 82356 0 0 43 251 3546 123 76 22 0 2
Interrupts drop down:
nic@stateless:~$ diff -u interrupts.3 interrupts.4
--- interrupts.3 2005-03-09 00:37:51.000000000 +1300
+++ interrupts.4 2005-03-09 00:37:56.000000000 +1300
@@ -2,26 +2,26 @@
1: 9 Phys-irq i8042
8: 4 Phys-irq rtc
15: 11 Phys-irq ide1
- 22: 45384 Phys-irq ioc0
- 24: 82320 Phys-irq eth0
+ 22: 45707 Phys-irq ioc0
+ 24: 82403 Phys-irq eth0
128: 1 Dynamic-irq misdirect
129: 102 Dynamic-irq ctrl-if
-130: 121707669 Dynamic-irq timer
+130: 121719090 Dynamic-irq timer
131: 0 Dynamic-irq console
132: 0 Dynamic-irq net-be-dbg
-133: 724 Dynamic-irq blkif-backend
+133: 726 Dynamic-irq blkif-backend
134: 25 Dynamic-irq vif1.0
135: 177 Dynamic-irq vif1.1
-136: 2352 Dynamic-irq blkif-backend
-137: 4300 Dynamic-irq vif2.0
+136: 2586 Dynamic-irq blkif-backend
+137: 5008 Dynamic-irq vif2.0
138: 1 Dynamic-irq vif2.1
139: 1936 Dynamic-irq blkif-backend
140: 2 Dynamic-irq vif5.0
-141: 4439 Dynamic-irq vif5.1
+141: 5151 Dynamic-irq vif5.1
142: 626 Dynamic-irq blkif-backend
-143: 451 Dynamic-irq vif4.0
-144: 640 Dynamic-irq vif4.1
-145: 901 Dynamic-irq blkif-backend
+143: 452 Dynamic-irq vif4.0
+144: 641 Dynamic-irq vif4.1
+145: 904 Dynamic-irq blkif-backend
146: 3 Dynamic-irq vif6.0
147: 97 Dynamic-irq vif6.1
NMI: 0
then start back up with the compile job is completed:
1 0 0 29848 56 67280 0 0 452 203 6867 397 62 34 0 4
1 0 0 29208 56 67520 0 0 451 33 11491 382 51 43 1 4
2 0 0 32472 56 67664 0 0 399 205 3197 370 53 46 0 2
1 0 0 29528 56 67908 0 0 416 64 5640 354 43 53 0 4
1 0 0 31320 56 68396 0 0 328 218 1633 326 46 53 1 1
1 0 0 22488 56 68612 0 0 192 183 2367 215 57 40 2 1
1 0 0 24984 56 68688 0 0 1987 61 2169 186 40 59 0 1
1 0 0 29976 56 69716 0 0 1805 161 3892 292 45 53 2 0
1 0 0 28136 56 70052 0 0 1972 108 11184 318 40 54 1 5
1 0 0 27504 56 70256 0 0 2363 660 4377 305 31 66 1 2
1 0 0 28144 56 70360 0 0 2396 186 2604 273 29 68 1 2
1 0 0 22256 56 70512 0 0 2277 396 1514 254 31 67 0 2
1 0 0 31664 56 70640 0 0 1860 482 11381 245 23 70 0 6
3 0 0 34736 56 70972 0 0 408 133 10282 352 36 58 3 2
1 0 0 17200 56 80856 0 0 647 566 45363 510 40 31 8 22
0 1 0 3072 56 98700 0 0 521 2509 56155 453 49 16 20 16
0 0 0 2752 56 99360 0 0 109 152 173391 94 0 1 83 16
Seems like a bug to me.
Nicholas
-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
next prev parent reply other threads:[~2005-03-08 11:45 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-03-08 7:52 Interrupt levels Ian Pratt
2005-03-08 8:50 ` Nicholas Lee
2005-03-08 9:14 ` Keir Fraser
2005-03-08 9:45 ` Nicholas Lee
2005-03-08 11:45 ` Nicholas Lee [this message]
-- strict thread matches above, loose matches on Subject: below --
2005-03-10 2:28 Ian Pratt
2005-03-10 5:41 ` Nicholas Lee
2005-03-10 10:47 ` James Bulpin
2005-03-09 22:24 Ian Pratt
2005-03-10 1:56 ` Nicholas Lee
2005-03-09 21:33 Ian Pratt
2005-03-09 22:12 ` Nicholas Lee
2005-03-09 15:20 Ian Pratt
2005-03-09 20:03 ` Nicholas Lee
2005-03-09 9:38 Ian Pratt
2005-03-09 10:59 ` Nicholas Lee
2005-03-09 8:47 Ian Pratt
2005-03-09 9:22 ` Nicholas Lee
2005-03-09 8:03 Ian Pratt
2005-03-09 8:17 ` Nicholas Lee
2005-03-08 23:24 Ian Pratt
2005-03-08 12:50 Ian Pratt
2005-03-09 0:48 ` Nicholas Lee
2005-03-09 3:32 ` Nicholas Lee
2005-03-08 11:07 Ian Pratt
2005-03-08 23:17 ` Nicholas Lee
2005-03-08 23:22 ` Nivedita Singhvi
2005-03-09 0:38 ` Nicholas Lee
2005-03-08 6:52 Nicholas Lee
2005-03-08 21:58 ` Kurt Garloff
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20050308114552.GA366@plumtree.co.nz \
--to=nic-lists@plumtree.co.nz \
--cc=ian.pratt@cl.cam.ac.uk \
--cc=m+Ian.Pratt@cl.cam.ac.uk \
--cc=xen-devel@lists.sourceforge.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.