Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1520614
| From | Jonathan Corbet <corbet@lwn.net> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | Re: [PATCH] Change the document about iowait |
| Date | 2016-11-13 16:50 +0100 |
| Message-ID | <sD5cB-27f-3@gated-at.bofh.it> (permalink) |
| References | <swmrT-5VS-5@gated-at.bofh.it> |
| Organization | LWN.net |
On Wed, 26 Oct 2016 10:41:28 +0800 Chao Fan <fanc.fnst@cn.fujitsu.com> wrote: > The iowait is not reliable by reading from /proc/stat, so this > method to get iowait is not suggested. And we mark it in the > document. Sorry for the delay on this. Life has been...challenging... > Signed-off-by: Cao Jin <caoj.fnst@cn.fujitsu.com> > Signed-off-by: Chao Fan <fanc.fnst@cn.fujitsu.com> What does this signoff chain mean? > --- > Documentation/filesystems/proc.txt | 11 ++++++++++- > 1 file changed, 10 insertions(+), 1 deletion(-) > > diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt > index 74329fd..71f5096 100644 > --- a/Documentation/filesystems/proc.txt > +++ b/Documentation/filesystems/proc.txt > @@ -1305,7 +1305,16 @@ second). The meanings of the columns are as follows, from left to right: > - nice: niced processes executing in user mode > - system: processes executing in kernel mode > - idle: twiddling thumbs > -- iowait: waiting for I/O to complete > +- iowait: In a word, iowait stands for waiting for I/O to complete. But there > + are several problems: > + 1. Cpu will not wait for I/O to complete, iowait is the time that a task is > + waiting for I/O to complete. When cpu goes into idle state for > + outstanding task io, another task will be scheduled on this CPU. > + 2. In a multi-core CPU, the task waiting for I/O to complete is not running > + on any CPU, so the iowait of each CPU is difficult to calculate. > + 3. The value of iowait field in /proc/stat will decrease in certain > + conditions. > + So, the iowait is not reliable by reading from /proc/stat. > - irq: servicing interrupts So I suppose I can apply this. But is there any chance of making it say what iowait actually measures, rather than just saying that it's unreliable? Thanks, jon
Back to linux.kernel | Previous | Next — Next in thread | Find similar | Unroll thread
Re: [PATCH] Change the document about iowait Jonathan Corbet <corbet@lwn.net> - 2016-11-13 16:50 +0100
Re: [PATCH] Change the document about iowait Chao Fan <fanc.fnst@cn.fujitsu.com> - 2016-11-14 04:50 +0100
Re: [PATCH] Change the document about iowait Jonathan Corbet <corbet@lwn.net> - 2016-11-19 18:40 +0100
csiph-web