From: Avi Kivity <avi@scylladb.com>
To: "Frank Ch. Eigler" <fche@redhat.com>
Cc: Mark Wielaard <mjw@redhat.com>, David Smith <dsmith@redhat.com>,
systemtap@sourceware.org
Subject: Re: Some newbie questions
Date: Wed, 17 Aug 2016 18:30:00 -0000 [thread overview]
Message-ID: <8af2d587-ebc0-730b-3b1c-a452c486af57@scylladb.com> (raw)
In-Reply-To: <y0minuze060.fsf@fche.csb>
On 08/17/2016 09:10 PM, Frank Ch. Eigler wrote:
> avi wrote:
>
>> [...]
>> My load indeed calls epoll_wait() at a high rate, with zero timeout.
> OK. That's an unusual and cpu-saturating method. Without systemtap
> active, the code must be doing millions of syscalls per second. Are
> you sure there's no better way?
Yes. It's not very common, but some high throughput or latency
sensitive applications will poll. In my case it's both high throughput
and latency sensitive.
In any case, asking users to do fewer syscalls because it hurts
systemtap will not work.
>
>> But now it's completely dominated by tracing overhead.
>> [...]
>> | |--48.48%--task_utrace_struct
>> | | |--25.21%--utrace_report_syscall_entry
>> | | | syscall_trace_enter_phase2
>> | | | syscall_trace_enter
>> | | | do_syscall_64
>> | | | return_from_SYSCALL_64
>> | | | |
>> | | | --23.47%--epoll_pwait
>> [...]
> OK, this implicates the systemtap runtime utrace-emulation code that
Why is it emulating? Is there something missing in Kconfig? This is a
4.6 Fedora kernel, so I expect the kitchen sink has been compiled in.
> looks for mmap/exec syscalls. Perhaps there could be a mode (a -D
> setting?) for selecting kprobes rather than syscall-tracepoints to
> catch those events.
>
Why not trace just mmap and exec syscalls, rather than everything?
Surely there's a static tracepoint for those.
I'd say defaulting to taking a spinlock on every syscall is going to
cause anyone who's using more than a laptop a lot of pain.
next prev parent reply other threads:[~2016-08-17 18:30 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-08-07 8:08 Avi Kivity
2016-08-07 9:16 ` Avi Kivity
2016-08-08 14:58 ` David Smith
2016-08-09 6:51 ` Avi Kivity
2016-08-09 13:54 ` Mark Wielaard
2016-08-09 14:11 ` Avi Kivity
2016-08-10 15:33 ` Frank Ch. Eigler
2016-08-10 15:39 ` Avi Kivity
2016-08-10 15:40 ` Avi Kivity
2016-08-10 16:47 ` Frank Ch. Eigler
2016-08-11 7:51 ` Avi Kivity
2016-08-11 8:13 ` Avi Kivity
2016-08-11 16:18 ` Frank Ch. Eigler
2016-08-17 15:15 ` Avi Kivity
2016-08-17 18:06 ` David Smith
2016-08-17 18:22 ` Avi Kivity
2016-08-17 18:10 ` Frank Ch. Eigler
2016-08-17 18:30 ` Avi Kivity [this message]
2016-08-17 18:35 ` Frank Ch. Eigler
2016-08-17 18:42 ` Avi Kivity
2016-08-25 16:21 ` Avi Kivity
2016-08-25 17:37 ` David Smith
2016-08-26 18:39 ` Avi Kivity
2016-08-26 20:08 ` David Smith
2016-08-26 21:21 ` Josh Stone
2016-08-28 13:30 ` Avi Kivity
2016-08-28 13:35 ` Avi Kivity
2016-08-28 13:45 ` Avi Kivity
2016-08-29 18:26 ` David Smith
2016-08-31 16:44 ` Avi Kivity
2016-08-09 15:17 ` David Smith
2016-08-09 15:28 ` Avi Kivity
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=8af2d587-ebc0-730b-3b1c-a452c486af57@scylladb.com \
--to=avi@scylladb.com \
--cc=dsmith@redhat.com \
--cc=fche@redhat.com \
--cc=mjw@redhat.com \
--cc=systemtap@sourceware.org \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).