From: "Ralph Böhme" <slow@samba.org>
To: Milian Wolff <milian.wolff@kdab.com>
Cc: linux-perf-users@vger.kernel.org
Subject: Re: Callgraphs split at the kernel boundary
Date: Mon, 4 Sep 2017 19:01:27 +0200 [thread overview]
Message-ID: <20170904170127.yuofaew6orzrxl5e@kazak> (raw)
In-Reply-To: <4314827.rDNsEPmXDr@milian-kdab2>
On Mon, Sep 04, 2017 at 06:44:23PM +0200, Milian Wolff wrote:
> In my case, it seems to help to remove the filter. I.e. by default perf
> nowadays uses `cycles:uppp` on my system. If I instead specify `cycles:P`
> explicitly at `perf record` time, the issue above seems to be reduced, albeit
> not completely. Does this help in your situation too?
nope, I still mostly see split callgraphs, when recording with
# perf record -e cycles:P -g -p 30027
Eg (from perf script):
smbd 30027 66331.499304: 250000 cpu-clock:HG:
9a8233 tcp_chrono_start (/usr/lib/debug/lib/modules/4.12.8-300.fc26.x86_64/vmlinux)
997565 skb_entail (/usr/lib/debug/lib/modules/4.12.8-300.fc26.x86_64/vmlinux)
99ba5d tcp_sendmsg (/usr/lib/debug/lib/modules/4.12.8-300.fc26.x86_64/vmlinux)
9c81c8 inet_sendmsg (/usr/lib/debug/lib/modules/4.12.8-300.fc26.x86_64/vmlinux)
920f08 sock_sendmsg (/usr/lib/debug/lib/modules/4.12.8-300.fc26.x86_64/vmlinux)
920fac sock_write_iter (/usr/lib/debug/lib/modules/4.12.8-300.fc26.x86_64/vmlinux)
463e72 __do_readv_writev (/usr/lib/debug/lib/modules/4.12.8-300.fc26.x86_64/vmlinux)
463fdc do_readv_writev (/usr/lib/debug/lib/modules/4.12.8-300.fc26.x86_64/vmlinux)
464260 vfs_writev (/usr/lib/debug/lib/modules/4.12.8-300.fc26.x86_64/vmlinux)
4642e1 do_writev (/usr/lib/debug/lib/modules/4.12.8-300.fc26.x86_64/vmlinux)
465510 sys_writev (/usr/lib/debug/lib/modules/4.12.8-300.fc26.x86_64/vmlinux)
a77737 entry_SYSCALL_64 (/usr/lib/debug/lib/modules/4.12.8-300.fc26.x86_64/vmlinux)
106680 __GI___writev (/usr/lib64/libc-2.25.so)
c846d36d60 [unknown] ([unknown])
c846d36f10 [unknown] ([unknown])
65742f746e657665 [unknown] ([unknown])
Also wondering about the bottom three unresolved symbols... *scratches head*
Cheerio!
-slow
next prev parent reply other threads:[~2017-09-04 17:08 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-09-03 9:49 Callgraphs split at the kernel boundary Ralph Böhme
2017-09-04 16:26 ` Milian Wolff
2017-09-04 16:44 ` Milian Wolff
2017-09-04 17:01 ` Ralph Böhme [this message]
2017-09-04 17:22 ` Milian Wolff
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=20170904170127.yuofaew6orzrxl5e@kazak \
--to=slow@samba.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=milian.wolff@kdab.com \
/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