From: Andi Kleen <andi@firstfloor.org>
To: Mike Galbraith <efault@gmx.de>
Cc: Andi Kleen <andi@firstfloor.org>,
Josh Poimboeuf <jpoimboe@redhat.com>,
x86@kernel.org, linux-kernel@vger.kernel.org,
live-patching@vger.kernel.org,
Linus Torvalds <torvalds@linux-foundation.org>,
Andy Lutomirski <luto@kernel.org>, Jiri Slaby <jslaby@suse.cz>,
Ingo Molnar <mingo@kernel.org>, "H. Peter Anvin" <hpa@zytor.com>,
Peter Zijlstra <peterz@infradead.org>
Subject: Re: [PATCH v3 00/10] x86: ORC unwinder (previously undwarf)
Date: Wed, 12 Jul 2017 21:15:25 -0700 [thread overview]
Message-ID: <20170713041524.GC3044@two.firstfloor.org> (raw)
In-Reply-To: <1499914980.5107.18.camel@gmx.de>
On Thu, Jul 13, 2017 at 05:03:00AM +0200, Mike Galbraith wrote:
> On Wed, 2017-07-12 at 15:30 -0700, Andi Kleen wrote:
> > Josh Poimboeuf <jpoimboe@redhat.com> writes:
> > >
> > > The ORC data format does have a few downsides compared to DWARF. The
> > > ORC unwind tables take up ~1MB more memory than DWARF eh_frame tables.
> > >
> > Can we have an option to just use dwarf instead? For people
> > who don't want to waste a MB+ to solve a problem that doesn't
> > exist (as proven by many years of opensuse kernel experience)
>
> Sure the dwarf unwinder works well for crashes, but at the price of
> demolishing ftrace/perf utility.
You mean the unwind performance?
That's a valid concern, but neither ORC nor dwarf are likely
to address it. However most usages of ftrace/perf shouldn't be that
depending on unwind performance -- just lower the frequency of your
events.
The only possible win is if the win from not using FP code is
significant enough. On the x86 side the only modern CPUs that should really
care about this are Atoms.
-Andi
next prev parent reply other threads:[~2017-07-13 4:15 UTC|newest]
Thread overview: 60+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-07-11 15:33 [PATCH v3 00/10] x86: ORC unwinder (previously undwarf) Josh Poimboeuf
2017-07-11 15:33 ` [PATCH v3 01/10] x86/entry/64: Refactor IRQ stacks and make them NMI-safe Josh Poimboeuf
2017-07-18 10:40 ` [tip:x86/asm] " tip-bot for Andy Lutomirski
2017-07-11 15:33 ` [PATCH v3 02/10] x86/entry/64: Initialize the top of the IRQ stack before switching stacks Josh Poimboeuf
2017-07-18 10:41 ` [tip:x86/asm] " tip-bot for Andy Lutomirski
2017-07-11 15:33 ` [PATCH v3 03/10] x86/dumpstack: fix occasionally missing registers Josh Poimboeuf
2017-07-18 10:41 ` [tip:x86/asm] x86/dumpstack: Fix " tip-bot for Josh Poimboeuf
2017-07-11 15:33 ` [PATCH v3 04/10] x86/dumpstack: fix interrupt and exception stack boundary checks Josh Poimboeuf
2017-07-18 10:42 ` [tip:x86/asm] x86/dumpstack: Fix " tip-bot for Josh Poimboeuf
2017-07-11 15:33 ` [PATCH v3 05/10] objtool: add ORC unwind table generation Josh Poimboeuf
2017-07-18 10:42 ` [tip:x86/asm] objtool: Add " tip-bot for Josh Poimboeuf
2017-07-11 15:33 ` [PATCH v3 06/10] objtool, x86: add facility for asm code to provide unwind hints Josh Poimboeuf
2017-07-18 10:43 ` [tip:x86/asm] objtool, x86: Add " tip-bot for Josh Poimboeuf
2017-07-11 15:33 ` [PATCH v3 07/10] x86/entry/64: add unwind hint annotations Josh Poimboeuf
2017-07-18 10:43 ` [tip:x86/asm] x86/entry/64: Add " tip-bot for Josh Poimboeuf
2017-07-11 15:33 ` [PATCH v3 08/10] x86/asm: add unwind hint annotations to sync_core() Josh Poimboeuf
2017-07-18 10:43 ` [tip:x86/asm] x86/asm: Add " tip-bot for Josh Poimboeuf
2017-07-11 15:33 ` [PATCH v3 09/10] x86/unwind: add ORC unwinder Josh Poimboeuf
2017-07-14 17:22 ` [PATCH v3.1 " Josh Poimboeuf
2017-07-20 7:12 ` Jiri Slaby
2017-07-20 21:16 ` Josh Poimboeuf
2017-07-11 15:33 ` [PATCH v3 10/10] x86/kconfig: make it easier to switch to the new " Josh Poimboeuf
2017-07-12 8:27 ` [PATCH v3 00/10] x86: ORC unwinder (previously undwarf) Ingo Molnar
2017-07-12 14:42 ` Josh Poimboeuf
2017-07-12 19:27 ` Ingo Molnar
2017-07-14 17:17 ` Josh Poimboeuf
2017-07-25 9:09 ` Ingo Molnar
2017-07-25 17:58 ` Josh Poimboeuf
2017-07-25 18:46 ` Kees Cook
2017-07-12 21:49 ` Andres Freund
2017-07-12 22:32 ` Josh Poimboeuf
2017-07-12 22:36 ` Andres Freund
2017-07-12 22:40 ` Josh Poimboeuf
2017-07-12 22:54 ` Andres Freund
2017-07-13 7:12 ` Peter Zijlstra
2017-07-13 8:50 ` Peter Zijlstra
2017-07-13 8:51 ` Peter Zijlstra
2017-07-13 9:19 ` Ingo Molnar
2017-07-13 12:17 ` Josh Poimboeuf
2017-07-13 12:21 ` Josh Poimboeuf
2017-07-13 12:35 ` Josh Poimboeuf
2017-07-14 8:33 ` Ingo Molnar
2017-07-14 8:29 ` Ingo Molnar
2017-07-25 11:55 ` [RFC] perf: Delayed userspace unwind (Was: [PATCH v3 00/10] x86: ORC unwinder) Peter Zijlstra
2017-07-28 14:13 ` Jiri Olsa
2017-07-28 14:21 ` Peter Zijlstra
2017-07-29 3:35 ` Andy Lutomirski
2017-07-29 9:28 ` Peter Zijlstra
2017-07-12 22:30 ` [PATCH v3 00/10] x86: ORC unwinder (previously undwarf) Andi Kleen
2017-07-12 22:47 ` Josh Poimboeuf
2017-07-13 4:29 ` Andi Kleen
2017-07-13 13:15 ` Josh Poimboeuf
2017-07-13 9:29 ` Ingo Molnar
2017-07-12 23:22 ` Andy Lutomirski
2017-07-13 3:03 ` Mike Galbraith
2017-07-13 4:15 ` Andi Kleen [this message]
2017-07-13 4:28 ` Mike Galbraith
2017-07-13 4:40 ` Andi Kleen
2017-07-13 5:22 ` Mike Galbraith
2017-07-13 12:02 ` Jiri Kosina
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=20170713041524.GC3044@two.firstfloor.org \
--to=andi@firstfloor.org \
--cc=efault@gmx.de \
--cc=hpa@zytor.com \
--cc=jpoimboe@redhat.com \
--cc=jslaby@suse.cz \
--cc=linux-kernel@vger.kernel.org \
--cc=live-patching@vger.kernel.org \
--cc=luto@kernel.org \
--cc=mingo@kernel.org \
--cc=peterz@infradead.org \
--cc=torvalds@linux-foundation.org \
--cc=x86@kernel.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