From: Peter Zijlstra <peterz@infradead.org>
To: Steven Rostedt <rostedt@goodmis.org>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>,
Thomas Gleixner <tglx@linutronix.de>,
Ingo Molnar <mingo@redhat.com>, "H. Peter Anvin" <hpa@zytor.com>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Linux Next Mailing List <linux-next@vger.kernel.org>,
mhiramat@kernel.org, ast@kernel.org, hjl.tools@gmail.com,
rick.p.edgecombe@intel.com, rppt@kernel.org,
linux-toolchains@vger.kernel.org, Andrew.Cooper3@citrix.com,
ndesaulniers@google.com
Subject: Re: linux-next: build warnings after merge of the tip tree
Date: Tue, 22 Mar 2022 16:48:02 +0100 [thread overview]
Message-ID: <Yjnvsrp8253bxWPA@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <20220322110438.25c2a760@gandalf.local.home>
On Tue, Mar 22, 2022 at 11:04:38AM -0400, Steven Rostedt wrote:
> > In recap:
> >
> > __fentry__ -- push on trace-stack
> > __ftail__ -- mark top-most entry complete
> > __fexit__ -- mark top-most entry complete;
> > pop all completed entries
>
> Again, this would require that the tail-calls are also being traced.
Which is why we should inhibit tail-calls if the function is notrace.
> > inhibit tail-calls to notrace.
>
> Just inhibiting tail-calls to notrace would work without any of the above.
I'm lost again; what? Without any of the above you got nothing because
return-trampoline will not work.
> But my fear is that will cause a noticeable performance impact.
Most code isn't in fact notrace, and call+ret aren't *that* expensive.
> > It's function graph tracing, kretprobes and whatever else this rethook
> > stuff is about that needs this because return trampolines will stop
> > working somewhere in the not too distant future.
>
> Another crazy solution is to have:
>
> func_A:
> call __fentry__
> ...
> tail: jmp 1f
> call 1f
> call __fexit__
> ret
> 1: jmp func_B
>
>
> where the compiler tells us about "tail:" and that we know that func_A ends
> with a tail call, and if we want to trace the end of func_A we convert that
> jmp 1f into a nop. And then we call the func_B and it's return comes back
> to where we call __fexit__ and then return normally.
At that point giving us something like:
1:
pushsection __ftail_loc
.long 1b - .
popsection
jmp.d32 func_B
call __fexit__
ret
is smaller and simpler, we can patch the jmp.d32 to call when tracing.
The only problem is SLS, that might wants an int3 after jmp too
( https://www.amd.com/en/corporate/product-security/bulletin/amd-sb-1026 ).
That does avoid the need for __ftail__ I suppose.
next prev parent reply other threads:[~2022-03-22 15:48 UTC|newest]
Thread overview: 93+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-03-21 3:03 linux-next: build warnings after merge of the tip tree Stephen Rothwell
2022-03-21 12:55 ` Peter Zijlstra
2022-03-21 13:04 ` Peter Zijlstra
2022-03-21 13:08 ` Peter Zijlstra
2022-03-21 13:45 ` Peter Zijlstra
2022-03-21 14:19 ` Mark Rutland
2022-03-21 15:28 ` Peter Zijlstra
2022-03-21 15:45 ` Peter Zijlstra
2022-03-21 16:37 ` Linus Torvalds
2022-03-21 16:44 ` Peter Zijlstra
2022-03-21 16:52 ` Linus Torvalds
2022-03-21 22:05 ` Stephen Rothwell
2022-03-21 22:12 ` Alexei Starovoitov
2022-03-21 22:46 ` Stephen Rothwell
2022-03-21 22:50 ` Alexei Starovoitov
2022-03-21 22:55 ` Steven Rostedt
2022-03-22 4:51 ` Masami Hiramatsu
2022-03-22 4:53 ` Alexei Starovoitov
2022-03-22 7:42 ` Peter Zijlstra
2022-03-22 4:38 ` Masami Hiramatsu
2022-03-21 15:28 ` Steven Rostedt
2022-03-21 16:04 ` Peter Zijlstra
2022-03-21 16:12 ` Steven Rostedt
2022-03-21 16:15 ` Steven Rostedt
2022-03-21 16:22 ` Steven Rostedt
2022-03-21 16:39 ` Steven Rostedt
2022-03-21 16:40 ` Peter Zijlstra
2022-03-21 16:45 ` Steven Rostedt
2022-03-21 16:50 ` Peter Zijlstra
2022-03-21 16:54 ` Steven Rostedt
2022-03-22 7:54 ` Peter Zijlstra
2022-03-22 13:12 ` Steven Rostedt
2022-03-22 14:35 ` Peter Zijlstra
2022-03-22 15:04 ` Steven Rostedt
2022-03-22 15:19 ` Peter Zijlstra
2022-03-22 15:48 ` Peter Zijlstra [this message]
2022-03-22 16:17 ` Steven Rostedt
2022-03-23 2:23 ` Masami Hiramatsu
2022-03-23 2:42 ` Steven Rostedt
2022-03-23 6:28 ` Masami Hiramatsu
2022-03-22 14:25 ` Masami Hiramatsu
2022-03-21 16:48 ` Peter Zijlstra
2022-03-22 5:31 ` Masami Hiramatsu
2022-03-22 8:08 ` Peter Zijlstra
2022-03-22 9:14 ` Masami Hiramatsu
2022-03-22 12:07 ` Peter Zijlstra
2022-03-22 12:17 ` Peter Zijlstra
2022-03-22 12:46 ` Masami Hiramatsu
2022-03-22 13:22 ` Steven Rostedt
2022-03-22 13:15 ` Mark Rutland
2022-03-22 13:51 ` Masami Hiramatsu
2022-03-22 10:46 ` Peter Zijlstra
2022-03-22 10:59 ` Peter Zijlstra
-- strict thread matches above, loose matches on Subject: below --
2024-02-02 3:59 Stephen Rothwell
2023-12-01 0:29 Stephen Rothwell
2023-12-01 12:09 ` Uros Bizjak
2023-12-04 4:08 ` Stephen Rothwell
2023-12-04 7:02 ` Uros Bizjak
2023-12-11 5:19 ` Stephen Rothwell
2023-12-11 7:06 ` Uros Bizjak
2023-06-02 3:12 Stephen Rothwell
2022-11-21 7:41 Stephen Rothwell
2022-05-20 7:49 Stephen Rothwell
2022-04-27 0:10 Stephen Rothwell
2022-04-27 11:31 ` Borislav Petkov
2022-04-27 13:43 ` Tom Lendacky
2022-03-22 3:51 Stephen Rothwell
2022-03-22 21:52 ` Peter Zijlstra
2022-03-22 23:11 ` Stephen Rothwell
2021-12-17 3:40 Stephen Rothwell
2022-01-21 23:58 ` Stephen Rothwell
2022-03-15 2:32 ` Stephen Rothwell
2022-04-04 3:26 ` Stephen Rothwell
2021-10-12 10:20 Stephen Rothwell
2021-10-12 13:58 ` André Almeida
2020-11-30 7:05 Stephen Rothwell
2020-11-30 10:17 ` Borislav Petkov
2020-11-30 21:56 ` Ernst, Justin
2020-11-30 22:18 ` Borislav Petkov
2020-11-23 7:19 Stephen Rothwell
2020-11-23 23:03 ` Jarkko Sakkinen
2017-11-02 2:53 Stephen Rothwell
2017-11-03 21:00 ` Masami Hiramatsu
2017-11-04 8:01 ` Ingo Molnar
2017-11-04 12:16 ` Masami Hiramatsu
2017-11-13 11:31 ` Stephen Rothwell
2017-06-23 4:19 Stephen Rothwell
2016-07-14 3:49 Stephen Rothwell
2016-07-14 3:37 Stephen Rothwell
2016-07-14 4:18 ` Stephen Rothwell
2012-10-12 5:11 Stephen Rothwell
2011-01-31 4:27 Stephen Rothwell
2011-01-31 5:08 ` Jaswinder Singh
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=Yjnvsrp8253bxWPA@hirez.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=Andrew.Cooper3@citrix.com \
--cc=ast@kernel.org \
--cc=hjl.tools@gmail.com \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-next@vger.kernel.org \
--cc=linux-toolchains@vger.kernel.org \
--cc=mhiramat@kernel.org \
--cc=mingo@redhat.com \
--cc=ndesaulniers@google.com \
--cc=rick.p.edgecombe@intel.com \
--cc=rostedt@goodmis.org \
--cc=rppt@kernel.org \
--cc=sfr@canb.auug.org.au \
--cc=tglx@linutronix.de \
/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 NNTP newsgroup(s).