From: Julien Thierry <jthierry@redhat.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Alexandre Chartre <alexandre.chartre@oracle.com>,
x86@kernel.org, linux-kernel@vger.kernel.org,
jpoimboe@redhat.com, tglx@linutronix.de
Subject: Re: [PATCH 4/7] objtool: Add support for return trampoline call
Date: Fri, 3 Apr 2020 09:11:55 +0100 [thread overview]
Message-ID: <bc3a31dc-9d59-5756-aad3-187533f05654@redhat.com> (raw)
In-Reply-To: <20200402154022.GG20730@hirez.programming.kicks-ass.net>
On 4/2/20 4:40 PM, Peter Zijlstra wrote:
> On Thu, Apr 02, 2020 at 04:31:05PM +0100, Julien Thierry wrote:
>> My understanding is that once you find an intra-function call, either you
>> hit a return, ending the branch, so the return should undo the modification
>> the intra-function call did (whether is it a retpoline return or not).
>> Otherwise, the intra-function call branch will need to reach an end in some
>> way (e.g. hitting a CONTEXT_SWITCH instruction, calling a
>> dead_end_function).
>>
>> Am I missing something?
>
> The thing is basically doing:
>
> mov $n, cx
> 1: call 2f
> 2: dec cx
> jnz 1b
> add 8*n, sp
>
> So it does N calls to self, then subtracts N words from the stack.
>
> The reason being that the CPU has a return-stack-buffer for predicting
> returns, and call/ret being naturally paired, that works. The above
> is a software flush of the RSB.
>
Ah, lovely... Maybe that's where SAVE/RESTORE unwind hints could be nice
;) .
Otherwise, I don't really have a good suggestion for this...
--
Julien Thierry
next prev parent reply other threads:[~2020-04-03 8:12 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-04-02 8:22 [PATCH 0/7] objtool changes to remove most ANNOTATE_NOSPEC_ALTERNATIVE Alexandre Chartre
2020-04-02 8:22 ` [PATCH 1/7] objtool: is_fentry_call() crashes if call has no destination Alexandre Chartre
2020-04-02 8:22 ` [PATCH 2/7] objtool: Allow branches within the same alternative Alexandre Chartre
2020-04-02 12:03 ` Julien Thierry
2020-04-02 12:38 ` Alexandre Chartre
2020-04-02 8:22 ` [PATCH 3/7] objtool: Add support for intra-function calls Alexandre Chartre
2020-04-02 12:53 ` Julien Thierry
2020-04-02 13:24 ` Alexandre Chartre
2020-04-02 13:38 ` Julien Thierry
2020-04-02 14:56 ` Alexandre Chartre
2020-04-02 15:04 ` Peter Zijlstra
2020-04-02 15:54 ` Josh Poimboeuf
2020-04-03 7:06 ` Alexandre Chartre
2020-04-02 15:49 ` Josh Poimboeuf
2020-04-02 17:27 ` Josh Poimboeuf
2020-04-03 8:01 ` Julien Thierry
2020-04-03 12:41 ` Peter Zijlstra
2020-04-03 12:49 ` Julien Thierry
2020-04-03 14:37 ` Peter Zijlstra
2020-04-03 14:44 ` Josh Poimboeuf
2020-04-02 8:22 ` [PATCH 4/7] objtool: Add support for return trampoline call Alexandre Chartre
2020-04-02 13:26 ` Julien Thierry
2020-04-02 14:46 ` Alexandre Chartre
2020-04-02 15:31 ` Julien Thierry
2020-04-02 15:40 ` Peter Zijlstra
2020-04-03 8:11 ` Julien Thierry [this message]
2020-04-03 15:17 ` Josh Poimboeuf
2020-04-03 15:22 ` Josh Poimboeuf
2020-04-03 15:32 ` Josh Poimboeuf
2020-04-03 15:46 ` Peter Zijlstra
2020-04-03 15:55 ` Josh Poimboeuf
2020-04-04 13:32 ` Peter Zijlstra
2020-04-04 14:22 ` Josh Poimboeuf
2020-04-04 15:51 ` Peter Zijlstra
2020-04-06 8:19 ` Alexandre Chartre
2020-04-06 9:31 ` Peter Zijlstra
2020-04-06 11:03 ` Alexandre Chartre
2020-04-06 14:16 ` Josh Poimboeuf
2020-04-02 15:27 ` Peter Zijlstra
2020-04-03 7:19 ` Alexandre Chartre
2020-04-06 14:34 ` Alexandre Chartre
2020-04-06 14:55 ` Alexandre Chartre
2020-04-02 8:22 ` [PATCH 5/7] x86/speculation: Annotate intra-function calls Alexandre Chartre
2020-04-03 16:05 ` Josh Poimboeuf
2020-04-03 16:16 ` Josh Poimboeuf
2020-04-03 17:14 ` Alexandre Chartre
2020-04-03 17:18 ` Peter Zijlstra
2020-04-03 17:24 ` Josh Poimboeuf
2020-04-03 18:20 ` Peter Zijlstra
2020-04-02 8:22 ` [PATCH 6/7] x86/speculation: Annotate retpoline return instructions Alexandre Chartre
2020-04-02 8:22 ` [PATCH 7/7] x86/speculation: Remove most ANNOTATE_NOSPEC_ALTERNATIVE Alexandre Chartre
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=bc3a31dc-9d59-5756-aad3-187533f05654@redhat.com \
--to=jthierry@redhat.com \
--cc=alexandre.chartre@oracle.com \
--cc=jpoimboe@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
--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;
as well as URLs for NNTP newsgroup(s).