From: Alan Cox <gnomes@lxorguk.ukuu.org.uk>
To: Brian Gerst <brgerst@gmail.com>
Cc: Andi Kleen <andi@firstfloor.org>,
Thomas Gleixner <tglx@linutronix.de>,
Linus Torvalds <torvalds@linux-foundation.org>,
gregkh@linux-foundation.org,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Tim Chen <tim.c.chen@linux.intel.com>,
Dave Hansen <dave.hansen@linux.intel.com>,
David Woodhouse <dwmw@amazon.co.uk>,
Andi Kleen <ak@linux.intel.com>
Subject: Re: [PATCH v2 01/12] x86/retpoline: Define retpoline indirect thunk and macros
Date: Thu, 4 Jan 2018 02:32:32 +0000 [thread overview]
Message-ID: <20180104023232.0460184b@alans-desktop> (raw)
In-Reply-To: <CAMzpN2gk1_mj7xZsj39yycctJqEdpHYOS0N6XoGjTtZHp=tmsg@mail.gmail.com>
> > +ENTRY(__x86.indirect_thunk)
> > + CFI_STARTPROC
> > + call retpoline_call_target
> > +2:
> > + lfence /* stop speculation */
> > + jmp 2b
> > +retpoline_call_target:
> > +#ifdef CONFIG_64BIT
> > + lea 8(%rsp), %rsp
> > +#else
> > + lea 4(%esp), %esp
> > +#endif
> > + ret
> > + CFI_ENDPROC
> > +ENDPROC(__x86.indirect_thunk)
> > +
> > + EXPORT_SYMBOL(__x86.indirect_thunk)
> > --
> > 2.14.3
> >
>
> Can someone actually explain WTF this mess is trying to accomplish?
Think of it as an 'indirect call that doesn't speculate' instruction.
There isn't one in the processor but this specific sequence happens to
make the micro-architecture do just that as efficiently as possible.
What it's actually doing on the non-speculated path (ie the reachable
code) is to call, put the address we want to hit over the existing return
address and then return, to the address we want to indirectly go to.
It's faster than doing a far branch or flushing branch predictors and the
like.
Alan
next prev parent reply other threads:[~2018-01-04 2:32 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-01-04 2:00 Avoid speculative indirect calls in kernel Andi Kleen
2018-01-04 2:00 ` [PATCH v2 01/12] x86/retpoline: Define retpoline indirect thunk and macros Andi Kleen
2018-01-04 2:15 ` Brian Gerst
2018-01-04 2:32 ` Alan Cox [this message]
2018-01-04 2:00 ` [PATCH v2 02/12] x86/retpoline/crypto: Convert crypto assembler indirect jumps Andi Kleen
2018-01-04 2:00 ` [PATCH v2 03/12] x86/retpoline/entry: Convert entry " Andi Kleen
2018-01-04 8:42 ` David Woodhouse
2018-01-04 2:00 ` [PATCH v2 04/12] x86/retpoline/ftrace: Convert ftrace " Andi Kleen
2018-01-04 2:00 ` [PATCH v2 05/12] x86/retpoline/hyperv: Convert " Andi Kleen
2018-01-04 2:00 ` [PATCH v2 06/12] x86/retpoline/crypto: Convert xen " Andi Kleen
2018-01-04 6:48 ` Juergen Gross
2018-01-04 6:50 ` Andi Kleen
2018-01-04 2:00 ` [PATCH v2 07/12] x86/retpoline/checksum32: Convert " Andi Kleen
2018-01-04 2:00 ` [PATCH v2 08/12] x86/retpoline/irq32: " Andi Kleen
2018-01-04 2:00 ` [PATCH v2 09/12] x86/retpoline: Finally enable retpoline for C code Andi Kleen
2018-01-04 2:00 ` [PATCH v2 10/12] retpoline/taint: Taint kernel for missing retpoline in compiler Andi Kleen
2018-01-04 2:00 ` [PATCH v2 11/12] retpoline/objtool: Disable some objtool warnings Andi Kleen
2018-01-04 14:38 ` Josh Poimboeuf
2018-01-04 14:46 ` David Woodhouse
2018-01-04 15:59 ` Andi Kleen
2018-01-04 16:06 ` Josh Poimboeuf
2018-01-04 16:13 ` Andi Kleen
2018-01-04 16:32 ` Josh Poimboeuf
2018-01-04 17:35 ` Josh Poimboeuf
2018-01-04 2:00 ` [PATCH v2 12/12] retpoline: Attempt to quiten objtool warning for unreachable code Andi Kleen
2018-01-04 11:49 ` Avoid speculative indirect calls in kernel Pavel Machek
2018-01-04 12:09 ` Alan Cox
2018-01-04 13:32 ` Pavel Machek
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=20180104023232.0460184b@alans-desktop \
--to=gnomes@lxorguk.ukuu.org.uk \
--cc=ak@linux.intel.com \
--cc=andi@firstfloor.org \
--cc=brgerst@gmail.com \
--cc=dave.hansen@linux.intel.com \
--cc=dwmw@amazon.co.uk \
--cc=gregkh@linux-foundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=tglx@linutronix.de \
--cc=tim.c.chen@linux.intel.com \
--cc=torvalds@linux-foundation.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