public inbox for linux-arm-kernel@lists.infradead.org 
 help / color / mirror / Atom feed
From: will.deacon@arm•com (Will Deacon)
To: linux-arm-kernel@lists•infradead.org
Subject: [PATCH v5 3/8] arm: fixmap: implement __set_fixmap()
Date: Wed, 10 Sep 2014 18:51:26 +0100	[thread overview]
Message-ID: <20140910175126.GJ1710@arm.com> (raw)
In-Reply-To: <CAGXu5jJoW-B7WwuFUhbN4e1aL-HVxa9x4da9hB9DcCGEXVLgzw@mail.gmail.com>

Hi Kees,

On Tue, Sep 09, 2014 at 03:33:11PM +0100, Kees Cook wrote:
> On Tue, Sep 9, 2014 at 5:38 AM, Will Deacon <will.deacon@arm•com> wrote:
> > On Mon, Sep 08, 2014 at 11:40:43PM +0100, Kees Cook wrote:
> >> Ah, so it was, yes! Will, which version of this logic would you prefer?
> >
> > I still don't think we're solving the general problem here -- we're actually
> > just making the ftrace case work. It is perfectly possible for another CPU
> > to undergo a TLB miss and refill whilst the page table is being modified by
> > the CPU with preemption disabled. In this case, a local tlb flush won't
> > invalidate that entry on the other core, and we have no way of knowing when
> > the original permissions are actually observed across the system.
> 
> The fixmap is used by anything doing patching _except_ ftrace,
> actually. It's used by jump labels, kprobes, and kgdb. This code is
> the general case. Access to set_fixmap is done via the kernel patching
> interface: patch_text().
> 
> Right now, the patch_text interface checks cache_ops_need_broadcast(),
> and conditionally runs under stop_machine(). We could make this
> unconditional, and we'll avoid any problem with TLB misses on another
> CPU.

Yes, it we always use stop_machine, that solves the TLB broadcast problem
and we could do that if CONFIG_ARM_ERRATA_798181 is set.

> > So I think we need to figure out a way to invalidate the TLB properly. What
> > do architectures that use IPIs for TLB broadcasting do (x86, some powerpc,
> > mips, ...)? They must have exactly the same problem.
> 
> I don't think this should be done at the set_fixmap level, as it is
> more a primitive. I think making sure patch_text() always works would
> be best. What do you think of using an unconditional stop_machine()
> instead?

Why not move the TLB invalidation into patch_text, then we can do
stop_machine if IS_ENABLED(CONFIG_ARM_ERRATA_798181) ||
tlb_ops_need_broadcast()?

Then that just leaves ftrace.

Will

  reply	other threads:[~2014-09-10 17:51 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-09-03 21:57 [PATCH v5 0/8] arm: support CONFIG_RODATA Kees Cook
2014-09-03 21:57 ` [PATCH v5 1/8] arm: use generic fixmap.h Kees Cook
2014-09-03 21:57 ` [PATCH v5 2/8] ARM: expand fixmap region to 3MB Kees Cook
2014-09-03 21:57 ` [PATCH v5 3/8] arm: fixmap: implement __set_fixmap() Kees Cook
2014-09-04 17:03   ` Will Deacon
2014-09-04 17:23     ` Kees Cook
2014-09-04 17:27       ` Will Deacon
2014-09-05 19:41         ` Kees Cook
2014-09-08 10:39           ` Will Deacon
2014-09-08 18:38             ` Kees Cook
2014-09-08 19:16         ` Kees Cook
2014-09-08 21:55           ` Rabin Vincent
2014-09-08 22:40             ` Kees Cook
2014-09-09 12:38               ` Will Deacon
2014-09-09 14:33                 ` Kees Cook
2014-09-10 17:51                   ` Will Deacon [this message]
2014-09-11 15:27                     ` Kees Cook
2014-09-11 16:05                       ` Kees Cook
2014-09-11 16:16                         ` Will Deacon
2014-09-11 16:27                           ` Kees Cook
2014-09-03 21:57 ` [PATCH v5 4/8] arm: use fixmap for text patching when text is RO Kees Cook
2014-09-03 21:57 ` [PATCH v5 5/8] ARM: kexec: Make .text R/W in machine_kexec Kees Cook
2014-09-03 21:57 ` [PATCH v5 6/8] arm: kgdb: Handle read-only text / modules Kees Cook
2014-09-03 21:57 ` [PATCH v5 7/8] ARM: mm: allow non-text sections to be non-executable Kees Cook
2014-09-03 21:57 ` [PATCH v5 8/8] ARM: mm: allow text and rodata sections to be read-only Kees Cook

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=20140910175126.GJ1710@arm.com \
    --to=will.deacon@arm$(echo .)com \
    --cc=linux-arm-kernel@lists$(echo .)infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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