From: "H. Peter Anvin" <hpa@zytor.com>
To: Will Deacon <will.deacon@arm.com>, Peter Zijlstra <peterz@infradead.org>
Cc: dvlasenk@redhat.com, jpoimboe@redhat.com,
Julien Thierry <julien.thierry@arm.com>,
catalin.marinas@arm.com, valentin.schneider@arm.com,
linux-kernel@vger.kernel.org,
Andy Lutomirski <luto@amacapital.net>,
mingo@redhat.com, james.morse@arm.com, luto@kernel.org,
brgerst@gmail.com, bp@alien8.de, tglx@linutronix.de,
torvalds@linux-foundation.org, Ingo Molnar <mingo@kernel.org>,
linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH] sched/x86: Save [ER]FLAGS on context switch
Date: Wed, 20 Feb 2019 14:55:59 -0800 [thread overview]
Message-ID: <f8e99706-4e3b-68ec-7a3e-6e6dd876ad2f@zytor.com> (raw)
In-Reply-To: <20190219124808.GG8501@fuggles.cambridge.arm.com>
On 2/19/19 4:48 AM, Will Deacon wrote:
>
> I think you'll still hate this, but could we not disable preemption during
> the uaccess-enabled region, re-enabling it on the fault path after we've
> toggled uaccess off and disable it again when we return back to the
> uaccess-enabled region? Doesn't help with tracing, but it should at least
> handle the common case.
>
There is a worse problem with this, I still realize: this would mean blocking
preemption across what could possibly be a *very* large copy_from_user(), for
example.
Exceptions *have* to handle this; there is no way around it. Perhaps the
scheduler isn't the right place to put these kinds of asserts, either.
Now, __fentry__ is kind of a special beast; in some ways it is an "exception
implemented as a function call"; on x86 one could even consider using an INT
instruction in order to reduce the NOP footprint in the unarmed case. Nor is
__fentry__ a C function; it has far more of an exception-like ABI.
*Regardless* of what else we do, I believe __fentry__ ought to
save/disable/restore AC, just like an exception does.
The idea of using s/a gcc plugin/objtool/ for this isn't really a bad idea.
Obviously the general problem is undecidable :) but the enforcement of some
simple, fairly draconian rules ("as tight as possible, but no tighter")
shouldn't be a huge problem.
An actual gcc plugin -- which would probably be quite complex -- could make
gcc itself aware of user space accesses and be able to rearrange them to
minimize STAC/CLAC and avoid kernel-space accesses inside those brackets.
Finally, of course, there is the option of simply outlawing this practice as a
matter of policy and require that all structures be accessed through a limited
set of APIs. As I recall, the number of places where there were
performance-critical regions which could not use the normal accessors are
fairly small (signal frames being the main one.) Doing bulk copy to/from
kernel memory and then accessing them from there would have some performance
cost, but would eliminate the need for this complexity entirely.
-hpa
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
next prev parent reply other threads:[~2019-02-20 22:56 UTC|newest]
Thread overview: 88+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-15 13:58 [PATCH v3 0/4] uaccess: Add unsafe accessors for arm64 Julien Thierry
2019-01-15 13:58 ` [PATCH v3 1/4] arm64: uaccess: Cleanup get/put_user() Julien Thierry
2019-01-15 13:58 ` [PATCH v3 2/4] arm64: uaccess: Implement unsafe accessors Julien Thierry
2019-01-15 13:58 ` [PATCH v3 3/4] uaccess: Check no rescheduling function is called in unsafe region Julien Thierry
2019-01-30 16:58 ` Valentin Schneider
2019-02-04 13:27 ` Julien Thierry
2019-02-11 13:45 ` Ingo Molnar
2019-02-11 13:51 ` Peter Zijlstra
2019-02-12 9:15 ` Julien Thierry
2019-02-13 8:21 ` Ingo Molnar
2019-02-13 10:35 ` Peter Zijlstra
2019-02-13 10:50 ` Julien Thierry
2019-02-13 13:17 ` Peter Zijlstra
2019-02-13 13:20 ` Peter Zijlstra
2019-02-13 14:00 ` Will Deacon
2019-02-13 14:07 ` Julien Thierry
2019-02-13 14:17 ` Peter Zijlstra
2019-02-13 14:24 ` Julien Thierry
2019-02-13 14:40 ` Peter Zijlstra
2019-02-13 15:08 ` Peter Zijlstra
2019-02-13 14:25 ` Peter Zijlstra
2019-02-13 14:39 ` Julien Thierry
2019-02-13 14:41 ` Peter Zijlstra
2019-02-13 15:45 ` Peter Zijlstra
2019-02-13 18:54 ` Peter Zijlstra
[not found] ` <D61C430D-4321-4114-AB85-671A3C7B8EAE@amacapital.net>
2019-02-13 22:21 ` Peter Zijlstra
2019-02-13 22:49 ` Andy Lutomirski
2019-02-14 10:14 ` [PATCH] sched/x86: Save [ER]FLAGS on context switch Peter Zijlstra
2019-02-14 16:18 ` Brian Gerst
2019-02-14 19:34 ` Peter Zijlstra
2019-02-15 14:34 ` Brian Gerst
2019-02-15 17:18 ` Linus Torvalds
2019-02-15 17:40 ` Peter Zijlstra
2019-02-15 18:28 ` Andy Lutomirski
2019-02-15 23:34 ` Peter Zijlstra
2019-02-16 0:21 ` Linus Torvalds
2019-02-16 10:32 ` Peter Zijlstra
2019-02-16 4:06 ` hpa
2019-02-16 10:30 ` Peter Zijlstra
2019-02-18 22:30 ` H. Peter Anvin
2019-02-19 0:24 ` Linus Torvalds
2019-02-19 2:20 ` Andy Lutomirski
2019-02-19 2:46 ` H. Peter Anvin
2019-02-19 9:07 ` Julien Thierry
2019-02-19 8:53 ` Julien Thierry
2019-02-19 9:15 ` Peter Zijlstra
2019-02-19 9:19 ` Peter Zijlstra
2019-02-19 9:04 ` Peter Zijlstra
2019-02-19 9:21 ` hpa
2019-02-19 9:44 ` Peter Zijlstra
2019-02-19 11:38 ` Thomas Gleixner
2019-02-19 11:58 ` Peter Zijlstra
2019-02-19 12:48 ` Will Deacon
2019-02-20 22:55 ` H. Peter Anvin [this message]
2019-02-21 12:06 ` Julien Thierry
2019-02-21 21:35 ` Thomas Gleixner
2019-02-21 22:08 ` Linus Torvalds
2019-02-22 12:58 ` Peter Zijlstra
2019-02-22 18:10 ` Thomas Gleixner
2019-02-22 22:26 ` [RFC][PATCH] objtool: STAC/CLAC validation Peter Zijlstra
2019-02-22 23:34 ` Linus Torvalds
2019-02-23 8:43 ` Peter Zijlstra
2019-02-22 23:39 ` hpa
2019-02-23 8:39 ` Peter Zijlstra
2019-02-25 8:47 ` hpa
2019-02-25 13:21 ` Peter Zijlstra
2019-03-01 15:07 ` Peter Zijlstra
2019-02-25 8:49 ` hpa
2019-02-22 23:55 ` Andy Lutomirski
2019-02-23 8:37 ` Peter Zijlstra
2019-02-23 10:52 ` Peter Zijlstra
2019-02-25 10:51 ` Peter Zijlstra
2019-02-25 11:53 ` Peter Zijlstra
2019-02-25 15:36 ` Andy Lutomirski
2019-02-23 0:34 ` Andy Lutomirski
2019-02-23 1:12 ` Linus Torvalds
2019-02-23 1:16 ` Andy Lutomirski
2019-02-23 1:33 ` Linus Torvalds
2019-02-23 1:40 ` Linus Torvalds
2019-02-25 8:33 ` Julien Thierry
2019-02-25 11:55 ` Peter Zijlstra
2019-02-21 12:46 ` [PATCH] sched/x86: Save [ER]FLAGS on context switch Will Deacon
2019-02-21 22:06 ` Andy Lutomirski
2019-02-18 9:03 ` [PATCH v2] " Peter Zijlstra
2019-02-13 23:19 ` [PATCH v3 3/4] uaccess: Check no rescheduling function is called in unsafe region Linus Torvalds
2019-01-15 13:58 ` [PATCH v3 4/4] arm64: uaccess: Implement user_access_region_active Julien Thierry
2019-01-25 14:27 ` [PATCH v3 0/4] uaccess: Add unsafe accessors for arm64 Catalin Marinas
2019-01-30 16:17 ` Julien Thierry
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=f8e99706-4e3b-68ec-7a3e-6e6dd876ad2f@zytor.com \
--to=hpa@zytor.com \
--cc=bp@alien8.de \
--cc=brgerst@gmail.com \
--cc=catalin.marinas@arm.com \
--cc=dvlasenk@redhat.com \
--cc=james.morse@arm.com \
--cc=jpoimboe@redhat.com \
--cc=julien.thierry@arm.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=luto@amacapital.net \
--cc=luto@kernel.org \
--cc=mingo@kernel.org \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
--cc=valentin.schneider@arm.com \
--cc=will.deacon@arm.com \
/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).