linux-kernel - RE: [PATCH bpf-next 08/13] uprobes/x86: Add support to optimize uprobes

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <0916e24539ba4bae9fb729198b033bd7@AcuMS.aculab.com>
Date: Mon, 16 Dec 2024 11:10:23 +0000
From: David Laight <David.Laight@...LAB.COM>
To: 'Oleg Nesterov' <oleg@...hat.com>, "linux-mm@...ck.org"
	<linux-mm@...ck.org>
CC: 'Jiri Olsa' <olsajiri@...il.com>, Peter Zijlstra <peterz@...radead.org>,
	Andrii Nakryiko <andrii@...nel.org>, "bpf@...r.kernel.org"
	<bpf@...r.kernel.org>, Song Liu <songliubraving@...com>, Yonghong Song
	<yhs@...com>, John Fastabend <john.fastabend@...il.com>, Hao Luo
	<haoluo@...gle.com>, Steven Rostedt <rostedt@...dmis.org>, Masami Hiramatsu
	<mhiramat@...nel.org>, Alan Maguire <alan.maguire@...cle.com>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	"linux-trace-kernel@...r.kernel.org" <linux-trace-kernel@...r.kernel.org>
Subject: RE: [PATCH bpf-next 08/13] uprobes/x86: Add support to optimize
 uprobes

From: Oleg Nesterov
> Sent: 16 December 2024 10:13
> 
> David,
> 
> let me say first that my understanding of this magic is very limited,
> please correct me.

I only (half) understand what the 'magic' has to accomplish and
some of the pitfalls.

I've copied linux-mm - someone there might know more.

> On 12/16, David Laight wrote:
> >
> > It all depends on how hard __replace_page() tries to be atomic.
> > The page has to change from one backed by the executable to a private
> > one backed by swap - otherwise you can't write to it.
> 
> This is what uprobe_write_opcode() does,

And will be enough for single byte changes - they'll be picked up
at some point after the change.

> > But the problems arise when the instruction prefetch unit has read
> > part of the 5-byte instruction (it might even only read half a cache
> > line at a time).
> > I'm not sure how long the pipeline can sit in that state - but I
> > can do a memory read of a PCIe address that takes ~3000 clocks.
> > (And a misaligned AVX-512 read is probably eight 8-byte transfers.)
> >
> > So I think you need to force an interrupt while the PTE is invalid.
> > And that need to be simultaneous on all cpu running that process.
> 
> __replace_page() does ptep_get_and_clear(old_pte) + flush_tlb_page().
> 
> That's not enough?

I doubt it. As I understand it.
The hardware page tables will be shared by all the threads of a process.
So unless you hard synchronise all the cpu (and flush the TLB) while the
PTE is being changed there is always the possibility of a cpu picking up
the new PTE before the IPI that (I presume) flush_tlb_page() generates
is processed.
If that happens when the instruction you are patching is part-read into
the instruction decode buffer then you'll execute a mismatch of the two
instructions.

I can't remember the outcome of discussions about live-patching kernel
code - and I'm sure that was aligned 32bit writes.

> 
> > Stopping the process using ptrace would do it.
> 
> Not an option :/

Thought you'd say that.

	David

> 
> Oleg.

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)