Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CACy=wMngd4vUwZCLQZPrN7fBFrJDZuNa_n=Qkc9Y3-_vqX_jqQ@mail.gmail.com>
Date: Thu, 5 Oct 2017 10:10:43 -0700
From: Dave Weinstein <olorin@...gle.com>
To: Linus Torvalds <torvalds@...ux-foundation.org>
Cc: "Roberts, William C" <william.c.roberts@...el.com>, "Tobin C. Harding" <me@...in.cc>, Tejun Heo <tj@...nel.org>, 
	Jordan Glover <Golden_Miller83@...tonmail.ch>, Greg KH <gregkh@...uxfoundation.org>, 
	Petr Mladek <pmladek@...e.com>, Joe Perches <joe@...ches.com>, Ian Campbell <ijc@...lion.org.uk>, 
	Sergey Senozhatsky <sergey.senozhatsky@...il.com>, 
	"kernel-hardening@...ts.openwall.com" <kernel-hardening@...ts.openwall.com>, 
	Catalin Marinas <catalin.marinas@....com>, Will Deacon <will.deacon@....com>, 
	Steven Rostedt <rostedt@...dmis.org>, Chris Fries <cfries@...gle.com>
Subject: Re: [RFC V2 0/6] add more kernel pointer filter options

> Does android set it to 2 right now?

Yes. And as of Android 8.0, it will attempt to set it to 4, and step back
to 2 if the higher restriction values are not supported (
https://android.googlesource.com/platform/system/core/+/44f7e4f42190fdb5309b818d5acc0ff6b0f87249%5E%21/#F0
)

>     We already don't get great bug reports. If we now start getting

>bug reports and tell users "recreate this with 'kptr_enabled' on the

>kernel command line" because some debug message didn't give proper

>data, that will just result in us getting even less bug reports.

This is why the original Android patch set whitelists the stack traces --
it was a deliberate choice to make sure that kernel bug reports were
generally meaningful by default.

>Honestly, what's the difference between setting kptr_restrict to 4 and

>just using a sed-script (or maybe some coccinelle) to remove all

>existing plain %p users?

The primary benefit from my perspective is that the userland component can
establish a general expectation of the level of information disclosure that
depends on a single feature being implemented, rather than an arbitrarily
large number of point fixes across multiple kernel versions.

--Dave

-- 
Dave Weinstein
Android SDL

Content of type "text/html" skipped

Powered by blists - more mailing lists

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.