linux-kernel - Re: [RFC][PATCH 1/2] x86: Allow breakpoints to emulate call functions

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: <20190507172159.5t3bm3mjkwagvite@treble>
Date:   Tue, 7 May 2019 12:21:59 -0500
From:   Josh Poimboeuf <jpoimboe@...hat.com>
To:     Linus Torvalds <torvalds@...ux-foundation.org>
Cc:     Peter Zijlstra <peterz@...radead.org>,
        Steven Rostedt <rostedt@...dmis.org>,
        Andy Lutomirski <luto@...capital.net>,
        Linux List Kernel Mailing <linux-kernel@...r.kernel.org>,
        Ingo Molnar <mingo@...nel.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Andy Lutomirski <luto@...nel.org>,
        Nicolai Stange <nstange@...e.de>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
        "H. Peter Anvin" <hpa@...or.com>,
        the arch/x86 maintainers <x86@...nel.org>,
        Jiri Kosina <jikos@...nel.org>,
        Miroslav Benes <mbenes@...e.cz>,
        Petr Mladek <pmladek@...e.com>,
        Joe Lawrence <joe.lawrence@...hat.com>,
        Shuah Khan <shuah@...nel.org>,
        Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>,
        Tim Chen <tim.c.chen@...ux.intel.com>,
        Sebastian Andrzej Siewior <bigeasy@...utronix.de>,
        Mimi Zohar <zohar@...ux.ibm.com>,
        Juergen Gross <jgross@...e.com>,
        Nick Desaulniers <ndesaulniers@...gle.com>,
        Nayna Jain <nayna@...ux.ibm.com>,
        Masahiro Yamada <yamada.masahiro@...ionext.com>,
        Joerg Roedel <jroedel@...e.de>,
        "open list:KERNEL SELFTEST FRAMEWORK" 
        <linux-kselftest@...r.kernel.org>, stable <stable@...r.kernel.org>,
        Masami Hiramatsu <mhiramat@...nel.org>
Subject: Re: [RFC][PATCH 1/2] x86: Allow breakpoints to emulate call functions

On Tue, May 07, 2019 at 10:08:50AM -0700, Linus Torvalds wrote:
> On Tue, May 7, 2019 at 9:34 AM Peter Zijlstra <peterz@...radead.org> wrote:
> >
> > Would you consider my approach later on, under the guise of unification?
> 
> WHY?
> 
> The *only* advantage of your patch is that trivial "look up kernel stack" macro.
> 
> Seriously. There's absolutely nothing else.
> 
> Is that macro ugly? Yes. But it's directly explainable by just
> pointing to the architecture documentation.
> 
> It's a one-liner hack.
> 
> And for that, you want to complicate the x86-32 entry and exit code?
> 
> Do we have different emulation for "push" on 32-bit and 64-bit? Yes.
> But again, that's just how the hardware works. This is not some
> "generic hw-independent code". This is literally emulating
> instructions that care about instruction encoding and bit size
> details, where there are certainly _similarities_ (and in the case of
> 'call', they look bit-identical), but it's also not like "same code"
> is a big argument. That's why we have a helper function, to hide the
> details.
> 
> I point to my diffstat once again. It's smaller, and I argue that it
> is actually conceptually *simpler* to simply say "this is how the
> architecture works".
> 
> And yes, I realize that I may be biased by the fact that I simply know
> i386 so well, so to me it simply makes more sense to just work with
> what the hardware gives us. The i386 exception model with the kernel
> stack nesting is a *hell* of a lot simpler than the x86-64 one. The
> fact is, x86-64 messed things up, and swapgs and friends are an
> abomination against God.
> 
> So the whole "let's clean up x86-32 to look like x86-64, which got
> things right" is to me a completely bogus argument. x86-64 got the
> "yes, push ss/sp unconditionally" part right, but got a lot of other
> things horribly wrong. So this is all just one small detail that
> differs, across two architectures that are similar but have very
> different warts.
> 
> But that diffstat is still hard, cold, unbiased data.

regs->sp is *undefined* on x86-32.  We're damning our future selves to
have to always remember to use that darn kernel_stack_pointer() helper
for eternity just because of x86-32.

There have already been several bugs related to that.  Because regs->sp
is there, so why wouldn't you use it?

If we truly want the code to reflect the HW, then we should have a
pt_regs_kernel and a pt_regs_user on 32-bit.  I'm pretty sure we don't
want to go there...

IMO, we either need to make the pt_regs struct(s) match the HW behavior,
or make entry code match pt_regs.  But this in-between thing just
creates a bunch of headaches.

-- 
Josh