linux-kernel - Re: [PATCH 2/3] objtool: Ignore retpoline alternatives

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] [day] [month] [year] [list]

Message-ID: <20180112203621.yjoymm3qfmsvpt76@gmail.com>
Date:   Fri, 12 Jan 2018 21:36:21 +0100
From:   Ingo Molnar <mingo@...nel.org>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     Linus Torvalds <torvalds@...ux-foundation.org>,
        David Woodhouse <dwmw2@...radead.org>,
        Josh Poimboeuf <jpoimboe@...hat.com>,
        the arch/x86 maintainers <x86@...nel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Andi Kleen <ak@...ux.intel.com>, Paul Turner <pjt@...gle.com>,
        Greg Kroah-Hartman <gregkh@...ux-foundation.org>,
        Tim Chen <tim.c.chen@...ux.intel.com>,
        Dave Hansen <dave.hansen@...el.com>,
        Kees Cook <keescook@...gle.com>,
        Andy Lutomirski <luto@...capital.net>,
        Jiri Kosina <jikos@...nel.org>,
        One Thousand Gnomes <gnomes@...rguk.ukuu.org.uk>
Subject: Re: [PATCH 2/3] objtool: Ignore retpoline alternatives


* Peter Zijlstra <peterz@...radead.org> wrote:

> > But if you're an actual human, the "small numbers as labels" is fine.
> 
> I find descriptive labels much nicer than random numbers, I'll take some
> crazy characters if so required.
> 
> 
> Consider the retpoline thing:
> 
>  call .Lset_up_target
> .Lcapture_spec:
>   pause
>   jmp .Lcapture_spec
> .Lset_up_target:
>   mov %r11, (%rsp);
>   ret;
> 
> 
> over:
> 
>  call 2f
> 1:
>  pause
>  jmp 1b
> 2:
>  mov %r11, (%rsp)
>  ret
> 
> 
> give me the first any day of the week.

Absolutely, in fact I detest what we have:

+#define __FILL_RETURN_BUFFER(reg, nr, sp)      \
+       mov     $(nr/2), reg;                   \
+771:                                           \
+       call    772f;                           \
+773:   /* speculation trap */                  \
+       pause;                                  \
+       jmp     773b;                           \
+772:                                           \
+       call    774f;                           \
+775:   /* speculation trap */                  \
+       pause;                                  \
+       jmp     775b;                           \
+774:                                           \
+       dec     reg;                            \
+       jnz     771b;                           \
+       add     $(BITS_PER_LONG/8) * nr, sp;

I mean, WTF??

Also, note that this:

>  call .Lset_up_target
> .Lcapture_spec:
>   pause
>   jmp .Lcapture_spec
> .Lset_up_target:
>   mov %r11, (%rsp);
>   ret;
> 

becomes even more readable with a bit more human-readability improvements:

	call .L_set_up_target

 .L_capture_spec:
	pause
	jmp .L_capture_spec

 .L_set_up_target:
	mov %r11, (%rsp)
	ret 

I.e. the underscores in the label names, the tabs and the newlines create better 
vertical and horizontal separation between the syntactical elements and provide 
grouping of larger constructs.

Thanks,

	Ingo