All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andy Lutomirski <luto@amacapital.net>
To: Ingo Molnar <mingo@kernel.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>,
	Denys Vlasenko <dvlasenk@redhat.com>,
	Kees Cook <keescook@chromium.org>, Borislav Petkov <bp@alien8.de>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Alexei Starovoitov <ast@plumgrid.com>,
	Oleg Nesterov <oleg@redhat.com>, Will Drewry <wad@chromium.org>,
	Steven Rostedt <rostedt@goodmis.org>, X86 ML <x86@kernel.org>,
	"H. Peter Anvin" <hpa@zytor.com>,
	Linus Torvalds <torvalds@linux-foundation.org>
Subject: Re: [PATCH 3/5] x86/asm/entry/32: Shorten __audit_syscall_entry args preparation
Date: Fri, 12 Jun 2015 16:28:06 -0700	[thread overview]
Message-ID: <CALCETrWp79ASjZ1f28_u6bVO4b0Mh4=n6H8VhbA=Y6qcQLG45w@mail.gmail.com> (raw)
In-Reply-To: <20150610062110.GA1165@gmail.com>

On Jun 9, 2015 11:21 PM, "Ingo Molnar" <mingo@kernel.org> wrote:
>
>
> * Denys Vlasenko <dvlasenk@redhat.com> wrote:
>
> > We use three MOVs to swap edx and ecx. We can use one XCHG instead.
> >
> > Expand the comments. It's difficult to keep track which arg# every register
> > corresponds to, so spell it out.
>
> > +     /*
> > +      * At this point, registers hold syscall args in 32-bit ABI:
> > +      * eax is syscall#, args are in ebx,ecx,edx,esi,edi,ebp.
> > +      * Shuffle them to match what __audit_syscall_entry() wants.
> > +      */
> > +     movl    %esi, %r8d              /* arg5 (r8): 4th syscall arg */
> > +     xchg    %ecx, %edx              /* arg4 (rcx): 3rd syscall arg (edx) */
> > +                                     /* arg3 (rdx): 2nd syscall arg (ecx) */
> > +     movl    %ebx, %esi              /* arg2 (rsi): 1st syscall arg */
> > +     movl    %eax, %edi              /* arg1 (rdi): syscall number */
> >       call    __audit_syscall_entry
>
> So while we are at it I improved this a bit more, to:
>
>         /*
>          * At this point, registers hold syscall args in 32-bit syscall ABI:
>          *   eax is syscall#, args are in ebx,ecx,edx,esi,edi,ebp.
>          *
>          * We want to pass them to __audit_syscall_entry(), which is a 64-bit
>          * C function with 5 parameters, so shuffle them to match what
>          * __audit_syscall_entry() expects: rdi,rsi,rdx,rcx,r8.
>          */
>         movl    %esi, %r8d              /* arg5 (r8 ) <= 4th syscall arg (esi) */
>         xchg    %ecx, %edx              /* arg4 (rcx) <= 3rd syscall arg (edx) */
>                                         /* arg3 (rdx) <= 2nd syscall arg (ecx) */
>         movl    %ebx, %esi              /* arg2 (rsi) <= 1st syscall arg (ebx) */
>         movl    %eax, %edi              /* arg1 (rdi) <= syscall number  (eax) */
>         call    __audit_syscall_entry
>
> Btw., syscall auditing is not auditing syscall arguments #5 and #6?

Indeed.  That's the least of its problems.  Don't ever read that code
or you might accidentally git rm it.

--Andy

  reply	other threads:[~2015-06-12 23:28 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-06-09 18:54 [PATCH 1/5] x86/asm/entry/32: Fix fallout from r9 trick removal in SYSCALL code Denys Vlasenko
2015-06-09 18:54 ` [PATCH 2/5] x86/asm/entry/32: Explain reloading of registers after __audit_syscall_entry Denys Vlasenko
2015-06-10  7:09   ` [tip:x86/asm] x86/asm/entry/32: Explain reloading of registers after __audit_syscall_entry() tip-bot for Denys Vlasenko
2015-06-09 18:54 ` [PATCH 3/5] x86/asm/entry/32: Shorten __audit_syscall_entry args preparation Denys Vlasenko
2015-06-10  6:21   ` Ingo Molnar
2015-06-12 23:28     ` Andy Lutomirski [this message]
2015-06-10  7:10   ` [tip:x86/asm] x86/asm/entry/32: Shorten __audit_syscall_entry() " tip-bot for Denys Vlasenko
2015-06-09 18:54 ` [PATCH 4/5] x86/asm/entry/32: Replace RESTORE_RSI_RDI[_RDX] with open-coded 32-bit reads Denys Vlasenko
2015-06-09 19:01   ` Andy Lutomirski
2015-06-09 19:03     ` Denys Vlasenko
2015-06-09 19:11       ` Andy Lutomirski
2015-06-09 19:18         ` Denys Vlasenko
2015-06-09 19:27           ` Andy Lutomirski
2015-06-14  8:40   ` Ingo Molnar
2015-06-14 15:21     ` Denys Vlasenko
2015-06-15 20:20       ` Ingo Molnar
2015-06-16  0:24         ` Denys Vlasenko
2015-06-18  9:31           ` Ingo Molnar
2015-06-18 10:59             ` Denys Vlasenko
2015-06-09 18:54 ` [PATCH 5/5] x86/asm/entry/32: Simplify ptrace register shuffling Denys Vlasenko
2015-06-09 18:59   ` Andy Lutomirski
2015-06-09 19:14     ` Denys Vlasenko
2015-06-18  9:33   ` Ingo Molnar
2015-06-10  7:09 ` [tip:x86/asm] x86/asm/entry/32: Fix fallout from the R9 trick removal in the SYSCALL code tip-bot for Denys Vlasenko

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='CALCETrWp79ASjZ1f28_u6bVO4b0Mh4=n6H8VhbA=Y6qcQLG45w@mail.gmail.com' \
    --to=luto@amacapital.net \
    --cc=ast@plumgrid.com \
    --cc=bp@alien8.de \
    --cc=dvlasenk@redhat.com \
    --cc=fweisbec@gmail.com \
    --cc=hpa@zytor.com \
    --cc=keescook@chromium.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=oleg@redhat.com \
    --cc=rostedt@goodmis.org \
    --cc=torvalds@linux-foundation.org \
    --cc=wad@chromium.org \
    --cc=x86@kernel.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.