Re: [PATCH v2 6/8] x86/entry: add unwind hint annotations

From	Andy Lutomirski <luto@kernel.org>
Newsgroups	linux.kernel
Subject	Re: [PATCH v2 6/8] x86/entry: add unwind hint annotations
Date	2017-06-29 23:20 +0200
Message-ID	<tXP0Z-4yl-5@gated-at.bofh.it> (permalink)
References	<tXmV4-8uj-15@gated-at.bofh.it> <tXmV4-8uj-23@gated-at.bofh.it> <tXLTs-2de-23@gated-at.bofh.it> <tXMPw-2Ms-23@gated-at.bofh.it> <tXMZd-354-49@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On Thu, Jun 29, 2017 at 12:05 PM, Josh Poimboeuf <jpoimboe@redhat.com> wrote:
> On Thu, Jun 29, 2017 at 11:50:18AM -0700, Andy Lutomirski wrote:
>> On Thu, Jun 29, 2017 at 10:53 AM, Josh Poimboeuf <jpoimboe@redhat.com> wrote:
>> > There's a bug here that will need a small change to the entry code.
>> >
>> > Mike Galbraith reported:
>> >
>> >   WARNING: can't dereference registers at ffffc900089d7e08 for ip ffffffff81740bbb
>> >
>> > After some looking I found that it's caused by the following code
>> > snippet in the 'interrupt' macro in entry_64.S:
>> >
>> >         /*
>> >          * Save previous stack pointer, optionally switch to interrupt stack.
>> >          * irq_count is used to check if a CPU is already on an interrupt stack
>> >          * or not. While this is essentially redundant with preempt_count it is
>> >          * a little cheaper to use a separate counter in the PDA (short of
>> >          * moving irq_enter into assembly, which would be too much work)
>> >          */
>> >         movq    %rsp, %rdi
>> >         incl    PER_CPU_VAR(irq_count)
>> >         cmovzq  PER_CPU_VAR(irq_stack_ptr), %rsp
>> >         UNWIND_HINT_REGS base=rdi
>> >         pushq   %rdi
>> >         UNWIND_HINT_REGS indirect=1
>> >
>> > The problem is that it's changing the stack pointer *before* writing the
>> > previous stack pointer (push %rdi).  So when unwinding from an NMI which
>> > hit between the rsp write and the rdi push, the unwinder tries to access
>> > the regs on the previous stack (by reading rdi), but the previous stack
>> > pointer isn't there yet, so the access is considered out of bounds.
>>
>> Ugh, that code.  Does this problem go away with this patch applied:
>>
>> https://git.kernel.org/pub/scm/linux/kernel/git/luto/linux.git/commit/?h=x86/entry_ist&id=2231ec7e0bcc1a2bc94a17081511ab54cc6badd1
>>
>> If so, want to update the patch for new kernels (shouldn't conflict
>> with anything except your unwind hints)?
>
> I don't think that patch will fix it, because it still updates rsp
> *before* writing the old rsp on the new stack.  So there's still a
> window where the "previous stack" pointer is missing.

But it's in a register.  Is undwarf not able to grok that?  I have no
fundamental problem with pushing it to the new stack first, but the
actual asm is nastier because we don't have an addressing mode that's
*(*(gs:blahblahblah)) = reg.

At least my patch makes all the copied of this code identical so the
problem can be solved only once.

Thread

[PATCH v2 0/8] x86: undwarf unwinder Josh Poimboeuf <jpoimboe@redhat.com> - 2017-06-28 17:20 +0200
  [PATCH v2 4/8] objtool: add undwarf debuginfo generation Josh Poimboeuf <jpoimboe@redhat.com> - 2017-06-28 17:20 +0200
    Re: [PATCH v2 4/8] objtool: add undwarf debuginfo generation Ingo Molnar <mingo@kernel.org> - 2017-06-29 09:20 +0200
      Re: [PATCH v2 4/8] objtool: add undwarf debuginfo generation Josh Poimboeuf <jpoimboe@redhat.com> - 2017-06-29 15:50 +0200
    Re: [PATCH v2 4/8] objtool: add undwarf debuginfo generation Ingo Molnar <mingo@kernel.org> - 2017-06-29 09:30 +0200
      Re: [PATCH v2 4/8] objtool: add undwarf debuginfo generation Josh Poimboeuf <jpoimboe@redhat.com> - 2017-06-29 16:10 +0200
        Re: [PATCH v2 4/8] objtool: add undwarf debuginfo generation Ingo Molnar <mingo@kernel.org> - 2017-06-29 16:50 +0200
          Re: [PATCH v2 4/8] objtool: add undwarf debuginfo generation Josh Poimboeuf <jpoimboe@redhat.com> - 2017-06-29 17:10 +0200
  [PATCH v2 7/8] x86/asm: add unwind hint annotations to sync_core() Josh Poimboeuf <jpoimboe@redhat.com> - 2017-06-28 17:20 +0200
  [PATCH v2 2/8] objtool, x86: add several functions and files to the objtool whitelist Josh Poimboeuf <jpoimboe@redhat.com> - 2017-06-28 17:20 +0200
    [tip:core/objtool] objtool, x86: Add several functions and files to  the objtool whitelist tip-bot for Josh Poimboeuf <tipbot@zytor.com> - 2017-06-30 15:20 +0200
  [PATCH v2 6/8] x86/entry: add unwind hint annotations Josh Poimboeuf <jpoimboe@redhat.com> - 2017-06-28 17:20 +0200
    Re: [PATCH v2 6/8] x86/entry: add unwind hint annotations Josh Poimboeuf <jpoimboe@redhat.com> - 2017-06-29 20:00 +0200
      Re: [PATCH v2 6/8] x86/entry: add unwind hint annotations Andy Lutomirski <luto@kernel.org> - 2017-06-29 21:00 +0200
        Re: [PATCH v2 6/8] x86/entry: add unwind hint annotations Josh Poimboeuf <jpoimboe@redhat.com> - 2017-06-29 21:10 +0200
          Re: [PATCH v2 6/8] x86/entry: add unwind hint annotations Andy Lutomirski <luto@kernel.org> - 2017-06-29 23:20 +0200
            Re: [PATCH v2 6/8] x86/entry: add unwind hint annotations Josh Poimboeuf <jpoimboe@redhat.com> - 2017-06-29 23:50 +0200
              Re: [PATCH v2 6/8] x86/entry: add unwind hint annotations Andy Lutomirski <luto@amacapital.net> - 2017-06-30 01:00 +0200
                Re: [PATCH v2 6/8] x86/entry: add unwind hint annotations Josh Poimboeuf <jpoimboe@redhat.com> - 2017-06-30 04:20 +0200
                Re: [PATCH v2 6/8] x86/entry: add unwind hint annotations Andy Lutomirski <luto@kernel.org> - 2017-06-30 07:10 +0200
                Re: [PATCH v2 6/8] x86/entry: add unwind hint annotations Andy Lutomirski <luto@kernel.org> - 2017-06-30 07:50 +0200
                Re: [PATCH v2 6/8] x86/entry: add unwind hint annotations Josh Poimboeuf <jpoimboe@redhat.com> - 2017-06-30 15:20 +0200
                Re: [PATCH v2 6/8] x86/entry: add unwind hint annotations Andy Lutomirski <luto@kernel.org> - 2017-06-30 17:50 +0200
                Re: [PATCH v2 6/8] x86/entry: add unwind hint annotations Josh Poimboeuf <jpoimboe@redhat.com> - 2017-06-30 18:00 +0200
                Re: [PATCH v2 6/8] x86/entry: add unwind hint annotations Andy Lutomirski <luto@kernel.org> - 2017-06-30 18:00 +0200
                Re: [PATCH v2 6/8] x86/entry: add unwind hint annotations Josh Poimboeuf <jpoimboe@redhat.com> - 2017-06-30 18:20 +0200
  [PATCH v2 1/8] objtool: move checking code to check.c Josh Poimboeuf <jpoimboe@redhat.com> - 2017-06-28 17:20 +0200
  Re: [PATCH v2 0/8] x86: undwarf unwinder Ingo Molnar <mingo@kernel.org> - 2017-06-29 10:00 +0200
    Re: [PATCH v2 0/8] x86: undwarf unwinder Josh Poimboeuf <jpoimboe@redhat.com> - 2017-06-29 16:20 +0200
      Re: [PATCH v2 0/8] x86: undwarf unwinder Josh Poimboeuf <jpoimboe@redhat.com> - 2017-06-29 21:20 +0200
  Re: [PATCH v2 3/8] objtool: stack validation 2.0 Ingo Molnar <mingo@kernel.org> - 2017-06-30 10:40 +0200
    Re: [PATCH v2 3/8] objtool: stack validation 2.0 Josh Poimboeuf <jpoimboe@redhat.com> - 2017-06-30 15:30 +0200
      Re: [PATCH v2 3/8] objtool: stack validation 2.0 Josh Poimboeuf <jpoimboe@redhat.com> - 2017-06-30 15:30 +0200
    [PATCH] objtool: silence warnings for functions which use iret Josh Poimboeuf <jpoimboe@redhat.com> - 2017-06-30 16:10 +0200
      [tip:core/objtool] objtool: Silence warnings for functions which  use IRET tip-bot for Josh Poimboeuf <tipbot@zytor.com> - 2017-06-30 20:00 +0200

csiph-web