All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ingo Molnar <mingo@kernel.org>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
	Dave Hansen <dave@sr71.net>, Kees Cook <keescook@google.com>,
	"x86@kernel.org" <x86@kernel.org>,
	LKML <linux-kernel@vger.kernel.org>,
	Linux-MM <linux-mm@kvack.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Peter Zijlstra <a.p.zijlstra@chello.nl>,
	Andy Lutomirski <luto@kernel.org>, Borislav Petkov <bp@alien8.de>
Subject: Re: [PATCH 26/26] x86, pkeys: Documentation
Date: Sat, 3 Oct 2015 08:46:37 +0200	[thread overview]
Message-ID: <20151003064637.GA23054@gmail.com> (raw)
In-Reply-To: <560E7508.6080006@redhat.com>


* Paolo Bonzini <pbonzini@redhat.com> wrote:

> 
> 
> On 02/10/2015 13:58, Linus Torvalds wrote:
> > On Fri, Oct 2, 2015 at 7:49 AM, Paolo Bonzini <pbonzini@redhat.com> wrote:
> >> On 02/10/2015 00:48, Linus Torvalds wrote:
> >>> It's quite likely that you will find that compilers put read-only
> >>> constants in the text section, knowing that executable means readable.
> >>
> >> Not on x86 (because it has large immediates; RISC machines and s390 do
> >> put large constants in the text section).
> >>
> >> But at the very least jump tables reside in the .text seection.
> > 
> > Yes, at least traditionally gcc put things like the jump tables for
> > switch() statements immediately next to the code. That caused lots of
> > pain on the P4, where the L1 I$ and D$ were exclusive. I think that
> > caused gcc to then put the jump tables further away, and it might be
> > in a separate section these days - but it might also just be
> > "sufficiently aligned" that the L1 cache issue isn't in play any more.
> > 
> > Anyway, because of the P4 exclusive L1 I/D$ issue we can pretty much
> > rest easy knowing that the data accesses and text accesses should be
> > separated by at least one cacheline (maybe even 128 bytes - I think
> > the L4 used 64-byte line size, but it was sub-sections of a 128-byte
> > bigger line - but that might have been in the L2 only).
> > 
> > But I could easily see the compiler/linker still putting them in the
> > same ELF segment.
> 
> You're entirely right, it puts them in .rodata actually.  But .rodata is
> in the same segment as .text:
> 
> $ readelf --segments /bin/true
> ...
>  Section to Segment mapping:
>   Segment Sections...
>    00     
>    01     .interp 
>    02     .interp .note.ABI-tag .note.gnu.build-id .gnu.hash .dynsym
>           .dynstr .gnu.version .gnu.version_r .rela.dyn .rela.plt .init
>           .plt .text .fini .rodata .eh_frame_hdr .eh_frame 
>    03     .init_array .fini_array .jcr .data.rel.ro .dynamic .got .data .bss 
>    04     .dynamic 
>    05     .note.ABI-tag .note.gnu.build-id 
>    06     .eh_frame_hdr 
>    07     
>    08     .init_array .fini_array .jcr .data.rel.ro .dynamic .got 

Is there an easy(-ish) way (i.e. using compiler/linker flags, not linker scripts) 
to build the ELF binary in such a way so that non-code data:

          .rodata .eh_frame_hdr .eh_frame 

... gets put into a separate (readonly and non-executable) segment? That would 
enable things from the distro side AFAICS, right?

(assuming I'm reading the ELF dump right.)

Or does this need binutils surgery?

Thanks,

	Ingo

WARNING: multiple messages have this Message-ID (diff)
From: Ingo Molnar <mingo@kernel.org>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
	Dave Hansen <dave@sr71.net>, Kees Cook <keescook@google.com>,
	"x86@kernel.org" <x86@kernel.org>,
	LKML <linux-kernel@vger.kernel.org>,
	Linux-MM <linux-mm@kvack.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Peter Zijlstra <a.p.zijlstra@chello.nl>,
	Andy Lutomirski <luto@kernel.org>, Borislav Petkov <bp@alien8.de>
Subject: Re: [PATCH 26/26] x86, pkeys: Documentation
Date: Sat, 3 Oct 2015 08:46:37 +0200	[thread overview]
Message-ID: <20151003064637.GA23054@gmail.com> (raw)
In-Reply-To: <560E7508.6080006@redhat.com>


* Paolo Bonzini <pbonzini@redhat.com> wrote:

> 
> 
> On 02/10/2015 13:58, Linus Torvalds wrote:
> > On Fri, Oct 2, 2015 at 7:49 AM, Paolo Bonzini <pbonzini@redhat.com> wrote:
> >> On 02/10/2015 00:48, Linus Torvalds wrote:
> >>> It's quite likely that you will find that compilers put read-only
> >>> constants in the text section, knowing that executable means readable.
> >>
> >> Not on x86 (because it has large immediates; RISC machines and s390 do
> >> put large constants in the text section).
> >>
> >> But at the very least jump tables reside in the .text seection.
> > 
> > Yes, at least traditionally gcc put things like the jump tables for
> > switch() statements immediately next to the code. That caused lots of
> > pain on the P4, where the L1 I$ and D$ were exclusive. I think that
> > caused gcc to then put the jump tables further away, and it might be
> > in a separate section these days - but it might also just be
> > "sufficiently aligned" that the L1 cache issue isn't in play any more.
> > 
> > Anyway, because of the P4 exclusive L1 I/D$ issue we can pretty much
> > rest easy knowing that the data accesses and text accesses should be
> > separated by at least one cacheline (maybe even 128 bytes - I think
> > the L4 used 64-byte line size, but it was sub-sections of a 128-byte
> > bigger line - but that might have been in the L2 only).
> > 
> > But I could easily see the compiler/linker still putting them in the
> > same ELF segment.
> 
> You're entirely right, it puts them in .rodata actually.  But .rodata is
> in the same segment as .text:
> 
> $ readelf --segments /bin/true
> ...
>  Section to Segment mapping:
>   Segment Sections...
>    00     
>    01     .interp 
>    02     .interp .note.ABI-tag .note.gnu.build-id .gnu.hash .dynsym
>           .dynstr .gnu.version .gnu.version_r .rela.dyn .rela.plt .init
>           .plt .text .fini .rodata .eh_frame_hdr .eh_frame 
>    03     .init_array .fini_array .jcr .data.rel.ro .dynamic .got .data .bss 
>    04     .dynamic 
>    05     .note.ABI-tag .note.gnu.build-id 
>    06     .eh_frame_hdr 
>    07     
>    08     .init_array .fini_array .jcr .data.rel.ro .dynamic .got 

Is there an easy(-ish) way (i.e. using compiler/linker flags, not linker scripts) 
to build the ELF binary in such a way so that non-code data:

          .rodata .eh_frame_hdr .eh_frame 

... gets put into a separate (readonly and non-executable) segment? That would 
enable things from the distro side AFAICS, right?

(assuming I'm reading the ELF dump right.)

Or does this need binutils surgery?

Thanks,

	Ingo

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  reply	other threads:[~2015-10-03  6:46 UTC|newest]

Thread overview: 172+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-09-16 17:49 [PATCH 00/26] [RFCv2] x86: Memory Protection Keys Dave Hansen
2015-09-16 17:49 ` Dave Hansen
2015-09-16 17:49 ` [PATCH 04/26] x86, pku: define new CR4 bit Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-16 17:49 ` [PATCH 03/26] x86, pkeys: cpuid bit definition Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-16 17:49 ` [PATCH 02/26] x86, pkeys: Add Kconfig option Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-16 17:49 ` [PATCH 01/26] x86, fpu: add placeholder for Processor Trace XSAVE state Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-16 17:49 ` [PATCH 07/26] x86, pkeys: new page fault error code bit: PF_PK Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-16 17:49 ` [PATCH 05/26] x86, pkey: add PKRU xsave fields and data structure(s) Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-22 19:53   ` Thomas Gleixner
2015-09-22 19:53     ` Thomas Gleixner
2015-09-22 19:58     ` Dave Hansen
2015-09-22 19:58       ` Dave Hansen
2015-09-16 17:49 ` [PATCH 06/26] x86, pkeys: PTE bits for storing protection key Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-16 17:49 ` [PATCH 10/26] x86, pkeys: notify userspace about protection key faults Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-22 20:03   ` Thomas Gleixner
2015-09-22 20:03     ` Thomas Gleixner
2015-09-22 20:21     ` Dave Hansen
2015-09-22 20:21       ` Dave Hansen
2015-09-22 20:27       ` Thomas Gleixner
2015-09-22 20:27         ` Thomas Gleixner
2015-09-22 20:29         ` Dave Hansen
2015-09-22 20:29           ` Dave Hansen
2015-09-23  8:05           ` Ingo Molnar
2015-09-23  8:05             ` Ingo Molnar
2015-09-24  9:23   ` Ingo Molnar
2015-09-24  9:23     ` Ingo Molnar
2015-09-24  9:30     ` Ingo Molnar
2015-09-24  9:30       ` Ingo Molnar
2015-09-24 17:41       ` Dave Hansen
2015-09-24 17:41         ` Dave Hansen
2015-09-25  7:11         ` Ingo Molnar
2015-09-25  7:11           ` Ingo Molnar
2015-09-25 23:18           ` Dave Hansen
2015-09-25 23:18             ` Dave Hansen
2015-09-26  6:20             ` Ingo Molnar
2015-09-26  6:20               ` Ingo Molnar
2015-09-27 22:39               ` Dave Hansen
2015-09-27 22:39                 ` Dave Hansen
2015-09-28  5:59                 ` Ingo Molnar
2015-09-28  5:59                   ` Ingo Molnar
2015-09-24 17:15     ` Dave Hansen
2015-09-24 17:15       ` Dave Hansen
2015-09-28 19:25       ` Christian Borntraeger
2015-09-28 19:25         ` Christian Borntraeger
2015-09-28 19:32         ` Dave Hansen
2015-09-28 19:32           ` Dave Hansen
2015-09-16 17:49 ` [PATCH 09/26] x86, pkeys: arch-specific protection bits Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-16 17:49 ` [PATCH 08/26] x86, pkeys: store protection in high VMA flags Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-16 17:49 ` [PATCH 11/26] x86, pkeys: add functions for set/fetch PKRU Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-22 20:05   ` Thomas Gleixner
2015-09-22 20:05     ` Thomas Gleixner
2015-09-22 20:22     ` Dave Hansen
2015-09-22 20:22       ` Dave Hansen
2015-09-16 17:49 ` [PATCH 14/26] x86, pkeys: check VMAs and PTEs for protection keys Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-16 17:49 ` [PATCH 12/26] mm: factor out VMA fault permission checking Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-16 17:49 ` [PATCH 13/26] mm: simplify get_user_pages() PTE bit handling Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-16 17:49 ` [PATCH 15/26] x86, pkeys: optimize fault handling in access_error() Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-16 17:49 ` [PATCH 16/26] x86, pkeys: dump PKRU with other kernel registers Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-16 17:49 ` [PATCH 17/26] x86, pkeys: dump PTE pkey in /proc/pid/smaps Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-16 17:49 ` [PATCH 19/26] [NEWSYSCALL] mm, multi-arch: pass a protection key in to calc_vm_flag_bits() Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-16 17:49 ` [PATCH 20/26] [NEWSYSCALL] mm: implement new mprotect_pkey() system call Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-16 17:49 ` [PATCH 18/26] x86, pkeys: add Kconfig prompt to existing config option Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-16 17:49 ` [PATCH 22/26] [HIJACKPROT] mm: Pass the 4-bit protection key in via PROT_ bits to syscalls Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-16 17:49 ` [PATCH 21/26] [NEWSYSCALL] x86: wire up mprotect_key() system call Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-16 17:49 ` [PATCH 24/26] [HIJACKPROT] x86, pkeys: mask off pkeys bits in mprotect() Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-16 17:49 ` [PATCH 25/26] x86, pkeys: actually enable Memory Protection Keys in CPU Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-16 17:49 ` [PATCH 23/26] [HIJACKPROT] x86, pkeys: add x86 version of arch_validate_prot() Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-16 17:49 ` [PATCH 26/26] x86, pkeys: Documentation Dave Hansen
2015-09-16 17:49   ` Dave Hansen
2015-09-20  8:55   ` Ingo Molnar
2015-09-20  8:55     ` Ingo Molnar
2015-09-21  4:34     ` Dave Hansen
2015-09-21  4:34       ` Dave Hansen
2015-09-24  9:49       ` Ingo Molnar
2015-09-24  9:49         ` Ingo Molnar
2015-09-24 19:10         ` Dave Hansen
2015-09-24 19:10           ` Dave Hansen
2015-09-24 19:17           ` Andy Lutomirski
2015-09-24 19:17             ` Andy Lutomirski
2015-09-25  7:16             ` Ingo Molnar
2015-09-25  7:16               ` Ingo Molnar
2015-09-25  6:15           ` Ingo Molnar
2015-09-25  6:15             ` Ingo Molnar
2015-10-01 11:17           ` Ingo Molnar
2015-10-01 11:17             ` Ingo Molnar
2015-10-01 20:39             ` Kees Cook
2015-10-01 20:39               ` Kees Cook
2015-10-01 20:45               ` Andy Lutomirski
2015-10-01 20:45                 ` Andy Lutomirski
2015-10-02  6:23                 ` Ingo Molnar
2015-10-02  6:23                   ` Ingo Molnar
2015-10-02 17:50                   ` Dave Hansen
2015-10-02 17:50                     ` Dave Hansen
2015-10-03  7:27                     ` Ingo Molnar
2015-10-03  7:27                       ` Ingo Molnar
2015-10-06 23:28                       ` Dave Hansen
2015-10-06 23:28                         ` Dave Hansen
2015-10-07  7:11                         ` Ingo Molnar
2015-10-07  7:11                           ` Ingo Molnar
2015-10-16 15:12                       ` Dave Hansen
2015-10-16 15:12                         ` Dave Hansen
2015-10-21 18:55                         ` Andy Lutomirski
2015-10-21 18:55                           ` Andy Lutomirski
2015-10-21 19:11                           ` Dave Hansen
2015-10-21 19:11                             ` Dave Hansen
2015-10-21 23:22                             ` Andy Lutomirski
2015-10-21 23:22                               ` Andy Lutomirski
2015-10-01 20:58               ` Dave Hansen
2015-10-01 20:58                 ` Dave Hansen
2015-10-01 22:33               ` Dave Hansen
2015-10-01 22:35                 ` Kees Cook
2015-10-01 22:35                   ` Kees Cook
2015-10-01 22:39                   ` Dave Hansen
2015-10-01 22:39                     ` Dave Hansen
2015-10-01 22:48                 ` Linus Torvalds
2015-10-01 22:48                   ` Linus Torvalds
2015-10-01 22:56                   ` Dave Hansen
2015-10-01 22:56                     ` Dave Hansen
2015-10-02  1:38                     ` Linus Torvalds
2015-10-02  1:38                       ` Linus Torvalds
2015-10-02 18:08                       ` Dave Hansen
2015-10-02 18:08                         ` Dave Hansen
2015-10-02  7:09                   ` Ingo Molnar
2015-10-02  7:09                     ` Ingo Molnar
2015-10-03  6:59                     ` Ingo Molnar
2015-10-03  6:59                       ` Ingo Molnar
2015-10-02 11:49                   ` Paolo Bonzini
2015-10-02 11:49                     ` Paolo Bonzini
2015-10-02 11:58                     ` Linus Torvalds
2015-10-02 11:58                       ` Linus Torvalds
2015-10-02 12:14                       ` Paolo Bonzini
2015-10-02 12:14                         ` Paolo Bonzini
2015-10-03  6:46                         ` Ingo Molnar [this message]
2015-10-03  6:46                           ` Ingo Molnar
2015-10-01 22:57                 ` Andy Lutomirski
2015-10-01 22:57                   ` Andy Lutomirski
2015-10-02  6:09                 ` Ingo Molnar
2015-10-02  6:09                   ` Ingo Molnar
2015-10-03  8:17         ` Ingo Molnar
2015-10-03  8:17           ` Ingo Molnar
2015-10-07 20:24           ` Dave Hansen
2015-10-07 20:24             ` Dave Hansen
2015-10-07 20:39             ` Andy Lutomirski
2015-10-07 20:39               ` Andy Lutomirski
2015-10-07 20:47               ` Dave Hansen
2015-10-07 20:47                 ` Dave Hansen
2015-09-16 17:51 ` Fwd: [PATCH 00/26] [RFCv2] x86: Memory Protection Keys Dave Hansen

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=20151003064637.GA23054@gmail.com \
    --to=mingo@kernel.org \
    --cc=a.p.zijlstra@chello.nl \
    --cc=akpm@linux-foundation.org \
    --cc=bp@alien8.de \
    --cc=dave@sr71.net \
    --cc=keescook@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=luto@kernel.org \
    --cc=pbonzini@redhat.com \
    --cc=torvalds@linux-foundation.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.