1. 29 Sep, 2017 2 commits
    • Andrey Ryabinin's avatar
      x86/asm: Use register variable to get stack pointer value · 196bd485
      Andrey Ryabinin authored
      Currently we use current_stack_pointer() function to get the value
      of the stack pointer register. Since commit:
        f5caf621 ("x86/asm: Fix inline asm call constraints for Clang")
      ... we have a stack register variable declared. It can be used instead of
      current_stack_pointer() function which allows to optimize away some
      excessive "mov %rsp, %<dst>" instructions:
       -mov    %rsp,%rdx
       -sub    %rdx,%rax
       -cmp    $0x3fff,%rax
       -ja     ffffffff810722fd <ist_begin_non_atomic+0x2d>
       +sub    %rsp,%rax
       +cmp    $0x3fff,%rax
       +ja     ffffffff810722fa <ist_begin_non_atomic+0x2a>
      Remove current_stack_pointer(), rename __asm_call_sp to current_stack_pointer
      and use it instead of the removed function.
      Signed-off-by: default avatarAndrey Ryabinin <aryabinin@virtuozzo.com>
      Reviewed-by: default avatarJosh Poimboeuf <jpoimboe@redhat.com>
      Cc: Andy Lutomirski <luto@kernel.org>
      Cc: Linus Torvalds <torvalds@linux-foundation.org>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Thomas Gleixner <tglx@linutronix.de>
      Link: http://lkml.kernel.org/r/20170929141537.29167-1-aryabinin@virtuozzo.comSigned-off-by: default avatarIngo Molnar <mingo@kernel.org>
    • Josh Poimboeuf's avatar
      x86/asm: Fix inline asm call constraints for GCC 4.4 · 520a13c5
      Josh Poimboeuf authored
      The kernel test bot (run by Xiaolong Ye) reported that the following commit:
        f5caf621 ("x86/asm: Fix inline asm call constraints for Clang")
      is causing double faults in a kernel compiled with GCC 4.4.
      Linus subsequently diagnosed the crash pattern and the buggy commit and found that
      the issue is with this code:
        register unsigned int __asm_call_sp asm("esp");
        #define ASM_CALL_CONSTRAINT "+r" (__asm_call_sp)
      Even on a 64-bit kernel, it's using ESP instead of RSP.  That causes GCC
      to produce the following bogus code:
        ffffffff8147461d:       89 e0                   mov    %esp,%eax
        ffffffff8147461f:       4c 89 f7                mov    %r14,%rdi
        ffffffff81474622:       4c 89 fe                mov    %r15,%rsi
        ffffffff81474625:       ba 20 00 00 00          mov    $0x20,%edx
        ffffffff8147462a:       89 c4                   mov    %eax,%esp
        ffffffff8147462c:       e8 bf 52 05 00          callq  ffffffff814c98f0 <copy_user_generic_unrolled>
      Despite the absurdity of it backing up and restoring the stack pointer
      for no reason, the bug is actually the fact that it's only backing up
      and restoring the lower 32 bits of the stack pointer.  The upper 32 bits
      are getting cleared out, corrupting the stack pointer.
      So change the '__asm_call_sp' register variable to be associated with
      the actual full-size stack pointer.
      This also requires changing the __ASM_SEL() macro to be based on the
      actual compiled arch size, rather than the CONFIG value, because
      CONFIG_X86_64 compiles some files with '-m32' (e.g., realmode and vdso).
      Otherwise Clang fails to build the kernel because it complains about the
      use of a 64-bit register (RSP) in a 32-bit file.
      Reported-and-Bisected-and-Tested-by: default avatarkernel test robot <xiaolong.ye@intel.com>
      Diagnosed-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      Signed-off-by: default avatarJosh Poimboeuf <jpoimboe@redhat.com>
      Cc: Alexander Potapenko <glider@google.com>
      Cc: Andrey Ryabinin <aryabinin@virtuozzo.com>
      Cc: Andy Lutomirski <luto@kernel.org>
      Cc: Arnd Bergmann <arnd@arndb.de>
      Cc: Dmitriy Vyukov <dvyukov@google.com>
      Cc: LKP <lkp@01.org>
      Cc: Linus Torvalds <torvalds@linux-foundation.org>
      Cc: Matthias Kaehlcke <mka@chromium.org>
      Cc: Miguel Bernal Marin <miguel.bernal.marin@linux.intel.com>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Thomas Gleixner <tglx@linutronix.de>
      Fixes: f5caf621 ("x86/asm: Fix inline asm call constraints for Clang")
      Link: http://lkml.kernel.org/r/20170928215826.6sdpmwtkiydiytim@trebleSigned-off-by: default avatarIngo Molnar <mingo@kernel.org>
  2. 23 Sep, 2017 1 commit
    • Josh Poimboeuf's avatar
      x86/asm: Fix inline asm call constraints for Clang · f5caf621
      Josh Poimboeuf authored
      For inline asm statements which have a CALL instruction, we list the
      stack pointer as a constraint to convince GCC to ensure the frame
      pointer is set up first:
        static inline void foo()
      	register void *__sp asm(_ASM_SP);
      	asm("call bar" : "+r" (__sp))
      Unfortunately, that pattern causes Clang to corrupt the stack pointer.
      The fix is easy: convert the stack pointer register variable to a global
      It should be noted that the end result is different based on the GCC
      version.  With GCC 6.4, this patch has exactly the same result as
      	defconfig	defconfig-nofp	distro		distro-nofp
       before	9820389		9491555		8816046		8516940
       after	9820389		9491555		8816046		8516940
      With GCC 7.2, however, GCC's behavior has changed.  It now changes its
      behavior based on the conversion of the register variable to a global.
      That somehow convinces it to *always* set up the frame pointer before
      inserting *any* inline asm.  (Therefore, listing the variable as an
      output constraint is a no-op and is no longer necessary.)  It's a bit
      overkill, but the performance impact should be negligible.  And in fact,
      there's a nice improvement with frame pointers disabled:
      	defconfig	defconfig-nofp	distro		distro-nofp
       before	9796316		9468236		9076191		8790305
       after	9796957		9464267		9076381		8785949
      So in summary, while listing the stack pointer as an output constraint
      is no longer necessary for newer versions of GCC, it's still needed for
      older versions.
      Suggested-by: default avatarAndrey Ryabinin <aryabinin@virtuozzo.com>
      Reported-by: default avatarMatthias Kaehlcke <mka@chromium.org>
      Signed-off-by: default avatarJosh Poimboeuf <jpoimboe@redhat.com>
      Cc: Alexander Potapenko <glider@google.com>
      Cc: Andy Lutomirski <luto@kernel.org>
      Cc: Arnd Bergmann <arnd@arndb.de>
      Cc: Dmitriy Vyukov <dvyukov@google.com>
      Cc: Linus Torvalds <torvalds@linux-foundation.org>
      Cc: Miguel Bernal Marin <miguel.bernal.marin@linux.intel.com>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Thomas Gleixner <tglx@linutronix.de>
      Link: http://lkml.kernel.org/r/3db862e970c432ae823cf515c52b54fec8270e0e.1505942196.git.jpoimboe@redhat.comSigned-off-by: default avatarIngo Molnar <mingo@kernel.org>
  3. 17 Aug, 2017 1 commit
    • Kees Cook's avatar
      locking/refcounts, x86/asm: Implement fast refcount overflow protection · 7a46ec0e
      Kees Cook authored
      This implements refcount_t overflow protection on x86 without a noticeable
      performance impact, though without the fuller checking of REFCOUNT_FULL.
      This is done by duplicating the existing atomic_t refcount implementation
      but with normally a single instruction added to detect if the refcount
      has gone negative (e.g. wrapped past INT_MAX or below zero). When detected,
      the handler saturates the refcount_t to INT_MIN / 2. With this overflow
      protection, the erroneous reference release that would follow a wrap back
      to zero is blocked from happening, avoiding the class of refcount-overflow
      use-after-free vulnerabilities entirely.
      Only the overflow case of refcounting can be perfectly protected, since
      it can be detected and stopped before the reference is freed and left to
      be abused by an attacker. There isn't a way to block early decrements,
      and while REFCOUNT_FULL stops increment-from-zero cases (which would
      be the state _after_ an early decrement and stops potential double-free
      conditions), this fast implementation does not, since it would require
      the more expensive cmpxchg loops. Since the overflow case is much more
      common (e.g. missing a "put" during an error path), this protection
      provides real-world protection. For example, the two public refcount
      overflow use-after-free exploits published in 2016 would have been
      rendered unexploitable:
      This implementation does, however, notice an unchecked decrement to zero
      (i.e. caller used refcount_dec() instead of refcount_dec_and_test() and it
      resulted in a zero). Decrements under zero are noticed (since they will
      have resulted in a negative value), though this only indicates that a
      use-after-free may have already happened. Such notifications are likely
      avoidable by an attacker that has already exploited a use-after-free
      vulnerability, but it's better to have them reported than allow such
      conditions to remain universally silent.
      On first overflow detection, the refcount value is reset to INT_MIN / 2
      (which serves as a saturation value) and a report and stack trace are
      produced. When operations detect only negative value results (such as
      changing an already saturated value), saturation still happens but no
      notification is performed (since the value was already saturated).
      On the matter of races, since the entire range beyond INT_MAX but before
      0 is negative, every operation at INT_MIN / 2 will trap, leaving no
      overflow-only race condition.
      As for performance, this implementation adds a single "js" instruction
      to the regular execution flow of a copy of the standard atomic_t refcount
      operations. (The non-"and_test" refcount_dec() function, which is uncommon
      in regular refcount design patterns, has an additional "jz" instruction
      to detect reaching exactly zero.) Since this is a forward jump, it is by
      default the non-predicted path, which will be reinforced by dynamic branch
      prediction. The result is this protection having virtually no measurable
      change in performance over standard atomic_t operations. The error path,
      located in .text.unlikely, saves the refcount location and then uses UD0
      to fire a refcount exception handler, which resets the refcount, handles
      reporting, and returns to regular execution. This keeps the changes to
      .text size minimal, avoiding return jumps and open-coded calls to the
      error reporting routine.
      Example assembly comparison:
      refcount_inc() before:
        ffffffff81546149:       f0 ff 45 f4             lock incl -0xc(%rbp)
      refcount_inc() after:
        ffffffff81546149:       f0 ff 45 f4             lock incl -0xc(%rbp)
        ffffffff8154614d:       0f 88 80 d5 17 00       js     ffffffff816c36d3
        ffffffff816c36d3:       48 8d 4d f4             lea    -0xc(%rbp),%rcx
        ffffffff816c36d7:       0f ff                   (bad)
      These are the cycle counts comparing a loop of refcount_inc() from 1
      to INT_MAX and back down to 0 (via refcount_dec_and_test()), between
      unprotected refcount_t (atomic_t), fully protected REFCOUNT_FULL
      (refcount_t-full), and this overflow-protected refcount (refcount_t-fast):
        2147483646 refcount_inc()s and 2147483647 refcount_dec_and_test()s:
      		    cycles		protections
        atomic_t           82249267387	none
        refcount_t-fast    82211446892	overflow, untested dec-to-zero
        refcount_t-full   144814735193	overflow, untested dec-to-zero, inc-from-zero
      This code is a modified version of the x86 PAX_REFCOUNT atomic_t
      overflow defense from the last public patch of PaX/grsecurity, based
      on my understanding of the code. Changes or omissions from the original
      code are mine and don't reflect the original grsecurity/PaX code. Thanks
      to PaX Team for various suggestions for improvement for repurposing this
      code to be a refcount-only protection.
      Signed-off-by: default avatarKees Cook <keescook@chromium.org>
      Reviewed-by: default avatarJosh Poimboeuf <jpoimboe@redhat.com>
      Cc: Alexey Dobriyan <adobriyan@gmail.com>
      Cc: Andrew Morton <akpm@linux-foundation.org>
      Cc: Arnd Bergmann <arnd@arndb.de>
      Cc: Christoph Hellwig <hch@infradead.org>
      Cc: David S. Miller <davem@davemloft.net>
      Cc: Davidlohr Bueso <dave@stgolabs.net>
      Cc: Elena Reshetova <elena.reshetova@intel.com>
      Cc: Eric Biggers <ebiggers3@gmail.com>
      Cc: Eric W. Biederman <ebiederm@xmission.com>
      Cc: Greg KH <gregkh@linuxfoundation.org>
      Cc: Hans Liljestrand <ishkamiel@gmail.com>
      Cc: James Bottomley <James.Bottomley@hansenpartnership.com>
      Cc: Jann Horn <jannh@google.com>
      Cc: Linus Torvalds <torvalds@linux-foundation.org>
      Cc: Manfred Spraul <manfred@colorfullife.com>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Rik van Riel <riel@redhat.com>
      Cc: Serge E. Hallyn <serge@hallyn.com>
      Cc: Thomas Gleixner <tglx@linutronix.de>
      Cc: arozansk@redhat.com
      Cc: axboe@kernel.dk
      Cc: kernel-hardening@lists.openwall.com
      Cc: linux-arch <linux-arch@vger.kernel.org>
      Link: http://lkml.kernel.org/r/20170815161924.GA133115@beastSigned-off-by: default avatarIngo Molnar <mingo@kernel.org>
  4. 05 May, 2017 1 commit
    • Matthias Kaehlcke's avatar
      x86/mm/kaslr: Use the _ASM_MUL macro for multiplication to work around Clang incompatibility · 121843eb
      Matthias Kaehlcke authored
      The constraint "rm" allows the compiler to put mix_const into memory.
      When the input operand is a memory location then MUL needs an operand
      size suffix, since Clang can't infer the multiplication width from the
      Add and use the _ASM_MUL macro which determines the operand size and
      resolves to the NUL instruction with the corresponding suffix.
      This fixes the following error when building with clang:
        CC      arch/x86/lib/kaslr.o
        /tmp/kaslr-dfe1ad.s: Assembler messages:
        /tmp/kaslr-dfe1ad.s:182: Error: no instruction mnemonic suffix given and no register operands; can't size instruction
      Signed-off-by: default avatarMatthias Kaehlcke <mka@chromium.org>
      Cc: Grant Grundler <grundler@chromium.org>
      Cc: Greg Hackmann <ghackmann@google.com>
      Cc: Kees Cook <keescook@chromium.org>
      Cc: Linus Torvalds <torvalds@linux-foundation.org>
      Cc: Michael Davidson <md@google.com>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Thomas Gleixner <tglx@linutronix.de>
      Link: http://lkml.kernel.org/r/20170501224741.133938-1-mka@chromium.orgSigned-off-by: default avatarIngo Molnar <mingo@kernel.org>
  5. 08 Jun, 2016 1 commit
  6. 18 Feb, 2016 1 commit
  7. 14 May, 2015 1 commit
  8. 24 Apr, 2014 1 commit
    • Masami Hiramatsu's avatar
      kprobes: Introduce NOKPROBE_SYMBOL() macro to maintain kprobes blacklist · 376e2424
      Masami Hiramatsu authored
      Introduce NOKPROBE_SYMBOL() macro which builds a kprobes
      blacklist at kernel build time.
      The usage of this macro is similar to EXPORT_SYMBOL(),
      placed after the function definition:
      Since this macro will inhibit inlining of static/inline
      functions, this patch also introduces a nokprobe_inline macro
      for static/inline functions. In this case, we must use
      NOKPROBE_SYMBOL() for the inline function caller.
      When CONFIG_KPROBES=y, the macro stores the given function
      address in the "_kprobe_blacklist" section.
      Since the data structures are not fully initialized by the
      macro (because there is no "size" information),  those
      are re-initialized at boot time by using kallsyms.
      Signed-off-by: default avatarMasami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
      Link: http://lkml.kernel.org/r/20140417081705.26341.96719.stgit@ltc230.yrl.intra.hitachi.co.jp
      Cc: Alok Kataria <akataria@vmware.com>
      Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
      Cc: Andrew Morton <akpm@linux-foundation.org>
      Cc: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
      Cc: Arnd Bergmann <arnd@arndb.de>
      Cc: Christopher Li <sparse@chrisli.org>
      Cc: Chris Wright <chrisw@sous-sol.org>
      Cc: David S. Miller <davem@davemloft.net>
      Cc: Jan-Simon Möller <dl9pf@gmx.de>
      Cc: Jeremy Fitzhardinge <jeremy@goop.org>
      Cc: Linus Torvalds <torvalds@linux-foundation.org>
      Cc: Randy Dunlap <rdunlap@infradead.org>
      Cc: Rusty Russell <rusty@rustcorp.com.au>
      Cc: linux-arch@vger.kernel.org
      Cc: linux-doc@vger.kernel.org
      Cc: linux-sparse@vger.kernel.org
      Cc: virtualization@lists.linux-foundation.org
      Signed-off-by: default avatarIngo Molnar <mingo@kernel.org>
  9. 29 Aug, 2013 1 commit
  10. 21 Apr, 2012 1 commit
    • H. Peter Anvin's avatar
      x86, extable: Switch to relative exception table entries · 70627654
      H. Peter Anvin authored
      Switch to using relative exception table entries on x86.  On i386,
      this has the advantage that the exception table entries don't need to
      be relocated; on x86-64 this means the exception table entries take up
      only half the space.
      In either case, a 32-bit delta is sufficient, as the range of kernel
      code addresses is limited.
      Since part of the goal is to avoid needing to adjust the entries when
      the kernel is relocated, the old trick of using addresses in the NULL
      pointer range to indicate uaccess_err no longer works (and unlike RISC
      architectures we can't use a flag bit); instead use an delta just
      below +2G to indicate these special entries.  The reach is still
      limited to a single instruction.
      Signed-off-by: default avatarH. Peter Anvin <hpa@zytor.com>
      Cc: David Daney <david.daney@cavium.com>
      Link: http://lkml.kernel.org/r/CA%2B55aFyijf43qSu3N9nWHEBwaGbb7T2Oq9A=9EyR=Jtyqfq_cQ@mail.gmail.com
  11. 20 Apr, 2012 3 commits
  12. 21 Jul, 2011 2 commits
    • Jan Beulich's avatar
      x86: Fix write lock scalability 64-bit issue · a750036f
      Jan Beulich authored
      With the write lock path simply subtracting RW_LOCK_BIAS there
      is, on large systems, the theoretical possibility of overflowing
      the 32-bit value that was used so far (namely if 128 or more
      CPUs manage to do the subtraction, but don't get to do the
      inverse addition in the failure path quickly enough).
      A first measure is to modify RW_LOCK_BIAS itself - with the new
      value chosen, it is good for up to 2048 CPUs each allowed to
      nest over 2048 times on the read path without causing an issue.
      Quite possibly it would even be sufficient to adjust the bias a
      little further, assuming that allowing for significantly less
      nesting would suffice.
      However, as the original value chosen allowed for even more
      nesting levels, to support more than 2048 CPUs (possible
      currently only for 64-bit kernels) the lock itself gets widened
      to 64 bits.
      Signed-off-by: default avatarJan Beulich <jbeulich@novell.com>
      Cc: Linus Torvalds <torvalds@linux-foundation.org>
      Cc: Andrew Morton <akpm@linux-foundation.org>
      Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
      Link: http://lkml.kernel.org/r/4E258E0D020000780004E3F0@nat28.tlf.novell.comSigned-off-by: default avatarIngo Molnar <mingo@elte.hu>
    • Jan Beulich's avatar
      x86: Unify rwlock assembly implementation · 4625cd63
      Jan Beulich authored
      Rather than having two functionally identical implementations
      for 32- and 64-bit configurations, extend the existing assembly
      abstractions enough to fold the two rwlock implementations into
      a shared one.
      Signed-off-by: default avatarJan Beulich <jbeulich@novell.com>
      Cc: Linus Torvalds <torvalds@linux-foundation.org>
      Cc: Andrew Morton <akpm@linux-foundation.org>
      Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
      Link: http://lkml.kernel.org/r/4E258DD7020000780004E3EA@nat28.tlf.novell.comSigned-off-by: default avatarIngo Molnar <mingo@elte.hu>
  13. 31 Aug, 2009 1 commit
  14. 23 Oct, 2008 2 commits
  15. 19 Aug, 2008 1 commit
  16. 22 Jul, 2008 1 commit
    • Vegard Nossum's avatar
      x86: consolidate header guards · 77ef50a5
      Vegard Nossum authored
      This patch is the result of an automatic script that consolidates the
      format of all the headers in include/asm-x86/.
      The format:
      1. No leading underscore. Names with leading underscores are reserved.
      2. Pathname components are separated by two underscores. So we can
         distinguish between mm_types.h and mm/types.h.
      3. Everything except letters and numbers are turned into single
      Signed-off-by: default avatarVegard Nossum <vegard.nossum@gmail.com>
  17. 09 Jul, 2008 3 commits
  18. 17 Jun, 2008 1 commit
  19. 04 Feb, 2008 1 commit
  20. 30 Jan, 2008 3 commits