musl - stdatomic library implementation

Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <1437983501.757.4.camel@inria.fr>
Date: Mon, 27 Jul 2015 09:51:41 +0200
From: Jens Gustedt <jens.gustedt@...ia.fr>
To: musl@...ts.openwall.com
Subject: stdatomic library implementation

Hello,
in two follow up mails I will send you the first version of all of
this as a patch and a documentation of it as .org and .pdf files.

You might be most interested in the implementation issues, first, so I
copy that part of the documentation below.

Thanks for considering

Jens

* The implementation


** Requirements

*** Compilers

  You should be able to compile this implementation with any version
  of modern gcc and clang. (Versions are hard to tell, gcc should work
  for 4.1) The quality of the resulting binary will depend on the
  implementation of atomic support by the compiler.

  There are three different implementations, for modern clang and gcc,
  and one for those compilers that only support the =__sync_=
  built-ins. They are only tested with clang and gcc, but might work
  with other compilers that implement one of the sets of built-ins and
  is otherwise compatible to some gcc extensions:

  - compound expressions with =({ })=
  - =__attribute__= with =__alias__= and =__unused__=
  - =__builtin_choose_expr= for the =__sync= version as a precursor of
    C11's =_Generic=

  There are some heuristics in place to decide at compile time which
  case applies, namely =__clang__= to detect clang, =__ATOMIC_...=
  macros to detect the C11 versions of the built-ins.

*** OS or C library support

    The library may work with different lock constructs. The musl
    integration uses =LOCK= and =UNLOCK= internal macros.

** Caveats

*** Symbol renaming

  There is one important difficulty when compiling this. The original
  =__atomic= library interface was developed with C++ in mind and not
  C. Therefore it freely uses function overloading for the built-ins
  versus the library interface. Since we also use the library
  functions as fallbacks in the implementation of some of the =_X=
  variants this naming scheme is not supportable with a C compiler.

  We get away with it by using internal names, prefixed with =__impl_=
  for all functions. Then we rename symbols to the intended ones using
  =objcopy=.

  - The current integration into musl does this with a *script* that
    you have to run *manually* after compilation.
  - You then have to launch make a *second time* to do the final link.

 This technique is certainly not ideal and subject to improvement.

*** Support of 16 byte atomic instructions

    The main difference for modern processors that is relevant here is
    if it supports 16 byte atomic instructions or not. There is no
    difficulty to detect this at compile time, but if the library is
    used with code that is compiled with a different compiler or just
    different compiler options, incompatible binary code may be
    produced.

    My plan is to freeze that feature at compile time of the library
    and reflect the capacity in the =<stdatomic.h>= that is
    provided. This then may result in code that is a bit less
    optimized than it could, but that is compatible.

    - If the library is *not* compiled with direct 16 byte support the
      application may not use it, and thus use a memory implementation
      for such operations.

    - If the library *is* compiled with direct 16 byte support but the
      application compiler doesn't support it, the user code should
      fallback to library calls, but which in turn use the atomic
      instructions. So such a variant would have a call overhead and
      would not be able to inline the atomics in the user binary.

    All of this is not yet, done, though. Be careful when using this
    preliminary version.


** Leftovers

   There are some leftovers that will hopefully disappear. In
   particular there are several hash functions and a instrumentation
   infrastructure for the hashes. I didn't have enough test cases yet
   to see what would be best, here.


-- 
:: INRIA Nancy Grand Est ::: Camus ::::::: ICube/ICPS :::
:: ::::::::::::::: office Strasbourg : +33 368854536   ::
:: :::::::::::::::::::::: gsm France : +33 651400183   ::
:: ::::::::::::::: gsm international : +49 15737185122 ::
:: http://icube-icps.unistra.fr/index.php/Jens_Gustedt ::




Download attachment "signature.asc" of type "application/pgp-signature" (182 bytes)
Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.