[commit: ghc] master: NUMA support (9e5ea67)

[git at git.haskell.org]

Repository : ssh://git@git.haskell.org/ghc

On branch  : master
Link       : http://ghc.haskell.org/trac/ghc/changeset/9e5ea67e268be2659cd30ebaed7044d298198ab0/ghc

>---------------------------------------------------------------

commit 9e5ea67e268be2659cd30ebaed7044d298198ab0
Author: Simon Marlow <marlowsd at gmail.com>
Date:   Sat Apr 23 21:14:49 2016 +0100

    NUMA support
    
    Summary:
    The aim here is to reduce the number of remote memory accesses on
    systems with a NUMA memory architecture, typically multi-socket servers.
    
    Linux provides a NUMA API for doing two things:
    * Allocating memory local to a particular node
    * Binding a thread to a particular node
    
    When given the +RTS --numa flag, the runtime will
    * Determine the number of NUMA nodes (N) by querying the OS
    * Assign capabilities to nodes, so cap C is on node C%N
    * Bind worker threads on a capability to the correct node
    * Keep a separate free lists in the block layer for each node
    * Allocate the nursery for a capability from node-local memory
    * Allocate blocks in the GC from node-local memory
    
    For example, using nofib/parallel/queens on a 24-core 2-socket machine:
    
    ```
    $ ./Main 15 +RTS -N24 -s -A64m
      Total   time  173.960s  (  7.467s elapsed)
    
    $ ./Main 15 +RTS -N24 -s -A64m --numa
      Total   time  150.836s  (  6.423s elapsed)
    ```
    
    The biggest win here is expected to be allocating from node-local
    memory, so that means programs using a large -A value (as here).
    
    According to perf, on this program the number of remote memory accesses
    were reduced by more than 50% by using `--numa`.
    
    Test Plan:
    * validate
    * There's a new flag --debug-numa=<n> that pretends to do NUMA without
      actually making the OS calls, which is useful for testing the code
      on non-NUMA systems.
    * TODO: I need to add some unit tests
    
    Reviewers: erikd, austin, rwbarton, ezyang, bgamari, hvr, niteria
    
    Subscribers: thomie
    
    Differential Revision: https://phabricator.haskell.org/D2199


>---------------------------------------------------------------

9e5ea67e268be2659cd30ebaed7044d298198ab0
 configure.ac                                  |   7 +
 docs/users_guide/runtime_control.rst          |  50 ++++
 includes/Cmm.h                                |   1 -
 includes/Rts.h                                |   1 -
 includes/RtsAPI.h                             |   6 +-
 includes/rts/Constants.h                      |   6 +
 includes/rts/Flags.h                          |   8 +-
 includes/rts/OSThreads.h                      |   4 +-
 includes/rts/Threads.h                        |   4 +-
 includes/rts/storage/Block.h                  |  20 +-
 includes/rts/storage/MBlock.h                 |   2 +
 rts/Capability.c                              |  38 +--
 rts/Capability.h                              |  14 +-
 rts/HeapStackCheck.cmm                        |   1 +
 rts/Inlines.c                                 |   1 +
 rts/Messages.h                                |   1 +
 rts/PrimOps.cmm                               |   1 +
 rts/ProfHeap.c                                |   1 +
 rts/RtsFlags.c                                | 118 +++++++--
 {includes/rts/storage => rts}/SMPClosureOps.h |   4 +
 rts/STM.c                                     |   1 +
 rts/Schedule.c                                |   5 +-
 rts/Task.c                                    |  26 +-
 rts/Task.h                                    |   6 +
 rts/eventlog/EventLog.c                       |   2 +-
 rts/package.conf.in                           |   4 +-
 rts/posix/OSMem.c                             |  64 +++++
 rts/posix/OSThreads.c                         |  30 ++-
 rts/sm/BlockAlloc.c                           | 350 ++++++++++++++++----------
 rts/sm/BlockAlloc.h                           |   1 +
 rts/sm/GC.c                                   |   8 +-
 rts/sm/GCUtils.c                              |  15 +-
 rts/sm/GCUtils.h                              |  14 +-
 rts/sm/MBlock.c                               |  19 +-
 rts/sm/MarkStack.h                            |   1 +
 rts/sm/OSMem.h                                |   4 +
 rts/sm/Storage.c                              | 190 ++++++++------
 rts/win32/OSMem.c                             |  22 ++
 rts/win32/OSThreads.c                         |   3 +
 testsuite/config/ghc                          |   6 +-
 testsuite/tests/codeGen/should_run/all.T      |   5 +-
 testsuite/tests/concurrent/prog001/all.T      |   0
 testsuite/tests/concurrent/should_run/all.T   |   5 +-
 43 files changed, 788 insertions(+), 281 deletions(-)

Diff suppressed because of size. To see it, use:

    git diff-tree --root --patch-with-stat --no-color --find-copies-harder --ignore-space-at-eol --cc 9e5ea67e268be2659cd30ebaed7044d298198ab0