guile-commits
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Guile-commits] branch master updated (c99f9ba -> ac5a05d)


From: Andy Wingo
Subject: [Guile-commits] branch master updated (c99f9ba -> ac5a05d)
Date: Wed, 11 Nov 2015 11:39:05 +0000

wingo pushed a change to branch master
in repository guile.

      from  c99f9ba   Release v2.1.1
       new  ce36fb1   Bootstrap build doesn't have to expand CPS optimizations
       new  8d79dfd   Revert "Bootstrap build doesn't have to expand CPS 
optimizations"
       new  04356da   CSE can run on first-order CPS
       new  3f345f5   Run CSE to clean up after closure conversion
       new  34f3fb7   Fix slot-allocation to make 'return' not need to 
alloc-frame
       new  7aee3c7   return-values opcode resets the frame
       new  7c9e477   Don't emit redundant reset-frame before return
       new  2f08838   Replace return primcalls with $values
       new  696339a   Always emit return-values
       new  8db54c8   rtl.test uses return-values
       new  c984432   Remove use of return in disassembler.scm
       new  9585508   Remove return opcode
       new  dd77a81   Treat tail $values as generating lazy allocations
       new  e7660a6   VM support for raw slots
       new  e3cc0ee   Reflection support for unboxed f64 slots
       new  fc87033   Stack slots can hold a double
       new  5bbc47b   Add VM ops to pack and unpack raw f64 values.
       new  6087539   Type inference distinguishes between untagged and tagged 
flonums
       new  b1ac8d6   bv-{f32,f64}-{ref,set!} operate on raw f64 values
       new  c438998   Scalar replacement for f64->scm
       new  3b4941f   Add fadd, fsub, fmul, fdiv instructions
       new  12e9e21   Add new pass to specialize "add" into "fadd" where 
possible
       new  f0594be   Fix slot representation computation for fadd, fmul, etc
       new  5b9835e   The compiler can unbox float64 loop variables
       new  80f2726   Better f64 unboxing for loop vars that might flow to 
$ktail
       new  7dc3e4b   Remove debug printout in specialize-numbers
       new  ac5a05d   Bump bytecode version

The 27 revisions listed above as "new" are entirely new to this
repository and will be described in separate emails.  The revisions
listed as "adds" were already present in the repository and have only
been added to this reference.


Summary of changes:
 bootstrap/Makefile.am                      |    1 +
 doc/ref/vm.texi                            |   11 +-
 libguile/_scm.h                            |    2 +-
 libguile/frames.c                          |   56 ++++-
 libguile/frames.h                          |    6 +-
 libguile/loader.c                          |    4 +-
 libguile/loader.h                          |    4 +-
 libguile/vm-engine.c                       |  141 +++++++++--
 libguile/vm.c                              |   77 ++++---
 module/Makefile.am                         |    1 +
 module/language/cps/compile-bytecode.scm   |   26 +--
 module/language/cps/contification.scm      |    7 +-
 module/language/cps/cse.scm                |  320 +++++++++++++-------------
 module/language/cps/effects-analysis.scm   |    9 +
 module/language/cps/optimize.scm           |    5 +-
 module/language/cps/slot-allocation.scm    |  103 +++++++--
 module/language/cps/specialize-numbers.scm |  345 ++++++++++++++++++++++++++++
 module/language/cps/type-fold.scm          |    2 +-
 module/language/cps/types.scm              |  141 ++++++++----
 module/language/cps/verify.scm             |    4 +-
 module/language/tree-il/compile-cps.scm    |   43 +++-
 module/system/repl/debug.scm               |    6 +-
 module/system/vm/assembler.scm             |   53 +++--
 module/system/vm/debug.scm                 |   13 +-
 module/system/vm/disassembler.scm          |    6 +-
 module/system/vm/frame.scm                 |   50 +++--
 test-suite/tests/rtl.test                  |   97 ++++----
 27 files changed, 1106 insertions(+), 427 deletions(-)
 create mode 100644 module/language/cps/specialize-numbers.scm



reply via email to

[Prev in Thread] Current Thread [Next in Thread]