1. CitooZz Banditozz
  2. pypy

Source

pypy / pypy / rpython / lltypesystem / opimpl.py

Author Commit Message Date Builds
Maciej Fijalkowski
hack differently - explicit getfield on vrefs (that does not necesarilly force them)
Branches
vref-copy
Maciej Fijalkowski
a branch to experiment with speeding up sys._getframe and friends.
Branches
vref-copy
Armin Rigo
Backout 161f9ca68f8e + 8a78c6bf2abb: un-merge improve-rbigint. There are failures left, e.g. pypy/rlib/test/test_rbigint on Linux32.
stian
Backed out changeset d65e8cef8bec
Maciej Fijalkowski
Backed out changeset fcdcec196a0b. This breaks tests at the very least. It should stay on a branch for now.
stian
Make _wide_digit use a cast instead of making a new value (slighty faster), make sure we support 32bit (I'm building a 32bit binary myself now, seems to work). And fix things those things arigo pointed out
Branches
improve-rbigint
stian
Introduce int128 and int cache. Also find a new karatsuba cutoff that is fine with the new settings. Some things are slower (especially creating new ints). But generally, it's a major speedup. 4 failing tests, mostly due to not being able to cast down to int. Hash is also wrong (critical). Not JIT support yet I suppose.
Branches
improve-rbigint
wlav
o) merge default into branch o) fix on my side the problems due to changes in array
Branches
reflex-support
wlav
rollback (translation fails in array)
Branches
reflex-support
Alex Gaynor
Added the operation for RPython.
Branches
float-bytes-2
Christian Tismer
Merge with default
Branches
win64-stage1
Alex Gaynor
fix a failing test
Christian Tismer
merge
Branches
win64-stage1
Christian Tismer
Merging the basic win64 changes which are without doubt into default
Christian Tismer
Merge with default
Branches
win64-stage1
Carl Friedrich Bolz
also add an explicit way to record a class
Christian Tismer
blaming myself about bad ideas. is_valid_int should never raise an exception, basta!
Branches
win64-stage1
Christian Tismer
three more typos
Branches
win64-stage1
Christian Tismer
changed all checks for (int, long) to is_valid_int. I'm pretty sure that I made mistakes, we'll see ;-)
Branches
win64-stage1
Christian Tismer
reverted merge with default
Branches
win64-stage1
Christian Tismer
Merge with default
Branches
win64-stage1
Armin Rigo
Attempt to kill 5225def4e322, which contains the broken merges 015d05dac523 and 4dcb9ca387a6.
Christian Tismer
merge
Branches
win64-stage1
Christian Tismer
'merge'
Christian Tismer
fixed test_compute_hash
Branches
win64_gborg
Christian Tismer
merge default
Branches
win64 test
Maciej Fijalkowski
(fijal, arigo reviewing) Merge lightweight-finalizers branch. This branch adds a possitibility for a finalizer to be "lightweight". This means it won't resurrect an object or call arbitrary code. Such finalizers are detected automatically and can be dealt with more efficiently (such objects can live in the nursery for example).
Maciej Fijalkowski
remove track_alloc_stop support, not needed any more
Armin Rigo
Fix the declaration of the lloperation 'cast_float_to_int' to no longer pretend it raises OverflowError, which is wrong. Fix the JIT by refusing to jtransform operationsl like 'cast_float_to_uint' and 'cast_uint_to_float' and their longlong and ulonglong equivalents. These would each require their own custom messy code in the backends. (If really needed, they are probably better written as '_ll_1_cast_x_to_y' support functions.)
Armin Rigo
Attempt to fix sys._current_frames() by returning fake frame objects.
  1. Prev
  2. 1
  3. 2
  4. 3
  5. 4
  6. Next