Commits

Show all
Author Commit Message Labels Comments Date
timfel
merge default
timfel
Drop to gdb on unix when pdb.set_trace is encountered
Armin Rigo
Test, and fixes for 32-bit
Branches
int-float-list-strategy
Richard Plangger
product assembler now uses SHUFPD instead of SHUFPS and moves two floats instead of one to the accumulator resolving test issues with the recent changes
Branches
vecopt
Armin Rigo
setslice() support and clean-ups
Branches
int-float-list-strategy
Armin Rigo
lst = [1.2]; lst += [0] lst = [1.2]; lst += [0, 3.4]
Branches
int-float-list-strategy
Armin Rigo
lst = [0]; lst += [1.2] lst = [0]; lst += [1.2, 3]
Branches
int-float-list-strategy
Armin Rigo
hg merge default
Branches
int-float-list-strategy
Armin Rigo
A simpler fix for "from x import *", with more tests showing the problem of the previous one.
Armin Rigo
hg backout e83ef97376fe: wrong fix
Richard Plangger
preventing signext from int16,int8 to greater or vice versa, the first direction was ignored up to now
Branches
vecopt
Manuel Jacob
hg merge default
Branches
py3k
Manuel Jacob
Fix syntax error.
Branches
py3k
Armin Rigo
Add comment
Branches
int-float-list-strategy
Armin Rigo
These two cases are now implemented
Branches
int-float-list-strategy
Armin Rigo
Support the (likely) most common mixed cases: [int-or-float].extend([int]) [int-or-float].extend([float])
Branches
int-float-list-strategy
Armin Rigo
int-or-float sort()
Branches
int-float-list-strategy
Armin Rigo
Some skipped tests (for later maybe)
Branches
int-float-list-strategy
Armin Rigo
More tests, and float -> int-or-float
Branches
int-float-list-strategy
Armin Rigo
int -> int-or-float
Branches
int-float-list-strategy
Armin Rigo
Start writing the strategy
Branches
int-float-list-strategy
Armin Rigo
more tests
Branches
int-float-list-strategy
Armin Rigo
in-progress
Branches
int-float-list-strategy
Armin Rigo
A branch to try out the list-of-ints-or-floats strategy
Branches
int-float-list-strategy
Armin Rigo
Minor optimization
Richard Plangger
restricted input size for vectorization to 75 operations max (dont see any chance for it to succeed with more than 20-30 operations) reduce resetting rval at the end of the loop
Branches
vecopt
Richard Plangger
updated docs
Branches
vecopt
Armin Rigo
Copy this thread initialization logic from module/_cffi_backend/ccallback.py
Armin Rigo
Don't use 'rffi.cast(rffi.UINT, pointer)'! This casts to a 32-bit integer, not a full machine-word integer.
Armin Rigo
Return 1, not '(int)&W_IntObject1'
  1. Prev
  2. Next