Source

pypy / pypy / rlib / jit.py

The branch 'autoreds' does not exist.
Author Commit Message Labels Comments Date
Antonio Cuni
crash very early if we try to use JitDriver hooks with reds='auto', and detect it anyway in warmspot.py, giving a better assertion message
Branches
autoreds
Antonio Cuni
make sure to set .autoreds and .numreds even when the reds are given as class attributes
Branches
autoreds
Antonio Cuni
the test to show the whole point of the excercise, i.e. the possibility to automatically create jit portals by iterating over a specially-decorated RPython iterator
Branches
autoreds
Antonio Cuni
crash even when unstralsated if you call can_enter_jit on an autoreds driver
Branches
autoreds
Antonio Cuni
make sure that we cannot use autoreds with confirm_enter_jit
Branches
autoreds
Antonio Cuni
it took 2 days of experiments, playing around and understand JIT details I never knew about :), but I finally managed to find a simple way to have red variables automatically detected by the JIT; still in-progress, but the basic idea should work
Branches
autoreds
Maciej Fijalkowski
fix someobject in jit
Branches
kill-someobject
Maciej Fijalkowski
revert the jit.py modification
Branches
kill-someobject
Maciej Fijalkowski
(arigo, fijal) Rename _freeze_ returning False to _cleanup_, now we can have hasattr(x, '_freeze_') without strange hacks to check for frozen PBCs.
Branches
kill-someobject
Alex Gaynor
Write some of the JIT parameter handling code more sanely (also of course makes it work with the new lack of SomeObject)
Branches
kill-someobject
Alex Gaynor
Make jit/codewriter tests up to date with the new APIs
Branches
kill-someobject
Armin Rigo
Maybe temporary: set max_unroll_loops to 0, which should completely prevent tracing more than 1 iteration. Maybe a better value would be 1 (retraces once, so trace at most 2 iterations), but it's hard to say a priori.
David Schneider
merge default
Branches
arm-backend-2
Armin Rigo
Work in progress: remove optimizeopt/fficall, and (plan to) replace it with just a test in pyjitpl.
Branches
ffi-backend
Maciej Fijalkowski
actually fix tests
Maciej Fijalkowski
fix argument naming
Maciej Fijalkowski
preserve the names for jit_unroll_iff, otherwise we end up with unreadable unwrap_spec
Maciej Fijalkowski
shift dict_unrolling_heuristics to dictmultiobject.py
Branches
virtual-arguments
Maciej Fijalkowski
typo
Branches
virtual-arguments
Maciej Fijalkowski
some more careful unrolling, start to write more test_pypy_c tests
Branches
virtual-arguments
Maciej Fijalkowski
merge virtual-args here, so I don't have to worry about conflicts
Branches
speedup-unpackiterable
Maciej Fijalkowski
a bit of refactoring but now one can get to details about counters from the interpreter if he really feels like it
Branches
even-more-jit-hooks
David Schneider
merge default
Branches
arm-backend-2
David Schneider
merge default
Branches
arm-backend-2
Armin Rigo
An even stronger wording of "don't use this option!", as discussed on irc with hakanardo :-)
Carl Friedrich Bolz
merge default
Branches
virtual-arguments
Carl Friedrich Bolz
give an error when the predicate of look_inside_iff has a different number of arguments than the decorated function.
Branches
virtual-arguments
Maciej Fijalkowski
hopefully a better specialization
Maciej Fijalkowski
(fijal, arigo) heuristic for unrolling loops, unroll a bit more
David Schneider
merge default
Branches
arm-backend-2
  1. Prev
  2. 1
  3. 2
  4. 3
  5. 4
  6. 5
  7. 6
  8. 7
  9. 8
  10. 9
  11. Next