Commits

Show all
Author Commit Message Labels Comments Date
Maciej Fijalkowski
An attempt to fix a race condition between reloading the frame and reacquiring
Maciej Fijalkowski
(agaynor, fijal) write a passing test
Armin Rigo
Bah? Another version of gcc 4.8 writes "rep; ret" instead of "rep ret".
Armin Rigo
in-progress
Branches
stm-gc-2
Alex Gaynor
really fix tests
Branches
virtualref-virtualizable
Alex Gaynor
fix test
Branches
virtualref-virtualizable
Alex Gaynor
merged default in
Branches
virtualref-virtualizable
David Schneider
kill tabs
Alex Gaynor
fix types
Branches
virtualref-virtualizable
Alex Gaynor
merged default in
Branches
virtualref-virtualizable
Alex Gaynor
if we put a guard here, we need liveness data
Branches
virtualref-virtualizable
Brian Kearns
close these sockets so test passes on pypy
Alex Gaynor
fix guard_str_value
Branches
virtualref-virtualizable
Alex Gaynor
Seems to be working implementation. Let's try tests.
Branches
virtualref-virtualizable
Alex Gaynor
Insert a place for our special logic
Branches
virtualref-virtualizable
Brian Kearns
null merge default past the backout of this branch
Branches
virtual-raw-mallocs
Brian Kearns
merge default up to 295a4447f6ba
Branches
virtual-raw-mallocs
Brian Kearns
finish backing out virtual-raw-mallocs
Alex Gaynor
a failing test
Branches
virtualref-virtualizable
Maciej Fijalkowski
Backout the merge of virtual-raw-mallocs, it has more than one issue.
Alex Gaynor
A branch in which to try to make:
Branches
virtualref-virtualizable
David Schneider
fix argument and return value handling in call_release_gil and call_reacquire_gil. Fixes test_zrpy_releasegil.py failures
David Schneider
move GIL related shared code to llsupport
David Schneider
preallocate list to hold the locations of arguments to a call
David Schneider
merge heads
David Schneider
reenable the debugging code injection, fixes test_ztranslation_jit_stats.py
Armin Rigo
Fixes
Branches
stm-gc-2
Armin Rigo
Fix for translator.backendopt.test.test_all:test_secondary_backendopt.
Armin Rigo
Restore the original support: when there is an inevitable transaction,
Branches
stm-gc-2
David Schneider
fix for _write_barrier_fastpath, missing a check to select the correct slowpath helper
  1. Prev
  2. Next