Commits

Show all
Author Commit Message Labels Comments Date
Maciej Fijalkowski
pointer comparisons
Branches
lightweight-finalizers
Maciej Fijalkowski
precise name of the llop
Branches
lightweight-finalizers
Maciej Fijalkowski
raw_free is also fine
Branches
lightweight-finalizers
Maciej Fijalkowski
track_allocation is just fine
Branches
lightweight-finalizers
Maciej Fijalkowski
Various operations and a test
Branches
lightweight-finalizers
Maciej Fijalkowski
all casts are ok
Branches
lightweight-finalizers
Maciej Fijalkowski
pass the extra flag in the JIT as well
Branches
lightweight-finalizers
Maciej Fijalkowski
fix translation
Branches
lightweight-finalizers
Maciej Fijalkowski
add this optimization to minimark. unclear if more tests are needed
Branches
lightweight-finalizers
Maciej Fijalkowski
fix test_transformed_gc
Branches
lightweight-finalizers
Maciej Fijalkowski
clean up previous attempts, make this one work
Branches
lightweight-finalizers
Maciej Fijalkowski
finish semispace
Branches
lightweight-finalizers
Maciej Fijalkowski
call RSocket.__del__ directly instead of close(). Will not raise
Branches
lightweight-finalizers
Maciej Fijalkowski
chained objects unallowed
Branches
lightweight-finalizers
Maciej Fijalkowski
remove most of the previous approach. Now attach T_HAS_LIGHT_FINALIZER
Branches
lightweight-finalizers
Maciej Fijalkowski
kill some unused imports, thanks pyflakes
Branches
lightweight-finalizers
Maciej Fijalkowski
change approach - remove the owns_raw_memory_decorator, we'll try to detect
Branches
lightweight-finalizers
Maciej Fijalkowski
don't call close on socket, instead almost copy the logic. The idea is that
Branches
lightweight-finalizers
Maciej Fijalkowski
add FinalizerAnalyzer that checks whether an RPython del is lightweight
Branches
lightweight-finalizers
Maciej Fijalkowski
leave a comment how this is buggy
Branches
lightweight-finalizers
Maciej Fijalkowski
remove one set of hacks to another in order for it to work after translation
Branches
lightweight-finalizers
Maciej Fijalkowski
Write a test that should remind that when we implement raw malloc in jit,
Branches
lightweight-finalizers
Maciej Fijalkowski
a test and a fix
Branches
lightweight-finalizers
Maciej Fijalkowski
support the case of also having a del
Branches
lightweight-finalizers
Maciej Fijalkowski
use the interface for tests
Branches
lightweight-finalizers
Maciej Fijalkowski
fix translation. I don't think this case can happen in practice though, no
Branches
lightweight-finalizers
Maciej Fijalkowski
make test_translated_gc happy
Branches
lightweight-finalizers
Maciej Fijalkowski
implement in minimark
Branches
lightweight-finalizers
Maciej Fijalkowski
improve the test and fix it
Branches
lightweight-finalizers
Maciej Fijalkowski
Make it work with generation & hybrid. Improve the test
Branches
lightweight-finalizers
  1. Prev
  2. Next