Commits

Show all
Author Commit Message Labels Comments Date
Remi Meier
fix forksupport for small objs (copy&paste mistake)
Branches
gc-small-uniform
Remi Meier
fix
Branches
gc-small-uniform
Remi Meier
merge default
Branches
gc-small-uniform
Remi Meier
(arigo,xoraxax,remi) remove (probably) bogus assert
Armin Rigo
Add some dprints
Armin Rigo
(remi, arigo) Maybe fix a bug, by making it explicit in each stm_thread_local if it is running or not (associated_segment_num >= 1 or == -1).
Armin Rigo
Change some dprints
Armin Rigo
Need to check more carefully that the transaction is running
Armin Rigo
improve comment
Armin Rigo
(remi, xoraxax, me around) Bug fix
Armin Rigo
Fix
Armin Rigo
Add an assert that fails
Remi Meier
test cleanup
Remi Meier
remove some xfail from test_demo.py
Remi Meier
phew, ensure that nobody applies our old backup copies after we just committed
Remi Meier
possibly fix a memory leak
Remi Meier
reduce diff between c7 and c8 a bit
Remi Meier
fix to make small old objs survive major gc
Armin Rigo
Fix: can't always read entry->index from the major GC, because the current segment might be bogus if the hashtable is still young.
Armin Rigo
Add stm_stop_all_other_threads() and stm_resume_all_other_threads()
Remi Meier
asserts and comments from trying to find a bug exposed by demo_random (even w/o forking)
Remi Meier
add forksupport (still buggy)
Remi Meier
don't mmap readmarkers as SHARED (seems to fix forking)
Remi Meier
fix for on-major-gc validation not doing anything when a transaction thought it was still uncommitted and inevitable
Remi Meier
fix demo_random.c
Remi Meier
Backed out changeset: 7540c1f155d0
Remi Meier
Backed out changeset: 60f7ccae893c
Armin Rigo
another quite unsure attempt
Armin Rigo
probably? quite unsure about this (see demo/demo_random)
Armin Rigo
Backed out changeset 756a54dd920b: does not work
  1. Prev
  2. Next