lp:~mwhudson/pypy/imported-gc-jit-hack

Branch merges

Propose for merging

No branches dependent on this one.

Related source package recipes

No recipes using this branch. (?)

Create packaging recipe

Related snap packages

Related bugs

Related blueprints

23277. By arigo on 2009-11-05

This is the best result so far. At least in this case I understand the
results: they seem to clearly show that a lot of the objects allocated
during jit-tracing actually survive. The results are not good enough --
they say that jit-tracing takes only half the time, but the truth is
that it's always soon followed by a lengthy minor collection, so that
the total time (as well as the total GC time) end up not really
improved. Closing this branch for now.

23276. By arigo on 2009-11-05

The current solution I think best. It still gives mixed
results.

23275. By arigo on 2009-11-05

Revert r68981.

23274. By arigo on 2009-11-04

Force a minor collection at the start of jit-tracing.
Experimental.

23273. By arigo on 2009-11-04

"import sys" missing.

23272. By arigo on 2009-11-04

Use resize_nursery() in the JIT.

23271. By arigo on 2009-11-04

Comment out the lines that bound the nursery size to half the initial space
size, with explanation.

23270. By arigo on 2009-11-04

In record_red_zone() we know that the nursery is going
to be allocated very soon, and it was already allocated
previously. So it should really be counted as part of
the allocated objects. The result is that we resize
the semispaces a bit more eagerly.

23269. By arigo on 2009-11-04

Interface resize_nursery() in framework.py, and test it in test_newgc.

23268. By arigo on 2009-11-04

Implement resizing the nursery at runtime.