Carl Friedrich Bolz-Tereick pushed to branch branch/default at PyPy / pypy
Commits: a55f4a8e by Carl Friedrich Bolz-Tereick at 2021-11-25T11:44:06+01:00 somewhat laboriously reproduce the problem of #3598 as a jit test --HG-- branch : shortpreamble-no-mutate - - - - - 61abc439 by Carl Friedrich Bolz-Tereick at 2021-11-25T16:06:51+01:00 two variants of the test: one with one without the erroneous bridge going to the main loop --HG-- branch : shortpreamble-no-mutate - - - - - dbcaa8fd by Carl Friedrich Bolz-Tereick at 2021-11-26T13:52:07+01:00 at least prevent mutation of the earlier trace when inlining the short preamble --HG-- branch : shortpreamble-no-mutate - - - - - e8fa427b by Carl Friedrich Bolz-Tereick at 2021-11-26T16:11:09+01:00 nonsense --HG-- branch : shortpreamble-no-mutate - - - - - 5c08dfef by Carl Friedrich Bolz-Tereick at 2021-12-01T12:25:43+01:00 merge shortpreamble-no-mutate: stop doing guard strengthening with guards that come from inlining the short preamble. doing that can lead to endless bridges. fixes #3598 - - - - - 5 changed files: - rpython/jit/metainterp/optimizeopt/optimizer.py - rpython/jit/metainterp/optimizeopt/rewrite.py - rpython/jit/metainterp/optimizeopt/test/test_optimizebridge.py - rpython/jit/metainterp/optimizeopt/unroll.py - rpython/jit/metainterp/test/test_loop.py View it on Heptapod: https://foss.heptapod.net/pypy/pypy/-/compare/520ddaecc0a729ee9f1a994a08bc1e0ada87be13...5c08dfefb9520026d7b6fefaf9a2ab0731307e75 -- View it on Heptapod: https://foss.heptapod.net/pypy/pypy/-/compare/520ddaecc0a729ee9f1a994a08bc1e0ada87be13...5c08dfefb9520026d7b6fefaf9a2ab0731307e75 You're receiving this email because of your account on foss.heptapod.net.
_______________________________________________ pypy-commit mailing list pypy-commit@python.org https://mail.python.org/mailman/listinfo/pypy-commit
