1. Pypy
  2. Untitled project
  3. pypy

Commits

Show all
Author Commit Message Date Builds
Ronan Lamy
Hack the functionptr when passing an llexternal as llimpl in register_external()
Branches
llimpl
Manuel Jacob
Kill running_on_llinterp. Instead we attach the fakeimpl to the function object and fish it in llinterp. (grafted from 39eb895a3a295c0a20a0fdd467e33dec0eb2fcfb)
Branches
llimpl
Manuel Jacob
Replace test using running_on_llinterp directly by another test that uses running_on_llinterp indirectly through register_external. (grafted from cd8bd484b255)
Branches
llimpl
Armin Rigo
borrowed results (copying from cpyext-gc-support)
Armin Rigo
Pass a few more tests
fijal
merge
fijal
start working on windows
fijal
somehow it ate my bracket
Richard Plangger
refactoring issue
Branches
s390x-backend
Richard Plangger
ffi call fixed in deprecated api that is still used (fix before that was not sufficient), fixed legacy tests test_libffi
Branches
s390x-backend
Richard Plangger
call int should provide rffi.INT instead of rffi.SIGNED, this works on little endian, but not big
Branches
s390x-backend
Ronan Lamy
forgot to add rpython/rtyper/debug.py
Ronan Lamy
Use regular register_external() call for attach_gdb()
Ronan Lamy
Create rpython.rtyper.debug and move ll_assert() and fatalerror() there. This reduces the number of imports from rpython/rtyper/ to rpython.rlib and avoids a cyclical import dependency.
Richard Plangger
fixed a bug where pool did not allocate space for constant parameter (gc_load_indexed, and store)
Branches
s390x-backend
Manuel Jacob
Skip vmprof jit tests on non-x86-64 CPUs.
Richard Plangger
test checks that trace is formed for a little endian read, forcing to read little endian on s390x
Branches
s390x-backend
Richard Plangger
must branch if not equal (error introduced in the last 2 commits)
Branches
s390x-backend
Richard Plangger
backchain was not correct (now it is, tested it), flush_cc must write 1, it must not be something != 1
Branches
s390x-backend
Richard Plangger
allocating the back chain correctly, one problem when hitting a memory error, stack size would have been incorrect!
Branches
s390x-backend
fijal
merge
fijal
remove the conftest that skipped all tests
fijal
run actually tests and make them pass
Carl Friedrich Bolz
just support exactly the same colors as graphviz
Carl Friedrich Bolz
the ll_assert has the condition wrong :-(
fijal
update whatsnew
fijal
merge vmprof-newstack
fijal
closed to be merged branch
Branches
vmprof-newstack
fijal
bump the version one level
Branches
vmprof-newstack
Ronan Lamy
Use the correct exceptions in rpython/annotator/builtin.py
  1. Prev
  2. Next