- 08 Oct, 2013 2 commits
-
-
Stefan Marr authored
Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Stefan Marr authored
Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
- 23 Sep, 2013 6 commits
-
-
Tobias Pape authored
Before: ./som -cp Smalltalk TestSuite/TestHarness.som 0,01s user 0,01s system 4% cpu 0,538 total After: ./som -cp Smalltalk TestSuite/TestHarness.som 0,01s user 0,01s system 56% cpu 0,027 total
-
Tobias Pape authored
-
Stefan Marr authored
- Double>>#round - Integer class>>#fromString: - System>>#fullGC: Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Stefan Marr authored
- need to be proper SOM objects Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Stefan Marr authored
The other elements such as number of locals, maximum number of stack elements, and number of method fields are not available on all SOM VMs. These information should be made available by a VM-specific mirror object. Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Stefan Marr authored
Looks like this code is not actually used, at least the reference was not a problem so far. Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
- 22 Sep, 2013 20 commits
-
-
Tobias Pape authored
-
Tobias Pape authored
-
Tobias Pape authored
Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Tobias Pape authored
-
Tobias Pape authored
-
Tobias Pape authored
first, _start_time in the primitve module would be frozen on translate time and not set when the binary starts. second, python random is not rpython, we use rrandom
-
Stefan Marr authored
See 81bb9103 Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Tobias Pape authored
-
Tobias Pape authored
-
Tobias Pape authored
Signed-off-by:
Stefan Marr <git@stefan-marr.de> Conflicts: src/som/vm/universe.py
-
Stefan Marr authored
Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Tobias Pape authored
for that, provide own raw_input. also: crtl-d works now.
-
Stefan Marr authored
Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Tobias Pape authored
-
Stefan Marr authored
types.MethodType is not used anymore to bind methods Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Stefan Marr authored
Let's try to have all theses things in one place, to reduce the differences between PySOM a RPySOM to a minimum. Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Stefan Marr authored
Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Stefan Marr authored
Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Stefan Marr authored
Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Stefan Marr authored
Bug introduced in b8d3cc26 , which was replacing os.path.split with RPython compatible equivalent Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
- 21 Sep, 2013 12 commits
-
-
Stefan Marr authored
- at the moment this is hard coded, there might be a better way Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Stefan Marr authored
Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Stefan Marr authored
Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Stefan Marr authored
- converted tuples to lists Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Stefan Marr authored
Changing to a tuple didn't did not solve the issue that RPython could not find a unique name for those function pointers. Now the traditional approach. This also allows us to get rid of the exception for the HALT bytecode. Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Stefan Marr authored
RPython complained about not being able to determine a unique name for the function pointers, using a tuple doesn't cause this warning anymore. Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Stefan Marr authored
- they used metaprogramming before, but that does not seem to work. - RPython gives me 'operation cannot succeed' and conflicting types for variables if I try to change the code slightly -> will just use simple constants instead, less maintainable but it works Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Stefan Marr authored
Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Tobias Pape authored
Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Stefan Marr authored
Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Stefan Marr authored
Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-
Stefan Marr authored
- think those are integers... Signed-off-by:
Stefan Marr <git@stefan-marr.de>
-