- 06 Jul, 2015 2 commits
-
-
Kevin Modzelewski authored
Add a new JIT tier which is tightly coupled to the AST Interpreter
-
Kevin Modzelewski authored
For issue #254, add lt, le, gt, ge to old-style class
-
- 05 Jul, 2015 1 commit
-
-
Kevin Modzelewski authored
assembler: Use 32bit moves if the immediate fits inside 32bit
-
- 02 Jul, 2015 7 commits
-
-
Marius Wachtler authored
-
Marius Wachtler authored
makes the encoding 4-5byte shorter and works because the upper 32bit will get auto cleared
-
Marius Wachtler authored
This JIT is tightly coupled to the ASTInterpreter, at every CFGBlock* entry/exit on can switch between interpreting and directly executing the generated code without having to do any translations. Generating the code is pretty fast compared to the LLVM tier but the generated code is not as fast as code generated by the higher LLVM tiers. But because the JITed can use runtime ICs, avoids a lot of interpretation overhead and stores CFGBlock locals sysbols inside register/stack slots, it's much faster than the interpreter.
-
Kevin Modzelewski authored
Use less large constants
-
Boxiang Sun authored
-
Boxiang Sun authored
-
Boxiang Sun authored
-
- 01 Jul, 2015 8 commits
-
-
Kevin Modzelewski authored
Preperations for the new JIT tier
-
Kevin Modzelewski authored
Conflicts: src/core/types.h
-
Marius Wachtler authored
IC sizes are guessed...
-
Marius Wachtler authored
It deleted the passed ICSlotRewrite* and there was no way for a caller to know this without looking at the source. Make the ownership explicit by using a std::unique_ptr
-
Marius Wachtler authored
and make code ready for the new JIT tier.
-
Marius Wachtler authored
-
Kevin Modzelewski authored
Conflicts: src/runtime/iterobject.cpp
-
Kevin Modzelewski authored
add -a flag which outputs assembly of ICs
-
- 30 Jun, 2015 13 commits
-
-
Chris Toshok authored
-
Chris Toshok authored
build sorted vectors for large/huge objects at gc time, and use binarySearch to power ::allocationFrom
-
Chris Toshok authored
-
Kevin Modzelewski authored
fastpath investigations
-
Kevin Modzelewski authored
I think this only gets hit through explicit calls to tuple(), but at one point it was crashing on this codepath which is how I noticed it.
-
Kevin Modzelewski authored
-
Kevin Modzelewski authored
-
Kevin Modzelewski authored
Our tuple format is now the same as CPython's, so we can just enable the fast macros again. Little bit of trickiness since they declare their storage array to have a size of 1; I'm not sure how important that is, but let's play it safe and match it.
-
Kevin Modzelewski authored
We already have this for the rest of our code, but it wasn't enabled for C files.
-
Kevin Modzelewski authored
- conservative references into the large/huge heaps are especially expensive - it's nice to know what kind of object we are scanning
-
Kevin Modzelewski authored
-
Kevin Modzelewski authored
So we can compare against their latest instead of whatever was in apt.
-
Kevin Modzelewski authored
Notably, to the rewriter. Turns out that a decent amount of the time that used to get allocated to slowpaths is actually time spent in the rewriter, doing successful rewrites.
-
- 29 Jun, 2015 2 commits
-
-
Kevin Modzelewski authored
Take another pass over the failing cpython tests
-
Kevin Modzelewski authored
-
- 27 Jun, 2015 5 commits
-
-
Kevin Modzelewski authored
Switch BoxedMethodDescriptor to tpp_call
-
Kevin Modzelewski authored
While it's bad that we do so poorly with large heaps, that's not the point of this benchmark.
-
Kevin Modzelewski authored
- "Override" the stattimer stack while we are throwing an exception. - Mark some more things as being back in "builtins" code.
-
Kevin Modzelewski authored
Minor, but should cut down on some overhead
-
Kevin Modzelewski authored
reduce api conversions
-
- 26 Jun, 2015 2 commits
-
-
Kevin Modzelewski authored
We got away with sloppy exception clearing in some places, since raiseExcHelper would end up calling typeCall which would re-throw the capi exception that hadn't been cleared. Now that typeCall doesn't do that (which is fine), those bugs got exposed.
-
Kevin Modzelewski authored
We can't rewrite into the object-creation behavior, but we can rewrite the decision to just use CPython's c slots.
-