6 ms·
PyPy 2.0 beta 1 released
- andybak 14y agoARM, eh? What's the memory usage like? That used to be a problem for anyone considering PyPy on mobile chipsets.
- firlefans 14y agoAlso wondering about this, would be v. cool to play with it on iOS.
- fijal 14y agoiOS does not allow you to run any sort of non-apple just-in-time compiler (there is no public API to get executable mmaped pages to be precise), so a no go, you cannot.
- gsnedders 14y agoIt's more than that: nothing outside of the built-in applications is allowed to allocated executable memory at all. You can't even use JSC's JIT (i.e., that of Apple's JS engine) in third-party applications.
- masklinn 14y ago> You can't even use JSC's JIT (i.e., that of Apple's JS engine) in third-party applications. Although this very specific one may get fixed through remote views, which apple is currently toying with (a webview would live in its own "webview process" and thus embedded webview could finally get native JIT-ed speeds)
- kkuduk 14y agoRegarding non-synthetic benchmarks, quite recently I've tested pypy as a "backend" to Gentoo's package manager (portage). System was quite not up-to-date, lots of packages and dependencies, this is how much time did it take portage to figure out what to update: python2.7: 323.13s user 6.50s system 99% cpu 5:30.59 total python3.2: 271.40s user 6.33s system 99% cpu 4:38.64 total pypy1.9: 168.28s user 5.95s system 99% cpu 2:55.16 total
- fijal 14y agoIf you feel like providing such a benchmark for speed.pypy.org, we would be very happy to accept it. I'm not sure how hard would that be though.
- dbaupp 14y ago(Your benchmark site seems to be broken: http://isitup.org/speed.pypy.org http://isitup.org/speed.pypy.org )
- fijal 14y agoit seems that it's just overloaded. EDIT: back for now
- kkuduk 14y agoWell, probably I would have to bundle whole portage tree, so it would be ~1 GB big, but I'll give it a shot this weekend.
- alberth 14y agoHow does PyPy 2.0 compares to LuaJIT 2.0 [1] with regards to memory usage and performance? I ask since they have similar goals for their respective languages. [1] http://luajit.org http://luajit.org
- fijal 14y agoThere is very limited data for non-artificial and non-numeric benchmarks for Lua and Python, so treat it all with a grain of salt. PyPy mostly cares about large scale programs (thanks to a large quantity of Python libraries it's not that hard), while LuaJIT usually cares for smaller stuff. Last time I tried to run richards (which is the largest example I found for both Lua and Python), PyPy was able to eventually outperform LuaJIT slightly, given enough iterations, however LuaJIT warmup time was very very impressive and it was much faster at the beginning. From a very unscientific perspective I would expect LuaJIT to mostly outperform PyPy, especially on stuff where it's feasible to compare, like by implementing small numeric kernels in both Lua and Python, but I haven't measured a whole lot in this area. LuaJIT definitely outperforms PyPy on the computer language shootout benchmarks. I would also expect the field to be more leveled if you implement say a webserver in both, caring about performance, but this is really on the guessing side.
- haberman 14y ago> PyPy was able to eventually outperform LuaJIT slightly, given enough iterations Is this benchmark data-intensive? One known weak point of LuaJIT is its GC, which is slated for a rewrite: http://wiki.luajit.org/New-Garbage-Collector http://wiki.luajit.org/New-Garbage-Collector If "richards" allocates lots of objects that accumulate and need to be GC'd, this could explain why PyPy wins eventually.
- igouy 14y ago"[Richards] simulates the task dispatcher in the kernel of an operating system." http://labs.oracle.com/people/mario/java_benchmarking/richards/richards.html http://labs.oracle.com/people/mario/java_benchmarking/richar...
- realsmart 14y agoThe OSX binary segfaults on a 64bit Lion. Can anybody confirm this?
- fijal 14y agoI've seen at least 3 independent people saying that. We're investigating (it works on OS X Mountain Lion though)
- continuations 14y ago> Software Transactional Memory Would pypy be able to take advantage of Haswell's hardware transactional memory?
- fijal 14y agoNot in the current incarnation, the HTM buffer is too small.
- continuations 14y agoIs this specific to pypy or is the HTM buffer too small for most languages? Are there any languages that will be able to use Haswell's HTM?
- arigo 14y agoThe HTM buffer is too small for any language for doing what I'm talking about, which is long transactions. Haswell uses its L1 Data cache as buffer --- which means at most 32KB or 64KB, and only if you're on a lucky day. (Actually it depends on cache collision and alignment issues: it can in theory overflow with only 9 memory accesses, but I'd expect the numbers to be generally more than half the full cache size.) The HTM buffer is big enough if the goal is "only" to remove the GIL. It's going to be a 10-lines patch to CPython or PyPy (yes, really!). The GIL is often made into a big issue, but in my opinion removing it isn't the final solution to all multicore troubles in Python: that's just the first step.
- lmm 14y agoThe options were covered in-depth on http://morepypy.blogspot.co.uk/2012/08/multicore-programming-in-pypy-and.html http://morepypy.blogspot.co.uk/2012/08/multicore-programming... , posted here a few months ago