Mailman 3 October 2014 - pypy-dev

Py3.3 nightly builds for Windows
by Ram Rachum Oct. 16, 2014

Oct. 16, 2014

Hi guys, Is there any progress on putting nightly builds of Py3.3 for Windows here: http://buildbot.pypy.org/nightly/py3.3/ It has only Linux binaries. Thanks, Ram.

2 1

Re: [pypy-dev] [pypy-commit] pypy default: investigate mark_opaque_ptr
by Armin Rigo Oct. 14, 2014

Oct. 14, 2014

Hi Hakan, On 13 October 2014 11:52, Hakan Ardo <hakan(a)debian.org> wrote: > On Mon, Oct 13, 2014 at 11:45 AM, Armin Rigo <arigo(a)tunes.org> wrote: >> Hi Hakan, >> >> On 13 October 2014 10:11, Hakan Ardo <hakan(a)debian.org> wrote: >>> mark_opaque_ptr is used by unrolling to prevent moving getfield_gc(p1) >>> into the short preamble is p1 if opaque. This is needed since the >>> pointer might be pointing to something of a different type than what >>> it was pointing to during the tracing. In which case the execution of >>> the trace will be aborted before the original position of the >>> getfield_gc but after the short preamble is executed. >> >> Thanks for the confirmation. This is the only issue, and the only >> reason for mark_opaque_ptr, right? I've already written it down in >> bbebe6918aa9. > > Yes as far as I can remember. You have ofcourse the exact same issue > with getarrayitem_* and friends... Ah, indeed. I'm thinking about a more involved fix, prompted by https://bitbucket.org/pypy/pypy/issue/1886 . Would it work? The idea would be to allow moving the getfield_gc, but in case it was on an opaque pointer, add a new "guard_gctype" operation in the short preamble. This is possible (and easy) with our own GCs, but wouldn't work with Boehm, so it would be conditional... Armin PS: you're still using pypy-dev(a)codespeak.net; I think this address stopped working by now. I fixed it to pypy-dev(a)python.org.

2 6

Status of SSE support?
by Timothy Baldridge Oct. 14, 2014

Oct. 14, 2014

I've read a few older articles about SIMD support in PyPy, what is the status of this? If I wanted to add something like a Vector3 type to my language (like mono did here http://tirania.org/blog/archive/2008/Nov-03.html) and wanted to take advantage of SSE are there primitives for this? Thanks, Timothy -- “One of the main causes of the fall of the Roman Empire was that–lacking zero–they had no way to indicate successful termination of their C programs.” (Robert Firth)

4 6

GCArray vs list
by Timothy Baldridge Oct. 11, 2014

Oct. 11, 2014

I'm in the process of tuning my JIT, and thus far it's going pretty well. However I can't mark the args in my Frame object a virtualizable array, because doing so throws this exception: [translation:info] File "/Users/tim/Dropbox/oss/pypy/rpython/jit/metainterp/warmspot.py", line 477, in make_virtualizable_infos [translation:info] vinfos[VTYPEPTR] = VirtualizableInfo(self, VTYPEPTR) [translation:info] File "/Users/tim/Dropbox/oss/pypy/rpython/jit/metainterp/virtualizable.py", line 52, in __init__ [translation:info] assert isinstance(ARRAY, lltype.GcArray) Digging into it I see that ARRAY is: <GcStruct list { length, items }>-- That makes me think that there's something I'm doing to my list of args that's flagging it as a list instead of a GcArray? Before I go and rework how args are handled in my interpreter I'd like a bit more information about what could be going on here. For the record, I make sure that every access to my args array is inside the valid range of the array, and is always a positive value. I'm then marking it as "args[*]" in my virtualizables. If I remove the [*] it translates fine, but then allocates a list on every invocation of an interpreter function. Thanks for all the help this far, my interpreter is coming along nicely. Timothy Baldridge

3 3

GPS vehicle tracker with 3G compatible /Attn: purchase manager与您共享了相册。
by GPS vehicle tracker with 3G compatible /Attn: purchase manager Oct. 7, 2014

Oct. 7, 2014

Dear Sir This is Anna,the sales manager of GPS in China. VT600 is an advanced GPS vehicle tracker, with 3G UMTS 900/2100Mhz or 850/1900Mhz compatible. What is more, VT600 can compatible with accelerator meter to detect harsh start and stop, can compatible with RFID to identify driver, can compatible with camera to take photo. Rich I/O features are available also. I would appreciate if you forward this letter to Technical Manager or to other expert responsible for technical integration of new products in your company, or provide me with his contact for we could discuss all the details of our future cooperation. Your early reply is highly appreciated. Best Regards Anna https://picasaweb.google.com/lh/sredir?uname=113068216911443794145&target=A…

1 0

kill Testrunner scratchbox support? (arm develoment)
by Ronny Pfannschmidt Oct. 6, 2014

Oct. 6, 2014

Hi, in the refine-testrunner branch i am currently refactoring the testrunner to allow easyer reasoning about its code and support junitxml (for better result metadata than the resultlog format) While doing so i noticed that we don't seem to use scratchbox anymore, but (according to a comment from armin on IRC) rather a modern alternative. So i am wondering if I can maybe scrap the support. However i would need some help of someone that handled the current arm setup, to roughly explain the current setup/needs. -- Ronny

3 2

Stacklet problems
by Timothy Baldridge Oct. 5, 2014

Oct. 5, 2014

I have an interpreter that takes advantage of the rstacklet api. My tests pass fine when running untranslated in CPython, but after translating via rpython I get a "Segmentation fault: 11", are there any major differences between cpython and rpython stacklets? Thanks, Timothy

2 2