tag:blogger.com,1999:blog-3971202189709462152.post5161284681004717142..comments2024-03-11T12:50:02.036+01:00Comments on PyPy Status Blog: A snake which bites its tail: PyPy JITting itselfCarl Friedrich Bolz-Tereickhttp://www.blogger.com/profile/00518922641059511014noreply@blogger.comBlogger21125tag:blogger.com,1999:blog-3971202189709462152.post-17617670415077514922010-11-23T11:10:56.257+01:002010-11-23T11:10:56.257+01:00Oh and a huge congratulations for this achievement...Oh and a huge congratulations for this achievement!!!Tim Parkinhttps://www.blogger.com/profile/12597110929479917422noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-78587538863307882772010-11-23T11:10:10.490+01:002010-11-23T11:10:10.490+01:00Yes I think the word you wanted was "uses&quo...Yes I think the word you wanted was "uses" instead of "leaks". The latter implies unforseen problems and errors, the former implies that memory usage hasn't been addressed yet... Just to reiterate - PyPy currently *uses* more memory than CPython.Tim Parkinhttps://www.blogger.com/profile/12597110929479917422noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-8795549826673999742010-11-12T08:58:35.198+01:002010-11-12T08:58:35.198+01:00Technically it's not "leaking". And ...Technically it's not "leaking". And yes, we're trying to address this issue.Maciej Fijalkowskihttps://www.blogger.com/profile/11410841070239382771noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-83965604912933750892010-11-12T08:29:42.518+01:002010-11-12T08:29:42.518+01:00In other words: The pypy jit compiler leaks a mass...In other words: The pypy jit compiler leaks a massive amount of memory. Will you address this issue?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-33765348484605692972010-11-12T08:13:47.171+01:002010-11-12T08:13:47.171+01:00Nice work !Nice work !adimascihttps://www.blogger.com/profile/15008729707348314739noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-29333336250465138762010-11-11T18:26:33.518+01:002010-11-11T18:26:33.518+01:00@cfbolz Well, but you sure can run the 64bit versi...@cfbolz Well, but you sure can run the 64bit version with the same module list as you did for 32bit... So if running the benchmark again in the same conditions isn't a lot of work, it'd provide yet another interesting data point ;)Victornoreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-66373807891732835012010-11-11T14:11:27.332+01:002010-11-11T14:11:27.332+01:00@Anonymous: Because the 64bit version is translati...@Anonymous: Because the 64bit version is translating all modules, which simply gives the translator a lot more to do. We cannot do that yet on 32bit due to memory problems.Carl Friedrich Bolz-Tereickhttps://www.blogger.com/profile/00518922641059511014noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-30552028684985791592010-11-11T14:03:36.908+01:002010-11-11T14:03:36.908+01:00How come the 64 bit timings are so much worse than...How come the 64 bit timings are so much worse than the 32 bit timings (both CPython and PyPy)?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-70390092306953461032010-11-11T11:28:54.656+01:002010-11-11T11:28:54.656+01:00@Anonymous: you are right when you say that compil...@Anonymous: you are right when you say that compile_c mostly invokes gcc, but also a python script called trackgcroot.py.<br /><br />The python script is run with the same interpreter using for translate.py (so pypy in this case), and it happens that it's slower than with cpython.Antonio Cunihttps://www.blogger.com/profile/17017456817083804792noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-20155896569240323062010-11-11T00:19:43.743+01:002010-11-11T00:19:43.743+01:00Very cool achievement. I'm curious however to...Very cool achievement. I'm curious however to know why compile_c section is slower. I thought it was mostly waiting on external programs to run and so should of been similar time cpython? Congratulations!Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-66006559603084506652010-11-10T18:02:07.817+01:002010-11-10T18:02:07.817+01:00For reference, at some point (long ago) I tried to...For reference, at some point (long ago) I tried to use Psyco to speed up translate.py on CPython; but i didn't make any difference -- I'm guessing it's because we have nested scope variables at a few critical points, which Psyco cannot optimize. Now I no longer have a need for that :-)Armin Rigohttps://www.blogger.com/profile/06300515270104686574noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-4335738186622906362010-11-10T17:04:14.304+01:002010-11-10T17:04:14.304+01:00PyPy continues to get more and more impressive.PyPy continues to get more and more impressive.Michael Foordhttps://www.blogger.com/profile/06229713779852499022noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-37513889255605337422010-11-10T08:21:33.098+01:002010-11-10T08:21:33.098+01:00@ArneBab: I'm not sure what you mean, but cons...@ArneBab: I'm not sure what you mean, but consider that at the moment PyPy does not support Python 3, so it does not make sense to compare against it.Antonio Cunihttps://www.blogger.com/profile/17017456817083804792noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-35644112004398290572010-11-10T00:30:13.596+01:002010-11-10T00:30:13.596+01:00Wow, looks great!
Many thanks for posting the be...Wow, looks great! <br /><br />Many thanks for posting the benchmark – and for your relentless work on pypy!<br /><br />One thing: Could you add tests comparing with programs converted to python3?ArneBabhttps://www.blogger.com/profile/16449390422848764481noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-40188452401547148492010-11-10T00:26:12.893+01:002010-11-10T00:26:12.893+01:00Excellent.. congratulations!Excellent.. congratulations!Eric van Riet Paaphttps://www.blogger.com/profile/04007960162946588987noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-70108197253150485162010-11-10T00:00:16.089+01:002010-11-10T00:00:16.089+01:00Wow, cool work!Wow, cool work!Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-18547648876278001762010-11-09T19:39:16.332+01:002010-11-09T19:39:16.332+01:00@Gary yes, that is correct@Gary yes, that is correctMaciej Fijalkowskihttps://www.blogger.com/profile/11410841070239382771noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-25499817301798073182010-11-09T19:36:47.955+01:002010-11-09T19:36:47.955+01:00There should be a way to not only throw away jit m...There should be a way to not only throw away jit memory but somehow tell pypy to try to not use more than say 3gb of ram so it will not hit swap on 4gb machines.Leonardo Santagadahttps://www.blogger.com/profile/01169581670273063873noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-52949227052792973602010-11-09T19:28:52.535+01:002010-11-09T19:28:52.535+01:00This is great!
One question: A while back, after ...This is great!<br /><br />One question: A while back, after the GSoC project for 64-bit, there was an issue with asmgcc-64 such that the 64-bit GC was slower than it should be.<br /><br />It appears from the performance described in this post, that that must be resolved now. Is that right?<br /><br />Thanks,<br />GaryAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-7548754254355485032010-11-09T19:05:27.997+01:002010-11-09T19:05:27.997+01:00@Victor: No, Håkan's branch has not been merge...@Victor: No, Håkan's branch has not been merged. It still has some problems that we don't quite know how to solve.<br /><br />The nightly builds include all other improvements though. We plan to do a release at some point soon.Carl Friedrich Bolz-Tereickhttps://www.blogger.com/profile/00518922641059511014noreply@blogger.comtag:blogger.com,1999:blog-3971202189709462152.post-4577708286614695142010-11-09T18:50:23.410+01:002010-11-09T18:50:23.410+01:00This is amazing, huge kudos to all PyPy developers...This is amazing, huge kudos to all PyPy developers!<br /><br />Do these results include "Håkan's jit-unroll-loops branch" you mentioned in sprint report? When are we going to get a release containing these improvements? And do the nightly builds include them?Victornoreply@blogger.com