Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
  • Dec 27 2019 23:49
  • Jan 17 2019 05:30
    xxllp commented #1426
  • Dec 13 2018 23:14
    deronnax closed #1421
  • Jul 14 2018 08:28
    corporatepiyush commented #1426
  • May 23 2018 21:37
    toshok closed #528
  • May 23 2018 21:37
    toshok closed #558
  • May 23 2018 21:34
    toshok closed #609
  • May 23 2018 21:34
    toshok closed #756
  • Feb 21 2018 05:25
    osman-masood commented #1426
  • Feb 03 2018 02:34
    skywind3000 commented #1426
  • Feb 01 2018 22:06
    danluu commented #1426
  • Jan 27 2018 19:57
    lesshaste commented #1426
  • Jan 12 2018 13:11
    skywind3000 opened #1426
  • Nov 12 2017 19:59
    xoviat commented #1421
  • Aug 25 2017 15:58
    Travis dropbox/pyston (py3) passed (4685)
  • Aug 25 2017 15:31
    kmod commented #1425
  • Aug 25 2017 15:31
    kmod closed #1425
  • Aug 24 2017 15:02
    kmod commented #1422
  • Aug 01 2017 17:06
    denji closed #378
  • Jul 28 2017 00:31
    Travis dropbox/pyston#1425 passed (4684)
Kevin Modzelewski
@kmod
and get that working first
-I will force it to stay in codegen/ast_interpreter.cpp and make things much simpler
if that doesn't help, it might be some of the static analysis we do; if you post the crash then I could make a more informed guess
Stefan O'Neil
@stefanoneil
Sure thing:
$ ./pyston_dbg -I
Pyston v0.7.0 (rev f7e8fcf7d57c624da5ac8a8be7d0d55c260ff32e), targeting Python 2.7.8
>> 
>> 
>> escal x


reached an escal statement

>> 
>> 
>> def foo():
...  escal y
...  return 4
... 
../../src/core/ast.h:1151: virtual bool pyston::ASTVisitor::visit_escal(pyston::AST_Escal *): Assertion `0' failed: 
Someone called abort!
Aborted (core dumped)
I also have gdb output, in case it is more informative:
(gdb) run happytest.py
Starting program: /home/escal3/debug_pyston/pyston/pyston_dbg happytest.py
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".


reached an escal statement

../../src/core/ast.h:1151: virtual bool pyston::ASTVisitor::visit_escal(pyston::AST_Escal *): Assertion `0' failed: 
Someone called abort!

Program received signal SIGABRT, Aborted.
0x00007ffff5185c37 in __GI_raise (sig=sig@entry=6)
    at ../nptl/sysdeps/unix/sysv/linux/raise.c:56
56    ../nptl/sysdeps/unix/sysv/linux/raise.c: No such file or directory.
(gdb) backtrace
#0  0x00007ffff5185c37 in __GI_raise (sig=sig@entry=6)
    at ../nptl/sysdeps/unix/sysv/linux/raise.c:56
#1  0x00007ffff5189028 in __GI_abort () at abort.c:89
#2  0x00000000007d0245 in abort () at ../../src/codegen/unwinding.cpp:1158
#3  0x00000000006167b7 in pyston::ASTVisitor::visit_escal (this=0x7fffffffaa88, 
    node=0x21f22d8) at ../../src/core/ast.h:1151
#4  0x00000000007dd1d1 in pyston::AST_Escal::accept (this=0x21f22d8, 
    v=0x7fffffffaa88) at ../../src/core/ast.cpp:557
#5  0x00000000014afdbc in pyston::NameCollectorVisitor::visit_functiondef (
    this=0x7fffffffaa88, node=0x21f2190)
    at ../../src/analysis/scoping_analysis.cpp:683
#6  0x00000000014e4ca5 in pyston::AST_FunctionDef::accept (this=0x21f2190, 
    v=0x7fffffffaa88) at ../../src/core/ast.cpp:521
#7  0x00000000014afcdb in pyston::NameCollectorVisitor::collect (node=0x21f2190, 
    map=0x7fffffffab18, scoping=0x7fffffffc3b0)
    at ../../src/analysis/scoping_analysis.cpp:795
#8  0x0000000000612f9e in pyston::ScopingAnalysis::analyzeSubtree (
    this=0x7fffffffc3b0, node=0x21f2190)
    at ../../src/analysis/scoping_analysis.cpp:977
#9  0x00000000014b0e85 in pyston::ScopingAnalysis::getScopeInfoForNode (
    this=0x7fffffffc3b0, node=0x21f2190)
    at ../../src/analysis/scoping_analysis.cpp:986
#10 0x00000000014f3142 in pyston::ModuleCFGProcessor::runRecursively (
    this=0x7fffffffc3b0, body=..., name=0x7ffff49af2e0, lineno=3, args=0x21f21f0, 
    orig_node=0x21f2190) at ../../src/core/cfg.cpp:3538
#11 0x00000000014f4271 in pyston::CFGVisitor::visit_functiondef (
    this=0x7fffffffbba8, node=0x21f2190) at ../../src/core/cfg.cpp:1966
#12 0x00000000014e4ca5 in pyston::AST_FunctionDef::accept (this=0x21f2190, 
    v=0x7fffffffbba8) at ../../src/core/ast.cpp:521
#13 0x00000000007f217f in pyston::computeCFG (body=..., 
    ast_type=pyston::AST_TYPE::Module, lineno=1, args=0x0, filename=0x7ffff49adf48, 
    source=0x251de40, param_names=..., scoping=0x23ed9c0, cfgizer=0x7fffffffc3b0)
    at ../../src/core/cfg.cpp:3380
#14 0x00000000014f36ec in pyston::ModuleCFGProcessor::runRecursively (
    this=0x7fffffffc3b0, body=..., name=0x7ffff7e8f2b0, lineno=1, args=0x0, 
    orig_node=0x21f20a0) at ../../src/core/cfg.cpp:3581
#15 0x00000000014f3d3f in pyston::computeAllCFGs (ast=0x21f20a0, 
    globals_from_module=true, future_flags=16, fn=0x7ffff49adf48, bm=0x23686e0)
    at ../../src/core/cfg.cpp:3597
#16 0x0000000000752df2 in pyston::compileAndRunModule (m=0x21f20a0, bm=0x23686e0)
    at ../../src/codegen/irgen/hooks.cpp:234
#17 0x00000000005fa84b in pyston::main (argc=2, argv=0x7fffffffdfc8)
    at ../../src/jit.cpp:536
#18 0x00000000005f95e2 in main (argc=2, argv=0x7fffffffdfc8)
    at ../../src/jit.cpp:591
Kevin Modzelewski
@kmod
Is that your new keyword?
Sounds like it just doesn't know what to do with it
Stefan O'Neil
@stefanoneil
my apologies for the late reply, i've been away
you were right about it being the scoping analysis
there were some places in src/analysis/scoping_analysis.cppwhere I didn't realize I needed to make additions
hence why the crash only showed up in function definitions
Stefan O'Neil
@stefanoneil
oh and yes, "escal" is just the name I am using for the keyword right now
Kevin Modzelewski
@kmod
Ah, yeah there are some differences between how we handle functions and modules, which is the source of some annoying quirks like this
Sun
@Daetalus
Hi, @stefanoneil you are working on a private repo?
Ali Ahsan
@aliahsan07
what are the instructions of running pyston after entering the docker?
Ali Ahsan
@aliahsan07
wow this project is effectively dead
Kevin Modzelewski
@kmod
yes unfortunately
When in the docker container, python will end up running Pyston
And you can pip install things as well
Ali Ahsan
@aliahsan07
@kmod can pyston be used to develop some sort of a control flow graph, given a python program?
Kevin Modzelewski
@kmod
Only inside of a single function
salvator lopez
@salvator_lopez_twitter
what about guido van rossum ? .. is he working on dropbox when u switched to Golang and Dont wanna Develop Pyston anyway ?
Tong
@tongzhou_gitlab
hi, is Pyston still under development?
our lab needs a LLVM based Python compiler, so far I only found Pyston and Numba
Kevin Modzelewski
@kmod
Nope, not under development
Python isn't really compile-able, so I think what you're asking for isn't a fit for Python
Probably the closest you can get is Cython->clang
Tong
@tongzhou_gitlab
hmm okay
thanks for the info!
Pratyush Das
@reikdas
Is anyone aware of any forks where people have continued working on this? (and still are)
Kevin Modzelewski
@kmod
I may have an as-yet-unannounced successor project in the works :)
Pratyush Das
@reikdas
:D
Sun
@Daetalus
@kmod Hi, it's your personal project?
Seo Sanghyeon
@sanxiyn
Also: is it related to kmod/dpro?
Kevin Modzelewski
@kmod
@sanxiyn yep you found it!
It's still in the very early stages, but Marius and I have been working on expanding some of the ideas from Pyston into their own project
Kevin Modzelewski
@kmod
I'll post about it in more detail when we have more to show, but until then feel free to take a look / ask questions!
An Long
@aisk
Cool!
An Long
@aisk
Dose the optimized codes will be working on the stuff which pep523 described, thus it can be support by CPython "officially" and the user can using it without any capalibility corcerns?
Kevin Modzelewski
@kmod
We use Cython, which is a different way of achieving CPython support
There are some small compatibility differences but for the most part it's just drop-in :)
An Long
@aisk
Just have a little experience on cython, but looks like cython can not aot compile the whole standard library (or runtime). So if the new project is based on cython, does it have the ability to optimize the runtime?
Kevin Modzelewski
@kmod
Oh interesting! Do you have any more info on this?
I've only seen https://cython.readthedocs.io/en/latest/src/userguide/limitations.html which makes their compatibility seem quite good
Also I'd say we "use" cython and are not "based on" cython -- we use cython to lower Python code to C code, and then do a number of C-level tricks to optimize the generated code
I'm not sure which "runtime" you're referring to, but we definitely optimize the CPython C runtime :)
Also if anyone is interested, I started putting up some issues on our issue tracker. They should be pretty approachable and would be helpful!
Sun
@Daetalus
I modified Cython source code for support new syntax, which is add a type called "nogil extension" to get rid of Cython environment, so we can enable multicore support for Python-like syntax. https://lab.nexedi.com/Daetalus/cython/commits/nogil_extension_lwan