Mailman 3 December 2016 - pypy-dev

Where should I look then !
by Shubha Ramani 12 Dec '16

12 Dec '16

To do this: > On Dec 9, 2016, at 9:34 AM, Shubha Ramani via pypy-dev <pypy-dev(a)python.org> wrote: > > Armin you articulated exactly what I want. Please tell me the right place to look then ? > > Shubha > > Sent from Shubha Ramani's iPhone 7 > >> On Dec 9, 2016, at 8:29 AM, Armin Rigo <armin.rigo(a)gmail.com> wrote: >> >> Hi, >> >> On 9 December 2016 at 15:55, Shubha Ramani via pypy-dev >> <pypy-dev(a)python.org> wrote: >>> I am no longer confused about that. >> >> I think the rest of your mail shows there is still confusion. >> >>> But so far, in order to be able to dump >>> jitcodes, I have to instrument python code with >>> "JitDriver", "merge_point", "meta_interp", etc...under the test directory - >> >> Maybe you should start again by describing in detail what you want to >> do, from the start. >> >> Let me take a wild guess, completely unrelated to the questions you're >> asking. Maybe >> your goal is to run a "pypy" binary, which JITs some parts of the user >> Python code. >> What you want then is to map the raw machine code emitted in memory, >> back to Python- >> level information. For example, from an IP, you want an answer like >> "this 'add' instruction >> comes from JIT-compiling through this Python function, precisely here >> [Python bytecode of a CALL], with the call inlined, and running this >> other Python >> function, precisely here [Python bytecode of a BINARY_ADD]. >> >> If that's anywhere close to what you're looking for, then you are >> looking at the wrong place. >> >> >> A bientôt, >> >> Armin. > Sent from Shubha Ramani's iPhone 7

2 2

PYPYLOG - get two at once jit-backend-dump+jit-backend-addr
by Shubha Ramani 12 Dec '16

12 Dec '16

I can do export PYPYLOG=jit-backend-dump:shubha.log and I can do export PYPYLOG=jit-backend-addr:shubha.logwhile running the attached file which computes Fibonacci numbers. I run the script just by pypy fib.py and I did addthe requisite jitdriver and merge_point stuff. My question is, is there a way to get both jit-backend-addr and jit-backend-dump in the same log ? Also what other options are there for PYPYLOG besides the 2 mentioned, jit-backend-dump and jit-backend-addr ? Thanks, Shubha

3 4

how do you parse jit log containing binary dump ?
by Shubha Ramani 09 Dec '16

09 Dec '16

I successfully created a file containing binary data, but don't know what to do with it.I produced it using test /opt/pypy/rpython/jit/backend/x86/test/test_jitlog.py

4 14

int_py_mod_nonnegargs(): one arg is negative LLAssertFailure
by Shubha Ramani 09 Dec '16

09 Dec '16

Why am I getting this error ?pidigits.py runs fine outside of the pytest environment. Please see attached files

2 2

why doesn't this code dump jit debug info ?
by Shubha Ramani 08 Dec '16

08 Dec '16

import py import binary_trees import os from rpython.rlib.rjitlog import rjitlog def test_binary_trees(): file = ('./jit.log') fileno = os.open(file, os.O_WRONLY | os.O_CREAT) enable_jitlog = lambda: rjitlog.enable_jitlog(fileno) os.system("./binary_trees.py 100 100") running it as /opt/pypy/pytest.py test_shubha.py The log gets created but it is empty

2 2

Request to make `makedev`, `major` and `minor` macros
by John Zhang 06 Dec '16

06 Dec '16

Hi all (Armin?), I would like to request making the definitions of `rpython.rlib.rposix.c_makedev/c_major/c_minor` include explicit macro functions, i.e., passing `macro=_MACRO_ON_POSIX` to the `rffi.llexternal` function. These functions are macros on both Linux and macOS (the comment above the code acknowledges this fact). Without explicitly saying they are macros, these functions can’t be called using rffi. The attached patch also adds another test in test_rposix.py, which exposes the problem. The reason that they don’t currently affect the existing PyPy is because PyPy compiles to C anyway. But I’m working on another backend (Mu Micro VM), so I’d appreciate the ability to call these external functions from shared libraries directly. I think the macro wrapper generation is a great idea and its existing work is something that I can try to incorporate in my backend. I would appreciate someone can patch the change (it’s a bit tricky on my end using git-remote-hg. :P ). Regards, John Zhang ------------------------------------------------------ John Zhang Research Assistant Programming Languages, Design & Implementation Division Computer Systems Group ANU College of Engineering & Computer Science 108 North Rd The Australian National University Acton ACT 2601 john.zhang(a)anu.edu.au<mailto:john.zhang@anu.edu.au>

2 3

How to run the tests under /opt/pypy/rpython/jit/metainterp/test ?
by Shubha Ramani 05 Dec '16

05 Dec '16

Shubha D. Ramanishubharamani(a)gmail.com shubharamani(a)yahoo.com

2 3

please help. Why can't I do a pdb.set_trace() in llmodel.py at said line ?
by Shubha Ramani 05 Dec '16

05 Dec '16

Shubha D. Ramanishubharamani(a)gmail.com shubharamani(a)yahoo.com

3 8

with rpython is there a way to not write out the "c" files ?
by Shubha Ramani 05 Dec '16

05 Dec '16

It takes a long time and I don't really care about the backend c implementation at the moment.Do I just do a -s as shown below ? Command I'm running: /opt/pypy/rpython$ ./bin/rpython -Ojit --jit-backend=auto ./translator/goal/targetsimpleread.py --annotate, -a Do type inference--no-annotate Don't do type inference--rtype, -t Do rtyping--no-rtype Don't do rtyping--pyjitpl JIT generation step--no-pyjitpl Don't JIT generation step--pyjittest JIT test with llgraph backend--no-jittest Don't JIT test with llgraph backend--backendopt Do backend optimizations--no-backendopt Don't do backend optimizations--source, -s Create source--no-source Don't create source--compile, -c Compile (default goal)--no-compile Don't compile--llinterpret Interpret the rtyped flow graphs--no-llinterpret Don't interpret the rtyped flow graphs Shubha D. Ramanishubharamani(a)gmail.com shubharamani(a)yahoo.com

3 3

Mercurial "general delta"
by Armin Rigo 04 Dec '16

04 Dec '16

Hi all, A quick note for people that have a PyPy repo since years and are using it like me without ever re-cloning from Bitbucket: the format of modern mercurial repos is more compact (500MB instead of almost 900MB), but convertion of existing repos is not automatic. The new format is called General Delta. If your directory ".hg/store/" is closer to 900 than 500 MB, try this: hg clone -U --config format.generaldelta=1 --pull OLDREPO NEWREPO You can then throw away the OLDREPO. Or like me you can (1) make sure OLDREPO is up-to-date with default without any local changes; (2) only throw away OLDREPO/.hg and replace it with NEWREPO/.hg; (3) optionally re-modify .hg/hgrc; (4) then do "hg update default" to resynchronize. I noticed this because Bitbucket says the pypy repo is not General Delta. I don't know how to fix that, or if it really matters at all: I guess newly cloned repos from Bitbucket will automatically use the latest format anyway. A bientôt, Armin.

3 6