Mailman 3 October 2015 - pypy-dev

[Fwd: Re: Building pypy from sources: partial build support ?]
by Yury V. Zaytsev Oct. 16, 2015

Oct. 16, 2015

-------- Forwarded Message -------- From: David Cournapeau <cournape(a)gmail.com> To: Yury V. Zaytsev <yury(a)shurup.com> Subject: Re: [pypy-dev] Building pypy from sources: partial build support ? Date: Thu, 15 Oct 2015 16:37:15 +0100 Thank Yury. It seems my error happens after that stage actually: * _tkinter/tklib_build.py /usr/bin/ld: cannot find -ltcl collect2: ld returned 1 exit status Traceback (most recent call last): File "_tkinter/tklib_build.py", line 51, in <… [View More]

3 2

Re: [pypy-dev] Building pypy from sources: partial build support ?
by Yury V. Zaytsev Oct. 15, 2015

Oct. 15, 2015

On Thu, 2015-10-15 at 16:37 +0100, David Cournapeau wrote: > > Is there any way for to fix my build environment and restart the build > without starting from scratch there ? Hi David, I will forward your original mail to the list, so that someone could correct me if I'm wrong, but I think that in your case since pypy-c was built successfully, you can just try to fix the tcl stuff and run pypy-c .../pypy/tool/build_cffi_imports.py as suggested here: http://pypy.org/download.… [View More]

1 0

Building pypy from sources: partial build support ?
by David Cournapeau Oct. 15, 2015

Oct. 15, 2015

Hi there, I am trying to build pypy from sources on Linux, but I get some compilation/linking errors at the end (related to gdbm, tcl/tk, etc...). Fixing those are easy, but running "pypy ../../rpython/bin/rpython -Ojit targetpypystandalone " seems to start a build from scratch, so every small fix has a 1.5 hour cycle. Is there a way to restart the compilation from the last point it failed ? The docs did not seem to mention anything, thanks, David

3 2

PPC stacklet switching code
by Vaibhav Sood Oct. 15, 2015

Oct. 15, 2015

Hi, The stack switching code (switch_xxx_gcc.h) for PPC is missing for stacklets under rpython/translator/c/src/stacklet, I am trying to work on adding that code but I am facing an issue when doing that due to the way PPC assembly works and needed some help/advice I added the switch_ppc64_gcc.h file for PPC with the slp_switch function similar to the one's in the other files (eg. switch_x86_gcc.h etc) and I run the test cases in tests.c to test the new code but these fail with segfaults. I … [View More]gdb'd the failures and find that the segfaults are due to PPC assembly using the calling functions stack to store the LR register but the g_save_state/g_restore_state functions in stacklet.c overwriting the stack contents causing the stored value to be destroyed Specifically, Running test_new(..) in tests.c calls stacklet_new(..) to create a new stacklet which in turn calls g_initialstub. _stacklet_switchstack is called twice here, first to save the old stack contents using g_initial_save_state in the heap memory and then again in the "if" code to restore the contents to the new stack using g_restore_state. The issue is when running the g_restore_state function, at the start the function prologue code saves the Link Register (LR) in the LR save area in the calling function's (slp_switch calls g_restore_state) stack frame as per PPC function calling conventions. This value is restored back into the LR register at the end by the epilogue code to return back to the calling function slp_switch. However in between the calling function's stack content is overwritten by the g_restore_state code (it memcpy's the stored heap memory data to the stack) causing the saved LR value to be destroyed and thus returning from the g_restore_state function (using blr) causes a segfault I am not sure how to get around this problem since it is specific to the way PPC assembly works and the same thing is not an issue on other platforms, any help/advice/comments would be a great help Thanks! Vaibhav DISCLAIMER ========== This e-mail may contain privileged and confidential information which is the property of Persistent Systems Ltd. It is intended only for the use of the individual or entity to which it is addressed. If you are not the intended recipient, you are not authorized to read, retain, copy, print, distribute or use this message. If you have received this communication in error, please notify the sender and delete all copies of this message. Persistent Systems Ltd. does not accept any liability for virus infected mails. [View Less]

2 4

p12 to p7b
by Roshan Cherian Oct. 14, 2015

Oct. 14, 2015

Hi Team, I have a requirement to convert from p12 to p7b. I am sorry for my little knowledge on this and I am not even sure I can ask this here. I am doing the following: p12 = load_pkcs12(file(self._pkcs12Path, 'rb').read()) pem = dump_certificate(FILETYPE_PEM, p12.get_certificate()) p7 = load_pkcs7_data(FILETYPE_PEM, pem) However I am getting an error: Error: [('asn1 encoding routines', 'ASN1_CHECK_TLEN', 'wrong tag'), ('asn1 encoding routines', 'ASN1_D2I_EX_PRIMITIVE', … [View More]

2 3

PEP 0484 - the Numeric Tower
by Laura Creighton Oct. 13, 2015

Oct. 13, 2015

Any chance of adding Decimal to the list of things that are also acceptable for things annotated float? Laura

1 1

Working version of pypy3 that support Python 3.3
by Ofir Shwartz Oct. 7, 2015

Oct. 7, 2015

Is there any update on that matter? I tried it now and got the same results for the most updated jit and no-jit pypy3.3. Obviously, pypy 3.2 runs perfectly good. Please assist. Thanks

3 4

Copy of list
by Tuom Larsen Oct. 7, 2015

Oct. 7, 2015

Dear PyPy developers! In CPython, a common idiom to copy a list is to use slice, as in: copy = original[:] I noticed that under PyPy 2.6.0 it seems quite a bit slower than using `list` constructor: from timeit import timeit print timeit('b = a[:]', 'a = list(range(100))') # 0.0732719898224 print timeit('b = list(a)', 'a = list(range(100))') # 0.00285792350769 Please, could some comment on what is the preferred way to copy a list under PyPy? Should I try to avoid the "slice way"? Thanks!

4 7

Crash in JIT
by Tuom Larsen Oct. 7, 2015

Oct. 7, 2015

Hello! I have a fairly long code here so I have yet to distill a short test case but I started to get a crash in 2.6.0 (I never saw this kind of error in earlier versions). Is there anything I can do to give you more helpful information besides the following error message? RPython traceback: File "pypy_module_pypyjit_interp_jit.c", line 149, in jump_absolute__AccessDirect_None File "rpython_jit_metainterp_warmstate.c", line 7443, in maybe_compile_and_run__star_5_1 File "… [View More]

2 2

Size limit
by Tuom Larsen Oct. 7, 2015

Oct. 7, 2015

Hello! I have a piece of auto-generated code and I noticed that if a function has more than certain number of expressions it starts to run much slower. I'm guessing this has something to do with what JIT considers to be still worth of compiling but I just would like to confirm this: is there any limit on how big (in term of operations) a function needs to be in order to get compiled by JIT? Is it possible to increase the limit from user code? Or is there any other reason for the slowdown? … [View More]Please find below an arficial example. Here it turn in ~0.13 seconds but when I uncomment the first line it drops to 5.3 seconds. Thanks, Tuom Larsen def f(x): a0 = x*0.1; a1 = x*0.1; a2 = x*0.2; a3 = x*0.3; a4 = x*0.4 a5 = x*0.5; a6 = x*0.6; a7 = x*0.7; a8 = x*0.8; a9 = x*0.9 b0 = x*0.1; b1 = x*0.1; b2 = x*0.2; b3 = x*0.3; b4 = x*0.4 b5 = x*0.5; b6 = x*0.6; b7 = x*0.7; b8 = x*0.8; b9 = x*0.9 return sum([ # (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9), (a0+a1+a2+a3+a4+a5+a6+a7+a8+a9)*(b0+b1+b2+b3+b4+b5+b6+b7+b8+b9)]) for i in range(100000): f(0.1) [View Less]

2 3