Re: [sage-devel] Re: Sage 8.0 Build Error on MacOS, [mpir-3.0.0.p0] Error building MPIR.

2017-07-29 Thread François Bissey 
Can we have a bit more of that log? I would like to see the full
command.

François

> On 30/07/2017, at 12:04, Michael Frey  wrote:
> 
> I tried AS_INTEGRATED_ASSEMBLER=1 make without success, I also tried adding 
> it to my environment and also from the debugging shell.
> 
> I also tried: export CFLAGS="-Wa,-q $CFLAGS" without success.
> 
> The make ends with:
> 
> vscanf.o scanf/.libs/vsscanf.o   -L/Users/michael/sage-8.0/local/lib  -m64 
> -O2 -march=skylake -mtune=skylake -Wl,-rpath 
> -Wl,/Users/michael/sage-8.0/local/lib   -install_name  
> /Users/michael/sage-8.0/local/lib/libmpir.23.dylib -compatibility_version 24 
> -current_version 24.3 -Wl,-single_module
> 
> ld: in section __DATA,__const reloc 0: X86_64_RELOC_SUBTRACTOR must have 
> r_extern=1 file 'mpn/.libs/addmul_1.o' for architecture x86_64
> 
> ld: in section __DATA,__const reloc 0: X86_64_RELOC_SUBTRACTOR must have 
> r_extern=1 file 'mpn/.libs/addmul_1.o' for architecture x86_64
> 
> clang: error: linker command failed with exit code 1 (use -v to see 
> invocation)
> 
> clang: error: linker command failed with exit code 1 (use -v to see 
> invocation)
> 
> make[2]: *** [libgmp.la] Error 1
> 
> make[2]: *** Waiting for unfinished jobs
> 
> make[2]: *** [libmpir.la] Error 1
> 
> make[1]: *** [all-recursive] Error 1
> 
> make: *** [all] Error 2
> 
> 
> 
> 
> -- 
> You received this message because you are subscribed to the Google Groups 
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to sage-devel+unsubscr...@googlegroups.com.
> To post to this group, send email to sage-devel@googlegroups.com.
> Visit this group at https://groups.google.com/group/sage-devel.
> For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Re: Sage 8.0 Build Error on MacOS, [mpir-3.0.0.p0] Error building MPIR.

2017-07-29 Thread Michael Frey 
I tried AS_INTEGRATED_ASSEMBLER=1 make without success, I also tried adding 
it to my environment and also from the debugging shell.

I also tried: export CFLAGS="-Wa,-q $CFLAGS" without success.

The make ends with:

vscanf.o scanf/.libs/vsscanf.o   -L/Users/michael/sage-8.0/local/lib  -m64 
-O2 -march=skylake -mtune=skylake -Wl,-rpath 
-Wl,/Users/michael/sage-8.0/local/lib   -install_name  
/Users/michael/sage-8.0/local/lib/libmpir.23.dylib -compatibility_version 
24 -current_version 24.3 -Wl,-single_module

ld: in section __DATA,__const reloc 0: X86_64_RELOC_SUBTRACTOR must have 
r_extern=1 file 'mpn/.libs/addmul_1.o' for architecture x86_64

ld: in section __DATA,__const reloc 0: X86_64_RELOC_SUBTRACTOR must have 
r_extern=1 file 'mpn/.libs/addmul_1.o' for architecture x86_64

clang: *error: *linker command failed with exit code 1 (use -v to see 
invocation)

clang: *error: *linker command failed with exit code 1 (use -v to see 
invocation)

make[2]: *** [libgmp.la] Error 1

make[2]: *** Waiting for unfinished jobs

make[2]: *** [libmpir.la] Error 1

make[1]: *** [all-recursive] Error 1

make: *** [all] Error 2


-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Followups to finite-field-element GAP-to-Sage conversion bug

2017-07-29 Thread Itay Bookstein 
Continuing the discussion from https://trac.sagemath.org/ticket/23153, 
there are a couple of directions to discuss:

   - Do we want to remove the LogFFE and replace it with an implementation 
   that rewrites the element as a polynomial over the primitive root?
   - Do we want to add support for GAP fields whose size is larger than 2^16 in 
   both conversion directions, so that things like SL(2, 2^24) work? Or maybe 
   we want a different/Sage-native SL implementation?

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Re: Sage 8.0 Build Error on MacOS, [mpir-3.0.0.p0] Error building MPIR.

2017-07-29 Thread Volker Braun 
I made https://trac.sagemath.org/ticket/23549


-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Re: Sage 8.0 Build Error on MacOS, [mpir-3.0.0.p0] Error building MPIR.

2017-07-29 Thread Volker Braun 
Can you try building sage with 

AS_INTEGRATED_ASSEMBLER=1 make



On Saturday, July 29, 2017 at 4:53:40 PM UTC+2, Michael Frey wrote:
>
> I found this thread on the MPIR issues list: 
> https://github.com/wbhart/mpir/issues/217
>
> It appears that the MacOS assembler is old.  Is there to configure the 
> sage build as they suggest?
>  
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Re: Sage 8.0 Build Error on MacOS, [mpir-3.0.0.p0] Error building MPIR.

2017-07-29 Thread Michael Frey 
I found this thread on the MPIR issues list: 
https://github.com/wbhart/mpir/issues/217

It appears that the MacOS assembler is old.  Is there to configure the 
sage build as they suggest?
 

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Re: Would it be a good idea to start working on a module on Population Dynamics?

2017-07-29 Thread Simon King 
On 2017-07-27, Travis Scrimshaw  wrote:
> I am assuming by "this process" is the Sage peer review. It also does 
> usually improve the quality of the code and the documentation.

+1

> That is precisely what could be a problem. Doing things one-by-one means 
> you may not see how one dependency affects another. Also, what are the 
> rules? If B is based on A, do we require that A does not break anything in 
> B or is B responsible for any downstream changes in A? If it is the former, 
> then you have a useless extra layer because there is no point in have 
> separate modules. It is functionally acting as one complete unit.

No. It just means that A has a stable and reliable API.

> How much of that code is really a maintenance burden? Would any of those 
> bugs actually be fixed if you go to external packages? How many of them 
> would actually be found, much less reported (to the correct place)? If code 
> being included too quickly is a problem, then why are you suggesting a 
> method to include code even faster with less QC measures?

+1

Cheers,
Simon

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Re: Would it be a good idea to start working on a module on Population Dynamics?

2017-07-29 Thread Simon King 
On 2017-07-25, Vincent Delecroix <20100.delecr...@gmail.com> wrote:
> The code quality is up to the project developers and not the consequence 
> of the peer review process that we have in Sage.

Of course, if the project developers decide on using peer review and unit/doc
testing, then they'll get a good quality. And now assume the development
team of some package consists of a single person. Difficult to do peer
review within the team, then...

> As a conclusion, I would be much happier with
>
>   1) A Sage kernel as small as possible with strictly specified API and 
> slow release cycle.
>
>   2) External packages registered in a database and tested on a regular 
> basis by patchbots...

... and which the Sage development team supports by providing a peer review,
please, unless the developer team of the external package has enough resources.

Kind regards,
Simon

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Re: Ergodic Theory

2017-07-29 Thread Simon King 
Hi Marc,

On 2017-07-27, Marc Masdeu  wrote:
> @Simon maybe it would help the OP if you elaborated on why the group 
> cohomology code was not included in the standard Sage...

1. It depends on an optional package with GPL-compatible licence
   at build time and on an optional package with GPL-incompatible
   licence at runtime. It thus can't be standard.
2. When I started the project (quite some time ago), Cython didn't provide
   optional extension modules.
3. Now we do have optional extension modules in the Sage src/ tree.
   I think it is a reasonable policy to reserve these for functionality
   that is provided by standard Sage that can optionally be based
   on a different backend. However, that is not the case for group
   cohomology.

> @Nikhil I would suggest that you concentrate in writing useful code in 
> ergodic theory. How to distribute it is secondary, IMO. For ease of 
> development, you definitely don't want to keep recompiling Sage every time 
> you make a small change in it, so I'd recommend you develop it outside of 
> the Sage source tree.

I don't agree on that recommendation. If you do a small change in your
code, then you need to recompile your code. But that's all. If you do a
small change in the Sage src/ tree, then recompilation of Sage will be
very quick (unless you touch src/structure/element.pxd). And if you use
ccache, the situation will be even easier.

So, in that aspect, there is no difference between src/sage and not
src/sage...

> Once you have code that you are happy to share, put 
> it on github.com or somewhere else for people to look at. You can then 
> worry about how to distribute it then.

If it is clear that the functionality you are adding fits within the
standard scope of Sage, does not depend on packages with GPL-incompatible
licence, and does depend on Sage infrastructure, I'd recommend doing it in
src/sage and open a trac ticket.

Best regards,
Simon

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.