| 1 |
Hi all, |
| 2 |
|
| 3 |
*Summary:* |
| 4 |
|
| 5 |
Last week's plans: |
| 6 |
1) Successfully export src_* and test with mpich. |
| 7 |
2) Continue testing against mpi-providers and adding more versions of |
| 8 |
openmpi and mpich to the overlay. |
| 9 |
3) Continue adding to wiki/README as functionality is added. Will talk to |
| 10 |
mentors about other means of documentation as well. |
| 11 |
|
| 12 |
There were some change in plans, as I got some clarification with regards |
| 13 |
to design. Essentially, mpi-select will have little focus on the mpi |
| 14 |
implementations themselves, as the implementations will have no need to |
| 15 |
inherit mpi-select, but rather the software the depends on various |
| 16 |
mpi-select. |
| 17 |
|
| 18 |
Next week's plan: |
| 19 |
1) Successfully test mpi-select against hpl, disregarding having only |
| 20 |
mpi-dependent libs installed. |
| 21 |
2) Implement proper DEPEND= function based on users' make.conf settings. |
| 22 |
3) Clarify design goals for mpi-dependent installations and start work on |
| 23 |
executing this. |
| 24 |
|
| 25 |
*Day-by-day breakdown:* |
| 26 |
|
| 27 |
*Monday's Goals:* |
| 28 |
|
| 29 |
Research and emulate PYTHON_TARGETS make.conf execution. |
| 30 |
Work in progress. From Justin's email, I gathered that I have some more |
| 31 |
research to do with Python's implementation in make.conf. Today I looked at |
| 32 |
python code and tested different make.conf setups to see how they were |
| 33 |
handled. I'm hoping this will give me a better framework to work with |
| 34 |
similar design when I go to code some of this tomorrow. I will likely have |
| 35 |
some questions about this when it comes to combining this with multilib's |
| 36 |
handling of different ABIs. |
| 37 |
|
| 38 |
Add userspace toggles (hopefully using eselect). I may have questions about |
| 39 |
how to do this exactly. |
| 40 |
Change of plans. This will be covered later with userspace MPI selection, |
| 41 |
which I will tackle later on after the 2nd eclass is more developed. |
| 42 |
However, I will have to keep this userspace implementation in mind when |
| 43 |
writing the 2nd eclass. |
| 44 |
|
| 45 |
|
| 46 |
*Tuesday's Goals:* |
| 47 |
|
| 48 |
Implement a mixture of Python's make.conf setup while being conscious of |
| 49 |
different multilib ABIs. |
| 50 |
|
| 51 |
I believe there exists a much better way to do this as opposed to what I |
| 52 |
have currently implemented. More specifically: is there a variable that |
| 53 |
tracks the location of make.conf on any given system? Right now I am |
| 54 |
searching for make.conf in a static grep search, which I assume is a bit |
| 55 |
gross. I plan on continuing the email discussion early tomorrow. |
| 56 |
|
| 57 |
- NOTE: as tomorrow is a national holiday in the US, while I will still be |
| 58 |
working, my hours will be shorter. I am planning on 4-6 hours of work |
| 59 |
tomorrow, and will continue regular shifts for the rest of the week to hit |
| 60 |
35+ hours. |
| 61 |
|
| 62 |
|
| 63 |
|
| 64 |
*Wednesday's Goals:* |
| 65 |
|
| 66 |
Properly test and debug everything in mpi-select. There's been quite a bit |
| 67 |
of code added that has not yet been tested, it is about time I got |
| 68 |
mpi-select back to a stable build. |
| 69 |
I am still struggling with gcc/fortran compilation issues. I have tried |
| 70 |
installing and reinstalling gcc (versions 4.8.5 & 5.4.0) with |
| 71 |
USE="+fortran" with no avail when it comes to compilation errors. Any help |
| 72 |
with this would be much appreciated, as this has been a bit of a barrier |
| 73 |
that I have held onto for too long thinking I could resolve it myself. |
| 74 |
|
| 75 |
Continue design conversations. While things are starting to come together |
| 76 |
more, some concepts are still foggy as there are many things to consider, |
| 77 |
making tasks more complex. |
| 78 |
Incomplete. I hope to set up a proper weekly meeting in the near future to |
| 79 |
settle this. |
| 80 |
|
| 81 |
|
| 82 |
*Thursday's Plans:* |
| 83 |
|
| 84 |
Make mpi_wrapper in particular more useful in terms of environment |
| 85 |
variables. Start deveopment on implementation-specific case statement in |
| 86 |
src_* steps by referencing recent ebuilds. |
| 87 |
|
| 88 |
Work in progress. I have started conditional src_* for primarily mpich. |
| 89 |
Will look for openmpi ebuild trends tomorrow. |
| 90 |
|
| 91 |
Resolve testing/compiler issues. |
| 92 |
Work in progress. I anticipate on using a different bare metal Gentoo |
| 93 |
environment for testing more. |
| 94 |
|
| 95 |
|
| 96 |
|
| 97 |
*Friday's Plans:* |
| 98 |
|
| 99 |
Add openmpi conditional src_* functionality to mpi-select. |
| 100 |
Change in plans, as per our weekly meeting. I was a bit confused as to how |
| 101 |
to execute this, but I will be working on the correct version this weekend. |
| 102 |
I want to work on it while our conversation is fresh in memory ;) |
| 103 |
|
| 104 |
Iron out bugs via testing, properly comment/document functions. Clean up & |
| 105 |
refactor code that could look prettier. |
| 106 |
Work in progress. Again, I will be using some of this weekend to work on |
| 107 |
this while it is fresh in my mind. |
| 108 |
|
| 109 |
|
| 110 |
Thanks, |
| 111 |
Michael |
Archives