portelli/Grid
1
0
Fork 0
mirror of https://github.com/paboyle/Grid.git synced 2025-10-31 03:54:33 +00:00
Code Releases Activity

Updated todo list

Browse Source
This commit is contained in:
Peter Boyle
2015-05-10 15:13:50 +01:00
parent 7b0dd6c5d6
commit 27c2d13968
1 changed files with 23 additions and 54 deletions
Download Patch File Download Diff File Expand all files Collapse all files

77
TODO
View File

@@ -2,44 +2,17 @@
- use protocol buffers? replace xmlReader/Writer ec..
- Binary use htonll, htonl
* Stencil -- DONE
* CovariantShift support -----Use a class to store gauge field? (parallel transport?)
* Strong test for norm2, conj and all primitive types. -- Grid_simd test is almost there
* Reduce implemention is poor
* Bug in SeedFixedIntegers gives same output on each site.
* Bug in RNG with complex numbers ; only filling real values; need helper function -- DONE
* Stencil operator support -----Initial thoughts, trial implementation DONE.
-----some simple tests that Stencil matches Cshift.
-----do all permute in comms phase, so that copy permute
-----cases move into a buffer.
-----allow transform in/out buffers spproj
* CovariantShift support -----Use a class to store gauge field? (parallel transport?)
* Strong test for norm2, conj and all primitive types.
* Consider switch std::vector to boost arrays or something lighter weight
boost::multi_array<type, 3> A()... to replace multi1d, multi2d etc..
* norm2l is a hack. figure out syntax error and make this norm2 c.f. tests/Grid_gamma.cc
**** std::vector replacement;
Had to change "reserve" back to "resize" on std::vector in Lattice class.
This forces the constructor call on EVERY element of the array with negative
performance effects on temporaries.
The reversion was required because copy constructur has to work.
CONCLUSION: I must implement a similar to vector without construction/fill on
resize. Find out if valarray or alternative works differently prior to
doing this since there may still be something I can use..
* Make the Tensor types and Complex etc... play more nicely.
** Make the Tensor types and Complex etc... play more nicely.
- TensorRemove is a hack, come up with a long term rationalised approach to Complex vs. Scalar<Scalar<Scalar<Complex > > >
QDP forces use of "toDouble" to get back to non tensor scalar. This role is presently taken TensorRemove, but I
@@ -60,24 +33,35 @@
- I have collated into single location at least.
- Need to use _mm_*insert/extract routines.
* Conformable test in Cshift routines.
* Flavour matrices?
* Pauli, SU subgroup, etc..
* su3 exponentiation & log etc.. [Jamie's code?]
* TaProj
* Fourspin, two spin project --- DONE
* FFTnD ?
* Parallel MPI2 IO
Plaquette checks into nersc reader.
* rb4d support for 5th dimension in Mobius.
* Check for missing functionality - partially audited against QDP++ layout
// Base class to share common code between vRealF, VComplexF etc...
//
// Unary functions
// cos,sin, tan, acos, asin, cosh, acosh, tanh, sinh, // Scalar<vReal> only arg
// exp, log, sqrt, fabs
//
// transposeColor, transposeSpin,
// adjColor, adjSpin,
//
// copyMask.
//
// localMaxAbs
//
// Fourier transform equivalent.
Algorithms
* LinearOperator
@@ -97,29 +81,14 @@ AUDITS:
* Replace vset with a call to merge.;
* care in Gmerge,Gextract over vset .
* extract / merge extra implementation removal
* Test infrastructure
// TODO
//
// Base class to share common code between vRealF, VComplexF etc...
//
// Unary functions
// cos,sin, tan, acos, asin, cosh, acosh, tanh, sinh, // Scalar<vReal> only arg
// exp, log, sqrt, fabs
//
// transposeColor, transposeSpin,
// adjColor, adjSpin,
//
// copyMask.
//
// localMaxAbs
//
// Fourier transform equivalent.
======================================================================================================
FUNCTIONALITY:
* Stencil -- DONE
* Test infrastructure -- DONE
* Fourspin, two spin project --- DONE
* Dirac Gamma/Dirac structures ---- DONE
* Conditional execution, where etc... -----DONE, simple test
* Integer relational support -----DONE