Allowed EOFA to accept different solvers for the L and R operations in the heatbath step
Fixed EOFA Meofa operating on member Phi rather than input field
Added derived EOFA pseudofermion variant that allows for mixed prec CG to be used in the heatbath
Added forces/Test_mobius_gparity_eofa_mixed testing the above reproduces the regular EOFA
To Test_gamma, added checks for the various properties of the charge conjugation matrix C=-gamma2*gamma4 in Grid basis
Broke up ExactOneFlavourRatio refresh into a virtual routine that generates eta and one that uses it as with the ratio and RHMC actions
Added accessors to the pseudofermion field to TwoFlavourEvenOddRatio and ExactOneFlavourRatio
Adapted GeneralEvenOddRationalRatio and Test_rhmc_EOWilsonRatio_doubleVsMixedPrec to recent changes that require passing in serial RNG
For GeneralEvenOddRationalRatio and TwoFlavourEvenOddRatio, broke refresh into two stages, the first of which generates the random field and the second that computes the pseudofermion field.
This allows derived classes to override the generation of the random field, for example in testing.
Test_dwf_gpforce now uses Gparity in x-direction and APBC in time as opposed to G-parity in time
Added Test_action_dwf_gparity2fvs1f that compares the DWF fermion action with the 2f and the 1f (doubled-lattice) implementations of Gparity
Audited the code conventions (again) with the CPS momentum denominator
and added anti periodic in time to the Test_mobius_force.cc and
tested the Test_dwf_gpforce.
Promoted thesee to test full HMC hamiltonian, tr P^2/2 + phidag MdagM phi
with the same pdot and Udot as audited in the Integrator.h etc...
With full comments and sources for factors.
Added a conjugate momentum "filter" that applies a phase to each site. With sites set to 1.0 or 0.0 this acts as a mask and enables, for example, the freezing of inactive gauge links in DDHMC
Added tests/forces/Test_momentum_filter demonstrating the use of the filter to freeze boundary links
This sector of the code does not matter for anything other than Guido's quenched HMC
studies, and any plaq specific optimisations should be retained in a private branch
instead of destroying the code simplicity.
Christopher Kelly
Peter Boyle
Guido Cossu
David Murphy