0
I Use This!
High Activity

Commits : Listings

Analyzed about 16 hours ago. based on code collected about 16 hours ago.
Jul 20, 2018 — Jul 20, 2019
Commit Message Contributor Files Modified Lines Added Lines Removed Code Location Date
[linalg] improvement for bw-gather, now counts non-zero coeffs. Anon32 More... over 11 years ago
[linalg] also output rlink for nfsforge. Anon32 More... over 11 years ago
[linalg] forgot chomp() in doit. Anon32 More... over 11 years ago
[linalg] bw-transpose now takes --out Anon32 More... over 11 years ago
[linalg] script adjustments for many machines. Anon32 More... over 11 years ago
[linalg] added tests/ directory. Not all tests pass for the moment (pfft fails, because actually ifft is used and it's buggy). Anon32 More... over 11 years ago
revamped the doit scripts. Anon32 More... over 11 years ago
updated for shipment of newer matlingen Anon32 More... over 11 years ago
[linalg/nfsforge] added functionality to nfsforge. Anon32 More... over 11 years ago
[linalg] minor changes to cxx/slave.cpp Anon32 More... over 11 years ago
[linalg] crash early when degeneracy is detected in master* Anon32 More... over 11 years ago
[linalg] tolerate short vectors in bw-mul, extend with zeroes. Anon32 More... over 11 years ago
[linalg] minor changes ; increase size of acceptable coeffs in bw-secure. Anon32 More... over 11 years ago
[linalg] changed sse2 accumulation thing to account for mildly larger modulus. Anon32 More... over 11 years ago
[linalg] helped diagnose bad-x problems Anon32 More... over 11 years ago
[linalg] changed nfsforge to include the unreduced matrix. Anon32 More... over 11 years ago
[linalg] added some features/bugfixes to the fft lingen thing most notably: - escape the computation when generators are found, and pi goes haywire. - also fixed a 6+-year-old bug triggered by small moduli - fixed an fft allocation bug if limbs_per_coeff > 1 ; the result is still incorrect though, I'll have to fix that. - added reduction steps here and there after addmuls. Really this addmul crap should go away. - when a linear generator is found early, no message was printed because we had ``check_chance'' activated only very late. Now it's done always, but does not force the quadratic algo. added a ``mul'' program. added some stuff for printmagma (A, pi, F_INIT) Anon32 More... over 11 years ago
master2 via integer FFT. Corrected a 6+-year-old bug in the way. being tested, seems to work so far. Added a basic fft test code. Anon32 More... over 11 years ago
starting a conversion of master2 to cxx -- no ambition so far. Anon32 More... over 11 years ago
[linalg] merged matlingen Anon32 More... over 11 years ago
[linalg] corrected copy bug. Anon32 More... over 11 years ago
[linalg] corrected bw-secure for small modulus, + corrected bw-gather Anon32 More... over 11 years ago
[linalg] bugfixed erroneous ``unsigned short'' (int16_t was meant), + console lock bw-random looks at the modulus recover_vector is smarter. Anon32 More... over 11 years ago
[linalg] added sse2 code for tiny modulus ; rocks, really. Anon32 More... over 11 years ago
missing cxx/transpose.cpp Anon32 More... over 11 years ago
factoring out traits on the scalar type. Anon32 More... over 11 years ago
[linalg] added nfsforge helper program. Anon32 More... over 11 years ago
[linalg] tiny scripts improvements. Anon32 More... over 11 years ago
[linalg] allow matrix rows to come in non-sorted order. Anon32 More... over 11 years ago
Imports for block wiedemann -- code as it was roughly one year ago Anon32 More... over 11 years ago