| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Scaling needs the log residual, PR needs the linear one
|
|
The gradients are correct, the test way wrong.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
This partially reverts 0bcfe3cdf917ae53764c423c7af865352af47920.
|
|
|
|
|
|
Necessary since we encourage users to interrupt indexamajig with Ctrl+C
Also, generally improve the signal handling.
|
|
|
|
Conflicts:
src/im-sandbox.c
|
|
... it certainly needed it.
|
|
|
|
|
|
|
|
|
|
|
|
D'oh! This must have crept in during a big tidyup before a commit
|
|
This trips *everyone* up.
|
|
|
|
|
|
|
|
|
|
|
|
hkl files.
|
|
|
|
|
|
Needed for OSX compatability
|
|
|
|
A mutex (in a shared memory segment) is the correct synchronisation
primitive here. I had confused myself...
|
|
Nice side-effect: indexamajig -j and --int-diag can now be used together
|
|
|
|
|
|
|
|
This allows a useful possibility of using only -y/-w to calculate a left
coset decomposition.
|
|
|
