Sanfilippo

The simpler that I can imagine concerning alignement is that the compiler is twisted to use the aligned read when it should not, but that should issue an exception. No...

> Another issue that I had, was that I did not allocate halos for fields which I used for checking the residual, something which didn't cause problems with AVX2 either...

In my Makefile.am attached to issue #364 I had made the install do what is expected from an autotools project (or at least my understanding of that...)

Yeah that will be my XMas present! I need to recheck that nothing has been changed and adjust the Makefile, plus, it was not supporting cuda yet. Which wrapper are...

I have pushed a modified version of the autotools files in [my fork](https://github.com/sunpho84/tmLQCD), I think there is no point in requesting a pull from you, because there are many "ripples...

I finished the pull request, and modified accordingly travis script and read_input.l internal parameters. Now the parser is working and the test is passing. Make works has aspected. GPU, SPI...

Yes I was using your suggestion, that is: https://github.com/Finkenrath/tmLQCD/tree/DDalphaAMG_nd linked against https://github.com/sbacchio/DDalphaAMG/commits/TM2p1p1

> @sunpho84 if you're still interested in the A40.40 run (or was it A30.40 ?) you can try with the master branch of sbacchio/DDalphaAMG and the master branch of Finkenrath/tmLQCD....

> @sunpho84 On Marconi A2, did you see the p->H(1,0) ... output? I can't remember. yes in the old logs, see e.g. /marconi_work/INF17_lqcd123_0/sanfo/hmcnf2p1p1/A40.40/logs/log_mg_1490524967 then I've tried a few variations of...

It looks to me as if it was happening on a random basis. Here you have a sample: ``` +----------------------------------------------------------+ | 2-level method | | postsmoothing K-cycle | | FGMRES...