Tim Haines

The requisite changes to the wiki are in https://github.com/dyninst/dyninst-wiki/pull/1. They should be merged at the same time this PR is.

The build system changes are working on at least two platforms now. Once we remove the old x86 decoding, I'll run this through the test suite.

>Dyninst instrumentation on x86-64 needs some low level instruction decoding interfaces implemented in libcommon for determining where are the register and immediate operands in an instruction, then change these operands...

https://bottle.cs.wisc.edu/branch/PR638

There are regressions on x86. You can ignore the first two runs on coriander from Sept 10.

@mxz297 Sorry, that interface is a bit confusing (it's on my list of things to fix... eventually). That first result indicates that tests which _passed_ on a machine with glibc-2.29...

@mxz297 I just pushed a fix for how the regressions are calculated. If you go back to the regressions page for the latest run on coriander, you can search for...

@mxz297 That's interesting. I'll try running it again. >The auto testing script will try to build Dyninst but I am not sure how to specify where to find Capstone in...

@mxz297 I re-ran both PR638 and the head of master on coriander, and found the following regressions: https://bottle.cs.wisc.edu/regressions/run?curid=105&previd=108

>On my end, I cannot reproduce the failure for test1_12. test1_12 is always passing on my end. Did you run this on coriander? Are you using group mode or running...