Code Inclusion

[Antony]

Hi,

I want to tread lightly as to not offend anyone, but I noticed in the revision history the following line: 

* Exelyx C code which will be rewritten in assembly

I am hoping that the plan for this is a "clean room" design where someone is looking at the code and writing a spec for another programmer to develop and not just looking at the code and rewriting in assembly. That is the best practice so that there is no appearance of code plagiarism (unless the original author gives consent).

The last thing from a publicity perspective that you would want is a claim that someone else's code was included in this manner, just think of all the "headache" AT&T caused the BSD line and later SCO with Linux. (There are examples, but those are probably two of the bigger ones)

-T

[Maarten Vermeulen]

Exelyx is a previous project from Mercury but it was discontinued around 2015 in favor of a new project, which became Night Kernel.

The project was open source.

[Antony]

Oh, nevermind then.

In other news, there is an open source ACPI library, the only downside is it's written in C. I'm sure you have probably seen it, it's called ACPIA. Linux uses it. Using that library would probably be helpful. I have done some studying on APM and I quit writing the code I was working on because it was giving me a headache because of some of the oddities I ran into looking at example code against RBIL (Ralf Brown's Interrupt List) I may revisit it soon and try to finish it up.

[Mercury Thirteen]

I was actually just reading about ACPIA in my recent revamp of the PS/2 Controller code. Granted, I haven't given the power management aspect of Night much thought yet, but I wonder if it would be simpler to use APM in the beginning then phase in ACPI over time? Or, maybe it would make more sense to simply skip APM altogether and go right for the end goal of ACPI but phased in in sections.

Decisions, decisions...