On Wed, 2009-06-10 at 21:59 -0400, Duane Ellis wrote: > ?? >> on The List here. If I can afford to take the time, OpenOCD will be > ?? >> fully decoupled from its various driver modules. > > duane> Debugging these things become a *ROYAL* *P*I*T*A* > > duane> Symbols don't work, etc, (ie: I've been debugging some SANE > drivers, ugh, and I've done quite a > duane> bit with these). > > Zach> The Linux kernel shows how modules can be done well: static or > dynamic. > Zach> I would settle for no less here. > > You missed a key point.
Nope, but you missed the one that I tried to make (tangentially). > How the *modules* are done - is well done, no questions asked. > > How modules are debugged, is another matter entirely. > That is my point. In terms of debugging, "static modules" would be no different than what we have today. Literally, I see "plugins" being another option that users can enable, with the default being the same code we have now. Use them or don't. I expect some users will always want to have a monolithic OpenOCD binary. That's fine, but let us have our cake too. Allowing for plug-ins should not change anything for developers, but they would reduce OpenOCD's run-time resource requirements without compromising on installed features. Users of binary packages should appreciate being able to load only those modules that they need into memory, not be forced to load them all. You don't have to debug this feature, but some of us would. --Z _______________________________________________ Openocd-development mailing list Openocd-development@lists.berlios.de https://lists.berlios.de/mailman/listinfo/openocd-development
