On Saturday 10 January 2009, Steve Sakoman wrote: > Perhaps I'm being a bit clueless, but I'm not sure I see the benefit > of specifying in the kernel config whether the OMAP SOC is a 3503, > 3515, 3525, or 3530. > > The processor is the same ARM Cortex-A8 in all cases. Code that may > want to use the graphics accelerator or DSP can easily check for the > presence of each at run time (Control OMAP Status Register @ > 0x4800244C) > > Is there something I am missing? I suspect the advantage would be the ability to remove some code at compile/link time rather having it linger in the executable image. If it's not got the C64x+ DSP then don't register the DSP bridge ... when done right, that ensures the bridge code won't even be in the target image. It's sort of like the flip side of being able to enable just a few boards out of a dozen possible. Code for the other boards isn't included, but the kernel is still powerful enough to run on more than a single board. (Which is great for testing, version management, and all that kind of stuff.) - Dave -- To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
