"ARM 6, ARM710, StrongARM, ARM9, ARM7500, VRPC and RPCEmu"
I really hope that you don't consider this a demonstration of portability. And even then, if you want to draw up a list of CPU core compatibility, I think you'll find the Pace/Castle fork wins. From the list in the sources:
ARM 600, ARM 610, ARM 700, ARM 710, ARM 710a, SA-110, ARM 7500, ARM 7500FE, SA-1100, SA-1110, ARM 720T, ARM 920T, ARM 922T, 80200, 80321, Cortex-A8
Or in terms of CPU architecture, your supported list is
ARMv3, ARMv4, ARMv4T
whereas Pace/Castle support
ARMv3, ARMv4, ARMv4T, ARMv5TE, ARMv6, ARMv7
But as I'm sure you appreciate, this is only a small part of the story. Granted, you have support for the ViewFinder and the Omega graphics chip, but at least as much effort goes into driving the rest of the I/O system, of which you only appear to support two varieties, IOMD and S3C24xxx (and even then, apparently, tragically, not via a unified source tree).
By comparison, the Pace/Castle fork has support for at least 7 (that I know of) radically different I/O architectures - not counting support for complex external silicon like Chrontel, DENC, MPEG or audio CODEC chips. And the HAL was designed as a separate binary component to enable OEMs to write support for their own I/O chipsets without having to pass the source back to Pace, so there may be others that I've never seen.
Please login before posting a comment. Use the form on the right to do so or create a free account.
Search the archives
Today's featured article
Qercus reviewed but renewed? Forty months after taking out an annual subscription, Martin Hansen ponders whether or not to continue his Qercus sub 28 comments, latest by hzn on 3/8/07 4:15PM. Published: 27 Jul 2007