From: J.P. N. <jn...@pr...> - 2006-02-03 16:50:27
|
I'm still curious about the ramifications of the 270's 7-stage pipeline. It introduces another level of error handling requirements in the CPU core, and it may require your ARM code to be recompiled with a 270 specific compiler even if it works with other ARMs: I would guess not, but if I were designing a mobile chip I might be inclined to take shortcuts and solve the problem in the compiler (the RISC mantra). That's really the focus of my concern, and I apologize for being too general in my original question. Even so, thanks for the clarification. -----Original Message----- From: gum...@li... [mailto:gum...@li...] On Behalf Of Doug Sutherland Sent: Friday, February 03, 2006 11:10 To: gum...@li... Subject: Re: [Gumstix-users] About the Gumstix next generation The Developer's manual for PXA270 states compatibility with ARMv5TE instruction set. This is the same instruction set used by PXA255. Every ARM7 and ARM9 I have looked at uses ARMv4T instruction set. I think there is upwards compatibility such that code written for ARMv4T will run on ARMv5TE. There may be some differences in power management or startup code, but that can be true for any given ARM chip, often specific to a vendor. -- Doug J.P. Norair wrote: > I guess this is a question to the people at Gumstix working on the=20 > 270. Does it require special consideration in porting low level code=20 > in order to account for the much different power management and=20 > seemingly non-ARM-like CPU architecture? One of the nice things about > the PXA255 is that it's relatively code-compatible with ARM7s and ARM9s. ------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://sel.as-us.falkag.net/sel?cmd=3Dlnk&kid=3D103432&bid=3D230486&dat=3D= 121642 _______________________________________________ gumstix-users mailing list gum...@li... https://lists.sourceforge.net/lists/listinfo/gumstix-users |