[Icc-430] V7.05 BETA0 - error in handling of longs in div32u/mod32u?

[Richard]

Eh? All the internal routines, including div32u etc have been modified. Do 
you have a test case showing it's not working?

At 08:43 PM 1/23/2007, Kris Heidenstrom wrote:
>Richard wrote
>
>
>>V7.05
>[...]
>>     - Interface to the 32 bit internal functions for longs and floating
>>       point now use R12/R13 in addition to R14/R15 for argument passing,
>>       resulting in smaller and faster code.
>[...]
>
>There seems to be a problem with division of longs. I'm not
>100% sure of the details, but what I think has happened is this.
>
>The canned code block for div32u (also used by mod32u) has
>not been updated to use R12,13 and is still assuming that one
>parameter will be passed on the stack. At the end of the canned
>code block, the code copies the return address from (SP+0) to
>(SP+4), then adds 4 to SP, then returns. This would be right if
>there was a 32-bit parameter on the stack, but this parameter is
>now passed in registers. As a result, some data further up the
>stack is clobbered, and the MCU may or may not return to
>somewhere meaningful, depending on the stack usage of the
>function that invokes div32u/mod32u. I'm not sure whether
>the canned code for div32u/mod32u comes from a library, or is
>hard-coded into the compiler, but in either case, I think it needs
>to be updated to be compatible with this new calling convention.
>
>Regards
>
>Kris
>--
>Kris Heidenstrom         Embedded systems designer / programmer
>kris at abbey.co.nz         Abbey Systems Ltd - Telemetry Specialists
>Wellington NEW ZEALAND   Voice +64-4-385-6611  Fax +64-4-385-6848
>
>
>_______________________________________________
>Icc-430 mailing list
>Icc-430 at imagecraft.com
>http://dragonsgate.net/mailman/listinfo/icc-430

// richard (This email is for mailing lists. To reach me directly, please 
use richard at imagecraft.com)