Dear sdcc users, the stm8 branch has been merged today, so sdcc from trunk (and thus starting tomorrow also the sdcc snapshots) can generate code for the stm8. The port passes the regression tests (except for very few exceptions - see below), and so far everything seems to work fine on real hardware, too.
The port still has a few shortcomings, but Valentin and me will work on them: * A few regression test failures (their number is comparable to those in other ports, and it seems they mostly affect floating-point functions, in particular asinf(), atanf(), and sinhf(). * No support for extended stack yet - local variables and parameters are limited to a total of about 240 Bytes per function (exception: for functions that have no parameters, local variables that are in aggregates or unions might not count against this limit). The compiler will report an error, if it cannot generate code due to this issue. * No peephole optimizer yet, code generation still lacks some optimizations, a few more high-level ones are still missing as well - the generated code is not yet as optimized as in other ports of sdcc or other compilers for the stm8. Philipp P.S.: Since this port is now in trunk, feel free to report any issues you discover as bugs, or open feature request for features you would like to see in the port. ------------------------------------------------------------------------------ Try New Relic Now & We'll Send You this Cool Shirt New Relic is the only SaaS-based application performance monitoring service that delivers powerful full stack analytics. Optimize and monitor your browser, app, & servers with just a few lines of code. Try New Relic and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may _______________________________________________ Sdcc-user mailing list Sdcc-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/sdcc-user
