On Tue, Dec 21, 2010 at 7:01 PM, Chris Liechti <cliec...@gmx.net> wrote:
> Am 22.12.2010 00:25, schrieb Peter Bigot:
> > -mmcu=msp430xgeneric, and only build relocatable object files (not
> > applications)
>
> hardware multiplier support will be switched with different command line
> options?
>
Yes, there are three primitive options -mivcnt (size of interrupt vector
table), -mcpu (msp430, msp430x, msp430xv2), and -mmpy (distinguish the five
choices for hardware multiplier support). These will be in the gcc
infrastructure and will be what controls code generation and, to some
extent, linker behavior. My intent is to have the device package (which
contains headers and linker command files) provide a specs file for gcc that
provides default values for these based on -mmcu parameters. This would be
a way to add the legacy __MSP430_1611__ defines, which is probably worth
doing (again so long as doing so does not introduce ambiguity).
>
> > Sorry, this is going to break some code. It'll probably be possible to
> work
> > around this with a legacy.h header that does some magic, but I won't get
> to
> > that in the initial releases, and will prioritize it below things like
> > integrating 20-bit pointer support and improving other optimizations
> unless
> > I hear a lot of really really loud objections.
>
> this sounds like a simple header file that maps (#defines) the
> MSP430X1121 to MSP430F1121 etc would suffice to get backwards
> compatibility.
>
That mostly works for the compiler, but the linker gets a little tricker,
especially when multiple devices with the same generic name have different
address maps.
> this would of course only make sense when this backwards compatibility
> would be automatically handled by io.h. if the users sources need to be
> touched to insert an other include, it's also easy to change the -mmcu
> option in the makefile. however it would be nice to have a solution that
> is compatible among several compiler versions.
>
> maybe you do not have to do this all by yourself and someone else can
> provide such helper files.
>
Sure. Anybody interested in this should review the ti (not pab/ti_headers)
branch of the msp430-libc git repository on the mspgcc4 sourceforge
project. The latest commits include a utility that analyzes the devices for
consistency with genericized names. What I'd like to see is a clear rule on
when a genericized name can be used, for what purpose, and which specific
device the name uses as its basis.
Peter
------------------------------------------------------------------------------
Forrester recently released a report on the Return on Investment (ROI) of
Google Apps. They found a 300% ROI, 38%-56% cost savings, and break-even
within 7 months. Over 3 million businesses have gone Google with Google Apps:
an online email calendar, and document program that's accessible from your
browser. Read the Forrester report: http://p.sf.net/sfu/googleapps-sfnew
_______________________________________________
Mspgcc-users mailing list
Mspgcc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mspgcc-users
