Why even remove the switch. This compiler is for small devices and I don't see any problem with just leaving it in. Unless it is needlessly complicating the code, removing it will break someone's program needlessly. I doubt there is much code involved, probably just a few line in the pre-processor.
Douglas On Jun 26, 2011, at 2:30 PM, csra...@bol.com.br wrote: > This list is not accepting my posts, so I'm cc'ing you Philipp in case my > thoughts get lost in the Net. > > I think you reasoning is sensible and correct and the best thing to do is to > remove --short-is-8bits. > I'd suggest a multi step approach: > 1) next release the sdcc will honour the switch but issue a deprecated > warning; > 2) next release after that will issue an error message. > 3) next after that will silently ignore the switch. > Meanwhile documentation to the users can be left in the docs and website > giving the users the ways of accomodating: > a) Stick with an older version of sdcc; > b) rewrite the code and correlated build scripts to bring code on par to the > present versions. > Regards, > -- > Cesar Rabak > > Em 25/06/2011 15:34, Philipp Klaus Krause < p...@spth.de > escreveu: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Am 25.06.2011 14:25, schrieb Kustaa Nyholm: > > > > If there is extra man power in the project to do something like this why > > not use it for something more productive like trying to optimize the code > > generation. > > It is being used for optimization. Just have a look at the table on > http://sourceforge.net/apps/trac/sdcc/wiki/Philipp%27s%20TODO%20list. > There's a 10% code size reduction between sdcc revision #6464 and #6604 > for the z80 port. Compared to sdcc 2.7.0, released just four years ago, > code size is down by over 20%. > > However unused and unecessary things are binding manpower: There's more > code to break when making changes, or alternatively more code to > understand to not break anything. The users have to go through > documentaion for s uch stuff when looking for the things useful for them. > Thus, removing feature that no one uses and that provide no advantage > over other existing alternatives is something that needs some manpower > in the short term. But in the long term it saves manpower, which can be > used for implementing and maintaining the useful stuff. > > Philipp > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.11 (GNU/Linux) > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ > > iEYEARECAAYFAk4GKh0ACgkQbtUV+xsoLppoKACfayt/3yc60C4MyPLkQdTZ6w27 > DOYAn1MQ0t2mfivI6Y9jskhFsghjJrIy > =Kk5x > -----END PGP SIGNATURE----- > > ------------------------------------------------------------------------------ > All the data continuously generated in your IT infrastructure contains a > definitive record of customers, application performance, security > threats, fraudulent activity and more. Splunk takes this data and makes > sense of it. Business sense. IT sense. Common sense.. > http://p.sf.net/sfu/splunk-d2d-c1 > _______________________________________________ > Sdcc-user mailing list > Sdcc-user@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/sdcc-user > > ------------------------------------------------------------------------------ > All of the data generated in your IT infrastructure is seriously valuable. > Why? It contains a definitive record of application performance, security > threats, fraudulent activity, and more. Splunk takes this data and makes > sense of it. IT sense. And common sense. > http://p.sf.net/sfu/splunk-d2d-c2 > _______________________________________________ > Sdcc-user mailing list > Sdcc-user@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/sdcc-user ------------------------------------------------------------------------------ All of the data generated in your IT infrastructure is seriously valuable. Why? It contains a definitive record of application performance, security threats, fraudulent activity, and more. Splunk takes this data and makes sense of it. IT sense. And common sense. http://p.sf.net/sfu/splunk-d2d-c2 _______________________________________________ Sdcc-user mailing list Sdcc-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/sdcc-user
