From: David B. <dba...@ed...> - 2007-03-02 19:46:01
|
----- Original Message ----- From: "Erik Petrich" <epe...@iv...> To: "Development chatter about sdcc" <sdc...@li...> Sent: Sunday, December 31, 2006 4:49 PM Subject: Re: [sdcc-devel] Dissection of a compiler > > On Thu, 28 Dec 2006, David Barnett wrote: > >> > I'm working on updating the grammar to handle the inline and restrict >> > keywords. >> Does that mean there will finally be support for inline functions? That >> would be terrific! >> >> Have you made much progress? > > 'restrict' is now in the grammar and I think 'inline' will also be there > in the next few days. At that point you should be able to use them in your > source code without any complaints of syntax errors. However, note that > the C standard doesn't require the compiler to do much in response to > them. Getting the compiler to actually inline a function will take some > more work after it's in the grammar. > > Erik Erik, I know you've been busy setting up the build server. Anyway, I'm still really excited to see some changes to the front end. If there's anything I can do to lend a hand, let me know and I'll try my best. Keep in mind I'm not very experienced with the source code, but maybe you could point me in the right direction... I've been trying to get involved with SDCC for some time now, but I never seem to get very far making sense of it. I'd be interested in working on the actual inlining, but that may not be the best place to start. On another note, the "internals and porting" seems to have stagnated. If we could keep adding to it here and there when we have a free moment, it would go a long way helping people like me to dive in (I say "we", but I don't know any more useful information to add yet, and I suspect some of what I did add isn't accurate). David |
From: David B. <dba...@ed...> - 2007-03-05 15:11:26
|
----- Original Message ----- From: "Jan Waclawek" <we...@ef...> To: "Development chatter about sdcc" <sdc...@li...> Sent: Saturday, March 03, 2007 2:32 AM Subject: Re: [sdcc-devel] Dissection of a compiler > David, > > That was me who started and attempted this sort of documentation. > > Meantime, I also attempted to set up for compiling the sources, even > started > to play with linux because of it, and failed repeatedly. This is simply > beyond my rather limited capabilities of a small embedded programmer... I certainly know the feeling. > However, I still have hope that I might be able to figure out whereabouts > of > the compiler simply by reading the sources. I plan to continue on the > documentation, when I find time and energy to do it. Don't feel too obligated. Free software is all about fun =). Just reading the sources without getting them running might be kinda rough on you unless you have some amazing source code analysis tools. Anyway, you did a great job nosing out some information. I think some of the problem is that the developers who really understand the guts of SDCC don't remember what they had trouble understanding to begin with. Of course, they're also busy enough with actually fixing bugs and now with setting up the build environment. I just hope we're not annoying them with the questions and discussion when they're trying to get other work done (are we?). > Meantime, any contribution is welcome... :-) > > Jan Waclawek > > > > > David Barnett wrote: > >> On another note, the "internals and porting" seems to have stagnated. If > we >> could keep adding to it here and there when we have a free moment, it > would >> go a long way helping people like me to dive in (I say "we", but I don't >> know any more useful information to add yet, and I suspect some of what I >> did add isn't accurate). |
From: Erik P. <epe...@iv...> - 2007-03-27 06:38:41
|
Revision 4719 has most of the needed support for function inlining. I normally don't like to commit partially complete work, but I haven't been able to make much progress lately. The main thing remaining is to generate non-inlined versions of inlined functions if they are referenced in a way that could not be inlined (such as recursion or use of the address-of operator). The numbering of the inlined code's sequence points also needs fixups, but they aren't used much yet, so it will probably work anyway. We should start considering upgrading the cdb debugging information to support discontinuous address ranges mapping to a single source line. Although this is already a minor problem with regular code, inlined functions become nearly undebugable if they are instantiated more than once. It would also be nice to figure out a good way to allow inline assembly to access the parameters and return value of an inlined function. Erik |
From: Jan W. <we...@ef...> - 2007-03-03 08:29:27
|
David, That was me who started and attempted this sort of documentation. Meantime, I also attempted to set up for compiling the sources, even started to play with linux because of it, and failed repeatedly. This is simply beyond my rather limited capabilities of a small embedded programmer... However, I still have hope that I might be able to figure out whereabouts of the compiler simply by reading the sources. I plan to continue on the documentation, when I find time and energy to do it. Meantime, any contribution is welcome... :-) Jan Waclawek David Barnett wrote: > On another note, the "internals and porting" seems to have stagnated. If we > could keep adding to it here and there when we have a free moment, it would > go a long way helping people like me to dive in (I say "we", but I don't > know any more useful information to add yet, and I suspect some of what I > did add isn't accurate). |
From: <we...@ef...> - 2007-03-03 13:28:48
|
Maarten, > See the manual for more details. Oh yes. These words indicate hours and hours of finding out the how's and whys... You certainly remember that this is not the first time you encourage me to start with some serious work on sdcc. I believe I could figure out the necessary details - just it takes some nonzero time. And this is the key trouble - my time for hobbies is rather limited, having a job for living, a family, a house to maintain (and finish in some respects)... So, I rather choose a project which yield results faster, than that which needs lengthy setup of the necessary tools. So, thanks, if I will feel like it I will get back, but not today, sorry. Meantime, I hope, others will contribute to the wiki... Jan Waclawek -----Original Message----- From: "Maarten Brock" <sou...@ds...> Subj: Re: [sdcc-devel] Dissection of a compiler Date: Sat 3. Mar 2007 9:48 Size: 1K To: Development chatter about sdcc <sdc...@li...> Jan, If you feel setting up linux and compiling the sources is too hard and you're more comfortable with windows, maybe you should try the microsoft visual studio way. Go to the microsoft website and download microsoft visual C++ express for free. It lacks many features but is enough for compiling command line programs like SDCC. After opening workspace sdcc.dsw it will convert it from v6 to v8. Also install UnxUtils. See the manual for more details. HTH, Maarten |
From: Jan W. <we...@ef...> - 2007-03-05 15:19:26
|
> David Barnett wrote: > I think some of the problem is that the >developers who really understand the guts of SDCC don't remember what they >had trouble understanding to begin with. Of course, they're also busy >enough with actually fixing bugs and now with setting up the build >environment. I just hope we're not annoying them with the questions and >discussion when they're trying to get other work done (are we?). Couldn't have said better... JW |