From peter at sensair.com Mon Dec 3 03:23:51 2007 From: peter at sensair.com (Peter Lissenburg) Date: Mon Dec 3 03:38:53 2007 Subject: [Icc-mot] Opentcp and ICC12 Now uIP In-Reply-To: <473C39F0.3060101@sensair.com> References: <00cd01c7b9ab$02b5e610$6401a8c0@ddt2> <200706282040.l5SKe4xd099242@dragonsgate2.imagecraft.com> <06aa01c8020e$66846710$6901a8c0@ddt2> <200709282124.l8SLO7sp083844@dragonsgate2.imagecraft.com> <4706EF23.5040600@sensair.com> <001001c8084b$039f4c80$a300a8c0@REKS> <47085444.9030404@sensair.com><4709CA25.3020101@sensair.com> <003401c80999$38559c00$0400a8c0@edvardo> <000601c8099d$fa5159d0$0400a8c0@edvardo> <47196356.7060405@sensair.com><200710200520.l9K5KDAf071472@dragonsgate2.imagecraft.com> <473A5F95.9030908@fiocca.net> <000601c8268c$8748c7f0$0400a8c0@edvardo> <473C39F0.3060101@sensair.com> Message-ID: <4753E747.8050809@sensair.com> Well just to answer my own musing. I have taken Pierre Morency's port http://www.ericengler.com/downloads/uIP-HCS12NE-release-1.0.zip of uIP http://www.sics.se/~adam/uip/ And got it to compile and work in ICC12 It seems that the version of uIP that Pierre used was older than the current version, which has used protothreads. What I have now is messy, but as I understand it more, I will clean it up. Anyone wanting what I have just drop me an email. I am very new to (all this C stuff!) porting code, and it is interesting to see how non portable some code is. So far, it seems that ICC12 helps in writing portable code because it has a minimum of "extensions" to C. Is this the case or am I missing something? Cheers. Peter L. Peter Lissenburg wrote: > Hi all, > working on getting opentcp working with ICC12 and NE64. > I am looking closely at > http://www.celestialaudio.com/MC9S12NE64/ > but I know too little about codewarrior projects to figure out how to > rearrange the project structure into something reasonable, that I may > use the info from Jesse. > I see a (non) link at > http://www.imagecraft.com/pub/code_samples/CPU12/NO_FILE > on the ICC12 demo download page, has anyone got some sort of "blank > project" for this? > Or a pointer to how I should proceed? > > As hardware, I have the NE64 demo board, and my own prototype. > I would be happy to "publish" a ICC12 project for the demo board once I > have got that far. > > Any thoughts most welcome. > Thanks. > Peter L. > > _______________________________________________ > Icc-mot mailing list > Icc-mot@imagecraft.com > http://dragonsgate.net/mailman/listinfo/icc-mot > > > From dean.bell at flightec.com Mon Dec 3 11:37:43 2007 From: dean.bell at flightec.com (Dean Bell) Date: Mon Dec 3 11:51:34 2007 Subject: [Icc-mot] Opentcp and ICC12 Now uIP In-Reply-To: <4753E747.8050809@sensair.com> References: <00cd01c7b9ab$02b5e610$6401a8c0@ddt2> <200706282040.l5SKe4xd099242@dragonsgate2.imagecraft.com> <06aa01c8020e$66846710$6901a8c0@ddt2> <200709282124.l8SLO7sp083844@dragonsgate2.imagecraft.com> <4706EF23.5040600@sensair.com> <001001c8084b$039f4c80$a300a8c0@REKS> <47085444.9030404@sensair.com> <4709CA25.3020101@sensair.com> <003401c80999$38559c00$0400a8c0@edvardo> <000601c8099d$fa5159d0$0400a8c0@edvardo> <47196356.7060405@sensair.com> <200710200520.l9K5KDAf071472@dragonsgate2.imagecraft.com> <473A5F95.9030908@fiocca.net> <000601c8268c$8748c7f0$0400a8c0@edvardo> <473C39F0.3060101@sensair.com> <4753E747.8050809@sensair.com> Message-ID: <009c01c835e3$f9614950$1101a8c0@Technicaldesgnxp1b12e69c9632> Hi Peter I sore your last post just as I was looking at uIP and in the mean time I have ported Ver 0.9 under ICC12, and have it running on a NeCore12 from TechArt. It is still work in progress but I have a SMTP client, HTTP server (with CGI), TIME client (port 37) and a simple TELNET server running, as well as the main app. It was not too hard and I have learned lot from the excise. I have only been working on it part time before I go to the office in the morning. The tricky part is the 64K and the size of the HTML pages, and the balance of the RAM between app requirements and the Ethernet buffers. I am currently using 1024 bytes each for the 2 x RX, 1 x Tx buffers. This was in part due to the largest HTMP page just over 900 bytes. Dean. -----Original Message----- From: icc-mot-bounces@imagecraft.com [mailto:icc-mot-bounces@imagecraft.com] On Behalf Of Peter Lissenburg Sent: Tuesday, 4 December 2007 00:24 To: Discussion List for ICC08/11/12/16 users. You do NOT need to subscribe to icc-announce if you are a member of this. Subject: Re: [Icc-mot] Opentcp and ICC12 Now uIP Well just to answer my own musing. I have taken Pierre Morency's port http://www.ericengler.com/downloads/uIP-HCS12NE-release-1.0.zip of uIP http://www.sics.se/~adam/uip/ And got it to compile and work in ICC12 It seems that the version of uIP that Pierre used was older than the current version, which has used protothreads. What I have now is messy, but as I understand it more, I will clean it up. Anyone wanting what I have just drop me an email. I am very new to (all this C stuff!) porting code, and it is interesting to see how non portable some code is. So far, it seems that ICC12 helps in writing portable code because it has a minimum of "extensions" to C. Is this the case or am I missing something? Cheers. Peter L. Peter Lissenburg wrote: > Hi all, > working on getting opentcp working with ICC12 and NE64. > I am looking closely at > http://www.celestialaudio.com/MC9S12NE64/ > but I know too little about codewarrior projects to figure out how to > rearrange the project structure into something reasonable, that I may > use the info from Jesse. > I see a (non) link at > http://www.imagecraft.com/pub/code_samples/CPU12/NO_FILE > on the ICC12 demo download page, has anyone got some sort of "blank > project" for this? > Or a pointer to how I should proceed? > > As hardware, I have the NE64 demo board, and my own prototype. > I would be happy to "publish" a ICC12 project for the demo board once I > have got that far. > > Any thoughts most welcome. > Thanks. > Peter L. > > _______________________________________________ > Icc-mot mailing list > Icc-mot@imagecraft.com > http://dragonsgate.net/mailman/listinfo/icc-mot > > > _______________________________________________ Icc-mot mailing list Icc-mot@imagecraft.com http://dragonsgate.net/mailman/listinfo/icc-mot From peter at sensair.com Mon Dec 3 13:05:30 2007 From: peter at sensair.com (Peter Lissenburg) Date: Mon Dec 3 13:20:35 2007 Subject: [Icc-mot] Opentcp and ICC12 Now uIP In-Reply-To: <009c01c835e3$f9614950$1101a8c0@Technicaldesgnxp1b12e69c9632> References: <00cd01c7b9ab$02b5e610$6401a8c0@ddt2> <200706282040.l5SKe4xd099242@dragonsgate2.imagecraft.com> <06aa01c8020e$66846710$6901a8c0@ddt2> <200709282124.l8SLO7sp083844@dragonsgate2.imagecraft.com> <4706EF23.5040600@sensair.com> <001001c8084b$039f4c80$a300a8c0@REKS> <47085444.9030404@sensair.com> <4709CA25.3020101@sensair.com> <003401c80999$38559c00$0400a8c0@edvardo> <000601c8099d$fa5159d0$0400a8c0@edvardo> <47196356.7060405@sensair.com> <200710200520.l9K5KDAf071472@dragonsgate2.imagecraft.com> <473A5F95.9030908@fiocca.net> <000601c8268c$8748c7f0$0400a8c0@edvardo> <473C39F0.3060101@sensair.com> <4753E747.8050809@sensair.com> <009c01c835e3$f9614950$1101a8c0@Technicaldesgnxp1b12e69c9632> Message-ID: <47546F9A.4030800@sensair.com> Hi Dean, sounds like you're way ahead of me. I just have a simple HTML server, and I think there is a sort of CGI there too. But telnet will be very useful. I'll look into it. Did version 0.9 of uIP use protothreads? I'm starting to set it up to use uexec from Richard. PL Dean Bell wrote: > Hi Peter > > I sore your last post just as I was looking at uIP and in the mean time I > have ported Ver 0.9 under ICC12, and have it running on a NeCore12 from > TechArt. > > It is still work in progress but I have a SMTP client, HTTP server (with > CGI), TIME client (port 37) and a simple TELNET server running, as well as > the main app. It was not too hard and I have learned lot from the excise. I > have only been working on it part time before I go to the office in the > morning. > > The tricky part is the 64K and the size of the HTML pages, and the balance > of the RAM between app requirements and the Ethernet buffers. I am currently > using 1024 bytes each for the 2 x RX, 1 x Tx buffers. This was in part due > to the largest HTMP page just over 900 bytes. > > Dean. > > -----Original Message----- > From: icc-mot-bounces@imagecraft.com [mailto:icc-mot-bounces@imagecraft.com] > On Behalf Of Peter Lissenburg > Sent: Tuesday, 4 December 2007 00:24 > To: Discussion List for ICC08/11/12/16 users. You do NOT need to subscribe > to icc-announce if you are a member of this. > Subject: Re: [Icc-mot] Opentcp and ICC12 Now uIP > > Well just to answer my own musing. > I have taken Pierre Morency's port > http://www.ericengler.com/downloads/uIP-HCS12NE-release-1.0.zip > of uIP > http://www.sics.se/~adam/uip/ > And got it to compile and work in ICC12 > It seems that the version of uIP that Pierre used was older than the > current version, which has used protothreads. What I have now is messy, > but as I understand it more, I will clean it up. Anyone wanting what I > have just drop me an email. > > I am very new to (all this C stuff!) porting code, and it is interesting > to see how non portable some code is. > So far, it seems that ICC12 helps in writing portable code because it > has a minimum of "extensions" to C. Is this the case or am I missing > something? > > Cheers. > Peter L. > > > > > > Peter Lissenburg wrote: >> Hi all, >> working on getting opentcp working with ICC12 and NE64. >> I am looking closely at >> http://www.celestialaudio.com/MC9S12NE64/ >> but I know too little about codewarrior projects to figure out how to >> rearrange the project structure into something reasonable, that I may >> use the info from Jesse. >> I see a (non) link at >> http://www.imagecraft.com/pub/code_samples/CPU12/NO_FILE >> on the ICC12 demo download page, has anyone got some sort of "blank >> project" for this? >> Or a pointer to how I should proceed? >> >> As hardware, I have the NE64 demo board, and my own prototype. >> I would be happy to "publish" a ICC12 project for the demo board once I >> have got that far. >> >> Any thoughts most welcome. >> Thanks. >> Peter L. >> >> _______________________________________________ >> Icc-mot mailing list >> Icc-mot@imagecraft.com >> http://dragonsgate.net/mailman/listinfo/icc-mot >> >> >> > > _______________________________________________ > Icc-mot mailing list > Icc-mot@imagecraft.com > http://dragonsgate.net/mailman/listinfo/icc-mot > > > _______________________________________________ > Icc-mot mailing list > Icc-mot@imagecraft.com > http://dragonsgate.net/mailman/listinfo/icc-mot > > > From richard at imagecraft.com Wed Dec 5 14:12:30 2007 From: richard at imagecraft.com (Richard) Date: Wed Dec 5 14:26:50 2007 Subject: [Icc-mot] Request for help - Dragon-12 users Message-ID: <200712052226.lB5MQmbh001047@dragonsgate2.imagecraft.com> If you are a Dragon-12 users, please send me a msg off list. I need to debug a VERY simple program. Thanks. // richard On-line orders, support, and listservers available on web site. [ For technical support on ImageCraft products, please include all previous replies in your msgs. ] From richard at imagecraft.com Wed Dec 5 17:56:49 2007 From: richard at imagecraft.com (Richard) Date: Wed Dec 5 18:11:10 2007 Subject: [Icc-mot] Compiler price increase expected Jan 01, 2008. Order now... Message-ID: <200712060211.lB62B88a003641@dragonsgate2.imagecraft.com> We expect to increase the list prices of the compilers in 2008. Order now to take advantage of the current lower prices. // richard On-line orders, support, and listservers available on web site. [ For technical support on ImageCraft products, please include all previous replies in your msgs. ] From robmilne at web.net Wed Dec 5 18:20:01 2007 From: robmilne at web.net (Rob Milne) Date: Wed Dec 5 18:32:24 2007 Subject: [Icc-mot] Compiler price increase expected Jan 01, 2008. Order now... In-Reply-To: <200712060211.lB62B88a003641@dragonsgate2.imagecraft.com> References: <200712060211.lB62B88a003641@dragonsgate2.imagecraft.com> Message-ID: <47575C51.1030505@web.net> Hi Richard, Is the existence of a 7.x PRO version of ICC12 a possibility before then? -rob Richard wrote: > We expect to increase the list prices of the compilers in 2008. Order > now to take advantage of the current lower prices. > > > // richard > On-line orders, support, > and listservers available on web site. > [ For technical support on ImageCraft products, please include all > previous replies in your msgs. ] > _______________________________________________ > Icc-mot mailing list > Icc-mot@imagecraft.com > http://dragonsgate.net/mailman/listinfo/icc-mot > From jpdi at free.fr Wed Dec 5 22:12:06 2007 From: jpdi at free.fr (jpdi) Date: Wed Dec 5 22:26:00 2007 Subject: [Icc-mot] Compiler price increase expected Jan 01, 2008. Ordernow... In-Reply-To: <47575C51.1030505@web.net> Message-ID: <20071206061207.1ACB13F6168@smtp5-g19.free.fr> Hi, Richard, Same question, about PRO Version ? I'm waiting about it (using Version 6). What about XGate assembler for S12X ? I don't find anything about it in demo version ? Joel Petrique > -----Message d'origine----- > De?: icc-mot-bounces@imagecraft.com [mailto:icc-mot- > bounces@imagecraft.com] De la part de Rob Milne > Envoy??: jeudi 6 d?cembre 2007 03:20 > ??: Discussion List for ICC08/11/12/16 users. You do NOT need to subscribe > toicc-announce if you are a member of this. > Objet?: Re: [Icc-mot] Compiler price increase expected Jan 01, 2008. > Ordernow... > > Hi Richard, > > Is the existence of a 7.x PRO version of ICC12 a possibility before then? > > -rob > > Richard wrote: > > We expect to increase the list prices of the compilers in 2008. Order > > now to take advantage of the current lower prices. > > > > > > // richard > > On-line orders, support, > > and listservers available on web site. > > [ For technical support on ImageCraft products, please include all > > previous replies in your msgs. ] > > _______________________________________________ > > Icc-mot mailing list > > Icc-mot@imagecraft.com > > http://dragonsgate.net/mailman/listinfo/icc-mot > > > > _______________________________________________ > Icc-mot mailing list > Icc-mot@imagecraft.com > http://dragonsgate.net/mailman/listinfo/icc-mot From ekarpicz at freemail.lt Thu Dec 6 00:13:43 2007 From: ekarpicz at freemail.lt (Edward Karpicz) Date: Thu Dec 6 00:27:26 2007 Subject: [Icc-mot] Compiler price increase expected Jan 01, 2008. Ordernow... References: <20071206061207.1ACB13F6168@smtp5-g19.free.fr> Message-ID: <000601c837df$eaa18df0$0400a8c0@edvardo> Hi XGATE assembler is here. .CPU = XGATE asm directive switches to XGATE mode. But XGATE isn't very usable without necessary linker mods :-(. These aren't yet done. Pages data pointer isn't done, double isn't done etc etc. New S12X instructions aren't supported by V7 asm. :-( Edward ----- Original Message ----- From: "jpdi" To: "'Discussion List for ICC08/11/12/16 users. You do NOT need tosubscribetoicc-announce if you are a member of this.'" Sent: Thursday, December 06, 2007 8:12 AM Subject: RE: [Icc-mot] Compiler price increase expected Jan 01,2008. Ordernow... Hi, Richard, Same question, about PRO Version ? I'm waiting about it (using Version 6). What about XGate assembler for S12X ? I don't find anything about it in demo version ? Joel Petrique > -----Message d'origine----- > De : icc-mot-bounces@imagecraft.com [mailto:icc-mot- > bounces@imagecraft.com] De la part de Rob Milne > Envoy? : jeudi 6 d?cembre 2007 03:20 > ? : Discussion List for ICC08/11/12/16 users. You do NOT need to subscribe > toicc-announce if you are a member of this. > Objet : Re: [Icc-mot] Compiler price increase expected Jan 01, 2008. > Ordernow... > > Hi Richard, > > Is the existence of a 7.x PRO version of ICC12 a possibility before then? > > -rob > > Richard wrote: > > We expect to increase the list prices of the compilers in 2008. Order > > now to take advantage of the current lower prices. > > > > > > // richard > > On-line orders, support, > > and listservers available on web site. > > [ For technical support on ImageCraft products, please include all > > previous replies in your msgs. ] > > _______________________________________________ > > Icc-mot mailing list > > Icc-mot@imagecraft.com > > http://dragonsgate.net/mailman/listinfo/icc-mot > > > > _______________________________________________ > Icc-mot mailing list > Icc-mot@imagecraft.com > http://dragonsgate.net/mailman/listinfo/icc-mot _______________________________________________ Icc-mot mailing list Icc-mot@imagecraft.com http://dragonsgate.net/mailman/listinfo/icc-mot From richard-lists at imagecraft.com Thu Dec 6 01:27:18 2007 From: richard-lists at imagecraft.com (Richard) Date: Thu Dec 6 01:41:39 2007 Subject: [Icc-mot] Compiler price increase expected Jan 01, 2008. Order now... In-Reply-To: <200712060211.lB62B88a003641@dragonsgate2.imagecraft.com> References: <200712060211.lB62B88a003641@dragonsgate2.imagecraft.com> Message-ID: <200712060941.lB69fbVg007266@dragonsgate2.imagecraft.com> [ Some questions about V7 PRO for ICC12 ] Good question, let me think about it. Certainly I don't mind if people pre-purchase V7 PRO. These features are for sure to be available in first PRO release, which should be fairly soon: - __far for paged function and paged pointers. This uses the same mechanism as __flash which we just added to the AVR compiler and that partoculare feature is working, so porting it to CPU12 should be fairly straight forward - MIO global optimizer **** XGate, S12X etc. has to wait until later. On a pragmatic level, ImageCraft's model is to target mass market chips so we can have a large installed base, it's the old Borland $50 Tubro-C model. Unfortunately, with Freescale aggressively giving away their software tools, the Freescale market is becoming increasing difficult. In the end, Freescale is shooting themselves in the foot, but time will tell. At 05:56 PM 12/5/2007, Richard wrote: >We expect to increase the list prices of the compilers in 2008. >Order now to take advantage of the current lower prices. > // richard (This email is for mailing lists. To reach me directly, please use richard at imagecraft.com) From barryc at rjlsystems.com Thu Dec 6 05:19:54 2007 From: barryc at rjlsystems.com (Barry Callahan) Date: Thu Dec 6 05:33:46 2007 Subject: [Icc-mot] Compiler price increase expected Jan 01, 2008. Order now... In-Reply-To: <200712060941.lB69fbVg007266@dragonsgate2.imagecraft.com> References: <200712060211.lB62B88a003641@dragonsgate2.imagecraft.com> <200712060941.lB69fbVg007266@dragonsgate2.imagecraft.com> Message-ID: <4757F6FA.8080803@rjlsystems.com> Does V7 allow for library functions (like atof, ftoa, (s)printf...) to be paged? Or do they have to be nonpaged functions as they do in V6.16? Richard wrote: > [ Some questions about V7 PRO for ICC12 ] > > Good question, let me think about it. Certainly I don't mind if people > pre-purchase V7 PRO. > > These features are for sure to be available in first PRO release, > which should be fairly soon: > - __far for paged function and paged pointers. This uses the same > mechanism as __flash which we just added to the AVR compiler and that > partoculare feature is working, so porting it to CPU12 should be > fairly straight forward > > - MIO global optimizer > > **** > XGate, S12X etc. has to wait until later. On a pragmatic level, > ImageCraft's model is to target mass market chips so we can have a > large installed base, it's the old Borland $50 Tubro-C model. > Unfortunately, with Freescale aggressively giving away their software > tools, the Freescale market is becoming increasing difficult. In the > end, Freescale is shooting themselves in the foot, but time will tell. > > At 05:56 PM 12/5/2007, Richard wrote: >> We expect to increase the list prices of the compilers in 2008. Order >> now to take advantage of the current lower prices. >> > > // richard (This email is for mailing lists. To reach me directly, > please use richard at imagecraft.com) > _______________________________________________ > Icc-mot mailing list > Icc-mot@imagecraft.com > http://dragonsgate.net/mailman/listinfo/icc-mot From richard-lists at imagecraft.com Thu Dec 6 12:17:17 2007 From: richard-lists at imagecraft.com (Richard) Date: Thu Dec 6 12:31:39 2007 Subject: [Icc-mot] Compiler price increase expected Jan 01, 2008. Order now... In-Reply-To: <4757F6FA.8080803@rjlsystems.com> References: <200712060211.lB62B88a003641@dragonsgate2.imagecraft.com> <200712060941.lB69fbVg007266@dragonsgate2.imagecraft.com> <4757F6FA.8080803@rjlsystems.com> Message-ID: <200712062031.lB6KVc3j013879@dragonsgate2.imagecraft.com> Actually, a number of library functions, including printf, are already in paged forms! The only exceptions are the ones written in asm since they cannot be converted to paged forms automatically. At 05:19 AM 12/6/2007, Barry Callahan wrote: >Does V7 allow for library functions (like atof, ftoa, (s)printf...) >to be paged? Or do they have to be nonpaged functions as they do in V6.16? > > // richard (This email is for mailing lists. To reach me directly, please use richard at imagecraft.com) From dean.bell at flightec.com Thu Dec 6 17:58:02 2007 From: dean.bell at flightec.com (Dean Bell) Date: Thu Dec 6 18:11:56 2007 Subject: [Icc-mot] Compiler price increase expected Jan 01, 2008. Order now... In-Reply-To: <200712062031.lB6KVc3j013879@dragonsgate2.imagecraft.com> References: <200712060211.lB62B88a003641@dragonsgate2.imagecraft.com> <200712060941.lB69fbVg007266@dragonsgate2.imagecraft.com> <4757F6FA.8080803@rjlsystems.com> <200712062031.lB6KVc3j013879@dragonsgate2.imagecraft.com> Message-ID: <00eb01c83874$99ee6b40$1101a8c0@Technicaldesgnxp1b12e69c9632> Hi Richard What will be the size of the expected price increase? Dean. From ekarpicz at freemail.lt Wed Dec 12 12:05:16 2007 From: ekarpicz at freemail.lt (Edward Karpicz) Date: Wed Dec 12 12:19:12 2007 Subject: [Icc-mot] CPU12 V7.03A released References: <200708170734.l7H7Y3Zj008551@dragonsgate2.imagecraft.com> Message-ID: <002601c83cfa$50c35bd0$a300a8c0@REKS> V7.03A changelog states that bug of converting unsigned int to FP was fixed. See messages dated August 20th, 2007. Bug has been here since August! http://dragonsgate.net/pipermail/icc-mot/2007-August/date.html Edward ----- Original Message ----- From: "Richard" To: ; Sent: Friday, August 17, 2007 9:21 AM Subject: [Icc-mot] CPU12 V7.03A released > V7.03A - August 17th, 2007 > IDE/Compiler > - Added "MIO Global Optimizations" to the Project->Options->Compiler > tab. As MIO has not been fully tuned for the CPU12 target, please > consider this as beta testing of this feature. As such, this is > enabled for all versions. When released, MIO will only be enabled > for the PRO version. Please notify us of any issues you may > encounter while beta testing MIO. > Compiler > - Occsionally the compiler optimizer incorrectly removed an > instruction > - compiler was incorrectly using int to floating point conversion in > the cases where it should be using unsigned to floating point > conversion. > > > // richard > On-line orders, support, and > listservers available on web site. > [ For technical support on ImageCraft products, please include all > previous replies in your msgs. ] > _______________________________________________ > Icc-mot mailing list > Icc-mot@imagecraft.com > http://dragonsgate.net/mailman/listinfo/icc-mot > From richard-lists at imagecraft.com Wed Dec 12 13:19:18 2007 From: richard-lists at imagecraft.com (Richard) Date: Wed Dec 12 13:33:57 2007 Subject: [Icc-mot] CPU12 V7.03A released In-Reply-To: <002601c83cfa$50c35bd0$a300a8c0@REKS> References: <200708170734.l7H7Y3Zj008551@dragonsgate2.imagecraft.com> <002601c83cfa$50c35bd0$a300a8c0@REKS> Message-ID: <200712122133.lBCLXuGs097579@dragonsgate2.imagecraft.com> Are you saying there is still a bug? Got a test case? I will check. At 12:05 PM 12/12/2007, Edward Karpicz wrote: >V7.03A changelog states that bug of converting unsigned int to FP >was fixed. See messages dated August 20th, 2007. Bug has been here >since August! > >http://dragonsgate.net/pipermail/icc-mot/2007-August/date.html > >Edward > >----- Original Message ----- From: "Richard" >To: ; >Sent: Friday, August 17, 2007 9:21 AM >Subject: [Icc-mot] CPU12 V7.03A released > > >>V7.03A - August 17th, 2007 >> IDE/Compiler >> - Added "MIO Global Optimizations" to the Project->Options->Compiler >> tab. As MIO has not been fully tuned for the CPU12 target, please >> consider this as beta testing of this feature. As such, this is >> enabled for all versions. When released, MIO will only be enabled >> for the PRO version. Please notify us of any issues you may >> encounter while beta testing MIO. >> Compiler >> - Occsionally the compiler optimizer incorrectly removed an instruction >> - compiler was incorrectly using int to floating point conversion in >> the cases where it should be using unsigned to floating point >> conversion. // richard (This email is for mailing lists. To reach me directly, please use richard at imagecraft.com) From ekarpicz at freemail.lt Wed Dec 12 13:47:20 2007 From: ekarpicz at freemail.lt (Edward Karpicz) Date: Wed Dec 12 14:01:15 2007 Subject: [Icc-mot] CPU12 V7.03A released References: <200708170734.l7H7Y3Zj008551@dragonsgate2.imagecraft.com><002601c83cfa$50c35bd0$a300a8c0@REKS> <200712122133.lBCLXuGs097579@dragonsgate2.imagecraft.com> Message-ID: <001401c83d08$930b6790$a300a8c0@REKS> Compiler uses the same runtime function int2fp when it converts signed and unsigned ints to FP. It can't be just lib bug, it must be compiler. unsigned short us; unsigned int ui; signed short ss; signed int si; float f; void foo(void) { f = us; f = ui; f = ss; f = si; } 10D9 ; unsigned short us; 10D9 ; unsigned int ui; 10D9 ; signed short ss; 10D9 ; signed int si; 10D9 ; float f; 10D9 ; 10D9 ; void foo(void) 10D9 ; { 10D9 .dbline 101 10D9 ; f = us; 10D9 FC020A ldd _us 10DC 16116B jsr int2fp 10DF 1805B10200 movw 2,S+,_f 10E4 1805B10202 movw 2,S+,_f+2 10E9 .dbline 102 10E9 ; f = ui; 10E9 FC0208 ldd _ui 10EC 16116B jsr int2fp 10EF 1805B10200 movw 2,S+,_f 10F4 1805B10202 movw 2,S+,_f+2 10F9 .dbline 103 10F9 ; f = ss; 10F9 FC0206 ldd _ss 10FC 16116B jsr int2fp 10FF 1805B10200 movw 2,S+,_f 1104 1805B10202 movw 2,S+,_f+2 1109 .dbline 104 1109 ; f = si; 1109 FC0204 ldd _si 110C 16116B jsr int2fp 110F 1805B10200 movw 2,S+,_f 1114 1805B10202 movw 2,S+,_f+2 Edward ----- Original Message ----- From: "Richard" To: Sent: Wednesday, December 12, 2007 11:19 PM Subject: Re: [Icc-mot] CPU12 V7.03A released > Are you saying there is still a bug? Got a test case? I will check. > > At 12:05 PM 12/12/2007, Edward Karpicz wrote: >>V7.03A changelog states that bug of converting unsigned int to FP was >>fixed. See messages dated August 20th, 2007. Bug has been here since >>August! >> >>http://dragonsgate.net/pipermail/icc-mot/2007-August/date.html >> >>Edward >> >>----- Original Message ----- From: "Richard" >>To: ; >>Sent: Friday, August 17, 2007 9:21 AM >>Subject: [Icc-mot] CPU12 V7.03A released >> >> >>>V7.03A - August 17th, 2007 >>> IDE/Compiler >>> - Added "MIO Global Optimizations" to the Project->Options->Compiler >>> tab. As MIO has not been fully tuned for the CPU12 target, please >>> consider this as beta testing of this feature. As such, this is >>> enabled for all versions. When released, MIO will only be enabled >>> for the PRO version. Please notify us of any issues you may >>> encounter while beta testing MIO. >>> Compiler >>> - Occsionally the compiler optimizer incorrectly removed an >>> instruction >>> - compiler was incorrectly using int to floating point conversion in >>> the cases where it should be using unsigned to floating point >>> conversion. > > // richard (This email is for mailing lists. To reach me directly, please > use richard at imagecraft.com) > _______________________________________________ > Icc-mot mailing list > Icc-mot@imagecraft.com > http://dragonsgate.net/mailman/listinfo/icc-mot > From richard-lists at imagecraft.com Wed Dec 12 14:36:29 2007 From: richard-lists at imagecraft.com (Richard) Date: Wed Dec 12 14:51:09 2007 Subject: [Icc-mot] CPU12 V7.03A released In-Reply-To: <001401c83d08$930b6790$a300a8c0@REKS> References: <200708170734.l7H7Y3Zj008551@dragonsgate2.imagecraft.com> <002601c83cfa$50c35bd0$a300a8c0@REKS> <200712122133.lBCLXuGs097579@dragonsgate2.imagecraft.com> <001401c83d08$930b6790$a300a8c0@REKS> Message-ID: <200712122251.lBCMp83L098784@dragonsgate2.imagecraft.com> AH, it appears that somehow the fixed compiler was not released in 7.03A. Wonder what happened. I will make a 7.03B soon. Thanks At 01:47 PM 12/12/2007, Edward Karpicz wrote: >Compiler uses the same runtime function int2fp when it converts >signed and unsigned ints to FP. It can't be just lib bug, it must be compiler. > >unsigned short us; >unsigned int ui; >signed short ss; >signed int si; >float f; > >void foo(void) >{ > f = us; > f = ui; > f = ss; > f = si; >} > > > >10D9 ; unsigned short us; >10D9 ; unsigned int ui; >10D9 ; signed short ss; >10D9 ; signed int si; >10D9 ; float f; >10D9 ; >10D9 ; void foo(void) >10D9 ; { >10D9 .dbline 101 >10D9 ; f = us; >10D9 FC020A ldd _us >10DC 16116B jsr int2fp >10DF 1805B10200 movw 2,S+,_f >10E4 1805B10202 movw 2,S+,_f+2 >10E9 .dbline 102 >10E9 ; f = ui; >10E9 FC0208 ldd _ui >10EC 16116B jsr int2fp >10EF 1805B10200 movw 2,S+,_f >10F4 1805B10202 movw 2,S+,_f+2 >10F9 .dbline 103 >10F9 ; f = ss; >10F9 FC0206 ldd _ss >10FC 16116B jsr int2fp >10FF 1805B10200 movw 2,S+,_f >1104 1805B10202 movw 2,S+,_f+2 >1109 .dbline 104 >1109 ; f = si; >1109 FC0204 ldd _si >110C 16116B jsr int2fp >110F 1805B10200 movw 2,S+,_f >1114 1805B10202 movw 2,S+,_f+2 > >Edward > > >----- Original Message ----- From: "Richard" >To: >Sent: Wednesday, December 12, 2007 11:19 PM >Subject: Re: [Icc-mot] CPU12 V7.03A released > > >>Are you saying there is still a bug? Got a test case? I will check. >> >>At 12:05 PM 12/12/2007, Edward Karpicz wrote: >>>V7.03A changelog states that bug of converting unsigned int to FP >>>was fixed. See messages dated August 20th, 2007. Bug has been here >>>since August! >>> >>>http://dragonsgate.net/pipermail/icc-mot/2007-August/date.html >>> >>>Edward >>> >>>----- Original Message ----- From: "Richard" >>>To: ; >>>Sent: Friday, August 17, 2007 9:21 AM >>>Subject: [Icc-mot] CPU12 V7.03A released >>> >>> >>>>V7.03A - August 17th, 2007 >>>> IDE/Compiler >>>> - Added "MIO Global Optimizations" to the Project->Options->Compiler >>>> tab. As MIO has not been fully tuned for the CPU12 target, please >>>> consider this as beta testing of this feature. As such, this is >>>> enabled for all versions. When released, MIO will only be enabled >>>> for the PRO version. Please notify us of any issues you may >>>> encounter while beta testing MIO. >>>> Compiler >>>> - Occsionally the compiler optimizer incorrectly removed an >>>> instruction >>>> - compiler was incorrectly using int to floating point conversion in >>>> the cases where it should be using unsigned to floating point >>>> conversion. >> >>// richard (This email is for mailing lists. To reach me directly, >>please use richard at imagecraft.com) >>_______________________________________________ >>Icc-mot mailing list >>Icc-mot@imagecraft.com >>http://dragonsgate.net/mailman/listinfo/icc-mot > > >_______________________________________________ >Icc-mot mailing list >Icc-mot@imagecraft.com >http://dragonsgate.net/mailman/listinfo/icc-mot // richard (This email is for mailing lists. To reach me directly, please use richard at imagecraft.com) From ekarpicz at freemail.lt Wed Dec 12 20:30:01 2007 From: ekarpicz at freemail.lt (Edward Karpicz) Date: Wed Dec 12 20:43:51 2007 Subject: [Icc-mot] CPU12 V7.03A released References: <200708170734.l7H7Y3Zj008551@dragonsgate2.imagecraft.com><002601c83cfa$50c35bd0$a300a8c0@REKS><200712122133.lBCLXuGs097579@dragonsgate2.imagecraft.com><001401c83d08$930b6790$a300a8c0@REKS> <200712122251.lBCMp83L098784@dragonsgate2.imagecraft.com> Message-ID: <001001c83d40$d3efab80$0200a8c0@leon> That could explain why we didn't see the benefits of MIO introduced in 7.03A :-). I wonder, can we try it with STD/ADV? Any example codes available? Thanks! Edward ----- Original Message ----- From: "Richard" To: "Discussion List for ICC08/11/12/16 users. You do NOT need to subscribe toicc-announce if you are a member of this." ; "Discussion List for ICC08/11/12/16 users. You do NOT need tosubscribetoicc-announce if you are a member of this." Sent: Thursday, December 13, 2007 12:36 AM Subject: Re: [Icc-mot] CPU12 V7.03A released > AH, it appears that somehow the fixed compiler was not released in 7.03A. > Wonder what happened. I will make a 7.03B soon. > > Thanks > > At 01:47 PM 12/12/2007, Edward Karpicz wrote: >>Compiler uses the same runtime function int2fp when it converts signed and >>unsigned ints to FP. It can't be just lib bug, it must be compiler. >> >>unsigned short us; >>unsigned int ui; >>signed short ss; >>signed int si; >>float f; >> >>void foo(void) >>{ >> f = us; >> f = ui; >> f = ss; >> f = si; >>} >> >> >> >>10D9 ; unsigned short us; >>10D9 ; unsigned int ui; >>10D9 ; signed short ss; >>10D9 ; signed int si; >>10D9 ; float f; >>10D9 ; >>10D9 ; void foo(void) >>10D9 ; { >>10D9 .dbline 101 >>10D9 ; f = us; >>10D9 FC020A ldd _us >>10DC 16116B jsr int2fp >>10DF 1805B10200 movw 2,S+,_f >>10E4 1805B10202 movw 2,S+,_f+2 >>10E9 .dbline 102 >>10E9 ; f = ui; >>10E9 FC0208 ldd _ui >>10EC 16116B jsr int2fp >>10EF 1805B10200 movw 2,S+,_f >>10F4 1805B10202 movw 2,S+,_f+2 >>10F9 .dbline 103 >>10F9 ; f = ss; >>10F9 FC0206 ldd _ss >>10FC 16116B jsr int2fp >>10FF 1805B10200 movw 2,S+,_f >>1104 1805B10202 movw 2,S+,_f+2 >>1109 .dbline 104 >>1109 ; f = si; >>1109 FC0204 ldd _si >>110C 16116B jsr int2fp >>110F 1805B10200 movw 2,S+,_f >>1114 1805B10202 movw 2,S+,_f+2 >> >>Edward >> >> >>----- Original Message ----- From: "Richard" >> >>To: >>Sent: Wednesday, December 12, 2007 11:19 PM >>Subject: Re: [Icc-mot] CPU12 V7.03A released >> >> >>>Are you saying there is still a bug? Got a test case? I will check. >>> >>>At 12:05 PM 12/12/2007, Edward Karpicz wrote: >>>>V7.03A changelog states that bug of converting unsigned int to FP was >>>>fixed. See messages dated August 20th, 2007. Bug has been here since >>>>August! >>>> >>>>http://dragonsgate.net/pipermail/icc-mot/2007-August/date.html >>>> >>>>Edward >>>> >>>>----- Original Message ----- From: "Richard" >>>>To: ; >>>>Sent: Friday, August 17, 2007 9:21 AM >>>>Subject: [Icc-mot] CPU12 V7.03A released >>>> >>>> >>>>>V7.03A - August 17th, 2007 >>>>> IDE/Compiler >>>>> - Added "MIO Global Optimizations" to the >>>>> Project->Options->Compiler >>>>> tab. As MIO has not been fully tuned for the CPU12 target, >>>>> please >>>>> consider this as beta testing of this feature. As such, this is >>>>> enabled for all versions. When released, MIO will only be >>>>> enabled >>>>> for the PRO version. Please notify us of any issues you may >>>>> encounter while beta testing MIO. >>>>> Compiler >>>>> - Occsionally the compiler optimizer incorrectly removed an >>>>> instruction >>>>> - compiler was incorrectly using int to floating point conversion >>>>> in >>>>> the cases where it should be using unsigned to floating point >>>>> conversion. >>> >>>// richard (This email is for mailing lists. To reach me directly, please >>>use richard at imagecraft.com) >>>_______________________________________________ >>>Icc-mot mailing list >>>Icc-mot@imagecraft.com >>>http://dragonsgate.net/mailman/listinfo/icc-mot >> >> >>_______________________________________________ >>Icc-mot mailing list >>Icc-mot@imagecraft.com >>http://dragonsgate.net/mailman/listinfo/icc-mot > > // richard (This email is for mailing lists. To reach me directly, please > use richard at imagecraft.com) > _______________________________________________ > Icc-mot mailing list > Icc-mot@imagecraft.com > http://dragonsgate.net/mailman/listinfo/icc-mot > From warm38 at juno.com Wed Dec 19 11:45:50 2007 From: warm38 at juno.com (Wade A Smith) Date: Wed Dec 19 12:01:00 2007 Subject: [Icc-mot] tables in Flash not RAM Message-ID: <20071219.144558.480.1.warm38@juno.com> How do I get the table into Flash? --------------------- code #define MENU_MAX_MSGS 3 const INTU8 *menuTable[MENU_MAX_MSGS+1] = { {"\r\nFirst Menu Line\r\n"}, {"\r\nThe Second Menu Line\r\n"}, {"\r\nThe Third Menu Line\r\n"}, {""} // end of table }; --------------------- generated code .area data 3020 _menuTable:: 0020 .blkw 1 .area idata --- 0020 03B8 .word L22 .area data 0022 .blkw 1 .area idata --- 0022 0375 .word L23 .area data 0024 .blkw 1 .area idata --- 0024 032D .word L24 .area data 0026 .blkw 1 .area idata --- 0026 03E7 .word L21 --------------------- menuTable is at $3020 which is a RAM address. Any problems and I lose the addresses of the strings. L22, L23, L24 and L21 are in Flash at 0x5###. How do I get menuTable to be in Flash and NOT RAM? 'const' does nothing. If I leave out 'const', it does the same -- the variable is in RAM and the strings are in Flash and I still cant see the variables. wade -------------- next part -------------- An HTML attachment was scrubbed... URL: http://dragonsgate.net/pipermail/icc-mot/attachments/20071219/92bbbda6/attachment.html From dean.bell at flightec.com Wed Dec 19 14:28:58 2007 From: dean.bell at flightec.com (Dean Bell) Date: Wed Dec 19 14:43:05 2007 Subject: [Icc-mot] tables in Flash not RAM In-Reply-To: <20071219.144558.480.1.warm38@juno.com> References: <20071219.144558.480.1.warm38@juno.com> Message-ID: <008201c8428e$8cbac9e0$1101a8c0@Technicaldesgnxp1b12e69c9632> Try this. const INTU8 MENU_1_MSG[] = {"\r\nFirst Menu Line\r\n"}; const INTU8 MENU_2_MSG[] = {"\r\nThe Second Menu Line\r\n"}; const INTU8 MENU_3_MSG[] = {"\r\nThe Third Menu Line\r\n"}; const INTU8 MENU_4_MSG[] = {""}; // end of table typedef const char* STR; const STR menuTable2[] = { MENU_1_MSG, MENU_2_MSG, MENU_3_MSG, MENU_4_MSG // end of table }; Dean. _____ From: icc-mot-bounces@imagecraft.com [mailto:icc-mot-bounces@imagecraft.com] On Behalf Of Wade A Smith Sent: Thursday, 20 December 2007 08:46 To: icc-mot@imagecraft.com Subject: [Icc-mot] tables in Flash not RAM How do I get the table into Flash? --------------------- code #define MENU_MAX_MSGS 3 const INTU8 *menuTable[MENU_MAX_MSGS+1] = { {"\r\nFirst Menu Line\r\n"}, {"\r\nThe Second Menu Line\r\n"}, {"\r\nThe Third Menu Line\r\n"}, {""} // end of table }; --------------------- generated code .area data 3020 _menuTable:: 0020 .blkw 1 .area idata --- 0020 03B8 .word L22 .area data 0022 .blkw 1 .area idata --- 0022 0375 .word L23 .area data 0024 .blkw 1 .area idata --- 0024 032D .word L24 .area data 0026 .blkw 1 .area idata --- 0026 03E7 .word L21 --------------------- menuTable is at $3020 which is a RAM address. Any problems and I lose the addresses of the strings. L22, L23, L24 and L21 are in Flash at 0x5###. How do I get menuTable to be in Flash and NOT RAM? 'const' does nothing. If I leave out 'const', it does the same -- the variable is in RAM and the strings are in Flash and I still cant see the variables. wade -------------- next part -------------- An HTML attachment was scrubbed... URL: http://dragonsgate.net/pipermail/icc-mot/attachments/20071220/25029463/attachment.html From engineering at data-acquisition.com.au Wed Dec 19 19:47:38 2007 From: engineering at data-acquisition.com.au (David Swinnerton) Date: Wed Dec 19 20:01:23 2007 Subject: [Icc-mot] tables in Flash not RAM References: <20071219.144558.480.1.warm38@juno.com> Message-ID: <009101c842bb$117ad7c0$9777ecdc@net> I'm not in front of a compiler to check but does the following not work? This should make both the strings and the table itself constant yes? ...James Brown Data Acquisition #define MENU_MAX_MSGS 3 const INTU8 * const menuTable[MENU_MAX_MSGS+1] = { {"\r\nFirst Menu Line\r\n"}, {"\r\nThe Second Menu Line\r\n"}, {"\r\nThe Third Menu Line\r\n"}, {""} // end of table }; ----- Original Message ----- From: Wade A Smith To: icc-mot@imagecraft.com Sent: Thursday, December 20, 2007 5:45 AM Subject: [Icc-mot] tables in Flash not RAM How do I get the table into Flash? --------------------- code #define MENU_MAX_MSGS 3 const INTU8 *menuTable[MENU_MAX_MSGS+1] = { {"\r\nFirst Menu Line\r\n"}, {"\r\nThe Second Menu Line\r\n"}, {"\r\nThe Third Menu Line\r\n"}, {""} // end of table }; --------------------- generated code .area data 3020 _menuTable:: 0020 .blkw 1 .area idata --- 0020 03B8 .word L22 .area data 0022 .blkw 1 .area idata --- 0022 0375 .word L23 .area data 0024 .blkw 1 .area idata --- 0024 032D .word L24 .area data 0026 .blkw 1 .area idata --- 0026 03E7 .word L21 --------------------- menuTable is at $3020 which is a RAM address. Any problems and I lose the addresses of the strings. L22, L23, L24 and L21 are in Flash at 0x5###. How do I get menuTable to be in Flash and NOT RAM? 'const' does nothing. If I leave out 'const', it does the same -- the variable is in RAM and the strings are in Flash and I still cant see the variables. wade ------------------------------------------------------------------------------ _______________________________________________ Icc-mot mailing list Icc-mot@imagecraft.com http://dragonsgate.net/mailman/listinfo/icc-mot -------------- next part -------------- An HTML attachment was scrubbed... URL: http://dragonsgate.net/pipermail/icc-mot/attachments/20071220/54485294/attachment.html From ekarpicz at freemail.lt Wed Dec 19 20:48:34 2007 From: ekarpicz at freemail.lt (Edward Karpicz) Date: Wed Dec 19 21:02:35 2007 Subject: [Icc-mot] tables in Flash not RAM References: <20071219.144558.480.1.warm38@juno.com> Message-ID: <006901c842c3$943caf00$0200a8c0@leon> Like David and Dean already suggested, you need two const's, one for chars, the string is made of, and another one for pointers to strings. Edward ----- Original Message ----- From: Wade A Smith To: icc-mot@imagecraft.com Sent: Wednesday, December 19, 2007 9:45 PM Subject: [Icc-mot] tables in Flash not RAM How do I get the table into Flash? --------------------- code #define MENU_MAX_MSGS 3 const INTU8 *menuTable[MENU_MAX_MSGS+1] = { {"\r\nFirst Menu Line\r\n"}, {"\r\nThe Second Menu Line\r\n"}, {"\r\nThe Third Menu Line\r\n"}, {""} // end of table }; --------------------- generated code .area data 3020 _menuTable:: 0020 .blkw 1 .area idata --- 0020 03B8 .word L22 .area data 0022 .blkw 1 .area idata --- 0022 0375 .word L23 .area data 0024 .blkw 1 .area idata --- 0024 032D .word L24 .area data 0026 .blkw 1 .area idata --- 0026 03E7 .word L21 --------------------- menuTable is at $3020 which is a RAM address. Any problems and I lose the addresses of the strings. L22, L23, L24 and L21 are in Flash at 0x5###. How do I get menuTable to be in Flash and NOT RAM? 'const' does nothing. If I leave out 'const', it does the same -- the variable is in RAM and the strings are in Flash and I still cant see the variables. wade ------------------------------------------------------------------------------ _______________________________________________ Icc-mot mailing list Icc-mot@imagecraft.com http://dragonsgate.net/mailman/listinfo/icc-mot -------------- next part -------------- An HTML attachment was scrubbed... URL: http://dragonsgate.net/pipermail/icc-mot/attachments/20071220/909ac42a/attachment.html From richard at imagecraft.com Mon Dec 24 20:25:14 2007 From: richard at imagecraft.com (Richard) Date: Mon Dec 24 20:40:03 2007 Subject: [Icc-mot] Year End Thanks and 2008 Plans Message-ID: <200712250439.lBP4di5J062077@dragonsgate2.imagecraft.com> Thank you for your support in the last year. In fact, a number of you have been our loyal customers for a number of years. For all of you, please accept our sincere thanks. The embedded tools market is becoming more challenging. We still believe we have the right model of "easy to use professional tools at an affordable price." Nevertheless, some tweaks are needed: - Minor pricing increase. The base prices of the tools will be: == STD - $249 == ADV - $369 == PRO - $549 Upgrade pricing etc. will be on the website as of Jan 1st, 2008. - More target support == Parallax Propeller C == Atmel AVR32 == > 64K support for MSP430X == S12X for CPU12 == ARM Thumb2 Of course we will continue to support and enhance our existing products. - More technological improvements == Tuned MIO == new Coloring Register allocator Our goal is to keep pushing the envelope re: price / performance, even on the high end of the market. We continue to invest in deep engineering projects in pursuing this goal. == Eclipse based IDE We will not forgo our simple to use IDE. However, as customers demand increase, we will release new IDEs based on Eclipse as alternatives. - More Product Integrations and Leveraging To further leverage our compiler expertise, we are planning to add: == A preemptive RTOS with optional TCP/IP, USB, file system, etc. The RTOS will be scalable from small 8K devices to ARM7/AVR32s. == More hardware development packages. We understand it is important to have a package options with easy to follow examples and complete tool integrations. If you have any software or hardware products that you think may be a fit for our company, please feel free to discuss it with me. Lastly, the office will be closed for a few days until the weekend, may your Heart and Hearth be warm, may the Light of your Spirit be Bright, and may the coming year bring you much joy! // richard