Sorry for top posting ... I had to hack this email together from the archive.
How can I see or find this patch that Orlando did for the SST 39VF1601. My board has it too and I can't flash probe 0. I don't want to duplicate effort by trying to make this work if Orlando submitted a patch that takes care of it. Regards, Brian From: Orlando <orlcp...@...<http://gmane.org/get-address.php?address=orlcp440%2dRe5JQEeQqe8AvxtiuMwx3w%40public.gmane.org> > Subject: Re: SST CFI flash programming<http://news.gmane.org/find-root.php?message_id=%3c881cea4b0812311334i21dee6f5h80874330e441081e%40mail.gmail.com%3e> Newsgroups: gmane.comp.debugging.openocd.devel<http://news.gmane.org/gmane.comp.debugging.openocd.devel> Date: 2008-12-31 21:34:57 GMT (15 weeks, 13 hours and 45 minutes ago) Ok guys I figured out how to do this. I have added support for non_cfi *SST * *flash* 39VF1601 and tested it. How can I upload my changes to the main branch? I imagine I cannot just submit it with svn. I have to send you a patch? Where? how? On Tue, Dec 30, 2008 at 11:29 PM, Orlando < [email protected]> wrote: > I looked through the source code and found this out > SVN Revision 155 of cfi.c > > - added support for non-CFI compatible *flash*es to cfi driver (currently > only *SST*39VFxxx devices supported) > This checkin is experimental, not suitable for general use > > That is great because my *flash* memory chip is a 39VF1601. How can I > enable this feature? I tried modifying the source code and changed > cfi_info->not_cfi = 0; to cfi_info->not_cfi = 1; > That partially did the trick since now OpenOCD seems to recognize my * > flash* but I get an error as well and none of the *flash* functions work. > This is the message I get > > *flash* probe 0 > cfi primary command set 0 unsupported > *flash* 'cfi' found at 0x00000000 > > I do not know what it means by command set 0 unsupported. then I tried > > *flash* info 0 > #0: cfi at 0x00000000, size 0x00200000, buswidth 2, chipwidth 2 > # 0: 0x00000000 (0x200000 2048kB) protection state unknown > cfi *flash* bank not probed yet > > Not proved yet? How come? I tried a *flash* erase check like this > > *flash* erase_check 0 > timeout waiting for SYSCOMP & DBGACK, last DBG_STATUS: 5 > Running slow fallback erase check - add working memory > timeout waiting for SYSCOMP & DBGACK, last DBG_STATUS: 5 > unknown error when checking erase state of *flash* bank #0 at 0x00000000 > # 0: 0x00000000 (0x200000 2048kB) erased > > From here on I keep getting these errors. Until I reboot. What is this > DBGACK and DBG_STATUS? > > I hope it does not have anything to do with my board not supporting the > DBGRQ and DBGACK JTAG pins. These are disconnected. The board uses MiniJTAG > it has all the pins but those two. > > Let me know if anybody knows how to properly enable this non-cfi > functionality. Thanks > > > On Tue, Dec 30, 2008 at 6:07 PM, Orlando < > [email protected]> wrote: > >> Hello, >> >> Does anybody know when OpenOCD is going to support *SST* "CFI" *flash*? I >> have a development board based on the s3c2440 ARM9 processor and it uses a >> 2MB *SST* *flash*. It says that it is CFI compatible but obviously that >> is not entirely true. >> > > _______________________________________________ Openocd-development mailing list openocd-developm...@... <http://gmane.org/get-address.php?address=Openocd%2ddevelopment%2d0fE9KPoRgkgATYTw5x5z8w%40public.gmane.org> https://lists.berlios.de/mailman/listinfo/openocd-development
_______________________________________________ Openocd-development mailing list [email protected] https://lists.berlios.de/mailman/listinfo/openocd-development
