Am 30.06.2013 22:05, schrieb Pete Batard: > Hi Oleksij, thanks for the reply. > > On 2013.06.30 07:15, Oleksij Rempel wrote: >> right now i'm fighting one of broadcom SOCs. But it seems like yours is >> other kind. Can you please tell which EJTAG version is it? > > I actually have no idea. Is there a way to find out that information > without having the datasheet?
enable debug mode and you will get some thing like "EJTAG: features ..." > From what I gather, Broadcom seems to be very tight lipped about > revealing any kind of useful data about its SoCs, at least publicly. yeah... china be with you ;) take a look at the forums behind the wall. May be you will find some. > >> Any way. About endians >> you can check CP0 Status and Config registers by using this commands: >> mips32 cp0 12 0 >> mips32 cp0 15 0 > > Not much luck here: > > > mips32 cp0 12 0 > couldn't access reg 12 > > mips32 cp0 15 0 > couldn't access reg 15 > > To be fair, it doesn't look like I can get any insightful data from any > registers when using the current OpenOCD against this SoC. Every single > register I poke reports zero, including PC (and I can't resume execution > either). is target actually halted? you should try openocd with my changes: https://github.com/olerem/openocd/commits/brake i would really like to have some feedback from you. >> Now to your flash. >> I see that you set 16bit bus mode and 16 bit flash mode. But manually >> you tests 0x1fc00020 which according to your flash doc should be 8bit mode. > > Aha, that's a good point. > >> Till now i never seen that manufacture decided to reduce flash size to >> performance. I mean if you use 16bit mode, you will get half size for >> same price ;) >> >> Try some thing like this: >> flash bank $_FLASHNAME cfi 0x1fc00000 0x00400000 1 1 $_TARGETNAME x16_as_x8 > > Actually the line above is what I started with, but it's only when I > used 16bit that I was able to get the expected: > Flash Manufacturer/Device: 0x227e 0x0001 > > However, I think you're right: there's something fishy with using x16 > access if the relevant data is provided from the x8 addresses, so the > flash is most likely in x8 or x16_as_x8 mode. > >> may be you'll need different address: 0xbe000000 > > That's a good suggestion. I'll try to poke different base addresses in > x8 mode and see what I can get. > > Many thanks for the pointers. > > Regards, > > /Pete > > ------------------------------------------------------------------------------ > This SF.net email is sponsored by Windows: > > Build for Windows Store. > > http://p.sf.net/sfu/windows-dev2dev > _______________________________________________ > OpenOCD-devel mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/openocd-devel > -- Regards, Oleksij ------------------------------------------------------------------------------ This SF.net email is sponsored by Windows: Build for Windows Store. http://p.sf.net/sfu/windows-dev2dev _______________________________________________ OpenOCD-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/openocd-devel
