Finally I could resuscitate the ppc os with the bring up script that Cesar recommends. It's a little difficult to get all the dependencies, but the script uploads the u-boot by the usb port using the python xmodem package, I had to comment the log functions because they are not supported by the pip default version. Also when installing the urjtag you have to add the location of the file liburjtag.so to the LD_LIBRARY_PATH or the script wont work. With the uboot running is easy transfer the uimage and romfs using a tftp server and the uboot has some handy commands to write them in flash. Thanks to everybody.
The jtag system brings me some questions... In the bringing up code there is an option to disable the FTDI, so I assume that if I select that option I would be able to connect to the ppc with my external jtag programmer? There is a more direct option to interface the ppc with another jtag? (It would be nice to have the possibility to play with a bare metal ppc) El lun., 14 sept. 2020 a las 6:10, Marc (<[email protected]>) escribió: > So one way of recovering a roach with erased flash is to jtag a very > small program into the cpu cache/SRAM. That doesn't require the DRAM to > be initialised. From there one can then program the flash. > > Such a program exists in > > > https://github.com/ska-sa/roach2_testing/roach2_production_test/support_files/program.mac > > here formatted in a way that the macgraigor tools can understand. > > It can speak xmodem, and writes whatever it receives (typically > uboot) to flash. > > But these things can be fiddly, so best not to erase uboot if at all > possible. > > regards > > marc > > On Sun, Sep 13, 2020 at 7:40 PM Sebastian Antonio Jorquera Tapia > <[email protected]> wrote: > > > > Thanks for the responses! > > I think the Matt guessing is a good starting point, also in page 26 of > the schematics says that to debug the ppc you need the halt signal, my jtag > programmer has it but I am going to take a closer look there. > > > > I had seen the binging up doc, but I had the feeling that it suppose > that you have the uboot running, they even configure the eth so they have > access to the os. > > Like I erased the uboot, the USB doesn't have a loopback so I don't see > any response there :( Anyway the ftdi should still work and I should be > able to run the scripts (crossed fingers) > > > > I also found some info about the xmd connection of the ppc in the edk > reference manual, and it says that for certain configurations the jtag > doesnt recognize the ppc as a xilinx device, so the user has to explicitly > define it to be recognized. So maybe I am in that case. > > > > > > > > El sáb., 12 sept. 2020 a las 19:43, Cesar Strauss (< > [email protected]>) escribió: > >> > >> Em 12/09/2020 18:08, Sebastian Antonio Jorquera Tapia escreveu: > >> > When I was playing trying to make a netbooting with the ROACH2 and I > >> > issue the command: "run tftpuboot" and making the story short I ended > up > >> > with no uboot :( > >> > >> I was able to successfully update the U-Boot on a ROACH2 by following > >> these instructions: > >> > >> > https://casper.ssl.berkeley.edu/wiki/ROACH-2_Revision_2#Testing.2FBringup_Status > >> > >> ... which point to this document: > >> > >> > https://docs.google.com/a/ska.ac.za/document/d/1tqw4C6uZ6EULl1OykTFL_vQTnK52UBr0aYqTg44E5wg/edit > >> > >> ... which point to this repository: > >> > >> https://github.com/ska-sa/roach2_testing > >> > >> .. which has the following bring-up script: > >> > >> > https://github.com/ska-sa/roach2_testing/blob/master/roach2_production_test/roach2_ats.py > >> > >> ... which has some low level commands: > >> > >> > https://github.com/ska-sa/roach2_testing/blob/master/roach2_production_test/roach2_ats.py#L1034 > >> > >> Interestingly, you only need a USB cable, since the FTDI chip is > >> connected to the JTAG chain internally. > >> > >> Having said that, the road was not smooth. And, in the end, updating > >> UBoot did not solve the root problem we had (DRAM check in U-Boot). We > >> ended up shipping the ROACH2 back to Digicom, where they restored it to > >> factory condition. > >> > >> Regards, > >> > >> Cesar > >> > >> -- > >> You received this message because you are subscribed to the Google > Groups "[email protected]" group. > >> To unsubscribe from this group and stop receiving emails from it, send > an email to [email protected]. > >> To view this discussion on the web visit > https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/5bc6f2ae-7dc5-868f-591a-da7415dfafcf%40inpe.br > . > > > > -- > > You received this message because you are subscribed to the Google > Groups "[email protected]" group. > > To unsubscribe from this group and stop receiving emails from it, send > an email to [email protected]. > > To view this discussion on the web visit > https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CAASoV%3DNXWk7v6OjSzrDFvxyx-9T6OEXvYymv5a2Nu_aAH2RsBA%40mail.gmail.com > . > > > > -- > https://katfs.kat.ac.za/~marc/ > > -- > You received this message because you are subscribed to the Google Groups " > [email protected]" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CAGrhWaR64THAnV8p3DXW4LzJf%2BsxLdzavAxb_EU4x9OUiKsOdQ%40mail.gmail.com > . > -- You received this message because you are subscribed to the Google Groups "[email protected]" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CAASoV%3DMsjMa2yYmMQehTJx2R4Xxth6%2BWbe3mAYZJvY7QKWJWXw%40mail.gmail.com.
