Howto get __log_buf on a Xilinx FPGA

Howto get __log_buf on a Xilinx FPGA

[Ricardo Ayres Severo]

Hi all,

I'm trying to boot linux 2.6.24 on a XUP Virtex II Pro evaluation
board. It uncompresses ok but doesn't show nothing more on the
console. I tried to change the console to ttyS0 and ttyUL0. Now I want
to see the __log_buf but I don't if I am doing it the right way, it
only returns trash. What I'm doing is the following:

$ grep __log_buf System.map
c01f3d5c b __log_buf

$ xmd -xmp system.xmp -opt etc/xmd_ppc405_0.opt
(xmd starts ok, mdm is not enabled)

$ powerpc-eabi-gdb arch/ppc/boot/images/zImage.elf
(click on run, the image is loaded and started)

this is the output on the terminal:
---output start---
loaded at:     00400000 004F119C

board data at: 00000000 0000007C

relocated to:  00404048 004040C4

zimage at:     00404E34 004EE12C

avail ram:     004F2000 7C9E2378


Linux/PPC load: console=ttyS0,9600

Uncompressing Linux...done.

Now booting the kernel
---output end---

after a while I stop the debugger and type rst in xmd.
then I ask gdb to dump the memory from position 0x1f3d5c to 0x1f3fff
and only trash is returned.

What am I doing wrong? Or the kernel really isn't being executed?

Thanks,

-- 
Ricardo Ayres Severo <severo.ricardo@gmail.com>

Re: Howto get __log_buf on a Xilinx FPGA

[Grant Likely]

On Feb 12, 2008 12:30 PM, Ricardo Ayres Severo <severo.ricardo@gmail.com> wrote:
>
> Linux/PPC load: console=ttyS0,9600
>
> Uncompressing Linux...done.
>
> Now booting the kernel
> ---output end---
>
> after a while I stop the debugger and type rst in xmd.
> then I ask gdb to dump the memory from position 0x1f3d5c to 0x1f3fff
> and only trash is returned.
>
> What am I doing wrong? Or the kernel really isn't being executed?

In this case it seems like the kernel isn't even booting.  You can try
inserting progress markers into the kernel boot path to spit single
characters out of the uartlite port.

g.

-- 
Grant Likely, B.Sc., P.Eng.
Secret Lab Technologies Ltd.

Re: Howto get __log_buf on a Xilinx FPGA

[Ricardo Ayres Severo]

Grant,

thanks, I'll do this.

On Feb 12, 2008 5:36 PM, Grant Likely <grant.likely@secretlab.ca> wrote:
> On Feb 12, 2008 12:30 PM, Ricardo Ayres Severo <severo.ricardo@gmail.com> wrote:
> >
> > Linux/PPC load: console=ttyS0,9600
> >
> > Uncompressing Linux...done.
> >
> > Now booting the kernel
> > ---output end---
> >
> > after a while I stop the debugger and type rst in xmd.
> > then I ask gdb to dump the memory from position 0x1f3d5c to 0x1f3fff
> > and only trash is returned.
> >
> > What am I doing wrong? Or the kernel really isn't being executed?
>
> In this case it seems like the kernel isn't even booting.  You can try
> inserting progress markers into the kernel boot path to spit single
> characters out of the uartlite port.
>
> g.
>
> --
> Grant Likely, B.Sc., P.Eng.
> Secret Lab Technologies Ltd.
>



-- 
Ricardo Ayres Severo <severo.ricardo@gmail.com>