Re: pt_regs.dbcr0/1

I am not seeing where the other debug registers that are associated with
the debug state are being state changed and managed.  For example
IAC{1-4} [breakpnt], DACs [watchpnts], DVCs are being handled.

The 4xx debug hardware looks to be similar to what is spec'd in Book-E.
If this solution is not solved at this time it will come up in the
future with new Book-E parts coming down the pipes.

- kumar

On Tuesday, June 4, 2002, at 12:22 PM, Dan Malek wrote:

Paul Mackerras wrote:

quoted

So has anyone done a 4xx-aware gdb yet?

Not to my knowledge.  Daniel Jacobowitz (drow) has discussed it a
little,
but I have not heard much about it lately.

quoted

....  I think we are in violent agreement. :)

Not again......It must be that Aussie English :-)

quoted

..... gdb + xmon/kgdb
may be possible but we will have to see how complex it looks once we
know what the kernel support for gdb looks like.

I believe we have tied the BDI context switch configuration option to
allowing external debugging by ensuring the kernel doesn't mess around
with the debug registers.  We have at least discussed doing it.  As
long as we properly context switch the debug registers, all of the
software
modes should coexist happily.

Thanks.


	-- Dan

** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/