From mboxrd@z Thu Jan  1 00:00:00 1970
From: Nick Duffek <nsd@redhat.com>
To: cagney@cygnus.com, fche@redhat.com
Cc: gdb-patches@sourceware.cygnus.com
Subject: Re: Possible remote.c patch for Z-packet breakpoints + Harvard + SID
Date: Tue, 17 Jul 2001 14:25:00 -0000
Message-id: <200107171937.f6HJb4v24007@rtl.cygnus.com>
References: <3B547E19.2070201@cygnus.com>
X-SW-Source: 2001-07/msg00439.html

On 17-Jul-2001, Andrew Cagney wrote:

>Have you looked at ``TRANSLATE_XFER_ADDRESS''?  The d10v uses it to translate
>between a virtual CORE_ADDR and a real CORE_ADDR.

I think that won't work, because only Z-packet breakpoint addresses need
to be translated to real addresses before being sent to SID.
TRANSLATE_XFER_ADDRESS translates memory read/write addresses, which must
not be translated before being sent to SID for the architecture in
question.

>However, as a general rule, I think GDB should be consistent and always send
>down down CORE_ADDR's.

Meaning virtual CORE_ADDRs?  I agree.

On 16-Jul-2001, Frank Ch . Eigler wrote:

>If this gdb-side approach is not deemed acceptable to gdb folks, we
>may be able to make complementary changes on the sid side without too
>much littering.  

Do we have a consensus that a SID-side approach is preferable to a
GDB-side one?

Nick