On Mon, Jun 02, 2003 at 10:33:29PM -0400, Andrew Cagney wrote:
>On Mon, Jun 02, 2003 at 08:59:45PM -0400, Andrew Cagney wrote:
>
>>Hello,
>>
>>This adds per-frame memory and architecture methods. Instead of using
>>the global variablecurrent_gdbarch and the context dependant
>>read_memory, architecture code can use these. Should help future proof
>>architecture code should thread/target changes start happening.
>>
>>It uses them in the d10v.
>>
>>I'll commit in a few days.
>
>
>Could you explain to us where you see this going? Right now it looks
>like a solution without a problem. Sure, someday they may need a
>target parameter, but until we're closer to that day I don't see why a
>frame parameter is the way to go.
Ref: [multi-arch] The frame as the global parameter (long, important),
http://sources.redhat.com/ml/gdb/2001-02/msg00335.html
Very enlightening, thank you!
[To paraphrase you, it sure must be gratifying to see how much of that
message looks "obvious" now, two years later.]