Re: store.exp failure on i686-linux with newer gcc's

[Tom Tromey <tromey at redhat dot com>]

>>>>> "Nathan" == Nathan Froyd <froydnj@codesourcery.com> writes:

>> So, perhaps it is better to continue to copy in the bits eagerly
>> instead of postponing them to read_pieced_value.

Nathan> I'm not exactly sure what you mean here.  At least for values where some
Nathan> of the pieces live in registers, if you don't have the frame for the
Nathan> value, you can't write to the register, correct?

Oh, never mind, I was confused.

I see now that read_pieced_value will only be called to fill in a lazy
value.  So, you are correct, this can't happen for values on the
history.

This patch is ok.

Tom