Re: [patch] Fix internal error on optimized-out values (regression by me)

[Jan Kratochvil <jan.kratochvil@redhat.com>]

On Tue, 11 Oct 2011 19:02:45 +0200, Tom Tromey wrote:
> Given this I tend to think that the patch that went in is incorrect,

Which patch do you refer to here?  Neither of the two proposed patch of mine
in this threads went it yet.


> in
> that it may break operations on an SRA'd structure where some bits are
> optimized away.  I thought I added tests for this, not sure though.

Missing DW_AT_data_member_location is defined as offset 0.  Not sure how an
optimized out field should look like.  Empty DWARF block will just not modify
the struct base address, therefore it may mean also the offset 0 instead of
optimized-out value.  And popping the base address keeping the stack empty is
not an empty DWARF block then.

I tried some testcase but I failed to force GCC making a field optimized out.


Thanks,
Jan