Re: [PATCH] Resolve dynamic types before computing pointer size

[Keith Seitz <keiths@redhat.com>]

On 10/19/21 09:21, Tom Tromey wrote:
>>>>>> "Keith" == Keith Seitz via Gdb-patches <gdb-patches@sourceware.org> writes:
> 
> Keith> This patch addresses this by changing the API for find_size_for_pointer
> Keith> so that it takes the actual struct value of the pointer variable,
> Keith> resolves the pointer's target type if it is dynamic, and returns
> Keith> a size based on this resolved type.
> 
> Thanks, this looks good to me.

Unfortunately, my patch to find_size_for_pointer_math interferes with your recently committed
patch ["Fix bug in dynamic type resolution"], which you "snuck" [:-)] in before I could finish
rebasing and retesting my patch.

With my patch applied:

+FAIL: gdb.ada/array_of_variant.exp: scenario=minimal: print first element of another_array
+FAIL: gdb.ada/array_of_variant.exp: scenario=minimal: print second element of another_array

In both cases, we see "Cannot access memory at address 0x4".

So until I figure this out, I'll have to put this patch into a holding pattern.

Thank you for the review!

Keith