Dwarf views broken on ia64?

[Alexandre Oliva]

On Jun  1, 2018, "Jan Beulich" <JBeulich@suse.com> wrote:

> Does either of you have any opinion on the proposed diagnostic
> improvements (without the conversion from as_bad() to as_warn()
> of course) then?

I agree having more information about the location of the problem is
desirable.  The problem is that the error was only expected to hit at
locview asserts, that check that the view at the point is zero.  In this
case, you don't even have a symbol name, just the literal constant zero.

That it hit in a different situations in which there was symbolic
information just goes to show how unexpected and broken it was for ia64
bundles.

-- 
Alexandre Oliva, freedom fighter   https://FSFLA.org/blogs/lxo
Be the change, be Free!         FSF Latin America board member
GNU Toolchain Engineer                Free Software Evangelist