Re: [PATCH] Add proper handling for non-local references in nested functions

[Pierre-Marie de Rodat <derodat at adacore dot com>]

On 07/23/2015 07:21 PM, Kevin Buettner wrote:
> I ran into another issue, both yesterday and today, when attempting to
> apply your patch.  Apparently, it adds some test cases,
> gdb.base/nested-subp1.c, gdb.base/nested-subp1.exp, etc. which already
> exist.  As can be seen in the log below, I told patch to not assume -R
> as well as to not apply the patch.  I'm using sources, current as of
> yesterday, upon which I first applied your other patch,
> 0001-Replace-the-block_found-global-with-explicit-data-fl.patch.

Huh, strange! I think you have a dirty source tree: I just fetched the 
upstream repository and I have this:

    $ git annotate fsf/master -- gdb/testsuite/gdb.base/nested-subp1.exp
    fatal: no such path gdb/testsuite/gdb.base/nested-subp1.exp in 
fsf/master

So I guess you should check the state of your source tree with "git status".

> Here's the log showing the issue I ran into:
>
> [kev@pinnacle gdb]$ patch -p2 </tmp/0001-DWARF-handle-non-local-references-in-nested-function.patch
> patching file ada-lang.c
> [...]
> The next patch would create the file testsuite/gdb.base/nested-subp1.c,
> which already exists!  Assume -R? [n] n
> Apply anyway? [n] n
> [...]

I would suggest you to use "git am" to import patches: this will keep 
track of which change comes from which commit. If you want a quick 
discussion for this matter, I'm on IRC (#gdb), by the way.

--
Pierre-Marie de Rodat