Re: [PATCH 0/3] Fix thread exited messages for remote targets

[John Baldwin <jhb@FreeBSD.org>]

On 8/21/23 2:31 AM, Andrew Burgess via Gdb-patches wrote:
> The point of this series is to fix the thread ... exited messages for
> remote targets.  I made a previous attempt to fix this issue:
> 
>    https://inbox.sourceware.org/gdb-patches/0215c520c680334efac7f0544d834ec1fb7aff8e.1669634536.git.aburgess@redhat.com/
>    https://inbox.sourceware.org/gdb-patches/d6d49338119d2a32e060f95219aacac31bdf4886.1674207665.git.aburgess@redhat.com/
> 
> But Pedro hinted that he had a better solution as part of his step
> over clone/exit work, the latest version of which can be found here:
> 
>    https://inbox.sourceware.org/gdb-patches/20221212203101.1034916-31-pedro@palves.net/
> 
> Unfortunately, the larger series, of which that commit is a part,
> seems to have stalled.  However, that particular patch doesn't really
> depend on anything else in the series, and so, in this series I've
> pulled out this one patch (3/3), along with one other patch on which
> the above sort-of depended (1/3 - this changes a similar area of GDB).
> 
> The remaining patch (2/3) is my own work, and is an additional
> refactor which I think follows naturally from the first patch.
> 
> Patches #1 and #3 are mostly Pedro's work, though, some updates were
> needed during the rebase.  Any bugs introduced by the rebase are mine.

Thanks for this cleanup.  The asymmetry of what targets had to do here
was kind of odd.

One question I have is why is the optional metadata specifically a
ULONGEST exit code rather than just a std::string suffix?  If it were
a string then windows-nat.c could use something like:

    std::string info = string_printf ("with code %u", (unsigned) exit_code).
    delete_thread_with_info (to_del, info, silent);

-- 
John Baldwin