Re: [PULL v2 05/25] error: add global &error_warn destination

[Markus Armbruster]

Stefan Berger <stefanb@linux.ibm.com> writes:

> On 4/6/23 09:17, Peter Maydell wrote:
>> On Thu, 6 Apr 2023 at 14:16, Peter Maydell <peter.maydell@linaro.org> wrote:
>>>
>>> On Mon, 13 Mar 2023 at 11:47, <marcandre.lureau@redhat.com> wrote:
>>>>
>>>> From: Marc-André Lureau <marcandre.lureau@redhat.com>
>>>>
>>>> This can help debugging issues or develop, when error handling is
>>>> introduced.
>>>>
>>>> Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
>>>> Reviewed-by: Stefan Berger <stefanb@linux.ibm.com>
>>>> Message-Id: <20230221124802.4103554-6-marcandre.lureau@redhat.com>
>>>
>>> Hi; Coverity points out that this introduces a use-after-free
>>> (CID 1507493):
>>
>> ...and also CID 1508179 (same issue, just one warning about the
>> callsite in error_setv() and one about the callsite in
>> error_propagate()).
>>
>> thanks
>> -- PMM
>> 
>
> I'll be out starting tomorrow. I don't see Marc-André online.

I'll also be out starting tomorrow, until April 17.

The patch went in without my review, because I was unable to review it
in time.

As Coverity points out, it broke a design invariant.  Red "rethink this"
flag.  Can't do until I'm back.  Can't review the Stefan's proposed fix
either, because that one needs just as much thinking.

I suggest to treat &error_warn exactly like NULL for the release.

[...]