[PATCH] Set behavior of sprintf-like functions with overlapping source and destination

[Siddhesh Poyarekar]

On 27/12/18 10:29 PM, Zack Weinberg wrote:
> On Thu, Dec 27, 2018 at 8:46 AM Gabriel F. T. Gomes
> <gabriel@inconstante.eti.br> wrote:
>> On Thu, Dec 27, 2018 at 11:15:06AM +0530, Siddhesh Poyarekar wrote:
>>> Thanks, to clarify, is your position that we revert to old behaviour (for
>>> now) for the default case only or for everything, including _FORTIFY_SOURCE?
>>
>> The _FORTIFY_SOURCE version did not change behaviour...  Even before
>> Zack's patches, it would overwrite overlapping buffers.  This means that
>> this patch, as it is today, reverts all behavior to what it used to be.
> 
> Yes.  My position right now is that we should merge Gabriel's patch as
> is -- including the test case! -- for 2.29, and consider whether we
> want to make any further changes after the release.

OK thaks for the clarification.  Paul, do you have a strong opposition 
to this or is it OK if we take this course for this release?

Siddhesh