>> It's not clear to me what you want: do you want to fix this issue on 
>> the emacs-29 branch or on the master branch?
>
> I suggested to install your fix on master.
>

Okay, but see below.

>> We now have two bug reports on this (the one by Frédéric Giquel in this 
>> bug, and the one by Eason Huang in bug#64989).  I just verified (again) 
>> that the Fix-combine-change-call-with-timestamps patch fixes these two 
>> recipes.
>
> If bug#64989 could be fixed by a simpler, safer change (which remains to 
> be seen), we might do that on emacs-29.  If not, at least we will have 
> this fixed in Emacs 30.
>

Bug#64989 is the same bug, only the recipe differs.  And the change I 
suggested six months ago is safe, it replaces an obscure and bugged piece 
of code with a refactored and clear piece of code.  I attach it again 
(that patch is for emacs-29, the one for master would be similar).

If Stefan (or anyone else) fixes the bug in a safer way, that patch is not 
necessary.