> Very well, it is obviously an improvement. The reason for the 
> current asymmetry was actually that the algorithm discarded the 
> low bits; what about fixing that as well? The improved accuracy 
> amounts to less than 1 % of difference in the return value; no 
> other code needs changing, and we get the symmetry for free. 
> Proposed patch attached.

Firstly, Mattias and Eli, thank you both for investigating and 
dealing with this bug so quickly and thoroughly.

I just noticed that the patch and commit 
7e8c1a671872ef8e45057f25912594cf548639ab to master both reference 
the bug incorrectly in test/src/xfaces-tests.el. The comment under 
the new test is "Check symmetry (bug#51455)" however the bug ID is 
41544.