[Bug c/67435] Large performance drop on apparently unrelated changes (potential cause : hot loop alignment)

[trippels at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67435

--- Comment #9 from Markus Trippelsdorf <trippels at gcc dot gnu.org> ---
(In reply to Yann Collet from comment #8)
> However, for some reason, in the table provided, both Sandy Bridge and
> Haswell get a default loop alignment value of 16. not 32.
> 
> Is there a reason for that choice ?

These values are normally strait out of the Vendors manuals.
And there are also drawbacks to high alignment values.

> Less precise but still great, having the ability to set this optimization
> parameter for a function or a section code would be great. But my experiment
> seem to show that using #pragma or __attribute__ with align-loops does not
> work, as if the optimization setting was simply ignored.

Well, there already is an aligned attribute for functions, variables and
fields,
see: 
http://gcc.gnu.org/onlinedocs/gcc/Common-Function-Attributes.html#Common-Function-Attributes