[PATCH] D15245: Use a higher inlining threshold for hot callees.

[Easwaran Raman via llvm-commits]

I plan to get rid of the cold attribute in a future patch.

Thanks,
Easwaran

On Wed, Dec 9, 2015 at 11:37 PM, Justin Bogner <mail at justinbogner.com>
wrote:

> Easwaran Raman <eraman at google.com> writes:
> > Justin and Duncan, does this patch look good or do you have any concerns?
>
> It bothers me a bit that this doesn't actually remove the special cases
> in profiling code, since they still need to deal with the "cold"
> attribute. I kind of feel like there's a more general way to handle all
> of this, but I'm okay with this as an incremental step. Thanks!
>
> > On Mon, Dec 7, 2015 at 10:31 AM, Justin Bogner <mail at justinbogner.com>
> wrote:
> >> Easwaran Raman <eraman at google.com> writes:
> >>> On Fri, Dec 4, 2015 at 4:13 PM, Justin Bogner <mail at justinbogner.com>
> wrote:
> >>>> Moving this out of the frontend to here makes sense, but I have a
> couple
> >>>> of questions.
> >>>>
> >>>> - Does this only apply when we've supplied actual profile data, or can
> >>>>   we hit it based off of the heuristic counts in no-profile-data mode?
> >>>>   If the latter, this isn't quite NFC.
> >>>
> >>> It applies only when there is profile data since entry count and
> maximum
> >>> function count come only from profile data.
> >>>
> >>>> - Can the same be done for the "cold" attribute the frontend also
> >>>>   applies?
> >>>
> >>> The difference between the inlinehint and cold attributes is that
> >>> the latter is used not just by the inliner. For instance, branch
> >>> probability info. I can move it to the backend and replace all calls
> >>> to hasFnAttr(Attribute::Cold) with a wrapper that also returns true
> >>> if the count is below the threshold. Do you want that change to be
> >>> in this patch?
> >>
> >> That sounds more invasive, so it probably makes sense to consider it
> >> separately.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20151210/18fbbb35/attachment.html>