Change the test case. Thank you.

Fix the test case. Thank you, Quentin.

Add the requested comment. Thank you for reviewing this.

Kindly Ping

Add one more AssertingVH

Add more AssertingVH.

Address Eli's comments. Thank you.

Address Eli's comments. Thank you very much, Eli.

Address Alexey's comments. Thank you.

Address Alexey's comments.

Check the users of Ext now.

Rebased the test case. Thank you for taking a look.

Kindly Ping

Kindly Ping

Fixed a segment fault bug with a new test case, fixed an existing test case, and fixed the format.

Update to address Serguei's comments. Thank you for taking a look.

Thank you, Piotr.

Rebase the test case.

Updated to address Alexey's latest comments. Thank you again.

Rebase and update the patch.

Update to address Alexey's comments. Thank you very much.

Abandon in favor of D44868.

Matt, what do you think?

Address Jun's comments.

Kindly Ping #3

LGTM

Kindly Ping #2

Kindly ping. I am appreciated to take any advice.

Made some changes to improve the readability.

• Completely remove the check of base types. So, any large data structure is supported.

I made two changes to address the comments.

LGTM

Hi Keno,

Now I only move the callsites whose callees also call the callers to the end of the inline list to delay the creation of recursive functions. Is it more acceptable?

Kindly Ping (#3)

Thank you very much, Easwaran. The new code looks much simpler. I will commit soon.

Thank you, Chad. I will wait for a couple of days before I commit this patch in case I will get some new comments.

Kindly Ping (#3)

Kindly Ping (#2).

Thank you, Jun.

Kindly Ping.

Kindly Ping (#2)

I ran the spec2006/17 performance again on top of the ToT. Here is the number:

Address Matt's comments. Thank you.

Kindly Ping (#8)

Kindly Ping.

Kindly Ping

In D40097#932995, @davidxl wrote:

Using heuristic like this for inline order decision is like tossing a coin. It is very likely that doing this can hurt some cases where inlining of inner edges are important but gets blocked.

Due to current practical limitation in the inliner such as lack of the ability to inline self recursive functions, this patch can help to workaround that limitation a little, so the patch looks fine for now, though we should not depend strongly on this inline behavior in the future.

Kindly Ping

Kindly Ping (#7)

Address most of Matt's comments. Thank you.

It occurs to me that the need for this might be removed by another potential improvement (suggested to me by Chandler at EuroLLVM this year): We can make CGSCC analysis wrapper for our current analysis passes in order to allow these analyses to look back through function arguments into the function's callers. To make this work, we'd:

1. Make CGSCC analysis passes corresponding to our current analysis passes. For example, AA, ValueTracking, etc. (I realize that ValueTracking is not really an analysis pass right now, and while we might want to change that at some point, I don't think it matters for this description).
2. These pass accept analysis queries and produce results by looking at the calling functions and calling the function/local analysis on the values at each call site, and then intersecting the results. If there are too many call sites, or unknown call sites, then you need to give up.
3. Provide these CGSCC analysis handles to the corresponding local analyses so that, as they do a recursive analysis, when they hit arguments, then can call into the CGSCC analyses to continue the analysis into the callers.

If we did that, do you think we'd still need this transformation?

Address Easwaran's comments. Thank you very much.