SFN+LVU+IEPM v4 (was: Re: Statement Frontier Notes, Location Views, and Inlined Entry Point Markers)

[Alexandre Oliva <aoliva at redhat dot com>]

On Sep 30, 2017, Alexandre Oliva <aoliva@redhat.com> wrote:

> On Aug 31, 2017, Alexandre Oliva <aoliva@redhat.com> wrote:
>> On Aug 23, 2017, Richard Biener <richard.guenther@gmail.com> wrote:
>>> Just separating the boilerplate changes out from the "meat" of the change
>>> into a separate patch for easier reviewing would be nice.

>> I've broken up the patch into a patchset with 10 patches.  I've already
>> posted the one that makes -g options negatable the other day;

> That one was approved and checked in.

>> I'll post the other 9 as a follow up to this message.

> I've refreshed and retested the others.  There was a conflict in
> find_bb_boundaries, and the changes for compare-debug made there seem to
> have obviated what was in the patchset, so I dropped mine.

As a followup to this message, I'll post an updated, fixed and refreshed
patchset.

The main changes to this patchset, aside from trivial conclift
resolution (e.g. SDB removal, various instances of code movement), are:

- the reversal of the final.c API changes that had required changes to
  many targets; they API changes are now kept internal to final.c

- the changes to gcc/cp/constprop.c requested by jason

- more comments before block_within_block_p and its caller, and an
  additional argument to decide whether to execute the test both ways

- new macros to create RTL patterns for markers, so that all of the
  decisions about their representation are at a single place

- stabilize find_bb_boundaries, fixes -fcompare-debug errors

Most of the patchset is already approved, part by richi, part by jeff.
The only points still requiring changes, AFAIK, are the ones mentioned
above.  So, I'd appreciate if richi, jeff and jason would let me know
whether these changes are enough to earn their approval.  Of course, if
anyone would like to raise any other issues and concerns, I'll do my
best to address them.

Nevertheless, for reference, I post the entire patchset.  The
incremental changes, aside from the trivial ones made in merges, can be
identified for inspection with git log trunk..aoliva/SFN (that's the
branch where I'm making incremental changes and merges, unlike
aoliva/SLI, where the patchset is being maintained with rebased
consolidated changes).  If there's interest in my posting the
incremental changes, I'd be glad to post them as well, just let me know.



There are a couple of compare-debug errors I'm aware of, but that are
preexisting problems unrelated with this patchset, so I'm not delaying
the posting of the patchset further so as to include fixes for those.
(they come up in unusual testing scenarios; I've already posted on Oct 5
to gcc-patches about adding different compiler flags to debug info in
different stages; the other shows up in a -fcompare-debug run of the
libstdc++-v3 testsuite, and it's an artifact of dumping rtl for
-fcompare-debug rather than an actual codegen difference)

-- 
Alexandre Oliva, freedom fighter    http://FSFLA.org/~lxoliva/
You must be the change you wish to see in the world. -- Gandhi
Be Free! -- http://FSFLA.org/   FSF Latin America board member
Free Software Evangelist|Red Hat Brasil GNU Toolchain Engineer