[Bug tree-optimization/98174] New: [11 Regression] Ranger takes too much memory

[Bug tree-optimization/98174] New: [11 Regression] Ranger takes too much memory

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98174

            Bug ID: 98174
           Summary: [11 Regression] Ranger takes too much memory
           Product: gcc
           Version: 11.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: rguenth at gcc dot gnu.org
  Target Milestone: ---

The testcase from PR98144 shows ranger eating 5GB memory which is excessive.

[Bug tree-optimization/98174] [11 Regression] Ranger takes too much memory

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98174

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P1
           Keywords|                            |missed-optimization
                 CC|                            |aldyh at gcc dot gnu.org
   Target Milestone|---                         |11.0

--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
It removes a workaround for excessive REE memory use during RTL checking
bootstrap (-fno-ree).  -fno-tree-vrp tames it.

[Bug tree-optimization/98174] [11 Regression] Ranger takes too much memory

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98174

--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> ---
Maybe related (didn't check yet), the testcases for PR91257 and PR93199 now OOM
at -O2

[Bug tree-optimization/98174] [11 Regression] Ranger takes too much memory

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98174

--- Comment #3 from Richard Biener <rguenth at gcc dot gnu.org> ---
(In reply to Richard Biener from comment #2)
> Maybe related (didn't check yet), the testcases for PR91257 and PR93199 now
> OOM at -O2

https://gcc.opensuse.org/gcc-old/c++bench-czerny/random/

[Bug tree-optimization/98174] [11 Regression] Ranger takes too much memory

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98174

--- Comment #4 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Andrew Macleod <amacleod@gcc.gnu.org>:

https://gcc.gnu.org/g:7f359556a772e26eabf8d31e53aae1de6f2f200d

commit r11-5921-g7f359556a772e26eabf8d31e53aae1de6f2f200d
Author: Andrew MacLeod <amacleod@redhat.com>
Date:   Thu Dec 10 14:59:14 2020 -0500

    Reduce memory requirements for ranger

    Calculate block exit info upfront, and then any SSA_NAME which is never
    used in an outgoing range calculation is a pure global and can bypass the
    on-entry cache.

            PR tree-optimization/98174
            * gimple-range-cache.cc (ranger_cache::ssa_range_in_bb): Only push
            poor values to be examined if it isn't a pure global.
            (ranger_cache::block_range): Don't process pure globals.
            (ranger_cache::fill_block_cache): Adjust has_edge_range call.
            * gimple-range-gori.cc (gori_map::all_outgoing): New bitmap.
            (gori_map::gori_map): Allocate all_outgoing.
            (gori_map::is_export_p): No specified BB returns global context.
            (gori_map::calculate_gori): Accumulate each block into global.
            (gori_compute::gori_compute): Preprocess each block for exports.
            (gori_compute::has_edge_range_p): No edge returns global context.
            * gimple-range-gori.h (has_edge_range_p): Provide default
parameter.

[Bug tree-optimization/98174] [11 Regression] Ranger takes too much memory

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98174

--- Comment #5 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Can this be closed now, or are there still testcases where Ranger is too much
compile time memory hungry?

[Bug tree-optimization/98174] [11 Regression] Ranger takes too much memory

[amacleod at redhat dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98174

Andrew Macleod <amacleod at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
             Status|UNCONFIRMED                 |RESOLVED

--- Comment #6 from Andrew Macleod <amacleod at redhat dot com> ---
I'm not aware of anything outrageous at the moment.... especially when combined
with the change from 98866.