[Bug ipa/98815] Redundant free_dominance_info in cgraph_node::analyze()

[marxin at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98815

Martin Liška  changed:

   What|Removed |Added

 Status|ASSIGNED|RESOLVED
 Resolution|--- |FIXED

--- Comment #6 from Martin Liška  ---
Fixed.

[Bug ipa/98815] Redundant free_dominance_info in cgraph_node::analyze()

[cvs-commit at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98815

--- Comment #5 from CVS Commits  ---
The master branch has been updated by Martin Liska :

https://gcc.gnu.org/g:021607e12cb9c40d0859b78490f44bb3f7da5812

commit r12-23-g021607e12cb9c40d0859b78490f44bb3f7da5812
Author: Martin Liska 
Date:   Tue Jan 26 12:55:52 2021 +0100

Test simlified call in cgraph_node::analyze().

gcc/ChangeLog:

PR ipa/98815
* cgraphunit.c (cgraph_node::analyze): Remove duplicate
free_dominance_info calls.

[Bug ipa/98815] Redundant free_dominance_info in cgraph_node::analyze()

[marxin at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98815

--- Comment #4 from Martin Liška  ---
I can confirm the patch survives bootstrap and regression tests.
I'm going to send it at the beginning of the next stage1.

[Bug ipa/98815] Redundant free_dominance_info in cgraph_node::analyze()

[marxin at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98815

Martin Liška  changed:

   What|Removed |Added

   Target Milestone|--- |12.0

[Bug ipa/98815] Redundant free_dominance_info in cgraph_node::analyze()

[marxin at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98815

Martin Liška  changed:

   What|Removed |Added

 Resolution|INVALID |---
 Status|RESOLVED|ASSIGNED
   Assignee|unassigned at gcc dot gnu.org  |marxin at gcc dot 
gnu.org
   Last reconfirmed||2021-01-26
 Ever confirmed|0   |1

--- Comment #3 from Martin Liška  ---
All right, let me take a look.

[Bug ipa/98815] Redundant free_dominance_info in cgraph_node::analyze()

[fxue at os dot amperecomputing.com via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98815

--- Comment #2 from Feng Xue  ---
If we step into free_dominance_info(dir), it is a wrapper of
free_dominance_info (cfun, dir), which means it assumes a non-NULL "cfun".
Additionally, please go through calling stack of free_dominance_info():

   free_dominance_info ()
 -> if (!dom_info_available_p ())
   -> dom_info_state (fn, dir) != DOM_NONE
  ->  if (!fn->cfg)
return DOM_NONE;

This shows when "fn->cfg" is NULL, free_dominance_info() does nothing.
Above all, if conditional check "cfun && fn->cfg" is true, two calls to
free_dominance_info() in cgraph_node::analyze() are redundant, and if the check
is false, these calls are trival. So we could remove those calls in
cgraph_node::analyze(), not execute_pass_list().

[Bug ipa/98815] Redundant free_dominance_info in cgraph_node::analyze()

[marxin at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98815

Martin Liška  changed:

   What|Removed |Added

 Resolution|--- |INVALID
 Status|UNCONFIRMED |RESOLVED

--- Comment #1 from Martin Liška  ---
Thanks for the report.

It's likely more complicated as execute_pass_list calls free_dominance_info
conditionally.

  if (cfun && fn->cfg)
{
  free_dominance_info (CDI_DOMINATORS);
  free_dominance_info (CDI_POST_DOMINATORS);
}

Moreover, there are more callers of execute_pass_list than the only
cgraph_node::analyze. That said, I tend to close it as invalid.