[Bug analyzer/122975] [13/14/15 Regression] -Wanalyzer-infinite-recursion false positive involving integer values

2026-01-14 Thread cvs-commit at gcc dot gnu.org via Gcc-bugs 
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122975

--- Comment #6 from GCC Commits  ---
The releases/gcc-15 branch has been updated by David Malcolm
:

https://gcc.gnu.org/g:e869244a134c84ef9f5ba15d8005bf66f9f2048c

commit r15-10673-ge869244a134c84ef9f5ba15d8005bf66f9f2048c
Author: David Malcolm 
Date:   Mon Dec 15 11:48:49 2025 -0500

analyzer: fold X + (-X) to zero [PR122975]

gcc/analyzer/ChangeLog:
PR analyzer/122975
* region-model-manager.cc (region_model_manager::maybe_fold_binop):
Fold
X + (-X) to zero.

gcc/testsuite/ChangeLog:
PR analyzer/122975
* c-c++-common/analyzer/arith-1.c: New test.
* c-c++-common/analyzer/infinite-recursion-pr122975.c: New test.

Signed-off-by: David Malcolm 

(cherry picked from commit c48970fc4a4d1747a50e6abd1634757212420c04)

[Bug analyzer/122975] [13/14/15 Regression] -Wanalyzer-infinite-recursion false positive involving integer values

2025-12-29 Thread fabio.porcedda at gmail dot com via Gcc-bugs 
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122975

--- Comment #5 from Fabio Porcedda  ---
(In reply to David Malcolm from comment #3)
> Should be fixed on trunk for gcc 16 by the above patch.
> 
> Keeping open to track backporting.

Hi David,
thanks for the fix.
I've tried the fix and i confirm that fixes the test.c but it is still present
with a slightly different code, please check test2.c

test2.c: In function ‘fun’:
test2.c:4:9: warning: infinite recursion [CWE-674]
[-Wanalyzer-infinite-recursion]
4 | fun(x, 8 - x);
  | ^
  ‘fun’: events 1-4
│
│1 | void fun(int x, int y)
│  |  ^~~
│  |  |
│  |  (1) entry to ‘fun’
│2 | {
│3 | if (x + y > 8)
│  |~
│  ||
│  |(2) following ‘true’ branch... ─>─┐
│  |  │
│  |  │
│  |┌─┘
│4 |│fun(x, 8 - x);
│  |│~
│  |│|
│  |└───>(3) ...to here
│  | (4) calling ‘fun’ from ‘fun’
│
└──> ‘fun’: events 5-8
   │
   │1 | void fun(int x, int y)
   │  |  ^~~
   │  |  |
   │  |  (5) initial entry to ‘fun’
   │2 | {
   │3 | if (x + y > 8)
   │  |~
   │  ||
   │  |(6) following ‘true’ branch... ─>─┐
   │  |  │
   │  |  │
   │  |┌─┘
   │4 |│fun(x, 8 - x);
   │  |│~
   │  |│|
   │  |└───>(7) ...to here
   │  | (8) calling ‘fun’ from ‘fun’
   │
   └──> ‘fun’: events 9-10
  │
  │1 | void fun(int x, int y)
  │  |  ^~~
  │  |  |
  │  |  (9) recursive entry to ‘fun’; previously
entered at (5)
  │  |  (10) ⚠️  apparently infinite recursion
  │

Could you please fix this as well?

Thanks

[Bug analyzer/122975] [13/14/15 Regression] -Wanalyzer-infinite-recursion false positive involving integer values

2025-12-29 Thread fabio.porcedda at gmail dot com via Gcc-bugs 
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122975

--- Comment #4 from Fabio Porcedda  ---
Created attachment 63180
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=63180&action=edit
test2.c test to reproduce the issue even after test.c pass

[Bug analyzer/122975] [13/14/15 Regression] -Wanalyzer-infinite-recursion false positive involving integer values

2025-12-15 Thread dmalcolm at gcc dot gnu.org via Gcc-bugs 
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122975

David Malcolm  changed:

   What|Removed |Added

Summary|[13/14/15/16 Regression]|[13/14/15 Regression]
   |-Wanalyzer-infinite-recursi |-Wanalyzer-infinite-recursi
   |on false positive involving |on false positive involving
   |integer values  |integer values

--- Comment #3 from David Malcolm  ---
Should be fixed on trunk for gcc 16 by the above patch.

Keeping open to track backporting.