public inbox for gcc-cvs@sourceware.org
help / color / mirror / Atom feed
From: Pierre-Marie de Rodat <pmderodat@gcc.gnu.org>
To: gcc-cvs@gcc.gnu.org
Subject: [gcc r12-1478] [Ada] Avoid unhelpful dead code warnings
Date: Tue, 15 Jun 2021 10:23:04 +0000 (GMT)	[thread overview]
Message-ID: <20210615102304.2EDE83840027@sourceware.org> (raw)

https://gcc.gnu.org/g:ba998f6407ba006a8f8055171ee8e8510f0bc46b

commit r12-1478-gba998f6407ba006a8f8055171ee8e8510f0bc46b
Author: Steve Baird <baird@adacore.com>
Date:   Fri Feb 12 16:34:04 2021 -0800

    [Ada] Avoid unhelpful dead code warnings
    
    gcc/ada/
    
            * exp_util.adb (Kill_Dead_Code): Generalize the existing
            handling of if statements to handle case statements similarly.

Diff:
---
 gcc/ada/exp_util.adb | 39 ++++++++++++++++++++++-----------------
 1 file changed, 22 insertions(+), 17 deletions(-)

diff --git a/gcc/ada/exp_util.adb b/gcc/ada/exp_util.adb
index 78797f64165..860525d12a1 100644
--- a/gcc/ada/exp_util.adb
+++ b/gcc/ada/exp_util.adb
@@ -9238,28 +9238,33 @@ package body Exp_Util is
          if W then
 
             --  We suppress the warning if this code is under control of an
-            --  if statement, whose condition is a simple identifier, and
-            --  either we are in an instance, or warnings off is set for this
-            --  identifier. The reason for killing it in the instance case is
-            --  that it is common and reasonable for code to be deleted in
-            --  instances for various reasons.
+            --  if/case statement and either
+            --    a) we are in an instance and the condition/selector
+            --       has a statically known value; or
+            --    b) the condition/selector is a simple identifier and
+            --       warnings off is set for this identifier.
+            --  Dead code is common and reasonable in instances, so we don't
+            --  want a warning in that case.
 
-            --  Could we use Is_Statically_Unevaluated here???
+            declare
+               C : Node_Id := Empty;
+            begin
+               if Nkind (Parent (N)) = N_If_Statement then
+                  C := Condition (Parent (N));
+               elsif Nkind (Parent (N)) = N_Case_Statement_Alternative then
+                  C := Expression (Parent (Parent (N)));
+               end if;
 
-            if Nkind (Parent (N)) = N_If_Statement then
-               declare
-                  C : constant Node_Id := Condition (Parent (N));
-               begin
-                  if Nkind (C) = N_Identifier
-                    and then
-                      (In_Instance
-                        or else (Present (Entity (C))
-                                  and then Has_Warnings_Off (Entity (C))))
+               if Present (C) then
+                  if (In_Instance and Compile_Time_Known_Value (C))
+                    or else (Nkind (C) = N_Identifier
+                             and then Present (Entity (C))
+                             and then Has_Warnings_Off (Entity (C)))
                   then
                      W := False;
                   end if;
-               end;
-            end if;
+               end if;
+            end;
 
             --  Generate warning if not suppressed


                 reply	other threads:[~2021-06-15 10:23 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20210615102304.2EDE83840027@sourceware.org \
    --to=pmderodat@gcc.gnu.org \
    --cc=gcc-cvs@gcc.gnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).