public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "jwerner at chromium dot org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug c/97434] New: Missed dead code optimization from data flow analysis
Date: Thu, 15 Oct 2020 01:38:10 +0000	[thread overview]
Message-ID: <bug-97434-4@http.gcc.gnu.org/bugzilla/> (raw)

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97434

            Bug ID: 97434
           Summary: Missed dead code optimization from data flow analysis
           Product: gcc
           Version: 8.3.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c
          Assignee: unassigned at gcc dot gnu.org
          Reporter: jwerner at chromium dot org
  Target Milestone: ---

I found a pretty simple case where GCC cannot optimize out a redundant check.
I've reduced it to the following minimal test case:


unsigned int random_number(void);                                               
void eliminate_me(void);                                                        

void main(void)                                                                 
{                                                                               
        unsigned int a = random_number();                                       
        unsigned int b = random_number();                                       

        if (b > a)                                                              
                return;                                                         

        int x = b - 8;                                                          

        if (x > 0 && x > a)                                                     
                eliminate_me();                                                 
}


I think it should be really easy to prove that eliminate_me() cannot be called,
because x can never be greater than a (otherwise b would have also been greater
than a and the function would have terminated earlier). I don't know anything
about how compilers do data flow analysis in detail, but GCC can usually figure
out so much that I'm surprised it cannot figure out this one.

             reply	other threads:[~2020-10-15  1:38 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-10-15  1:38 jwerner at chromium dot org [this message]
2020-10-15  7:32 ` [Bug tree-optimization/97434] " rguenth at gcc dot gnu.org
2020-11-21  0:48 ` amacleod at redhat dot com
2021-12-16  0:12 ` pinskia at gcc dot gnu.org

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=bug-97434-4@http.gcc.gnu.org/bugzilla/ \
    --to=gcc-bugzilla@gcc.gnu.org \
    --cc=gcc-bugs@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).