linux-kernel - Re: [PATCH] objtool: prevent memory leak in error paths

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <YH26xMfn04KATRow@kernel.org>
Date:   Mon, 19 Apr 2021 14:15:48 -0300
From:   Arnaldo Carvalho de Melo <acme@...nel.org>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     Dan Carpenter <dan.carpenter@...cle.com>,
        Muhammad Usama Anjum <musamaanjum@...il.com>,
        Josh Poimboeuf <jpoimboe@...hat.com>,
        Ingo Molnar <mingo@...nel.org>,
        Miroslav Benes <mbenes@...e.cz>, Borislav Petkov <bp@...e.de>,
        open list <linux-kernel@...r.kernel.org>,
        kernel-janitors@...r.kernel.org, colin.king@...onical.com
Subject: Re: [PATCH] objtool: prevent memory leak in error paths

Em Thu, Apr 15, 2021 at 09:24:41AM +0200, Peter Zijlstra escreveu:
> On Wed, Apr 14, 2021 at 11:47:09AM +0300, Dan Carpenter wrote:
> > On Wed, Apr 14, 2021 at 01:45:11AM +0500, Muhammad Usama Anjum wrote:
> > > Memory allocated by sym and sym->name isn't being freed if some error
> > > occurs in elf_create_undef_symbol(). Free the sym and sym->name if error
> > > is detected before returning NULL.
> > > 
> > > Addresses-Coverity: ("Prevent memory leak")
> > > Fixes: 2f2f7e47f052 ("objtool: Add elf_create_undef_symbol()")
> > > Signed-off-by: Muhammad Usama Anjum <musamaanjum@...il.com>
> > > ---
> > > Only build has been tested.
> > > 
> > 
> > Just ignore leaks from the tools/ directory.  These things run and then
> > exit and all the memory is freed.  #OldSchoolGarbageCollector
> 
> Mostly true; but I suspect tools/perf might care, it has some longer
> running things in.

Yes, and now we have 'perf daemon' that is long running.

- Arnaldo