Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Fri, 2 Jul 2021 14:57:51 -0300
From:   Arnaldo Carvalho de Melo <acme@kernel.org>
To:     Leo Yan <leo.yan@linaro.org>
Cc:     gushengxian <gushengxian507419@gmail.com>, will@kernel.org,
        mathieu.poirier@linaro.org, peterz@infradead.org, mingo@redhat.com,
        mark.rutland@arm.com, alexander.shishkin@linux.intel.com,
        jolsa@redhat.com, namhyung@kernel.org,
        linux-arm-kernel@lists.infradead.org,
        linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org,
        gushengxian <gushengxian@yulong.com>
Subject: Re: [PATCH] perf: tests: fix some mmemory leak issues
Message-ID: <YN9TnyUMd0VesywP@kernel.org>
References: <20210702040955.631618-1-gushengxian507419@gmail.com>
 <YN8NjJLHoB+wzLXa@kernel.org>
 <20210702135641.GA22592@leoy-ThinkPad-X240s>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20210702135641.GA22592@leoy-ThinkPad-X240s>
Precedence: bulk

Em Fri, Jul 02, 2021 at 09:56:41PM +0800, Leo Yan escreveu:
> Hi Arnaldo,
> 
> On Fri, Jul 02, 2021 at 09:58:52AM -0300, Arnaldo Carvalho de Melo wrote:
> > Em Thu, Jul 01, 2021 at 09:09:55PM -0700, gushengxian escreveu:
> > > From: gushengxian <gushengxian@yulong.com>
> > > 
> > > Some memory leak issues should be fixed by free().
> > > Reported by cppcheck.
> 
> I don't think this patch does the right thing.  You could see that the
> memory is allocated in arch specific function sample_ustack(), and the
> "buf" pointer is assigned to sample->user_stack.data; and the memory
> actually is released in the caller function test_dwarf_unwind__thread:
> 
> noinline int test_dwarf_unwind__thread(struct thread *thread)
> {
>         struct perf_sample sample;
>         unsigned long cnt = 0;
>         int err = -1;
> 
>         memset(&sample, 0, sizeof(sample));
> 
>         if (test__arch_unwind_sample(&sample, thread)) {
>                 pr_debug("failed to get unwind sample\n");
>                 goto out;
>         } 
> 
>         [...]
> 
>  out:
>         zfree(&sample.user_stack.data);
>         zfree(&sample.user_regs.regs);
>         return err;
> }
> 
> So this patch will break the testing and doesn't fix any memory leak
> issue.

You are right, those buffers are allocated in those functions and then
_returned_ via sample->user_stack, the tool (and myself, ugh) got
fooled, I'll remove that from my local tree.

Thanks Leo!

- Arnaldo