Re: [PATCH 1/2] perf kvm: Fix memory leak in process_sample_event()

[Namhyung Kim]

Hello,

On Fri, Aug 07, 2026 at 05:08:35PM +0200, Michalis Niarchos via B4 Relay wrote:
> From: Michalis Niarchos <michael.niarchos@xxxxxxxxx>
> 
> machine__resolve() indirectly acquires a thread reference via
> machine__findnew_thread(). Release it, as suggested by the documentation
> of the former.
> 
> Signed-off-by: Michalis Niarchos <michael.niarchos@xxxxxxxxx>
> ---
>  tools/perf/builtin-kvm.c | 7 +++++--
>  1 file changed, 5 insertions(+), 2 deletions(-)
> 
> diff --git a/tools/perf/builtin-kvm.c b/tools/perf/builtin-kvm.c
> index 00baafb05603..8e7bc7b2db18 100644
> --- a/tools/perf/builtin-kvm.c
> +++ b/tools/perf/builtin-kvm.c
> @@ -1145,7 +1145,7 @@ static int process_sample_event(const struct perf_tool *tool,
>  	if (machine__resolve(machine, &kvm->al, sample) < 0) {
>  		pr_warning("WARNING: at offset %#" PRIx64 ": fail to resolve address location, skipping sample\n",
>  			   sample->file_offset);
> -		return 0;
> +		goto out;

I think it's ok to return here when machine__resolve() returns -1.
I'll remove this hunk.

Thanks,
Namhyung


>  	}
>  
>  	thread = machine__findnew_thread(machine, sample->pid, sample->tid);
> @@ -1153,13 +1153,16 @@ static int process_sample_event(const struct perf_tool *tool,
>  		pr_debug("problem processing %s (%u) event at offset %#" PRIx64 ", skipping it.\n",
>  			 perf_event__name(event->header.type), event->header.type,
>  			 sample->file_offset);
> -		return -1;
> +		err = -1;
> +		goto out;
>  	}
>  
>  	if (!handle_kvm_event(kvm, thread, sample))
>  		err = -1;
>  
>  	thread__put(thread);
> +out:
> +	addr_location__exit(&kvm->al);
>  	return err;
>  }
>  
> 
> -- 
> 2.55.0
> 
>