linux-kernel - Re: [PATCH v2 2/2] perf report: Fix memory leaks in the hierarchy mode

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

Message-ID: <403e3848770cef1d70357452082ddda0e92275fc.camel@intel.com>
Date: Fri, 7 Mar 2025 16:28:41 +0000
From: "Falcon, Thomas" <thomas.falcon@...el.com>
To: "namhyung@...nel.org" <namhyung@...nel.org>, "acme@...nel.org"
	<acme@...nel.org>, "irogers@...gle.com" <irogers@...gle.com>,
	"kan.liang@...ux.intel.com" <kan.liang@...ux.intel.com>
CC: "jolsa@...nel.org" <jolsa@...nel.org>, "peterz@...radead.org"
	<peterz@...radead.org>, "Hunter, Adrian" <adrian.hunter@...el.com>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	"linux-perf-users@...r.kernel.org" <linux-perf-users@...r.kernel.org>,
	"mingo@...nel.org" <mingo@...nel.org>
Subject: Re: [PATCH v2 2/2] perf report: Fix memory leaks in the hierarchy
 mode

On Thu, 2025-03-06 at 22:12 -0800, Namhyung Kim wrote:
> Ian told me that there are many memory leaks in the hierarchy mode. 
> I
> can easily reproduce it with the follwing command.
> 
>   $ make DEBUG=1 EXTRA_CFLAGS=-fsanitize=leak
> 
>   $ perf record --latency -g -- ./perf test -w thloop
> 
>   $ perf report -H --stdio
>   ...
>   Indirect leak of 168 byte(s) in 21 object(s) allocated from:
>       #0 0x7f3414c16c65 in malloc
> ../../../../src/libsanitizer/lsan/lsan_interceptors.cpp:75
>       #1 0x55ed3602346e in map__get util/map.h:189
>       #2 0x55ed36024cc4 in hist_entry__init util/hist.c:476
>       #3 0x55ed36025208 in hist_entry__new util/hist.c:588
>       #4 0x55ed36027c05 in hierarchy_insert_entry util/hist.c:1587
>       #5 0x55ed36027e2e in hists__hierarchy_insert_entry
> util/hist.c:1638
>       #6 0x55ed36027fa4 in hists__collapse_insert_entry
> util/hist.c:1685
>       #7 0x55ed360283e8 in hists__collapse_resort util/hist.c:1776
>       #8 0x55ed35de0323 in report__collapse_hists
> /home/namhyung/project/linux/tools/perf/builtin-report.c:735
>       #9 0x55ed35de15b4 in __cmd_report
> /home/namhyung/project/linux/tools/perf/builtin-report.c:1119
>       #10 0x55ed35de43dc in cmd_report
> /home/namhyung/project/linux/tools/perf/builtin-report.c:1867
>       #11 0x55ed35e66767 in run_builtin
> /home/namhyung/project/linux/tools/perf/perf.c:351
>       #12 0x55ed35e66a0e in handle_internal_command
> /home/namhyung/project/linux/tools/perf/perf.c:404
>       #13 0x55ed35e66b67 in run_argv
> /home/namhyung/project/linux/tools/perf/perf.c:448
>       #14 0x55ed35e66eb0 in main
> /home/namhyung/project/linux/tools/perf/perf.c:556
>       #15 0x7f340ac33d67 in __libc_start_call_main
> ../sysdeps/nptl/libc_start_call_main.h:58
>   ...
> 
>   $ perf report -H --stdio 2>&1 | grep -c '^Indirect leak'
>   93
> 
> I found that hist_entry__delete() missed to release child entries in
> the
> hierarchy tree (hroot_{in,out}).  It needs to iterate the child
> entries
> and call hist_entry__delete() recursively.
> 
> After this change:
> 
>   $ perf report -H --stdio 2>&1 | grep -c '^Indirect leak'
>   0
> 
> Reported-by: Ian Rogers <irogers@...gle.com>
> Signed-off-by: Namhyung Kim <namhyung@...nel.org>

Tested on an Arrow Lake system.

Tested-by Thomas Falcon <thomas.falcon@...el.com>

> ---
> v2) Use rbtree_postorder_for_each_entry_safe()  (Ian)
> 
>  tools/perf/util/hist.c | 10 ++++++++++
>  1 file changed, 10 insertions(+)
> 
> diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c
> index fbf131aeae7ffe9b..d65228c1141251fb 100644
> --- a/tools/perf/util/hist.c
> +++ b/tools/perf/util/hist.c
> @@ -1385,6 +1385,16 @@ void hist_entry__delete(struct hist_entry *he)
>  {
>  	struct hist_entry_ops *ops = he->ops;
>  
> +	if (symbol_conf.report_hierarchy) {
> +		struct rb_root *root = &he->hroot_out.rb_root;
> +		struct hist_entry *child, *tmp;
> +
> +		rbtree_postorder_for_each_entry_safe(child, tmp,
> root, rb_node)
> +			hist_entry__delete(child);
> +
> +		*root = RB_ROOT;
> +	}
> +
>  	thread__zput(he->thread);
>  	map_symbol__exit(&he->ms);
>