samples/bpf: Fix fout leak in hbm's run_bpf_prog
authorHao Zeng <zenghao@kylinos.cn>
Tue, 11 Apr 2023 08:43:49 +0000 (16:43 +0800)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 24 May 2023 16:32:38 +0000 (17:32 +0100)
[ Upstream commit 23acb14af1914010dd0aae1bbb7fab28bf518b8e ]

Fix fout being fopen'ed but then not subsequently fclose'd. In the affected
branch, fout is otherwise going out of scope.

Signed-off-by: Hao Zeng <zenghao@kylinos.cn>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20230411084349.1999628-1-zenghao@kylinos.cn
Signed-off-by: Sasha Levin <sashal@kernel.org>
samples/bpf/hbm.c

index 516fbac..7f89700 100644 (file)
@@ -315,6 +315,7 @@ static int run_bpf_prog(char *prog, int cg_id)
                fout = fopen(fname, "w");
                fprintf(fout, "id:%d\n", cg_id);
                fprintf(fout, "ERROR: Could not lookup queue_stats\n");
+               fclose(fout);
        } else if (stats_flag && qstats.lastPacketTime >
                   qstats.firstPacketTime) {
                long long delta_us = (qstats.lastPacketTime -