On 08/01, Jiri Olsa wrote:
>
> > Note the additional path_put() in testmod_unregister_uprobe(). Does it need
> > a separate patch or can it come with 5/5 ?
>
> I think it'd be better to have it separately, the test is already
> released.. so people might want to backport just the fix
OK, I'll rebase and add the patch below to v4. OK?
Oleg.
---
>From f6bf42015048938d826880e3bf4a318bb64a05b4 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <o...@redhat.com>
Date: Thu, 1 Aug 2024 14:21:47 +0200
Subject: [PATCH] selftests/bpf: fix uprobe.path leak in bpf_testmod
From: Jiri Olsa <olsaj...@gmail.com>
testmod_unregister_uprobe() forgets to path_put(&uprobe.path).
Signed-off-by: Jiri Olsa <olsaj...@gmail.com>
Signed-off-by: Oleg Nesterov <o...@redhat.com>
---
tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
index 86babdd6f850..55f6905de743 100644
--- a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
+++ b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
@@ -477,6 +477,7 @@ static void testmod_unregister_uprobe(void)
if (uprobe.offset) {
uprobe_unregister(d_real_inode(uprobe.path.dentry),
uprobe.offset, &uprobe.consumer);
+ path_put(&uprobe.path);
uprobe.offset = 0;
}
--
2.25.1.362.g51ebf55
