linux-kernel - Re: [PATCH] sample: bpf: syscall_tp_user: print result of verify

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: <882349c0-123d-3deb-88e8-d400ec702d1f@fb.com>
Date:   Thu, 31 Mar 2022 20:01:52 -0700
From:   Yonghong Song <yhs@...com>
To:     Song Chen <chensong_2000@....cn>, ast@...nel.org,
        daniel@...earbox.net, andrii@...nel.org, kafai@...com,
        songliubraving@...com, john.fastabend@...il.com,
        kpsingh@...nel.org, netdev@...r.kernel.org, bpf@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] sample: bpf: syscall_tp_user: print result of verify_map



On 3/31/22 6:41 PM, Song Chen wrote:
> syscall_tp only prints the map id and messages when something goes wrong,
> but it doesn't print the value passed from bpf map. I think it's better
> to show that value to users.
> 
> What's more, i also added a 2-second sleep before calling verify_map,
> to make the value more obvious.
> 
> Signed-off-by: Song Chen <chensong_2000@....cn>
> ---
>   samples/bpf/syscall_tp_user.c | 4 ++++
>   1 file changed, 4 insertions(+)
> 
> diff --git a/samples/bpf/syscall_tp_user.c b/samples/bpf/syscall_tp_user.c
> index a0ebf1833ed3..1faa7f08054e 100644
> --- a/samples/bpf/syscall_tp_user.c
> +++ b/samples/bpf/syscall_tp_user.c
> @@ -36,6 +36,9 @@ static void verify_map(int map_id)
>   		fprintf(stderr, "failed: map #%d returns value 0\n", map_id);
>   		return;
>   	}
> +
> +	printf("verify map:%d val: %d\n", map_id, val);

I am not sure how useful it is or anybody really cares.
This is just a sample to demonstrate how bpf tracepoint works.
The error path has error print out already.

> +
>   	val = 0;
>   	if (bpf_map_update_elem(map_id, &key, &val, BPF_ANY) != 0) {
>   		fprintf(stderr, "map_update failed: %s\n", strerror(errno));
> @@ -98,6 +101,7 @@ static int test(char *filename, int num_progs)
>   	}
>   	close(fd);
>   
> +	sleep(2);

The commit message mentioned this sleep(2) is
to make the value more obvious. I don't know what does this mean.
sleep(2) can be added only if it fixed a bug.

>   	/* verify the map */
>   	for (i = 0; i < num_progs; i++) {
>   		verify_map(map0_fds[i]);