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: <20220425214559.55988956@gandalf.local.home>
Date:   Mon, 25 Apr 2022 21:45:59 -0400
From:   Steven Rostedt <rostedt@...dmis.org>
To:     Daniel Bristot de Oliveira <bristot@...nel.org>
Cc:     linux-trace-devel@...r.kernel.org, linux-kernel@...r.kernel.org,
        John Kacur <jkacur@...hat.com>, Daniel Wagner <dwagner@...e.de>
Subject: Re: [PATCH 1/2] rtla: Remove procps-ng dependency

On Fri, 22 Apr 2022 12:01:31 +0200
Daniel Bristot de Oliveira <bristot@...nel.org> wrote:


> +
> +/*
> + * procfs_is_workload_pid - check if a procfs entry contains a workload_prefix* comm
> + *
> + * Check if the procfs entry is a directory of a process, and then check if the
> + * process has a comm with the prefix set in char *workload_prefix. As the
> + * current users of this function only check for kernel threads, there is no
> + * need to check for the threads for the process.
> + *
> + * Return: True if the proc_entry contains a comm file with workload_prefix*.
> + * Otherwise returns false.
> + */
> +static int procfs_is_workload_pid(const char *workload_prefix, struct dirent *proc_entry)
> +{
> +	char comm_path[MAX_PATH], comm[MAX_PATH];

This is probably fine (but there is one issue), but I would have done this
a little different.

	int len = strlen(workload_prefix);
	char comm[len + 1];

> +	int comm_fd, retval;
> +	char *t_name;
> +
> +	if (proc_entry->d_type != DT_DIR)
> +		return 0;
> +
> +	if (*proc_entry->d_name == '.')
> +		return 0;
> +
> +	/* check if the string is a pid */
> +	for (t_name = proc_entry->d_name; t_name; t_name++) {
> +		if (!isdigit(*t_name))
> +			break;
> +	}
> +
> +	if (*t_name != '\0')
> +		return 0;
> +
> +	snprintf(comm_path, MAX_PATH, "/proc/%s/comm", proc_entry->d_name);
> +	comm_fd = open(comm_path, O_RDONLY);
> +	if (comm_fd < 0)
> +		return 0;
> +
> +	memset(comm, 0, MAX_PATH);

No need for the memset.

> +	retval = read(comm_fd, comm, MAX_PATH);

	retval = read(comm_fd, comm, len + 1);

> +
> +	close(comm_fd);
> +
> +	if (retval <= 0)
> +		return 0;

	if (comm[len] != '\n')
		return 0;

> +
> +	retval = !strncmp(workload_prefix, comm, strlen(workload_prefix));

What happens if strlen(workload_prefix) is greater than MAX_PATH? ;-)

	retval = !strncmp(workload_prefix, comm, len);

But that's me. If you want to keep this as is, let me know.

-- Steve



> +	if (!retval)
> +		return 0;
> +
> +	/* comm already have \n */
> +	debug_msg("Found workload pid:%s comm:%s", proc_entry->d_name, comm);
> +
> +	return 1;
> +}
> +

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ