linux-kernel - Re: [PATCH v4 1/5] elfcore: use TASK_COMM

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: <20211013091104.4ea0e05b@gandalf.local.home>
Date:   Wed, 13 Oct 2021 09:11:04 -0400
From:   Steven Rostedt <rostedt@...dmis.org>
To:     Yafang Shao <laoar.shao@...il.com>
Cc:     keescook@...omium.org, peterz@...radead.org, pmladek@...e.com,
        viro@...iv.linux.org.uk, akpm@...ux-foundation.org,
        valentin.schneider@....com, mathieu.desnoyers@...icios.com,
        qiang.zhang@...driver.com, robdclark@...omium.org,
        christian@...uner.io, dietmar.eggemann@....com, mingo@...hat.com,
        juri.lelli@...hat.com, vincent.guittot@...aro.org,
        bsegall@...gle.com, mgorman@...e.de, bristot@...hat.com,
        davem@...emloft.net, kuba@...nel.org, linux-kernel@...r.kernel.org,
        kernel test robot <lkp@...el.com>
Subject: Re: [PATCH v4 1/5] elfcore: use TASK_COMM_LEN instead of 16 in
 prpsinfo

On Wed, 13 Oct 2021 10:23:42 +0000
Yafang Shao <laoar.shao@...il.com> wrote:

> kernel test robot reported a -Wstringop-truncation warning after I
> extend task comm from 16 to 24. Below is the detailed warning:
> 
>    fs/binfmt_elf.c: In function 'fill_psinfo.isra':
> >> fs/binfmt_elf.c:1575:9: warning: 'strncpy' output may be truncated copying 16 bytes from a string of length 23 [-Wstringop-truncation]  
>     1575 |         strncpy(psinfo->pr_fname, p->comm, sizeof(psinfo->pr_fname));
>          |         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> 
> This patch can fix this warning.
> 
> struct elf_prpsinfo was moved from include/uapi/linux/elfcore.h into
> include/linux/elfcore.h in commit
> 1e6b57d6421f ("unexport linux/elfcore.h")
> 
> As it is not UAPI code, we can replace 16 with TASK_COMM_LEN without
> worrying about breaking userspace things.
> 
> struct elf_prpsinfo is used to dump the task information in userspace
> coredump or kernel vmcore. So I verified what will happen to vmcore if
> I extend the size of TASK_COMM_LEN to 24. The result is that the vmcore
> still work fine as expected, for example:
> 
> crash> ps  
>    PID    PPID  CPU       TASK        ST  %MEM     VSZ    RSS  COMM
> >     0      0   0  ffffffff8501a940  RU   0.0       0      0  [swapper/0]
> >     0      0   1  ffff996e00f81f80  RU   0.0       0      0  [swapper/1]
> >     0      0   2  ffff996e00f80000  RU   0.0       0      0  [swapper/2]
> >     0      0   3  ffff996e00f85e80  RU   0.0       0      0  [swapper/3]
> >     0      0   4  ffff996e00f83f00  RU   0.0       0      0  [swapper/4]  
>       0      0   5  ffff996e00f8de80  RU   0.0       0      0  [swapper/5]
> >     0      0   6  ffff996e00f8bf00  RU   0.0       0      0  [swapper/6]
> >     0      0   7  ffff996e00f89f80  RU   0.0       0      0  [swapper/7]
> >     0      0   8  ffff996e00f88000  RU   0.0       0      0  [swapper/8]
> >     0      0   9  ffff996e00f93f00  RU   0.0       0      0  [swapper/9]
> >     0      0  10  ffff996e00f91f80  RU   0.0       0      0  [swapper/10]
> >     0      0  11  ffff996e00f90000  RU   0.0       0      0  [swapper/11]
> >     0      0  12  ffff996e00f95e80  RU   0.0       0      0  [swapper/12]
> >     0      0  13  ffff996e00f98000  RU   0.0       0      0  [swapper/13]
> >     0      0  14  ffff996e00f9de80  RU   0.0       0      0  [swapper/14]
> >     0      0  15  ffff996e00f9bf00  RU   0.0       0      0  [swapper/15]  
> 
> Reported-by: kernel test robot <lkp@...el.com>
> Signed-off-by: Yafang Shao <laoar.shao@...il.com>
> Cc: Peter Zijlstra <peterz@...radead.org>
> Cc: Steven Rostedt <rostedt@...dmis.org>
> Cc: Kees Cook <keescook@...omium.org>
> Cc: Al Viro <viro@...iv.linux.org.uk>
> Cc: Petr Mladek <pmladek@...e.com>
> ---
>  include/linux/elfcore-compat.h | 2 +-
>  include/linux/elfcore.h        | 4 ++--
>  2 files changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/include/linux/elfcore-compat.h b/include/linux/elfcore-compat.h
> index e272c3d452ce..8a52a782161d 100644
> --- a/include/linux/elfcore-compat.h
> +++ b/include/linux/elfcore-compat.h
> @@ -43,7 +43,7 @@ struct compat_elf_prpsinfo
>  	__compat_uid_t			pr_uid;
>  	__compat_gid_t			pr_gid;
>  	compat_pid_t			pr_pid, pr_ppid, pr_pgrp, pr_sid;
> -	char				pr_fname[16];
> +	char				pr_fname[TASK_COMM_LEN];
>  	char				pr_psargs[ELF_PRARGSZ];
>  };

Nice clean up, but should we add "#include <linux/sched.h>" to this header,
to make sure that it pulls in TASK_COMM_LEN define and not just hope it
gets pulled in beforehand by chance?

>  
> diff --git a/include/linux/elfcore.h b/include/linux/elfcore.h
> index 2aaa15779d50..ff4e4e455160 100644
> --- a/include/linux/elfcore.h
> +++ b/include/linux/elfcore.h
> @@ -65,8 +65,8 @@ struct elf_prpsinfo
>  	__kernel_gid_t	pr_gid;
>  	pid_t	pr_pid, pr_ppid, pr_pgrp, pr_sid;
>  	/* Lots missing */
> -	char	pr_fname[16];	/* filename of executable */
> -	char	pr_psargs[ELF_PRARGSZ];	/* initial part of arg list */
> +	char	pr_fname[TASK_COMM_LEN]; /* filename of executable */
> +	char	pr_psargs[ELF_PRARGSZ];	 /* initial part of arg list */

This header is fine, as it pulls in sched/task_stack.h which includes
sched.h.

-- Steve

>  };
>  
>  static inline void elf_core_copy_regs(elf_gregset_t *elfregs, struct pt_regs *regs)