lkml.org 
[lkml]   [2015]   [Jun]   [28]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
SubjectRe: [PATCH v2 2/2] tty: check tcsetpgrp p is a process group
From
On Sun, Jun 28, 2015 at 12:07 PM, Peter Hurley <peter@hurleysoftware.com> wrote:
> On 06/27/2015 08:51 PM, Patrick Donnelly wrote:
>> This fixes a bug where a process can set the foreground process group to its
>> pid even if its pid is not a valid pgrp.
>>
>> Signed-off-by: Patrick Donnelly <batrick@batbytes.com>
>> ---
>> drivers/tty/tty_io.c | 3 +++
>> 1 file changed, 3 insertions(+)
>>
>> diff --git a/drivers/tty/tty_io.c b/drivers/tty/tty_io.c
>> index fbb55db..01b4769 100644
>> --- a/drivers/tty/tty_io.c
>> +++ b/drivers/tty/tty_io.c
>> @@ -2579,6 +2579,9 @@ static int tiocspgrp(struct tty_struct *tty, struct tty_struct *real_tty, pid_t
>> retval = -ESRCH;
>> if (!pgrp)
>> goto out_unlock;
>> + retval = -EINVAL;
>> + if (!pid_task(pgrp, PIDTYPE_PGID))
>> + goto out_unlock;
>
> This change implies that the sequence in session_of_pgrp() that specifically
> checks for pid_task(pgrp, PIDTYPE_PGID) == NULL is not doing anything
> useful. However, that hypothesis is directly contradicted by the
> comment above session_of_pgrp()
>
> "* This checks not only the pgrp, but falls back on the pid if no
> * satisfactory pgrp is found. I dunno - gdb doesn't work correctly
> * without this..."
>
> Regards,
> Peter Hurley
>
>> retval = -EPERM;
>> if (session_of_pgrp(pgrp) != task_session(current))
>> goto out_unlock;
>>

Ah, missed that. Good catch! I guess this patch is no good since it
was already accounted for and it breaks gdb.

--
Patrick Donnelly


\
 
 \ /
  Last update: 2015-06-29 03:41    [W:0.074 / U:0.292 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site