RE: memory leak in call_usermodehelper()

From: Srikumar Subramanian (SrikumarS_at_ami.com)
Date: 12/31/03

  • Next message: Andrew Morton: "Re: 2.6.1-rc1-mm1 - ieee1394 broken again?"
    To: "'arjanv@redhat.com'" <arjanv@redhat.com>, Srikumar Subramanian <SrikumarS@ami.com>
    Date:	Wed, 31 Dec 2003 14:43:18 -0500
    
    

    Basically, I am not intended to have my own syscall.

    I am generating a report on all deleted file and directory. For that I
    trapped sys_unlink() function and calling a external program using
    call_usermodehelper(). Since sys_unlink() is called very frequently in my
    case, memory leak caused by calling call_usermodehelper kills all the
    process in the system.

    Just to narrow down the problem, I introduced my own syscall.

    -----Original Message-----
    From: Arjan van de Ven [mailto:arjanv@redhat.com]
    Sent: Wednesday, December 31, 2003 7:13 AM
    To: Srikumar Subramanian
    Cc: 'Andrew Morton'; 'linux-kernel@vger.kernel.org'; Boopathi Veerappan
    Subject: RE: memory leak in call_usermodehelper()

    On Wed, 2003-12-31 at 06:22, Srikumar Subramanian wrote:

    >
    > Is there any alternative to call_usermodehelper in kernel 2.4.20?
    >
    most of all don't implement your own syscalls!

    > Hi,
    > I am using 2.4.20-8 Redhat 9 kernel.
    >

    well clearly not quite since you're adding syscalls to it.
    Also 2.4.20-8 isn't the current Red Hat Linux 9 kernel; 2.4.20-27.9 is.
    -
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.kernel.org
    More majordomo info at http://vger.kernel.org/majordomo-info.html
    Please read the FAQ at http://www.tux.org/lkml/


  • Next message: Andrew Morton: "Re: 2.6.1-rc1-mm1 - ieee1394 broken again?"

    Relevant Pages

    • Re: is killing zombies possible w/o a reboot?
      ... You couldn't rmmod smb, since it was ... I'm not going to pretend to be a kernel expert, ... is there a way to simply signal a syscall that it is to be interrupted ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • perfctr-2.7.2 released
      ... This version changes the user/kernel interface from ioctls ... Requested by Andrew Morton (2.6-mm kernel maintainer). ... The new user-space syscall() wrappers only work with certain ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: how do you call userspace syscalls (e.g. sys_rename) from inside kernel
      ... >> the kernel which calls one of the userspace system calls from inside the ... > without library support (usually that brand new syscall you added). ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: Process system call access list.
      ... > When the kernel loaded a process, it would keep a copy of the bit mask ... But in your case where the compiler gives the syscall list itself, ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: how do you call userspace syscalls (e.g. sys_rename) from inside kernel
      ... On Fri, 2004-10-08 at 14:04 +0100, Luke Kenneth Casson Leighton wrote: ... > the kernel which calls one of the userspace system calls from inside the ... without library support (usually that brand new syscall you added). ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)