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] [day] [month] [year] [list]
Message-ID: <CAFZWReYF83JPvnqqVPgw88UBwQwJkeDk0wMTAeOdWDe-xoYtzA@mail.gmail.com>
Date:   Mon, 13 Jan 2020 17:35:56 +0530
From:   Uday R <opensource.linuz@...il.com>
To:     linux-kernel@...r.kernel.org
Subject: Re: Issue with Kexec memory allocation

Any help ?

Thanks,
RK

On Mon, Jan 6, 2020 at 12:43 PM Uday R <opensource.linuz@...il.com> wrote:
>
> Hello folks,
>
> I am using kexec to load into the new kernel from the existing kernel.
> My vmlinuz image size is 6.6MB, initrd size is 18MB, and the root file
> system size is 715MB.  I use kexec command to load the new kernel into
> memory. I observed that before calling kexec if the available memory
> is less than 1.33GB then kexec lands into an out of memory issue
> effectively kexec being killed by OOM-KILLER.  I dont get into out of
> memory issue if i have memory >=1.33GB.
>
> My question is why kexec tries to over allocate memory when the
> available memory is less than 1.33GB and gets the available memory to
> ZERO.
> In case if the available memory is >= 1.33GB before calling kexec,
> After kexec load, i see the memory consumed by kexec load is
> approximately equal to size of "linux+initrd+root file system".
>
> Is there a bug in the 'Kexec' code and had been fixed ?
>
> Note: i am using the kernel from a vendor, Just wanted to know if
> there is no such issue in mainstream code.
>
> Thanks in Advance,
> Uday

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ