[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <48ACF9E8.6020203@cs.columbia.edu>
Date: Thu, 21 Aug 2008 01:15:20 -0400
From: Oren Laadan <orenl@...columbia.edu>
To: dave@...ux.vnet.ibm.com
CC: arnd@...db.de, jeremy@...p.org, linux-kernel@...r.kernel.org,
containers@...ts.linux-foundation.org
Subject: Re: [RFC v2][PATCH 1/9] kernel based checkpoint-restart
Subject line should have been 's/PATCH 1/PATCH 0/' ...
I left cr_debug() in for now to provide more info than pr_debug();
eventually that will be changed back to pr_debug()
In the mini-conference we considered doing CR in a kernel module,
but decided against because we needed a system call. It is still
possible to put the bulk of the code in a module. This is useful,
besides reducing debug time (recompile, unload, reload), to reduce
the kernel memory footprint. Also, an administrator can load/unload
the module to enable/disable this feature. Any thoughts ?
Oren.
Oren Laadan wrote:
>
> These patches implement checkpoint-restart [CR v2]. This version adds
> save and restore of open files state (regular files and directories)
> which makes it more usable. Other changes address the feedback given
> for the previous version. It is also refactored (along Dave's posting)
> for easier reviewing.
>
> Todo:
> - Add support for x86-64 and improve ABI
> - Refine or change syscall interface
> - Extend to handle (multiple) tasks in a container
> - Security (without CAPS_SYS_ADMIN files restore may fail)
>
> Changelog:
>
> [2008-Aug-20] v2:
> - Added dump and restore of open files (regular and directories);
> see the changes in the test program (ckpt.c)
> - Added basic handling of shared objects, and use 'parent tag'
> - Added documentation
> - Improved ABI, add 64bit padding for image data
> - Improved locking when saving/restoring memory
> - Added UTS information to header (release, version, machine)
> - Cleanup extraction of filename from a file pointer
> - Refactor to allow easier reviewing
> - Remove requirement for CAPS_SYS_ADMIN until we come up with a
> security policy (this means that file restore may fail)
> - Other cleanup in response to comments for v1
>
> [2008-Jul-29] v1:
> - Initial version: support a single task with address space of only
> private anonymous or file-mapped VMAs; syscalls ignore pid/crid
> argument and act on current process.
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists