[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAKYAXd8p9tmEQ8uBz6YTUfBSWsSYWTnJc0iu7e5OGshMKoBbwQ@mail.gmail.com>
Date: Wed, 28 Apr 2021 06:38:55 +0900
From: Namjae Jeon <linkinjeon@...nel.org>
To: "J. Bruce Fields" <bfields@...ldses.org>
Cc: Namjae Jeon <namjae.jeon@...sung.com>,
linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org,
smfrench@...il.com, senozhatsky@...omium.org, hyc.lee@...il.com,
viro@...iv.linux.org.uk, hch@....de, hch@...radead.org,
ronniesahlberg@...il.com, aurelien.aptel@...il.com,
aaptel@...e.com, sandeen@...deen.net, dan.carpenter@...cle.com,
colin.king@...onical.com, rdunlap@...radead.org,
willy@...radead.org
Subject: Re: [PATCH v2 00/10] cifsd: introduce new SMB3 kernel server
2021-04-28 5:53 GMT+09:00, J. Bruce Fields <bfields@...ldses.org>:
> On Thu, Apr 22, 2021 at 09:28:14AM +0900, Namjae Jeon wrote:
>> This is the patch series for cifsd(ksmbd) kernel server.
>>
>> What is cifsd(ksmbd) ?
>> ======================
>>
>> The SMB family of protocols is the most widely deployed
>> network filesystem protocol, the default on Windows and Macs (and even
>> on many phones and tablets), with clients and servers on all major
>> operating systems, but lacked a kernel server for Linux. For many
>> cases the current userspace server choices were suboptimal
>> either due to memory footprint, performance or difficulty integrating
>> well with advanced Linux features.
>>
>> ksmbd is a new kernel module which implements the server-side of the SMB3
>> protocol.
>> The target is to provide optimized performance, GPLv2 SMB server, better
>> lease handling (distributed caching). The bigger goal is to add new
>> features more rapidly (e.g. RDMA aka "smbdirect", and recent encryption
>> and signing improvements to the protocol) which are easier to develop
>> on a smaller, more tightly optimized kernel server than for example
>> in Samba. The Samba project is much broader in scope (tools, security
>> services,
>> LDAP, Active Directory Domain Controller, and a cross platform file
>> server
>> for a wider variety of purposes) but the user space file server portion
>> of Samba has proved hard to optimize for some Linux workloads, including
>> for smaller devices. This is not meant to replace Samba, but rather be
>> an extension to allow better optimizing for Linux, and will continue to
>> integrate well with Samba user space tools and libraries where
>> appropriate.
>> Working with the Samba team we have already made sure that the
>> configuration
>> files and xattrs are in a compatible format between the kernel and
>> user space server.
>>
>>
>> Architecture
>> ============
>>
>> |--- ...
>> --------|--- ksmbd/3 - Client 3
>> |-------|--- ksmbd/2 - Client 2
>> | |
>> ____________________________________________________
>> | | |- Client 1
>> |
>> <--- Socket ---|--- ksmbd/1 <<= Authentication : NTLM/NTLM2, Kerberos
>> |
>> | | | | <<= SMB engine : SMB2, SMB2.1, SMB3,
>> SMB3.0.2, |
>> | | | | SMB3.1.1
>> |
>> | | |
>> |____________________________________________________|
>> | | |
>> | | |--- VFS --- Local Filesystem
>> | |
>> KERNEL |--- ksmbd/0(forker kthread)
>> ---------------||---------------------------------------------------------------
>> USER ||
>> || communication using NETLINK
>> || ______________________________________________
>> || | |
>> ksmbd.mountd <<= DCE/RPC(srvsvc, wkssvc, samr, lsarpc) |
>> ^ | <<= configure shares setting, user accounts |
>> | |______________________________________________|
>> |
>> |------ smb.conf(config file)
>> |
>> |------ ksmbdpwd.db(user account/password file)
>> ^
>> ksmbd.adduser ---------------|
>>
>> The subset of performance related operations(open/read/write/close etc.)
>> belong
>> in kernelspace(ksmbd) and the other subset which belong to
>> operations(DCE/RPC,
>> user account/share database) which are not really related with performance
>> are
>> handled in userspace(ksmbd.mountd).
>>
>> When the ksmbd.mountd is started, It starts up a forker thread at
>> initialization
>> time and opens a dedicated port 445 for listening to SMB requests.
>> Whenever new
>> clients make request, Forker thread will accept the client connection and
>> fork
>> a new thread for dedicated communication channel between the client and
>> the server.
>
> Judging from the diagram above, all those threads are kernel threads, is
> that right? So a kernel thread gets each call first, then uses netlink
> to get help from ksmbd.mountd if necessary, is that right?
Yes, That's right.
>
> --b.
>
Powered by blists - more mailing lists