Received-SPF: pass (google.com: domain of linux-kernel+bounces-78887-linux.lists.archive=gmail.com@vger.kernel.org designates 139.178.88.99 as permitted sender) client-ip=139.178.88.99;
Sender: John Groves <grovesaustin@gmail.com>
From: John Groves <John@Groves.net>
To: John Groves <jgroves@micron.com>,
	Jonathan Corbet <corbet@lwn.net>,
	Dan Williams <dan.j.williams@intel.com>,
	Vishal Verma <vishal.l.verma@intel.com>,
	Dave Jiang <dave.jiang@intel.com>,
	Alexander Viro <viro@zeniv.linux.org.uk>,
	Christian Brauner <brauner@kernel.org>,
	Jan Kara <jack@suse.cz>,
	Matthew Wilcox <willy@infradead.org>,
	linux-cxl@vger.kernel.org,
	linux-fsdevel@vger.kernel.org,
	linux-doc@vger.kernel.org,
	linux-kernel@vger.kernel.org,
	nvdimm@lists.linux.dev
Cc: John@Groves.net,
	john@jagalactic.com,
	Dave Chinner <david@fromorbit.com>,
	Christoph Hellwig <hch@infradead.org>,
	dave.hansen@linux.intel.com,
	gregory.price@memverge.com,
	John Groves <john@groves.net>
Subject: [RFC PATCH 01/20] famfs: Documentation
Date: Fri, 23 Feb 2024 11:41:45 -0600
Message-Id: <ec08aa5a00bef714b44d61be8b3e7ed58ce7f66a.1708709155.git.john@groves.net>
In-Reply-To: <cover.1708709155.git.john@groves.net>
References: <cover.1708709155.git.john@groves.net>
Precedence: bulk
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit

Introduce Documentation/filesystems/famfs.rst into the Documentation
tree

Signed-off-by: John Groves <john@groves.net>
---
 Documentation/filesystems/famfs.rst | 124 ++++++++++++++++++++++++++++
 1 file changed, 124 insertions(+)
 create mode 100644 Documentation/filesystems/famfs.rst

diff --git a/Documentation/filesystems/famfs.rst b/Documentation/filesystems/famfs.rst
new file mode 100644
index 000000000000..c2cc50c10d03
--- /dev/null
+++ b/Documentation/filesystems/famfs.rst
@@ -0,0 +1,124 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+.. _famfs_index:
+
+==================================================================
+famfs: The kernel component of the famfs shared memory file system
+==================================================================
+
+- Copyright (C) 2024 Micron Technology, Inc.
+
+Introduction
+============
+Compute Express Link (CXL) provides a mechanism for disaggregated or
+fabric-attached memory (FAM). This creates opportunities for data sharing;
+clustered apps that would otherwise have to shard or replicate data can
+share one copy in disaggregated memory.
+
+Famfs, which is not CXL-specific in any way, provides a mechanism for
+multiple hosts to use data in shared memory, by giving it a file system
+interface. With famfs, any app that understands files (which is all of
+them, right?) can access data sets in shared memory. Although famfs
+supports read and write calls, the real point is to support mmap, which
+provides direct (dax) access to the memory - either writable or read-only.
+
+Shared memory can pose complex coherency and synchronization issues, but
+there are also simple cases. Two simple and eminently useful patterns that
+occur frequently in data analytics and AI are:
+
+* Serial Sharing - Only one host or process at a time has access to a file
+* Read-only Sharing - Multiple hosts or processes share read-only access
+  to a file
+
+The famfs kernel file system is part of the famfs framework; User space
+components [1] handle metadata allocation and distribution, and direct the
+famfs kernel module to instantiate files that map to specific memory.
+
+The famfs framework manages coherency of its own metadata and structures,
+but does not attempt to manage coherency for applications.
+
+Famfs also provides data isolation between files. That is, even though
+the host has access to an entire memory "device" (as a dax device), apps
+cannot write to memory for which the file is read-only, and mapping one
+file provides isolation from the memory of all other files. This is pretty
+basic, but some experimental shared memory usage patterns provide no such
+isolation.
+
+Principles of Operation
+=======================
+
+Without its user space components, the famfs kernel module is just a
+semi-functional clone of ramfs with latent fs-dax support. The user space
+components maintain superblocks and metadata logs, and use the famfs kernel
+component to provide a file system view of shared memory across multiple
+hosts.
+
+Each host has an independent instance of the famfs kernel module. After
+mount, files are not visible until the user space component instantiates
+them (normally by playing the famfs metadata log).
+
+Once instantiated, files on each host can point to the same shared memory,
+but in-memory metadata (inodes, etc.) is ephemeral on each host that has a
+famfs instance mounted. Like ramfs, the famfs in-kernel file system has no
+backing store for metadata modifications. If metadata is ever persisted,
+that must be done by the user space components. However, mutations to file
+data are saved to the shared memory - subject to write permission and
+processor cache behavior.
+
+
+Famfs is Not a Conventional File System
+---------------------------------------
+
+Famfs files can be accessed by conventional means, but there are
+limitations. The kernel component of famfs is not involved in the
+allocation of backing memory for files at all; the famfs user space
+creates files and passes the allocation extent lists into the kernel via
+the per-file FAMFSIOC_MAP_CREATE ioctl. A file that lacks this metadata is
+treated as invalid by the famfs kernel module. As a practical matter files
+must be created via the famfs library or cli, but they can be consumed as
+if they were conventional files.
+
+Famfs differs in some important ways from conventional file systems:
+
+* Files must be pre-allocated by the famfs framework; Allocation is never
+  performed on write.
+* Any operation that changes a file's size is considered to put the file
+  in an invalid state, disabling access to the data. It may be possible to
+  revisit this in the future.
+* (Typically the famfs user space can restore files to a valid state by
+  replaying the famfs metadata log.)
+
+Famfs exists to apply the existing file system abstractions on top of
+shared memory so applications and workflows can more easily consume it.
+
+Key Requirements
+================
+
+The primary requirements for famfs are:
+
+1. Must support a file system abstraction backed by sharable dax memory
+2. Files must efficiently handle VMA faults
+3. Must support metadata distribution in a sharable way
+4. Must handle clients with a stale copy of metadata
+
+The famfs kernel component takes care of 1-2 above.
+
+Requirements 3 and 4 are handled by the user space components, and are
+largely orthogonal to the functionality of the famfs kernel module.
+
+Requirements 3 and 4 cannot be met by conventional fs-dax file systems
+(e.g. xfs and ext4) because they use write-back metadata; it is not valid
+to mount such a file system on two hosts from the same in-memory image.
+
+
+Famfs Usage
+===========
+
+Famfs usage is documented at [1].
+
+
+References
+==========
+
+- [1] Famfs user space repository and documentation
+      https://github.com/cxl-micron-reskit/famfs
-- 
2.43.0