linux/Documentation/filesystems/fuse-io.rst
<<
>>
Prefs
   1.. SPDX-License-Identifier: GPL-2.0
   2
   3==============
   4Fuse I/O Modes
   5==============
   6
   7Fuse supports the following I/O modes:
   8
   9- direct-io
  10- cached
  11  + write-through
  12  + writeback-cache
  13
  14The direct-io mode can be selected with the FOPEN_DIRECT_IO flag in the
  15FUSE_OPEN reply.
  16
  17In direct-io mode the page cache is completely bypassed for reads and writes.
  18No read-ahead takes place. Shared mmap is disabled.
  19
  20In cached mode reads may be satisfied from the page cache, and data may be
  21read-ahead by the kernel to fill the cache.  The cache is always kept consistent
  22after any writes to the file.  All mmap modes are supported.
  23
  24The cached mode has two sub modes controlling how writes are handled.  The
  25write-through mode is the default and is supported on all kernels.  The
  26writeback-cache mode may be selected by the FUSE_WRITEBACK_CACHE flag in the
  27FUSE_INIT reply.
  28
  29In write-through mode each write is immediately sent to userspace as one or more
  30WRITE requests, as well as updating any cached pages (and caching previously
  31uncached, but fully written pages).  No READ requests are ever sent for writes,
  32so when an uncached page is partially written, the page is discarded.
  33
  34In writeback-cache mode (enabled by the FUSE_WRITEBACK_CACHE flag) writes go to
  35the cache only, which means that the write(2) syscall can often complete very
  36fast.  Dirty pages are written back implicitly (background writeback or page
  37reclaim on memory pressure) or explicitly (invoked by close(2), fsync(2) and
  38when the last ref to the file is being released on munmap(2)).  This mode
  39assumes that all changes to the filesystem go through the FUSE kernel module
  40(size and atime/ctime/mtime attributes are kept up-to-date by the kernel), so
  41it's generally not suitable for network filesystems.  If a partial page is
  42written, then the page needs to be first read from userspace.  This means, that
  43even for files opened for O_WRONLY it is possible that READ requests will be
  44generated by the kernel.
  45