1===============
   2Pathname lookup
   3===============
   4
   5This write-up is based on three articles published at lwn.net:
   6
   7- <https://lwn.net/Articles/649115/> Pathname lookup in Linux
   8- <https://lwn.net/Articles/649729/> RCU-walk: faster pathname lookup in Linux
   9- <https://lwn.net/Articles/650786/> A walk among the symlinks
  10
  11Written by Neil Brown with help from Al Viro and Jon Corbet.
  12It has subsequently been updated to reflect changes in the kernel
  13including:
  14
  15- per-directory parallel name lookup.
 
  16
  17Introduction to pathname lookup
  18===============================
  19
  20The most obvious aspect of pathname lookup, which very little
  21exploration is needed to discover, is that it is complex.  There are
  22many rules, special cases, and implementation alternatives that all
  23combine to confuse the unwary reader.  Computer science has long been
  24acquainted with such complexity and has tools to help manage it.  One
  25tool that we will make extensive use of is "divide and conquer".  For
  26the early parts of the analysis we will divide off symlinks - leaving
  27them until the final part.  Well before we get to symlinks we have
  28another major division based on the VFS's approach to locking which
  29will allow us to review "REF-walk" and "RCU-walk" separately.  But we
  30are getting ahead of ourselves.  There are some important low level
  31distinctions we need to clarify first.
  32
  33There are two sorts of ...
  34--------------------------
  35
  36.. _openat: http://man7.org/linux/man-pages/man2/openat.2.html
  37
  38Pathnames (sometimes "file names"), used to identify objects in the
  39filesystem, will be familiar to most readers.  They contain two sorts
  40of elements: "slashes" that are sequences of one or more "``/``"
  41characters, and "components" that are sequences of one or more
  42non-"``/``" characters.  These form two kinds of paths.  Those that
  43start with slashes are "absolute" and start from the filesystem root.
  44The others are "relative" and start from the current directory, or
  45from some other location specified by a file descriptor given to a
  46"``XXXat``" system call such as `openat() <openat_>`_.
  47
  48.. _execveat: http://man7.org/linux/man-pages/man2/execveat.2.html
  49
  50It is tempting to describe the second kind as starting with a
  51component, but that isn't always accurate: a pathname can lack both
  52slashes and components, it can be empty, in other words.  This is
  53generally forbidden in POSIX, but some of those "xxx``at``" system calls
  54in Linux permit it when the ``AT_EMPTY_PATH`` flag is given.  For
  55example, if you have an open file descriptor on an executable file you
  56can execute it by calling `execveat() <execveat_>`_ passing
  57the file descriptor, an empty path, and the ``AT_EMPTY_PATH`` flag.
  58
  59These paths can be divided into two sections: the final component and
  60everything else.  The "everything else" is the easy bit.  In all cases
  61it must identify a directory that already exists, otherwise an error
  62such as ``ENOENT`` or ``ENOTDIR`` will be reported.
  63
  64The final component is not so simple.  Not only do different system
  65calls interpret it quite differently (e.g. some create it, some do
  66not), but it might not even exist: neither the empty pathname nor the
  67pathname that is just slashes have a final component.  If it does
  68exist, it could be "``.``" or "``..``" which are handled quite differently
  69from other components.
  70
  71.. _POSIX: http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap04.html#tag_04_12
  72
  73If a pathname ends with a slash, such as "``/tmp/foo/``" it might be
  74tempting to consider that to have an empty final component.  In many
  75ways that would lead to correct results, but not always.  In
  76particular, ``mkdir()`` and ``rmdir()`` each create or remove a directory named
  77by the final component, and they are required to work with pathnames
  78ending in "``/``".  According to POSIX_
  79
  80  A pathname that contains at least one non- &lt;slash> character and
  81  that ends with one or more trailing &lt;slash> characters shall not
  82  be resolved successfully unless the last pathname component before
  83  the trailing <slash> characters names an existing directory or a
  84  directory entry that is to be created for a directory immediately
  85  after the pathname is resolved.
  86
  87The Linux pathname walking code (mostly in ``fs/namei.c``) deals with
  88all of these issues: breaking the path into components, handling the
  89"everything else" quite separately from the final component, and
  90checking that the trailing slash is not used where it isn't
  91permitted.  It also addresses the important issue of concurrent
  92access.
  93
  94While one process is looking up a pathname, another might be making
  95changes that affect that lookup.  One fairly extreme case is that if
  96"a/b" were renamed to "a/c/b" while another process were looking up
  97"a/b/..", that process might successfully resolve on "a/c".
  98Most races are much more subtle, and a big part of the task of
  99pathname lookup is to prevent them from having damaging effects.  Many
 100of the possible races are seen most clearly in the context of the
 101"dcache" and an understanding of that is central to understanding
 102pathname lookup.
 103
 104More than just a cache
 105----------------------
 106
 107The "dcache" caches information about names in each filesystem to
 108make them quickly available for lookup.  Each entry (known as a
 109"dentry") contains three significant fields: a component name, a
 110pointer to a parent dentry, and a pointer to the "inode" which
 111contains further information about the object in that parent with
 112the given name.  The inode pointer can be ``NULL`` indicating that the
 113name doesn't exist in the parent.  While there can be linkage in the
 114dentry of a directory to the dentries of the children, that linkage is
 115not used for pathname lookup, and so will not be considered here.
 116
 117The dcache has a number of uses apart from accelerating lookup.  One
 118that will be particularly relevant is that it is closely integrated
 119with the mount table that records which filesystem is mounted where.
 120What the mount table actually stores is which dentry is mounted on top
 121of which other dentry.
 122
 123When considering the dcache, we have another of our "two types"
 124distinctions: there are two types of filesystems.
 125
 126Some filesystems ensure that the information in the dcache is always
 127completely accurate (though not necessarily complete).  This can allow
 128the VFS to determine if a particular file does or doesn't exist
 129without checking with the filesystem, and means that the VFS can
 130protect the filesystem against certain races and other problems.
 131These are typically "local" filesystems such as ext3, XFS, and Btrfs.
 132
 133Other filesystems don't provide that guarantee because they cannot.
 134These are typically filesystems that are shared across a network,
 135whether remote filesystems like NFS and 9P, or cluster filesystems
 136like ocfs2 or cephfs.  These filesystems allow the VFS to revalidate
 137cached information, and must provide their own protection against
 138awkward races.  The VFS can detect these filesystems by the
 139``DCACHE_OP_REVALIDATE`` flag being set in the dentry.
 140
 141REF-walk: simple concurrency management with refcounts and spinlocks
 142--------------------------------------------------------------------
 143
 144With all of those divisions carefully classified, we can now start
 145looking at the actual process of walking along a path.  In particular
 146we will start with the handling of the "everything else" part of a
 147pathname, and focus on the "REF-walk" approach to concurrency
 148management.  This code is found in the ``link_path_walk()`` function, if
 149you ignore all the places that only run when "``LOOKUP_RCU``"
 150(indicating the use of RCU-walk) is set.
 151
 152.. _Meet the Lockers: https://lwn.net/Articles/453685/
 153
 154REF-walk is fairly heavy-handed with locks and reference counts.  Not
 155as heavy-handed as in the old "big kernel lock" days, but certainly not
 156afraid of taking a lock when one is needed.  It uses a variety of
 157different concurrency controls.  A background understanding of the
 158various primitives is assumed, or can be gleaned from elsewhere such
 159as in `Meet the Lockers`_.
 160
 161The locking mechanisms used by REF-walk include:
 162
 163dentry->d_lockref
 164~~~~~~~~~~~~~~~~~
 165
 166This uses the lockref primitive to provide both a spinlock and a
 167reference count.  The special-sauce of this primitive is that the
 168conceptual sequence "lock; inc_ref; unlock;" can often be performed
 169with a single atomic memory operation.
 170
 171Holding a reference on a dentry ensures that the dentry won't suddenly
 172be freed and used for something else, so the values in various fields
 173will behave as expected.  It also protects the ``->d_inode`` reference
 174to the inode to some extent.
 175
 176The association between a dentry and its inode is fairly permanent.
 177For example, when a file is renamed, the dentry and inode move
 178together to the new location.  When a file is created the dentry will
 179initially be negative (i.e. ``d_inode`` is ``NULL``), and will be assigned
 180to the new inode as part of the act of creation.
 181
 182When a file is deleted, this can be reflected in the cache either by
 183setting ``d_inode`` to ``NULL``, or by removing it from the hash table
 184(described shortly) used to look up the name in the parent directory.
 185If the dentry is still in use the second option is used as it is
 186perfectly legal to keep using an open file after it has been deleted
 187and having the dentry around helps.  If the dentry is not otherwise in
 188use (i.e. if the refcount in ``d_lockref`` is one), only then will
 189``d_inode`` be set to ``NULL``.  Doing it this way is more efficient for a
 190very common case.
 191
 192So as long as a counted reference is held to a dentry, a non-``NULL`` ``->d_inode``
 193value will never be changed.
 194
 195dentry->d_lock
 196~~~~~~~~~~~~~~
 197
 198``d_lock`` is a synonym for the spinlock that is part of ``d_lockref`` above.
 199For our purposes, holding this lock protects against the dentry being
 200renamed or unlinked.  In particular, its parent (``d_parent``), and its
 201name (``d_name``) cannot be changed, and it cannot be removed from the
 202dentry hash table.
 203
 204When looking for a name in a directory, REF-walk takes ``d_lock`` on
 205each candidate dentry that it finds in the hash table and then checks
 206that the parent and name are correct.  So it doesn't lock the parent
 207while searching in the cache; it only locks children.
 208
 209When looking for the parent for a given name (to handle "``..``"),
 210REF-walk can take ``d_lock`` to get a stable reference to ``d_parent``,
 211but it first tries a more lightweight approach.  As seen in
 212``dget_parent()``, if a reference can be claimed on the parent, and if
 213subsequently ``d_parent`` can be seen to have not changed, then there is
 214no need to actually take the lock on the child.
 215
 216rename_lock
 217~~~~~~~~~~~
 218
 219Looking up a given name in a given directory involves computing a hash
 220from the two values (the name and the dentry of the directory),
 221accessing that slot in a hash table, and searching the linked list
 222that is found there.
 223
 224When a dentry is renamed, the name and the parent dentry can both
 225change so the hash will almost certainly change too.  This would move the
 226dentry to a different chain in the hash table.  If a filename search
 227happened to be looking at a dentry that was moved in this way,
 228it might end up continuing the search down the wrong chain,
 229and so miss out on part of the correct chain.
 230
 231The name-lookup process (``d_lookup()``) does _not_ try to prevent this
 232from happening, but only to detect when it happens.
 233``rename_lock`` is a seqlock that is updated whenever any dentry is
 234renamed.  If ``d_lookup`` finds that a rename happened while it
 235unsuccessfully scanned a chain in the hash table, it simply tries
 236again.
 237
 
 
 
 
 
 
 
 238inode->i_rwsem
 239~~~~~~~~~~~~~~
 240
 241``i_rwsem`` is a read/write semaphore that serializes all changes to a particular
 242directory.  This ensures that, for example, an ``unlink()`` and a ``rename()``
 243cannot both happen at the same time.  It also keeps the directory
 244stable while the filesystem is asked to look up a name that is not
 245currently in the dcache or, optionally, when the list of entries in a
 246directory is being retrieved with ``readdir()``.
 247
 248This has a complementary role to that of ``d_lock``: ``i_rwsem`` on a
 249directory protects all of the names in that directory, while ``d_lock``
 250on a name protects just one name in a directory.  Most changes to the
 251dcache hold ``i_rwsem`` on the relevant directory inode and briefly take
 252``d_lock`` on one or more the dentries while the change happens.  One
 253exception is when idle dentries are removed from the dcache due to
 254memory pressure.  This uses ``d_lock``, but ``i_rwsem`` plays no role.
 255
 256The semaphore affects pathname lookup in two distinct ways.  Firstly it
 257prevents changes during lookup of a name in a directory.  ``walk_component()`` uses
 258``lookup_fast()`` first which, in turn, checks to see if the name is in the cache,
 259using only ``d_lock`` locking.  If the name isn't found, then ``walk_component()``
 260falls back to ``lookup_slow()`` which takes a shared lock on ``i_rwsem``, checks again that
 261the name isn't in the cache, and then calls in to the filesystem to get a
 262definitive answer.  A new dentry will be added to the cache regardless of
 263the result.
 264
 265Secondly, when pathname lookup reaches the final component, it will
 266sometimes need to take an exclusive lock on ``i_rwsem`` before performing the last lookup so
 267that the required exclusion can be achieved.  How path lookup chooses
 268to take, or not take, ``i_rwsem`` is one of the
 269issues addressed in a subsequent section.
 270
 271If two threads attempt to look up the same name at the same time - a
 272name that is not yet in the dcache - the shared lock on ``i_rwsem`` will
 273not prevent them both adding new dentries with the same name.  As this
 274would result in confusion an extra level of interlocking is used,
 275based around a secondary hash table (``in_lookup_hashtable``) and a
 276per-dentry flag bit (``DCACHE_PAR_LOOKUP``).
 277
 278To add a new dentry to the cache while only holding a shared lock on
 279``i_rwsem``, a thread must call ``d_alloc_parallel()``.  This allocates a
 280dentry, stores the required name and parent in it, checks if there
 281is already a matching dentry in the primary or secondary hash
 282tables, and if not, stores the newly allocated dentry in the secondary
 283hash table, with ``DCACHE_PAR_LOOKUP`` set.
 284
 285If a matching dentry was found in the primary hash table then that is
 286returned and the caller can know that it lost a race with some other
 287thread adding the entry.  If no matching dentry is found in either
 288cache, the newly allocated dentry is returned and the caller can
 289detect this from the presence of ``DCACHE_PAR_LOOKUP``.  In this case it
 290knows that it has won any race and now is responsible for asking the
 291filesystem to perform the lookup and find the matching inode.  When
 292the lookup is complete, it must call ``d_lookup_done()`` which clears
 293the flag and does some other house keeping, including removing the
 294dentry from the secondary hash table - it will normally have been
 295added to the primary hash table already.  Note that a ``struct
 296waitqueue_head`` is passed to ``d_alloc_parallel()``, and
 297``d_lookup_done()`` must be called while this ``waitqueue_head`` is still
 298in scope.
 299
 300If a matching dentry is found in the secondary hash table,
 301``d_alloc_parallel()`` has a little more work to do. It first waits for
 302``DCACHE_PAR_LOOKUP`` to be cleared, using a wait_queue that was passed
 303to the instance of ``d_alloc_parallel()`` that won the race and that
 304will be woken by the call to ``d_lookup_done()``.  It then checks to see
 305if the dentry has now been added to the primary hash table.  If it
 306has, the dentry is returned and the caller just sees that it lost any
 307race.  If it hasn't been added to the primary hash table, the most
 308likely explanation is that some other dentry was added instead using
 309``d_splice_alias()``.  In any case, ``d_alloc_parallel()`` repeats all the
 310look ups from the start and will normally return something from the
 311primary hash table.
 312
 313mnt->mnt_count
 314~~~~~~~~~~~~~~
 315
 316``mnt_count`` is a per-CPU reference counter on "``mount``" structures.
 317Per-CPU here means that incrementing the count is cheap as it only
 318uses CPU-local memory, but checking if the count is zero is expensive as
 319it needs to check with every CPU.  Taking a ``mnt_count`` reference
 320prevents the mount structure from disappearing as the result of regular
 321unmount operations, but does not prevent a "lazy" unmount.  So holding
 322``mnt_count`` doesn't ensure that the mount remains in the namespace and,
 323in particular, doesn't stabilize the link to the mounted-on dentry.  It
 324does, however, ensure that the ``mount`` data structure remains coherent,
 325and it provides a reference to the root dentry of the mounted
 326filesystem.  So a reference through ``->mnt_count`` provides a stable
 327reference to the mounted dentry, but not the mounted-on dentry.
 328
 329mount_lock
 330~~~~~~~~~~
 331
 332``mount_lock`` is a global seqlock, a bit like ``rename_lock``.  It can be used to
 333check if any change has been made to any mount points.
 334
 335While walking down the tree (away from the root) this lock is used when
 336crossing a mount point to check that the crossing was safe.  That is,
 337the value in the seqlock is read, then the code finds the mount that
 338is mounted on the current directory, if there is one, and increments
 339the ``mnt_count``.  Finally the value in ``mount_lock`` is checked against
 340the old value.  If there is no change, then the crossing was safe.  If there
 341was a change, the ``mnt_count`` is decremented and the whole process is
 342retried.
 343
 344When walking up the tree (towards the root) by following a ".." link,
 345a little more care is needed.  In this case the seqlock (which
 346contains both a counter and a spinlock) is fully locked to prevent
 347any changes to any mount points while stepping up.  This locking is
 348needed to stabilize the link to the mounted-on dentry, which the
 349refcount on the mount itself doesn't ensure.
 350
 
 
 
 
 
 
 
 351RCU
 352~~~
 353
 354Finally the global (but extremely lightweight) RCU read lock is held
 355from time to time to ensure certain data structures don't get freed
 356unexpectedly.
 357
 358In particular it is held while scanning chains in the dcache hash
 359table, and the mount point hash table.
 360
 361Bringing it together with ``struct nameidata``
 362----------------------------------------------
 363
 364.. _First edition Unix: http://minnie.tuhs.org/cgi-bin/utree.pl?file=V1/u2.s
 365
 366Throughout the process of walking a path, the current status is stored
 367in a ``struct nameidata``, "namei" being the traditional name - dating
 368all the way back to `First Edition Unix`_ - of the function that
 369converts a "name" to an "inode".  ``struct nameidata`` contains (among
 370other fields):
 371
 372``struct path path``
 373~~~~~~~~~~~~~~~~~~~~
 374
 375A ``path`` contains a ``struct vfsmount`` (which is
 376embedded in a ``struct mount``) and a ``struct dentry``.  Together these
 377record the current status of the walk.  They start out referring to the
 378starting point (the current working directory, the root directory, or some other
 379directory identified by a file descriptor), and are updated on each
 380step.  A reference through ``d_lockref`` and ``mnt_count`` is always
 381held.
 382
 383``struct qstr last``
 384~~~~~~~~~~~~~~~~~~~~
 385
 386This is a string together with a length (i.e. _not_ ``nul`` terminated)
 387that is the "next" component in the pathname.
 388
 389``int last_type``
 390~~~~~~~~~~~~~~~~~
 391
 392This is one of ``LAST_NORM``, ``LAST_ROOT``, ``LAST_DOT``, ``LAST_DOTDOT``, or
 393``LAST_BIND``.  The ``last`` field is only valid if the type is
 394``LAST_NORM``.  ``LAST_BIND`` is used when following a symlink and no
 395components of the symlink have been processed yet.  Others should be
 396fairly self-explanatory.
 397
 398``struct path root``
 399~~~~~~~~~~~~~~~~~~~~
 400
 401This is used to hold a reference to the effective root of the
 402filesystem.  Often that reference won't be needed, so this field is
 403only assigned the first time it is used, or when a non-standard root
 404is requested.  Keeping a reference in the ``nameidata`` ensures that
 405only one root is in effect for the entire path walk, even if it races
 406with a ``chroot()`` system call.
 407
 
 
 
 
 408The root is needed when either of two conditions holds: (1) either the
 409pathname or a symbolic link starts with a "'/'", or (2) a "``..``"
 410component is being handled, since "``..``" from the root must always stay
 411at the root.  The value used is usually the current root directory of
 412the calling process.  An alternate root can be provided as when
 413``sysctl()`` calls ``file_open_root()``, and when NFSv4 or Btrfs call
 414``mount_subtree()``.  In each case a pathname is being looked up in a very
 415specific part of the filesystem, and the lookup must not be allowed to
 416escape that subtree.  It works a bit like a local ``chroot()``.
 417
 418Ignoring the handling of symbolic links, we can now describe the
 419"``link_path_walk()``" function, which handles the lookup of everything
 420except the final component as:
 421
 422   Given a path (``name``) and a nameidata structure (``nd``), check that the
 423   current directory has execute permission and then advance ``name``
 424   over one component while updating ``last_type`` and ``last``.  If that
 425   was the final component, then return, otherwise call
 426   ``walk_component()`` and repeat from the top.
 427
 428``walk_component()`` is even easier.  If the component is ``LAST_DOTS``,
 429it calls ``handle_dots()`` which does the necessary locking as already
 430described.  If it finds a ``LAST_NORM`` component it first calls
 431"``lookup_fast()``" which only looks in the dcache, but will ask the
 432filesystem to revalidate the result if it is that sort of filesystem.
 433If that doesn't get a good result, it calls "``lookup_slow()``" which
 434takes ``i_rwsem``, rechecks the cache, and then asks the filesystem
 435to find a definitive answer.  Each of these will call
 436``follow_managed()`` (as described below) to handle any mount points.
 437
 438In the absence of symbolic links, ``walk_component()`` creates a new
 439``struct path`` containing a counted reference to the new dentry and a
 440reference to the new ``vfsmount`` which is only counted if it is
 441different from the previous ``vfsmount``.  It then calls
 442``path_to_nameidata()`` to install the new ``struct path`` in the
 443``struct nameidata`` and drop the unneeded references.
 444
 445This "hand-over-hand" sequencing of getting a reference to the new
 446dentry before dropping the reference to the previous dentry may
 447seem obvious, but is worth pointing out so that we will recognize its
 448analogue in the "RCU-walk" version.
 449
 450Handling the final component
 451----------------------------
 452
 453``link_path_walk()`` only walks as far as setting ``nd->last`` and
 454``nd->last_type`` to refer to the final component of the path.  It does
 455not call ``walk_component()`` that last time.  Handling that final
 456component remains for the caller to sort out. Those callers are
 457``path_lookupat()``, ``path_parentat()``, ``path_mountpoint()`` and
 458``path_openat()`` each of which handles the differing requirements of
 459different system calls.
 460
 461``path_parentat()`` is clearly the simplest - it just wraps a little bit
 462of housekeeping around ``link_path_walk()`` and returns the parent
 463directory and final component to the caller.  The caller will be either
 464aiming to create a name (via ``filename_create()``) or remove or rename
 465a name (in which case ``user_path_parent()`` is used).  They will use
 466``i_rwsem`` to exclude other changes while they validate and then
 467perform their operation.
 468
 469``path_lookupat()`` is nearly as simple - it is used when an existing
 470object is wanted such as by ``stat()`` or ``chmod()``.  It essentially just
 471calls ``walk_component()`` on the final component through a call to
 472``lookup_last()``.  ``path_lookupat()`` returns just the final dentry.
 473
 474``path_mountpoint()`` handles the special case of unmounting which must
 475not try to revalidate the mounted filesystem.  It effectively
 476contains, through a call to ``mountpoint_last()``, an alternate
 477implementation of ``lookup_slow()`` which skips that step.  This is
 478important when unmounting a filesystem that is inaccessible, such as
 479one provided by a dead NFS server.
 480
 481Finally ``path_openat()`` is used for the ``open()`` system call; it
 482contains, in support functions starting with "``do_last()``", all the
 483complexity needed to handle the different subtleties of O_CREAT (with
 484or without O_EXCL), final "``/``" characters, and trailing symbolic
 485links.  We will revisit this in the final part of this series, which
 486focuses on those symbolic links.  "``do_last()``" will sometimes, but
 487not always, take ``i_rwsem``, depending on what it finds.
 488
 489Each of these, or the functions which call them, need to be alert to
 490the possibility that the final component is not ``LAST_NORM``.  If the
 491goal of the lookup is to create something, then any value for
 492``last_type`` other than ``LAST_NORM`` will result in an error.  For
 493example if ``path_parentat()`` reports ``LAST_DOTDOT``, then the caller
 494won't try to create that name.  They also check for trailing slashes
 495by testing ``last.name[last.len]``.  If there is any character beyond
 496the final component, it must be a trailing slash.
 497
 498Revalidation and automounts
 499---------------------------
 500
 501Apart from symbolic links, there are only two parts of the "REF-walk"
 502process not yet covered.  One is the handling of stale cache entries
 503and the other is automounts.
 504
 505On filesystems that require it, the lookup routines will call the
 506``->d_revalidate()`` dentry method to ensure that the cached information
 507is current.  This will often confirm validity or update a few details
 508from a server.  In some cases it may find that there has been change
 509further up the path and that something that was thought to be valid
 510previously isn't really.  When this happens the lookup of the whole
 511path is aborted and retried with the "``LOOKUP_REVAL``" flag set.  This
 512forces revalidation to be more thorough.  We will see more details of
 513this retry process in the next article.
 514
 515Automount points are locations in the filesystem where an attempt to
 516lookup a name can trigger changes to how that lookup should be
 517handled, in particular by mounting a filesystem there.  These are
 518covered in greater detail in autofs.txt in the Linux documentation
 519tree, but a few notes specifically related to path lookup are in order
 520here.
 521
 522The Linux VFS has a concept of "managed" dentries which is reflected
 523in function names such as "``follow_managed()``".  There are three
 524potentially interesting things about these dentries corresponding
 525to three different flags that might be set in ``dentry->d_flags``:
 526
 527``DCACHE_MANAGE_TRANSIT``
 528~~~~~~~~~~~~~~~~~~~~~~~~~
 529
 530If this flag has been set, then the filesystem has requested that the
 531``d_manage()`` dentry operation be called before handling any possible
 532mount point.  This can perform two particular services:
 533
 534It can block to avoid races.  If an automount point is being
 535unmounted, the ``d_manage()`` function will usually wait for that
 536process to complete before letting the new lookup proceed and possibly
 537trigger a new automount.
 538
 539It can selectively allow only some processes to transit through a
 540mount point.  When a server process is managing automounts, it may
 541need to access a directory without triggering normal automount
 542processing.  That server process can identify itself to the ``autofs``
 543filesystem, which will then give it a special pass through
 544``d_manage()`` by returning ``-EISDIR``.
 545
 546``DCACHE_MOUNTED``
 547~~~~~~~~~~~~~~~~~~
 548
 549This flag is set on every dentry that is mounted on.  As Linux
 550supports multiple filesystem namespaces, it is possible that the
 551dentry may not be mounted on in *this* namespace, just in some
 552other.  So this flag is seen as a hint, not a promise.
 553
 554If this flag is set, and ``d_manage()`` didn't return ``-EISDIR``,
 555``lookup_mnt()`` is called to examine the mount hash table (honoring the
 556``mount_lock`` described earlier) and possibly return a new ``vfsmount``
 557and a new ``dentry`` (both with counted references).
 558
 559``DCACHE_NEED_AUTOMOUNT``
 560~~~~~~~~~~~~~~~~~~~~~~~~~
 561
 562If ``d_manage()`` allowed us to get this far, and ``lookup_mnt()`` didn't
 563find a mount point, then this flag causes the ``d_automount()`` dentry
 564operation to be called.
 565
 566The ``d_automount()`` operation can be arbitrarily complex and may
 567communicate with server processes etc. but it should ultimately either
 568report that there was an error, that there was nothing to mount, or
 569should provide an updated ``struct path`` with new ``dentry`` and ``vfsmount``.
 570
 571In the latter case, ``finish_automount()`` will be called to safely
 572install the new mount point into the mount table.
 573
 574There is no new locking of import here and it is important that no
 575locks (only counted references) are held over this processing due to
 576the very real possibility of extended delays.
 577This will become more important next time when we examine RCU-walk
 578which is particularly sensitive to delays.
 579
 580RCU-walk - faster pathname lookup in Linux
 581==========================================
 582
 583RCU-walk is another algorithm for performing pathname lookup in Linux.
 584It is in many ways similar to REF-walk and the two share quite a bit
 585of code.  The significant difference in RCU-walk is how it allows for
 586the possibility of concurrent access.
 587
 588We noted that REF-walk is complex because there are numerous details
 589and special cases.  RCU-walk reduces this complexity by simply
 590refusing to handle a number of cases -- it instead falls back to
 591REF-walk.  The difficulty with RCU-walk comes from a different
 592direction: unfamiliarity.  The locking rules when depending on RCU are
 593quite different from traditional locking, so we will spend a little extra
 594time when we come to those.
 595
 596Clear demarcation of roles
 597--------------------------
 598
 599The easiest way to manage concurrency is to forcibly stop any other
 600thread from changing the data structures that a given thread is
 601looking at.  In cases where no other thread would even think of
 602changing the data and lots of different threads want to read at the
 603same time, this can be very costly.  Even when using locks that permit
 604multiple concurrent readers, the simple act of updating the count of
 605the number of current readers can impose an unwanted cost.  So the
 606goal when reading a shared data structure that no other process is
 607changing is to avoid writing anything to memory at all.  Take no
 608locks, increment no counts, leave no footprints.
 609
 610The REF-walk mechanism already described certainly doesn't follow this
 611principle, but then it is really designed to work when there may well
 612be other threads modifying the data.  RCU-walk, in contrast, is
 613designed for the common situation where there are lots of frequent
 614readers and only occasional writers.  This may not be common in all
 615parts of the filesystem tree, but in many parts it will be.  For the
 616other parts it is important that RCU-walk can quickly fall back to
 617using REF-walk.
 618
 619Pathname lookup always starts in RCU-walk mode but only remains there
 620as long as what it is looking for is in the cache and is stable.  It
 621dances lightly down the cached filesystem image, leaving no footprints
 622and carefully watching where it is, to be sure it doesn't trip.  If it
 623notices that something has changed or is changing, or if something
 624isn't in the cache, then it tries to stop gracefully and switch to
 625REF-walk.
 626
 627This stopping requires getting a counted reference on the current
 628``vfsmount`` and ``dentry``, and ensuring that these are still valid -
 629that a path walk with REF-walk would have found the same entries.
 630This is an invariant that RCU-walk must guarantee.  It can only make
 631decisions, such as selecting the next step, that are decisions which
 632REF-walk could also have made if it were walking down the tree at the
 633same time.  If the graceful stop succeeds, the rest of the path is
 634processed with the reliable, if slightly sluggish, REF-walk.  If
 635RCU-walk finds it cannot stop gracefully, it simply gives up and
 636restarts from the top with REF-walk.
 637
 638This pattern of "try RCU-walk, if that fails try REF-walk" can be
 639clearly seen in functions like ``filename_lookup()``,
 640``filename_parentat()``, ``filename_mountpoint()``,
 641``do_filp_open()``, and ``do_file_open_root()``.  These five
 642correspond roughly to the four ``path_``* functions we met earlier,
 643each of which calls ``link_path_walk()``.  The ``path_*`` functions are
 644called using different mode flags until a mode is found which works.
 645They are first called with ``LOOKUP_RCU`` set to request "RCU-walk".  If
 646that fails with the error ``ECHILD`` they are called again with no
 647special flag to request "REF-walk".  If either of those report the
 648error ``ESTALE`` a final attempt is made with ``LOOKUP_REVAL`` set (and no
 649``LOOKUP_RCU``) to ensure that entries found in the cache are forcibly
 650revalidated - normally entries are only revalidated if the filesystem
 651determines that they are too old to trust.
 652
 653The ``LOOKUP_RCU`` attempt may drop that flag internally and switch to
 654REF-walk, but will never then try to switch back to RCU-walk.  Places
 655that trip up RCU-walk are much more likely to be near the leaves and
 656so it is very unlikely that there will be much, if any, benefit from
 657switching back.
 658
 659RCU and seqlocks: fast and light
 660--------------------------------
 661
 662RCU is, unsurprisingly, critical to RCU-walk mode.  The
 663``rcu_read_lock()`` is held for the entire time that RCU-walk is walking
 664down a path.  The particular guarantee it provides is that the key
 665data structures - dentries, inodes, super_blocks, and mounts - will
 666not be freed while the lock is held.  They might be unlinked or
 667invalidated in one way or another, but the memory will not be
 668repurposed so values in various fields will still be meaningful.  This
 669is the only guarantee that RCU provides; everything else is done using
 670seqlocks.
 671
 672As we saw above, REF-walk holds a counted reference to the current
 673dentry and the current vfsmount, and does not release those references
 674before taking references to the "next" dentry or vfsmount.  It also
 675sometimes takes the ``d_lock`` spinlock.  These references and locks are
 676taken to prevent certain changes from happening.  RCU-walk must not
 677take those references or locks and so cannot prevent such changes.
 678Instead, it checks to see if a change has been made, and aborts or
 679retries if it has.
 680
 681To preserve the invariant mentioned above (that RCU-walk may only make
 682decisions that REF-walk could have made), it must make the checks at
 683or near the same places that REF-walk holds the references.  So, when
 684REF-walk increments a reference count or takes a spinlock, RCU-walk
 685samples the status of a seqlock using ``read_seqcount_begin()`` or a
 686similar function.  When REF-walk decrements the count or drops the
 687lock, RCU-walk checks if the sampled status is still valid using
 688``read_seqcount_retry()`` or similar.
 689
 690However, there is a little bit more to seqlocks than that.  If
 691RCU-walk accesses two different fields in a seqlock-protected
 692structure, or accesses the same field twice, there is no a priori
 693guarantee of any consistency between those accesses.  When consistency
 694is needed - which it usually is - RCU-walk must take a copy and then
 695use ``read_seqcount_retry()`` to validate that copy.
 696
 697``read_seqcount_retry()`` not only checks the sequence number, but also
 698imposes a memory barrier so that no memory-read instruction from
 699*before* the call can be delayed until *after* the call, either by the
 700CPU or by the compiler.  A simple example of this can be seen in
 701``slow_dentry_cmp()`` which, for filesystems which do not use simple
 702byte-wise name equality, calls into the filesystem to compare a name
 703against a dentry.  The length and name pointer are copied into local
 704variables, then ``read_seqcount_retry()`` is called to confirm the two
 705are consistent, and only then is ``->d_compare()`` called.  When
 706standard filename comparison is used, ``dentry_cmp()`` is called
 707instead.  Notably it does _not_ use ``read_seqcount_retry()``, but
 708instead has a large comment explaining why the consistency guarantee
 709isn't necessary.  A subsequent ``read_seqcount_retry()`` will be
 710sufficient to catch any problem that could occur at this point.
 711
 712With that little refresher on seqlocks out of the way we can look at
 713the bigger picture of how RCU-walk uses seqlocks.
 714
 715``mount_lock`` and ``nd->m_seq``
 716~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 717
 718We already met the ``mount_lock`` seqlock when REF-walk used it to
 719ensure that crossing a mount point is performed safely.  RCU-walk uses
 720it for that too, but for quite a bit more.
 721
 722Instead of taking a counted reference to each ``vfsmount`` as it
 723descends the tree, RCU-walk samples the state of ``mount_lock`` at the
 724start of the walk and stores this initial sequence number in the
 725``struct nameidata`` in the ``m_seq`` field.  This one lock and one
 726sequence number are used to validate all accesses to all ``vfsmounts``,
 727and all mount point crossings.  As changes to the mount table are
 728relatively rare, it is reasonable to fall back on REF-walk any time
 729that any "mount" or "unmount" happens.
 730
 731``m_seq`` is checked (using ``read_seqretry()``) at the end of an RCU-walk
 732sequence, whether switching to REF-walk for the rest of the path or
 733when the end of the path is reached.  It is also checked when stepping
 734down over a mount point (in ``__follow_mount_rcu()``) or up (in
 735``follow_dotdot_rcu()``).  If it is ever found to have changed, the
 736whole RCU-walk sequence is aborted and the path is processed again by
 737REF-walk.
 738
 739If RCU-walk finds that ``mount_lock`` hasn't changed then it can be sure
 740that, had REF-walk taken counted references on each vfsmount, the
 741results would have been the same.  This ensures the invariant holds,
 742at least for vfsmount structures.
 743
 744``dentry->d_seq`` and ``nd->seq``
 745~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 746
 747In place of taking a count or lock on ``d_reflock``, RCU-walk samples
 748the per-dentry ``d_seq`` seqlock, and stores the sequence number in the
 749``seq`` field of the nameidata structure, so ``nd->seq`` should always be
 750the current sequence number of ``nd->dentry``.  This number needs to be
 751revalidated after copying, and before using, the name, parent, or
 752inode of the dentry.
 753
 754The handling of the name we have already looked at, and the parent is
 755only accessed in ``follow_dotdot_rcu()`` which fairly trivially follows
 756the required pattern, though it does so for three different cases.
 757
 758When not at a mount point, ``d_parent`` is followed and its ``d_seq`` is
 759collected.  When we are at a mount point, we instead follow the
 760``mnt->mnt_mountpoint`` link to get a new dentry and collect its
 761``d_seq``.  Then, after finally finding a ``d_parent`` to follow, we must
 762check if we have landed on a mount point and, if so, must find that
 763mount point and follow the ``mnt->mnt_root`` link.  This would imply a
 764somewhat unusual, but certainly possible, circumstance where the
 765starting point of the path lookup was in part of the filesystem that
 766was mounted on, and so not visible from the root.
 767
 768The inode pointer, stored in ``->d_inode``, is a little more
 769interesting.  The inode will always need to be accessed at least
 770twice, once to determine if it is NULL and once to verify access
 771permissions.  Symlink handling requires a validated inode pointer too.
 772Rather than revalidating on each access, a copy is made on the first
 773access and it is stored in the ``inode`` field of ``nameidata`` from where
 774it can be safely accessed without further validation.
 775
 776``lookup_fast()`` is the only lookup routine that is used in RCU-mode,
 777``lookup_slow()`` being too slow and requiring locks.  It is in
 778``lookup_fast()`` that we find the important "hand over hand" tracking
 779of the current dentry.
 780
 781The current ``dentry`` and current ``seq`` number are passed to
 782``__d_lookup_rcu()`` which, on success, returns a new ``dentry`` and a
 783new ``seq`` number.  ``lookup_fast()`` then copies the inode pointer and
 784revalidates the new ``seq`` number.  It then validates the old ``dentry``
 785with the old ``seq`` number one last time and only then continues.  This
 786process of getting the ``seq`` number of the new dentry and then
 787checking the ``seq`` number of the old exactly mirrors the process of
 788getting a counted reference to the new dentry before dropping that for
 789the old dentry which we saw in REF-walk.
 790
 791No ``inode->i_rwsem`` or even ``rename_lock``
 792~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 793
 794A semaphore is a fairly heavyweight lock that can only be taken when it is
 795permissible to sleep.  As ``rcu_read_lock()`` forbids sleeping,
 796``inode->i_rwsem`` plays no role in RCU-walk.  If some other thread does
 797take ``i_rwsem`` and modifies the directory in a way that RCU-walk needs
 798to notice, the result will be either that RCU-walk fails to find the
 799dentry that it is looking for, or it will find a dentry which
 800``read_seqretry()`` won't validate.  In either case it will drop down to
 801REF-walk mode which can take whatever locks are needed.
 802
 803Though ``rename_lock`` could be used by RCU-walk as it doesn't require
 804any sleeping, RCU-walk doesn't bother.  REF-walk uses ``rename_lock`` to
 805protect against the possibility of hash chains in the dcache changing
 806while they are being searched.  This can result in failing to find
 807something that actually is there.  When RCU-walk fails to find
 808something in the dentry cache, whether it is really there or not, it
 809already drops down to REF-walk and tries again with appropriate
 810locking.  This neatly handles all cases, so adding extra checks on
 811rename_lock would bring no significant value.
 812
 813``unlazy walk()`` and ``complete_walk()``
 814-----------------------------------------
 815
 816That "dropping down to REF-walk" typically involves a call to
 817``unlazy_walk()``, so named because "RCU-walk" is also sometimes
 818referred to as "lazy walk".  ``unlazy_walk()`` is called when
 819following the path down to the current vfsmount/dentry pair seems to
 820have proceeded successfully, but the next step is problematic.  This
 821can happen if the next name cannot be found in the dcache, if
 822permission checking or name revalidation couldn't be achieved while
 823the ``rcu_read_lock()`` is held (which forbids sleeping), if an
 824automount point is found, or in a couple of cases involving symlinks.
 825It is also called from ``complete_walk()`` when the lookup has reached
 826the final component, or the very end of the path, depending on which
 827particular flavor of lookup is used.
 828
 829Other reasons for dropping out of RCU-walk that do not trigger a call
 830to ``unlazy_walk()`` are when some inconsistency is found that cannot be
 831handled immediately, such as ``mount_lock`` or one of the ``d_seq``
 832seqlocks reporting a change.  In these cases the relevant function
 833will return ``-ECHILD`` which will percolate up until it triggers a new
 834attempt from the top using REF-walk.
 835
 836For those cases where ``unlazy_walk()`` is an option, it essentially
 837takes a reference on each of the pointers that it holds (vfsmount,
 838dentry, and possibly some symbolic links) and then verifies that the
 839relevant seqlocks have not been changed.  If there have been changes,
 840it, too, aborts with ``-ECHILD``, otherwise the transition to REF-walk
 841has been a success and the lookup process continues.
 842
 843Taking a reference on those pointers is not quite as simple as just
 844incrementing a counter.  That works to take a second reference if you
 845already have one (often indirectly through another object), but it
 846isn't sufficient if you don't actually have a counted reference at
 847all.  For ``dentry->d_lockref``, it is safe to increment the reference
 848counter to get a reference unless it has been explicitly marked as
 849"dead" which involves setting the counter to ``-128``.
 850``lockref_get_not_dead()`` achieves this.
 851
 852For ``mnt->mnt_count`` it is safe to take a reference as long as
 853``mount_lock`` is then used to validate the reference.  If that
 854validation fails, it may *not* be safe to just drop that reference in
 855the standard way of calling ``mnt_put()`` - an unmount may have
 856progressed too far.  So the code in ``legitimize_mnt()``, when it
 857finds that the reference it got might not be safe, checks the
 858``MNT_SYNC_UMOUNT`` flag to determine if a simple ``mnt_put()`` is
 859correct, or if it should just decrement the count and pretend none of
 860this ever happened.
 861
 862Taking care in filesystems
 863--------------------------
 864
 865RCU-walk depends almost entirely on cached information and often will
 866not call into the filesystem at all.  However there are two places,
 867besides the already-mentioned component-name comparison, where the
 868file system might be included in RCU-walk, and it must know to be
 869careful.
 870
 871If the filesystem has non-standard permission-checking requirements -
 872such as a networked filesystem which may need to check with the server
 873- the ``i_op->permission`` interface might be called during RCU-walk.
 874In this case an extra "``MAY_NOT_BLOCK``" flag is passed so that it
 875knows not to sleep, but to return ``-ECHILD`` if it cannot complete
 876promptly.  ``i_op->permission`` is given the inode pointer, not the
 877dentry, so it doesn't need to worry about further consistency checks.
 878However if it accesses any other filesystem data structures, it must
 879ensure they are safe to be accessed with only the ``rcu_read_lock()``
 880held.  This typically means they must be freed using ``kfree_rcu()`` or
 881similar.
 882
 883.. _READ_ONCE: https://lwn.net/Articles/624126/
 884
 885If the filesystem may need to revalidate dcache entries, then
 886``d_op->d_revalidate`` may be called in RCU-walk too.  This interface
 887*is* passed the dentry but does not have access to the ``inode`` or the
 888``seq`` number from the ``nameidata``, so it needs to be extra careful
 889when accessing fields in the dentry.  This "extra care" typically
 890involves using  `READ_ONCE() <READ_ONCE_>`_ to access fields, and verifying the
 891result is not NULL before using it.  This pattern can be seen in
 892``nfs_lookup_revalidate()``.
 893
 894A pair of patterns
 895------------------
 896
 897In various places in the details of REF-walk and RCU-walk, and also in
 898the big picture, there are a couple of related patterns that are worth
 899being aware of.
 900
 901The first is "try quickly and check, if that fails try slowly".  We
 902can see that in the high-level approach of first trying RCU-walk and
 903then trying REF-walk, and in places where ``unlazy_walk()`` is used to
 904switch to REF-walk for the rest of the path.  We also saw it earlier
 905in ``dget_parent()`` when following a "``..``" link.  It tries a quick way
 906to get a reference, then falls back to taking locks if needed.
 907
 908The second pattern is "try quickly and check, if that fails try
 909again - repeatedly".  This is seen with the use of ``rename_lock`` and
 910``mount_lock`` in REF-walk.  RCU-walk doesn't make use of this pattern -
 911if anything goes wrong it is much safer to just abort and try a more
 912sedate approach.
 913
 914The emphasis here is "try quickly and check".  It should probably be
 915"try quickly _and carefully,_ then check".  The fact that checking is
 916needed is a reminder that the system is dynamic and only a limited
 917number of things are safe at all.  The most likely cause of errors in
 918this whole process is assuming something is safe when in reality it
 919isn't.  Careful consideration of what exactly guarantees the safety of
 920each access is sometimes necessary.
 921
 922A walk among the symlinks
 923=========================
 924
 925There are several basic issues that we will examine to understand the
 926handling of symbolic links:  the symlink stack, together with cache
 927lifetimes, will help us understand the overall recursive handling of
 928symlinks and lead to the special care needed for the final component.
 929Then a consideration of access-time updates and summary of the various
 930flags controlling lookup will finish the story.
 931
 932The symlink stack
 933-----------------
 934
 935There are only two sorts of filesystem objects that can usefully
 936appear in a path prior to the final component: directories and symlinks.
 937Handling directories is quite straightforward: the new directory
 938simply becomes the starting point at which to interpret the next
 939component on the path.  Handling symbolic links requires a bit more
 940work.
 941
 942Conceptually, symbolic links could be handled by editing the path.  If
 943a component name refers to a symbolic link, then that component is
 944replaced by the body of the link and, if that body starts with a '/',
 945then all preceding parts of the path are discarded.  This is what the
 946"``readlink -f``" command does, though it also edits out "``.``" and
 947"``..``" components.
 948
 949Directly editing the path string is not really necessary when looking
 950up a path, and discarding early components is pointless as they aren't
 951looked at anyway.  Keeping track of all remaining components is
 952important, but they can of course be kept separately; there is no need
 953to concatenate them.  As one symlink may easily refer to another,
 954which in turn can refer to a third, we may need to keep the remaining
 955components of several paths, each to be processed when the preceding
 956ones are completed.  These path remnants are kept on a stack of
 957limited size.
 958
 959There are two reasons for placing limits on how many symlinks can
 960occur in a single path lookup.  The most obvious is to avoid loops.
 961If a symlink referred to itself either directly or through
 962intermediaries, then following the symlink can never complete
 963successfully - the error ``ELOOP`` must be returned.  Loops can be
 964detected without imposing limits, but limits are the simplest solution
 965and, given the second reason for restriction, quite sufficient.
 966
 967.. _outlined recently: http://thread.gmane.org/gmane.linux.kernel/1934390/focus=1934550
 968
 969The second reason was `outlined recently`_ by Linus:
 970
 971   Because it's a latency and DoS issue too. We need to react well to
 972   true loops, but also to "very deep" non-loops. It's not about memory
 973   use, it's about users triggering unreasonable CPU resources.
 974
 975Linux imposes a limit on the length of any pathname: ``PATH_MAX``, which
 976is 4096.  There are a number of reasons for this limit; not letting the
 977kernel spend too much time on just one path is one of them.  With
 978symbolic links you can effectively generate much longer paths so some
 979sort of limit is needed for the same reason.  Linux imposes a limit of
 980at most 40 symlinks in any one path lookup.  It previously imposed a
 981further limit of eight on the maximum depth of recursion, but that was
 982raised to 40 when a separate stack was implemented, so there is now
 983just the one limit.
 984
 985The ``nameidata`` structure that we met in an earlier article contains a
 986small stack that can be used to store the remaining part of up to two
 987symlinks.  In many cases this will be sufficient.  If it isn't, a
 988separate stack is allocated with room for 40 symlinks.  Pathname
 989lookup will never exceed that stack as, once the 40th symlink is
 990detected, an error is returned.
 991
 992It might seem that the name remnants are all that needs to be stored on
 993this stack, but we need a bit more.  To see that, we need to move on to
 994cache lifetimes.
 995
 996Storage and lifetime of cached symlinks
 997---------------------------------------
 998
 999Like other filesystem resources, such as inodes and directory
1000entries, symlinks are cached by Linux to avoid repeated costly access
1001to external storage.  It is particularly important for RCU-walk to be
1002able to find and temporarily hold onto these cached entries, so that
1003it doesn't need to drop down into REF-walk.
1004
1005.. _object-oriented design pattern: https://lwn.net/Articles/446317/
1006
1007While each filesystem is free to make its own choice, symlinks are
1008typically stored in one of two places.  Short symlinks are often
1009stored directly in the inode.  When a filesystem allocates a ``struct
1010inode`` it typically allocates extra space to store private data (a
1011common `object-oriented design pattern`_ in the kernel).  This will
1012sometimes include space for a symlink.  The other common location is
1013in the page cache, which normally stores the content of files.  The
1014pathname in a symlink can be seen as the content of that symlink and
1015can easily be stored in the page cache just like file content.
1016
1017When neither of these is suitable, the next most likely scenario is
1018that the filesystem will allocate some temporary memory and copy or
1019construct the symlink content into that memory whenever it is needed.
1020
1021When the symlink is stored in the inode, it has the same lifetime as
1022the inode which, itself, is protected by RCU or by a counted reference
1023on the dentry.  This means that the mechanisms that pathname lookup
1024uses to access the dcache and icache (inode cache) safely are quite
1025sufficient for accessing some cached symlinks safely.  In these cases,
1026the ``i_link`` pointer in the inode is set to point to wherever the
1027symlink is stored and it can be accessed directly whenever needed.
1028
1029When the symlink is stored in the page cache or elsewhere, the
1030situation is not so straightforward.  A reference on a dentry or even
1031on an inode does not imply any reference on cached pages of that
1032inode, and even an ``rcu_read_lock()`` is not sufficient to ensure that
1033a page will not disappear.  So for these symlinks the pathname lookup
1034code needs to ask the filesystem to provide a stable reference and,
1035significantly, needs to release that reference when it is finished
1036with it.
1037
1038Taking a reference to a cache page is often possible even in RCU-walk
1039mode.  It does require making changes to memory, which is best avoided,
1040but that isn't necessarily a big cost and it is better than dropping
1041out of RCU-walk mode completely.  Even filesystems that allocate
1042space to copy the symlink into can use ``GFP_ATOMIC`` to often successfully
1043allocate memory without the need to drop out of RCU-walk.  If a
1044filesystem cannot successfully get a reference in RCU-walk mode, it
1045must return ``-ECHILD`` and ``unlazy_walk()`` will be called to return to
1046REF-walk mode in which the filesystem is allowed to sleep.
1047
1048The place for all this to happen is the ``i_op->follow_link()`` inode
1049method.  In the present mainline code this is never actually called in
1050RCU-walk mode as the rewrite is not quite complete.  It is likely that
1051in a future release this method will be passed an ``inode`` pointer when
1052called in RCU-walk mode so it both (1) knows to be careful, and (2) has the
1053validated pointer.  Much like the ``i_op->permission()`` method we
1054looked at previously, ``->follow_link()`` would need to be careful that
1055all the data structures it references are safe to be accessed while
1056holding no counted reference, only the RCU lock.  Though getting a
1057reference with ``->follow_link()`` is not yet done in RCU-walk mode, the
1058code is ready to release the reference when that does happen.
1059
1060This need to drop the reference to a symlink adds significant
1061complexity.  It requires a reference to the inode so that the
1062``i_op->put_link()`` inode operation can be called.  In REF-walk, that
1063reference is kept implicitly through a reference to the dentry, so
1064keeping the ``struct path`` of the symlink is easiest.  For RCU-walk,
1065the pointer to the inode is kept separately.  To allow switching from
1066RCU-walk back to REF-walk in the middle of processing nested symlinks
1067we also need the seq number for the dentry so we can confirm that
1068switching back was safe.
1069
1070Finally, when providing a reference to a symlink, the filesystem also
1071provides an opaque "cookie" that must be passed to ``->put_link()`` so that it
1072knows what to free.  This might be the allocated memory area, or a
1073pointer to the ``struct page`` in the page cache, or something else
1074completely.  Only the filesystem knows what it is.
1075
1076In order for the reference to each symlink to be dropped when the walk completes,
1077whether in RCU-walk or REF-walk, the symlink stack needs to contain,
1078along with the path remnants:
1079
1080- the ``struct path`` to provide a reference to the inode in REF-walk
1081- the ``struct inode *`` to provide a reference to the inode in RCU-walk
1082- the ``seq`` to allow the path to be safely switched from RCU-walk to REF-walk
1083- the ``cookie`` that tells ``->put_path()`` what to put.
1084
1085This means that each entry in the symlink stack needs to hold five
1086pointers and an integer instead of just one pointer (the path
1087remnant).  On a 64-bit system, this is about 40 bytes per entry;
1088with 40 entries it adds up to 1600 bytes total, which is less than
1089half a page.  So it might seem like a lot, but is by no means
1090excessive.
1091
1092Note that, in a given stack frame, the path remnant (``name``) is not
1093part of the symlink that the other fields refer to.  It is the remnant
1094to be followed once that symlink has been fully parsed.
1095
1096Following the symlink
1097---------------------
1098
1099The main loop in ``link_path_walk()`` iterates seamlessly over all
1100components in the path and all of the non-final symlinks.  As symlinks
1101are processed, the ``name`` pointer is adjusted to point to a new
1102symlink, or is restored from the stack, so that much of the loop
1103doesn't need to notice.  Getting this ``name`` variable on and off the
1104stack is very straightforward; pushing and popping the references is
1105a little more complex.
1106
1107When a symlink is found, ``walk_component()`` returns the value ``1``
1108(``0`` is returned for any other sort of success, and a negative number
1109is, as usual, an error indicator).  This causes ``get_link()`` to be
1110called; it then gets the link from the filesystem.  Providing that
1111operation is successful, the old path ``name`` is placed on the stack,
1112and the new value is used as the ``name`` for a while.  When the end of
1113the path is found (i.e. ``*name`` is ``'\0'``) the old ``name`` is restored
1114off the stack and path walking continues.
1115
1116Pushing and popping the reference pointers (inode, cookie, etc.) is more
1117complex in part because of the desire to handle tail recursion.  When
1118the last component of a symlink itself points to a symlink, we
1119want to pop the symlink-just-completed off the stack before pushing
1120the symlink-just-found to avoid leaving empty path remnants that would
1121just get in the way.
1122
1123It is most convenient to push the new symlink references onto the
1124stack in ``walk_component()`` immediately when the symlink is found;
1125``walk_component()`` is also the last piece of code that needs to look at the
1126old symlink as it walks that last component.  So it is quite
1127convenient for ``walk_component()`` to release the old symlink and pop
1128the references just before pushing the reference information for the
1129new symlink.  It is guided in this by two flags; ``WALK_GET``, which
1130gives it permission to follow a symlink if it finds one, and
1131``WALK_PUT``, which tells it to release the current symlink after it has been
1132followed.  ``WALK_PUT`` is tested first, leading to a call to
1133``put_link()``.  ``WALK_GET`` is tested subsequently (by
1134``should_follow_link()``) leading to a call to ``pick_link()`` which sets
1135up the stack frame.
1136
1137Symlinks with no final component
1138~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1139
1140A pair of special-case symlinks deserve a little further explanation.
1141Both result in a new ``struct path`` (with mount and dentry) being set
1142up in the ``nameidata``, and result in ``get_link()`` returning ``NULL``.
1143
1144The more obvious case is a symlink to "``/``".  All symlinks starting
1145with "``/``" are detected in ``get_link()`` which resets the ``nameidata``
1146to point to the effective filesystem root.  If the symlink only
1147contains "``/``" then there is nothing more to do, no components at all,
1148so ``NULL`` is returned to indicate that the symlink can be released and
1149the stack frame discarded.
1150
1151The other case involves things in ``/proc`` that look like symlinks but
1152aren't really::
1153
1154     $ ls -l /proc/self/fd/1
1155     lrwx------ 1 neilb neilb 64 Jun 13 10:19 /proc/self/fd/1 -> /dev/pts/4
1156
1157Every open file descriptor in any process is represented in ``/proc`` by
1158something that looks like a symlink.  It is really a reference to the
1159target file, not just the name of it.  When you ``readlink`` these
1160objects you get a name that might refer to the same file - unless it
1161has been unlinked or mounted over.  When ``walk_component()`` follows
1162one of these, the ``->follow_link()`` method in "procfs" doesn't return
1163a string name, but instead calls ``nd_jump_link()`` which updates the
1164``nameidata`` in place to point to that target.  ``->follow_link()`` then
1165returns ``NULL``.  Again there is no final component and ``get_link()``
1166reports this by leaving the ``last_type`` field of ``nameidata`` as
1167``LAST_BIND``.
1168
1169Following the symlink in the final component
1170--------------------------------------------
1171
1172All this leads to ``link_path_walk()`` walking down every component, and
1173following all symbolic links it finds, until it reaches the final
1174component.  This is just returned in the ``last`` field of ``nameidata``.
1175For some callers, this is all they need; they want to create that
1176``last`` name if it doesn't exist or give an error if it does.  Other
1177callers will want to follow a symlink if one is found, and possibly
1178apply special handling to the last component of that symlink, rather
1179than just the last component of the original file name.  These callers
1180potentially need to call ``link_path_walk()`` again and again on
1181successive symlinks until one is found that doesn't point to another
1182symlink.
1183
1184This case is handled by the relevant caller of ``link_path_walk()``, such as
1185``path_lookupat()`` using a loop that calls ``link_path_walk()``, and then
1186handles the final component.  If the final component is a symlink
1187that needs to be followed, then ``trailing_symlink()`` is called to set
1188things up properly and the loop repeats, calling ``link_path_walk()``
1189again.  This could loop as many as 40 times if the last component of
1190each symlink is another symlink.
1191
1192The various functions that examine the final component and possibly
1193report that it is a symlink are ``lookup_last()``, ``mountpoint_last()``
1194and ``do_last()``, each of which use the same convention as
1195``walk_component()`` of returning ``1`` if a symlink was found that needs
1196to be followed.
1197
1198Of these, ``do_last()`` is the most interesting as it is used for
1199opening a file.  Part of ``do_last()`` runs with ``i_rwsem`` held and this
1200part is in a separate function: ``lookup_open()``.
1201
1202Explaining ``do_last()`` completely is beyond the scope of this article,
1203but a few highlights should help those interested in exploring the
1204code.
1205
12061. Rather than just finding the target file, ``do_last()`` needs to open
1207   it.  If the file was found in the dcache, then ``vfs_open()`` is used for
1208   this.  If not, then ``lookup_open()`` will either call ``atomic_open()`` (if
1209   the filesystem provides it) to combine the final lookup with the open, or
1210   will perform the separate ``lookup_real()`` and ``vfs_create()`` steps
1211   directly.  In the later case the actual "open" of this newly found or
1212   created file will be performed by ``vfs_open()``, just as if the name
1213   were found in the dcache.
1214
12152. ``vfs_open()`` can fail with ``-EOPENSTALE`` if the cached information
1216   wasn't quite current enough.  Rather than restarting the lookup from
1217   the top with ``LOOKUP_REVAL`` set, ``lookup_open()`` is called instead,
1218   giving the filesystem a chance to resolve small inconsistencies.
1219   If that doesn't work, only then is the lookup restarted from the top.
1220
12213. An open with O_CREAT **does** follow a symlink in the final component,
1222   unlike other creation system calls (like ``mkdir``).  So the sequence::
1223
1224          ln -s bar /tmp/foo
1225          echo hello > /tmp/foo
1226
1227   will create a file called ``/tmp/bar``.  This is not permitted if
1228   ``O_EXCL`` is set but otherwise is handled for an O_CREAT open much
1229   like for a non-creating open: ``should_follow_link()`` returns ``1``, and
1230   so does ``do_last()`` so that ``trailing_symlink()`` gets called and the
1231   open process continues on the symlink that was found.
1232
1233Updating the access time
1234------------------------
1235
1236We previously said of RCU-walk that it would "take no locks, increment
1237no counts, leave no footprints."  We have since seen that some
1238"footprints" can be needed when handling symlinks as a counted
1239reference (or even a memory allocation) may be needed.  But these
1240footprints are best kept to a minimum.
1241
1242One other place where walking down a symlink can involve leaving
1243footprints in a way that doesn't affect directories is in updating access times.
1244In Unix (and Linux) every filesystem object has a "last accessed
1245time", or "``atime``".  Passing through a directory to access a file
1246within is not considered to be an access for the purposes of
1247``atime``; only listing the contents of a directory can update its ``atime``.
1248Symlinks are different it seems.  Both reading a symlink (with ``readlink()``)
1249and looking up a symlink on the way to some other destination can
1250update the atime on that symlink.
1251
1252.. _clearest statement: http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap04.html#tag_04_08
1253
1254It is not clear why this is the case; POSIX has little to say on the
1255subject.  The `clearest statement`_ is that, if a particular implementation
1256updates a timestamp in a place not specified by POSIX, this must be
1257documented "except that any changes caused by pathname resolution need
1258not be documented".  This seems to imply that POSIX doesn't really
1259care about access-time updates during pathname lookup.
1260
1261.. _Linux 1.3.87: https://git.kernel.org/cgit/linux/kernel/git/history/history.git/diff/fs/ext2/symlink.c?id=f806c6db77b8eaa6e00dcfb6b567706feae8dbb8
1262
1263An examination of history shows that prior to `Linux 1.3.87`_, the ext2
1264filesystem, at least, didn't update atime when following a link.
1265Unfortunately we have no record of why that behavior was changed.
1266
1267In any case, access time must now be updated and that operation can be
1268quite complex.  Trying to stay in RCU-walk while doing it is best
1269avoided.  Fortunately it is often permitted to skip the ``atime``
1270update.  Because ``atime`` updates cause performance problems in various
1271areas, Linux supports the ``relatime`` mount option, which generally
1272limits the updates of ``atime`` to once per day on files that aren't
1273being changed (and symlinks never change once created).  Even without
1274``relatime``, many filesystems record ``atime`` with a one-second
1275granularity, so only one update per second is required.
1276
1277It is easy to test if an ``atime`` update is needed while in RCU-walk
1278mode and, if it isn't, the update can be skipped and RCU-walk mode
1279continues.  Only when an ``atime`` update is actually required does the
1280path walk drop down to REF-walk.  All of this is handled in the
1281``get_link()`` function.
1282
1283A few flags
1284-----------
1285
1286A suitable way to wrap up this tour of pathname walking is to list
1287the various flags that can be stored in the ``nameidata`` to guide the
1288lookup process.  Many of these are only meaningful on the final
1289component, others reflect the current state of the pathname lookup.
 
 
1290And then there is ``LOOKUP_EMPTY``, which doesn't fit conceptually with
1291the others.  If this is not set, an empty pathname causes an error
1292very early on.  If it is set, empty pathnames are not considered to be
1293an error.
1294
1295Global state flags
1296~~~~~~~~~~~~~~~~~~
1297
1298We have already met two global state flags: ``LOOKUP_RCU`` and
1299``LOOKUP_REVAL``.  These select between one of three overall approaches
1300to lookup: RCU-walk, REF-walk, and REF-walk with forced revalidation.
1301
1302``LOOKUP_PARENT`` indicates that the final component hasn't been reached
1303yet.  This is primarily used to tell the audit subsystem the full
1304context of a particular access being audited.
1305
1306``LOOKUP_ROOT`` indicates that the ``root`` field in the ``nameidata`` was
1307provided by the caller, so it shouldn't be released when it is no
1308longer needed.
1309
1310``LOOKUP_JUMPED`` means that the current dentry was chosen not because
1311it had the right name but for some other reason.  This happens when
1312following "``..``", following a symlink to ``/``, crossing a mount point
1313or accessing a "``/proc/$PID/fd/$FD``" symlink.  In this case the
1314filesystem has not been asked to revalidate the name (with
1315``d_revalidate()``).  In such cases the inode may still need to be
1316revalidated, so ``d_op->d_weak_revalidate()`` is called if
1317``LOOKUP_JUMPED`` is set when the look completes - which may be at the
1318final component or, when creating, unlinking, or renaming, at the penultimate component.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1319
1320Final-component flags
1321~~~~~~~~~~~~~~~~~~~~~
1322
1323Some of these flags are only set when the final component is being
1324considered.  Others are only checked for when considering that final
1325component.
1326
1327``LOOKUP_AUTOMOUNT`` ensures that, if the final component is an automount
1328point, then the mount is triggered.  Some operations would trigger it
1329anyway, but operations like ``stat()`` deliberately don't.  ``statfs()``
1330needs to trigger the mount but otherwise behaves a lot like ``stat()``, so
1331it sets ``LOOKUP_AUTOMOUNT``, as does "``quotactl()``" and the handling of
1332"``mount --bind``".
1333
1334``LOOKUP_FOLLOW`` has a similar function to ``LOOKUP_AUTOMOUNT`` but for
1335symlinks.  Some system calls set or clear it implicitly, while
1336others have API flags such as ``AT_SYMLINK_FOLLOW`` and
1337``UMOUNT_NOFOLLOW`` to control it.  Its effect is similar to
1338``WALK_GET`` that we already met, but it is used in a different way.
1339
1340``LOOKUP_DIRECTORY`` insists that the final component is a directory.
1341Various callers set this and it is also set when the final component
1342is found to be followed by a slash.
1343
1344Finally ``LOOKUP_OPEN``, ``LOOKUP_CREATE``, ``LOOKUP_EXCL``, and
1345``LOOKUP_RENAME_TARGET`` are not used directly by the VFS but are made
1346available to the filesystem and particularly the ``->d_revalidate()``
1347method.  A filesystem can choose not to bother revalidating too hard
1348if it knows that it will be asked to open or create the file soon.
1349These flags were previously useful for ``->lookup()`` too but with the
1350introduction of ``->atomic_open()`` they are less relevant there.
1351
1352End of the road
1353---------------
1354
1355Despite its complexity, all this pathname lookup code appears to be
1356in good shape - various parts are certainly easier to understand now
1357than even a couple of releases ago.  But that doesn't mean it is
1358"finished".   As already mentioned, RCU-walk currently only follows
1359symlinks that are stored in the inode so, while it handles many ext4
1360symlinks, it doesn't help with NFS, XFS, or Btrfs.  That support
1361is not likely to be long delayed.

   1===============
   2Pathname lookup
   3===============
   4
   5This write-up is based on three articles published at lwn.net:
   6
   7- <https://lwn.net/Articles/649115/> Pathname lookup in Linux
   8- <https://lwn.net/Articles/649729/> RCU-walk: faster pathname lookup in Linux
   9- <https://lwn.net/Articles/650786/> A walk among the symlinks
  10
  11Written by Neil Brown with help from Al Viro and Jon Corbet.
  12It has subsequently been updated to reflect changes in the kernel
  13including:
  14
  15- per-directory parallel name lookup.
  16- ``openat2()`` resolution restriction flags.
  17
  18Introduction to pathname lookup
  19===============================
  20
  21The most obvious aspect of pathname lookup, which very little
  22exploration is needed to discover, is that it is complex.  There are
  23many rules, special cases, and implementation alternatives that all
  24combine to confuse the unwary reader.  Computer science has long been
  25acquainted with such complexity and has tools to help manage it.  One
  26tool that we will make extensive use of is "divide and conquer".  For
  27the early parts of the analysis we will divide off symlinks - leaving
  28them until the final part.  Well before we get to symlinks we have
  29another major division based on the VFS's approach to locking which
  30will allow us to review "REF-walk" and "RCU-walk" separately.  But we
  31are getting ahead of ourselves.  There are some important low level
  32distinctions we need to clarify first.
  33
  34There are two sorts of ...
  35--------------------------
  36
  37.. _openat: http://man7.org/linux/man-pages/man2/openat.2.html
  38
  39Pathnames (sometimes "file names"), used to identify objects in the
  40filesystem, will be familiar to most readers.  They contain two sorts
  41of elements: "slashes" that are sequences of one or more "``/``"
  42characters, and "components" that are sequences of one or more
  43non-"``/``" characters.  These form two kinds of paths.  Those that
  44start with slashes are "absolute" and start from the filesystem root.
  45The others are "relative" and start from the current directory, or
  46from some other location specified by a file descriptor given to
  47"``*at()``" system calls such as `openat() <openat_>`_.
  48
  49.. _execveat: http://man7.org/linux/man-pages/man2/execveat.2.html
  50
  51It is tempting to describe the second kind as starting with a
  52component, but that isn't always accurate: a pathname can lack both
  53slashes and components, it can be empty, in other words.  This is
  54generally forbidden in POSIX, but some of those "``*at()``" system calls
  55in Linux permit it when the ``AT_EMPTY_PATH`` flag is given.  For
  56example, if you have an open file descriptor on an executable file you
  57can execute it by calling `execveat() <execveat_>`_ passing
  58the file descriptor, an empty path, and the ``AT_EMPTY_PATH`` flag.
  59
  60These paths can be divided into two sections: the final component and
  61everything else.  The "everything else" is the easy bit.  In all cases
  62it must identify a directory that already exists, otherwise an error
  63such as ``ENOENT`` or ``ENOTDIR`` will be reported.
  64
  65The final component is not so simple.  Not only do different system
  66calls interpret it quite differently (e.g. some create it, some do
  67not), but it might not even exist: neither the empty pathname nor the
  68pathname that is just slashes have a final component.  If it does
  69exist, it could be "``.``" or "``..``" which are handled quite differently
  70from other components.
  71
  72.. _POSIX: https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap04.html#tag_04_12
  73
  74If a pathname ends with a slash, such as "``/tmp/foo/``" it might be
  75tempting to consider that to have an empty final component.  In many
  76ways that would lead to correct results, but not always.  In
  77particular, ``mkdir()`` and ``rmdir()`` each create or remove a directory named
  78by the final component, and they are required to work with pathnames
  79ending in "``/``".  According to POSIX_:
  80
  81  A pathname that contains at least one non-<slash> character and
  82  that ends with one or more trailing <slash> characters shall not
  83  be resolved successfully unless the last pathname component before
  84  the trailing <slash> characters names an existing directory or a
  85  directory entry that is to be created for a directory immediately
  86  after the pathname is resolved.
  87
  88The Linux pathname walking code (mostly in ``fs/namei.c``) deals with
  89all of these issues: breaking the path into components, handling the
  90"everything else" quite separately from the final component, and
  91checking that the trailing slash is not used where it isn't
  92permitted.  It also addresses the important issue of concurrent
  93access.
  94
  95While one process is looking up a pathname, another might be making
  96changes that affect that lookup.  One fairly extreme case is that if
  97"a/b" were renamed to "a/c/b" while another process were looking up
  98"a/b/..", that process might successfully resolve on "a/c".
  99Most races are much more subtle, and a big part of the task of
 100pathname lookup is to prevent them from having damaging effects.  Many
 101of the possible races are seen most clearly in the context of the
 102"dcache" and an understanding of that is central to understanding
 103pathname lookup.
 104
 105More than just a cache
 106----------------------
 107
 108The "dcache" caches information about names in each filesystem to
 109make them quickly available for lookup.  Each entry (known as a
 110"dentry") contains three significant fields: a component name, a
 111pointer to a parent dentry, and a pointer to the "inode" which
 112contains further information about the object in that parent with
 113the given name.  The inode pointer can be ``NULL`` indicating that the
 114name doesn't exist in the parent.  While there can be linkage in the
 115dentry of a directory to the dentries of the children, that linkage is
 116not used for pathname lookup, and so will not be considered here.
 117
 118The dcache has a number of uses apart from accelerating lookup.  One
 119that will be particularly relevant is that it is closely integrated
 120with the mount table that records which filesystem is mounted where.
 121What the mount table actually stores is which dentry is mounted on top
 122of which other dentry.
 123
 124When considering the dcache, we have another of our "two types"
 125distinctions: there are two types of filesystems.
 126
 127Some filesystems ensure that the information in the dcache is always
 128completely accurate (though not necessarily complete).  This can allow
 129the VFS to determine if a particular file does or doesn't exist
 130without checking with the filesystem, and means that the VFS can
 131protect the filesystem against certain races and other problems.
 132These are typically "local" filesystems such as ext3, XFS, and Btrfs.
 133
 134Other filesystems don't provide that guarantee because they cannot.
 135These are typically filesystems that are shared across a network,
 136whether remote filesystems like NFS and 9P, or cluster filesystems
 137like ocfs2 or cephfs.  These filesystems allow the VFS to revalidate
 138cached information, and must provide their own protection against
 139awkward races.  The VFS can detect these filesystems by the
 140``DCACHE_OP_REVALIDATE`` flag being set in the dentry.
 141
 142REF-walk: simple concurrency management with refcounts and spinlocks
 143--------------------------------------------------------------------
 144
 145With all of those divisions carefully classified, we can now start
 146looking at the actual process of walking along a path.  In particular
 147we will start with the handling of the "everything else" part of a
 148pathname, and focus on the "REF-walk" approach to concurrency
 149management.  This code is found in the ``link_path_walk()`` function, if
 150you ignore all the places that only run when "``LOOKUP_RCU``"
 151(indicating the use of RCU-walk) is set.
 152
 153.. _Meet the Lockers: https://lwn.net/Articles/453685/
 154
 155REF-walk is fairly heavy-handed with locks and reference counts.  Not
 156as heavy-handed as in the old "big kernel lock" days, but certainly not
 157afraid of taking a lock when one is needed.  It uses a variety of
 158different concurrency controls.  A background understanding of the
 159various primitives is assumed, or can be gleaned from elsewhere such
 160as in `Meet the Lockers`_.
 161
 162The locking mechanisms used by REF-walk include:
 163
 164dentry->d_lockref
 165~~~~~~~~~~~~~~~~~
 166
 167This uses the lockref primitive to provide both a spinlock and a
 168reference count.  The special-sauce of this primitive is that the
 169conceptual sequence "lock; inc_ref; unlock;" can often be performed
 170with a single atomic memory operation.
 171
 172Holding a reference on a dentry ensures that the dentry won't suddenly
 173be freed and used for something else, so the values in various fields
 174will behave as expected.  It also protects the ``->d_inode`` reference
 175to the inode to some extent.
 176
 177The association between a dentry and its inode is fairly permanent.
 178For example, when a file is renamed, the dentry and inode move
 179together to the new location.  When a file is created the dentry will
 180initially be negative (i.e. ``d_inode`` is ``NULL``), and will be assigned
 181to the new inode as part of the act of creation.
 182
 183When a file is deleted, this can be reflected in the cache either by
 184setting ``d_inode`` to ``NULL``, or by removing it from the hash table
 185(described shortly) used to look up the name in the parent directory.
 186If the dentry is still in use the second option is used as it is
 187perfectly legal to keep using an open file after it has been deleted
 188and having the dentry around helps.  If the dentry is not otherwise in
 189use (i.e. if the refcount in ``d_lockref`` is one), only then will
 190``d_inode`` be set to ``NULL``.  Doing it this way is more efficient for a
 191very common case.
 192
 193So as long as a counted reference is held to a dentry, a non-``NULL`` ``->d_inode``
 194value will never be changed.
 195
 196dentry->d_lock
 197~~~~~~~~~~~~~~
 198
 199``d_lock`` is a synonym for the spinlock that is part of ``d_lockref`` above.
 200For our purposes, holding this lock protects against the dentry being
 201renamed or unlinked.  In particular, its parent (``d_parent``), and its
 202name (``d_name``) cannot be changed, and it cannot be removed from the
 203dentry hash table.
 204
 205When looking for a name in a directory, REF-walk takes ``d_lock`` on
 206each candidate dentry that it finds in the hash table and then checks
 207that the parent and name are correct.  So it doesn't lock the parent
 208while searching in the cache; it only locks children.
 209
 210When looking for the parent for a given name (to handle "``..``"),
 211REF-walk can take ``d_lock`` to get a stable reference to ``d_parent``,
 212but it first tries a more lightweight approach.  As seen in
 213``dget_parent()``, if a reference can be claimed on the parent, and if
 214subsequently ``d_parent`` can be seen to have not changed, then there is
 215no need to actually take the lock on the child.
 216
 217rename_lock
 218~~~~~~~~~~~
 219
 220Looking up a given name in a given directory involves computing a hash
 221from the two values (the name and the dentry of the directory),
 222accessing that slot in a hash table, and searching the linked list
 223that is found there.
 224
 225When a dentry is renamed, the name and the parent dentry can both
 226change so the hash will almost certainly change too.  This would move the
 227dentry to a different chain in the hash table.  If a filename search
 228happened to be looking at a dentry that was moved in this way,
 229it might end up continuing the search down the wrong chain,
 230and so miss out on part of the correct chain.
 231
 232The name-lookup process (``d_lookup()``) does *not* try to prevent this
 233from happening, but only to detect when it happens.
 234``rename_lock`` is a seqlock that is updated whenever any dentry is
 235renamed.  If ``d_lookup`` finds that a rename happened while it
 236unsuccessfully scanned a chain in the hash table, it simply tries
 237again.
 238
 239``rename_lock`` is also used to detect and defend against potential attacks
 240against ``LOOKUP_BENEATH`` and ``LOOKUP_IN_ROOT`` when resolving ".." (where
 241the parent directory is moved outside the root, bypassing the ``path_equal()``
 242check). If ``rename_lock`` is updated during the lookup and the path encounters
 243a "..", a potential attack occurred and ``handle_dots()`` will bail out with
 244``-EAGAIN``.
 245
 246inode->i_rwsem
 247~~~~~~~~~~~~~~
 248
 249``i_rwsem`` is a read/write semaphore that serializes all changes to a particular
 250directory.  This ensures that, for example, an ``unlink()`` and a ``rename()``
 251cannot both happen at the same time.  It also keeps the directory
 252stable while the filesystem is asked to look up a name that is not
 253currently in the dcache or, optionally, when the list of entries in a
 254directory is being retrieved with ``readdir()``.
 255
 256This has a complementary role to that of ``d_lock``: ``i_rwsem`` on a
 257directory protects all of the names in that directory, while ``d_lock``
 258on a name protects just one name in a directory.  Most changes to the
 259dcache hold ``i_rwsem`` on the relevant directory inode and briefly take
 260``d_lock`` on one or more the dentries while the change happens.  One
 261exception is when idle dentries are removed from the dcache due to
 262memory pressure.  This uses ``d_lock``, but ``i_rwsem`` plays no role.
 263
 264The semaphore affects pathname lookup in two distinct ways.  Firstly it
 265prevents changes during lookup of a name in a directory.  ``walk_component()`` uses
 266``lookup_fast()`` first which, in turn, checks to see if the name is in the cache,
 267using only ``d_lock`` locking.  If the name isn't found, then ``walk_component()``
 268falls back to ``lookup_slow()`` which takes a shared lock on ``i_rwsem``, checks again that
 269the name isn't in the cache, and then calls in to the filesystem to get a
 270definitive answer.  A new dentry will be added to the cache regardless of
 271the result.
 272
 273Secondly, when pathname lookup reaches the final component, it will
 274sometimes need to take an exclusive lock on ``i_rwsem`` before performing the last lookup so
 275that the required exclusion can be achieved.  How path lookup chooses
 276to take, or not take, ``i_rwsem`` is one of the
 277issues addressed in a subsequent section.
 278
 279If two threads attempt to look up the same name at the same time - a
 280name that is not yet in the dcache - the shared lock on ``i_rwsem`` will
 281not prevent them both adding new dentries with the same name.  As this
 282would result in confusion an extra level of interlocking is used,
 283based around a secondary hash table (``in_lookup_hashtable``) and a
 284per-dentry flag bit (``DCACHE_PAR_LOOKUP``).
 285
 286To add a new dentry to the cache while only holding a shared lock on
 287``i_rwsem``, a thread must call ``d_alloc_parallel()``.  This allocates a
 288dentry, stores the required name and parent in it, checks if there
 289is already a matching dentry in the primary or secondary hash
 290tables, and if not, stores the newly allocated dentry in the secondary
 291hash table, with ``DCACHE_PAR_LOOKUP`` set.
 292
 293If a matching dentry was found in the primary hash table then that is
 294returned and the caller can know that it lost a race with some other
 295thread adding the entry.  If no matching dentry is found in either
 296cache, the newly allocated dentry is returned and the caller can
 297detect this from the presence of ``DCACHE_PAR_LOOKUP``.  In this case it
 298knows that it has won any race and now is responsible for asking the
 299filesystem to perform the lookup and find the matching inode.  When
 300the lookup is complete, it must call ``d_lookup_done()`` which clears
 301the flag and does some other house keeping, including removing the
 302dentry from the secondary hash table - it will normally have been
 303added to the primary hash table already.  Note that a ``struct
 304waitqueue_head`` is passed to ``d_alloc_parallel()``, and
 305``d_lookup_done()`` must be called while this ``waitqueue_head`` is still
 306in scope.
 307
 308If a matching dentry is found in the secondary hash table,
 309``d_alloc_parallel()`` has a little more work to do. It first waits for
 310``DCACHE_PAR_LOOKUP`` to be cleared, using a wait_queue that was passed
 311to the instance of ``d_alloc_parallel()`` that won the race and that
 312will be woken by the call to ``d_lookup_done()``.  It then checks to see
 313if the dentry has now been added to the primary hash table.  If it
 314has, the dentry is returned and the caller just sees that it lost any
 315race.  If it hasn't been added to the primary hash table, the most
 316likely explanation is that some other dentry was added instead using
 317``d_splice_alias()``.  In any case, ``d_alloc_parallel()`` repeats all the
 318look ups from the start and will normally return something from the
 319primary hash table.
 320
 321mnt->mnt_count
 322~~~~~~~~~~~~~~
 323
 324``mnt_count`` is a per-CPU reference counter on "``mount``" structures.
 325Per-CPU here means that incrementing the count is cheap as it only
 326uses CPU-local memory, but checking if the count is zero is expensive as
 327it needs to check with every CPU.  Taking a ``mnt_count`` reference
 328prevents the mount structure from disappearing as the result of regular
 329unmount operations, but does not prevent a "lazy" unmount.  So holding
 330``mnt_count`` doesn't ensure that the mount remains in the namespace and,
 331in particular, doesn't stabilize the link to the mounted-on dentry.  It
 332does, however, ensure that the ``mount`` data structure remains coherent,
 333and it provides a reference to the root dentry of the mounted
 334filesystem.  So a reference through ``->mnt_count`` provides a stable
 335reference to the mounted dentry, but not the mounted-on dentry.
 336
 337mount_lock
 338~~~~~~~~~~
 339
 340``mount_lock`` is a global seqlock, a bit like ``rename_lock``.  It can be used to
 341check if any change has been made to any mount points.
 342
 343While walking down the tree (away from the root) this lock is used when
 344crossing a mount point to check that the crossing was safe.  That is,
 345the value in the seqlock is read, then the code finds the mount that
 346is mounted on the current directory, if there is one, and increments
 347the ``mnt_count``.  Finally the value in ``mount_lock`` is checked against
 348the old value.  If there is no change, then the crossing was safe.  If there
 349was a change, the ``mnt_count`` is decremented and the whole process is
 350retried.
 351
 352When walking up the tree (towards the root) by following a ".." link,
 353a little more care is needed.  In this case the seqlock (which
 354contains both a counter and a spinlock) is fully locked to prevent
 355any changes to any mount points while stepping up.  This locking is
 356needed to stabilize the link to the mounted-on dentry, which the
 357refcount on the mount itself doesn't ensure.
 358
 359``mount_lock`` is also used to detect and defend against potential attacks
 360against ``LOOKUP_BENEATH`` and ``LOOKUP_IN_ROOT`` when resolving ".." (where
 361the parent directory is moved outside the root, bypassing the ``path_equal()``
 362check). If ``mount_lock`` is updated during the lookup and the path encounters
 363a "..", a potential attack occurred and ``handle_dots()`` will bail out with
 364``-EAGAIN``.
 365
 366RCU
 367~~~
 368
 369Finally the global (but extremely lightweight) RCU read lock is held
 370from time to time to ensure certain data structures don't get freed
 371unexpectedly.
 372
 373In particular it is held while scanning chains in the dcache hash
 374table, and the mount point hash table.
 375
 376Bringing it together with ``struct nameidata``
 377----------------------------------------------
 378
 379.. _First edition Unix: https://minnie.tuhs.org/cgi-bin/utree.pl?file=V1/u2.s
 380
 381Throughout the process of walking a path, the current status is stored
 382in a ``struct nameidata``, "namei" being the traditional name - dating
 383all the way back to `First Edition Unix`_ - of the function that
 384converts a "name" to an "inode".  ``struct nameidata`` contains (among
 385other fields):
 386
 387``struct path path``
 388~~~~~~~~~~~~~~~~~~~~
 389
 390A ``path`` contains a ``struct vfsmount`` (which is
 391embedded in a ``struct mount``) and a ``struct dentry``.  Together these
 392record the current status of the walk.  They start out referring to the
 393starting point (the current working directory, the root directory, or some other
 394directory identified by a file descriptor), and are updated on each
 395step.  A reference through ``d_lockref`` and ``mnt_count`` is always
 396held.
 397
 398``struct qstr last``
 399~~~~~~~~~~~~~~~~~~~~
 400
 401This is a string together with a length (i.e. *not* ``nul`` terminated)
 402that is the "next" component in the pathname.
 403
 404``int last_type``
 405~~~~~~~~~~~~~~~~~
 406
 407This is one of ``LAST_NORM``, ``LAST_ROOT``, ``LAST_DOT`` or ``LAST_DOTDOT``.
 408The ``last`` field is only valid if the type is ``LAST_NORM``.
 
 
 
 409
 410``struct path root``
 411~~~~~~~~~~~~~~~~~~~~
 412
 413This is used to hold a reference to the effective root of the
 414filesystem.  Often that reference won't be needed, so this field is
 415only assigned the first time it is used, or when a non-standard root
 416is requested.  Keeping a reference in the ``nameidata`` ensures that
 417only one root is in effect for the entire path walk, even if it races
 418with a ``chroot()`` system call.
 419
 420It should be noted that in the case of ``LOOKUP_IN_ROOT`` or
 421``LOOKUP_BENEATH``, the effective root becomes the directory file descriptor
 422passed to ``openat2()`` (which exposes these ``LOOKUP_`` flags).
 423
 424The root is needed when either of two conditions holds: (1) either the
 425pathname or a symbolic link starts with a "'/'", or (2) a "``..``"
 426component is being handled, since "``..``" from the root must always stay
 427at the root.  The value used is usually the current root directory of
 428the calling process.  An alternate root can be provided as when
 429``sysctl()`` calls ``file_open_root()``, and when NFSv4 or Btrfs call
 430``mount_subtree()``.  In each case a pathname is being looked up in a very
 431specific part of the filesystem, and the lookup must not be allowed to
 432escape that subtree.  It works a bit like a local ``chroot()``.
 433
 434Ignoring the handling of symbolic links, we can now describe the
 435"``link_path_walk()``" function, which handles the lookup of everything
 436except the final component as:
 437
 438   Given a path (``name``) and a nameidata structure (``nd``), check that the
 439   current directory has execute permission and then advance ``name``
 440   over one component while updating ``last_type`` and ``last``.  If that
 441   was the final component, then return, otherwise call
 442   ``walk_component()`` and repeat from the top.
 443
 444``walk_component()`` is even easier.  If the component is ``LAST_DOTS``,
 445it calls ``handle_dots()`` which does the necessary locking as already
 446described.  If it finds a ``LAST_NORM`` component it first calls
 447"``lookup_fast()``" which only looks in the dcache, but will ask the
 448filesystem to revalidate the result if it is that sort of filesystem.
 449If that doesn't get a good result, it calls "``lookup_slow()``" which
 450takes ``i_rwsem``, rechecks the cache, and then asks the filesystem
 451to find a definitive answer.  Each of these will call
 452``follow_managed()`` (as described below) to handle any mount points.
 453
 454In the absence of symbolic links, ``walk_component()`` creates a new
 455``struct path`` containing a counted reference to the new dentry and a
 456reference to the new ``vfsmount`` which is only counted if it is
 457different from the previous ``vfsmount``.  It then calls
 458``path_to_nameidata()`` to install the new ``struct path`` in the
 459``struct nameidata`` and drop the unneeded references.
 460
 461This "hand-over-hand" sequencing of getting a reference to the new
 462dentry before dropping the reference to the previous dentry may
 463seem obvious, but is worth pointing out so that we will recognize its
 464analogue in the "RCU-walk" version.
 465
 466Handling the final component
 467----------------------------
 468
 469``link_path_walk()`` only walks as far as setting ``nd->last`` and
 470``nd->last_type`` to refer to the final component of the path.  It does
 471not call ``walk_component()`` that last time.  Handling that final
 472component remains for the caller to sort out. Those callers are
 473``path_lookupat()``, ``path_parentat()``, ``path_mountpoint()`` and
 474``path_openat()`` each of which handles the differing requirements of
 475different system calls.
 476
 477``path_parentat()`` is clearly the simplest - it just wraps a little bit
 478of housekeeping around ``link_path_walk()`` and returns the parent
 479directory and final component to the caller.  The caller will be either
 480aiming to create a name (via ``filename_create()``) or remove or rename
 481a name (in which case ``user_path_parent()`` is used).  They will use
 482``i_rwsem`` to exclude other changes while they validate and then
 483perform their operation.
 484
 485``path_lookupat()`` is nearly as simple - it is used when an existing
 486object is wanted such as by ``stat()`` or ``chmod()``.  It essentially just
 487calls ``walk_component()`` on the final component through a call to
 488``lookup_last()``.  ``path_lookupat()`` returns just the final dentry.
 489
 490``path_mountpoint()`` handles the special case of unmounting which must
 491not try to revalidate the mounted filesystem.  It effectively
 492contains, through a call to ``mountpoint_last()``, an alternate
 493implementation of ``lookup_slow()`` which skips that step.  This is
 494important when unmounting a filesystem that is inaccessible, such as
 495one provided by a dead NFS server.
 496
 497Finally ``path_openat()`` is used for the ``open()`` system call; it
 498contains, in support functions starting with "``do_last()``", all the
 499complexity needed to handle the different subtleties of O_CREAT (with
 500or without O_EXCL), final "``/``" characters, and trailing symbolic
 501links.  We will revisit this in the final part of this series, which
 502focuses on those symbolic links.  "``do_last()``" will sometimes, but
 503not always, take ``i_rwsem``, depending on what it finds.
 504
 505Each of these, or the functions which call them, need to be alert to
 506the possibility that the final component is not ``LAST_NORM``.  If the
 507goal of the lookup is to create something, then any value for
 508``last_type`` other than ``LAST_NORM`` will result in an error.  For
 509example if ``path_parentat()`` reports ``LAST_DOTDOT``, then the caller
 510won't try to create that name.  They also check for trailing slashes
 511by testing ``last.name[last.len]``.  If there is any character beyond
 512the final component, it must be a trailing slash.
 513
 514Revalidation and automounts
 515---------------------------
 516
 517Apart from symbolic links, there are only two parts of the "REF-walk"
 518process not yet covered.  One is the handling of stale cache entries
 519and the other is automounts.
 520
 521On filesystems that require it, the lookup routines will call the
 522``->d_revalidate()`` dentry method to ensure that the cached information
 523is current.  This will often confirm validity or update a few details
 524from a server.  In some cases it may find that there has been change
 525further up the path and that something that was thought to be valid
 526previously isn't really.  When this happens the lookup of the whole
 527path is aborted and retried with the "``LOOKUP_REVAL``" flag set.  This
 528forces revalidation to be more thorough.  We will see more details of
 529this retry process in the next article.
 530
 531Automount points are locations in the filesystem where an attempt to
 532lookup a name can trigger changes to how that lookup should be
 533handled, in particular by mounting a filesystem there.  These are
 534covered in greater detail in autofs.txt in the Linux documentation
 535tree, but a few notes specifically related to path lookup are in order
 536here.
 537
 538The Linux VFS has a concept of "managed" dentries which is reflected
 539in function names such as "``follow_managed()``".  There are three
 540potentially interesting things about these dentries corresponding
 541to three different flags that might be set in ``dentry->d_flags``:
 542
 543``DCACHE_MANAGE_TRANSIT``
 544~~~~~~~~~~~~~~~~~~~~~~~~~
 545
 546If this flag has been set, then the filesystem has requested that the
 547``d_manage()`` dentry operation be called before handling any possible
 548mount point.  This can perform two particular services:
 549
 550It can block to avoid races.  If an automount point is being
 551unmounted, the ``d_manage()`` function will usually wait for that
 552process to complete before letting the new lookup proceed and possibly
 553trigger a new automount.
 554
 555It can selectively allow only some processes to transit through a
 556mount point.  When a server process is managing automounts, it may
 557need to access a directory without triggering normal automount
 558processing.  That server process can identify itself to the ``autofs``
 559filesystem, which will then give it a special pass through
 560``d_manage()`` by returning ``-EISDIR``.
 561
 562``DCACHE_MOUNTED``
 563~~~~~~~~~~~~~~~~~~
 564
 565This flag is set on every dentry that is mounted on.  As Linux
 566supports multiple filesystem namespaces, it is possible that the
 567dentry may not be mounted on in *this* namespace, just in some
 568other.  So this flag is seen as a hint, not a promise.
 569
 570If this flag is set, and ``d_manage()`` didn't return ``-EISDIR``,
 571``lookup_mnt()`` is called to examine the mount hash table (honoring the
 572``mount_lock`` described earlier) and possibly return a new ``vfsmount``
 573and a new ``dentry`` (both with counted references).
 574
 575``DCACHE_NEED_AUTOMOUNT``
 576~~~~~~~~~~~~~~~~~~~~~~~~~
 577
 578If ``d_manage()`` allowed us to get this far, and ``lookup_mnt()`` didn't
 579find a mount point, then this flag causes the ``d_automount()`` dentry
 580operation to be called.
 581
 582The ``d_automount()`` operation can be arbitrarily complex and may
 583communicate with server processes etc. but it should ultimately either
 584report that there was an error, that there was nothing to mount, or
 585should provide an updated ``struct path`` with new ``dentry`` and ``vfsmount``.
 586
 587In the latter case, ``finish_automount()`` will be called to safely
 588install the new mount point into the mount table.
 589
 590There is no new locking of import here and it is important that no
 591locks (only counted references) are held over this processing due to
 592the very real possibility of extended delays.
 593This will become more important next time when we examine RCU-walk
 594which is particularly sensitive to delays.
 595
 596RCU-walk - faster pathname lookup in Linux
 597==========================================
 598
 599RCU-walk is another algorithm for performing pathname lookup in Linux.
 600It is in many ways similar to REF-walk and the two share quite a bit
 601of code.  The significant difference in RCU-walk is how it allows for
 602the possibility of concurrent access.
 603
 604We noted that REF-walk is complex because there are numerous details
 605and special cases.  RCU-walk reduces this complexity by simply
 606refusing to handle a number of cases -- it instead falls back to
 607REF-walk.  The difficulty with RCU-walk comes from a different
 608direction: unfamiliarity.  The locking rules when depending on RCU are
 609quite different from traditional locking, so we will spend a little extra
 610time when we come to those.
 611
 612Clear demarcation of roles
 613--------------------------
 614
 615The easiest way to manage concurrency is to forcibly stop any other
 616thread from changing the data structures that a given thread is
 617looking at.  In cases where no other thread would even think of
 618changing the data and lots of different threads want to read at the
 619same time, this can be very costly.  Even when using locks that permit
 620multiple concurrent readers, the simple act of updating the count of
 621the number of current readers can impose an unwanted cost.  So the
 622goal when reading a shared data structure that no other process is
 623changing is to avoid writing anything to memory at all.  Take no
 624locks, increment no counts, leave no footprints.
 625
 626The REF-walk mechanism already described certainly doesn't follow this
 627principle, but then it is really designed to work when there may well
 628be other threads modifying the data.  RCU-walk, in contrast, is
 629designed for the common situation where there are lots of frequent
 630readers and only occasional writers.  This may not be common in all
 631parts of the filesystem tree, but in many parts it will be.  For the
 632other parts it is important that RCU-walk can quickly fall back to
 633using REF-walk.
 634
 635Pathname lookup always starts in RCU-walk mode but only remains there
 636as long as what it is looking for is in the cache and is stable.  It
 637dances lightly down the cached filesystem image, leaving no footprints
 638and carefully watching where it is, to be sure it doesn't trip.  If it
 639notices that something has changed or is changing, or if something
 640isn't in the cache, then it tries to stop gracefully and switch to
 641REF-walk.
 642
 643This stopping requires getting a counted reference on the current
 644``vfsmount`` and ``dentry``, and ensuring that these are still valid -
 645that a path walk with REF-walk would have found the same entries.
 646This is an invariant that RCU-walk must guarantee.  It can only make
 647decisions, such as selecting the next step, that are decisions which
 648REF-walk could also have made if it were walking down the tree at the
 649same time.  If the graceful stop succeeds, the rest of the path is
 650processed with the reliable, if slightly sluggish, REF-walk.  If
 651RCU-walk finds it cannot stop gracefully, it simply gives up and
 652restarts from the top with REF-walk.
 653
 654This pattern of "try RCU-walk, if that fails try REF-walk" can be
 655clearly seen in functions like ``filename_lookup()``,
 656``filename_parentat()``, ``filename_mountpoint()``,
 657``do_filp_open()``, and ``do_file_open_root()``.  These five
 658correspond roughly to the four ``path_*()`` functions we met earlier,
 659each of which calls ``link_path_walk()``.  The ``path_*()`` functions are
 660called using different mode flags until a mode is found which works.
 661They are first called with ``LOOKUP_RCU`` set to request "RCU-walk".  If
 662that fails with the error ``ECHILD`` they are called again with no
 663special flag to request "REF-walk".  If either of those report the
 664error ``ESTALE`` a final attempt is made with ``LOOKUP_REVAL`` set (and no
 665``LOOKUP_RCU``) to ensure that entries found in the cache are forcibly
 666revalidated - normally entries are only revalidated if the filesystem
 667determines that they are too old to trust.
 668
 669The ``LOOKUP_RCU`` attempt may drop that flag internally and switch to
 670REF-walk, but will never then try to switch back to RCU-walk.  Places
 671that trip up RCU-walk are much more likely to be near the leaves and
 672so it is very unlikely that there will be much, if any, benefit from
 673switching back.
 674
 675RCU and seqlocks: fast and light
 676--------------------------------
 677
 678RCU is, unsurprisingly, critical to RCU-walk mode.  The
 679``rcu_read_lock()`` is held for the entire time that RCU-walk is walking
 680down a path.  The particular guarantee it provides is that the key
 681data structures - dentries, inodes, super_blocks, and mounts - will
 682not be freed while the lock is held.  They might be unlinked or
 683invalidated in one way or another, but the memory will not be
 684repurposed so values in various fields will still be meaningful.  This
 685is the only guarantee that RCU provides; everything else is done using
 686seqlocks.
 687
 688As we saw above, REF-walk holds a counted reference to the current
 689dentry and the current vfsmount, and does not release those references
 690before taking references to the "next" dentry or vfsmount.  It also
 691sometimes takes the ``d_lock`` spinlock.  These references and locks are
 692taken to prevent certain changes from happening.  RCU-walk must not
 693take those references or locks and so cannot prevent such changes.
 694Instead, it checks to see if a change has been made, and aborts or
 695retries if it has.
 696
 697To preserve the invariant mentioned above (that RCU-walk may only make
 698decisions that REF-walk could have made), it must make the checks at
 699or near the same places that REF-walk holds the references.  So, when
 700REF-walk increments a reference count or takes a spinlock, RCU-walk
 701samples the status of a seqlock using ``read_seqcount_begin()`` or a
 702similar function.  When REF-walk decrements the count or drops the
 703lock, RCU-walk checks if the sampled status is still valid using
 704``read_seqcount_retry()`` or similar.
 705
 706However, there is a little bit more to seqlocks than that.  If
 707RCU-walk accesses two different fields in a seqlock-protected
 708structure, or accesses the same field twice, there is no a priori
 709guarantee of any consistency between those accesses.  When consistency
 710is needed - which it usually is - RCU-walk must take a copy and then
 711use ``read_seqcount_retry()`` to validate that copy.
 712
 713``read_seqcount_retry()`` not only checks the sequence number, but also
 714imposes a memory barrier so that no memory-read instruction from
 715*before* the call can be delayed until *after* the call, either by the
 716CPU or by the compiler.  A simple example of this can be seen in
 717``slow_dentry_cmp()`` which, for filesystems which do not use simple
 718byte-wise name equality, calls into the filesystem to compare a name
 719against a dentry.  The length and name pointer are copied into local
 720variables, then ``read_seqcount_retry()`` is called to confirm the two
 721are consistent, and only then is ``->d_compare()`` called.  When
 722standard filename comparison is used, ``dentry_cmp()`` is called
 723instead.  Notably it does *not* use ``read_seqcount_retry()``, but
 724instead has a large comment explaining why the consistency guarantee
 725isn't necessary.  A subsequent ``read_seqcount_retry()`` will be
 726sufficient to catch any problem that could occur at this point.
 727
 728With that little refresher on seqlocks out of the way we can look at
 729the bigger picture of how RCU-walk uses seqlocks.
 730
 731``mount_lock`` and ``nd->m_seq``
 732~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 733
 734We already met the ``mount_lock`` seqlock when REF-walk used it to
 735ensure that crossing a mount point is performed safely.  RCU-walk uses
 736it for that too, but for quite a bit more.
 737
 738Instead of taking a counted reference to each ``vfsmount`` as it
 739descends the tree, RCU-walk samples the state of ``mount_lock`` at the
 740start of the walk and stores this initial sequence number in the
 741``struct nameidata`` in the ``m_seq`` field.  This one lock and one
 742sequence number are used to validate all accesses to all ``vfsmounts``,
 743and all mount point crossings.  As changes to the mount table are
 744relatively rare, it is reasonable to fall back on REF-walk any time
 745that any "mount" or "unmount" happens.
 746
 747``m_seq`` is checked (using ``read_seqretry()``) at the end of an RCU-walk
 748sequence, whether switching to REF-walk for the rest of the path or
 749when the end of the path is reached.  It is also checked when stepping
 750down over a mount point (in ``__follow_mount_rcu()``) or up (in
 751``follow_dotdot_rcu()``).  If it is ever found to have changed, the
 752whole RCU-walk sequence is aborted and the path is processed again by
 753REF-walk.
 754
 755If RCU-walk finds that ``mount_lock`` hasn't changed then it can be sure
 756that, had REF-walk taken counted references on each vfsmount, the
 757results would have been the same.  This ensures the invariant holds,
 758at least for vfsmount structures.
 759
 760``dentry->d_seq`` and ``nd->seq``
 761~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 762
 763In place of taking a count or lock on ``d_reflock``, RCU-walk samples
 764the per-dentry ``d_seq`` seqlock, and stores the sequence number in the
 765``seq`` field of the nameidata structure, so ``nd->seq`` should always be
 766the current sequence number of ``nd->dentry``.  This number needs to be
 767revalidated after copying, and before using, the name, parent, or
 768inode of the dentry.
 769
 770The handling of the name we have already looked at, and the parent is
 771only accessed in ``follow_dotdot_rcu()`` which fairly trivially follows
 772the required pattern, though it does so for three different cases.
 773
 774When not at a mount point, ``d_parent`` is followed and its ``d_seq`` is
 775collected.  When we are at a mount point, we instead follow the
 776``mnt->mnt_mountpoint`` link to get a new dentry and collect its
 777``d_seq``.  Then, after finally finding a ``d_parent`` to follow, we must
 778check if we have landed on a mount point and, if so, must find that
 779mount point and follow the ``mnt->mnt_root`` link.  This would imply a
 780somewhat unusual, but certainly possible, circumstance where the
 781starting point of the path lookup was in part of the filesystem that
 782was mounted on, and so not visible from the root.
 783
 784The inode pointer, stored in ``->d_inode``, is a little more
 785interesting.  The inode will always need to be accessed at least
 786twice, once to determine if it is NULL and once to verify access
 787permissions.  Symlink handling requires a validated inode pointer too.
 788Rather than revalidating on each access, a copy is made on the first
 789access and it is stored in the ``inode`` field of ``nameidata`` from where
 790it can be safely accessed without further validation.
 791
 792``lookup_fast()`` is the only lookup routine that is used in RCU-mode,
 793``lookup_slow()`` being too slow and requiring locks.  It is in
 794``lookup_fast()`` that we find the important "hand over hand" tracking
 795of the current dentry.
 796
 797The current ``dentry`` and current ``seq`` number are passed to
 798``__d_lookup_rcu()`` which, on success, returns a new ``dentry`` and a
 799new ``seq`` number.  ``lookup_fast()`` then copies the inode pointer and
 800revalidates the new ``seq`` number.  It then validates the old ``dentry``
 801with the old ``seq`` number one last time and only then continues.  This
 802process of getting the ``seq`` number of the new dentry and then
 803checking the ``seq`` number of the old exactly mirrors the process of
 804getting a counted reference to the new dentry before dropping that for
 805the old dentry which we saw in REF-walk.
 806
 807No ``inode->i_rwsem`` or even ``rename_lock``
 808~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 809
 810A semaphore is a fairly heavyweight lock that can only be taken when it is
 811permissible to sleep.  As ``rcu_read_lock()`` forbids sleeping,
 812``inode->i_rwsem`` plays no role in RCU-walk.  If some other thread does
 813take ``i_rwsem`` and modifies the directory in a way that RCU-walk needs
 814to notice, the result will be either that RCU-walk fails to find the
 815dentry that it is looking for, or it will find a dentry which
 816``read_seqretry()`` won't validate.  In either case it will drop down to
 817REF-walk mode which can take whatever locks are needed.
 818
 819Though ``rename_lock`` could be used by RCU-walk as it doesn't require
 820any sleeping, RCU-walk doesn't bother.  REF-walk uses ``rename_lock`` to
 821protect against the possibility of hash chains in the dcache changing
 822while they are being searched.  This can result in failing to find
 823something that actually is there.  When RCU-walk fails to find
 824something in the dentry cache, whether it is really there or not, it
 825already drops down to REF-walk and tries again with appropriate
 826locking.  This neatly handles all cases, so adding extra checks on
 827rename_lock would bring no significant value.
 828
 829``unlazy walk()`` and ``complete_walk()``
 830-----------------------------------------
 831
 832That "dropping down to REF-walk" typically involves a call to
 833``unlazy_walk()``, so named because "RCU-walk" is also sometimes
 834referred to as "lazy walk".  ``unlazy_walk()`` is called when
 835following the path down to the current vfsmount/dentry pair seems to
 836have proceeded successfully, but the next step is problematic.  This
 837can happen if the next name cannot be found in the dcache, if
 838permission checking or name revalidation couldn't be achieved while
 839the ``rcu_read_lock()`` is held (which forbids sleeping), if an
 840automount point is found, or in a couple of cases involving symlinks.
 841It is also called from ``complete_walk()`` when the lookup has reached
 842the final component, or the very end of the path, depending on which
 843particular flavor of lookup is used.
 844
 845Other reasons for dropping out of RCU-walk that do not trigger a call
 846to ``unlazy_walk()`` are when some inconsistency is found that cannot be
 847handled immediately, such as ``mount_lock`` or one of the ``d_seq``
 848seqlocks reporting a change.  In these cases the relevant function
 849will return ``-ECHILD`` which will percolate up until it triggers a new
 850attempt from the top using REF-walk.
 851
 852For those cases where ``unlazy_walk()`` is an option, it essentially
 853takes a reference on each of the pointers that it holds (vfsmount,
 854dentry, and possibly some symbolic links) and then verifies that the
 855relevant seqlocks have not been changed.  If there have been changes,
 856it, too, aborts with ``-ECHILD``, otherwise the transition to REF-walk
 857has been a success and the lookup process continues.
 858
 859Taking a reference on those pointers is not quite as simple as just
 860incrementing a counter.  That works to take a second reference if you
 861already have one (often indirectly through another object), but it
 862isn't sufficient if you don't actually have a counted reference at
 863all.  For ``dentry->d_lockref``, it is safe to increment the reference
 864counter to get a reference unless it has been explicitly marked as
 865"dead" which involves setting the counter to ``-128``.
 866``lockref_get_not_dead()`` achieves this.
 867
 868For ``mnt->mnt_count`` it is safe to take a reference as long as
 869``mount_lock`` is then used to validate the reference.  If that
 870validation fails, it may *not* be safe to just drop that reference in
 871the standard way of calling ``mnt_put()`` - an unmount may have
 872progressed too far.  So the code in ``legitimize_mnt()``, when it
 873finds that the reference it got might not be safe, checks the
 874``MNT_SYNC_UMOUNT`` flag to determine if a simple ``mnt_put()`` is
 875correct, or if it should just decrement the count and pretend none of
 876this ever happened.
 877
 878Taking care in filesystems
 879--------------------------
 880
 881RCU-walk depends almost entirely on cached information and often will
 882not call into the filesystem at all.  However there are two places,
 883besides the already-mentioned component-name comparison, where the
 884file system might be included in RCU-walk, and it must know to be
 885careful.
 886
 887If the filesystem has non-standard permission-checking requirements -
 888such as a networked filesystem which may need to check with the server
 889- the ``i_op->permission`` interface might be called during RCU-walk.
 890In this case an extra "``MAY_NOT_BLOCK``" flag is passed so that it
 891knows not to sleep, but to return ``-ECHILD`` if it cannot complete
 892promptly.  ``i_op->permission`` is given the inode pointer, not the
 893dentry, so it doesn't need to worry about further consistency checks.
 894However if it accesses any other filesystem data structures, it must
 895ensure they are safe to be accessed with only the ``rcu_read_lock()``
 896held.  This typically means they must be freed using ``kfree_rcu()`` or
 897similar.
 898
 899.. _READ_ONCE: https://lwn.net/Articles/624126/
 900
 901If the filesystem may need to revalidate dcache entries, then
 902``d_op->d_revalidate`` may be called in RCU-walk too.  This interface
 903*is* passed the dentry but does not have access to the ``inode`` or the
 904``seq`` number from the ``nameidata``, so it needs to be extra careful
 905when accessing fields in the dentry.  This "extra care" typically
 906involves using  `READ_ONCE() <READ_ONCE_>`_ to access fields, and verifying the
 907result is not NULL before using it.  This pattern can be seen in
 908``nfs_lookup_revalidate()``.
 909
 910A pair of patterns
 911------------------
 912
 913In various places in the details of REF-walk and RCU-walk, and also in
 914the big picture, there are a couple of related patterns that are worth
 915being aware of.
 916
 917The first is "try quickly and check, if that fails try slowly".  We
 918can see that in the high-level approach of first trying RCU-walk and
 919then trying REF-walk, and in places where ``unlazy_walk()`` is used to
 920switch to REF-walk for the rest of the path.  We also saw it earlier
 921in ``dget_parent()`` when following a "``..``" link.  It tries a quick way
 922to get a reference, then falls back to taking locks if needed.
 923
 924The second pattern is "try quickly and check, if that fails try
 925again - repeatedly".  This is seen with the use of ``rename_lock`` and
 926``mount_lock`` in REF-walk.  RCU-walk doesn't make use of this pattern -
 927if anything goes wrong it is much safer to just abort and try a more
 928sedate approach.
 929
 930The emphasis here is "try quickly and check".  It should probably be
 931"try quickly *and carefully*, then check".  The fact that checking is
 932needed is a reminder that the system is dynamic and only a limited
 933number of things are safe at all.  The most likely cause of errors in
 934this whole process is assuming something is safe when in reality it
 935isn't.  Careful consideration of what exactly guarantees the safety of
 936each access is sometimes necessary.
 937
 938A walk among the symlinks
 939=========================
 940
 941There are several basic issues that we will examine to understand the
 942handling of symbolic links:  the symlink stack, together with cache
 943lifetimes, will help us understand the overall recursive handling of
 944symlinks and lead to the special care needed for the final component.
 945Then a consideration of access-time updates and summary of the various
 946flags controlling lookup will finish the story.
 947
 948The symlink stack
 949-----------------
 950
 951There are only two sorts of filesystem objects that can usefully
 952appear in a path prior to the final component: directories and symlinks.
 953Handling directories is quite straightforward: the new directory
 954simply becomes the starting point at which to interpret the next
 955component on the path.  Handling symbolic links requires a bit more
 956work.
 957
 958Conceptually, symbolic links could be handled by editing the path.  If
 959a component name refers to a symbolic link, then that component is
 960replaced by the body of the link and, if that body starts with a '/',
 961then all preceding parts of the path are discarded.  This is what the
 962"``readlink -f``" command does, though it also edits out "``.``" and
 963"``..``" components.
 964
 965Directly editing the path string is not really necessary when looking
 966up a path, and discarding early components is pointless as they aren't
 967looked at anyway.  Keeping track of all remaining components is
 968important, but they can of course be kept separately; there is no need
 969to concatenate them.  As one symlink may easily refer to another,
 970which in turn can refer to a third, we may need to keep the remaining
 971components of several paths, each to be processed when the preceding
 972ones are completed.  These path remnants are kept on a stack of
 973limited size.
 974
 975There are two reasons for placing limits on how many symlinks can
 976occur in a single path lookup.  The most obvious is to avoid loops.
 977If a symlink referred to itself either directly or through
 978intermediaries, then following the symlink can never complete
 979successfully - the error ``ELOOP`` must be returned.  Loops can be
 980detected without imposing limits, but limits are the simplest solution
 981and, given the second reason for restriction, quite sufficient.
 982
 983.. _outlined recently: http://thread.gmane.org/gmane.linux.kernel/1934390/focus=1934550
 984
 985The second reason was `outlined recently`_ by Linus:
 986
 987   Because it's a latency and DoS issue too. We need to react well to
 988   true loops, but also to "very deep" non-loops. It's not about memory
 989   use, it's about users triggering unreasonable CPU resources.
 990
 991Linux imposes a limit on the length of any pathname: ``PATH_MAX``, which
 992is 4096.  There are a number of reasons for this limit; not letting the
 993kernel spend too much time on just one path is one of them.  With
 994symbolic links you can effectively generate much longer paths so some
 995sort of limit is needed for the same reason.  Linux imposes a limit of
 996at most 40 symlinks in any one path lookup.  It previously imposed a
 997further limit of eight on the maximum depth of recursion, but that was
 998raised to 40 when a separate stack was implemented, so there is now
 999just the one limit.
1000
1001The ``nameidata`` structure that we met in an earlier article contains a
1002small stack that can be used to store the remaining part of up to two
1003symlinks.  In many cases this will be sufficient.  If it isn't, a
1004separate stack is allocated with room for 40 symlinks.  Pathname
1005lookup will never exceed that stack as, once the 40th symlink is
1006detected, an error is returned.
1007
1008It might seem that the name remnants are all that needs to be stored on
1009this stack, but we need a bit more.  To see that, we need to move on to
1010cache lifetimes.
1011
1012Storage and lifetime of cached symlinks
1013---------------------------------------
1014
1015Like other filesystem resources, such as inodes and directory
1016entries, symlinks are cached by Linux to avoid repeated costly access
1017to external storage.  It is particularly important for RCU-walk to be
1018able to find and temporarily hold onto these cached entries, so that
1019it doesn't need to drop down into REF-walk.
1020
1021.. _object-oriented design pattern: https://lwn.net/Articles/446317/
1022
1023While each filesystem is free to make its own choice, symlinks are
1024typically stored in one of two places.  Short symlinks are often
1025stored directly in the inode.  When a filesystem allocates a ``struct
1026inode`` it typically allocates extra space to store private data (a
1027common `object-oriented design pattern`_ in the kernel).  This will
1028sometimes include space for a symlink.  The other common location is
1029in the page cache, which normally stores the content of files.  The
1030pathname in a symlink can be seen as the content of that symlink and
1031can easily be stored in the page cache just like file content.
1032
1033When neither of these is suitable, the next most likely scenario is
1034that the filesystem will allocate some temporary memory and copy or
1035construct the symlink content into that memory whenever it is needed.
1036
1037When the symlink is stored in the inode, it has the same lifetime as
1038the inode which, itself, is protected by RCU or by a counted reference
1039on the dentry.  This means that the mechanisms that pathname lookup
1040uses to access the dcache and icache (inode cache) safely are quite
1041sufficient for accessing some cached symlinks safely.  In these cases,
1042the ``i_link`` pointer in the inode is set to point to wherever the
1043symlink is stored and it can be accessed directly whenever needed.
1044
1045When the symlink is stored in the page cache or elsewhere, the
1046situation is not so straightforward.  A reference on a dentry or even
1047on an inode does not imply any reference on cached pages of that
1048inode, and even an ``rcu_read_lock()`` is not sufficient to ensure that
1049a page will not disappear.  So for these symlinks the pathname lookup
1050code needs to ask the filesystem to provide a stable reference and,
1051significantly, needs to release that reference when it is finished
1052with it.
1053
1054Taking a reference to a cache page is often possible even in RCU-walk
1055mode.  It does require making changes to memory, which is best avoided,
1056but that isn't necessarily a big cost and it is better than dropping
1057out of RCU-walk mode completely.  Even filesystems that allocate
1058space to copy the symlink into can use ``GFP_ATOMIC`` to often successfully
1059allocate memory without the need to drop out of RCU-walk.  If a
1060filesystem cannot successfully get a reference in RCU-walk mode, it
1061must return ``-ECHILD`` and ``unlazy_walk()`` will be called to return to
1062REF-walk mode in which the filesystem is allowed to sleep.
1063
1064The place for all this to happen is the ``i_op->follow_link()`` inode
1065method.  In the present mainline code this is never actually called in
1066RCU-walk mode as the rewrite is not quite complete.  It is likely that
1067in a future release this method will be passed an ``inode`` pointer when
1068called in RCU-walk mode so it both (1) knows to be careful, and (2) has the
1069validated pointer.  Much like the ``i_op->permission()`` method we
1070looked at previously, ``->follow_link()`` would need to be careful that
1071all the data structures it references are safe to be accessed while
1072holding no counted reference, only the RCU lock.  Though getting a
1073reference with ``->follow_link()`` is not yet done in RCU-walk mode, the
1074code is ready to release the reference when that does happen.
1075
1076This need to drop the reference to a symlink adds significant
1077complexity.  It requires a reference to the inode so that the
1078``i_op->put_link()`` inode operation can be called.  In REF-walk, that
1079reference is kept implicitly through a reference to the dentry, so
1080keeping the ``struct path`` of the symlink is easiest.  For RCU-walk,
1081the pointer to the inode is kept separately.  To allow switching from
1082RCU-walk back to REF-walk in the middle of processing nested symlinks
1083we also need the seq number for the dentry so we can confirm that
1084switching back was safe.
1085
1086Finally, when providing a reference to a symlink, the filesystem also
1087provides an opaque "cookie" that must be passed to ``->put_link()`` so that it
1088knows what to free.  This might be the allocated memory area, or a
1089pointer to the ``struct page`` in the page cache, or something else
1090completely.  Only the filesystem knows what it is.
1091
1092In order for the reference to each symlink to be dropped when the walk completes,
1093whether in RCU-walk or REF-walk, the symlink stack needs to contain,
1094along with the path remnants:
1095
1096- the ``struct path`` to provide a reference to the inode in REF-walk
1097- the ``struct inode *`` to provide a reference to the inode in RCU-walk
1098- the ``seq`` to allow the path to be safely switched from RCU-walk to REF-walk
1099- the ``cookie`` that tells ``->put_path()`` what to put.
1100
1101This means that each entry in the symlink stack needs to hold five
1102pointers and an integer instead of just one pointer (the path
1103remnant).  On a 64-bit system, this is about 40 bytes per entry;
1104with 40 entries it adds up to 1600 bytes total, which is less than
1105half a page.  So it might seem like a lot, but is by no means
1106excessive.
1107
1108Note that, in a given stack frame, the path remnant (``name``) is not
1109part of the symlink that the other fields refer to.  It is the remnant
1110to be followed once that symlink has been fully parsed.
1111
1112Following the symlink
1113---------------------
1114
1115The main loop in ``link_path_walk()`` iterates seamlessly over all
1116components in the path and all of the non-final symlinks.  As symlinks
1117are processed, the ``name`` pointer is adjusted to point to a new
1118symlink, or is restored from the stack, so that much of the loop
1119doesn't need to notice.  Getting this ``name`` variable on and off the
1120stack is very straightforward; pushing and popping the references is
1121a little more complex.
1122
1123When a symlink is found, ``walk_component()`` returns the value ``1``
1124(``0`` is returned for any other sort of success, and a negative number
1125is, as usual, an error indicator).  This causes ``get_link()`` to be
1126called; it then gets the link from the filesystem.  Providing that
1127operation is successful, the old path ``name`` is placed on the stack,
1128and the new value is used as the ``name`` for a while.  When the end of
1129the path is found (i.e. ``*name`` is ``'\0'``) the old ``name`` is restored
1130off the stack and path walking continues.
1131
1132Pushing and popping the reference pointers (inode, cookie, etc.) is more
1133complex in part because of the desire to handle tail recursion.  When
1134the last component of a symlink itself points to a symlink, we
1135want to pop the symlink-just-completed off the stack before pushing
1136the symlink-just-found to avoid leaving empty path remnants that would
1137just get in the way.
1138
1139It is most convenient to push the new symlink references onto the
1140stack in ``walk_component()`` immediately when the symlink is found;
1141``walk_component()`` is also the last piece of code that needs to look at the
1142old symlink as it walks that last component.  So it is quite
1143convenient for ``walk_component()`` to release the old symlink and pop
1144the references just before pushing the reference information for the
1145new symlink.  It is guided in this by two flags; ``WALK_GET``, which
1146gives it permission to follow a symlink if it finds one, and
1147``WALK_PUT``, which tells it to release the current symlink after it has been
1148followed.  ``WALK_PUT`` is tested first, leading to a call to
1149``put_link()``.  ``WALK_GET`` is tested subsequently (by
1150``should_follow_link()``) leading to a call to ``pick_link()`` which sets
1151up the stack frame.
1152
1153Symlinks with no final component
1154~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1155
1156A pair of special-case symlinks deserve a little further explanation.
1157Both result in a new ``struct path`` (with mount and dentry) being set
1158up in the ``nameidata``, and result in ``get_link()`` returning ``NULL``.
1159
1160The more obvious case is a symlink to "``/``".  All symlinks starting
1161with "``/``" are detected in ``get_link()`` which resets the ``nameidata``
1162to point to the effective filesystem root.  If the symlink only
1163contains "``/``" then there is nothing more to do, no components at all,
1164so ``NULL`` is returned to indicate that the symlink can be released and
1165the stack frame discarded.
1166
1167The other case involves things in ``/proc`` that look like symlinks but
1168aren't really (and are therefore commonly referred to as "magic-links")::
1169
1170     $ ls -l /proc/self/fd/1
1171     lrwx------ 1 neilb neilb 64 Jun 13 10:19 /proc/self/fd/1 -> /dev/pts/4
1172
1173Every open file descriptor in any process is represented in ``/proc`` by
1174something that looks like a symlink.  It is really a reference to the
1175target file, not just the name of it.  When you ``readlink`` these
1176objects you get a name that might refer to the same file - unless it
1177has been unlinked or mounted over.  When ``walk_component()`` follows
1178one of these, the ``->follow_link()`` method in "procfs" doesn't return
1179a string name, but instead calls ``nd_jump_link()`` which updates the
1180``nameidata`` in place to point to that target.  ``->follow_link()`` then
1181returns ``NULL``.  Again there is no final component and ``get_link()``
1182reports this by leaving the ``last_type`` field of ``nameidata`` as
1183``LAST_BIND``.
1184
1185Following the symlink in the final component
1186--------------------------------------------
1187
1188All this leads to ``link_path_walk()`` walking down every component, and
1189following all symbolic links it finds, until it reaches the final
1190component.  This is just returned in the ``last`` field of ``nameidata``.
1191For some callers, this is all they need; they want to create that
1192``last`` name if it doesn't exist or give an error if it does.  Other
1193callers will want to follow a symlink if one is found, and possibly
1194apply special handling to the last component of that symlink, rather
1195than just the last component of the original file name.  These callers
1196potentially need to call ``link_path_walk()`` again and again on
1197successive symlinks until one is found that doesn't point to another
1198symlink.
1199
1200This case is handled by the relevant caller of ``link_path_walk()``, such as
1201``path_lookupat()`` using a loop that calls ``link_path_walk()``, and then
1202handles the final component.  If the final component is a symlink
1203that needs to be followed, then ``trailing_symlink()`` is called to set
1204things up properly and the loop repeats, calling ``link_path_walk()``
1205again.  This could loop as many as 40 times if the last component of
1206each symlink is another symlink.
1207
1208The various functions that examine the final component and possibly
1209report that it is a symlink are ``lookup_last()``, ``mountpoint_last()``
1210and ``do_last()``, each of which use the same convention as
1211``walk_component()`` of returning ``1`` if a symlink was found that needs
1212to be followed.
1213
1214Of these, ``do_last()`` is the most interesting as it is used for
1215opening a file.  Part of ``do_last()`` runs with ``i_rwsem`` held and this
1216part is in a separate function: ``lookup_open()``.
1217
1218Explaining ``do_last()`` completely is beyond the scope of this article,
1219but a few highlights should help those interested in exploring the
1220code.
1221
12221. Rather than just finding the target file, ``do_last()`` needs to open
1223   it.  If the file was found in the dcache, then ``vfs_open()`` is used for
1224   this.  If not, then ``lookup_open()`` will either call ``atomic_open()`` (if
1225   the filesystem provides it) to combine the final lookup with the open, or
1226   will perform the separate ``lookup_real()`` and ``vfs_create()`` steps
1227   directly.  In the later case the actual "open" of this newly found or
1228   created file will be performed by ``vfs_open()``, just as if the name
1229   were found in the dcache.
1230
12312. ``vfs_open()`` can fail with ``-EOPENSTALE`` if the cached information
1232   wasn't quite current enough.  Rather than restarting the lookup from
1233   the top with ``LOOKUP_REVAL`` set, ``lookup_open()`` is called instead,
1234   giving the filesystem a chance to resolve small inconsistencies.
1235   If that doesn't work, only then is the lookup restarted from the top.
1236
12373. An open with O_CREAT **does** follow a symlink in the final component,
1238   unlike other creation system calls (like ``mkdir``).  So the sequence::
1239
1240          ln -s bar /tmp/foo
1241          echo hello > /tmp/foo
1242
1243   will create a file called ``/tmp/bar``.  This is not permitted if
1244   ``O_EXCL`` is set but otherwise is handled for an O_CREAT open much
1245   like for a non-creating open: ``should_follow_link()`` returns ``1``, and
1246   so does ``do_last()`` so that ``trailing_symlink()`` gets called and the
1247   open process continues on the symlink that was found.
1248
1249Updating the access time
1250------------------------
1251
1252We previously said of RCU-walk that it would "take no locks, increment
1253no counts, leave no footprints."  We have since seen that some
1254"footprints" can be needed when handling symlinks as a counted
1255reference (or even a memory allocation) may be needed.  But these
1256footprints are best kept to a minimum.
1257
1258One other place where walking down a symlink can involve leaving
1259footprints in a way that doesn't affect directories is in updating access times.
1260In Unix (and Linux) every filesystem object has a "last accessed
1261time", or "``atime``".  Passing through a directory to access a file
1262within is not considered to be an access for the purposes of
1263``atime``; only listing the contents of a directory can update its ``atime``.
1264Symlinks are different it seems.  Both reading a symlink (with ``readlink()``)
1265and looking up a symlink on the way to some other destination can
1266update the atime on that symlink.
1267
1268.. _clearest statement: https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap04.html#tag_04_08
1269
1270It is not clear why this is the case; POSIX has little to say on the
1271subject.  The `clearest statement`_ is that, if a particular implementation
1272updates a timestamp in a place not specified by POSIX, this must be
1273documented "except that any changes caused by pathname resolution need
1274not be documented".  This seems to imply that POSIX doesn't really
1275care about access-time updates during pathname lookup.
1276
1277.. _Linux 1.3.87: https://git.kernel.org/cgit/linux/kernel/git/history/history.git/diff/fs/ext2/symlink.c?id=f806c6db77b8eaa6e00dcfb6b567706feae8dbb8
1278
1279An examination of history shows that prior to `Linux 1.3.87`_, the ext2
1280filesystem, at least, didn't update atime when following a link.
1281Unfortunately we have no record of why that behavior was changed.
1282
1283In any case, access time must now be updated and that operation can be
1284quite complex.  Trying to stay in RCU-walk while doing it is best
1285avoided.  Fortunately it is often permitted to skip the ``atime``
1286update.  Because ``atime`` updates cause performance problems in various
1287areas, Linux supports the ``relatime`` mount option, which generally
1288limits the updates of ``atime`` to once per day on files that aren't
1289being changed (and symlinks never change once created).  Even without
1290``relatime``, many filesystems record ``atime`` with a one-second
1291granularity, so only one update per second is required.
1292
1293It is easy to test if an ``atime`` update is needed while in RCU-walk
1294mode and, if it isn't, the update can be skipped and RCU-walk mode
1295continues.  Only when an ``atime`` update is actually required does the
1296path walk drop down to REF-walk.  All of this is handled in the
1297``get_link()`` function.
1298
1299A few flags
1300-----------
1301
1302A suitable way to wrap up this tour of pathname walking is to list
1303the various flags that can be stored in the ``nameidata`` to guide the
1304lookup process.  Many of these are only meaningful on the final
1305component, others reflect the current state of the pathname lookup, and some
1306apply restrictions to all path components encountered in the path lookup.
1307
1308And then there is ``LOOKUP_EMPTY``, which doesn't fit conceptually with
1309the others.  If this is not set, an empty pathname causes an error
1310very early on.  If it is set, empty pathnames are not considered to be
1311an error.
1312
1313Global state flags
1314~~~~~~~~~~~~~~~~~~
1315
1316We have already met two global state flags: ``LOOKUP_RCU`` and
1317``LOOKUP_REVAL``.  These select between one of three overall approaches
1318to lookup: RCU-walk, REF-walk, and REF-walk with forced revalidation.
1319
1320``LOOKUP_PARENT`` indicates that the final component hasn't been reached
1321yet.  This is primarily used to tell the audit subsystem the full
1322context of a particular access being audited.
1323
1324``LOOKUP_ROOT`` indicates that the ``root`` field in the ``nameidata`` was
1325provided by the caller, so it shouldn't be released when it is no
1326longer needed.
1327
1328``LOOKUP_JUMPED`` means that the current dentry was chosen not because
1329it had the right name but for some other reason.  This happens when
1330following "``..``", following a symlink to ``/``, crossing a mount point
1331or accessing a "``/proc/$PID/fd/$FD``" symlink (also known as a "magic
1332link"). In this case the filesystem has not been asked to revalidate the
1333name (with ``d_revalidate()``).  In such cases the inode may still need
1334to be revalidated, so ``d_op->d_weak_revalidate()`` is called if
1335``LOOKUP_JUMPED`` is set when the look completes - which may be at the
1336final component or, when creating, unlinking, or renaming, at the penultimate component.
1337
1338Resolution-restriction flags
1339~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1340
1341In order to allow userspace to protect itself against certain race conditions
1342and attack scenarios involving changing path components, a series of flags are
1343available which apply restrictions to all path components encountered during
1344path lookup. These flags are exposed through ``openat2()``'s ``resolve`` field.
1345
1346``LOOKUP_NO_SYMLINKS`` blocks all symlink traversals (including magic-links).
1347This is distinctly different from ``LOOKUP_FOLLOW``, because the latter only
1348relates to restricting the following of trailing symlinks.
1349
1350``LOOKUP_NO_MAGICLINKS`` blocks all magic-link traversals. Filesystems must
1351ensure that they return errors from ``nd_jump_link()``, because that is how
1352``LOOKUP_NO_MAGICLINKS`` and other magic-link restrictions are implemented.
1353
1354``LOOKUP_NO_XDEV`` blocks all ``vfsmount`` traversals (this includes both
1355bind-mounts and ordinary mounts). Note that the ``vfsmount`` which contains the
1356lookup is determined by the first mountpoint the path lookup reaches --
1357absolute paths start with the ``vfsmount`` of ``/``, and relative paths start
1358with the ``dfd``'s ``vfsmount``. Magic-links are only permitted if the
1359``vfsmount`` of the path is unchanged.
1360
1361``LOOKUP_BENEATH`` blocks any path components which resolve outside the
1362starting point of the resolution. This is done by blocking ``nd_jump_root()``
1363as well as blocking ".." if it would jump outside the starting point.
1364``rename_lock`` and ``mount_lock`` are used to detect attacks against the
1365resolution of "..". Magic-links are also blocked.
1366
1367``LOOKUP_IN_ROOT`` resolves all path components as though the starting point
1368were the filesystem root. ``nd_jump_root()`` brings the resolution back to
1369the starting point, and ".." at the starting point will act as a no-op. As with
1370``LOOKUP_BENEATH``, ``rename_lock`` and ``mount_lock`` are used to detect
1371attacks against ".." resolution. Magic-links are also blocked.
1372
1373Final-component flags
1374~~~~~~~~~~~~~~~~~~~~~
1375
1376Some of these flags are only set when the final component is being
1377considered.  Others are only checked for when considering that final
1378component.
1379
1380``LOOKUP_AUTOMOUNT`` ensures that, if the final component is an automount
1381point, then the mount is triggered.  Some operations would trigger it
1382anyway, but operations like ``stat()`` deliberately don't.  ``statfs()``
1383needs to trigger the mount but otherwise behaves a lot like ``stat()``, so
1384it sets ``LOOKUP_AUTOMOUNT``, as does "``quotactl()``" and the handling of
1385"``mount --bind``".
1386
1387``LOOKUP_FOLLOW`` has a similar function to ``LOOKUP_AUTOMOUNT`` but for
1388symlinks.  Some system calls set or clear it implicitly, while
1389others have API flags such as ``AT_SYMLINK_FOLLOW`` and
1390``UMOUNT_NOFOLLOW`` to control it.  Its effect is similar to
1391``WALK_GET`` that we already met, but it is used in a different way.
1392
1393``LOOKUP_DIRECTORY`` insists that the final component is a directory.
1394Various callers set this and it is also set when the final component
1395is found to be followed by a slash.
1396
1397Finally ``LOOKUP_OPEN``, ``LOOKUP_CREATE``, ``LOOKUP_EXCL``, and
1398``LOOKUP_RENAME_TARGET`` are not used directly by the VFS but are made
1399available to the filesystem and particularly the ``->d_revalidate()``
1400method.  A filesystem can choose not to bother revalidating too hard
1401if it knows that it will be asked to open or create the file soon.
1402These flags were previously useful for ``->lookup()`` too but with the
1403introduction of ``->atomic_open()`` they are less relevant there.
1404
1405End of the road
1406---------------
1407
1408Despite its complexity, all this pathname lookup code appears to be
1409in good shape - various parts are certainly easier to understand now
1410than even a couple of releases ago.  But that doesn't mean it is
1411"finished".   As already mentioned, RCU-walk currently only follows
1412symlinks that are stored in the inode so, while it handles many ext4
1413symlinks, it doesn't help with NFS, XFS, or Btrfs.  That support
1414is not likely to be long delayed.