whiterose

linux unikernel
Log | Files | Refs | README | LICENSE | git clone https://git.ne02ptzero.me/git/whiterose

commit dad4f140edaa3f6bb452b6913d41af1ffd672e45
parent 69d5b97c597307773fe6c59775a5d5a88bb7e6b3
Author: Linus Torvalds <torvalds@linux-foundation.org>
Date:   Sun, 28 Oct 2018 11:35:40 -0700

Merge branch 'xarray' of git://git.infradead.org/users/willy/linux-dax

Pull XArray conversion from Matthew Wilcox:
 "The XArray provides an improved interface to the radix tree data
  structure, providing locking as part of the API, specifying GFP flags
  at allocation time, eliminating preloading, less re-walking the tree,
  more efficient iterations and not exposing RCU-protected pointers to
  its users.

  This patch set

   1. Introduces the XArray implementation

   2. Converts the pagecache to use it

   3. Converts memremap to use it

  The page cache is the most complex and important user of the radix
  tree, so converting it was most important. Converting the memremap
  code removes the only other user of the multiorder code, which allows
  us to remove the radix tree code that supported it.

  I have 40+ followup patches to convert many other users of the radix
  tree over to the XArray, but I'd like to get this part in first. The
  other conversions haven't been in linux-next and aren't suitable for
  applying yet, but you can see them in the xarray-conv branch if you're
  interested"

* 'xarray' of git://git.infradead.org/users/willy/linux-dax: (90 commits)
  radix tree: Remove multiorder support
  radix tree test: Convert multiorder tests to XArray
  radix tree tests: Convert item_delete_rcu to XArray
  radix tree tests: Convert item_kill_tree to XArray
  radix tree tests: Move item_insert_order
  radix tree test suite: Remove multiorder benchmarking
  radix tree test suite: Remove __item_insert
  memremap: Convert to XArray
  xarray: Add range store functionality
  xarray: Move multiorder_check to in-kernel tests
  xarray: Move multiorder_shrink to kernel tests
  xarray: Move multiorder account test in-kernel
  radix tree test suite: Convert iteration test to XArray
  radix tree test suite: Convert tag_tagged_items to XArray
  radix tree: Remove radix_tree_clear_tags
  radix tree: Remove radix_tree_maybe_preload_order
  radix tree: Remove split/join code
  radix tree: Remove radix_tree_update_node_t
  page cache: Finish XArray conversion
  dax: Convert page fault handlers to XArray
  ...

Diffstat:
M.clang-format | 1-
M.mailmap | 7+++++++
MDocumentation/core-api/index.rst | 1+
ADocumentation/core-api/xarray.rst | 435+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
MMAINTAINERS | 17++++++++++++++---
March/parisc/kernel/syscall.S | 2+-
March/powerpc/include/asm/book3s/64/pgtable.h | 4+---
March/powerpc/include/asm/nohash/64/pgtable.h | 4+---
Mdrivers/gpu/drm/i915/i915_gem.c | 17+++++++----------
Mdrivers/input/keyboard/hilkbd.c | 2+-
Mdrivers/pci/hotplug/acpiphp.h | 2+-
Mdrivers/pci/hotplug/acpiphp_core.c | 4++--
Mdrivers/pci/hotplug/acpiphp_glue.c | 2+-
Mdrivers/staging/erofs/utils.c | 18++++++------------
Mfs/btrfs/compression.c | 6++----
Mfs/btrfs/extent_io.c | 12+++++-------
Mfs/buffer.c | 14+++++++-------
Mfs/dax.c | 917++++++++++++++++++++++++++++++++++---------------------------------------------
Mfs/ext4/inode.c | 2+-
Mfs/f2fs/data.c | 6+++---
Mfs/f2fs/dir.c | 2+-
Mfs/f2fs/f2fs.h | 2+-
Mfs/f2fs/inline.c | 2+-
Mfs/f2fs/node.c | 6++----
Mfs/fs-writeback.c | 25+++++++++----------------
Mfs/gfs2/aops.c | 2+-
Mfs/inode.c | 2+-
Mfs/isofs/dir.c | 2+-
Mfs/nfs/blocklayout/blocklayout.c | 2+-
Mfs/nilfs2/btnode.c | 26+++++++++-----------------
Mfs/nilfs2/page.c | 29+++++++++++++----------------
Mfs/proc/task_mmu.c | 2+-
Minclude/linux/fs.h | 63+++++++++++++++++++++++++++++++++++++++++----------------------
Minclude/linux/idr.h | 18+++++++-----------
Minclude/linux/pagemap.h | 10+++++-----
Minclude/linux/pagevec.h | 8+++++---
Minclude/linux/radix-tree.h | 178+++++++++++--------------------------------------------------------------------
Minclude/linux/swap.h | 22+++++++++-------------
Minclude/linux/swapops.h | 19+++++++------------
Minclude/linux/xarray.h | 1293++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++-
Mkernel/memremap.c | 75+++++++++++++++------------------------------------------------------------
Mlib/Kconfig | 5++++-
Mlib/Kconfig.debug | 3+++
Mlib/Makefile | 3++-
Mlib/idr.c | 401+++++++++++++++++++++++++++++++++++++++++--------------------------------------
Mlib/radix-tree.c | 834+++++++++----------------------------------------------------------------------
Alib/test_xarray.c | 1238+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Alib/xarray.c | 2036+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Mmm/Kconfig | 4++--
Mmm/filemap.c | 724++++++++++++++++++++++++++++++++++---------------------------------------------
Mmm/huge_memory.c | 17+++++++----------
Mmm/khugepaged.c | 178++++++++++++++++++++++++++++++++-----------------------------------------------
Mmm/madvise.c | 2+-
Mmm/memcontrol.c | 2+-
Mmm/memfd.c | 105++++++++++++++++++++++++++++++++-----------------------------------------------
Mmm/migrate.c | 48++++++++++++++++++------------------------------
Mmm/mincore.c | 2+-
Mmm/page-writeback.c | 72++++++++++++++++++++++++++----------------------------------------------
Mmm/readahead.c | 10++++------
Mmm/shmem.c | 193++++++++++++++++++++++++++++++-------------------------------------------------
Mmm/swap.c | 6+++---
Mmm/swap_state.c | 119+++++++++++++++++++++++++++----------------------------------------------------
Mmm/truncate.c | 27++++++++++++---------------
Mmm/vmscan.c | 10+++++-----
Mmm/workingset.c | 68+++++++++++++++++++++++++++++---------------------------------------
Mtools/include/asm-generic/bitops.h | 1+
Mtools/include/asm-generic/bitops/atomic.h | 9---------
Atools/include/asm-generic/bitops/non-atomic.h | 109+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Mtools/include/linux/bitmap.h | 1+
Mtools/include/linux/kernel.h | 1+
Mtools/include/linux/spinlock.h | 12++++++++++--
Mtools/testing/radix-tree/.gitignore | 1+
Mtools/testing/radix-tree/Makefile | 11++++++++---
Mtools/testing/radix-tree/benchmark.c | 141++++++++++++-------------------------------------------------------------------
Atools/testing/radix-tree/bitmap.c | 23+++++++++++++++++++++++
Mtools/testing/radix-tree/generated/autoconf.h | 2+-
Mtools/testing/radix-tree/idr-test.c | 71++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++-----
Mtools/testing/radix-tree/iteration_check.c | 109++++++++++++++++++++++++++++++++++++++-----------------------------------------
Mtools/testing/radix-tree/linux/bug.h | 1+
Atools/testing/radix-tree/linux/kconfig.h | 1+
Mtools/testing/radix-tree/linux/kernel.h | 5+++++
Atools/testing/radix-tree/linux/lockdep.h | 11+++++++++++
Mtools/testing/radix-tree/linux/radix-tree.h | 1-
Mtools/testing/radix-tree/linux/rcupdate.h | 2++
Mtools/testing/radix-tree/main.c | 66+++---------------------------------------------------------------
Mtools/testing/radix-tree/multiorder.c | 609+++++++++----------------------------------------------------------------------
Mtools/testing/radix-tree/regression1.c | 75+++++++++++++++++++++++++++------------------------------------------------
Mtools/testing/radix-tree/regression2.c | 8++++----
Mtools/testing/radix-tree/regression3.c | 23-----------------------
Mtools/testing/radix-tree/tag_check.c | 33++-------------------------------
Mtools/testing/radix-tree/test.c | 131++++++++++++++++++++++++++-----------------------------------------------------
Mtools/testing/radix-tree/test.h | 13+++++--------
Atools/testing/radix-tree/xarray.c | 35+++++++++++++++++++++++++++++++++++
93 files changed, 7052 insertions(+), 3821 deletions(-)

diff --git a/.clang-format b/.clang-format @@ -323,7 +323,6 @@ ForEachMacros: - 'protocol_for_each_card' - 'protocol_for_each_dev' - 'queue_for_each_hw_ctx' - - 'radix_tree_for_each_contig' - 'radix_tree_for_each_slot' - 'radix_tree_for_each_tagged' - 'rbtree_postorder_for_each_entry_safe' diff --git a/.mailmap b/.mailmap @@ -119,6 +119,13 @@ Mark Brown <broonie@sirena.org.uk> Mark Yao <markyao0591@gmail.com> <mark.yao@rock-chips.com> Martin Kepplinger <martink@posteo.de> <martin.kepplinger@theobroma-systems.com> Martin Kepplinger <martink@posteo.de> <martin.kepplinger@ginzinger.com> +Matthew Wilcox <willy@infradead.org> <matthew.r.wilcox@intel.com> +Matthew Wilcox <willy@infradead.org> <matthew@wil.cx> +Matthew Wilcox <willy@infradead.org> <mawilcox@linuxonhyperv.com> +Matthew Wilcox <willy@infradead.org> <mawilcox@microsoft.com> +Matthew Wilcox <willy@infradead.org> <willy@debian.org> +Matthew Wilcox <willy@infradead.org> <willy@linux.intel.com> +Matthew Wilcox <willy@infradead.org> <willy@parisc-linux.org> Matthieu CASTET <castet.matthieu@free.fr> Mauro Carvalho Chehab <mchehab@kernel.org> <mchehab@brturbo.com.br> Mauro Carvalho Chehab <mchehab@kernel.org> <maurochehab@gmail.com> diff --git a/Documentation/core-api/index.rst b/Documentation/core-api/index.rst @@ -21,6 +21,7 @@ Core utilities local_ops workqueue genericirq + xarray flexible-arrays librs genalloc diff --git a/Documentation/core-api/xarray.rst b/Documentation/core-api/xarray.rst @@ -0,0 +1,435 @@ +.. SPDX-License-Identifier: GPL-2.0+ + +====== +XArray +====== + +:Author: Matthew Wilcox + +Overview +======== + +The XArray is an abstract data type which behaves like a very large array +of pointers. It meets many of the same needs as a hash or a conventional +resizable array. Unlike a hash, it allows you to sensibly go to the +next or previous entry in a cache-efficient manner. In contrast to a +resizable array, there is no need to copy data or change MMU mappings in +order to grow the array. It is more memory-efficient, parallelisable +and cache friendly than a doubly-linked list. It takes advantage of +RCU to perform lookups without locking. + +The XArray implementation is efficient when the indices used are densely +clustered; hashing the object and using the hash as the index will not +perform well. The XArray is optimised for small indices, but still has +good performance with large indices. If your index can be larger than +``ULONG_MAX`` then the XArray is not the data type for you. The most +important user of the XArray is the page cache. + +Each non-``NULL`` entry in the array has three bits associated with +it called marks. Each mark may be set or cleared independently of +the others. You can iterate over entries which are marked. + +Normal pointers may be stored in the XArray directly. They must be 4-byte +aligned, which is true for any pointer returned from :c:func:`kmalloc` and +:c:func:`alloc_page`. It isn't true for arbitrary user-space pointers, +nor for function pointers. You can store pointers to statically allocated +objects, as long as those objects have an alignment of at least 4. + +You can also store integers between 0 and ``LONG_MAX`` in the XArray. +You must first convert it into an entry using :c:func:`xa_mk_value`. +When you retrieve an entry from the XArray, you can check whether it is +a value entry by calling :c:func:`xa_is_value`, and convert it back to +an integer by calling :c:func:`xa_to_value`. + +Some users want to store tagged pointers instead of using the marks +described above. They can call :c:func:`xa_tag_pointer` to create an +entry with a tag, :c:func:`xa_untag_pointer` to turn a tagged entry +back into an untagged pointer and :c:func:`xa_pointer_tag` to retrieve +the tag of an entry. Tagged pointers use the same bits that are used +to distinguish value entries from normal pointers, so each user must +decide whether they want to store value entries or tagged pointers in +any particular XArray. + +The XArray does not support storing :c:func:`IS_ERR` pointers as some +conflict with value entries or internal entries. + +An unusual feature of the XArray is the ability to create entries which +occupy a range of indices. Once stored to, looking up any index in +the range will return the same entry as looking up any other index in +the range. Setting a mark on one index will set it on all of them. +Storing to any index will store to all of them. Multi-index entries can +be explicitly split into smaller entries, or storing ``NULL`` into any +entry will cause the XArray to forget about the range. + +Normal API +========== + +Start by initialising an XArray, either with :c:func:`DEFINE_XARRAY` +for statically allocated XArrays or :c:func:`xa_init` for dynamically +allocated ones. A freshly-initialised XArray contains a ``NULL`` +pointer at every index. + +You can then set entries using :c:func:`xa_store` and get entries +using :c:func:`xa_load`. xa_store will overwrite any entry with the +new entry and return the previous entry stored at that index. You can +use :c:func:`xa_erase` instead of calling :c:func:`xa_store` with a +``NULL`` entry. There is no difference between an entry that has never +been stored to and one that has most recently had ``NULL`` stored to it. + +You can conditionally replace an entry at an index by using +:c:func:`xa_cmpxchg`. Like :c:func:`cmpxchg`, it will only succeed if +the entry at that index has the 'old' value. It also returns the entry +which was at that index; if it returns the same entry which was passed as +'old', then :c:func:`xa_cmpxchg` succeeded. + +If you want to only store a new entry to an index if the current entry +at that index is ``NULL``, you can use :c:func:`xa_insert` which +returns ``-EEXIST`` if the entry is not empty. + +You can enquire whether a mark is set on an entry by using +:c:func:`xa_get_mark`. If the entry is not ``NULL``, you can set a mark +on it by using :c:func:`xa_set_mark` and remove the mark from an entry by +calling :c:func:`xa_clear_mark`. You can ask whether any entry in the +XArray has a particular mark set by calling :c:func:`xa_marked`. + +You can copy entries out of the XArray into a plain array by calling +:c:func:`xa_extract`. Or you can iterate over the present entries in +the XArray by calling :c:func:`xa_for_each`. You may prefer to use +:c:func:`xa_find` or :c:func:`xa_find_after` to move to the next present +entry in the XArray. + +Calling :c:func:`xa_store_range` stores the same entry in a range +of indices. If you do this, some of the other operations will behave +in a slightly odd way. For example, marking the entry at one index +may result in the entry being marked at some, but not all of the other +indices. Storing into one index may result in the entry retrieved by +some, but not all of the other indices changing. + +Finally, you can remove all entries from an XArray by calling +:c:func:`xa_destroy`. If the XArray entries are pointers, you may wish +to free the entries first. You can do this by iterating over all present +entries in the XArray using the :c:func:`xa_for_each` iterator. + +ID assignment +------------- + +You can call :c:func:`xa_alloc` to store the entry at any unused index +in the XArray. If you need to modify the array from interrupt context, +you can use :c:func:`xa_alloc_bh` or :c:func:`xa_alloc_irq` to disable +interrupts while allocating the ID. Unlike :c:func:`xa_store`, allocating +a ``NULL`` pointer does not delete an entry. Instead it reserves an +entry like :c:func:`xa_reserve` and you can release it using either +:c:func:`xa_erase` or :c:func:`xa_release`. To use ID assignment, the +XArray must be defined with :c:func:`DEFINE_XARRAY_ALLOC`, or initialised +by passing ``XA_FLAGS_ALLOC`` to :c:func:`xa_init_flags`, + +Memory allocation +----------------- + +The :c:func:`xa_store`, :c:func:`xa_cmpxchg`, :c:func:`xa_alloc`, +:c:func:`xa_reserve` and :c:func:`xa_insert` functions take a gfp_t +parameter in case the XArray needs to allocate memory to store this entry. +If the entry is being deleted, no memory allocation needs to be performed, +and the GFP flags specified will be ignored. + +It is possible for no memory to be allocatable, particularly if you pass +a restrictive set of GFP flags. In that case, the functions return a +special value which can be turned into an errno using :c:func:`xa_err`. +If you don't need to know exactly which error occurred, using +:c:func:`xa_is_err` is slightly more efficient. + +Locking +------- + +When using the Normal API, you do not have to worry about locking. +The XArray uses RCU and an internal spinlock to synchronise access: + +No lock needed: + * :c:func:`xa_empty` + * :c:func:`xa_marked` + +Takes RCU read lock: + * :c:func:`xa_load` + * :c:func:`xa_for_each` + * :c:func:`xa_find` + * :c:func:`xa_find_after` + * :c:func:`xa_extract` + * :c:func:`xa_get_mark` + +Takes xa_lock internally: + * :c:func:`xa_store` + * :c:func:`xa_insert` + * :c:func:`xa_erase` + * :c:func:`xa_erase_bh` + * :c:func:`xa_erase_irq` + * :c:func:`xa_cmpxchg` + * :c:func:`xa_store_range` + * :c:func:`xa_alloc` + * :c:func:`xa_alloc_bh` + * :c:func:`xa_alloc_irq` + * :c:func:`xa_destroy` + * :c:func:`xa_set_mark` + * :c:func:`xa_clear_mark` + +Assumes xa_lock held on entry: + * :c:func:`__xa_store` + * :c:func:`__xa_insert` + * :c:func:`__xa_erase` + * :c:func:`__xa_cmpxchg` + * :c:func:`__xa_alloc` + * :c:func:`__xa_set_mark` + * :c:func:`__xa_clear_mark` + +If you want to take advantage of the lock to protect the data structures +that you are storing in the XArray, you can call :c:func:`xa_lock` +before calling :c:func:`xa_load`, then take a reference count on the +object you have found before calling :c:func:`xa_unlock`. This will +prevent stores from removing the object from the array between looking +up the object and incrementing the refcount. You can also use RCU to +avoid dereferencing freed memory, but an explanation of that is beyond +the scope of this document. + +The XArray does not disable interrupts or softirqs while modifying +the array. It is safe to read the XArray from interrupt or softirq +context as the RCU lock provides enough protection. + +If, for example, you want to store entries in the XArray in process +context and then erase them in softirq context, you can do that this way:: + + void foo_init(struct foo *foo) + { + xa_init_flags(&foo->array, XA_FLAGS_LOCK_BH); + } + + int foo_store(struct foo *foo, unsigned long index, void *entry) + { + int err; + + xa_lock_bh(&foo->array); + err = xa_err(__xa_store(&foo->array, index, entry, GFP_KERNEL)); + if (!err) + foo->count++; + xa_unlock_bh(&foo->array); + return err; + } + + /* foo_erase() is only called from softirq context */ + void foo_erase(struct foo *foo, unsigned long index) + { + xa_lock(&foo->array); + __xa_erase(&foo->array, index); + foo->count--; + xa_unlock(&foo->array); + } + +If you are going to modify the XArray from interrupt or softirq context, +you need to initialise the array using :c:func:`xa_init_flags`, passing +``XA_FLAGS_LOCK_IRQ`` or ``XA_FLAGS_LOCK_BH``. + +The above example also shows a common pattern of wanting to extend the +coverage of the xa_lock on the store side to protect some statistics +associated with the array. + +Sharing the XArray with interrupt context is also possible, either +using :c:func:`xa_lock_irqsave` in both the interrupt handler and process +context, or :c:func:`xa_lock_irq` in process context and :c:func:`xa_lock` +in the interrupt handler. Some of the more common patterns have helper +functions such as :c:func:`xa_erase_bh` and :c:func:`xa_erase_irq`. + +Sometimes you need to protect access to the XArray with a mutex because +that lock sits above another mutex in the locking hierarchy. That does +not entitle you to use functions like :c:func:`__xa_erase` without taking +the xa_lock; the xa_lock is used for lockdep validation and will be used +for other purposes in the future. + +The :c:func:`__xa_set_mark` and :c:func:`__xa_clear_mark` functions are also +available for situations where you look up an entry and want to atomically +set or clear a mark. It may be more efficient to use the advanced API +in this case, as it will save you from walking the tree twice. + +Advanced API +============ + +The advanced API offers more flexibility and better performance at the +cost of an interface which can be harder to use and has fewer safeguards. +No locking is done for you by the advanced API, and you are required +to use the xa_lock while modifying the array. You can choose whether +to use the xa_lock or the RCU lock while doing read-only operations on +the array. You can mix advanced and normal operations on the same array; +indeed the normal API is implemented in terms of the advanced API. The +advanced API is only available to modules with a GPL-compatible license. + +The advanced API is based around the xa_state. This is an opaque data +structure which you declare on the stack using the :c:func:`XA_STATE` +macro. This macro initialises the xa_state ready to start walking +around the XArray. It is used as a cursor to maintain the position +in the XArray and let you compose various operations together without +having to restart from the top every time. + +The xa_state is also used to store errors. You can call +:c:func:`xas_error` to retrieve the error. All operations check whether +the xa_state is in an error state before proceeding, so there's no need +for you to check for an error after each call; you can make multiple +calls in succession and only check at a convenient point. The only +errors currently generated by the XArray code itself are ``ENOMEM`` and +``EINVAL``, but it supports arbitrary errors in case you want to call +:c:func:`xas_set_err` yourself. + +If the xa_state is holding an ``ENOMEM`` error, calling :c:func:`xas_nomem` +will attempt to allocate more memory using the specified gfp flags and +cache it in the xa_state for the next attempt. The idea is that you take +the xa_lock, attempt the operation and drop the lock. The operation +attempts to allocate memory while holding the lock, but it is more +likely to fail. Once you have dropped the lock, :c:func:`xas_nomem` +can try harder to allocate more memory. It will return ``true`` if it +is worth retrying the operation (i.e. that there was a memory error *and* +more memory was allocated). If it has previously allocated memory, and +that memory wasn't used, and there is no error (or some error that isn't +``ENOMEM``), then it will free the memory previously allocated. + +Internal Entries +---------------- + +The XArray reserves some entries for its own purposes. These are never +exposed through the normal API, but when using the advanced API, it's +possible to see them. Usually the best way to handle them is to pass them +to :c:func:`xas_retry`, and retry the operation if it returns ``true``. + +.. flat-table:: + :widths: 1 1 6 + + * - Name + - Test + - Usage + + * - Node + - :c:func:`xa_is_node` + - An XArray node. May be visible when using a multi-index xa_state. + + * - Sibling + - :c:func:`xa_is_sibling` + - A non-canonical entry for a multi-index entry. The value indicates + which slot in this node has the canonical entry. + + * - Retry + - :c:func:`xa_is_retry` + - This entry is currently being modified by a thread which has the + xa_lock. The node containing this entry may be freed at the end + of this RCU period. You should restart the lookup from the head + of the array. + + * - Zero + - :c:func:`xa_is_zero` + - Zero entries appear as ``NULL`` through the Normal API, but occupy + an entry in the XArray which can be used to reserve the index for + future use. + +Other internal entries may be added in the future. As far as possible, they +will be handled by :c:func:`xas_retry`. + +Additional functionality +------------------------ + +The :c:func:`xas_create_range` function allocates all the necessary memory +to store every entry in a range. It will set ENOMEM in the xa_state if +it cannot allocate memory. + +You can use :c:func:`xas_init_marks` to reset the marks on an entry +to their default state. This is usually all marks clear, unless the +XArray is marked with ``XA_FLAGS_TRACK_FREE``, in which case mark 0 is set +and all other marks are clear. Replacing one entry with another using +:c:func:`xas_store` will not reset the marks on that entry; if you want +the marks reset, you should do that explicitly. + +The :c:func:`xas_load` will walk the xa_state as close to the entry +as it can. If you know the xa_state has already been walked to the +entry and need to check that the entry hasn't changed, you can use +:c:func:`xas_reload` to save a function call. + +If you need to move to a different index in the XArray, call +:c:func:`xas_set`. This resets the cursor to the top of the tree, which +will generally make the next operation walk the cursor to the desired +spot in the tree. If you want to move to the next or previous index, +call :c:func:`xas_next` or :c:func:`xas_prev`. Setting the index does +not walk the cursor around the array so does not require a lock to be +held, while moving to the next or previous index does. + +You can search for the next present entry using :c:func:`xas_find`. This +is the equivalent of both :c:func:`xa_find` and :c:func:`xa_find_after`; +if the cursor has been walked to an entry, then it will find the next +entry after the one currently referenced. If not, it will return the +entry at the index of the xa_state. Using :c:func:`xas_next_entry` to +move to the next present entry instead of :c:func:`xas_find` will save +a function call in the majority of cases at the expense of emitting more +inline code. + +The :c:func:`xas_find_marked` function is similar. If the xa_state has +not been walked, it will return the entry at the index of the xa_state, +if it is marked. Otherwise, it will return the first marked entry after +the entry referenced by the xa_state. The :c:func:`xas_next_marked` +function is the equivalent of :c:func:`xas_next_entry`. + +When iterating over a range of the XArray using :c:func:`xas_for_each` +or :c:func:`xas_for_each_marked`, it may be necessary to temporarily stop +the iteration. The :c:func:`xas_pause` function exists for this purpose. +After you have done the necessary work and wish to resume, the xa_state +is in an appropriate state to continue the iteration after the entry +you last processed. If you have interrupts disabled while iterating, +then it is good manners to pause the iteration and reenable interrupts +every ``XA_CHECK_SCHED`` entries. + +The :c:func:`xas_get_mark`, :c:func:`xas_set_mark` and +:c:func:`xas_clear_mark` functions require the xa_state cursor to have +been moved to the appropriate location in the xarray; they will do +nothing if you have called :c:func:`xas_pause` or :c:func:`xas_set` +immediately before. + +You can call :c:func:`xas_set_update` to have a callback function +called each time the XArray updates a node. This is used by the page +cache workingset code to maintain its list of nodes which contain only +shadow entries. + +Multi-Index Entries +------------------- + +The XArray has the ability to tie multiple indices together so that +operations on one index affect all indices. For example, storing into +any index will change the value of the entry retrieved from any index. +Setting or clearing a mark on any index will set or clear the mark +on every index that is tied together. The current implementation +only allows tying ranges which are aligned powers of two together; +eg indices 64-127 may be tied together, but 2-6 may not be. This may +save substantial quantities of memory; for example tying 512 entries +together will save over 4kB. + +You can create a multi-index entry by using :c:func:`XA_STATE_ORDER` +or :c:func:`xas_set_order` followed by a call to :c:func:`xas_store`. +Calling :c:func:`xas_load` with a multi-index xa_state will walk the +xa_state to the right location in the tree, but the return value is not +meaningful, potentially being an internal entry or ``NULL`` even when there +is an entry stored within the range. Calling :c:func:`xas_find_conflict` +will return the first entry within the range or ``NULL`` if there are no +entries in the range. The :c:func:`xas_for_each_conflict` iterator will +iterate over every entry which overlaps the specified range. + +If :c:func:`xas_load` encounters a multi-index entry, the xa_index +in the xa_state will not be changed. When iterating over an XArray +or calling :c:func:`xas_find`, if the initial index is in the middle +of a multi-index entry, it will not be altered. Subsequent calls +or iterations will move the index to the first index in the range. +Each entry will only be returned once, no matter how many indices it +occupies. + +Using :c:func:`xas_next` or :c:func:`xas_prev` with a multi-index xa_state +is not supported. Using either of these functions on a multi-index entry +will reveal sibling entries; these should be skipped over by the caller. + +Storing ``NULL`` into any index of a multi-index entry will set the entry +at every index to ``NULL`` and dissolve the tie. Splitting a multi-index +entry into entries occupying smaller ranges is not yet supported. + +Functions and structures +======================== + +.. kernel-doc:: include/linux/xarray.h +.. kernel-doc:: lib/xarray.c diff --git a/MAINTAINERS b/MAINTAINERS @@ -535,7 +535,7 @@ F: Documentation/hwmon/adt7475 F: drivers/hwmon/adt7475.c ADVANSYS SCSI DRIVER -M: Matthew Wilcox <matthew@wil.cx> +M: Matthew Wilcox <willy@infradead.org> M: Hannes Reinecke <hare@suse.com> L: linux-scsi@vger.kernel.org S: Maintained @@ -4393,7 +4393,7 @@ S: Maintained F: drivers/i2c/busses/i2c-diolan-u2c.c FILESYSTEM DIRECT ACCESS (DAX) -M: Matthew Wilcox <mawilcox@microsoft.com> +M: Matthew Wilcox <willy@infradead.org> M: Ross Zwisler <zwisler@kernel.org> M: Jan Kara <jack@suse.cz> L: linux-fsdevel@vger.kernel.org @@ -8697,7 +8697,7 @@ F: drivers/message/fusion/ F: drivers/scsi/mpt3sas/ LSILOGIC/SYMBIOS/NCR 53C8XX and 53C1010 PCI-SCSI drivers -M: Matthew Wilcox <matthew@wil.cx> +M: Matthew Wilcox <willy@infradead.org> L: linux-scsi@vger.kernel.org S: Maintained F: drivers/scsi/sym53c8xx_2/ @@ -16137,6 +16137,17 @@ T: git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git x86/vdso S: Maintained F: arch/x86/entry/vdso/ +XARRAY +M: Matthew Wilcox <willy@infradead.org> +L: linux-fsdevel@vger.kernel.org +S: Supported +F: Documentation/core-api/xarray.rst +F: lib/idr.c +F: lib/xarray.c +F: include/linux/idr.h +F: include/linux/xarray.h +F: tools/testing/radix-tree + XC2028/3028 TUNER DRIVER M: Mauro Carvalho Chehab <mchehab@kernel.org> L: linux-media@vger.kernel.org diff --git a/arch/parisc/kernel/syscall.S b/arch/parisc/kernel/syscall.S @@ -2,7 +2,7 @@ * Linux/PA-RISC Project (http://www.parisc-linux.org/) * * System call entry code / Linux gateway page - * Copyright (c) Matthew Wilcox 1999 <willy@bofh.ai> + * Copyright (c) Matthew Wilcox 1999 <willy@infradead.org> * Licensed under the GNU GPL. * thanks to Philipp Rumpf, Mike Shaver and various others * sorry about the wall, puffin.. diff --git a/arch/powerpc/include/asm/book3s/64/pgtable.h b/arch/powerpc/include/asm/book3s/64/pgtable.h @@ -716,9 +716,7 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot) BUILD_BUG_ON(_PAGE_HPTEFLAGS & (0x1f << _PAGE_BIT_SWAP_TYPE)); \ BUILD_BUG_ON(_PAGE_HPTEFLAGS & _PAGE_SWP_SOFT_DIRTY); \ } while (0) -/* - * on pte we don't need handle RADIX_TREE_EXCEPTIONAL_SHIFT; - */ + #define SWP_TYPE_BITS 5 #define __swp_type(x) (((x).val >> _PAGE_BIT_SWAP_TYPE) \ & ((1UL << SWP_TYPE_BITS) - 1)) diff --git a/arch/powerpc/include/asm/nohash/64/pgtable.h b/arch/powerpc/include/asm/nohash/64/pgtable.h @@ -350,9 +350,7 @@ static inline void __ptep_set_access_flags(struct vm_area_struct *vma, #define MAX_SWAPFILES_CHECK() do { \ BUILD_BUG_ON(MAX_SWAPFILES_SHIFT > SWP_TYPE_BITS); \ } while (0) -/* - * on pte we don't need handle RADIX_TREE_EXCEPTIONAL_SHIFT; - */ + #define SWP_TYPE_BITS 5 #define __swp_type(x) (((x).val >> _PAGE_BIT_SWAP_TYPE) \ & ((1UL << SWP_TYPE_BITS) - 1)) diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c @@ -5996,7 +5996,8 @@ i915_gem_object_get_sg(struct drm_i915_gem_object *obj, count = __sg_page_count(sg); while (idx + count <= n) { - unsigned long exception, i; + void *entry; + unsigned long i; int ret; /* If we cannot allocate and insert this entry, or the @@ -6011,12 +6012,9 @@ i915_gem_object_get_sg(struct drm_i915_gem_object *obj, if (ret && ret != -EEXIST) goto scan; - exception = - RADIX_TREE_EXCEPTIONAL_ENTRY | - idx << RADIX_TREE_EXCEPTIONAL_SHIFT; + entry = xa_mk_value(idx); for (i = 1; i < count; i++) { - ret = radix_tree_insert(&iter->radix, idx + i, - (void *)exception); + ret = radix_tree_insert(&iter->radix, idx + i, entry); if (ret && ret != -EEXIST) goto scan; } @@ -6054,15 +6052,14 @@ lookup: GEM_BUG_ON(!sg); /* If this index is in the middle of multi-page sg entry, - * the radixtree will contain an exceptional entry that points + * the radix tree will contain a value entry that points * to the start of that range. We will return the pointer to * the base page and the offset of this page within the * sg entry's range. */ *offset = 0; - if (unlikely(radix_tree_exception(sg))) { - unsigned long base = - (unsigned long)sg >> RADIX_TREE_EXCEPTIONAL_SHIFT; + if (unlikely(xa_is_value(sg))) { + unsigned long base = xa_to_value(sg); sg = radix_tree_lookup(&iter->radix, base); GEM_BUG_ON(!sg); diff --git a/drivers/input/keyboard/hilkbd.c b/drivers/input/keyboard/hilkbd.c @@ -2,7 +2,7 @@ * linux/drivers/hil/hilkbd.c * * Copyright (C) 1998 Philip Blundell <philb@gnu.org> - * Copyright (C) 1999 Matthew Wilcox <willy@bofh.ai> + * Copyright (C) 1999 Matthew Wilcox <willy@infradead.org> * Copyright (C) 1999-2007 Helge Deller <deller@gmx.de> * * Very basic HP Human Interface Loop (HIL) driver. diff --git a/drivers/pci/hotplug/acpiphp.h b/drivers/pci/hotplug/acpiphp.h @@ -8,7 +8,7 @@ * Copyright (C) 2002 Hiroshi Aono (h-aono@ap.jp.nec.com) * Copyright (C) 2002,2003 Takayoshi Kochi (t-kochi@bq.jp.nec.com) * Copyright (C) 2002,2003 NEC Corporation - * Copyright (C) 2003-2005 Matthew Wilcox (matthew.wilcox@hp.com) + * Copyright (C) 2003-2005 Matthew Wilcox (willy@infradead.org) * Copyright (C) 2003-2005 Hewlett Packard * * All rights reserved. diff --git a/drivers/pci/hotplug/acpiphp_core.c b/drivers/pci/hotplug/acpiphp_core.c @@ -8,7 +8,7 @@ * Copyright (C) 2002 Hiroshi Aono (h-aono@ap.jp.nec.com) * Copyright (C) 2002,2003 Takayoshi Kochi (t-kochi@bq.jp.nec.com) * Copyright (C) 2002,2003 NEC Corporation - * Copyright (C) 2003-2005 Matthew Wilcox (matthew.wilcox@hp.com) + * Copyright (C) 2003-2005 Matthew Wilcox (willy@infradead.org) * Copyright (C) 2003-2005 Hewlett Packard * * All rights reserved. @@ -40,7 +40,7 @@ bool acpiphp_disabled; static struct acpiphp_attention_info *attention_info; #define DRIVER_VERSION "0.5" -#define DRIVER_AUTHOR "Greg Kroah-Hartman <gregkh@us.ibm.com>, Takayoshi Kochi <t-kochi@bq.jp.nec.com>, Matthew Wilcox <willy@hp.com>" +#define DRIVER_AUTHOR "Greg Kroah-Hartman <gregkh@us.ibm.com>, Takayoshi Kochi <t-kochi@bq.jp.nec.com>, Matthew Wilcox <willy@infradead.org>" #define DRIVER_DESC "ACPI Hot Plug PCI Controller Driver" MODULE_AUTHOR(DRIVER_AUTHOR); diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c @@ -5,7 +5,7 @@ * Copyright (C) 2002,2003 Takayoshi Kochi (t-kochi@bq.jp.nec.com) * Copyright (C) 2002 Hiroshi Aono (h-aono@ap.jp.nec.com) * Copyright (C) 2002,2003 NEC Corporation - * Copyright (C) 2003-2005 Matthew Wilcox (matthew.wilcox@hp.com) + * Copyright (C) 2003-2005 Matthew Wilcox (willy@infradead.org) * Copyright (C) 2003-2005 Hewlett Packard * Copyright (C) 2005 Rajesh Shah (rajesh.shah@intel.com) * Copyright (C) 2005 Intel Corporation diff --git a/drivers/staging/erofs/utils.c b/drivers/staging/erofs/utils.c @@ -35,7 +35,6 @@ static atomic_long_t erofs_global_shrink_cnt; #ifdef CONFIG_EROFS_FS_ZIP -/* radix_tree and the future XArray both don't use tagptr_t yet */ struct erofs_workgroup *erofs_find_workgroup( struct super_block *sb, pgoff_t index, bool *tag) { @@ -47,9 +46,8 @@ repeat: rcu_read_lock(); grp = radix_tree_lookup(&sbi->workstn_tree, index); if (grp != NULL) { - *tag = radix_tree_exceptional_entry(grp); - grp = (void *)((unsigned long)grp & - ~RADIX_TREE_EXCEPTIONAL_ENTRY); + *tag = xa_pointer_tag(grp); + grp = xa_untag_pointer(grp); if (erofs_workgroup_get(grp, &oldcount)) { /* prefer to relax rcu read side */ @@ -83,9 +81,7 @@ int erofs_register_workgroup(struct super_block *sb, sbi = EROFS_SB(sb); erofs_workstn_lock(sbi); - if (tag) - grp = (void *)((unsigned long)grp | - 1UL << RADIX_TREE_EXCEPTIONAL_SHIFT); + grp = xa_tag_pointer(grp, tag); err = radix_tree_insert(&sbi->workstn_tree, grp->index, grp); @@ -131,9 +127,7 @@ repeat: for (i = 0; i < found; ++i) { int cnt; - struct erofs_workgroup *grp = (void *) - ((unsigned long)batch[i] & - ~RADIX_TREE_EXCEPTIONAL_ENTRY); + struct erofs_workgroup *grp = xa_untag_pointer(batch[i]); first_index = grp->index + 1; @@ -150,8 +144,8 @@ repeat: #endif continue; - if (radix_tree_delete(&sbi->workstn_tree, - grp->index) != grp) { + if (xa_untag_pointer(radix_tree_delete(&sbi->workstn_tree, + grp->index)) != grp) { #ifdef EROFS_FS_HAS_MANAGED_CACHE skip: erofs_workgroup_unfreeze(grp, 1); diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c @@ -437,10 +437,8 @@ static noinline int add_ra_bio_pages(struct inode *inode, if (pg_index > end_index) break; - rcu_read_lock(); - page = radix_tree_lookup(&mapping->i_pages, pg_index); - rcu_read_unlock(); - if (page && !radix_tree_exceptional_entry(page)) { + page = xa_load(&mapping->i_pages, pg_index); + if (page && !xa_is_value(page)) { misses++; if (misses > 4) break; diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c @@ -3784,7 +3784,7 @@ int btree_write_cache_pages(struct address_space *mapping, pgoff_t index; pgoff_t end; /* Inclusive */ int scanned = 0; - int tag; + xa_mark_t tag; pagevec_init(&pvec); if (wbc->range_cyclic) { @@ -3909,7 +3909,7 @@ static int extent_write_cache_pages(struct address_space *mapping, pgoff_t done_index; int range_whole = 0; int scanned = 0; - int tag; + xa_mark_t tag; /* * We have to hold onto the inode so that ordered extents can do their @@ -5159,11 +5159,9 @@ void clear_extent_buffer_dirty(struct extent_buffer *eb) clear_page_dirty_for_io(page); xa_lock_irq(&page->mapping->i_pages); - if (!PageDirty(page)) { - radix_tree_tag_clear(&page->mapping->i_pages, - page_index(page), - PAGECACHE_TAG_DIRTY); - } + if (!PageDirty(page)) + __xa_clear_mark(&page->mapping->i_pages, + page_index(page), PAGECACHE_TAG_DIRTY); xa_unlock_irq(&page->mapping->i_pages); ClearPageError(page); unlock_page(page); diff --git a/fs/buffer.c b/fs/buffer.c @@ -562,7 +562,7 @@ void mark_buffer_dirty_inode(struct buffer_head *bh, struct inode *inode) EXPORT_SYMBOL(mark_buffer_dirty_inode); /* - * Mark the page dirty, and set it dirty in the radix tree, and mark the inode + * Mark the page dirty, and set it dirty in the page cache, and mark the inode * dirty. * * If warn is true, then emit a warning if the page is not uptodate and has @@ -579,8 +579,8 @@ void __set_page_dirty(struct page *page, struct address_space *mapping, if (page->mapping) { /* Race with truncate? */ WARN_ON_ONCE(warn && !PageUptodate(page)); account_page_dirtied(page, mapping); - radix_tree_tag_set(&mapping->i_pages, - page_index(page), PAGECACHE_TAG_DIRTY); + __xa_set_mark(&mapping->i_pages, page_index(page), + PAGECACHE_TAG_DIRTY); } xa_unlock_irqrestore(&mapping->i_pages, flags); } @@ -1050,7 +1050,7 @@ __getblk_slow(struct block_device *bdev, sector_t block, * The relationship between dirty buffers and dirty pages: * * Whenever a page has any dirty buffers, the page's dirty bit is set, and - * the page is tagged dirty in its radix tree. + * the page is tagged dirty in the page cache. * * At all times, the dirtiness of the buffers represents the dirtiness of * subsections of the page. If the page has buffers, the page dirty bit is @@ -1073,9 +1073,9 @@ __getblk_slow(struct block_device *bdev, sector_t block, * mark_buffer_dirty - mark a buffer_head as needing writeout * @bh: the buffer_head to mark dirty * - * mark_buffer_dirty() will set the dirty bit against the buffer, then set its - * backing page dirty, then tag the page as dirty in its address_space's radix - * tree and then attach the address_space's inode to its superblock's dirty + * mark_buffer_dirty() will set the dirty bit against the buffer, then set + * its backing page dirty, then tag the page as dirty in the page cache + * and then attach the address_space's inode to its superblock's dirty * inode list. * * mark_buffer_dirty() is atomic. It takes bh->b_page->mapping->private_lock, diff --git a/fs/dax.c b/fs/dax.c @@ -38,6 +38,17 @@ #define CREATE_TRACE_POINTS #include <trace/events/fs_dax.h> +static inline unsigned int pe_order(enum page_entry_size pe_size) +{ + if (pe_size == PE_SIZE_PTE) + return PAGE_SHIFT - PAGE_SHIFT; + if (pe_size == PE_SIZE_PMD) + return PMD_SHIFT - PAGE_SHIFT; + if (pe_size == PE_SIZE_PUD) + return PUD_SHIFT - PAGE_SHIFT; + return ~0; +} + /* We choose 4096 entries - same as per-zone page wait tables */ #define DAX_WAIT_TABLE_BITS 12 #define DAX_WAIT_TABLE_ENTRIES (1 << DAX_WAIT_TABLE_BITS) @@ -46,6 +57,9 @@ #define PG_PMD_COLOUR ((PMD_SIZE >> PAGE_SHIFT) - 1) #define PG_PMD_NR (PMD_SIZE >> PAGE_SHIFT) +/* The order of a PMD entry */ +#define PMD_ORDER (PMD_SHIFT - PAGE_SHIFT) + static wait_queue_head_t wait_table[DAX_WAIT_TABLE_ENTRIES]; static int __init init_dax_wait_table(void) @@ -59,63 +73,74 @@ static int __init init_dax_wait_table(void) fs_initcall(init_dax_wait_table); /* - * We use lowest available bit in exceptional entry for locking, one bit for - * the entry size (PMD) and two more to tell us if the entry is a zero page or - * an empty entry that is just used for locking. In total four special bits. + * DAX pagecache entries use XArray value entries so they can't be mistaken + * for pages. We use one bit for locking, one bit for the entry size (PMD) + * and two more to tell us if the entry is a zero page or an empty entry that + * is just used for locking. In total four special bits. * * If the PMD bit isn't set the entry has size PAGE_SIZE, and if the ZERO_PAGE * and EMPTY bits aren't set the entry is a normal DAX entry with a filesystem * block allocation. */ -#define RADIX_DAX_SHIFT (RADIX_TREE_EXCEPTIONAL_SHIFT + 4) -#define RADIX_DAX_ENTRY_LOCK (1 << RADIX_TREE_EXCEPTIONAL_SHIFT) -#define RADIX_DAX_PMD (1 << (RADIX_TREE_EXCEPTIONAL_SHIFT + 1)) -#define RADIX_DAX_ZERO_PAGE (1 << (RADIX_TREE_EXCEPTIONAL_SHIFT + 2)) -#define RADIX_DAX_EMPTY (1 << (RADIX_TREE_EXCEPTIONAL_SHIFT + 3)) +#define DAX_SHIFT (4) +#define DAX_LOCKED (1UL << 0) +#define DAX_PMD (1UL << 1) +#define DAX_ZERO_PAGE (1UL << 2) +#define DAX_EMPTY (1UL << 3) -static unsigned long dax_radix_pfn(void *entry) +static unsigned long dax_to_pfn(void *entry) { - return (unsigned long)entry >> RADIX_DAX_SHIFT; + return xa_to_value(entry) >> DAX_SHIFT; } -static void *dax_radix_locked_entry(unsigned long pfn, unsigned long flags) +static void *dax_make_entry(pfn_t pfn, unsigned long flags) { - return (void *)(RADIX_TREE_EXCEPTIONAL_ENTRY | flags | - (pfn << RADIX_DAX_SHIFT) | RADIX_DAX_ENTRY_LOCK); + return xa_mk_value(flags | (pfn_t_to_pfn(pfn) << DAX_SHIFT)); } -static unsigned int dax_radix_order(void *entry) +static void *dax_make_page_entry(struct page *page) { - if ((unsigned long)entry & RADIX_DAX_PMD) - return PMD_SHIFT - PAGE_SHIFT; + pfn_t pfn = page_to_pfn_t(page); + return dax_make_entry(pfn, PageHead(page) ? DAX_PMD : 0); +} + +static bool dax_is_locked(void *entry) +{ + return xa_to_value(entry) & DAX_LOCKED; +} + +static unsigned int dax_entry_order(void *entry) +{ + if (xa_to_value(entry) & DAX_PMD) + return PMD_ORDER; return 0; } static int dax_is_pmd_entry(void *entry) { - return (unsigned long)entry & RADIX_DAX_PMD; + return xa_to_value(entry) & DAX_PMD; } static int dax_is_pte_entry(void *entry) { - return !((unsigned long)entry & RADIX_DAX_PMD); + return !(xa_to_value(entry) & DAX_PMD); } static int dax_is_zero_entry(void *entry) { - return (unsigned long)entry & RADIX_DAX_ZERO_PAGE; + return xa_to_value(entry) & DAX_ZERO_PAGE; } static int dax_is_empty_entry(void *entry) { - return (unsigned long)entry & RADIX_DAX_EMPTY; + return xa_to_value(entry) & DAX_EMPTY; } /* - * DAX radix tree locking + * DAX page cache entry locking */ struct exceptional_entry_key { - struct address_space *mapping; + struct xarray *xa; pgoff_t entry_start; }; @@ -124,10 +149,11 @@ struct wait_exceptional_entry_queue { struct exceptional_entry_key key; }; -static wait_queue_head_t *dax_entry_waitqueue(struct address_space *mapping, - pgoff_t index, void *entry, struct exceptional_entry_key *key) +static wait_queue_head_t *dax_entry_waitqueue(struct xa_state *xas, + void *entry, struct exceptional_entry_key *key) { unsigned long hash; + unsigned long index = xas->xa_index; /* * If 'entry' is a PMD, align the 'index' that we use for the wait @@ -136,22 +162,21 @@ static wait_queue_head_t *dax_entry_waitqueue(struct address_space *mapping, */ if (dax_is_pmd_entry(entry)) index &= ~PG_PMD_COLOUR; - - key->mapping = mapping; + key->xa = xas->xa; key->entry_start = index; - hash = hash_long((unsigned long)mapping ^ index, DAX_WAIT_TABLE_BITS); + hash = hash_long((unsigned long)xas->xa ^ index, DAX_WAIT_TABLE_BITS); return wait_table + hash; } -static int wake_exceptional_entry_func(wait_queue_entry_t *wait, unsigned int mode, - int sync, void *keyp) +static int wake_exceptional_entry_func(wait_queue_entry_t *wait, + unsigned int mode, int sync, void *keyp) { struct exceptional_entry_key *key = keyp; struct wait_exceptional_entry_queue *ewait = container_of(wait, struct wait_exceptional_entry_queue, wait); - if (key->mapping != ewait->key.mapping || + if (key->xa != ewait->key.xa || key->entry_start != ewait->key.entry_start) return 0; return autoremove_wake_function(wait, mode, sync, NULL); @@ -162,13 +187,12 @@ static int wake_exceptional_entry_func(wait_queue_entry_t *wait, unsigned int mo * The important information it's conveying is whether the entry at * this index used to be a PMD entry. */ -static void dax_wake_mapping_entry_waiter(struct address_space *mapping, - pgoff_t index, void *entry, bool wake_all) +static void dax_wake_entry(struct xa_state *xas, void *entry, bool wake_all) { struct exceptional_entry_key key; wait_queue_head_t *wq; - wq = dax_entry_waitqueue(mapping, index, entry, &key); + wq = dax_entry_waitqueue(xas, entry, &key); /* * Checking for locked entry and prepare_to_wait_exclusive() happens @@ -181,55 +205,16 @@ static void dax_wake_mapping_entry_waiter(struct address_space *mapping, } /* - * Check whether the given slot is locked. Must be called with the i_pages - * lock held. - */ -static inline int slot_locked(struct address_space *mapping, void **slot) -{ - unsigned long entry = (unsigned long) - radix_tree_deref_slot_protected(slot, &mapping->i_pages.xa_lock); - return entry & RADIX_DAX_ENTRY_LOCK; -} - -/* - * Mark the given slot as locked. Must be called with the i_pages lock held. - */ -static inline void *lock_slot(struct address_space *mapping, void **slot) -{ - unsigned long entry = (unsigned long) - radix_tree_deref_slot_protected(slot, &mapping->i_pages.xa_lock); - - entry |= RADIX_DAX_ENTRY_LOCK; - radix_tree_replace_slot(&mapping->i_pages, slot, (void *)entry); - return (void *)entry; -} - -/* - * Mark the given slot as unlocked. Must be called with the i_pages lock held. - */ -static inline void *unlock_slot(struct address_space *mapping, void **slot) -{ - unsigned long entry = (unsigned long) - radix_tree_deref_slot_protected(slot, &mapping->i_pages.xa_lock); - - entry &= ~(unsigned long)RADIX_DAX_ENTRY_LOCK; - radix_tree_replace_slot(&mapping->i_pages, slot, (void *)entry); - return (void *)entry; -} - -/* - * Lookup entry in radix tree, wait for it to become unlocked if it is - * exceptional entry and return it. The caller must call - * put_unlocked_mapping_entry() when he decided not to lock the entry or - * put_locked_mapping_entry() when he locked the entry and now wants to - * unlock it. + * Look up entry in page cache, wait for it to become unlocked if it + * is a DAX entry and return it. The caller must subsequently call + * put_unlocked_entry() if it did not lock the entry or dax_unlock_entry() + * if it did. * * Must be called with the i_pages lock held. */ -static void *__get_unlocked_mapping_entry(struct address_space *mapping, - pgoff_t index, void ***slotp, bool (*wait_fn)(void)) +static void *get_unlocked_entry(struct xa_state *xas) { - void *entry, **slot; + void *entry; struct wait_exceptional_entry_queue ewait; wait_queue_head_t *wq; @@ -237,80 +222,54 @@ static void *__get_unlocked_mapping_entry(struct address_space *mapping, ewait.wait.func = wake_exceptional_entry_func; for (;;) { - bool revalidate; - - entry = __radix_tree_lookup(&mapping->i_pages, index, NULL, - &slot); - if (!entry || - WARN_ON_ONCE(!radix_tree_exceptional_entry(entry)) || - !slot_locked(mapping, slot)) { - if (slotp) - *slotp = slot; + entry = xas_load(xas); + if (!entry || xa_is_internal(entry) || + WARN_ON_ONCE(!xa_is_value(entry)) || + !dax_is_locked(entry)) return entry; - } - wq = dax_entry_waitqueue(mapping, index, entry, &ewait.key); + wq = dax_entry_waitqueue(xas, entry, &ewait.key); prepare_to_wait_exclusive(wq, &ewait.wait, TASK_UNINTERRUPTIBLE); - xa_unlock_irq(&mapping->i_pages); - revalidate = wait_fn(); + xas_unlock_irq(xas); + xas_reset(xas); + schedule(); finish_wait(wq, &ewait.wait); - xa_lock_irq(&mapping->i_pages); - if (revalidate) - return ERR_PTR(-EAGAIN); + xas_lock_irq(xas); } } -static bool entry_wait(void) -{ - schedule(); - /* - * Never return an ERR_PTR() from - * __get_unlocked_mapping_entry(), just keep looping. - */ - return false; -} - -static void *get_unlocked_mapping_entry(struct address_space *mapping, - pgoff_t index, void ***slotp) +static void put_unlocked_entry(struct xa_state *xas, void *entry) { - return __get_unlocked_mapping_entry(mapping, index, slotp, entry_wait); -} - -static void unlock_mapping_entry(struct address_space *mapping, pgoff_t index) -{ - void *entry, **slot; - - xa_lock_irq(&mapping->i_pages); - entry = __radix_tree_lookup(&mapping->i_pages, index, NULL, &slot); - if (WARN_ON_ONCE(!entry || !radix_tree_exceptional_entry(entry) || - !slot_locked(mapping, slot))) { - xa_unlock_irq(&mapping->i_pages); - return; - } - unlock_slot(mapping, slot); - xa_unlock_irq(&mapping->i_pages); - dax_wake_mapping_entry_waiter(mapping, index, entry, false); + /* If we were the only waiter woken, wake the next one */ + if (entry) + dax_wake_entry(xas, entry, false); } -static void put_locked_mapping_entry(struct address_space *mapping, - pgoff_t index) +/* + * We used the xa_state to get the entry, but then we locked the entry and + * dropped the xa_lock, so we know the xa_state is stale and must be reset + * before use. + */ +static void dax_unlock_entry(struct xa_state *xas, void *entry) { - unlock_mapping_entry(mapping, index); + void *old; + + xas_reset(xas); + xas_lock_irq(xas); + old = xas_store(xas, entry); + xas_unlock_irq(xas); + BUG_ON(!dax_is_locked(old)); + dax_wake_entry(xas, entry, false); } /* - * Called when we are done with radix tree entry we looked up via - * get_unlocked_mapping_entry() and which we didn't lock in the end. + * Return: The entry stored at this location before it was locked. */ -static void put_unlocked_mapping_entry(struct address_space *mapping, - pgoff_t index, void *entry) +static void *dax_lock_entry(struct xa_state *xas, void *entry) { - if (!entry) - return; - - /* We have to wake up next waiter for the radix tree entry lock */ - dax_wake_mapping_entry_waiter(mapping, index, entry, false); + unsigned long v = xa_to_value(entry); + return xas_store(xas, xa_mk_value(v | DAX_LOCKED)); } static unsigned long dax_entry_size(void *entry) @@ -325,9 +284,9 @@ static unsigned long dax_entry_size(void *entry) return PAGE_SIZE; } -static unsigned long dax_radix_end_pfn(void *entry) +static unsigned long dax_end_pfn(void *entry) { - return dax_radix_pfn(entry) + dax_entry_size(entry) / PAGE_SIZE; + return dax_to_pfn(entry) + dax_entry_size(entry) / PAGE_SIZE; } /* @@ -335,8 +294,8 @@ static unsigned long dax_radix_end_pfn(void *entry) * 'empty' and 'zero' entries. */ #define for_each_mapped_pfn(entry, pfn) \ - for (pfn = dax_radix_pfn(entry); \ - pfn < dax_radix_end_pfn(entry); pfn++) + for (pfn = dax_to_pfn(entry); \ + pfn < dax_end_pfn(entry); pfn++) /* * TODO: for reflink+dax we need a way to associate a single page with @@ -393,33 +352,16 @@ static struct page *dax_busy_page(void *entry) return NULL; } -static bool entry_wait_revalidate(void) -{ - rcu_read_unlock(); - schedule(); - rcu_read_lock(); - - /* - * Tell __get_unlocked_mapping_entry() to take a break, we need - * to revalidate page->mapping after dropping locks - */ - return true; -} - bool dax_lock_mapping_entry(struct page *page) { - pgoff_t index; - struct inode *inode; - bool did_lock = false; - void *entry = NULL, **slot; - struct address_space *mapping; + XA_STATE(xas, NULL, 0); + void *entry; - rcu_read_lock(); for (;;) { - mapping = READ_ONCE(page->mapping); + struct address_space *mapping = READ_ONCE(page->mapping); if (!dax_mapping(mapping)) - break; + return false; /* * In the device-dax case there's no need to lock, a @@ -428,98 +370,94 @@ bool dax_lock_mapping_entry(struct page *page) * otherwise we would not have a valid pfn_to_page() * translation. */ - inode = mapping->host; - if (S_ISCHR(inode->i_mode)) { - did_lock = true; - break; - } + if (S_ISCHR(mapping->host->i_mode)) + return true; - xa_lock_irq(&mapping->i_pages); + xas.xa = &mapping->i_pages; + xas_lock_irq(&xas); if (mapping != page->mapping) { - xa_unlock_irq(&mapping->i_pages); + xas_unlock_irq(&xas); continue; } - index = page->index; - - entry = __get_unlocked_mapping_entry(mapping, index, &slot, - entry_wait_revalidate); - if (!entry) { - xa_unlock_irq(&mapping->i_pages); - break; - } else if (IS_ERR(entry)) { - xa_unlock_irq(&mapping->i_pages); - WARN_ON_ONCE(PTR_ERR(entry) != -EAGAIN); - continue; + xas_set(&xas, page->index); + entry = xas_load(&xas); + if (dax_is_locked(entry)) { + entry = get_unlocked_entry(&xas); + /* Did the page move while we slept? */ + if (dax_to_pfn(entry) != page_to_pfn(page)) { + xas_unlock_irq(&xas); + continue; + } } - lock_slot(mapping, slot); - did_lock = true; - xa_unlock_irq(&mapping->i_pages); - break; + dax_lock_entry(&xas, entry); + xas_unlock_irq(&xas); + return true; } - rcu_read_unlock(); - - return did_lock; } void dax_unlock_mapping_entry(struct page *page) { struct address_space *mapping = page->mapping; - struct inode *inode = mapping->host; + XA_STATE(xas, &mapping->i_pages, page->index); - if (S_ISCHR(inode->i_mode)) + if (S_ISCHR(mapping->host->i_mode)) return; - unlock_mapping_entry(mapping, page->index); + dax_unlock_entry(&xas, dax_make_page_entry(page)); } /* - * Find radix tree entry at given index. If it points to an exceptional entry, - * return it with the radix tree entry locked. If the radix tree doesn't - * contain given index, create an empty exceptional entry for the index and - * return with it locked. + * Find page cache entry at given index. If it is a DAX entry, return it + * with the entry locked. If the page cache doesn't contain an entry at + * that index, add a locked empty entry. * - * When requesting an entry with size RADIX_DAX_PMD, grab_mapping_entry() will - * either return that locked entry or will return an error. This error will - * happen if there are any 4k entries within the 2MiB range that we are - * requesting. + * When requesting an entry with size DAX_PMD, grab_mapping_entry() will + * either return that locked entry or will return VM_FAULT_FALLBACK. + * This will happen if there are any PTE entries within the PMD range + * that we are requesting. * - * We always favor 4k entries over 2MiB entries. There isn't a flow where we - * evict 4k entries in order to 'upgrade' them to a 2MiB entry. A 2MiB - * insertion will fail if it finds any 4k entries already in the tree, and a - * 4k insertion will cause an existing 2MiB entry to be unmapped and - * downgraded to 4k entries. This happens for both 2MiB huge zero pages as - * well as 2MiB empty entries. + * We always favor PTE entries over PMD entries. There isn't a flow where we + * evict PTE entries in order to 'upgrade' them to a PMD entry. A PMD + * insertion will fail if it finds any PTE entries already in the tree, and a + * PTE insertion will cause an existing PMD entry to be unmapped and + * downgraded to PTE entries. This happens for both PMD zero pages as + * well as PMD empty entries. * - * The exception to this downgrade path is for 2MiB DAX PMD entries that have - * real storage backing them. We will leave these real 2MiB DAX entries in - * the tree, and PTE writes will simply dirty the entire 2MiB DAX entry. + * The exception to this downgrade path is for PMD entries that have + * real storage backing them. We will leave these real PMD entries in + * the tree, and PTE writes will simply dirty the entire PMD entry. * * Note: Unlike filemap_fault() we don't honor FAULT_FLAG_RETRY flags. For * persistent memory the benefit is doubtful. We can add that later if we can * show it helps. + * + * On error, this function does not return an ERR_PTR. Instead it returns + * a VM_FAULT code, encoded as an xarray internal entry. The ERR_PTR values + * overlap with xarray value entries. */ -static void *grab_mapping_entry(struct address_space *mapping, pgoff_t index, - unsigned long size_flag) +static void *grab_mapping_entry(struct xa_state *xas, + struct address_space *mapping, unsigned long size_flag) { - bool pmd_downgrade = false; /* splitting 2MiB entry into 4k entries? */ - void *entry, **slot; - -restart: - xa_lock_irq(&mapping->i_pages); - entry = get_unlocked_mapping_entry(mapping, index, &slot); + unsigned long index = xas->xa_index; + bool pmd_downgrade = false; /* splitting PMD entry into PTE entries? */ + void *entry; - if (WARN_ON_ONCE(entry && !radix_tree_exceptional_entry(entry))) { - entry = ERR_PTR(-EIO); - goto out_unlock; - } +retry: + xas_lock_irq(xas); + entry = get_unlocked_entry(xas); + if (xa_is_internal(entry)) + goto fallback; if (entry) { - if (size_flag & RADIX_DAX_PMD) { + if (WARN_ON_ONCE(!xa_is_value(entry))) { + xas_set_err(xas, EIO); + goto out_unlock; + } + + if (size_flag & DAX_PMD) { if (dax_is_pte_entry(entry)) { - put_unlocked_mapping_entry(mapping, index, - entry); - entry = ERR_PTR(-EEXIST); - goto out_unlock; + put_unlocked_entry(xas, entry); + goto fallback; } } else { /* trying to grab a PTE entry */ if (dax_is_pmd_entry(entry) && @@ -530,87 +468,57 @@ restart: } } - /* No entry for given index? Make sure radix tree is big enough. */ - if (!entry || pmd_downgrade) { - int err; - - if (pmd_downgrade) { - /* - * Make sure 'entry' remains valid while we drop - * the i_pages lock. - */ - entry = lock_slot(mapping, slot); - } + if (pmd_downgrade) { + /* + * Make sure 'entry' remains valid while we drop + * the i_pages lock. + */ + dax_lock_entry(xas, entry); - xa_unlock_irq(&mapping->i_pages); /* * Besides huge zero pages the only other thing that gets * downgraded are empty entries which don't need to be * unmapped. */ - if (pmd_downgrade && dax_is_zero_entry(entry)) - unmap_mapping_pages(mapping, index & ~PG_PMD_COLOUR, - PG_PMD_NR, false); - - err = radix_tree_preload( - mapping_gfp_mask(mapping) & ~__GFP_HIGHMEM); - if (err) { - if (pmd_downgrade) - put_locked_mapping_entry(mapping, index); - return ERR_PTR(err); - } - xa_lock_irq(&mapping->i_pages); - - if (!entry) { - /* - * We needed to drop the i_pages lock while calling - * radix_tree_preload() and we didn't have an entry to - * lock. See if another thread inserted an entry at - * our index during this time. - */ - entry = __radix_tree_lookup(&mapping->i_pages, index, - NULL, &slot); - if (entry) { - radix_tree_preload_end(); - xa_unlock_irq(&mapping->i_pages); - goto restart; - } + if (dax_is_zero_entry(entry)) { + xas_unlock_irq(xas); + unmap_mapping_pages(mapping, + xas->xa_index & ~PG_PMD_COLOUR, + PG_PMD_NR, false); + xas_reset(xas); + xas_lock_irq(xas); } - if (pmd_downgrade) { - dax_disassociate_entry(entry, mapping, false); - radix_tree_delete(&mapping->i_pages, index); - mapping->nrexceptional--; - dax_wake_mapping_entry_waiter(mapping, index, entry, - true); - } + dax_disassociate_entry(entry, mapping, false); + xas_store(xas, NULL); /* undo the PMD join */ + dax_wake_entry(xas, entry, true); + mapping->nrexceptional--; + entry = NULL; + xas_set(xas, index); + } - entry = dax_radix_locked_entry(0, size_flag | RADIX_DAX_EMPTY); - - err = __radix_tree_insert(&mapping->i_pages, index, - dax_radix_order(entry), entry); - radix_tree_preload_end(); - if (err) { - xa_unlock_irq(&mapping->i_pages); - /* - * Our insertion of a DAX entry failed, most likely - * because we were inserting a PMD entry and it - * collided with a PTE sized entry at a different - * index in the PMD range. We haven't inserted - * anything into the radix tree and have no waiters to - * wake. - */ - return ERR_PTR(err); - } - /* Good, we have inserted empty locked entry into the tree. */ + if (entry) { + dax_lock_entry(xas, entry); + } else { + entry = dax_make_entry(pfn_to_pfn_t(0), size_flag | DAX_EMPTY); + dax_lock_entry(xas, entry); + if (xas_error(xas)) + goto out_unlock; mapping->nrexceptional++; - xa_unlock_irq(&mapping->i_pages); - return entry; } - entry = lock_slot(mapping, slot); - out_unlock: - xa_unlock_irq(&mapping->i_pages); + +out_unlock: + xas_unlock_irq(xas); + if (xas_nomem(xas, mapping_gfp_mask(mapping) & ~__GFP_HIGHMEM)) + goto retry; + if (xas->xa_node == XA_ERROR(-ENOMEM)) + return xa_mk_internal(VM_FAULT_OOM); + if (xas_error(xas)) + return xa_mk_internal(VM_FAULT_SIGBUS); return entry; +fallback: + xas_unlock_irq(xas); + return xa_mk_internal(VM_FAULT_FALLBACK); } /** @@ -630,11 +538,10 @@ restart: */ struct page *dax_layout_busy_page(struct address_space *mapping) { - pgoff_t indices[PAGEVEC_SIZE]; + XA_STATE(xas, &mapping->i_pages, 0); + void *entry; + unsigned int scanned = 0; struct page *page = NULL; - struct pagevec pvec; - pgoff_t index, end; - unsigned i; /* * In the 'limited' case get_user_pages() for dax is disabled. @@ -645,13 +552,9 @@ struct page *dax_layout_busy_page(struct address_space *mapping) if (!dax_mapping(mapping) || !mapping_mapped(mapping)) return NULL; - pagevec_init(&pvec); - index = 0; - end = -1; - /* * If we race get_user_pages_fast() here either we'll see the - * elevated page count in the pagevec_lookup and wait, or + * elevated page count in the iteration and wait, or * get_user_pages_fast() will see that the page it took a reference * against is no longer mapped in the page tables and bail to the * get_user_pages() slow path. The slow path is protected by @@ -663,94 +566,68 @@ struct page *dax_layout_busy_page(struct address_space *mapping) */ unmap_mapping_range(mapping, 0, 0, 1); - while (index < end && pagevec_lookup_entries(&pvec, mapping, index, - min(end - index, (pgoff_t)PAGEVEC_SIZE), - indices)) { - pgoff_t nr_pages = 1; - - for (i = 0; i < pagevec_count(&pvec); i++) { - struct page *pvec_ent = pvec.pages[i]; - void *entry; - - index = indices[i]; - if (index >= end) - break; - - if (WARN_ON_ONCE( - !radix_tree_exceptional_entry(pvec_ent))) - continue; - - xa_lock_irq(&mapping->i_pages); - entry = get_unlocked_mapping_entry(mapping, index, NULL); - if (entry) { - page = dax_busy_page(entry); - /* - * Account for multi-order entries at - * the end of the pagevec. - */ - if (i + 1 >= pagevec_count(&pvec)) - nr_pages = 1UL << dax_radix_order(entry); - } - put_unlocked_mapping_entry(mapping, index, entry); - xa_unlock_irq(&mapping->i_pages); - if (page) - break; - } - - /* - * We don't expect normal struct page entries to exist in our - * tree, but we keep these pagevec calls so that this code is - * consistent with the common pattern for handling pagevecs - * throughout the kernel. - */ - pagevec_remove_exceptionals(&pvec); - pagevec_release(&pvec); - index += nr_pages; - + xas_lock_irq(&xas); + xas_for_each(&xas, entry, ULONG_MAX) { + if (WARN_ON_ONCE(!xa_is_value(entry))) + continue; + if (unlikely(dax_is_locked(entry))) + entry = get_unlocked_entry(&xas); + if (entry) + page = dax_busy_page(entry); + put_unlocked_entry(&xas, entry); if (page) break; + if (++scanned % XA_CHECK_SCHED) + continue; + + xas_pause(&xas); + xas_unlock_irq(&xas); + cond_resched(); + xas_lock_irq(&xas); } + xas_unlock_irq(&xas); return page; } EXPORT_SYMBOL_GPL(dax_layout_busy_page); -static int __dax_invalidate_mapping_entry(struct address_space *mapping, +static int __dax_invalidate_entry(struct address_space *mapping, pgoff_t index, bool trunc) { + XA_STATE(xas, &mapping->i_pages, index); int ret = 0; void *entry; - struct radix_tree_root *pages = &mapping->i_pages; - xa_lock_irq(pages); - entry = get_unlocked_mapping_entry(mapping, index, NULL); - if (!entry || WARN_ON_ONCE(!radix_tree_exceptional_entry(entry))) + xas_lock_irq(&xas); + entry = get_unlocked_entry(&xas); + if (!entry || WARN_ON_ONCE(!xa_is_value(entry))) goto out; if (!trunc && - (radix_tree_tag_get(pages, index, PAGECACHE_TAG_DIRTY) || - radix_tree_tag_get(pages, index, PAGECACHE_TAG_TOWRITE))) + (xas_get_mark(&xas, PAGECACHE_TAG_DIRTY) || + xas_get_mark(&xas, PAGECACHE_TAG_TOWRITE))) goto out; dax_disassociate_entry(entry, mapping, trunc); - radix_tree_delete(pages, index); + xas_store(&xas, NULL); mapping->nrexceptional--; ret = 1; out: - put_unlocked_mapping_entry(mapping, index, entry); - xa_unlock_irq(pages); + put_unlocked_entry(&xas, entry); + xas_unlock_irq(&xas); return ret; } + /* - * Delete exceptional DAX entry at @index from @mapping. Wait for radix tree - * entry to get unlocked before deleting it. + * Delete DAX entry at @index from @mapping. Wait for it + * to be unlocked before deleting it. */ int dax_delete_mapping_entry(struct address_space *mapping, pgoff_t index) { - int ret = __dax_invalidate_mapping_entry(mapping, index, true); + int ret = __dax_invalidate_entry(mapping, index, true); /* * This gets called from truncate / punch_hole path. As such, the caller * must hold locks protecting against concurrent modifications of the - * radix tree (usually fs-private i_mmap_sem for writing). Since the - * caller has seen exceptional entry for this index, we better find it + * page cache (usually fs-private i_mmap_sem for writing). Since the + * caller has seen a DAX entry for this index, we better find it * at that index as well... */ WARN_ON_ONCE(!ret); @@ -758,12 +635,12 @@ int dax_delete_mapping_entry(struct address_space *mapping, pgoff_t index) } /* - * Invalidate exceptional DAX entry if it is clean. + * Invalidate DAX entry if it is clean. */ int dax_invalidate_mapping_entry_sync(struct address_space *mapping, pgoff_t index) { - return __dax_invalidate_mapping_entry(mapping, index, false); + return __dax_invalidate_entry(mapping, index, false); } static int copy_user_dax(struct block_device *bdev, struct dax_device *dax_dev, @@ -799,30 +676,27 @@ static int copy_user_dax(struct block_device *bdev, struct dax_device *dax_dev, * already in the tree, we will skip the insertion and just dirty the PMD as * appropriate. */ -static void *dax_insert_mapping_entry(struct address_space *mapping, - struct vm_fault *vmf, - void *entry, pfn_t pfn_t, - unsigned long flags, bool dirty) +static void *dax_insert_entry(struct xa_state *xas, + struct address_space *mapping, struct vm_fault *vmf, + void *entry, pfn_t pfn, unsigned long flags, bool dirty) { - struct radix_tree_root *pages = &mapping->i_pages; - unsigned long pfn = pfn_t_to_pfn(pfn_t); - pgoff_t index = vmf->pgoff; - void *new_entry; + void *new_entry = dax_make_entry(pfn, flags); if (dirty) __mark_inode_dirty(mapping->host, I_DIRTY_PAGES); - if (dax_is_zero_entry(entry) && !(flags & RADIX_DAX_ZERO_PAGE)) { + if (dax_is_zero_entry(entry) && !(flags & DAX_ZERO_PAGE)) { + unsigned long index = xas->xa_index; /* we are replacing a zero page with block mapping */ if (dax_is_pmd_entry(entry)) unmap_mapping_pages(mapping, index & ~PG_PMD_COLOUR, - PG_PMD_NR, false); + PG_PMD_NR, false); else /* pte entry */ - unmap_mapping_pages(mapping, vmf->pgoff, 1, false); + unmap_mapping_pages(mapping, index, 1, false); } - xa_lock_irq(pages); - new_entry = dax_radix_locked_entry(pfn, flags); + xas_reset(xas); + xas_lock_irq(xas); if (dax_entry_size(entry) != dax_entry_size(new_entry)) { dax_disassociate_entry(entry, mapping, false); dax_associate_entry(new_entry, mapping, vmf->vma, vmf->address); @@ -830,33 +704,30 @@ static void *dax_insert_mapping_entry(struct address_space *mapping, if (dax_is_zero_entry(entry) || dax_is_empty_entry(entry)) { /* - * Only swap our new entry into the radix tree if the current + * Only swap our new entry into the page cache if the current * entry is a zero page or an empty entry. If a normal PTE or - * PMD entry is already in the tree, we leave it alone. This + * PMD entry is already in the cache, we leave it alone. This * means that if we are trying to insert a PTE and the * existing entry is a PMD, we will just leave the PMD in the * tree and dirty it if necessary. */ - struct radix_tree_node *node; - void **slot; - void *ret; - - ret = __radix_tree_lookup(pages, index, &node, &slot); - WARN_ON_ONCE(ret != entry); - __radix_tree_replace(pages, node, slot, - new_entry, NULL); + void *old = dax_lock_entry(xas, new_entry); + WARN_ON_ONCE(old != xa_mk_value(xa_to_value(entry) | + DAX_LOCKED)); entry = new_entry; + } else { + xas_load(xas); /* Walk the xa_state */ } if (dirty) - radix_tree_tag_set(pages, index, PAGECACHE_TAG_DIRTY); + xas_set_mark(xas, PAGECACHE_TAG_DIRTY); - xa_unlock_irq(pages); + xas_unlock_irq(xas); return entry; } -static inline unsigned long -pgoff_address(pgoff_t pgoff, struct vm_area_struct *vma) +static inline +unsigned long pgoff_address(pgoff_t pgoff, struct vm_area_struct *vma) { unsigned long address; @@ -866,8 +737,8 @@ pgoff_address(pgoff_t pgoff, struct vm_area_struct *vma) } /* Walk all mappings of a given index of a file and writeprotect them */ -static void dax_mapping_entry_mkclean(struct address_space *mapping, - pgoff_t index, unsigned long pfn) +static void dax_entry_mkclean(struct address_space *mapping, pgoff_t index, + unsigned long pfn) { struct vm_area_struct *vma; pte_t pte, *ptep = NULL; @@ -937,11 +808,9 @@ unlock_pte: i_mmap_unlock_read(mapping); } -static int dax_writeback_one(struct dax_device *dax_dev, - struct address_space *mapping, pgoff_t index, void *entry) +static int dax_writeback_one(struct xa_state *xas, struct dax_device *dax_dev, + struct address_space *mapping, void *entry) { - struct radix_tree_root *pages = &mapping->i_pages; - void *entry2, **slot; unsigned long pfn; long ret = 0; size_t size; @@ -950,32 +819,38 @@ static int dax_writeback_one(struct dax_device *dax_dev, * A page got tagged dirty in DAX mapping? Something is seriously * wrong. */ - if (WARN_ON(!radix_tree_exceptional_entry(entry))) + if (WARN_ON(!xa_is_value(entry))) return -EIO; - xa_lock_irq(pages); - entry2 = get_unlocked_mapping_entry(mapping, index, &slot); - /* Entry got punched out / reallocated? */ - if (!entry2 || WARN_ON_ONCE(!radix_tree_exceptional_entry(entry2))) - goto put_unlocked; - /* - * Entry got reallocated elsewhere? No need to writeback. We have to - * compare pfns as we must not bail out due to difference in lockbit - * or entry type. - */ - if (dax_radix_pfn(entry2) != dax_radix_pfn(entry)) - goto put_unlocked; - if (WARN_ON_ONCE(dax_is_empty_entry(entry) || - dax_is_zero_entry(entry))) { - ret = -EIO; - goto put_unlocked; + if (unlikely(dax_is_locked(entry))) { + void *old_entry = entry; + + entry = get_unlocked_entry(xas); + + /* Entry got punched out / reallocated? */ + if (!entry || WARN_ON_ONCE(!xa_is_value(entry))) + goto put_unlocked; + /* + * Entry got reallocated elsewhere? No need to writeback. + * We have to compare pfns as we must not bail out due to + * difference in lockbit or entry type. + */ + if (dax_to_pfn(old_entry) != dax_to_pfn(entry)) + goto put_unlocked; + if (WARN_ON_ONCE(dax_is_empty_entry(entry) || + dax_is_zero_entry(entry))) { + ret = -EIO; + goto put_unlocked; + } + + /* Another fsync thread may have already done this entry */ + if (!xas_get_mark(xas, PAGECACHE_TAG_TOWRITE)) + goto put_unlocked; } - /* Another fsync thread may have already written back this entry */ - if (!radix_tree_tag_get(pages, index, PAGECACHE_TAG_TOWRITE)) - goto put_unlocked; /* Lock the entry to serialize with page faults */ - entry = lock_slot(mapping, slot); + dax_lock_entry(xas, entry); + /* * We can clear the tag now but we have to be careful so that concurrent * dax_writeback_one() calls for the same index cannot finish before we @@ -983,8 +858,8 @@ static int dax_writeback_one(struct dax_device *dax_dev, * at the entry only under the i_pages lock and once they do that * they will see the entry locked and wait for it to unlock. */ - radix_tree_tag_clear(pages, index, PAGECACHE_TAG_TOWRITE); - xa_unlock_irq(pages); + xas_clear_mark(xas, PAGECACHE_TAG_TOWRITE); + xas_unlock_irq(xas); /* * Even if dax_writeback_mapping_range() was given a wbc->range_start @@ -993,10 +868,10 @@ static int dax_writeback_one(struct dax_device *dax_dev, * This allows us to flush for PMD_SIZE and not have to worry about * partial PMD writebacks. */ - pfn = dax_radix_pfn(entry); - size = PAGE_SIZE << dax_radix_order(entry); + pfn = dax_to_pfn(entry); + size = PAGE_SIZE << dax_entry_order(entry); - dax_mapping_entry_mkclean(mapping, index, pfn); + dax_entry_mkclean(mapping, xas->xa_index, pfn); dax_flush(dax_dev, page_address(pfn_to_page(pfn)), size); /* * After we have flushed the cache, we can clear the dirty tag. There @@ -1004,16 +879,18 @@ static int dax_writeback_one(struct dax_device *dax_dev, * the pfn mappings are writeprotected and fault waits for mapping * entry lock. */ - xa_lock_irq(pages); - radix_tree_tag_clear(pages, index, PAGECACHE_TAG_DIRTY); - xa_unlock_irq(pages); - trace_dax_writeback_one(mapping->host, index, size >> PAGE_SHIFT); - put_locked_mapping_entry(mapping, index); + xas_reset(xas); + xas_lock_irq(xas); + xas_store(xas, entry); + xas_clear_mark(xas, PAGECACHE_TAG_DIRTY); + dax_wake_entry(xas, entry, false); + + trace_dax_writeback_one(mapping->host, xas->xa_index, + size >> PAGE_SHIFT); return ret; put_unlocked: - put_unlocked_mapping_entry(mapping, index, entry2); - xa_unlock_irq(pages); + put_unlocked_entry(xas, entry); return ret; } @@ -1025,13 +902,13 @@ static int dax_writeback_one(struct dax_device *dax_dev, int dax_writeback_mapping_range(struct address_space *mapping, struct block_device *bdev, struct writeback_control *wbc) { + XA_STATE(xas, &mapping->i_pages, wbc->range_start >> PAGE_SHIFT); struct inode *inode = mapping->host; - pgoff_t start_index, end_index; - pgoff_t indices[PAGEVEC_SIZE]; + pgoff_t end_index = wbc->range_end >> PAGE_SHIFT; struct dax_device *dax_dev; - struct pagevec pvec; - bool done = false; - int i, ret = 0; + void *entry; + int ret = 0; + unsigned int scanned = 0; if (WARN_ON_ONCE(inode->i_blkbits != PAGE_SHIFT)) return -EIO; @@ -1043,41 +920,29 @@ int dax_writeback_mapping_range(struct address_space *mapping, if (!dax_dev) return -EIO; - start_index = wbc->range_start >> PAGE_SHIFT; - end_index = wbc->range_end >> PAGE_SHIFT; - - trace_dax_writeback_range(inode, start_index, end_index); + trace_dax_writeback_range(inode, xas.xa_index, end_index); - tag_pages_for_writeback(mapping, start_index, end_index); + tag_pages_for_writeback(mapping, xas.xa_index, end_index); - pagevec_init(&pvec); - while (!done) { - pvec.nr = find_get_entries_tag(mapping, start_index, - PAGECACHE_TAG_TOWRITE, PAGEVEC_SIZE, - pvec.pages, indices); - - if (pvec.nr == 0) + xas_lock_irq(&xas); + xas_for_each_marked(&xas, entry, end_index, PAGECACHE_TAG_TOWRITE) { + ret = dax_writeback_one(&xas, dax_dev, mapping, entry); + if (ret < 0) { + mapping_set_error(mapping, ret); break; - - for (i = 0; i < pvec.nr; i++) { - if (indices[i] > end_index) { - done = true; - break; - } - - ret = dax_writeback_one(dax_dev, mapping, indices[i], - pvec.pages[i]); - if (ret < 0) { - mapping_set_error(mapping, ret); - goto out; - } } - start_index = indices[pvec.nr - 1] + 1; + if (++scanned % XA_CHECK_SCHED) + continue; + + xas_pause(&xas); + xas_unlock_irq(&xas); + cond_resched(); + xas_lock_irq(&xas); } -out: + xas_unlock_irq(&xas); put_dax(dax_dev); - trace_dax_writeback_range_done(inode, start_index, end_index); - return (ret < 0 ? ret : 0); + trace_dax_writeback_range_done(inode, xas.xa_index, end_index); + return ret; } EXPORT_SYMBOL_GPL(dax_writeback_mapping_range); @@ -1125,16 +990,18 @@ out: * If this page is ever written to we will re-fault and change the mapping to * point to real DAX storage instead. */ -static vm_fault_t dax_load_hole(struct address_space *mapping, void *entry, - struct vm_fault *vmf) +static vm_fault_t dax_load_hole(struct xa_state *xas, + struct address_space *mapping, void **entry, + struct vm_fault *vmf) { struct inode *inode = mapping->host; unsigned long vaddr = vmf->address; pfn_t pfn = pfn_to_pfn_t(my_zero_pfn(vaddr)); vm_fault_t ret; - dax_insert_mapping_entry(mapping, vmf, entry, pfn, RADIX_DAX_ZERO_PAGE, - false); + *entry = dax_insert_entry(xas, mapping, vmf, *entry, pfn, + DAX_ZERO_PAGE, false); + ret = vmf_insert_mixed(vmf->vma, vaddr, pfn); trace_dax_load_hole(inode, vmf, ret); return ret; @@ -1342,6 +1209,7 @@ static vm_fault_t dax_iomap_pte_fault(struct vm_fault *vmf, pfn_t *pfnp, { struct vm_area_struct *vma = vmf->vma; struct address_space *mapping = vma->vm_file->f_mapping; + XA_STATE(xas, &mapping->i_pages, vmf->pgoff); struct inode *inode = mapping->host; unsigned long vaddr = vmf->address; loff_t pos = (loff_t)vmf->pgoff << PAGE_SHIFT; @@ -1368,9 +1236,9 @@ static vm_fault_t dax_iomap_pte_fault(struct vm_fault *vmf, pfn_t *pfnp, if (write && !vmf->cow_page) flags |= IOMAP_WRITE; - entry = grab_mapping_entry(mapping, vmf->pgoff, 0); - if (IS_ERR(entry)) { - ret = dax_fault_return(PTR_ERR(entry)); + entry = grab_mapping_entry(&xas, mapping, 0); + if (xa_is_internal(entry)) { + ret = xa_to_internal(entry); goto out; } @@ -1443,7 +1311,7 @@ static vm_fault_t dax_iomap_pte_fault(struct vm_fault *vmf, pfn_t *pfnp, if (error < 0) goto error_finish_iomap; - entry = dax_insert_mapping_entry(mapping, vmf, entry, pfn, + entry = dax_insert_entry(&xas, mapping, vmf, entry, pfn, 0, write && !sync); /* @@ -1471,7 +1339,7 @@ static vm_fault_t dax_iomap_pte_fault(struct vm_fault *vmf, pfn_t *pfnp, case IOMAP_UNWRITTEN: case IOMAP_HOLE: if (!write) { - ret = dax_load_hole(mapping, entry, vmf); + ret = dax_load_hole(&xas, mapping, &entry, vmf); goto finish_iomap; } /*FALLTHRU*/ @@ -1498,21 +1366,20 @@ static vm_fault_t dax_iomap_pte_fault(struct vm_fault *vmf, pfn_t *pfnp, ops->iomap_end(inode, pos, PAGE_SIZE, copied, flags, &iomap); } unlock_entry: - put_locked_mapping_entry(mapping, vmf->pgoff); + dax_unlock_entry(&xas, entry); out: trace_dax_pte_fault_done(inode, vmf, ret); return ret | major; } #ifdef CONFIG_FS_DAX_PMD -static vm_fault_t dax_pmd_load_hole(struct vm_fault *vmf, struct iomap *iomap, - void *entry) +static vm_fault_t dax_pmd_load_hole(struct xa_state *xas, struct vm_fault *vmf, + struct iomap *iomap, void **entry) { struct address_space *mapping = vmf->vma->vm_file->f_mapping; unsigned long pmd_addr = vmf->address & PMD_MASK; struct inode *inode = mapping->host; struct page *zero_page; - void *ret = NULL; spinlock_t *ptl; pmd_t pmd_entry; pfn_t pfn; @@ -1523,8 +1390,8 @@ static vm_fault_t dax_pmd_load_hole(struct vm_fault *vmf, struct iomap *iomap, goto fallback; pfn = page_to_pfn_t(zero_page); - ret = dax_insert_mapping_entry(mapping, vmf, entry, pfn, - RADIX_DAX_PMD | RADIX_DAX_ZERO_PAGE, false); + *entry = dax_insert_entry(xas, mapping, vmf, *entry, pfn, + DAX_PMD | DAX_ZERO_PAGE, false); ptl = pmd_lock(vmf->vma->vm_mm, vmf->pmd); if (!pmd_none(*(vmf->pmd))) { @@ -1536,11 +1403,11 @@ static vm_fault_t dax_pmd_load_hole(struct vm_fault *vmf, struct iomap *iomap, pmd_entry = pmd_mkhuge(pmd_entry); set_pmd_at(vmf->vma->vm_mm, pmd_addr, vmf->pmd, pmd_entry); spin_unlock(ptl); - trace_dax_pmd_load_hole(inode, vmf, zero_page, ret); + trace_dax_pmd_load_hole(inode, vmf, zero_page, *entry); return VM_FAULT_NOPAGE; fallback: - trace_dax_pmd_load_hole_fallback(inode, vmf, zero_page, ret); + trace_dax_pmd_load_hole_fallback(inode, vmf, zero_page, *entry); return VM_FAULT_FALLBACK; } @@ -1549,6 +1416,7 @@ static vm_fault_t dax_iomap_pmd_fault(struct vm_fault *vmf, pfn_t *pfnp, { struct vm_area_struct *vma = vmf->vma; struct address_space *mapping = vma->vm_file->f_mapping; + XA_STATE_ORDER(xas, &mapping->i_pages, vmf->pgoff, PMD_ORDER); unsigned long pmd_addr = vmf->address & PMD_MASK; bool write = vmf->flags & FAULT_FLAG_WRITE; bool sync; @@ -1556,7 +1424,7 @@ static vm_fault_t dax_iomap_pmd_fault(struct vm_fault *vmf, pfn_t *pfnp, struct inode *inode = mapping->host; vm_fault_t result = VM_FAULT_FALLBACK; struct iomap iomap = { 0 }; - pgoff_t max_pgoff, pgoff; + pgoff_t max_pgoff; void *entry; loff_t pos; int error; @@ -1567,7 +1435,6 @@ static vm_fault_t dax_iomap_pmd_fault(struct vm_fault *vmf, pfn_t *pfnp, * supposed to hold locks serializing us with truncate / punch hole so * this is a reliable test. */ - pgoff = linear_page_index(vma, pmd_addr); max_pgoff = DIV_ROUND_UP(i_size_read(inode), PAGE_SIZE); trace_dax_pmd_fault(inode, vmf, max_pgoff, 0); @@ -1576,7 +1443,7 @@ static vm_fault_t dax_iomap_pmd_fault(struct vm_fault *vmf, pfn_t *pfnp, * Make sure that the faulting address's PMD offset (color) matches * the PMD offset from the start of the file. This is necessary so * that a PMD range in the page table overlaps exactly with a PMD - * range in the radix tree. + * range in the page cache. */ if ((vmf->pgoff & PG_PMD_COLOUR) != ((vmf->address >> PAGE_SHIFT) & PG_PMD_COLOUR)) @@ -1592,24 +1459,26 @@ static vm_fault_t dax_iomap_pmd_fault(struct vm_fault *vmf, pfn_t *pfnp, if ((pmd_addr + PMD_SIZE) > vma->vm_end) goto fallback; - if (pgoff >= max_pgoff) { + if (xas.xa_index >= max_pgoff) { result = VM_FAULT_SIGBUS; goto out; } /* If the PMD would extend beyond the file size */ - if ((pgoff | PG_PMD_COLOUR) >= max_pgoff) + if ((xas.xa_index | PG_PMD_COLOUR) >= max_pgoff) goto fallback; /* - * grab_mapping_entry() will make sure we get a 2MiB empty entry, a - * 2MiB zero page entry or a DAX PMD. If it can't (because a 4k page - * is already in the tree, for instance), it will return -EEXIST and - * we just fall back to 4k entries. + * grab_mapping_entry() will make sure we get an empty PMD entry, + * a zero PMD entry or a DAX PMD. If it can't (because a PTE + * entry is already in the array, for instance), it will return + * VM_FAULT_FALLBACK. */ - entry = grab_mapping_entry(mapping, pgoff, RADIX_DAX_PMD); - if (IS_ERR(entry)) + entry = grab_mapping_entry(&xas, mapping, DAX_PMD); + if (xa_is_internal(entry)) { + result = xa_to_internal(entry); goto fallback; + } /* * It is possible, particularly with mixed reads & writes to private @@ -1628,7 +1497,7 @@ static vm_fault_t dax_iomap_pmd_fault(struct vm_fault *vmf, pfn_t *pfnp, * setting up a mapping, so really we're using iomap_begin() as a way * to look up our filesystem block. */ - pos = (loff_t)pgoff << PAGE_SHIFT; + pos = (loff_t)xas.xa_index << PAGE_SHIFT; error = ops->iomap_begin(inode, pos, PMD_SIZE, iomap_flags, &iomap); if (error) goto unlock_entry; @@ -1644,8 +1513,8 @@ static vm_fault_t dax_iomap_pmd_fault(struct vm_fault *vmf, pfn_t *pfnp, if (error < 0) goto finish_iomap; - entry = dax_insert_mapping_entry(mapping, vmf, entry, pfn, - RADIX_DAX_PMD, write && !sync); + entry = dax_insert_entry(&xas, mapping, vmf, entry, pfn, + DAX_PMD, write && !sync); /* * If we are doing synchronous page fault and inode needs fsync, @@ -1669,7 +1538,7 @@ static vm_fault_t dax_iomap_pmd_fault(struct vm_fault *vmf, pfn_t *pfnp, case IOMAP_HOLE: if (WARN_ON_ONCE(write)) break; - result = dax_pmd_load_hole(vmf, &iomap, entry); + result = dax_pmd_load_hole(&xas, vmf, &iomap, &entry); break; default: WARN_ON_ONCE(1); @@ -1692,7 +1561,7 @@ static vm_fault_t dax_iomap_pmd_fault(struct vm_fault *vmf, pfn_t *pfnp, &iomap); } unlock_entry: - put_locked_mapping_entry(mapping, pgoff); + dax_unlock_entry(&xas, entry); fallback: if (result == VM_FAULT_FALLBACK) { split_huge_pmd(vma, vmf->pmd, vmf->address); @@ -1737,54 +1606,49 @@ vm_fault_t dax_iomap_fault(struct vm_fault *vmf, enum page_entry_size pe_size, } EXPORT_SYMBOL_GPL(dax_iomap_fault); -/** +/* * dax_insert_pfn_mkwrite - insert PTE or PMD entry into page tables * @vmf: The description of the fault - * @pe_size: Size of entry to be inserted * @pfn: PFN to insert + * @order: Order of entry to insert. * - * This function inserts writeable PTE or PMD entry into page tables for mmaped - * DAX file. It takes care of marking corresponding radix tree entry as dirty - * as well. + * This function inserts a writeable PTE or PMD entry into the page tables + * for an mmaped DAX file. It also marks the page cache entry as dirty. */ -static vm_fault_t dax_insert_pfn_mkwrite(struct vm_fault *vmf, - enum page_entry_size pe_size, - pfn_t pfn) +static vm_fault_t +dax_insert_pfn_mkwrite(struct vm_fault *vmf, pfn_t pfn, unsigned int order) { struct address_space *mapping = vmf->vma->vm_file->f_mapping; - void *entry, **slot; - pgoff_t index = vmf->pgoff; + XA_STATE_ORDER(xas, &mapping->i_pages, vmf->pgoff, order); + void *entry; vm_fault_t ret; - xa_lock_irq(&mapping->i_pages); - entry = get_unlocked_mapping_entry(mapping, index, &slot); + xas_lock_irq(&xas); + entry = get_unlocked_entry(&xas); /* Did we race with someone splitting entry or so? */ if (!entry || - (pe_size == PE_SIZE_PTE && !dax_is_pte_entry(entry)) || - (pe_size == PE_SIZE_PMD && !dax_is_pmd_entry(entry))) { - put_unlocked_mapping_entry(mapping, index, entry); - xa_unlock_irq(&mapping->i_pages); + (order == 0 && !dax_is_pte_entry(entry)) || + (order == PMD_ORDER && (xa_is_internal(entry) || + !dax_is_pmd_entry(entry)))) { + put_unlocked_entry(&xas, entry); + xas_unlock_irq(&xas); trace_dax_insert_pfn_mkwrite_no_entry(mapping->host, vmf, VM_FAULT_NOPAGE); return VM_FAULT_NOPAGE; } - radix_tree_tag_set(&mapping->i_pages, index, PAGECACHE_TAG_DIRTY); - entry = lock_slot(mapping, slot); - xa_unlock_irq(&mapping->i_pages); - switch (pe_size) { - case PE_SIZE_PTE: + xas_set_mark(&xas, PAGECACHE_TAG_DIRTY); + dax_lock_entry(&xas, entry); + xas_unlock_irq(&xas); + if (order == 0) ret = vmf_insert_mixed_mkwrite(vmf->vma, vmf->address, pfn); - break; #ifdef CONFIG_FS_DAX_PMD - case PE_SIZE_PMD: + else if (order == PMD_ORDER) ret = vmf_insert_pfn_pmd(vmf->vma, vmf->address, vmf->pmd, pfn, true); - break; #endif - default: + else ret = VM_FAULT_FALLBACK; - } - put_locked_mapping_entry(mapping, index); + dax_unlock_entry(&xas, entry); trace_dax_insert_pfn_mkwrite(mapping->host, vmf, ret); return ret; } @@ -1804,17 +1668,12 @@ vm_fault_t dax_finish_sync_fault(struct vm_fault *vmf, { int err; loff_t start = ((loff_t)vmf->pgoff) << PAGE_SHIFT; - size_t len = 0; + unsigned int order = pe_order(pe_size); + size_t len = PAGE_SIZE << order; - if (pe_size == PE_SIZE_PTE) - len = PAGE_SIZE; - else if (pe_size == PE_SIZE_PMD) - len = PMD_SIZE; - else - WARN_ON_ONCE(1); err = vfs_fsync_range(vmf->vma->vm_file, start, start + len - 1, 1); if (err) return VM_FAULT_SIGBUS; - return dax_insert_pfn_mkwrite(vmf, pe_size, pfn); + return dax_insert_pfn_mkwrite(vmf, pfn, order); } EXPORT_SYMBOL_GPL(dax_finish_sync_fault); diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c @@ -2643,7 +2643,7 @@ static int mpage_prepare_extent_to_map(struct mpage_da_data *mpd) long left = mpd->wbc->nr_to_write; pgoff_t index = mpd->first_page; pgoff_t end = mpd->last_page; - int tag; + xa_mark_t tag; int i, err = 0; int blkbits = mpd->inode->i_blkbits; ext4_lblk_t lblk; diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c @@ -2071,7 +2071,7 @@ static int f2fs_write_cache_pages(struct address_space *mapping, pgoff_t done_index; int cycled; int range_whole = 0; - int tag; + xa_mark_t tag; int nwritten = 0; pagevec_init(&pvec); @@ -2787,13 +2787,13 @@ const struct address_space_operations f2fs_dblock_aops = { #endif }; -void f2fs_clear_radix_tree_dirty_tag(struct page *page) +void f2fs_clear_page_cache_dirty_tag(struct page *page) { struct address_space *mapping = page_mapping(page); unsigned long flags; xa_lock_irqsave(&mapping->i_pages, flags); - radix_tree_tag_clear(&mapping->i_pages, page_index(page), + __xa_clear_mark(&mapping->i_pages, page_index(page), PAGECACHE_TAG_DIRTY); xa_unlock_irqrestore(&mapping->i_pages, flags); } diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c @@ -726,7 +726,7 @@ void f2fs_delete_entry(struct f2fs_dir_entry *dentry, struct page *page, if (bit_pos == NR_DENTRY_IN_BLOCK && !f2fs_truncate_hole(dir, page->index, page->index + 1)) { - f2fs_clear_radix_tree_dirty_tag(page); + f2fs_clear_page_cache_dirty_tag(page); clear_page_dirty_for_io(page); ClearPagePrivate(page); ClearPageUptodate(page); diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h @@ -3108,7 +3108,7 @@ int f2fs_migrate_page(struct address_space *mapping, struct page *newpage, struct page *page, enum migrate_mode mode); #endif bool f2fs_overwrite_io(struct inode *inode, loff_t pos, size_t len); -void f2fs_clear_radix_tree_dirty_tag(struct page *page); +void f2fs_clear_page_cache_dirty_tag(struct page *page); /* * gc.c diff --git a/fs/f2fs/inline.c b/fs/f2fs/inline.c @@ -243,7 +243,7 @@ int f2fs_write_inline_data(struct inode *inode, struct page *page) kunmap_atomic(src_addr); set_page_dirty(dn.inode_page); - f2fs_clear_radix_tree_dirty_tag(page); + f2fs_clear_page_cache_dirty_tag(page); set_inode_flag(inode, FI_APPEND_WRITE); set_inode_flag(inode, FI_DATA_EXIST); diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c @@ -101,7 +101,7 @@ bool f2fs_available_free_memory(struct f2fs_sb_info *sbi, int type) static void clear_node_page_dirty(struct page *page) { if (PageDirty(page)) { - f2fs_clear_radix_tree_dirty_tag(page); + f2fs_clear_page_cache_dirty_tag(page); clear_page_dirty_for_io(page); dec_page_count(F2FS_P_SB(page), F2FS_DIRTY_NODES); } @@ -1306,9 +1306,7 @@ void f2fs_ra_node_page(struct f2fs_sb_info *sbi, nid_t nid) if (f2fs_check_nid_range(sbi, nid)) return; - rcu_read_lock(); - apage = radix_tree_lookup(&NODE_MAPPING(sbi)->i_pages, nid); - rcu_read_unlock(); + apage = xa_load(&NODE_MAPPING(sbi)->i_pages, nid); if (apage) return; diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c @@ -339,9 +339,9 @@ static void inode_switch_wbs_work_fn(struct work_struct *work) struct address_space *mapping = inode->i_mapping; struct bdi_writeback *old_wb = inode->i_wb; struct bdi_writeback *new_wb = isw->new_wb; - struct radix_tree_iter iter; + XA_STATE(xas, &mapping->i_pages, 0); + struct page *page; bool switched = false; - void **slot; /* * By the time control reaches here, RCU grace period has passed @@ -375,25 +375,18 @@ static void inode_switch_wbs_work_fn(struct work_struct *work) * to possibly dirty pages while PAGECACHE_TAG_WRITEBACK points to * pages actually under writeback. */ - radix_tree_for_each_tagged(slot, &mapping->i_pages, &iter, 0, - PAGECACHE_TAG_DIRTY) { - struct page *page = radix_tree_deref_slot_protected(slot, - &mapping->i_pages.xa_lock); - if (likely(page) && PageDirty(page)) { + xas_for_each_marked(&xas, page, ULONG_MAX, PAGECACHE_TAG_DIRTY) { + if (PageDirty(page)) { dec_wb_stat(old_wb, WB_RECLAIMABLE); inc_wb_stat(new_wb, WB_RECLAIMABLE); } } - radix_tree_for_each_tagged(slot, &mapping->i_pages, &iter, 0, - PAGECACHE_TAG_WRITEBACK) { - struct page *page = radix_tree_deref_slot_protected(slot, - &mapping->i_pages.xa_lock); - if (likely(page)) { - WARN_ON_ONCE(!PageWriteback(page)); - dec_wb_stat(old_wb, WB_WRITEBACK); - inc_wb_stat(new_wb, WB_WRITEBACK); - } + xas_set(&xas, 0); + xas_for_each_marked(&xas, page, ULONG_MAX, PAGECACHE_TAG_WRITEBACK) { + WARN_ON_ONCE(!PageWriteback(page)); + dec_wb_stat(old_wb, WB_WRITEBACK); + inc_wb_stat(new_wb, WB_WRITEBACK); } wb_get(new_wb); diff --git a/fs/gfs2/aops.c b/fs/gfs2/aops.c @@ -366,7 +366,7 @@ static int gfs2_write_cache_jdata(struct address_space *mapping, pgoff_t done_index; int cycled; int range_whole = 0; - int tag; + xa_mark_t tag; pagevec_init(&pvec); if (wbc->range_cyclic) { diff --git a/fs/inode.c b/fs/inode.c @@ -349,7 +349,7 @@ EXPORT_SYMBOL(inc_nlink); static void __address_space_init_once(struct address_space *mapping) { - INIT_RADIX_TREE(&mapping->i_pages, GFP_ATOMIC | __GFP_ACCOUNT); + xa_init_flags(&mapping->i_pages, XA_FLAGS_LOCK_IRQ); init_rwsem(&mapping->i_mmap_rwsem); INIT_LIST_HEAD(&mapping->private_list); spin_lock_init(&mapping->private_lock); diff --git a/fs/isofs/dir.c b/fs/isofs/dir.c @@ -46,7 +46,7 @@ int isofs_name_translate(struct iso_directory_record *de, char *new, struct inod return i; } -/* Acorn extensions written by Matthew Wilcox <willy@bofh.ai> 1998 */ +/* Acorn extensions written by Matthew Wilcox <willy@infradead.org> 1998 */ int get_acorn_filename(struct iso_directory_record *de, char *retname, struct inode *inode) { diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c @@ -896,7 +896,7 @@ static u64 pnfs_num_cont_bytes(struct inode *inode, pgoff_t idx) end = DIV_ROUND_UP(i_size_read(inode), PAGE_SIZE); if (end != inode->i_mapping->nrpages) { rcu_read_lock(); - end = page_cache_next_hole(mapping, idx + 1, ULONG_MAX); + end = page_cache_next_miss(mapping, idx + 1, ULONG_MAX); rcu_read_unlock(); } diff --git a/fs/nilfs2/btnode.c b/fs/nilfs2/btnode.c @@ -168,24 +168,18 @@ int nilfs_btnode_prepare_change_key(struct address_space *btnc, ctxt->newbh = NULL; if (inode->i_blkbits == PAGE_SHIFT) { - lock_page(obh->b_page); - /* - * We cannot call radix_tree_preload for the kernels older - * than 2.6.23, because it is not exported for modules. - */ + struct page *opage = obh->b_page; + lock_page(opage); retry: - err = radix_tree_preload(GFP_NOFS & ~__GFP_HIGHMEM); - if (err) - goto failed_unlock; /* BUG_ON(oldkey != obh->b_page->index); */ - if (unlikely(oldkey != obh->b_page->index)) - NILFS_PAGE_BUG(obh->b_page, + if (unlikely(oldkey != opage->index)) + NILFS_PAGE_BUG(opage, "invalid oldkey %lld (newkey=%lld)", (unsigned long long)oldkey, (unsigned long long)newkey); xa_lock_irq(&btnc->i_pages); - err = radix_tree_insert(&btnc->i_pages, newkey, obh->b_page); + err = __xa_insert(&btnc->i_pages, newkey, opage, GFP_NOFS); xa_unlock_irq(&btnc->i_pages); /* * Note: page->index will not change to newkey until @@ -193,7 +187,6 @@ retry: * To protect the page in intermediate state, the page lock * is held. */ - radix_tree_preload_end(); if (!err) return 0; else if (err != -EEXIST) @@ -203,7 +196,7 @@ retry: if (!err) goto retry; /* fallback to copy mode */ - unlock_page(obh->b_page); + unlock_page(opage); } nbh = nilfs_btnode_create_block(btnc, newkey); @@ -243,9 +236,8 @@ void nilfs_btnode_commit_change_key(struct address_space *btnc, mark_buffer_dirty(obh); xa_lock_irq(&btnc->i_pages); - radix_tree_delete(&btnc->i_pages, oldkey); - radix_tree_tag_set(&btnc->i_pages, newkey, - PAGECACHE_TAG_DIRTY); + __xa_erase(&btnc->i_pages, oldkey); + __xa_set_mark(&btnc->i_pages, newkey, PAGECACHE_TAG_DIRTY); xa_unlock_irq(&btnc->i_pages); opage->index = obh->b_blocknr = newkey; @@ -275,7 +267,7 @@ void nilfs_btnode_abort_change_key(struct address_space *btnc, if (nbh == NULL) { /* blocksize == pagesize */ xa_lock_irq(&btnc->i_pages); - radix_tree_delete(&btnc->i_pages, newkey); + __xa_erase(&btnc->i_pages, newkey); xa_unlock_irq(&btnc->i_pages); unlock_page(ctxt->bh->b_page); } else diff --git a/fs/nilfs2/page.c b/fs/nilfs2/page.c @@ -289,7 +289,7 @@ repeat: * @dmap: destination page cache * @smap: source page cache * - * No pages must no be added to the cache during this process. + * No pages must be added to the cache during this process. * This must be ensured by the caller. */ void nilfs_copy_back_pages(struct address_space *dmap, @@ -298,7 +298,6 @@ void nilfs_copy_back_pages(struct address_space *dmap, struct pagevec pvec; unsigned int i, n; pgoff_t index = 0; - int err; pagevec_init(&pvec); repeat: @@ -313,35 +312,34 @@ repeat: lock_page(page); dpage = find_lock_page(dmap, offset); if (dpage) { - /* override existing page on the destination cache */ + /* overwrite existing page in the destination cache */ WARN_ON(PageDirty(dpage)); nilfs_copy_page(dpage, page, 0); unlock_page(dpage); put_page(dpage); + /* Do we not need to remove page from smap here? */ } else { - struct page *page2; + struct page *p; /* move the page to the destination cache */ xa_lock_irq(&smap->i_pages); - page2 = radix_tree_delete(&smap->i_pages, offset); - WARN_ON(page2 != page); - + p = __xa_erase(&smap->i_pages, offset); + WARN_ON(page != p); smap->nrpages--; xa_unlock_irq(&smap->i_pages); xa_lock_irq(&dmap->i_pages); - err = radix_tree_insert(&dmap->i_pages, offset, page); - if (unlikely(err < 0)) { - WARN_ON(err == -EEXIST); + p = __xa_store(&dmap->i_pages, offset, page, GFP_NOFS); + if (unlikely(p)) { + /* Probably -ENOMEM */ page->mapping = NULL; - put_page(page); /* for cache */ + put_page(page); } else { page->mapping = dmap; dmap->nrpages++; if (PageDirty(page)) - radix_tree_tag_set(&dmap->i_pages, - offset, - PAGECACHE_TAG_DIRTY); + __xa_set_mark(&dmap->i_pages, offset, + PAGECACHE_TAG_DIRTY); } xa_unlock_irq(&dmap->i_pages); } @@ -467,8 +465,7 @@ int __nilfs_clear_page_dirty(struct page *page) if (mapping) { xa_lock_irq(&mapping->i_pages); if (test_bit(PG_dirty, &page->flags)) { - radix_tree_tag_clear(&mapping->i_pages, - page_index(page), + __xa_clear_mark(&mapping->i_pages, page_index(page), PAGECACHE_TAG_DIRTY); xa_unlock_irq(&mapping->i_pages); return clear_page_dirty_for_io(page); diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c @@ -521,7 +521,7 @@ static void smaps_pte_entry(pte_t *pte, unsigned long addr, if (!page) return; - if (radix_tree_exceptional_entry(page)) + if (xa_is_value(page)) mss->swap += PAGE_SIZE; else put_page(page); diff --git a/include/linux/fs.h b/include/linux/fs.h @@ -403,24 +403,40 @@ int pagecache_write_end(struct file *, struct address_space *mapping, loff_t pos, unsigned len, unsigned copied, struct page *page, void *fsdata); +/** + * struct address_space - Contents of a cacheable, mappable object. + * @host: Owner, either the inode or the block_device. + * @i_pages: Cached pages. + * @gfp_mask: Memory allocation flags to use for allocating pages. + * @i_mmap_writable: Number of VM_SHARED mappings. + * @i_mmap: Tree of private and shared mappings. + * @i_mmap_rwsem: Protects @i_mmap and @i_mmap_writable. + * @nrpages: Number of page entries, protected by the i_pages lock. + * @nrexceptional: Shadow or DAX entries, protected by the i_pages lock. + * @writeback_index: Writeback starts here. + * @a_ops: Methods. + * @flags: Error bits and flags (AS_*). + * @wb_err: The most recent error which has occurred. + * @private_lock: For use by the owner of the address_space. + * @private_list: For use by the owner of the address_space. + * @private_data: For use by the owner of the address_space. + */ struct address_space { - struct inode *host; /* owner: inode, block_device */ - struct radix_tree_root i_pages; /* cached pages */ - atomic_t i_mmap_writable;/* count VM_SHARED mappings */ - struct rb_root_cached i_mmap; /* tree of private and shared mappings */ - struct rw_semaphore i_mmap_rwsem; /* protect tree, count, list */ - /* Protected by the i_pages lock */ - unsigned long nrpages; /* number of total pages */ - /* number of shadow or DAX exceptional entries */ + struct inode *host; + struct xarray i_pages; + gfp_t gfp_mask; + atomic_t i_mmap_writable; + struct rb_root_cached i_mmap; + struct rw_semaphore i_mmap_rwsem; + unsigned long nrpages; unsigned long nrexceptional; - pgoff_t writeback_index;/* writeback starts here */ - const struct address_space_operations *a_ops; /* methods */ - unsigned long flags; /* error bits */ - spinlock_t private_lock; /* for use by the address_space */ - gfp_t gfp_mask; /* implicit gfp mask for allocations */ - struct list_head private_list; /* for use by the address_space */ - void *private_data; /* ditto */ + pgoff_t writeback_index; + const struct address_space_operations *a_ops; + unsigned long flags; errseq_t wb_err; + spinlock_t private_lock; + struct list_head private_list; + void *private_data; } __attribute__((aligned(sizeof(long)))) __randomize_layout; /* * On most architectures that alignment is already the case; but @@ -467,15 +483,18 @@ struct block_device { struct mutex bd_fsfreeze_mutex; } __randomize_layout; +/* XArray tags, for tagging dirty and writeback pages in the pagecache. */ +#define PAGECACHE_TAG_DIRTY XA_MARK_0 +#define PAGECACHE_TAG_WRITEBACK XA_MARK_1 +#define PAGECACHE_TAG_TOWRITE XA_MARK_2 + /* - * Radix-tree tags, for tagging dirty and writeback pages within the pagecache - * radix trees + * Returns true if any of the pages in the mapping are marked with the tag. */ -#define PAGECACHE_TAG_DIRTY 0 -#define PAGECACHE_TAG_WRITEBACK 1 -#define PAGECACHE_TAG_TOWRITE 2 - -int mapping_tagged(struct address_space *mapping, int tag); +static inline bool mapping_tagged(struct address_space *mapping, xa_mark_t tag) +{ + return xa_marked(&mapping->i_pages, tag); +} static inline void i_mmap_lock_write(struct address_space *mapping) { diff --git a/include/linux/idr.h b/include/linux/idr.h @@ -214,8 +214,7 @@ static inline void idr_preload_end(void) ++id, (entry) = idr_get_next((idr), &(id))) /* - * IDA - IDR based id allocator, use when translation from id to - * pointer isn't necessary. + * IDA - ID Allocator, use when translation from id to pointer isn't necessary. */ #define IDA_CHUNK_SIZE 128 /* 128 bytes per chunk */ #define IDA_BITMAP_LONGS (IDA_CHUNK_SIZE / sizeof(long)) @@ -225,14 +224,14 @@ struct ida_bitmap { unsigned long bitmap[IDA_BITMAP_LONGS]; }; -DECLARE_PER_CPU(struct ida_bitmap *, ida_bitmap); - struct ida { - struct radix_tree_root ida_rt; + struct xarray xa; }; +#define IDA_INIT_FLAGS (XA_FLAGS_LOCK_IRQ | XA_FLAGS_ALLOC) + #define IDA_INIT(name) { \ - .ida_rt = RADIX_TREE_INIT(name, IDR_RT_MARKER | GFP_NOWAIT), \ + .xa = XARRAY_INIT(name, IDA_INIT_FLAGS) \ } #define DEFINE_IDA(name) struct ida name = IDA_INIT(name) @@ -292,7 +291,7 @@ static inline int ida_alloc_max(struct ida *ida, unsigned int max, gfp_t gfp) static inline void ida_init(struct ida *ida) { - INIT_RADIX_TREE(&ida->ida_rt, IDR_RT_MARKER | GFP_NOWAIT); + xa_init_flags(&ida->xa, IDA_INIT_FLAGS); } #define ida_simple_get(ida, start, end, gfp) \ @@ -301,9 +300,6 @@ static inline void ida_init(struct ida *ida) static inline bool ida_is_empty(const struct ida *ida) { - return radix_tree_empty(&ida->ida_rt); + return xa_empty(&ida->xa); } - -/* in lib/radix-tree.c */ -int ida_pre_get(struct ida *ida, gfp_t gfp_mask); #endif /* __IDR_H__ */ diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h @@ -241,9 +241,9 @@ static inline gfp_t readahead_gfp_mask(struct address_space *x) typedef int filler_t(void *, struct page *); -pgoff_t page_cache_next_hole(struct address_space *mapping, +pgoff_t page_cache_next_miss(struct address_space *mapping, pgoff_t index, unsigned long max_scan); -pgoff_t page_cache_prev_hole(struct address_space *mapping, +pgoff_t page_cache_prev_miss(struct address_space *mapping, pgoff_t index, unsigned long max_scan); #define FGP_ACCESSED 0x00000001 @@ -363,17 +363,17 @@ static inline unsigned find_get_pages(struct address_space *mapping, unsigned find_get_pages_contig(struct address_space *mapping, pgoff_t start, unsigned int nr_pages, struct page **pages); unsigned find_get_pages_range_tag(struct address_space *mapping, pgoff_t *index, - pgoff_t end, int tag, unsigned int nr_pages, + pgoff_t end, xa_mark_t tag, unsigned int nr_pages, struct page **pages); static inline unsigned find_get_pages_tag(struct address_space *mapping, - pgoff_t *index, int tag, unsigned int nr_pages, + pgoff_t *index, xa_mark_t tag, unsigned int nr_pages, struct page **pages) { return find_get_pages_range_tag(mapping, index, (pgoff_t)-1, tag, nr_pages, pages); } unsigned find_get_entries_tag(struct address_space *mapping, pgoff_t start, - int tag, unsigned int nr_entries, + xa_mark_t tag, unsigned int nr_entries, struct page **entries, pgoff_t *indices); struct page *grab_cache_page_write_begin(struct address_space *mapping, diff --git a/include/linux/pagevec.h b/include/linux/pagevec.h @@ -9,6 +9,8 @@ #ifndef _LINUX_PAGEVEC_H #define _LINUX_PAGEVEC_H +#include <linux/xarray.h> + /* 15 pointers + header align the pagevec structure to a power of two */ #define PAGEVEC_SIZE 15 @@ -40,12 +42,12 @@ static inline unsigned pagevec_lookup(struct pagevec *pvec, unsigned pagevec_lookup_range_tag(struct pagevec *pvec, struct address_space *mapping, pgoff_t *index, pgoff_t end, - int tag); + xa_mark_t tag); unsigned pagevec_lookup_range_nr_tag(struct pagevec *pvec, struct address_space *mapping, pgoff_t *index, pgoff_t end, - int tag, unsigned max_pages); + xa_mark_t tag, unsigned max_pages); static inline unsigned pagevec_lookup_tag(struct pagevec *pvec, - struct address_space *mapping, pgoff_t *index, int tag) + struct address_space *mapping, pgoff_t *index, xa_mark_t tag) { return pagevec_lookup_range_tag(pvec, mapping, index, (pgoff_t)-1, tag); } diff --git a/include/linux/radix-tree.h b/include/linux/radix-tree.h @@ -28,34 +28,30 @@ #include <linux/rcupdate.h> #include <linux/spinlock.h> #include <linux/types.h> +#include <linux/xarray.h> + +/* Keep unconverted code working */ +#define radix_tree_root xarray +#define radix_tree_node xa_node /* * The bottom two bits of the slot determine how the remaining bits in the * slot are interpreted: * * 00 - data pointer - * 01 - internal entry - * 10 - exceptional entry - * 11 - this bit combination is currently unused/reserved + * 10 - internal entry + * x1 - value entry * * The internal entry may be a pointer to the next level in the tree, a * sibling entry, or an indicator that the entry in this slot has been moved * to another location in the tree and the lookup should be restarted. While * NULL fits the 'data pointer' pattern, it means that there is no entry in * the tree for this index (no matter what level of the tree it is found at). - * This means that you cannot store NULL in the tree as a value for the index. + * This means that storing a NULL entry in the tree is the same as deleting + * the entry from the tree. */ #define RADIX_TREE_ENTRY_MASK 3UL -#define RADIX_TREE_INTERNAL_NODE 1UL - -/* - * Most users of the radix tree store pointers but shmem/tmpfs stores swap - * entries in the same tree. They are marked as exceptional entries to - * distinguish them from pointers to struct page. - * EXCEPTIONAL_ENTRY tests the bit, EXCEPTIONAL_SHIFT shifts content past it. - */ -#define RADIX_TREE_EXCEPTIONAL_ENTRY 2 -#define RADIX_TREE_EXCEPTIONAL_SHIFT 2 +#define RADIX_TREE_INTERNAL_NODE 2UL static inline bool radix_tree_is_internal_node(void *ptr) { @@ -65,75 +61,32 @@ static inline bool radix_tree_is_internal_node(void *ptr) /*** radix-tree API starts here ***/ -#define RADIX_TREE_MAX_TAGS 3 - -#ifndef RADIX_TREE_MAP_SHIFT -#define RADIX_TREE_MAP_SHIFT (CONFIG_BASE_SMALL ? 4 : 6) -#endif - +#define RADIX_TREE_MAP_SHIFT XA_CHUNK_SHIFT #define RADIX_TREE_MAP_SIZE (1UL << RADIX_TREE_MAP_SHIFT) #define RADIX_TREE_MAP_MASK (RADIX_TREE_MAP_SIZE-1) -#define RADIX_TREE_TAG_LONGS \ - ((RADIX_TREE_MAP_SIZE + BITS_PER_LONG - 1) / BITS_PER_LONG) +#define RADIX_TREE_MAX_TAGS XA_MAX_MARKS +#define RADIX_TREE_TAG_LONGS XA_MARK_LONGS #define RADIX_TREE_INDEX_BITS (8 /* CHAR_BIT */ * sizeof(unsigned long)) #define RADIX_TREE_MAX_PATH (DIV_ROUND_UP(RADIX_TREE_INDEX_BITS, \ RADIX_TREE_MAP_SHIFT)) -/* - * @count is the count of every non-NULL element in the ->slots array - * whether that is an exceptional entry, a retry entry, a user pointer, - * a sibling entry or a pointer to the next level of the tree. - * @exceptional is the count of every element in ->slots which is - * either radix_tree_exceptional_entry() or is a sibling entry for an - * exceptional entry. - */ -struct radix_tree_node { - unsigned char shift; /* Bits remaining in each slot */ - unsigned char offset; /* Slot offset in parent */ - unsigned char count; /* Total entry count */ - unsigned char exceptional; /* Exceptional entry count */ - struct radix_tree_node *parent; /* Used when ascending tree */ - struct radix_tree_root *root; /* The tree we belong to */ - union { - struct list_head private_list; /* For tree user */ - struct rcu_head rcu_head; /* Used when freeing node */ - }; - void __rcu *slots[RADIX_TREE_MAP_SIZE]; - unsigned long tags[RADIX_TREE_MAX_TAGS][RADIX_TREE_TAG_LONGS]; -}; - -/* The IDR tag is stored in the low bits of the GFP flags */ +/* The IDR tag is stored in the low bits of xa_flags */ #define ROOT_IS_IDR ((__force gfp_t)4) -/* The top bits of gfp_mask are used to store the root tags */ +/* The top bits of xa_flags are used to store the root tags */ #define ROOT_TAG_SHIFT (__GFP_BITS_SHIFT) -struct radix_tree_root { - spinlock_t xa_lock; - gfp_t gfp_mask; - struct radix_tree_node __rcu *rnode; -}; - -#define RADIX_TREE_INIT(name, mask) { \ - .xa_lock = __SPIN_LOCK_UNLOCKED(name.xa_lock), \ - .gfp_mask = (mask), \ - .rnode = NULL, \ -} +#define RADIX_TREE_INIT(name, mask) XARRAY_INIT(name, mask) #define RADIX_TREE(name, mask) \ struct radix_tree_root name = RADIX_TREE_INIT(name, mask) -#define INIT_RADIX_TREE(root, mask) \ -do { \ - spin_lock_init(&(root)->xa_lock); \ - (root)->gfp_mask = (mask); \ - (root)->rnode = NULL; \ -} while (0) +#define INIT_RADIX_TREE(root, mask) xa_init_flags(root, mask) static inline bool radix_tree_empty(const struct radix_tree_root *root) { - return root->rnode == NULL; + return root->xa_head == NULL; } /** @@ -143,7 +96,6 @@ static inline bool radix_tree_empty(const struct radix_tree_root *root) * @next_index: one beyond the last index for this chunk * @tags: bit-mask for tag-iterating * @node: node that contains current slot - * @shift: shift for the node that holds our slots * * This radix tree iterator works in terms of "chunks" of slots. A chunk is a * subinterval of slots contained within one radix tree leaf node. It is @@ -157,20 +109,8 @@ struct radix_tree_iter { unsigned long next_index; unsigned long tags; struct radix_tree_node *node; -#ifdef CONFIG_RADIX_TREE_MULTIORDER - unsigned int shift; -#endif }; -static inline unsigned int iter_shift(const struct radix_tree_iter *iter) -{ -#ifdef CONFIG_RADIX_TREE_MULTIORDER - return iter->shift; -#else - return 0; -#endif -} - /** * Radix-tree synchronization * @@ -194,12 +134,11 @@ static inline unsigned int iter_shift(const struct radix_tree_iter *iter) * radix_tree_lookup_slot * radix_tree_tag_get * radix_tree_gang_lookup - * radix_tree_gang_lookup_slot * radix_tree_gang_lookup_tag * radix_tree_gang_lookup_tag_slot * radix_tree_tagged * - * The first 8 functions are able to be called locklessly, using RCU. The + * The first 7 functions are able to be called locklessly, using RCU. The * caller must ensure calls to these functions are made within rcu_read_lock() * regions. Other readers (lock-free or otherwise) and modifications may be * running concurrently. @@ -269,17 +208,6 @@ static inline int radix_tree_deref_retry(void *arg) } /** - * radix_tree_exceptional_entry - radix_tree_deref_slot gave exceptional entry? - * @arg: value returned by radix_tree_deref_slot - * Returns: 0 if well-aligned pointer, non-0 if exceptional entry. - */ -static inline int radix_tree_exceptional_entry(void *arg) -{ - /* Not unlikely because radix_tree_exception often tested first */ - return (unsigned long)arg & RADIX_TREE_EXCEPTIONAL_ENTRY; -} - -/** * radix_tree_exception - radix_tree_deref_slot returned either exception? * @arg: value returned by radix_tree_deref_slot * Returns: 0 if well-aligned pointer, non-0 if either kind of exception. @@ -289,47 +217,28 @@ static inline int radix_tree_exception(void *arg) return unlikely((unsigned long)arg & RADIX_TREE_ENTRY_MASK); } -int __radix_tree_create(struct radix_tree_root *, unsigned long index, - unsigned order, struct radix_tree_node **nodep, - void __rcu ***slotp); -int __radix_tree_insert(struct radix_tree_root *, unsigned long index, - unsigned order, void *); -static inline int radix_tree_insert(struct radix_tree_root *root, - unsigned long index, void *entry) -{ - return __radix_tree_insert(root, index, 0, entry); -} +int radix_tree_insert(struct radix_tree_root *, unsigned long index, + void *); void *__radix_tree_lookup(const struct radix_tree_root *, unsigned long index, struct radix_tree_node **nodep, void __rcu ***slotp); void *radix_tree_lookup(const struct radix_tree_root *, unsigned long); void __rcu **radix_tree_lookup_slot(const struct radix_tree_root *, unsigned long index); -typedef void (*radix_tree_update_node_t)(struct radix_tree_node *); void __radix_tree_replace(struct radix_tree_root *, struct radix_tree_node *, - void __rcu **slot, void *entry, - radix_tree_update_node_t update_node); + void __rcu **slot, void *entry); void radix_tree_iter_replace(struct radix_tree_root *, const struct radix_tree_iter *, void __rcu **slot, void *entry); void radix_tree_replace_slot(struct radix_tree_root *, void __rcu **slot, void *entry); -void __radix_tree_delete_node(struct radix_tree_root *, - struct radix_tree_node *, - radix_tree_update_node_t update_node); void radix_tree_iter_delete(struct radix_tree_root *, struct radix_tree_iter *iter, void __rcu **slot); void *radix_tree_delete_item(struct radix_tree_root *, unsigned long, void *); void *radix_tree_delete(struct radix_tree_root *, unsigned long); -void radix_tree_clear_tags(struct radix_tree_root *, struct radix_tree_node *, - void __rcu **slot); unsigned int radix_tree_gang_lookup(const struct radix_tree_root *, void **results, unsigned long first_index, unsigned int max_items); -unsigned int radix_tree_gang_lookup_slot(const struct radix_tree_root *, - void __rcu ***results, unsigned long *indices, - unsigned long first_index, unsigned int max_items); int radix_tree_preload(gfp_t gfp_mask); int radix_tree_maybe_preload(gfp_t gfp_mask); -int radix_tree_maybe_preload_order(gfp_t gfp_mask, int order); void radix_tree_init(void); void *radix_tree_tag_set(struct radix_tree_root *, unsigned long index, unsigned int tag); @@ -337,8 +246,6 @@ void *radix_tree_tag_clear(struct radix_tree_root *, unsigned long index, unsigned int tag); int radix_tree_tag_get(const struct radix_tree_root *, unsigned long index, unsigned int tag); -void radix_tree_iter_tag_set(struct radix_tree_root *, - const struct radix_tree_iter *iter, unsigned int tag); void radix_tree_iter_tag_clear(struct radix_tree_root *, const struct radix_tree_iter *iter, unsigned int tag); unsigned int radix_tree_gang_lookup_tag(const struct radix_tree_root *, @@ -354,12 +261,6 @@ static inline void radix_tree_preload_end(void) preempt_enable(); } -int radix_tree_split_preload(unsigned old_order, unsigned new_order, gfp_t); -int radix_tree_split(struct radix_tree_root *, unsigned long index, - unsigned new_order); -int radix_tree_join(struct radix_tree_root *, unsigned long index, - unsigned new_order, void *); - void __rcu **idr_get_free(struct radix_tree_root *root, struct radix_tree_iter *iter, gfp_t gfp, unsigned long max); @@ -465,7 +366,7 @@ void __rcu **radix_tree_iter_retry(struct radix_tree_iter *iter) static inline unsigned long __radix_tree_iter_add(struct radix_tree_iter *iter, unsigned long slots) { - return iter->index + (slots << iter_shift(iter)); + return iter->index + slots; } /** @@ -490,21 +391,9 @@ void __rcu **__must_check radix_tree_iter_resume(void __rcu **slot, static __always_inline long radix_tree_chunk_size(struct radix_tree_iter *iter) { - return (iter->next_index - iter->index) >> iter_shift(iter); + return iter->next_index - iter->index; } -#ifdef CONFIG_RADIX_TREE_MULTIORDER -void __rcu **__radix_tree_next_slot(void __rcu **slot, - struct radix_tree_iter *iter, unsigned flags); -#else -/* Can't happen without sibling entries, but the compiler can't tell that */ -static inline void __rcu **__radix_tree_next_slot(void __rcu **slot, - struct radix_tree_iter *iter, unsigned flags) -{ - return slot; -} -#endif - /** * radix_tree_next_slot - find next slot in chunk * @@ -563,8 +452,6 @@ static __always_inline void __rcu **radix_tree_next_slot(void __rcu **slot, return NULL; found: - if (unlikely(radix_tree_is_internal_node(rcu_dereference_raw(*slot)))) - return __radix_tree_next_slot(slot, iter, flags); return slot; } @@ -584,23 +471,6 @@ static __always_inline void __rcu **radix_tree_next_slot(void __rcu **slot, slot = radix_tree_next_slot(slot, iter, 0)) /** - * radix_tree_for_each_contig - iterate over contiguous slots - * - * @slot: the void** variable for pointer to slot - * @root: the struct radix_tree_root pointer - * @iter: the struct radix_tree_iter pointer - * @start: iteration starting index - * - * @slot points to radix tree slot, @iter->index contains its index. - */ -#define radix_tree_for_each_contig(slot, root, iter, start) \ - for (slot = radix_tree_iter_init(iter, start) ; \ - slot || (slot = radix_tree_next_chunk(root, iter, \ - RADIX_TREE_ITER_CONTIG)) ; \ - slot = radix_tree_next_slot(slot, iter, \ - RADIX_TREE_ITER_CONTIG)) - -/** * radix_tree_for_each_tagged - iterate over tagged slots * * @slot: the void** variable for pointer to slot diff --git a/include/linux/swap.h b/include/linux/swap.h @@ -300,17 +300,12 @@ void *workingset_eviction(struct address_space *mapping, struct page *page); void workingset_refault(struct page *page, void *shadow); void workingset_activation(struct page *page); -/* Do not use directly, use workingset_lookup_update */ -void workingset_update_node(struct radix_tree_node *node); - -/* Returns workingset_update_node() if the mapping has shadow entries. */ -#define workingset_lookup_update(mapping) \ -({ \ - radix_tree_update_node_t __helper = workingset_update_node; \ - if (dax_mapping(mapping) || shmem_mapping(mapping)) \ - __helper = NULL; \ - __helper; \ -}) +/* Only track the nodes of mappings with shadow entries */ +void workingset_update_node(struct xa_node *node); +#define mapping_set_update(xas, mapping) do { \ + if (!dax_mapping(mapping) && !shmem_mapping(mapping)) \ + xas_set_update(xas, workingset_update_node); \ +} while (0) /* linux/mm/page_alloc.c */ extern unsigned long totalram_pages; @@ -409,7 +404,7 @@ extern void show_swap_cache_info(void); extern int add_to_swap(struct page *page); extern int add_to_swap_cache(struct page *, swp_entry_t, gfp_t); extern int __add_to_swap_cache(struct page *page, swp_entry_t entry); -extern void __delete_from_swap_cache(struct page *); +extern void __delete_from_swap_cache(struct page *, swp_entry_t entry); extern void delete_from_swap_cache(struct page *); extern void free_page_and_swap_cache(struct page *); extern void free_pages_and_swap_cache(struct page **, int); @@ -563,7 +558,8 @@ static inline int add_to_swap_cache(struct page *page, swp_entry_t entry, return -1; } -static inline void __delete_from_swap_cache(struct page *page) +static inline void __delete_from_swap_cache(struct page *page, + swp_entry_t entry) { } diff --git a/include/linux/swapops.h b/include/linux/swapops.h @@ -18,9 +18,8 @@ * * swp_entry_t's are *never* stored anywhere in their arch-dependent format. */ -#define SWP_TYPE_SHIFT(e) ((sizeof(e.val) * 8) - \ - (MAX_SWAPFILES_SHIFT + RADIX_TREE_EXCEPTIONAL_SHIFT)) -#define SWP_OFFSET_MASK(e) ((1UL << SWP_TYPE_SHIFT(e)) - 1) +#define SWP_TYPE_SHIFT (BITS_PER_XA_VALUE - MAX_SWAPFILES_SHIFT) +#define SWP_OFFSET_MASK ((1UL << SWP_TYPE_SHIFT) - 1) /* * Store a type+offset into a swp_entry_t in an arch-independent format @@ -29,8 +28,7 @@ static inline swp_entry_t swp_entry(unsigned long type, pgoff_t offset) { swp_entry_t ret; - ret.val = (type << SWP_TYPE_SHIFT(ret)) | - (offset & SWP_OFFSET_MASK(ret)); + ret.val = (type << SWP_TYPE_SHIFT) | (offset & SWP_OFFSET_MASK); return ret; } @@ -40,7 +38,7 @@ static inline swp_entry_t swp_entry(unsigned long type, pgoff_t offset) */ static inline unsigned swp_type(swp_entry_t entry) { - return (entry.val >> SWP_TYPE_SHIFT(entry)); + return (entry.val >> SWP_TYPE_SHIFT); } /* @@ -49,7 +47,7 @@ static inline unsigned swp_type(swp_entry_t entry) */ static inline pgoff_t swp_offset(swp_entry_t entry) { - return entry.val & SWP_OFFSET_MASK(entry); + return entry.val & SWP_OFFSET_MASK; } #ifdef CONFIG_MMU @@ -90,16 +88,13 @@ static inline swp_entry_t radix_to_swp_entry(void *arg) { swp_entry_t entry; - entry.val = (unsigned long)arg >> RADIX_TREE_EXCEPTIONAL_SHIFT; + entry.val = xa_to_value(arg); return entry; } static inline void *swp_to_radix_entry(swp_entry_t entry) { - unsigned long value; - - value = entry.val << RADIX_TREE_EXCEPTIONAL_SHIFT; - return (void *)(value | RADIX_TREE_EXCEPTIONAL_ENTRY); + return xa_mk_value(entry.val); } #if IS_ENABLED(CONFIG_DEVICE_PRIVATE) diff --git a/include/linux/xarray.h b/include/linux/xarray.h @@ -4,10 +4,432 @@ /* * eXtensible Arrays * Copyright (c) 2017 Microsoft Corporation - * Author: Matthew Wilcox <mawilcox@microsoft.com> + * Author: Matthew Wilcox <willy@infradead.org> + * + * See Documentation/core-api/xarray.rst for how to use the XArray. */ +#include <linux/bug.h> +#include <linux/compiler.h> +#include <linux/gfp.h> +#include <linux/kconfig.h> +#include <linux/kernel.h> +#include <linux/rcupdate.h> #include <linux/spinlock.h> +#include <linux/types.h> + +/* + * The bottom two bits of the entry determine how the XArray interprets + * the contents: + * + * 00: Pointer entry + * 10: Internal entry + * x1: Value entry or tagged pointer + * + * Attempting to store internal entries in the XArray is a bug. + * + * Most internal entries are pointers to the next node in the tree. + * The following internal entries have a special meaning: + * + * 0-62: Sibling entries + * 256: Zero entry + * 257: Retry entry + * + * Errors are also represented as internal entries, but use the negative + * space (-4094 to -2). They're never stored in the slots array; only + * returned by the normal API. + */ + +#define BITS_PER_XA_VALUE (BITS_PER_LONG - 1) + +/** + * xa_mk_value() - Create an XArray entry from an integer. + * @v: Value to store in XArray. + * + * Context: Any context. + * Return: An entry suitable for storing in the XArray. + */ +static inline void *xa_mk_value(unsigned long v) +{ + WARN_ON((long)v < 0); + return (void *)((v << 1) | 1); +} + +/** + * xa_to_value() - Get value stored in an XArray entry. + * @entry: XArray entry. + * + * Context: Any context. + * Return: The value stored in the XArray entry. + */ +static inline unsigned long xa_to_value(const void *entry) +{ + return (unsigned long)entry >> 1; +} + +/** + * xa_is_value() - Determine if an entry is a value. + * @entry: XArray entry. + * + * Context: Any context. + * Return: True if the entry is a value, false if it is a pointer. + */ +static inline bool xa_is_value(const void *entry) +{ + return (unsigned long)entry & 1; +} + +/** + * xa_tag_pointer() - Create an XArray entry for a tagged pointer. + * @p: Plain pointer. + * @tag: Tag value (0, 1 or 3). + * + * If the user of the XArray prefers, they can tag their pointers instead + * of storing value entries. Three tags are available (0, 1 and 3). + * These are distinct from the xa_mark_t as they are not replicated up + * through the array and cannot be searched for. + * + * Context: Any context. + * Return: An XArray entry. + */ +static inline void *xa_tag_pointer(void *p, unsigned long tag) +{ + return (void *)((unsigned long)p | tag); +} + +/** + * xa_untag_pointer() - Turn an XArray entry into a plain pointer. + * @entry: XArray entry. + * + * If you have stored a tagged pointer in the XArray, call this function + * to get the untagged version of the pointer. + * + * Context: Any context. + * Return: A pointer. + */ +static inline void *xa_untag_pointer(void *entry) +{ + return (void *)((unsigned long)entry & ~3UL); +} + +/** + * xa_pointer_tag() - Get the tag stored in an XArray entry. + * @entry: XArray entry. + * + * If you have stored a tagged pointer in the XArray, call this function + * to get the tag of that pointer. + * + * Context: Any context. + * Return: A tag. + */ +static inline unsigned int xa_pointer_tag(void *entry) +{ + return (unsigned long)entry & 3UL; +} + +/* + * xa_mk_internal() - Create an internal entry. + * @v: Value to turn into an internal entry. + * + * Context: Any context. + * Return: An XArray internal entry corresponding to this value. + */ +static inline void *xa_mk_internal(unsigned long v) +{ + return (void *)((v << 2) | 2); +} + +/* + * xa_to_internal() - Extract the value from an internal entry. + * @entry: XArray entry. + * + * Context: Any context. + * Return: The value which was stored in the internal entry. + */ +static inline unsigned long xa_to_internal(const void *entry) +{ + return (unsigned long)entry >> 2; +} + +/* + * xa_is_internal() - Is the entry an internal entry? + * @entry: XArray entry. + * + * Context: Any context. + * Return: %true if the entry is an internal entry. + */ +static inline bool xa_is_internal(const void *entry) +{ + return ((unsigned long)entry & 3) == 2; +} + +/** + * xa_is_err() - Report whether an XArray operation returned an error + * @entry: Result from calling an XArray function + * + * If an XArray operation cannot complete an operation, it will return + * a special value indicating an error. This function tells you + * whether an error occurred; xa_err() tells you which error occurred. + * + * Context: Any context. + * Return: %true if the entry indicates an error. + */ +static inline bool xa_is_err(const void *entry) +{ + return unlikely(xa_is_internal(entry)); +} + +/** + * xa_err() - Turn an XArray result into an errno. + * @entry: Result from calling an XArray function. + * + * If an XArray operation cannot complete an operation, it will return + * a special pointer value which encodes an errno. This function extracts + * the errno from the pointer value, or returns 0 if the pointer does not + * represent an errno. + * + * Context: Any context. + * Return: A negative errno or 0. + */ +static inline int xa_err(void *entry) +{ + /* xa_to_internal() would not do sign extension. */ + if (xa_is_err(entry)) + return (long)entry >> 2; + return 0; +} + +typedef unsigned __bitwise xa_mark_t; +#define XA_MARK_0 ((__force xa_mark_t)0U) +#define XA_MARK_1 ((__force xa_mark_t)1U) +#define XA_MARK_2 ((__force xa_mark_t)2U) +#define XA_PRESENT ((__force xa_mark_t)8U) +#define XA_MARK_MAX XA_MARK_2 +#define XA_FREE_MARK XA_MARK_0 + +enum xa_lock_type { + XA_LOCK_IRQ = 1, + XA_LOCK_BH = 2, +}; + +/* + * Values for xa_flags. The radix tree stores its GFP flags in the xa_flags, + * and we remain compatible with that. + */ +#define XA_FLAGS_LOCK_IRQ ((__force gfp_t)XA_LOCK_IRQ) +#define XA_FLAGS_LOCK_BH ((__force gfp_t)XA_LOCK_BH) +#define XA_FLAGS_TRACK_FREE ((__force gfp_t)4U) +#define XA_FLAGS_MARK(mark) ((__force gfp_t)((1U << __GFP_BITS_SHIFT) << \ + (__force unsigned)(mark))) + +#define XA_FLAGS_ALLOC (XA_FLAGS_TRACK_FREE | XA_FLAGS_MARK(XA_FREE_MARK)) + +/** + * struct xarray - The anchor of the XArray. + * @xa_lock: Lock that protects the contents of the XArray. + * + * To use the xarray, define it statically or embed it in your data structure. + * It is a very small data structure, so it does not usually make sense to + * allocate it separately and keep a pointer to it in your data structure. + * + * You may use the xa_lock to protect your own data structures as well. + */ +/* + * If all of the entries in the array are NULL, @xa_head is a NULL pointer. + * If the only non-NULL entry in the array is at index 0, @xa_head is that + * entry. If any other entry in the array is non-NULL, @xa_head points + * to an @xa_node. + */ +struct xarray { + spinlock_t xa_lock; +/* private: The rest of the data structure is not to be used directly. */ + gfp_t xa_flags; + void __rcu * xa_head; +}; + +#define XARRAY_INIT(name, flags) { \ + .xa_lock = __SPIN_LOCK_UNLOCKED(name.xa_lock), \ + .xa_flags = flags, \ + .xa_head = NULL, \ +} + +/** + * DEFINE_XARRAY_FLAGS() - Define an XArray with custom flags. + * @name: A string that names your XArray. + * @flags: XA_FLAG values. + * + * This is intended for file scope definitions of XArrays. It declares + * and initialises an empty XArray with the chosen name and flags. It is + * equivalent to calling xa_init_flags() on the array, but it does the + * initialisation at compiletime instead of runtime. + */ +#define DEFINE_XARRAY_FLAGS(name, flags) \ + struct xarray name = XARRAY_INIT(name, flags) + +/** + * DEFINE_XARRAY() - Define an XArray. + * @name: A string that names your XArray. + * + * This is intended for file scope definitions of XArrays. It declares + * and initialises an empty XArray with the chosen name. It is equivalent + * to calling xa_init() on the array, but it does the initialisation at + * compiletime instead of runtime. + */ +#define DEFINE_XARRAY(name) DEFINE_XARRAY_FLAGS(name, 0) + +/** + * DEFINE_XARRAY_ALLOC() - Define an XArray which can allocate IDs. + * @name: A string that names your XArray. + * + * This is intended for file scope definitions of allocating XArrays. + * See also DEFINE_XARRAY(). + */ +#define DEFINE_XARRAY_ALLOC(name) DEFINE_XARRAY_FLAGS(name, XA_FLAGS_ALLOC) + +void xa_init_flags(struct xarray *, gfp_t flags); +void *xa_load(struct xarray *, unsigned long index); +void *xa_store(struct xarray *, unsigned long index, void *entry, gfp_t); +void *xa_cmpxchg(struct xarray *, unsigned long index, + void *old, void *entry, gfp_t); +int xa_reserve(struct xarray *, unsigned long index, gfp_t); +void *xa_store_range(struct xarray *, unsigned long first, unsigned long last, + void *entry, gfp_t); +bool xa_get_mark(struct xarray *, unsigned long index, xa_mark_t); +void xa_set_mark(struct xarray *, unsigned long index, xa_mark_t); +void xa_clear_mark(struct xarray *, unsigned long index, xa_mark_t); +void *xa_find(struct xarray *xa, unsigned long *index, + unsigned long max, xa_mark_t) __attribute__((nonnull(2))); +void *xa_find_after(struct xarray *xa, unsigned long *index, + unsigned long max, xa_mark_t) __attribute__((nonnull(2))); +unsigned int xa_extract(struct xarray *, void **dst, unsigned long start, + unsigned long max, unsigned int n, xa_mark_t); +void xa_destroy(struct xarray *); + +/** + * xa_init() - Initialise an empty XArray. + * @xa: XArray. + * + * An empty XArray is full of NULL entries. + * + * Context: Any context. + */ +static inline void xa_init(struct xarray *xa) +{ + xa_init_flags(xa, 0); +} + +/** + * xa_empty() - Determine if an array has any present entries. + * @xa: XArray. + * + * Context: Any context. + * Return: %true if the array contains only NULL pointers. + */ +static inline bool xa_empty(const struct xarray *xa) +{ + return xa->xa_head == NULL; +} + +/** + * xa_marked() - Inquire whether any entry in this array has a mark set + * @xa: Array + * @mark: Mark value + * + * Context: Any context. + * Return: %true if any entry has this mark set. + */ +static inline bool xa_marked(const struct xarray *xa, xa_mark_t mark) +{ + return xa->xa_flags & XA_FLAGS_MARK(mark); +} + +/** + * xa_erase() - Erase this entry from the XArray. + * @xa: XArray. + * @index: Index of entry. + * + * This function is the equivalent of calling xa_store() with %NULL as + * the third argument. The XArray does not need to allocate memory, so + * the user does not need to provide GFP flags. + * + * Context: Process context. Takes and releases the xa_lock. + * Return: The entry which used to be at this index. + */ +static inline void *xa_erase(struct xarray *xa, unsigned long index) +{ + return xa_store(xa, index, NULL, 0); +} + +/** + * xa_insert() - Store this entry in the XArray unless another entry is + * already present. + * @xa: XArray. + * @index: Index into array. + * @entry: New entry. + * @gfp: Memory allocation flags. + * + * If you would rather see the existing entry in the array, use xa_cmpxchg(). + * This function is for users who don't care what the entry is, only that + * one is present. + * + * Context: Process context. Takes and releases the xa_lock. + * May sleep if the @gfp flags permit. + * Return: 0 if the store succeeded. -EEXIST if another entry was present. + * -ENOMEM if memory could not be allocated. + */ +static inline int xa_insert(struct xarray *xa, unsigned long index, + void *entry, gfp_t gfp) +{ + void *curr = xa_cmpxchg(xa, index, NULL, entry, gfp); + if (!curr) + return 0; + if (xa_is_err(curr)) + return xa_err(curr); + return -EEXIST; +} + +/** + * xa_release() - Release a reserved entry. + * @xa: XArray. + * @index: Index of entry. + * + * After calling xa_reserve(), you can call this function to release the + * reservation. If the entry at @index has been stored to, this function + * will do nothing. + */ +static inline void xa_release(struct xarray *xa, unsigned long index) +{ + xa_cmpxchg(xa, index, NULL, NULL, 0); +} + +/** + * xa_for_each() - Iterate over a portion of an XArray. + * @xa: XArray. + * @entry: Entry retrieved from array. + * @index: Index of @entry. + * @max: Maximum index to retrieve from array. + * @filter: Selection criterion. + * + * Initialise @index to the lowest index you want to retrieve from the + * array. During the iteration, @entry will have the value of the entry + * stored in @xa at @index. The iteration will skip all entries in the + * array which do not match @filter. You may modify @index during the + * iteration if you want to skip or reprocess indices. It is safe to modify + * the array during the iteration. At the end of the iteration, @entry will + * be set to NULL and @index will have a value less than or equal to max. + * + * xa_for_each() is O(n.log(n)) while xas_for_each() is O(n). You have + * to handle your own locking with xas_for_each(), and if you have to unlock + * after each iteration, it will also end up being O(n.log(n)). xa_for_each() + * will spin if it hits a retry entry; if you intend to see retry entries, + * you should use the xas_for_each() iterator instead. The xas_for_each() + * iterator will expand into more inline code than xa_for_each(). + * + * Context: Any context. Takes and releases the RCU lock. + */ +#define xa_for_each(xa, entry, index, max, filter) \ + for (entry = xa_find(xa, &index, max, filter); entry; \ + entry = xa_find_after(xa, &index, max, filter)) #define xa_trylock(xa) spin_trylock(&(xa)->xa_lock) #define xa_lock(xa) spin_lock(&(xa)->xa_lock) @@ -21,4 +443,873 @@ #define xa_unlock_irqrestore(xa, flags) \ spin_unlock_irqrestore(&(xa)->xa_lock, flags) +/* + * Versions of the normal API which require the caller to hold the + * xa_lock. If the GFP flags allow it, they will drop the lock to + * allocate memory, then reacquire it afterwards. These functions + * may also re-enable interrupts if the XArray flags indicate the + * locking should be interrupt safe. + */ +void *__xa_erase(struct xarray *, unsigned long index); +void *__xa_store(struct xarray *, unsigned long index, void *entry, gfp_t); +void *__xa_cmpxchg(struct xarray *, unsigned long index, void *old, + void *entry, gfp_t); +int __xa_alloc(struct xarray *, u32 *id, u32 max, void *entry, gfp_t); +void __xa_set_mark(struct xarray *, unsigned long index, xa_mark_t); +void __xa_clear_mark(struct xarray *, unsigned long index, xa_mark_t); + +/** + * __xa_insert() - Store this entry in the XArray unless another entry is + * already present. + * @xa: XArray. + * @index: Index into array. + * @entry: New entry. + * @gfp: Memory allocation flags. + * + * If you would rather see the existing entry in the array, use __xa_cmpxchg(). + * This function is for users who don't care what the entry is, only that + * one is present. + * + * Context: Any context. Expects xa_lock to be held on entry. May + * release and reacquire xa_lock if the @gfp flags permit. + * Return: 0 if the store succeeded. -EEXIST if another entry was present. + * -ENOMEM if memory could not be allocated. + */ +static inline int __xa_insert(struct xarray *xa, unsigned long index, + void *entry, gfp_t gfp) +{ + void *curr = __xa_cmpxchg(xa, index, NULL, entry, gfp); + if (!curr) + return 0; + if (xa_is_err(curr)) + return xa_err(curr); + return -EEXIST; +} + +/** + * xa_erase_bh() - Erase this entry from the XArray. + * @xa: XArray. + * @index: Index of entry. + * + * This function is the equivalent of calling xa_store() with %NULL as + * the third argument. The XArray does not need to allocate memory, so + * the user does not need to provide GFP flags. + * + * Context: Process context. Takes and releases the xa_lock while + * disabling softirqs. + * Return: The entry which used to be at this index. + */ +static inline void *xa_erase_bh(struct xarray *xa, unsigned long index) +{ + void *entry; + + xa_lock_bh(xa); + entry = __xa_erase(xa, index); + xa_unlock_bh(xa); + + return entry; +} + +/** + * xa_erase_irq() - Erase this entry from the XArray. + * @xa: XArray. + * @index: Index of entry. + * + * This function is the equivalent of calling xa_store() with %NULL as + * the third argument. The XArray does not need to allocate memory, so + * the user does not need to provide GFP flags. + * + * Context: Process context. Takes and releases the xa_lock while + * disabling interrupts. + * Return: The entry which used to be at this index. + */ +static inline void *xa_erase_irq(struct xarray *xa, unsigned long index) +{ + void *entry; + + xa_lock_irq(xa); + entry = __xa_erase(xa, index); + xa_unlock_irq(xa); + + return entry; +} + +/** + * xa_alloc() - Find somewhere to store this entry in the XArray. + * @xa: XArray. + * @id: Pointer to ID. + * @max: Maximum ID to allocate (inclusive). + * @entry: New entry. + * @gfp: Memory allocation flags. + * + * Allocates an unused ID in the range specified by @id and @max. + * Updates the @id pointer with the index, then stores the entry at that + * index. A concurrent lookup will not see an uninitialised @id. + * + * Context: Process context. Takes and releases the xa_lock. May sleep if + * the @gfp flags permit. + * Return: 0 on success, -ENOMEM if memory allocation fails or -ENOSPC if + * there is no more space in the XArray. + */ +static inline int xa_alloc(struct xarray *xa, u32 *id, u32 max, void *entry, + gfp_t gfp) +{ + int err; + + xa_lock(xa); + err = __xa_alloc(xa, id, max, entry, gfp); + xa_unlock(xa); + + return err; +} + +/** + * xa_alloc_bh() - Find somewhere to store this entry in the XArray. + * @xa: XArray. + * @id: Pointer to ID. + * @max: Maximum ID to allocate (inclusive). + * @entry: New entry. + * @gfp: Memory allocation flags. + * + * Allocates an unused ID in the range specified by @id and @max. + * Updates the @id pointer with the index, then stores the entry at that + * index. A concurrent lookup will not see an uninitialised @id. + * + * Context: Process context. Takes and releases the xa_lock while + * disabling softirqs. May sleep if the @gfp flags permit. + * Return: 0 on success, -ENOMEM if memory allocation fails or -ENOSPC if + * there is no more space in the XArray. + */ +static inline int xa_alloc_bh(struct xarray *xa, u32 *id, u32 max, void *entry, + gfp_t gfp) +{ + int err; + + xa_lock_bh(xa); + err = __xa_alloc(xa, id, max, entry, gfp); + xa_unlock_bh(xa); + + return err; +} + +/** + * xa_alloc_irq() - Find somewhere to store this entry in the XArray. + * @xa: XArray. + * @id: Pointer to ID. + * @max: Maximum ID to allocate (inclusive). + * @entry: New entry. + * @gfp: Memory allocation flags. + * + * Allocates an unused ID in the range specified by @id and @max. + * Updates the @id pointer with the index, then stores the entry at that + * index. A concurrent lookup will not see an uninitialised @id. + * + * Context: Process context. Takes and releases the xa_lock while + * disabling interrupts. May sleep if the @gfp flags permit. + * Return: 0 on success, -ENOMEM if memory allocation fails or -ENOSPC if + * there is no more space in the XArray. + */ +static inline int xa_alloc_irq(struct xarray *xa, u32 *id, u32 max, void *entry, + gfp_t gfp) +{ + int err; + + xa_lock_irq(xa); + err = __xa_alloc(xa, id, max, entry, gfp); + xa_unlock_irq(xa); + + return err; +} + +/* Everything below here is the Advanced API. Proceed with caution. */ + +/* + * The xarray is constructed out of a set of 'chunks' of pointers. Choosing + * the best chunk size requires some tradeoffs. A power of two recommends + * itself so that we can walk the tree based purely on shifts and masks. + * Generally, the larger the better; as the number of slots per level of the + * tree increases, the less tall the tree needs to be. But that needs to be + * balanced against the memory consumption of each node. On a 64-bit system, + * xa_node is currently 576 bytes, and we get 7 of them per 4kB page. If we + * doubled the number of slots per node, we'd get only 3 nodes per 4kB page. + */ +#ifndef XA_CHUNK_SHIFT +#define XA_CHUNK_SHIFT (CONFIG_BASE_SMALL ? 4 : 6) +#endif +#define XA_CHUNK_SIZE (1UL << XA_CHUNK_SHIFT) +#define XA_CHUNK_MASK (XA_CHUNK_SIZE - 1) +#define XA_MAX_MARKS 3 +#define XA_MARK_LONGS DIV_ROUND_UP(XA_CHUNK_SIZE, BITS_PER_LONG) + +/* + * @count is the count of every non-NULL element in the ->slots array + * whether that is a value entry, a retry entry, a user pointer, + * a sibling entry or a pointer to the next level of the tree. + * @nr_values is the count of every element in ->slots which is + * either a value entry or a sibling of a value entry. + */ +struct xa_node { + unsigned char shift; /* Bits remaining in each slot */ + unsigned char offset; /* Slot offset in parent */ + unsigned char count; /* Total entry count */ + unsigned char nr_values; /* Value entry count */ + struct xa_node __rcu *parent; /* NULL at top of tree */ + struct xarray *array; /* The array we belong to */ + union { + struct list_head private_list; /* For tree user */ + struct rcu_head rcu_head; /* Used when freeing node */ + }; + void __rcu *slots[XA_CHUNK_SIZE]; + union { + unsigned long tags[XA_MAX_MARKS][XA_MARK_LONGS]; + unsigned long marks[XA_MAX_MARKS][XA_MARK_LONGS]; + }; +}; + +void xa_dump(const struct xarray *); +void xa_dump_node(const struct xa_node *); + +#ifdef XA_DEBUG +#define XA_BUG_ON(xa, x) do { \ + if (x) { \ + xa_dump(xa); \ + BUG(); \ + } \ + } while (0) +#define XA_NODE_BUG_ON(node, x) do { \ + if (x) { \ + if (node) xa_dump_node(node); \ + BUG(); \ + } \ + } while (0) +#else +#define XA_BUG_ON(xa, x) do { } while (0) +#define XA_NODE_BUG_ON(node, x) do { } while (0) +#endif + +/* Private */ +static inline void *xa_head(const struct xarray *xa) +{ + return rcu_dereference_check(xa->xa_head, + lockdep_is_held(&xa->xa_lock)); +} + +/* Private */ +static inline void *xa_head_locked(const struct xarray *xa) +{ + return rcu_dereference_protected(xa->xa_head, + lockdep_is_held(&xa->xa_lock)); +} + +/* Private */ +static inline void *xa_entry(const struct xarray *xa, + const struct xa_node *node, unsigned int offset) +{ + XA_NODE_BUG_ON(node, offset >= XA_CHUNK_SIZE); + return rcu_dereference_check(node->slots[offset], + lockdep_is_held(&xa->xa_lock)); +} + +/* Private */ +static inline void *xa_entry_locked(const struct xarray *xa, + const struct xa_node *node, unsigned int offset) +{ + XA_NODE_BUG_ON(node, offset >= XA_CHUNK_SIZE); + return rcu_dereference_protected(node->slots[offset], + lockdep_is_held(&xa->xa_lock)); +} + +/* Private */ +static inline struct xa_node *xa_parent(const struct xarray *xa, + const struct xa_node *node) +{ + return rcu_dereference_check(node->parent, + lockdep_is_held(&xa->xa_lock)); +} + +/* Private */ +static inline struct xa_node *xa_parent_locked(const struct xarray *xa, + const struct xa_node *node) +{ + return rcu_dereference_protected(node->parent, + lockdep_is_held(&xa->xa_lock)); +} + +/* Private */ +static inline void *xa_mk_node(const struct xa_node *node) +{ + return (void *)((unsigned long)node | 2); +} + +/* Private */ +static inline struct xa_node *xa_to_node(const void *entry) +{ + return (struct xa_node *)((unsigned long)entry - 2); +} + +/* Private */ +static inline bool xa_is_node(const void *entry) +{ + return xa_is_internal(entry) && (unsigned long)entry > 4096; +} + +/* Private */ +static inline void *xa_mk_sibling(unsigned int offset) +{ + return xa_mk_internal(offset); +} + +/* Private */ +static inline unsigned long xa_to_sibling(const void *entry) +{ + return xa_to_internal(entry); +} + +/** + * xa_is_sibling() - Is the entry a sibling entry? + * @entry: Entry retrieved from the XArray + * + * Return: %true if the entry is a sibling entry. + */ +static inline bool xa_is_sibling(const void *entry) +{ + return IS_ENABLED(CONFIG_XARRAY_MULTI) && xa_is_internal(entry) && + (entry < xa_mk_sibling(XA_CHUNK_SIZE - 1)); +} + +#define XA_ZERO_ENTRY xa_mk_internal(256) +#define XA_RETRY_ENTRY xa_mk_internal(257) + +/** + * xa_is_zero() - Is the entry a zero entry? + * @entry: Entry retrieved from the XArray + * + * Return: %true if the entry is a zero entry. + */ +static inline bool xa_is_zero(const void *entry) +{ + return unlikely(entry == XA_ZERO_ENTRY); +} + +/** + * xa_is_retry() - Is the entry a retry entry? + * @entry: Entry retrieved from the XArray + * + * Return: %true if the entry is a retry entry. + */ +static inline bool xa_is_retry(const void *entry) +{ + return unlikely(entry == XA_RETRY_ENTRY); +} + +/** + * typedef xa_update_node_t - A callback function from the XArray. + * @node: The node which is being processed + * + * This function is called every time the XArray updates the count of + * present and value entries in a node. It allows advanced users to + * maintain the private_list in the node. + * + * Context: The xa_lock is held and interrupts may be disabled. + * Implementations should not drop the xa_lock, nor re-enable + * interrupts. + */ +typedef void (*xa_update_node_t)(struct xa_node *node); + +/* + * The xa_state is opaque to its users. It contains various different pieces + * of state involved in the current operation on the XArray. It should be + * declared on the stack and passed between the various internal routines. + * The various elements in it should not be accessed directly, but only + * through the provided accessor functions. The below documentation is for + * the benefit of those working on the code, not for users of the XArray. + * + * @xa_node usually points to the xa_node containing the slot we're operating + * on (and @xa_offset is the offset in the slots array). If there is a + * single entry in the array at index 0, there are no allocated xa_nodes to + * point to, and so we store %NULL in @xa_node. @xa_node is set to + * the value %XAS_RESTART if the xa_state is not walked to the correct + * position in the tree of nodes for this operation. If an error occurs + * during an operation, it is set to an %XAS_ERROR value. If we run off the + * end of the allocated nodes, it is set to %XAS_BOUNDS. + */ +struct xa_state { + struct xarray *xa; + unsigned long xa_index; + unsigned char xa_shift; + unsigned char xa_sibs; + unsigned char xa_offset; + unsigned char xa_pad; /* Helps gcc generate better code */ + struct xa_node *xa_node; + struct xa_node *xa_alloc; + xa_update_node_t xa_update; +}; + +/* + * We encode errnos in the xas->xa_node. If an error has happened, we need to + * drop the lock to fix it, and once we've done so the xa_state is invalid. + */ +#define XA_ERROR(errno) ((struct xa_node *)(((unsigned long)errno << 2) | 2UL)) +#define XAS_BOUNDS ((struct xa_node *)1UL) +#define XAS_RESTART ((struct xa_node *)3UL) + +#define __XA_STATE(array, index, shift, sibs) { \ + .xa = array, \ + .xa_index = index, \ + .xa_shift = shift, \ + .xa_sibs = sibs, \ + .xa_offset = 0, \ + .xa_pad = 0, \ + .xa_node = XAS_RESTART, \ + .xa_alloc = NULL, \ + .xa_update = NULL \ +} + +/** + * XA_STATE() - Declare an XArray operation state. + * @name: Name of this operation state (usually xas). + * @array: Array to operate on. + * @index: Initial index of interest. + * + * Declare and initialise an xa_state on the stack. + */ +#define XA_STATE(name, array, index) \ + struct xa_state name = __XA_STATE(array, index, 0, 0) + +/** + * XA_STATE_ORDER() - Declare an XArray operation state. + * @name: Name of this operation state (usually xas). + * @array: Array to operate on. + * @index: Initial index of interest. + * @order: Order of entry. + * + * Declare and initialise an xa_state on the stack. This variant of + * XA_STATE() allows you to specify the 'order' of the element you + * want to operate on.` + */ +#define XA_STATE_ORDER(name, array, index, order) \ + struct xa_state name = __XA_STATE(array, \ + (index >> order) << order, \ + order - (order % XA_CHUNK_SHIFT), \ + (1U << (order % XA_CHUNK_SHIFT)) - 1) + +#define xas_marked(xas, mark) xa_marked((xas)->xa, (mark)) +#define xas_trylock(xas) xa_trylock((xas)->xa) +#define xas_lock(xas) xa_lock((xas)->xa) +#define xas_unlock(xas) xa_unlock((xas)->xa) +#define xas_lock_bh(xas) xa_lock_bh((xas)->xa) +#define xas_unlock_bh(xas) xa_unlock_bh((xas)->xa) +#define xas_lock_irq(xas) xa_lock_irq((xas)->xa) +#define xas_unlock_irq(xas) xa_unlock_irq((xas)->xa) +#define xas_lock_irqsave(xas, flags) \ + xa_lock_irqsave((xas)->xa, flags) +#define xas_unlock_irqrestore(xas, flags) \ + xa_unlock_irqrestore((xas)->xa, flags) + +/** + * xas_error() - Return an errno stored in the xa_state. + * @xas: XArray operation state. + * + * Return: 0 if no error has been noted. A negative errno if one has. + */ +static inline int xas_error(const struct xa_state *xas) +{ + return xa_err(xas->xa_node); +} + +/** + * xas_set_err() - Note an error in the xa_state. + * @xas: XArray operation state. + * @err: Negative error number. + * + * Only call this function with a negative @err; zero or positive errors + * will probably not behave the way you think they should. If you want + * to clear the error from an xa_state, use xas_reset(). + */ +static inline void xas_set_err(struct xa_state *xas, long err) +{ + xas->xa_node = XA_ERROR(err); +} + +/** + * xas_invalid() - Is the xas in a retry or error state? + * @xas: XArray operation state. + * + * Return: %true if the xas cannot be used for operations. + */ +static inline bool xas_invalid(const struct xa_state *xas) +{ + return (unsigned long)xas->xa_node & 3; +} + +/** + * xas_valid() - Is the xas a valid cursor into the array? + * @xas: XArray operation state. + * + * Return: %true if the xas can be used for operations. + */ +static inline bool xas_valid(const struct xa_state *xas) +{ + return !xas_invalid(xas); +} + +/** + * xas_is_node() - Does the xas point to a node? + * @xas: XArray operation state. + * + * Return: %true if the xas currently references a node. + */ +static inline bool xas_is_node(const struct xa_state *xas) +{ + return xas_valid(xas) && xas->xa_node; +} + +/* True if the pointer is something other than a node */ +static inline bool xas_not_node(struct xa_node *node) +{ + return ((unsigned long)node & 3) || !node; +} + +/* True if the node represents RESTART or an error */ +static inline bool xas_frozen(struct xa_node *node) +{ + return (unsigned long)node & 2; +} + +/* True if the node represents head-of-tree, RESTART or BOUNDS */ +static inline bool xas_top(struct xa_node *node) +{ + return node <= XAS_RESTART; +} + +/** + * xas_reset() - Reset an XArray operation state. + * @xas: XArray operation state. + * + * Resets the error or walk state of the @xas so future walks of the + * array will start from the root. Use this if you have dropped the + * xarray lock and want to reuse the xa_state. + * + * Context: Any context. + */ +static inline void xas_reset(struct xa_state *xas) +{ + xas->xa_node = XAS_RESTART; +} + +/** + * xas_retry() - Retry the operation if appropriate. + * @xas: XArray operation state. + * @entry: Entry from xarray. + * + * The advanced functions may sometimes return an internal entry, such as + * a retry entry or a zero entry. This function sets up the @xas to restart + * the walk from the head of the array if needed. + * + * Context: Any context. + * Return: true if the operation needs to be retried. + */ +static inline bool xas_retry(struct xa_state *xas, const void *entry) +{ + if (xa_is_zero(entry)) + return true; + if (!xa_is_retry(entry)) + return false; + xas_reset(xas); + return true; +} + +void *xas_load(struct xa_state *); +void *xas_store(struct xa_state *, void *entry); +void *xas_find(struct xa_state *, unsigned long max); +void *xas_find_conflict(struct xa_state *); + +bool xas_get_mark(const struct xa_state *, xa_mark_t); +void xas_set_mark(const struct xa_state *, xa_mark_t); +void xas_clear_mark(const struct xa_state *, xa_mark_t); +void *xas_find_marked(struct xa_state *, unsigned long max, xa_mark_t); +void xas_init_marks(const struct xa_state *); + +bool xas_nomem(struct xa_state *, gfp_t); +void xas_pause(struct xa_state *); + +void xas_create_range(struct xa_state *); + +/** + * xas_reload() - Refetch an entry from the xarray. + * @xas: XArray operation state. + * + * Use this function to check that a previously loaded entry still has + * the same value. This is useful for the lockless pagecache lookup where + * we walk the array with only the RCU lock to protect us, lock the page, + * then check that the page hasn't moved since we looked it up. + * + * The caller guarantees that @xas is still valid. If it may be in an + * error or restart state, call xas_load() instead. + * + * Return: The entry at this location in the xarray. + */ +static inline void *xas_reload(struct xa_state *xas) +{ + struct xa_node *node = xas->xa_node; + + if (node) + return xa_entry(xas->xa, node, xas->xa_offset); + return xa_head(xas->xa); +} + +/** + * xas_set() - Set up XArray operation state for a different index. + * @xas: XArray operation state. + * @index: New index into the XArray. + * + * Move the operation state to refer to a different index. This will + * have the effect of starting a walk from the top; see xas_next() + * to move to an adjacent index. + */ +static inline void xas_set(struct xa_state *xas, unsigned long index) +{ + xas->xa_index = index; + xas->xa_node = XAS_RESTART; +} + +/** + * xas_set_order() - Set up XArray operation state for a multislot entry. + * @xas: XArray operation state. + * @index: Target of the operation. + * @order: Entry occupies 2^@order indices. + */ +static inline void xas_set_order(struct xa_state *xas, unsigned long index, + unsigned int order) +{ +#ifdef CONFIG_XARRAY_MULTI + xas->xa_index = order < BITS_PER_LONG ? (index >> order) << order : 0; + xas->xa_shift = order - (order % XA_CHUNK_SHIFT); + xas->xa_sibs = (1 << (order % XA_CHUNK_SHIFT)) - 1; + xas->xa_node = XAS_RESTART; +#else + BUG_ON(order > 0); + xas_set(xas, index); +#endif +} + +/** + * xas_set_update() - Set up XArray operation state for a callback. + * @xas: XArray operation state. + * @update: Function to call when updating a node. + * + * The XArray can notify a caller after it has updated an xa_node. + * This is advanced functionality and is only needed by the page cache. + */ +static inline void xas_set_update(struct xa_state *xas, xa_update_node_t update) +{ + xas->xa_update = update; +} + +/** + * xas_next_entry() - Advance iterator to next present entry. + * @xas: XArray operation state. + * @max: Highest index to return. + * + * xas_next_entry() is an inline function to optimise xarray traversal for + * speed. It is equivalent to calling xas_find(), and will call xas_find() + * for all the hard cases. + * + * Return: The next present entry after the one currently referred to by @xas. + */ +static inline void *xas_next_entry(struct xa_state *xas, unsigned long max) +{ + struct xa_node *node = xas->xa_node; + void *entry; + + if (unlikely(xas_not_node(node) || node->shift || + xas->xa_offset != (xas->xa_index & XA_CHUNK_MASK))) + return xas_find(xas, max); + + do { + if (unlikely(xas->xa_index >= max)) + return xas_find(xas, max); + if (unlikely(xas->xa_offset == XA_CHUNK_MASK)) + return xas_find(xas, max); + entry = xa_entry(xas->xa, node, xas->xa_offset + 1); + if (unlikely(xa_is_internal(entry))) + return xas_find(xas, max); + xas->xa_offset++; + xas->xa_index++; + } while (!entry); + + return entry; +} + +/* Private */ +static inline unsigned int xas_find_chunk(struct xa_state *xas, bool advance, + xa_mark_t mark) +{ + unsigned long *addr = xas->xa_node->marks[(__force unsigned)mark]; + unsigned int offset = xas->xa_offset; + + if (advance) + offset++; + if (XA_CHUNK_SIZE == BITS_PER_LONG) { + if (offset < XA_CHUNK_SIZE) { + unsigned long data = *addr & (~0UL << offset); + if (data) + return __ffs(data); + } + return XA_CHUNK_SIZE; + } + + return find_next_bit(addr, XA_CHUNK_SIZE, offset); +} + +/** + * xas_next_marked() - Advance iterator to next marked entry. + * @xas: XArray operation state. + * @max: Highest index to return. + * @mark: Mark to search for. + * + * xas_next_marked() is an inline function to optimise xarray traversal for + * speed. It is equivalent to calling xas_find_marked(), and will call + * xas_find_marked() for all the hard cases. + * + * Return: The next marked entry after the one currently referred to by @xas. + */ +static inline void *xas_next_marked(struct xa_state *xas, unsigned long max, + xa_mark_t mark) +{ + struct xa_node *node = xas->xa_node; + unsigned int offset; + + if (unlikely(xas_not_node(node) || node->shift)) + return xas_find_marked(xas, max, mark); + offset = xas_find_chunk(xas, true, mark); + xas->xa_offset = offset; + xas->xa_index = (xas->xa_index & ~XA_CHUNK_MASK) + offset; + if (xas->xa_index > max) + return NULL; + if (offset == XA_CHUNK_SIZE) + return xas_find_marked(xas, max, mark); + return xa_entry(xas->xa, node, offset); +} + +/* + * If iterating while holding a lock, drop the lock and reschedule + * every %XA_CHECK_SCHED loops. + */ +enum { + XA_CHECK_SCHED = 4096, +}; + +/** + * xas_for_each() - Iterate over a range of an XArray. + * @xas: XArray operation state. + * @entry: Entry retrieved from the array. + * @max: Maximum index to retrieve from array. + * + * The loop body will be executed for each entry present in the xarray + * between the current xas position and @max. @entry will be set to + * the entry retrieved from the xarray. It is safe to delete entries + * from the array in the loop body. You should hold either the RCU lock + * or the xa_lock while iterating. If you need to drop the lock, call + * xas_pause() first. + */ +#define xas_for_each(xas, entry, max) \ + for (entry = xas_find(xas, max); entry; \ + entry = xas_next_entry(xas, max)) + +/** + * xas_for_each_marked() - Iterate over a range of an XArray. + * @xas: XArray operation state. + * @entry: Entry retrieved from the array. + * @max: Maximum index to retrieve from array. + * @mark: Mark to search for. + * + * The loop body will be executed for each marked entry in the xarray + * between the current xas position and @max. @entry will be set to + * the entry retrieved from the xarray. It is safe to delete entries + * from the array in the loop body. You should hold either the RCU lock + * or the xa_lock while iterating. If you need to drop the lock, call + * xas_pause() first. + */ +#define xas_for_each_marked(xas, entry, max, mark) \ + for (entry = xas_find_marked(xas, max, mark); entry; \ + entry = xas_next_marked(xas, max, mark)) + +/** + * xas_for_each_conflict() - Iterate over a range of an XArray. + * @xas: XArray operation state. + * @entry: Entry retrieved from the array. + * + * The loop body will be executed for each entry in the XArray that lies + * within the range specified by @xas. If the loop completes successfully, + * any entries that lie in this range will be replaced by @entry. The caller + * may break out of the loop; if they do so, the contents of the XArray will + * be unchanged. The operation may fail due to an out of memory condition. + * The caller may also call xa_set_err() to exit the loop while setting an + * error to record the reason. + */ +#define xas_for_each_conflict(xas, entry) \ + while ((entry = xas_find_conflict(xas))) + +void *__xas_next(struct xa_state *); +void *__xas_prev(struct xa_state *); + +/** + * xas_prev() - Move iterator to previous index. + * @xas: XArray operation state. + * + * If the @xas was in an error state, it will remain in an error state + * and this function will return %NULL. If the @xas has never been walked, + * it will have the effect of calling xas_load(). Otherwise one will be + * subtracted from the index and the state will be walked to the correct + * location in the array for the next operation. + * + * If the iterator was referencing index 0, this function wraps + * around to %ULONG_MAX. + * + * Return: The entry at the new index. This may be %NULL or an internal + * entry. + */ +static inline void *xas_prev(struct xa_state *xas) +{ + struct xa_node *node = xas->xa_node; + + if (unlikely(xas_not_node(node) || node->shift || + xas->xa_offset == 0)) + return __xas_prev(xas); + + xas->xa_index--; + xas->xa_offset--; + return xa_entry(xas->xa, node, xas->xa_offset); +} + +/** + * xas_next() - Move state to next index. + * @xas: XArray operation state. + * + * If the @xas was in an error state, it will remain in an error state + * and this function will return %NULL. If the @xas has never been walked, + * it will have the effect of calling xas_load(). Otherwise one will be + * added to the index and the state will be walked to the correct + * location in the array for the next operation. + * + * If the iterator was referencing index %ULONG_MAX, this function wraps + * around to 0. + * + * Return: The entry at the new index. This may be %NULL or an internal + * entry. + */ +static inline void *xas_next(struct xa_state *xas) +{ + struct xa_node *node = xas->xa_node; + + if (unlikely(xas_not_node(node) || node->shift || + xas->xa_offset == XA_CHUNK_MASK)) + return __xas_next(xas); + + xas->xa_index++; + xas->xa_offset++; + return xa_entry(xas->xa, node, xas->xa_offset); +} + #endif /* _LINUX_XARRAY_H */ diff --git a/kernel/memremap.c b/kernel/memremap.c @@ -1,47 +1,21 @@ /* SPDX-License-Identifier: GPL-2.0 */ /* Copyright(c) 2015 Intel Corporation. All rights reserved. */ -#include <linux/radix-tree.h> #include <linux/device.h> -#include <linux/types.h> -#include <linux/pfn_t.h> #include <linux/io.h> #include <linux/kasan.h> -#include <linux/mm.h> #include <linux/memory_hotplug.h> +#include <linux/mm.h> +#include <linux/pfn_t.h> #include <linux/swap.h> #include <linux/swapops.h> +#include <linux/types.h> #include <linux/wait_bit.h> +#include <linux/xarray.h> -static DEFINE_MUTEX(pgmap_lock); -static RADIX_TREE(pgmap_radix, GFP_KERNEL); +static DEFINE_XARRAY(pgmap_array); #define SECTION_MASK ~((1UL << PA_SECTION_SHIFT) - 1) #define SECTION_SIZE (1UL << PA_SECTION_SHIFT) -static unsigned long order_at(struct resource *res, unsigned long pgoff) -{ - unsigned long phys_pgoff = PHYS_PFN(res->start) + pgoff; - unsigned long nr_pages, mask; - - nr_pages = PHYS_PFN(resource_size(res)); - if (nr_pages == pgoff) - return ULONG_MAX; - - /* - * What is the largest aligned power-of-2 range available from - * this resource pgoff to the end of the resource range, - * considering the alignment of the current pgoff? - */ - mask = phys_pgoff | rounddown_pow_of_two(nr_pages - pgoff); - if (!mask) - return ULONG_MAX; - - return find_first_bit(&mask, BITS_PER_LONG); -} - -#define foreach_order_pgoff(res, order, pgoff) \ - for (pgoff = 0, order = order_at((res), pgoff); order < ULONG_MAX; \ - pgoff += 1UL << order, order = order_at((res), pgoff)) - #if IS_ENABLED(CONFIG_DEVICE_PRIVATE) vm_fault_t device_private_entry_fault(struct vm_area_struct *vma, unsigned long addr, @@ -70,18 +44,10 @@ vm_fault_t device_private_entry_fault(struct vm_area_struct *vma, EXPORT_SYMBOL(device_private_entry_fault); #endif /* CONFIG_DEVICE_PRIVATE */ -static void pgmap_radix_release(struct resource *res, unsigned long end_pgoff) +static void pgmap_array_delete(struct resource *res) { - unsigned long pgoff, order; - - mutex_lock(&pgmap_lock); - foreach_order_pgoff(res, order, pgoff) { - if (pgoff >= end_pgoff) - break; - radix_tree_delete(&pgmap_radix, PHYS_PFN(res->start) + pgoff); - } - mutex_unlock(&pgmap_lock); - + xa_store_range(&pgmap_array, PHYS_PFN(res->start), PHYS_PFN(res->end), + NULL, GFP_KERNEL); synchronize_rcu(); } @@ -142,7 +108,7 @@ static void devm_memremap_pages_release(void *data) mem_hotplug_done(); untrack_pfn(NULL, PHYS_PFN(align_start), align_size); - pgmap_radix_release(res, -1); + pgmap_array_delete(res); dev_WARN_ONCE(dev, pgmap->altmap.alloc, "%s: failed to free all reserved pages\n", __func__); } @@ -177,7 +143,6 @@ void *devm_memremap_pages(struct device *dev, struct dev_pagemap *pgmap) struct resource *res = &pgmap->res; struct dev_pagemap *conflict_pgmap; pgprot_t pgprot = PAGE_KERNEL; - unsigned long pgoff, order; int error, nid, is_ram; align_start = res->start & ~(SECTION_SIZE - 1); @@ -216,20 +181,10 @@ void *devm_memremap_pages(struct device *dev, struct dev_pagemap *pgmap) pgmap->dev = dev; - mutex_lock(&pgmap_lock); - error = 0; - - foreach_order_pgoff(res, order, pgoff) { - error = __radix_tree_insert(&pgmap_radix, - PHYS_PFN(res->start) + pgoff, order, pgmap); - if (error) { - dev_err(dev, "%s: failed: %d\n", __func__, error); - break; - } - } - mutex_unlock(&pgmap_lock); + error = xa_err(xa_store_range(&pgmap_array, PHYS_PFN(res->start), + PHYS_PFN(res->end), pgmap, GFP_KERNEL)); if (error) - goto err_radix; + goto err_array; nid = dev_to_node(dev); if (nid < 0) @@ -274,8 +229,8 @@ void *devm_memremap_pages(struct device *dev, struct dev_pagemap *pgmap) err_kasan: untrack_pfn(NULL, PHYS_PFN(align_start), align_size); err_pfn_remap: - err_radix: - pgmap_radix_release(res, pgoff); + pgmap_array_delete(res); + err_array: return ERR_PTR(error); } EXPORT_SYMBOL(devm_memremap_pages); @@ -315,7 +270,7 @@ struct dev_pagemap *get_dev_pagemap(unsigned long pfn, /* fall back to slow path lookup */ rcu_read_lock(); - pgmap = radix_tree_lookup(&pgmap_radix, PHYS_PFN(phys)); + pgmap = xa_load(&pgmap_array, PHYS_PFN(phys)); if (pgmap && !percpu_ref_tryget_live(pgmap->ref)) pgmap = NULL; rcu_read_unlock(); diff --git a/lib/Kconfig b/lib/Kconfig @@ -399,8 +399,11 @@ config INTERVAL_TREE for more information. -config RADIX_TREE_MULTIORDER +config XARRAY_MULTI bool + help + Support entries which occupy multiple consecutive indices in the + XArray. config ASSOCIATIVE_ARRAY bool diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug @@ -1813,6 +1813,9 @@ config TEST_BITFIELD config TEST_UUID tristate "Test functions located in the uuid module at runtime" +config TEST_XARRAY + tristate "Test the XArray code at runtime" + config TEST_OVERFLOW tristate "Test check_*_overflow() functions at runtime" diff --git a/lib/Makefile b/lib/Makefile @@ -18,7 +18,7 @@ KCOV_INSTRUMENT_debugobjects.o := n KCOV_INSTRUMENT_dynamic_debug.o := n lib-y := ctype.o string.o vsprintf.o cmdline.o \ - rbtree.o radix-tree.o timerqueue.o\ + rbtree.o radix-tree.o timerqueue.o xarray.o \ idr.o int_sqrt.o extable.o \ sha1.o chacha20.o irq_regs.o argv_split.o \ flex_proportions.o ratelimit.o show_mem.o \ @@ -68,6 +68,7 @@ obj-$(CONFIG_TEST_PRINTF) += test_printf.o obj-$(CONFIG_TEST_BITMAP) += test_bitmap.o obj-$(CONFIG_TEST_BITFIELD) += test_bitfield.o obj-$(CONFIG_TEST_UUID) += test_uuid.o +obj-$(CONFIG_TEST_XARRAY) += test_xarray.o obj-$(CONFIG_TEST_PARMAN) += test_parman.o obj-$(CONFIG_TEST_KMOD) += test_kmod.o obj-$(CONFIG_TEST_DEBUG_VIRTUAL) += test_debug_virtual.o diff --git a/lib/idr.c b/lib/idr.c @@ -6,8 +6,6 @@ #include <linux/spinlock.h> #include <linux/xarray.h> -DEFINE_PER_CPU(struct ida_bitmap *, ida_bitmap); - /** * idr_alloc_u32() - Allocate an ID. * @idr: IDR handle. @@ -39,10 +37,8 @@ int idr_alloc_u32(struct idr *idr, void *ptr, u32 *nextid, unsigned int base = idr->idr_base; unsigned int id = *nextid; - if (WARN_ON_ONCE(radix_tree_is_internal_node(ptr))) - return -EINVAL; - if (WARN_ON_ONCE(!(idr->idr_rt.gfp_mask & ROOT_IS_IDR))) - idr->idr_rt.gfp_mask |= IDR_RT_MARKER; + if (WARN_ON_ONCE(!(idr->idr_rt.xa_flags & ROOT_IS_IDR))) + idr->idr_rt.xa_flags |= IDR_RT_MARKER; id = (id < base) ? 0 : id - base; radix_tree_iter_init(&iter, id); @@ -295,15 +291,13 @@ void *idr_replace(struct idr *idr, void *ptr, unsigned long id) void __rcu **slot = NULL; void *entry; - if (WARN_ON_ONCE(radix_tree_is_internal_node(ptr))) - return ERR_PTR(-EINVAL); id -= idr->idr_base; entry = __radix_tree_lookup(&idr->idr_rt, id, &node, &slot); if (!slot || radix_tree_tag_get(&idr->idr_rt, id, IDR_FREE)) return ERR_PTR(-ENOENT); - __radix_tree_replace(&idr->idr_rt, node, slot, ptr, NULL); + __radix_tree_replace(&idr->idr_rt, node, slot, ptr); return entry; } @@ -324,6 +318,9 @@ EXPORT_SYMBOL(idr_replace); * free the individual IDs in it. You can use ida_is_empty() to find * out whether the IDA has any IDs currently allocated. * + * The IDA handles its own locking. It is safe to call any of the IDA + * functions without synchronisation in your code. + * * IDs are currently limited to the range [0-INT_MAX]. If this is an awkward * limitation, it should be quite straightforward to raise the maximum. */ @@ -331,161 +328,197 @@ EXPORT_SYMBOL(idr_replace); /* * Developer's notes: * - * The IDA uses the functionality provided by the IDR & radix tree to store - * bitmaps in each entry. The IDR_FREE tag means there is at least one bit - * free, unlike the IDR where it means at least one entry is free. + * The IDA uses the functionality provided by the XArray to store bitmaps in + * each entry. The XA_FREE_MARK is only cleared when all bits in the bitmap + * have been set. * - * I considered telling the radix tree that each slot is an order-10 node - * and storing the bit numbers in the radix tree, but the radix tree can't - * allow a single multiorder entry at index 0, which would significantly - * increase memory consumption for the IDA. So instead we divide the index - * by the number of bits in the leaf bitmap before doing a radix tree lookup. + * I considered telling the XArray that each slot is an order-10 node + * and indexing by bit number, but the XArray can't allow a single multi-index + * entry in the head, which would significantly increase memory consumption + * for the IDA. So instead we divide the index by the number of bits in the + * leaf bitmap before doing a radix tree lookup. * * As an optimisation, if there are only a few low bits set in any given - * leaf, instead of allocating a 128-byte bitmap, we use the 'exceptional - * entry' functionality of the radix tree to store BITS_PER_LONG - 2 bits - * directly in the entry. By being really tricksy, we could store - * BITS_PER_LONG - 1 bits, but there're diminishing returns after optimising - * for 0-3 allocated IDs. - * - * We allow the radix tree 'exceptional' count to get out of date. Nothing - * in the IDA nor the radix tree code checks it. If it becomes important - * to maintain an accurate exceptional count, switch the rcu_assign_pointer() - * calls to radix_tree_iter_replace() which will correct the exceptional - * count. - * - * The IDA always requires a lock to alloc/free. If we add a 'test_bit' + * leaf, instead of allocating a 128-byte bitmap, we store the bits + * as a value entry. Value entries never have the XA_FREE_MARK cleared + * because we can always convert them into a bitmap entry. + * + * It would be possible to optimise further; once we've run out of a + * single 128-byte bitmap, we currently switch to a 576-byte node, put + * the 128-byte bitmap in the first entry and then start allocating extra + * 128-byte entries. We could instead use the 512 bytes of the node's + * data as a bitmap before moving to that scheme. I do not believe this + * is a worthwhile optimisation; Rasmus Villemoes surveyed the current + * users of the IDA and almost none of them use more than 1024 entries. + * Those that do use more than the 8192 IDs that the 512 bytes would + * provide. + * + * The IDA always uses a lock to alloc/free. If we add a 'test_bit' * equivalent, it will still need locking. Going to RCU lookup would require * using RCU to free bitmaps, and that's not trivial without embedding an * RCU head in the bitmap, which adds a 2-pointer overhead to each 128-byte * bitmap, which is excessive. */ -#define IDA_MAX (0x80000000U / IDA_BITMAP_BITS - 1) - -static int ida_get_new_above(struct ida *ida, int start) +/** + * ida_alloc_range() - Allocate an unused ID. + * @ida: IDA handle. + * @min: Lowest ID to allocate. + * @max: Highest ID to allocate. + * @gfp: Memory allocation flags. + * + * Allocate an ID between @min and @max, inclusive. The allocated ID will + * not exceed %INT_MAX, even if @max is larger. + * + * Context: Any context. + * Return: The allocated ID, or %-ENOMEM if memory could not be allocated, + * or %-ENOSPC if there are no free IDs. + */ +int ida_alloc_range(struct ida *ida, unsigned int min, unsigned int max, + gfp_t gfp) { - struct radix_tree_root *root = &ida->ida_rt; - void __rcu **slot; - struct radix_tree_iter iter; - struct ida_bitmap *bitmap; - unsigned long index; - unsigned bit, ebit; - int new; - - index = start / IDA_BITMAP_BITS; - bit = start % IDA_BITMAP_BITS; - ebit = bit + RADIX_TREE_EXCEPTIONAL_SHIFT; - - slot = radix_tree_iter_init(&iter, index); - for (;;) { - if (slot) - slot = radix_tree_next_slot(slot, &iter, - RADIX_TREE_ITER_TAGGED); - if (!slot) { - slot = idr_get_free(root, &iter, GFP_NOWAIT, IDA_MAX); - if (IS_ERR(slot)) { - if (slot == ERR_PTR(-ENOMEM)) - return -EAGAIN; - return PTR_ERR(slot); + XA_STATE(xas, &ida->xa, min / IDA_BITMAP_BITS); + unsigned bit = min % IDA_BITMAP_BITS; + unsigned long flags; + struct ida_bitmap *bitmap, *alloc = NULL; + + if ((int)min < 0) + return -ENOSPC; + + if ((int)max < 0) + max = INT_MAX; + +retry: + xas_lock_irqsave(&xas, flags); +next: + bitmap = xas_find_marked(&xas, max / IDA_BITMAP_BITS, XA_FREE_MARK); + if (xas.xa_index > min / IDA_BITMAP_BITS) + bit = 0; + if (xas.xa_index * IDA_BITMAP_BITS + bit > max) + goto nospc; + + if (xa_is_value(bitmap)) { + unsigned long tmp = xa_to_value(bitmap); + + if (bit < BITS_PER_XA_VALUE) { + bit = find_next_zero_bit(&tmp, BITS_PER_XA_VALUE, bit); + if (xas.xa_index * IDA_BITMAP_BITS + bit > max) + goto nospc; + if (bit < BITS_PER_XA_VALUE) { + tmp |= 1UL << bit; + xas_store(&xas, xa_mk_value(tmp)); + goto out; } } - if (iter.index > index) { - bit = 0; - ebit = RADIX_TREE_EXCEPTIONAL_SHIFT; - } - new = iter.index * IDA_BITMAP_BITS; - bitmap = rcu_dereference_raw(*slot); - if (radix_tree_exception(bitmap)) { - unsigned long tmp = (unsigned long)bitmap; - ebit = find_next_zero_bit(&tmp, BITS_PER_LONG, ebit); - if (ebit < BITS_PER_LONG) { - tmp |= 1UL << ebit; - rcu_assign_pointer(*slot, (void *)tmp); - return new + ebit - - RADIX_TREE_EXCEPTIONAL_SHIFT; - } - bitmap = this_cpu_xchg(ida_bitmap, NULL); - if (!bitmap) - return -EAGAIN; - bitmap->bitmap[0] = tmp >> RADIX_TREE_EXCEPTIONAL_SHIFT; - rcu_assign_pointer(*slot, bitmap); + bitmap = alloc; + if (!bitmap) + bitmap = kzalloc(sizeof(*bitmap), GFP_NOWAIT); + if (!bitmap) + goto alloc; + bitmap->bitmap[0] = tmp; + xas_store(&xas, bitmap); + if (xas_error(&xas)) { + bitmap->bitmap[0] = 0; + goto out; } + } - if (bitmap) { - bit = find_next_zero_bit(bitmap->bitmap, - IDA_BITMAP_BITS, bit); - new += bit; - if (new < 0) - return -ENOSPC; - if (bit == IDA_BITMAP_BITS) - continue; + if (bitmap) { + bit = find_next_zero_bit(bitmap->bitmap, IDA_BITMAP_BITS, bit); + if (xas.xa_index * IDA_BITMAP_BITS + bit > max) + goto nospc; + if (bit == IDA_BITMAP_BITS) + goto next; - __set_bit(bit, bitmap->bitmap); - if (bitmap_full(bitmap->bitmap, IDA_BITMAP_BITS)) - radix_tree_iter_tag_clear(root, &iter, - IDR_FREE); + __set_bit(bit, bitmap->bitmap); + if (bitmap_full(bitmap->bitmap, IDA_BITMAP_BITS)) + xas_clear_mark(&xas, XA_FREE_MARK); + } else { + if (bit < BITS_PER_XA_VALUE) { + bitmap = xa_mk_value(1UL << bit); } else { - new += bit; - if (new < 0) - return -ENOSPC; - if (ebit < BITS_PER_LONG) { - bitmap = (void *)((1UL << ebit) | - RADIX_TREE_EXCEPTIONAL_ENTRY); - radix_tree_iter_replace(root, &iter, slot, - bitmap); - return new; - } - bitmap = this_cpu_xchg(ida_bitmap, NULL); + bitmap = alloc; if (!bitmap) - return -EAGAIN; + bitmap = kzalloc(sizeof(*bitmap), GFP_NOWAIT); + if (!bitmap) + goto alloc; __set_bit(bit, bitmap->bitmap); - radix_tree_iter_replace(root, &iter, slot, bitmap); } - - return new; + xas_store(&xas, bitmap); + } +out: + xas_unlock_irqrestore(&xas, flags); + if (xas_nomem(&xas, gfp)) { + xas.xa_index = min / IDA_BITMAP_BITS; + bit = min % IDA_BITMAP_BITS; + goto retry; } + if (bitmap != alloc) + kfree(alloc); + if (xas_error(&xas)) + return xas_error(&xas); + return xas.xa_index * IDA_BITMAP_BITS + bit; +alloc: + xas_unlock_irqrestore(&xas, flags); + alloc = kzalloc(sizeof(*bitmap), gfp); + if (!alloc) + return -ENOMEM; + xas_set(&xas, min / IDA_BITMAP_BITS); + bit = min % IDA_BITMAP_BITS; + goto retry; +nospc: + xas_unlock_irqrestore(&xas, flags); + return -ENOSPC; } +EXPORT_SYMBOL(ida_alloc_range); -static void ida_remove(struct ida *ida, int id) +/** + * ida_free() - Release an allocated ID. + * @ida: IDA handle. + * @id: Previously allocated ID. + * + * Context: Any context. + */ +void ida_free(struct ida *ida, unsigned int id) { - unsigned long index = id / IDA_BITMAP_BITS; - unsigned offset = id % IDA_BITMAP_BITS; + XA_STATE(xas, &ida->xa, id / IDA_BITMAP_BITS); + unsigned bit = id % IDA_BITMAP_BITS; struct ida_bitmap *bitmap; - unsigned long *btmp; - struct radix_tree_iter iter; - void __rcu **slot; + unsigned long flags; - slot = radix_tree_iter_lookup(&ida->ida_rt, &iter, index); - if (!slot) - goto err; + BUG_ON((int)id < 0); + + xas_lock_irqsave(&xas, flags); + bitmap = xas_load(&xas); - bitmap = rcu_dereference_raw(*slot); - if (radix_tree_exception(bitmap)) { - btmp = (unsigned long *)slot; - offset += RADIX_TREE_EXCEPTIONAL_SHIFT; - if (offset >= BITS_PER_LONG) + if (xa_is_value(bitmap)) { + unsigned long v = xa_to_value(bitmap); + if (bit >= BITS_PER_XA_VALUE) goto err; + if (!(v & (1UL << bit))) + goto err; + v &= ~(1UL << bit); + if (!v) + goto delete; + xas_store(&xas, xa_mk_value(v)); } else { - btmp = bitmap->bitmap; - } - if (!test_bit(offset, btmp)) - goto err; - - __clear_bit(offset, btmp); - radix_tree_iter_tag_set(&ida->ida_rt, &iter, IDR_FREE); - if (radix_tree_exception(bitmap)) { - if (rcu_dereference_raw(*slot) == - (void *)RADIX_TREE_EXCEPTIONAL_ENTRY) - radix_tree_iter_delete(&ida->ida_rt, &iter, slot); - } else if (bitmap_empty(btmp, IDA_BITMAP_BITS)) { - kfree(bitmap); - radix_tree_iter_delete(&ida->ida_rt, &iter, slot); + if (!test_bit(bit, bitmap->bitmap)) + goto err; + __clear_bit(bit, bitmap->bitmap); + xas_set_mark(&xas, XA_FREE_MARK); + if (bitmap_empty(bitmap->bitmap, IDA_BITMAP_BITS)) { + kfree(bitmap); +delete: + xas_store(&xas, NULL); + } } + xas_unlock_irqrestore(&xas, flags); return; err: + xas_unlock_irqrestore(&xas, flags); WARN(1, "ida_free called for id=%d which is not allocated.\n", id); } +EXPORT_SYMBOL(ida_free); /** * ida_destroy() - Free all IDs. @@ -500,80 +533,60 @@ static void ida_remove(struct ida *ida, int id) */ void ida_destroy(struct ida *ida) { + XA_STATE(xas, &ida->xa, 0); + struct ida_bitmap *bitmap; unsigned long flags; - struct radix_tree_iter iter; - void __rcu **slot; - xa_lock_irqsave(&ida->ida_rt, flags); - radix_tree_for_each_slot(slot, &ida->ida_rt, &iter, 0) { - struct ida_bitmap *bitmap = rcu_dereference_raw(*slot); - if (!radix_tree_exception(bitmap)) + xas_lock_irqsave(&xas, flags); + xas_for_each(&xas, bitmap, ULONG_MAX) { + if (!xa_is_value(bitmap)) kfree(bitmap); - radix_tree_iter_delete(&ida->ida_rt, &iter, slot); + xas_store(&xas, NULL); } - xa_unlock_irqrestore(&ida->ida_rt, flags); + xas_unlock_irqrestore(&xas, flags); } EXPORT_SYMBOL(ida_destroy); -/** - * ida_alloc_range() - Allocate an unused ID. - * @ida: IDA handle. - * @min: Lowest ID to allocate. - * @max: Highest ID to allocate. - * @gfp: Memory allocation flags. - * - * Allocate an ID between @min and @max, inclusive. The allocated ID will - * not exceed %INT_MAX, even if @max is larger. - * - * Context: Any context. - * Return: The allocated ID, or %-ENOMEM if memory could not be allocated, - * or %-ENOSPC if there are no free IDs. - */ -int ida_alloc_range(struct ida *ida, unsigned int min, unsigned int max, - gfp_t gfp) -{ - int id = 0; - unsigned long flags; +#ifndef __KERNEL__ +extern void xa_dump_index(unsigned long index, unsigned int shift); +#define IDA_CHUNK_SHIFT ilog2(IDA_BITMAP_BITS) - if ((int)min < 0) - return -ENOSPC; - - if ((int)max < 0) - max = INT_MAX; - -again: - xa_lock_irqsave(&ida->ida_rt, flags); - id = ida_get_new_above(ida, min); - if (id > (int)max) { - ida_remove(ida, id); - id = -ENOSPC; - } - xa_unlock_irqrestore(&ida->ida_rt, flags); +static void ida_dump_entry(void *entry, unsigned long index) +{ + unsigned long i; + + if (!entry) + return; + + if (xa_is_node(entry)) { + struct xa_node *node = xa_to_node(entry); + unsigned int shift = node->shift + IDA_CHUNK_SHIFT + + XA_CHUNK_SHIFT; + + xa_dump_index(index * IDA_BITMAP_BITS, shift); + xa_dump_node(node); + for (i = 0; i < XA_CHUNK_SIZE; i++) + ida_dump_entry(node->slots[i], + index | (i << node->shift)); + } else if (xa_is_value(entry)) { + xa_dump_index(index * IDA_BITMAP_BITS, ilog2(BITS_PER_LONG)); + pr_cont("value: data %lx [%px]\n", xa_to_value(entry), entry); + } else { + struct ida_bitmap *bitmap = entry; - if (unlikely(id == -EAGAIN)) { - if (!ida_pre_get(ida, gfp)) - return -ENOMEM; - goto again; + xa_dump_index(index * IDA_BITMAP_BITS, IDA_CHUNK_SHIFT); + pr_cont("bitmap: %p data", bitmap); + for (i = 0; i < IDA_BITMAP_LONGS; i++) + pr_cont(" %lx", bitmap->bitmap[i]); + pr_cont("\n"); } - - return id; } -EXPORT_SYMBOL(ida_alloc_range); -/** - * ida_free() - Release an allocated ID. - * @ida: IDA handle. - * @id: Previously allocated ID. - * - * Context: Any context. - */ -void ida_free(struct ida *ida, unsigned int id) +static void ida_dump(struct ida *ida) { - unsigned long flags; - - BUG_ON((int)id < 0); - xa_lock_irqsave(&ida->ida_rt, flags); - ida_remove(ida, id); - xa_unlock_irqrestore(&ida->ida_rt, flags); + struct xarray *xa = &ida->xa; + pr_debug("ida: %p node %p free %d\n", ida, xa->xa_head, + xa->xa_flags >> ROOT_TAG_SHIFT); + ida_dump_entry(xa->xa_head, 0); } -EXPORT_SYMBOL(ida_free); +#endif diff --git a/lib/radix-tree.c b/lib/radix-tree.c @@ -38,15 +38,13 @@ #include <linux/rcupdate.h> #include <linux/slab.h> #include <linux/string.h> +#include <linux/xarray.h> -/* Number of nodes in fully populated tree of given height */ -static unsigned long height_to_maxnodes[RADIX_TREE_MAX_PATH + 1] __read_mostly; - /* * Radix tree node cache. */ -static struct kmem_cache *radix_tree_node_cachep; +struct kmem_cache *radix_tree_node_cachep; /* * The radix tree is variable-height, so an insert operation not only has @@ -98,24 +96,7 @@ static inline void *node_to_entry(void *ptr) return (void *)((unsigned long)ptr | RADIX_TREE_INTERNAL_NODE); } -#define RADIX_TREE_RETRY node_to_entry(NULL) - -#ifdef CONFIG_RADIX_TREE_MULTIORDER -/* Sibling slots point directly to another slot in the same node */ -static inline -bool is_sibling_entry(const struct radix_tree_node *parent, void *node) -{ - void __rcu **ptr = node; - return (parent->slots <= ptr) && - (ptr < parent->slots + RADIX_TREE_MAP_SIZE); -} -#else -static inline -bool is_sibling_entry(const struct radix_tree_node *parent, void *node) -{ - return false; -} -#endif +#define RADIX_TREE_RETRY XA_RETRY_ENTRY static inline unsigned long get_slot_offset(const struct radix_tree_node *parent, void __rcu **slot) @@ -129,24 +110,13 @@ static unsigned int radix_tree_descend(const struct radix_tree_node *parent, unsigned int offset = (index >> parent->shift) & RADIX_TREE_MAP_MASK; void __rcu **entry = rcu_dereference_raw(parent->slots[offset]); -#ifdef CONFIG_RADIX_TREE_MULTIORDER - if (radix_tree_is_internal_node(entry)) { - if (is_sibling_entry(parent, entry)) { - void __rcu **sibentry; - sibentry = (void __rcu **) entry_to_node(entry); - offset = get_slot_offset(parent, sibentry); - entry = rcu_dereference_raw(*sibentry); - } - } -#endif - *nodep = (void *)entry; return offset; } static inline gfp_t root_gfp_mask(const struct radix_tree_root *root) { - return root->gfp_mask & (__GFP_BITS_MASK & ~GFP_ZONEMASK); + return root->xa_flags & (__GFP_BITS_MASK & ~GFP_ZONEMASK); } static inline void tag_set(struct radix_tree_node *node, unsigned int tag, @@ -169,32 +139,32 @@ static inline int tag_get(const struct radix_tree_node *node, unsigned int tag, static inline void root_tag_set(struct radix_tree_root *root, unsigned tag) { - root->gfp_mask |= (__force gfp_t)(1 << (tag + ROOT_TAG_SHIFT)); + root->xa_flags |= (__force gfp_t)(1 << (tag + ROOT_TAG_SHIFT)); } static inline void root_tag_clear(struct radix_tree_root *root, unsigned tag) { - root->gfp_mask &= (__force gfp_t)~(1 << (tag + ROOT_TAG_SHIFT)); + root->xa_flags &= (__force gfp_t)~(1 << (tag + ROOT_TAG_SHIFT)); } static inline void root_tag_clear_all(struct radix_tree_root *root) { - root->gfp_mask &= (1 << ROOT_TAG_SHIFT) - 1; + root->xa_flags &= (__force gfp_t)((1 << ROOT_TAG_SHIFT) - 1); } static inline int root_tag_get(const struct radix_tree_root *root, unsigned tag) { - return (__force int)root->gfp_mask & (1 << (tag + ROOT_TAG_SHIFT)); + return (__force int)root->xa_flags & (1 << (tag + ROOT_TAG_SHIFT)); } static inline unsigned root_tags_get(const struct radix_tree_root *root) { - return (__force unsigned)root->gfp_mask >> ROOT_TAG_SHIFT; + return (__force unsigned)root->xa_flags >> ROOT_TAG_SHIFT; } static inline bool is_idr(const struct radix_tree_root *root) { - return !!(root->gfp_mask & ROOT_IS_IDR); + return !!(root->xa_flags & ROOT_IS_IDR); } /* @@ -254,7 +224,7 @@ radix_tree_find_next_bit(struct radix_tree_node *node, unsigned int tag, static unsigned int iter_offset(const struct radix_tree_iter *iter) { - return (iter->index >> iter_shift(iter)) & RADIX_TREE_MAP_MASK; + return iter->index & RADIX_TREE_MAP_MASK; } /* @@ -277,99 +247,6 @@ static unsigned long next_index(unsigned long index, return (index & ~node_maxindex(node)) + (offset << node->shift); } -#ifndef __KERNEL__ -static void dump_node(struct radix_tree_node *node, unsigned long index) -{ - unsigned long i; - - pr_debug("radix node: %p offset %d indices %lu-%lu parent %p tags %lx %lx %lx shift %d count %d exceptional %d\n", - node, node->offset, index, index | node_maxindex(node), - node->parent, - node->tags[0][0], node->tags[1][0], node->tags[2][0], - node->shift, node->count, node->exceptional); - - for (i = 0; i < RADIX_TREE_MAP_SIZE; i++) { - unsigned long first = index | (i << node->shift); - unsigned long last = first | ((1UL << node->shift) - 1); - void *entry = node->slots[i]; - if (!entry) - continue; - if (entry == RADIX_TREE_RETRY) { - pr_debug("radix retry offset %ld indices %lu-%lu parent %p\n", - i, first, last, node); - } else if (!radix_tree_is_internal_node(entry)) { - pr_debug("radix entry %p offset %ld indices %lu-%lu parent %p\n", - entry, i, first, last, node); - } else if (is_sibling_entry(node, entry)) { - pr_debug("radix sblng %p offset %ld indices %lu-%lu parent %p val %p\n", - entry, i, first, last, node, - *(void **)entry_to_node(entry)); - } else { - dump_node(entry_to_node(entry), first); - } - } -} - -/* For debug */ -static void radix_tree_dump(struct radix_tree_root *root) -{ - pr_debug("radix root: %p rnode %p tags %x\n", - root, root->rnode, - root->gfp_mask >> ROOT_TAG_SHIFT); - if (!radix_tree_is_internal_node(root->rnode)) - return; - dump_node(entry_to_node(root->rnode), 0); -} - -static void dump_ida_node(void *entry, unsigned long index) -{ - unsigned long i; - - if (!entry) - return; - - if (radix_tree_is_internal_node(entry)) { - struct radix_tree_node *node = entry_to_node(entry); - - pr_debug("ida node: %p offset %d indices %lu-%lu parent %p free %lx shift %d count %d\n", - node, node->offset, index * IDA_BITMAP_BITS, - ((index | node_maxindex(node)) + 1) * - IDA_BITMAP_BITS - 1, - node->parent, node->tags[0][0], node->shift, - node->count); - for (i = 0; i < RADIX_TREE_MAP_SIZE; i++) - dump_ida_node(node->slots[i], - index | (i << node->shift)); - } else if (radix_tree_exceptional_entry(entry)) { - pr_debug("ida excp: %p offset %d indices %lu-%lu data %lx\n", - entry, (int)(index & RADIX_TREE_MAP_MASK), - index * IDA_BITMAP_BITS, - index * IDA_BITMAP_BITS + BITS_PER_LONG - - RADIX_TREE_EXCEPTIONAL_SHIFT, - (unsigned long)entry >> - RADIX_TREE_EXCEPTIONAL_SHIFT); - } else { - struct ida_bitmap *bitmap = entry; - - pr_debug("ida btmp: %p offset %d indices %lu-%lu data", bitmap, - (int)(index & RADIX_TREE_MAP_MASK), - index * IDA_BITMAP_BITS, - (index + 1) * IDA_BITMAP_BITS - 1); - for (i = 0; i < IDA_BITMAP_LONGS; i++) - pr_cont(" %lx", bitmap->bitmap[i]); - pr_cont("\n"); - } -} - -static void ida_dump(struct ida *ida) -{ - struct radix_tree_root *root = &ida->ida_rt; - pr_debug("ida: %p node %p free %d\n", ida, root->rnode, - root->gfp_mask >> ROOT_TAG_SHIFT); - dump_ida_node(root->rnode, 0); -} -#endif - /* * This assumes that the caller has performed appropriate preallocation, and * that the caller has pinned this thread of control to the current CPU. @@ -378,7 +255,7 @@ static struct radix_tree_node * radix_tree_node_alloc(gfp_t gfp_mask, struct radix_tree_node *parent, struct radix_tree_root *root, unsigned int shift, unsigned int offset, - unsigned int count, unsigned int exceptional) + unsigned int count, unsigned int nr_values) { struct radix_tree_node *ret = NULL; @@ -425,14 +302,14 @@ out: ret->shift = shift; ret->offset = offset; ret->count = count; - ret->exceptional = exceptional; + ret->nr_values = nr_values; ret->parent = parent; - ret->root = root; + ret->array = root; } return ret; } -static void radix_tree_node_rcu_free(struct rcu_head *head) +void radix_tree_node_rcu_free(struct rcu_head *head) { struct radix_tree_node *node = container_of(head, struct radix_tree_node, rcu_head); @@ -530,77 +407,10 @@ int radix_tree_maybe_preload(gfp_t gfp_mask) } EXPORT_SYMBOL(radix_tree_maybe_preload); -#ifdef CONFIG_RADIX_TREE_MULTIORDER -/* - * Preload with enough objects to ensure that we can split a single entry - * of order @old_order into many entries of size @new_order - */ -int radix_tree_split_preload(unsigned int old_order, unsigned int new_order, - gfp_t gfp_mask) -{ - unsigned top = 1 << (old_order % RADIX_TREE_MAP_SHIFT); - unsigned layers = (old_order / RADIX_TREE_MAP_SHIFT) - - (new_order / RADIX_TREE_MAP_SHIFT); - unsigned nr = 0; - - WARN_ON_ONCE(!gfpflags_allow_blocking(gfp_mask)); - BUG_ON(new_order >= old_order); - - while (layers--) - nr = nr * RADIX_TREE_MAP_SIZE + 1; - return __radix_tree_preload(gfp_mask, top * nr); -} -#endif - -/* - * The same as function above, but preload number of nodes required to insert - * (1 << order) continuous naturally-aligned elements. - */ -int radix_tree_maybe_preload_order(gfp_t gfp_mask, int order) -{ - unsigned long nr_subtrees; - int nr_nodes, subtree_height; - - /* Preloading doesn't help anything with this gfp mask, skip it */ - if (!gfpflags_allow_blocking(gfp_mask)) { - preempt_disable(); - return 0; - } - - /* - * Calculate number and height of fully populated subtrees it takes to - * store (1 << order) elements. - */ - nr_subtrees = 1 << order; - for (subtree_height = 0; nr_subtrees > RADIX_TREE_MAP_SIZE; - subtree_height++) - nr_subtrees >>= RADIX_TREE_MAP_SHIFT; - - /* - * The worst case is zero height tree with a single item at index 0 and - * then inserting items starting at ULONG_MAX - (1 << order). - * - * This requires RADIX_TREE_MAX_PATH nodes to build branch from root to - * 0-index item. - */ - nr_nodes = RADIX_TREE_MAX_PATH; - - /* Plus branch to fully populated subtrees. */ - nr_nodes += RADIX_TREE_MAX_PATH - subtree_height; - - /* Root node is shared. */ - nr_nodes--; - - /* Plus nodes required to build subtrees. */ - nr_nodes += nr_subtrees * height_to_maxnodes[subtree_height]; - - return __radix_tree_preload(gfp_mask, nr_nodes); -} - static unsigned radix_tree_load_root(const struct radix_tree_root *root, struct radix_tree_node **nodep, unsigned long *maxindex) { - struct radix_tree_node *node = rcu_dereference_raw(root->rnode); + struct radix_tree_node *node = rcu_dereference_raw(root->xa_head); *nodep = node; @@ -629,7 +439,7 @@ static int radix_tree_extend(struct radix_tree_root *root, gfp_t gfp, while (index > shift_maxindex(maxshift)) maxshift += RADIX_TREE_MAP_SHIFT; - entry = rcu_dereference_raw(root->rnode); + entry = rcu_dereference_raw(root->xa_head); if (!entry && (!is_idr(root) || root_tag_get(root, IDR_FREE))) goto out; @@ -656,9 +466,9 @@ static int radix_tree_extend(struct radix_tree_root *root, gfp_t gfp, BUG_ON(shift > BITS_PER_LONG); if (radix_tree_is_internal_node(entry)) { entry_to_node(entry)->parent = node; - } else if (radix_tree_exceptional_entry(entry)) { - /* Moving an exceptional root->rnode to a node */ - node->exceptional = 1; + } else if (xa_is_value(entry)) { + /* Moving a value entry root->xa_head to a node */ + node->nr_values = 1; } /* * entry was already in the radix tree, so we do not need @@ -666,7 +476,7 @@ static int radix_tree_extend(struct radix_tree_root *root, gfp_t gfp, */ node->slots[0] = (void __rcu *)entry; entry = node_to_entry(node); - rcu_assign_pointer(root->rnode, entry); + rcu_assign_pointer(root->xa_head, entry); shift += RADIX_TREE_MAP_SHIFT; } while (shift <= maxshift); out: @@ -677,13 +487,12 @@ out: * radix_tree_shrink - shrink radix tree to minimum height * @root radix tree root */ -static inline bool radix_tree_shrink(struct radix_tree_root *root, - radix_tree_update_node_t update_node) +static inline bool radix_tree_shrink(struct radix_tree_root *root) { bool shrunk = false; for (;;) { - struct radix_tree_node *node = rcu_dereference_raw(root->rnode); + struct radix_tree_node *node = rcu_dereference_raw(root->xa_head); struct radix_tree_node *child; if (!radix_tree_is_internal_node(node)) @@ -692,15 +501,20 @@ static inline bool radix_tree_shrink(struct radix_tree_root *root, /* * The candidate node has more than one child, or its child - * is not at the leftmost slot, or the child is a multiorder - * entry, we cannot shrink. + * is not at the leftmost slot, we cannot shrink. */ if (node->count != 1) break; child = rcu_dereference_raw(node->slots[0]); if (!child) break; - if (!radix_tree_is_internal_node(child) && node->shift) + + /* + * For an IDR, we must not shrink entry 0 into the root in + * case somebody calls idr_replace() with a pointer that + * appears to be an internal entry + */ + if (!node->shift && is_idr(root)) break; if (radix_tree_is_internal_node(child)) @@ -711,9 +525,9 @@ static inline bool radix_tree_shrink(struct radix_tree_root *root, * moving the node from one part of the tree to another: if it * was safe to dereference the old pointer to it * (node->slots[0]), it will be safe to dereference the new - * one (root->rnode) as far as dependent read barriers go. + * one (root->xa_head) as far as dependent read barriers go. */ - root->rnode = (void __rcu *)child; + root->xa_head = (void __rcu *)child; if (is_idr(root) && !tag_get(node, IDR_FREE, 0)) root_tag_clear(root, IDR_FREE); @@ -738,8 +552,6 @@ static inline bool radix_tree_shrink(struct radix_tree_root *root, node->count = 0; if (!radix_tree_is_internal_node(child)) { node->slots[0] = (void __rcu *)RADIX_TREE_RETRY; - if (update_node) - update_node(node); } WARN_ON_ONCE(!list_empty(&node->private_list)); @@ -751,8 +563,7 @@ static inline bool radix_tree_shrink(struct radix_tree_root *root, } static bool delete_node(struct radix_tree_root *root, - struct radix_tree_node *node, - radix_tree_update_node_t update_node) + struct radix_tree_node *node) { bool deleted = false; @@ -761,9 +572,8 @@ static bool delete_node(struct radix_tree_root *root, if (node->count) { if (node_to_entry(node) == - rcu_dereference_raw(root->rnode)) - deleted |= radix_tree_shrink(root, - update_node); + rcu_dereference_raw(root->xa_head)) + deleted |= radix_tree_shrink(root); return deleted; } @@ -778,7 +588,7 @@ static bool delete_node(struct radix_tree_root *root, */ if (!is_idr(root)) root_tag_clear_all(root); - root->rnode = NULL; + root->xa_head = NULL; } WARN_ON_ONCE(!list_empty(&node->private_list)); @@ -795,7 +605,6 @@ static bool delete_node(struct radix_tree_root *root, * __radix_tree_create - create a slot in a radix tree * @root: radix tree root * @index: index key - * @order: index occupies 2^order aligned slots * @nodep: returns node * @slotp: returns slot * @@ -803,36 +612,34 @@ static bool delete_node(struct radix_tree_root *root, * at position @index in the radix tree @root. * * Until there is more than one item in the tree, no nodes are - * allocated and @root->rnode is used as a direct slot instead of + * allocated and @root->xa_head is used as a direct slot instead of * pointing to a node, in which case *@nodep will be NULL. * * Returns -ENOMEM, or 0 for success. */ -int __radix_tree_create(struct radix_tree_root *root, unsigned long index, - unsigned order, struct radix_tree_node **nodep, - void __rcu ***slotp) +static int __radix_tree_create(struct radix_tree_root *root, + unsigned long index, struct radix_tree_node **nodep, + void __rcu ***slotp) { struct radix_tree_node *node = NULL, *child; - void __rcu **slot = (void __rcu **)&root->rnode; + void __rcu **slot = (void __rcu **)&root->xa_head; unsigned long maxindex; unsigned int shift, offset = 0; - unsigned long max = index | ((1UL << order) - 1); + unsigned long max = index; gfp_t gfp = root_gfp_mask(root); shift = radix_tree_load_root(root, &child, &maxindex); /* Make sure the tree is high enough. */ - if (order > 0 && max == ((1UL << order) - 1)) - max++; if (max > maxindex) { int error = radix_tree_extend(root, gfp, max, shift); if (error < 0) return error; shift = error; - child = rcu_dereference_raw(root->rnode); + child = rcu_dereference_raw(root->xa_head); } - while (shift > order) { + while (shift > 0) { shift -= RADIX_TREE_MAP_SHIFT; if (child == NULL) { /* Have to add a child node. */ @@ -875,8 +682,7 @@ static void radix_tree_free_nodes(struct radix_tree_node *node) for (;;) { void *entry = rcu_dereference_raw(child->slots[offset]); - if (radix_tree_is_internal_node(entry) && - !is_sibling_entry(child, entry)) { + if (xa_is_node(entry) && child->shift) { child = entry_to_node(entry); offset = 0; continue; @@ -894,96 +700,30 @@ static void radix_tree_free_nodes(struct radix_tree_node *node) } } -#ifdef CONFIG_RADIX_TREE_MULTIORDER static inline int insert_entries(struct radix_tree_node *node, - void __rcu **slot, void *item, unsigned order, bool replace) -{ - struct radix_tree_node *child; - unsigned i, n, tag, offset, tags = 0; - - if (node) { - if (order > node->shift) - n = 1 << (order - node->shift); - else - n = 1; - offset = get_slot_offset(node, slot); - } else { - n = 1; - offset = 0; - } - - if (n > 1) { - offset = offset & ~(n - 1); - slot = &node->slots[offset]; - } - child = node_to_entry(slot); - - for (i = 0; i < n; i++) { - if (slot[i]) { - if (replace) { - node->count--; - for (tag = 0; tag < RADIX_TREE_MAX_TAGS; tag++) - if (tag_get(node, tag, offset + i)) - tags |= 1 << tag; - } else - return -EEXIST; - } - } - - for (i = 0; i < n; i++) { - struct radix_tree_node *old = rcu_dereference_raw(slot[i]); - if (i) { - rcu_assign_pointer(slot[i], child); - for (tag = 0; tag < RADIX_TREE_MAX_TAGS; tag++) - if (tags & (1 << tag)) - tag_clear(node, tag, offset + i); - } else { - rcu_assign_pointer(slot[i], item); - for (tag = 0; tag < RADIX_TREE_MAX_TAGS; tag++) - if (tags & (1 << tag)) - tag_set(node, tag, offset); - } - if (radix_tree_is_internal_node(old) && - !is_sibling_entry(node, old) && - (old != RADIX_TREE_RETRY)) - radix_tree_free_nodes(old); - if (radix_tree_exceptional_entry(old)) - node->exceptional--; - } - if (node) { - node->count += n; - if (radix_tree_exceptional_entry(item)) - node->exceptional += n; - } - return n; -} -#else -static inline int insert_entries(struct radix_tree_node *node, - void __rcu **slot, void *item, unsigned order, bool replace) + void __rcu **slot, void *item, bool replace) { if (*slot) return -EEXIST; rcu_assign_pointer(*slot, item); if (node) { node->count++; - if (radix_tree_exceptional_entry(item)) - node->exceptional++; + if (xa_is_value(item)) + node->nr_values++; } return 1; } -#endif /** * __radix_tree_insert - insert into a radix tree * @root: radix tree root * @index: index key - * @order: key covers the 2^order indices around index * @item: item to insert * * Insert an item into the radix tree at position @index. */ -int __radix_tree_insert(struct radix_tree_root *root, unsigned long index, - unsigned order, void *item) +int radix_tree_insert(struct radix_tree_root *root, unsigned long index, + void *item) { struct radix_tree_node *node; void __rcu **slot; @@ -991,11 +731,11 @@ int __radix_tree_insert(struct radix_tree_root *root, unsigned long index, BUG_ON(radix_tree_is_internal_node(item)); - error = __radix_tree_create(root, index, order, &node, &slot); + error = __radix_tree_create(root, index, &node, &slot); if (error) return error; - error = insert_entries(node, slot, item, order, false); + error = insert_entries(node, slot, item, false); if (error < 0) return error; @@ -1010,7 +750,7 @@ int __radix_tree_insert(struct radix_tree_root *root, unsigned long index, return 0; } -EXPORT_SYMBOL(__radix_tree_insert); +EXPORT_SYMBOL(radix_tree_insert); /** * __radix_tree_lookup - lookup an item in a radix tree @@ -1023,7 +763,7 @@ EXPORT_SYMBOL(__radix_tree_insert); * tree @root. * * Until there is more than one item in the tree, no nodes are - * allocated and @root->rnode is used as a direct slot instead of + * allocated and @root->xa_head is used as a direct slot instead of * pointing to a node, in which case *@nodep will be NULL. */ void *__radix_tree_lookup(const struct radix_tree_root *root, @@ -1036,7 +776,7 @@ void *__radix_tree_lookup(const struct radix_tree_root *root, restart: parent = NULL; - slot = (void __rcu **)&root->rnode; + slot = (void __rcu **)&root->xa_head; radix_tree_load_root(root, &node, &maxindex); if (index > maxindex) return NULL; @@ -1049,6 +789,8 @@ void *__radix_tree_lookup(const struct radix_tree_root *root, parent = entry_to_node(node); offset = radix_tree_descend(parent, &node, index); slot = parent->slots + offset; + if (parent->shift == 0) + break; } if (nodep) @@ -1100,36 +842,12 @@ void *radix_tree_lookup(const struct radix_tree_root *root, unsigned long index) } EXPORT_SYMBOL(radix_tree_lookup); -static inline void replace_sibling_entries(struct radix_tree_node *node, - void __rcu **slot, int count, int exceptional) -{ -#ifdef CONFIG_RADIX_TREE_MULTIORDER - void *ptr = node_to_entry(slot); - unsigned offset = get_slot_offset(node, slot) + 1; - - while (offset < RADIX_TREE_MAP_SIZE) { - if (rcu_dereference_raw(node->slots[offset]) != ptr) - break; - if (count < 0) { - node->slots[offset] = NULL; - node->count--; - } - node->exceptional += exceptional; - offset++; - } -#endif -} - static void replace_slot(void __rcu **slot, void *item, - struct radix_tree_node *node, int count, int exceptional) + struct radix_tree_node *node, int count, int values) { - if (WARN_ON_ONCE(radix_tree_is_internal_node(item))) - return; - - if (node && (count || exceptional)) { + if (node && (count || values)) { node->count += count; - node->exceptional += exceptional; - replace_sibling_entries(node, slot, count, exceptional); + node->nr_values += values; } rcu_assign_pointer(*slot, item); @@ -1172,37 +890,31 @@ static int calculate_count(struct radix_tree_root *root, * @node: pointer to tree node * @slot: pointer to slot in @node * @item: new item to store in the slot. - * @update_node: callback for changing leaf nodes * * For use with __radix_tree_lookup(). Caller must hold tree write locked * across slot lookup and replacement. */ void __radix_tree_replace(struct radix_tree_root *root, struct radix_tree_node *node, - void __rcu **slot, void *item, - radix_tree_update_node_t update_node) + void __rcu **slot, void *item) { void *old = rcu_dereference_raw(*slot); - int exceptional = !!radix_tree_exceptional_entry(item) - - !!radix_tree_exceptional_entry(old); + int values = !!xa_is_value(item) - !!xa_is_value(old); int count = calculate_count(root, node, slot, item, old); /* - * This function supports replacing exceptional entries and + * This function supports replacing value entries and * deleting entries, but that needs accounting against the - * node unless the slot is root->rnode. + * node unless the slot is root->xa_head. */ - WARN_ON_ONCE(!node && (slot != (void __rcu **)&root->rnode) && - (count || exceptional)); - replace_slot(slot, item, node, count, exceptional); + WARN_ON_ONCE(!node && (slot != (void __rcu **)&root->xa_head) && + (count || values)); + replace_slot(slot, item, node, count, values); if (!node) return; - if (update_node) - update_node(node); - - delete_node(root, node, update_node); + delete_node(root, node); } /** @@ -1211,12 +923,12 @@ void __radix_tree_replace(struct radix_tree_root *root, * @slot: pointer to slot * @item: new item to store in the slot. * - * For use with radix_tree_lookup_slot(), radix_tree_gang_lookup_slot(), + * For use with radix_tree_lookup_slot() and * radix_tree_gang_lookup_tag_slot(). Caller must hold tree write locked * across slot lookup and replacement. * * NOTE: This cannot be used to switch between non-entries (empty slots), - * regular entries, and exceptional entries, as that requires accounting + * regular entries, and value entries, as that requires accounting * inside the radix tree node. When switching from one type of entry or * deleting, use __radix_tree_lookup() and __radix_tree_replace() or * radix_tree_iter_replace(). @@ -1224,7 +936,7 @@ void __radix_tree_replace(struct radix_tree_root *root, void radix_tree_replace_slot(struct radix_tree_root *root, void __rcu **slot, void *item) { - __radix_tree_replace(root, NULL, slot, item, NULL); + __radix_tree_replace(root, NULL, slot, item); } EXPORT_SYMBOL(radix_tree_replace_slot); @@ -1234,162 +946,16 @@ EXPORT_SYMBOL(radix_tree_replace_slot); * @slot: pointer to slot * @item: new item to store in the slot. * - * For use with radix_tree_split() and radix_tree_for_each_slot(). - * Caller must hold tree write locked across split and replacement. + * For use with radix_tree_for_each_slot(). + * Caller must hold tree write locked. */ void radix_tree_iter_replace(struct radix_tree_root *root, const struct radix_tree_iter *iter, void __rcu **slot, void *item) { - __radix_tree_replace(root, iter->node, slot, item, NULL); + __radix_tree_replace(root, iter->node, slot, item); } -#ifdef CONFIG_RADIX_TREE_MULTIORDER -/** - * radix_tree_join - replace multiple entries with one multiorder entry - * @root: radix tree root - * @index: an index inside the new entry - * @order: order of the new entry - * @item: new entry - * - * Call this function to replace several entries with one larger entry. - * The existing entries are presumed to not need freeing as a result of - * this call. - * - * The replacement entry will have all the tags set on it that were set - * on any of the entries it is replacing. - */ -int radix_tree_join(struct radix_tree_root *root, unsigned long index, - unsigned order, void *item) -{ - struct radix_tree_node *node; - void __rcu **slot; - int error; - - BUG_ON(radix_tree_is_internal_node(item)); - - error = __radix_tree_create(root, index, order, &node, &slot); - if (!error) - error = insert_entries(node, slot, item, order, true); - if (error > 0) - error = 0; - - return error; -} - -/** - * radix_tree_split - Split an entry into smaller entries - * @root: radix tree root - * @index: An index within the large entry - * @order: Order of new entries - * - * Call this function as the first step in replacing a multiorder entry - * with several entries of lower order. After this function returns, - * loop over the relevant portion of the tree using radix_tree_for_each_slot() - * and call radix_tree_iter_replace() to set up each new entry. - * - * The tags from this entry are replicated to all the new entries. - * - * The radix tree should be locked against modification during the entire - * replacement operation. Lock-free lookups will see RADIX_TREE_RETRY which - * should prompt RCU walkers to restart the lookup from the root. - */ -int radix_tree_split(struct radix_tree_root *root, unsigned long index, - unsigned order) -{ - struct radix_tree_node *parent, *node, *child; - void __rcu **slot; - unsigned int offset, end; - unsigned n, tag, tags = 0; - gfp_t gfp = root_gfp_mask(root); - - if (!__radix_tree_lookup(root, index, &parent, &slot)) - return -ENOENT; - if (!parent) - return -ENOENT; - - offset = get_slot_offset(parent, slot); - - for (tag = 0; tag < RADIX_TREE_MAX_TAGS; tag++) - if (tag_get(parent, tag, offset)) - tags |= 1 << tag; - - for (end = offset + 1; end < RADIX_TREE_MAP_SIZE; end++) { - if (!is_sibling_entry(parent, - rcu_dereference_raw(parent->slots[end]))) - break; - for (tag = 0; tag < RADIX_TREE_MAX_TAGS; tag++) - if (tags & (1 << tag)) - tag_set(parent, tag, end); - /* rcu_assign_pointer ensures tags are set before RETRY */ - rcu_assign_pointer(parent->slots[end], RADIX_TREE_RETRY); - } - rcu_assign_pointer(parent->slots[offset], RADIX_TREE_RETRY); - parent->exceptional -= (end - offset); - - if (order == parent->shift) - return 0; - if (order > parent->shift) { - while (offset < end) - offset += insert_entries(parent, &parent->slots[offset], - RADIX_TREE_RETRY, order, true); - return 0; - } - - node = parent; - - for (;;) { - if (node->shift > order) { - child = radix_tree_node_alloc(gfp, node, root, - node->shift - RADIX_TREE_MAP_SHIFT, - offset, 0, 0); - if (!child) - goto nomem; - if (node != parent) { - node->count++; - rcu_assign_pointer(node->slots[offset], - node_to_entry(child)); - for (tag = 0; tag < RADIX_TREE_MAX_TAGS; tag++) - if (tags & (1 << tag)) - tag_set(node, tag, offset); - } - - node = child; - offset = 0; - continue; - } - - n = insert_entries(node, &node->slots[offset], - RADIX_TREE_RETRY, order, false); - BUG_ON(n > RADIX_TREE_MAP_SIZE); - - for (tag = 0; tag < RADIX_TREE_MAX_TAGS; tag++) - if (tags & (1 << tag)) - tag_set(node, tag, offset); - offset += n; - - while (offset == RADIX_TREE_MAP_SIZE) { - if (node == parent) - break; - offset = node->offset; - child = node; - node = node->parent; - rcu_assign_pointer(node->slots[offset], - node_to_entry(child)); - offset++; - } - if ((node == parent) && (offset == end)) - return 0; - } - - nomem: - /* Shouldn't happen; did user forget to preload? */ - /* TODO: free all the allocated nodes */ - WARN_ON(1); - return -ENOMEM; -} -#endif - static void node_tag_set(struct radix_tree_root *root, struct radix_tree_node *node, unsigned int tag, unsigned int offset) @@ -1447,18 +1013,6 @@ void *radix_tree_tag_set(struct radix_tree_root *root, } EXPORT_SYMBOL(radix_tree_tag_set); -/** - * radix_tree_iter_tag_set - set a tag on the current iterator entry - * @root: radix tree root - * @iter: iterator state - * @tag: tag to set - */ -void radix_tree_iter_tag_set(struct radix_tree_root *root, - const struct radix_tree_iter *iter, unsigned int tag) -{ - node_tag_set(root, iter->node, tag, iter_offset(iter)); -} - static void node_tag_clear(struct radix_tree_root *root, struct radix_tree_node *node, unsigned int tag, unsigned int offset) @@ -1574,14 +1128,6 @@ int radix_tree_tag_get(const struct radix_tree_root *root, } EXPORT_SYMBOL(radix_tree_tag_get); -static inline void __set_iter_shift(struct radix_tree_iter *iter, - unsigned int shift) -{ -#ifdef CONFIG_RADIX_TREE_MULTIORDER - iter->shift = shift; -#endif -} - /* Construct iter->tags bit-mask from node->tags[tag] array */ static void set_iter_tags(struct radix_tree_iter *iter, struct radix_tree_node *node, unsigned offset, @@ -1608,92 +1154,11 @@ static void set_iter_tags(struct radix_tree_iter *iter, } } -#ifdef CONFIG_RADIX_TREE_MULTIORDER -static void __rcu **skip_siblings(struct radix_tree_node **nodep, - void __rcu **slot, struct radix_tree_iter *iter) -{ - while (iter->index < iter->next_index) { - *nodep = rcu_dereference_raw(*slot); - if (*nodep && !is_sibling_entry(iter->node, *nodep)) - return slot; - slot++; - iter->index = __radix_tree_iter_add(iter, 1); - iter->tags >>= 1; - } - - *nodep = NULL; - return NULL; -} - -void __rcu **__radix_tree_next_slot(void __rcu **slot, - struct radix_tree_iter *iter, unsigned flags) -{ - unsigned tag = flags & RADIX_TREE_ITER_TAG_MASK; - struct radix_tree_node *node; - - slot = skip_siblings(&node, slot, iter); - - while (radix_tree_is_internal_node(node)) { - unsigned offset; - unsigned long next_index; - - if (node == RADIX_TREE_RETRY) - return slot; - node = entry_to_node(node); - iter->node = node; - iter->shift = node->shift; - - if (flags & RADIX_TREE_ITER_TAGGED) { - offset = radix_tree_find_next_bit(node, tag, 0); - if (offset == RADIX_TREE_MAP_SIZE) - return NULL; - slot = &node->slots[offset]; - iter->index = __radix_tree_iter_add(iter, offset); - set_iter_tags(iter, node, offset, tag); - node = rcu_dereference_raw(*slot); - } else { - offset = 0; - slot = &node->slots[0]; - for (;;) { - node = rcu_dereference_raw(*slot); - if (node) - break; - slot++; - offset++; - if (offset == RADIX_TREE_MAP_SIZE) - return NULL; - } - iter->index = __radix_tree_iter_add(iter, offset); - } - if ((flags & RADIX_TREE_ITER_CONTIG) && (offset > 0)) - goto none; - next_index = (iter->index | shift_maxindex(iter->shift)) + 1; - if (next_index < iter->next_index) - iter->next_index = next_index; - } - - return slot; - none: - iter->next_index = 0; - return NULL; -} -EXPORT_SYMBOL(__radix_tree_next_slot); -#else -static void __rcu **skip_siblings(struct radix_tree_node **nodep, - void __rcu **slot, struct radix_tree_iter *iter) -{ - return slot; -} -#endif - void __rcu **radix_tree_iter_resume(void __rcu **slot, struct radix_tree_iter *iter) { - struct radix_tree_node *node; - slot++; iter->index = __radix_tree_iter_add(iter, 1); - skip_siblings(&node, slot, iter); iter->next_index = iter->index; iter->tags = 0; return NULL; @@ -1744,8 +1209,7 @@ void __rcu **radix_tree_next_chunk(const struct radix_tree_root *root, iter->next_index = maxindex + 1; iter->tags = 1; iter->node = NULL; - __set_iter_shift(iter, 0); - return (void __rcu **)&root->rnode; + return (void __rcu **)&root->xa_head; } do { @@ -1765,8 +1229,6 @@ void __rcu **radix_tree_next_chunk(const struct radix_tree_root *root, while (++offset < RADIX_TREE_MAP_SIZE) { void *slot = rcu_dereference_raw( node->slots[offset]); - if (is_sibling_entry(node, slot)) - continue; if (slot) break; } @@ -1784,13 +1246,12 @@ void __rcu **radix_tree_next_chunk(const struct radix_tree_root *root, goto restart; if (child == RADIX_TREE_RETRY) break; - } while (radix_tree_is_internal_node(child)); + } while (node->shift && radix_tree_is_internal_node(child)); /* Update the iterator state */ - iter->index = (index &~ node_maxindex(node)) | (offset << node->shift); + iter->index = (index &~ node_maxindex(node)) | offset; iter->next_index = (index | node_maxindex(node)) + 1; iter->node = node; - __set_iter_shift(iter, node->shift); if (flags & RADIX_TREE_ITER_TAGGED) set_iter_tags(iter, node, offset, tag); @@ -1847,48 +1308,6 @@ radix_tree_gang_lookup(const struct radix_tree_root *root, void **results, EXPORT_SYMBOL(radix_tree_gang_lookup); /** - * radix_tree_gang_lookup_slot - perform multiple slot lookup on radix tree - * @root: radix tree root - * @results: where the results of the lookup are placed - * @indices: where their indices should be placed (but usually NULL) - * @first_index: start the lookup from this key - * @max_items: place up to this many items at *results - * - * Performs an index-ascending scan of the tree for present items. Places - * their slots at *@results and returns the number of items which were - * placed at *@results. - * - * The implementation is naive. - * - * Like radix_tree_gang_lookup as far as RCU and locking goes. Slots must - * be dereferenced with radix_tree_deref_slot, and if using only RCU - * protection, radix_tree_deref_slot may fail requiring a retry. - */ -unsigned int -radix_tree_gang_lookup_slot(const struct radix_tree_root *root, - void __rcu ***results, unsigned long *indices, - unsigned long first_index, unsigned int max_items) -{ - struct radix_tree_iter iter; - void __rcu **slot; - unsigned int ret = 0; - - if (unlikely(!max_items)) - return 0; - - radix_tree_for_each_slot(slot, root, &iter, first_index) { - results[ret] = slot; - if (indices) - indices[ret] = iter.index; - if (++ret == max_items) - break; - } - - return ret; -} -EXPORT_SYMBOL(radix_tree_gang_lookup_slot); - -/** * radix_tree_gang_lookup_tag - perform multiple lookup on a radix tree * based on a tag * @root: radix tree root @@ -1964,28 +1383,11 @@ radix_tree_gang_lookup_tag_slot(const struct radix_tree_root *root, } EXPORT_SYMBOL(radix_tree_gang_lookup_tag_slot); -/** - * __radix_tree_delete_node - try to free node after clearing a slot - * @root: radix tree root - * @node: node containing @index - * @update_node: callback for changing leaf nodes - * - * After clearing the slot at @index in @node from radix tree - * rooted at @root, call this function to attempt freeing the - * node and shrinking the tree. - */ -void __radix_tree_delete_node(struct radix_tree_root *root, - struct radix_tree_node *node, - radix_tree_update_node_t update_node) -{ - delete_node(root, node, update_node); -} - static bool __radix_tree_delete(struct radix_tree_root *root, struct radix_tree_node *node, void __rcu **slot) { void *old = rcu_dereference_raw(*slot); - int exceptional = radix_tree_exceptional_entry(old) ? -1 : 0; + int values = xa_is_value(old) ? -1 : 0; unsigned offset = get_slot_offset(node, slot); int tag; @@ -1995,8 +1397,8 @@ static bool __radix_tree_delete(struct radix_tree_root *root, for (tag = 0; tag < RADIX_TREE_MAX_TAGS; tag++) node_tag_clear(root, node, tag, offset); - replace_slot(slot, NULL, node, -1, exceptional); - return node && delete_node(root, node, NULL); + replace_slot(slot, NULL, node, -1, values); + return node && delete_node(root, node); } /** @@ -2068,19 +1470,6 @@ void *radix_tree_delete(struct radix_tree_root *root, unsigned long index) } EXPORT_SYMBOL(radix_tree_delete); -void radix_tree_clear_tags(struct radix_tree_root *root, - struct radix_tree_node *node, - void __rcu **slot) -{ - if (node) { - unsigned int tag, offset = get_slot_offset(node, slot); - for (tag = 0; tag < RADIX_TREE_MAX_TAGS; tag++) - node_tag_clear(root, node, tag, offset); - } else { - root_tag_clear_all(root); - } -} - /** * radix_tree_tagged - test whether any items in the tree are tagged * @root: radix tree root @@ -2106,33 +1495,12 @@ void idr_preload(gfp_t gfp_mask) } EXPORT_SYMBOL(idr_preload); -int ida_pre_get(struct ida *ida, gfp_t gfp) -{ - /* - * The IDA API has no preload_end() equivalent. Instead, - * ida_get_new() can return -EAGAIN, prompting the caller - * to return to the ida_pre_get() step. - */ - if (!__radix_tree_preload(gfp, IDA_PRELOAD_SIZE)) - preempt_enable(); - - if (!this_cpu_read(ida_bitmap)) { - struct ida_bitmap *bitmap = kzalloc(sizeof(*bitmap), gfp); - if (!bitmap) - return 0; - if (this_cpu_cmpxchg(ida_bitmap, NULL, bitmap)) - kfree(bitmap); - } - - return 1; -} - void __rcu **idr_get_free(struct radix_tree_root *root, struct radix_tree_iter *iter, gfp_t gfp, unsigned long max) { struct radix_tree_node *node = NULL, *child; - void __rcu **slot = (void __rcu **)&root->rnode; + void __rcu **slot = (void __rcu **)&root->xa_head; unsigned long maxindex, start = iter->next_index; unsigned int shift, offset = 0; @@ -2148,8 +1516,10 @@ void __rcu **idr_get_free(struct radix_tree_root *root, if (error < 0) return ERR_PTR(error); shift = error; - child = rcu_dereference_raw(root->rnode); + child = rcu_dereference_raw(root->xa_head); } + if (start == 0 && shift == 0) + shift = RADIX_TREE_MAP_SHIFT; while (shift) { shift -= RADIX_TREE_MAP_SHIFT; @@ -2192,7 +1562,6 @@ void __rcu **idr_get_free(struct radix_tree_root *root, else iter->next_index = 1; iter->node = node; - __set_iter_shift(iter, shift); set_iter_tags(iter, node, offset, IDR_FREE); return slot; @@ -2211,10 +1580,10 @@ void __rcu **idr_get_free(struct radix_tree_root *root, */ void idr_destroy(struct idr *idr) { - struct radix_tree_node *node = rcu_dereference_raw(idr->idr_rt.rnode); + struct radix_tree_node *node = rcu_dereference_raw(idr->idr_rt.xa_head); if (radix_tree_is_internal_node(node)) radix_tree_free_nodes(node); - idr->idr_rt.rnode = NULL; + idr->idr_rt.xa_head = NULL; root_tag_set(&idr->idr_rt, IDR_FREE); } EXPORT_SYMBOL(idr_destroy); @@ -2228,31 +1597,6 @@ radix_tree_node_ctor(void *arg) INIT_LIST_HEAD(&node->private_list); } -static __init unsigned long __maxindex(unsigned int height) -{ - unsigned int width = height * RADIX_TREE_MAP_SHIFT; - int shift = RADIX_TREE_INDEX_BITS - width; - - if (shift < 0) - return ~0UL; - if (shift >= BITS_PER_LONG) - return 0UL; - return ~0UL >> shift; -} - -static __init void radix_tree_init_maxnodes(void) -{ - unsigned long height_to_maxindex[RADIX_TREE_MAX_PATH + 1]; - unsigned int i, j; - - for (i = 0; i < ARRAY_SIZE(height_to_maxindex); i++) - height_to_maxindex[i] = __maxindex(i); - for (i = 0; i < ARRAY_SIZE(height_to_maxnodes); i++) { - for (j = i; j > 0; j--) - height_to_maxnodes[i] += height_to_maxindex[j - 1] + 1; - } -} - static int radix_tree_cpu_dead(unsigned int cpu) { struct radix_tree_preload *rtp; @@ -2266,8 +1610,6 @@ static int radix_tree_cpu_dead(unsigned int cpu) kmem_cache_free(radix_tree_node_cachep, node); rtp->nr--; } - kfree(per_cpu(ida_bitmap, cpu)); - per_cpu(ida_bitmap, cpu) = NULL; return 0; } @@ -2277,11 +1619,11 @@ void __init radix_tree_init(void) BUILD_BUG_ON(RADIX_TREE_MAX_TAGS + __GFP_BITS_SHIFT > 32); BUILD_BUG_ON(ROOT_IS_IDR & ~GFP_ZONEMASK); + BUILD_BUG_ON(XA_CHUNK_SIZE > 255); radix_tree_node_cachep = kmem_cache_create("radix_tree_node", sizeof(struct radix_tree_node), 0, SLAB_PANIC | SLAB_RECLAIM_ACCOUNT, radix_tree_node_ctor); - radix_tree_init_maxnodes(); ret = cpuhp_setup_state_nocalls(CPUHP_RADIX_DEAD, "lib/radix:dead", NULL, radix_tree_cpu_dead); WARN_ON(ret < 0); diff --git a/lib/test_xarray.c b/lib/test_xarray.c @@ -0,0 +1,1238 @@ +// SPDX-License-Identifier: GPL-2.0+ +/* + * test_xarray.c: Test the XArray API + * Copyright (c) 2017-2018 Microsoft Corporation + * Author: Matthew Wilcox <willy@infradead.org> + */ + +#include <linux/xarray.h> +#include <linux/module.h> + +static unsigned int tests_run; +static unsigned int tests_passed; + +#ifndef XA_DEBUG +# ifdef __KERNEL__ +void xa_dump(const struct xarray *xa) { } +# endif +#undef XA_BUG_ON +#define XA_BUG_ON(xa, x) do { \ + tests_run++; \ + if (x) { \ + printk("BUG at %s:%d\n", __func__, __LINE__); \ + xa_dump(xa); \ + dump_stack(); \ + } else { \ + tests_passed++; \ + } \ +} while (0) +#endif + +static void *xa_store_index(struct xarray *xa, unsigned long index, gfp_t gfp) +{ + return xa_store(xa, index, xa_mk_value(index & LONG_MAX), gfp); +} + +static void xa_alloc_index(struct xarray *xa, unsigned long index, gfp_t gfp) +{ + u32 id = 0; + + XA_BUG_ON(xa, xa_alloc(xa, &id, UINT_MAX, xa_mk_value(index & LONG_MAX), + gfp) != 0); + XA_BUG_ON(xa, id != index); +} + +static void xa_erase_index(struct xarray *xa, unsigned long index) +{ + XA_BUG_ON(xa, xa_erase(xa, index) != xa_mk_value(index & LONG_MAX)); + XA_BUG_ON(xa, xa_load(xa, index) != NULL); +} + +/* + * If anyone needs this, please move it to xarray.c. We have no current + * users outside the test suite because all current multislot users want + * to use the advanced API. + */ +static void *xa_store_order(struct xarray *xa, unsigned long index, + unsigned order, void *entry, gfp_t gfp) +{ + XA_STATE_ORDER(xas, xa, index, order); + void *curr; + + do { + xas_lock(&xas); + curr = xas_store(&xas, entry); + xas_unlock(&xas); + } while (xas_nomem(&xas, gfp)); + + return curr; +} + +static noinline void check_xa_err(struct xarray *xa) +{ + XA_BUG_ON(xa, xa_err(xa_store_index(xa, 0, GFP_NOWAIT)) != 0); + XA_BUG_ON(xa, xa_err(xa_erase(xa, 0)) != 0); +#ifndef __KERNEL__ + /* The kernel does not fail GFP_NOWAIT allocations */ + XA_BUG_ON(xa, xa_err(xa_store_index(xa, 1, GFP_NOWAIT)) != -ENOMEM); + XA_BUG_ON(xa, xa_err(xa_store_index(xa, 1, GFP_NOWAIT)) != -ENOMEM); +#endif + XA_BUG_ON(xa, xa_err(xa_store_index(xa, 1, GFP_KERNEL)) != 0); + XA_BUG_ON(xa, xa_err(xa_store(xa, 1, xa_mk_value(0), GFP_KERNEL)) != 0); + XA_BUG_ON(xa, xa_err(xa_erase(xa, 1)) != 0); +// kills the test-suite :-( +// XA_BUG_ON(xa, xa_err(xa_store(xa, 0, xa_mk_internal(0), 0)) != -EINVAL); +} + +static noinline void check_xas_retry(struct xarray *xa) +{ + XA_STATE(xas, xa, 0); + void *entry; + + xa_store_index(xa, 0, GFP_KERNEL); + xa_store_index(xa, 1, GFP_KERNEL); + + rcu_read_lock(); + XA_BUG_ON(xa, xas_find(&xas, ULONG_MAX) != xa_mk_value(0)); + xa_erase_index(xa, 1); + XA_BUG_ON(xa, !xa_is_retry(xas_reload(&xas))); + XA_BUG_ON(xa, xas_retry(&xas, NULL)); + XA_BUG_ON(xa, xas_retry(&xas, xa_mk_value(0))); + xas_reset(&xas); + XA_BUG_ON(xa, xas.xa_node != XAS_RESTART); + XA_BUG_ON(xa, xas_next_entry(&xas, ULONG_MAX) != xa_mk_value(0)); + XA_BUG_ON(xa, xas.xa_node != NULL); + + XA_BUG_ON(xa, xa_store_index(xa, 1, GFP_KERNEL) != NULL); + XA_BUG_ON(xa, !xa_is_internal(xas_reload(&xas))); + xas.xa_node = XAS_RESTART; + XA_BUG_ON(xa, xas_next_entry(&xas, ULONG_MAX) != xa_mk_value(0)); + rcu_read_unlock(); + + /* Make sure we can iterate through retry entries */ + xas_lock(&xas); + xas_set(&xas, 0); + xas_store(&xas, XA_RETRY_ENTRY); + xas_set(&xas, 1); + xas_store(&xas, XA_RETRY_ENTRY); + + xas_set(&xas, 0); + xas_for_each(&xas, entry, ULONG_MAX) { + xas_store(&xas, xa_mk_value(xas.xa_index)); + } + xas_unlock(&xas); + + xa_erase_index(xa, 0); + xa_erase_index(xa, 1); +} + +static noinline void check_xa_load(struct xarray *xa) +{ + unsigned long i, j; + + for (i = 0; i < 1024; i++) { + for (j = 0; j < 1024; j++) { + void *entry = xa_load(xa, j); + if (j < i) + XA_BUG_ON(xa, xa_to_value(entry) != j); + else + XA_BUG_ON(xa, entry); + } + XA_BUG_ON(xa, xa_store_index(xa, i, GFP_KERNEL) != NULL); + } + + for (i = 0; i < 1024; i++) { + for (j = 0; j < 1024; j++) { + void *entry = xa_load(xa, j); + if (j >= i) + XA_BUG_ON(xa, xa_to_value(entry) != j); + else + XA_BUG_ON(xa, entry); + } + xa_erase_index(xa, i); + } + XA_BUG_ON(xa, !xa_empty(xa)); +} + +static noinline void check_xa_mark_1(struct xarray *xa, unsigned long index) +{ + unsigned int order; + unsigned int max_order = IS_ENABLED(CONFIG_XARRAY_MULTI) ? 8 : 1; + + /* NULL elements have no marks set */ + XA_BUG_ON(xa, xa_get_mark(xa, index, XA_MARK_0)); + xa_set_mark(xa, index, XA_MARK_0); + XA_BUG_ON(xa, xa_get_mark(xa, index, XA_MARK_0)); + + /* Storing a pointer will not make a mark appear */ + XA_BUG_ON(xa, xa_store_index(xa, index, GFP_KERNEL) != NULL); + XA_BUG_ON(xa, xa_get_mark(xa, index, XA_MARK_0)); + xa_set_mark(xa, index, XA_MARK_0); + XA_BUG_ON(xa, !xa_get_mark(xa, index, XA_MARK_0)); + + /* Setting one mark will not set another mark */ + XA_BUG_ON(xa, xa_get_mark(xa, index + 1, XA_MARK_0)); + XA_BUG_ON(xa, xa_get_mark(xa, index, XA_MARK_1)); + + /* Storing NULL clears marks, and they can't be set again */ + xa_erase_index(xa, index); + XA_BUG_ON(xa, !xa_empty(xa)); + XA_BUG_ON(xa, xa_get_mark(xa, index, XA_MARK_0)); + xa_set_mark(xa, index, XA_MARK_0); + XA_BUG_ON(xa, xa_get_mark(xa, index, XA_MARK_0)); + + /* + * Storing a multi-index entry over entries with marks gives the + * entire entry the union of the marks + */ + BUG_ON((index % 4) != 0); + for (order = 2; order < max_order; order++) { + unsigned long base = round_down(index, 1UL << order); + unsigned long next = base + (1UL << order); + unsigned long i; + + XA_BUG_ON(xa, xa_store_index(xa, index + 1, GFP_KERNEL)); + xa_set_mark(xa, index + 1, XA_MARK_0); + XA_BUG_ON(xa, xa_store_index(xa, index + 2, GFP_KERNEL)); + xa_set_mark(xa, index + 2, XA_MARK_1); + XA_BUG_ON(xa, xa_store_index(xa, next, GFP_KERNEL)); + xa_store_order(xa, index, order, xa_mk_value(index), + GFP_KERNEL); + for (i = base; i < next; i++) { + XA_STATE(xas, xa, i); + unsigned int seen = 0; + void *entry; + + XA_BUG_ON(xa, !xa_get_mark(xa, i, XA_MARK_0)); + XA_BUG_ON(xa, !xa_get_mark(xa, i, XA_MARK_1)); + XA_BUG_ON(xa, xa_get_mark(xa, i, XA_MARK_2)); + + /* We should see two elements in the array */ + xas_for_each(&xas, entry, ULONG_MAX) + seen++; + XA_BUG_ON(xa, seen != 2); + + /* One of which is marked */ + xas_set(&xas, 0); + seen = 0; + xas_for_each_marked(&xas, entry, ULONG_MAX, XA_MARK_0) + seen++; + XA_BUG_ON(xa, seen != 1); + } + XA_BUG_ON(xa, xa_get_mark(xa, next, XA_MARK_0)); + XA_BUG_ON(xa, xa_get_mark(xa, next, XA_MARK_1)); + XA_BUG_ON(xa, xa_get_mark(xa, next, XA_MARK_2)); + xa_erase_index(xa, index); + xa_erase_index(xa, next); + XA_BUG_ON(xa, !xa_empty(xa)); + } + XA_BUG_ON(xa, !xa_empty(xa)); +} + +static noinline void check_xa_mark_2(struct xarray *xa) +{ + XA_STATE(xas, xa, 0); + unsigned long index; + unsigned int count = 0; + void *entry; + + xa_store_index(xa, 0, GFP_KERNEL); + xa_set_mark(xa, 0, XA_MARK_0); + xas_lock(&xas); + xas_load(&xas); + xas_init_marks(&xas); + xas_unlock(&xas); + XA_BUG_ON(xa, !xa_get_mark(xa, 0, XA_MARK_0) == 0); + + for (index = 3500; index < 4500; index++) { + xa_store_index(xa, index, GFP_KERNEL); + xa_set_mark(xa, index, XA_MARK_0); + } + + xas_reset(&xas); + rcu_read_lock(); + xas_for_each_marked(&xas, entry, ULONG_MAX, XA_MARK_0) + count++; + rcu_read_unlock(); + XA_BUG_ON(xa, count != 1000); + + xas_lock(&xas); + xas_for_each(&xas, entry, ULONG_MAX) { + xas_init_marks(&xas); + XA_BUG_ON(xa, !xa_get_mark(xa, xas.xa_index, XA_MARK_0)); + XA_BUG_ON(xa, !xas_get_mark(&xas, XA_MARK_0)); + } + xas_unlock(&xas); + + xa_destroy(xa); +} + +static noinline void check_xa_mark(struct xarray *xa) +{ + unsigned long index; + + for (index = 0; index < 16384; index += 4) + check_xa_mark_1(xa, index); + + check_xa_mark_2(xa); +} + +static noinline void check_xa_shrink(struct xarray *xa) +{ + XA_STATE(xas, xa, 1); + struct xa_node *node; + unsigned int order; + unsigned int max_order = IS_ENABLED(CONFIG_XARRAY_MULTI) ? 15 : 1; + + XA_BUG_ON(xa, !xa_empty(xa)); + XA_BUG_ON(xa, xa_store_index(xa, 0, GFP_KERNEL) != NULL); + XA_BUG_ON(xa, xa_store_index(xa, 1, GFP_KERNEL) != NULL); + + /* + * Check that erasing the entry at 1 shrinks the tree and properly + * marks the node as being deleted. + */ + xas_lock(&xas); + XA_BUG_ON(xa, xas_load(&xas) != xa_mk_value(1)); + node = xas.xa_node; + XA_BUG_ON(xa, xa_entry_locked(xa, node, 0) != xa_mk_value(0)); + XA_BUG_ON(xa, xas_store(&xas, NULL) != xa_mk_value(1)); + XA_BUG_ON(xa, xa_load(xa, 1) != NULL); + XA_BUG_ON(xa, xas.xa_node != XAS_BOUNDS); + XA_BUG_ON(xa, xa_entry_locked(xa, node, 0) != XA_RETRY_ENTRY); + XA_BUG_ON(xa, xas_load(&xas) != NULL); + xas_unlock(&xas); + XA_BUG_ON(xa, xa_load(xa, 0) != xa_mk_value(0)); + xa_erase_index(xa, 0); + XA_BUG_ON(xa, !xa_empty(xa)); + + for (order = 0; order < max_order; order++) { + unsigned long max = (1UL << order) - 1; + xa_store_order(xa, 0, order, xa_mk_value(0), GFP_KERNEL); + XA_BUG_ON(xa, xa_load(xa, max) != xa_mk_value(0)); + XA_BUG_ON(xa, xa_load(xa, max + 1) != NULL); + rcu_read_lock(); + node = xa_head(xa); + rcu_read_unlock(); + XA_BUG_ON(xa, xa_store_index(xa, ULONG_MAX, GFP_KERNEL) != + NULL); + rcu_read_lock(); + XA_BUG_ON(xa, xa_head(xa) == node); + rcu_read_unlock(); + XA_BUG_ON(xa, xa_load(xa, max + 1) != NULL); + xa_erase_index(xa, ULONG_MAX); + XA_BUG_ON(xa, xa->xa_head != node); + xa_erase_index(xa, 0); + } +} + +static noinline void check_cmpxchg(struct xarray *xa) +{ + void *FIVE = xa_mk_value(5); + void *SIX = xa_mk_value(6); + void *LOTS = xa_mk_value(12345678); + + XA_BUG_ON(xa, !xa_empty(xa)); + XA_BUG_ON(xa, xa_store_index(xa, 12345678, GFP_KERNEL) != NULL); + XA_BUG_ON(xa, xa_insert(xa, 12345678, xa, GFP_KERNEL) != -EEXIST); + XA_BUG_ON(xa, xa_cmpxchg(xa, 12345678, SIX, FIVE, GFP_KERNEL) != LOTS); + XA_BUG_ON(xa, xa_cmpxchg(xa, 12345678, LOTS, FIVE, GFP_KERNEL) != LOTS); + XA_BUG_ON(xa, xa_cmpxchg(xa, 12345678, FIVE, LOTS, GFP_KERNEL) != FIVE); + XA_BUG_ON(xa, xa_cmpxchg(xa, 5, FIVE, NULL, GFP_KERNEL) != NULL); + XA_BUG_ON(xa, xa_cmpxchg(xa, 5, NULL, FIVE, GFP_KERNEL) != NULL); + xa_erase_index(xa, 12345678); + xa_erase_index(xa, 5); + XA_BUG_ON(xa, !xa_empty(xa)); +} + +static noinline void check_reserve(struct xarray *xa) +{ + void *entry; + unsigned long index = 0; + + /* An array with a reserved entry is not empty */ + XA_BUG_ON(xa, !xa_empty(xa)); + xa_reserve(xa, 12345678, GFP_KERNEL); + XA_BUG_ON(xa, xa_empty(xa)); + XA_BUG_ON(xa, xa_load(xa, 12345678)); + xa_release(xa, 12345678); + XA_BUG_ON(xa, !xa_empty(xa)); + + /* Releasing a used entry does nothing */ + xa_reserve(xa, 12345678, GFP_KERNEL); + XA_BUG_ON(xa, xa_store_index(xa, 12345678, GFP_NOWAIT) != NULL); + xa_release(xa, 12345678); + xa_erase_index(xa, 12345678); + XA_BUG_ON(xa, !xa_empty(xa)); + + /* cmpxchg sees a reserved entry as NULL */ + xa_reserve(xa, 12345678, GFP_KERNEL); + XA_BUG_ON(xa, xa_cmpxchg(xa, 12345678, NULL, xa_mk_value(12345678), + GFP_NOWAIT) != NULL); + xa_release(xa, 12345678); + xa_erase_index(xa, 12345678); + XA_BUG_ON(xa, !xa_empty(xa)); + + /* Can iterate through a reserved entry */ + xa_store_index(xa, 5, GFP_KERNEL); + xa_reserve(xa, 6, GFP_KERNEL); + xa_store_index(xa, 7, GFP_KERNEL); + + xa_for_each(xa, entry, index, ULONG_MAX, XA_PRESENT) { + XA_BUG_ON(xa, index != 5 && index != 7); + } + xa_destroy(xa); +} + +static noinline void check_xas_erase(struct xarray *xa) +{ + XA_STATE(xas, xa, 0); + void *entry; + unsigned long i, j; + + for (i = 0; i < 200; i++) { + for (j = i; j < 2 * i + 17; j++) { + xas_set(&xas, j); + do { + xas_lock(&xas); + xas_store(&xas, xa_mk_value(j)); + xas_unlock(&xas); + } while (xas_nomem(&xas, GFP_KERNEL)); + } + + xas_set(&xas, ULONG_MAX); + do { + xas_lock(&xas); + xas_store(&xas, xa_mk_value(0)); + xas_unlock(&xas); + } while (xas_nomem(&xas, GFP_KERNEL)); + + xas_lock(&xas); + xas_store(&xas, NULL); + + xas_set(&xas, 0); + j = i; + xas_for_each(&xas, entry, ULONG_MAX) { + XA_BUG_ON(xa, entry != xa_mk_value(j)); + xas_store(&xas, NULL); + j++; + } + xas_unlock(&xas); + XA_BUG_ON(xa, !xa_empty(xa)); + } +} + +#ifdef CONFIG_XARRAY_MULTI +static noinline void check_multi_store_1(struct xarray *xa, unsigned long index, + unsigned int order) +{ + XA_STATE(xas, xa, index); + unsigned long min = index & ~((1UL << order) - 1); + unsigned long max = min + (1UL << order); + + xa_store_order(xa, index, order, xa_mk_value(index), GFP_KERNEL); + XA_BUG_ON(xa, xa_load(xa, min) != xa_mk_value(index)); + XA_BUG_ON(xa, xa_load(xa, max - 1) != xa_mk_value(index)); + XA_BUG_ON(xa, xa_load(xa, max) != NULL); + XA_BUG_ON(xa, xa_load(xa, min - 1) != NULL); + + XA_BUG_ON(xa, xas_store(&xas, xa_mk_value(min)) != xa_mk_value(index)); + XA_BUG_ON(xa, xa_load(xa, min) != xa_mk_value(min)); + XA_BUG_ON(xa, xa_load(xa, max - 1) != xa_mk_value(min)); + XA_BUG_ON(xa, xa_load(xa, max) != NULL); + XA_BUG_ON(xa, xa_load(xa, min - 1) != NULL); + + xa_erase_index(xa, min); + XA_BUG_ON(xa, !xa_empty(xa)); +} + +static noinline void check_multi_store_2(struct xarray *xa, unsigned long index, + unsigned int order) +{ + XA_STATE(xas, xa, index); + xa_store_order(xa, index, order, xa_mk_value(0), GFP_KERNEL); + + XA_BUG_ON(xa, xas_store(&xas, xa_mk_value(1)) != xa_mk_value(0)); + XA_BUG_ON(xa, xas.xa_index != index); + XA_BUG_ON(xa, xas_store(&xas, NULL) != xa_mk_value(1)); + XA_BUG_ON(xa, !xa_empty(xa)); +} +#endif + +static noinline void check_multi_store(struct xarray *xa) +{ +#ifdef CONFIG_XARRAY_MULTI + unsigned long i, j, k; + unsigned int max_order = (sizeof(long) == 4) ? 30 : 60; + + /* Loading from any position returns the same value */ + xa_store_order(xa, 0, 1, xa_mk_value(0), GFP_KERNEL); + XA_BUG_ON(xa, xa_load(xa, 0) != xa_mk_value(0)); + XA_BUG_ON(xa, xa_load(xa, 1) != xa_mk_value(0)); + XA_BUG_ON(xa, xa_load(xa, 2) != NULL); + rcu_read_lock(); + XA_BUG_ON(xa, xa_to_node(xa_head(xa))->count != 2); + XA_BUG_ON(xa, xa_to_node(xa_head(xa))->nr_values != 2); + rcu_read_unlock(); + + /* Storing adjacent to the value does not alter the value */ + xa_store(xa, 3, xa, GFP_KERNEL); + XA_BUG_ON(xa, xa_load(xa, 0) != xa_mk_value(0)); + XA_BUG_ON(xa, xa_load(xa, 1) != xa_mk_value(0)); + XA_BUG_ON(xa, xa_load(xa, 2) != NULL); + rcu_read_lock(); + XA_BUG_ON(xa, xa_to_node(xa_head(xa))->count != 3); + XA_BUG_ON(xa, xa_to_node(xa_head(xa))->nr_values != 2); + rcu_read_unlock(); + + /* Overwriting multiple indexes works */ + xa_store_order(xa, 0, 2, xa_mk_value(1), GFP_KERNEL); + XA_BUG_ON(xa, xa_load(xa, 0) != xa_mk_value(1)); + XA_BUG_ON(xa, xa_load(xa, 1) != xa_mk_value(1)); + XA_BUG_ON(xa, xa_load(xa, 2) != xa_mk_value(1)); + XA_BUG_ON(xa, xa_load(xa, 3) != xa_mk_value(1)); + XA_BUG_ON(xa, xa_load(xa, 4) != NULL); + rcu_read_lock(); + XA_BUG_ON(xa, xa_to_node(xa_head(xa))->count != 4); + XA_BUG_ON(xa, xa_to_node(xa_head(xa))->nr_values != 4); + rcu_read_unlock(); + + /* We can erase multiple values with a single store */ + xa_store_order(xa, 0, 63, NULL, GFP_KERNEL); + XA_BUG_ON(xa, !xa_empty(xa)); + + /* Even when the first slot is empty but the others aren't */ + xa_store_index(xa, 1, GFP_KERNEL); + xa_store_index(xa, 2, GFP_KERNEL); + xa_store_order(xa, 0, 2, NULL, GFP_KERNEL); + XA_BUG_ON(xa, !xa_empty(xa)); + + for (i = 0; i < max_order; i++) { + for (j = 0; j < max_order; j++) { + xa_store_order(xa, 0, i, xa_mk_value(i), GFP_KERNEL); + xa_store_order(xa, 0, j, xa_mk_value(j), GFP_KERNEL); + + for (k = 0; k < max_order; k++) { + void *entry = xa_load(xa, (1UL << k) - 1); + if ((i < k) && (j < k)) + XA_BUG_ON(xa, entry != NULL); + else + XA_BUG_ON(xa, entry != xa_mk_value(j)); + } + + xa_erase(xa, 0); + XA_BUG_ON(xa, !xa_empty(xa)); + } + } + + for (i = 0; i < 20; i++) { + check_multi_store_1(xa, 200, i); + check_multi_store_1(xa, 0, i); + check_multi_store_1(xa, (1UL << i) + 1, i); + } + check_multi_store_2(xa, 4095, 9); +#endif +} + +static DEFINE_XARRAY_ALLOC(xa0); + +static noinline void check_xa_alloc(void) +{ + int i; + u32 id; + + /* An empty array should assign 0 to the first alloc */ + xa_alloc_index(&xa0, 0, GFP_KERNEL); + + /* Erasing it should make the array empty again */ + xa_erase_index(&xa0, 0); + XA_BUG_ON(&xa0, !xa_empty(&xa0)); + + /* And it should assign 0 again */ + xa_alloc_index(&xa0, 0, GFP_KERNEL); + + /* The next assigned ID should be 1 */ + xa_alloc_index(&xa0, 1, GFP_KERNEL); + xa_erase_index(&xa0, 1); + + /* Storing a value should mark it used */ + xa_store_index(&xa0, 1, GFP_KERNEL); + xa_alloc_index(&xa0, 2, GFP_KERNEL); + + /* If we then erase 0, it should be free */ + xa_erase_index(&xa0, 0); + xa_alloc_index(&xa0, 0, GFP_KERNEL); + + xa_erase_index(&xa0, 1); + xa_erase_index(&xa0, 2); + + for (i = 1; i < 5000; i++) { + xa_alloc_index(&xa0, i, GFP_KERNEL); + } + + xa_destroy(&xa0); + + id = 0xfffffffeU; + XA_BUG_ON(&xa0, xa_alloc(&xa0, &id, UINT_MAX, xa_mk_value(0), + GFP_KERNEL) != 0); + XA_BUG_ON(&xa0, id != 0xfffffffeU); + XA_BUG_ON(&xa0, xa_alloc(&xa0, &id, UINT_MAX, xa_mk_value(0), + GFP_KERNEL) != 0); + XA_BUG_ON(&xa0, id != 0xffffffffU); + XA_BUG_ON(&xa0, xa_alloc(&xa0, &id, UINT_MAX, xa_mk_value(0), + GFP_KERNEL) != -ENOSPC); + XA_BUG_ON(&xa0, id != 0xffffffffU); + xa_destroy(&xa0); +} + +static noinline void __check_store_iter(struct xarray *xa, unsigned long start, + unsigned int order, unsigned int present) +{ + XA_STATE_ORDER(xas, xa, start, order); + void *entry; + unsigned int count = 0; + +retry: + xas_lock(&xas); + xas_for_each_conflict(&xas, entry) { + XA_BUG_ON(xa, !xa_is_value(entry)); + XA_BUG_ON(xa, entry < xa_mk_value(start)); + XA_BUG_ON(xa, entry > xa_mk_value(start + (1UL << order) - 1)); + count++; + } + xas_store(&xas, xa_mk_value(start)); + xas_unlock(&xas); + if (xas_nomem(&xas, GFP_KERNEL)) { + count = 0; + goto retry; + } + XA_BUG_ON(xa, xas_error(&xas)); + XA_BUG_ON(xa, count != present); + XA_BUG_ON(xa, xa_load(xa, start) != xa_mk_value(start)); + XA_BUG_ON(xa, xa_load(xa, start + (1UL << order) - 1) != + xa_mk_value(start)); + xa_erase_index(xa, start); +} + +static noinline void check_store_iter(struct xarray *xa) +{ + unsigned int i, j; + unsigned int max_order = IS_ENABLED(CONFIG_XARRAY_MULTI) ? 20 : 1; + + for (i = 0; i < max_order; i++) { + unsigned int min = 1 << i; + unsigned int max = (2 << i) - 1; + __check_store_iter(xa, 0, i, 0); + XA_BUG_ON(xa, !xa_empty(xa)); + __check_store_iter(xa, min, i, 0); + XA_BUG_ON(xa, !xa_empty(xa)); + + xa_store_index(xa, min, GFP_KERNEL); + __check_store_iter(xa, min, i, 1); + XA_BUG_ON(xa, !xa_empty(xa)); + xa_store_index(xa, max, GFP_KERNEL); + __check_store_iter(xa, min, i, 1); + XA_BUG_ON(xa, !xa_empty(xa)); + + for (j = 0; j < min; j++) + xa_store_index(xa, j, GFP_KERNEL); + __check_store_iter(xa, 0, i, min); + XA_BUG_ON(xa, !xa_empty(xa)); + for (j = 0; j < min; j++) + xa_store_index(xa, min + j, GFP_KERNEL); + __check_store_iter(xa, min, i, min); + XA_BUG_ON(xa, !xa_empty(xa)); + } +#ifdef CONFIG_XARRAY_MULTI + xa_store_index(xa, 63, GFP_KERNEL); + xa_store_index(xa, 65, GFP_KERNEL); + __check_store_iter(xa, 64, 2, 1); + xa_erase_index(xa, 63); +#endif + XA_BUG_ON(xa, !xa_empty(xa)); +} + +static noinline void check_multi_find(struct xarray *xa) +{ +#ifdef CONFIG_XARRAY_MULTI + unsigned long index; + + xa_store_order(xa, 12, 2, xa_mk_value(12), GFP_KERNEL); + XA_BUG_ON(xa, xa_store_index(xa, 16, GFP_KERNEL) != NULL); + + index = 0; + XA_BUG_ON(xa, xa_find(xa, &index, ULONG_MAX, XA_PRESENT) != + xa_mk_value(12)); + XA_BUG_ON(xa, index != 12); + index = 13; + XA_BUG_ON(xa, xa_find(xa, &index, ULONG_MAX, XA_PRESENT) != + xa_mk_value(12)); + XA_BUG_ON(xa, (index < 12) || (index >= 16)); + XA_BUG_ON(xa, xa_find_after(xa, &index, ULONG_MAX, XA_PRESENT) != + xa_mk_value(16)); + XA_BUG_ON(xa, index != 16); + + xa_erase_index(xa, 12); + xa_erase_index(xa, 16); + XA_BUG_ON(xa, !xa_empty(xa)); +#endif +} + +static noinline void check_multi_find_2(struct xarray *xa) +{ + unsigned int max_order = IS_ENABLED(CONFIG_XARRAY_MULTI) ? 10 : 1; + unsigned int i, j; + void *entry; + + for (i = 0; i < max_order; i++) { + unsigned long index = 1UL << i; + for (j = 0; j < index; j++) { + XA_STATE(xas, xa, j + index); + xa_store_index(xa, index - 1, GFP_KERNEL); + xa_store_order(xa, index, i, xa_mk_value(index), + GFP_KERNEL); + rcu_read_lock(); + xas_for_each(&xas, entry, ULONG_MAX) { + xa_erase_index(xa, index); + } + rcu_read_unlock(); + xa_erase_index(xa, index - 1); + XA_BUG_ON(xa, !xa_empty(xa)); + } + } +} + +static noinline void check_find(struct xarray *xa) +{ + unsigned long i, j, k; + + XA_BUG_ON(xa, !xa_empty(xa)); + + /* + * Check xa_find with all pairs between 0 and 99 inclusive, + * starting at every index between 0 and 99 + */ + for (i = 0; i < 100; i++) { + XA_BUG_ON(xa, xa_store_index(xa, i, GFP_KERNEL) != NULL); + xa_set_mark(xa, i, XA_MARK_0); + for (j = 0; j < i; j++) { + XA_BUG_ON(xa, xa_store_index(xa, j, GFP_KERNEL) != + NULL); + xa_set_mark(xa, j, XA_MARK_0); + for (k = 0; k < 100; k++) { + unsigned long index = k; + void *entry = xa_find(xa, &index, ULONG_MAX, + XA_PRESENT); + if (k <= j) + XA_BUG_ON(xa, index != j); + else if (k <= i) + XA_BUG_ON(xa, index != i); + else + XA_BUG_ON(xa, entry != NULL); + + index = k; + entry = xa_find(xa, &index, ULONG_MAX, + XA_MARK_0); + if (k <= j) + XA_BUG_ON(xa, index != j); + else if (k <= i) + XA_BUG_ON(xa, index != i); + else + XA_BUG_ON(xa, entry != NULL); + } + xa_erase_index(xa, j); + XA_BUG_ON(xa, xa_get_mark(xa, j, XA_MARK_0)); + XA_BUG_ON(xa, !xa_get_mark(xa, i, XA_MARK_0)); + } + xa_erase_index(xa, i); + XA_BUG_ON(xa, xa_get_mark(xa, i, XA_MARK_0)); + } + XA_BUG_ON(xa, !xa_empty(xa)); + check_multi_find(xa); + check_multi_find_2(xa); +} + +/* See find_swap_entry() in mm/shmem.c */ +static noinline unsigned long xa_find_entry(struct xarray *xa, void *item) +{ + XA_STATE(xas, xa, 0); + unsigned int checked = 0; + void *entry; + + rcu_read_lock(); + xas_for_each(&xas, entry, ULONG_MAX) { + if (xas_retry(&xas, entry)) + continue; + if (entry == item) + break; + checked++; + if ((checked % 4) != 0) + continue; + xas_pause(&xas); + } + rcu_read_unlock(); + + return entry ? xas.xa_index : -1; +} + +static noinline void check_find_entry(struct xarray *xa) +{ +#ifdef CONFIG_XARRAY_MULTI + unsigned int order; + unsigned long offset, index; + + for (order = 0; order < 20; order++) { + for (offset = 0; offset < (1UL << (order + 3)); + offset += (1UL << order)) { + for (index = 0; index < (1UL << (order + 5)); + index += (1UL << order)) { + xa_store_order(xa, index, order, + xa_mk_value(index), GFP_KERNEL); + XA_BUG_ON(xa, xa_load(xa, index) != + xa_mk_value(index)); + XA_BUG_ON(xa, xa_find_entry(xa, + xa_mk_value(index)) != index); + } + XA_BUG_ON(xa, xa_find_entry(xa, xa) != -1); + xa_destroy(xa); + } + } +#endif + + XA_BUG_ON(xa, xa_find_entry(xa, xa) != -1); + xa_store_index(xa, ULONG_MAX, GFP_KERNEL); + XA_BUG_ON(xa, xa_find_entry(xa, xa) != -1); + XA_BUG_ON(xa, xa_find_entry(xa, xa_mk_value(LONG_MAX)) != -1); + xa_erase_index(xa, ULONG_MAX); + XA_BUG_ON(xa, !xa_empty(xa)); +} + +static noinline void check_move_small(struct xarray *xa, unsigned long idx) +{ + XA_STATE(xas, xa, 0); + unsigned long i; + + xa_store_index(xa, 0, GFP_KERNEL); + xa_store_index(xa, idx, GFP_KERNEL); + + rcu_read_lock(); + for (i = 0; i < idx * 4; i++) { + void *entry = xas_next(&xas); + if (i <= idx) + XA_BUG_ON(xa, xas.xa_node == XAS_RESTART); + XA_BUG_ON(xa, xas.xa_index != i); + if (i == 0 || i == idx) + XA_BUG_ON(xa, entry != xa_mk_value(i)); + else + XA_BUG_ON(xa, entry != NULL); + } + xas_next(&xas); + XA_BUG_ON(xa, xas.xa_index != i); + + do { + void *entry = xas_prev(&xas); + i--; + if (i <= idx) + XA_BUG_ON(xa, xas.xa_node == XAS_RESTART); + XA_BUG_ON(xa, xas.xa_index != i); + if (i == 0 || i == idx) + XA_BUG_ON(xa, entry != xa_mk_value(i)); + else + XA_BUG_ON(xa, entry != NULL); + } while (i > 0); + + xas_set(&xas, ULONG_MAX); + XA_BUG_ON(xa, xas_next(&xas) != NULL); + XA_BUG_ON(xa, xas.xa_index != ULONG_MAX); + XA_BUG_ON(xa, xas_next(&xas) != xa_mk_value(0)); + XA_BUG_ON(xa, xas.xa_index != 0); + XA_BUG_ON(xa, xas_prev(&xas) != NULL); + XA_BUG_ON(xa, xas.xa_index != ULONG_MAX); + rcu_read_unlock(); + + xa_erase_index(xa, 0); + xa_erase_index(xa, idx); + XA_BUG_ON(xa, !xa_empty(xa)); +} + +static noinline void check_move(struct xarray *xa) +{ + XA_STATE(xas, xa, (1 << 16) - 1); + unsigned long i; + + for (i = 0; i < (1 << 16); i++) + XA_BUG_ON(xa, xa_store_index(xa, i, GFP_KERNEL) != NULL); + + rcu_read_lock(); + do { + void *entry = xas_prev(&xas); + i--; + XA_BUG_ON(xa, entry != xa_mk_value(i)); + XA_BUG_ON(xa, i != xas.xa_index); + } while (i != 0); + + XA_BUG_ON(xa, xas_prev(&xas) != NULL); + XA_BUG_ON(xa, xas.xa_index != ULONG_MAX); + + do { + void *entry = xas_next(&xas); + XA_BUG_ON(xa, entry != xa_mk_value(i)); + XA_BUG_ON(xa, i != xas.xa_index); + i++; + } while (i < (1 << 16)); + rcu_read_unlock(); + + for (i = (1 << 8); i < (1 << 15); i++) + xa_erase_index(xa, i); + + i = xas.xa_index; + + rcu_read_lock(); + do { + void *entry = xas_prev(&xas); + i--; + if ((i < (1 << 8)) || (i >= (1 << 15))) + XA_BUG_ON(xa, entry != xa_mk_value(i)); + else + XA_BUG_ON(xa, entry != NULL); + XA_BUG_ON(xa, i != xas.xa_index); + } while (i != 0); + + XA_BUG_ON(xa, xas_prev(&xas) != NULL); + XA_BUG_ON(xa, xas.xa_index != ULONG_MAX); + + do { + void *entry = xas_next(&xas); + if ((i < (1 << 8)) || (i >= (1 << 15))) + XA_BUG_ON(xa, entry != xa_mk_value(i)); + else + XA_BUG_ON(xa, entry != NULL); + XA_BUG_ON(xa, i != xas.xa_index); + i++; + } while (i < (1 << 16)); + rcu_read_unlock(); + + xa_destroy(xa); + + for (i = 0; i < 16; i++) + check_move_small(xa, 1UL << i); + + for (i = 2; i < 16; i++) + check_move_small(xa, (1UL << i) - 1); +} + +static noinline void xa_store_many_order(struct xarray *xa, + unsigned long index, unsigned order) +{ + XA_STATE_ORDER(xas, xa, index, order); + unsigned int i = 0; + + do { + xas_lock(&xas); + XA_BUG_ON(xa, xas_find_conflict(&xas)); + xas_create_range(&xas); + if (xas_error(&xas)) + goto unlock; + for (i = 0; i < (1U << order); i++) { + XA_BUG_ON(xa, xas_store(&xas, xa_mk_value(index + i))); + xas_next(&xas); + } +unlock: + xas_unlock(&xas); + } while (xas_nomem(&xas, GFP_KERNEL)); + + XA_BUG_ON(xa, xas_error(&xas)); +} + +static noinline void check_create_range_1(struct xarray *xa, + unsigned long index, unsigned order) +{ + unsigned long i; + + xa_store_many_order(xa, index, order); + for (i = index; i < index + (1UL << order); i++) + xa_erase_index(xa, i); + XA_BUG_ON(xa, !xa_empty(xa)); +} + +static noinline void check_create_range_2(struct xarray *xa, unsigned order) +{ + unsigned long i; + unsigned long nr = 1UL << order; + + for (i = 0; i < nr * nr; i += nr) + xa_store_many_order(xa, i, order); + for (i = 0; i < nr * nr; i++) + xa_erase_index(xa, i); + XA_BUG_ON(xa, !xa_empty(xa)); +} + +static noinline void check_create_range_3(void) +{ + XA_STATE(xas, NULL, 0); + xas_set_err(&xas, -EEXIST); + xas_create_range(&xas); + XA_BUG_ON(NULL, xas_error(&xas) != -EEXIST); +} + +static noinline void check_create_range_4(struct xarray *xa, + unsigned long index, unsigned order) +{ + XA_STATE_ORDER(xas, xa, index, order); + unsigned long base = xas.xa_index; + unsigned long i = 0; + + xa_store_index(xa, index, GFP_KERNEL); + do { + xas_lock(&xas); + xas_create_range(&xas); + if (xas_error(&xas)) + goto unlock; + for (i = 0; i < (1UL << order); i++) { + void *old = xas_store(&xas, xa_mk_value(base + i)); + if (xas.xa_index == index) + XA_BUG_ON(xa, old != xa_mk_value(base + i)); + else + XA_BUG_ON(xa, old != NULL); + xas_next(&xas); + } +unlock: + xas_unlock(&xas); + } while (xas_nomem(&xas, GFP_KERNEL)); + + XA_BUG_ON(xa, xas_error(&xas)); + + for (i = base; i < base + (1UL << order); i++) + xa_erase_index(xa, i); + XA_BUG_ON(xa, !xa_empty(xa)); +} + +static noinline void check_create_range(struct xarray *xa) +{ + unsigned int order; + unsigned int max_order = IS_ENABLED(CONFIG_XARRAY_MULTI) ? 12 : 1; + + for (order = 0; order < max_order; order++) { + check_create_range_1(xa, 0, order); + check_create_range_1(xa, 1U << order, order); + check_create_range_1(xa, 2U << order, order); + check_create_range_1(xa, 3U << order, order); + check_create_range_1(xa, 1U << 24, order); + if (order < 10) + check_create_range_2(xa, order); + + check_create_range_4(xa, 0, order); + check_create_range_4(xa, 1U << order, order); + check_create_range_4(xa, 2U << order, order); + check_create_range_4(xa, 3U << order, order); + check_create_range_4(xa, 1U << 24, order); + + check_create_range_4(xa, 1, order); + check_create_range_4(xa, (1U << order) + 1, order); + check_create_range_4(xa, (2U << order) + 1, order); + check_create_range_4(xa, (2U << order) - 1, order); + check_create_range_4(xa, (3U << order) + 1, order); + check_create_range_4(xa, (3U << order) - 1, order); + check_create_range_4(xa, (1U << 24) + 1, order); + } + + check_create_range_3(); +} + +static noinline void __check_store_range(struct xarray *xa, unsigned long first, + unsigned long last) +{ +#ifdef CONFIG_XARRAY_MULTI + xa_store_range(xa, first, last, xa_mk_value(first), GFP_KERNEL); + + XA_BUG_ON(xa, xa_load(xa, first) != xa_mk_value(first)); + XA_BUG_ON(xa, xa_load(xa, last) != xa_mk_value(first)); + XA_BUG_ON(xa, xa_load(xa, first - 1) != NULL); + XA_BUG_ON(xa, xa_load(xa, last + 1) != NULL); + + xa_store_range(xa, first, last, NULL, GFP_KERNEL); +#endif + + XA_BUG_ON(xa, !xa_empty(xa)); +} + +static noinline void check_store_range(struct xarray *xa) +{ + unsigned long i, j; + + for (i = 0; i < 128; i++) { + for (j = i; j < 128; j++) { + __check_store_range(xa, i, j); + __check_store_range(xa, 128 + i, 128 + j); + __check_store_range(xa, 4095 + i, 4095 + j); + __check_store_range(xa, 4096 + i, 4096 + j); + __check_store_range(xa, 123456 + i, 123456 + j); + __check_store_range(xa, UINT_MAX + i, UINT_MAX + j); + } + } +} + +static LIST_HEAD(shadow_nodes); + +static void test_update_node(struct xa_node *node) +{ + if (node->count && node->count == node->nr_values) { + if (list_empty(&node->private_list)) + list_add(&shadow_nodes, &node->private_list); + } else { + if (!list_empty(&node->private_list)) + list_del_init(&node->private_list); + } +} + +static noinline void shadow_remove(struct xarray *xa) +{ + struct xa_node *node; + + xa_lock(xa); + while ((node = list_first_entry_or_null(&shadow_nodes, + struct xa_node, private_list))) { + XA_STATE(xas, node->array, 0); + XA_BUG_ON(xa, node->array != xa); + list_del_init(&node->private_list); + xas.xa_node = xa_parent_locked(node->array, node); + xas.xa_offset = node->offset; + xas.xa_shift = node->shift + XA_CHUNK_SHIFT; + xas_set_update(&xas, test_update_node); + xas_store(&xas, NULL); + } + xa_unlock(xa); +} + +static noinline void check_workingset(struct xarray *xa, unsigned long index) +{ + XA_STATE(xas, xa, index); + xas_set_update(&xas, test_update_node); + + do { + xas_lock(&xas); + xas_store(&xas, xa_mk_value(0)); + xas_next(&xas); + xas_store(&xas, xa_mk_value(1)); + xas_unlock(&xas); + } while (xas_nomem(&xas, GFP_KERNEL)); + + XA_BUG_ON(xa, list_empty(&shadow_nodes)); + + xas_lock(&xas); + xas_next(&xas); + xas_store(&xas, &xas); + XA_BUG_ON(xa, !list_empty(&shadow_nodes)); + + xas_store(&xas, xa_mk_value(2)); + xas_unlock(&xas); + XA_BUG_ON(xa, list_empty(&shadow_nodes)); + + shadow_remove(xa); + XA_BUG_ON(xa, !list_empty(&shadow_nodes)); + XA_BUG_ON(xa, !xa_empty(xa)); +} + +/* + * Check that the pointer / value / sibling entries are accounted the + * way we expect them to be. + */ +static noinline void check_account(struct xarray *xa) +{ +#ifdef CONFIG_XARRAY_MULTI + unsigned int order; + + for (order = 1; order < 12; order++) { + XA_STATE(xas, xa, 1 << order); + + xa_store_order(xa, 0, order, xa, GFP_KERNEL); + xas_load(&xas); + XA_BUG_ON(xa, xas.xa_node->count == 0); + XA_BUG_ON(xa, xas.xa_node->count > (1 << order)); + XA_BUG_ON(xa, xas.xa_node->nr_values != 0); + + xa_store_order(xa, 1 << order, order, xa_mk_value(1 << order), + GFP_KERNEL); + XA_BUG_ON(xa, xas.xa_node->count != xas.xa_node->nr_values * 2); + + xa_erase(xa, 1 << order); + XA_BUG_ON(xa, xas.xa_node->nr_values != 0); + + xa_erase(xa, 0); + XA_BUG_ON(xa, !xa_empty(xa)); + } +#endif +} + +static noinline void check_destroy(struct xarray *xa) +{ + unsigned long index; + + XA_BUG_ON(xa, !xa_empty(xa)); + + /* Destroying an empty array is a no-op */ + xa_destroy(xa); + XA_BUG_ON(xa, !xa_empty(xa)); + + /* Destroying an array with a single entry */ + for (index = 0; index < 1000; index++) { + xa_store_index(xa, index, GFP_KERNEL); + XA_BUG_ON(xa, xa_empty(xa)); + xa_destroy(xa); + XA_BUG_ON(xa, !xa_empty(xa)); + } + + /* Destroying an array with a single entry at ULONG_MAX */ + xa_store(xa, ULONG_MAX, xa, GFP_KERNEL); + XA_BUG_ON(xa, xa_empty(xa)); + xa_destroy(xa); + XA_BUG_ON(xa, !xa_empty(xa)); + +#ifdef CONFIG_XARRAY_MULTI + /* Destroying an array with a multi-index entry */ + xa_store_order(xa, 1 << 11, 11, xa, GFP_KERNEL); + XA_BUG_ON(xa, xa_empty(xa)); + xa_destroy(xa); + XA_BUG_ON(xa, !xa_empty(xa)); +#endif +} + +static DEFINE_XARRAY(array); + +static int xarray_checks(void) +{ + check_xa_err(&array); + check_xas_retry(&array); + check_xa_load(&array); + check_xa_mark(&array); + check_xa_shrink(&array); + check_xas_erase(&array); + check_cmpxchg(&array); + check_reserve(&array); + check_multi_store(&array); + check_xa_alloc(); + check_find(&array); + check_find_entry(&array); + check_account(&array); + check_destroy(&array); + check_move(&array); + check_create_range(&array); + check_store_range(&array); + check_store_iter(&array); + + check_workingset(&array, 0); + check_workingset(&array, 64); + check_workingset(&array, 4096); + + printk("XArray: %u of %u tests passed\n", tests_passed, tests_run); + return (tests_run == tests_passed) ? 0 : -EINVAL; +} + +static void xarray_exit(void) +{ +} + +module_init(xarray_checks); +module_exit(xarray_exit); +MODULE_AUTHOR("Matthew Wilcox <willy@infradead.org>"); +MODULE_LICENSE("GPL"); diff --git a/lib/xarray.c b/lib/xarray.c @@ -0,0 +1,2036 @@ +// SPDX-License-Identifier: GPL-2.0+ +/* + * XArray implementation + * Copyright (c) 2017 Microsoft Corporation + * Author: Matthew Wilcox <willy@infradead.org> + */ + +#include <linux/bitmap.h> +#include <linux/export.h> +#include <linux/list.h> +#include <linux/slab.h> +#include <linux/xarray.h> + +/* + * Coding conventions in this file: + * + * @xa is used to refer to the entire xarray. + * @xas is the 'xarray operation state'. It may be either a pointer to + * an xa_state, or an xa_state stored on the stack. This is an unfortunate + * ambiguity. + * @index is the index of the entry being operated on + * @mark is an xa_mark_t; a small number indicating one of the mark bits. + * @node refers to an xa_node; usually the primary one being operated on by + * this function. + * @offset is the index into the slots array inside an xa_node. + * @parent refers to the @xa_node closer to the head than @node. + * @entry refers to something stored in a slot in the xarray + */ + +static inline unsigned int xa_lock_type(const struct xarray *xa) +{ + return (__force unsigned int)xa->xa_flags & 3; +} + +static inline void xas_lock_type(struct xa_state *xas, unsigned int lock_type) +{ + if (lock_type == XA_LOCK_IRQ) + xas_lock_irq(xas); + else if (lock_type == XA_LOCK_BH) + xas_lock_bh(xas); + else + xas_lock(xas); +} + +static inline void xas_unlock_type(struct xa_state *xas, unsigned int lock_type) +{ + if (lock_type == XA_LOCK_IRQ) + xas_unlock_irq(xas); + else if (lock_type == XA_LOCK_BH) + xas_unlock_bh(xas); + else + xas_unlock(xas); +} + +static inline bool xa_track_free(const struct xarray *xa) +{ + return xa->xa_flags & XA_FLAGS_TRACK_FREE; +} + +static inline void xa_mark_set(struct xarray *xa, xa_mark_t mark) +{ + if (!(xa->xa_flags & XA_FLAGS_MARK(mark))) + xa->xa_flags |= XA_FLAGS_MARK(mark); +} + +static inline void xa_mark_clear(struct xarray *xa, xa_mark_t mark) +{ + if (xa->xa_flags & XA_FLAGS_MARK(mark)) + xa->xa_flags &= ~(XA_FLAGS_MARK(mark)); +} + +static inline unsigned long *node_marks(struct xa_node *node, xa_mark_t mark) +{ + return node->marks[(__force unsigned)mark]; +} + +static inline bool node_get_mark(struct xa_node *node, + unsigned int offset, xa_mark_t mark) +{ + return test_bit(offset, node_marks(node, mark)); +} + +/* returns true if the bit was set */ +static inline bool node_set_mark(struct xa_node *node, unsigned int offset, + xa_mark_t mark) +{ + return __test_and_set_bit(offset, node_marks(node, mark)); +} + +/* returns true if the bit was set */ +static inline bool node_clear_mark(struct xa_node *node, unsigned int offset, + xa_mark_t mark) +{ + return __test_and_clear_bit(offset, node_marks(node, mark)); +} + +static inline bool node_any_mark(struct xa_node *node, xa_mark_t mark) +{ + return !bitmap_empty(node_marks(node, mark), XA_CHUNK_SIZE); +} + +static inline void node_mark_all(struct xa_node *node, xa_mark_t mark) +{ + bitmap_fill(node_marks(node, mark), XA_CHUNK_SIZE); +} + +#define mark_inc(mark) do { \ + mark = (__force xa_mark_t)((__force unsigned)(mark) + 1); \ +} while (0) + +/* + * xas_squash_marks() - Merge all marks to the first entry + * @xas: Array operation state. + * + * Set a mark on the first entry if any entry has it set. Clear marks on + * all sibling entries. + */ +static void xas_squash_marks(const struct xa_state *xas) +{ + unsigned int mark = 0; + unsigned int limit = xas->xa_offset + xas->xa_sibs + 1; + + if (!xas->xa_sibs) + return; + + do { + unsigned long *marks = xas->xa_node->marks[mark]; + if (find_next_bit(marks, limit, xas->xa_offset + 1) == limit) + continue; + __set_bit(xas->xa_offset, marks); + bitmap_clear(marks, xas->xa_offset + 1, xas->xa_sibs); + } while (mark++ != (__force unsigned)XA_MARK_MAX); +} + +/* extracts the offset within this node from the index */ +static unsigned int get_offset(unsigned long index, struct xa_node *node) +{ + return (index >> node->shift) & XA_CHUNK_MASK; +} + +static void xas_set_offset(struct xa_state *xas) +{ + xas->xa_offset = get_offset(xas->xa_index, xas->xa_node); +} + +/* move the index either forwards (find) or backwards (sibling slot) */ +static void xas_move_index(struct xa_state *xas, unsigned long offset) +{ + unsigned int shift = xas->xa_node->shift; + xas->xa_index &= ~XA_CHUNK_MASK << shift; + xas->xa_index += offset << shift; +} + +static void xas_advance(struct xa_state *xas) +{ + xas->xa_offset++; + xas_move_index(xas, xas->xa_offset); +} + +static void *set_bounds(struct xa_state *xas) +{ + xas->xa_node = XAS_BOUNDS; + return NULL; +} + +/* + * Starts a walk. If the @xas is already valid, we assume that it's on + * the right path and just return where we've got to. If we're in an + * error state, return NULL. If the index is outside the current scope + * of the xarray, return NULL without changing @xas->xa_node. Otherwise + * set @xas->xa_node to NULL and return the current head of the array. + */ +static void *xas_start(struct xa_state *xas) +{ + void *entry; + + if (xas_valid(xas)) + return xas_reload(xas); + if (xas_error(xas)) + return NULL; + + entry = xa_head(xas->xa); + if (!xa_is_node(entry)) { + if (xas->xa_index) + return set_bounds(xas); + } else { + if ((xas->xa_index >> xa_to_node(entry)->shift) > XA_CHUNK_MASK) + return set_bounds(xas); + } + + xas->xa_node = NULL; + return entry; +} + +static void *xas_descend(struct xa_state *xas, struct xa_node *node) +{ + unsigned int offset = get_offset(xas->xa_index, node); + void *entry = xa_entry(xas->xa, node, offset); + + xas->xa_node = node; + if (xa_is_sibling(entry)) { + offset = xa_to_sibling(entry); + entry = xa_entry(xas->xa, node, offset); + } + + xas->xa_offset = offset; + return entry; +} + +/** + * xas_load() - Load an entry from the XArray (advanced). + * @xas: XArray operation state. + * + * Usually walks the @xas to the appropriate state to load the entry + * stored at xa_index. However, it will do nothing and return %NULL if + * @xas is in an error state. xas_load() will never expand the tree. + * + * If the xa_state is set up to operate on a multi-index entry, xas_load() + * may return %NULL or an internal entry, even if there are entries + * present within the range specified by @xas. + * + * Context: Any context. The caller should hold the xa_lock or the RCU lock. + * Return: Usually an entry in the XArray, but see description for exceptions. + */ +void *xas_load(struct xa_state *xas) +{ + void *entry = xas_start(xas); + + while (xa_is_node(entry)) { + struct xa_node *node = xa_to_node(entry); + + if (xas->xa_shift > node->shift) + break; + entry = xas_descend(xas, node); + } + return entry; +} +EXPORT_SYMBOL_GPL(xas_load); + +/* Move the radix tree node cache here */ +extern struct kmem_cache *radix_tree_node_cachep; +extern void radix_tree_node_rcu_free(struct rcu_head *head); + +#define XA_RCU_FREE ((struct xarray *)1) + +static void xa_node_free(struct xa_node *node) +{ + XA_NODE_BUG_ON(node, !list_empty(&node->private_list)); + node->array = XA_RCU_FREE; + call_rcu(&node->rcu_head, radix_tree_node_rcu_free); +} + +/* + * xas_destroy() - Free any resources allocated during the XArray operation. + * @xas: XArray operation state. + * + * This function is now internal-only. + */ +static void xas_destroy(struct xa_state *xas) +{ + struct xa_node *node = xas->xa_alloc; + + if (!node) + return; + XA_NODE_BUG_ON(node, !list_empty(&node->private_list)); + kmem_cache_free(radix_tree_node_cachep, node); + xas->xa_alloc = NULL; +} + +/** + * xas_nomem() - Allocate memory if needed. + * @xas: XArray operation state. + * @gfp: Memory allocation flags. + * + * If we need to add new nodes to the XArray, we try to allocate memory + * with GFP_NOWAIT while holding the lock, which will usually succeed. + * If it fails, @xas is flagged as needing memory to continue. The caller + * should drop the lock and call xas_nomem(). If xas_nomem() succeeds, + * the caller should retry the operation. + * + * Forward progress is guaranteed as one node is allocated here and + * stored in the xa_state where it will be found by xas_alloc(). More + * nodes will likely be found in the slab allocator, but we do not tie + * them up here. + * + * Return: true if memory was needed, and was successfully allocated. + */ +bool xas_nomem(struct xa_state *xas, gfp_t gfp) +{ + if (xas->xa_node != XA_ERROR(-ENOMEM)) { + xas_destroy(xas); + return false; + } + xas->xa_alloc = kmem_cache_alloc(radix_tree_node_cachep, gfp); + if (!xas->xa_alloc) + return false; + XA_NODE_BUG_ON(xas->xa_alloc, !list_empty(&xas->xa_alloc->private_list)); + xas->xa_node = XAS_RESTART; + return true; +} +EXPORT_SYMBOL_GPL(xas_nomem); + +/* + * __xas_nomem() - Drop locks and allocate memory if needed. + * @xas: XArray operation state. + * @gfp: Memory allocation flags. + * + * Internal variant of xas_nomem(). + * + * Return: true if memory was needed, and was successfully allocated. + */ +static bool __xas_nomem(struct xa_state *xas, gfp_t gfp) + __must_hold(xas->xa->xa_lock) +{ + unsigned int lock_type = xa_lock_type(xas->xa); + + if (xas->xa_node != XA_ERROR(-ENOMEM)) { + xas_destroy(xas); + return false; + } + if (gfpflags_allow_blocking(gfp)) { + xas_unlock_type(xas, lock_type); + xas->xa_alloc = kmem_cache_alloc(radix_tree_node_cachep, gfp); + xas_lock_type(xas, lock_type); + } else { + xas->xa_alloc = kmem_cache_alloc(radix_tree_node_cachep, gfp); + } + if (!xas->xa_alloc) + return false; + XA_NODE_BUG_ON(xas->xa_alloc, !list_empty(&xas->xa_alloc->private_list)); + xas->xa_node = XAS_RESTART; + return true; +} + +static void xas_update(struct xa_state *xas, struct xa_node *node) +{ + if (xas->xa_update) + xas->xa_update(node); + else + XA_NODE_BUG_ON(node, !list_empty(&node->private_list)); +} + +static void *xas_alloc(struct xa_state *xas, unsigned int shift) +{ + struct xa_node *parent = xas->xa_node; + struct xa_node *node = xas->xa_alloc; + + if (xas_invalid(xas)) + return NULL; + + if (node) { + xas->xa_alloc = NULL; + } else { + node = kmem_cache_alloc(radix_tree_node_cachep, + GFP_NOWAIT | __GFP_NOWARN); + if (!node) { + xas_set_err(xas, -ENOMEM); + return NULL; + } + } + + if (parent) { + node->offset = xas->xa_offset; + parent->count++; + XA_NODE_BUG_ON(node, parent->count > XA_CHUNK_SIZE); + xas_update(xas, parent); + } + XA_NODE_BUG_ON(node, shift > BITS_PER_LONG); + XA_NODE_BUG_ON(node, !list_empty(&node->private_list)); + node->shift = shift; + node->count = 0; + node->nr_values = 0; + RCU_INIT_POINTER(node->parent, xas->xa_node); + node->array = xas->xa; + + return node; +} + +#ifdef CONFIG_XARRAY_MULTI +/* Returns the number of indices covered by a given xa_state */ +static unsigned long xas_size(const struct xa_state *xas) +{ + return (xas->xa_sibs + 1UL) << xas->xa_shift; +} +#endif + +/* + * Use this to calculate the maximum index that will need to be created + * in order to add the entry described by @xas. Because we cannot store a + * multiple-index entry at index 0, the calculation is a little more complex + * than you might expect. + */ +static unsigned long xas_max(struct xa_state *xas) +{ + unsigned long max = xas->xa_index; + +#ifdef CONFIG_XARRAY_MULTI + if (xas->xa_shift || xas->xa_sibs) { + unsigned long mask = xas_size(xas) - 1; + max |= mask; + if (mask == max) + max++; + } +#endif + + return max; +} + +/* The maximum index that can be contained in the array without expanding it */ +static unsigned long max_index(void *entry) +{ + if (!xa_is_node(entry)) + return 0; + return (XA_CHUNK_SIZE << xa_to_node(entry)->shift) - 1; +} + +static void xas_shrink(struct xa_state *xas) +{ + struct xarray *xa = xas->xa; + struct xa_node *node = xas->xa_node; + + for (;;) { + void *entry; + + XA_NODE_BUG_ON(node, node->count > XA_CHUNK_SIZE); + if (node->count != 1) + break; + entry = xa_entry_locked(xa, node, 0); + if (!entry) + break; + if (!xa_is_node(entry) && node->shift) + break; + xas->xa_node = XAS_BOUNDS; + + RCU_INIT_POINTER(xa->xa_head, entry); + if (xa_track_free(xa) && !node_get_mark(node, 0, XA_FREE_MARK)) + xa_mark_clear(xa, XA_FREE_MARK); + + node->count = 0; + node->nr_values = 0; + if (!xa_is_node(entry)) + RCU_INIT_POINTER(node->slots[0], XA_RETRY_ENTRY); + xas_update(xas, node); + xa_node_free(node); + if (!xa_is_node(entry)) + break; + node = xa_to_node(entry); + node->parent = NULL; + } +} + +/* + * xas_delete_node() - Attempt to delete an xa_node + * @xas: Array operation state. + * + * Attempts to delete the @xas->xa_node. This will fail if xa->node has + * a non-zero reference count. + */ +static void xas_delete_node(struct xa_state *xas) +{ + struct xa_node *node = xas->xa_node; + + for (;;) { + struct xa_node *parent; + + XA_NODE_BUG_ON(node, node->count > XA_CHUNK_SIZE); + if (node->count) + break; + + parent = xa_parent_locked(xas->xa, node); + xas->xa_node = parent; + xas->xa_offset = node->offset; + xa_node_free(node); + + if (!parent) { + xas->xa->xa_head = NULL; + xas->xa_node = XAS_BOUNDS; + return; + } + + parent->slots[xas->xa_offset] = NULL; + parent->count--; + XA_NODE_BUG_ON(parent, parent->count > XA_CHUNK_SIZE); + node = parent; + xas_update(xas, node); + } + + if (!node->parent) + xas_shrink(xas); +} + +/** + * xas_free_nodes() - Free this node and all nodes that it references + * @xas: Array operation state. + * @top: Node to free + * + * This node has been removed from the tree. We must now free it and all + * of its subnodes. There may be RCU walkers with references into the tree, + * so we must replace all entries with retry markers. + */ +static void xas_free_nodes(struct xa_state *xas, struct xa_node *top) +{ + unsigned int offset = 0; + struct xa_node *node = top; + + for (;;) { + void *entry = xa_entry_locked(xas->xa, node, offset); + + if (xa_is_node(entry)) { + node = xa_to_node(entry); + offset = 0; + continue; + } + if (entry) + RCU_INIT_POINTER(node->slots[offset], XA_RETRY_ENTRY); + offset++; + while (offset == XA_CHUNK_SIZE) { + struct xa_node *parent; + + parent = xa_parent_locked(xas->xa, node); + offset = node->offset + 1; + node->count = 0; + node->nr_values = 0; + xas_update(xas, node); + xa_node_free(node); + if (node == top) + return; + node = parent; + } + } +} + +/* + * xas_expand adds nodes to the head of the tree until it has reached + * sufficient height to be able to contain @xas->xa_index + */ +static int xas_expand(struct xa_state *xas, void *head) +{ + struct xarray *xa = xas->xa; + struct xa_node *node = NULL; + unsigned int shift = 0; + unsigned long max = xas_max(xas); + + if (!head) { + if (max == 0) + return 0; + while ((max >> shift) >= XA_CHUNK_SIZE) + shift += XA_CHUNK_SHIFT; + return shift + XA_CHUNK_SHIFT; + } else if (xa_is_node(head)) { + node = xa_to_node(head); + shift = node->shift + XA_CHUNK_SHIFT; + } + xas->xa_node = NULL; + + while (max > max_index(head)) { + xa_mark_t mark = 0; + + XA_NODE_BUG_ON(node, shift > BITS_PER_LONG); + node = xas_alloc(xas, shift); + if (!node) + return -ENOMEM; + + node->count = 1; + if (xa_is_value(head)) + node->nr_values = 1; + RCU_INIT_POINTER(node->slots[0], head); + + /* Propagate the aggregated mark info to the new child */ + for (;;) { + if (xa_track_free(xa) && mark == XA_FREE_MARK) { + node_mark_all(node, XA_FREE_MARK); + if (!xa_marked(xa, XA_FREE_MARK)) { + node_clear_mark(node, 0, XA_FREE_MARK); + xa_mark_set(xa, XA_FREE_MARK); + } + } else if (xa_marked(xa, mark)) { + node_set_mark(node, 0, mark); + } + if (mark == XA_MARK_MAX) + break; + mark_inc(mark); + } + + /* + * Now that the new node is fully initialised, we can add + * it to the tree + */ + if (xa_is_node(head)) { + xa_to_node(head)->offset = 0; + rcu_assign_pointer(xa_to_node(head)->parent, node); + } + head = xa_mk_node(node); + rcu_assign_pointer(xa->xa_head, head); + xas_update(xas, node); + + shift += XA_CHUNK_SHIFT; + } + + xas->xa_node = node; + return shift; +} + +/* + * xas_create() - Create a slot to store an entry in. + * @xas: XArray operation state. + * + * Most users will not need to call this function directly, as it is called + * by xas_store(). It is useful for doing conditional store operations + * (see the xa_cmpxchg() implementation for an example). + * + * Return: If the slot already existed, returns the contents of this slot. + * If the slot was newly created, returns NULL. If it failed to create the + * slot, returns NULL and indicates the error in @xas. + */ +static void *xas_create(struct xa_state *xas) +{ + struct xarray *xa = xas->xa; + void *entry; + void __rcu **slot; + struct xa_node *node = xas->xa_node; + int shift; + unsigned int order = xas->xa_shift; + + if (xas_top(node)) { + entry = xa_head_locked(xa); + xas->xa_node = NULL; + shift = xas_expand(xas, entry); + if (shift < 0) + return NULL; + entry = xa_head_locked(xa); + slot = &xa->xa_head; + } else if (xas_error(xas)) { + return NULL; + } else if (node) { + unsigned int offset = xas->xa_offset; + + shift = node->shift; + entry = xa_entry_locked(xa, node, offset); + slot = &node->slots[offset]; + } else { + shift = 0; + entry = xa_head_locked(xa); + slot = &xa->xa_head; + } + + while (shift > order) { + shift -= XA_CHUNK_SHIFT; + if (!entry) { + node = xas_alloc(xas, shift); + if (!node) + break; + if (xa_track_free(xa)) + node_mark_all(node, XA_FREE_MARK); + rcu_assign_pointer(*slot, xa_mk_node(node)); + } else if (xa_is_node(entry)) { + node = xa_to_node(entry); + } else { + break; + } + entry = xas_descend(xas, node); + slot = &node->slots[xas->xa_offset]; + } + + return entry; +} + +/** + * xas_create_range() - Ensure that stores to this range will succeed + * @xas: XArray operation state. + * + * Creates all of the slots in the range covered by @xas. Sets @xas to + * create single-index entries and positions it at the beginning of the + * range. This is for the benefit of users which have not yet been + * converted to use multi-index entries. + */ +void xas_create_range(struct xa_state *xas) +{ + unsigned long index = xas->xa_index; + unsigned char shift = xas->xa_shift; + unsigned char sibs = xas->xa_sibs; + + xas->xa_index |= ((sibs + 1) << shift) - 1; + if (xas_is_node(xas) && xas->xa_node->shift == xas->xa_shift) + xas->xa_offset |= sibs; + xas->xa_shift = 0; + xas->xa_sibs = 0; + + for (;;) { + xas_create(xas); + if (xas_error(xas)) + goto restore; + if (xas->xa_index <= (index | XA_CHUNK_MASK)) + goto success; + xas->xa_index -= XA_CHUNK_SIZE; + + for (;;) { + struct xa_node *node = xas->xa_node; + xas->xa_node = xa_parent_locked(xas->xa, node); + xas->xa_offset = node->offset - 1; + if (node->offset != 0) + break; + } + } + +restore: + xas->xa_shift = shift; + xas->xa_sibs = sibs; + xas->xa_index = index; + return; +success: + xas->xa_index = index; + if (xas->xa_node) + xas_set_offset(xas); +} +EXPORT_SYMBOL_GPL(xas_create_range); + +static void update_node(struct xa_state *xas, struct xa_node *node, + int count, int values) +{ + if (!node || (!count && !values)) + return; + + node->count += count; + node->nr_values += values; + XA_NODE_BUG_ON(node, node->count > XA_CHUNK_SIZE); + XA_NODE_BUG_ON(node, node->nr_values > XA_CHUNK_SIZE); + xas_update(xas, node); + if (count < 0) + xas_delete_node(xas); +} + +/** + * xas_store() - Store this entry in the XArray. + * @xas: XArray operation state. + * @entry: New entry. + * + * If @xas is operating on a multi-index entry, the entry returned by this + * function is essentially meaningless (it may be an internal entry or it + * may be %NULL, even if there are non-NULL entries at some of the indices + * covered by the range). This is not a problem for any current users, + * and can be changed if needed. + * + * Return: The old entry at this index. + */ +void *xas_store(struct xa_state *xas, void *entry) +{ + struct xa_node *node; + void __rcu **slot = &xas->xa->xa_head; + unsigned int offset, max; + int count = 0; + int values = 0; + void *first, *next; + bool value = xa_is_value(entry); + + if (entry) + first = xas_create(xas); + else + first = xas_load(xas); + + if (xas_invalid(xas)) + return first; + node = xas->xa_node; + if (node && (xas->xa_shift < node->shift)) + xas->xa_sibs = 0; + if ((first == entry) && !xas->xa_sibs) + return first; + + next = first; + offset = xas->xa_offset; + max = xas->xa_offset + xas->xa_sibs; + if (node) { + slot = &node->slots[offset]; + if (xas->xa_sibs) + xas_squash_marks(xas); + } + if (!entry) + xas_init_marks(xas); + + for (;;) { + /* + * Must clear the marks before setting the entry to NULL, + * otherwise xas_for_each_marked may find a NULL entry and + * stop early. rcu_assign_pointer contains a release barrier + * so the mark clearing will appear to happen before the + * entry is set to NULL. + */ + rcu_assign_pointer(*slot, entry); + if (xa_is_node(next)) + xas_free_nodes(xas, xa_to_node(next)); + if (!node) + break; + count += !next - !entry; + values += !xa_is_value(first) - !value; + if (entry) { + if (offset == max) + break; + if (!xa_is_sibling(entry)) + entry = xa_mk_sibling(xas->xa_offset); + } else { + if (offset == XA_CHUNK_MASK) + break; + } + next = xa_entry_locked(xas->xa, node, ++offset); + if (!xa_is_sibling(next)) { + if (!entry && (offset > max)) + break; + first = next; + } + slot++; + } + + update_node(xas, node, count, values); + return first; +} +EXPORT_SYMBOL_GPL(xas_store); + +/** + * xas_get_mark() - Returns the state of this mark. + * @xas: XArray operation state. + * @mark: Mark number. + * + * Return: true if the mark is set, false if the mark is clear or @xas + * is in an error state. + */ +bool xas_get_mark(const struct xa_state *xas, xa_mark_t mark) +{ + if (xas_invalid(xas)) + return false; + if (!xas->xa_node) + return xa_marked(xas->xa, mark); + return node_get_mark(xas->xa_node, xas->xa_offset, mark); +} +EXPORT_SYMBOL_GPL(xas_get_mark); + +/** + * xas_set_mark() - Sets the mark on this entry and its parents. + * @xas: XArray operation state. + * @mark: Mark number. + * + * Sets the specified mark on this entry, and walks up the tree setting it + * on all the ancestor entries. Does nothing if @xas has not been walked to + * an entry, or is in an error state. + */ +void xas_set_mark(const struct xa_state *xas, xa_mark_t mark) +{ + struct xa_node *node = xas->xa_node; + unsigned int offset = xas->xa_offset; + + if (xas_invalid(xas)) + return; + + while (node) { + if (node_set_mark(node, offset, mark)) + return; + offset = node->offset; + node = xa_parent_locked(xas->xa, node); + } + + if (!xa_marked(xas->xa, mark)) + xa_mark_set(xas->xa, mark); +} +EXPORT_SYMBOL_GPL(xas_set_mark); + +/** + * xas_clear_mark() - Clears the mark on this entry and its parents. + * @xas: XArray operation state. + * @mark: Mark number. + * + * Clears the specified mark on this entry, and walks back to the head + * attempting to clear it on all the ancestor entries. Does nothing if + * @xas has not been walked to an entry, or is in an error state. + */ +void xas_clear_mark(const struct xa_state *xas, xa_mark_t mark) +{ + struct xa_node *node = xas->xa_node; + unsigned int offset = xas->xa_offset; + + if (xas_invalid(xas)) + return; + + while (node) { + if (!node_clear_mark(node, offset, mark)) + return; + if (node_any_mark(node, mark)) + return; + + offset = node->offset; + node = xa_parent_locked(xas->xa, node); + } + + if (xa_marked(xas->xa, mark)) + xa_mark_clear(xas->xa, mark); +} +EXPORT_SYMBOL_GPL(xas_clear_mark); + +/** + * xas_init_marks() - Initialise all marks for the entry + * @xas: Array operations state. + * + * Initialise all marks for the entry specified by @xas. If we're tracking + * free entries with a mark, we need to set it on all entries. All other + * marks are cleared. + * + * This implementation is not as efficient as it could be; we may walk + * up the tree multiple times. + */ +void xas_init_marks(const struct xa_state *xas) +{ + xa_mark_t mark = 0; + + for (;;) { + if (xa_track_free(xas->xa) && mark == XA_FREE_MARK) + xas_set_mark(xas, mark); + else + xas_clear_mark(xas, mark); + if (mark == XA_MARK_MAX) + break; + mark_inc(mark); + } +} +EXPORT_SYMBOL_GPL(xas_init_marks); + +/** + * xas_pause() - Pause a walk to drop a lock. + * @xas: XArray operation state. + * + * Some users need to pause a walk and drop the lock they're holding in + * order to yield to a higher priority thread or carry out an operation + * on an entry. Those users should call this function before they drop + * the lock. It resets the @xas to be suitable for the next iteration + * of the loop after the user has reacquired the lock. If most entries + * found during a walk require you to call xas_pause(), the xa_for_each() + * iterator may be more appropriate. + * + * Note that xas_pause() only works for forward iteration. If a user needs + * to pause a reverse iteration, we will need a xas_pause_rev(). + */ +void xas_pause(struct xa_state *xas) +{ + struct xa_node *node = xas->xa_node; + + if (xas_invalid(xas)) + return; + + if (node) { + unsigned int offset = xas->xa_offset; + while (++offset < XA_CHUNK_SIZE) { + if (!xa_is_sibling(xa_entry(xas->xa, node, offset))) + break; + } + xas->xa_index += (offset - xas->xa_offset) << node->shift; + } else { + xas->xa_index++; + } + xas->xa_node = XAS_RESTART; +} +EXPORT_SYMBOL_GPL(xas_pause); + +/* + * __xas_prev() - Find the previous entry in the XArray. + * @xas: XArray operation state. + * + * Helper function for xas_prev() which handles all the complex cases + * out of line. + */ +void *__xas_prev(struct xa_state *xas) +{ + void *entry; + + if (!xas_frozen(xas->xa_node)) + xas->xa_index--; + if (xas_not_node(xas->xa_node)) + return xas_load(xas); + + if (xas->xa_offset != get_offset(xas->xa_index, xas->xa_node)) + xas->xa_offset--; + + while (xas->xa_offset == 255) { + xas->xa_offset = xas->xa_node->offset - 1; + xas->xa_node = xa_parent(xas->xa, xas->xa_node); + if (!xas->xa_node) + return set_bounds(xas); + } + + for (;;) { + entry = xa_entry(xas->xa, xas->xa_node, xas->xa_offset); + if (!xa_is_node(entry)) + return entry; + + xas->xa_node = xa_to_node(entry); + xas_set_offset(xas); + } +} +EXPORT_SYMBOL_GPL(__xas_prev); + +/* + * __xas_next() - Find the next entry in the XArray. + * @xas: XArray operation state. + * + * Helper function for xas_next() which handles all the complex cases + * out of line. + */ +void *__xas_next(struct xa_state *xas) +{ + void *entry; + + if (!xas_frozen(xas->xa_node)) + xas->xa_index++; + if (xas_not_node(xas->xa_node)) + return xas_load(xas); + + if (xas->xa_offset != get_offset(xas->xa_index, xas->xa_node)) + xas->xa_offset++; + + while (xas->xa_offset == XA_CHUNK_SIZE) { + xas->xa_offset = xas->xa_node->offset + 1; + xas->xa_node = xa_parent(xas->xa, xas->xa_node); + if (!xas->xa_node) + return set_bounds(xas); + } + + for (;;) { + entry = xa_entry(xas->xa, xas->xa_node, xas->xa_offset); + if (!xa_is_node(entry)) + return entry; + + xas->xa_node = xa_to_node(entry); + xas_set_offset(xas); + } +} +EXPORT_SYMBOL_GPL(__xas_next); + +/** + * xas_find() - Find the next present entry in the XArray. + * @xas: XArray operation state. + * @max: Highest index to return. + * + * If the @xas has not yet been walked to an entry, return the entry + * which has an index >= xas.xa_index. If it has been walked, the entry + * currently being pointed at has been processed, and so we move to the + * next entry. + * + * If no entry is found and the array is smaller than @max, the iterator + * is set to the smallest index not yet in the array. This allows @xas + * to be immediately passed to xas_store(). + * + * Return: The entry, if found, otherwise %NULL. + */ +void *xas_find(struct xa_state *xas, unsigned long max) +{ + void *entry; + + if (xas_error(xas)) + return NULL; + + if (!xas->xa_node) { + xas->xa_index = 1; + return set_bounds(xas); + } else if (xas_top(xas->xa_node)) { + entry = xas_load(xas); + if (entry || xas_not_node(xas->xa_node)) + return entry; + } else if (!xas->xa_node->shift && + xas->xa_offset != (xas->xa_index & XA_CHUNK_MASK)) { + xas->xa_offset = ((xas->xa_index - 1) & XA_CHUNK_MASK) + 1; + } + + xas_advance(xas); + + while (xas->xa_node && (xas->xa_index <= max)) { + if (unlikely(xas->xa_offset == XA_CHUNK_SIZE)) { + xas->xa_offset = xas->xa_node->offset + 1; + xas->xa_node = xa_parent(xas->xa, xas->xa_node); + continue; + } + + entry = xa_entry(xas->xa, xas->xa_node, xas->xa_offset); + if (xa_is_node(entry)) { + xas->xa_node = xa_to_node(entry); + xas->xa_offset = 0; + continue; + } + if (entry && !xa_is_sibling(entry)) + return entry; + + xas_advance(xas); + } + + if (!xas->xa_node) + xas->xa_node = XAS_BOUNDS; + return NULL; +} +EXPORT_SYMBOL_GPL(xas_find); + +/** + * xas_find_marked() - Find the next marked entry in the XArray. + * @xas: XArray operation state. + * @max: Highest index to return. + * @mark: Mark number to search for. + * + * If the @xas has not yet been walked to an entry, return the marked entry + * which has an index >= xas.xa_index. If it has been walked, the entry + * currently being pointed at has been processed, and so we return the + * first marked entry with an index > xas.xa_index. + * + * If no marked entry is found and the array is smaller than @max, @xas is + * set to the bounds state and xas->xa_index is set to the smallest index + * not yet in the array. This allows @xas to be immediately passed to + * xas_store(). + * + * If no entry is found before @max is reached, @xas is set to the restart + * state. + * + * Return: The entry, if found, otherwise %NULL. + */ +void *xas_find_marked(struct xa_state *xas, unsigned long max, xa_mark_t mark) +{ + bool advance = true; + unsigned int offset; + void *entry; + + if (xas_error(xas)) + return NULL; + + if (!xas->xa_node) { + xas->xa_index = 1; + goto out; + } else if (xas_top(xas->xa_node)) { + advance = false; + entry = xa_head(xas->xa); + xas->xa_node = NULL; + if (xas->xa_index > max_index(entry)) + goto bounds; + if (!xa_is_node(entry)) { + if (xa_marked(xas->xa, mark)) + return entry; + xas->xa_index = 1; + goto out; + } + xas->xa_node = xa_to_node(entry); + xas->xa_offset = xas->xa_index >> xas->xa_node->shift; + } + + while (xas->xa_index <= max) { + if (unlikely(xas->xa_offset == XA_CHUNK_SIZE)) { + xas->xa_offset = xas->xa_node->offset + 1; + xas->xa_node = xa_parent(xas->xa, xas->xa_node); + if (!xas->xa_node) + break; + advance = false; + continue; + } + + if (!advance) { + entry = xa_entry(xas->xa, xas->xa_node, xas->xa_offset); + if (xa_is_sibling(entry)) { + xas->xa_offset = xa_to_sibling(entry); + xas_move_index(xas, xas->xa_offset); + } + } + + offset = xas_find_chunk(xas, advance, mark); + if (offset > xas->xa_offset) { + advance = false; + xas_move_index(xas, offset); + /* Mind the wrap */ + if ((xas->xa_index - 1) >= max) + goto max; + xas->xa_offset = offset; + if (offset == XA_CHUNK_SIZE) + continue; + } + + entry = xa_entry(xas->xa, xas->xa_node, xas->xa_offset); + if (!xa_is_node(entry)) + return entry; + xas->xa_node = xa_to_node(entry); + xas_set_offset(xas); + } + +out: + if (!max) + goto max; +bounds: + xas->xa_node = XAS_BOUNDS; + return NULL; +max: + xas->xa_node = XAS_RESTART; + return NULL; +} +EXPORT_SYMBOL_GPL(xas_find_marked); + +/** + * xas_find_conflict() - Find the next present entry in a range. + * @xas: XArray operation state. + * + * The @xas describes both a range and a position within that range. + * + * Context: Any context. Expects xa_lock to be held. + * Return: The next entry in the range covered by @xas or %NULL. + */ +void *xas_find_conflict(struct xa_state *xas) +{ + void *curr; + + if (xas_error(xas)) + return NULL; + + if (!xas->xa_node) + return NULL; + + if (xas_top(xas->xa_node)) { + curr = xas_start(xas); + if (!curr) + return NULL; + while (xa_is_node(curr)) { + struct xa_node *node = xa_to_node(curr); + curr = xas_descend(xas, node); + } + if (curr) + return curr; + } + + if (xas->xa_node->shift > xas->xa_shift) + return NULL; + + for (;;) { + if (xas->xa_node->shift == xas->xa_shift) { + if ((xas->xa_offset & xas->xa_sibs) == xas->xa_sibs) + break; + } else if (xas->xa_offset == XA_CHUNK_MASK) { + xas->xa_offset = xas->xa_node->offset; + xas->xa_node = xa_parent_locked(xas->xa, xas->xa_node); + if (!xas->xa_node) + break; + continue; + } + curr = xa_entry_locked(xas->xa, xas->xa_node, ++xas->xa_offset); + if (xa_is_sibling(curr)) + continue; + while (xa_is_node(curr)) { + xas->xa_node = xa_to_node(curr); + xas->xa_offset = 0; + curr = xa_entry_locked(xas->xa, xas->xa_node, 0); + } + if (curr) + return curr; + } + xas->xa_offset -= xas->xa_sibs; + return NULL; +} +EXPORT_SYMBOL_GPL(xas_find_conflict); + +/** + * xa_init_flags() - Initialise an empty XArray with flags. + * @xa: XArray. + * @flags: XA_FLAG values. + * + * If you need to initialise an XArray with special flags (eg you need + * to take the lock from interrupt context), use this function instead + * of xa_init(). + * + * Context: Any context. + */ +void xa_init_flags(struct xarray *xa, gfp_t flags) +{ + unsigned int lock_type; + static struct lock_class_key xa_lock_irq; + static struct lock_class_key xa_lock_bh; + + spin_lock_init(&xa->xa_lock); + xa->xa_flags = flags; + xa->xa_head = NULL; + + lock_type = xa_lock_type(xa); + if (lock_type == XA_LOCK_IRQ) + lockdep_set_class(&xa->xa_lock, &xa_lock_irq); + else if (lock_type == XA_LOCK_BH) + lockdep_set_class(&xa->xa_lock, &xa_lock_bh); +} +EXPORT_SYMBOL(xa_init_flags); + +/** + * xa_load() - Load an entry from an XArray. + * @xa: XArray. + * @index: index into array. + * + * Context: Any context. Takes and releases the RCU lock. + * Return: The entry at @index in @xa. + */ +void *xa_load(struct xarray *xa, unsigned long index) +{ + XA_STATE(xas, xa, index); + void *entry; + + rcu_read_lock(); + do { + entry = xas_load(&xas); + if (xa_is_zero(entry)) + entry = NULL; + } while (xas_retry(&xas, entry)); + rcu_read_unlock(); + + return entry; +} +EXPORT_SYMBOL(xa_load); + +static void *xas_result(struct xa_state *xas, void *curr) +{ + if (xa_is_zero(curr)) + return NULL; + XA_NODE_BUG_ON(xas->xa_node, xa_is_internal(curr)); + if (xas_error(xas)) + curr = xas->xa_node; + return curr; +} + +/** + * __xa_erase() - Erase this entry from the XArray while locked. + * @xa: XArray. + * @index: Index into array. + * + * If the entry at this index is a multi-index entry then all indices will + * be erased, and the entry will no longer be a multi-index entry. + * This function expects the xa_lock to be held on entry. + * + * Context: Any context. Expects xa_lock to be held on entry. May + * release and reacquire xa_lock if @gfp flags permit. + * Return: The old entry at this index. + */ +void *__xa_erase(struct xarray *xa, unsigned long index) +{ + XA_STATE(xas, xa, index); + return xas_result(&xas, xas_store(&xas, NULL)); +} +EXPORT_SYMBOL_GPL(__xa_erase); + +/** + * xa_store() - Store this entry in the XArray. + * @xa: XArray. + * @index: Index into array. + * @entry: New entry. + * @gfp: Memory allocation flags. + * + * After this function returns, loads from this index will return @entry. + * Storing into an existing multislot entry updates the entry of every index. + * The marks associated with @index are unaffected unless @entry is %NULL. + * + * Context: Process context. Takes and releases the xa_lock. May sleep + * if the @gfp flags permit. + * Return: The old entry at this index on success, xa_err(-EINVAL) if @entry + * cannot be stored in an XArray, or xa_err(-ENOMEM) if memory allocation + * failed. + */ +void *xa_store(struct xarray *xa, unsigned long index, void *entry, gfp_t gfp) +{ + XA_STATE(xas, xa, index); + void *curr; + + if (WARN_ON_ONCE(xa_is_internal(entry))) + return XA_ERROR(-EINVAL); + + do { + xas_lock(&xas); + curr = xas_store(&xas, entry); + if (xa_track_free(xa) && entry) + xas_clear_mark(&xas, XA_FREE_MARK); + xas_unlock(&xas); + } while (xas_nomem(&xas, gfp)); + + return xas_result(&xas, curr); +} +EXPORT_SYMBOL(xa_store); + +/** + * __xa_store() - Store this entry in the XArray. + * @xa: XArray. + * @index: Index into array. + * @entry: New entry. + * @gfp: Memory allocation flags. + * + * You must already be holding the xa_lock when calling this function. + * It will drop the lock if needed to allocate memory, and then reacquire + * it afterwards. + * + * Context: Any context. Expects xa_lock to be held on entry. May + * release and reacquire xa_lock if @gfp flags permit. + * Return: The old entry at this index or xa_err() if an error happened. + */ +void *__xa_store(struct xarray *xa, unsigned long index, void *entry, gfp_t gfp) +{ + XA_STATE(xas, xa, index); + void *curr; + + if (WARN_ON_ONCE(xa_is_internal(entry))) + return XA_ERROR(-EINVAL); + + do { + curr = xas_store(&xas, entry); + if (xa_track_free(xa) && entry) + xas_clear_mark(&xas, XA_FREE_MARK); + } while (__xas_nomem(&xas, gfp)); + + return xas_result(&xas, curr); +} +EXPORT_SYMBOL(__xa_store); + +/** + * xa_cmpxchg() - Conditionally replace an entry in the XArray. + * @xa: XArray. + * @index: Index into array. + * @old: Old value to test against. + * @entry: New value to place in array. + * @gfp: Memory allocation flags. + * + * If the entry at @index is the same as @old, replace it with @entry. + * If the return value is equal to @old, then the exchange was successful. + * + * Context: Process context. Takes and releases the xa_lock. May sleep + * if the @gfp flags permit. + * Return: The old value at this index or xa_err() if an error happened. + */ +void *xa_cmpxchg(struct xarray *xa, unsigned long index, + void *old, void *entry, gfp_t gfp) +{ + XA_STATE(xas, xa, index); + void *curr; + + if (WARN_ON_ONCE(xa_is_internal(entry))) + return XA_ERROR(-EINVAL); + + do { + xas_lock(&xas); + curr = xas_load(&xas); + if (curr == XA_ZERO_ENTRY) + curr = NULL; + if (curr == old) { + xas_store(&xas, entry); + if (xa_track_free(xa) && entry) + xas_clear_mark(&xas, XA_FREE_MARK); + } + xas_unlock(&xas); + } while (xas_nomem(&xas, gfp)); + + return xas_result(&xas, curr); +} +EXPORT_SYMBOL(xa_cmpxchg); + +/** + * __xa_cmpxchg() - Store this entry in the XArray. + * @xa: XArray. + * @index: Index into array. + * @old: Old value to test against. + * @entry: New entry. + * @gfp: Memory allocation flags. + * + * You must already be holding the xa_lock when calling this function. + * It will drop the lock if needed to allocate memory, and then reacquire + * it afterwards. + * + * Context: Any context. Expects xa_lock to be held on entry. May + * release and reacquire xa_lock if @gfp flags permit. + * Return: The old entry at this index or xa_err() if an error happened. + */ +void *__xa_cmpxchg(struct xarray *xa, unsigned long index, + void *old, void *entry, gfp_t gfp) +{ + XA_STATE(xas, xa, index); + void *curr; + + if (WARN_ON_ONCE(xa_is_internal(entry))) + return XA_ERROR(-EINVAL); + + do { + curr = xas_load(&xas); + if (curr == XA_ZERO_ENTRY) + curr = NULL; + if (curr == old) { + xas_store(&xas, entry); + if (xa_track_free(xa) && entry) + xas_clear_mark(&xas, XA_FREE_MARK); + } + } while (__xas_nomem(&xas, gfp)); + + return xas_result(&xas, curr); +} +EXPORT_SYMBOL(__xa_cmpxchg); + +/** + * xa_reserve() - Reserve this index in the XArray. + * @xa: XArray. + * @index: Index into array. + * @gfp: Memory allocation flags. + * + * Ensures there is somewhere to store an entry at @index in the array. + * If there is already something stored at @index, this function does + * nothing. If there was nothing there, the entry is marked as reserved. + * Loads from @index will continue to see a %NULL pointer until a + * subsequent store to @index. + * + * If you do not use the entry that you have reserved, call xa_release() + * or xa_erase() to free any unnecessary memory. + * + * Context: Process context. Takes and releases the xa_lock, IRQ or BH safe + * if specified in XArray flags. May sleep if the @gfp flags permit. + * Return: 0 if the reservation succeeded or -ENOMEM if it failed. + */ +int xa_reserve(struct xarray *xa, unsigned long index, gfp_t gfp) +{ + XA_STATE(xas, xa, index); + unsigned int lock_type = xa_lock_type(xa); + void *curr; + + do { + xas_lock_type(&xas, lock_type); + curr = xas_load(&xas); + if (!curr) + xas_store(&xas, XA_ZERO_ENTRY); + xas_unlock_type(&xas, lock_type); + } while (xas_nomem(&xas, gfp)); + + return xas_error(&xas); +} +EXPORT_SYMBOL(xa_reserve); + +#ifdef CONFIG_XARRAY_MULTI +static void xas_set_range(struct xa_state *xas, unsigned long first, + unsigned long last) +{ + unsigned int shift = 0; + unsigned long sibs = last - first; + unsigned int offset = XA_CHUNK_MASK; + + xas_set(xas, first); + + while ((first & XA_CHUNK_MASK) == 0) { + if (sibs < XA_CHUNK_MASK) + break; + if ((sibs == XA_CHUNK_MASK) && (offset < XA_CHUNK_MASK)) + break; + shift += XA_CHUNK_SHIFT; + if (offset == XA_CHUNK_MASK) + offset = sibs & XA_CHUNK_MASK; + sibs >>= XA_CHUNK_SHIFT; + first >>= XA_CHUNK_SHIFT; + } + + offset = first & XA_CHUNK_MASK; + if (offset + sibs > XA_CHUNK_MASK) + sibs = XA_CHUNK_MASK - offset; + if ((((first + sibs + 1) << shift) - 1) > last) + sibs -= 1; + + xas->xa_shift = shift; + xas->xa_sibs = sibs; +} + +/** + * xa_store_range() - Store this entry at a range of indices in the XArray. + * @xa: XArray. + * @first: First index to affect. + * @last: Last index to affect. + * @entry: New entry. + * @gfp: Memory allocation flags. + * + * After this function returns, loads from any index between @first and @last, + * inclusive will return @entry. + * Storing into an existing multislot entry updates the entry of every index. + * The marks associated with @index are unaffected unless @entry is %NULL. + * + * Context: Process context. Takes and releases the xa_lock. May sleep + * if the @gfp flags permit. + * Return: %NULL on success, xa_err(-EINVAL) if @entry cannot be stored in + * an XArray, or xa_err(-ENOMEM) if memory allocation failed. + */ +void *xa_store_range(struct xarray *xa, unsigned long first, + unsigned long last, void *entry, gfp_t gfp) +{ + XA_STATE(xas, xa, 0); + + if (WARN_ON_ONCE(xa_is_internal(entry))) + return XA_ERROR(-EINVAL); + if (last < first) + return XA_ERROR(-EINVAL); + + do { + xas_lock(&xas); + if (entry) { + unsigned int order = (last == ~0UL) ? 64 : + ilog2(last + 1); + xas_set_order(&xas, last, order); + xas_create(&xas); + if (xas_error(&xas)) + goto unlock; + } + do { + xas_set_range(&xas, first, last); + xas_store(&xas, entry); + if (xas_error(&xas)) + goto unlock; + first += xas_size(&xas); + } while (first <= last); +unlock: + xas_unlock(&xas); + } while (xas_nomem(&xas, gfp)); + + return xas_result(&xas, NULL); +} +EXPORT_SYMBOL(xa_store_range); +#endif /* CONFIG_XARRAY_MULTI */ + +/** + * __xa_alloc() - Find somewhere to store this entry in the XArray. + * @xa: XArray. + * @id: Pointer to ID. + * @max: Maximum ID to allocate (inclusive). + * @entry: New entry. + * @gfp: Memory allocation flags. + * + * Allocates an unused ID in the range specified by @id and @max. + * Updates the @id pointer with the index, then stores the entry at that + * index. A concurrent lookup will not see an uninitialised @id. + * + * Context: Any context. Expects xa_lock to be held on entry. May + * release and reacquire xa_lock if @gfp flags permit. + * Return: 0 on success, -ENOMEM if memory allocation fails or -ENOSPC if + * there is no more space in the XArray. + */ +int __xa_alloc(struct xarray *xa, u32 *id, u32 max, void *entry, gfp_t gfp) +{ + XA_STATE(xas, xa, 0); + int err; + + if (WARN_ON_ONCE(xa_is_internal(entry))) + return -EINVAL; + if (WARN_ON_ONCE(!xa_track_free(xa))) + return -EINVAL; + + if (!entry) + entry = XA_ZERO_ENTRY; + + do { + xas.xa_index = *id; + xas_find_marked(&xas, max, XA_FREE_MARK); + if (xas.xa_node == XAS_RESTART) + xas_set_err(&xas, -ENOSPC); + xas_store(&xas, entry); + xas_clear_mark(&xas, XA_FREE_MARK); + } while (__xas_nomem(&xas, gfp)); + + err = xas_error(&xas); + if (!err) + *id = xas.xa_index; + return err; +} +EXPORT_SYMBOL(__xa_alloc); + +/** + * __xa_set_mark() - Set this mark on this entry while locked. + * @xa: XArray. + * @index: Index of entry. + * @mark: Mark number. + * + * Attempting to set a mark on a NULL entry does not succeed. + * + * Context: Any context. Expects xa_lock to be held on entry. + */ +void __xa_set_mark(struct xarray *xa, unsigned long index, xa_mark_t mark) +{ + XA_STATE(xas, xa, index); + void *entry = xas_load(&xas); + + if (entry) + xas_set_mark(&xas, mark); +} +EXPORT_SYMBOL_GPL(__xa_set_mark); + +/** + * __xa_clear_mark() - Clear this mark on this entry while locked. + * @xa: XArray. + * @index: Index of entry. + * @mark: Mark number. + * + * Context: Any context. Expects xa_lock to be held on entry. + */ +void __xa_clear_mark(struct xarray *xa, unsigned long index, xa_mark_t mark) +{ + XA_STATE(xas, xa, index); + void *entry = xas_load(&xas); + + if (entry) + xas_clear_mark(&xas, mark); +} +EXPORT_SYMBOL_GPL(__xa_clear_mark); + +/** + * xa_get_mark() - Inquire whether this mark is set on this entry. + * @xa: XArray. + * @index: Index of entry. + * @mark: Mark number. + * + * This function uses the RCU read lock, so the result may be out of date + * by the time it returns. If you need the result to be stable, use a lock. + * + * Context: Any context. Takes and releases the RCU lock. + * Return: True if the entry at @index has this mark set, false if it doesn't. + */ +bool xa_get_mark(struct xarray *xa, unsigned long index, xa_mark_t mark) +{ + XA_STATE(xas, xa, index); + void *entry; + + rcu_read_lock(); + entry = xas_start(&xas); + while (xas_get_mark(&xas, mark)) { + if (!xa_is_node(entry)) + goto found; + entry = xas_descend(&xas, xa_to_node(entry)); + } + rcu_read_unlock(); + return false; + found: + rcu_read_unlock(); + return true; +} +EXPORT_SYMBOL(xa_get_mark); + +/** + * xa_set_mark() - Set this mark on this entry. + * @xa: XArray. + * @index: Index of entry. + * @mark: Mark number. + * + * Attempting to set a mark on a NULL entry does not succeed. + * + * Context: Process context. Takes and releases the xa_lock. + */ +void xa_set_mark(struct xarray *xa, unsigned long index, xa_mark_t mark) +{ + xa_lock(xa); + __xa_set_mark(xa, index, mark); + xa_unlock(xa); +} +EXPORT_SYMBOL(xa_set_mark); + +/** + * xa_clear_mark() - Clear this mark on this entry. + * @xa: XArray. + * @index: Index of entry. + * @mark: Mark number. + * + * Clearing a mark always succeeds. + * + * Context: Process context. Takes and releases the xa_lock. + */ +void xa_clear_mark(struct xarray *xa, unsigned long index, xa_mark_t mark) +{ + xa_lock(xa); + __xa_clear_mark(xa, index, mark); + xa_unlock(xa); +} +EXPORT_SYMBOL(xa_clear_mark); + +/** + * xa_find() - Search the XArray for an entry. + * @xa: XArray. + * @indexp: Pointer to an index. + * @max: Maximum index to search to. + * @filter: Selection criterion. + * + * Finds the entry in @xa which matches the @filter, and has the lowest + * index that is at least @indexp and no more than @max. + * If an entry is found, @indexp is updated to be the index of the entry. + * This function is protected by the RCU read lock, so it may not find + * entries which are being simultaneously added. It will not return an + * %XA_RETRY_ENTRY; if you need to see retry entries, use xas_find(). + * + * Context: Any context. Takes and releases the RCU lock. + * Return: The entry, if found, otherwise %NULL. + */ +void *xa_find(struct xarray *xa, unsigned long *indexp, + unsigned long max, xa_mark_t filter) +{ + XA_STATE(xas, xa, *indexp); + void *entry; + + rcu_read_lock(); + do { + if ((__force unsigned int)filter < XA_MAX_MARKS) + entry = xas_find_marked(&xas, max, filter); + else + entry = xas_find(&xas, max); + } while (xas_retry(&xas, entry)); + rcu_read_unlock(); + + if (entry) + *indexp = xas.xa_index; + return entry; +} +EXPORT_SYMBOL(xa_find); + +/** + * xa_find_after() - Search the XArray for a present entry. + * @xa: XArray. + * @indexp: Pointer to an index. + * @max: Maximum index to search to. + * @filter: Selection criterion. + * + * Finds the entry in @xa which matches the @filter and has the lowest + * index that is above @indexp and no more than @max. + * If an entry is found, @indexp is updated to be the index of the entry. + * This function is protected by the RCU read lock, so it may miss entries + * which are being simultaneously added. It will not return an + * %XA_RETRY_ENTRY; if you need to see retry entries, use xas_find(). + * + * Context: Any context. Takes and releases the RCU lock. + * Return: The pointer, if found, otherwise %NULL. + */ +void *xa_find_after(struct xarray *xa, unsigned long *indexp, + unsigned long max, xa_mark_t filter) +{ + XA_STATE(xas, xa, *indexp + 1); + void *entry; + + rcu_read_lock(); + for (;;) { + if ((__force unsigned int)filter < XA_MAX_MARKS) + entry = xas_find_marked(&xas, max, filter); + else + entry = xas_find(&xas, max); + if (xas.xa_shift) { + if (xas.xa_index & ((1UL << xas.xa_shift) - 1)) + continue; + } else { + if (xas.xa_offset < (xas.xa_index & XA_CHUNK_MASK)) + continue; + } + if (!xas_retry(&xas, entry)) + break; + } + rcu_read_unlock(); + + if (entry) + *indexp = xas.xa_index; + return entry; +} +EXPORT_SYMBOL(xa_find_after); + +static unsigned int xas_extract_present(struct xa_state *xas, void **dst, + unsigned long max, unsigned int n) +{ + void *entry; + unsigned int i = 0; + + rcu_read_lock(); + xas_for_each(xas, entry, max) { + if (xas_retry(xas, entry)) + continue; + dst[i++] = entry; + if (i == n) + break; + } + rcu_read_unlock(); + + return i; +} + +static unsigned int xas_extract_marked(struct xa_state *xas, void **dst, + unsigned long max, unsigned int n, xa_mark_t mark) +{ + void *entry; + unsigned int i = 0; + + rcu_read_lock(); + xas_for_each_marked(xas, entry, max, mark) { + if (xas_retry(xas, entry)) + continue; + dst[i++] = entry; + if (i == n) + break; + } + rcu_read_unlock(); + + return i; +} + +/** + * xa_extract() - Copy selected entries from the XArray into a normal array. + * @xa: The source XArray to copy from. + * @dst: The buffer to copy entries into. + * @start: The first index in the XArray eligible to be selected. + * @max: The last index in the XArray eligible to be selected. + * @n: The maximum number of entries to copy. + * @filter: Selection criterion. + * + * Copies up to @n entries that match @filter from the XArray. The + * copied entries will have indices between @start and @max, inclusive. + * + * The @filter may be an XArray mark value, in which case entries which are + * marked with that mark will be copied. It may also be %XA_PRESENT, in + * which case all entries which are not NULL will be copied. + * + * The entries returned may not represent a snapshot of the XArray at a + * moment in time. For example, if another thread stores to index 5, then + * index 10, calling xa_extract() may return the old contents of index 5 + * and the new contents of index 10. Indices not modified while this + * function is running will not be skipped. + * + * If you need stronger guarantees, holding the xa_lock across calls to this + * function will prevent concurrent modification. + * + * Context: Any context. Takes and releases the RCU lock. + * Return: The number of entries copied. + */ +unsigned int xa_extract(struct xarray *xa, void **dst, unsigned long start, + unsigned long max, unsigned int n, xa_mark_t filter) +{ + XA_STATE(xas, xa, start); + + if (!n) + return 0; + + if ((__force unsigned int)filter < XA_MAX_MARKS) + return xas_extract_marked(&xas, dst, max, n, filter); + return xas_extract_present(&xas, dst, max, n); +} +EXPORT_SYMBOL(xa_extract); + +/** + * xa_destroy() - Free all internal data structures. + * @xa: XArray. + * + * After calling this function, the XArray is empty and has freed all memory + * allocated for its internal data structures. You are responsible for + * freeing the objects referenced by the XArray. + * + * Context: Any context. Takes and releases the xa_lock, interrupt-safe. + */ +void xa_destroy(struct xarray *xa) +{ + XA_STATE(xas, xa, 0); + unsigned long flags; + void *entry; + + xas.xa_node = NULL; + xas_lock_irqsave(&xas, flags); + entry = xa_head_locked(xa); + RCU_INIT_POINTER(xa->xa_head, NULL); + xas_init_marks(&xas); + /* lockdep checks we're still holding the lock in xas_free_nodes() */ + if (xa_is_node(entry)) + xas_free_nodes(&xas, xa_to_node(entry)); + xas_unlock_irqrestore(&xas, flags); +} +EXPORT_SYMBOL(xa_destroy); + +#ifdef XA_DEBUG +void xa_dump_node(const struct xa_node *node) +{ + unsigned i, j; + + if (!node) + return; + if ((unsigned long)node & 3) { + pr_cont("node %px\n", node); + return; + } + + pr_cont("node %px %s %d parent %px shift %d count %d values %d " + "array %px list %px %px marks", + node, node->parent ? "offset" : "max", node->offset, + node->parent, node->shift, node->count, node->nr_values, + node->array, node->private_list.prev, node->private_list.next); + for (i = 0; i < XA_MAX_MARKS; i++) + for (j = 0; j < XA_MARK_LONGS; j++) + pr_cont(" %lx", node->marks[i][j]); + pr_cont("\n"); +} + +void xa_dump_index(unsigned long index, unsigned int shift) +{ + if (!shift) + pr_info("%lu: ", index); + else if (shift >= BITS_PER_LONG) + pr_info("0-%lu: ", ~0UL); + else + pr_info("%lu-%lu: ", index, index | ((1UL << shift) - 1)); +} + +void xa_dump_entry(const void *entry, unsigned long index, unsigned long shift) +{ + if (!entry) + return; + + xa_dump_index(index, shift); + + if (xa_is_node(entry)) { + if (shift == 0) { + pr_cont("%px\n", entry); + } else { + unsigned long i; + struct xa_node *node = xa_to_node(entry); + xa_dump_node(node); + for (i = 0; i < XA_CHUNK_SIZE; i++) + xa_dump_entry(node->slots[i], + index + (i << node->shift), node->shift); + } + } else if (xa_is_value(entry)) + pr_cont("value %ld (0x%lx) [%px]\n", xa_to_value(entry), + xa_to_value(entry), entry); + else if (!xa_is_internal(entry)) + pr_cont("%px\n", entry); + else if (xa_is_retry(entry)) + pr_cont("retry (%ld)\n", xa_to_internal(entry)); + else if (xa_is_sibling(entry)) + pr_cont("sibling (slot %ld)\n", xa_to_sibling(entry)); + else if (xa_is_zero(entry)) + pr_cont("zero (%ld)\n", xa_to_internal(entry)); + else + pr_cont("UNKNOWN ENTRY (%px)\n", entry); +} + +void xa_dump(const struct xarray *xa) +{ + void *entry = xa->xa_head; + unsigned int shift = 0; + + pr_info("xarray: %px head %px flags %x marks %d %d %d\n", xa, entry, + xa->xa_flags, xa_marked(xa, XA_MARK_0), + xa_marked(xa, XA_MARK_1), xa_marked(xa, XA_MARK_2)); + if (xa_is_node(entry)) + shift = xa_to_node(entry)->shift + XA_CHUNK_SHIFT; + xa_dump_entry(entry, 0, shift); +} +#endif diff --git a/mm/Kconfig b/mm/Kconfig @@ -379,7 +379,7 @@ config TRANSPARENT_HUGEPAGE bool "Transparent Hugepage Support" depends on HAVE_ARCH_TRANSPARENT_HUGEPAGE select COMPACTION - select RADIX_TREE_MULTIORDER + select XARRAY_MULTI help Transparent Hugepages allows the kernel to use huge pages and huge tlb transparently to the applications whenever possible. @@ -671,7 +671,7 @@ config ZONE_DEVICE depends on MEMORY_HOTREMOVE depends on SPARSEMEM_VMEMMAP depends on ARCH_HAS_ZONE_DEVICE - select RADIX_TREE_MULTIORDER + select XARRAY_MULTI help Device memory hotplug support allows for establishing pmem, diff --git a/mm/filemap.c b/mm/filemap.c @@ -113,60 +113,26 @@ * ->tasklist_lock (memory_failure, collect_procs_ao) */ -static int page_cache_tree_insert(struct address_space *mapping, - struct page *page, void **shadowp) -{ - struct radix_tree_node *node; - void **slot; - int error; - - error = __radix_tree_create(&mapping->i_pages, page->index, 0, - &node, &slot); - if (error) - return error; - if (*slot) { - void *p; - - p = radix_tree_deref_slot_protected(slot, - &mapping->i_pages.xa_lock); - if (!radix_tree_exceptional_entry(p)) - return -EEXIST; - - mapping->nrexceptional--; - if (shadowp) - *shadowp = p; - } - __radix_tree_replace(&mapping->i_pages, node, slot, page, - workingset_lookup_update(mapping)); - mapping->nrpages++; - return 0; -} - -static void page_cache_tree_delete(struct address_space *mapping, +static void page_cache_delete(struct address_space *mapping, struct page *page, void *shadow) { - int i, nr; + XA_STATE(xas, &mapping->i_pages, page->index); + unsigned int nr = 1; + + mapping_set_update(&xas, mapping); - /* hugetlb pages are represented by one entry in the radix tree */ - nr = PageHuge(page) ? 1 : hpage_nr_pages(page); + /* hugetlb pages are represented by a single entry in the xarray */ + if (!PageHuge(page)) { + xas_set_order(&xas, page->index, compound_order(page)); + nr = 1U << compound_order(page); + } VM_BUG_ON_PAGE(!PageLocked(page), page); VM_BUG_ON_PAGE(PageTail(page), page); VM_BUG_ON_PAGE(nr != 1 && shadow, page); - for (i = 0; i < nr; i++) { - struct radix_tree_node *node; - void **slot; - - __radix_tree_lookup(&mapping->i_pages, page->index + i, - &node, &slot); - - VM_BUG_ON_PAGE(!node && nr != 1, page); - - radix_tree_clear_tags(&mapping->i_pages, node, slot); - __radix_tree_replace(&mapping->i_pages, node, slot, shadow, - workingset_lookup_update(mapping)); - } + xas_store(&xas, shadow); + xas_init_marks(&xas); page->mapping = NULL; /* Leave page->index set: truncation lookup relies upon it */ @@ -265,7 +231,7 @@ void __delete_from_page_cache(struct page *page, void *shadow) trace_mm_filemap_delete_from_page_cache(page); unaccount_page_cache_page(mapping, page); - page_cache_tree_delete(mapping, page, shadow); + page_cache_delete(mapping, page, shadow); } static void page_cache_free_page(struct address_space *mapping, @@ -308,7 +274,7 @@ void delete_from_page_cache(struct page *page) EXPORT_SYMBOL(delete_from_page_cache); /* - * page_cache_tree_delete_batch - delete several pages from page cache + * page_cache_delete_batch - delete several pages from page cache * @mapping: the mapping to which pages belong * @pvec: pagevec with pages to delete * @@ -321,24 +287,19 @@ EXPORT_SYMBOL(delete_from_page_cache); * * The function expects the i_pages lock to be held. */ -static void -page_cache_tree_delete_batch(struct address_space *mapping, +static void page_cache_delete_batch(struct address_space *mapping, struct pagevec *pvec) { - struct radix_tree_iter iter; - void **slot; + XA_STATE(xas, &mapping->i_pages, pvec->pages[0]->index); int total_pages = 0; int i = 0, tail_pages = 0; struct page *page; - pgoff_t start; - start = pvec->pages[0]->index; - radix_tree_for_each_slot(slot, &mapping->i_pages, &iter, start) { + mapping_set_update(&xas, mapping); + xas_for_each(&xas, page, ULONG_MAX) { if (i >= pagevec_count(pvec) && !tail_pages) break; - page = radix_tree_deref_slot_protected(slot, - &mapping->i_pages.xa_lock); - if (radix_tree_exceptional_entry(page)) + if (xa_is_value(page)) continue; if (!tail_pages) { /* @@ -346,8 +307,11 @@ page_cache_tree_delete_batch(struct address_space *mapping, * have our pages locked so they are protected from * being removed. */ - if (page != pvec->pages[i]) + if (page != pvec->pages[i]) { + VM_BUG_ON_PAGE(page->index > + pvec->pages[i]->index, page); continue; + } WARN_ON_ONCE(!PageLocked(page)); if (PageTransHuge(page) && !PageHuge(page)) tail_pages = HPAGE_PMD_NR - 1; @@ -358,11 +322,11 @@ page_cache_tree_delete_batch(struct address_space *mapping, */ i++; } else { + VM_BUG_ON_PAGE(page->index + HPAGE_PMD_NR - tail_pages + != pvec->pages[i]->index, page); tail_pages--; } - radix_tree_clear_tags(&mapping->i_pages, iter.node, slot); - __radix_tree_replace(&mapping->i_pages, iter.node, slot, NULL, - workingset_lookup_update(mapping)); + xas_store(&xas, NULL); total_pages++; } mapping->nrpages -= total_pages; @@ -383,7 +347,7 @@ void delete_from_page_cache_batch(struct address_space *mapping, unaccount_page_cache_page(mapping, pvec->pages[i]); } - page_cache_tree_delete_batch(mapping, pvec); + page_cache_delete_batch(mapping, pvec); xa_unlock_irqrestore(&mapping->i_pages, flags); for (i = 0; i < pagevec_count(pvec); i++) @@ -493,20 +457,31 @@ EXPORT_SYMBOL(filemap_flush); bool filemap_range_has_page(struct address_space *mapping, loff_t start_byte, loff_t end_byte) { - pgoff_t index = start_byte >> PAGE_SHIFT; - pgoff_t end = end_byte >> PAGE_SHIFT; struct page *page; + XA_STATE(xas, &mapping->i_pages, start_byte >> PAGE_SHIFT); + pgoff_t max = end_byte >> PAGE_SHIFT; if (end_byte < start_byte) return false; - if (mapping->nrpages == 0) - return false; + rcu_read_lock(); + for (;;) { + page = xas_find(&xas, max); + if (xas_retry(&xas, page)) + continue; + /* Shadow entries don't count */ + if (xa_is_value(page)) + continue; + /* + * We don't need to try to pin this page; we're about to + * release the RCU lock anyway. It is enough to know that + * there was a page here recently. + */ + break; + } + rcu_read_unlock(); - if (!find_get_pages_range(mapping, &index, end, 1, &page)) - return false; - put_page(page); - return true; + return page != NULL; } EXPORT_SYMBOL(filemap_range_has_page); @@ -777,51 +752,44 @@ EXPORT_SYMBOL(file_write_and_wait_range); * locked. This function does not add the new page to the LRU, the * caller must do that. * - * The remove + add is atomic. The only way this function can fail is - * memory allocation failure. + * The remove + add is atomic. This function cannot fail. */ int replace_page_cache_page(struct page *old, struct page *new, gfp_t gfp_mask) { - int error; + struct address_space *mapping = old->mapping; + void (*freepage)(struct page *) = mapping->a_ops->freepage; + pgoff_t offset = old->index; + XA_STATE(xas, &mapping->i_pages, offset); + unsigned long flags; VM_BUG_ON_PAGE(!PageLocked(old), old); VM_BUG_ON_PAGE(!PageLocked(new), new); VM_BUG_ON_PAGE(new->mapping, new); - error = radix_tree_preload(gfp_mask & GFP_RECLAIM_MASK); - if (!error) { - struct address_space *mapping = old->mapping; - void (*freepage)(struct page *); - unsigned long flags; - - pgoff_t offset = old->index; - freepage = mapping->a_ops->freepage; - - get_page(new); - new->mapping = mapping; - new->index = offset; + get_page(new); + new->mapping = mapping; + new->index = offset; - xa_lock_irqsave(&mapping->i_pages, flags); - __delete_from_page_cache(old, NULL); - error = page_cache_tree_insert(mapping, new, NULL); - BUG_ON(error); + xas_lock_irqsave(&xas, flags); + xas_store(&xas, new); - /* - * hugetlb pages do not participate in page cache accounting. - */ - if (!PageHuge(new)) - __inc_node_page_state(new, NR_FILE_PAGES); - if (PageSwapBacked(new)) - __inc_node_page_state(new, NR_SHMEM); - xa_unlock_irqrestore(&mapping->i_pages, flags); - mem_cgroup_migrate(old, new); - radix_tree_preload_end(); - if (freepage) - freepage(old); - put_page(old); - } + old->mapping = NULL; + /* hugetlb pages do not participate in page cache accounting. */ + if (!PageHuge(old)) + __dec_node_page_state(new, NR_FILE_PAGES); + if (!PageHuge(new)) + __inc_node_page_state(new, NR_FILE_PAGES); + if (PageSwapBacked(old)) + __dec_node_page_state(new, NR_SHMEM); + if (PageSwapBacked(new)) + __inc_node_page_state(new, NR_SHMEM); + xas_unlock_irqrestore(&xas, flags); + mem_cgroup_migrate(old, new); + if (freepage) + freepage(old); + put_page(old); - return error; + return 0; } EXPORT_SYMBOL_GPL(replace_page_cache_page); @@ -830,12 +798,15 @@ static int __add_to_page_cache_locked(struct page *page, pgoff_t offset, gfp_t gfp_mask, void **shadowp) { + XA_STATE(xas, &mapping->i_pages, offset); int huge = PageHuge(page); struct mem_cgroup *memcg; int error; + void *old; VM_BUG_ON_PAGE(!PageLocked(page), page); VM_BUG_ON_PAGE(PageSwapBacked(page), page); + mapping_set_update(&xas, mapping); if (!huge) { error = mem_cgroup_try_charge(page, current->mm, @@ -844,39 +815,47 @@ static int __add_to_page_cache_locked(struct page *page, return error; } - error = radix_tree_maybe_preload(gfp_mask & GFP_RECLAIM_MASK); - if (error) { - if (!huge) - mem_cgroup_cancel_charge(page, memcg, false); - return error; - } - get_page(page); page->mapping = mapping; page->index = offset; - xa_lock_irq(&mapping->i_pages); - error = page_cache_tree_insert(mapping, page, shadowp); - radix_tree_preload_end(); - if (unlikely(error)) - goto err_insert; + do { + xas_lock_irq(&xas); + old = xas_load(&xas); + if (old && !xa_is_value(old)) + xas_set_err(&xas, -EEXIST); + xas_store(&xas, page); + if (xas_error(&xas)) + goto unlock; + + if (xa_is_value(old)) { + mapping->nrexceptional--; + if (shadowp) + *shadowp = old; + } + mapping->nrpages++; + + /* hugetlb pages do not participate in page cache accounting */ + if (!huge) + __inc_node_page_state(page, NR_FILE_PAGES); +unlock: + xas_unlock_irq(&xas); + } while (xas_nomem(&xas, gfp_mask & GFP_RECLAIM_MASK)); + + if (xas_error(&xas)) + goto error; - /* hugetlb pages do not participate in page cache accounting. */ - if (!huge) - __inc_node_page_state(page, NR_FILE_PAGES); - xa_unlock_irq(&mapping->i_pages); if (!huge) mem_cgroup_commit_charge(page, memcg, false, false); trace_mm_filemap_add_to_page_cache(page); return 0; -err_insert: +error: page->mapping = NULL; /* Leave page->index set: truncation relies upon it */ - xa_unlock_irq(&mapping->i_pages); if (!huge) mem_cgroup_cancel_charge(page, memcg, false); put_page(page); - return error; + return xas_error(&xas); } /** @@ -1341,86 +1320,76 @@ int __lock_page_or_retry(struct page *page, struct mm_struct *mm, } /** - * page_cache_next_hole - find the next hole (not-present entry) - * @mapping: mapping - * @index: index - * @max_scan: maximum range to search - * - * Search the set [index, min(index+max_scan-1, MAX_INDEX)] for the - * lowest indexed hole. - * - * Returns: the index of the hole if found, otherwise returns an index - * outside of the set specified (in which case 'return - index >= - * max_scan' will be true). In rare cases of index wrap-around, 0 will - * be returned. - * - * page_cache_next_hole may be called under rcu_read_lock. However, - * like radix_tree_gang_lookup, this will not atomically search a - * snapshot of the tree at a single point in time. For example, if a - * hole is created at index 5, then subsequently a hole is created at - * index 10, page_cache_next_hole covering both indexes may return 10 - * if called under rcu_read_lock. + * page_cache_next_miss() - Find the next gap in the page cache. + * @mapping: Mapping. + * @index: Index. + * @max_scan: Maximum range to search. + * + * Search the range [index, min(index + max_scan - 1, ULONG_MAX)] for the + * gap with the lowest index. + * + * This function may be called under the rcu_read_lock. However, this will + * not atomically search a snapshot of the cache at a single point in time. + * For example, if a gap is created at index 5, then subsequently a gap is + * created at index 10, page_cache_next_miss covering both indices may + * return 10 if called under the rcu_read_lock. + * + * Return: The index of the gap if found, otherwise an index outside the + * range specified (in which case 'return - index >= max_scan' will be true). + * In the rare case of index wrap-around, 0 will be returned. */ -pgoff_t page_cache_next_hole(struct address_space *mapping, +pgoff_t page_cache_next_miss(struct address_space *mapping, pgoff_t index, unsigned long max_scan) { - unsigned long i; - - for (i = 0; i < max_scan; i++) { - struct page *page; + XA_STATE(xas, &mapping->i_pages, index); - page = radix_tree_lookup(&mapping->i_pages, index); - if (!page || radix_tree_exceptional_entry(page)) + while (max_scan--) { + void *entry = xas_next(&xas); + if (!entry || xa_is_value(entry)) break; - index++; - if (index == 0) + if (xas.xa_index == 0) break; } - return index; + return xas.xa_index; } -EXPORT_SYMBOL(page_cache_next_hole); +EXPORT_SYMBOL(page_cache_next_miss); /** - * page_cache_prev_hole - find the prev hole (not-present entry) - * @mapping: mapping - * @index: index - * @max_scan: maximum range to search - * - * Search backwards in the range [max(index-max_scan+1, 0), index] for - * the first hole. - * - * Returns: the index of the hole if found, otherwise returns an index - * outside of the set specified (in which case 'index - return >= - * max_scan' will be true). In rare cases of wrap-around, ULONG_MAX - * will be returned. - * - * page_cache_prev_hole may be called under rcu_read_lock. However, - * like radix_tree_gang_lookup, this will not atomically search a - * snapshot of the tree at a single point in time. For example, if a - * hole is created at index 10, then subsequently a hole is created at - * index 5, page_cache_prev_hole covering both indexes may return 5 if - * called under rcu_read_lock. + * page_cache_prev_miss() - Find the next gap in the page cache. + * @mapping: Mapping. + * @index: Index. + * @max_scan: Maximum range to search. + * + * Search the range [max(index - max_scan + 1, 0), index] for the + * gap with the highest index. + * + * This function may be called under the rcu_read_lock. However, this will + * not atomically search a snapshot of the cache at a single point in time. + * For example, if a gap is created at index 10, then subsequently a gap is + * created at index 5, page_cache_prev_miss() covering both indices may + * return 5 if called under the rcu_read_lock. + * + * Return: The index of the gap if found, otherwise an index outside the + * range specified (in which case 'index - return >= max_scan' will be true). + * In the rare case of wrap-around, ULONG_MAX will be returned. */ -pgoff_t page_cache_prev_hole(struct address_space *mapping, +pgoff_t page_cache_prev_miss(struct address_space *mapping, pgoff_t index, unsigned long max_scan) { - unsigned long i; - - for (i = 0; i < max_scan; i++) { - struct page *page; + XA_STATE(xas, &mapping->i_pages, index); - page = radix_tree_lookup(&mapping->i_pages, index); - if (!page || radix_tree_exceptional_entry(page)) + while (max_scan--) { + void *entry = xas_prev(&xas); + if (!entry || xa_is_value(entry)) break; - index--; - if (index == ULONG_MAX) + if (xas.xa_index == ULONG_MAX) break; } - return index; + return xas.xa_index; } -EXPORT_SYMBOL(page_cache_prev_hole); +EXPORT_SYMBOL(page_cache_prev_miss); /** * find_get_entry - find and get a page cache entry @@ -1437,47 +1406,40 @@ EXPORT_SYMBOL(page_cache_prev_hole); */ struct page *find_get_entry(struct address_space *mapping, pgoff_t offset) { - void **pagep; + XA_STATE(xas, &mapping->i_pages, offset); struct page *head, *page; rcu_read_lock(); repeat: - page = NULL; - pagep = radix_tree_lookup_slot(&mapping->i_pages, offset); - if (pagep) { - page = radix_tree_deref_slot(pagep); - if (unlikely(!page)) - goto out; - if (radix_tree_exception(page)) { - if (radix_tree_deref_retry(page)) - goto repeat; - /* - * A shadow entry of a recently evicted page, - * or a swap entry from shmem/tmpfs. Return - * it without attempting to raise page count. - */ - goto out; - } + xas_reset(&xas); + page = xas_load(&xas); + if (xas_retry(&xas, page)) + goto repeat; + /* + * A shadow entry of a recently evicted page, or a swap entry from + * shmem/tmpfs. Return it without attempting to raise page count. + */ + if (!page || xa_is_value(page)) + goto out; - head = compound_head(page); - if (!page_cache_get_speculative(head)) - goto repeat; + head = compound_head(page); + if (!page_cache_get_speculative(head)) + goto repeat; - /* The page was split under us? */ - if (compound_head(page) != head) { - put_page(head); - goto repeat; - } + /* The page was split under us? */ + if (compound_head(page) != head) { + put_page(head); + goto repeat; + } - /* - * Has the page moved? - * This is part of the lockless pagecache protocol. See - * include/linux/pagemap.h for details. - */ - if (unlikely(page != *pagep)) { - put_page(head); - goto repeat; - } + /* + * Has the page moved? + * This is part of the lockless pagecache protocol. See + * include/linux/pagemap.h for details. + */ + if (unlikely(page != xas_reload(&xas))) { + put_page(head); + goto repeat; } out: rcu_read_unlock(); @@ -1508,7 +1470,7 @@ struct page *find_lock_entry(struct address_space *mapping, pgoff_t offset) repeat: page = find_get_entry(mapping, offset); - if (page && !radix_tree_exception(page)) { + if (page && !xa_is_value(page)) { lock_page(page); /* Has the page been truncated? */ if (unlikely(page_mapping(page) != mapping)) { @@ -1554,7 +1516,7 @@ struct page *pagecache_get_page(struct address_space *mapping, pgoff_t offset, repeat: page = find_get_entry(mapping, offset); - if (radix_tree_exceptional_entry(page)) + if (xa_is_value(page)) page = NULL; if (!page) goto no_page; @@ -1640,53 +1602,48 @@ unsigned find_get_entries(struct address_space *mapping, pgoff_t start, unsigned int nr_entries, struct page **entries, pgoff_t *indices) { - void **slot; + XA_STATE(xas, &mapping->i_pages, start); + struct page *page; unsigned int ret = 0; - struct radix_tree_iter iter; if (!nr_entries) return 0; rcu_read_lock(); - radix_tree_for_each_slot(slot, &mapping->i_pages, &iter, start) { - struct page *head, *page; -repeat: - page = radix_tree_deref_slot(slot); - if (unlikely(!page)) + xas_for_each(&xas, page, ULONG_MAX) { + struct page *head; + if (xas_retry(&xas, page)) continue; - if (radix_tree_exception(page)) { - if (radix_tree_deref_retry(page)) { - slot = radix_tree_iter_retry(&iter); - continue; - } - /* - * A shadow entry of a recently evicted page, a swap - * entry from shmem/tmpfs or a DAX entry. Return it - * without attempting to raise page count. - */ + /* + * A shadow entry of a recently evicted page, a swap + * entry from shmem/tmpfs or a DAX entry. Return it + * without attempting to raise page count. + */ + if (xa_is_value(page)) goto export; - } head = compound_head(page); if (!page_cache_get_speculative(head)) - goto repeat; + goto retry; /* The page was split under us? */ - if (compound_head(page) != head) { - put_page(head); - goto repeat; - } + if (compound_head(page) != head) + goto put_page; /* Has the page moved? */ - if (unlikely(page != *slot)) { - put_page(head); - goto repeat; - } + if (unlikely(page != xas_reload(&xas))) + goto put_page; + export: - indices[ret] = iter.index; + indices[ret] = xas.xa_index; entries[ret] = page; if (++ret == nr_entries) break; + continue; +put_page: + put_page(head); +retry: + xas_reset(&xas); } rcu_read_unlock(); return ret; @@ -1717,64 +1674,50 @@ unsigned find_get_pages_range(struct address_space *mapping, pgoff_t *start, pgoff_t end, unsigned int nr_pages, struct page **pages) { - struct radix_tree_iter iter; - void **slot; + XA_STATE(xas, &mapping->i_pages, *start); + struct page *page; unsigned ret = 0; if (unlikely(!nr_pages)) return 0; rcu_read_lock(); - radix_tree_for_each_slot(slot, &mapping->i_pages, &iter, *start) { - struct page *head, *page; - - if (iter.index > end) - break; -repeat: - page = radix_tree_deref_slot(slot); - if (unlikely(!page)) + xas_for_each(&xas, page, end) { + struct page *head; + if (xas_retry(&xas, page)) continue; - - if (radix_tree_exception(page)) { - if (radix_tree_deref_retry(page)) { - slot = radix_tree_iter_retry(&iter); - continue; - } - /* - * A shadow entry of a recently evicted page, - * or a swap entry from shmem/tmpfs. Skip - * over it. - */ + /* Skip over shadow, swap and DAX entries */ + if (xa_is_value(page)) continue; - } head = compound_head(page); if (!page_cache_get_speculative(head)) - goto repeat; + goto retry; /* The page was split under us? */ - if (compound_head(page) != head) { - put_page(head); - goto repeat; - } + if (compound_head(page) != head) + goto put_page; /* Has the page moved? */ - if (unlikely(page != *slot)) { - put_page(head); - goto repeat; - } + if (unlikely(page != xas_reload(&xas))) + goto put_page; pages[ret] = page; if (++ret == nr_pages) { - *start = pages[ret - 1]->index + 1; + *start = page->index + 1; goto out; } + continue; +put_page: + put_page(head); +retry: + xas_reset(&xas); } /* * We come here when there is no page beyond @end. We take care to not * overflow the index @start as it confuses some of the callers. This - * breaks the iteration when there is page at index -1 but that is + * breaks the iteration when there is a page at index -1 but that is * already broken anyway. */ if (end == (pgoff_t)-1) @@ -1802,57 +1745,43 @@ out: unsigned find_get_pages_contig(struct address_space *mapping, pgoff_t index, unsigned int nr_pages, struct page **pages) { - struct radix_tree_iter iter; - void **slot; + XA_STATE(xas, &mapping->i_pages, index); + struct page *page; unsigned int ret = 0; if (unlikely(!nr_pages)) return 0; rcu_read_lock(); - radix_tree_for_each_contig(slot, &mapping->i_pages, &iter, index) { - struct page *head, *page; -repeat: - page = radix_tree_deref_slot(slot); - /* The hole, there no reason to continue */ - if (unlikely(!page)) - break; - - if (radix_tree_exception(page)) { - if (radix_tree_deref_retry(page)) { - slot = radix_tree_iter_retry(&iter); - continue; - } - /* - * A shadow entry of a recently evicted page, - * or a swap entry from shmem/tmpfs. Stop - * looking for contiguous pages. - */ + for (page = xas_load(&xas); page; page = xas_next(&xas)) { + struct page *head; + if (xas_retry(&xas, page)) + continue; + /* + * If the entry has been swapped out, we can stop looking. + * No current caller is looking for DAX entries. + */ + if (xa_is_value(page)) break; - } head = compound_head(page); if (!page_cache_get_speculative(head)) - goto repeat; + goto retry; /* The page was split under us? */ - if (compound_head(page) != head) { - put_page(head); - goto repeat; - } + if (compound_head(page) != head) + goto put_page; /* Has the page moved? */ - if (unlikely(page != *slot)) { - put_page(head); - goto repeat; - } + if (unlikely(page != xas_reload(&xas))) + goto put_page; /* * must check mapping and index after taking the ref. * otherwise we can get both false positives and false * negatives, which is just confusing to the caller. */ - if (page->mapping == NULL || page_to_pgoff(page) != iter.index) { + if (!page->mapping || page_to_pgoff(page) != xas.xa_index) { put_page(page); break; } @@ -1860,6 +1789,11 @@ repeat: pages[ret] = page; if (++ret == nr_pages) break; + continue; +put_page: + put_page(head); +retry: + xas_reset(&xas); } rcu_read_unlock(); return ret; @@ -1879,74 +1813,58 @@ EXPORT_SYMBOL(find_get_pages_contig); * @tag. We update @index to index the next page for the traversal. */ unsigned find_get_pages_range_tag(struct address_space *mapping, pgoff_t *index, - pgoff_t end, int tag, unsigned int nr_pages, + pgoff_t end, xa_mark_t tag, unsigned int nr_pages, struct page **pages) { - struct radix_tree_iter iter; - void **slot; + XA_STATE(xas, &mapping->i_pages, *index); + struct page *page; unsigned ret = 0; if (unlikely(!nr_pages)) return 0; rcu_read_lock(); - radix_tree_for_each_tagged(slot, &mapping->i_pages, &iter, *index, tag) { - struct page *head, *page; - - if (iter.index > end) - break; -repeat: - page = radix_tree_deref_slot(slot); - if (unlikely(!page)) + xas_for_each_marked(&xas, page, end, tag) { + struct page *head; + if (xas_retry(&xas, page)) continue; - - if (radix_tree_exception(page)) { - if (radix_tree_deref_retry(page)) { - slot = radix_tree_iter_retry(&iter); - continue; - } - /* - * A shadow entry of a recently evicted page. - * - * Those entries should never be tagged, but - * this tree walk is lockless and the tags are - * looked up in bulk, one radix tree node at a - * time, so there is a sizable window for page - * reclaim to evict a page we saw tagged. - * - * Skip over it. - */ + /* + * Shadow entries should never be tagged, but this iteration + * is lockless so there is a window for page reclaim to evict + * a page we saw tagged. Skip over it. + */ + if (xa_is_value(page)) continue; - } head = compound_head(page); if (!page_cache_get_speculative(head)) - goto repeat; + goto retry; /* The page was split under us? */ - if (compound_head(page) != head) { - put_page(head); - goto repeat; - } + if (compound_head(page) != head) + goto put_page; /* Has the page moved? */ - if (unlikely(page != *slot)) { - put_page(head); - goto repeat; - } + if (unlikely(page != xas_reload(&xas))) + goto put_page; pages[ret] = page; if (++ret == nr_pages) { - *index = pages[ret - 1]->index + 1; + *index = page->index + 1; goto out; } + continue; +put_page: + put_page(head); +retry: + xas_reset(&xas); } /* - * We come here when we got at @end. We take care to not overflow the + * We come here when we got to @end. We take care to not overflow the * index @index as it confuses some of the callers. This breaks the - * iteration when there is page at index -1 but that is already broken - * anyway. + * iteration when there is a page at index -1 but that is already + * broken anyway. */ if (end == (pgoff_t)-1) *index = (pgoff_t)-1; @@ -1972,57 +1890,51 @@ EXPORT_SYMBOL(find_get_pages_range_tag); * @tag. */ unsigned find_get_entries_tag(struct address_space *mapping, pgoff_t start, - int tag, unsigned int nr_entries, + xa_mark_t tag, unsigned int nr_entries, struct page **entries, pgoff_t *indices) { - void **slot; + XA_STATE(xas, &mapping->i_pages, start); + struct page *page; unsigned int ret = 0; - struct radix_tree_iter iter; if (!nr_entries) return 0; rcu_read_lock(); - radix_tree_for_each_tagged(slot, &mapping->i_pages, &iter, start, tag) { - struct page *head, *page; -repeat: - page = radix_tree_deref_slot(slot); - if (unlikely(!page)) + xas_for_each_marked(&xas, page, ULONG_MAX, tag) { + struct page *head; + if (xas_retry(&xas, page)) continue; - if (radix_tree_exception(page)) { - if (radix_tree_deref_retry(page)) { - slot = radix_tree_iter_retry(&iter); - continue; - } - - /* - * A shadow entry of a recently evicted page, a swap - * entry from shmem/tmpfs or a DAX entry. Return it - * without attempting to raise page count. - */ + /* + * A shadow entry of a recently evicted page, a swap + * entry from shmem/tmpfs or a DAX entry. Return it + * without attempting to raise page count. + */ + if (xa_is_value(page)) goto export; - } head = compound_head(page); if (!page_cache_get_speculative(head)) - goto repeat; + goto retry; /* The page was split under us? */ - if (compound_head(page) != head) { - put_page(head); - goto repeat; - } + if (compound_head(page) != head) + goto put_page; /* Has the page moved? */ - if (unlikely(page != *slot)) { - put_page(head); - goto repeat; - } + if (unlikely(page != xas_reload(&xas))) + goto put_page; + export: - indices[ret] = iter.index; + indices[ret] = xas.xa_index; entries[ret] = page; if (++ret == nr_entries) break; + continue; +put_page: + put_page(head); +retry: + xas_reset(&xas); } rcu_read_unlock(); return ret; @@ -2626,45 +2538,31 @@ EXPORT_SYMBOL(filemap_fault); void filemap_map_pages(struct vm_fault *vmf, pgoff_t start_pgoff, pgoff_t end_pgoff) { - struct radix_tree_iter iter; - void **slot; struct file *file = vmf->vma->vm_file; struct address_space *mapping = file->f_mapping; pgoff_t last_pgoff = start_pgoff; unsigned long max_idx; + XA_STATE(xas, &mapping->i_pages, start_pgoff); struct page *head, *page; rcu_read_lock(); - radix_tree_for_each_slot(slot, &mapping->i_pages, &iter, start_pgoff) { - if (iter.index > end_pgoff) - break; -repeat: - page = radix_tree_deref_slot(slot); - if (unlikely(!page)) - goto next; - if (radix_tree_exception(page)) { - if (radix_tree_deref_retry(page)) { - slot = radix_tree_iter_retry(&iter); - continue; - } + xas_for_each(&xas, page, end_pgoff) { + if (xas_retry(&xas, page)) + continue; + if (xa_is_value(page)) goto next; - } head = compound_head(page); if (!page_cache_get_speculative(head)) - goto repeat; + goto next; /* The page was split under us? */ - if (compound_head(page) != head) { - put_page(head); - goto repeat; - } + if (compound_head(page) != head) + goto skip; /* Has the page moved? */ - if (unlikely(page != *slot)) { - put_page(head); - goto repeat; - } + if (unlikely(page != xas_reload(&xas))) + goto skip; if (!PageUptodate(page) || PageReadahead(page) || @@ -2683,10 +2581,10 @@ repeat: if (file->f_ra.mmap_miss > 0) file->f_ra.mmap_miss--; - vmf->address += (iter.index - last_pgoff) << PAGE_SHIFT; + vmf->address += (xas.xa_index - last_pgoff) << PAGE_SHIFT; if (vmf->pte) - vmf->pte += iter.index - last_pgoff; - last_pgoff = iter.index; + vmf->pte += xas.xa_index - last_pgoff; + last_pgoff = xas.xa_index; if (alloc_set_pte(vmf, NULL, page)) goto unlock; unlock_page(page); @@ -2699,8 +2597,6 @@ next: /* Huge page is mapped? No need to proceed. */ if (pmd_trans_huge(*vmf->pmd)) break; - if (iter.index == end_pgoff) - break; } rcu_read_unlock(); } @@ -2810,7 +2706,7 @@ repeat: put_page(page); if (err == -EEXIST) goto repeat; - /* Presumably ENOMEM for radix tree node */ + /* Presumably ENOMEM for xarray node */ return ERR_PTR(err); } diff --git a/mm/huge_memory.c b/mm/huge_memory.c @@ -2450,13 +2450,13 @@ static void __split_huge_page(struct page *page, struct list_head *list, ClearPageCompound(head); /* See comment in __split_huge_page_tail() */ if (PageAnon(head)) { - /* Additional pin to radix tree of swap cache */ + /* Additional pin to swap cache */ if (PageSwapCache(head)) page_ref_add(head, 2); else page_ref_inc(head); } else { - /* Additional pin to radix tree */ + /* Additional pin to page cache */ page_ref_add(head, 2); xa_unlock(&head->mapping->i_pages); } @@ -2568,7 +2568,7 @@ bool can_split_huge_page(struct page *page, int *pextra_pins) { int extra_pins; - /* Additional pins from radix tree */ + /* Additional pins from page cache */ if (PageAnon(page)) extra_pins = PageSwapCache(page) ? HPAGE_PMD_NR : 0; else @@ -2664,17 +2664,14 @@ int split_huge_page_to_list(struct page *page, struct list_head *list) spin_lock_irqsave(zone_lru_lock(page_zone(head)), flags); if (mapping) { - void **pslot; + XA_STATE(xas, &mapping->i_pages, page_index(head)); - xa_lock(&mapping->i_pages); - pslot = radix_tree_lookup_slot(&mapping->i_pages, - page_index(head)); /* - * Check if the head page is present in radix tree. + * Check if the head page is present in page cache. * We assume all tail are present too, if head is there. */ - if (radix_tree_deref_slot_protected(pslot, - &mapping->i_pages.xa_lock) != head) + xa_lock(&mapping->i_pages); + if (xas_load(&xas) != head) goto fail; } diff --git a/mm/khugepaged.c b/mm/khugepaged.c @@ -1288,17 +1288,17 @@ static void retract_page_tables(struct address_space *mapping, pgoff_t pgoff) * * Basic scheme is simple, details are more complex: * - allocate and freeze a new huge page; - * - scan over radix tree replacing old pages the new one + * - scan page cache replacing old pages with the new one * + swap in pages if necessary; * + fill in gaps; - * + keep old pages around in case if rollback is required; - * - if replacing succeed: + * + keep old pages around in case rollback is required; + * - if replacing succeeds: * + copy data over; * + free old pages; * + unfreeze huge page; * - if replacing failed; * + put all pages back and unfreeze them; - * + restore gaps in the radix-tree; + * + restore gaps in the page cache; * + free huge page; */ static void collapse_shmem(struct mm_struct *mm, @@ -1306,12 +1306,11 @@ static void collapse_shmem(struct mm_struct *mm, struct page **hpage, int node) { gfp_t gfp; - struct page *page, *new_page, *tmp; + struct page *new_page; struct mem_cgroup *memcg; pgoff_t index, end = start + HPAGE_PMD_NR; LIST_HEAD(pagelist); - struct radix_tree_iter iter; - void **slot; + XA_STATE_ORDER(xas, &mapping->i_pages, start, HPAGE_PMD_ORDER); int nr_none = 0, result = SCAN_SUCCEED; VM_BUG_ON(start & (HPAGE_PMD_NR - 1)); @@ -1336,48 +1335,49 @@ static void collapse_shmem(struct mm_struct *mm, __SetPageLocked(new_page); BUG_ON(!page_ref_freeze(new_page, 1)); - /* - * At this point the new_page is 'frozen' (page_count() is zero), locked - * and not up-to-date. It's safe to insert it into radix tree, because - * nobody would be able to map it or use it in other way until we - * unfreeze it. + * At this point the new_page is 'frozen' (page_count() is zero), + * locked and not up-to-date. It's safe to insert it into the page + * cache, because nobody would be able to map it or use it in other + * way until we unfreeze it. */ - index = start; - xa_lock_irq(&mapping->i_pages); - radix_tree_for_each_slot(slot, &mapping->i_pages, &iter, start) { - int n = min(iter.index, end) - index; - - /* - * Handle holes in the radix tree: charge it from shmem and - * insert relevant subpage of new_page into the radix-tree. - */ - if (n && !shmem_charge(mapping->host, n)) { - result = SCAN_FAIL; + /* This will be less messy when we use multi-index entries */ + do { + xas_lock_irq(&xas); + xas_create_range(&xas); + if (!xas_error(&xas)) break; - } - nr_none += n; - for (; index < min(iter.index, end); index++) { - radix_tree_insert(&mapping->i_pages, index, - new_page + (index % HPAGE_PMD_NR)); - } + xas_unlock_irq(&xas); + if (!xas_nomem(&xas, GFP_KERNEL)) + goto out; + } while (1); - /* We are done. */ - if (index >= end) - break; + xas_set(&xas, start); + for (index = start; index < end; index++) { + struct page *page = xas_next(&xas); + + VM_BUG_ON(index != xas.xa_index); + if (!page) { + if (!shmem_charge(mapping->host, 1)) { + result = SCAN_FAIL; + break; + } + xas_store(&xas, new_page + (index % HPAGE_PMD_NR)); + nr_none++; + continue; + } - page = radix_tree_deref_slot_protected(slot, - &mapping->i_pages.xa_lock); - if (radix_tree_exceptional_entry(page) || !PageUptodate(page)) { - xa_unlock_irq(&mapping->i_pages); + if (xa_is_value(page) || !PageUptodate(page)) { + xas_unlock_irq(&xas); /* swap in or instantiate fallocated page */ if (shmem_getpage(mapping->host, index, &page, SGP_NOHUGE)) { result = SCAN_FAIL; - goto tree_unlocked; + goto xa_unlocked; } - xa_lock_irq(&mapping->i_pages); + xas_lock_irq(&xas); + xas_set(&xas, index); } else if (trylock_page(page)) { get_page(page); } else { @@ -1397,7 +1397,7 @@ static void collapse_shmem(struct mm_struct *mm, result = SCAN_TRUNCATED; goto out_unlock; } - xa_unlock_irq(&mapping->i_pages); + xas_unlock_irq(&xas); if (isolate_lru_page(page)) { result = SCAN_DEL_PAGE_LRU; @@ -1407,17 +1407,16 @@ static void collapse_shmem(struct mm_struct *mm, if (page_mapped(page)) unmap_mapping_pages(mapping, index, 1, false); - xa_lock_irq(&mapping->i_pages); + xas_lock_irq(&xas); + xas_set(&xas, index); - slot = radix_tree_lookup_slot(&mapping->i_pages, index); - VM_BUG_ON_PAGE(page != radix_tree_deref_slot_protected(slot, - &mapping->i_pages.xa_lock), page); + VM_BUG_ON_PAGE(page != xas_load(&xas), page); VM_BUG_ON_PAGE(page_mapped(page), page); /* * The page is expected to have page_count() == 3: * - we hold a pin on it; - * - one reference from radix tree; + * - one reference from page cache; * - one from isolate_lru_page; */ if (!page_ref_freeze(page, 3)) { @@ -1432,56 +1431,30 @@ static void collapse_shmem(struct mm_struct *mm, list_add_tail(&page->lru, &pagelist); /* Finally, replace with the new page. */ - radix_tree_replace_slot(&mapping->i_pages, slot, - new_page + (index % HPAGE_PMD_NR)); - - slot = radix_tree_iter_resume(slot, &iter); - index++; + xas_store(&xas, new_page + (index % HPAGE_PMD_NR)); continue; out_lru: - xa_unlock_irq(&mapping->i_pages); + xas_unlock_irq(&xas); putback_lru_page(page); out_isolate_failed: unlock_page(page); put_page(page); - goto tree_unlocked; + goto xa_unlocked; out_unlock: unlock_page(page); put_page(page); break; } + xas_unlock_irq(&xas); - /* - * Handle hole in radix tree at the end of the range. - * This code only triggers if there's nothing in radix tree - * beyond 'end'. - */ - if (result == SCAN_SUCCEED && index < end) { - int n = end - index; - - if (!shmem_charge(mapping->host, n)) { - result = SCAN_FAIL; - goto tree_locked; - } - - for (; index < end; index++) { - radix_tree_insert(&mapping->i_pages, index, - new_page + (index % HPAGE_PMD_NR)); - } - nr_none += n; - } - -tree_locked: - xa_unlock_irq(&mapping->i_pages); -tree_unlocked: - +xa_unlocked: if (result == SCAN_SUCCEED) { - unsigned long flags; + struct page *page, *tmp; struct zone *zone = page_zone(new_page); /* - * Replacing old pages with new one has succeed, now we need to - * copy the content and free old pages. + * Replacing old pages with new one has succeeded, now we + * need to copy the content and free the old pages. */ list_for_each_entry_safe(page, tmp, &pagelist, lru) { copy_highpage(new_page + (page->index % HPAGE_PMD_NR), @@ -1495,16 +1468,16 @@ tree_unlocked: put_page(page); } - local_irq_save(flags); + local_irq_disable(); __inc_node_page_state(new_page, NR_SHMEM_THPS); if (nr_none) { __mod_node_page_state(zone->zone_pgdat, NR_FILE_PAGES, nr_none); __mod_node_page_state(zone->zone_pgdat, NR_SHMEM, nr_none); } - local_irq_restore(flags); + local_irq_enable(); /* - * Remove pte page tables, so we can re-faulti + * Remove pte page tables, so we can re-fault * the page as huge. */ retract_page_tables(mapping, start); @@ -1521,37 +1494,37 @@ tree_unlocked: khugepaged_pages_collapsed++; } else { - /* Something went wrong: rollback changes to the radix-tree */ + struct page *page; + /* Something went wrong: roll back page cache changes */ shmem_uncharge(mapping->host, nr_none); - xa_lock_irq(&mapping->i_pages); - radix_tree_for_each_slot(slot, &mapping->i_pages, &iter, start) { - if (iter.index >= end) - break; + xas_lock_irq(&xas); + xas_set(&xas, start); + xas_for_each(&xas, page, end - 1) { page = list_first_entry_or_null(&pagelist, struct page, lru); - if (!page || iter.index < page->index) { + if (!page || xas.xa_index < page->index) { if (!nr_none) break; nr_none--; /* Put holes back where they were */ - radix_tree_delete(&mapping->i_pages, iter.index); + xas_store(&xas, NULL); continue; } - VM_BUG_ON_PAGE(page->index != iter.index, page); + VM_BUG_ON_PAGE(page->index != xas.xa_index, page); /* Unfreeze the page. */ list_del(&page->lru); page_ref_unfreeze(page, 2); - radix_tree_replace_slot(&mapping->i_pages, slot, page); - slot = radix_tree_iter_resume(slot, &iter); - xa_unlock_irq(&mapping->i_pages); + xas_store(&xas, page); + xas_pause(&xas); + xas_unlock_irq(&xas); putback_lru_page(page); unlock_page(page); - xa_lock_irq(&mapping->i_pages); + xas_lock_irq(&xas); } VM_BUG_ON(nr_none); - xa_unlock_irq(&mapping->i_pages); + xas_unlock_irq(&xas); /* Unfreeze new_page, caller would take care about freeing it */ page_ref_unfreeze(new_page, 1); @@ -1569,8 +1542,7 @@ static void khugepaged_scan_shmem(struct mm_struct *mm, pgoff_t start, struct page **hpage) { struct page *page = NULL; - struct radix_tree_iter iter; - void **slot; + XA_STATE(xas, &mapping->i_pages, start); int present, swap; int node = NUMA_NO_NODE; int result = SCAN_SUCCEED; @@ -1579,17 +1551,11 @@ static void khugepaged_scan_shmem(struct mm_struct *mm, swap = 0; memset(khugepaged_node_load, 0, sizeof(khugepaged_node_load)); rcu_read_lock(); - radix_tree_for_each_slot(slot, &mapping->i_pages, &iter, start) { - if (iter.index >= start + HPAGE_PMD_NR) - break; - - page = radix_tree_deref_slot(slot); - if (radix_tree_deref_retry(page)) { - slot = radix_tree_iter_retry(&iter); + xas_for_each(&xas, page, start + HPAGE_PMD_NR - 1) { + if (xas_retry(&xas, page)) continue; - } - if (radix_tree_exception(page)) { + if (xa_is_value(page)) { if (++swap > khugepaged_max_ptes_swap) { result = SCAN_EXCEED_SWAP_PTE; break; @@ -1628,7 +1594,7 @@ static void khugepaged_scan_shmem(struct mm_struct *mm, present++; if (need_resched()) { - slot = radix_tree_iter_resume(slot, &iter); + xas_pause(&xas); cond_resched_rcu(); } } diff --git a/mm/madvise.c b/mm/madvise.c @@ -251,7 +251,7 @@ static void force_shm_swapin_readahead(struct vm_area_struct *vma, index = ((start - vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff; page = find_get_entry(mapping, index); - if (!radix_tree_exceptional_entry(page)) { + if (!xa_is_value(page)) { if (page) put_page(page); continue; diff --git a/mm/memcontrol.c b/mm/memcontrol.c @@ -4728,7 +4728,7 @@ static struct page *mc_handle_file_pte(struct vm_area_struct *vma, /* shmem/tmpfs may report page out on swap: account for that too. */ if (shmem_mapping(mapping)) { page = find_get_entry(mapping, pgoff); - if (radix_tree_exceptional_entry(page)) { + if (xa_is_value(page)) { swp_entry_t swp = radix_to_swp_entry(page); if (do_memsw_account()) *entry = swp; diff --git a/mm/memfd.c b/mm/memfd.c @@ -21,44 +21,36 @@ #include <uapi/linux/memfd.h> /* - * We need a tag: a new tag would expand every radix_tree_node by 8 bytes, + * We need a tag: a new tag would expand every xa_node by 8 bytes, * so reuse a tag which we firmly believe is never set or cleared on tmpfs * or hugetlbfs because they are memory only filesystems. */ #define MEMFD_TAG_PINNED PAGECACHE_TAG_TOWRITE #define LAST_SCAN 4 /* about 150ms max */ -static void memfd_tag_pins(struct address_space *mapping) +static void memfd_tag_pins(struct xa_state *xas) { - struct radix_tree_iter iter; - void __rcu **slot; - pgoff_t start; struct page *page; + unsigned int tagged = 0; lru_add_drain(); - start = 0; - rcu_read_lock(); - - radix_tree_for_each_slot(slot, &mapping->i_pages, &iter, start) { - page = radix_tree_deref_slot(slot); - if (!page || radix_tree_exception(page)) { - if (radix_tree_deref_retry(page)) { - slot = radix_tree_iter_retry(&iter); - continue; - } - } else if (page_count(page) - page_mapcount(page) > 1) { - xa_lock_irq(&mapping->i_pages); - radix_tree_tag_set(&mapping->i_pages, iter.index, - MEMFD_TAG_PINNED); - xa_unlock_irq(&mapping->i_pages); - } - if (need_resched()) { - slot = radix_tree_iter_resume(slot, &iter); - cond_resched_rcu(); - } + xas_lock_irq(xas); + xas_for_each(xas, page, ULONG_MAX) { + if (xa_is_value(page)) + continue; + if (page_count(page) - page_mapcount(page) > 1) + xas_set_mark(xas, MEMFD_TAG_PINNED); + + if (++tagged % XA_CHECK_SCHED) + continue; + + xas_pause(xas); + xas_unlock_irq(xas); + cond_resched(); + xas_lock_irq(xas); } - rcu_read_unlock(); + xas_unlock_irq(xas); } /* @@ -72,17 +64,17 @@ static void memfd_tag_pins(struct address_space *mapping) */ static int memfd_wait_for_pins(struct address_space *mapping) { - struct radix_tree_iter iter; - void __rcu **slot; - pgoff_t start; + XA_STATE(xas, &mapping->i_pages, 0); struct page *page; int error, scan; - memfd_tag_pins(mapping); + memfd_tag_pins(&xas); error = 0; for (scan = 0; scan <= LAST_SCAN; scan++) { - if (!radix_tree_tagged(&mapping->i_pages, MEMFD_TAG_PINNED)) + unsigned int tagged = 0; + + if (!xas_marked(&xas, MEMFD_TAG_PINNED)) break; if (!scan) @@ -90,45 +82,34 @@ static int memfd_wait_for_pins(struct address_space *mapping) else if (schedule_timeout_killable((HZ << scan) / 200)) scan = LAST_SCAN; - start = 0; - rcu_read_lock(); - radix_tree_for_each_tagged(slot, &mapping->i_pages, &iter, - start, MEMFD_TAG_PINNED) { - - page = radix_tree_deref_slot(slot); - if (radix_tree_exception(page)) { - if (radix_tree_deref_retry(page)) { - slot = radix_tree_iter_retry(&iter); - continue; - } - - page = NULL; - } - - if (page && - page_count(page) - page_mapcount(page) != 1) { - if (scan < LAST_SCAN) - goto continue_resched; - + xas_set(&xas, 0); + xas_lock_irq(&xas); + xas_for_each_marked(&xas, page, ULONG_MAX, MEMFD_TAG_PINNED) { + bool clear = true; + if (xa_is_value(page)) + continue; + if (page_count(page) - page_mapcount(page) != 1) { /* * On the last scan, we clean up all those tags * we inserted; but make a note that we still * found pages pinned. */ - error = -EBUSY; + if (scan == LAST_SCAN) + error = -EBUSY; + else + clear = false; } + if (clear) + xas_clear_mark(&xas, MEMFD_TAG_PINNED); + if (++tagged % XA_CHECK_SCHED) + continue; - xa_lock_irq(&mapping->i_pages); - radix_tree_tag_clear(&mapping->i_pages, - iter.index, MEMFD_TAG_PINNED); - xa_unlock_irq(&mapping->i_pages); -continue_resched: - if (need_resched()) { - slot = radix_tree_iter_resume(slot, &iter); - cond_resched_rcu(); - } + xas_pause(&xas); + xas_unlock_irq(&xas); + cond_resched(); + xas_lock_irq(&xas); } - rcu_read_unlock(); + xas_unlock_irq(&xas); } return error; diff --git a/mm/migrate.c b/mm/migrate.c @@ -326,7 +326,7 @@ void __migration_entry_wait(struct mm_struct *mm, pte_t *ptep, page = migration_entry_to_page(entry); /* - * Once radix-tree replacement of page migration started, page_count + * Once page cache replacement of page migration started, page_count * *must* be zero. And, we don't want to call wait_on_page_locked() * against a page without get_page(). * So, we use get_page_unless_zero(), here. Even failed, page fault @@ -441,10 +441,10 @@ int migrate_page_move_mapping(struct address_space *mapping, struct buffer_head *head, enum migrate_mode mode, int extra_count) { + XA_STATE(xas, &mapping->i_pages, page_index(page)); struct zone *oldzone, *newzone; int dirty; int expected_count = 1 + extra_count; - void **pslot; /* * Device public or private pages have an extra refcount as they are @@ -470,21 +470,16 @@ int migrate_page_move_mapping(struct address_space *mapping, oldzone = page_zone(page); newzone = page_zone(newpage); - xa_lock_irq(&mapping->i_pages); - - pslot = radix_tree_lookup_slot(&mapping->i_pages, - page_index(page)); + xas_lock_irq(&xas); expected_count += hpage_nr_pages(page) + page_has_private(page); - if (page_count(page) != expected_count || - radix_tree_deref_slot_protected(pslot, - &mapping->i_pages.xa_lock) != page) { - xa_unlock_irq(&mapping->i_pages); + if (page_count(page) != expected_count || xas_load(&xas) != page) { + xas_unlock_irq(&xas); return -EAGAIN; } if (!page_ref_freeze(page, expected_count)) { - xa_unlock_irq(&mapping->i_pages); + xas_unlock_irq(&xas); return -EAGAIN; } @@ -498,7 +493,7 @@ int migrate_page_move_mapping(struct address_space *mapping, if (mode == MIGRATE_ASYNC && head && !buffer_migrate_lock_buffers(head, mode)) { page_ref_unfreeze(page, expected_count); - xa_unlock_irq(&mapping->i_pages); + xas_unlock_irq(&xas); return -EAGAIN; } @@ -526,16 +521,13 @@ int migrate_page_move_mapping(struct address_space *mapping, SetPageDirty(newpage); } - radix_tree_replace_slot(&mapping->i_pages, pslot, newpage); + xas_store(&xas, newpage); if (PageTransHuge(page)) { int i; - int index = page_index(page); for (i = 1; i < HPAGE_PMD_NR; i++) { - pslot = radix_tree_lookup_slot(&mapping->i_pages, - index + i); - radix_tree_replace_slot(&mapping->i_pages, pslot, - newpage + i); + xas_next(&xas); + xas_store(&xas, newpage + i); } } @@ -546,7 +538,7 @@ int migrate_page_move_mapping(struct address_space *mapping, */ page_ref_unfreeze(page, expected_count - hpage_nr_pages(page)); - xa_unlock(&mapping->i_pages); + xas_unlock(&xas); /* Leave irq disabled to prevent preemption while updating stats */ /* @@ -586,22 +578,18 @@ EXPORT_SYMBOL(migrate_page_move_mapping); int migrate_huge_page_move_mapping(struct address_space *mapping, struct page *newpage, struct page *page) { + XA_STATE(xas, &mapping->i_pages, page_index(page)); int expected_count; - void **pslot; - - xa_lock_irq(&mapping->i_pages); - - pslot = radix_tree_lookup_slot(&mapping->i_pages, page_index(page)); + xas_lock_irq(&xas); expected_count = 2 + page_has_private(page); - if (page_count(page) != expected_count || - radix_tree_deref_slot_protected(pslot, &mapping->i_pages.xa_lock) != page) { - xa_unlock_irq(&mapping->i_pages); + if (page_count(page) != expected_count || xas_load(&xas) != page) { + xas_unlock_irq(&xas); return -EAGAIN; } if (!page_ref_freeze(page, expected_count)) { - xa_unlock_irq(&mapping->i_pages); + xas_unlock_irq(&xas); return -EAGAIN; } @@ -610,11 +598,11 @@ int migrate_huge_page_move_mapping(struct address_space *mapping, get_page(newpage); - radix_tree_replace_slot(&mapping->i_pages, pslot, newpage); + xas_store(&xas, newpage); page_ref_unfreeze(page, expected_count - 1); - xa_unlock_irq(&mapping->i_pages); + xas_unlock_irq(&xas); return MIGRATEPAGE_SUCCESS; } diff --git a/mm/mincore.c b/mm/mincore.c @@ -66,7 +66,7 @@ static unsigned char mincore_page(struct address_space *mapping, pgoff_t pgoff) * shmem/tmpfs may return swap: account for swapcache * page too. */ - if (radix_tree_exceptional_entry(page)) { + if (xa_is_value(page)) { swp_entry_t swp = radix_to_swp_entry(page); page = find_get_page(swap_address_space(swp), swp_offset(swp)); diff --git a/mm/page-writeback.c b/mm/page-writeback.c @@ -2097,34 +2097,25 @@ void __init page_writeback_init(void) * dirty pages in the file (thus it is important for this function to be quick * so that it can tag pages faster than a dirtying process can create them). */ -/* - * We tag pages in batches of WRITEBACK_TAG_BATCH to reduce the i_pages lock - * latency. - */ void tag_pages_for_writeback(struct address_space *mapping, pgoff_t start, pgoff_t end) { -#define WRITEBACK_TAG_BATCH 4096 - unsigned long tagged = 0; - struct radix_tree_iter iter; - void **slot; + XA_STATE(xas, &mapping->i_pages, start); + unsigned int tagged = 0; + void *page; - xa_lock_irq(&mapping->i_pages); - radix_tree_for_each_tagged(slot, &mapping->i_pages, &iter, start, - PAGECACHE_TAG_DIRTY) { - if (iter.index > end) - break; - radix_tree_iter_tag_set(&mapping->i_pages, &iter, - PAGECACHE_TAG_TOWRITE); - tagged++; - if ((tagged % WRITEBACK_TAG_BATCH) != 0) + xas_lock_irq(&xas); + xas_for_each_marked(&xas, page, end, PAGECACHE_TAG_DIRTY) { + xas_set_mark(&xas, PAGECACHE_TAG_TOWRITE); + if (++tagged % XA_CHECK_SCHED) continue; - slot = radix_tree_iter_resume(slot, &iter); - xa_unlock_irq(&mapping->i_pages); + + xas_pause(&xas); + xas_unlock_irq(&xas); cond_resched(); - xa_lock_irq(&mapping->i_pages); + xas_lock_irq(&xas); } - xa_unlock_irq(&mapping->i_pages); + xas_unlock_irq(&xas); } EXPORT_SYMBOL(tag_pages_for_writeback); @@ -2170,7 +2161,7 @@ int write_cache_pages(struct address_space *mapping, pgoff_t end; /* Inclusive */ pgoff_t done_index; int range_whole = 0; - int tag; + xa_mark_t tag; pagevec_init(&pvec); if (wbc->range_cyclic) { @@ -2442,7 +2433,7 @@ void account_page_cleaned(struct page *page, struct address_space *mapping, /* * For address_spaces which do not use buffers. Just tag the page as dirty in - * its radix tree. + * the xarray. * * This is also used when a single buffer is being dirtied: we want to set the * page dirty in that case, but not all the buffers. This is a "bottom-up" @@ -2468,7 +2459,7 @@ int __set_page_dirty_nobuffers(struct page *page) BUG_ON(page_mapping(page) != mapping); WARN_ON_ONCE(!PagePrivate(page) && !PageUptodate(page)); account_page_dirtied(page, mapping); - radix_tree_tag_set(&mapping->i_pages, page_index(page), + __xa_set_mark(&mapping->i_pages, page_index(page), PAGECACHE_TAG_DIRTY); xa_unlock_irqrestore(&mapping->i_pages, flags); unlock_page_memcg(page); @@ -2631,13 +2622,13 @@ EXPORT_SYMBOL(__cancel_dirty_page); * Returns true if the page was previously dirty. * * This is for preparing to put the page under writeout. We leave the page - * tagged as dirty in the radix tree so that a concurrent write-for-sync + * tagged as dirty in the xarray so that a concurrent write-for-sync * can discover it via a PAGECACHE_TAG_DIRTY walk. The ->writepage * implementation will run either set_page_writeback() or set_page_dirty(), - * at which stage we bring the page's dirty flag and radix-tree dirty tag + * at which stage we bring the page's dirty flag and xarray dirty tag * back into sync. * - * This incoherency between the page's dirty flag and radix-tree tag is + * This incoherency between the page's dirty flag and xarray tag is * unfortunate, but it only exists while the page is locked. */ int clear_page_dirty_for_io(struct page *page) @@ -2718,7 +2709,7 @@ int test_clear_page_writeback(struct page *page) xa_lock_irqsave(&mapping->i_pages, flags); ret = TestClearPageWriteback(page); if (ret) { - radix_tree_tag_clear(&mapping->i_pages, page_index(page), + __xa_clear_mark(&mapping->i_pages, page_index(page), PAGECACHE_TAG_WRITEBACK); if (bdi_cap_account_writeback(bdi)) { struct bdi_writeback *wb = inode_to_wb(inode); @@ -2758,11 +2749,13 @@ int __test_set_page_writeback(struct page *page, bool keep_write) lock_page_memcg(page); if (mapping && mapping_use_writeback_tags(mapping)) { + XA_STATE(xas, &mapping->i_pages, page_index(page)); struct inode *inode = mapping->host; struct backing_dev_info *bdi = inode_to_bdi(inode); unsigned long flags; - xa_lock_irqsave(&mapping->i_pages, flags); + xas_lock_irqsave(&xas, flags); + xas_load(&xas); ret = TestSetPageWriteback(page); if (!ret) { bool on_wblist; @@ -2770,8 +2763,7 @@ int __test_set_page_writeback(struct page *page, bool keep_write) on_wblist = mapping_tagged(mapping, PAGECACHE_TAG_WRITEBACK); - radix_tree_tag_set(&mapping->i_pages, page_index(page), - PAGECACHE_TAG_WRITEBACK); + xas_set_mark(&xas, PAGECACHE_TAG_WRITEBACK); if (bdi_cap_account_writeback(bdi)) inc_wb_stat(inode_to_wb(inode), WB_WRITEBACK); @@ -2784,12 +2776,10 @@ int __test_set_page_writeback(struct page *page, bool keep_write) sb_mark_inode_writeback(mapping->host); } if (!PageDirty(page)) - radix_tree_tag_clear(&mapping->i_pages, page_index(page), - PAGECACHE_TAG_DIRTY); + xas_clear_mark(&xas, PAGECACHE_TAG_DIRTY); if (!keep_write) - radix_tree_tag_clear(&mapping->i_pages, page_index(page), - PAGECACHE_TAG_TOWRITE); - xa_unlock_irqrestore(&mapping->i_pages, flags); + xas_clear_mark(&xas, PAGECACHE_TAG_TOWRITE); + xas_unlock_irqrestore(&xas, flags); } else { ret = TestSetPageWriteback(page); } @@ -2803,16 +2793,6 @@ int __test_set_page_writeback(struct page *page, bool keep_write) } EXPORT_SYMBOL(__test_set_page_writeback); -/* - * Return true if any of the pages in the mapping are marked with the - * passed tag. - */ -int mapping_tagged(struct address_space *mapping, int tag) -{ - return radix_tree_tagged(&mapping->i_pages, tag); -} -EXPORT_SYMBOL(mapping_tagged); - /** * wait_for_stable_page() - wait for writeback to finish, if necessary. * @page: The page to wait on. diff --git a/mm/readahead.c b/mm/readahead.c @@ -176,10 +176,8 @@ unsigned int __do_page_cache_readahead(struct address_space *mapping, if (page_offset > end_index) break; - rcu_read_lock(); - page = radix_tree_lookup(&mapping->i_pages, page_offset); - rcu_read_unlock(); - if (page && !radix_tree_exceptional_entry(page)) { + page = xa_load(&mapping->i_pages, page_offset); + if (page && !xa_is_value(page)) { /* * Page already present? Kick off the current batch of * contiguous pages before continuing with the next @@ -336,7 +334,7 @@ static pgoff_t count_history_pages(struct address_space *mapping, pgoff_t head; rcu_read_lock(); - head = page_cache_prev_hole(mapping, offset - 1, max); + head = page_cache_prev_miss(mapping, offset - 1, max); rcu_read_unlock(); return offset - 1 - head; @@ -425,7 +423,7 @@ ondemand_readahead(struct address_space *mapping, pgoff_t start; rcu_read_lock(); - start = page_cache_next_hole(mapping, offset + 1, max_pages); + start = page_cache_next_miss(mapping, offset + 1, max_pages); rcu_read_unlock(); if (!start || start - offset > max_pages) diff --git a/mm/shmem.c b/mm/shmem.c @@ -322,24 +322,20 @@ void shmem_uncharge(struct inode *inode, long pages) } /* - * Replace item expected in radix tree by a new item, while holding tree lock. + * Replace item expected in xarray by a new item, while holding xa_lock. */ -static int shmem_radix_tree_replace(struct address_space *mapping, +static int shmem_replace_entry(struct address_space *mapping, pgoff_t index, void *expected, void *replacement) { - struct radix_tree_node *node; - void __rcu **pslot; + XA_STATE(xas, &mapping->i_pages, index); void *item; VM_BUG_ON(!expected); VM_BUG_ON(!replacement); - item = __radix_tree_lookup(&mapping->i_pages, index, &node, &pslot); - if (!item) - return -ENOENT; + item = xas_load(&xas); if (item != expected) return -ENOENT; - __radix_tree_replace(&mapping->i_pages, node, pslot, - replacement, NULL); + xas_store(&xas, replacement); return 0; } @@ -353,12 +349,7 @@ static int shmem_radix_tree_replace(struct address_space *mapping, static bool shmem_confirm_swap(struct address_space *mapping, pgoff_t index, swp_entry_t swap) { - void *item; - - rcu_read_lock(); - item = radix_tree_lookup(&mapping->i_pages, index); - rcu_read_unlock(); - return item == swp_to_radix_entry(swap); + return xa_load(&mapping->i_pages, index) == swp_to_radix_entry(swap); } /* @@ -586,9 +577,11 @@ static inline bool is_huge_enabled(struct shmem_sb_info *sbinfo) */ static int shmem_add_to_page_cache(struct page *page, struct address_space *mapping, - pgoff_t index, void *expected) + pgoff_t index, void *expected, gfp_t gfp) { - int error, nr = hpage_nr_pages(page); + XA_STATE_ORDER(xas, &mapping->i_pages, index, compound_order(page)); + unsigned long i = 0; + unsigned long nr = 1UL << compound_order(page); VM_BUG_ON_PAGE(PageTail(page), page); VM_BUG_ON_PAGE(index != round_down(index, nr), page); @@ -600,47 +593,39 @@ static int shmem_add_to_page_cache(struct page *page, page->mapping = mapping; page->index = index; - xa_lock_irq(&mapping->i_pages); - if (PageTransHuge(page)) { - void __rcu **results; - pgoff_t idx; - int i; - - error = 0; - if (radix_tree_gang_lookup_slot(&mapping->i_pages, - &results, &idx, index, 1) && - idx < index + HPAGE_PMD_NR) { - error = -EEXIST; + do { + void *entry; + xas_lock_irq(&xas); + entry = xas_find_conflict(&xas); + if (entry != expected) + xas_set_err(&xas, -EEXIST); + xas_create_range(&xas); + if (xas_error(&xas)) + goto unlock; +next: + xas_store(&xas, page + i); + if (++i < nr) { + xas_next(&xas); + goto next; } - - if (!error) { - for (i = 0; i < HPAGE_PMD_NR; i++) { - error = radix_tree_insert(&mapping->i_pages, - index + i, page + i); - VM_BUG_ON(error); - } + if (PageTransHuge(page)) { count_vm_event(THP_FILE_ALLOC); + __inc_node_page_state(page, NR_SHMEM_THPS); } - } else if (!expected) { - error = radix_tree_insert(&mapping->i_pages, index, page); - } else { - error = shmem_radix_tree_replace(mapping, index, expected, - page); - } - - if (!error) { mapping->nrpages += nr; - if (PageTransHuge(page)) - __inc_node_page_state(page, NR_SHMEM_THPS); __mod_node_page_state(page_pgdat(page), NR_FILE_PAGES, nr); __mod_node_page_state(page_pgdat(page), NR_SHMEM, nr); - xa_unlock_irq(&mapping->i_pages); - } else { +unlock: + xas_unlock_irq(&xas); + } while (xas_nomem(&xas, gfp)); + + if (xas_error(&xas)) { page->mapping = NULL; - xa_unlock_irq(&mapping->i_pages); page_ref_sub(page, nr); + return xas_error(&xas); } - return error; + + return 0; } /* @@ -654,7 +639,7 @@ static void shmem_delete_from_page_cache(struct page *page, void *radswap) VM_BUG_ON_PAGE(PageCompound(page), page); xa_lock_irq(&mapping->i_pages); - error = shmem_radix_tree_replace(mapping, page->index, page, radswap); + error = shmem_replace_entry(mapping, page->index, page, radswap); page->mapping = NULL; mapping->nrpages--; __dec_node_page_state(page, NR_FILE_PAGES); @@ -665,7 +650,7 @@ static void shmem_delete_from_page_cache(struct page *page, void *radswap) } /* - * Remove swap entry from radix tree, free the swap and its page cache. + * Remove swap entry from page cache, free the swap and its page cache. */ static int shmem_free_swap(struct address_space *mapping, pgoff_t index, void *radswap) @@ -673,7 +658,7 @@ static int shmem_free_swap(struct address_space *mapping, void *old; xa_lock_irq(&mapping->i_pages); - old = radix_tree_delete_item(&mapping->i_pages, index, radswap); + old = __xa_cmpxchg(&mapping->i_pages, index, radswap, NULL, 0); xa_unlock_irq(&mapping->i_pages); if (old != radswap) return -ENOENT; @@ -691,29 +676,19 @@ static int shmem_free_swap(struct address_space *mapping, unsigned long shmem_partial_swap_usage(struct address_space *mapping, pgoff_t start, pgoff_t end) { - struct radix_tree_iter iter; - void __rcu **slot; + XA_STATE(xas, &mapping->i_pages, start); struct page *page; unsigned long swapped = 0; rcu_read_lock(); - - radix_tree_for_each_slot(slot, &mapping->i_pages, &iter, start) { - if (iter.index >= end) - break; - - page = radix_tree_deref_slot(slot); - - if (radix_tree_deref_retry(page)) { - slot = radix_tree_iter_retry(&iter); + xas_for_each(&xas, page, end - 1) { + if (xas_retry(&xas, page)) continue; - } - - if (radix_tree_exceptional_entry(page)) + if (xa_is_value(page)) swapped++; if (need_resched()) { - slot = radix_tree_iter_resume(slot, &iter); + xas_pause(&xas); cond_resched_rcu(); } } @@ -788,7 +763,7 @@ void shmem_unlock_mapping(struct address_space *mapping) } /* - * Remove range of pages and swap entries from radix tree, and free them. + * Remove range of pages and swap entries from page cache, and free them. * If !unfalloc, truncate or punch hole; if unfalloc, undo failed fallocate. */ static void shmem_undo_range(struct inode *inode, loff_t lstart, loff_t lend, @@ -824,7 +799,7 @@ static void shmem_undo_range(struct inode *inode, loff_t lstart, loff_t lend, if (index >= end) break; - if (radix_tree_exceptional_entry(page)) { + if (xa_is_value(page)) { if (unfalloc) continue; nr_swaps_freed += !shmem_free_swap(mapping, @@ -921,7 +896,7 @@ static void shmem_undo_range(struct inode *inode, loff_t lstart, loff_t lend, if (index >= end) break; - if (radix_tree_exceptional_entry(page)) { + if (xa_is_value(page)) { if (unfalloc) continue; if (shmem_free_swap(mapping, index, page)) { @@ -1110,34 +1085,27 @@ static void shmem_evict_inode(struct inode *inode) clear_inode(inode); } -static unsigned long find_swap_entry(struct radix_tree_root *root, void *item) +static unsigned long find_swap_entry(struct xarray *xa, void *item) { - struct radix_tree_iter iter; - void __rcu **slot; - unsigned long found = -1; + XA_STATE(xas, xa, 0); unsigned int checked = 0; + void *entry; rcu_read_lock(); - radix_tree_for_each_slot(slot, root, &iter, 0) { - void *entry = radix_tree_deref_slot(slot); - - if (radix_tree_deref_retry(entry)) { - slot = radix_tree_iter_retry(&iter); + xas_for_each(&xas, entry, ULONG_MAX) { + if (xas_retry(&xas, entry)) continue; - } - if (entry == item) { - found = iter.index; + if (entry == item) break; - } checked++; - if ((checked % 4096) != 0) + if ((checked % XA_CHECK_SCHED) != 0) continue; - slot = radix_tree_iter_resume(slot, &iter); + xas_pause(&xas); cond_resched_rcu(); } - rcu_read_unlock(); - return found; + + return entry ? xas.xa_index : -1; } /* @@ -1175,10 +1143,10 @@ static int shmem_unuse_inode(struct shmem_inode_info *info, * We needed to drop mutex to make that restrictive page * allocation, but the inode might have been freed while we * dropped it: although a racing shmem_evict_inode() cannot - * complete without emptying the radix_tree, our page lock + * complete without emptying the page cache, our page lock * on this swapcache page is not enough to prevent that - * free_swap_and_cache() of our swap entry will only - * trylock_page(), removing swap from radix_tree whatever. + * trylock_page(), removing swap from page cache whatever. * * We must not proceed to shmem_add_to_page_cache() if the * inode has been freed, but of course we cannot rely on @@ -1200,7 +1168,7 @@ static int shmem_unuse_inode(struct shmem_inode_info *info, */ if (!error) error = shmem_add_to_page_cache(*pagep, mapping, index, - radswap); + radswap, gfp); if (error != -ENOMEM) { /* * Truncation and eviction use free_swap_and_cache(), which @@ -1244,7 +1212,7 @@ int shmem_unuse(swp_entry_t swap, struct page *page) &memcg, false); if (error) goto out; - /* No radix_tree_preload: swap entry keeps a place for page in tree */ + /* No memory allocation: swap entry occupies the slot for the page */ error = -EAGAIN; mutex_lock(&shmem_swaplist_mutex); @@ -1453,23 +1421,17 @@ static struct page *shmem_alloc_hugepage(gfp_t gfp, struct shmem_inode_info *info, pgoff_t index) { struct vm_area_struct pvma; - struct inode *inode = &info->vfs_inode; - struct address_space *mapping = inode->i_mapping; - pgoff_t idx, hindex; - void __rcu **results; + struct address_space *mapping = info->vfs_inode.i_mapping; + pgoff_t hindex; struct page *page; if (!IS_ENABLED(CONFIG_TRANSPARENT_HUGE_PAGECACHE)) return NULL; hindex = round_down(index, HPAGE_PMD_NR); - rcu_read_lock(); - if (radix_tree_gang_lookup_slot(&mapping->i_pages, &results, &idx, - hindex, 1) && idx < hindex + HPAGE_PMD_NR) { - rcu_read_unlock(); + if (xa_find(&mapping->i_pages, &hindex, hindex + HPAGE_PMD_NR - 1, + XA_PRESENT)) return NULL; - } - rcu_read_unlock(); shmem_pseudo_vma_init(&pvma, info, hindex); page = alloc_pages_vma(gfp | __GFP_COMP | __GFP_NORETRY | __GFP_NOWARN, @@ -1578,8 +1540,7 @@ static int shmem_replace_page(struct page **pagep, gfp_t gfp, * a nice clean interface for us to replace oldpage by newpage there. */ xa_lock_irq(&swap_mapping->i_pages); - error = shmem_radix_tree_replace(swap_mapping, swap_index, oldpage, - newpage); + error = shmem_replace_entry(swap_mapping, swap_index, oldpage, newpage); if (!error) { __inc_node_page_state(newpage, NR_FILE_PAGES); __dec_node_page_state(oldpage, NR_FILE_PAGES); @@ -1643,7 +1604,7 @@ static int shmem_getpage_gfp(struct inode *inode, pgoff_t index, repeat: swap.val = 0; page = find_lock_entry(mapping, index); - if (radix_tree_exceptional_entry(page)) { + if (xa_is_value(page)) { swap = radix_to_swp_entry(page); page = NULL; } @@ -1718,7 +1679,7 @@ repeat: false); if (!error) { error = shmem_add_to_page_cache(page, mapping, index, - swp_to_radix_entry(swap)); + swp_to_radix_entry(swap), gfp); /* * We already confirmed swap under page lock, and make * no memory allocation here, so usually no possibility @@ -1824,13 +1785,8 @@ alloc_nohuge: page = shmem_alloc_and_acct_page(gfp, inode, PageTransHuge(page)); if (error) goto unacct; - error = radix_tree_maybe_preload_order(gfp & GFP_RECLAIM_MASK, - compound_order(page)); - if (!error) { - error = shmem_add_to_page_cache(page, mapping, hindex, - NULL); - radix_tree_preload_end(); - } + error = shmem_add_to_page_cache(page, mapping, hindex, + NULL, gfp & GFP_RECLAIM_MASK); if (error) { mem_cgroup_cancel_charge(page, memcg, PageTransHuge(page)); @@ -1931,7 +1887,7 @@ unlock: spin_unlock_irq(&info->lock); goto repeat; } - if (error == -EEXIST) /* from above or from radix_tree_insert */ + if (error == -EEXIST) goto repeat; return error; } @@ -2299,11 +2255,8 @@ static int shmem_mfill_atomic_pte(struct mm_struct *dst_mm, if (ret) goto out_release; - ret = radix_tree_maybe_preload(gfp & GFP_RECLAIM_MASK); - if (!ret) { - ret = shmem_add_to_page_cache(page, mapping, pgoff, NULL); - radix_tree_preload_end(); - } + ret = shmem_add_to_page_cache(page, mapping, pgoff, NULL, + gfp & GFP_RECLAIM_MASK); if (ret) goto out_release_uncharge; @@ -2548,7 +2501,7 @@ static ssize_t shmem_file_read_iter(struct kiocb *iocb, struct iov_iter *to) } /* - * llseek SEEK_DATA or SEEK_HOLE through the radix_tree. + * llseek SEEK_DATA or SEEK_HOLE through the page cache. */ static pgoff_t shmem_seek_hole_data(struct address_space *mapping, pgoff_t index, pgoff_t end, int whence) @@ -2578,7 +2531,7 @@ static pgoff_t shmem_seek_hole_data(struct address_space *mapping, index = indices[i]; } page = pvec.pages[i]; - if (page && !radix_tree_exceptional_entry(page)) { + if (page && !xa_is_value(page)) { if (!PageUptodate(page)) page = NULL; } diff --git a/mm/swap.c b/mm/swap.c @@ -964,7 +964,7 @@ void pagevec_remove_exceptionals(struct pagevec *pvec) for (i = 0, j = 0; i < pagevec_count(pvec); i++) { struct page *page = pvec->pages[i]; - if (!radix_tree_exceptional_entry(page)) + if (!xa_is_value(page)) pvec->pages[j++] = page; } pvec->nr = j; @@ -1001,7 +1001,7 @@ EXPORT_SYMBOL(pagevec_lookup_range); unsigned pagevec_lookup_range_tag(struct pagevec *pvec, struct address_space *mapping, pgoff_t *index, pgoff_t end, - int tag) + xa_mark_t tag) { pvec->nr = find_get_pages_range_tag(mapping, index, end, tag, PAGEVEC_SIZE, pvec->pages); @@ -1011,7 +1011,7 @@ EXPORT_SYMBOL(pagevec_lookup_range_tag); unsigned pagevec_lookup_range_nr_tag(struct pagevec *pvec, struct address_space *mapping, pgoff_t *index, pgoff_t end, - int tag, unsigned max_pages) + xa_mark_t tag, unsigned max_pages) { pvec->nr = find_get_pages_range_tag(mapping, index, end, tag, min_t(unsigned int, max_pages, PAGEVEC_SIZE), pvec->pages); diff --git a/mm/swap_state.c b/mm/swap_state.c @@ -107,14 +107,15 @@ void show_swap_cache_info(void) } /* - * __add_to_swap_cache resembles add_to_page_cache_locked on swapper_space, + * add_to_swap_cache resembles add_to_page_cache_locked on swapper_space, * but sets SwapCache flag and private instead of mapping and index. */ -int __add_to_swap_cache(struct page *page, swp_entry_t entry) +int add_to_swap_cache(struct page *page, swp_entry_t entry, gfp_t gfp) { - int error, i, nr = hpage_nr_pages(page); - struct address_space *address_space; + struct address_space *address_space = swap_address_space(entry); pgoff_t idx = swp_offset(entry); + XA_STATE_ORDER(xas, &address_space->i_pages, idx, compound_order(page)); + unsigned long i, nr = 1UL << compound_order(page); VM_BUG_ON_PAGE(!PageLocked(page), page); VM_BUG_ON_PAGE(PageSwapCache(page), page); @@ -123,73 +124,52 @@ int __add_to_swap_cache(struct page *page, swp_entry_t entry) page_ref_add(page, nr); SetPageSwapCache(page); - address_space = swap_address_space(entry); - xa_lock_irq(&address_space->i_pages); - for (i = 0; i < nr; i++) { - set_page_private(page + i, entry.val + i); - error = radix_tree_insert(&address_space->i_pages, - idx + i, page + i); - if (unlikely(error)) - break; - } - if (likely(!error)) { + do { + xas_lock_irq(&xas); + xas_create_range(&xas); + if (xas_error(&xas)) + goto unlock; + for (i = 0; i < nr; i++) { + VM_BUG_ON_PAGE(xas.xa_index != idx + i, page); + set_page_private(page + i, entry.val + i); + xas_store(&xas, page + i); + xas_next(&xas); + } address_space->nrpages += nr; __mod_node_page_state(page_pgdat(page), NR_FILE_PAGES, nr); ADD_CACHE_INFO(add_total, nr); - } else { - /* - * Only the context which have set SWAP_HAS_CACHE flag - * would call add_to_swap_cache(). - * So add_to_swap_cache() doesn't returns -EEXIST. - */ - VM_BUG_ON(error == -EEXIST); - set_page_private(page + i, 0UL); - while (i--) { - radix_tree_delete(&address_space->i_pages, idx + i); - set_page_private(page + i, 0UL); - } - ClearPageSwapCache(page); - page_ref_sub(page, nr); - } - xa_unlock_irq(&address_space->i_pages); +unlock: + xas_unlock_irq(&xas); + } while (xas_nomem(&xas, gfp)); - return error; -} - - -int add_to_swap_cache(struct page *page, swp_entry_t entry, gfp_t gfp_mask) -{ - int error; + if (!xas_error(&xas)) + return 0; - error = radix_tree_maybe_preload_order(gfp_mask, compound_order(page)); - if (!error) { - error = __add_to_swap_cache(page, entry); - radix_tree_preload_end(); - } - return error; + ClearPageSwapCache(page); + page_ref_sub(page, nr); + return xas_error(&xas); } /* * This must be called only on pages that have * been verified to be in the swap cache. */ -void __delete_from_swap_cache(struct page *page) +void __delete_from_swap_cache(struct page *page, swp_entry_t entry) { - struct address_space *address_space; + struct address_space *address_space = swap_address_space(entry); int i, nr = hpage_nr_pages(page); - swp_entry_t entry; - pgoff_t idx; + pgoff_t idx = swp_offset(entry); + XA_STATE(xas, &address_space->i_pages, idx); VM_BUG_ON_PAGE(!PageLocked(page), page); VM_BUG_ON_PAGE(!PageSwapCache(page), page); VM_BUG_ON_PAGE(PageWriteback(page), page); - entry.val = page_private(page); - address_space = swap_address_space(entry); - idx = swp_offset(entry); for (i = 0; i < nr; i++) { - radix_tree_delete(&address_space->i_pages, idx + i); + void *entry = xas_store(&xas, NULL); + VM_BUG_ON_PAGE(entry != page + i, entry); set_page_private(page + i, 0); + xas_next(&xas); } ClearPageSwapCache(page); address_space->nrpages -= nr; @@ -217,7 +197,7 @@ int add_to_swap(struct page *page) return 0; /* - * Radix-tree node allocations from PF_MEMALLOC contexts could + * XArray node allocations from PF_MEMALLOC contexts could * completely exhaust the page allocator. __GFP_NOMEMALLOC * stops emergency reserves from being allocated. * @@ -229,7 +209,6 @@ int add_to_swap(struct page *page) */ err = add_to_swap_cache(page, entry, __GFP_HIGH|__GFP_NOMEMALLOC|__GFP_NOWARN); - /* -ENOMEM radix-tree allocation failure */ if (err) /* * add_to_swap_cache() doesn't return -EEXIST, so we can safely @@ -263,14 +242,11 @@ fail: */ void delete_from_swap_cache(struct page *page) { - swp_entry_t entry; - struct address_space *address_space; + swp_entry_t entry = { .val = page_private(page) }; + struct address_space *address_space = swap_address_space(entry); - entry.val = page_private(page); - - address_space = swap_address_space(entry); xa_lock_irq(&address_space->i_pages); - __delete_from_swap_cache(page); + __delete_from_swap_cache(page, entry); xa_unlock_irq(&address_space->i_pages); put_swap_page(page, entry); @@ -414,18 +390,10 @@ struct page *__read_swap_cache_async(swp_entry_t entry, gfp_t gfp_mask, } /* - * call radix_tree_preload() while we can wait. - */ - err = radix_tree_maybe_preload(gfp_mask & GFP_KERNEL); - if (err) - break; - - /* * Swap entry may have been freed since our caller observed it. */ err = swapcache_prepare(entry); if (err == -EEXIST) { - radix_tree_preload_end(); /* * We might race against get_swap_page() and stumble * across a SWAP_HAS_CACHE swap_map entry whose page @@ -433,27 +401,20 @@ struct page *__read_swap_cache_async(swp_entry_t entry, gfp_t gfp_mask, */ cond_resched(); continue; - } - if (err) { /* swp entry is obsolete ? */ - radix_tree_preload_end(); + } else if (err) /* swp entry is obsolete ? */ break; - } - /* May fail (-ENOMEM) if radix-tree node allocation failed. */ + /* May fail (-ENOMEM) if XArray node allocation failed. */ __SetPageLocked(new_page); __SetPageSwapBacked(new_page); - err = __add_to_swap_cache(new_page, entry); + err = add_to_swap_cache(new_page, entry, gfp_mask & GFP_KERNEL); if (likely(!err)) { - radix_tree_preload_end(); - /* - * Initiate read into locked page and return. - */ + /* Initiate read into locked page */ SetPageWorkingset(new_page); lru_cache_add_anon(new_page); *new_page_allocated = true; return new_page; } - radix_tree_preload_end(); __ClearPageLocked(new_page); /* * add_to_swap_cache() doesn't return -EEXIST, so we can safely @@ -626,7 +587,7 @@ int init_swap_address_space(unsigned int type, unsigned long nr_pages) return -ENOMEM; for (i = 0; i < nr; i++) { space = spaces + i; - INIT_RADIX_TREE(&space->i_pages, GFP_ATOMIC|__GFP_NOWARN); + xa_init_flags(&space->i_pages, XA_FLAGS_LOCK_IRQ); atomic_set(&space->i_mmap_writable, 0); space->a_ops = &swap_aops; /* swap cache doesn't use writeback related tags */ diff --git a/mm/truncate.c b/mm/truncate.c @@ -33,15 +33,12 @@ static inline void __clear_shadow_entry(struct address_space *mapping, pgoff_t index, void *entry) { - struct radix_tree_node *node; - void **slot; + XA_STATE(xas, &mapping->i_pages, index); - if (!__radix_tree_lookup(&mapping->i_pages, index, &node, &slot)) + xas_set_update(&xas, workingset_update_node); + if (xas_load(&xas) != entry) return; - if (*slot != entry) - return; - __radix_tree_replace(&mapping->i_pages, node, slot, NULL, - workingset_update_node); + xas_store(&xas, NULL); mapping->nrexceptional--; } @@ -70,7 +67,7 @@ static void truncate_exceptional_pvec_entries(struct address_space *mapping, return; for (j = 0; j < pagevec_count(pvec); j++) - if (radix_tree_exceptional_entry(pvec->pages[j])) + if (xa_is_value(pvec->pages[j])) break; if (j == pagevec_count(pvec)) @@ -85,7 +82,7 @@ static void truncate_exceptional_pvec_entries(struct address_space *mapping, struct page *page = pvec->pages[i]; pgoff_t index = indices[i]; - if (!radix_tree_exceptional_entry(page)) { + if (!xa_is_value(page)) { pvec->pages[j++] = page; continue; } @@ -347,7 +344,7 @@ void truncate_inode_pages_range(struct address_space *mapping, if (index >= end) break; - if (radix_tree_exceptional_entry(page)) + if (xa_is_value(page)) continue; if (!trylock_page(page)) @@ -442,7 +439,7 @@ void truncate_inode_pages_range(struct address_space *mapping, break; } - if (radix_tree_exceptional_entry(page)) + if (xa_is_value(page)) continue; lock_page(page); @@ -561,7 +558,7 @@ unsigned long invalidate_mapping_pages(struct address_space *mapping, if (index > end) break; - if (radix_tree_exceptional_entry(page)) { + if (xa_is_value(page)) { invalidate_exceptional_entry(mapping, index, page); continue; @@ -692,7 +689,7 @@ int invalidate_inode_pages2_range(struct address_space *mapping, if (index > end) break; - if (radix_tree_exceptional_entry(page)) { + if (xa_is_value(page)) { if (!invalidate_exceptional_entry2(mapping, index, page)) ret = -EBUSY; @@ -738,10 +735,10 @@ int invalidate_inode_pages2_range(struct address_space *mapping, index++; } /* - * For DAX we invalidate page tables after invalidating radix tree. We + * For DAX we invalidate page tables after invalidating page cache. We * could invalidate page tables while invalidating each entry however * that would be expensive. And doing range unmapping before doesn't - * work as we have no cheap way to find whether radix tree entry didn't + * work as we have no cheap way to find whether page cache entry didn't * get remapped later. */ if (dax_mapping(mapping)) { diff --git a/mm/vmscan.c b/mm/vmscan.c @@ -751,12 +751,12 @@ static inline int is_page_cache_freeable(struct page *page) { /* * A freeable page cache page is referenced only by the caller - * that isolated the page, the page cache radix tree and - * optional buffer heads at page->private. + * that isolated the page, the page cache and optional buffer + * heads at page->private. */ - int radix_pins = PageTransHuge(page) && PageSwapCache(page) ? + int page_cache_pins = PageTransHuge(page) && PageSwapCache(page) ? HPAGE_PMD_NR : 1; - return page_count(page) - page_has_private(page) == 1 + radix_pins; + return page_count(page) - page_has_private(page) == 1 + page_cache_pins; } static int may_write_to_inode(struct inode *inode, struct scan_control *sc) @@ -932,7 +932,7 @@ static int __remove_mapping(struct address_space *mapping, struct page *page, if (PageSwapCache(page)) { swp_entry_t swap = { .val = page_private(page) }; mem_cgroup_swapout(page, swap); - __delete_from_swap_cache(page); + __delete_from_swap_cache(page, swap); xa_unlock_irqrestore(&mapping->i_pages, flags); put_swap_page(page, swap); } else { diff --git a/mm/workingset.c b/mm/workingset.c @@ -160,20 +160,20 @@ * and activations is maintained (node->inactive_age). * * On eviction, a snapshot of this counter (along with some bits to - * identify the node) is stored in the now empty page cache radix tree + * identify the node) is stored in the now empty page cache * slot of the evicted page. This is called a shadow entry. * * On cache misses for which there are shadow entries, an eligible * refault distance will immediately activate the refaulting page. */ -#define EVICTION_SHIFT (RADIX_TREE_EXCEPTIONAL_ENTRY + \ +#define EVICTION_SHIFT ((BITS_PER_LONG - BITS_PER_XA_VALUE) + \ 1 + NODES_SHIFT + MEM_CGROUP_ID_SHIFT) #define EVICTION_MASK (~0UL >> EVICTION_SHIFT) /* * Eviction timestamps need to be able to cover the full range of - * actionable refaults. However, bits are tight in the radix tree + * actionable refaults. However, bits are tight in the xarray * entry, and after storing the identifier for the lruvec there might * not be enough left to represent every single actionable refault. In * that case, we have to sacrifice granularity for distance, and group @@ -185,22 +185,21 @@ static void *pack_shadow(int memcgid, pg_data_t *pgdat, unsigned long eviction, bool workingset) { eviction >>= bucket_order; + eviction &= EVICTION_MASK; eviction = (eviction << MEM_CGROUP_ID_SHIFT) | memcgid; eviction = (eviction << NODES_SHIFT) | pgdat->node_id; eviction = (eviction << 1) | workingset; - eviction = (eviction << RADIX_TREE_EXCEPTIONAL_SHIFT); - return (void *)(eviction | RADIX_TREE_EXCEPTIONAL_ENTRY); + return xa_mk_value(eviction); } static void unpack_shadow(void *shadow, int *memcgidp, pg_data_t **pgdat, unsigned long *evictionp, bool *workingsetp) { - unsigned long entry = (unsigned long)shadow; + unsigned long entry = xa_to_value(shadow); int memcgid, nid; bool workingset; - entry >>= RADIX_TREE_EXCEPTIONAL_SHIFT; workingset = entry & 1; entry >>= 1; nid = entry & ((1UL << NODES_SHIFT) - 1); @@ -367,7 +366,7 @@ out: static struct list_lru shadow_nodes; -void workingset_update_node(struct radix_tree_node *node) +void workingset_update_node(struct xa_node *node) { /* * Track non-empty nodes that contain only shadow entries; @@ -379,7 +378,7 @@ void workingset_update_node(struct radix_tree_node *node) */ VM_WARN_ON_ONCE(!irqs_disabled()); /* For __inc_lruvec_page_state */ - if (node->count && node->count == node->exceptional) { + if (node->count && node->count == node->nr_values) { if (list_empty(&node->private_list)) { list_lru_add(&shadow_nodes, &node->private_list); __inc_lruvec_page_state(virt_to_page(node), @@ -404,7 +403,7 @@ static unsigned long count_shadow_nodes(struct shrinker *shrinker, nodes = list_lru_shrink_count(&shadow_nodes, sc); /* - * Approximate a reasonable limit for the radix tree nodes + * Approximate a reasonable limit for the nodes * containing shadow entries. We don't need to keep more * shadow entries than possible pages on the active list, * since refault distances bigger than that are dismissed. @@ -419,11 +418,11 @@ static unsigned long count_shadow_nodes(struct shrinker *shrinker, * worst-case density of 1/8th. Below that, not all eligible * refaults can be detected anymore. * - * On 64-bit with 7 radix_tree_nodes per page and 64 slots + * On 64-bit with 7 xa_nodes per page and 64 slots * each, this will reclaim shadow entries when they consume * ~1.8% of available memory: * - * PAGE_SIZE / radix_tree_nodes / node_entries * 8 / PAGE_SIZE + * PAGE_SIZE / xa_nodes / node_entries * 8 / PAGE_SIZE */ #ifdef CONFIG_MEMCG if (sc->memcg) { @@ -438,7 +437,7 @@ static unsigned long count_shadow_nodes(struct shrinker *shrinker, #endif pages = node_present_pages(sc->nid); - max_nodes = pages >> (RADIX_TREE_MAP_SHIFT - 3); + max_nodes = pages >> (XA_CHUNK_SHIFT - 3); if (!nodes) return SHRINK_EMPTY; @@ -451,11 +450,11 @@ static unsigned long count_shadow_nodes(struct shrinker *shrinker, static enum lru_status shadow_lru_isolate(struct list_head *item, struct list_lru_one *lru, spinlock_t *lru_lock, - void *arg) + void *arg) __must_hold(lru_lock) { + struct xa_node *node = container_of(item, struct xa_node, private_list); + XA_STATE(xas, node->array, 0); struct address_space *mapping; - struct radix_tree_node *node; - unsigned int i; int ret; /* @@ -463,15 +462,14 @@ static enum lru_status shadow_lru_isolate(struct list_head *item, * the shadow node LRU under the i_pages lock and the * lru_lock. Because the page cache tree is emptied before * the inode can be destroyed, holding the lru_lock pins any - * address_space that has radix tree nodes on the LRU. + * address_space that has nodes on the LRU. * * We can then safely transition to the i_pages lock to * pin only the address_space of the particular node we want * to reclaim, take the node off-LRU, and drop the lru_lock. */ - node = container_of(item, struct radix_tree_node, private_list); - mapping = container_of(node->root, struct address_space, i_pages); + mapping = container_of(node->array, struct address_space, i_pages); /* Coming from the list, invert the lock order */ if (!xa_trylock(&mapping->i_pages)) { @@ -490,29 +488,21 @@ static enum lru_status shadow_lru_isolate(struct list_head *item, * no pages, so we expect to be able to remove them all and * delete and free the empty node afterwards. */ - if (WARN_ON_ONCE(!node->exceptional)) + if (WARN_ON_ONCE(!node->nr_values)) goto out_invalid; - if (WARN_ON_ONCE(node->count != node->exceptional)) - goto out_invalid; - for (i = 0; i < RADIX_TREE_MAP_SIZE; i++) { - if (node->slots[i]) { - if (WARN_ON_ONCE(!radix_tree_exceptional_entry(node->slots[i]))) - goto out_invalid; - if (WARN_ON_ONCE(!node->exceptional)) - goto out_invalid; - if (WARN_ON_ONCE(!mapping->nrexceptional)) - goto out_invalid; - node->slots[i] = NULL; - node->exceptional--; - node->count--; - mapping->nrexceptional--; - } - } - if (WARN_ON_ONCE(node->exceptional)) + if (WARN_ON_ONCE(node->count != node->nr_values)) goto out_invalid; + mapping->nrexceptional -= node->nr_values; + xas.xa_node = xa_parent_locked(&mapping->i_pages, node); + xas.xa_offset = node->offset; + xas.xa_shift = node->shift + XA_CHUNK_SHIFT; + xas_set_update(&xas, workingset_update_node); + /* + * We could store a shadow entry here which was the minimum of the + * shadow entries we were tracking ... + */ + xas_store(&xas, NULL); __inc_lruvec_page_state(virt_to_page(node), WORKINGSET_NODERECLAIM); - __radix_tree_delete_node(&mapping->i_pages, node, - workingset_lookup_update(mapping)); out_invalid: xa_unlock_irq(&mapping->i_pages); diff --git a/tools/include/asm-generic/bitops.h b/tools/include/asm-generic/bitops.h @@ -27,5 +27,6 @@ #include <asm-generic/bitops/hweight.h> #include <asm-generic/bitops/atomic.h> +#include <asm-generic/bitops/non-atomic.h> #endif /* __TOOLS_ASM_GENERIC_BITOPS_H */ diff --git a/tools/include/asm-generic/bitops/atomic.h b/tools/include/asm-generic/bitops/atomic.h @@ -15,13 +15,4 @@ static inline void clear_bit(int nr, unsigned long *addr) addr[nr / __BITS_PER_LONG] &= ~(1UL << (nr % __BITS_PER_LONG)); } -static __always_inline int test_bit(unsigned int nr, const unsigned long *addr) -{ - return ((1UL << (nr % __BITS_PER_LONG)) & - (((unsigned long *)addr)[nr / __BITS_PER_LONG])) != 0; -} - -#define __set_bit(nr, addr) set_bit(nr, addr) -#define __clear_bit(nr, addr) clear_bit(nr, addr) - #endif /* _TOOLS_LINUX_ASM_GENERIC_BITOPS_ATOMIC_H_ */ diff --git a/tools/include/asm-generic/bitops/non-atomic.h b/tools/include/asm-generic/bitops/non-atomic.h @@ -0,0 +1,109 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _ASM_GENERIC_BITOPS_NON_ATOMIC_H_ +#define _ASM_GENERIC_BITOPS_NON_ATOMIC_H_ + +#include <asm/types.h> + +/** + * __set_bit - Set a bit in memory + * @nr: the bit to set + * @addr: the address to start counting from + * + * Unlike set_bit(), this function is non-atomic and may be reordered. + * If it's called on the same region of memory simultaneously, the effect + * may be that only one operation succeeds. + */ +static inline void __set_bit(int nr, volatile unsigned long *addr) +{ + unsigned long mask = BIT_MASK(nr); + unsigned long *p = ((unsigned long *)addr) + BIT_WORD(nr); + + *p |= mask; +} + +static inline void __clear_bit(int nr, volatile unsigned long *addr) +{ + unsigned long mask = BIT_MASK(nr); + unsigned long *p = ((unsigned long *)addr) + BIT_WORD(nr); + + *p &= ~mask; +} + +/** + * __change_bit - Toggle a bit in memory + * @nr: the bit to change + * @addr: the address to start counting from + * + * Unlike change_bit(), this function is non-atomic and may be reordered. + * If it's called on the same region of memory simultaneously, the effect + * may be that only one operation succeeds. + */ +static inline void __change_bit(int nr, volatile unsigned long *addr) +{ + unsigned long mask = BIT_MASK(nr); + unsigned long *p = ((unsigned long *)addr) + BIT_WORD(nr); + + *p ^= mask; +} + +/** + * __test_and_set_bit - Set a bit and return its old value + * @nr: Bit to set + * @addr: Address to count from + * + * This operation is non-atomic and can be reordered. + * If two examples of this operation race, one can appear to succeed + * but actually fail. You must protect multiple accesses with a lock. + */ +static inline int __test_and_set_bit(int nr, volatile unsigned long *addr) +{ + unsigned long mask = BIT_MASK(nr); + unsigned long *p = ((unsigned long *)addr) + BIT_WORD(nr); + unsigned long old = *p; + + *p = old | mask; + return (old & mask) != 0; +} + +/** + * __test_and_clear_bit - Clear a bit and return its old value + * @nr: Bit to clear + * @addr: Address to count from + * + * This operation is non-atomic and can be reordered. + * If two examples of this operation race, one can appear to succeed + * but actually fail. You must protect multiple accesses with a lock. + */ +static inline int __test_and_clear_bit(int nr, volatile unsigned long *addr) +{ + unsigned long mask = BIT_MASK(nr); + unsigned long *p = ((unsigned long *)addr) + BIT_WORD(nr); + unsigned long old = *p; + + *p = old & ~mask; + return (old & mask) != 0; +} + +/* WARNING: non atomic and it can be reordered! */ +static inline int __test_and_change_bit(int nr, + volatile unsigned long *addr) +{ + unsigned long mask = BIT_MASK(nr); + unsigned long *p = ((unsigned long *)addr) + BIT_WORD(nr); + unsigned long old = *p; + + *p = old ^ mask; + return (old & mask) != 0; +} + +/** + * test_bit - Determine whether a bit is set + * @nr: bit number to test + * @addr: Address to start counting from + */ +static inline int test_bit(int nr, const volatile unsigned long *addr) +{ + return 1UL & (addr[BIT_WORD(nr)] >> (nr & (BITS_PER_LONG-1))); +} + +#endif /* _ASM_GENERIC_BITOPS_NON_ATOMIC_H_ */ diff --git a/tools/include/linux/bitmap.h b/tools/include/linux/bitmap.h @@ -15,6 +15,7 @@ void __bitmap_or(unsigned long *dst, const unsigned long *bitmap1, const unsigned long *bitmap2, int bits); int __bitmap_and(unsigned long *dst, const unsigned long *bitmap1, const unsigned long *bitmap2, unsigned int bits); +void bitmap_clear(unsigned long *map, unsigned int start, int len); #define BITMAP_FIRST_WORD_MASK(start) (~0UL << ((start) & (BITS_PER_LONG - 1))) diff --git a/tools/include/linux/kernel.h b/tools/include/linux/kernel.h @@ -70,6 +70,7 @@ #define BUG_ON(cond) assert(!(cond)) #endif #endif +#define BUG() BUG_ON(1) #if __BYTE_ORDER == __BIG_ENDIAN #define cpu_to_le16 bswap_16 diff --git a/tools/include/linux/spinlock.h b/tools/include/linux/spinlock.h @@ -8,8 +8,14 @@ #define spinlock_t pthread_mutex_t #define DEFINE_SPINLOCK(x) pthread_mutex_t x = PTHREAD_MUTEX_INITIALIZER #define __SPIN_LOCK_UNLOCKED(x) (pthread_mutex_t)PTHREAD_MUTEX_INITIALIZER -#define spin_lock_init(x) pthread_mutex_init(x, NULL) - +#define spin_lock_init(x) pthread_mutex_init(x, NULL) + +#define spin_lock(x) pthread_mutex_lock(x) +#define spin_unlock(x) pthread_mutex_unlock(x) +#define spin_lock_bh(x) pthread_mutex_lock(x) +#define spin_unlock_bh(x) pthread_mutex_unlock(x) +#define spin_lock_irq(x) pthread_mutex_lock(x) +#define spin_unlock_irq(x) pthread_mutex_unlock(x) #define spin_lock_irqsave(x, f) (void)f, pthread_mutex_lock(x) #define spin_unlock_irqrestore(x, f) (void)f, pthread_mutex_unlock(x) @@ -31,4 +37,6 @@ static inline bool arch_spin_is_locked(arch_spinlock_t *mutex) return true; } +#include <linux/lockdep.h> + #endif diff --git a/tools/testing/radix-tree/.gitignore b/tools/testing/radix-tree/.gitignore @@ -4,3 +4,4 @@ idr-test main multiorder radix-tree.c +xarray diff --git a/tools/testing/radix-tree/Makefile b/tools/testing/radix-tree/Makefile @@ -4,8 +4,8 @@ CFLAGS += -I. -I../../include -g -Og -Wall -D_LGPL_SOURCE -fsanitize=address \ -fsanitize=undefined LDFLAGS += -fsanitize=address -fsanitize=undefined LDLIBS+= -lpthread -lurcu -TARGETS = main idr-test multiorder -CORE_OFILES := radix-tree.o idr.o linux.o test.o find_bit.o +TARGETS = main idr-test multiorder xarray +CORE_OFILES := xarray.o radix-tree.o idr.o linux.o test.o find_bit.o bitmap.o OFILES = main.o $(CORE_OFILES) regression1.o regression2.o regression3.o \ tag_check.o multiorder.o idr-test.o iteration_check.o benchmark.o @@ -25,6 +25,8 @@ main: $(OFILES) idr-test.o: ../../../lib/test_ida.c idr-test: idr-test.o $(CORE_OFILES) +xarray: $(CORE_OFILES) + multiorder: multiorder.o $(CORE_OFILES) clean: @@ -35,6 +37,7 @@ vpath %.c ../../lib $(OFILES): Makefile *.h */*.h generated/map-shift.h \ ../../include/linux/*.h \ ../../include/asm/*.h \ + ../../../include/linux/xarray.h \ ../../../include/linux/radix-tree.h \ ../../../include/linux/idr.h @@ -44,8 +47,10 @@ radix-tree.c: ../../../lib/radix-tree.c idr.c: ../../../lib/idr.c sed -e 's/^static //' -e 's/__always_inline //' -e 's/inline //' < $< > $@ +xarray.o: ../../../lib/xarray.c ../../../lib/test_xarray.c + generated/map-shift.h: @if ! grep -qws $(SHIFT) generated/map-shift.h; then \ - echo "#define RADIX_TREE_MAP_SHIFT $(SHIFT)" > \ + echo "#define XA_CHUNK_SHIFT $(SHIFT)" > \ generated/map-shift.h; \ fi diff --git a/tools/testing/radix-tree/benchmark.c b/tools/testing/radix-tree/benchmark.c @@ -17,9 +17,6 @@ #include <time.h> #include "test.h" -#define for_each_index(i, base, order) \ - for (i = base; i < base + (1 << order); i++) - #define NSEC_PER_SEC 1000000000L static long long benchmark_iter(struct radix_tree_root *root, bool tagged) @@ -61,7 +58,7 @@ again: } static void benchmark_insert(struct radix_tree_root *root, - unsigned long size, unsigned long step, int order) + unsigned long size, unsigned long step) { struct timespec start, finish; unsigned long index; @@ -70,19 +67,19 @@ static void benchmark_insert(struct radix_tree_root *root, clock_gettime(CLOCK_MONOTONIC, &start); for (index = 0 ; index < size ; index += step) - item_insert_order(root, index, order); + item_insert(root, index); clock_gettime(CLOCK_MONOTONIC, &finish); nsec = (finish.tv_sec - start.tv_sec) * NSEC_PER_SEC + (finish.tv_nsec - start.tv_nsec); - printv(2, "Size: %8ld, step: %8ld, order: %d, insertion: %15lld ns\n", - size, step, order, nsec); + printv(2, "Size: %8ld, step: %8ld, insertion: %15lld ns\n", + size, step, nsec); } static void benchmark_tagging(struct radix_tree_root *root, - unsigned long size, unsigned long step, int order) + unsigned long size, unsigned long step) { struct timespec start, finish; unsigned long index; @@ -98,138 +95,53 @@ static void benchmark_tagging(struct radix_tree_root *root, nsec = (finish.tv_sec - start.tv_sec) * NSEC_PER_SEC + (finish.tv_nsec - start.tv_nsec); - printv(2, "Size: %8ld, step: %8ld, order: %d, tagging: %17lld ns\n", - size, step, order, nsec); + printv(2, "Size: %8ld, step: %8ld, tagging: %17lld ns\n", + size, step, nsec); } static void benchmark_delete(struct radix_tree_root *root, - unsigned long size, unsigned long step, int order) + unsigned long size, unsigned long step) { struct timespec start, finish; - unsigned long index, i; + unsigned long index; long long nsec; clock_gettime(CLOCK_MONOTONIC, &start); for (index = 0 ; index < size ; index += step) - for_each_index(i, index, order) - item_delete(root, i); + item_delete(root, index); clock_gettime(CLOCK_MONOTONIC, &finish); nsec = (finish.tv_sec - start.tv_sec) * NSEC_PER_SEC + (finish.tv_nsec - start.tv_nsec); - printv(2, "Size: %8ld, step: %8ld, order: %d, deletion: %16lld ns\n", - size, step, order, nsec); + printv(2, "Size: %8ld, step: %8ld, deletion: %16lld ns\n", + size, step, nsec); } -static void benchmark_size(unsigned long size, unsigned long step, int order) +static void benchmark_size(unsigned long size, unsigned long step) { RADIX_TREE(tree, GFP_KERNEL); long long normal, tagged; - benchmark_insert(&tree, size, step, order); - benchmark_tagging(&tree, size, step, order); + benchmark_insert(&tree, size, step); + benchmark_tagging(&tree, size, step); tagged = benchmark_iter(&tree, true); normal = benchmark_iter(&tree, false); - printv(2, "Size: %8ld, step: %8ld, order: %d, tagged iteration: %8lld ns\n", - size, step, order, tagged); - printv(2, "Size: %8ld, step: %8ld, order: %d, normal iteration: %8lld ns\n", - size, step, order, normal); + printv(2, "Size: %8ld, step: %8ld, tagged iteration: %8lld ns\n", + size, step, tagged); + printv(2, "Size: %8ld, step: %8ld, normal iteration: %8lld ns\n", + size, step, normal); - benchmark_delete(&tree, size, step, order); + benchmark_delete(&tree, size, step); item_kill_tree(&tree); rcu_barrier(); } -static long long __benchmark_split(unsigned long index, - int old_order, int new_order) -{ - struct timespec start, finish; - long long nsec; - RADIX_TREE(tree, GFP_ATOMIC); - - item_insert_order(&tree, index, old_order); - - clock_gettime(CLOCK_MONOTONIC, &start); - radix_tree_split(&tree, index, new_order); - clock_gettime(CLOCK_MONOTONIC, &finish); - nsec = (finish.tv_sec - start.tv_sec) * NSEC_PER_SEC + - (finish.tv_nsec - start.tv_nsec); - - item_kill_tree(&tree); - - return nsec; - -} - -static void benchmark_split(unsigned long size, unsigned long step) -{ - int i, j, idx; - long long nsec = 0; - - - for (idx = 0; idx < size; idx += step) { - for (i = 3; i < 11; i++) { - for (j = 0; j < i; j++) { - nsec += __benchmark_split(idx, i, j); - } - } - } - - printv(2, "Size %8ld, step %8ld, split time %10lld ns\n", - size, step, nsec); - -} - -static long long __benchmark_join(unsigned long index, - unsigned order1, unsigned order2) -{ - unsigned long loc; - struct timespec start, finish; - long long nsec; - void *item, *item2 = item_create(index + 1, order1); - RADIX_TREE(tree, GFP_KERNEL); - - item_insert_order(&tree, index, order2); - item = radix_tree_lookup(&tree, index); - - clock_gettime(CLOCK_MONOTONIC, &start); - radix_tree_join(&tree, index + 1, order1, item2); - clock_gettime(CLOCK_MONOTONIC, &finish); - nsec = (finish.tv_sec - start.tv_sec) * NSEC_PER_SEC + - (finish.tv_nsec - start.tv_nsec); - - loc = find_item(&tree, item); - if (loc == -1) - free(item); - - item_kill_tree(&tree); - - return nsec; -} - -static void benchmark_join(unsigned long step) -{ - int i, j, idx; - long long nsec = 0; - - for (idx = 0; idx < 1 << 10; idx += step) { - for (i = 1; i < 15; i++) { - for (j = 0; j < i; j++) { - nsec += __benchmark_join(idx, i, j); - } - } - } - - printv(2, "Size %8d, step %8ld, join time %10lld ns\n", - 1 << 10, step, nsec); -} - void benchmark(void) { unsigned long size[] = {1 << 10, 1 << 20, 0}; @@ -242,16 +154,5 @@ void benchmark(void) for (c = 0; size[c]; c++) for (s = 0; step[s]; s++) - benchmark_size(size[c], step[s], 0); - - for (c = 0; size[c]; c++) - for (s = 0; step[s]; s++) - benchmark_size(size[c], step[s] << 9, 9); - - for (c = 0; size[c]; c++) - for (s = 0; step[s]; s++) - benchmark_split(size[c], step[s]); - - for (s = 0; step[s]; s++) - benchmark_join(step[s]); + benchmark_size(size[c], step[s]); } diff --git a/tools/testing/radix-tree/bitmap.c b/tools/testing/radix-tree/bitmap.c @@ -0,0 +1,23 @@ +/* lib/bitmap.c pulls in at least two other files. */ + +#include <linux/bitmap.h> + +void bitmap_clear(unsigned long *map, unsigned int start, int len) +{ + unsigned long *p = map + BIT_WORD(start); + const unsigned int size = start + len; + int bits_to_clear = BITS_PER_LONG - (start % BITS_PER_LONG); + unsigned long mask_to_clear = BITMAP_FIRST_WORD_MASK(start); + + while (len - bits_to_clear >= 0) { + *p &= ~mask_to_clear; + len -= bits_to_clear; + bits_to_clear = BITS_PER_LONG; + mask_to_clear = ~0UL; + p++; + } + if (len) { + mask_to_clear &= BITMAP_LAST_WORD_MASK(size); + *p &= ~mask_to_clear; + } +} diff --git a/tools/testing/radix-tree/generated/autoconf.h b/tools/testing/radix-tree/generated/autoconf.h @@ -1 +1 @@ -#define CONFIG_RADIX_TREE_MULTIORDER 1 +#define CONFIG_XARRAY_MULTI 1 diff --git a/tools/testing/radix-tree/idr-test.c b/tools/testing/radix-tree/idr-test.c @@ -19,7 +19,7 @@ #include "test.h" -#define DUMMY_PTR ((void *)0x12) +#define DUMMY_PTR ((void *)0x10) int item_idr_free(int id, void *p, void *data) { @@ -227,6 +227,66 @@ void idr_u32_test(int base) idr_u32_test1(&idr, 0xffffffff); } +static void idr_align_test(struct idr *idr) +{ + char name[] = "Motorola 68000"; + int i, id; + void *entry; + + for (i = 0; i < 9; i++) { + BUG_ON(idr_alloc(idr, &name[i], 0, 0, GFP_KERNEL) != i); + idr_for_each_entry(idr, entry, id); + } + idr_destroy(idr); + + for (i = 1; i < 10; i++) { + BUG_ON(idr_alloc(idr, &name[i], 0, 0, GFP_KERNEL) != i - 1); + idr_for_each_entry(idr, entry, id); + } + idr_destroy(idr); + + for (i = 2; i < 11; i++) { + BUG_ON(idr_alloc(idr, &name[i], 0, 0, GFP_KERNEL) != i - 2); + idr_for_each_entry(idr, entry, id); + } + idr_destroy(idr); + + for (i = 3; i < 12; i++) { + BUG_ON(idr_alloc(idr, &name[i], 0, 0, GFP_KERNEL) != i - 3); + idr_for_each_entry(idr, entry, id); + } + idr_destroy(idr); + + for (i = 0; i < 8; i++) { + BUG_ON(idr_alloc(idr, &name[i], 0, 0, GFP_KERNEL) != 0); + BUG_ON(idr_alloc(idr, &name[i + 1], 0, 0, GFP_KERNEL) != 1); + idr_for_each_entry(idr, entry, id); + idr_remove(idr, 1); + idr_for_each_entry(idr, entry, id); + idr_remove(idr, 0); + BUG_ON(!idr_is_empty(idr)); + } + + for (i = 0; i < 8; i++) { + BUG_ON(idr_alloc(idr, NULL, 0, 0, GFP_KERNEL) != 0); + idr_for_each_entry(idr, entry, id); + idr_replace(idr, &name[i], 0); + idr_for_each_entry(idr, entry, id); + BUG_ON(idr_find(idr, 0) != &name[i]); + idr_remove(idr, 0); + } + + for (i = 0; i < 8; i++) { + BUG_ON(idr_alloc(idr, &name[i], 0, 0, GFP_KERNEL) != 0); + BUG_ON(idr_alloc(idr, NULL, 0, 0, GFP_KERNEL) != 1); + idr_remove(idr, 1); + idr_for_each_entry(idr, entry, id); + idr_replace(idr, &name[i + 1], 0); + idr_for_each_entry(idr, entry, id); + idr_remove(idr, 0); + } +} + void idr_checks(void) { unsigned long i; @@ -307,6 +367,7 @@ void idr_checks(void) idr_u32_test(4); idr_u32_test(1); idr_u32_test(0); + idr_align_test(&idr); } #define module_init(x) @@ -344,16 +405,16 @@ void ida_check_conv_user(void) DEFINE_IDA(ida); unsigned long i; - radix_tree_cpu_dead(1); for (i = 0; i < 1000000; i++) { int id = ida_alloc(&ida, GFP_NOWAIT); if (id == -ENOMEM) { - IDA_BUG_ON(&ida, (i % IDA_BITMAP_BITS) != - BITS_PER_LONG - 2); + IDA_BUG_ON(&ida, ((i % IDA_BITMAP_BITS) != + BITS_PER_XA_VALUE) && + ((i % IDA_BITMAP_BITS) != 0)); id = ida_alloc(&ida, GFP_KERNEL); } else { IDA_BUG_ON(&ida, (i % IDA_BITMAP_BITS) == - BITS_PER_LONG - 2); + BITS_PER_XA_VALUE); } IDA_BUG_ON(&ida, id != i); } diff --git a/tools/testing/radix-tree/iteration_check.c b/tools/testing/radix-tree/iteration_check.c @@ -1,5 +1,5 @@ /* - * iteration_check.c: test races having to do with radix tree iteration + * iteration_check.c: test races having to do with xarray iteration * Copyright (c) 2016 Intel Corporation * Author: Ross Zwisler <ross.zwisler@linux.intel.com> * @@ -12,41 +12,54 @@ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for * more details. */ -#include <linux/radix-tree.h> #include <pthread.h> #include "test.h" #define NUM_THREADS 5 #define MAX_IDX 100 -#define TAG 0 -#define NEW_TAG 1 +#define TAG XA_MARK_0 +#define NEW_TAG XA_MARK_1 -static pthread_mutex_t tree_lock = PTHREAD_MUTEX_INITIALIZER; static pthread_t threads[NUM_THREADS]; static unsigned int seeds[3]; -static RADIX_TREE(tree, GFP_KERNEL); +static DEFINE_XARRAY(array); static bool test_complete; static int max_order; -/* relentlessly fill the tree with tagged entries */ +void my_item_insert(struct xarray *xa, unsigned long index) +{ + XA_STATE(xas, xa, index); + struct item *item = item_create(index, 0); + int order; + +retry: + xas_lock(&xas); + for (order = max_order; order >= 0; order--) { + xas_set_order(&xas, index, order); + item->order = order; + if (xas_find_conflict(&xas)) + continue; + xas_store(&xas, item); + xas_set_mark(&xas, TAG); + break; + } + xas_unlock(&xas); + if (xas_nomem(&xas, GFP_KERNEL)) + goto retry; + if (order < 0) + free(item); +} + +/* relentlessly fill the array with tagged entries */ static void *add_entries_fn(void *arg) { rcu_register_thread(); while (!test_complete) { unsigned long pgoff; - int order; for (pgoff = 0; pgoff < MAX_IDX; pgoff++) { - pthread_mutex_lock(&tree_lock); - for (order = max_order; order >= 0; order--) { - if (item_insert_order(&tree, pgoff, order) - == 0) { - item_tag_set(&tree, pgoff, TAG); - break; - } - } - pthread_mutex_unlock(&tree_lock); + my_item_insert(&array, pgoff); } } @@ -56,33 +69,25 @@ static void *add_entries_fn(void *arg) } /* - * Iterate over the tagged entries, doing a radix_tree_iter_retry() as we find - * things that have been removed and randomly resetting our iteration to the - * next chunk with radix_tree_iter_resume(). Both radix_tree_iter_retry() and - * radix_tree_iter_resume() cause radix_tree_next_slot() to be called with a - * NULL 'slot' variable. + * Iterate over tagged entries, retrying when we find ourselves in a deleted + * node and randomly pausing the iteration. */ static void *tagged_iteration_fn(void *arg) { - struct radix_tree_iter iter; - void **slot; + XA_STATE(xas, &array, 0); + void *entry; rcu_register_thread(); while (!test_complete) { + xas_set(&xas, 0); rcu_read_lock(); - radix_tree_for_each_tagged(slot, &tree, &iter, 0, TAG) { - void *entry = radix_tree_deref_slot(slot); - if (unlikely(!entry)) + xas_for_each_marked(&xas, entry, ULONG_MAX, TAG) { + if (xas_retry(&xas, entry)) continue; - if (radix_tree_deref_retry(entry)) { - slot = radix_tree_iter_retry(&iter); - continue; - } - if (rand_r(&seeds[0]) % 50 == 0) { - slot = radix_tree_iter_resume(slot, &iter); + xas_pause(&xas); rcu_read_unlock(); rcu_barrier(); rcu_read_lock(); @@ -97,33 +102,25 @@ static void *tagged_iteration_fn(void *arg) } /* - * Iterate over the entries, doing a radix_tree_iter_retry() as we find things - * that have been removed and randomly resetting our iteration to the next - * chunk with radix_tree_iter_resume(). Both radix_tree_iter_retry() and - * radix_tree_iter_resume() cause radix_tree_next_slot() to be called with a - * NULL 'slot' variable. + * Iterate over the entries, retrying when we find ourselves in a deleted + * node and randomly pausing the iteration. */ static void *untagged_iteration_fn(void *arg) { - struct radix_tree_iter iter; - void **slot; + XA_STATE(xas, &array, 0); + void *entry; rcu_register_thread(); while (!test_complete) { + xas_set(&xas, 0); rcu_read_lock(); - radix_tree_for_each_slot(slot, &tree, &iter, 0) { - void *entry = radix_tree_deref_slot(slot); - if (unlikely(!entry)) + xas_for_each(&xas, entry, ULONG_MAX) { + if (xas_retry(&xas, entry)) continue; - if (radix_tree_deref_retry(entry)) { - slot = radix_tree_iter_retry(&iter); - continue; - } - if (rand_r(&seeds[1]) % 50 == 0) { - slot = radix_tree_iter_resume(slot, &iter); + xas_pause(&xas); rcu_read_unlock(); rcu_barrier(); rcu_read_lock(); @@ -138,7 +135,7 @@ static void *untagged_iteration_fn(void *arg) } /* - * Randomly remove entries to help induce radix_tree_iter_retry() calls in the + * Randomly remove entries to help induce retries in the * two iteration functions. */ static void *remove_entries_fn(void *arg) @@ -147,12 +144,13 @@ static void *remove_entries_fn(void *arg) while (!test_complete) { int pgoff; + struct item *item; pgoff = rand_r(&seeds[2]) % MAX_IDX; - pthread_mutex_lock(&tree_lock); - item_delete(&tree, pgoff); - pthread_mutex_unlock(&tree_lock); + item = xa_erase(&array, pgoff); + if (item) + item_free(item, pgoff); } rcu_unregister_thread(); @@ -165,8 +163,7 @@ static void *tag_entries_fn(void *arg) rcu_register_thread(); while (!test_complete) { - tag_tagged_items(&tree, &tree_lock, 0, MAX_IDX, 10, TAG, - NEW_TAG); + tag_tagged_items(&array, 0, MAX_IDX, 10, TAG, NEW_TAG); } rcu_unregister_thread(); return NULL; @@ -217,5 +214,5 @@ void iteration_test(unsigned order, unsigned test_duration) } } - item_kill_tree(&tree); + item_kill_tree(&array); } diff --git a/tools/testing/radix-tree/linux/bug.h b/tools/testing/radix-tree/linux/bug.h @@ -1 +1,2 @@ +#include <stdio.h> #include "asm/bug.h" diff --git a/tools/testing/radix-tree/linux/kconfig.h b/tools/testing/radix-tree/linux/kconfig.h @@ -0,0 +1 @@ +#include "../../../../include/linux/kconfig.h" diff --git a/tools/testing/radix-tree/linux/kernel.h b/tools/testing/radix-tree/linux/kernel.h @@ -14,7 +14,12 @@ #include "../../../include/linux/kconfig.h" #define printk printf +#define pr_info printk #define pr_debug printk #define pr_cont printk +#define __acquires(x) +#define __releases(x) +#define __must_hold(x) + #endif /* _KERNEL_H */ diff --git a/tools/testing/radix-tree/linux/lockdep.h b/tools/testing/radix-tree/linux/lockdep.h @@ -0,0 +1,11 @@ +#ifndef _LINUX_LOCKDEP_H +#define _LINUX_LOCKDEP_H +struct lock_class_key { + unsigned int a; +}; + +static inline void lockdep_set_class(spinlock_t *lock, + struct lock_class_key *key) +{ +} +#endif /* _LINUX_LOCKDEP_H */ diff --git a/tools/testing/radix-tree/linux/radix-tree.h b/tools/testing/radix-tree/linux/radix-tree.h @@ -2,7 +2,6 @@ #ifndef _TEST_RADIX_TREE_H #define _TEST_RADIX_TREE_H -#include "generated/map-shift.h" #include "../../../../include/linux/radix-tree.h" extern int kmalloc_verbose; diff --git a/tools/testing/radix-tree/linux/rcupdate.h b/tools/testing/radix-tree/linux/rcupdate.h @@ -6,5 +6,7 @@ #define rcu_dereference_raw(p) rcu_dereference(p) #define rcu_dereference_protected(p, cond) rcu_dereference(p) +#define rcu_dereference_check(p, cond) rcu_dereference(p) +#define RCU_INIT_POINTER(p, v) (p) = (v) #endif diff --git a/tools/testing/radix-tree/main.c b/tools/testing/radix-tree/main.c @@ -214,7 +214,7 @@ void copy_tag_check(void) } // printf("\ncopying tags...\n"); - tagged = tag_tagged_items(&tree, NULL, start, end, ITEMS, 0, 1); + tagged = tag_tagged_items(&tree, start, end, ITEMS, XA_MARK_0, XA_MARK_1); // printf("checking copied tags\n"); assert(tagged == count); @@ -223,7 +223,7 @@ void copy_tag_check(void) /* Copy tags in several rounds */ // printf("\ncopying tags...\n"); tmp = rand() % (count / 10 + 2); - tagged = tag_tagged_items(&tree, NULL, start, end, tmp, 0, 2); + tagged = tag_tagged_items(&tree, start, end, tmp, XA_MARK_0, XA_MARK_2); assert(tagged == count); // printf("%lu %lu %lu\n", tagged, tmp, count); @@ -236,63 +236,6 @@ void copy_tag_check(void) item_kill_tree(&tree); } -static void __locate_check(struct radix_tree_root *tree, unsigned long index, - unsigned order) -{ - struct item *item; - unsigned long index2; - - item_insert_order(tree, index, order); - item = item_lookup(tree, index); - index2 = find_item(tree, item); - if (index != index2) { - printv(2, "index %ld order %d inserted; found %ld\n", - index, order, index2); - abort(); - } -} - -static void __order_0_locate_check(void) -{ - RADIX_TREE(tree, GFP_KERNEL); - int i; - - for (i = 0; i < 50; i++) - __locate_check(&tree, rand() % INT_MAX, 0); - - item_kill_tree(&tree); -} - -static void locate_check(void) -{ - RADIX_TREE(tree, GFP_KERNEL); - unsigned order; - unsigned long offset, index; - - __order_0_locate_check(); - - for (order = 0; order < 20; order++) { - for (offset = 0; offset < (1 << (order + 3)); - offset += (1UL << order)) { - for (index = 0; index < (1UL << (order + 5)); - index += (1UL << order)) { - __locate_check(&tree, index + offset, order); - } - if (find_item(&tree, &tree) != -1) - abort(); - - item_kill_tree(&tree); - } - } - - if (find_item(&tree, &tree) != -1) - abort(); - __locate_check(&tree, -1, 0); - if (find_item(&tree, &tree) != -1) - abort(); - item_kill_tree(&tree); -} - static void single_thread_tests(bool long_run) { int i; @@ -303,10 +246,6 @@ static void single_thread_tests(bool long_run) rcu_barrier(); printv(2, "after multiorder_check: %d allocated, preempt %d\n", nr_allocated, preempt_count); - locate_check(); - rcu_barrier(); - printv(2, "after locate_check: %d allocated, preempt %d\n", - nr_allocated, preempt_count); tag_check(); rcu_barrier(); printv(2, "after tag_check: %d allocated, preempt %d\n", @@ -365,6 +304,7 @@ int main(int argc, char **argv) rcu_register_thread(); radix_tree_init(); + xarray_tests(); regression1_test(); regression2_test(); regression3_test(); diff --git a/tools/testing/radix-tree/multiorder.c b/tools/testing/radix-tree/multiorder.c @@ -20,230 +20,39 @@ #include "test.h" -#define for_each_index(i, base, order) \ - for (i = base; i < base + (1 << order); i++) - -static void __multiorder_tag_test(int index, int order) -{ - RADIX_TREE(tree, GFP_KERNEL); - int base, err, i; - - /* our canonical entry */ - base = index & ~((1 << order) - 1); - - printv(2, "Multiorder tag test with index %d, canonical entry %d\n", - index, base); - - err = item_insert_order(&tree, index, order); - assert(!err); - - /* - * Verify we get collisions for covered indices. We try and fail to - * insert an exceptional entry so we don't leak memory via - * item_insert_order(). - */ - for_each_index(i, base, order) { - err = __radix_tree_insert(&tree, i, order, - (void *)(0xA0 | RADIX_TREE_EXCEPTIONAL_ENTRY)); - assert(err == -EEXIST); - } - - for_each_index(i, base, order) { - assert(!radix_tree_tag_get(&tree, i, 0)); - assert(!radix_tree_tag_get(&tree, i, 1)); - } - - assert(radix_tree_tag_set(&tree, index, 0)); - - for_each_index(i, base, order) { - assert(radix_tree_tag_get(&tree, i, 0)); - assert(!radix_tree_tag_get(&tree, i, 1)); - } - - assert(tag_tagged_items(&tree, NULL, 0, ~0UL, 10, 0, 1) == 1); - assert(radix_tree_tag_clear(&tree, index, 0)); - - for_each_index(i, base, order) { - assert(!radix_tree_tag_get(&tree, i, 0)); - assert(radix_tree_tag_get(&tree, i, 1)); - } - - assert(radix_tree_tag_clear(&tree, index, 1)); - - assert(!radix_tree_tagged(&tree, 0)); - assert(!radix_tree_tagged(&tree, 1)); - - item_kill_tree(&tree); -} - -static void __multiorder_tag_test2(unsigned order, unsigned long index2) +static int item_insert_order(struct xarray *xa, unsigned long index, + unsigned order) { - RADIX_TREE(tree, GFP_KERNEL); - unsigned long index = (1 << order); - index2 += index; - - assert(item_insert_order(&tree, 0, order) == 0); - assert(item_insert(&tree, index2) == 0); - - assert(radix_tree_tag_set(&tree, 0, 0)); - assert(radix_tree_tag_set(&tree, index2, 0)); - - assert(tag_tagged_items(&tree, NULL, 0, ~0UL, 10, 0, 1) == 2); - - item_kill_tree(&tree); -} - -static void multiorder_tag_tests(void) -{ - int i, j; - - /* test multi-order entry for indices 0-7 with no sibling pointers */ - __multiorder_tag_test(0, 3); - __multiorder_tag_test(5, 3); - - /* test multi-order entry for indices 8-15 with no sibling pointers */ - __multiorder_tag_test(8, 3); - __multiorder_tag_test(15, 3); - - /* - * Our order 5 entry covers indices 0-31 in a tree with height=2. - * This is broken up as follows: - * 0-7: canonical entry - * 8-15: sibling 1 - * 16-23: sibling 2 - * 24-31: sibling 3 - */ - __multiorder_tag_test(0, 5); - __multiorder_tag_test(29, 5); - - /* same test, but with indices 32-63 */ - __multiorder_tag_test(32, 5); - __multiorder_tag_test(44, 5); - - /* - * Our order 8 entry covers indices 0-255 in a tree with height=3. - * This is broken up as follows: - * 0-63: canonical entry - * 64-127: sibling 1 - * 128-191: sibling 2 - * 192-255: sibling 3 - */ - __multiorder_tag_test(0, 8); - __multiorder_tag_test(190, 8); - - /* same test, but with indices 256-511 */ - __multiorder_tag_test(256, 8); - __multiorder_tag_test(300, 8); - - __multiorder_tag_test(0x12345678UL, 8); - - for (i = 1; i < 10; i++) - for (j = 0; j < (10 << i); j++) - __multiorder_tag_test2(i, j); -} - -static void multiorder_check(unsigned long index, int order) -{ - unsigned long i; - unsigned long min = index & ~((1UL << order) - 1); - unsigned long max = min + (1UL << order); - void **slot; - struct item *item2 = item_create(min, order); - RADIX_TREE(tree, GFP_KERNEL); - - printv(2, "Multiorder index %ld, order %d\n", index, order); - - assert(item_insert_order(&tree, index, order) == 0); - - for (i = min; i < max; i++) { - struct item *item = item_lookup(&tree, i); - assert(item != 0); - assert(item->index == index); - } - for (i = 0; i < min; i++) - item_check_absent(&tree, i); - for (i = max; i < 2*max; i++) - item_check_absent(&tree, i); - for (i = min; i < max; i++) - assert(radix_tree_insert(&tree, i, item2) == -EEXIST); - - slot = radix_tree_lookup_slot(&tree, index); - free(*slot); - radix_tree_replace_slot(&tree, slot, item2); - for (i = min; i < max; i++) { - struct item *item = item_lookup(&tree, i); - assert(item != 0); - assert(item->index == min); - } - - assert(item_delete(&tree, min) != 0); - - for (i = 0; i < 2*max; i++) - item_check_absent(&tree, i); -} - -static void multiorder_shrink(unsigned long index, int order) -{ - unsigned long i; - unsigned long max = 1 << order; - RADIX_TREE(tree, GFP_KERNEL); - struct radix_tree_node *node; - - printv(2, "Multiorder shrink index %ld, order %d\n", index, order); + XA_STATE_ORDER(xas, xa, index, order); + struct item *item = item_create(index, order); - assert(item_insert_order(&tree, 0, order) == 0); - - node = tree.rnode; - - assert(item_insert(&tree, index) == 0); - assert(node != tree.rnode); - - assert(item_delete(&tree, index) != 0); - assert(node == tree.rnode); - - for (i = 0; i < max; i++) { - struct item *item = item_lookup(&tree, i); - assert(item != 0); - assert(item->index == 0); - } - for (i = max; i < 2*max; i++) - item_check_absent(&tree, i); - - if (!item_delete(&tree, 0)) { - printv(2, "failed to delete index %ld (order %d)\n", index, order); - abort(); - } - - for (i = 0; i < 2*max; i++) - item_check_absent(&tree, i); -} - -static void multiorder_insert_bug(void) -{ - RADIX_TREE(tree, GFP_KERNEL); + do { + xas_lock(&xas); + xas_store(&xas, item); + xas_unlock(&xas); + } while (xas_nomem(&xas, GFP_KERNEL)); - item_insert(&tree, 0); - radix_tree_tag_set(&tree, 0, 0); - item_insert_order(&tree, 3 << 6, 6); + if (!xas_error(&xas)) + return 0; - item_kill_tree(&tree); + free(item); + return xas_error(&xas); } -void multiorder_iteration(void) +void multiorder_iteration(struct xarray *xa) { - RADIX_TREE(tree, GFP_KERNEL); - struct radix_tree_iter iter; - void **slot; + XA_STATE(xas, xa, 0); + struct item *item; int i, j, err; - printv(1, "Multiorder iteration test\n"); - #define NUM_ENTRIES 11 int index[NUM_ENTRIES] = {0, 2, 4, 8, 16, 32, 34, 36, 64, 72, 128}; int order[NUM_ENTRIES] = {1, 1, 2, 3, 4, 1, 0, 1, 3, 0, 7}; + printv(1, "Multiorder iteration test\n"); + for (i = 0; i < NUM_ENTRIES; i++) { - err = item_insert_order(&tree, index[i], order[i]); + err = item_insert_order(xa, index[i], order[i]); assert(!err); } @@ -252,14 +61,14 @@ void multiorder_iteration(void) if (j <= (index[i] | ((1 << order[i]) - 1))) break; - radix_tree_for_each_slot(slot, &tree, &iter, j) { - int height = order[i] / RADIX_TREE_MAP_SHIFT; - int shift = height * RADIX_TREE_MAP_SHIFT; + xas_set(&xas, j); + xas_for_each(&xas, item, ULONG_MAX) { + int height = order[i] / XA_CHUNK_SHIFT; + int shift = height * XA_CHUNK_SHIFT; unsigned long mask = (1UL << order[i]) - 1; - struct item *item = *slot; - assert((iter.index | mask) == (index[i] | mask)); - assert(iter.shift == shift); + assert((xas.xa_index | mask) == (index[i] | mask)); + assert(xas.xa_node->shift == shift); assert(!radix_tree_is_internal_node(item)); assert((item->index | mask) == (index[i] | mask)); assert(item->order == order[i]); @@ -267,18 +76,15 @@ void multiorder_iteration(void) } } - item_kill_tree(&tree); + item_kill_tree(xa); } -void multiorder_tagged_iteration(void) +void multiorder_tagged_iteration(struct xarray *xa) { - RADIX_TREE(tree, GFP_KERNEL); - struct radix_tree_iter iter; - void **slot; + XA_STATE(xas, xa, 0); + struct item *item; int i, j; - printv(1, "Multiorder tagged iteration test\n"); - #define MT_NUM_ENTRIES 9 int index[MT_NUM_ENTRIES] = {0, 2, 4, 16, 32, 40, 64, 72, 128}; int order[MT_NUM_ENTRIES] = {1, 0, 2, 4, 3, 1, 3, 0, 7}; @@ -286,13 +92,15 @@ void multiorder_tagged_iteration(void) #define TAG_ENTRIES 7 int tag_index[TAG_ENTRIES] = {0, 4, 16, 40, 64, 72, 128}; + printv(1, "Multiorder tagged iteration test\n"); + for (i = 0; i < MT_NUM_ENTRIES; i++) - assert(!item_insert_order(&tree, index[i], order[i])); + assert(!item_insert_order(xa, index[i], order[i])); - assert(!radix_tree_tagged(&tree, 1)); + assert(!xa_marked(xa, XA_MARK_1)); for (i = 0; i < TAG_ENTRIES; i++) - assert(radix_tree_tag_set(&tree, tag_index[i], 1)); + xa_set_mark(xa, tag_index[i], XA_MARK_1); for (j = 0; j < 256; j++) { int k; @@ -304,23 +112,23 @@ void multiorder_tagged_iteration(void) break; } - radix_tree_for_each_tagged(slot, &tree, &iter, j, 1) { + xas_set(&xas, j); + xas_for_each_marked(&xas, item, ULONG_MAX, XA_MARK_1) { unsigned long mask; - struct item *item = *slot; for (k = i; index[k] < tag_index[i]; k++) ; mask = (1UL << order[k]) - 1; - assert((iter.index | mask) == (tag_index[i] | mask)); - assert(!radix_tree_is_internal_node(item)); + assert((xas.xa_index | mask) == (tag_index[i] | mask)); + assert(!xa_is_internal(item)); assert((item->index | mask) == (tag_index[i] | mask)); assert(item->order == order[k]); i++; } } - assert(tag_tagged_items(&tree, NULL, 0, ~0UL, TAG_ENTRIES, 1, 2) == - TAG_ENTRIES); + assert(tag_tagged_items(xa, 0, ULONG_MAX, TAG_ENTRIES, XA_MARK_1, + XA_MARK_2) == TAG_ENTRIES); for (j = 0; j < 256; j++) { int mask, k; @@ -332,297 +140,31 @@ void multiorder_tagged_iteration(void) break; } - radix_tree_for_each_tagged(slot, &tree, &iter, j, 2) { - struct item *item = *slot; + xas_set(&xas, j); + xas_for_each_marked(&xas, item, ULONG_MAX, XA_MARK_2) { for (k = i; index[k] < tag_index[i]; k++) ; mask = (1 << order[k]) - 1; - assert((iter.index | mask) == (tag_index[i] | mask)); - assert(!radix_tree_is_internal_node(item)); + assert((xas.xa_index | mask) == (tag_index[i] | mask)); + assert(!xa_is_internal(item)); assert((item->index | mask) == (tag_index[i] | mask)); assert(item->order == order[k]); i++; } } - assert(tag_tagged_items(&tree, NULL, 1, ~0UL, MT_NUM_ENTRIES * 2, 1, 0) - == TAG_ENTRIES); + assert(tag_tagged_items(xa, 1, ULONG_MAX, MT_NUM_ENTRIES * 2, XA_MARK_1, + XA_MARK_0) == TAG_ENTRIES); i = 0; - radix_tree_for_each_tagged(slot, &tree, &iter, 0, 0) { - assert(iter.index == tag_index[i]); + xas_set(&xas, 0); + xas_for_each_marked(&xas, item, ULONG_MAX, XA_MARK_0) { + assert(xas.xa_index == tag_index[i]); i++; } + assert(i == TAG_ENTRIES); - item_kill_tree(&tree); -} - -/* - * Basic join checks: make sure we can't find an entry in the tree after - * a larger entry has replaced it - */ -static void multiorder_join1(unsigned long index, - unsigned order1, unsigned order2) -{ - unsigned long loc; - void *item, *item2 = item_create(index + 1, order1); - RADIX_TREE(tree, GFP_KERNEL); - - item_insert_order(&tree, index, order2); - item = radix_tree_lookup(&tree, index); - radix_tree_join(&tree, index + 1, order1, item2); - loc = find_item(&tree, item); - if (loc == -1) - free(item); - item = radix_tree_lookup(&tree, index + 1); - assert(item == item2); - item_kill_tree(&tree); -} - -/* - * Check that the accounting of exceptional entries is handled correctly - * by joining an exceptional entry to a normal pointer. - */ -static void multiorder_join2(unsigned order1, unsigned order2) -{ - RADIX_TREE(tree, GFP_KERNEL); - struct radix_tree_node *node; - void *item1 = item_create(0, order1); - void *item2; - - item_insert_order(&tree, 0, order2); - radix_tree_insert(&tree, 1 << order2, (void *)0x12UL); - item2 = __radix_tree_lookup(&tree, 1 << order2, &node, NULL); - assert(item2 == (void *)0x12UL); - assert(node->exceptional == 1); - - item2 = radix_tree_lookup(&tree, 0); - free(item2); - - radix_tree_join(&tree, 0, order1, item1); - item2 = __radix_tree_lookup(&tree, 1 << order2, &node, NULL); - assert(item2 == item1); - assert(node->exceptional == 0); - item_kill_tree(&tree); -} - -/* - * This test revealed an accounting bug for exceptional entries at one point. - * Nodes were being freed back into the pool with an elevated exception count - * by radix_tree_join() and then radix_tree_split() was failing to zero the - * count of exceptional entries. - */ -static void multiorder_join3(unsigned int order) -{ - RADIX_TREE(tree, GFP_KERNEL); - struct radix_tree_node *node; - void **slot; - struct radix_tree_iter iter; - unsigned long i; - - for (i = 0; i < (1 << order); i++) { - radix_tree_insert(&tree, i, (void *)0x12UL); - } - - radix_tree_join(&tree, 0, order, (void *)0x16UL); - rcu_barrier(); - - radix_tree_split(&tree, 0, 0); - - radix_tree_for_each_slot(slot, &tree, &iter, 0) { - radix_tree_iter_replace(&tree, &iter, slot, (void *)0x12UL); - } - - __radix_tree_lookup(&tree, 0, &node, NULL); - assert(node->exceptional == node->count); - - item_kill_tree(&tree); -} - -static void multiorder_join(void) -{ - int i, j, idx; - - for (idx = 0; idx < 1024; idx = idx * 2 + 3) { - for (i = 1; i < 15; i++) { - for (j = 0; j < i; j++) { - multiorder_join1(idx, i, j); - } - } - } - - for (i = 1; i < 15; i++) { - for (j = 0; j < i; j++) { - multiorder_join2(i, j); - } - } - - for (i = 3; i < 10; i++) { - multiorder_join3(i); - } -} - -static void check_mem(unsigned old_order, unsigned new_order, unsigned alloc) -{ - struct radix_tree_preload *rtp = &radix_tree_preloads; - if (rtp->nr != 0) - printv(2, "split(%u %u) remaining %u\n", old_order, new_order, - rtp->nr); - /* - * Can't check for equality here as some nodes may have been - * RCU-freed while we ran. But we should never finish with more - * nodes allocated since they should have all been preloaded. - */ - if (nr_allocated > alloc) - printv(2, "split(%u %u) allocated %u %u\n", old_order, new_order, - alloc, nr_allocated); -} - -static void __multiorder_split(int old_order, int new_order) -{ - RADIX_TREE(tree, GFP_ATOMIC); - void **slot; - struct radix_tree_iter iter; - unsigned alloc; - struct item *item; - - radix_tree_preload(GFP_KERNEL); - assert(item_insert_order(&tree, 0, old_order) == 0); - radix_tree_preload_end(); - - /* Wipe out the preloaded cache or it'll confuse check_mem() */ - radix_tree_cpu_dead(0); - - item = radix_tree_tag_set(&tree, 0, 2); - - radix_tree_split_preload(old_order, new_order, GFP_KERNEL); - alloc = nr_allocated; - radix_tree_split(&tree, 0, new_order); - check_mem(old_order, new_order, alloc); - radix_tree_for_each_slot(slot, &tree, &iter, 0) { - radix_tree_iter_replace(&tree, &iter, slot, - item_create(iter.index, new_order)); - } - radix_tree_preload_end(); - - item_kill_tree(&tree); - free(item); -} - -static void __multiorder_split2(int old_order, int new_order) -{ - RADIX_TREE(tree, GFP_KERNEL); - void **slot; - struct radix_tree_iter iter; - struct radix_tree_node *node; - void *item; - - __radix_tree_insert(&tree, 0, old_order, (void *)0x12); - - item = __radix_tree_lookup(&tree, 0, &node, NULL); - assert(item == (void *)0x12); - assert(node->exceptional > 0); - - radix_tree_split(&tree, 0, new_order); - radix_tree_for_each_slot(slot, &tree, &iter, 0) { - radix_tree_iter_replace(&tree, &iter, slot, - item_create(iter.index, new_order)); - } - - item = __radix_tree_lookup(&tree, 0, &node, NULL); - assert(item != (void *)0x12); - assert(node->exceptional == 0); - - item_kill_tree(&tree); -} - -static void __multiorder_split3(int old_order, int new_order) -{ - RADIX_TREE(tree, GFP_KERNEL); - void **slot; - struct radix_tree_iter iter; - struct radix_tree_node *node; - void *item; - - __radix_tree_insert(&tree, 0, old_order, (void *)0x12); - - item = __radix_tree_lookup(&tree, 0, &node, NULL); - assert(item == (void *)0x12); - assert(node->exceptional > 0); - - radix_tree_split(&tree, 0, new_order); - radix_tree_for_each_slot(slot, &tree, &iter, 0) { - radix_tree_iter_replace(&tree, &iter, slot, (void *)0x16); - } - - item = __radix_tree_lookup(&tree, 0, &node, NULL); - assert(item == (void *)0x16); - assert(node->exceptional > 0); - - item_kill_tree(&tree); - - __radix_tree_insert(&tree, 0, old_order, (void *)0x12); - - item = __radix_tree_lookup(&tree, 0, &node, NULL); - assert(item == (void *)0x12); - assert(node->exceptional > 0); - - radix_tree_split(&tree, 0, new_order); - radix_tree_for_each_slot(slot, &tree, &iter, 0) { - if (iter.index == (1 << new_order)) - radix_tree_iter_replace(&tree, &iter, slot, - (void *)0x16); - else - radix_tree_iter_replace(&tree, &iter, slot, NULL); - } - - item = __radix_tree_lookup(&tree, 1 << new_order, &node, NULL); - assert(item == (void *)0x16); - assert(node->count == node->exceptional); - do { - node = node->parent; - if (!node) - break; - assert(node->count == 1); - assert(node->exceptional == 0); - } while (1); - - item_kill_tree(&tree); -} - -static void multiorder_split(void) -{ - int i, j; - - for (i = 3; i < 11; i++) - for (j = 0; j < i; j++) { - __multiorder_split(i, j); - __multiorder_split2(i, j); - __multiorder_split3(i, j); - } -} - -static void multiorder_account(void) -{ - RADIX_TREE(tree, GFP_KERNEL); - struct radix_tree_node *node; - void **slot; - - item_insert_order(&tree, 0, 5); - - __radix_tree_insert(&tree, 1 << 5, 5, (void *)0x12); - __radix_tree_lookup(&tree, 0, &node, NULL); - assert(node->count == node->exceptional * 2); - radix_tree_delete(&tree, 1 << 5); - assert(node->exceptional == 0); - - __radix_tree_insert(&tree, 1 << 5, 5, (void *)0x12); - __radix_tree_lookup(&tree, 1 << 5, &node, &slot); - assert(node->count == node->exceptional * 2); - __radix_tree_replace(&tree, node, slot, NULL, NULL); - assert(node->exceptional == 0); - - item_kill_tree(&tree); + item_kill_tree(xa); } bool stop_iteration = false; @@ -645,68 +187,45 @@ static void *creator_func(void *ptr) static void *iterator_func(void *ptr) { - struct radix_tree_root *tree = ptr; - struct radix_tree_iter iter; + XA_STATE(xas, ptr, 0); struct item *item; - void **slot; while (!stop_iteration) { rcu_read_lock(); - radix_tree_for_each_slot(slot, tree, &iter, 0) { - item = radix_tree_deref_slot(slot); - - if (!item) + xas_for_each(&xas, item, ULONG_MAX) { + if (xas_retry(&xas, item)) continue; - if (radix_tree_deref_retry(item)) { - slot = radix_tree_iter_retry(&iter); - continue; - } - item_sanity(item, iter.index); + item_sanity(item, xas.xa_index); } rcu_read_unlock(); } return NULL; } -static void multiorder_iteration_race(void) +static void multiorder_iteration_race(struct xarray *xa) { const int num_threads = sysconf(_SC_NPROCESSORS_ONLN); pthread_t worker_thread[num_threads]; - RADIX_TREE(tree, GFP_KERNEL); int i; - pthread_create(&worker_thread[0], NULL, &creator_func, &tree); + pthread_create(&worker_thread[0], NULL, &creator_func, xa); for (i = 1; i < num_threads; i++) - pthread_create(&worker_thread[i], NULL, &iterator_func, &tree); + pthread_create(&worker_thread[i], NULL, &iterator_func, xa); for (i = 0; i < num_threads; i++) pthread_join(worker_thread[i], NULL); - item_kill_tree(&tree); + item_kill_tree(xa); } +static DEFINE_XARRAY(array); + void multiorder_checks(void) { - int i; - - for (i = 0; i < 20; i++) { - multiorder_check(200, i); - multiorder_check(0, i); - multiorder_check((1UL << i) + 1, i); - } - - for (i = 0; i < 15; i++) - multiorder_shrink((1UL << (i + RADIX_TREE_MAP_SHIFT)), i); - - multiorder_insert_bug(); - multiorder_tag_tests(); - multiorder_iteration(); - multiorder_tagged_iteration(); - multiorder_join(); - multiorder_split(); - multiorder_account(); - multiorder_iteration_race(); + multiorder_iteration(&array); + multiorder_tagged_iteration(&array); + multiorder_iteration_race(&array); radix_tree_cpu_dead(0); } diff --git a/tools/testing/radix-tree/regression1.c b/tools/testing/radix-tree/regression1.c @@ -44,7 +44,6 @@ #include "regression.h" static RADIX_TREE(mt_tree, GFP_KERNEL); -static pthread_mutex_t mt_lock = PTHREAD_MUTEX_INITIALIZER; struct page { pthread_mutex_t lock; @@ -53,12 +52,12 @@ struct page { unsigned long index; }; -static struct page *page_alloc(void) +static struct page *page_alloc(int index) { struct page *p; p = malloc(sizeof(struct page)); p->count = 1; - p->index = 1; + p->index = index; pthread_mutex_init(&p->lock, NULL); return p; @@ -80,53 +79,33 @@ static void page_free(struct page *p) static unsigned find_get_pages(unsigned long start, unsigned int nr_pages, struct page **pages) { - unsigned int i; - unsigned int ret; - unsigned int nr_found; + XA_STATE(xas, &mt_tree, start); + struct page *page; + unsigned int ret = 0; rcu_read_lock(); -restart: - nr_found = radix_tree_gang_lookup_slot(&mt_tree, - (void ***)pages, NULL, start, nr_pages); - ret = 0; - for (i = 0; i < nr_found; i++) { - struct page *page; -repeat: - page = radix_tree_deref_slot((void **)pages[i]); - if (unlikely(!page)) + xas_for_each(&xas, page, ULONG_MAX) { + if (xas_retry(&xas, page)) continue; - if (radix_tree_exception(page)) { - if (radix_tree_deref_retry(page)) { - /* - * Transient condition which can only trigger - * when entry at index 0 moves out of or back - * to root: none yet gotten, safe to restart. - */ - assert((start | i) == 0); - goto restart; - } - /* - * No exceptional entries are inserted in this test. - */ - assert(0); - } - pthread_mutex_lock(&page->lock); - if (!page->count) { - pthread_mutex_unlock(&page->lock); - goto repeat; - } + if (!page->count) + goto unlock; + /* don't actually update page refcount */ pthread_mutex_unlock(&page->lock); /* Has the page moved? */ - if (unlikely(page != *((void **)pages[i]))) { - goto repeat; - } + if (unlikely(page != xas_reload(&xas))) + goto put_page; pages[ret] = page; ret++; + continue; +unlock: + pthread_mutex_unlock(&page->lock); +put_page: + xas_reset(&xas); } rcu_read_unlock(); return ret; @@ -145,30 +124,30 @@ static void *regression1_fn(void *arg) for (j = 0; j < 1000000; j++) { struct page *p; - p = page_alloc(); - pthread_mutex_lock(&mt_lock); + p = page_alloc(0); + xa_lock(&mt_tree); radix_tree_insert(&mt_tree, 0, p); - pthread_mutex_unlock(&mt_lock); + xa_unlock(&mt_tree); - p = page_alloc(); - pthread_mutex_lock(&mt_lock); + p = page_alloc(1); + xa_lock(&mt_tree); radix_tree_insert(&mt_tree, 1, p); - pthread_mutex_unlock(&mt_lock); + xa_unlock(&mt_tree); - pthread_mutex_lock(&mt_lock); + xa_lock(&mt_tree); p = radix_tree_delete(&mt_tree, 1); pthread_mutex_lock(&p->lock); p->count--; pthread_mutex_unlock(&p->lock); - pthread_mutex_unlock(&mt_lock); + xa_unlock(&mt_tree); page_free(p); - pthread_mutex_lock(&mt_lock); + xa_lock(&mt_tree); p = radix_tree_delete(&mt_tree, 0); pthread_mutex_lock(&p->lock); p->count--; pthread_mutex_unlock(&p->lock); - pthread_mutex_unlock(&mt_lock); + xa_unlock(&mt_tree); page_free(p); } } else { diff --git a/tools/testing/radix-tree/regression2.c b/tools/testing/radix-tree/regression2.c @@ -53,9 +53,9 @@ #include "regression.h" #include "test.h" -#define PAGECACHE_TAG_DIRTY 0 -#define PAGECACHE_TAG_WRITEBACK 1 -#define PAGECACHE_TAG_TOWRITE 2 +#define PAGECACHE_TAG_DIRTY XA_MARK_0 +#define PAGECACHE_TAG_WRITEBACK XA_MARK_1 +#define PAGECACHE_TAG_TOWRITE XA_MARK_2 static RADIX_TREE(mt_tree, GFP_KERNEL); unsigned long page_count = 0; @@ -92,7 +92,7 @@ void regression2_test(void) /* 1. */ start = 0; end = max_slots - 2; - tag_tagged_items(&mt_tree, NULL, start, end, 1, + tag_tagged_items(&mt_tree, start, end, 1, PAGECACHE_TAG_DIRTY, PAGECACHE_TAG_TOWRITE); /* 2. */ diff --git a/tools/testing/radix-tree/regression3.c b/tools/testing/radix-tree/regression3.c @@ -69,21 +69,6 @@ void regression3_test(void) continue; } } - radix_tree_delete(&root, 1); - - first = true; - radix_tree_for_each_contig(slot, &root, &iter, 0) { - printv(2, "contig %ld %p\n", iter.index, *slot); - if (first) { - radix_tree_insert(&root, 1, ptr); - first = false; - } - if (radix_tree_deref_retry(*slot)) { - printv(2, "retry at %ld\n", iter.index); - slot = radix_tree_iter_retry(&iter); - continue; - } - } radix_tree_for_each_slot(slot, &root, &iter, 0) { printv(2, "slot %ld %p\n", iter.index, *slot); @@ -93,14 +78,6 @@ void regression3_test(void) } } - radix_tree_for_each_contig(slot, &root, &iter, 0) { - printv(2, "contig %ld %p\n", iter.index, *slot); - if (!iter.index) { - printv(2, "next at %ld\n", iter.index); - slot = radix_tree_iter_resume(slot, &iter); - } - } - radix_tree_tag_set(&root, 0, 0); radix_tree_tag_set(&root, 1, 0); radix_tree_for_each_tagged(slot, &root, &iter, 0, 0) { diff --git a/tools/testing/radix-tree/tag_check.c b/tools/testing/radix-tree/tag_check.c @@ -24,7 +24,7 @@ __simple_checks(struct radix_tree_root *tree, unsigned long index, int tag) item_tag_set(tree, index, tag); ret = item_tag_get(tree, index, tag); assert(ret != 0); - ret = tag_tagged_items(tree, NULL, first, ~0UL, 10, tag, !tag); + ret = tag_tagged_items(tree, first, ~0UL, 10, tag, !tag); assert(ret == 1); ret = item_tag_get(tree, index, !tag); assert(ret != 0); @@ -321,7 +321,7 @@ static void single_check(void) assert(ret == 0); verify_tag_consistency(&tree, 0); verify_tag_consistency(&tree, 1); - ret = tag_tagged_items(&tree, NULL, first, 10, 10, 0, 1); + ret = tag_tagged_items(&tree, first, 10, 10, XA_MARK_0, XA_MARK_1); assert(ret == 1); ret = radix_tree_gang_lookup_tag(&tree, (void **)items, 0, BATCH, 1); assert(ret == 1); @@ -331,34 +331,6 @@ static void single_check(void) item_kill_tree(&tree); } -void radix_tree_clear_tags_test(void) -{ - unsigned long index; - struct radix_tree_node *node; - struct radix_tree_iter iter; - void **slot; - - RADIX_TREE(tree, GFP_KERNEL); - - item_insert(&tree, 0); - item_tag_set(&tree, 0, 0); - __radix_tree_lookup(&tree, 0, &node, &slot); - radix_tree_clear_tags(&tree, node, slot); - assert(item_tag_get(&tree, 0, 0) == 0); - - for (index = 0; index < 1000; index++) { - item_insert(&tree, index); - item_tag_set(&tree, index, 0); - } - - radix_tree_for_each_slot(slot, &tree, &iter, 0) { - radix_tree_clear_tags(&tree, iter.node, slot); - assert(item_tag_get(&tree, iter.index, 0) == 0); - } - - item_kill_tree(&tree); -} - void tag_check(void) { single_check(); @@ -376,5 +348,4 @@ void tag_check(void) thrash_tags(); rcu_barrier(); printv(2, "after thrash_tags: %d allocated\n", nr_allocated); - radix_tree_clear_tags_test(); } diff --git a/tools/testing/radix-tree/test.c b/tools/testing