[PATCH v6 58/99] dax: Convert lock_slot to XArray

From: Matthew Wilcox Signed-off-by: Matthew Wilcox --- fs/dax.c | 22 -- 1 file changed, 12 insertions(+), 10 deletions(-) diff --git a/fs/dax.c b/fs/dax.c index f3463d93a6ce..8eab0b56f7f9 100644 --- a/fs/dax.c +++ b/fs/dax.c @@ -188,12 +188,11 @@ static void

[PATCH v6 65/99] dax: Fix sparse warning

From: Matthew Wilcox sparse doesn't know that follow_pte_pmd() conditionally acquires the ptl, because it's in a separate compilation unit. Move follow_pte_pmd() to mm.h where sparse can see it. Signed-off-by: Matthew Wilcox --- include/linux/mm.h | 15 ++- m

[PATCH v6 34/99] mm: Convert delete_from_swap_cache to XArray

From: Matthew Wilcox Both callers of __delete_from_swap_cache have the swp_entry_t already, so pass that in to make constructing the XA_STATE easier. Signed-off-by: Matthew Wilcox --- include/linux/swap.h | 5 +++-- mm/swap_state.c | 24 ++-- mm/vmscan.c

[PATCH v6 40/99] pagevec: Use xa_tag_t

From: Matthew Wilcox Removes sparse warnings. Signed-off-by: Matthew Wilcox --- fs/btrfs/extent_io.c| 4 ++-- fs/ext4/inode.c | 2 +- fs/f2fs/data.c | 2 +- fs/gfs2/aops.c | 2 +- include/linux/pagevec.h | 8 +--- mm/swap.c | 4 ++-- 6 files

[PATCH v6 59/99] dax: More XArray conversion

From: Matthew Wilcox This time, we want to convert get_unlocked_mapping_entry() to use the XArray. That has a ripple effect, causing us to change the waitqueues to hash on the address of the xarray rather than the address of the mapping (functionally equivalent), and create a lot of on-the

[PATCH v6 41/99] shmem: Convert replace to XArray

From: Matthew Wilcox shmem_radix_tree_replace() is renamed to shmem_xa_replace() and converted to use the XArray API. Signed-off-by: Matthew Wilcox --- mm/shmem.c | 22 -- 1 file changed, 8 insertions(+), 14 deletions(-) diff --git a/mm/shmem.c b/mm/shmem.c index

[PATCH v6 90/99] btrfs: Convert delayed_nodes_tree to XArray

From: Matthew Wilcox Rename it to just 'delayed_nodes' and remove it from the protection of btrfs_root->inode_lock. Signed-off-by: Matthew Wilcox --- fs/btrfs/ctree.h | 8 +++--- fs/btrfs/delayed-inode.c | 65 fs/bt

[PATCH v6 83/99] hwspinlock: Convert to XArray

From: Matthew Wilcox I had to mess with the locking a bit as I converted the code from a mutex to the xa_lock. Signed-off-by: Matthew Wilcox --- drivers/hwspinlock/hwspinlock_core.c | 151 --- 1 file changed, 52 insertions(+), 99 deletions(-) diff --git a

[PATCH v6 69/99] brd: Convert to XArray

From: Matthew Wilcox Convert brd_pages from a radix tree to an XArray. Simpler and smaller code; in particular another user of radix_tree_preload is eliminated. Signed-off-by: Matthew Wilcox --- drivers/block/brd.c | 93 - 1 file changed

[PATCH v6 85/99] btrfs: Remove unused spinlock

From: Matthew Wilcox The reada_lock in struct btrfs_device was only initialised, and not actually used. That's good because there's another lock also called reada_lock in the btrfs_fs_info that was quite heavily used. Remove this one. Signed-off-by: Matthew Wilcox --- fs/btrfs/vol

[PATCH v6 70/99] xfs: Convert m_perag_tree to XArray

From: Matthew Wilcox Getting rid of the m_perag_lock lets us also get rid of the call to radix_tree_preload(). This is a relatively naive conversion; we could improve performance over the radix tree implementation by passing around xa_state pointers instead of indices, possibly at the expense

[PATCH v6 86/99] btrfs: Convert reada_zones to XArray

From: Matthew Wilcox The use of the reada_lock means we have to use the xa_reserve() API. If we can avoid using reada_lock to protect this xarray, we can drop the use of that function. Signed-off-by: Matthew Wilcox --- fs/btrfs/reada.c | 54

[PATCH v6 44/99] shmem: Convert shmem_tag_pins to XArray

From: Matthew Wilcox Simplify the locking by taking the spinlock while we walk the tree on the assumption that many acquires and releases of the lock will be worse than holding the lock for a (potentially) long time. We could replicate the same locking behaviour with the xarray, but would have

[PATCH v6 84/99] btrfs: Convert fs_roots_radix to XArray

From: Matthew Wilcox Most of the gang lookups being done can be expressed just as efficiently and somewhat more naturally as xa_for_each() loops. I opted not to change the one in btrfs_cleanup_fs_roots() as it's using SRCU which is subtle and quick to anger. Signed-off-by: Matthew W

[PATCH v6 98/99] qrtr: Convert to XArray

From: Matthew Wilcox Moved the kref protection under the xa_lock too. Signed-off-by: Matthew Wilcox --- net/qrtr/qrtr.c | 21 +++-- 1 file changed, 11 insertions(+), 10 deletions(-) diff --git a/net/qrtr/qrtr.c b/net/qrtr/qrtr.c index 77ab05e23001..7de9a06d2aa2 100644 --- a

[PATCH v6 45/99] shmem: Convert shmem_wait_for_pins to XArray

From: Matthew Wilcox As with shmem_tag_pins(), hold the lock around the entire loop instead of acquiring & dropping it for each entry we're going to untag. Signed-off-by: Matthew Wilcox --- mm/shmem.c | 59 --- 1 file ch

[PATCH v6 99/99] null_blk: Convert to XArray

From: Matthew Wilcox We can probably avoid the call to xa_reserve() by changing the locking, but I didn't feel confident enough to do that. Signed-off-by: Matthew Wilcox --- drivers/block/null_blk.c | 87 +--- 1 file changed, 38 insertions(+

[PATCH v6 46/99] shmem: Convert shmem_add_to_page_cache to XArray

From: Matthew Wilcox This removes the last caller of radix_tree_maybe_preload_order(). Simpler code, unless we run out of memory for new xa_nodes partway through inserting entries into the xarray. Hopefully we can support multi-index entries in the page cache soon and all the awful code goes

[PATCH v6 47/99] shmem: Convert shmem_alloc_hugepage to XArray

From: Matthew Wilcox xa_find() is a slightly easier API to use than radix_tree_gang_lookup_slot() because it contains its own RCU locking. Signed-off-by: Matthew Wilcox --- mm/shmem.c | 14 -- 1 file changed, 4 insertions(+), 10 deletions(-) diff --git a/mm/shmem.c b/mm/shmem.c

[PATCH v6 60/99] dax: Convert __dax_invalidate_mapping_entry to XArray

From: Matthew Wilcox Simple now that we already have an xa_state! Signed-off-by: Matthew Wilcox --- fs/dax.c | 10 +- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/fs/dax.c b/fs/dax.c index d3fe61b95216..9a30224da4d6 100644 --- a/fs/dax.c +++ b/fs/dax.c @@ -413,24

[PATCH v6 61/99] dax: Convert dax_writeback_one to XArray

From: Matthew Wilcox Likewise easy Signed-off-by: Matthew Wilcox --- fs/dax.c | 17 +++-- 1 file changed, 7 insertions(+), 10 deletions(-) diff --git a/fs/dax.c b/fs/dax.c index 9a30224da4d6..b66b8c896ed8 100644 --- a/fs/dax.c +++ b/fs/dax.c @@ -632,8 +632,7 @@ static int

[PATCH v6 48/99] shmem: Convert shmem_free_swap to XArray

From: Matthew Wilcox This is a perfect use for xa_cmpxchg(). Note the use of 0 for GFP flags; we won't be allocating memory. Signed-off-by: Matthew Wilcox --- mm/shmem.c | 7 ++- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/mm/shmem.c b/mm/shmem.c index e8233cb

[PATCH v6 67/99] mm: Convert cgroup writeback to XArray

From: Matthew Wilcox This is a fairly naive conversion, leaving in place the GFP_ATOMIC allocation. By switching the locking around, we could use GFP_KERNEL and probably simplify the error handling. Signed-off-by: Matthew Wilcox --- include/linux/backing-dev-defs.h | 2 +- include/linux

[PATCH v6 68/99] vmalloc: Convert to XArray

From: Matthew Wilcox The radix tree of vmap blocks is simpler to express as an XArray. Saves a couple of hundred bytes of text and eliminates a user of the radix tree preload API. Signed-off-by: Matthew Wilcox --- mm/vmalloc.c | 39 +-- 1 file changed, 13

[PATCH v6 29/99] page cache: Convert filemap_range_has_page to XArray

From: Matthew Wilcox Instead of calling find_get_pages_range() and putting any reference, just use xa_find() to look for a page in the right range. Signed-off-by: Matthew Wilcox --- mm/filemap.c | 9 + 1 file changed, 1 insertion(+), 8 deletions(-) diff --git a/mm/filemap.c b/mm

[PATCH v6 96/99] dma-debug: Convert to XArray

From: Matthew Wilcox This is an unusual way to use the xarray tags. If any other users come up, we can add an xas_get_tags() / xas_set_tags() API, but until then I don't want to encourage this kind of abuse. Signed-off-by: Matthew Wilcox --- lib/dma-debug.c

[PATCH v6 97/99] xen: Convert pvcalls-back to XArray

From: Matthew Wilcox This is a straightforward conversion. Signed-off-by: Matthew Wilcox --- drivers/xen/pvcalls-back.c | 51 ++ 1 file changed, 15 insertions(+), 36 deletions(-) diff --git a/drivers/xen/pvcalls-back.c b/drivers/xen/pvcalls-back.c

[PATCH v6 95/99] f2fs: Convert gclist.iroot to XArray

From: Matthew Wilcox Straightforward conversion. Signed-off-by: Matthew Wilcox --- fs/f2fs/gc.c | 14 +++--- fs/f2fs/gc.h | 2 +- 2 files changed, 8 insertions(+), 8 deletions(-) diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c index aac1e02f75df..2b33068dc36b 100644 --- a/fs/f2fs/gc.c

[PATCH v6 19/99] idr: Convert to XArray

From: Matthew Wilcox The IDR distinguishes between unallocated entries (read as NULL) and entries where the user has chosen to store NULL. The radix tree was modified to consider NULL entries which had tag 0 _clear_ as being allocated, but it added a lot of complexity. Instead, the XArray has

[PATCH v6 22/99] page cache: Convert hole search to XArray

From: Matthew Wilcox The page cache offers the ability to search for a miss in the previous or next N locations. Rather than teach the XArray about the page cache's definition of a miss, use xas_prev() and xas_next() to search the page array. This should be more efficient as it does not

[PATCH v6 25/99] page cache: Convert page deletion to XArray

From: Matthew Wilcox The code is slightly shorter and simpler. Signed-off-by: Matthew Wilcox --- mm/filemap.c | 30 ++ 1 file changed, 14 insertions(+), 16 deletions(-) diff --git a/mm/filemap.c b/mm/filemap.c index e6371b551de1..ed30d5310e50 100644 --- a/mm

[PATCH v6 26/99] page cache: Convert page cache lookups to XArray

From: Matthew Wilcox Introduce page_cache_pin() to factor out the common logic between the various lookup routines: find_get_entry find_get_entries find_get_pages_range find_get_pages_contig find_get_pages_range_tag find_get_entries_tag filemap_map_pages By using the xa_state to control the

[PATCH v6 28/99] page cache: Remove stray radix comment

From: Matthew Wilcox Signed-off-by: Matthew Wilcox --- mm/filemap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/filemap.c b/mm/filemap.c index d2a0031d61f5..2536fcacb5bc 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -2606,7 +2606,7 @@ static struct page

[PATCH v6 23/99] page cache: Add page_cache_range_empty function

From: Matthew Wilcox btrfs has its own custom function for determining whether the page cache has any pages in a particular range. Move this functionality to the page cache, and call it from btrfs. Signed-off-by: Matthew Wilcox --- fs/btrfs/btrfs_inode.h | 7 - fs/btrfs/inode.c

[PATCH v6 27/99] page cache: Convert delete_batch to XArray

From: Matthew Wilcox Rename the function from page_cache_tree_delete_batch to just page_cache_delete_batch. Signed-off-by: Matthew Wilcox --- mm/filemap.c | 28 +--- 1 file changed, 13 insertions(+), 15 deletions(-) diff --git a/mm/filemap.c b/mm/filemap.c index

[PATCH v6 02/99] page cache: Use xa_lock

From: Matthew Wilcox Remove the address_space ->tree_lock and use the xa_lock newly added to the radix_tree_root. Rename the address_space ->page_tree to ->pages, since we don't really care that it's a tree. Take the opportunity to rearrange the elements of address_space to

[PATCH v6 03/99] xarray: Replace exceptional entries

From: Matthew Wilcox Introduce xarray value entries to replace the radix tree exceptional entry code. This is a slight change in encoding to allow the use of an extra bit (we can now store BITS_PER_LONG - 1 bits in a value entry). It is also a change in emphasis; exceptional entries are

[PATCH v6 21/99] xarray: Add xa_reserve and xa_release

From: Matthew Wilcox This function simply creates a slot in the XArray for users which need to acquire multiple locks before storing their entry in the tree and so cannot use a plain xa_store(). Signed-off-by: Matthew Wilcox --- include/linux/xarray.h | 14 ++ lib

[PATCH v6 10/99] xarray: Add xa_store

From: Matthew Wilcox xa_store() differs from radix_tree_insert() in that it will overwrite an existing element in the array rather than returning an error. This is the behaviour which most users want, and those that want more complex behaviour generally want to use the xas family of routines

[PATCH v6 11/99] xarray: Add xa_cmpxchg and xa_insert

From: Matthew Wilcox Like cmpxchg(), xa_cmpxchg will only store to the index if the current entry matches the old entry. It returns the current entry, which is usually more useful than the errno returned by radix_tree_insert(). For the users who really only want the errno, the xa_insert

[PATCH v6 15/99] xarray: Add xas_next and xas_prev

From: Matthew Wilcox These two functions move the xas index by one position, and adjust the rest of the iterator state to match it. This is more efficient than calling xas_set() as it keeps the iterator at the leaves of the tree instead of walking the iterator from the root each time. Signed

[PATCH v6 12/99] xarray: Add xa_for_each

From: Matthew Wilcox This iterator allows the user to efficiently walk a range of the array, executing the loop body once for each entry in that range that matches the filter. This commit also includes xa_find() and xa_find_above() which are helper functions for xa_for_each() but may also be

[PATCH v6 18/99] xarray: Add ability to store errno values

From: Matthew Wilcox While the radix tree offers no ability to store IS_ERR pointers, documenting that the XArray does not led to some concern. Here is a sanctioned way to store errnos in the XArray. I'm concerned that it will confuse people who can't tell the difference between

[PATCH v6 14/99] xarray: Add xa_destroy

From: Matthew Wilcox This function frees all the internal memory allocated to the xarray and reinitialises it to be empty. Signed-off-by: Matthew Wilcox --- include/linux/xarray.h | 1 + lib/xarray.c | 26 ++ 2 files changed, 27 insertions(+) diff --git a

[PATCH v6 17/99] xarray: Add MAINTAINERS entry

From: Matthew Wilcox Add myself as XArray and IDR maintainer. Signed-off-by: Matthew Wilcox --- MAINTAINERS | 12 1 file changed, 12 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 18994806e441..55ae4c0b38d5 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -14893,6

[PATCH v6 13/99] xarray: Add xa_extract

From: Matthew Wilcox This function combines the functionality of radix_tree_gang_lookup() and radix_tree_gang_lookup_tagged(). It extracts entries matching the specified filter into a normal array. Signed-off-by: Matthew Wilcox --- include/linux/xarray.h | 2 ++ lib/xarray.c | 80

[PATCH v6 05/99] xarray: Add definition of struct xarray

From: Matthew Wilcox This is a direct replacement for struct radix_tree_root. Some of the struct members have changed name; convert those, and use a #define so that radix_tree users continue to work without change. Signed-off-by: Matthew Wilcox --- include/linux/radix-tree.h

[PATCH v6 16/99] xarray: Add xas_create_range

From: Matthew Wilcox This hopefully temporary function is useful for users who have not yet been converted to multi-index entries. Signed-off-by: Matthew Wilcox --- include/linux/xarray.h | 2 ++ lib/xarray.c | 22 ++ 2 files changed, 24 insertions(+) diff

[PATCH v6 09/99] xarray: Add xa_get_tag, xa_set_tag and xa_clear_tag

From: Matthew Wilcox XArray tags are slightly more strongly typed than the radix tree tags, but occupy the same bits. This commit also adds the xas_ family of tag operations, for cases where the caller is already holding the lock, and xa_tagged() to ask whether any array member has a particular

[PATCH v6 07/99] xarray: Add documentation

From: Matthew Wilcox This is documentation on how to use the XArray, not details about its internal implementation. Signed-off-by: Matthew Wilcox --- Documentation/core-api/index.rst | 1 + Documentation/core-api/xarray.rst | 361 ++ 2 files changed, 362

[PATCH v6 08/99] xarray: Add xa_load

From: Matthew Wilcox This first function in the XArray API brings with it a lot of support infrastructure. The advanced API is based around the xa_state which is a more capable version of the radix_tree_iter. As the test-suite demonstrates, it is possible to use the xarray and radix tree APIs

[PATCH v6 06/99] xarray: Define struct xa_node

From: Matthew Wilcox This is a direct replacement for struct radix_tree_node. A couple of struct members have changed name, so convert those. Use a #define so that radix tree users continue to work without change. Signed-off-by: Matthew Wilcox --- include/linux/radix-tree.h| 29

[PATCH v6 00/99] XArray version 6

From: Matthew Wilcox This version of the XArray has no known bugs. I have converted the radix tree test suite entirely over to the XArray and fixed all bugs that it has uncovered. There are additional tests in the test suite for the XArray, so I now claim the XArray has better test coverage

[PATCH v6 01/99] xarray: Add the xa_lock to the radix_tree_root

From: Matthew Wilcox This results in no change in structure size on 64-bit x86 as it fits in the padding between the gfp_t and the void *. Initialising the spinlock requires a name for the benefit of lockdep, so RADIX_TREE_INIT() now needs to know the name of the radix tree it's initial

[PATCH v6 04/99] xarray: Change definition of sibling entries

From: Matthew Wilcox Instead of storing a pointer to the slot containing the canonical entry, store the offset of the slot. Produces slightly more efficient code (~300 bytes) and simplifies the implementation. Signed-off-by: Matthew Wilcox --- include/linux/xarray.h | 90

<    1   2