linux_dsm_epyc7002/lib/idr.c

593 lines
17 KiB
C
Raw Normal View History

#include <linux/bitmap.h>
#include <linux/bug.h>
#include <linux/export.h>
#include <linux/idr.h>
#include <linux/slab.h>
#include <linux/spinlock.h>
#include <linux/xarray.h>
/**
* idr_alloc_u32() - Allocate an ID.
* @idr: IDR handle.
* @ptr: Pointer to be associated with the new ID.
* @nextid: Pointer to an ID.
* @max: The maximum ID to allocate (inclusive).
* @gfp: Memory allocation flags.
*
* Allocates an unused ID in the range specified by @nextid and @max.
* Note that @max is inclusive whereas the @end parameter to idr_alloc()
* is exclusive. The new ID is assigned to @nextid before the pointer
* is inserted into the IDR, so if @nextid points into the object pointed
* to by @ptr, a concurrent lookup will not find an uninitialised ID.
*
* The caller should provide their own locking to ensure that two
* concurrent modifications to the IDR are not possible. Read-only
* accesses to the IDR may be done under the RCU read lock or may
* exclude simultaneous writers.
*
* Return: 0 if an ID was allocated, -ENOMEM if memory allocation failed,
* or -ENOSPC if no free IDs could be found. If an error occurred,
* @nextid is unchanged.
*/
int idr_alloc_u32(struct idr *idr, void *ptr, u32 *nextid,
unsigned long max, gfp_t gfp)
idr: implement idr_preload[_end]() and idr_alloc() The current idr interface is very cumbersome. * For all allocations, two function calls - idr_pre_get() and idr_get_new*() - should be made. * idr_pre_get() doesn't guarantee that the following idr_get_new*() will not fail from memory shortage. If idr_get_new*() returns -EAGAIN, the caller is expected to retry pre_get and allocation. * idr_get_new*() can't enforce upper limit. Upper limit can only be enforced by allocating and then freeing if above limit. * idr_layer buffer is unnecessarily per-idr. Each idr ends up keeping around MAX_IDR_FREE idr_layers. The memory consumed per idr is under two pages but it makes it difficult to make idr_layer larger. This patch implements the following new set of allocation functions. * idr_preload[_end]() - Similar to radix preload but doesn't fail. The first idr_alloc() inside preload section can be treated as if it were called with @gfp_mask used for idr_preload(). * idr_alloc() - Allocate an ID w/ lower and upper limits. Takes @gfp_flags and can be used w/o preloading. When used inside preloaded section, the allocation mask of preloading can be assumed. If idr_alloc() can be called from a context which allows sufficiently relaxed @gfp_mask, it can be used by itself. If, for example, idr_alloc() is called inside spinlock protected region, preloading can be used like the following. idr_preload(GFP_KERNEL); spin_lock(lock); id = idr_alloc(idr, ptr, start, end, GFP_NOWAIT); spin_unlock(lock); idr_preload_end(); if (id < 0) error; which is much simpler and less error-prone than idr_pre_get and idr_get_new*() loop. The new interface uses per-pcu idr_layer buffer and thus the number of idr's in the system doesn't affect the amount of memory used for preloading. idr_layer_alloc() is introduced to handle idr_layer allocations for both old and new ID allocation paths. This is a bit hairy now but the new interface is expected to replace the old and the internal implementation eventually will become simpler. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-28 08:03:55 +07:00
{
struct radix_tree_iter iter;
void __rcu **slot;
unsigned int base = idr->idr_base;
unsigned int id = *nextid;
idr: implement idr_preload[_end]() and idr_alloc() The current idr interface is very cumbersome. * For all allocations, two function calls - idr_pre_get() and idr_get_new*() - should be made. * idr_pre_get() doesn't guarantee that the following idr_get_new*() will not fail from memory shortage. If idr_get_new*() returns -EAGAIN, the caller is expected to retry pre_get and allocation. * idr_get_new*() can't enforce upper limit. Upper limit can only be enforced by allocating and then freeing if above limit. * idr_layer buffer is unnecessarily per-idr. Each idr ends up keeping around MAX_IDR_FREE idr_layers. The memory consumed per idr is under two pages but it makes it difficult to make idr_layer larger. This patch implements the following new set of allocation functions. * idr_preload[_end]() - Similar to radix preload but doesn't fail. The first idr_alloc() inside preload section can be treated as if it were called with @gfp_mask used for idr_preload(). * idr_alloc() - Allocate an ID w/ lower and upper limits. Takes @gfp_flags and can be used w/o preloading. When used inside preloaded section, the allocation mask of preloading can be assumed. If idr_alloc() can be called from a context which allows sufficiently relaxed @gfp_mask, it can be used by itself. If, for example, idr_alloc() is called inside spinlock protected region, preloading can be used like the following. idr_preload(GFP_KERNEL); spin_lock(lock); id = idr_alloc(idr, ptr, start, end, GFP_NOWAIT); spin_unlock(lock); idr_preload_end(); if (id < 0) error; which is much simpler and less error-prone than idr_pre_get and idr_get_new*() loop. The new interface uses per-pcu idr_layer buffer and thus the number of idr's in the system doesn't affect the amount of memory used for preloading. idr_layer_alloc() is introduced to handle idr_layer allocations for both old and new ID allocation paths. This is a bit hairy now but the new interface is expected to replace the old and the internal implementation eventually will become simpler. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-28 08:03:55 +07:00
if (WARN_ON_ONCE(!(idr->idr_rt.xa_flags & ROOT_IS_IDR)))
idr->idr_rt.xa_flags |= IDR_RT_MARKER;
idr: implement idr_preload[_end]() and idr_alloc() The current idr interface is very cumbersome. * For all allocations, two function calls - idr_pre_get() and idr_get_new*() - should be made. * idr_pre_get() doesn't guarantee that the following idr_get_new*() will not fail from memory shortage. If idr_get_new*() returns -EAGAIN, the caller is expected to retry pre_get and allocation. * idr_get_new*() can't enforce upper limit. Upper limit can only be enforced by allocating and then freeing if above limit. * idr_layer buffer is unnecessarily per-idr. Each idr ends up keeping around MAX_IDR_FREE idr_layers. The memory consumed per idr is under two pages but it makes it difficult to make idr_layer larger. This patch implements the following new set of allocation functions. * idr_preload[_end]() - Similar to radix preload but doesn't fail. The first idr_alloc() inside preload section can be treated as if it were called with @gfp_mask used for idr_preload(). * idr_alloc() - Allocate an ID w/ lower and upper limits. Takes @gfp_flags and can be used w/o preloading. When used inside preloaded section, the allocation mask of preloading can be assumed. If idr_alloc() can be called from a context which allows sufficiently relaxed @gfp_mask, it can be used by itself. If, for example, idr_alloc() is called inside spinlock protected region, preloading can be used like the following. idr_preload(GFP_KERNEL); spin_lock(lock); id = idr_alloc(idr, ptr, start, end, GFP_NOWAIT); spin_unlock(lock); idr_preload_end(); if (id < 0) error; which is much simpler and less error-prone than idr_pre_get and idr_get_new*() loop. The new interface uses per-pcu idr_layer buffer and thus the number of idr's in the system doesn't affect the amount of memory used for preloading. idr_layer_alloc() is introduced to handle idr_layer allocations for both old and new ID allocation paths. This is a bit hairy now but the new interface is expected to replace the old and the internal implementation eventually will become simpler. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-28 08:03:55 +07:00
id = (id < base) ? 0 : id - base;
radix_tree_iter_init(&iter, id);
slot = idr_get_free(&idr->idr_rt, &iter, gfp, max - base);
if (IS_ERR(slot))
return PTR_ERR(slot);
idr: implement idr_preload[_end]() and idr_alloc() The current idr interface is very cumbersome. * For all allocations, two function calls - idr_pre_get() and idr_get_new*() - should be made. * idr_pre_get() doesn't guarantee that the following idr_get_new*() will not fail from memory shortage. If idr_get_new*() returns -EAGAIN, the caller is expected to retry pre_get and allocation. * idr_get_new*() can't enforce upper limit. Upper limit can only be enforced by allocating and then freeing if above limit. * idr_layer buffer is unnecessarily per-idr. Each idr ends up keeping around MAX_IDR_FREE idr_layers. The memory consumed per idr is under two pages but it makes it difficult to make idr_layer larger. This patch implements the following new set of allocation functions. * idr_preload[_end]() - Similar to radix preload but doesn't fail. The first idr_alloc() inside preload section can be treated as if it were called with @gfp_mask used for idr_preload(). * idr_alloc() - Allocate an ID w/ lower and upper limits. Takes @gfp_flags and can be used w/o preloading. When used inside preloaded section, the allocation mask of preloading can be assumed. If idr_alloc() can be called from a context which allows sufficiently relaxed @gfp_mask, it can be used by itself. If, for example, idr_alloc() is called inside spinlock protected region, preloading can be used like the following. idr_preload(GFP_KERNEL); spin_lock(lock); id = idr_alloc(idr, ptr, start, end, GFP_NOWAIT); spin_unlock(lock); idr_preload_end(); if (id < 0) error; which is much simpler and less error-prone than idr_pre_get and idr_get_new*() loop. The new interface uses per-pcu idr_layer buffer and thus the number of idr's in the system doesn't affect the amount of memory used for preloading. idr_layer_alloc() is introduced to handle idr_layer allocations for both old and new ID allocation paths. This is a bit hairy now but the new interface is expected to replace the old and the internal implementation eventually will become simpler. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-28 08:03:55 +07:00
*nextid = iter.index + base;
/* there is a memory barrier inside radix_tree_iter_replace() */
radix_tree_iter_replace(&idr->idr_rt, &iter, slot, ptr);
radix_tree_iter_tag_clear(&idr->idr_rt, &iter, IDR_FREE);
return 0;
idr: implement idr_preload[_end]() and idr_alloc() The current idr interface is very cumbersome. * For all allocations, two function calls - idr_pre_get() and idr_get_new*() - should be made. * idr_pre_get() doesn't guarantee that the following idr_get_new*() will not fail from memory shortage. If idr_get_new*() returns -EAGAIN, the caller is expected to retry pre_get and allocation. * idr_get_new*() can't enforce upper limit. Upper limit can only be enforced by allocating and then freeing if above limit. * idr_layer buffer is unnecessarily per-idr. Each idr ends up keeping around MAX_IDR_FREE idr_layers. The memory consumed per idr is under two pages but it makes it difficult to make idr_layer larger. This patch implements the following new set of allocation functions. * idr_preload[_end]() - Similar to radix preload but doesn't fail. The first idr_alloc() inside preload section can be treated as if it were called with @gfp_mask used for idr_preload(). * idr_alloc() - Allocate an ID w/ lower and upper limits. Takes @gfp_flags and can be used w/o preloading. When used inside preloaded section, the allocation mask of preloading can be assumed. If idr_alloc() can be called from a context which allows sufficiently relaxed @gfp_mask, it can be used by itself. If, for example, idr_alloc() is called inside spinlock protected region, preloading can be used like the following. idr_preload(GFP_KERNEL); spin_lock(lock); id = idr_alloc(idr, ptr, start, end, GFP_NOWAIT); spin_unlock(lock); idr_preload_end(); if (id < 0) error; which is much simpler and less error-prone than idr_pre_get and idr_get_new*() loop. The new interface uses per-pcu idr_layer buffer and thus the number of idr's in the system doesn't affect the amount of memory used for preloading. idr_layer_alloc() is introduced to handle idr_layer allocations for both old and new ID allocation paths. This is a bit hairy now but the new interface is expected to replace the old and the internal implementation eventually will become simpler. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-28 08:03:55 +07:00
}
EXPORT_SYMBOL_GPL(idr_alloc_u32);
idr: implement idr_preload[_end]() and idr_alloc() The current idr interface is very cumbersome. * For all allocations, two function calls - idr_pre_get() and idr_get_new*() - should be made. * idr_pre_get() doesn't guarantee that the following idr_get_new*() will not fail from memory shortage. If idr_get_new*() returns -EAGAIN, the caller is expected to retry pre_get and allocation. * idr_get_new*() can't enforce upper limit. Upper limit can only be enforced by allocating and then freeing if above limit. * idr_layer buffer is unnecessarily per-idr. Each idr ends up keeping around MAX_IDR_FREE idr_layers. The memory consumed per idr is under two pages but it makes it difficult to make idr_layer larger. This patch implements the following new set of allocation functions. * idr_preload[_end]() - Similar to radix preload but doesn't fail. The first idr_alloc() inside preload section can be treated as if it were called with @gfp_mask used for idr_preload(). * idr_alloc() - Allocate an ID w/ lower and upper limits. Takes @gfp_flags and can be used w/o preloading. When used inside preloaded section, the allocation mask of preloading can be assumed. If idr_alloc() can be called from a context which allows sufficiently relaxed @gfp_mask, it can be used by itself. If, for example, idr_alloc() is called inside spinlock protected region, preloading can be used like the following. idr_preload(GFP_KERNEL); spin_lock(lock); id = idr_alloc(idr, ptr, start, end, GFP_NOWAIT); spin_unlock(lock); idr_preload_end(); if (id < 0) error; which is much simpler and less error-prone than idr_pre_get and idr_get_new*() loop. The new interface uses per-pcu idr_layer buffer and thus the number of idr's in the system doesn't affect the amount of memory used for preloading. idr_layer_alloc() is introduced to handle idr_layer allocations for both old and new ID allocation paths. This is a bit hairy now but the new interface is expected to replace the old and the internal implementation eventually will become simpler. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-28 08:03:55 +07:00
idr: introduce idr_alloc_cyclic() As Tejun points out, there are several users of the IDR facility that attempt to use it in a cyclic fashion. These users are likely to see -ENOSPC errors after the counter wraps one or more times however. This patchset adds a new idr_alloc_cyclic routine and converts several of these users to it. Many of these users are in obscure parts of the kernel, and I don't have a good way to test some of them. The change is pretty straightforward though, so hopefully it won't be an issue. There is one other cyclic user of idr_alloc that I didn't touch in ipc/util.c. That one is doing some strange stuff that I didn't quite understand, but it looks like it should probably be converted later somehow. This patch: Thus spake Tejun Heo: Ooh, BTW, the cyclic allocation is broken. It's prone to -ENOSPC after the first wraparound. There are several cyclic users in the kernel and I think it probably would be best to implement cyclic support in idr. This patch does that by adding new idr_alloc_cyclic function that such users in the kernel can use. With this, there's no need for a caller to keep track of the last value used as that's now tracked internally. This should prevent the ENOSPC problems that can hit when the "last allocated" counter exceeds INT_MAX. Later patches will convert existing cyclic users to the new interface. Signed-off-by: Jeff Layton <jlayton@redhat.com> Reviewed-by: Tejun Heo <tj@kernel.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: Eric Paris <eparis@parisplace.org> Cc: Jack Morgenstein <jackm@dev.mellanox.co.il> Cc: John McCutchan <john@johnmccutchan.com> Cc: Neil Horman <nhorman@tuxdriver.com> Cc: Or Gerlitz <ogerlitz@mellanox.com> Cc: Robert Love <rlove@rlove.org> Cc: Roland Dreier <roland@purestorage.com> Cc: Sridhar Samudrala <sri@us.ibm.com> Cc: Steve Wise <swise@opengridcomputing.com> Cc: Tom Tucker <tom@opengridcomputing.com> Cc: Vlad Yasevich <vyasevich@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-04-30 06:21:16 +07:00
/**
* idr_alloc() - Allocate an ID.
* @idr: IDR handle.
* @ptr: Pointer to be associated with the new ID.
* @start: The minimum ID (inclusive).
* @end: The maximum ID (exclusive).
* @gfp: Memory allocation flags.
*
* Allocates an unused ID in the range specified by @start and @end. If
* @end is <= 0, it is treated as one larger than %INT_MAX. This allows
* callers to use @start + N as @end as long as N is within integer range.
*
* The caller should provide their own locking to ensure that two
* concurrent modifications to the IDR are not possible. Read-only
* accesses to the IDR may be done under the RCU read lock or may
* exclude simultaneous writers.
*
* Return: The newly allocated ID, -ENOMEM if memory allocation failed,
* or -ENOSPC if no free IDs could be found.
*/
int idr_alloc(struct idr *idr, void *ptr, int start, int end, gfp_t gfp)
{
u32 id = start;
int ret;
if (WARN_ON_ONCE(start < 0))
return -EINVAL;
ret = idr_alloc_u32(idr, ptr, &id, end > 0 ? end - 1 : INT_MAX, gfp);
if (ret)
return ret;
return id;
}
EXPORT_SYMBOL_GPL(idr_alloc);
/**
* idr_alloc_cyclic() - Allocate an ID cyclically.
* @idr: IDR handle.
* @ptr: Pointer to be associated with the new ID.
* @start: The minimum ID (inclusive).
* @end: The maximum ID (exclusive).
* @gfp: Memory allocation flags.
*
* Allocates an unused ID in the range specified by @nextid and @end. If
* @end is <= 0, it is treated as one larger than %INT_MAX. This allows
* callers to use @start + N as @end as long as N is within integer range.
* The search for an unused ID will start at the last ID allocated and will
* wrap around to @start if no free IDs are found before reaching @end.
*
* The caller should provide their own locking to ensure that two
* concurrent modifications to the IDR are not possible. Read-only
* accesses to the IDR may be done under the RCU read lock or may
* exclude simultaneous writers.
*
* Return: The newly allocated ID, -ENOMEM if memory allocation failed,
* or -ENOSPC if no free IDs could be found.
*/
int idr_alloc_cyclic(struct idr *idr, void *ptr, int start, int end, gfp_t gfp)
{
u32 id = idr->idr_next;
int err, max = end > 0 ? end - 1 : INT_MAX;
idr: remove MAX_IDR_MASK and move left MAX_IDR_* into idr.c MAX_IDR_MASK is another weirdness in the idr interface. As idr covers whole positive integer range, it's defined as 0x7fffffff or INT_MAX. Its usage in idr_find(), idr_replace() and idr_remove() is bizarre. They basically mask off the sign bit and operate on the rest, so if the caller, by accident, passes in a negative number, the sign bit will be masked off and the remaining part will be used as if that was the input, which is worse than crashing. The constant is visible in idr.h and there are several users in the kernel. * drivers/i2c/i2c-core.c:i2c_add_numbered_adapter() Basically used to test if adap->nr is a negative number which isn't -1 and returns -EINVAL if so. idr_alloc() already has negative @start checking (w/ WARN_ON_ONCE), so this can go away. * drivers/infiniband/core/cm.c:cm_alloc_id() drivers/infiniband/hw/mlx4/cm.c:id_map_alloc() Used to wrap cyclic @start. Can be replaced with max(next, 0). Note that this type of cyclic allocation using idr is buggy. These are prone to spurious -ENOSPC failure after the first wraparound. * fs/super.c:get_anon_bdev() The ID allocated from ida is masked off before being tested whether it's inside valid range. ida allocated ID can never be a negative number and the masking is unnecessary. Update idr_*() functions to fail with -EINVAL when negative @id is specified and update other MAX_IDR_MASK users as described above. This leaves MAX_IDR_MASK without any user, remove it and relocate other MAX_IDR_* constants to lib/idr.c. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: Jean Delvare <khali@linux-fr.org> Cc: Roland Dreier <roland@kernel.org> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: "Marciniszyn, Mike" <mike.marciniszyn@intel.com> Cc: Jack Morgenstein <jackm@dev.mellanox.co.il> Cc: Or Gerlitz <ogerlitz@mellanox.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Wolfram Sang <wolfram@the-dreams.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-28 08:05:04 +07:00
if ((int)id < start)
id = start;
err = idr_alloc_u32(idr, ptr, &id, max, gfp);
if ((err == -ENOSPC) && (id > start)) {
id = start;
err = idr_alloc_u32(idr, ptr, &id, max, gfp);
}
if (err)
return err;
idr->idr_next = id + 1;
return id;
}
EXPORT_SYMBOL(idr_alloc_cyclic);
/**
* idr_remove() - Remove an ID from the IDR.
* @idr: IDR handle.
* @id: Pointer ID.
*
* Removes this ID from the IDR. If the ID was not previously in the IDR,
* this function returns %NULL.
*
* Since this function modifies the IDR, the caller should provide their
* own locking to ensure that concurrent modification of the same IDR is
* not possible.
*
* Return: The pointer formerly associated with this ID.
*/
void *idr_remove(struct idr *idr, unsigned long id)
{
return radix_tree_delete_item(&idr->idr_rt, id - idr->idr_base, NULL);
}
EXPORT_SYMBOL_GPL(idr_remove);
/**
* idr_find() - Return pointer for given ID.
* @idr: IDR handle.
* @id: Pointer ID.
*
* Looks up the pointer associated with this ID. A %NULL pointer may
* indicate that @id is not allocated or that the %NULL pointer was
* associated with this ID.
*
* This function can be called under rcu_read_lock(), given that the leaf
* pointers lifetimes are correctly managed.
*
* Return: The pointer associated with this ID.
*/
void *idr_find(const struct idr *idr, unsigned long id)
{
return radix_tree_lookup(&idr->idr_rt, id - idr->idr_base);
}
EXPORT_SYMBOL_GPL(idr_find);
/**
* idr_for_each() - Iterate through all stored pointers.
* @idr: IDR handle.
* @fn: Function to be called for each pointer.
* @data: Data passed to callback function.
*
* The callback function will be called for each entry in @idr, passing
* the ID, the entry and @data.
*
* If @fn returns anything other than %0, the iteration stops and that
* value is returned from this function.
*
* idr_for_each() can be called concurrently with idr_alloc() and
* idr_remove() if protected by RCU. Newly added entries may not be
* seen and deleted entries may be seen, but adding and removing entries
* will not cause other entries to be skipped, nor spurious ones to be seen.
*/
int idr_for_each(const struct idr *idr,
int (*fn)(int id, void *p, void *data), void *data)
{
struct radix_tree_iter iter;
void __rcu **slot;
int base = idr->idr_base;
radix_tree_for_each_slot(slot, &idr->idr_rt, &iter, 0) {
int ret;
unsigned long id = iter.index + base;
if (WARN_ON_ONCE(id > INT_MAX))
break;
ret = fn(id, rcu_dereference_raw(*slot), data);
if (ret)
return ret;
}
return 0;
}
EXPORT_SYMBOL(idr_for_each);
cgroup: CSS ID support Patch for Per-CSS(Cgroup Subsys State) ID and private hierarchy code. This patch attaches unique ID to each css and provides following. - css_lookup(subsys, id) returns pointer to struct cgroup_subysys_state of id. - css_get_next(subsys, id, rootid, depth, foundid) returns the next css under "root" by scanning When cgroup_subsys->use_id is set, an id for css is maintained. The cgroup framework only parepares - css_id of root css for subsys - id is automatically attached at creation of css. - id is *not* freed automatically. Because the cgroup framework don't know lifetime of cgroup_subsys_state. free_css_id() function is provided. This must be called by subsys. There are several reasons to develop this. - Saving space .... For example, memcg's swap_cgroup is array of pointers to cgroup. But it is not necessary to be very fast. By replacing pointers(8bytes per ent) to ID (2byes per ent), we can reduce much amount of memory usage. - Scanning without lock. CSS_ID provides "scan id under this ROOT" function. By this, scanning css under root can be written without locks. ex) do { rcu_read_lock(); next = cgroup_get_next(subsys, id, root, &found); /* check sanity of next here */ css_tryget(); rcu_read_unlock(); id = found + 1 } while(...) Characteristics: - Each css has unique ID under subsys. - Lifetime of ID is controlled by subsys. - css ID contains "ID" and "Depth in hierarchy" and stack of hierarchy - Allowed ID is 1-65535, ID 0 is UNUSED ID. Design Choices: - scan-by-ID v.s. scan-by-tree-walk. As /proc's pid scan does, scan-by-ID is robust when scanning is done by following kind of routine. scan -> rest a while(release a lock) -> conitunue from interrupted memcg's hierarchical reclaim does this. - When subsys->use_id is set, # of css in the system is limited to 65535. [bharata@linux.vnet.ibm.com: remove rcu_read_lock() from css_get_next()] Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Acked-by: Paul Menage <menage@google.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Balbir Singh <balbir@in.ibm.com> Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp> Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-04-03 06:57:25 +07:00
/**
* idr_get_next() - Find next populated entry.
* @idr: IDR handle.
* @nextid: Pointer to an ID.
*
* Returns the next populated entry in the tree with an ID greater than
* or equal to the value pointed to by @nextid. On exit, @nextid is updated
* to the ID of the found value. To use in a loop, the value pointed to by
* nextid must be incremented by the user.
cgroup: CSS ID support Patch for Per-CSS(Cgroup Subsys State) ID and private hierarchy code. This patch attaches unique ID to each css and provides following. - css_lookup(subsys, id) returns pointer to struct cgroup_subysys_state of id. - css_get_next(subsys, id, rootid, depth, foundid) returns the next css under "root" by scanning When cgroup_subsys->use_id is set, an id for css is maintained. The cgroup framework only parepares - css_id of root css for subsys - id is automatically attached at creation of css. - id is *not* freed automatically. Because the cgroup framework don't know lifetime of cgroup_subsys_state. free_css_id() function is provided. This must be called by subsys. There are several reasons to develop this. - Saving space .... For example, memcg's swap_cgroup is array of pointers to cgroup. But it is not necessary to be very fast. By replacing pointers(8bytes per ent) to ID (2byes per ent), we can reduce much amount of memory usage. - Scanning without lock. CSS_ID provides "scan id under this ROOT" function. By this, scanning css under root can be written without locks. ex) do { rcu_read_lock(); next = cgroup_get_next(subsys, id, root, &found); /* check sanity of next here */ css_tryget(); rcu_read_unlock(); id = found + 1 } while(...) Characteristics: - Each css has unique ID under subsys. - Lifetime of ID is controlled by subsys. - css ID contains "ID" and "Depth in hierarchy" and stack of hierarchy - Allowed ID is 1-65535, ID 0 is UNUSED ID. Design Choices: - scan-by-ID v.s. scan-by-tree-walk. As /proc's pid scan does, scan-by-ID is robust when scanning is done by following kind of routine. scan -> rest a while(release a lock) -> conitunue from interrupted memcg's hierarchical reclaim does this. - When subsys->use_id is set, # of css in the system is limited to 65535. [bharata@linux.vnet.ibm.com: remove rcu_read_lock() from css_get_next()] Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Acked-by: Paul Menage <menage@google.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Balbir Singh <balbir@in.ibm.com> Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp> Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-04-03 06:57:25 +07:00
*/
void *idr_get_next(struct idr *idr, int *nextid)
cgroup: CSS ID support Patch for Per-CSS(Cgroup Subsys State) ID and private hierarchy code. This patch attaches unique ID to each css and provides following. - css_lookup(subsys, id) returns pointer to struct cgroup_subysys_state of id. - css_get_next(subsys, id, rootid, depth, foundid) returns the next css under "root" by scanning When cgroup_subsys->use_id is set, an id for css is maintained. The cgroup framework only parepares - css_id of root css for subsys - id is automatically attached at creation of css. - id is *not* freed automatically. Because the cgroup framework don't know lifetime of cgroup_subsys_state. free_css_id() function is provided. This must be called by subsys. There are several reasons to develop this. - Saving space .... For example, memcg's swap_cgroup is array of pointers to cgroup. But it is not necessary to be very fast. By replacing pointers(8bytes per ent) to ID (2byes per ent), we can reduce much amount of memory usage. - Scanning without lock. CSS_ID provides "scan id under this ROOT" function. By this, scanning css under root can be written without locks. ex) do { rcu_read_lock(); next = cgroup_get_next(subsys, id, root, &found); /* check sanity of next here */ css_tryget(); rcu_read_unlock(); id = found + 1 } while(...) Characteristics: - Each css has unique ID under subsys. - Lifetime of ID is controlled by subsys. - css ID contains "ID" and "Depth in hierarchy" and stack of hierarchy - Allowed ID is 1-65535, ID 0 is UNUSED ID. Design Choices: - scan-by-ID v.s. scan-by-tree-walk. As /proc's pid scan does, scan-by-ID is robust when scanning is done by following kind of routine. scan -> rest a while(release a lock) -> conitunue from interrupted memcg's hierarchical reclaim does this. - When subsys->use_id is set, # of css in the system is limited to 65535. [bharata@linux.vnet.ibm.com: remove rcu_read_lock() from css_get_next()] Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Acked-by: Paul Menage <menage@google.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Balbir Singh <balbir@in.ibm.com> Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp> Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-04-03 06:57:25 +07:00
{
struct radix_tree_iter iter;
void __rcu **slot;
unsigned long base = idr->idr_base;
unsigned long id = *nextid;
cgroup: CSS ID support Patch for Per-CSS(Cgroup Subsys State) ID and private hierarchy code. This patch attaches unique ID to each css and provides following. - css_lookup(subsys, id) returns pointer to struct cgroup_subysys_state of id. - css_get_next(subsys, id, rootid, depth, foundid) returns the next css under "root" by scanning When cgroup_subsys->use_id is set, an id for css is maintained. The cgroup framework only parepares - css_id of root css for subsys - id is automatically attached at creation of css. - id is *not* freed automatically. Because the cgroup framework don't know lifetime of cgroup_subsys_state. free_css_id() function is provided. This must be called by subsys. There are several reasons to develop this. - Saving space .... For example, memcg's swap_cgroup is array of pointers to cgroup. But it is not necessary to be very fast. By replacing pointers(8bytes per ent) to ID (2byes per ent), we can reduce much amount of memory usage. - Scanning without lock. CSS_ID provides "scan id under this ROOT" function. By this, scanning css under root can be written without locks. ex) do { rcu_read_lock(); next = cgroup_get_next(subsys, id, root, &found); /* check sanity of next here */ css_tryget(); rcu_read_unlock(); id = found + 1 } while(...) Characteristics: - Each css has unique ID under subsys. - Lifetime of ID is controlled by subsys. - css ID contains "ID" and "Depth in hierarchy" and stack of hierarchy - Allowed ID is 1-65535, ID 0 is UNUSED ID. Design Choices: - scan-by-ID v.s. scan-by-tree-walk. As /proc's pid scan does, scan-by-ID is robust when scanning is done by following kind of routine. scan -> rest a while(release a lock) -> conitunue from interrupted memcg's hierarchical reclaim does this. - When subsys->use_id is set, # of css in the system is limited to 65535. [bharata@linux.vnet.ibm.com: remove rcu_read_lock() from css_get_next()] Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Acked-by: Paul Menage <menage@google.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Balbir Singh <balbir@in.ibm.com> Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp> Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-04-03 06:57:25 +07:00
id = (id < base) ? 0 : id - base;
slot = radix_tree_iter_find(&idr->idr_rt, &iter, id);
if (!slot)
cgroup: CSS ID support Patch for Per-CSS(Cgroup Subsys State) ID and private hierarchy code. This patch attaches unique ID to each css and provides following. - css_lookup(subsys, id) returns pointer to struct cgroup_subysys_state of id. - css_get_next(subsys, id, rootid, depth, foundid) returns the next css under "root" by scanning When cgroup_subsys->use_id is set, an id for css is maintained. The cgroup framework only parepares - css_id of root css for subsys - id is automatically attached at creation of css. - id is *not* freed automatically. Because the cgroup framework don't know lifetime of cgroup_subsys_state. free_css_id() function is provided. This must be called by subsys. There are several reasons to develop this. - Saving space .... For example, memcg's swap_cgroup is array of pointers to cgroup. But it is not necessary to be very fast. By replacing pointers(8bytes per ent) to ID (2byes per ent), we can reduce much amount of memory usage. - Scanning without lock. CSS_ID provides "scan id under this ROOT" function. By this, scanning css under root can be written without locks. ex) do { rcu_read_lock(); next = cgroup_get_next(subsys, id, root, &found); /* check sanity of next here */ css_tryget(); rcu_read_unlock(); id = found + 1 } while(...) Characteristics: - Each css has unique ID under subsys. - Lifetime of ID is controlled by subsys. - css ID contains "ID" and "Depth in hierarchy" and stack of hierarchy - Allowed ID is 1-65535, ID 0 is UNUSED ID. Design Choices: - scan-by-ID v.s. scan-by-tree-walk. As /proc's pid scan does, scan-by-ID is robust when scanning is done by following kind of routine. scan -> rest a while(release a lock) -> conitunue from interrupted memcg's hierarchical reclaim does this. - When subsys->use_id is set, # of css in the system is limited to 65535. [bharata@linux.vnet.ibm.com: remove rcu_read_lock() from css_get_next()] Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Acked-by: Paul Menage <menage@google.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Balbir Singh <balbir@in.ibm.com> Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp> Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-04-03 06:57:25 +07:00
return NULL;
id = iter.index + base;
cgroup: CSS ID support Patch for Per-CSS(Cgroup Subsys State) ID and private hierarchy code. This patch attaches unique ID to each css and provides following. - css_lookup(subsys, id) returns pointer to struct cgroup_subysys_state of id. - css_get_next(subsys, id, rootid, depth, foundid) returns the next css under "root" by scanning When cgroup_subsys->use_id is set, an id for css is maintained. The cgroup framework only parepares - css_id of root css for subsys - id is automatically attached at creation of css. - id is *not* freed automatically. Because the cgroup framework don't know lifetime of cgroup_subsys_state. free_css_id() function is provided. This must be called by subsys. There are several reasons to develop this. - Saving space .... For example, memcg's swap_cgroup is array of pointers to cgroup. But it is not necessary to be very fast. By replacing pointers(8bytes per ent) to ID (2byes per ent), we can reduce much amount of memory usage. - Scanning without lock. CSS_ID provides "scan id under this ROOT" function. By this, scanning css under root can be written without locks. ex) do { rcu_read_lock(); next = cgroup_get_next(subsys, id, root, &found); /* check sanity of next here */ css_tryget(); rcu_read_unlock(); id = found + 1 } while(...) Characteristics: - Each css has unique ID under subsys. - Lifetime of ID is controlled by subsys. - css ID contains "ID" and "Depth in hierarchy" and stack of hierarchy - Allowed ID is 1-65535, ID 0 is UNUSED ID. Design Choices: - scan-by-ID v.s. scan-by-tree-walk. As /proc's pid scan does, scan-by-ID is robust when scanning is done by following kind of routine. scan -> rest a while(release a lock) -> conitunue from interrupted memcg's hierarchical reclaim does this. - When subsys->use_id is set, # of css in the system is limited to 65535. [bharata@linux.vnet.ibm.com: remove rcu_read_lock() from css_get_next()] Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Acked-by: Paul Menage <menage@google.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Balbir Singh <balbir@in.ibm.com> Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp> Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-04-03 06:57:25 +07:00
if (WARN_ON_ONCE(id > INT_MAX))
return NULL;
*nextid = id;
return rcu_dereference_raw(*slot);
cgroup: CSS ID support Patch for Per-CSS(Cgroup Subsys State) ID and private hierarchy code. This patch attaches unique ID to each css and provides following. - css_lookup(subsys, id) returns pointer to struct cgroup_subysys_state of id. - css_get_next(subsys, id, rootid, depth, foundid) returns the next css under "root" by scanning When cgroup_subsys->use_id is set, an id for css is maintained. The cgroup framework only parepares - css_id of root css for subsys - id is automatically attached at creation of css. - id is *not* freed automatically. Because the cgroup framework don't know lifetime of cgroup_subsys_state. free_css_id() function is provided. This must be called by subsys. There are several reasons to develop this. - Saving space .... For example, memcg's swap_cgroup is array of pointers to cgroup. But it is not necessary to be very fast. By replacing pointers(8bytes per ent) to ID (2byes per ent), we can reduce much amount of memory usage. - Scanning without lock. CSS_ID provides "scan id under this ROOT" function. By this, scanning css under root can be written without locks. ex) do { rcu_read_lock(); next = cgroup_get_next(subsys, id, root, &found); /* check sanity of next here */ css_tryget(); rcu_read_unlock(); id = found + 1 } while(...) Characteristics: - Each css has unique ID under subsys. - Lifetime of ID is controlled by subsys. - css ID contains "ID" and "Depth in hierarchy" and stack of hierarchy - Allowed ID is 1-65535, ID 0 is UNUSED ID. Design Choices: - scan-by-ID v.s. scan-by-tree-walk. As /proc's pid scan does, scan-by-ID is robust when scanning is done by following kind of routine. scan -> rest a while(release a lock) -> conitunue from interrupted memcg's hierarchical reclaim does this. - When subsys->use_id is set, # of css in the system is limited to 65535. [bharata@linux.vnet.ibm.com: remove rcu_read_lock() from css_get_next()] Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Acked-by: Paul Menage <menage@google.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Balbir Singh <balbir@in.ibm.com> Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp> Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-04-03 06:57:25 +07:00
}
EXPORT_SYMBOL(idr_get_next);
cgroup: CSS ID support Patch for Per-CSS(Cgroup Subsys State) ID and private hierarchy code. This patch attaches unique ID to each css and provides following. - css_lookup(subsys, id) returns pointer to struct cgroup_subysys_state of id. - css_get_next(subsys, id, rootid, depth, foundid) returns the next css under "root" by scanning When cgroup_subsys->use_id is set, an id for css is maintained. The cgroup framework only parepares - css_id of root css for subsys - id is automatically attached at creation of css. - id is *not* freed automatically. Because the cgroup framework don't know lifetime of cgroup_subsys_state. free_css_id() function is provided. This must be called by subsys. There are several reasons to develop this. - Saving space .... For example, memcg's swap_cgroup is array of pointers to cgroup. But it is not necessary to be very fast. By replacing pointers(8bytes per ent) to ID (2byes per ent), we can reduce much amount of memory usage. - Scanning without lock. CSS_ID provides "scan id under this ROOT" function. By this, scanning css under root can be written without locks. ex) do { rcu_read_lock(); next = cgroup_get_next(subsys, id, root, &found); /* check sanity of next here */ css_tryget(); rcu_read_unlock(); id = found + 1 } while(...) Characteristics: - Each css has unique ID under subsys. - Lifetime of ID is controlled by subsys. - css ID contains "ID" and "Depth in hierarchy" and stack of hierarchy - Allowed ID is 1-65535, ID 0 is UNUSED ID. Design Choices: - scan-by-ID v.s. scan-by-tree-walk. As /proc's pid scan does, scan-by-ID is robust when scanning is done by following kind of routine. scan -> rest a while(release a lock) -> conitunue from interrupted memcg's hierarchical reclaim does this. - When subsys->use_id is set, # of css in the system is limited to 65535. [bharata@linux.vnet.ibm.com: remove rcu_read_lock() from css_get_next()] Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Acked-by: Paul Menage <menage@google.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Balbir Singh <balbir@in.ibm.com> Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp> Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-04-03 06:57:25 +07:00
/**
* idr_get_next_ul() - Find next populated entry.
* @idr: IDR handle.
* @nextid: Pointer to an ID.
*
* Returns the next populated entry in the tree with an ID greater than
* or equal to the value pointed to by @nextid. On exit, @nextid is updated
* to the ID of the found value. To use in a loop, the value pointed to by
* nextid must be incremented by the user.
*/
void *idr_get_next_ul(struct idr *idr, unsigned long *nextid)
{
struct radix_tree_iter iter;
void __rcu **slot;
unsigned long base = idr->idr_base;
unsigned long id = *nextid;
id = (id < base) ? 0 : id - base;
slot = radix_tree_iter_find(&idr->idr_rt, &iter, id);
if (!slot)
return NULL;
*nextid = iter.index + base;
return rcu_dereference_raw(*slot);
}
EXPORT_SYMBOL(idr_get_next_ul);
/**
* idr_replace() - replace pointer for given ID.
* @idr: IDR handle.
* @ptr: New pointer to associate with the ID.
* @id: ID to change.
*
* Replace the pointer registered with an ID and return the old value.
* This function can be called under the RCU read lock concurrently with
* idr_alloc() and idr_remove() (as long as the ID being removed is not
* the one being replaced!).
*
* Returns: the old value on success. %-ENOENT indicates that @id was not
* found. %-EINVAL indicates that @ptr was not valid.
*/
void *idr_replace(struct idr *idr, void *ptr, unsigned long id)
{
struct radix_tree_node *node;
void __rcu **slot = NULL;
void *entry;
id -= idr->idr_base;
idr: remove MAX_IDR_MASK and move left MAX_IDR_* into idr.c MAX_IDR_MASK is another weirdness in the idr interface. As idr covers whole positive integer range, it's defined as 0x7fffffff or INT_MAX. Its usage in idr_find(), idr_replace() and idr_remove() is bizarre. They basically mask off the sign bit and operate on the rest, so if the caller, by accident, passes in a negative number, the sign bit will be masked off and the remaining part will be used as if that was the input, which is worse than crashing. The constant is visible in idr.h and there are several users in the kernel. * drivers/i2c/i2c-core.c:i2c_add_numbered_adapter() Basically used to test if adap->nr is a negative number which isn't -1 and returns -EINVAL if so. idr_alloc() already has negative @start checking (w/ WARN_ON_ONCE), so this can go away. * drivers/infiniband/core/cm.c:cm_alloc_id() drivers/infiniband/hw/mlx4/cm.c:id_map_alloc() Used to wrap cyclic @start. Can be replaced with max(next, 0). Note that this type of cyclic allocation using idr is buggy. These are prone to spurious -ENOSPC failure after the first wraparound. * fs/super.c:get_anon_bdev() The ID allocated from ida is masked off before being tested whether it's inside valid range. ida allocated ID can never be a negative number and the masking is unnecessary. Update idr_*() functions to fail with -EINVAL when negative @id is specified and update other MAX_IDR_MASK users as described above. This leaves MAX_IDR_MASK without any user, remove it and relocate other MAX_IDR_* constants to lib/idr.c. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: Jean Delvare <khali@linux-fr.org> Cc: Roland Dreier <roland@kernel.org> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: "Marciniszyn, Mike" <mike.marciniszyn@intel.com> Cc: Jack Morgenstein <jackm@dev.mellanox.co.il> Cc: Or Gerlitz <ogerlitz@mellanox.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Wolfram Sang <wolfram@the-dreams.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-28 08:05:04 +07:00
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);
mm, truncate: do not check mapping for every page being truncated During truncation, the mapping has already been checked for shmem and dax so it's known that workingset_update_node is required. This patch avoids the checks on mapping for each page being truncated. In all other cases, a lookup helper is used to determine if workingset_update_node() needs to be called. The one danger is that the API is slightly harder to use as calling workingset_update_node directly without checking for dax or shmem mappings could lead to surprises. However, the API rarely needs to be used and hopefully the comment is enough to give people the hint. sparsetruncate (tiny) 4.14.0-rc4 4.14.0-rc4 oneirq-v1r1 pickhelper-v1r1 Min Time 141.00 ( 0.00%) 140.00 ( 0.71%) 1st-qrtle Time 142.00 ( 0.00%) 141.00 ( 0.70%) 2nd-qrtle Time 142.00 ( 0.00%) 142.00 ( 0.00%) 3rd-qrtle Time 143.00 ( 0.00%) 143.00 ( 0.00%) Max-90% Time 144.00 ( 0.00%) 144.00 ( 0.00%) Max-95% Time 147.00 ( 0.00%) 145.00 ( 1.36%) Max-99% Time 195.00 ( 0.00%) 191.00 ( 2.05%) Max Time 230.00 ( 0.00%) 205.00 ( 10.87%) Amean Time 144.37 ( 0.00%) 143.82 ( 0.38%) Stddev Time 10.44 ( 0.00%) 9.00 ( 13.74%) Coeff Time 7.23 ( 0.00%) 6.26 ( 13.41%) Best99%Amean Time 143.72 ( 0.00%) 143.34 ( 0.26%) Best95%Amean Time 142.37 ( 0.00%) 142.00 ( 0.26%) Best90%Amean Time 142.19 ( 0.00%) 141.85 ( 0.24%) Best75%Amean Time 141.92 ( 0.00%) 141.58 ( 0.24%) Best50%Amean Time 141.69 ( 0.00%) 141.31 ( 0.27%) Best25%Amean Time 141.38 ( 0.00%) 140.97 ( 0.29%) As you'd expect, the gain is marginal but it can be detected. The differences in bonnie are all within the noise which is not surprising given the impact on the microbenchmark. radix_tree_update_node_t is a callback for some radix operations that optionally passes in a private field. The only user of the callback is workingset_update_node and as it no longer requires a mapping, the private field is removed. Link: http://lkml.kernel.org/r/20171018075952.10627-3-mgorman@techsingularity.net Signed-off-by: Mel Gorman <mgorman@techsingularity.net> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Andi Kleen <ak@linux.intel.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-16 08:37:41 +07:00
__radix_tree_replace(&idr->idr_rt, node, slot, ptr, NULL);
return entry;
}
EXPORT_SYMBOL(idr_replace);
/**
* DOC: IDA description
*
* The IDA is an ID allocator which does not provide the ability to
* associate an ID with a pointer. As such, it only needs to store one
* bit per ID, and so is more space efficient than an IDR. To use an IDA,
* define it using DEFINE_IDA() (or embed a &struct ida in a data structure,
* then initialise it using ida_init()). To allocate a new ID, call
* ida_alloc(), ida_alloc_min(), ida_alloc_max() or ida_alloc_range().
* To free an ID, call ida_free().
*
* ida_destroy() can be used to dispose of an IDA without needing to
* 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.
*/
/*
* Developer's notes:
*
* 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 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 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.
*/
/**
* 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)
{
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;
}
}
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);
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))
xas_clear_mark(&xas, XA_FREE_MARK);
} else {
if (bit < BITS_PER_XA_VALUE) {
bitmap = xa_mk_value(1UL << bit);
} else {
bitmap = alloc;
if (!bitmap)
bitmap = kzalloc(sizeof(*bitmap), GFP_NOWAIT);
if (!bitmap)
goto alloc;
__set_bit(bit, bitmap->bitmap);
}
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);
/**
* 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)
{
XA_STATE(xas, &ida->xa, id / IDA_BITMAP_BITS);
unsigned bit = id % IDA_BITMAP_BITS;
struct ida_bitmap *bitmap;
unsigned long flags;
BUG_ON((int)id < 0);
xas_lock_irqsave(&xas, flags);
bitmap = xas_load(&xas);
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 {
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.
* @ida: IDA handle.
*
* Calling this function frees all IDs and releases all resources used
* by an IDA. When this call returns, the IDA is empty and can be reused
* or freed. If the IDA is already empty, there is no need to call this
* function.
*
* Context: Any context.
*/
void ida_destroy(struct ida *ida)
{
XA_STATE(xas, &ida->xa, 0);
struct ida_bitmap *bitmap;
unsigned long flags;
xas_lock_irqsave(&xas, flags);
xas_for_each(&xas, bitmap, ULONG_MAX) {
if (!xa_is_value(bitmap))
kfree(bitmap);
xas_store(&xas, NULL);
}
xas_unlock_irqrestore(&xas, flags);
}
EXPORT_SYMBOL(ida_destroy);
#ifndef __KERNEL__
extern void xa_dump_index(unsigned long index, unsigned int shift);
#define IDA_CHUNK_SHIFT ilog2(IDA_BITMAP_BITS)
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;
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");
}
}
static void ida_dump(struct ida *ida)
{
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);
}
#endif