mirror of
https://github.com/openvswitch/ovs
synced 2025-10-13 14:07:02 +00:00
704a1e09e9
One of the goals for Open vSwitch is to decouple kernel and userspace software, so that either one can be upgraded or rolled back independent of the other. To do this in full generality, it must be possible to change the kernel's idea of the flow key separately from the userspace version. In turn, that means that flow keys must become variable-length. This does not, however, fit in well with the ODP_FLOW_LIST ioctl in its current form, because that would require userspace to know how much space to allocate for each flow's key in advance, or to allocate as much space as could possibly be needed. Neither choice is very attractive. This commit prepares for a different solution, by replacing ODP_FLOW_LIST by a new ioctl ODP_FLOW_DUMP that retrieves a single flow from the datapath on each call. It is much cleaner to allocate the maximum amount of space for a single flow key than to do so for possibly a very large number of flow keys. As a side effect, this patch also fixes a race condition that sometimes made "ovs-dpctl dump-flows" print an error: previously, flows were listed and then their actions were retrieved, which left a window in which ovs-vswitchd could delete the flow. Now dumping a flow and its actions is a single step, closing that window. Dumping all of the flows in a datapath is no longer an atomic step, so now it is possible to miss some flows or see a single flow twice during iteration, if the flow table is modified by another process. It doesn't look like this should be a problem for ovs-vswitchd. It would be faster to retrieve a number of flows in batch instead of just one at a time, but that will naturally happen later when the kernel datapath interface is changed to use Netlink, so this patch does not bother with it. Signed-off-by: Ben Pfaff <blp@nicira.com> Acked-by: Jesse Gross <jesse@nicira.com>
72 lines
2.4 KiB
C
72 lines
2.4 KiB
C
/*
|
|
* Copyright (c) 2010 Nicira Networks.
|
|
* Distributed under the terms of the GNU GPL version 2.
|
|
*
|
|
* Significant portions of this file may be copied from parts of the Linux
|
|
* kernel, by Linus Torvalds and others.
|
|
*/
|
|
|
|
#ifndef TABLE_H
|
|
#define TABLE_H 1
|
|
|
|
struct tbl_bucket;
|
|
|
|
struct tbl_node {
|
|
u32 hash;
|
|
};
|
|
|
|
/**
|
|
* struct tbl - hash table
|
|
* @n_buckets: number of buckets (a power of 2 between %TBL_L1_SIZE and
|
|
* %TBL_MAX_BUCKETS)
|
|
* @buckets: pointer to @n_buckets/%TBL_L1_SIZE pointers to %TBL_L1_SIZE pointers
|
|
* to buckets
|
|
* @rcu: RCU callback structure
|
|
* @obj_destructor: Called on each element when the table is destroyed.
|
|
*
|
|
* The @buckets array is logically an array of pointers to buckets. It is
|
|
* broken into two levels to avoid the need to kmalloc() any object larger than
|
|
* a single page or to use vmalloc(). @buckets is always nonnull, as is each
|
|
* @buckets[i], but each @buckets[i][j] is nonnull only if the specified hash
|
|
* bucket is nonempty (for 0 <= i < @n_buckets/%TBL_L1_SIZE, 0 <= j <
|
|
* %TBL_L1_SIZE).
|
|
*/
|
|
struct tbl {
|
|
struct rcu_head rcu;
|
|
unsigned int n_buckets;
|
|
struct tbl_bucket __rcu ***buckets;
|
|
unsigned int count;
|
|
void (*obj_destructor)(struct tbl_node *);
|
|
};
|
|
|
|
#define TBL_L2_BITS (PAGE_SHIFT - ilog2(sizeof(struct tbl_bucket *)))
|
|
#define TBL_L2_SIZE (1 << TBL_L2_BITS)
|
|
#define TBL_L2_SHIFT 0
|
|
|
|
#define TBL_L1_BITS (PAGE_SHIFT - ilog2(sizeof(struct tbl_bucket **)))
|
|
#define TBL_L1_SIZE (1 << TBL_L1_BITS)
|
|
#define TBL_L1_SHIFT TBL_L2_BITS
|
|
|
|
/* For 4 kB pages, this is 1,024 on 32-bit or 512 on 64-bit. */
|
|
#define TBL_MIN_BUCKETS TBL_L2_SIZE
|
|
|
|
/* For 4 kB pages, this is 1,048,576 on 32-bit or 262,144 on 64-bit. */
|
|
#define TBL_MAX_BUCKETS (TBL_L1_SIZE * TBL_L2_SIZE)
|
|
|
|
struct tbl *tbl_create(unsigned int n_buckets);
|
|
void tbl_destroy(struct tbl *, void (*destructor)(struct tbl_node *));
|
|
struct tbl_node *tbl_lookup(struct tbl *, void *target, u32 hash,
|
|
int (*cmp)(const struct tbl_node *, void *target));
|
|
int tbl_insert(struct tbl *, struct tbl_node *, u32 hash);
|
|
int tbl_remove(struct tbl *, struct tbl_node *);
|
|
unsigned int tbl_count(struct tbl *);
|
|
int tbl_foreach(struct tbl *,
|
|
int (*callback)(struct tbl_node *, void *aux), void *aux);
|
|
struct tbl_node *tbl_next(struct tbl *, u32 *bucketp, u32 *objp);
|
|
|
|
int tbl_n_buckets(struct tbl *);
|
|
struct tbl *tbl_expand(struct tbl *);
|
|
void tbl_deferred_destroy(struct tbl *, void (*destructor)(struct tbl_node *));
|
|
|
|
#endif /* table.h */
|