Revision 9e2b7fa2df4365e99934901da4fb4af52d81e820 authored by Martin Willi on 06 November 2020, 07:30:30 UTC, committed by Jakub Kicinski on 12 November 2020, 15:47:06 UTC
VRF devices use an optimized direct path on output if a default qdisc is involved, calling Netfilter hooks directly. This path, however, does not consider Netfilter rules completing asynchronously, such as with NFQUEUE. The Netfilter okfn() is called for asynchronously accepted packets, but the VRF never passes that packet down the stack to send it out over the slave device. Using the slower redirect path for this seems not feasible, as we do not know beforehand if a Netfilter hook has asynchronously completing rules. Fix the use of asynchronously completing Netfilter rules in OUTPUT and POSTROUTING by using a special completion function that additionally calls dst_output() to pass the packet down the stack. Also, slightly adjust the use of nf_reset_ct() so that is called in the asynchronous case, too. Fixes: dcdd43c41e60 ("net: vrf: performance improvements for IPv4") Fixes: a9ec54d1b0cd ("net: vrf: performance improvements for IPv6") Signed-off-by: Martin Willi <martin@strongswan.org> Link: https://lore.kernel.org/r/20201106073030.3974927-1-martin@strongswan.org Signed-off-by: Jakub Kicinski <kuba@kernel.org>
1 parent 52755b6
bucket_locks.c
#include <linux/export.h>
#include <linux/kernel.h>
#include <linux/mm.h>
#include <linux/slab.h>
#include <linux/vmalloc.h>
/* Allocate an array of spinlocks to be accessed by a hash. Two arguments
* indicate the number of elements to allocate in the array. max_size
* gives the maximum number of elements to allocate. cpu_mult gives
* the number of locks per CPU to allocate. The size is rounded up
* to a power of 2 to be suitable as a hash table.
*/
int __alloc_bucket_spinlocks(spinlock_t **locks, unsigned int *locks_mask,
size_t max_size, unsigned int cpu_mult, gfp_t gfp,
const char *name, struct lock_class_key *key)
{
spinlock_t *tlocks = NULL;
unsigned int i, size;
#if defined(CONFIG_PROVE_LOCKING)
unsigned int nr_pcpus = 2;
#else
unsigned int nr_pcpus = num_possible_cpus();
#endif
if (cpu_mult) {
nr_pcpus = min_t(unsigned int, nr_pcpus, 64UL);
size = min_t(unsigned int, nr_pcpus * cpu_mult, max_size);
} else {
size = max_size;
}
if (sizeof(spinlock_t) != 0) {
tlocks = kvmalloc_array(size, sizeof(spinlock_t), gfp);
if (!tlocks)
return -ENOMEM;
for (i = 0; i < size; i++) {
spin_lock_init(&tlocks[i]);
lockdep_init_map(&tlocks[i].dep_map, name, key, 0);
}
}
*locks = tlocks;
*locks_mask = size - 1;
return 0;
}
EXPORT_SYMBOL(__alloc_bucket_spinlocks);
void free_bucket_spinlocks(spinlock_t *locks)
{
kvfree(locks);
}
EXPORT_SYMBOL(free_bucket_spinlocks);
![swh spinner](/static/img/swh-spinner.gif)
Computing file changes ...