linux/kernel/bpf/local_storage.c

615 lines
15 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0
#include <linux/bpf-cgroup.h>
#include <linux/bpf.h>
#include <linux/bpf_local_storage.h>
#include <linux/btf.h>
#include <linux/bug.h>
#include <linux/filter.h>
#include <linux/mm.h>
#include <linux/rbtree.h>
#include <linux/slab.h>
#include <uapi/linux/btf.h>
#include <linux/btf_ids.h>
#ifdef CONFIG_CGROUP_BPF
#include "../cgroup/cgroup-internal.h"
#define LOCAL_STORAGE_CREATE_FLAG_MASK \
(BPF_F_NUMA_NODE | BPF_F_ACCESS_MASK)
struct bpf_cgroup_storage_map {
struct bpf_map map;
spinlock_t lock;
struct rb_root root;
struct list_head list;
};
static struct bpf_cgroup_storage_map *map_to_storage(struct bpf_map *map)
{
return container_of(map, struct bpf_cgroup_storage_map, map);
}
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
static bool attach_type_isolated(const struct bpf_map *map)
{
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
return map->key_size == sizeof(struct bpf_cgroup_storage_key);
}
static int bpf_cgroup_storage_key_cmp(const struct bpf_cgroup_storage_map *map,
const void *_key1, const void *_key2)
{
if (attach_type_isolated(&map->map)) {
const struct bpf_cgroup_storage_key *key1 = _key1;
const struct bpf_cgroup_storage_key *key2 = _key2;
if (key1->cgroup_inode_id < key2->cgroup_inode_id)
return -1;
else if (key1->cgroup_inode_id > key2->cgroup_inode_id)
return 1;
else if (key1->attach_type < key2->attach_type)
return -1;
else if (key1->attach_type > key2->attach_type)
return 1;
} else {
const __u64 *cgroup_inode_id1 = _key1;
const __u64 *cgroup_inode_id2 = _key2;
if (*cgroup_inode_id1 < *cgroup_inode_id2)
return -1;
else if (*cgroup_inode_id1 > *cgroup_inode_id2)
return 1;
}
return 0;
}
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
struct bpf_cgroup_storage *
cgroup_storage_lookup(struct bpf_cgroup_storage_map *map,
void *key, bool locked)
{
struct rb_root *root = &map->root;
struct rb_node *node;
if (!locked)
spin_lock_bh(&map->lock);
node = root->rb_node;
while (node) {
struct bpf_cgroup_storage *storage;
storage = container_of(node, struct bpf_cgroup_storage, node);
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
switch (bpf_cgroup_storage_key_cmp(map, key, &storage->key)) {
case -1:
node = node->rb_left;
break;
case 1:
node = node->rb_right;
break;
default:
if (!locked)
spin_unlock_bh(&map->lock);
return storage;
}
}
if (!locked)
spin_unlock_bh(&map->lock);
return NULL;
}
static int cgroup_storage_insert(struct bpf_cgroup_storage_map *map,
struct bpf_cgroup_storage *storage)
{
struct rb_root *root = &map->root;
struct rb_node **new = &(root->rb_node), *parent = NULL;
while (*new) {
struct bpf_cgroup_storage *this;
this = container_of(*new, struct bpf_cgroup_storage, node);
parent = *new;
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
switch (bpf_cgroup_storage_key_cmp(map, &storage->key, &this->key)) {
case -1:
new = &((*new)->rb_left);
break;
case 1:
new = &((*new)->rb_right);
break;
default:
return -EEXIST;
}
}
rb_link_node(&storage->node, parent, new);
rb_insert_color(&storage->node, root);
return 0;
}
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
static void *cgroup_storage_lookup_elem(struct bpf_map *_map, void *key)
{
struct bpf_cgroup_storage_map *map = map_to_storage(_map);
struct bpf_cgroup_storage *storage;
storage = cgroup_storage_lookup(map, key, false);
if (!storage)
return NULL;
return &READ_ONCE(storage->buf)->data[0];
}
bpf: return long from bpf_map_ops funcs This patch changes the return types of bpf_map_ops functions to long, where previously int was returned. Using long allows for bpf programs to maintain the sign bit in the absence of sign extension during situations where inlined bpf helper funcs make calls to the bpf_map_ops funcs and a negative error is returned. The definitions of the helper funcs are generated from comments in the bpf uapi header at `include/uapi/linux/bpf.h`. The return type of these helpers was previously changed from int to long in commit bdb7b79b4ce8. For any case where one of the map helpers call the bpf_map_ops funcs that are still returning 32-bit int, a compiler might not include sign extension instructions to properly convert the 32-bit negative value a 64-bit negative value. For example: bpf assembly excerpt of an inlined helper calling a kernel function and checking for a specific error: ; err = bpf_map_update_elem(&mymap, &key, &val, BPF_NOEXIST); ... 46: call 0xffffffffe103291c ; htab_map_update_elem ; if (err && err != -EEXIST) { 4b: cmp $0xffffffffffffffef,%rax ; cmp -EEXIST,%rax kernel function assembly excerpt of return value from `htab_map_update_elem` returning 32-bit int: movl $0xffffffef, %r9d ... movl %r9d, %eax ...results in the comparison: cmp $0xffffffffffffffef, $0x00000000ffffffef Fixes: bdb7b79b4ce8 ("bpf: Switch most helper return values from 32-bit int to 64-bit long") Tested-by: Eduard Zingerman <eddyz87@gmail.com> Signed-off-by: JP Kobryn <inwardvessel@gmail.com> Link: https://lore.kernel.org/r/20230322194754.185781-3-inwardvessel@gmail.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2023-03-22 19:47:54 +00:00
static long cgroup_storage_update_elem(struct bpf_map *map, void *key,
void *value, u64 flags)
{
struct bpf_cgroup_storage *storage;
struct bpf_storage_buffer *new;
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
if (unlikely(flags & ~(BPF_F_LOCK | BPF_EXIST)))
return -EINVAL;
if (unlikely((flags & BPF_F_LOCK) &&
!btf_record_has_field(map->record, BPF_SPIN_LOCK)))
return -EINVAL;
storage = cgroup_storage_lookup((struct bpf_cgroup_storage_map *)map,
key, false);
if (!storage)
return -ENOENT;
if (flags & BPF_F_LOCK) {
copy_map_value_locked(map, storage->buf->data, value, false);
return 0;
}
new = bpf_map_kmalloc_node(map, struct_size(new, data, map->value_size),
bpf: Make non-preallocated allocation low priority GFP_ATOMIC doesn't cooperate well with memcg pressure so far, especially if we allocate too much GFP_ATOMIC memory. For example, when we set the memcg limit to limit a non-preallocated bpf memory, the GFP_ATOMIC can easily break the memcg limit by force charge. So it is very dangerous to use GFP_ATOMIC in non-preallocated case. One way to make it safe is to remove __GFP_HIGH from GFP_ATOMIC, IOW, use (__GFP_ATOMIC | __GFP_KSWAPD_RECLAIM) instead, then it will be limited if we allocate too much memory. There's a plan to completely remove __GFP_ATOMIC in the mm side[1], so let's use GFP_NOWAIT instead. We introduced BPF_F_NO_PREALLOC is because full map pre-allocation is too memory expensive for some cases. That means removing __GFP_HIGH doesn't break the rule of BPF_F_NO_PREALLOC, but has the same goal with it-avoiding issues caused by too much memory. So let's remove it. This fix can also apply to other run-time allocations, for example, the allocation in lpm trie, local storage and devmap. So let fix it consistently over the bpf code It also fixes a typo in the comment. [1]. https://lore.kernel.org/linux-mm/163712397076.13692.4727608274002939094@noble.neil.brown.name/ Cc: Roman Gushchin <roman.gushchin@linux.dev> Cc: Shakeel Butt <shakeelb@google.com> Cc: NeilBrown <neilb@suse.de> Signed-off-by: Yafang Shao <laoar.shao@gmail.com> Reviewed-by: Shakeel Butt <shakeelb@google.com> Link: https://lore.kernel.org/r/20220709154457.57379-2-laoar.shao@gmail.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2022-07-09 15:44:56 +00:00
__GFP_ZERO | GFP_NOWAIT | __GFP_NOWARN,
map->numa_node);
if (!new)
return -ENOMEM;
memcpy(&new->data[0], value, map->value_size);
bpf: Add map side support for bpf timers. Restrict bpf timers to array, hash (both preallocated and kmalloced), and lru map types. The per-cpu maps with timers don't make sense, since 'struct bpf_timer' is a part of map value. bpf timers in per-cpu maps would mean that the number of timers depends on number of possible cpus and timers would not be accessible from all cpus. lpm map support can be added in the future. The timers in inner maps are supported. The bpf_map_update/delete_elem() helpers and sys_bpf commands cancel and free bpf_timer in a given map element. Similar to 'struct bpf_spin_lock' BTF is required and it is used to validate that map element indeed contains 'struct bpf_timer'. Make check_and_init_map_value() init both bpf_spin_lock and bpf_timer when map element data is reused in preallocated htab and lru maps. Teach copy_map_value() to support both bpf_spin_lock and bpf_timer in a single map element. There could be one of each, but not more than one. Due to 'one bpf_timer in one element' restriction do not support timers in global data, since global data is a map of single element, but from bpf program side it's seen as many global variables and restriction of single global timer would be odd. The sys_bpf map_freeze and sys_mmap syscalls are not allowed on maps with timers, since user space could have corrupted mmap element and crashed the kernel. The maps with timers cannot be readonly. Due to these restrictions search for bpf_timer in datasec BTF in case it was placed in the global data to report clear error. The previous patch allowed 'struct bpf_timer' as a first field in a map element only. Relax this restriction. Refactor lru map to s/bpf_lru_push_free/htab_lru_push_free/ to cancel and free the timer when lru map deletes an element as a part of it eviction algorithm. Make sure that bpf program cannot access 'struct bpf_timer' via direct load/store. The timer operation are done through helpers only. This is similar to 'struct bpf_spin_lock'. Signed-off-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Yonghong Song <yhs@fb.com> Acked-by: Martin KaFai Lau <kafai@fb.com> Acked-by: Andrii Nakryiko <andrii@kernel.org> Acked-by: Toke Høiland-Jørgensen <toke@redhat.com> Link: https://lore.kernel.org/bpf/20210715005417.78572-5-alexei.starovoitov@gmail.com
2021-07-15 00:54:10 +00:00
check_and_init_map_value(map, new->data);
new = xchg(&storage->buf, new);
kfree_rcu(new, rcu);
return 0;
}
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
int bpf_percpu_cgroup_storage_copy(struct bpf_map *_map, void *key,
void *value)
{
struct bpf_cgroup_storage_map *map = map_to_storage(_map);
struct bpf_cgroup_storage *storage;
int cpu, off = 0;
u32 size;
rcu_read_lock();
storage = cgroup_storage_lookup(map, key, false);
if (!storage) {
rcu_read_unlock();
return -ENOENT;
}
/* per_cpu areas are zero-filled and bpf programs can only
* access 'value_size' of them, so copying rounded areas
* will not leak any kernel data
*/
size = round_up(_map->value_size, 8);
for_each_possible_cpu(cpu) {
bpf_long_memcpy(value + off,
per_cpu_ptr(storage->percpu_buf, cpu), size);
off += size;
}
rcu_read_unlock();
return 0;
}
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
int bpf_percpu_cgroup_storage_update(struct bpf_map *_map, void *key,
void *value, u64 map_flags)
{
struct bpf_cgroup_storage_map *map = map_to_storage(_map);
struct bpf_cgroup_storage *storage;
int cpu, off = 0;
u32 size;
if (map_flags != BPF_ANY && map_flags != BPF_EXIST)
return -EINVAL;
rcu_read_lock();
storage = cgroup_storage_lookup(map, key, false);
if (!storage) {
rcu_read_unlock();
return -ENOENT;
}
/* the user space will provide round_up(value_size, 8) bytes that
* will be copied into per-cpu area. bpf programs can only access
* value_size of it. During lookup the same extra bytes will be
* returned or zeros which were zero-filled by percpu_alloc,
* so no kernel data leaks possible
*/
size = round_up(_map->value_size, 8);
for_each_possible_cpu(cpu) {
bpf_long_memcpy(per_cpu_ptr(storage->percpu_buf, cpu),
value + off, size);
off += size;
}
rcu_read_unlock();
return 0;
}
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
static int cgroup_storage_get_next_key(struct bpf_map *_map, void *key,
void *_next_key)
{
struct bpf_cgroup_storage_map *map = map_to_storage(_map);
struct bpf_cgroup_storage *storage;
spin_lock_bh(&map->lock);
if (list_empty(&map->list))
goto enoent;
if (key) {
storage = cgroup_storage_lookup(map, key, true);
if (!storage)
goto enoent;
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
storage = list_next_entry(storage, list_map);
if (!storage)
goto enoent;
} else {
storage = list_first_entry(&map->list,
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
struct bpf_cgroup_storage, list_map);
}
spin_unlock_bh(&map->lock);
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
if (attach_type_isolated(&map->map)) {
struct bpf_cgroup_storage_key *next = _next_key;
*next = storage->key;
} else {
__u64 *next = _next_key;
*next = storage->key.cgroup_inode_id;
}
return 0;
enoent:
spin_unlock_bh(&map->lock);
return -ENOENT;
}
static struct bpf_map *cgroup_storage_map_alloc(union bpf_attr *attr)
{
__u32 max_value_size = BPF_LOCAL_STORAGE_MAX_VALUE_SIZE;
int numa_node = bpf_map_attr_numa_node(attr);
struct bpf_cgroup_storage_map *map;
/* percpu is bound by PCPU_MIN_UNIT_SIZE, non-percu
* is the same as other local storages.
*/
if (attr->map_type == BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE)
max_value_size = min_t(__u32, max_value_size,
PCPU_MIN_UNIT_SIZE);
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
if (attr->key_size != sizeof(struct bpf_cgroup_storage_key) &&
attr->key_size != sizeof(__u64))
return ERR_PTR(-EINVAL);
if (attr->value_size == 0)
return ERR_PTR(-EINVAL);
if (attr->value_size > max_value_size)
return ERR_PTR(-E2BIG);
if (attr->map_flags & ~LOCAL_STORAGE_CREATE_FLAG_MASK ||
!bpf_map_flags_access_ok(attr->map_flags))
return ERR_PTR(-EINVAL);
if (attr->max_entries)
/* max_entries is not used and enforced to be 0 */
return ERR_PTR(-EINVAL);
map = bpf_map_area_alloc(sizeof(struct bpf_cgroup_storage_map), numa_node);
if (!map)
return ERR_PTR(-ENOMEM);
/* copy mandatory map attributes */
bpf_map_init_from_attr(&map->map, attr);
spin_lock_init(&map->lock);
map->root = RB_ROOT;
INIT_LIST_HEAD(&map->list);
return &map->map;
}
static void cgroup_storage_map_free(struct bpf_map *_map)
{
struct bpf_cgroup_storage_map *map = map_to_storage(_map);
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
struct list_head *storages = &map->list;
struct bpf_cgroup_storage *storage, *stmp;
cgroup_lock();
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
list_for_each_entry_safe(storage, stmp, storages, list_map) {
bpf_cgroup_storage_unlink(storage);
bpf_cgroup_storage_free(storage);
}
cgroup_unlock();
WARN_ON(!RB_EMPTY_ROOT(&map->root));
WARN_ON(!list_empty(&map->list));
bpf_map_area_free(map);
}
bpf: return long from bpf_map_ops funcs This patch changes the return types of bpf_map_ops functions to long, where previously int was returned. Using long allows for bpf programs to maintain the sign bit in the absence of sign extension during situations where inlined bpf helper funcs make calls to the bpf_map_ops funcs and a negative error is returned. The definitions of the helper funcs are generated from comments in the bpf uapi header at `include/uapi/linux/bpf.h`. The return type of these helpers was previously changed from int to long in commit bdb7b79b4ce8. For any case where one of the map helpers call the bpf_map_ops funcs that are still returning 32-bit int, a compiler might not include sign extension instructions to properly convert the 32-bit negative value a 64-bit negative value. For example: bpf assembly excerpt of an inlined helper calling a kernel function and checking for a specific error: ; err = bpf_map_update_elem(&mymap, &key, &val, BPF_NOEXIST); ... 46: call 0xffffffffe103291c ; htab_map_update_elem ; if (err && err != -EEXIST) { 4b: cmp $0xffffffffffffffef,%rax ; cmp -EEXIST,%rax kernel function assembly excerpt of return value from `htab_map_update_elem` returning 32-bit int: movl $0xffffffef, %r9d ... movl %r9d, %eax ...results in the comparison: cmp $0xffffffffffffffef, $0x00000000ffffffef Fixes: bdb7b79b4ce8 ("bpf: Switch most helper return values from 32-bit int to 64-bit long") Tested-by: Eduard Zingerman <eddyz87@gmail.com> Signed-off-by: JP Kobryn <inwardvessel@gmail.com> Link: https://lore.kernel.org/r/20230322194754.185781-3-inwardvessel@gmail.com Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2023-03-22 19:47:54 +00:00
static long cgroup_storage_delete_elem(struct bpf_map *map, void *key)
{
return -EINVAL;
}
static int cgroup_storage_check_btf(const struct bpf_map *map,
const struct btf *btf,
const struct btf_type *key_type,
const struct btf_type *value_type)
{
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
if (attach_type_isolated(map)) {
struct btf_member *m;
u32 offset, size;
/* Key is expected to be of struct bpf_cgroup_storage_key type,
* which is:
* struct bpf_cgroup_storage_key {
* __u64 cgroup_inode_id;
* __u32 attach_type;
* };
*/
/*
* Key_type must be a structure with two fields.
*/
if (BTF_INFO_KIND(key_type->info) != BTF_KIND_STRUCT ||
BTF_INFO_VLEN(key_type->info) != 2)
return -EINVAL;
/*
* The first field must be a 64 bit integer at 0 offset.
*/
m = (struct btf_member *)(key_type + 1);
size = sizeof_field(struct bpf_cgroup_storage_key, cgroup_inode_id);
if (!btf_member_is_reg_int(btf, key_type, m, 0, size))
return -EINVAL;
/*
* The second field must be a 32 bit integer at 64 bit offset.
*/
m++;
offset = offsetof(struct bpf_cgroup_storage_key, attach_type);
size = sizeof_field(struct bpf_cgroup_storage_key, attach_type);
if (!btf_member_is_reg_int(btf, key_type, m, offset, size))
return -EINVAL;
} else {
u32 int_data;
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
/*
* Key is expected to be u64, which stores the cgroup_inode_id
*/
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
if (BTF_INFO_KIND(key_type->info) != BTF_KIND_INT)
return -EINVAL;
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
int_data = *(u32 *)(key_type + 1);
if (BTF_INT_BITS(int_data) != 64 || BTF_INT_OFFSET(int_data))
return -EINVAL;
}
return 0;
}
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
static void cgroup_storage_seq_show_elem(struct bpf_map *map, void *key,
struct seq_file *m)
{
enum bpf_cgroup_storage_type stype;
struct bpf_cgroup_storage *storage;
int cpu;
rcu_read_lock();
storage = cgroup_storage_lookup(map_to_storage(map), key, false);
if (!storage) {
rcu_read_unlock();
return;
}
btf_type_seq_show(map->btf, map->btf_key_type_id, key, m);
stype = cgroup_storage_type(map);
if (stype == BPF_CGROUP_STORAGE_SHARED) {
seq_puts(m, ": ");
btf_type_seq_show(map->btf, map->btf_value_type_id,
&READ_ONCE(storage->buf)->data[0], m);
seq_puts(m, "\n");
} else {
seq_puts(m, ": {\n");
for_each_possible_cpu(cpu) {
seq_printf(m, "\tcpu%d: ", cpu);
btf_type_seq_show(map->btf, map->btf_value_type_id,
per_cpu_ptr(storage->percpu_buf, cpu),
m);
seq_puts(m, "\n");
}
seq_puts(m, "}\n");
}
rcu_read_unlock();
}
static u64 cgroup_storage_map_usage(const struct bpf_map *map)
{
/* Currently the dynamically allocated elements are not counted. */
return sizeof(struct bpf_cgroup_storage_map);
}
BTF_ID_LIST_SINGLE(cgroup_storage_map_btf_ids, struct,
bpf_cgroup_storage_map)
const struct bpf_map_ops cgroup_storage_map_ops = {
.map_alloc = cgroup_storage_map_alloc,
.map_free = cgroup_storage_map_free,
.map_get_next_key = cgroup_storage_get_next_key,
.map_lookup_elem = cgroup_storage_lookup_elem,
.map_update_elem = cgroup_storage_update_elem,
.map_delete_elem = cgroup_storage_delete_elem,
.map_check_btf = cgroup_storage_check_btf,
.map_seq_show_elem = cgroup_storage_seq_show_elem,
.map_mem_usage = cgroup_storage_map_usage,
.map_btf_id = &cgroup_storage_map_btf_ids[0],
};
int bpf_cgroup_storage_assign(struct bpf_prog_aux *aux, struct bpf_map *_map)
{
enum bpf_cgroup_storage_type stype = cgroup_storage_type(_map);
if (aux->cgroup_storage[stype] &&
aux->cgroup_storage[stype] != _map)
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
return -EBUSY;
aux->cgroup_storage[stype] = _map;
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
return 0;
}
static size_t bpf_cgroup_storage_calculate_size(struct bpf_map *map, u32 *pages)
{
size_t size;
if (cgroup_storage_type(map) == BPF_CGROUP_STORAGE_SHARED) {
size = sizeof(struct bpf_storage_buffer) + map->value_size;
*pages = round_up(sizeof(struct bpf_cgroup_storage) + size,
PAGE_SIZE) >> PAGE_SHIFT;
} else {
size = map->value_size;
*pages = round_up(round_up(size, 8) * num_possible_cpus(),
PAGE_SIZE) >> PAGE_SHIFT;
}
return size;
}
struct bpf_cgroup_storage *bpf_cgroup_storage_alloc(struct bpf_prog *prog,
enum bpf_cgroup_storage_type stype)
{
const gfp_t gfp = __GFP_ZERO | GFP_USER;
struct bpf_cgroup_storage *storage;
struct bpf_map *map;
size_t size;
u32 pages;
map = prog->aux->cgroup_storage[stype];
if (!map)
return NULL;
size = bpf_cgroup_storage_calculate_size(map, &pages);
storage = bpf_map_kmalloc_node(map, sizeof(struct bpf_cgroup_storage),
gfp, map->numa_node);
if (!storage)
goto enomem;
if (stype == BPF_CGROUP_STORAGE_SHARED) {
storage->buf = bpf_map_kmalloc_node(map, size, gfp,
map->numa_node);
if (!storage->buf)
goto enomem;
bpf: Add map side support for bpf timers. Restrict bpf timers to array, hash (both preallocated and kmalloced), and lru map types. The per-cpu maps with timers don't make sense, since 'struct bpf_timer' is a part of map value. bpf timers in per-cpu maps would mean that the number of timers depends on number of possible cpus and timers would not be accessible from all cpus. lpm map support can be added in the future. The timers in inner maps are supported. The bpf_map_update/delete_elem() helpers and sys_bpf commands cancel and free bpf_timer in a given map element. Similar to 'struct bpf_spin_lock' BTF is required and it is used to validate that map element indeed contains 'struct bpf_timer'. Make check_and_init_map_value() init both bpf_spin_lock and bpf_timer when map element data is reused in preallocated htab and lru maps. Teach copy_map_value() to support both bpf_spin_lock and bpf_timer in a single map element. There could be one of each, but not more than one. Due to 'one bpf_timer in one element' restriction do not support timers in global data, since global data is a map of single element, but from bpf program side it's seen as many global variables and restriction of single global timer would be odd. The sys_bpf map_freeze and sys_mmap syscalls are not allowed on maps with timers, since user space could have corrupted mmap element and crashed the kernel. The maps with timers cannot be readonly. Due to these restrictions search for bpf_timer in datasec BTF in case it was placed in the global data to report clear error. The previous patch allowed 'struct bpf_timer' as a first field in a map element only. Relax this restriction. Refactor lru map to s/bpf_lru_push_free/htab_lru_push_free/ to cancel and free the timer when lru map deletes an element as a part of it eviction algorithm. Make sure that bpf program cannot access 'struct bpf_timer' via direct load/store. The timer operation are done through helpers only. This is similar to 'struct bpf_spin_lock'. Signed-off-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Yonghong Song <yhs@fb.com> Acked-by: Martin KaFai Lau <kafai@fb.com> Acked-by: Andrii Nakryiko <andrii@kernel.org> Acked-by: Toke Høiland-Jørgensen <toke@redhat.com> Link: https://lore.kernel.org/bpf/20210715005417.78572-5-alexei.starovoitov@gmail.com
2021-07-15 00:54:10 +00:00
check_and_init_map_value(map, storage->buf->data);
} else {
storage->percpu_buf = bpf_map_alloc_percpu(map, size, 8, gfp);
if (!storage->percpu_buf)
goto enomem;
}
storage->map = (struct bpf_cgroup_storage_map *)map;
return storage;
enomem:
kfree(storage);
return ERR_PTR(-ENOMEM);
}
static void free_shared_cgroup_storage_rcu(struct rcu_head *rcu)
{
struct bpf_cgroup_storage *storage =
container_of(rcu, struct bpf_cgroup_storage, rcu);
kfree(storage->buf);
kfree(storage);
}
static void free_percpu_cgroup_storage_rcu(struct rcu_head *rcu)
{
struct bpf_cgroup_storage *storage =
container_of(rcu, struct bpf_cgroup_storage, rcu);
free_percpu(storage->percpu_buf);
kfree(storage);
}
void bpf_cgroup_storage_free(struct bpf_cgroup_storage *storage)
{
enum bpf_cgroup_storage_type stype;
struct bpf_map *map;
if (!storage)
return;
map = &storage->map->map;
stype = cgroup_storage_type(map);
if (stype == BPF_CGROUP_STORAGE_SHARED)
call_rcu(&storage->rcu, free_shared_cgroup_storage_rcu);
else
call_rcu(&storage->rcu, free_percpu_cgroup_storage_rcu);
}
void bpf_cgroup_storage_link(struct bpf_cgroup_storage *storage,
struct cgroup *cgroup,
enum bpf_attach_type type)
{
struct bpf_cgroup_storage_map *map;
if (!storage)
return;
storage->key.attach_type = type;
storage->key.cgroup_inode_id = cgroup_id(cgroup);
map = storage->map;
spin_lock_bh(&map->lock);
WARN_ON(cgroup_storage_insert(map, storage));
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
list_add(&storage->list_map, &map->list);
list_add(&storage->list_cg, &cgroup->bpf.storages);
spin_unlock_bh(&map->lock);
}
void bpf_cgroup_storage_unlink(struct bpf_cgroup_storage *storage)
{
struct bpf_cgroup_storage_map *map;
struct rb_root *root;
if (!storage)
return;
map = storage->map;
spin_lock_bh(&map->lock);
root = &map->root;
rb_erase(&storage->node, root);
bpf: Make cgroup storages shared between programs on the same cgroup This change comes in several parts: One, the restriction that the CGROUP_STORAGE map can only be used by one program is removed. This results in the removal of the field 'aux' in struct bpf_cgroup_storage_map, and removal of relevant code associated with the field, and removal of now-noop functions bpf_free_cgroup_storage and bpf_cgroup_storage_release. Second, we permit a key of type u64 as the key to the map. Providing such a key type indicates that the map should ignore attach type when comparing map keys. However, for simplicity newly linked storage will still have the attach type at link time in its key struct. cgroup_storage_check_btf is adapted to accept u64 as the type of the key. Third, because the storages are now shared, the storages cannot be unconditionally freed on program detach. There could be two ways to solve this issue: * A. Reference count the usage of the storages, and free when the last program is detached. * B. Free only when the storage is impossible to be referred to again, i.e. when either the cgroup_bpf it is attached to, or the map itself, is freed. Option A has the side effect that, when the user detach and reattach a program, whether the program gets a fresh storage depends on whether there is another program attached using that storage. This could trigger races if the user is multi-threaded, and since nondeterminism in data races is evil, go with option B. The both the map and the cgroup_bpf now tracks their associated storages, and the storage unlink and free are removed from cgroup_bpf_detach and added to cgroup_bpf_release and cgroup_storage_map_free. The latter also new holds the cgroup_mutex to prevent any races with the former. Fourth, on attach, we reuse the old storage if the key already exists in the map, via cgroup_storage_lookup. If the storage does not exist yet, we create a new one, and publish it at the last step in the attach process. This does not create a race condition because for the whole attach the cgroup_mutex is held. We keep track of an array of new storages that was allocated and if the process fails only the new storages would get freed. Signed-off-by: YiFei Zhu <zhuyifei@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/d5401c6106728a00890401190db40020a1f84ff1.1595565795.git.zhuyifei@google.com
2020-07-24 04:47:43 +00:00
list_del(&storage->list_map);
list_del(&storage->list_cg);
spin_unlock_bh(&map->lock);
}
#endif