From da2577fdd0932ea4eefe73903f1130ee366767d2 Mon Sep 17 00:00:00 2001
From: Jonathan Lemon <jonathan.lemon@gmail.com>
Date: Sat, 8 Jun 2019 12:54:19 -0700
Subject: [PATCH 01/13] bpf: lpm_trie: check left child of last leftmost node
for NULL
If the leftmost parent node of the tree has does not have a child
on the left side, then trie_get_next_key (and bpftool map dump) will
not look at the child on the right. This leads to the traversal
missing elements.
Lookup is not affected.
Update selftest to handle this case.
Reproducer:
bpftool map create /sys/fs/bpf/lpm type lpm_trie key 6 \
value 1 entries 256 name test_lpm flags 1
bpftool map update pinned /sys/fs/bpf/lpm key 8 0 0 0 0 0 value 1
bpftool map update pinned /sys/fs/bpf/lpm key 16 0 0 0 0 128 value 2
bpftool map dump pinned /sys/fs/bpf/lpm
Returns only 1 element. (2 expected)
Fixes: b471f2f1de8b ("bpf: implement MAP_GET_NEXT_KEY command for LPM_TRIE")
Signed-off-by: Jonathan Lemon <jonathan.lemon@gmail.com>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
---
kernel/bpf/lpm_trie.c | 9 +++--
tools/testing/selftests/bpf/test_lpm_map.c | 41 ++++++++++++++++++++--
2 files changed, 45 insertions(+), 5 deletions(-)
diff --git a/kernel/bpf/lpm_trie.c b/kernel/bpf/lpm_trie.c
index e61630c2e50b2..864e2a4963762 100644
--- a/kernel/bpf/lpm_trie.c
+++ b/kernel/bpf/lpm_trie.c
@@ -716,9 +716,14 @@ static int trie_get_next_key(struct bpf_map *map, void *_key, void *_next_key)
* have exact two children, so this function will never return NULL.
*/
for (node = search_root; node;) {
- if (!(node->flags & LPM_TREE_NODE_FLAG_IM))
+ if (node->flags & LPM_TREE_NODE_FLAG_IM) {
+ node = rcu_dereference(node->child[0]);
+ } else {
next_node = node;
- node = rcu_dereference(node->child[0]);
+ node = rcu_dereference(node->child[0]);
+ if (!node)
+ node = rcu_dereference(next_node->child[1]);
+ }
}
do_copy:
next_key->prefixlen = next_node->prefixlen;
diff --git a/tools/testing/selftests/bpf/test_lpm_map.c b/tools/testing/selftests/bpf/test_lpm_map.c
index 02d7c871862af..006be39639778 100644
--- a/tools/testing/selftests/bpf/test_lpm_map.c
+++ b/tools/testing/selftests/bpf/test_lpm_map.c
@@ -573,13 +573,13 @@ static void test_lpm_get_next_key(void)
/* add one more element (total two) */
key_p->prefixlen = 24;
- inet_pton(AF_INET, "192.168.0.0", key_p->data);
+ inet_pton(AF_INET, "192.168.128.0", key_p->data);
assert(bpf_map_update_elem(map_fd, key_p, &value, 0) == 0);
memset(key_p, 0, key_size);
assert(bpf_map_get_next_key(map_fd, NULL, key_p) == 0);
assert(key_p->prefixlen == 24 && key_p->data[0] == 192 &&
- key_p->data[1] == 168 && key_p->data[2] == 0);
+ key_p->data[1] == 168 && key_p->data[2] == 128);
memset(next_key_p, 0, key_size);
assert(bpf_map_get_next_key(map_fd, key_p, next_key_p) == 0);
@@ -592,7 +592,7 @@ static void test_lpm_get_next_key(void)
/* Add one more element (total three) */
key_p->prefixlen = 24;
- inet_pton(AF_INET, "192.168.128.0", key_p->data);
+ inet_pton(AF_INET, "192.168.0.0", key_p->data);
assert(bpf_map_update_elem(map_fd, key_p, &value, 0) == 0);
memset(key_p, 0, key_size);
@@ -643,6 +643,41 @@ static void test_lpm_get_next_key(void)
assert(bpf_map_get_next_key(map_fd, key_p, next_key_p) == -1 &&
errno == ENOENT);
+ /* Add one more element (total five) */
+ key_p->prefixlen = 28;
+ inet_pton(AF_INET, "192.168.1.128", key_p->data);
+ assert(bpf_map_update_elem(map_fd, key_p, &value, 0) == 0);
+
+ memset(key_p, 0, key_size);
+ assert(bpf_map_get_next_key(map_fd, NULL, key_p) == 0);
+ assert(key_p->prefixlen == 24 && key_p->data[0] == 192 &&
+ key_p->data[1] == 168 && key_p->data[2] == 0);
+
+ memset(next_key_p, 0, key_size);
+ assert(bpf_map_get_next_key(map_fd, key_p, next_key_p) == 0);
+ assert(next_key_p->prefixlen == 28 && next_key_p->data[0] == 192 &&
+ next_key_p->data[1] == 168 && next_key_p->data[2] == 1 &&
+ next_key_p->data[3] == 128);
+
+ memcpy(key_p, next_key_p, key_size);
+ assert(bpf_map_get_next_key(map_fd, key_p, next_key_p) == 0);
+ assert(next_key_p->prefixlen == 24 && next_key_p->data[0] == 192 &&
+ next_key_p->data[1] == 168 && next_key_p->data[2] == 1);
+
+ memcpy(key_p, next_key_p, key_size);
+ assert(bpf_map_get_next_key(map_fd, key_p, next_key_p) == 0);
+ assert(next_key_p->prefixlen == 24 && next_key_p->data[0] == 192 &&
+ next_key_p->data[1] == 168 && next_key_p->data[2] == 128);
+
+ memcpy(key_p, next_key_p, key_size);
+ assert(bpf_map_get_next_key(map_fd, key_p, next_key_p) == 0);
+ assert(next_key_p->prefixlen == 16 && next_key_p->data[0] == 192 &&
+ next_key_p->data[1] == 168);
+
+ memcpy(key_p, next_key_p, key_size);
+ assert(bpf_map_get_next_key(map_fd, key_p, next_key_p) == -1 &&
+ errno == ENOENT);
+
/* no exact matching key should return the first one in post order */
key_p->prefixlen = 22;
inet_pton(AF_INET, "192.168.1.0", key_p->data);
From f12dd75959b0138f94da8ddcf43f2f3cf277216d Mon Sep 17 00:00:00 2001
From: Martin KaFai Lau <kafai@fb.com>
Date: Tue, 11 Jun 2019 14:45:57 -0700
Subject: [PATCH 02/13] bpf: net: Set sk_bpf_storage back to NULL for cloned sk
The cloned sk should not carry its parent-listener's sk_bpf_storage.
This patch fixes it by setting it back to NULL.
Fixes: 6ac99e8f23d4 ("bpf: Introduce bpf sk local storage")
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
---
net/core/sock.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/net/core/sock.c b/net/core/sock.c
index 2b37019584862..d90fd04622e58 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1850,6 +1850,9 @@ struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority)
goto out;
}
RCU_INIT_POINTER(newsk->sk_reuseport_cb, NULL);
+#ifdef CONFIG_BPF_SYSCALL
+ RCU_INIT_POINTER(newsk->sk_bpf_storage, NULL);
+#endif
newsk->sk_err = 0;
newsk->sk_err_soft = 0;
From 01d76b5317003e019ace561a9b775f51aafdfdc4 Mon Sep 17 00:00:00 2001
From: Ilya Maximets <i.maximets@samsung.com>
Date: Fri, 7 Jun 2019 20:27:32 +0300
Subject: [PATCH 03/13] xdp: check device pointer before clearing
We should not call 'ndo_bpf()' or 'dev_put()' with NULL argument.
Fixes: c9b47cc1fabc ("xsk: fix bug when trying to use both copy and zero-copy on one queue id")
Signed-off-by: Ilya Maximets <i.maximets@samsung.com>
Acked-by: Jonathan Lemon <jonathan.lemon@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
---
net/xdp/xdp_umem.c | 11 ++++++-----
1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/net/xdp/xdp_umem.c b/net/xdp/xdp_umem.c
index 2b18223e7eb8c..9c6de4f114f84 100644
--- a/net/xdp/xdp_umem.c
+++ b/net/xdp/xdp_umem.c
@@ -143,6 +143,9 @@ static void xdp_umem_clear_dev(struct xdp_umem *umem)
struct netdev_bpf bpf;
int err;
+ if (!umem->dev)
+ return;
+
if (umem->zc) {
bpf.command = XDP_SETUP_XSK_UMEM;
bpf.xsk.umem = NULL;
@@ -156,11 +159,9 @@ static void xdp_umem_clear_dev(struct xdp_umem *umem)
WARN(1, "failed to disable umem!\n");
}
- if (umem->dev) {
- rtnl_lock();
- xdp_clear_umem_at_qid(umem->dev, umem->queue_id);
- rtnl_unlock();
- }
+ rtnl_lock();
+ xdp_clear_umem_at_qid(umem->dev, umem->queue_id);
+ rtnl_unlock();
if (umem->zc) {
dev_put(umem->dev);
From b1d6c15b9d824a58c5415673f374fac19e8eccdf Mon Sep 17 00:00:00 2001
From: Martynas Pumputis <m@lambda.lt>
Date: Wed, 12 Jun 2019 18:05:40 +0200
Subject: [PATCH 04/13] bpf: simplify definition of BPF_FIB_LOOKUP related
flags
Previously, the BPF_FIB_LOOKUP_{DIRECT,OUTPUT} flags in the BPF UAPI
were defined with the help of BIT macro. This had the following issues:
- In order to use any of the flags, a user was required to depend
on <linux/bits.h>.
- No other flag in bpf.h uses the macro, so it seems that an unwritten
convention is to use (1 << (nr)) to define BPF-related flags.
Fixes: 87f5fc7e48dd ("bpf: Provide helper to do forwarding lookups in kernel FIB table")
Signed-off-by: Martynas Pumputis <m@lambda.lt>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
---
include/uapi/linux/bpf.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index e4114a7e44511..a8b823c30b434 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -3378,8 +3378,8 @@ struct bpf_raw_tracepoint_args {
/* DIRECT: Skip the FIB rules and go to FIB table associated with device
* OUTPUT: Do lookup from egress perspective; default is ingress
*/
-#define BPF_FIB_LOOKUP_DIRECT BIT(0)
-#define BPF_FIB_LOOKUP_OUTPUT BIT(1)
+#define BPF_FIB_LOOKUP_DIRECT (1U << 0)
+#define BPF_FIB_LOOKUP_OUTPUT (1U << 1)
enum {
BPF_FIB_LKUP_RET_SUCCESS, /* lookup successful */
From 0e265747491cb182429adfbef7d002164566f25c Mon Sep 17 00:00:00 2001
From: Martynas Pumputis <m@lambda.lt>
Date: Wed, 12 Jun 2019 18:05:41 +0200
Subject: [PATCH 05/13] bpf: sync BPF_FIB_LOOKUP flag changes with BPF uapi
Sync the changes to the flags made in "bpf: simplify definition of
BPF_FIB_LOOKUP related flags" with the BPF UAPI headers.
Doing in a separate commit to ease syncing of github/libbpf.
Signed-off-by: Martynas Pumputis <m@lambda.lt>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
---
tools/include/uapi/linux/bpf.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index e4114a7e44511..a8b823c30b434 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -3378,8 +3378,8 @@ struct bpf_raw_tracepoint_args {
/* DIRECT: Skip the FIB rules and go to FIB table associated with device
* OUTPUT: Do lookup from egress perspective; default is ingress
*/
-#define BPF_FIB_LOOKUP_DIRECT BIT(0)
-#define BPF_FIB_LOOKUP_OUTPUT BIT(1)
+#define BPF_FIB_LOOKUP_DIRECT (1U << 0)
+#define BPF_FIB_LOOKUP_OUTPUT (1U << 1)
enum {
BPF_FIB_LKUP_RET_SUCCESS, /* lookup successful */
From 3e0682695199bad51dd898fe064d1564637ff77a Mon Sep 17 00:00:00 2001
From: "Naveen N. Rao" <naveen.n.rao@linux.vnet.ibm.com>
Date: Thu, 13 Jun 2019 00:21:39 +0530
Subject: [PATCH 06/13] bpf: fix div64 overflow tests to properly detect errors
If the result of the division is LLONG_MIN, current tests do not detect
the error since the return value is truncated to a 32-bit value and ends
up being 0.
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
---
.../testing/selftests/bpf/verifier/div_overflow.c | 14 ++++++++++----
1 file changed, 10 insertions(+), 4 deletions(-)
diff --git a/tools/testing/selftests/bpf/verifier/div_overflow.c b/tools/testing/selftests/bpf/verifier/div_overflow.c
index bd3f38dbe7964..acab4f00819f3 100644
--- a/tools/testing/selftests/bpf/verifier/div_overflow.c
+++ b/tools/testing/selftests/bpf/verifier/div_overflow.c
@@ -29,8 +29,11 @@
"DIV64 overflow, check 1",
.insns = {
BPF_MOV64_IMM(BPF_REG_1, -1),
- BPF_LD_IMM64(BPF_REG_0, LLONG_MIN),
- BPF_ALU64_REG(BPF_DIV, BPF_REG_0, BPF_REG_1),
+ BPF_LD_IMM64(BPF_REG_2, LLONG_MIN),
+ BPF_ALU64_REG(BPF_DIV, BPF_REG_2, BPF_REG_1),
+ BPF_MOV32_IMM(BPF_REG_0, 0),
+ BPF_JMP_REG(BPF_JEQ, BPF_REG_0, BPF_REG_2, 1),
+ BPF_MOV32_IMM(BPF_REG_0, 1),
BPF_EXIT_INSN(),
},
.prog_type = BPF_PROG_TYPE_SCHED_CLS,
@@ -40,8 +43,11 @@
{
"DIV64 overflow, check 2",
.insns = {
- BPF_LD_IMM64(BPF_REG_0, LLONG_MIN),
- BPF_ALU64_IMM(BPF_DIV, BPF_REG_0, -1),
+ BPF_LD_IMM64(BPF_REG_1, LLONG_MIN),
+ BPF_ALU64_IMM(BPF_DIV, BPF_REG_1, -1),
+ BPF_MOV32_IMM(BPF_REG_0, 0),
+ BPF_JMP_REG(BPF_JEQ, BPF_REG_0, BPF_REG_1, 1),
+ BPF_MOV32_IMM(BPF_REG_0, 1),
BPF_EXIT_INSN(),
},
.prog_type = BPF_PROG_TYPE_SCHED_CLS,
From 758f2046ea040773ae8ea7f72dd3bbd8fa984501 Mon Sep 17 00:00:00 2001
From: "Naveen N. Rao" <naveen.n.rao@linux.vnet.ibm.com>
Date: Thu, 13 Jun 2019 00:21:40 +0530
Subject: [PATCH 07/13] powerpc/bpf: use unsigned division instruction for
64-bit operations
BPF_ALU64 div/mod operations are currently using signed division, unlike
BPF_ALU32 operations. Fix the same. DIV64 and MOD64 overflow tests pass
with this fix.
Fixes: 156d0e290e969c ("powerpc/ebpf/jit: Implement JIT compiler for extended BPF")
Cc: stable@vger.kernel.org # v4.8+
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
---
arch/powerpc/include/asm/ppc-opcode.h | 1 +
arch/powerpc/net/bpf_jit.h | 2 +-
arch/powerpc/net/bpf_jit_comp64.c | 8 ++++----
3 files changed, 6 insertions(+), 5 deletions(-)
diff --git a/arch/powerpc/include/asm/ppc-opcode.h b/arch/powerpc/include/asm/ppc-opcode.h
index 493c5c943acd0..2291daf39cd15 100644
--- a/arch/powerpc/include/asm/ppc-opcode.h
+++ b/arch/powerpc/include/asm/ppc-opcode.h
@@ -338,6 +338,7 @@
#define PPC_INST_MADDLD 0x10000033
#define PPC_INST_DIVWU 0x7c000396
#define PPC_INST_DIVD 0x7c0003d2
+#define PPC_INST_DIVDU 0x7c000392
#define PPC_INST_RLWINM 0x54000000
#define PPC_INST_RLWINM_DOT 0x54000001
#define PPC_INST_RLWIMI 0x50000000
diff --git a/arch/powerpc/net/bpf_jit.h b/arch/powerpc/net/bpf_jit.h
index dcac37745b05c..1e932898d430c 100644
--- a/arch/powerpc/net/bpf_jit.h
+++ b/arch/powerpc/net/bpf_jit.h
@@ -116,7 +116,7 @@
___PPC_RA(a) | IMM_L(i))
#define PPC_DIVWU(d, a, b) EMIT(PPC_INST_DIVWU | ___PPC_RT(d) | \
___PPC_RA(a) | ___PPC_RB(b))
-#define PPC_DIVD(d, a, b) EMIT(PPC_INST_DIVD | ___PPC_RT(d) | \
+#define PPC_DIVDU(d, a, b) EMIT(PPC_INST_DIVDU | ___PPC_RT(d) | \
___PPC_RA(a) | ___PPC_RB(b))
#define PPC_AND(d, a, b) EMIT(PPC_INST_AND | ___PPC_RA(d) | \
___PPC_RS(a) | ___PPC_RB(b))
diff --git a/arch/powerpc/net/bpf_jit_comp64.c b/arch/powerpc/net/bpf_jit_comp64.c
index 21a1dcd4b156c..e3fedeffe40f0 100644
--- a/arch/powerpc/net/bpf_jit_comp64.c
+++ b/arch/powerpc/net/bpf_jit_comp64.c
@@ -399,12 +399,12 @@ static int bpf_jit_build_body(struct bpf_prog *fp, u32 *image,
case BPF_ALU64 | BPF_DIV | BPF_X: /* dst /= src */
case BPF_ALU64 | BPF_MOD | BPF_X: /* dst %= src */
if (BPF_OP(code) == BPF_MOD) {
- PPC_DIVD(b2p[TMP_REG_1], dst_reg, src_reg);
+ PPC_DIVDU(b2p[TMP_REG_1], dst_reg, src_reg);
PPC_MULD(b2p[TMP_REG_1], src_reg,
b2p[TMP_REG_1]);
PPC_SUB(dst_reg, dst_reg, b2p[TMP_REG_1]);
} else
- PPC_DIVD(dst_reg, dst_reg, src_reg);
+ PPC_DIVDU(dst_reg, dst_reg, src_reg);
break;
case BPF_ALU | BPF_MOD | BPF_K: /* (u32) dst %= (u32) imm */
case BPF_ALU | BPF_DIV | BPF_K: /* (u32) dst /= (u32) imm */
@@ -432,7 +432,7 @@ static int bpf_jit_build_body(struct bpf_prog *fp, u32 *image,
break;
case BPF_ALU64:
if (BPF_OP(code) == BPF_MOD) {
- PPC_DIVD(b2p[TMP_REG_2], dst_reg,
+ PPC_DIVDU(b2p[TMP_REG_2], dst_reg,
b2p[TMP_REG_1]);
PPC_MULD(b2p[TMP_REG_1],
b2p[TMP_REG_1],
@@ -440,7 +440,7 @@ static int bpf_jit_build_body(struct bpf_prog *fp, u32 *image,
PPC_SUB(dst_reg, dst_reg,
b2p[TMP_REG_1]);
} else
- PPC_DIVD(dst_reg, dst_reg,
+ PPC_DIVDU(dst_reg, dst_reg,
b2p[TMP_REG_1]);
break;
}
From d4dd153d551634683fccf8881f606fa9f3dfa1ef Mon Sep 17 00:00:00 2001
From: Toshiaki Makita <toshiaki.makita1@gmail.com>
Date: Fri, 14 Jun 2019 17:20:13 +0900
Subject: [PATCH 08/13] bpf, devmap: Fix premature entry free on destroying map
dev_map_free() waits for flush_needed bitmap to be empty in order to
ensure all flush operations have completed before freeing its entries.
However the corresponding clear_bit() was called before using the
entries, so the entries could be used after free.
All access to the entries needs to be done before clearing the bit.
It seems commit a5e2da6e9787 ("bpf: netdev is never null in
__dev_map_flush") accidentally changed the clear_bit() and memory access
order.
Note that the problem happens only in __dev_map_flush(), not in
dev_map_flush_old(). dev_map_flush_old() is called only after nulling
out the corresponding netdev_map entry, so dev_map_free() never frees
the entry thus no such race happens there.
Fixes: a5e2da6e9787 ("bpf: netdev is never null in __dev_map_flush")
Signed-off-by: Toshiaki Makita <toshiaki.makita1@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
---
kernel/bpf/devmap.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index 1e525d70f8335..e001fb1a96b10 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -291,10 +291,10 @@ void __dev_map_flush(struct bpf_map *map)
if (unlikely(!dev))
continue;
- __clear_bit(bit, bitmap);
-
bq = this_cpu_ptr(dev->bulkq);
bq_xmit_all(dev, bq, XDP_XMIT_FLUSH, true);
+
+ __clear_bit(bit, bitmap);
}
}
From edabf4d9dd905acd60048ea1579943801e3a4876 Mon Sep 17 00:00:00 2001
From: Toshiaki Makita <toshiaki.makita1@gmail.com>
Date: Fri, 14 Jun 2019 17:20:14 +0900
Subject: [PATCH 09/13] bpf, devmap: Add missing bulk queue free
dev_map_free() forgot to free bulk queue when freeing its entries.
Fixes: 5d053f9da431 ("bpf: devmap prepare xdp frames for bulking")
Signed-off-by: Toshiaki Makita <toshiaki.makita1@gmail.com>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
---
kernel/bpf/devmap.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index e001fb1a96b10..a126d95d12de6 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -186,6 +186,7 @@ static void dev_map_free(struct bpf_map *map)
if (!dev)
continue;
+ free_percpu(dev->bulkq);
dev_put(dev->dev);
kfree(dev);
}
From 86723c8640633bee4b4588d3c7784ee7a0032f65 Mon Sep 17 00:00:00 2001
From: Toshiaki Makita <toshiaki.makita1@gmail.com>
Date: Fri, 14 Jun 2019 17:20:15 +0900
Subject: [PATCH 10/13] bpf, devmap: Add missing RCU read lock on flush
.ndo_xdp_xmit() assumes it is called under RCU. For example virtio_net
uses RCU to detect it has setup the resources for tx. The assumption
accidentally broke when introducing bulk queue in devmap.
Fixes: 5d053f9da431 ("bpf: devmap prepare xdp frames for bulking")
Reported-by: David Ahern <dsahern@gmail.com>
Signed-off-by: Toshiaki Makita <toshiaki.makita1@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
---
kernel/bpf/devmap.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index a126d95d12de6..1defea4b27553 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -282,6 +282,7 @@ void __dev_map_flush(struct bpf_map *map)
unsigned long *bitmap = this_cpu_ptr(dtab->flush_needed);
u32 bit;
+ rcu_read_lock();
for_each_set_bit(bit, bitmap, map->max_entries) {
struct bpf_dtab_netdev *dev = READ_ONCE(dtab->netdev_map[bit]);
struct xdp_bulk_queue *bq;
@@ -297,6 +298,7 @@ void __dev_map_flush(struct bpf_map *map)
__clear_bit(bit, bitmap);
}
+ rcu_read_unlock();
}
/* rcu_read_lock (from syscall and BPF contexts) ensures that if a delete and/or
@@ -389,6 +391,7 @@ static void dev_map_flush_old(struct bpf_dtab_netdev *dev)
int cpu;
+ rcu_read_lock();
for_each_online_cpu(cpu) {
bitmap = per_cpu_ptr(dev->dtab->flush_needed, cpu);
__clear_bit(dev->bit, bitmap);
@@ -396,6 +399,7 @@ static void dev_map_flush_old(struct bpf_dtab_netdev *dev)
bq = per_cpu_ptr(dev->bulkq, cpu);
bq_xmit_all(dev, bq, XDP_XMIT_FLUSH, false);
}
+ rcu_read_unlock();
}
}
From fe8d9571dc50232b569242fac7ea6332a654f186 Mon Sep 17 00:00:00 2001
From: Alexei Starovoitov <ast@kernel.org>
Date: Fri, 14 Jun 2019 15:43:28 -0700
Subject: [PATCH 11/13] bpf, x64: fix stack layout of JITed bpf code
Since commit 177366bf7ceb the %rbp stopped pointing to %rbp of the
previous stack frame. That broke frame pointer based stack unwinding.
This commit is a partial revert of it.
Note that the location of tail_call_cnt is fixed, since the verifier
enforces MAX_BPF_STACK stack size for programs with tail calls.
Fixes: 177366bf7ceb ("bpf: change x86 JITed program stack layout")
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
arch/x86/net/bpf_jit_comp.c | 74 +++++++++++--------------------------
1 file changed, 21 insertions(+), 53 deletions(-)
diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c
index afabf597c8557..d88bc09358864 100644
--- a/arch/x86/net/bpf_jit_comp.c
+++ b/arch/x86/net/bpf_jit_comp.c
@@ -190,9 +190,7 @@ struct jit_context {
#define BPF_MAX_INSN_SIZE 128
#define BPF_INSN_SAFETY 64
-#define AUX_STACK_SPACE 40 /* Space for RBX, R13, R14, R15, tailcnt */
-
-#define PROLOGUE_SIZE 37
+#define PROLOGUE_SIZE 20
/*
* Emit x86-64 prologue code for BPF program and check its size.
@@ -203,44 +201,19 @@ static void emit_prologue(u8 **pprog, u32 stack_depth, bool ebpf_from_cbpf)
u8 *prog = *pprog;
int cnt = 0;
- /* push rbp */
- EMIT1(0x55);
-
- /* mov rbp,rsp */
- EMIT3(0x48, 0x89, 0xE5);
-
- /* sub rsp, rounded_stack_depth + AUX_STACK_SPACE */
- EMIT3_off32(0x48, 0x81, 0xEC,
- round_up(stack_depth, 8) + AUX_STACK_SPACE);
-
- /* sub rbp, AUX_STACK_SPACE */
- EMIT4(0x48, 0x83, 0xED, AUX_STACK_SPACE);
-
- /* mov qword ptr [rbp+0],rbx */
- EMIT4(0x48, 0x89, 0x5D, 0);
- /* mov qword ptr [rbp+8],r13 */
- EMIT4(0x4C, 0x89, 0x6D, 8);
- /* mov qword ptr [rbp+16],r14 */
- EMIT4(0x4C, 0x89, 0x75, 16);
- /* mov qword ptr [rbp+24],r15 */
- EMIT4(0x4C, 0x89, 0x7D, 24);
-
+ EMIT1(0x55); /* push rbp */
+ EMIT3(0x48, 0x89, 0xE5); /* mov rbp, rsp */
+ /* sub rsp, rounded_stack_depth */
+ EMIT3_off32(0x48, 0x81, 0xEC, round_up(stack_depth, 8));
+ EMIT1(0x53); /* push rbx */
+ EMIT2(0x41, 0x55); /* push r13 */
+ EMIT2(0x41, 0x56); /* push r14 */
+ EMIT2(0x41, 0x57); /* push r15 */
if (!ebpf_from_cbpf) {
- /*
- * Clear the tail call counter (tail_call_cnt): for eBPF tail
- * calls we need to reset the counter to 0. It's done in two
- * instructions, resetting RAX register to 0, and moving it
- * to the counter location.
- */
-
- /* xor eax, eax */
- EMIT2(0x31, 0xc0);
- /* mov qword ptr [rbp+32], rax */
- EMIT4(0x48, 0x89, 0x45, 32);
-
+ /* zero init tail_call_cnt */
+ EMIT2(0x6a, 0x00);
BUILD_BUG_ON(cnt != PROLOGUE_SIZE);
}
-
*pprog = prog;
}
@@ -285,13 +258,13 @@ static void emit_bpf_tail_call(u8 **pprog)
* if (tail_call_cnt > MAX_TAIL_CALL_CNT)
* goto out;
*/
- EMIT2_off32(0x8B, 0x85, 36); /* mov eax, dword ptr [rbp + 36] */
+ EMIT2_off32(0x8B, 0x85, -36 - MAX_BPF_STACK); /* mov eax, dword ptr [rbp - 548] */
EMIT3(0x83, 0xF8, MAX_TAIL_CALL_CNT); /* cmp eax, MAX_TAIL_CALL_CNT */
#define OFFSET2 (30 + RETPOLINE_RAX_BPF_JIT_SIZE)
EMIT2(X86_JA, OFFSET2); /* ja out */
label2 = cnt;
EMIT3(0x83, 0xC0, 0x01); /* add eax, 1 */
- EMIT2_off32(0x89, 0x85, 36); /* mov dword ptr [rbp + 36], eax */
+ EMIT2_off32(0x89, 0x85, -36 - MAX_BPF_STACK); /* mov dword ptr [rbp -548], eax */
/* prog = array->ptrs[index]; */
EMIT4_off32(0x48, 0x8B, 0x84, 0xD6, /* mov rax, [rsi + rdx * 8 + offsetof(...)] */
@@ -1040,19 +1013,14 @@ xadd: if (is_imm8(insn->off))
seen_exit = true;
/* Update cleanup_addr */
ctx->cleanup_addr = proglen;
- /* mov rbx, qword ptr [rbp+0] */
- EMIT4(0x48, 0x8B, 0x5D, 0);
- /* mov r13, qword ptr [rbp+8] */
- EMIT4(0x4C, 0x8B, 0x6D, 8);
- /* mov r14, qword ptr [rbp+16] */
- EMIT4(0x4C, 0x8B, 0x75, 16);
- /* mov r15, qword ptr [rbp+24] */
- EMIT4(0x4C, 0x8B, 0x7D, 24);
-
- /* add rbp, AUX_STACK_SPACE */
- EMIT4(0x48, 0x83, 0xC5, AUX_STACK_SPACE);
- EMIT1(0xC9); /* leave */
- EMIT1(0xC3); /* ret */
+ if (!bpf_prog_was_classic(bpf_prog))
+ EMIT1(0x5B); /* get rid of tail_call_cnt */
+ EMIT2(0x41, 0x5F); /* pop r15 */
+ EMIT2(0x41, 0x5E); /* pop r14 */
+ EMIT2(0x41, 0x5D); /* pop r13 */
+ EMIT1(0x5B); /* pop rbx */
+ EMIT1(0xC9); /* leave */
+ EMIT1(0xC3); /* ret */
break;
default:
From 85749218e3a6a1ab4fb3c698394f40f07b80be5e Mon Sep 17 00:00:00 2001
From: Arthur Fabre <afabre@cloudflare.com>
Date: Sat, 15 Jun 2019 14:36:27 -0700
Subject: [PATCH 12/13] bpf: Fix out of bounds memory access in bpf_sk_storage
bpf_sk_storage maps use multiple spin locks to reduce contention.
The number of locks to use is determined by the number of possible CPUs.
With only 1 possible CPU, bucket_log == 0, and 2^0 = 1 locks are used.
When updating elements, the correct lock is determined with hash_ptr().
Calling hash_ptr() with 0 bits is undefined behavior, as it does:
x >> (64 - bits)
Using the value results in an out of bounds memory access.
In my case, this manifested itself as a page fault when raw_spin_lock_bh()
is called later, when running the self tests:
./tools/testing/selftests/bpf/test_verifier 773 775
[ 16.366342] BUG: unable to handle page fault for address: ffff8fe7a66f93f8
Force the minimum number of locks to two.
Signed-off-by: Arthur Fabre <afabre@cloudflare.com>
Fixes: 6ac99e8f23d4 ("bpf: Introduce bpf sk local storage")
Acked-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
net/core/bpf_sk_storage.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/net/core/bpf_sk_storage.c b/net/core/bpf_sk_storage.c
index cc9597a877707..d1c4e1f3be2c7 100644
--- a/net/core/bpf_sk_storage.c
+++ b/net/core/bpf_sk_storage.c
@@ -633,7 +633,8 @@ static struct bpf_map *bpf_sk_storage_map_alloc(union bpf_attr *attr)
return ERR_PTR(-ENOMEM);
bpf_map_init_from_attr(&smap->map, attr);
- smap->bucket_log = ilog2(roundup_pow_of_two(num_possible_cpus()));
+ /* Use at least 2 buckets, select_bucket() is undefined behavior with 1 bucket */
+ smap->bucket_log = max_t(u32, 1, ilog2(roundup_pow_of_two(num_possible_cpus())));
nbuckets = 1U << smap->bucket_log;
smap->buckets = kvcalloc(sizeof(*smap->buckets), nbuckets,
GFP_USER | __GFP_NOWARN);
From 9594dc3c7e71b9f52bee1d7852eb3d4e3aea9e99 Mon Sep 17 00:00:00 2001
From: Matt Mullins <mmullins@fb.com>
Date: Tue, 11 Jun 2019 14:53:04 -0700
Subject: [PATCH 13/13] bpf: fix nested bpf tracepoints with per-cpu data
BPF_PROG_TYPE_RAW_TRACEPOINTs can be executed nested on the same CPU, as
they do not increment bpf_prog_active while executing.
This enables three levels of nesting, to support
- a kprobe or raw tp or perf event,
- another one of the above that irq context happens to call, and
- another one in nmi context
(at most one of which may be a kprobe or perf event).
Fixes: 20b9d7ac4852 ("bpf: avoid excessive stack usage for perf_sample_data")
Signed-off-by: Matt Mullins <mmullins@fb.com>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
---
kernel/trace/bpf_trace.c | 100 ++++++++++++++++++++++++++++++++-------
1 file changed, 84 insertions(+), 16 deletions(-)
diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
index f92d6ad5e0806..1c9a4745e596d 100644
--- a/kernel/trace/bpf_trace.c
+++ b/kernel/trace/bpf_trace.c
@@ -410,8 +410,6 @@ static const struct bpf_func_proto bpf_perf_event_read_value_proto = {
.arg4_type = ARG_CONST_SIZE,
};
-static DEFINE_PER_CPU(struct perf_sample_data, bpf_trace_sd);
-
static __always_inline u64
__bpf_perf_event_output(struct pt_regs *regs, struct bpf_map *map,
u64 flags, struct perf_sample_data *sd)
@@ -442,24 +440,50 @@ __bpf_perf_event_output(struct pt_regs *regs, struct bpf_map *map,
return perf_event_output(event, sd, regs);
}
+/*
+ * Support executing tracepoints in normal, irq, and nmi context that each call
+ * bpf_perf_event_output
+ */
+struct bpf_trace_sample_data {
+ struct perf_sample_data sds[3];
+};
+
+static DEFINE_PER_CPU(struct bpf_trace_sample_data, bpf_trace_sds);
+static DEFINE_PER_CPU(int, bpf_trace_nest_level);
BPF_CALL_5(bpf_perf_event_output, struct pt_regs *, regs, struct bpf_map *, map,
u64, flags, void *, data, u64, size)
{
- struct perf_sample_data *sd = this_cpu_ptr(&bpf_trace_sd);
+ struct bpf_trace_sample_data *sds = this_cpu_ptr(&bpf_trace_sds);
+ int nest_level = this_cpu_inc_return(bpf_trace_nest_level);
struct perf_raw_record raw = {
.frag = {
.size = size,
.data = data,
},
};
+ struct perf_sample_data *sd;
+ int err;
- if (unlikely(flags & ~(BPF_F_INDEX_MASK)))
- return -EINVAL;
+ if (WARN_ON_ONCE(nest_level > ARRAY_SIZE(sds->sds))) {
+ err = -EBUSY;
+ goto out;
+ }
+
+ sd = &sds->sds[nest_level - 1];
+
+ if (unlikely(flags & ~(BPF_F_INDEX_MASK))) {
+ err = -EINVAL;
+ goto out;
+ }
perf_sample_data_init(sd, 0, 0);
sd->raw = &raw;
- return __bpf_perf_event_output(regs, map, flags, sd);
+ err = __bpf_perf_event_output(regs, map, flags, sd);
+
+out:
+ this_cpu_dec(bpf_trace_nest_level);
+ return err;
}
static const struct bpf_func_proto bpf_perf_event_output_proto = {
@@ -822,16 +846,48 @@ pe_prog_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
/*
* bpf_raw_tp_regs are separate from bpf_pt_regs used from skb/xdp
* to avoid potential recursive reuse issue when/if tracepoints are added
- * inside bpf_*_event_output, bpf_get_stackid and/or bpf_get_stack
+ * inside bpf_*_event_output, bpf_get_stackid and/or bpf_get_stack.
+ *
+ * Since raw tracepoints run despite bpf_prog_active, support concurrent usage
+ * in normal, irq, and nmi context.
*/
-static DEFINE_PER_CPU(struct pt_regs, bpf_raw_tp_regs);
+struct bpf_raw_tp_regs {
+ struct pt_regs regs[3];
+};
+static DEFINE_PER_CPU(struct bpf_raw_tp_regs, bpf_raw_tp_regs);
+static DEFINE_PER_CPU(int, bpf_raw_tp_nest_level);
+static struct pt_regs *get_bpf_raw_tp_regs(void)
+{
+ struct bpf_raw_tp_regs *tp_regs = this_cpu_ptr(&bpf_raw_tp_regs);
+ int nest_level = this_cpu_inc_return(bpf_raw_tp_nest_level);
+
+ if (WARN_ON_ONCE(nest_level > ARRAY_SIZE(tp_regs->regs))) {
+ this_cpu_dec(bpf_raw_tp_nest_level);
+ return ERR_PTR(-EBUSY);
+ }
+
+ return &tp_regs->regs[nest_level - 1];
+}
+
+static void put_bpf_raw_tp_regs(void)
+{
+ this_cpu_dec(bpf_raw_tp_nest_level);
+}
+
BPF_CALL_5(bpf_perf_event_output_raw_tp, struct bpf_raw_tracepoint_args *, args,
struct bpf_map *, map, u64, flags, void *, data, u64, size)
{
- struct pt_regs *regs = this_cpu_ptr(&bpf_raw_tp_regs);
+ struct pt_regs *regs = get_bpf_raw_tp_regs();
+ int ret;
+
+ if (IS_ERR(regs))
+ return PTR_ERR(regs);
perf_fetch_caller_regs(regs);
- return ____bpf_perf_event_output(regs, map, flags, data, size);
+ ret = ____bpf_perf_event_output(regs, map, flags, data, size);
+
+ put_bpf_raw_tp_regs();
+ return ret;
}
static const struct bpf_func_proto bpf_perf_event_output_proto_raw_tp = {
@@ -848,12 +904,18 @@ static const struct bpf_func_proto bpf_perf_event_output_proto_raw_tp = {
BPF_CALL_3(bpf_get_stackid_raw_tp, struct bpf_raw_tracepoint_args *, args,
struct bpf_map *, map, u64, flags)
{
- struct pt_regs *regs = this_cpu_ptr(&bpf_raw_tp_regs);
+ struct pt_regs *regs = get_bpf_raw_tp_regs();
+ int ret;
+
+ if (IS_ERR(regs))
+ return PTR_ERR(regs);
perf_fetch_caller_regs(regs);
/* similar to bpf_perf_event_output_tp, but pt_regs fetched differently */
- return bpf_get_stackid((unsigned long) regs, (unsigned long) map,
- flags, 0, 0);
+ ret = bpf_get_stackid((unsigned long) regs, (unsigned long) map,
+ flags, 0, 0);
+ put_bpf_raw_tp_regs();
+ return ret;
}
static const struct bpf_func_proto bpf_get_stackid_proto_raw_tp = {
@@ -868,11 +930,17 @@ static const struct bpf_func_proto bpf_get_stackid_proto_raw_tp = {
BPF_CALL_4(bpf_get_stack_raw_tp, struct bpf_raw_tracepoint_args *, args,
void *, buf, u32, size, u64, flags)
{
- struct pt_regs *regs = this_cpu_ptr(&bpf_raw_tp_regs);
+ struct pt_regs *regs = get_bpf_raw_tp_regs();
+ int ret;
+
+ if (IS_ERR(regs))
+ return PTR_ERR(regs);
perf_fetch_caller_regs(regs);
- return bpf_get_stack((unsigned long) regs, (unsigned long) buf,
- (unsigned long) size, flags, 0);
+ ret = bpf_get_stack((unsigned long) regs, (unsigned long) buf,
+ (unsigned long) size, flags, 0);
+ put_bpf_raw_tp_regs();
+ return ret;
}
static const struct bpf_func_proto bpf_get_stack_proto_raw_tp = {