Merge branch 'queue_stack_maps'

Mauricio Vasquez says:

====================
In some applications this is needed have a pool of free elements, for
example the list of free L4 ports in a SNAT.  None of the current maps allow
to do it as it is not possible to get any element without having they key
it is associated to, even if it were possible, the lack of locking mecanishms in
eBPF would do it almost impossible to be implemented without data races.

This patchset implements two new kind of eBPF maps: queue and stack.
Those maps provide to eBPF programs the peek, push and pop operations, and for
userspace applications a new bpf_map_lookup_and_delete_elem() is added.

Signed-off-by: Mauricio Vasquez B <mauricio.vasquez@polito.it>

v2 -> v3:
 - Remove "almost dead code" in syscall.c
 - Remove unnecessary copy_from_user in bpf_map_lookup_and_delete_elem
 - Rebase

v1 -> v2:
 - Put ARG_PTR_TO_UNINIT_MAP_VALUE logic into a separated patch
 - Fix missing __this_cpu_dec & preempt_enable calls in kernel/bpf/syscall.c

RFC v4 -> v1:
 - Remove roundup to power of 2 in memory allocation
 - Remove count and use a free slot to check if queue/stack is empty
 - Use if + assigment for wrapping indexes
 - Fix some minor style issues
 - Squash two patches together

RFC v3 -> RFC v4:
 - Revert renaming of kernel/bpf/stackmap.c
 - Remove restriction on value size
 - Remove len arguments from peek/pop helpers
 - Add new ARG_PTR_TO_UNINIT_MAP_VALUE

RFC v2 -> RFC v3:
 - Return elements by value instead that by reference
 - Implement queue/stack base on array and head + tail indexes
 - Rename stack trace related files to avoid confusion and conflicts

RFC v1 -> RFC v2:
 - Create two separate maps instead of single one + flags
 - Implement bpf_map_lookup_and_delete syscall
 - Support peek operation
 - Define replacement policy through flags in the update() method
 - Add eBPF side tests
====================

Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

10 files changed, 342 insertions, 2 deletions

diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index 5e46f6732781..a2fb333290dc 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -103,6 +103,7 @@ enum bpf_cmd {
 	BPF_BTF_LOAD,
 	BPF_BTF_GET_FD_BY_ID,
 	BPF_TASK_FD_QUERY,
+	BPF_MAP_LOOKUP_AND_DELETE_ELEM,
 };
 
 enum bpf_map_type {
@@ -128,6 +129,8 @@ enum bpf_map_type {
 	BPF_MAP_TYPE_CGROUP_STORAGE,
 	BPF_MAP_TYPE_REUSEPORT_SOCKARRAY,
 	BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE,
+	BPF_MAP_TYPE_QUEUE,
+	BPF_MAP_TYPE_STACK,
 };
 
 enum bpf_prog_type {
@@ -462,6 +465,28 @@ union bpf_attr {
  * 	Return
  * 		0 on success, or a negative error in case of failure.
  *
+ * int bpf_map_push_elem(struct bpf_map *map, const void *value, u64 flags)
+ * 	Description
+ * 		Push an element *value* in *map*. *flags* is one of:
+ *
+ * 		**BPF_EXIST**
+ * 		If the queue/stack is full, the oldest element is removed to
+ * 		make room for this.
+ * 	Return
+ * 		0 on success, or a negative error in case of failure.
+ *
+ * int bpf_map_pop_elem(struct bpf_map *map, void *value)
+ * 	Description
+ * 		Pop an element from *map*.
+ * Return
+ * 		0 on success, or a negative error in case of failure.
+ *
+ * int bpf_map_peek_elem(struct bpf_map *map, void *value)
+ * 	Description
+ * 		Get an element from *map* without removing it.
+ * Return
+ * 		0 on success, or a negative error in case of failure.
+ *
  * int bpf_probe_read(void *dst, u32 size, const void *src)
  * 	Description
  * 		For tracing programs, safely attempt to read *size* bytes from
@@ -2303,7 +2328,10 @@ union bpf_attr {
 	FN(skb_ancestor_cgroup_id),	\
 	FN(sk_lookup_tcp),		\
 	FN(sk_lookup_udp),		\
-	FN(sk_release),
+	FN(sk_release),			\
+	FN(map_push_elem),		\
+	FN(map_pop_elem),		\
+	FN(map_peek_elem),
 
 /* integer value in 'imm' field of BPF_CALL instruction selects which helper
  * function eBPF program intends to call
diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c
index d70a255cb05e..03f9bcc4ef50 100644
--- a/tools/lib/bpf/bpf.c
+++ b/tools/lib/bpf/bpf.c
@@ -278,6 +278,18 @@ int bpf_map_lookup_elem(int fd, const void *key, void *value)
 	return sys_bpf(BPF_MAP_LOOKUP_ELEM, &attr, sizeof(attr));
 }
 
+int bpf_map_lookup_and_delete_elem(int fd, const void *key, void *value)
+{
+	union bpf_attr attr;
+
+	bzero(&attr, sizeof(attr));
+	attr.map_fd = fd;
+	attr.key = ptr_to_u64(key);
+	attr.value = ptr_to_u64(value);
+
+	return sys_bpf(BPF_MAP_LOOKUP_AND_DELETE_ELEM, &attr, sizeof(attr));
+}
+
 int bpf_map_delete_elem(int fd, const void *key)
 {
 	union bpf_attr attr;
diff --git a/tools/lib/bpf/bpf.h b/tools/lib/bpf/bpf.h
index 258c3c178333..26a51538213c 100644
--- a/tools/lib/bpf/bpf.h
+++ b/tools/lib/bpf/bpf.h
@@ -99,6 +99,8 @@ LIBBPF_API int bpf_map_update_elem(int fd, const void *key, const void *value,
 				   __u64 flags);
 
 LIBBPF_API int bpf_map_lookup_elem(int fd, const void *key, void *value);
+LIBBPF_API int bpf_map_lookup_and_delete_elem(int fd, const void *key,
+					      void *value);
 LIBBPF_API int bpf_map_delete_elem(int fd, const void *key);
 LIBBPF_API int bpf_map_get_next_key(int fd, const void *key, void *next_key);
 LIBBPF_API int bpf_obj_pin(int fd, const char *pathname);
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index d99dd6fc3fbe..e39dfb4e7970 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -37,7 +37,7 @@ TEST_GEN_FILES = test_pkt_access.o test_xdp.o test_l4lb.o test_tcp_estats.o test
 	test_lwt_seg6local.o sendmsg4_prog.o sendmsg6_prog.o test_lirc_mode2_kern.o \
 	get_cgroup_id_kern.o socket_cookie_prog.o test_select_reuseport_kern.o \
 	test_skb_cgroup_id_kern.o bpf_flow.o netcnt_prog.o \
-	test_sk_lookup_kern.o test_xdp_vlan.o
+	test_sk_lookup_kern.o test_xdp_vlan.o test_queue_map.o test_stack_map.o
 
 # Order correspond to 'make run_tests' order
 TEST_PROGS := test_kmod.sh \
@@ -118,6 +118,9 @@ CLANG_FLAGS = -I. -I./include/uapi -I../../../include/uapi \
 $(OUTPUT)/test_l4lb_noinline.o: CLANG_FLAGS += -fno-inline
 $(OUTPUT)/test_xdp_noinline.o: CLANG_FLAGS += -fno-inline
 
+$(OUTPUT)/test_queue_map.o: test_queue_stack_map.h
+$(OUTPUT)/test_stack_map.o: test_queue_stack_map.h
+
 BTF_LLC_PROBE := $(shell $(LLC) -march=bpf -mattr=help 2>&1 | grep dwarfris)
 BTF_PAHOLE_PROBE := $(shell $(BTF_PAHOLE) --help 2>&1 | grep BTF)
 BTF_OBJCOPY_PROBE := $(shell $(LLVM_OBJCOPY) --help 2>&1 | grep -i 'usage.*llvm')
diff --git a/tools/testing/selftests/bpf/bpf_helpers.h b/tools/testing/selftests/bpf/bpf_helpers.h
index fda8c162d0df..6407a3df0f3b 100644
--- a/tools/testing/selftests/bpf/bpf_helpers.h
+++ b/tools/testing/selftests/bpf/bpf_helpers.h
@@ -16,6 +16,13 @@ static int (*bpf_map_update_elem)(void *map, void *key, void *value,
 	(void *) BPF_FUNC_map_update_elem;
 static int (*bpf_map_delete_elem)(void *map, void *key) =
 	(void *) BPF_FUNC_map_delete_elem;
+static int (*bpf_map_push_elem)(void *map, void *value,
+				unsigned long long flags) =
+	(void *) BPF_FUNC_map_push_elem;
+static int (*bpf_map_pop_elem)(void *map, void *value) =
+	(void *) BPF_FUNC_map_pop_elem;
+static int (*bpf_map_peek_elem)(void *map, void *value) =
+	(void *) BPF_FUNC_map_peek_elem;
 static int (*bpf_probe_read)(void *dst, int size, void *unsafe_ptr) =
 	(void *) BPF_FUNC_probe_read;
 static unsigned long long (*bpf_ktime_get_ns)(void) =
diff --git a/tools/testing/selftests/bpf/test_maps.c b/tools/testing/selftests/bpf/test_maps.c
index 9b552c0fc47d..4db2116e52be 100644
--- a/tools/testing/selftests/bpf/test_maps.c
+++ b/tools/testing/selftests/bpf/test_maps.c
@@ -15,6 +15,7 @@
 #include <string.h>
 #include <assert.h>
 #include <stdlib.h>
+#include <time.h>
 
 #include <sys/wait.h>
 #include <sys/socket.h>
@@ -471,6 +472,122 @@ static void test_devmap(int task, void *data)
 	close(fd);
 }
 
+static void test_queuemap(int task, void *data)
+{
+	const int MAP_SIZE = 32;
+	__u32 vals[MAP_SIZE + MAP_SIZE/2], val;
+	int fd, i;
+
+	/* Fill test values to be used */
+	for (i = 0; i < MAP_SIZE + MAP_SIZE/2; i++)
+		vals[i] = rand();
+
+	/* Invalid key size */
+	fd = bpf_create_map(BPF_MAP_TYPE_QUEUE, 4, sizeof(val), MAP_SIZE,
+			    map_flags);
+	assert(fd < 0 && errno == EINVAL);
+
+	fd = bpf_create_map(BPF_MAP_TYPE_QUEUE, 0, sizeof(val), MAP_SIZE,
+			    map_flags);
+	/* Queue map does not support BPF_F_NO_PREALLOC */
+	if (map_flags & BPF_F_NO_PREALLOC) {
+		assert(fd < 0 && errno == EINVAL);
+		return;
+	}
+	if (fd < 0) {
+		printf("Failed to create queuemap '%s'!\n", strerror(errno));
+		exit(1);
+	}
+
+	/* Push MAP_SIZE elements */
+	for (i = 0; i < MAP_SIZE; i++)
+		assert(bpf_map_update_elem(fd, NULL, &vals[i], 0) == 0);
+
+	/* Check that element cannot be pushed due to max_entries limit */
+	assert(bpf_map_update_elem(fd, NULL, &val, 0) == -1 &&
+	       errno == E2BIG);
+
+	/* Peek element */
+	assert(bpf_map_lookup_elem(fd, NULL, &val) == 0 && val == vals[0]);
+
+	/* Replace half elements */
+	for (i = MAP_SIZE; i < MAP_SIZE + MAP_SIZE/2; i++)
+		assert(bpf_map_update_elem(fd, NULL, &vals[i], BPF_EXIST) == 0);
+
+	/* Pop all elements */
+	for (i = MAP_SIZE/2; i < MAP_SIZE + MAP_SIZE/2; i++)
+		assert(bpf_map_lookup_and_delete_elem(fd, NULL, &val) == 0 &&
+		       val == vals[i]);
+
+	/* Check that there are not elements left */
+	assert(bpf_map_lookup_and_delete_elem(fd, NULL, &val) == -1 &&
+	       errno == ENOENT);
+
+	/* Check that non supported functions set errno to EINVAL */
+	assert(bpf_map_delete_elem(fd, NULL) == -1 && errno == EINVAL);
+	assert(bpf_map_get_next_key(fd, NULL, NULL) == -1 && errno == EINVAL);
+
+	close(fd);
+}
+
+static void test_stackmap(int task, void *data)
+{
+	const int MAP_SIZE = 32;
+	__u32 vals[MAP_SIZE + MAP_SIZE/2], val;
+	int fd, i;
+
+	/* Fill test values to be used */
+	for (i = 0; i < MAP_SIZE + MAP_SIZE/2; i++)
+		vals[i] = rand();
+
+	/* Invalid key size */
+	fd = bpf_create_map(BPF_MAP_TYPE_STACK, 4, sizeof(val), MAP_SIZE,
+			    map_flags);
+	assert(fd < 0 && errno == EINVAL);
+
+	fd = bpf_create_map(BPF_MAP_TYPE_STACK, 0, sizeof(val), MAP_SIZE,
+			    map_flags);
+	/* Stack map does not support BPF_F_NO_PREALLOC */
+	if (map_flags & BPF_F_NO_PREALLOC) {
+		assert(fd < 0 && errno == EINVAL);
+		return;
+	}
+	if (fd < 0) {
+		printf("Failed to create stackmap '%s'!\n", strerror(errno));
+		exit(1);
+	}
+
+	/* Push MAP_SIZE elements */
+	for (i = 0; i < MAP_SIZE; i++)
+		assert(bpf_map_update_elem(fd, NULL, &vals[i], 0) == 0);
+
+	/* Check that element cannot be pushed due to max_entries limit */
+	assert(bpf_map_update_elem(fd, NULL, &val, 0) == -1 &&
+	       errno == E2BIG);
+
+	/* Peek element */
+	assert(bpf_map_lookup_elem(fd, NULL, &val) == 0 && val == vals[i - 1]);
+
+	/* Replace half elements */
+	for (i = MAP_SIZE; i < MAP_SIZE + MAP_SIZE/2; i++)
+		assert(bpf_map_update_elem(fd, NULL, &vals[i], BPF_EXIST) == 0);
+
+	/* Pop all elements */
+	for (i = MAP_SIZE + MAP_SIZE/2 - 1; i >= MAP_SIZE/2; i--)
+		assert(bpf_map_lookup_and_delete_elem(fd, NULL, &val) == 0 &&
+		       val == vals[i]);
+
+	/* Check that there are not elements left */
+	assert(bpf_map_lookup_and_delete_elem(fd, NULL, &val) == -1 &&
+	       errno == ENOENT);
+
+	/* Check that non supported functions set errno to EINVAL */
+	assert(bpf_map_delete_elem(fd, NULL) == -1 && errno == EINVAL);
+	assert(bpf_map_get_next_key(fd, NULL, NULL) == -1 && errno == EINVAL);
+
+	close(fd);
+}
+
 #include <sys/socket.h>
 #include <sys/ioctl.h>
 #include <arpa/inet.h>
@@ -1434,10 +1551,15 @@ static void run_all_tests(void)
 	test_map_wronly();
 
 	test_reuseport_array();
+
+	test_queuemap(0, NULL);
+	test_stackmap(0, NULL);
 }
 
 int main(void)
 {
+	srand(time(NULL));
+
 	map_flags = 0;
 	run_all_tests();
 
diff --git a/tools/testing/selftests/bpf/test_progs.c b/tools/testing/selftests/bpf/test_progs.c
index e8becca9c521..2d3c04f45530 100644
--- a/tools/testing/selftests/bpf/test_progs.c
+++ b/tools/testing/selftests/bpf/test_progs.c
@@ -1735,8 +1735,105 @@ static void test_reference_tracking()
 	bpf_object__close(obj);
 }
 
+enum {
+	QUEUE,
+	STACK,
+};
+
+static void test_queue_stack_map(int type)
+{
+	const int MAP_SIZE = 32;
+	__u32 vals[MAP_SIZE], duration, retval, size, val;
+	int i, err, prog_fd, map_in_fd, map_out_fd;
+	char file[32], buf[128];
+	struct bpf_object *obj;
+	struct iphdr *iph = (void *)buf + sizeof(struct ethhdr);
+
+	/* Fill test values to be used */
+	for (i = 0; i < MAP_SIZE; i++)
+		vals[i] = rand();
+
+	if (type == QUEUE)
+		strncpy(file, "./test_queue_map.o", sizeof(file));
+	else if (type == STACK)
+		strncpy(file, "./test_stack_map.o", sizeof(file));
+	else
+		return;
+
+	err = bpf_prog_load(file, BPF_PROG_TYPE_SCHED_CLS, &obj, &prog_fd);
+	if (err) {
+		error_cnt++;
+		return;
+	}
+
+	map_in_fd = bpf_find_map(__func__, obj, "map_in");
+	if (map_in_fd < 0)
+		goto out;
+
+	map_out_fd = bpf_find_map(__func__, obj, "map_out");
+	if (map_out_fd < 0)
+		goto out;
+
+	/* Push 32 elements to the input map */
+	for (i = 0; i < MAP_SIZE; i++) {
+		err = bpf_map_update_elem(map_in_fd, NULL, &vals[i], 0);
+		if (err) {
+			error_cnt++;
+			goto out;
+		}
+	}
+
+	/* The eBPF program pushes iph.saddr in the output map,
+	 * pops the input map and saves this value in iph.daddr
+	 */
+	for (i = 0; i < MAP_SIZE; i++) {
+		if (type == QUEUE) {
+			val = vals[i];
+			pkt_v4.iph.saddr = vals[i] * 5;
+		} else if (type == STACK) {
+			val = vals[MAP_SIZE - 1 - i];
+			pkt_v4.iph.saddr = vals[MAP_SIZE - 1 - i] * 5;
+		}
+
+		err = bpf_prog_test_run(prog_fd, 1, &pkt_v4, sizeof(pkt_v4),
+					buf, &size, &retval, &duration);
+		if (err || retval || size != sizeof(pkt_v4) ||
+		    iph->daddr != val)
+			break;
+	}
+
+	CHECK(err || retval || size != sizeof(pkt_v4) || iph->daddr != val,
+	      "bpf_map_pop_elem",
+	      "err %d errno %d retval %d size %d iph->daddr %u\n",
+	      err, errno, retval, size, iph->daddr);
+
+	/* Queue is empty, program should return TC_ACT_SHOT */
+	err = bpf_prog_test_run(prog_fd, 1, &pkt_v4, sizeof(pkt_v4),
+				buf, &size, &retval, &duration);
+	CHECK(err || retval != 2 /* TC_ACT_SHOT */|| size != sizeof(pkt_v4),
+	      "check-queue-stack-map-empty",
+	      "err %d errno %d retval %d size %d\n",
+	      err, errno, retval, size);
+
+	/* Check that the program pushed elements correctly */
+	for (i = 0; i < MAP_SIZE; i++) {
+		err = bpf_map_lookup_and_delete_elem(map_out_fd, NULL, &val);
+		if (err || val != vals[i] * 5)
+			break;
+	}
+
+	CHECK(i != MAP_SIZE && (err || val != vals[i] * 5),
+	      "bpf_map_push_elem", "err %d value %u\n", err, val);
+
+out:
+	pkt_v4.iph.saddr = 0;
+	bpf_object__close(obj);
+}
+
 int main(void)
 {
+	srand(time(NULL));
+
 	jit_enabled = is_jit_enabled();
 
 	test_pkt_access();
@@ -1757,6 +1854,8 @@ int main(void)
 	test_task_fd_query_rawtp();
 	test_task_fd_query_tp();
 	test_reference_tracking();
+	test_queue_stack_map(QUEUE);
+	test_queue_stack_map(STACK);
 
 	printf("Summary: %d PASSED, %d FAILED\n", pass_cnt, error_cnt);
 	return error_cnt ? EXIT_FAILURE : EXIT_SUCCESS;
diff --git a/tools/testing/selftests/bpf/test_queue_map.c b/tools/testing/selftests/bpf/test_queue_map.c
new file mode 100644
index 000000000000..87db1f9da33d
--- /dev/null
+++ b/tools/testing/selftests/bpf/test_queue_map.c
@@ -0,0 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2018 Politecnico di Torino
+#define MAP_TYPE BPF_MAP_TYPE_QUEUE
+#include "test_queue_stack_map.h"
diff --git a/tools/testing/selftests/bpf/test_queue_stack_map.h b/tools/testing/selftests/bpf/test_queue_stack_map.h
new file mode 100644
index 000000000000..295b9b3bc5c7
--- /dev/null
+++ b/tools/testing/selftests/bpf/test_queue_stack_map.h
@@ -0,0 +1,59 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+// Copyright (c) 2018 Politecnico di Torino
+#include <stddef.h>
+#include <string.h>
+#include <linux/bpf.h>
+#include <linux/if_ether.h>
+#include <linux/ip.h>
+#include <linux/pkt_cls.h>
+#include "bpf_helpers.h"
+
+int _version SEC("version") = 1;
+
+struct bpf_map_def __attribute__ ((section("maps"), used)) map_in = {
+	.type = MAP_TYPE,
+	.key_size = 0,
+	.value_size = sizeof(__u32),
+	.max_entries = 32,
+	.map_flags = 0,
+};
+
+struct bpf_map_def __attribute__ ((section("maps"), used)) map_out = {
+	.type = MAP_TYPE,
+	.key_size = 0,
+	.value_size = sizeof(__u32),
+	.max_entries = 32,
+	.map_flags = 0,
+};
+
+SEC("test")
+int _test(struct __sk_buff *skb)
+{
+	void *data_end = (void *)(long)skb->data_end;
+	void *data = (void *)(long)skb->data;
+	struct ethhdr *eth = (struct ethhdr *)(data);
+	__u32 value;
+	int err;
+
+	if (eth + 1 > data_end)
+		return TC_ACT_SHOT;
+
+	struct iphdr *iph = (struct iphdr *)(eth + 1);
+
+	if (iph + 1 > data_end)
+		return TC_ACT_SHOT;
+
+	err = bpf_map_pop_elem(&map_in, &value);
+	if (err)
+		return TC_ACT_SHOT;
+
+	iph->daddr = value;
+
+	err = bpf_map_push_elem(&map_out, &iph->saddr, 0);
+	if (err)
+		return TC_ACT_SHOT;
+
+	return TC_ACT_OK;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/test_stack_map.c b/tools/testing/selftests/bpf/test_stack_map.c
new file mode 100644
index 000000000000..31c3880e6da0
--- /dev/null
+++ b/tools/testing/selftests/bpf/test_stack_map.c
@@ -0,0 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2018 Politecnico di Torino
+#define MAP_TYPE BPF_MAP_TYPE_STACK
+#include "test_queue_stack_map.h"