Skip to contentSkip to conversation

bpf: Add support for XDP programs in DEVMAP entries

PATCH Superseded David Ahern 4 participants 20 messages May 27, 2020

Subsystems (9)

v1 of 4 Superseded Latest is v4 →

lore
Share & tools
lore
#0 lore raw

cover bpf: Add support for XDP programs in DEVMAP entries

Implementation of Daniel's proposal for allowing DEVMAP entries to be a device index, program fd pair.

Programs are run after XDP_REDIRECT and have access to both Rx device and Tx device.

v1
- fixed prog put on invalid program - Toke
- changed write value from id to fd per Toke's comments about capabilities
- add test cases

David Ahern (5):
bpf: Handle 8-byte values in DEVMAP and DEVMAP_HASH
bpf: Add support to attach bpf program to a devmap entry
xdp: Add xdp_txq_info to xdp_buff
bpftool: Add SEC name for xdp programs attached to device map
selftest: Add tests for XDP programs in devmap entries

 include/linux/bpf.h                           |   5 +
 include/net/xdp.h                             |   5 +
 include/uapi/linux/bpf.h                      |   3 +
 kernel/bpf/devmap.c                           | 143 +++++++++++++++---
 net/core/dev.c                                |  18 +++
 net/core/filter.c                             |  17 +++
 tools/include/uapi/linux/bpf.h                |   3 +
 tools/lib/bpf/libbpf.c                        |   2 +
 .../bpf/prog_tests/xdp_devmap_attach.c        | 101 +++++++++++++
 .../selftests/bpf/progs/test_xdp_devmap.c     |  19 +++
 .../bpf/progs/test_xdp_with_devmap.c          |  17 +++
 11 files changed, 315 insertions(+), 18 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/xdp_devmap_attach.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_xdp_devmap.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_xdp_with_devmap.c
Signature · 1 line
2.21.1 (Apple Git-122.3)
#9 lore raw

PATCH 2/5 bpf: Add support to attach bpf program to a devmap entry

Add BPF_XDP_DEVMAP attach type for use with programs associated with a DEVMAP entry.

DEVMAPs can associate a program with a device entry by setting the value to <index, fd> pair. The program associated with the fd must have type XDP with expected attach type BPF_XDP_DEVMAP. When a program is associated with a device index, the program is run on an XDP_REDIRECT and before the buffer is added to the per-cpu queue. At this point rxq data is still valid; the next patch adds tx device information allowing the prorgam to see both ingress and egress device indices.

XDP generic is skb based and XDP programs do not work with skb's. Block the use case by walking maps used by a program that is to be attached via xdpgeneric and fail if any of them are DEVMAP / DEVMAP_HASH with 8-bytes values.

Block attach of BPF_XDP_DEVMAP programs to devices.

---
 include/linux/bpf.h            |  5 ++
 include/uapi/linux/bpf.h       |  1 +
 kernel/bpf/devmap.c            | 92 ++++++++++++++++++++++++++++++----
 net/core/dev.c                 | 18 +++++++
 tools/include/uapi/linux/bpf.h |  1 +
 5 files changed, 108 insertions(+), 9 deletions(-)
include/linux/bpf.h +5 -0
index efe8836b5c48..088751bc09aa 100644
@@ -1242,6 +1242,7 @@ int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
 		    struct net_device *dev_rx);
 int dev_map_generic_redirect(struct bpf_dtab_netdev *dst, struct sk_buff *skb,
 			     struct bpf_prog *xdp_prog);
+bool dev_map_can_have_prog(struct bpf_map *map);
 
 struct bpf_cpu_map_entry *__cpu_map_lookup_elem(struct bpf_map *map, u32 key);
 void __cpu_map_flush(void);
@@ -1355,6 +1356,10 @@ static inline struct net_device  *__dev_map_hash_lookup_elem(struct bpf_map *map
 {
 	return NULL;
 }
+static inline bool dev_map_can_have_prog(struct bpf_map *map)
+{
+	return false;
+}
 
 static inline void __dev_flush(void)
 {
include/uapi/linux/bpf.h +1 -0
index 97e1fd19ff58..8c2c0d0c9a0e 100644
@@ -224,6 +224,7 @@ enum bpf_attach_type {
 	BPF_CGROUP_INET6_GETPEERNAME,
 	BPF_CGROUP_INET4_GETSOCKNAME,
 	BPF_CGROUP_INET6_GETSOCKNAME,
+	BPF_XDP_DEVMAP,
 	__MAX_BPF_ATTACH_TYPE
 };
 
kernel/bpf/devmap.c +83 -9

191 lines not on this page · raw message

net/core/dev.c +18 -0
index ae37586f6ee8..10684833f864 100644
@@ -5420,6 +5420,18 @@ static int generic_xdp_install(struct net_device *dev, struct netdev_bpf *xdp)
 	struct bpf_prog *new = xdp->prog;
 	int ret = 0;
 
+	if (new) {
+		u32 i;
+
+		/* generic XDP does not work with DEVMAPs that can
+		 * have a bpf_prog installed on an entry
+		 */
+		for (i = 0; i < new->aux->used_map_cnt; i++) {
+			if (dev_map_can_have_prog(new->aux->used_maps[i]))
+				return -EINVAL;
+		}
+	}
+
 	switch (xdp->command) {
 	case XDP_SETUP_PROG:
 		rcu_assign_pointer(dev->xdp_prog, new);
@@ -8835,6 +8847,12 @@ int dev_change_xdp_fd(struct net_device *dev, struct netlink_ext_ack *extack,
 			return -EINVAL;
 		}
 
+		if (prog->expected_attach_type == BPF_XDP_DEVMAP) {
+			NL_SET_ERR_MSG(extack, "BPF_XDP_DEVMAP programs can not be attached to a device");
+			bpf_prog_put(prog);
+			return -EINVAL;
+		}
+
 		/* prog->aux->id may be 0 for orphaned device-bound progs */
 		if (prog->aux->id && prog->aux->id == prog_id) {
 			bpf_prog_put(prog);
tools/include/uapi/linux/bpf.h +1 -0
index 97e1fd19ff58..8c2c0d0c9a0e 100644
@@ -224,6 +224,7 @@ enum bpf_attach_type {
 	BPF_CGROUP_INET6_GETPEERNAME,
 	BPF_CGROUP_INET4_GETSOCKNAME,
 	BPF_CGROUP_INET6_GETSOCKNAME,
+	BPF_XDP_DEVMAP,
 	__MAX_BPF_ATTACH_TYPE
 };
 
Signature · 1 line
2.21.1 (Apple Git-122.3)
#19 lore raw

PATCH 5/5 selftest: Add tests for XDP programs in devmap entries

In reply to David Ahern

Add tests to verify ability to add an XDP program to a entry in a DEVMAP.

Add negative tests to show DEVMAP programs can not be attached to devices as a normal XDP program, and accesses to egress_ifindex require BPF_XDP_DEVMAP attach type.

---
 .../bpf/prog_tests/xdp_devmap_attach.c        | 101 ++++++++++++++++++
 .../selftests/bpf/progs/test_xdp_devmap.c     |  19 ++++
 .../bpf/progs/test_xdp_with_devmap.c          |  17 +++
 3 files changed, 137 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/xdp_devmap_attach.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_xdp_devmap.c
 create mode 100644 tools/testing/selftests/bpf/progs/test_xdp_with_devmap.c
tools/testing/selftests/bpf/prog_tests/xdp_devmap_attach.c new file +101 -0
new file mode 100644
index 000000000000..4f608bde9cbe
@@ -0,0 +1,101 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/if_link.h>
+#include <test_progs.h>
+
+#define IFINDEX_LO 1
+
+struct devmap_ext_val {
+	u32 ifindex;    /* must be first for compat with 4-byte values */
+	union {
+		int prog_fd;  /* prog fd on write */
+		u32 prog_id;  /* prog id on read */
+	};
+};
+
+void test_xdp_devmap_attach(void)
+{
+	struct bpf_prog_load_attr attr = {
+		.prog_type = BPF_PROG_TYPE_XDP,
+	};
+	struct bpf_object *obj, *dm_obj = NULL;
+	int err, dm_fd = -1, fd = -1, map_fd;
+	struct bpf_prog_info info = {};
+	struct devmap_ext_val val = {
+		.ifindex = IFINDEX_LO,
+	};
+	__u32 id, len = sizeof(info);
+	__u32 duration = 0, idx = 0;
+
+	attr.file = "./test_xdp_with_devmap.o",
+	err = bpf_prog_load_xattr(&attr, &obj, &fd);
+	if (CHECK(err, "load of xdp program with 8-byte devmap",
+		  "err %d errno %d\n", err, errno))
+		return;
+
+	/* can not attach program with DEVMAPs that allow programs */
+	err = bpf_set_link_xdp_fd(IFINDEX_LO, fd, XDP_FLAGS_SKB_MODE);
+	CHECK(err == 0, "Generic attach of program with 8-byte devmap",
+	      "should have failed\n");
+
+	map_fd = bpf_object__find_map_fd_by_name(obj, "dm_ports");
+	if (CHECK(map_fd < 0, "Lookup dm_ports map",
+		  "err %d errno %d\n", err, errno))
+		goto out_close;
+
+	/*
+	 * basic test - load program to be installed in devmap;
+	 * also verifies access to ingress and egress ifindices.
+	 */
+	attr.expected_attach_type = BPF_XDP_DEVMAP;
+	attr.file = "./test_xdp_devmap.o";
+	err = bpf_prog_load_xattr(&attr, &dm_obj, &dm_fd);
+	if (CHECK(err, "Load of BPF_XDP_DEVMAP program",
+		  "err %d errno %d\n", err, errno))
+		goto out_close;
+
+	err = bpf_obj_get_info_by_fd(dm_fd, &info, &len);
+	if (CHECK_FAIL(err))
+		goto out_close;
+
+	val.prog_fd = dm_fd;
+	err = bpf_map_update_elem(map_fd, &idx, &val, 0);
+	CHECK(err, "Add program to devmap entry",
+	      "err %d errno %d\n", err, errno);
+
+	err = bpf_map_lookup_elem(map_fd, &id, &val);
+	CHECK(err, "Read devmap entry",
+	      "err %d errno %d\n", err, errno);
+	CHECK(info.id != val.prog_id, "Expected program id in devmap entry",
+	      "expected %u read %u\n", info.id, val.prog_id);
+
+	/* can not attach BPF_XDP_DEVMAP program to a device */
+	err = bpf_set_link_xdp_fd(IFINDEX_LO, dm_fd, XDP_FLAGS_SKB_MODE);
+	CHECK(err == 0, "Attach of BPF_XDP_DEVMAP program",
+	      "should have failed\n");
+
+	bpf_object__close(dm_obj);
+
+	/* Load of BPF_XDP_DEVMAP as XDP should fail because of egress index */
+	attr.expected_attach_type = 0;
+	err = bpf_prog_load_xattr(&attr, &dm_obj, &dm_fd);
+	if (CHECK(err == 0,
+		  "Load of XDP program accessing egress ifindex without attach type",
+		  "should have failed\n"))
+		bpf_object__close(dm_obj);
+
+	attr.file = "./xdp_dummy.o";
+	err = bpf_prog_load_xattr(&attr, &dm_obj, &dm_fd);
+	if (CHECK_FAIL(err))
+		goto out_close;
+
+	val.ifindex = 1;
+	val.prog_fd = dm_fd;
+	err = bpf_map_update_elem(map_fd, &idx, &val, 0);
+	CHECK(err == 0, "Add non-BPF_XDP_DEVMAP program to devmap entry",
+	      "should have failed\n");
+
+	bpf_object__close(dm_obj);
+
+out_close:
+	bpf_object__close(obj);
+}
tools/testing/selftests/bpf/progs/test_xdp_devmap.c new file +19 -0
new file mode 100644
index 000000000000..64fc2c3cae01
@@ -0,0 +1,19 @@
+// SPDX-License-Identifier: GPL-2.0
+/* program inserted into devmap entry */
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+
+SEC("xdp_devmap_log")
+int xdpdm_devlog(struct xdp_md *ctx)
+{
+	char fmt[] = "devmap redirect: dev %u -> dev %u len %u\n";
+	void *data_end = (void *)(long)ctx->data_end;
+	void *data = (void *)(long)ctx->data;
+	unsigned int len = data_end - data;
+
+	bpf_trace_printk(fmt, sizeof(fmt), ctx->ingress_ifindex, ctx->egress_ifindex, len);
+
+	return XDP_PASS;
+}
+
+char _license[] SEC("license") = "GPL";
tools/testing/selftests/bpf/progs/test_xdp_with_devmap.c new file +17 -0
new file mode 100644
index 000000000000..cb76c9859e80
@@ -0,0 +1,17 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+
+struct bpf_map_def SEC("maps") dm_ports = {
+	.type = BPF_MAP_TYPE_DEVMAP,
+	.key_size = sizeof(__u32),
+	.value_size = sizeof(__u32) * 2,
+	.max_entries = 4,
+};
+
+SEC("xdp_devmap")
+int xdp_with_devmap(struct xdp_md *ctx)
+{
+	return bpf_redirect_map(&dm_ports, 1, 0);
+}
Signature · 1 line
2.21.1 (Apple Git-122.3)
#1 lore raw

PATCH 1/5 bpf: Handle 8-byte values in DEVMAP and DEVMAP_HASH

In reply to David Ahern

Add support to DEVMAP and DEVMAP_HASH to support 8-byte values as a <device index, program id> pair. To do this, a new struct is needed in bpf_dtab_netdev to hold the values to return on lookup.

---
 kernel/bpf/devmap.c | 56 ++++++++++++++++++++++++++++++++++-----------
 1 file changed, 43 insertions(+), 13 deletions(-)
kernel/bpf/devmap.c +43 -13
index a51d9fb7a359..95db6d8beebc 100644
@@ -60,12 +60,22 @@ struct xdp_dev_bulk_queue {
 	unsigned int count;
 };
 
+/* devmap value can be dev index or dev index + prog fd/id */
+struct dev_map_ext_val {
+	u32 ifindex;	/* must be first for compat with 4-byte values */
+	union {
+		int prog_fd;  /* prog fd on write */
+		u32 prog_id;  /* prog id on read */
+	};
+};
+
 struct bpf_dtab_netdev {
 	struct net_device *dev; /* must be first member, due to tracepoint */
 	struct hlist_node index_hlist;
 	struct bpf_dtab *dtab;
 	struct rcu_head rcu;
 	unsigned int idx;
+	struct dev_map_ext_val val;
 };
 
 struct bpf_dtab {
@@ -108,9 +118,13 @@ static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr)
 	u64 cost = 0;
 	int err;
 
-	/* check sanity of attributes */
+	/* check sanity of attributes. 2 value sizes supported:
+	 * 4 bytes: ifindex
+	 * 8 bytes: ifindex + prog fd
+	 */
 	if (attr->max_entries == 0 || attr->key_size != 4 ||
-	    attr->value_size != 4 || attr->map_flags & ~DEV_CREATE_FLAG_MASK)
+	    (attr->value_size != 4 && attr->value_size != 8) ||
+	    attr->map_flags & ~DEV_CREATE_FLAG_MASK)
 		return -EINVAL;
 
 	/* Lookup returns a pointer straight to dev->ifindex, so make sure the
@@ -472,18 +486,15 @@ int dev_map_generic_redirect(struct bpf_dtab_netdev *dst, struct sk_buff *skb,
 static void *dev_map_lookup_elem(struct bpf_map *map, void *key)
 {
 	struct bpf_dtab_netdev *obj = __dev_map_lookup_elem(map, *(u32 *)key);
-	struct net_device *dev = obj ? obj->dev : NULL;
 
-	return dev ? &dev->ifindex : NULL;
+	return obj ? &obj->val : NULL;
 }
 
 static void *dev_map_hash_lookup_elem(struct bpf_map *map, void *key)
 {
 	struct bpf_dtab_netdev *obj = __dev_map_hash_lookup_elem(map,
 								*(u32 *)key);
-	struct net_device *dev = obj ? obj->dev : NULL;
-
-	return dev ? &dev->ifindex : NULL;
+	return obj ? &obj->val : NULL;
 }
 
 static void __dev_map_entry_free(struct rcu_head *rcu)
@@ -552,15 +563,18 @@ static struct bpf_dtab_netdev *__dev_map_alloc_node(struct net *net,
 		return ERR_PTR(-ENOMEM);
 
 	dev->dev = dev_get_by_index(net, ifindex);
-	if (!dev->dev) {
-		kfree(dev);
-		return ERR_PTR(-EINVAL);
-	}
+	if (!dev->dev)
+		goto err_out;
 
 	dev->idx = idx;
 	dev->dtab = dtab;
 
+	dev->val.ifindex = ifindex;
+
 	return dev;
+err_out:
+	kfree(dev);
+	return ERR_PTR(-EINVAL);
 }
 
 static int __dev_map_update_elem(struct net *net, struct bpf_map *map,
@@ -568,8 +582,16 @@ static int __dev_map_update_elem(struct net *net, struct bpf_map *map,
 {
 	struct bpf_dtab *dtab = container_of(map, struct bpf_dtab, map);
 	struct bpf_dtab_netdev *dev, *old_dev;
-	u32 ifindex = *(u32 *)value;
 	u32 i = *(u32 *)key;
+	u32 ifindex;
+
+	if (map->value_size == 4) {
+		ifindex = *(u32 *)value;
+	} else {
+		struct dev_map_ext_val *val = value;
+
+		ifindex = val->ifindex;
+	}
 
 	if (unlikely(map_flags > BPF_EXIST))
 		return -EINVAL;
@@ -609,10 +631,18 @@ static int __dev_map_hash_update_elem(struct net *net, struct bpf_map *map,
 {
 	struct bpf_dtab *dtab = container_of(map, struct bpf_dtab, map);
 	struct bpf_dtab_netdev *dev, *old_dev;
-	u32 ifindex = *(u32 *)value;
 	u32 idx = *(u32 *)key;
 	unsigned long flags;
 	int err = -EEXIST;
+	u32 ifindex;
+
+	if (map->value_size == 4) {
+		ifindex = *(u32 *)value;
+	} else {
+		struct dev_map_ext_val *val = value;
+
+		ifindex = val->ifindex;
+	}
 
 	if (unlikely(map_flags > BPF_EXIST || !ifindex))
 		return -EINVAL;
Signature · 1 line
2.21.1 (Apple Git-122.3)
#14 lore raw

PATCH 4/5 bpftool: Add SEC name for xdp programs attached to device map

In reply to David Ahern

Support SEC("xdp_dm*") as a short cut for loading the program with type BPF_PROG_TYPE_XDP and expected attach type BPF_XDP_DEVMAP.

---
 tools/lib/bpf/libbpf.c | 2 ++
 1 file changed, 2 insertions(+)
tools/lib/bpf/libbpf.c +2 -0
index fa04cbe547ed..563827b260e8 100644
@@ -6657,6 +6657,8 @@ static const struct bpf_sec_def section_defs[] = {
 		.expected_attach_type = BPF_TRACE_ITER,
 		.is_attach_btf = true,
 		.attach_fn = attach_iter),
+	BPF_EAPROG_SEC("xdp_dm",		BPF_PROG_TYPE_XDP,
+						BPF_XDP_DEVMAP),
 	BPF_PROG_SEC("xdp",			BPF_PROG_TYPE_XDP),
 	BPF_PROG_SEC("perf_event",		BPF_PROG_TYPE_PERF_EVENT),
 	BPF_PROG_SEC("lwt_in",			BPF_PROG_TYPE_LWT_IN),
Signature · 1 line
2.21.1 (Apple Git-122.3)
#13 lore raw

PATCH 3/5 xdp: Add xdp_txq_info to xdp_buff

In reply to David Ahern

Add xdp_txq_info as the Tx counterpart to xdp_rxq_info. At the moment only the device is added. Other fields (queue_index) can be added as use cases arise.

From a UAPI perspective, add egress_ifindex to xdp context for bpf programs to see the Tx device.

Update the verifier to only allow accesses to egress_ifindex by XDP programs with BPF_XDP_DEVMAP expected attach type.

---
 include/net/xdp.h              |  5 +++++
 include/uapi/linux/bpf.h       |  2 ++
 kernel/bpf/devmap.c            |  3 +++
 net/core/filter.c              | 17 +++++++++++++++++
 tools/include/uapi/linux/bpf.h |  2 ++
 5 files changed, 29 insertions(+)
include/net/xdp.h +5 -0
index 90f11760bd12..d54022959491 100644
@@ -61,12 +61,17 @@ struct xdp_rxq_info {
 	struct xdp_mem_info mem;
 } ____cacheline_aligned; /* perf critical, avoid false-sharing */
 
+struct xdp_txq_info {
+	struct net_device *dev;
+};
+
 struct xdp_buff {
 	void *data;
 	void *data_end;
 	void *data_meta;
 	void *data_hard_start;
 	struct xdp_rxq_info *rxq;
+	struct xdp_txq_info *txq;
 	u32 frame_sz; /* frame size to deduce data_hard_end/reserved tailroom*/
 };
 
include/uapi/linux/bpf.h +2 -0
index 8c2c0d0c9a0e..264de1484a66 100644
@@ -3624,6 +3624,8 @@ struct xdp_md {
 	/* Below access go through struct xdp_rxq_info */
 	__u32 ingress_ifindex; /* rxq->dev->ifindex */
 	__u32 rx_queue_index;  /* rxq->queue_index  */
+
+	__u32 egress_ifindex;  /* txq->dev->ifindex */
 };
 
 enum sk_action {
kernel/bpf/devmap.c +3 -0
index 7658b3e2e7fc..2fefa7f65d90 100644
@@ -474,8 +474,11 @@ static struct xdp_buff *dev_map_run_prog(struct net_device *dev,
 					 struct xdp_buff *xdp,
 					 struct bpf_prog *xdp_prog)
 {
+	struct xdp_txq_info txq = { .dev = dev };
 	u32 act;
 
+	xdp->txq = &txq;
+
 	act = bpf_prog_run_xdp(xdp_prog, xdp);
 	switch (act) {
 	case XDP_DROP:
net/core/filter.c +17 -0
index bd2853d23b50..199e02a30381 100644
@@ -6994,6 +6994,13 @@ static bool xdp_is_valid_access(int off, int size,
 				const struct bpf_prog *prog,
 				struct bpf_insn_access_aux *info)
 {
+	if (prog->expected_attach_type != BPF_XDP_DEVMAP) {
+		switch (off) {
+		case offsetof(struct xdp_md, egress_ifindex):
+			return false;
+		}
+	}
+
 	if (type == BPF_WRITE) {
 		if (bpf_prog_is_dev_bound(prog->aux)) {
 			switch (off) {
@@ -7942,6 +7949,16 @@ static u32 xdp_convert_ctx_access(enum bpf_access_type type,
 				      offsetof(struct xdp_rxq_info,
 					       queue_index));
 		break;
+	case offsetof(struct xdp_md, egress_ifindex):
+		*insn++ = BPF_LDX_MEM(BPF_FIELD_SIZEOF(struct xdp_buff, txq),
+				      si->dst_reg, si->src_reg,
+				      offsetof(struct xdp_buff, txq));
+		*insn++ = BPF_LDX_MEM(BPF_FIELD_SIZEOF(struct xdp_txq_info, dev),
+				      si->dst_reg, si->dst_reg,
+				      offsetof(struct xdp_txq_info, dev));
+		*insn++ = BPF_LDX_MEM(BPF_W, si->dst_reg, si->dst_reg,
+				      offsetof(struct net_device, ifindex));
+		break;
 	}
 
 	return insn - insn_buf;
tools/include/uapi/linux/bpf.h +2 -0
index 8c2c0d0c9a0e..264de1484a66 100644
@@ -3624,6 +3624,8 @@ struct xdp_md {
 	/* Below access go through struct xdp_rxq_info */
 	__u32 ingress_ifindex; /* rxq->dev->ifindex */
 	__u32 rx_queue_index;  /* rxq->queue_index  */
+
+	__u32 egress_ifindex;  /* txq->dev->ifindex */
 };
 
 enum sk_action {
Signature · 1 line
2.21.1 (Apple Git-122.3)
#10 lore raw

Re: [PATCH bpf-next 2/5] bpf: Add support to attach bpf program to a devmap entry

In reply to David Ahern
Quoted 146 lines · David Ahern

Add BPF_XDP_DEVMAP attach type for use with programs associated with a DEVMAP entry.

DEVMAPs can associate a program with a device entry by setting the value to <index, fd> pair. The program associated with the fd must have type XDP with expected attach type BPF_XDP_DEVMAP. When a program is associated with a device index, the program is run on an XDP_REDIRECT and before the buffer is added to the per-cpu queue. At this point rxq data is still valid; the next patch adds tx device information allowing the prorgam to see both ingress and egress device indices.

XDP generic is skb based and XDP programs do not work with skb's. Block the use case by walking maps used by a program that is to be attached via xdpgeneric and fail if any of them are DEVMAP / DEVMAP_HASH with 8-bytes values.

Block attach of BPF_XDP_DEVMAP programs to devices.

---
 include/linux/bpf.h            |  5 ++
 include/uapi/linux/bpf.h       |  1 +
 kernel/bpf/devmap.c            | 92 ++++++++++++++++++++++++++++++----
 net/core/dev.c                 | 18 +++++++
 tools/include/uapi/linux/bpf.h |  1 +
 5 files changed, 108 insertions(+), 9 deletions(-)
include/linux/bpf.h +5 -0
index efe8836b5c48..088751bc09aa 100644
@@ -1242,6 +1242,7 @@ int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
 		    struct net_device *dev_rx);
 int dev_map_generic_redirect(struct bpf_dtab_netdev *dst, struct sk_buff *skb,
 			     struct bpf_prog *xdp_prog);
+bool dev_map_can_have_prog(struct bpf_map *map);
 
 struct bpf_cpu_map_entry *__cpu_map_lookup_elem(struct bpf_map *map, u32 key);
 void __cpu_map_flush(void);
@@ -1355,6 +1356,10 @@ static inline struct net_device  *__dev_map_hash_lookup_elem(struct bpf_map *map
 {
 	return NULL;
 }
+static inline bool dev_map_can_have_prog(struct bpf_map *map)
+{
+	return false;
+}
 
 static inline void __dev_flush(void)
 {
include/uapi/linux/bpf.h +1 -0
index 97e1fd19ff58..8c2c0d0c9a0e 100644
@@ -224,6 +224,7 @@ enum bpf_attach_type {
 	BPF_CGROUP_INET6_GETPEERNAME,
 	BPF_CGROUP_INET4_GETSOCKNAME,
 	BPF_CGROUP_INET6_GETSOCKNAME,
+	BPF_XDP_DEVMAP,
 	__MAX_BPF_ATTACH_TYPE
 };
 
kernel/bpf/devmap.c +44 -0
index 95db6d8beebc..7658b3e2e7fc 100644
@@ -73,6 +73,7 @@ struct bpf_dtab_netdev {
 	struct net_device *dev; /* must be first member, due to tracepoint */
 	struct hlist_node index_hlist;
 	struct bpf_dtab *dtab;
+	struct bpf_prog *xdp_prog;
 	struct rcu_head rcu;
 	unsigned int idx;
 	struct dev_map_ext_val val;
@@ -231,6 +232,8 @@ static void dev_map_free(struct bpf_map *map)
 
 			hlist_for_each_entry_safe(dev, next, head, index_hlist) {
 				hlist_del_rcu(&dev->index_hlist);
+				if (dev->xdp_prog)
+					bpf_prog_put(dev->xdp_prog);
 				dev_put(dev->dev);
 				kfree(dev);
 			}
@@ -245,6 +248,8 @@ static void dev_map_free(struct bpf_map *map)
 			if (!dev)
 				continue;
 
+			if (dev->xdp_prog)
+				bpf_prog_put(dev->xdp_prog);
 			dev_put(dev->dev);
 			kfree(dev);
 		}
@@ -331,6 +336,16 @@ static int dev_map_hash_get_next_key(struct bpf_map *map, void *key,
 	return -ENOENT;
 }
 
+bool dev_map_can_have_prog(struct bpf_map *map)
+{
+	if ((map->map_type == BPF_MAP_TYPE_DEVMAP ||
+	     map->map_type == BPF_MAP_TYPE_DEVMAP_HASH) &&
+	    map->value_size != 4)
+		return true;
+
+	return false;
+}
+
 static int bq_xmit_all(struct xdp_dev_bulk_queue *bq, u32 flags)
 {
 	struct net_device *dev = bq->dev;
@@ -455,6 +470,35 @@ static inline int __xdp_enqueue(struct net_device *dev, struct xdp_buff *xdp,
 	return bq_enqueue(dev, xdpf, dev_rx);
 }
 
+static struct xdp_buff *dev_map_run_prog(struct net_device *dev,
+					 struct xdp_buff *xdp,
+					 struct bpf_prog *xdp_prog)
+{
+	u32 act;
+
+	act = bpf_prog_run_xdp(xdp_prog, xdp);
+	switch (act) {
+	case XDP_DROP:
+		fallthrough;
+	case XDP_PASS:
+		break;
+	default:
+		bpf_warn_invalid_xdp_action(act);
+		fallthrough;
+	case XDP_ABORTED:
+		trace_xdp_exception(dev, xdp_prog, act);
+		act = XDP_DROP;
+		break;
+	}
+
+	if (act == XDP_DROP) {
+		xdp_return_buff(xdp);
+		xdp = NULL;
+	}
+
+	return xdp;
+}
+
 int dev_xdp_enqueue(struct net_device *dev, struct xdp_buff *xdp,
 		    struct net_device *dev_rx)
 {

kernel/bpf/devmap.c int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp, · line 519

Quoted 10 lines
kernel/bpf/devmap.c +5 -0
@@ -466,6 +510,11 @@ int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
 {
 	struct net_device *dev = dst->dev;
 
+	if (dst->xdp_prog) {
+		xdp = dev_map_run_prog(dev, xdp, dst->xdp_prog);
+		if (!xdp)
+			return 0;
+	}
 	return __xdp_enqueue(dev, xdp, dev_rx);
 }

Did you give any special consideration to where the hook should be? I'm asking because my immediate thought was that it should be on flush (i.e., in bq_xmit_all()), but now that I see this I'm so sure anymore. What were your thoughts around this?

-Toke

#15 lore raw

Re: [PATCH bpf-next 4/5] bpftool: Add SEC name for xdp programs attached to device map

In reply to David Ahern
Quoted 2 lines · David Ahern

Support SEC("xdp_dm*") as a short cut for loading the program with type BPF_PROG_TYPE_XDP and expected attach type BPF_XDP_DEVMAP.

You're not using this in the selftest; shouldn't you be? Also, the prefix should be libbpf: not bpftool:, no?

-Toke

#2 lore raw

Re: [PATCH bpf-next 1/5] bpf: Handle 8-byte values in DEVMAP and DEVMAP_HASH

In reply to David Ahern
Quoted 15 lines · On Tue, 26 May 2020 19:09:01 -0600 David Ahern <[email protected]> wrote:

Add support to DEVMAP and DEVMAP_HASH to support 8-byte values as a <device index, program id> pair. To do this, a new struct is needed in bpf_dtab_netdev to hold the values to return on lookup.

---
 kernel/bpf/devmap.c | 56 ++++++++++++++++++++++++++++++++++-----------
 1 file changed, 43 insertions(+), 13 deletions(-)
kernel/bpf/devmap.c +0 -0
index a51d9fb7a359..95db6d8beebc 100644
@@ -60,12 +60,22 @@ struct xdp_dev_bulk_queue {
 	unsigned int count;

kernel/bpf/devmap.c struct xdp_dev_bulk_queue { · line 70

Quoted 10 lines
kernel/bpf/devmap.c +8 -0
 };
 
+/* devmap value can be dev index or dev index + prog fd/id */
+struct dev_map_ext_val {
+	u32 ifindex;	/* must be first for compat with 4-byte values */
+	union {
+		int prog_fd;  /* prog fd on write */
+		u32 prog_id;  /* prog id on read */
+	};
+};

This smells like a BTF structure.
Andrii BTF does support union's right?

Quoted 12 lines
kernel/bpf/devmap.c +1 -0
 struct bpf_dtab_netdev {
 	struct net_device *dev; /* must be first member, due to tracepoint */
 	struct hlist_node index_hlist;
 	struct bpf_dtab *dtab;
 	struct rcu_head rcu;
 	unsigned int idx;
+	struct dev_map_ext_val val;
 };
 
 struct bpf_dtab {
@@ -108,9 +118,13 @@ static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr)
 	u64 cost = 0;

kernel/bpf/devmap.c static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr) · line 126

Quoted 10 lines
kernel/bpf/devmap.c +5 -2
 	int err;
 
-	/* check sanity of attributes */
+	/* check sanity of attributes. 2 value sizes supported:
+	 * 4 bytes: ifindex
+	 * 8 bytes: ifindex + prog fd
+	 */
 	if (attr->max_entries == 0 || attr->key_size != 4 ||
-	    attr->value_size != 4 || attr->map_flags & ~DEV_CREATE_FLAG_MASK)
+	    (attr->value_size != 4 && attr->value_size != 8) ||

IMHO we really need to leverage BTF here, as I'm sure we need to do more extensions, and this size matching will get more and more unmaintainable.

With BTF in place, dumping the map via bpftool, will also make the fields "self-documenting".

I will try to implement something that uses BTF for this case (and cpumap).

Signature · 4 lines
Best regards,
  Jesper Dangaard Brouer
  MSc.CS, Principal Kernel Engineer at Red Hat
  LinkedIn: http://www.linkedin.com/in/brouer
#3 lore raw

Re: [PATCH bpf-next 1/5] bpf: Handle 8-byte values in DEVMAP and DEVMAP_HASH

Quoted 16 lines · Jesper Dangaard Brouer

kernel/bpf/devmap.c static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr) · line 126

Quoted 12 lines
kernel/bpf/devmap.c +5 -2
@@ -108,9 +118,13 @@ static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr)
 	u64 cost = 0;
 	int err;
 
-	/* check sanity of attributes */
+	/* check sanity of attributes. 2 value sizes supported:
+	 * 4 bytes: ifindex
+	 * 8 bytes: ifindex + prog fd
+	 */
 	if (attr->max_entries == 0 || attr->key_size != 4 ||
-	    attr->value_size != 4 || attr->map_flags & ~DEV_CREATE_FLAG_MASK)
+	    (attr->value_size != 4 && attr->value_size != 8) ||

IMHO we really need to leverage BTF here, as I'm sure we need to do more extensions, and this size matching will get more and more unmaintainable.

Quoted 5 lines

With BTF in place, dumping the map via bpftool, will also make the fields "self-documenting".

I will try to implement something that uses BTF for this case (and cpumap).

as mentioned in a past response, BTF does not make any fields special and this code should not assume it either. You need to know precisely which 4 bytes is the program fd that needs to be looked up, and that AFAIK is beyond the scope of BTF.

#11 lore raw

Re: [PATCH bpf-next 2/5] bpf: Add support to attach bpf program to a devmap entry

In reply to Toke Høiland-Jørgensen
Quoted 4 lines · Toke Høiland-Jørgensen

Did you give any special consideration to where the hook should be? I'm asking because my immediate thought was that it should be on flush (i.e., in bq_xmit_all()), but now that I see this I'm so sure anymore. What were your thoughts around this?

I chose this spot for many reasons:

1. dev_map_enqueue has the bpf_dtab_netdev structure which holds the program

2. programs take xdp_buff, and dev_map_enqueue still has the xdp_buff with the rx information; no need to convert from buff to frame losing rx data, enqueue, back to buff to run program, back to frame to hand off to the driver.

3. no sense enqueuing if the device program drops the frame.

#16 lore raw

Re: [PATCH bpf-next 4/5] bpftool: Add SEC name for xdp programs attached to device map

In reply to Toke Høiland-Jørgensen
Quoted 8 lines · Toke Høiland-Jørgensen
Quoted 2 lines · David Ahern

Support SEC("xdp_dm*") as a short cut for loading the program with type BPF_PROG_TYPE_XDP and expected attach type BPF_XDP_DEVMAP.

You're not using this in the selftest; shouldn't you be? Also, the prefix should be libbpf: not bpftool:, no?

The selftest is exercising kernel APIs - what is allowed and what is not.

Yes, the subject should be libbpf.

#6 lore raw

Re: [PATCH bpf-next 1/5] bpf: Handle 8-byte values in DEVMAP and DEVMAP_HASH

In reply to Jesper Dangaard Brouer
Quoted 5 lines · Jesper Dangaard Brouer

IMHO we really need to leverage BTF here, as I'm sure we need to do more extensions, and this size matching will get more and more unmaintainable.

With BTF in place, dumping the map via bpftool, will also make the fields "self-documenting".

furthermore, the kernel is changing the value - an fd is passed in and an id is returned. I do not see how any of this fits into BTF.

#4 lore raw

Re: [PATCH bpf-next 1/5] bpf: Handle 8-byte values in DEVMAP and DEVMAP_HASH

In reply to David Ahern
Quoted 17 lines · David Ahern
Quoted 16 lines · Jesper Dangaard Brouer

kernel/bpf/devmap.c static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr) · line 126

Quoted 12 lines
kernel/bpf/devmap.c +5 -2
@@ -108,9 +118,13 @@ static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr)
 	u64 cost = 0;
 	int err;
 
-	/* check sanity of attributes */
+	/* check sanity of attributes. 2 value sizes supported:
+	 * 4 bytes: ifindex
+	 * 8 bytes: ifindex + prog fd
+	 */
 	if (attr->max_entries == 0 || attr->key_size != 4 ||
-	    attr->value_size != 4 || attr->map_flags & ~DEV_CREATE_FLAG_MASK)
+	    (attr->value_size != 4 && attr->value_size != 8) ||

IMHO we really need to leverage BTF here, as I'm sure we need to do more extensions, and this size matching will get more and more unmaintainable.

Quoted 8 lines
Quoted 5 lines

With BTF in place, dumping the map via bpftool, will also make the fields "self-documenting".

I will try to implement something that uses BTF for this case (and cpumap).

as mentioned in a past response, BTF does not make any fields special and this code should not assume it either.

Either way you're creating a contract where the kernel says "first four bytes is the ifindex, second four bytes is the fd/id". BTF just makes that explicit, and allows userspace to declare that it agrees this is what the fields should mean. And gives us more flexibility when extending the API later than just adding stuff at the end and looking at the size...

Quoted 2 lines

You need to know precisely which 4 bytes is the program fd that needs to be looked up, and that AFAIK is beyond the scope of BTF.

Exactly - BTF is a way for userspace to explicitly tell the kernel "I am going to put the fd into these four bytes of the value field", instead of the kernel implicitly assuming it's always bytes 5-8.

-Toke

#12 lore raw

Re: [PATCH bpf-next 2/5] bpf: Add support to attach bpf program to a devmap entry

In reply to David Ahern
Quoted 8 lines · David Ahern
Quoted 4 lines · Toke Høiland-Jørgensen

Did you give any special consideration to where the hook should be? I'm asking because my immediate thought was that it should be on flush (i.e., in bq_xmit_all()), but now that I see this I'm so sure anymore. What were your thoughts around this?

I chose this spot for many reasons:

Quoted 8 lines

1. dev_map_enqueue has the bpf_dtab_netdev structure which holds the program

2. programs take xdp_buff, and dev_map_enqueue still has the xdp_buff with the rx information; no need to convert from buff to frame losing rx data, enqueue, back to buff to run program, back to frame to hand off to the driver.

3. no sense enqueuing if the device program drops the frame.

Right, makes sense; thank you for explaining :)

-Toke

#17 lore raw

Re: [PATCH bpf-next 4/5] bpftool: Add SEC name for xdp programs attached to device map

In reply to David Ahern
Quoted 6 lines · David Ahern
Quoted 5 lines · Toke Høiland-Jørgensen
Quoted 2 lines · David Ahern

Support SEC("xdp_dm*") as a short cut for loading the program with type BPF_PROG_TYPE_XDP and expected attach type BPF_XDP_DEVMAP.

Quoted 6 lines
Quoted 3 lines

You're not using this in the selftest; shouldn't you be? Also, the prefix should be libbpf: not bpftool:, no?

The selftest is exercising kernel APIs - what is allowed and what is not.

Sure, but they also de facto serve as example code for features that are not documented anywhere else, so just seemed a bit odd to me that you were not using this to mark the programs.

Anyway, not going to insist if you prefer explicitly setting expected_attach_type...

-Toke

#5 lore raw

Re: [PATCH bpf-next 1/5] bpf: Handle 8-byte values in DEVMAP and DEVMAP_HASH

In reply to Toke Høiland-Jørgensen
Quoted 8 lines · Toke Høiland-Jørgensen

Either way you're creating a contract where the kernel says "first four bytes is the ifindex, second four bytes is the fd/id". BTF just makes that explicit, and allows userspace to declare that it agrees this is what the fields should mean. And gives us more flexibility when extending the API later than just adding stuff at the end and looking at the size...

Quoted 7 lines
Quoted 2 lines

You need to know precisely which 4 bytes is the program fd that needs to be looked up, and that AFAIK is beyond the scope of BTF.

Exactly - BTF is a way for userspace to explicitly tell the kernel "I am going to put the fd into these four bytes of the value field", instead of the kernel implicitly assuming it's always bytes 5-8.

Really, I should define that struct in uapi/linux/bpf.h. The ifindex field has special meaning: the kernel needs to convert it to a net_device. The prog_fd field has special meaning: it should map to a bpf program.

This use case is not in BTF's scope. But, prove me wrong. Ideas are cheap; code is hard. Show me the code that implements your idea.

#7 lore raw

Re: [PATCH bpf-next 1/5] bpf: Handle 8-byte values in DEVMAP and DEVMAP_HASH

In reply to David Ahern
Quoted 9 lines · On Wed, 27 May 2020 08:27:36 -0600 David Ahern <[email protected]> wrote:
Quoted 5 lines · Jesper Dangaard Brouer

IMHO we really need to leverage BTF here, as I'm sure we need to do more extensions, and this size matching will get more and more unmaintainable.

With BTF in place, dumping the map via bpftool, will also make the fields "self-documenting".

furthermore, the kernel is changing the value - an fd is passed in and an id is returned. I do not see how any of this fits into BTF.

It can, as BTF actually support union's (I just tested that).

For the sake of end-users understanding this, I do wonder if it is better to define the struct without the union, and have longer names that will be part of BTF description, e.g (dumped via bpftool):

 struct dev_map_ext_val {
        u32 ifindex;
        int bpf_prog_fd_write;
        u32 bpf_prog_id_read;

};

But a union would also work (also tested via BPF loading and BTF dumpinmg):

 struct dev_map_ext_val {
        u32 ifindex;
        union {
                int bpf_prog_fd_write;
                u32 bpf_prog_id_read;
        };

};

Signature · 4 lines
Best regards,
  Jesper Dangaard Brouer
  MSc.CS, Principal Kernel Engineer at Red Hat
  LinkedIn: http://www.linkedin.com/in/brouer
#18 lore raw

Re: [PATCH bpf-next 4/5] bpftool: Add SEC name for xdp programs attached to device map

In reply to Toke Høiland-Jørgensen
Quoted 15 lines · On Wed, 27 May 2020 17:01:10 +0200 Toke Høiland-Jørgensen <[email protected]>…
Quoted 6 lines · David Ahern
Quoted 5 lines · Toke Høiland-Jørgensen
Quoted 2 lines · David Ahern

Support SEC("xdp_dm*") as a short cut for loading the program with type BPF_PROG_TYPE_XDP and expected attach type BPF_XDP_DEVMAP.

Quoted 6 lines
Quoted 3 lines

You're not using this in the selftest; shouldn't you be? Also, the prefix should be libbpf: not bpftool:, no?

The selftest is exercising kernel APIs - what is allowed and what is not.

Quoted 6 lines

Sure, but they also de facto serve as example code for features that are not documented anywhere else, so just seemed a bit odd to me that you were not using this to mark the programs.

Anyway, not going to insist if you prefer explicitly setting expected_attach_type...

I actually think that it is better to demonstrate that it is possible to set:

attr.expected_attach_type = BPF_XDP_DEVMAP;

Because people will be grepping the source code to find examples ;-)

We could add a comment, that say SEC("xdp_dm") can be used as short cut.

Signature · 4 lines
Best regards,
  Jesper Dangaard Brouer
  MSc.CS, Principal Kernel Engineer at Red Hat
  LinkedIn: http://www.linkedin.com/in/brouer
#8 lore raw

Re: [PATCH bpf-next 1/5] bpf: Handle 8-byte values in DEVMAP and DEVMAP_HASH

In reply to Jesper Dangaard Brouer
Quoted 7 lines · Jesper Dangaard Brouer
Quoted 4 lines · On Wed, 27 May 2020 08:27:36 -0600 David Ahern <[email protected]> wrote:
Quoted 3 lines · Jesper Dangaard Brouer

IMHO we really need to leverage BTF here, as I'm sure we need to do more extensions, and this size matching will get more and more unmaintainable.

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines

With BTF in place, dumping the map via bpftool, will also make the fields "self-documenting".

furthermore, the kernel is changing the value - an fd is passed in and an id is returned. I do not see how any of this fits into BTF.

It can, as BTF actually support union's (I just tested that).

You are trying to force an arbitrary kernel API with BTF, and it adds no value over a static struct definition that grows as new capability is added.

DEVMAPs (and CPUMAP) are not like other maps - the fields in the entries have meaning to *the kernel*. That means the kernel needs to know the fields and what they represent - be it a device index, a program fd converted to an id, a cpuid, or any future extension. If you add 'u32 foo' to this interface, you still need kernel code to understand 'foo' maps to resource 'bar' via lookup function 'f'. You can not do this dynamically. BTF does not make sense here.

Furthermore, if the kernel does not understand a field then it better return an error code - EINVAL so the user knows expected functionality is not supported by that kernel.

Quoted 11 lines

But a union would also work (also tested via BPF loading and BTF dumpinmg):

 struct dev_map_ext_val {
        u32 ifindex;
        union {
                int bpf_prog_fd_write;
                u32 bpf_prog_id_read;
        };

};

I prefer the union and without the verb; comments convey the same meaning. The fd has no meaning beyond the process that created the entry so it does not need to be kept around bloating the interface.

For v2, I moved the struct to uapi/linux/bpf.h and added comments.