Merge branch 'Notify user space when a struct_ops object is detached/unregistered'

Kui-Feng Lee says:

====================
The subsystems managing struct_ops objects may need to detach a
struct_ops object due to errors or other reasons. It would be useful
to notify user space programs so that error recovery or logging can be
carried out.

This patch set enables the detach feature for struct_ops links and
send an event to epoll when a link is detached.  Subsystems could call
link->ops->detach() to detach a link and notify user space programs
through epoll.

The signatures of callback functions in "struct bpf_struct_ops" have
been changed as well to pass an extra link argument to
subsystems. Subsystems could detach the links received from reg() and
update() callbacks if there is. This also provides a way that
subsystems can distinguish registrations for an object that has been
registered multiple times for several links.

However, bpf struct_ops maps without BPF_F_LINK have no any link.
Subsystems will receive NULL link pointer for this case.
---
Changes from v6:

 - Fix the missing header at patch 5.

 - Move RCU_INIT_POINTER() back to its original position.

Changes from v5:

 - Change the commit title of the patch for bpftool.

Changes from v4:

 - Change error code for bpf_struct_ops_map_link_update()

 - Always return 0 for bpf_struct_ops_map_link_detach()

 - Hold update_mutex in bpf_struct_ops_link_create()

 - Add a separated instance of file_operations for links supporting
    poll.

 - Fix bpftool for bpf_link_fops_poll.

Changes from v3:

 - Add a comment to explain why holding update_mutex is not necessary
    in bpf_struct_ops_link_create()

 - Use rcu_access_pointer() in bpf_struct_ops_map_link_poll().

Changes from v2:

 - Rephrased commit logs and comments.

 - Addressed some mistakes from patch splitting.

 - Replace mutex with spinlock in bpf_testmod.c to address lockdep
    Splat and simplify the implementation.

 - Fix an argument passing to rcu_dereference_protected().

Changes from v1:

 - Pass a link to reg, unreg, and update callbacks.

 - Provide a function to detach a link from underlying subsystems.

 - Add a kfunc to mimic detachments from subsystems, and provide a
    flexible way to control when to do detachments.

 - Add two tests to detach a link from the subsystem after the refcount
    of the link drops to zero.

v6: https://lore.kernel.org/bpf/20240524223036.318800-1-thinker.li@gmail.com/
v5: https://lore.kernel.org/all/20240523230848.2022072-1-thinker.li@gmail.com/
v4: https://lore.kernel.org/all/20240521225121.770930-1-thinker.li@gmail.com/
v3: https://lore.kernel.org/all/20240510002942.1253354-1-thinker.li@gmail.com/
v2: https://lore.kernel.org/all/20240507055600.2382627-1-thinker.li@gmail.com/
v1: https://lore.kernel.org/all/20240429213609.487820-1-thinker.li@gmail.com/
====================

Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>

5 files changed, 78 insertions, 6 deletions

diff --git a/tools/bpf/bpftool/skeleton/pid_iter.bpf.c b/tools/bpf/bpftool/skeleton/pid_iter.bpf.c
index 7bdbcac3cf62..948dde25034e 100644
--- a/tools/bpf/bpftool/skeleton/pid_iter.bpf.c
+++ b/tools/bpf/bpftool/skeleton/pid_iter.bpf.c
@@ -29,6 +29,7 @@ enum bpf_link_type___local {
 };
 
 extern const void bpf_link_fops __ksym;
+extern const void bpf_link_fops_poll __ksym __weak;
 extern const void bpf_map_fops __ksym;
 extern const void bpf_prog_fops __ksym;
 extern const void btf_fops __ksym;
@@ -84,7 +85,11 @@ int iter(struct bpf_iter__task_file *ctx)
 		fops = &btf_fops;
 		break;
 	case BPF_OBJ_LINK:
-		fops = &bpf_link_fops;
+		if (&bpf_link_fops_poll &&
+		    file->f_op == &bpf_link_fops_poll)
+			fops = &bpf_link_fops_poll;
+		else
+			fops = &bpf_link_fops;
 		break;
 	default:
 		return 0;
diff --git a/tools/testing/selftests/bpf/bpf_test_no_cfi/bpf_test_no_cfi.c b/tools/testing/selftests/bpf/bpf_test_no_cfi/bpf_test_no_cfi.c
index b1dd889d5d7d..948eb3962732 100644
--- a/tools/testing/selftests/bpf/bpf_test_no_cfi/bpf_test_no_cfi.c
+++ b/tools/testing/selftests/bpf/bpf_test_no_cfi/bpf_test_no_cfi.c
@@ -22,12 +22,12 @@ static int dummy_init_member(const struct btf_type *t,
 	return 0;
 }
 
-static int dummy_reg(void *kdata)
+static int dummy_reg(void *kdata, struct bpf_link *link)
 {
 	return 0;
 }
 
-static void dummy_unreg(void *kdata)
+static void dummy_unreg(void *kdata, struct bpf_link *link)
 {
 }
 
diff --git a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
index 2a18bd320e92..0a09732cde4b 100644
--- a/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
+++ b/tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c
@@ -820,7 +820,7 @@ static const struct bpf_verifier_ops bpf_testmod_verifier_ops = {
 	.is_valid_access = bpf_testmod_ops_is_valid_access,
 };
 
-static int bpf_dummy_reg(void *kdata)
+static int bpf_dummy_reg(void *kdata, struct bpf_link *link)
 {
 	struct bpf_testmod_ops *ops = kdata;
 
@@ -835,7 +835,7 @@ static int bpf_dummy_reg(void *kdata)
 	return 0;
 }
 
-static void bpf_dummy_unreg(void *kdata)
+static void bpf_dummy_unreg(void *kdata, struct bpf_link *link)
 {
 }
 
@@ -871,7 +871,7 @@ struct bpf_struct_ops bpf_bpf_testmod_ops = {
 	.owner = THIS_MODULE,
 };
 
-static int bpf_dummy_reg2(void *kdata)
+static int bpf_dummy_reg2(void *kdata, struct bpf_link *link)
 {
 	struct bpf_testmod_ops2 *ops = kdata;
 
diff --git a/tools/testing/selftests/bpf/prog_tests/test_struct_ops_module.c b/tools/testing/selftests/bpf/prog_tests/test_struct_ops_module.c
index 29e183a80f49..bbcf12696a6b 100644
--- a/tools/testing/selftests/bpf/prog_tests/test_struct_ops_module.c
+++ b/tools/testing/selftests/bpf/prog_tests/test_struct_ops_module.c
@@ -3,9 +3,12 @@
 #include <test_progs.h>
 #include <time.h>
 
+#include <sys/epoll.h>
+
 #include "struct_ops_module.skel.h"
 #include "struct_ops_nulled_out_cb.skel.h"
 #include "struct_ops_forgotten_cb.skel.h"
+#include "struct_ops_detach.skel.h"
 
 static void check_map_info(struct bpf_map_info *info)
 {
@@ -242,6 +245,58 @@ cleanup:
 	struct_ops_forgotten_cb__destroy(skel);
 }
 
+/* Detach a link from a user space program */
+static void test_detach_link(void)
+{
+	struct epoll_event ev, events[2];
+	struct struct_ops_detach *skel;
+	struct bpf_link *link = NULL;
+	int fd, epollfd = -1, nfds;
+	int err;
+
+	skel = struct_ops_detach__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "struct_ops_detach__open_and_load"))
+		return;
+
+	link = bpf_map__attach_struct_ops(skel->maps.testmod_do_detach);
+	if (!ASSERT_OK_PTR(link, "attach_struct_ops"))
+		goto cleanup;
+
+	fd = bpf_link__fd(link);
+	if (!ASSERT_GE(fd, 0, "link_fd"))
+		goto cleanup;
+
+	epollfd = epoll_create1(0);
+	if (!ASSERT_GE(epollfd, 0, "epoll_create1"))
+		goto cleanup;
+
+	ev.events = EPOLLHUP;
+	ev.data.fd = fd;
+	err = epoll_ctl(epollfd, EPOLL_CTL_ADD, fd, &ev);
+	if (!ASSERT_OK(err, "epoll_ctl"))
+		goto cleanup;
+
+	err = bpf_link__detach(link);
+	if (!ASSERT_OK(err, "detach_link"))
+		goto cleanup;
+
+	/* Wait for EPOLLHUP */
+	nfds = epoll_wait(epollfd, events, 2, 500);
+	if (!ASSERT_EQ(nfds, 1, "epoll_wait"))
+		goto cleanup;
+
+	if (!ASSERT_EQ(events[0].data.fd, fd, "epoll_wait_fd"))
+		goto cleanup;
+	if (!ASSERT_TRUE(events[0].events & EPOLLHUP, "events[0].events"))
+		goto cleanup;
+
+cleanup:
+	if (epollfd >= 0)
+		close(epollfd);
+	bpf_link__destroy(link);
+	struct_ops_detach__destroy(skel);
+}
+
 void serial_test_struct_ops_module(void)
 {
 	if (test__start_subtest("struct_ops_load"))
@@ -254,5 +309,7 @@ void serial_test_struct_ops_module(void)
 		test_struct_ops_nulled_out_cb();
 	if (test__start_subtest("struct_ops_forgotten_cb"))
 		test_struct_ops_forgotten_cb();
+	if (test__start_subtest("test_detach_link"))
+		test_detach_link();
 }
 
diff --git a/tools/testing/selftests/bpf/progs/struct_ops_detach.c b/tools/testing/selftests/bpf/progs/struct_ops_detach.c
new file mode 100644
index 000000000000..56b787a89876
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/struct_ops_detach.c
@@ -0,0 +1,10 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2024 Meta Platforms, Inc. and affiliates. */
+#include <vmlinux.h>
+#include <bpf/bpf_helpers.h>
+#include "../bpf_testmod/bpf_testmod.h"
+
+char _license[] SEC("license") = "GPL";
+
+SEC(".struct_ops.link")
+struct bpf_testmod_ops testmod_do_detach;