Skip to contentSkip to conversation

Fix attaching fentry/fexit/fmod_ret/lsm to modules

PATCH Superseded Viktor Malik 2 participants 6 messages Feb 28, 2023 – Mar 9, 2023

Subsystems (7)
Networking Kselftest BPF Arm STM32 +3 more subsystems

v9 of 10 Superseded Latest is v10 →

lore
Share & tools
lore
#0 lore raw

v9 cover Fix attaching fentry/fexit/fmod_ret/lsm to modules

I noticed that the verifier behaves incorrectly when attaching to fentry of multiple functions of the same name located in different modules (or in vmlinux). The reason for this is that if the target program is not specified, the verifier will search kallsyms for the trampoline address to attach to. The entire kallsyms is always searched, not respecting the module in which the function to attach to is located.

As Yonghong correctly pointed out, there is yet another issue - the trampoline acquires the module reference in register_fentry which means that if the module is unloaded between the place where the address is found in the verifier and register_fentry, it is possible that another module is loaded to the same address in the meantime, which may lead to errors.

This patch fixes the above issues by extracting the module name from the BTF of the attachment target (which must be specified) and by doing the search in kallsyms of the correct module. At the same time, the module reference is acquired right after the address is found and only released right before the program itself is unloaded.

---
Changes in v9:
- two small changes suggested by Jiri Olsa and Jiri's ack

Changes in v8:
- added module_put to error paths in bpf_check_attach_target after the module reference is acquired

Changes in v7:
- refactored the module reference manipulation (comments by Jiri Olsa)
- cleaned up the test (comments by Andrii Nakryiko)

Changes in v6:
- storing the module reference inside bpf_prog_aux instead of bpf_trampoline and releasing it when the program is unloaded (suggested by Jiri Olsa)

Changes in v5:
- fixed acquiring and releasing of module references by trampolines to prevent modules being unloaded between address lookup and trampoline allocation

Changes in v4:
- reworked module kallsyms lookup approach using existing functions, verifier now calls btf_try_get_module to retrieve the module and find_kallsyms_symbol_value to get the symbol address (suggested by Alexei)
- included Jiri Olsa's comments
- improved description of the new test and added it as a comment into the test source

Changes in v3:
- added trivial implementation for kallsyms_lookup_name_in_module() for !CONFIG_MODULES (noticed by test robot, fix suggested by Hao Luo)

Changes in v2:
- introduced and used more space-efficient kallsyms lookup function, suggested by Jiri Olsa
- included Hao Luo's comments

Viktor Malik (2):
bpf: Fix attaching fentry/fexit/fmod_ret/lsm to modules bpf/selftests: Test fentry attachment to shadowed functions

 include/linux/bpf.h                           |   2 +
 kernel/bpf/syscall.c                          |   6 +
 kernel/bpf/trampoline.c                       |  28 ----
 kernel/bpf/verifier.c                         |  18 ++-
 kernel/module/internal.h                      |   5 +
 net/bpf/test_run.c                            |   5 +
 .../selftests/bpf/bpf_testmod/bpf_testmod.c   |   6 +
 .../bpf/prog_tests/module_attach_shadow.c     | 128 ++++++++++++++++++
 8 files changed, 169 insertions(+), 29 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/module_attach_shadow.c
Signature · 1 line
2.39.1
#1 lore raw

PATCH v9 1/2 bpf: Fix attaching fentry/fexit/fmod_ret/lsm to modules

This resolves two problems with attachment of fentry/fexit/fmod_ret/lsm to functions located in modules:

1. The verifier tries to find the address to attach to in kallsyms. This is always done by searching the entire kallsyms, not respecting the module in which the function is located. Such approach causes an incorrect attachment address to be computed if the function to attach to is shadowed by a function of the same name located earlier in kallsyms.

2. If the address to attach to is located in a module, the module reference is only acquired in register_fentry. If the module is unloaded between the place where the address is found (bpf_check_attach_target in the verifier) and register_fentry, it is possible that another module is loaded to the same address which may lead to potential errors.

Since the attachment must contain the BTF of the program to attach to, we extract the module from it and search for the function address in the correct module (resolving problem no. 1). Then, the module reference is taken directly in bpf_check_attach_target and stored in the bpf program (in bpf_prog_aux). The reference is only released when the program is unloaded (resolving problem no. 2).

---
 include/linux/bpf.h      |  2 ++
 kernel/bpf/syscall.c     |  6 ++++++
 kernel/bpf/trampoline.c  | 28 ----------------------------
 kernel/bpf/verifier.c    | 18 +++++++++++++++++-
 kernel/module/internal.h |  5 +++++
 5 files changed, 30 insertions(+), 29 deletions(-)
include/linux/bpf.h +2 -0
index 520b238abd5a..ad3333088da5 100644
@@ -1089,6 +1089,7 @@ struct bpf_trampoline {
 struct bpf_attach_target_info {
 	struct btf_func_model fmodel;
 	long tgt_addr;
+	struct module *tgt_mod;
 	const char *tgt_name;
 	const struct btf_type *tgt_type;
 };
@@ -1361,6 +1362,7 @@ struct bpf_prog_aux {
 	 * main prog always has linfo_idx == 0
 	 */
 	u32 linfo_idx;
+	struct module *mod;
 	u32 num_exentries;
 	struct exception_table_entry *extable;
 	union {
kernel/bpf/syscall.c +6 -0
index e3fcdc9836a6..3c9de31c892f 100644
@@ -2049,6 +2049,7 @@ static void __bpf_prog_put_noref(struct bpf_prog *prog, bool deferred)
 {
 	bpf_prog_kallsyms_del_all(prog);
 	btf_put(prog->aux->btf);
+	module_put(prog->aux->mod);
 	kvfree(prog->aux->jited_linfo);
 	kvfree(prog->aux->linfo);
 	kfree(prog->aux->kfunc_tab);
@@ -3095,6 +3096,11 @@ static int bpf_tracing_prog_attach(struct bpf_prog *prog,
 		if (err)
 			goto out_unlock;
 
+		if (tgt_info.tgt_mod) {
+			module_put(prog->aux->mod);
+			prog->aux->mod = tgt_info.tgt_mod;
+		}
+
 		tr = bpf_trampoline_get(key, &tgt_info);
 		if (!tr) {
 			err = -ENOMEM;
kernel/bpf/trampoline.c +0 -28
index d0ed7d6f5eec..f61d5138b12b 100644
@@ -9,7 +9,6 @@
 #include <linux/btf.h>
 #include <linux/rcupdate_trace.h>
 #include <linux/rcupdate_wait.h>
-#include <linux/module.h>
 #include <linux/static_call.h>
 #include <linux/bpf_verifier.h>
 #include <linux/bpf_lsm.h>
@@ -172,26 +171,6 @@ static struct bpf_trampoline *bpf_trampoline_lookup(u64 key)
 	return tr;
 }
 
-static int bpf_trampoline_module_get(struct bpf_trampoline *tr)
-{
-	struct module *mod;
-	int err = 0;
-
-	preempt_disable();
-	mod = __module_text_address((unsigned long) tr->func.addr);
-	if (mod && !try_module_get(mod))
-		err = -ENOENT;
-	preempt_enable();
-	tr->mod = mod;
-	return err;
-}
-
-static void bpf_trampoline_module_put(struct bpf_trampoline *tr)
-{
-	module_put(tr->mod);
-	tr->mod = NULL;
-}
-
 static int unregister_fentry(struct bpf_trampoline *tr, void *old_addr)
 {
 	void *ip = tr->func.addr;
@@ -202,8 +181,6 @@ static int unregister_fentry(struct bpf_trampoline *tr, void *old_addr)
 	else
 		ret = bpf_arch_text_poke(ip, BPF_MOD_CALL, old_addr, NULL);
 
-	if (!ret)
-		bpf_trampoline_module_put(tr);
 	return ret;
 }
 
@@ -238,9 +215,6 @@ static int register_fentry(struct bpf_trampoline *tr, void *new_addr)
 		tr->func.ftrace_managed = true;
 	}
 
-	if (bpf_trampoline_module_get(tr))
-		return -ENOENT;
-
 	if (tr->func.ftrace_managed) {
 		ftrace_set_filter_ip(tr->fops, (unsigned long)ip, 0, 1);
 		ret = register_ftrace_direct_multi(tr->fops, (long)new_addr);
@@ -248,8 +222,6 @@ static int register_fentry(struct bpf_trampoline *tr, void *new_addr)
 		ret = bpf_arch_text_poke(ip, BPF_MOD_CALL, NULL, new_addr);
 	}
 
-	if (ret)
-		bpf_trampoline_module_put(tr);
 	return ret;
 }
 
kernel/bpf/verifier.c +17 -1
index 5cb8b623f639..51104284c1ac 100644
@@ -24,6 +24,7 @@
 #include <linux/bpf_lsm.h>
 #include <linux/btf_ids.h>
 #include <linux/poison.h>
+#include "../module/internal.h"
 
 #include "disasm.h"
 
@@ -17302,6 +17303,7 @@ int bpf_check_attach_target(struct bpf_verifier_log *log,
 	const char *tname;
 	struct btf *btf;
 	long addr = 0;
+	struct module *mod = NULL;
 
 	if (!btf_id) {
 		bpf_log(log, "Tracing programs must provide btf_id\n");
@@ -17475,8 +17477,17 @@ int bpf_check_attach_target(struct bpf_verifier_log *log,
 			else
 				addr = (long) tgt_prog->aux->func[subprog]->bpf_func;
 		} else {
-			addr = kallsyms_lookup_name(tname);
+			if (btf_is_module(btf)) {
+				mod = btf_try_get_module(btf);
+				if (mod)
+					addr = find_kallsyms_symbol_value(mod, tname);
+				else
+					addr = 0;
+			} else {
+				addr = kallsyms_lookup_name(tname);
+			}
 			if (!addr) {
+				module_put(mod);
 				bpf_log(log,
 					"The address of function %s cannot be found\n",
 					tname);
@@ -17516,11 +17527,13 @@ int bpf_check_attach_target(struct bpf_verifier_log *log,
 				break;
 			}
 			if (ret) {
+				module_put(mod);
 				bpf_log(log, "%s is not sleepable\n", tname);
 				return ret;
 			}
 		} else if (prog->expected_attach_type == BPF_MODIFY_RETURN) {
 			if (tgt_prog) {
+				module_put(mod);
 				bpf_log(log, "can't modify return codes of BPF programs\n");
 				return -EINVAL;
 			}
@@ -17529,6 +17542,7 @@ int bpf_check_attach_target(struct bpf_verifier_log *log,
 			    !check_attach_modify_return(addr, tname))
 				ret = 0;
 			if (ret) {
+				module_put(mod);
 				bpf_log(log, "%s() is not modifiable\n", tname);
 				return ret;
 			}
@@ -17539,6 +17553,7 @@ int bpf_check_attach_target(struct bpf_verifier_log *log,
 	tgt_info->tgt_addr = addr;
 	tgt_info->tgt_name = tname;
 	tgt_info->tgt_type = t;
+	tgt_info->tgt_mod = mod;
 	return 0;
 }
 
@@ -17618,6 +17633,7 @@ static int check_attach_btf_id(struct bpf_verifier_env *env)
 	/* store info about the attachment target that will be used later */
 	prog->aux->attach_func_proto = tgt_info.tgt_type;
 	prog->aux->attach_func_name = tgt_info.tgt_name;
+	prog->aux->mod = tgt_info.tgt_mod;
 
 	if (tgt_prog) {
 		prog->aux->saved_dst_prog_type = tgt_prog->type;
kernel/module/internal.h +5 -0
index 2e2bf236f558..5cb103a46018 100644
@@ -256,6 +256,11 @@ static inline bool sect_empty(const Elf_Shdr *sect)
 static inline void init_build_id(struct module *mod, const struct load_info *info) { }
 static inline void layout_symtab(struct module *mod, struct load_info *info) { }
 static inline void add_kallsyms(struct module *mod, const struct load_info *info) { }
+static inline unsigned long find_kallsyms_symbol_value(struct module *mod
+						       const char *name)
+{
+	return 0;
+}
 #endif /* CONFIG_KALLSYMS */
 
 #ifdef CONFIG_SYSFS
Signature · 1 line
2.39.1
#2 lore raw

PATCH v9 2/2 bpf/selftests: Test fentry attachment to shadowed functions

In reply to Viktor Malik

Adds a new test that tries to attach a program to fentry of two functions of the same name, one located in vmlinux and the other in bpf_testmod.

To avoid conflicts with existing tests, a new function "bpf_fentry_shadow_test" was created both in vmlinux and in bpf_testmod.

The previous commit fixed a bug which caused this test to fail. The verifier would always use the vmlinux function's address as the target trampoline address, hence trying to create two trampolines for a single address, which is forbidden.

---
 net/bpf/test_run.c                            |   5 +
 .../selftests/bpf/bpf_testmod/bpf_testmod.c   |   6 +
 .../bpf/prog_tests/module_attach_shadow.c     | 128 ++++++++++++++++++
 3 files changed, 139 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/module_attach_shadow.c
net/bpf/test_run.c +5 -0
index 6f3d654b3339..aa3a4e6ba701 100644
@@ -560,6 +560,11 @@ long noinline bpf_kfunc_call_test4(signed char a, short b, int c, long d)
 	return (long)a + (long)b + (long)c + d;
 }
 
+int noinline bpf_fentry_shadow_test(int a)
+{
+	return a + 1;
+}
+
 struct prog_test_member1 {
 	int a;
 };
tools/testing/selftests/bpf/bpf_testmod/bpf_testmod.c +6 -0
index 46500636d8cd..c478b14fdea1 100644
@@ -229,6 +229,12 @@ static const struct btf_kfunc_id_set bpf_testmod_kfunc_set = {
 	.set   = &bpf_testmod_check_kfunc_ids,
 };
 
+noinline int bpf_fentry_shadow_test(int a)
+{
+	return a + 2;
+}
+EXPORT_SYMBOL_GPL(bpf_fentry_shadow_test);
+
 extern int bpf_fentry_test1(int a);
 
 static int bpf_testmod_init(void)
tools/testing/selftests/bpf/prog_tests/module_attach_shadow.c new file +128 -0

128 lines not on this page · raw message

Signature · 1 line
2.39.1
#3 lore raw
In reply to Viktor Malik
Quoted 15 lines · Viktor Malik

I noticed that the verifier behaves incorrectly when attaching to fentry of multiple functions of the same name located in different modules (or in vmlinux). The reason for this is that if the target program is not specified, the verifier will search kallsyms for the trampoline address to attach to. The entire kallsyms is always searched, not respecting the module in which the function to attach to is located.

As Yonghong correctly pointed out, there is yet another issue - the trampoline acquires the module reference in register_fentry which means that if the module is unloaded between the place where the address is found in the verifier and register_fentry, it is possible that another module is loaded to the same address in the meantime, which may lead to errors.

Quoted 6 lines

This patch fixes the above issues by extracting the module name from the BTF of the attachment target (which must be specified) and by doing the search in kallsyms of the correct module. At the same time, the module reference is acquired right after the address is found and only released right before the program itself is unloaded.

is it expected that your newly added test fails on arm64? See [0]

[0] https://github.com/kernel-patches/bpf/actions/runs/4359596129/jobs/7621687719

Quoted 59 lines

---
Changes in v9:
- two small changes suggested by Jiri Olsa and Jiri's ack

Changes in v8:
- added module_put to error paths in bpf_check_attach_target after the module reference is acquired

Changes in v7:
- refactored the module reference manipulation (comments by Jiri Olsa)
- cleaned up the test (comments by Andrii Nakryiko)

Changes in v6:
- storing the module reference inside bpf_prog_aux instead of bpf_trampoline and releasing it when the program is unloaded (suggested by Jiri Olsa)

Changes in v5:
- fixed acquiring and releasing of module references by trampolines to prevent modules being unloaded between address lookup and trampoline allocation

Changes in v4:
- reworked module kallsyms lookup approach using existing functions, verifier now calls btf_try_get_module to retrieve the module and find_kallsyms_symbol_value to get the symbol address (suggested by Alexei)
- included Jiri Olsa's comments
- improved description of the new test and added it as a comment into the test source

Changes in v3:
- added trivial implementation for kallsyms_lookup_name_in_module() for !CONFIG_MODULES (noticed by test robot, fix suggested by Hao Luo)

Changes in v2:
- introduced and used more space-efficient kallsyms lookup function, suggested by Jiri Olsa
- included Hao Luo's comments

Viktor Malik (2):
bpf: Fix attaching fentry/fexit/fmod_ret/lsm to modules bpf/selftests: Test fentry attachment to shadowed functions

 include/linux/bpf.h                           |   2 +
 kernel/bpf/syscall.c                          |   6 +
 kernel/bpf/trampoline.c                       |  28 ----
 kernel/bpf/verifier.c                         |  18 ++-
 kernel/module/internal.h                      |   5 +
 net/bpf/test_run.c                            |   5 +
 .../selftests/bpf/bpf_testmod/bpf_testmod.c   |   6 +
 .../bpf/prog_tests/module_attach_shadow.c     | 128 ++++++++++++++++++
 8 files changed, 169 insertions(+), 29 deletions(-)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/module_attach_shadow.c

--
2.39.1

#4 lore raw
Quoted 23 lines · Andrii Nakryiko
Quoted 15 lines · Viktor Malik

I noticed that the verifier behaves incorrectly when attaching to fentry of multiple functions of the same name located in different modules (or in vmlinux). The reason for this is that if the target program is not specified, the verifier will search kallsyms for the trampoline address to attach to. The entire kallsyms is always searched, not respecting the module in which the function to attach to is located.

As Yonghong correctly pointed out, there is yet another issue - the trampoline acquires the module reference in register_fentry which means that if the module is unloaded between the place where the address is found in the verifier and register_fentry, it is possible that another module is loaded to the same address in the meantime, which may lead to errors.

Quoted 6 lines

This patch fixes the above issues by extracting the module name from the BTF of the attachment target (which must be specified) and by doing the search in kallsyms of the correct module. At the same time, the module reference is acquired right after the address is found and only released right before the program itself is unloaded.

Quoted 3 lines

is it expected that your newly added test fails on arm64? See [0]

   [0] https://github.com/kernel-patches/bpf/actions/runs/4359596129/jobs/7621687719

I believe so, the test uses fentry and all fentry/fexit tests are failing on arm64 with the same error (524) and are disabled in the CI.

Quoted 61 lines
Quoted 60 lines

---
Changes in v9:
- two small changes suggested by Jiri Olsa and Jiri's ack

Changes in v8:
- added module_put to error paths in bpf_check_attach_target after the module reference is acquired

Changes in v7:
- refactored the module reference manipulation (comments by Jiri Olsa)
- cleaned up the test (comments by Andrii Nakryiko)

Changes in v6:
- storing the module reference inside bpf_prog_aux instead of bpf_trampoline and releasing it when the program is unloaded (suggested by Jiri Olsa)

Changes in v5:
- fixed acquiring and releasing of module references by trampolines to prevent modules being unloaded between address lookup and trampoline allocation

Changes in v4:
- reworked module kallsyms lookup approach using existing functions, verifier now calls btf_try_get_module to retrieve the module and find_kallsyms_symbol_value to get the symbol address (suggested by Alexei)
- included Jiri Olsa's comments
- improved description of the new test and added it as a comment into the test source

Changes in v3:
- added trivial implementation for kallsyms_lookup_name_in_module() for !CONFIG_MODULES (noticed by test robot, fix suggested by Hao Luo)

Changes in v2:
- introduced and used more space-efficient kallsyms lookup function, suggested by Jiri Olsa
- included Hao Luo's comments

Viktor Malik (2):
bpf: Fix attaching fentry/fexit/fmod_ret/lsm to modules bpf/selftests: Test fentry attachment to shadowed functions

  include/linux/bpf.h                           |   2 +
  kernel/bpf/syscall.c                          |   6 +
  kernel/bpf/trampoline.c                       |  28 ----
  kernel/bpf/verifier.c                         |  18 ++-
  kernel/module/internal.h                      |   5 +
  net/bpf/test_run.c                            |   5 +
  .../selftests/bpf/bpf_testmod/bpf_testmod.c   |   6 +
  .../bpf/prog_tests/module_attach_shadow.c     | 128 ++++++++++++++++++
  8 files changed, 169 insertions(+), 29 deletions(-)

create mode 100644 tools/testing/selftests/bpf/prog_tests/module_attach_shadow.c

--
2.39.1

#5 lore raw
Quoted 25 lines · Viktor Malik
Quoted 23 lines · Andrii Nakryiko
Quoted 21 lines · Viktor Malik

I noticed that the verifier behaves incorrectly when attaching to fentry of multiple functions of the same name located in different modules (or in vmlinux). The reason for this is that if the target program is not specified, the verifier will search kallsyms for the trampoline address to attach to. The entire kallsyms is always searched, not respecting the module in which the function to attach to is located.

As Yonghong correctly pointed out, there is yet another issue - the trampoline acquires the module reference in register_fentry which means that if the module is unloaded between the place where the address is found in the verifier and register_fentry, it is possible that another module is loaded to the same address in the meantime, which may lead to errors.

This patch fixes the above issues by extracting the module name from the BTF of the attachment target (which must be specified) and by doing the search in kallsyms of the correct module. At the same time, the module reference is acquired right after the address is found and only released right before the program itself is unloaded.

Quoted 6 lines
Quoted 3 lines

is it expected that your newly added test fails on arm64? See [0]

   [0] https://github.com/kernel-patches/bpf/actions/runs/4359596129/jobs/7621687719

I believe so, the test uses fentry and all fentry/fexit tests are failing on arm64 with the same error (524) and are disabled in the CI.

Then you have to add newly added tests into DENYLIST.aarch64

Quoted 63 lines
Quoted 62 lines
Quoted 61 lines

---
Changes in v9:
- two small changes suggested by Jiri Olsa and Jiri's ack

Changes in v8:
- added module_put to error paths in bpf_check_attach_target after the module reference is acquired

Changes in v7:
- refactored the module reference manipulation (comments by Jiri Olsa)
- cleaned up the test (comments by Andrii Nakryiko)

Changes in v6:
- storing the module reference inside bpf_prog_aux instead of bpf_trampoline and releasing it when the program is unloaded (suggested by Jiri Olsa)

Changes in v5:
- fixed acquiring and releasing of module references by trampolines to prevent modules being unloaded between address lookup and trampoline allocation

Changes in v4:
- reworked module kallsyms lookup approach using existing functions, verifier now calls btf_try_get_module to retrieve the module and find_kallsyms_symbol_value to get the symbol address (suggested by Alexei)
- included Jiri Olsa's comments
- improved description of the new test and added it as a comment into the test source

Changes in v3:
- added trivial implementation for kallsyms_lookup_name_in_module() for !CONFIG_MODULES (noticed by test robot, fix suggested by Hao Luo)

Changes in v2:
- introduced and used more space-efficient kallsyms lookup function, suggested by Jiri Olsa
- included Hao Luo's comments

Viktor Malik (2):
bpf: Fix attaching fentry/fexit/fmod_ret/lsm to modules bpf/selftests: Test fentry attachment to shadowed functions

  include/linux/bpf.h                           |   2 +
  kernel/bpf/syscall.c                          |   6 +
  kernel/bpf/trampoline.c                       |  28 ----
  kernel/bpf/verifier.c                         |  18 ++-
  kernel/module/internal.h                      |   5 +
  net/bpf/test_run.c                            |   5 +
  .../selftests/bpf/bpf_testmod/bpf_testmod.c   |   6 +
  .../bpf/prog_tests/module_attach_shadow.c     | 128 ++++++++++++++++++
  8 files changed, 169 insertions(+), 29 deletions(-)

create mode 100644 tools/testing/selftests/bpf/prog_tests/module_attach_shadow.c

--
2.39.1