From 7472d5a642c94a0ee1882ff3038de72ffe803a01 Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Thu, 27 Jan 2022 07:46:00 -0800 Subject: compiler_types: define __user as __attribute__((btf_type_tag("user"))) The __user attribute is currently mainly used by sparse for type checking. The attribute indicates whether a memory access is in user memory address space or not. Such information is important during tracing kernel internal functions or data structures as accessing user memory often has different mechanisms compared to accessing kernel memory. For example, the perf-probe needs explicit command line specification to indicate a particular argument or string in user-space memory ([1], [2], [3]). Currently, vmlinux BTF is available in kernel with many distributions. If __user attribute information is available in vmlinux BTF, the explicit user memory access information from users will not be necessary as the kernel can figure it out by itself with vmlinux BTF. Besides the above possible use for perf/probe, another use case is for bpf verifier. Currently, for bpf BPF_PROG_TYPE_TRACING type of bpf programs, users can write direct code like p->m1->m2 and "p" could be a function parameter. Without __user information in BTF, the verifier will assume p->m1 accessing kernel memory and will generate normal loads. Let us say "p" actually tagged with __user in the source code. In such cases, p->m1 is actually accessing user memory and direct load is not right and may produce incorrect result. For such cases, bpf_probe_read_user() will be the correct way to read p->m1. To support encoding __user information in BTF, a new attribute __attribute__((btf_type_tag(""))) is implemented in clang ([4]). For example, if we have #define __user __attribute__((btf_type_tag("user"))) during kernel compilation, the attribute "user" information will be preserved in dwarf. After pahole converting dwarf to BTF, __user information will be available in vmlinux BTF. The following is an example with latest upstream clang (clang14) and pahole 1.23: [$ ~] cat test.c #define __user __attribute__((btf_type_tag("user"))) int foo(int __user *arg) { return *arg; } [$ ~] clang -O2 -g -c test.c [$ ~] pahole -JV test.o ... [1] INT int size=4 nr_bits=32 encoding=SIGNED [2] TYPE_TAG user type_id=1 [3] PTR (anon) type_id=2 [4] FUNC_PROTO (anon) return=1 args=(3 arg) [5] FUNC foo type_id=4 [$ ~] You can see for the function argument "int __user *arg", its type is described as PTR -> TYPE_TAG(user) -> INT The kernel can use this information for bpf verification or other use cases. Current btf_type_tag is only supported in clang (>= clang14) and pahole (>= 1.23). gcc support is also proposed and under development ([5]). [1] http://lkml.kernel.org/r/155789874562.26965.10836126971405890891.stgit@devnote2 [2] http://lkml.kernel.org/r/155789872187.26965.4468456816590888687.stgit@devnote2 [3] http://lkml.kernel.org/r/155789871009.26965.14167558859557329331.stgit@devnote2 [4] https://reviews.llvm.org/D111199 [5] https://lore.kernel.org/bpf/0cbeb2fb-1a18-f690-e360-24b1c90c2a91@fb.com/ Signed-off-by: Yonghong Song Link: https://lore.kernel.org/r/20220127154600.652613-1-yhs@fb.com Signed-off-by: Alexei Starovoitov --- lib/Kconfig.debug | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'lib/Kconfig.debug') diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 14b89aa37c5c..6159859769fa 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -325,6 +325,14 @@ config DEBUG_INFO_BTF config PAHOLE_HAS_SPLIT_BTF def_bool $(success, test `$(PAHOLE) --version | sed -E 's/v([0-9]+)\.([0-9]+)/\1\2/'` -ge "119") +config PAHOLE_HAS_BTF_TAG + def_bool $(success, test `$(PAHOLE) --version | sed -E 's/v([0-9]+)\.([0-9]+)/\1\2/'` -ge "123") + depends on CC_IS_CLANG + help + Decide whether pahole emits btf_tag attributes (btf_type_tag and + btf_decl_tag) or not. Currently only clang compiler implements + these attributes, so make the config depend on CC_IS_CLANG. + config DEBUG_INFO_BTF_MODULES def_bool y depends on DEBUG_INFO_BTF && MODULES && PAHOLE_HAS_SPLIT_BTF -- cgit v1.2.3 From 6323c81350b73a4569cf52df85f80273faa64071 Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Tue, 1 Feb 2022 13:56:23 -0700 Subject: lib/Kconfig.debug: Use CONFIG_PAHOLE_VERSION Now that CONFIG_PAHOLE_VERSION exists, use it in the definition of CONFIG_PAHOLE_HAS_SPLIT_BTF and CONFIG_PAHOLE_HAS_BTF_TAG to reduce the amount of duplication across the tree. Signed-off-by: Nathan Chancellor Signed-off-by: Daniel Borkmann Acked-by: Andrii Nakryiko Link: https://lore.kernel.org/bpf/20220201205624.652313-5-nathan@kernel.org --- lib/Kconfig.debug | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'lib/Kconfig.debug') diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 6159859769fa..bd487d480902 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -323,10 +323,10 @@ config DEBUG_INFO_BTF DWARF type info into equivalent deduplicated BTF type info. config PAHOLE_HAS_SPLIT_BTF - def_bool $(success, test `$(PAHOLE) --version | sed -E 's/v([0-9]+)\.([0-9]+)/\1\2/'` -ge "119") + def_bool PAHOLE_VERSION >= 119 config PAHOLE_HAS_BTF_TAG - def_bool $(success, test `$(PAHOLE) --version | sed -E 's/v([0-9]+)\.([0-9]+)/\1\2/'` -ge "123") + def_bool PAHOLE_VERSION >= 123 depends on CC_IS_CLANG help Decide whether pahole emits btf_tag attributes (btf_type_tag and -- cgit v1.2.3 From 42d9b379e3e1790eafb87c799c9edfd0b37a37c7 Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Tue, 1 Feb 2022 13:56:24 -0700 Subject: lib/Kconfig.debug: Allow BTF + DWARF5 with pahole 1.21+ Commit 98cd6f521f10 ("Kconfig: allow explicit opt in to DWARF v5") prevented CONFIG_DEBUG_INFO_DWARF5 from being selected when CONFIG_DEBUG_INFO_BTF is enabled because pahole had issues with clang's DWARF5 info. This was resolved by [1], which is in pahole v1.21. Allow DEBUG_INFO_DWARF5 to be selected with DEBUG_INFO_BTF when using pahole v1.21 or newer. [1]: https://git.kernel.org/pub/scm/devel/pahole/pahole.git/commit/?id=7d8e829f636f47aba2e1b6eda57e74d8e31f733c Signed-off-by: Nathan Chancellor Signed-off-by: Daniel Borkmann Acked-by: Andrii Nakryiko Link: https://lore.kernel.org/bpf/20220201205624.652313-6-nathan@kernel.org --- lib/Kconfig.debug | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'lib/Kconfig.debug') diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index bd487d480902..1555da672275 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -296,7 +296,7 @@ config DEBUG_INFO_DWARF4 config DEBUG_INFO_DWARF5 bool "Generate DWARF Version 5 debuginfo" depends on !CC_IS_CLANG || (CC_IS_CLANG && (AS_IS_LLVM || (AS_IS_GNU && AS_VERSION >= 23502))) - depends on !DEBUG_INFO_BTF + depends on !DEBUG_INFO_BTF || PAHOLE_VERSION >= 121 help Generate DWARF v5 debug info. Requires binutils 2.35.2, gcc 5.0+ (gcc 5.0+ accepts the -gdwarf-5 flag but only had partial support for some -- cgit v1.2.3 From 5e214f2e43e453d862ebbbd2a4f7ee3fe650f209 Mon Sep 17 00:00:00 2001 From: Connor O'Brien Date: Wed, 23 Feb 2022 01:28:14 +0000 Subject: bpf: Add config to allow loading modules with BTF mismatches MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit BTF mismatch can occur for a separately-built module even when the ABI is otherwise compatible and nothing else would prevent successfully loading. Add a new Kconfig to control how mismatches are handled. By default, preserve the current behavior of refusing to load the module. If MODULE_ALLOW_BTF_MISMATCH is enabled, load the module but ignore its BTF information. Suggested-by: Yonghong Song Suggested-by: Michal Suchánek Signed-off-by: Connor O'Brien Signed-off-by: Daniel Borkmann Acked-by: Shung-Hsi Yu Acked-by: Song Liu Link: https://lore.kernel.org/bpf/CAADnVQJ+OVPnBz8z3vNu8gKXX42jCUqfuvhWAyCQDu8N_yqqwQ@mail.gmail.com Link: https://lore.kernel.org/bpf/20220223012814.1898677-1-connoro@google.com --- kernel/bpf/btf.c | 3 ++- lib/Kconfig.debug | 10 ++++++++++ 2 files changed, 12 insertions(+), 1 deletion(-) (limited to 'lib/Kconfig.debug') diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c index 0890e56e8b08..b472cf0c8fdb 100644 --- a/kernel/bpf/btf.c +++ b/kernel/bpf/btf.c @@ -6398,7 +6398,8 @@ static int btf_module_notify(struct notifier_block *nb, unsigned long op, pr_warn("failed to validate module [%s] BTF: %ld\n", mod->name, PTR_ERR(btf)); kfree(btf_mod); - err = PTR_ERR(btf); + if (!IS_ENABLED(CONFIG_MODULE_ALLOW_BTF_MISMATCH)) + err = PTR_ERR(btf); goto out; } err = btf_alloc_id(btf); diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 1555da672275..72ca4684beda 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -339,6 +339,16 @@ config DEBUG_INFO_BTF_MODULES help Generate compact split BTF type information for kernel modules. +config MODULE_ALLOW_BTF_MISMATCH + bool "Allow loading modules with non-matching BTF type info" + depends on DEBUG_INFO_BTF_MODULES + help + For modules whose split BTF does not match vmlinux, load without + BTF rather than refusing to load. The default behavior with + module BTF enabled is to reject modules with such mismatches; + this option will still load module BTF where possible but ignore + it when a mismatch is found. + config GDB_SCRIPTS bool "Provide GDB scripts for kernel debugging" help -- cgit v1.2.3 From f4616fabab39e084b5d7e15a32151d9a9aeab537 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Tue, 15 Mar 2022 23:02:35 +0900 Subject: fprobe: Add a selftest for fprobe Add a KUnit based selftest for fprobe interface. Signed-off-by: Masami Hiramatsu Signed-off-by: Steven Rostedt (Google) Tested-by: Steven Rostedt (Google) Signed-off-by: Alexei Starovoitov Link: https://lore.kernel.org/bpf/164735295554.1084943.18347620679928750960.stgit@devnote2 --- lib/Kconfig.debug | 12 ++++ lib/Makefile | 2 + lib/test_fprobe.c | 174 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 188 insertions(+) create mode 100644 lib/test_fprobe.c (limited to 'lib/Kconfig.debug') diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 72ca4684beda..b0bf0d224b2c 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -2118,6 +2118,18 @@ config KPROBES_SANITY_TEST Say N if you are unsure. +config FPROBE_SANITY_TEST + bool "Self test for fprobe" + depends on DEBUG_KERNEL + depends on FPROBE + depends on KUNIT=y + help + This option will enable testing the fprobe when the system boot. + A series of tests are made to verify that the fprobe is functioning + properly. + + Say N if you are unsure. + config BACKTRACE_SELF_TEST tristate "Self test for the backtrace code" depends on DEBUG_KERNEL diff --git a/lib/Makefile b/lib/Makefile index 300f569c626b..154008764b16 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -103,6 +103,8 @@ obj-$(CONFIG_TEST_HMM) += test_hmm.o obj-$(CONFIG_TEST_FREE_PAGES) += test_free_pages.o obj-$(CONFIG_KPROBES_SANITY_TEST) += test_kprobes.o obj-$(CONFIG_TEST_REF_TRACKER) += test_ref_tracker.o +CFLAGS_test_fprobe.o += $(CC_FLAGS_FTRACE) +obj-$(CONFIG_FPROBE_SANITY_TEST) += test_fprobe.o # # CFLAGS for compiling floating point code inside the kernel. x86/Makefile turns # off the generation of FPU/SSE* instructions for kernel proper but FPU_FLAGS diff --git a/lib/test_fprobe.c b/lib/test_fprobe.c new file mode 100644 index 000000000000..ed70637a2ffa --- /dev/null +++ b/lib/test_fprobe.c @@ -0,0 +1,174 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * test_fprobe.c - simple sanity test for fprobe + */ + +#include +#include +#include +#include + +#define div_factor 3 + +static struct kunit *current_test; + +static u32 rand1, entry_val, exit_val; + +/* Use indirect calls to avoid inlining the target functions */ +static u32 (*target)(u32 value); +static u32 (*target2)(u32 value); +static unsigned long target_ip; +static unsigned long target2_ip; + +static noinline u32 fprobe_selftest_target(u32 value) +{ + return (value / div_factor); +} + +static noinline u32 fprobe_selftest_target2(u32 value) +{ + return (value / div_factor) + 1; +} + +static notrace void fp_entry_handler(struct fprobe *fp, unsigned long ip, struct pt_regs *regs) +{ + KUNIT_EXPECT_FALSE(current_test, preemptible()); + /* This can be called on the fprobe_selftest_target and the fprobe_selftest_target2 */ + if (ip != target_ip) + KUNIT_EXPECT_EQ(current_test, ip, target2_ip); + entry_val = (rand1 / div_factor); +} + +static notrace void fp_exit_handler(struct fprobe *fp, unsigned long ip, struct pt_regs *regs) +{ + unsigned long ret = regs_return_value(regs); + + KUNIT_EXPECT_FALSE(current_test, preemptible()); + if (ip != target_ip) { + KUNIT_EXPECT_EQ(current_test, ip, target2_ip); + KUNIT_EXPECT_EQ(current_test, ret, (rand1 / div_factor) + 1); + } else + KUNIT_EXPECT_EQ(current_test, ret, (rand1 / div_factor)); + KUNIT_EXPECT_EQ(current_test, entry_val, (rand1 / div_factor)); + exit_val = entry_val + div_factor; +} + +/* Test entry only (no rethook) */ +static void test_fprobe_entry(struct kunit *test) +{ + struct fprobe fp_entry = { + .entry_handler = fp_entry_handler, + }; + + current_test = test; + + /* Before register, unregister should be failed. */ + KUNIT_EXPECT_NE(test, 0, unregister_fprobe(&fp_entry)); + KUNIT_EXPECT_EQ(test, 0, register_fprobe(&fp_entry, "fprobe_selftest_target*", NULL)); + + entry_val = 0; + exit_val = 0; + target(rand1); + KUNIT_EXPECT_NE(test, 0, entry_val); + KUNIT_EXPECT_EQ(test, 0, exit_val); + + entry_val = 0; + exit_val = 0; + target2(rand1); + KUNIT_EXPECT_NE(test, 0, entry_val); + KUNIT_EXPECT_EQ(test, 0, exit_val); + + KUNIT_EXPECT_EQ(test, 0, unregister_fprobe(&fp_entry)); +} + +static void test_fprobe(struct kunit *test) +{ + struct fprobe fp = { + .entry_handler = fp_entry_handler, + .exit_handler = fp_exit_handler, + }; + + current_test = test; + KUNIT_EXPECT_EQ(test, 0, register_fprobe(&fp, "fprobe_selftest_target*", NULL)); + + entry_val = 0; + exit_val = 0; + target(rand1); + KUNIT_EXPECT_NE(test, 0, entry_val); + KUNIT_EXPECT_EQ(test, entry_val + div_factor, exit_val); + + entry_val = 0; + exit_val = 0; + target2(rand1); + KUNIT_EXPECT_NE(test, 0, entry_val); + KUNIT_EXPECT_EQ(test, entry_val + div_factor, exit_val); + + KUNIT_EXPECT_EQ(test, 0, unregister_fprobe(&fp)); +} + +static void test_fprobe_syms(struct kunit *test) +{ + static const char *syms[] = {"fprobe_selftest_target", "fprobe_selftest_target2"}; + struct fprobe fp = { + .entry_handler = fp_entry_handler, + .exit_handler = fp_exit_handler, + }; + + current_test = test; + KUNIT_EXPECT_EQ(test, 0, register_fprobe_syms(&fp, syms, 2)); + + entry_val = 0; + exit_val = 0; + target(rand1); + KUNIT_EXPECT_NE(test, 0, entry_val); + KUNIT_EXPECT_EQ(test, entry_val + div_factor, exit_val); + + entry_val = 0; + exit_val = 0; + target2(rand1); + KUNIT_EXPECT_NE(test, 0, entry_val); + KUNIT_EXPECT_EQ(test, entry_val + div_factor, exit_val); + + KUNIT_EXPECT_EQ(test, 0, unregister_fprobe(&fp)); +} + +static unsigned long get_ftrace_location(void *func) +{ + unsigned long size, addr = (unsigned long)func; + + if (!kallsyms_lookup_size_offset(addr, &size, NULL) || !size) + return 0; + + return ftrace_location_range(addr, addr + size - 1); +} + +static int fprobe_test_init(struct kunit *test) +{ + do { + rand1 = prandom_u32(); + } while (rand1 <= div_factor); + + target = fprobe_selftest_target; + target2 = fprobe_selftest_target2; + target_ip = get_ftrace_location(target); + target2_ip = get_ftrace_location(target2); + + return 0; +} + +static struct kunit_case fprobe_testcases[] = { + KUNIT_CASE(test_fprobe_entry), + KUNIT_CASE(test_fprobe), + KUNIT_CASE(test_fprobe_syms), + {} +}; + +static struct kunit_suite fprobe_test_suite = { + .name = "fprobe_test", + .init = fprobe_test_init, + .test_cases = fprobe_testcases, +}; + +kunit_test_suites(&fprobe_test_suite); + +MODULE_LICENSE("GPL"); -- cgit v1.2.3