From a5f7a5bb3b7f28ba7e4fa246775b29a0e5537255 Mon Sep 17 00:00:00 2001 From: Jinke Han Date: Tue, 8 Sep 2026 15:37:42 +0800 Subject: x86/kprobes: Fix crash when probing CS CALL instructions When using eBPF to probe CS CALL instructions within a function, a crash can be triggered. The eBPF tool probes offset 257 of the __hrtimer_run_queues() function: <__hrtimer_run_queues+249>: nopl 0x0(%rax,%rax,1) <__hrtimer_run_queues+254>: mov %r14,%rdi <__hrtimer_run_queues+257>: cs call <__x86_indirect_thunk_r12> <__hrtimer_run_queues+263>: mov %eax,%r12d <__hrtimer_run_queues+266>: xchg %ax,%ax <__hrtimer_run_queues+268>: mov %r13,%rdi Which triggers this crash: BUG: unable to handle page fault for address: 00000000000f41c9 #PF: supervisor write access in kernel mode #PF: error_code(0x0002) - not-present page PGD 0 P4D 0 Oops: 0002 [#1] SMP NOPTI CPU: 1 PID: 0 Comm: swapper/1 Kdump: loaded Tainted: P RIP: 0010:__hrtimer_run_queues+0x106/0x230 Note that __hrtimer_run_queues+0x106 is __hrtimer_run_queues+262, which is at the 6th byte of the above CS CALL instruction. Since the CS CALL instruction occupies 6 bytes, the exception occurred in the middle of that call instruction. The root cause is that when using eBPF tools to probe in the middle of a function, a kprobe with INT3 is used as the underlying implementation. During single-step emulation of the original CALL instruction, int3_emulate_call() assumes that the probed CALL instruction is 5 bytes long. However, the actual CS-prefixed CALL instruction occupies 6 bytes, so it constructs an incorrect exception return address. When the CPU returns from the kprobe handler, the next instruction to be executed is at the address of the last byte of that CS CALL instruction. Coincidentally, starting from that address, the CPU fetches and decodes a completely different instruction, which ultimately triggers a kernel crash. Fix the issue by using the actual instruction length obtained from the instruction decoder when constructing the exception return address, rather than relying on the hardcoded CALL_INSN_SIZE macro. [ mingo: Refined the changelog ] Fixes: 6256e668b7af ("x86/kprobes: Use int3 instead of debug trap for single-step") Suggested-by: Masami Hiramatsu (Google) Signed-off-by: Jinke Han Signed-off-by: Ingo Molnar Reviewed-by: Masami Hiramatsu (Google) Acked-by: Yafang Shao Acked-by: Borislav Petkov Cc: Peter Zijlstra Link: https://patch.msgid.link/20260908073742.GA10517@didi-ThinkCentre-M920t-N000 --- lib/crc/crc32-main.c | 105 +++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 105 insertions(+) create mode 100644 lib/crc/crc32-main.c (limited to 'lib/crc/crc32-main.c') diff --git a/lib/crc/crc32-main.c b/lib/crc/crc32-main.c new file mode 100644 index 000000000..fbb90c900 --- /dev/null +++ b/lib/crc/crc32-main.c @@ -0,0 +1,105 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Aug 8, 2011 Bob Pearson with help from Joakim Tjernlund and George Spelvin + * cleaned up code to current version of sparse and added the slicing-by-8 + * algorithm to the closely similar existing slicing-by-4 algorithm. + * + * Oct 15, 2000 Matt Domsch + * Nicer crc32 functions/docs submitted by linux@horizon.com. Thanks! + * Code was from the public domain, copyright abandoned. Code was + * subsequently included in the kernel, thus was re-licensed under the + * GNU GPL v2. + * + * Oct 12, 2000 Matt Domsch + * Same crc32 function was used in 5 other places in the kernel. + * I made one version, and deleted the others. + * There are various incantations of crc32(). Some use a seed of 0 or ~0. + * Some xor at the end with ~0. The generic crc32() function takes + * seed as an argument, and doesn't xor at the end. Then individual + * users can do whatever they need. + * drivers/net/smc9194.c uses seed ~0, doesn't xor with ~0. + * fs/jffs2 uses seed 0, doesn't xor with ~0. + * fs/partitions/efi.c uses seed ~0, xor's with ~0. + */ + +/* see: Documentation/staging/crc32.rst for a description of algorithms */ + +#include +#include +#include +#include + +#include "crc32table.h" + +static inline u32 __maybe_unused +crc32_le_base(u32 crc, const u8 *p, size_t len) +{ + while (len--) + crc = (crc >> 8) ^ crc32table_le[(crc & 255) ^ *p++]; + return crc; +} + +static inline u32 __maybe_unused +crc32_be_base(u32 crc, const u8 *p, size_t len) +{ + while (len--) + crc = (crc << 8) ^ crc32table_be[(crc >> 24) ^ *p++]; + return crc; +} + +static inline u32 __maybe_unused +crc32c_base(u32 crc, const u8 *p, size_t len) +{ + while (len--) + crc = (crc >> 8) ^ crc32ctable_le[(crc & 255) ^ *p++]; + return crc; +} + +#ifdef CONFIG_CRC32_ARCH +#include "crc32.h" /* $(SRCARCH)/crc32.h */ + +u32 crc32_optimizations(void) +{ + return crc32_optimizations_arch(); +} +EXPORT_SYMBOL(crc32_optimizations); +#else +#define crc32_le_arch crc32_le_base +#define crc32_be_arch crc32_be_base +#define crc32c_arch crc32c_base +#endif + +u32 crc32_le(u32 crc, const void *p, size_t len) +{ + return crc32_le_arch(crc, p, len); +} +EXPORT_SYMBOL(crc32_le); + +u32 crc32_be(u32 crc, const void *p, size_t len) +{ + return crc32_be_arch(crc, p, len); +} +EXPORT_SYMBOL(crc32_be); + +u32 crc32c(u32 crc, const void *p, size_t len) +{ + return crc32c_arch(crc, p, len); +} +EXPORT_SYMBOL(crc32c); + +#ifdef crc32_mod_init_arch +static int __init crc32_mod_init(void) +{ + crc32_mod_init_arch(); + return 0; +} +subsys_initcall(crc32_mod_init); + +static void __exit crc32_mod_exit(void) +{ +} +module_exit(crc32_mod_exit); +#endif + +MODULE_DESCRIPTION("CRC32 library functions"); +MODULE_LICENSE("GPL"); -- cgit v1.3.1