x86/kprobes: Fix crash when probing CS CALL instructionsgrafted

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) <mhiramat@kernel.org>
Signed-off-by: Jinke Han <jinkehan@didiglobal.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Acked-by: Yafang Shao <laoar.shao@gmail.com>
Acked-by: Borislav Petkov <bp@alien8.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://patch.msgid.link/20260908073742.GA10517@didi-ThinkCentre-M920t-N000

1 files changed, 219 insertions, 0 deletions

diff --git a/scripts/Makefile.warn b/scripts/Makefile.warn
new file mode 100644
index 000000000..35af7d6c6
--- /dev/null
+++ b/scripts/Makefile.warn
@@ -0,0 +1,219 @@
+# SPDX-License-Identifier: GPL-2.0
+# ==========================================================================
+# make W=... settings
+#
+# There are four warning groups enabled by W=1, W=2, W=3, and W=e
+# They are independent, and can be combined like W=12 or W=123e.
+# ==========================================================================
+
+# Default set of warnings, always enabled
+KBUILD_CFLAGS += -Wall
+KBUILD_CFLAGS += -Wextra
+KBUILD_CFLAGS += -Wundef
+KBUILD_CFLAGS += -Werror=implicit-function-declaration
+KBUILD_CFLAGS += -Werror=implicit-int
+KBUILD_CFLAGS += -Werror=return-type
+KBUILD_CFLAGS += -Werror=strict-prototypes
+KBUILD_CFLAGS += -Wno-format-security
+KBUILD_CFLAGS += -Wno-trigraphs
+KBUILD_CFLAGS += -Wno-frame-address
+KBUILD_CFLAGS += $(call cc-option, -Wno-address-of-packed-member)
+KBUILD_CFLAGS += -Wmissing-declarations
+KBUILD_CFLAGS += -Wmissing-prototypes
+
+ifneq ($(CONFIG_FRAME_WARN),0)
+KBUILD_CFLAGS += -Wframe-larger-than=$(CONFIG_FRAME_WARN)
+endif
+
+KBUILD_CFLAGS-$(CONFIG_CC_NO_ARRAY_BOUNDS) += -Wno-array-bounds
+
+ifdef CONFIG_CC_IS_CLANG
+# Clang checks for overflow/truncation with '%p', while GCC does not:
+# https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111219
+KBUILD_CFLAGS += $(call cc-option, -Wno-format-overflow-non-kprintf)
+KBUILD_CFLAGS += $(call cc-option, -Wno-format-truncation-non-kprintf)
+
+# Clang may emit a warning when a const variable, such as the dummy variables
+# in typecheck(), or const member of an aggregate type are not initialized,
+# which can result in unexpected behavior. However, in many audited cases of
+# the "field" variant of the warning, this is intentional because the field is
+# never used within a particular call path, the field is within a union with
+# other non-const members, or the containing object is not const so the field
+# can be modified via memcpy() / memset(). While the variable warning also gets
+# disabled with this same switch, there should not be too much coverage lost
+# because -Wuninitialized will still flag when an uninitialized const variable
+# is used.
+KBUILD_CFLAGS += $(call cc-option, -Wno-default-const-init-unsafe)
+else
+
+# gcc inanely warns about local variables called 'main'
+KBUILD_CFLAGS += -Wno-main
+endif
+
+# Too noisy on range checks and in macros handling both signed and unsigned.
+KBUILD_CFLAGS += -Wno-type-limits
+
+# These result in bogus false positives
+KBUILD_CFLAGS += $(call cc-option, -Wno-dangling-pointer)
+
+# Stack Variable Length Arrays (VLAs) must not be used in the kernel.
+# Function array parameters should, however, be usable, but -Wvla will
+# warn for those. Clang has no way yet to distinguish between the VLA
+# types, so depend on GCC for now to keep stack VLAs out of the tree.
+# https://github.com/llvm/llvm-project/issues/57098
+# https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98217
+KBUILD_CFLAGS += $(call cc-option,-Wvla-larger-than=1)
+
+# disable pointer signed / unsigned warnings in gcc 4.0
+KBUILD_CFLAGS += -Wno-pointer-sign
+
+# In order to make sure new function cast mismatches are not introduced
+# in the kernel (to avoid tripping CFI checking), the kernel should be
+# globally built with -Wcast-function-type.
+KBUILD_CFLAGS += -Wcast-function-type
+
+# Currently, disable -Wstringop-overflow for GCC 11, globally.
+KBUILD_CFLAGS-$(CONFIG_CC_NO_STRINGOP_OVERFLOW) += $(call cc-option, -Wno-stringop-overflow)
+KBUILD_CFLAGS-$(CONFIG_CC_STRINGOP_OVERFLOW) += $(call cc-option, -Wstringop-overflow)
+
+# Currently, disable -Wunterminated-string-initialization as broken
+KBUILD_CFLAGS += $(call cc-option, -Wno-unterminated-string-initialization)
+
+# The allocators already balk at large sizes, so silence the compiler
+# warnings for bounds checks involving those possible values. While
+# -Wno-alloc-size-larger-than would normally be used here, earlier versions
+# of gcc (<9.1) weirdly don't handle the option correctly when _other_
+# warnings are produced (?!). Using -Walloc-size-larger-than=SIZE_MAX
+# doesn't work (as it is documented to), silently resolving to "0" prior to
+# version 9.1 (and producing an error more recently). Numeric values larger
+# than PTRDIFF_MAX also don't work prior to version 9.1, which are silently
+# ignored, continuing to default to PTRDIFF_MAX. So, left with no other
+# choice, we must perform a versioned check to disable this warning.
+# https://lore.kernel.org/lkml/20210824115859.187f272f@canb.auug.org.au
+KBUILD_CFLAGS-$(call gcc-min-version, 90100) += -Wno-alloc-size-larger-than
+KBUILD_CFLAGS += $(KBUILD_CFLAGS-y) $(CONFIG_CC_IMPLICIT_FALLTHROUGH)
+
+# Prohibit date/time macros, which would make the build non-deterministic
+KBUILD_CFLAGS += -Werror=date-time
+
+# enforce correct pointer usage
+KBUILD_CFLAGS += -Werror=incompatible-pointer-types
+
+# Require designated initializers for all marked structures
+KBUILD_CFLAGS += $(call cc-option,-Werror=designated-init)
+
+# Warn if there is an enum types mismatch
+KBUILD_CFLAGS += $(call cc-option,-Wenum-conversion)
+
+KBUILD_CFLAGS += -Wunused
+
+#
+# W=1 - warnings which may be relevant and do not occur too often
+#
+ifneq ($(findstring 1, $(KBUILD_EXTRA_WARN)),)
+
+KBUILD_CFLAGS += -Wmissing-format-attribute
+KBUILD_CFLAGS += -Wmissing-include-dirs
+KBUILD_CFLAGS += -Wunused-const-variable
+
+KBUILD_CPPFLAGS += -Wundef
+KBUILD_CPPFLAGS += -DKBUILD_EXTRA_WARN1
+
+else
+
+# Some diagnostics enabled by default are noisy.
+# Suppress them by using -Wno... except for W=1.
+KBUILD_CFLAGS += -Wno-unused-but-set-variable
+KBUILD_CFLAGS += -Wno-unused-const-variable
+KBUILD_CFLAGS += $(call cc-option, -Wno-packed-not-aligned)
+KBUILD_CFLAGS += $(call cc-option, -Wno-format-overflow)
+ifdef CONFIG_CC_IS_GCC
+KBUILD_CFLAGS += -Wno-format-truncation
+endif
+KBUILD_CFLAGS += $(call cc-option, -Wno-stringop-truncation)
+
+KBUILD_CFLAGS += -Wno-override-init # alias for -Wno-initializer-overrides in clang
+
+ifdef CONFIG_CC_IS_CLANG
+KBUILD_CFLAGS += -Wno-pointer-to-enum-cast
+KBUILD_CFLAGS += -Wno-tautological-constant-out-of-range-compare
+KBUILD_CFLAGS += -Wno-unaligned-access
+KBUILD_CFLAGS += -Wno-enum-compare-conditional
+endif
+
+endif
+
+#
+# W=2 - warnings which occur quite often but may still be relevant
+#
+ifneq ($(findstring 2, $(KBUILD_EXTRA_WARN)),)
+
+KBUILD_CFLAGS += -Wdisabled-optimization
+KBUILD_CFLAGS += -Wshadow
+KBUILD_CFLAGS += $(call cc-option, -Wlogical-op)
+KBUILD_CFLAGS += -Wunused-macros
+
+KBUILD_CPPFLAGS += -DKBUILD_EXTRA_WARN2
+
+else
+
+# The following turn off the warnings enabled by -Wextra
+KBUILD_CFLAGS += -Wno-missing-field-initializers
+KBUILD_CFLAGS += -Wno-shift-negative-value
+
+ifdef CONFIG_CC_IS_CLANG
+KBUILD_CFLAGS += -Wno-enum-enum-conversion
+endif
+
+ifdef CONFIG_CC_IS_GCC
+KBUILD_CFLAGS += -Wno-maybe-uninitialized
+endif
+
+endif
+
+#
+# W=3 - more obscure warnings, can most likely be ignored
+#
+ifneq ($(findstring 3, $(KBUILD_EXTRA_WARN)),)
+
+KBUILD_CFLAGS += -Wbad-function-cast
+KBUILD_CFLAGS += -Wcast-align
+KBUILD_CFLAGS += -Wcast-qual
+KBUILD_CFLAGS += -Wconversion
+KBUILD_CFLAGS += -Wpacked
+KBUILD_CFLAGS += -Wpadded
+KBUILD_CFLAGS += -Wpointer-arith
+KBUILD_CFLAGS += -Wredundant-decls
+KBUILD_CFLAGS += -Wsign-compare
+KBUILD_CFLAGS += -Wswitch-default
+
+KBUILD_CPPFLAGS += -DKBUILD_EXTRA_WARN3
+
+else
+
+# The following turn off the warnings enabled by -Wextra
+KBUILD_CFLAGS += -Wno-sign-compare
+KBUILD_CFLAGS += -Wno-unused-parameter
+
+endif
+
+#
+# W=e and CONFIG_WERROR - error out on warnings
+#
+ifneq ($(findstring e, $(KBUILD_EXTRA_WARN))$(CONFIG_WERROR),)
+
+KBUILD_CPPFLAGS		+= -Werror
+KBUILD_AFLAGS		+= -Wa,--fatal-warnings
+KBUILD_LDFLAGS		+= --fatal-warnings
+KBUILD_USERCFLAGS	+= -Werror
+KBUILD_USERLDFLAGS	+= -Wl,--fatal-warnings
+KBUILD_RUSTFLAGS	+= -Dwarnings
+
+# While hostprog flags are used during build bootstrapping (thus should not
+# depend on CONFIG_ symbols), -Werror is disruptive and should be opted into.
+# Only apply -Werror to hostprogs built after the initial Kconfig stage.
+KBUILD_HOSTCFLAGS	+= -Werror
+KBUILD_HOSTLDFLAGS	+= -Wl,--fatal-warnings
+KBUILD_HOSTRUSTFLAGS	+= -Dwarnings
+
+endif