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, 176 insertions, 0 deletions

diff --git a/lib/zstd/compress/hist.c b/lib/zstd/compress/hist.c
new file mode 100644
index 000000000..87145a2d9
--- /dev/null
+++ b/lib/zstd/compress/hist.c
@@ -0,0 +1,176 @@
+// SPDX-License-Identifier: GPL-2.0+ OR BSD-3-Clause
+/* ******************************************************************
+ * hist : Histogram functions
+ * part of Finite State Entropy project
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ *
+ *  You can contact the author at :
+ *  - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy
+ *  - Public forum : https://groups.google.com/forum/#!forum/lz4c
+ *
+ * This source code is licensed under both the BSD-style license (found in the
+ * LICENSE file in the root directory of this source tree) and the GPLv2 (found
+ * in the COPYING file in the root directory of this source tree).
+ * You may select, at your option, one of the above-listed licenses.
+****************************************************************** */
+
+/* --- dependencies --- */
+#include "../common/mem.h"             /* U32, BYTE, etc. */
+#include "../common/debug.h"           /* assert, DEBUGLOG */
+#include "../common/error_private.h"   /* ERROR */
+#include "hist.h"
+
+
+/* --- Error management --- */
+unsigned HIST_isError(size_t code) { return ERR_isError(code); }
+
+/*-**************************************************************
+ *  Histogram functions
+ ****************************************************************/
+void HIST_add(unsigned* count, const void* src, size_t srcSize)
+{
+    const BYTE* ip = (const BYTE*)src;
+    const BYTE* const end = ip + srcSize;
+
+    while (ip<end) {
+        count[*ip++]++;
+    }
+}
+
+unsigned HIST_count_simple(unsigned* count, unsigned* maxSymbolValuePtr,
+                           const void* src, size_t srcSize)
+{
+    const BYTE* ip = (const BYTE*)src;
+    const BYTE* const end = ip + srcSize;
+    unsigned maxSymbolValue = *maxSymbolValuePtr;
+    unsigned largestCount=0;
+
+    ZSTD_memset(count, 0, (maxSymbolValue+1) * sizeof(*count));
+    if (srcSize==0) { *maxSymbolValuePtr = 0; return 0; }
+
+    while (ip<end) {
+        assert(*ip <= maxSymbolValue);
+        count[*ip++]++;
+    }
+
+    while (!count[maxSymbolValue]) maxSymbolValue--;
+    *maxSymbolValuePtr = maxSymbolValue;
+
+    {   U32 s;
+        for (s=0; s<=maxSymbolValue; s++)
+            if (count[s] > largestCount) largestCount = count[s];
+    }
+
+    return largestCount;
+}
+
+typedef enum { trustInput, checkMaxSymbolValue } HIST_checkInput_e;
+
+/* HIST_count_parallel_wksp() :
+ * store histogram into 4 intermediate tables, recombined at the end.
+ * this design makes better use of OoO cpus,
+ * and is noticeably faster when some values are heavily repeated.
+ * But it needs some additional workspace for intermediate tables.
+ * `workSpace` must be a U32 table of size >= HIST_WKSP_SIZE_U32.
+ * @return : largest histogram frequency,
+ *           or an error code (notably when histogram's alphabet is larger than *maxSymbolValuePtr) */
+static size_t HIST_count_parallel_wksp(
+                                unsigned* count, unsigned* maxSymbolValuePtr,
+                                const void* source, size_t sourceSize,
+                                HIST_checkInput_e check,
+                                U32* const workSpace)
+{
+    const BYTE* ip = (const BYTE*)source;
+    const BYTE* const iend = ip+sourceSize;
+    size_t const countSize = (*maxSymbolValuePtr + 1) * sizeof(*count);
+    unsigned max=0;
+    U32* const Counting1 = workSpace;
+    U32* const Counting2 = Counting1 + 256;
+    U32* const Counting3 = Counting2 + 256;
+    U32* const Counting4 = Counting3 + 256;
+
+    /* safety checks */
+    assert(*maxSymbolValuePtr <= 255);
+    if (!sourceSize) {
+        ZSTD_memset(count, 0, countSize);
+        *maxSymbolValuePtr = 0;
+        return 0;
+    }
+    ZSTD_memset(workSpace, 0, 4*256*sizeof(unsigned));
+
+    /* by stripes of 16 bytes */
+    {   U32 cached = MEM_read32(ip); ip += 4;
+        while (ip < iend-15) {
+            U32 c = cached; cached = MEM_read32(ip); ip += 4;
+            Counting1[(BYTE) c     ]++;
+            Counting2[(BYTE)(c>>8) ]++;
+            Counting3[(BYTE)(c>>16)]++;
+            Counting4[       c>>24 ]++;
+            c = cached; cached = MEM_read32(ip); ip += 4;
+            Counting1[(BYTE) c     ]++;
+            Counting2[(BYTE)(c>>8) ]++;
+            Counting3[(BYTE)(c>>16)]++;
+            Counting4[       c>>24 ]++;
+            c = cached; cached = MEM_read32(ip); ip += 4;
+            Counting1[(BYTE) c     ]++;
+            Counting2[(BYTE)(c>>8) ]++;
+            Counting3[(BYTE)(c>>16)]++;
+            Counting4[       c>>24 ]++;
+            c = cached; cached = MEM_read32(ip); ip += 4;
+            Counting1[(BYTE) c     ]++;
+            Counting2[(BYTE)(c>>8) ]++;
+            Counting3[(BYTE)(c>>16)]++;
+            Counting4[       c>>24 ]++;
+        }
+        ip-=4;
+    }
+
+    /* finish last symbols */
+    while (ip<iend) Counting1[*ip++]++;
+
+    {   U32 s;
+        for (s=0; s<256; s++) {
+            Counting1[s] += Counting2[s] + Counting3[s] + Counting4[s];
+            if (Counting1[s] > max) max = Counting1[s];
+    }   }
+
+    {   unsigned maxSymbolValue = 255;
+        while (!Counting1[maxSymbolValue]) maxSymbolValue--;
+        if (check && maxSymbolValue > *maxSymbolValuePtr) return ERROR(maxSymbolValue_tooSmall);
+        *maxSymbolValuePtr = maxSymbolValue;
+        ZSTD_memmove(count, Counting1, countSize);   /* in case count & Counting1 are overlapping */
+    }
+    return (size_t)max;
+}
+
+/* HIST_countFast_wksp() :
+ * Same as HIST_countFast(), but using an externally provided scratch buffer.
+ * `workSpace` is a writable buffer which must be 4-bytes aligned,
+ * `workSpaceSize` must be >= HIST_WKSP_SIZE
+ */
+size_t HIST_countFast_wksp(unsigned* count, unsigned* maxSymbolValuePtr,
+                          const void* source, size_t sourceSize,
+                          void* workSpace, size_t workSpaceSize)
+{
+    if (sourceSize < 1500) /* heuristic threshold */
+        return HIST_count_simple(count, maxSymbolValuePtr, source, sourceSize);
+    if ((size_t)workSpace & 3) return ERROR(GENERIC);  /* must be aligned on 4-bytes boundaries */
+    if (workSpaceSize < HIST_WKSP_SIZE) return ERROR(workSpace_tooSmall);
+    return HIST_count_parallel_wksp(count, maxSymbolValuePtr, source, sourceSize, trustInput, (U32*)workSpace);
+}
+
+/* HIST_count_wksp() :
+ * Same as HIST_count(), but using an externally provided scratch buffer.
+ * `workSpace` size must be table of >= HIST_WKSP_SIZE_U32 unsigned */
+size_t HIST_count_wksp(unsigned* count, unsigned* maxSymbolValuePtr,
+                       const void* source, size_t sourceSize,
+                       void* workSpace, size_t workSpaceSize)
+{
+    if ((size_t)workSpace & 3) return ERROR(GENERIC);  /* must be aligned on 4-bytes boundaries */
+    if (workSpaceSize < HIST_WKSP_SIZE) return ERROR(workSpace_tooSmall);
+    if (*maxSymbolValuePtr < 255)
+        return HIST_count_parallel_wksp(count, maxSymbolValuePtr, source, sourceSize, checkMaxSymbolValue, (U32*)workSpace);
+    *maxSymbolValuePtr = 255;
+    return HIST_countFast_wksp(count, maxSymbolValuePtr, source, sourceSize, workSpace, workSpaceSize);
+}
+