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, 281 insertions, 0 deletions

diff --git a/arch/nios2/mm/cacheflush.c b/arch/nios2/mm/cacheflush.c
new file mode 100644
index 000000000..102324760
--- /dev/null
+++ b/arch/nios2/mm/cacheflush.c
@@ -0,0 +1,281 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2009, Wind River Systems Inc
+ * Implemented by fredrik.markstrom@gmail.com and ivarholmqvist@gmail.com
+ */
+
+#include <linux/export.h>
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/fs.h>
+#include <linux/pagemap.h>
+
+#include <asm/cacheflush.h>
+#include <asm/cpuinfo.h>
+
+static void __flush_dcache(unsigned long start, unsigned long end)
+{
+	unsigned long addr;
+
+	start &= ~(cpuinfo.dcache_line_size - 1);
+	end += (cpuinfo.dcache_line_size - 1);
+	end &= ~(cpuinfo.dcache_line_size - 1);
+
+	if (end > start + cpuinfo.dcache_size)
+		end = start + cpuinfo.dcache_size;
+
+	for (addr = start; addr < end; addr += cpuinfo.dcache_line_size) {
+		__asm__ __volatile__ ("   flushd 0(%0)\n"
+					: /* Outputs */
+					: /* Inputs  */ "r"(addr)
+					/* : No clobber */);
+	}
+}
+
+static void __invalidate_dcache(unsigned long start, unsigned long end)
+{
+	unsigned long addr;
+
+	start &= ~(cpuinfo.dcache_line_size - 1);
+	end += (cpuinfo.dcache_line_size - 1);
+	end &= ~(cpuinfo.dcache_line_size - 1);
+
+	for (addr = start; addr < end; addr += cpuinfo.dcache_line_size) {
+		__asm__ __volatile__ ("   initda 0(%0)\n"
+					: /* Outputs */
+					: /* Inputs  */ "r"(addr)
+					/* : No clobber */);
+	}
+}
+
+static void __flush_icache(unsigned long start, unsigned long end)
+{
+	unsigned long addr;
+
+	start &= ~(cpuinfo.icache_line_size - 1);
+	end += (cpuinfo.icache_line_size - 1);
+	end &= ~(cpuinfo.icache_line_size - 1);
+
+	if (end > start + cpuinfo.icache_size)
+		end = start + cpuinfo.icache_size;
+
+	for (addr = start; addr < end; addr += cpuinfo.icache_line_size) {
+		__asm__ __volatile__ ("   flushi %0\n"
+					: /* Outputs */
+					: /* Inputs  */ "r"(addr)
+					/* : No clobber */);
+	}
+	__asm__ __volatile(" flushp\n");
+}
+
+static void flush_aliases(struct address_space *mapping, struct folio *folio)
+{
+	struct mm_struct *mm = current->active_mm;
+	struct vm_area_struct *vma;
+	unsigned long flags;
+	pgoff_t pgoff;
+	unsigned long nr = folio_nr_pages(folio);
+
+	pgoff = folio->index;
+
+	flush_dcache_mmap_lock_irqsave(mapping, flags);
+	mapping_rmap_tree_foreach(vma, mapping, pgoff, pgoff + nr - 1) {
+		unsigned long start;
+
+		if (vma->vm_mm != mm)
+			continue;
+		if (!(vma->vm_flags & VM_MAYSHARE))
+			continue;
+
+		start = vma->vm_start + ((pgoff - vma->vm_pgoff) << PAGE_SHIFT);
+		flush_cache_range(vma, start, start + nr * PAGE_SIZE);
+	}
+	flush_dcache_mmap_unlock_irqrestore(mapping, flags);
+}
+
+void flush_cache_all(void)
+{
+	__flush_dcache(0, cpuinfo.dcache_size);
+	__flush_icache(0, cpuinfo.icache_size);
+}
+
+void flush_cache_mm(struct mm_struct *mm)
+{
+	flush_cache_all();
+}
+
+void flush_cache_dup_mm(struct mm_struct *mm)
+{
+	flush_cache_all();
+}
+
+void flush_icache_range(unsigned long start, unsigned long end)
+{
+	__flush_dcache(start, end);
+	__flush_icache(start, end);
+}
+
+void flush_dcache_range(unsigned long start, unsigned long end)
+{
+	__flush_dcache(start, end);
+	__flush_icache(start, end);
+}
+EXPORT_SYMBOL(flush_dcache_range);
+
+void invalidate_dcache_range(unsigned long start, unsigned long end)
+{
+	__invalidate_dcache(start, end);
+}
+EXPORT_SYMBOL(invalidate_dcache_range);
+
+void flush_cache_range(struct vm_area_struct *vma, unsigned long start,
+			unsigned long end)
+{
+	__flush_dcache(start, end);
+	if (vma == NULL || (vma->vm_flags & VM_EXEC))
+		__flush_icache(start, end);
+}
+
+void flush_icache_pages(struct vm_area_struct *vma, struct page *page,
+		unsigned int nr)
+{
+	unsigned long start = (unsigned long) page_address(page);
+	unsigned long end = start + nr * PAGE_SIZE;
+
+	__flush_dcache(start, end);
+	__flush_icache(start, end);
+}
+
+void flush_cache_page(struct vm_area_struct *vma, unsigned long vmaddr,
+			unsigned long pfn)
+{
+	unsigned long start = vmaddr;
+	unsigned long end = start + PAGE_SIZE;
+
+	__flush_dcache(start, end);
+	if (vma->vm_flags & VM_EXEC)
+		__flush_icache(start, end);
+}
+
+static void __flush_dcache_folio(struct folio *folio)
+{
+	/*
+	 * Writeback any data associated with the kernel mapping of this
+	 * page.  This ensures that data in the physical page is mutually
+	 * coherent with the kernels mapping.
+	 */
+	unsigned long start = (unsigned long)folio_address(folio);
+
+	__flush_dcache(start, start + folio_size(folio));
+}
+
+void flush_dcache_folio(struct folio *folio)
+{
+	struct address_space *mapping;
+
+	/*
+	 * The zero page is never written to, so never has any dirty
+	 * cache lines, and therefore never needs to be flushed.
+	 */
+	if (is_zero_pfn(folio_pfn(folio)))
+		return;
+
+	mapping = folio_flush_mapping(folio);
+
+	/* Flush this page if there are aliases. */
+	if (mapping && !mapping_mapped(mapping)) {
+		clear_bit(PG_dcache_clean, &folio->flags.f);
+	} else {
+		__flush_dcache_folio(folio);
+		if (mapping) {
+			unsigned long start = (unsigned long)folio_address(folio);
+			flush_aliases(mapping, folio);
+			flush_icache_range(start, start + folio_size(folio));
+		}
+		set_bit(PG_dcache_clean, &folio->flags.f);
+	}
+}
+EXPORT_SYMBOL(flush_dcache_folio);
+
+void flush_dcache_page(struct page *page)
+{
+	flush_dcache_folio(page_folio(page));
+}
+EXPORT_SYMBOL(flush_dcache_page);
+
+void update_mmu_cache_range(struct vm_fault *vmf, struct vm_area_struct *vma,
+		unsigned long address, pte_t *ptep, unsigned int nr)
+{
+	pte_t pte = *ptep;
+	unsigned long pfn = pte_pfn(pte);
+	struct folio *folio;
+	struct address_space *mapping;
+
+	reload_tlb_page(vma, address, pte);
+
+	if (!pfn_valid(pfn))
+		return;
+
+	/*
+	* The zero page is never written to, so never has any dirty
+	* cache lines, and therefore never needs to be flushed.
+	*/
+	if (is_zero_pfn(pfn))
+		return;
+
+	folio = page_folio(pfn_to_page(pfn));
+	if (!test_and_set_bit(PG_dcache_clean, &folio->flags.f))
+		__flush_dcache_folio(folio);
+
+	mapping = folio_flush_mapping(folio);
+	if (mapping) {
+		flush_aliases(mapping, folio);
+		if (vma->vm_flags & VM_EXEC)
+			flush_icache_pages(vma, &folio->page,
+					folio_nr_pages(folio));
+	}
+}
+
+void copy_user_page(void *vto, void *vfrom, unsigned long vaddr,
+		    struct page *to)
+{
+	__flush_dcache(vaddr, vaddr + PAGE_SIZE);
+	__flush_icache(vaddr, vaddr + PAGE_SIZE);
+	copy_page(vto, vfrom);
+	__flush_dcache((unsigned long)vto, (unsigned long)vto + PAGE_SIZE);
+	__flush_icache((unsigned long)vto, (unsigned long)vto + PAGE_SIZE);
+}
+
+void clear_user_page(void *addr, unsigned long vaddr, struct page *page)
+{
+	__flush_dcache(vaddr, vaddr + PAGE_SIZE);
+	__flush_icache(vaddr, vaddr + PAGE_SIZE);
+	clear_page(addr);
+	__flush_dcache((unsigned long)addr, (unsigned long)addr + PAGE_SIZE);
+	__flush_icache((unsigned long)addr, (unsigned long)addr + PAGE_SIZE);
+}
+
+void copy_from_user_page(struct vm_area_struct *vma, struct page *page,
+			unsigned long user_vaddr,
+			void *dst, void *src, int len)
+{
+	flush_cache_page(vma, user_vaddr, page_to_pfn(page));
+	memcpy(dst, src, len);
+	__flush_dcache((unsigned long)src, (unsigned long)src + len);
+	if (vma->vm_flags & VM_EXEC)
+		__flush_icache((unsigned long)src, (unsigned long)src + len);
+}
+
+void copy_to_user_page(struct vm_area_struct *vma, struct page *page,
+			unsigned long user_vaddr,
+			void *dst, void *src, int len)
+{
+	flush_cache_page(vma, user_vaddr, page_to_pfn(page));
+	memcpy(dst, src, len);
+	__flush_dcache((unsigned long)dst, (unsigned long)dst + len);
+	if (vma->vm_flags & VM_EXEC)
+		__flush_icache((unsigned long)dst, (unsigned long)dst + len);
+}