Commit c4bed4b9 authored by Lai Jiangshan's avatar Lai Jiangshan Committed by Thomas Gleixner
Browse files

x86/debug: Prevent data breakpoints on __per_cpu_offset 



When FSGSBASE is enabled, paranoid_entry() fetches the per-CPU GSBASE value
via __per_cpu_offset or pcpu_unit_offsets.

When a data breakpoint is set on __per_cpu_offset[cpu] (read-write
operation), the specific CPU will be stuck in an infinite #DB loop.

RCU will try to send an NMI to the specific CPU, but it is not working
either since NMI also relies on paranoid_entry(). Which means it's
undebuggable.

Fixes: eaad9812("x86/entry/64: Introduce the FIND_PERCPU_BASE macro")
Signed-off-by: default avatarLai Jiangshan <laijs@linux.alibaba.com>
Signed-off-by: default avatarThomas Gleixner <tglx@linutronix.de>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20210204152708.21308-1-jiangshanlai@gmail.com
parent 25a068b8

arch/x86/kernel/hw_breakpoint.c

+14 −0
Original line number Diff line number Diff line
@@ -269,6 +269,20 @@ static inline bool within_cpu_entry(unsigned long addr, unsigned long end) 
			CPU_ENTRY_AREA_TOTAL_SIZE))

		return true;



	/*

	 * When FSGSBASE is enabled, paranoid_entry() fetches the per-CPU

	 * GSBASE value via __per_cpu_offset or pcpu_unit_offsets.

	 */

#ifdef CONFIG_SMP

	if (within_area(addr, end, (unsigned long)__per_cpu_offset,

			sizeof(unsigned long) * nr_cpu_ids))

		return true;

#else

	if (within_area(addr, end, (unsigned long)&pcpu_unit_offsets,

			sizeof(pcpu_unit_offsets)))

		return true;

#endif



	for_each_possible_cpu(cpu) {

		/* The original rw GDT is being used after load_direct_gdt() */

		if (within_area(addr, end, (unsigned long)get_cpu_gdt_rw(cpu),