31da94c25a
This patch adds stack overflow detection to riscv, usable when CONFIG_VMAP_STACK=y. Overflow is detected in kernel exception entry(kernel/entry.S), if the kernel stack is overflow and been detected, the overflow handler is invoked on a per-cpu overflow stack. This approach preserves GPRs and the original exception information. The overflow detect is performed before any attempt is made to access the stack and the principle of stack overflow detection: kernel stacks are aligned to double their size, enabling overflow to be detected with a single bit test. For example, a 16K stack is aligned to 32K, ensuring that bit 14 of the SP must be zero. On an overflow (or underflow), this bit is flipped. Thus, overflow (of less than the size of the stack) can be detected by testing whether this bit is set. This gives us a useful error message on stack overflow, as can be trigger with the LKDTM overflow test: [ 388.053267] lkdtm: Performing direct entry EXHAUST_STACK [ 388.053663] lkdtm: Calling function with 1024 frame size to depth 32 ... [ 388.054016] lkdtm: loop 32/32 ... [ 388.054186] lkdtm: loop 31/32 ... [ 388.054491] lkdtm: loop 30/32 ... [ 388.054672] lkdtm: loop 29/32 ... [ 388.054859] lkdtm: loop 28/32 ... [ 388.055010] lkdtm: loop 27/32 ... [ 388.055163] lkdtm: loop 26/32 ... [ 388.055309] lkdtm: loop 25/32 ... [ 388.055481] lkdtm: loop 24/32 ... [ 388.055653] lkdtm: loop 23/32 ... [ 388.055837] lkdtm: loop 22/32 ... [ 388.056015] lkdtm: loop 21/32 ... [ 388.056188] lkdtm: loop 20/32 ... [ 388.058145] Insufficient stack space to handle exception! [ 388.058153] Task stack: [0xffffffd014260000..0xffffffd014264000] [ 388.058160] Overflow stack: [0xffffffe1f8d2c220..0xffffffe1f8d2d220] [ 388.058168] CPU: 0 PID: 89 Comm: bash Not tainted 5.12.0-rc8-dirty #90 [ 388.058175] Hardware name: riscv-virtio,qemu (DT) [ 388.058187] epc : number+0x32/0x2c0 [ 388.058247] ra : vsnprintf+0x2ae/0x3f0 [ 388.058255] epc : ffffffe0002d38f6 ra : ffffffe0002d814e sp : ffffffd01425ffc0 [ 388.058263] gp : ffffffe0012e4010 tp : ffffffe08014da00 t0 : ffffffd0142606e8 [ 388.058271] t1 : 0000000000000000 t2 : 0000000000000000 s0 : ffffffd014260070 [ 388.058303] s1 : ffffffd014260158 a0 : ffffffd01426015e a1 : ffffffd014260158 [ 388.058311] a2 : 0000000000000013 a3 : ffff0a01ffffff10 a4 : ffffffe000c398e0 [ 388.058319] a5 : 511b02ec65f3e300 a6 : 0000000000a1749a a7 : 0000000000000000 [ 388.058327] s2 : ffffffff000000ff s3 : 00000000ffff0a01 s4 : ffffffe0012e50a8 [ 388.058335] s5 : 0000000000ffff0a s6 : ffffffe0012e50a8 s7 : ffffffe000da1cc0 [ 388.058343] s8 : ffffffffffffffff s9 : ffffffd0142602b0 s10: ffffffd0142602a8 [ 388.058351] s11: ffffffd01426015e t3 : 00000000000f0000 t4 : ffffffffffffffff [ 388.058359] t5 : 000000000000002f t6 : ffffffd014260158 [ 388.058366] status: 0000000000000100 badaddr: ffffffd01425fff8 cause: 000000000000000f [ 388.058374] Kernel panic - not syncing: Kernel stack overflow [ 388.058381] CPU: 0 PID: 89 Comm: bash Not tainted 5.12.0-rc8-dirty #90 [ 388.058387] Hardware name: riscv-virtio,qemu (DT) [ 388.058393] Call Trace: [ 388.058400] [<ffffffe000004944>] walk_stackframe+0x0/0xce [ 388.058406] [<ffffffe0006f0b28>] dump_backtrace+0x38/0x46 [ 388.058412] [<ffffffe0006f0b46>] show_stack+0x10/0x18 [ 388.058418] [<ffffffe0006f3690>] dump_stack+0x74/0x8e [ 388.058424] [<ffffffe0006f0d52>] panic+0xfc/0x2b2 [ 388.058430] [<ffffffe0006f0acc>] print_trace_address+0x0/0x24 [ 388.058436] [<ffffffe0002d814e>] vsnprintf+0x2ae/0x3f0 [ 388.058956] SMP: stopping secondary CPUs Signed-off-by: Tong Tiangen <tongtiangen@huawei.com> Reviewed-by: Kefeng Wang <wangkefeng.wang@huawei.com> Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
114 lines
3.6 KiB
C
114 lines
3.6 KiB
C
/* SPDX-License-Identifier: GPL-2.0-only */
|
|
/*
|
|
* Copyright (C) 2009 Chen Liqin <liqin.chen@sunplusct.com>
|
|
* Copyright (C) 2012 Regents of the University of California
|
|
* Copyright (C) 2017 SiFive
|
|
*/
|
|
|
|
#ifndef _ASM_RISCV_THREAD_INFO_H
|
|
#define _ASM_RISCV_THREAD_INFO_H
|
|
|
|
#include <asm/page.h>
|
|
#include <linux/const.h>
|
|
|
|
/* thread information allocation */
|
|
#ifdef CONFIG_64BIT
|
|
#define THREAD_SIZE_ORDER (2)
|
|
#else
|
|
#define THREAD_SIZE_ORDER (1)
|
|
#endif
|
|
#define THREAD_SIZE (PAGE_SIZE << THREAD_SIZE_ORDER)
|
|
|
|
/*
|
|
* By aligning VMAP'd stacks to 2 * THREAD_SIZE, we can detect overflow by
|
|
* checking sp & (1 << THREAD_SHIFT), which we can do cheaply in the entry
|
|
* assembly.
|
|
*/
|
|
#ifdef CONFIG_VMAP_STACK
|
|
#define THREAD_ALIGN (2 * THREAD_SIZE)
|
|
#else
|
|
#define THREAD_ALIGN THREAD_SIZE
|
|
#endif
|
|
|
|
#define THREAD_SHIFT (PAGE_SHIFT + THREAD_SIZE_ORDER)
|
|
#define OVERFLOW_STACK_SIZE SZ_4K
|
|
#define SHADOW_OVERFLOW_STACK_SIZE (1024)
|
|
|
|
#ifndef __ASSEMBLY__
|
|
|
|
#include <asm/processor.h>
|
|
#include <asm/csr.h>
|
|
|
|
/*
|
|
* low level task data that entry.S needs immediate access to
|
|
* - this struct should fit entirely inside of one cache line
|
|
* - if the members of this struct changes, the assembly constants
|
|
* in asm-offsets.c must be updated accordingly
|
|
* - thread_info is included in task_struct at an offset of 0. This means that
|
|
* tp points to both thread_info and task_struct.
|
|
*/
|
|
struct thread_info {
|
|
unsigned long flags; /* low level flags */
|
|
int preempt_count; /* 0=>preemptible, <0=>BUG */
|
|
/*
|
|
* These stack pointers are overwritten on every system call or
|
|
* exception. SP is also saved to the stack it can be recovered when
|
|
* overwritten.
|
|
*/
|
|
long kernel_sp; /* Kernel stack pointer */
|
|
long user_sp; /* User stack pointer */
|
|
int cpu;
|
|
};
|
|
|
|
/*
|
|
* macros/functions for gaining access to the thread information structure
|
|
*
|
|
* preempt_count needs to be 1 initially, until the scheduler is functional.
|
|
*/
|
|
#define INIT_THREAD_INFO(tsk) \
|
|
{ \
|
|
.flags = 0, \
|
|
.preempt_count = INIT_PREEMPT_COUNT, \
|
|
}
|
|
|
|
#endif /* !__ASSEMBLY__ */
|
|
|
|
/*
|
|
* thread information flags
|
|
* - these are process state flags that various assembly files may need to
|
|
* access
|
|
* - pending work-to-be-done flags are in lowest half-word
|
|
* - other flags in upper half-word(s)
|
|
*/
|
|
#define TIF_SYSCALL_TRACE 0 /* syscall trace active */
|
|
#define TIF_NOTIFY_RESUME 1 /* callback before returning to user */
|
|
#define TIF_SIGPENDING 2 /* signal pending */
|
|
#define TIF_NEED_RESCHED 3 /* rescheduling necessary */
|
|
#define TIF_RESTORE_SIGMASK 4 /* restore signal mask in do_signal() */
|
|
#define TIF_MEMDIE 5 /* is terminating due to OOM killer */
|
|
#define TIF_SYSCALL_TRACEPOINT 6 /* syscall tracepoint instrumentation */
|
|
#define TIF_SYSCALL_AUDIT 7 /* syscall auditing */
|
|
#define TIF_SECCOMP 8 /* syscall secure computing */
|
|
#define TIF_NOTIFY_SIGNAL 9 /* signal notifications exist */
|
|
#define TIF_UPROBE 10 /* uprobe breakpoint or singlestep */
|
|
|
|
#define _TIF_SYSCALL_TRACE (1 << TIF_SYSCALL_TRACE)
|
|
#define _TIF_NOTIFY_RESUME (1 << TIF_NOTIFY_RESUME)
|
|
#define _TIF_SIGPENDING (1 << TIF_SIGPENDING)
|
|
#define _TIF_NEED_RESCHED (1 << TIF_NEED_RESCHED)
|
|
#define _TIF_SYSCALL_TRACEPOINT (1 << TIF_SYSCALL_TRACEPOINT)
|
|
#define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT)
|
|
#define _TIF_SECCOMP (1 << TIF_SECCOMP)
|
|
#define _TIF_NOTIFY_SIGNAL (1 << TIF_NOTIFY_SIGNAL)
|
|
#define _TIF_UPROBE (1 << TIF_UPROBE)
|
|
|
|
#define _TIF_WORK_MASK \
|
|
(_TIF_NOTIFY_RESUME | _TIF_SIGPENDING | _TIF_NEED_RESCHED | \
|
|
_TIF_NOTIFY_SIGNAL | _TIF_UPROBE)
|
|
|
|
#define _TIF_SYSCALL_WORK \
|
|
(_TIF_SYSCALL_TRACE | _TIF_SYSCALL_TRACEPOINT | _TIF_SYSCALL_AUDIT | \
|
|
_TIF_SECCOMP)
|
|
|
|
#endif /* _ASM_RISCV_THREAD_INFO_H */
|