OpenCloudOS-Kernel/arch/riscv/kernel/stacktrace.c

146 lines
3.3 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-only
/*
* Copyright (C) 2008 ARM Limited
* Copyright (C) 2014 Regents of the University of California
*/
#include <linux/export.h>
#include <linux/kallsyms.h>
#include <linux/sched.h>
#include <linux/sched/debug.h>
#include <linux/sched/task_stack.h>
#include <linux/stacktrace.h>
#include <linux/ftrace.h>
#include <asm/stacktrace.h>
#ifdef CONFIG_FRAME_POINTER
void notrace walk_stackframe(struct task_struct *task, struct pt_regs *regs,
bool (*fn)(void *, unsigned long), void *arg)
{
unsigned long fp, sp, pc;
riscv: eliminate unreliable __builtin_frame_address(1) I tried different pieces of code which uses __builtin_frame_address(1) (with both gcc version 7.5.0 and 10.3.0) to verify whether it works as expected on riscv64. The result is negative. What the compiler had generated is as below: 31 fp = (unsigned long)__builtin_frame_address(1); 0xffffffff80006024 <+200>: ld s1,0(s0) It takes '0(s0)' as the address of frame 1 (caller), but the actual address should be '-16(s0)'. | ... | <-+ +-----------------+ | | return address | | | previous fp | | | saved registers | | | local variables | | $fp --> | ... | | +-----------------+ | | return address | | | previous fp --------+ | saved registers | $sp --> | local variables | +-----------------+ This leads the kernel can not dump the full stack trace on riscv. [ 7.222126][ T1] Call Trace: [ 7.222804][ T1] [<ffffffff80006058>] dump_backtrace+0x2c/0x3a This problem is not exposed on most riscv builds just because the '0(s0)' occasionally is the address frame 2 (caller's caller), if only ra and fp are stored in frame 1 (caller). | ... | <-+ +-----------------+ | | return address | | $fp --> | previous fp | | +-----------------+ | | return address | | | previous fp --------+ | saved registers | $sp --> | local variables | +-----------------+ This could be a *bug* of gcc that should be fixed. But as noted in gcc manual "Calling this function with a nonzero argument can have unpredictable effects, including crashing the calling program.", let's remove the '__builtin_frame_address(1)' in backtrace code. With this fix now it can show full stack trace: [ 10.444838][ T1] Call Trace: [ 10.446199][ T1] [<ffffffff8000606c>] dump_backtrace+0x2c/0x3a [ 10.447711][ T1] [<ffffffff800060ac>] show_stack+0x32/0x3e [ 10.448710][ T1] [<ffffffff80a005c0>] dump_stack_lvl+0x58/0x7a [ 10.449941][ T1] [<ffffffff80a005f6>] dump_stack+0x14/0x1c [ 10.450929][ T1] [<ffffffff804c04ee>] ubsan_epilogue+0x10/0x5a [ 10.451869][ T1] [<ffffffff804c092e>] __ubsan_handle_load_invalid_value+0x6c/0x78 [ 10.453049][ T1] [<ffffffff8018f834>] __pagevec_release+0x62/0x64 [ 10.455476][ T1] [<ffffffff80190830>] truncate_inode_pages_range+0x132/0x5be [ 10.456798][ T1] [<ffffffff80190ce0>] truncate_inode_pages+0x24/0x30 [ 10.457853][ T1] [<ffffffff8045bb04>] kill_bdev+0x32/0x3c ... Signed-off-by: Changbin Du <changbin.du@gmail.com> Fixes: eac2f3059e02 ("riscv: stacktrace: fix the riscv stacktrace when CONFIG_FRAME_POINTER enabled") Cc: stable@vger.kernel.org Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
2022-01-17 23:44:33 +08:00
int level = 0;
if (regs) {
fp = frame_pointer(regs);
sp = user_stack_pointer(regs);
pc = instruction_pointer(regs);
riscv: stacktrace: Fix NULL pointer dereference When CONFIG_FRAME_POINTER=y, calling dump_stack() can always trigger NULL pointer dereference panic similar as below: [ 0.396060] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.13.0-rc5+ #47 [ 0.396692] Hardware name: riscv-virtio,qemu (DT) [ 0.397176] Call Trace: [ 0.398191] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000960 [ 0.399487] Oops [#1] [ 0.399739] Modules linked in: [ 0.400135] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.13.0-rc5+ #47 [ 0.400570] Hardware name: riscv-virtio,qemu (DT) [ 0.400926] epc : walk_stackframe+0xc4/0xdc [ 0.401291] ra : dump_backtrace+0x30/0x38 [ 0.401630] epc : ffffffff80004922 ra : ffffffff8000496a sp : ffffffe000f3bd00 [ 0.402115] gp : ffffffff80cfdcb8 tp : ffffffe000f30000 t0 : ffffffff80d0b0cf [ 0.402602] t1 : ffffffff80d0b0c0 t2 : 0000000000000000 s0 : ffffffe000f3bd60 [ 0.403071] s1 : ffffffff808bc2e8 a0 : 0000000000001000 a1 : 0000000000000000 [ 0.403448] a2 : ffffffff803d7088 a3 : ffffffff808bc2e8 a4 : 6131725dbc24d400 [ 0.403820] a5 : 0000000000001000 a6 : 0000000000000002 a7 : ffffffffffffffff [ 0.404226] s2 : 0000000000000000 s3 : 0000000000000000 s4 : 0000000000000000 [ 0.404634] s5 : ffffffff803d7088 s6 : ffffffff808bc2e8 s7 : ffffffff80630650 [ 0.405085] s8 : ffffffff80912a80 s9 : 0000000000000008 s10: ffffffff804000fc [ 0.405388] s11: 0000000000000000 t3 : 0000000000000043 t4 : ffffffffffffffff [ 0.405616] t5 : 000000000000003d t6 : ffffffe000f3baa8 [ 0.405793] status: 0000000000000100 badaddr: 0000000000000960 cause: 000000000000000d [ 0.406135] [<ffffffff80004922>] walk_stackframe+0xc4/0xdc [ 0.407032] [<ffffffff8000496a>] dump_backtrace+0x30/0x38 [ 0.407797] [<ffffffff803d7100>] show_stack+0x40/0x4c [ 0.408234] [<ffffffff803d9e5c>] dump_stack+0x90/0xb6 [ 0.409019] [<ffffffff8040423e>] ptdump_init+0x20/0xc4 [ 0.409681] [<ffffffff800015b6>] do_one_initcall+0x4c/0x226 [ 0.410110] [<ffffffff80401094>] kernel_init_freeable+0x1f4/0x258 [ 0.410562] [<ffffffff803dba88>] kernel_init+0x22/0x148 [ 0.410959] [<ffffffff800029e2>] ret_from_exception+0x0/0x14 [ 0.412241] ---[ end trace b2ab92c901b96251 ]--- [ 0.413099] Kernel panic - not syncing: Attempted to kill init! exitcode=0x0000000b The reason is the task is NULL when we finally call walk_stackframe() the NULL is passed from __dump_stack(): |static void __dump_stack(void) |{ | dump_stack_print_info(KERN_DEFAULT); | show_stack(NULL, NULL, KERN_DEFAULT); |} Fix this issue by checking "task == NULL" case in walk_stackframe(). Fixes: eac2f3059e02 ("riscv: stacktrace: fix the riscv stacktrace when CONFIG_FRAME_POINTER enabled") Signed-off-by: Jisheng Zhang <jszhang@kernel.org> Reviewed-by: Atish Patra <atish.patra@wdc.com> Tested-by: Wende Tan <twd2.me@gmail.com> Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
2021-07-16 21:40:51 +08:00
} else if (task == NULL || task == current) {
riscv: eliminate unreliable __builtin_frame_address(1) I tried different pieces of code which uses __builtin_frame_address(1) (with both gcc version 7.5.0 and 10.3.0) to verify whether it works as expected on riscv64. The result is negative. What the compiler had generated is as below: 31 fp = (unsigned long)__builtin_frame_address(1); 0xffffffff80006024 <+200>: ld s1,0(s0) It takes '0(s0)' as the address of frame 1 (caller), but the actual address should be '-16(s0)'. | ... | <-+ +-----------------+ | | return address | | | previous fp | | | saved registers | | | local variables | | $fp --> | ... | | +-----------------+ | | return address | | | previous fp --------+ | saved registers | $sp --> | local variables | +-----------------+ This leads the kernel can not dump the full stack trace on riscv. [ 7.222126][ T1] Call Trace: [ 7.222804][ T1] [<ffffffff80006058>] dump_backtrace+0x2c/0x3a This problem is not exposed on most riscv builds just because the '0(s0)' occasionally is the address frame 2 (caller's caller), if only ra and fp are stored in frame 1 (caller). | ... | <-+ +-----------------+ | | return address | | $fp --> | previous fp | | +-----------------+ | | return address | | | previous fp --------+ | saved registers | $sp --> | local variables | +-----------------+ This could be a *bug* of gcc that should be fixed. But as noted in gcc manual "Calling this function with a nonzero argument can have unpredictable effects, including crashing the calling program.", let's remove the '__builtin_frame_address(1)' in backtrace code. With this fix now it can show full stack trace: [ 10.444838][ T1] Call Trace: [ 10.446199][ T1] [<ffffffff8000606c>] dump_backtrace+0x2c/0x3a [ 10.447711][ T1] [<ffffffff800060ac>] show_stack+0x32/0x3e [ 10.448710][ T1] [<ffffffff80a005c0>] dump_stack_lvl+0x58/0x7a [ 10.449941][ T1] [<ffffffff80a005f6>] dump_stack+0x14/0x1c [ 10.450929][ T1] [<ffffffff804c04ee>] ubsan_epilogue+0x10/0x5a [ 10.451869][ T1] [<ffffffff804c092e>] __ubsan_handle_load_invalid_value+0x6c/0x78 [ 10.453049][ T1] [<ffffffff8018f834>] __pagevec_release+0x62/0x64 [ 10.455476][ T1] [<ffffffff80190830>] truncate_inode_pages_range+0x132/0x5be [ 10.456798][ T1] [<ffffffff80190ce0>] truncate_inode_pages+0x24/0x30 [ 10.457853][ T1] [<ffffffff8045bb04>] kill_bdev+0x32/0x3c ... Signed-off-by: Changbin Du <changbin.du@gmail.com> Fixes: eac2f3059e02 ("riscv: stacktrace: fix the riscv stacktrace when CONFIG_FRAME_POINTER enabled") Cc: stable@vger.kernel.org Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
2022-01-17 23:44:33 +08:00
fp = (unsigned long)__builtin_frame_address(0);
sp = current_stack_pointer;
riscv: eliminate unreliable __builtin_frame_address(1) I tried different pieces of code which uses __builtin_frame_address(1) (with both gcc version 7.5.0 and 10.3.0) to verify whether it works as expected on riscv64. The result is negative. What the compiler had generated is as below: 31 fp = (unsigned long)__builtin_frame_address(1); 0xffffffff80006024 <+200>: ld s1,0(s0) It takes '0(s0)' as the address of frame 1 (caller), but the actual address should be '-16(s0)'. | ... | <-+ +-----------------+ | | return address | | | previous fp | | | saved registers | | | local variables | | $fp --> | ... | | +-----------------+ | | return address | | | previous fp --------+ | saved registers | $sp --> | local variables | +-----------------+ This leads the kernel can not dump the full stack trace on riscv. [ 7.222126][ T1] Call Trace: [ 7.222804][ T1] [<ffffffff80006058>] dump_backtrace+0x2c/0x3a This problem is not exposed on most riscv builds just because the '0(s0)' occasionally is the address frame 2 (caller's caller), if only ra and fp are stored in frame 1 (caller). | ... | <-+ +-----------------+ | | return address | | $fp --> | previous fp | | +-----------------+ | | return address | | | previous fp --------+ | saved registers | $sp --> | local variables | +-----------------+ This could be a *bug* of gcc that should be fixed. But as noted in gcc manual "Calling this function with a nonzero argument can have unpredictable effects, including crashing the calling program.", let's remove the '__builtin_frame_address(1)' in backtrace code. With this fix now it can show full stack trace: [ 10.444838][ T1] Call Trace: [ 10.446199][ T1] [<ffffffff8000606c>] dump_backtrace+0x2c/0x3a [ 10.447711][ T1] [<ffffffff800060ac>] show_stack+0x32/0x3e [ 10.448710][ T1] [<ffffffff80a005c0>] dump_stack_lvl+0x58/0x7a [ 10.449941][ T1] [<ffffffff80a005f6>] dump_stack+0x14/0x1c [ 10.450929][ T1] [<ffffffff804c04ee>] ubsan_epilogue+0x10/0x5a [ 10.451869][ T1] [<ffffffff804c092e>] __ubsan_handle_load_invalid_value+0x6c/0x78 [ 10.453049][ T1] [<ffffffff8018f834>] __pagevec_release+0x62/0x64 [ 10.455476][ T1] [<ffffffff80190830>] truncate_inode_pages_range+0x132/0x5be [ 10.456798][ T1] [<ffffffff80190ce0>] truncate_inode_pages+0x24/0x30 [ 10.457853][ T1] [<ffffffff8045bb04>] kill_bdev+0x32/0x3c ... Signed-off-by: Changbin Du <changbin.du@gmail.com> Fixes: eac2f3059e02 ("riscv: stacktrace: fix the riscv stacktrace when CONFIG_FRAME_POINTER enabled") Cc: stable@vger.kernel.org Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
2022-01-17 23:44:33 +08:00
pc = (unsigned long)walk_stackframe;
} else {
/* task blocked in __switch_to */
fp = task->thread.s[0];
sp = task->thread.sp;
pc = task->thread.ra;
}
for (;;) {
unsigned long low, high;
struct stackframe *frame;
riscv: eliminate unreliable __builtin_frame_address(1) I tried different pieces of code which uses __builtin_frame_address(1) (with both gcc version 7.5.0 and 10.3.0) to verify whether it works as expected on riscv64. The result is negative. What the compiler had generated is as below: 31 fp = (unsigned long)__builtin_frame_address(1); 0xffffffff80006024 <+200>: ld s1,0(s0) It takes '0(s0)' as the address of frame 1 (caller), but the actual address should be '-16(s0)'. | ... | <-+ +-----------------+ | | return address | | | previous fp | | | saved registers | | | local variables | | $fp --> | ... | | +-----------------+ | | return address | | | previous fp --------+ | saved registers | $sp --> | local variables | +-----------------+ This leads the kernel can not dump the full stack trace on riscv. [ 7.222126][ T1] Call Trace: [ 7.222804][ T1] [<ffffffff80006058>] dump_backtrace+0x2c/0x3a This problem is not exposed on most riscv builds just because the '0(s0)' occasionally is the address frame 2 (caller's caller), if only ra and fp are stored in frame 1 (caller). | ... | <-+ +-----------------+ | | return address | | $fp --> | previous fp | | +-----------------+ | | return address | | | previous fp --------+ | saved registers | $sp --> | local variables | +-----------------+ This could be a *bug* of gcc that should be fixed. But as noted in gcc manual "Calling this function with a nonzero argument can have unpredictable effects, including crashing the calling program.", let's remove the '__builtin_frame_address(1)' in backtrace code. With this fix now it can show full stack trace: [ 10.444838][ T1] Call Trace: [ 10.446199][ T1] [<ffffffff8000606c>] dump_backtrace+0x2c/0x3a [ 10.447711][ T1] [<ffffffff800060ac>] show_stack+0x32/0x3e [ 10.448710][ T1] [<ffffffff80a005c0>] dump_stack_lvl+0x58/0x7a [ 10.449941][ T1] [<ffffffff80a005f6>] dump_stack+0x14/0x1c [ 10.450929][ T1] [<ffffffff804c04ee>] ubsan_epilogue+0x10/0x5a [ 10.451869][ T1] [<ffffffff804c092e>] __ubsan_handle_load_invalid_value+0x6c/0x78 [ 10.453049][ T1] [<ffffffff8018f834>] __pagevec_release+0x62/0x64 [ 10.455476][ T1] [<ffffffff80190830>] truncate_inode_pages_range+0x132/0x5be [ 10.456798][ T1] [<ffffffff80190ce0>] truncate_inode_pages+0x24/0x30 [ 10.457853][ T1] [<ffffffff8045bb04>] kill_bdev+0x32/0x3c ... Signed-off-by: Changbin Du <changbin.du@gmail.com> Fixes: eac2f3059e02 ("riscv: stacktrace: fix the riscv stacktrace when CONFIG_FRAME_POINTER enabled") Cc: stable@vger.kernel.org Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
2022-01-17 23:44:33 +08:00
if (unlikely(!__kernel_text_address(pc) || (level++ >= 1 && !fn(arg, pc))))
break;
/* Validate frame pointer */
low = sp + sizeof(struct stackframe);
high = ALIGN(sp, THREAD_SIZE);
if (unlikely(fp < low || fp > high || fp & 0x7))
break;
/* Unwind stack frame */
frame = (struct stackframe *)fp - 1;
sp = fp;
if (regs && (regs->epc == pc) && (frame->fp & 0x7)) {
fp = frame->ra;
pc = regs->ra;
} else {
fp = frame->fp;
pc = ftrace_graph_ret_addr(current, NULL, frame->ra,
(unsigned long *)(fp - 8));
}
}
}
#else /* !CONFIG_FRAME_POINTER */
void notrace walk_stackframe(struct task_struct *task,
struct pt_regs *regs, bool (*fn)(void *, unsigned long), void *arg)
{
unsigned long sp, pc;
unsigned long *ksp;
if (regs) {
sp = user_stack_pointer(regs);
pc = instruction_pointer(regs);
} else if (task == NULL || task == current) {
sp = current_stack_pointer;
pc = (unsigned long)walk_stackframe;
} else {
/* task blocked in __switch_to */
sp = task->thread.sp;
pc = task->thread.ra;
}
if (unlikely(sp & 0x7))
return;
ksp = (unsigned long *)sp;
while (!kstack_end(ksp)) {
if (__kernel_text_address(pc) && unlikely(!fn(arg, pc)))
break;
pc = (*ksp++) - 0x4;
}
}
#endif /* CONFIG_FRAME_POINTER */
static bool print_trace_address(void *arg, unsigned long pc)
{
const char *loglvl = arg;
print_ip_sym(loglvl, pc);
return true;
}
noinline void dump_backtrace(struct pt_regs *regs, struct task_struct *task,
const char *loglvl)
{
walk_stackframe(task, regs, print_trace_address, (void *)loglvl);
}
void show_stack(struct task_struct *task, unsigned long *sp, const char *loglvl)
{
pr_cont("%sCall Trace:\n", loglvl);
dump_backtrace(NULL, task, loglvl);
}
static bool save_wchan(void *arg, unsigned long pc)
{
if (!in_sched_functions(pc)) {
unsigned long *p = arg;
*p = pc;
return false;
}
return true;
}
unsigned long __get_wchan(struct task_struct *task)
{
unsigned long pc = 0;
if (!try_get_task_stack(task))
return 0;
walk_stackframe(task, NULL, save_wchan, &pc);
put_task_stack(task);
return pc;
}
noinline void arch_stack_walk(stack_trace_consume_fn consume_entry, void *cookie,
struct task_struct *task, struct pt_regs *regs)
{
walk_stackframe(task, regs, consume_entry, cookie);
}