gdb定位linux程序栈越界异常方法

现象：用gdb跑程序出现以下异常，怀疑变量过多导致栈越界

Program received signal SIGSEGV, Segmentation fault.

0x0000000000a1ef2d in AESL_WRAP_load_1_n (

insn=<error reading variable: Cannot access memory at address 0x7fffff796c78>,

channel=<error reading variable: Cannot access memory at address 0x7fffff796c70>,

pu_in_buf=<error reading variable: Cannot access memory at address 0x7fffff796c68>,

config=<error reading variable: Cannot access memory at address 0x7fffff796c60>,

buffer=<error reading variable: Cannot access memory at address 0x7fffff796c58>,

insn_pu=<error reading variable: Cannot access memory at address 0x7fffff796c50>)

at apatb_load_1_n.cpp:20219

定位方法：

1、设置main的断点，断住查询寄存器

(gdb) b main

(gdb) info registers

rax 0x53d106 5492998

rbx 0x0 0

rcx 0x340 832

rdx 0x7fffffffd238 140737488343608

rsi 0x7fffffffd228 140737488343592

rdi 0x1 1

rbp 0x7fffffffd140 0x7fffffffd140

rsp 0x7fffffffd030 0x7fffffffd030

r8 0x1ee24620 518145568

r9 0x0 0

r10 0x0 0

r11 0x6e72656b24242424 7958535002359342116

r12 0x40834e 4227918

r13 0x7fffffffd220 140737488343584

r14 0x0 0

r15 0x0 0

rip 0x53d11f 0x53d11f <main(int, char const**)+25>

eflags 0x206 [ PF IF ]

cs 0x33 51

ss 0x2b 43

ds 0x0 0

es 0x0 0

fs 0x0 0

gs 0x0 0

2.异常时打印寄存器

(gdb) info registers

rax 0x7ffffffee500 140737488282880

rbx 0x7ffffffee620 140737488283168

rcx 0xe6d2f20 242036512

rdx 0xc91a280 210870912

rsi 0xe4a4060 239747168

rdi 0x7ffffffee500 140737488282880

rbp 0x7ffffffee370 0x7ffffffee370

rsp 0x7fffff796c50 0x7fffff796c50

r8 0x18923f20 412237600

r9 0x7ffffffee430 140737488282672

r10 0x0 0

r11 0x0 0

r12 0xffffffffffffffff -1

r13 0x7ffffffee670 140737488283248

r14 0x0 0

r15 0x18485680 407393920

rip 0xa1ef2d 0xa1ef2d <AESL_WRAP_load_1_n(InstructionLoad&, Channel&, Load2PU*, LCUConfig&, LCUInput&, InstructionPU*)+18>

eflags 0x10206 [ PF IF RF ]

cs 0x33 51

ss 0x2b 43

ds 0x0 0

es 0x0 0

fs 0x0 0

gs 0x0 0

3.计算栈的大约大小。0x7fffffffd030 - 0x7fffff796c50=0x8663E0

4.查询默认的栈大小,只有8M，而当前的栈大小已超过为8M多，越界。

[wangjsh@localhost /]$ ulimit -s [110/110]

8192

5.修改当前栈大小，运行成功，不会出现异常

[wangjsh@localhost /]$ ulimit -s 102400

[wangjsh@localhost /]$ ulimit -s

102400