arm64汇编学习

arm64汇编学习

arm64汇编学习-（3）算术与移位指令

• 1 数据处理指令
• • 1.1 check the C condition of adds, adc,cmp
  • • 1.1.1 测试示例程序
    • 1.1.2 执行之前
    • 1.1.3 执行之后
    • • 1.1.3.1 ldr和mov指令之后
      • 1.1.3.2 ads和adc指令之后
      • 1.1.3.3 cmp和adc指令之后
  • 1.2 cmp和sbc指令的综合运用
  • • 1.2.1 示例代码
    • 1.2.2 compare_and_return(10, 9)测试之前
    • 1.2.3 compare_and_return(10, 9)测试之后
    • 1.2.4 compare_and_return(9, 10)测试之前
    • 1.2.5 compare_and_return(9, 10)测试之后
  • 1.3 测试ands指令对Z标志位的影响
  • • 1.3.2 测试之前
    • 1.3.3 测试之后
  • 1.4 测试位段bitfield指令
  • • 1.4.1 示例代码
    • 1.4.2 测试之前
    • 1.4.3 bfi x0, x1, 8, 4之后
    • 1.4.4 ubfx and sbfx之后
  • 1.5 读取寄存器的域
  • • 1.5.1 示例代码

本篇博客是基于对苯叔的第三季视频的学习整理而得，大家如果想深入学习可以购买《arm64体系结构编程与实践》以及购买苯叔出品的第三季视频。

1 数据处理指令

1.1 check the C condition of adds, adc,cmp

1.1.1 测试示例程序

/*测试adds, cmp以及adc指令的条件标志位
*/
.global add_inst_test
add_inst_test:mov x0, #0ldr x1, =0xffffffffffffffffmov x2, #3/*测试adds的进位功能，当有溢出发生时，C=1*/adds x0, x1, x1adc x3, xzr, xzr/*测试cmp的条件标志位，当x1 > x2时，C=1当x1 < x2是， C=0*/cmp x1, x2adc x4, xzr, xzrret

1.1.2 执行之前

1.1.3 执行之后

1.1.3.1 ldr和mov指令之后

1.1.3.2 ads和adc指令之后

ADDS: Add (extended register), setting flags, adds a register value and a sign or zero-extended register value, followed by an optional left shift amount, and writes the result to the destination register. The argument that is extended from the register can be a byte, halfword, word, or doubleword. It updates the condition flags based on the result.

• ADDS , <Xn|SP>, {, {#}}
• (result, nzcv) = AddWithCarry(operand1, operand2, ‘0’);
• PSTATE.<N,Z,C,V> = nzcv;
• X[d] = result;
  ADC：Add with Carry adds two register values and the Carry flag value, and writes the result to the destination register.
• (result, -) = AddWithCarry(operand1, operand2, PSTATE.C);
• X[d] = result;
  

1.1.3.3 cmp和adc指令之后

CMP:This instruction is an alias of the SUBS (shifted register) instruction.

• The encodings in this description are named to match the encodings of SUBS (shifted register).
• The description of SUBS (shifted register) gives the operational pseudocode for this instruction.
  

1.2 cmp和sbc指令的综合运用

1.2.1 示例代码

ret = compare_and_return(10, 9);
val = compare_and_return(9, 10);/*当arg1 >= arg2 时， 返回 0当arg1 < arg2 时，返回0xffffffffffffffff
*/
.global compare_and_return
compare_and_return:cmp x0, x1sbc x0, xzr, xzrret

1.2.2 compare_and_return(10, 9)测试之前

1.2.3 compare_and_return(10, 9)测试之后

1.2.4 compare_and_return(9, 10)测试之前

1.2.5 compare_and_return(9, 10)测试之后

1.3 测试ands指令对Z标志位的影响

2.1.3.1 示例代码

/*data_process_instr lab3: 测试ands指令对Z标志位的影响
*/
.global ands_test
ands_test:mov x1, #0x3mov x2, #0//mov x2, #2ands x3, x1, x2/*读取NZCV寄存器，查看Z标志位是否为1， 见armv8.6手册第C5.2.9章*/mrs x0, nzcvret

1.3.2 测试之前

1.3.3 测试之后

1.4 测试位段bitfield指令

1.4.1 示例代码

/*data_process_instr lab4: 测试位段bitfield指令*/
.global bitfield_test
bitfield_test:/* 位段插入*/mov x1, 0x345mov x0, 0bfi x0, x1, 8, 4/*无符号数的bitfield提取，提取x2，从第4个bit开始，提取8位,x3其他位都是0，最终：0xbc*/ldr x2, =0x5678abcdubfx x3, x2, #4, #8/*有符号数的bitfield提取，提取x2，从第4个bit开始，提取8bit。x4的其他比特位都是f， 最终：0xffffffffffffffbc*/sbfx x4, x2, #4, #8ret

1.4.2 测试之前

1.4.3 bfi x0, x1, 8, 4之后

BFI ：Bitfield Insert copies a bitfield of bits from the least significant bits of the source register to bit position of the destination register, leaving the other destination bits unchanged.

• bfi x0, x1, 8, 4 该汇编指令的含义为：将x1寄存器的最低4位拷贝到x0寄存器的bit 8开始的4位上去，x0寄存器的其他位不变，所以最终x0寄存器的值被设置为0x500
  

1.4.4 ubfx and sbfx之后

UBFX ：Unsigned Bitfield Extract copies a bitfield of bits starting from bit position in the source register to the least significant bits of the destination register, and sets destination bits above the bitfield to zero.

• UBFX , , #, #
  SBFX ：Signed Bitfield Extract copies a bitfield of bits starting from bit position in the source register to the least significant bits of the destination register, and sets destination bits above the bitfield to a copy of the most significant bit of the bitfield.
• SBFX , , #, #
  This content is only supported in a Docs.
  
  

1.5 读取寄存器的域

1.5.1 示例代码

/*data_process_instr lab4: 测试位段bitfield指令*/
.global bitfield_test
bitfield_test:/* 位段插入*/mov x1, 0x345mov x0, 0bfi x0, x1, 8, 4/*无符号数的bitfield提取，提取x2，从第4个bit开始，提取8位,x3其他位都是0，最终：0xbc*/ldr x2, =0x5678abcdubfx x3, x2, #4, #8/*有符号数的bitfield提取，提取x2，从第4个bit开始，提取8bit。x4的其他比特位都是f， 最终：0xffffffffffffffbc*/sbfx x4, x2, #4, #8/*lab5: 使用ubfx指令来读取 寄存器的位域*/mrs x1, ID_AA64ISAR0_EL1/*读取atomic域的值：判断是否支持LSE指令*/ubfx x0, x1, #20, #4/*读取AES域的值：判断是否支持AES指令*/ubfx x2, x1, #4, #4ret

2.5.1.2 测试之前的状态

2.5.1.3 测试之后的状态值