优化打印计数器的循环

时间:2016-12-07 22:17:24

标签: linux assembly optimization nasm x86-64

我有一个非常小的循环程序,可以打印从5000000到1的数字。我想让它尽可能快地运行。

我正在学习使用NASM的linux x86-64程序集。

global  main
extern  printf
main:
    push    rbx                     
    mov rax,5000000d
print:
    push    rax                     
    push    rcx                     
    mov     rdi, format             
    mov     rsi, rax                
    call    printf                  
    pop     rcx                     
    pop     rax                     
    dec     rax                     
    jnz     print                   
    pop     rbx                     
    ret

format:
db  "%ld", 10, 0

2 个答案:

答案 0 :(得分:3)

对printf的调用完全支配了即使是非常低效的循环的运行时间。 (你有没有注意到你推/弹rcx,即使你从来没有在任何地方使用它?也许是因为使用the slow LOOP instruction而遗留下来的。)

要了解有关编写高效x86 asm的更多信息,请参阅Agner Fog's Optimizing Assembly guide。 (以及他的微体系结构指南,如果你想深入了解特定CPU的细节以及它们之间的区别:在一个uarch CPU上最佳的可能不在另一个上。例如,IMUL r64在英特尔上具有更好的吞吐量和延迟CPU比AMD要好,但CMOV和ADC在英特尔前Broadwell上是2 uop,2周期延迟。与AMD相比为1,因为3输入ALU m-ops(FLAGS +两个寄存器)对AMD来说不是问题。)另请参阅标记wiki中的其他链接。

纯粹优化循环而不更改对printf的5M调用仅作为如何正确编写循环的示例,而不是实际加速此代码。但让我们从那开始:

; trivial fixes to loop efficiently while calling the same slow function
global  main
extern  printf
main:
    push    rbx
    mov     ebx, 5000000         ; don't waste a REX prefix for constants that fit in 32 bits
.print:
    ;; removed the push/pops from inside the loop.
    ; Use call-preserved regs instead of saving/restoring stuff inside a loop yourself.
    mov     edi, format          ; static data / code always has a 32-bit address
    mov     esi, ebx
    xor     eax, eax             ; The x86-64 SysV ABI requires al = number of FP args passed in FP registers for variadic functions
    call    printf                  
    dec     ebx
    jnz     .print

    pop     rbx                ; restore rbx, the one call-preserved reg we actually used.
    xor     eax,eax            ; successful exit status.
    ret

section .rodata       ; it's usually best to put constant data in a separate section of the text segment, not right next to code.
format:
db  "%ld", 10, 0

要加快速度,我们应该利用冗余将连续整数转换为字符串。由于"5000000\n"只有8个字节长(包括换行符),因此字符串表示适合64位寄存器。

我们可以将该字符串存储到缓冲区中,并按字符串长度递增指针。 (因为对于较小的数字它会变短,只需将当前字符串长度保存在寄存器中,您可以在特殊情况分支中更新它。)

我们可以就地减少字符串表示,以避免(重新)进行除以10的过程,将整数转换为十进制字符串。

由于进位/借位不会在寄存器内自然传播,并且AAS指令在64位模式下不可用(并且仅在AX上工作,甚至不在EAX中,并且速度很慢),我们必须自己做。我们每次都减1,所以我们知道会发生什么。我们可以通过展开10次来处理最不重要的数字,因此没有分支来处理它。

另请注意,由于我们想要打印顺序中的数字,所以进位方向错误,因为x86是little-endian。如果有一个很好的方法来利用我们的字符串在另一个字节顺序,我们可以使用BSWAP或MOVBE。 (但请注意,MOVBE r64是Skylake上的3个融合域uops,其中2个是ALU uops.BSWAP r64也是2 uops。)

也许我们应该在XMM向量寄存器的两半中并行执行奇数/偶数计数器。但是一旦字符串短于8B,那就停止工作了。一次存储一个数字串,我们可以轻松重叠。我们仍然可以在向量寄存器中执行进位传播,并使用MOVQ和MOVHPS分别存储两半。或者因为从0到5M的数字的4/5是7位数,所以我们可以存储一个特殊情况的代码,我们可以存储两个数字的整个16B向量。

处理较短字符串的一种更好的方法: SSSE3 PSHUFB将两个字符串混合到一个向量寄存器中的左包装,然后单个MOVUPS一次存储两个字符串。当字符串长度(位数)发生变化时,只需要更新shuffle掩码,因此不经常执行的进位处理特殊情况代码也可以这样做。

循环的热点部分的矢量化应该非常简单和便宜,并且应该只是性能的两倍。

;;; Optimized version: keep the string data in a register and modify it
;;; instead of doing the whole int->string conversion every time.

section  .bss
printbuf:  resb 1024*128 + 4096     ;  Buffer size ~= half L2 cache size on Intel SnB-family.  Or use a giant buffer that we write() once.  Or maybe vmsplice to give it away to the kernel, since we only run once.

global  main
extern  printf
main:
    push    rbx

    ; use some REX-only regs for values that we're always going to use a REX prefix with anyway for 64-bit operand size.
    mov     rdx, `5000000\n`   ; (NASM string constants as integers work like little-endian, so AL = '5' = 0x35 and the high byte holds '\n' = 10).  Note that YASM doesn't support back-ticks for C-style backslash processing.
    mov     r9, 1<<56         ; decrement by 1 in the 2nd-last byte: LSB of the decimal string
    ;xor     r9d, r9d
    ;bts      r9, 56           ; IDK if this code-size optimization outside the loop would help or not.

    mov     eax, 8            ; string length.
    mov     edi, printbuf

.storeloop:

    ;;  rdx = "????x9\n".  We compute the start value for the next iteration, i.e. counter -= 10 in rdx.

    mov     r8, rdx
    ;;  r8 = rdx.  We modify it to have each last digit from 9 down to 0 in sequence, and store those strings in the buffer.
    ;;  The string could be any length, always with the first ASCII digit in the low byte; our other constants are adjusted correctly for it
    ;; narrower than 8B means that our stores overlap, but that's fine.

    ;; Starting from here to compute the next unrolled iteration's starting value takes the `sub r8, r9` instructions off the critical path, vs. if we started from r8 at the bottom of the loop.  This gives out-of-order execution more to play with.
    ;;  It means each loop iteration's sequence of subs and stores are a separate dependency chain (except for the store addresses, but OOO can get ahead on those because we only pointer-increment every 2 stores).

    mov     [rdi], r8
    sub     r8, r9             ; r8 = "xxx8\n"

    mov     [rdi + rax], r8    ; defer p += len by using a 2-reg addressing mode
    sub     r8, r9             ; r8 = "xxx7\n"

    lea     edi, [rdi + rax*2]  ; if we had len*3 in another reg, we could defer this longer
           ;; our static buffer is guaranteed to be in the low 31 bits of address space so we can safely save a REX prefix on the LEA here.  Normally you shouldn't truncate pointers to 32-bits, but you asked for the fastest possible.  This won't hurt, and might help on some CPUs, especially with possible decode bottlenecks.

    ;; repeat that block 3 more times.
    ;; using a short inner loop for the 9..0 last digit might be a win on some CPUs (like maybe Core2), depending on their front-end loop-buffer capabilities if the frontend is a bottleneck at all here.

    ;; anyway, then for the last one:
    mov     [rdi], r8             ; r8 = "xxx1\n"
    sub     r8, r9
    mov     [rdi + rax], r8       ; r8 = "xxx0\n"

    lea     edi, [rdi + rax*2]


    ;; compute next iteration's RDX.  It's probably a win to interleave some of this into the loop body, but out-of-order execution should do a reasonably good job here.
    mov     rcx, r9
    shr     rcx, 8      ; maybe hoist this constant out, too
    ; rcx = 1 in the second-lowest digit
    sub     rdx, rcx

    ; detect carry when '0' (0x30) - 1 = 0x2F by checking the low bit of the high nibble in that byte.
    shl     rcx, 5
    test    rdx, rcx
    jz      .carry_second_digit
    ; .carry_second_digit is some complicated code to propagate carry as far as it needs to go, up to the most-significant digit.
    ; when it's done, it re-enters the loop at the top, with eax and r9 set appropriately.
    ; it only runs once per 100 digits, so it doesn't have to be super-fast

    ; maybe only do buffer-length checks in the carry-handling branch,
    ; in which case the jz .carry  can be  jnz .storeloop
    cmp     edi, esi              ; } while(p < endp)
    jbe     .storeloop

    ; write() system call on the buffer.
    ; Maybe need a loop around this instead of doing all 5M integer-strings in one giant buffer.

    pop     rbx
    xor     eax,eax            ; successful exit status.
    ret

这并没有完全充实,但应该知道什么可能效果很好。

如果使用SSE2进行矢量化,可能会使用标量整数寄存器来跟踪何时需要突破并处理进位。即从10开始的反击。

即使这个标量版本可能接近每个时钟维持一个商店,这使商店端口饱和。它们只有8B个存储(当字符串变短时,有用的部分比这个短),所以如果我们不在缓存未命中的瓶颈,我们肯定会把性能留在桌面上。但是使用3GHz CPU和双通道DDR3-1600(理论最大带宽约为25.6GB / s),每个时钟8B大约足以使单个内核饱和主存储器。

我们可以将其并行化,并将5M .. 1范围分解为块。通过一些聪明的数学运算,我们可以找出写"2500000\n"的第一个字符的字节,或者我们可以让每个线程以正确的顺序调用write()。 (或者使用相同的聪明数学让他们独立地使用不同的文件偏移调用pwrite(2),因此内核负责处理同一文件的多个写入者的所有同步。)

答案 1 :(得分:1)

你实际上是在打印一个固定的字符串。我将该字符串预先生成一个长常量。

然后该程序成为write的单个调用(或处理不完整写入的短循环)。