RGBA 到 ABGR:适用于 iOS/Xcode 的内联臂霓虹灯 asm

问题描述

这段代码(非常相似的代码，没有尝试过完全这段代码)使用Android NDK编译，但不适用于Xcode/armv7+arm64/iOS

This code(very similar code, haven't tried exactly this code) compiles using Android NDK, but not with Xcode/armv7+arm64/iOS

评论错误:

uint32_t *src;
uint32_t *dst;

#ifdef __ARM_NEON
__asm__ volatile(
    "vld1.32 {d0, d1}, [%[src]] 
" // error: Vector register expected
    "vrev32.8 q0, q0            
" // error: Unrecognized instruction mnemonic
    "vst1.32 {d0, d1}, [%[dst]] 
" // error: Vector register expected
    :
    : [src]"r"(src), [dst]"r"(dst)
    : "d0", "d1"
    );
#endif

这段代码有什么问题?

What's wrong with this code?

我使用内在函数重写了代码:

I rewrote the code using intrinsics:

uint8x16_t x = vreinterpretq_u8_u32(vld1q_u32(src));
uint8x16_t y = vrev32q_u8(x);
vst1q_u32(dst, vreinterpretq_u32_u8(y));

拆解后，我得到以下，这是我已经尝试过的一个变体:

After disassembling, I get the following, which is a variation I have already tried:

vld1.32 {d16, d17}, [r0]!
vrev32.8    q8, q8
vst1.32 {d16, d17}, [r1]!

所以我的代码现在看起来像这样，但给出了完全相同的错误:

So my code looks like this now, but gives the exact same errors:

__asm__ volatile("vld1.32 {d0, d1}, [%0]! 
"
                 "vrev32.8 q0, q0         
"
                 "vst1.32 {d0, d1}, [%1]! 
"
                 :
                 : "r"(src), "r"(dst)
                 : "d0", "d1"
                 );

通过反汇编阅读，我实际上找到了该函数的第二个版本.事实证明，arm64 使用的指令集略有不同.例如，arm64 程序集使用 rev32.16b v0, v0 代替.整个函数列表(我无法正面或反面)如下:

Reading through the disassembly, I actually found a second version of the function. It turns out that arm64 uses a slightly different instruction set. For example, the arm64 assembly uses rev32.16b v0, v0 instead. The whole function listing(which I can't make heads or tails of) is below:

_My_Function:
cmp     w2, #0
add w9, w2, #3
csel    w8, w9, w2, lt
cmp     w9, #7
b.lo    0x3f4
asr w9, w8, #2
ldr     x8, [x0]
mov  w9, w9
lsl x9, x9, #2
ldr q0, [x8], #16
rev32.16b   v0, v0
str q0, [x1], #16
sub x9, x9, #16
cbnz    x9, 0x3e0
ret

推荐答案

正如对原始问题的编辑中所述，我需要为 arm64 和 armv7 提供不同的程序集实现.

As stated in the edits to the original question, it turned out that I needed a different assembly implementation for arm64 and armv7.

#ifdef __ARM_NEON
  #if __LP64__
asm volatile("ldr q0, [%0], #16  
"
             "rev32.16b v0, v0   
"
             "str q0, [%1], #16  
"
             : "=r"(src), "=r"(dst)
             : "r"(src), "r"(dst)
             : "d0", "d1"
             );
  #else
asm volatile("vld1.32 {d0, d1}, [%0]! 
"
             "vrev32.8 q0, q0         
"
             "vst1.32 {d0, d1}, [%1]! 
"
             : "=r"(src), "=r"(dst)
             : "r"(src), "r"(dst)
             : "d0", "d1"
             );
  #endif
#else

我在原始帖子中发布的内在函数代码生成了令人惊讶的好汇编，并且还为我生成了 arm64 版本，因此将来使用内在函数可能是一个更好的主意.

The intrinsics code that I posted in the original post generated surprisingly good assembly though, and also generated the arm64 version for me, so it may be a better idea to use intrinsics instead in the future.

这篇关于RGBA 到 ABGR:适用于 iOS/Xcode 的内联臂霓虹灯 asm的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持跟版网！