Mirrored_Repos/LookingGlass
1
0
Fork 0
mirror of https://github.com/gnif/LookingGlass.git synced 2024-11-22 05:27:20 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

[common] more SSE improvements

Browse Source 
* 32bit inlined is slow for only large copies, warn if memcpySSE is
used when it shouldn't be.

* Removed 64bit memcpySSE as native inlined is faster

See: https://stackoverflow.com/questions/50422510/why-is-i386-memcpy-slow-on-x86-64
This commit is contained in:
Geoffrey McRae 2018-05-19 18:27:04 +10:00
parent 778af24d82
commit 56f0a8525b
1 changed files with 104 additions and 168 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

118
common/memcpySSE2.h
View File

@ -26,30 +26,32 @@ Place, Suite 330, Boston, MA 02111-1307 USA
#include "debug.h"
#if defined(__GNUC__) || defined(__GNUG__)
#define OP(...) #__VA_ARGS__ "\n\t"
inline static void memcpySSE(void *dst, const void * src, size_t length)
{
#if !defined(NATIVE_MEMCPY) && (defined(__x86_64__) || defined(__i386__))
#if defined(NATIVE_MEMCPY)
#define memcpySSE memcpy
#elif defined(MSVC)
extern "C" void memcpySSE(void *dst, const void * src, size_t length);
#elif (defined(__GNUC__) || defined(__GNUG__)) && defined(__i386__)
inline static void memcpySSE(void *dst, const void * src, size_t length)
{
if (length == 0 || dst == src)
return;
#ifdef __x86_64__
const void * end = dst + (length & ~0xFF);
size_t off = (15 - ((length & 0xFF) >> 4));
off = (off < 8) ? off * 16 : 7 * 16 + (off - 7) * 10;
#else
// copies under 1MB are faster with the inlined memcpy
// tell the dev to use that instead
if (length < 1048576)
{
static bool smallBufferWarn = false;
if (!smallBufferWarn)
{
DEBUG_WARN("Do not use memcpySSE for copies under 1MB in size!");
smallBufferWarn = true;
}
memcpy(dst, src, length);
return;
}
const void * end = dst + (length & ~0x7F);
const size_t off = (7 - ((length & 0x7F) >> 4)) * 10;
#endif
#ifdef __x86_64__
#define REG "rax"
#else
#define REG "eax"
#endif
__asm__ __volatile__ (
"cmp %[dst],%[end] \n\t"
@ -65,16 +67,6 @@ inline static void memcpySSE(void *dst, const void * src, size_t length)
"vmovaps 0x50(%[src]),%%xmm5 \n\t"
"vmovaps 0x60(%[src]),%%xmm6 \n\t"
"vmovaps 0x70(%[src]),%%xmm7 \n\t"
#ifdef __x86_64__
"vmovaps 0x80(%[src]),%%xmm8 \n\t"
"vmovaps 0x90(%[src]),%%xmm9 \n\t"
"vmovaps 0xA0(%[src]),%%xmm10 \n\t"
"vmovaps 0xB0(%[src]),%%xmm11 \n\t"
"vmovaps 0xC0(%[src]),%%xmm12 \n\t"
"vmovaps 0xD0(%[src]),%%xmm13 \n\t"
"vmovaps 0xE0(%[src]),%%xmm14 \n\t"
"vmovaps 0xF0(%[src]),%%xmm15 \n\t"
#endif
"vmovntdq %%xmm0 ,0x00(%[dst]) \n\t"
"vmovntdq %%xmm1 ,0x10(%[dst]) \n\t"
"vmovntdq %%xmm2 ,0x20(%[dst]) \n\t"
@ -83,63 +75,25 @@ inline static void memcpySSE(void *dst, const void * src, size_t length)
"vmovntdq %%xmm5 ,0x50(%[dst]) \n\t"
"vmovntdq %%xmm6 ,0x60(%[dst]) \n\t"
"vmovntdq %%xmm7 ,0x70(%[dst]) \n\t"
#ifdef __x86_64__
"vmovntdq %%xmm8 ,0x80(%[dst]) \n\t"
"vmovntdq %%xmm9 ,0x90(%[dst]) \n\t"
"vmovntdq %%xmm10,0xA0(%[dst]) \n\t"
"vmovntdq %%xmm11,0xB0(%[dst]) \n\t"
"vmovntdq %%xmm12,0xC0(%[dst]) \n\t"
"vmovntdq %%xmm13,0xD0(%[dst]) \n\t"
"vmovntdq %%xmm14,0xE0(%[dst]) \n\t"
"vmovntdq %%xmm15,0xF0(%[dst]) \n\t"
"add $0x100,%[dst] \n\t"
"add $0x100,%[src] \n\t"
#else
"add $0x80,%[dst] \n\t"
"add $0x80,%[src] \n\t"
#endif
"cmp %[dst],%[end] \n\t"
"jne loop_%= \n\t"
"Remain_%=: \n\t"
// copy any remaining 16 byte blocks
#ifdef __x86_64__
"leaq (%%rip), %%rax\n\t"
#else
"call GetPC_%=\n\t"
#endif
"Offset_%=:\n\t"
"add $(BlockTable_%= - Offset_%=), %%" REG "\n\t"
"add %[off],%%" REG " \n\t"
"jmp *%%" REG " \n\t"
"add $(BlockTable_%= - Offset_%=), %%eax \n\t"
"add %[off],%%eax \n\t"
"jmp *%%eax \n\t"
#ifdef __i386__
"GetPC_%=:\n\t"
"mov (%%esp), %%eax \n\t"
"ret \n\t"
#endif
"BlockTable_%=:\n\t"
#ifdef __x86_64__
"vmovaps 0xE0(%[src]),%%xmm14 \n\t"
"vmovntdq %%xmm14,0xE0(%[dst]) \n\t"
"vmovaps 0xD0(%[src]),%%xmm13 \n\t"
"vmovntdq %%xmm13,0xD0(%[dst]) \n\t"
"vmovaps 0xC0(%[src]),%%xmm12 \n\t"
"vmovntdq %%xmm12,0xC0(%[dst]) \n\t"
"vmovaps 0xB0(%[src]),%%xmm11 \n\t"
"vmovntdq %%xmm11,0xB0(%[dst]) \n\t"
"vmovaps 0xA0(%[src]),%%xmm10 \n\t"
"vmovntdq %%xmm10,0xA0(%[dst]) \n\t"
"vmovaps 0x90(%[src]),%%xmm9 \n\t"
"vmovntdq %%xmm9 ,0x90(%[dst]) \n\t"
"vmovaps 0x80(%[src]),%%xmm8 \n\t"
"vmovntdq %%xmm8 ,0x80(%[dst]) \n\t"
"vmovaps 0x70(%[src]),%%xmm7 \n\t"
"vmovntdq %%xmm7 ,0x70(%[dst]) \n\t"
#endif
"vmovaps 0x60(%[src]),%%xmm6 \n\t"
"vmovntdq %%xmm6 ,0x60(%[dst]) \n\t"
"vmovaps 0x50(%[src]),%%xmm5 \n\t"
@ -161,7 +115,7 @@ inline static void memcpySSE(void *dst, const void * src, size_t length)
[src]"+r" (src)
: [off]"r" (off),
[end]"r" (end)
: REG,
: "eax",
"xmm0",
"xmm1",
"xmm2",
@ -170,32 +124,14 @@ inline static void memcpySSE(void *dst, const void * src, size_t length)
"xmm5",
"xmm6",
"xmm7",
#ifdef __x86_64__
"xmm8",
"xmm9",
"xmm10",
"xmm11",
"xmm12",
"xmm13",
"xmm14",
"xmm15",
"rax",
#else
"eax",
#endif
"memory"
);
#undef REG
//copy any remaining bytes
for(size_t i = (length & 0xF); i; --i)
((uint8_t *)dst)[length - i] =
((uint8_t *)src)[length - i];
}
#else
memcpy(dst, src, length);
#endif
}
#else
extern "C" void memcpySSE(void *dst, const void * src, size_t length);
#define memcpySSE memcpy
#endif