diff options
| -rw-r--r-- | HISTORY | 3 | ||||
| -rw-r--r-- | dxr3memcpy.c | 54 |
2 files changed, 36 insertions, 21 deletions
@@ -243,6 +243,9 @@ NOTE: I havent found time to include all of the languages, will be done in pre2 (Christian Gmeiner, Paavo Hartikainen) - added many comments into source (Christian Gmeiner) - using doxygen for docs (Christian Gmeiner) +- extended cDxr3MemcpyBench::Rdtsc(uint32_t config_flags): support for + non-x86 archs, support for cpu's, which dont support rdtsc timing + (Christian Gmeiner) - added support for VDR 1.3.13 and later (Luca Olivetti, Peter Dittmann) - removed -lz from makefile (Christian Gmeiner) - compiles now with 3.4.x gcc's (Christian Gmeiner, Ville Skyttä) diff --git a/dxr3memcpy.c b/dxr3memcpy.c index 979d2c4..374cc32 100644 --- a/dxr3memcpy.c +++ b/dxr3memcpy.c @@ -34,6 +34,8 @@ #include "dxr3log.h" #include "dxr3cpu.h" #include "dxr3memcpy.h" +#include <sys/times.h> +#include <limits.h> // ================================== @@ -43,6 +45,9 @@ void *(* dxr3_memcpy)(void *to, const void *from, size_t len); #if defined(__i386__) || defined(__x86_64__) // ================================== // for small memory blocks (<256 bytes) this version is faster +#define small_memcpy(to,from,n) { register unsigned long int dummy; __asm__ __volatile__("rep; movsb":"=&D"(to), "=&S"(from), "=&c"(dummy) :"0" (to), "1" (from),"2" (n) : "memory"); } +/* +// -- doesn't compile with 2.95 gcc -- #define small_memcpy(to,from,n)\ {\ register unsigned long int dummy;\ @@ -52,9 +57,9 @@ __asm__ __volatile__(\ :"0" (to), "1" (from),"2" (n)\ : "memory");\ } - +*/ // ================================== -// linux kernel __memcpy (from: /include/asm/string.h) +//! linux kernel __memcpy (from: /include/asm/string.h) static __inline__ void * __memcpy ( void * to, const void * from, @@ -141,7 +146,7 @@ static void * sse_memcpy(void * to, const void * from, size_t len) "movntps %%xmm1, 16(%1)\n" "movntps %%xmm2, 32(%1)\n" "movntps %%xmm3, 48(%1)\n" - :: "r" (from), "r" (to) : "memory"); + : : "r" (from), "r" (to) : "memory"); from = ((const unsigned char *)from) + 64; to = ((unsigned char *)to) + 64; } @@ -164,15 +169,15 @@ static void * sse_memcpy(void * to, const void * from, size_t len) "movntps %%xmm1, 16(%1)\n" "movntps %%xmm2, 32(%1)\n" "movntps %%xmm3, 48(%1)\n" - :: "r" (from), "r" (to) : "memory"); + : : "r" (from), "r" (to) : "memory"); from = ((const unsigned char *)from) + 64; to = ((unsigned char *)to) + 64; } /* since movntq is weakly-ordered, a "sfence" * is needed to become ordered again. */ - __asm__ __volatile__ ("sfence":::"memory"); + __asm__ __volatile__ ("sfence": : :"memory"); /* enables to use FPU */ - __asm__ __volatile__ ("emms":::"memory"); + __asm__ __volatile__ ("emms": : :"memory"); } /* * Now do the tail of the block @@ -220,11 +225,11 @@ static void * mmx_memcpy(void * to, const void * from, size_t len) "movq %%mm5, 40(%1)\n" "movq %%mm6, 48(%1)\n" "movq %%mm7, 56(%1)\n" - :: "r" (from), "r" (to) : "memory"); + : : "r" (from), "r" (to) : "memory"); from = ((const unsigned char *)from) + 64; to = ((unsigned char *)to) + 64; } - __asm__ __volatile__ ("emms":::"memory"); + __asm__ __volatile__ ("emms": : :"memory"); } /* * Now do the tail of the block @@ -252,7 +257,7 @@ static void * mmx2_memcpy(void * to, const void * from, size_t len) " prefetchnta 224(%0)\n" " prefetchnta 256(%0)\n" " prefetchnta 288(%0)\n" - :: "r" (from) ); + : : "r" (from) ); if(len >= MIN_LEN) { @@ -288,14 +293,14 @@ static void * mmx2_memcpy(void * to, const void * from, size_t len) "movntq %%mm5, 40(%1)\n" "movntq %%mm6, 48(%1)\n" "movntq %%mm7, 56(%1)\n" - :: "r" (from), "r" (to) : "memory"); + : : "r" (from), "r" (to) : "memory"); from = ((const unsigned char *)from) + 64; to = ((unsigned char *)to) + 64; } /* since movntq is weakly-ordered, a "sfence" * is needed to become ordered again. */ - __asm__ __volatile__ ("sfence":::"memory"); - __asm__ __volatile__ ("emms":::"memory"); + __asm__ __volatile__ ("sfence": : :"memory"); + __asm__ __volatile__ ("emms": : :"memory"); } /* * Now do the tail of the block @@ -388,13 +393,13 @@ cDxr3MemcpyBench::cDxr3MemcpyBench(uint32_t config_flags) } // count 100 runs of the memcpy function - t = Rdtsc(); + t = Rdtsc(config_flags); for (j = 0; j < 50; j++) { m_methods[i].function(buf2,buf1,BUFSIZE); m_methods[i].function(buf1,buf2,BUFSIZE); } - t = Rdtsc() - t; + t = Rdtsc(config_flags) - t; m_methods[i].time = t; @@ -417,15 +422,22 @@ cDxr3MemcpyBench::cDxr3MemcpyBench(uint32_t config_flags) // ================================== //! needed for exact timing -unsigned long long int cDxr3MemcpyBench::Rdtsc() +unsigned long long int cDxr3MemcpyBench::Rdtsc(uint32_t config_flags) { #if defined(__i386__) || defined(__x86_64__) - unsigned long long int x; - __asm__ volatile (".byte 0x0f, 0x31" : "=A" (x)); - return x; + // we need rdtsc support + if (config_flags && CC_MMX) + { + unsigned long long int x; + __asm__ volatile (".byte 0x0f, 0x31" : "=A" (x)); + return x; + } + else + { + return times(NULL); + } #else - /* FIXME: implement an equivalent for using optimized memcpy on other - architectures */ - return 0; + struct tms tp; + return times(&tp); #endif /* __i386__ || __x86_64__ */ } |