// Copyright (c) the JPEG XL Project Authors. All rights reserved. // // Use of this source code is governed by a BSD-style // license that can be found in the LICENSE file. #ifndef LIB_JXL_BASE_TSC_TIMER_H_ #define LIB_JXL_BASE_TSC_TIMER_H_ // High-resolution (~10 ns) timestamps, using fences to prevent reordering and // ensure exactly the desired regions are measured. #include #include // clock_gettime #if defined(_WIN32) || defined(_WIN64) #ifndef WIN32_LEAN_AND_MEAN #define WIN32_LEAN_AND_MEAN #endif // WIN32_LEAN_AND_MEAN #ifndef NOMINMAX #define NOMINMAX #endif // NOMINMAX #ifndef NOGDI #define NOGDI #endif // NOGDI #include // Undef macros to avoid collisions #undef LoadFence #endif #if defined(__APPLE__) #include #include #endif #if defined(__HAIKU__) #include #endif #include #include #include // LoadFence namespace jxl { namespace profiler { // Ticks := platform-specific timer values (CPU cycles on x86). Must be // unsigned to guarantee wraparound on overflow. using Ticks = uint64_t; // TicksBefore/After return absolute timestamps and must be placed immediately // before and after the region to measure. We provide separate Before/After // functions because they use different fences. // // Background: RDTSC is not 'serializing'; earlier instructions may complete // after it, and/or later instructions may complete before it. 'Fences' ensure // regions' elapsed times are independent of such reordering. The only // documented unprivileged serializing instruction is CPUID, which acts as a // full fence (no reordering across it in either direction). Unfortunately // the latency of CPUID varies wildly (perhaps made worse by not initializing // its EAX input). Because it cannot reliably be deducted from the region's // elapsed time, it must not be included in the region to measure (i.e. // between the two RDTSC). // // The newer RDTSCP is sometimes described as serializing, but it actually // only serves as a half-fence with release semantics. Although all // instructions in the region will complete before the final timestamp is // captured, subsequent instructions may leak into the region and increase the // elapsed time. Inserting another fence after the final RDTSCP would prevent // such reordering without affecting the measured region. // // Fortunately, such a fence exists. The LFENCE instruction is only documented // to delay later loads until earlier loads are visible. However, Intel's // reference manual says it acts as a full fence (waiting until all earlier // instructions have completed, and delaying later instructions until it // completes). AMD assigns the same behavior to MFENCE. // // We need a fence before the initial RDTSC to prevent earlier instructions // from leaking into the region, and arguably another after RDTSC to avoid // region instructions from completing before the timestamp is recorded. // When surrounded by fences, the additional RDTSCP half-fence provides no // benefit, so the initial timestamp can be recorded via RDTSC, which has // lower overhead than RDTSCP because it does not read TSC_AUX. In summary, // we define Before = LFENCE/RDTSC/LFENCE; After = RDTSCP/LFENCE. // // Using Before+Before leads to higher variance and overhead than After+After. // However, After+After includes an LFENCE in the region measurements, which // adds a delay dependent on earlier loads. The combination of Before+After // is faster than Before+Before and more consistent than After+After because // the first LFENCE already delayed subsequent loads before the measured // region. This combination seems not to have been considered in prior work: // http://akaros.cs.berkeley.edu/lxr/akaros/kern/arch/x86/rdtsc_test.c // // Note: performance counters can measure 'exact' instructions-retired or // (unhalted) cycle counts. The RDPMC instruction is not serializing and also // requires fences. Unfortunately, it is not accessible on all OSes and we // prefer to avoid kernel-mode drivers. Performance counters are also affected // by several under/over-count errata, so we use the TSC instead. // Returns a 64-bit timestamp in unit of 'ticks'; to convert to seconds, // divide by InvariantTicksPerSecond. static HWY_INLINE HWY_MAYBE_UNUSED Ticks TicksBefore() { Ticks t; #if HWY_ARCH_PPC && defined(__GLIBC__) asm volatile("mfspr %0, %1" : "=r"(t) : "i"(268)); #elif HWY_ARCH_X86 && HWY_COMPILER_MSVC hwy::LoadFence(); HWY_FENCE; t = __rdtsc(); hwy::LoadFence(); HWY_FENCE; #elif HWY_ARCH_X86_64 asm volatile( "lfence\n\t" "rdtsc\n\t" "shl $32, %%rdx\n\t" "or %%rdx, %0\n\t" "lfence" : "=a"(t) : // "memory" avoids reordering. rdx = TSC >> 32. // "cc" = flags modified by SHL. : "rdx", "memory", "cc"); #elif HWY_ARCH_RVV asm volatile("rdcycle %0" : "=r"(t)); #elif defined(_WIN32) || defined(_WIN64) LARGE_INTEGER counter; (void)QueryPerformanceCounter(&counter); t = counter.QuadPart; #elif defined(__APPLE__) t = mach_absolute_time(); #elif defined(__HAIKU__) t = system_time_nsecs(); // since boot #else // POSIX timespec ts; clock_gettime(CLOCK_MONOTONIC, &ts); t = static_cast(ts.tv_sec * 1000000000LL + ts.tv_nsec); #endif return t; } static HWY_INLINE HWY_MAYBE_UNUSED Ticks TicksAfter() { Ticks t; #if HWY_ARCH_PPC && defined(__GLIBC__) asm volatile("mfspr %0, %1" : "=r"(t) : "i"(268)); #elif HWY_ARCH_X86 && HWY_COMPILER_MSVC HWY_FENCE; unsigned aux; t = __rdtscp(&aux); hwy::LoadFence(); HWY_FENCE; #elif HWY_ARCH_X86_64 // Use inline asm because __rdtscp generates code to store TSC_AUX (ecx). asm volatile( "rdtscp\n\t" "shl $32, %%rdx\n\t" "or %%rdx, %0\n\t" "lfence" : "=a"(t) : // "memory" avoids reordering. rcx = TSC_AUX. rdx = TSC >> 32. // "cc" = flags modified by SHL. : "rcx", "rdx", "memory", "cc"); #else t = TicksBefore(); // no difference on other platforms. #endif return t; } } // namespace profiler } // namespace jxl #endif // LIB_JXL_BASE_TSC_TIMER_H_