baalajimaestro/ollama
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity
ollama/llama/ggml-cuda/vendors/hip.h

204 lines
8.3 KiB
C
Raw Normal View History

Re-introduce the `llama` package (#5034) * Re-introduce the llama package This PR brings back the llama package, making it possible to call llama.cpp and ggml APIs from Go directly via CGo. This has a few advantages: - C APIs can be called directly from Go without needing to use the previous "server" REST API - On macOS and for CPU builds on Linux and Windows, Ollama can be built without a go generate ./... step, making it easy to get up and running to hack on parts of Ollama that don't require fast inference - Faster build times for AVX,AVX2,CUDA and ROCM (a full build of all runners takes <5 min on a fast CPU) - No git submodule making it easier to clone and build from source This is a big PR, but much of it is vendor code except for: - llama.go CGo bindings - example/: a simple example of running inference - runner/: a subprocess server designed to replace the llm/ext_server package - Makefile an as minimal as possible Makefile to build the runner package for different targets (cpu, avx, avx2, cuda, rocm) Co-authored-by: Jesse Gross <jesse@ollama.com> Co-authored-by: Daniel Hiltgen <daniel@ollama.com> * cache: Clear old KV cache entries when evicting a slot When forking a cache entry, if no empty slots are available we evict the least recently used one and copy over the KV entries from the closest match. However, this copy does not overwrite existing values but only adds new ones. Therefore, we need to clear the old slot first. This change fixes two issues: - The KV cache fills up and runs out of space even though we think we are managing it correctly - Performance gets worse over time as we use new cache entries that are not hot in the processor caches * doc: explain golang objc linker warning (#6830) * llama: gather transitive dependencies for rocm for dist packaging (#6848) * Refine go server makefiles to be more DRY (#6924) This breaks up the monolithic Makefile for the Go based runners into a set of utility files as well as recursive Makefiles for the runners. Files starting with the name "Makefile" are buildable, while files that end with ".make" are utilities to include in other Makefiles. This reduces the amount of nearly identical targets and helps set a pattern for future community contributions for new GPU runner architectures. When we are ready to switch over to the Go runners, these files should move to the top of the repo, and we should add targets for the main CLI, as well as a helper "install" (put all the built binaries on the local system in a runnable state) and "dist" target (generate the various tar/zip files for distribution) for local developer use. * llama: don't create extraneous directories (#6988) * llama: Exercise the new build in CI (#6989) Wire up some basic sanity testing in CI for the Go runner. GPU runners are not covered yet. * llama: Refine developer docs for Go server (#6842) This enhances the documentation for development focusing on the new Go server. After we complete the transition further doc refinements can remove the "transition" discussion. * runner.go: Allocate batches for all sequences during init We should tell the model that we could have full batches for all sequences. We already do this when we allocate the batches but it was missed during initialization. * llama.go: Don't return nil from Tokenize on zero length input Potentially receiving nil in a non-error condition is surprising to most callers - it's better to return an empty slice. * runner.go: Remove stop tokens from cache If the last token is EOG then we don't return this and it isn't present in the cache (because it was never submitted to Decode). This works well for extending the cache entry with a new sequence. However, for multi-token stop sequences, we won't return any of the tokens but all but the last one will be in the cache. This means when the conversation continues the cache will contain tokens that don't overlap with the new prompt. This works (we will pick up the portion where there is overlap) but it causes unnecessary cache thrashing because we will fork the original cache entry as it is not a perfect match. By trimming the cache to the tokens that we actually return this issue can be avoided. * runner.go: Simplify flushing of pending tokens * runner.go: Update TODOs * runner.go: Don't panic when processing sequences If there is an error processing a sequence, we should return a clean HTTP error back to Ollama rather than panicing. This will make us more resilient to transient failures. Panics can still occur during startup as there is no way to serve requests if that fails. Co-authored-by: jmorganca <jmorganca@gmail.com> * runner.go: More accurately capture timings Currently prompt processing time doesn't capture the that it takes to tokenize the input, only decoding time. We should capture the full process to more accurately reflect reality. This is especially true once we start processing images where the initial processing can take significant time. This is also more consistent with the existing C++ runner. * runner.go: Support for vision models In addition to bringing feature parity with the C++ runner, this also incorporates several improvements: - Cache prompting works with images, avoiding the need to re-decode embeddings for every message in a conversation - Parallelism is supported, avoiding the need to restrict to one sequence at a time. (Though for now Ollama will not schedule them while we might need to fall back to the old runner.) Co-authored-by: jmorganca <jmorganca@gmail.com> * runner.go: Move Unicode checking code and add tests * runner.go: Export external cache members Runner and cache are in the same package so the change doesn't affect anything but it is more internally consistent. * runner.go: Image embedding cache Generating embeddings from images can take significant time (on my machine between 100ms and 8s depending on the model). Although we already cache the result of decoding these images, the embeddings need to be regenerated every time. This is not necessary if we get the same image over and over again, for example, during a conversation. This currently uses a very small cache with a very simple algorithm but it is easy to improve as is warranted. * llama: catch up on patches Carry forward solar-pro and cli-unicode patches * runner.go: Don't re-allocate memory for every batch We can reuse memory allocated from batch to batch since batch size is fixed. This both saves the cost of reallocation as well keeps the cache lines hot. This results in a roughly 1% performance improvement for token generation with Nvidia GPUs on Linux. * runner.go: Default to classic input cache policy The input cache as part of the go runner implemented a cache policy that aims to maximize hit rate in both single and multi- user scenarios. When there is a cache hit, the response is very fast. However, performance is actually slower when there is an input cache miss due to worse GPU VRAM locality. This means that performance is generally better overall for multi-user scenarios (better input cache hit rate, locality was relatively poor already). But worse for single users (input cache hit rate is about the same, locality is now worse). This defaults the policy back to the old one to avoid a regression but keeps the new one available through an environment variable OLLAMA_MULTIUSER_CACHE. This is left undocumented as the goal is to improve this in the future to get the best of both worlds without user configuration. For inputs that result in cache misses, on Nvidia/Linux this change improves performance by 31% for prompt processing and 13% for token generation. * runner.go: Increase size of response channel Generally the CPU can easily keep up with handling reponses that are generated but there's no reason not to let generation continue and handle things in larger batches if needed. * llama: Add CI to verify all vendored changes have patches (#7066) Make sure we don't accidentally merge changes in the vendored code that aren't also reflected in the patches. * llama: adjust clip patch for mingw utf-16 (#7065) * llama: adjust clip patch for mingw utf-16 * llama: ensure static linking of runtime libs Avoid runtime dependencies on non-standard libraries * runner.go: Enable llamafile (all platforms) and BLAS (Mac OS) These are two features that are shown on llama.cpp's system info that are currently different between the two runners. On my test systems the performance difference is very small to negligible but it is probably still good to equalize the features. * llm: Don't add BOS/EOS for tokenize requests This is consistent with what server.cpp currently does. It affects things like token processing counts for embedding requests. * runner.go: Don't cache prompts for embeddings Our integration with server.cpp implicitly disables prompt caching because it is not part of the JSON object being parsed, this makes the Go runner behavior similarly. Prompt caching has been seen to affect the results of text completions on certain hardware. The results are not wrong either way but they are non-deterministic. However, embeddings seem to be affected even on hardware that does not show this behavior for completions. For now, it is best to maintain consistency with the existing behavior. * runner.go: Adjust debug log levels Add system info printed at startup and quiet down noisier logging. * llama: fix compiler flag differences (#7082) Adjust the flags for the new Go server to more closely match the generate flow * llama: refine developer docs (#7121) * llama: doc and example clean up (#7122) * llama: doc and example clean up * llama: Move new dockerfile into llama dir Temporary home until we fully transition to the Go server * llama: runner doc cleanup * llama.go: Add description for Tokenize error case --------- Co-authored-by: Jesse Gross <jesse@ollama.com> Co-authored-by: Daniel Hiltgen <daniel@ollama.com> Co-authored-by: Daniel Hiltgen <dhiltgen@users.noreply.github.com>
2024-10-08 15:53:54 +00:00
/**
* llama.cpp - commit 8962422b1c6f9b8b15f5aeaea42600bcc2d44177 - do not edit this file
*
* MIT License
*
* Copyright (c) 2023-2024 The ggml authors
*
* Permission is hereby granted, free of charge, to any person obtaining a copy
* of this software and associated documentation files (the "Software"), to deal
* in the Software without restriction, including without limitation the rights
* to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
* copies of the Software, and to permit persons to whom the Software is
* furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in all
* copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#pragma once
#include <hip/hip_runtime.h>
#include <hipblas/hipblas.h>
#include <hip/hip_fp16.h>
#ifdef __HIP_PLATFORM_AMD__
// for rocblas_initialize()
#include "rocblas/rocblas.h"
#endif // __HIP_PLATFORM_AMD__
#define CUBLAS_COMPUTE_16F HIPBLAS_R_16F
#define CUBLAS_COMPUTE_32F HIPBLAS_R_32F
#define CUBLAS_COMPUTE_32F_FAST_16F HIPBLAS_R_32F
#define CUBLAS_GEMM_DEFAULT HIPBLAS_GEMM_DEFAULT
#define CUBLAS_GEMM_DEFAULT_TENSOR_OP HIPBLAS_GEMM_DEFAULT
#define CUBLAS_OP_N HIPBLAS_OP_N
#define CUBLAS_OP_T HIPBLAS_OP_T
#define CUBLAS_STATUS_SUCCESS HIPBLAS_STATUS_SUCCESS
#define CUBLAS_TF32_TENSOR_OP_MATH 0
#define CUDA_R_16F HIPBLAS_R_16F
#define CUDA_R_32F HIPBLAS_R_32F
#define __shfl_xor_sync(mask, var, laneMask, width) __shfl_xor(var, laneMask, width)
#define cublasComputeType_t hipblasDatatype_t //deprecated, new hipblasComputeType_t not in 5.6
#define cublasCreate hipblasCreate
#define cublasDestroy hipblasDestroy
#define cublasGemmEx hipblasGemmEx
#define cublasGemmBatchedEx hipblasGemmBatchedEx
#define cublasGemmStridedBatchedEx hipblasGemmStridedBatchedEx
#define cublasHandle_t hipblasHandle_t
#define cublasSetMathMode(handle, mode) CUBLAS_STATUS_SUCCESS
#define cublasSetStream hipblasSetStream
#define cublasSgemm hipblasSgemm
#define cublasStatus_t hipblasStatus_t
#define cudaDataType_t hipblasDatatype_t //deprecated, new hipblasDatatype not in 5.6
#define cudaDeviceCanAccessPeer hipDeviceCanAccessPeer
#define cudaDeviceDisablePeerAccess hipDeviceDisablePeerAccess
#define cudaDeviceEnablePeerAccess hipDeviceEnablePeerAccess
#define cudaDeviceProp hipDeviceProp_t
#define cudaDeviceSynchronize hipDeviceSynchronize
#define cudaError_t hipError_t
#define cudaErrorPeerAccessAlreadyEnabled hipErrorPeerAccessAlreadyEnabled
#define cudaErrorPeerAccessNotEnabled hipErrorPeerAccessNotEnabled
#define cudaEventCreateWithFlags hipEventCreateWithFlags
#define cudaEventDisableTiming hipEventDisableTiming
#define cudaEventRecord hipEventRecord
#define cudaEventSynchronize hipEventSynchronize
#define cudaEvent_t hipEvent_t
#define cudaEventDestroy hipEventDestroy
#define cudaFree hipFree
#define cudaFreeHost hipHostFree
#define cudaGetDevice hipGetDevice
#define cudaGetDeviceCount hipGetDeviceCount
#define cudaGetDeviceProperties hipGetDeviceProperties
#define cudaGetErrorString hipGetErrorString
#define cudaGetLastError hipGetLastError
#define cudaHostRegister hipHostRegister
#define cudaHostRegisterPortable hipHostRegisterPortable
#define cudaHostRegisterReadOnly hipHostRegisterReadOnly
#define cudaHostUnregister hipHostUnregister
#define cudaLaunchHostFunc hipLaunchHostFunc
#define cudaMalloc hipMalloc
#define cudaMallocHost(ptr, size) hipHostMalloc(ptr, size, hipHostMallocDefault)
#define cudaMemcpy hipMemcpy
#define cudaMemcpyAsync hipMemcpyAsync
#define cudaMemcpyPeerAsync hipMemcpyPeerAsync
#define cudaMemcpy2DAsync hipMemcpy2DAsync
#define cudaMemcpyDeviceToDevice hipMemcpyDeviceToDevice
#define cudaMemcpyDeviceToHost hipMemcpyDeviceToHost
#define cudaMemcpyHostToDevice hipMemcpyHostToDevice
#define cudaMemcpyKind hipMemcpyKind
#define cudaMemset hipMemset
#define cudaMemsetAsync hipMemsetAsync
#define cudaMemGetInfo hipMemGetInfo
#define cudaOccupancyMaxPotentialBlockSize hipOccupancyMaxPotentialBlockSize
#define cudaSetDevice hipSetDevice
#define cudaStreamCreateWithFlags hipStreamCreateWithFlags
#define cudaStreamDestroy hipStreamDestroy
#define cudaStreamFireAndForget hipStreamFireAndForget
#define cudaStreamNonBlocking hipStreamNonBlocking
#define cudaStreamPerThread hipStreamPerThread
#define cudaStreamSynchronize hipStreamSynchronize
#define cudaStreamWaitEvent(stream, event, flags) hipStreamWaitEvent(stream, event, flags)
#define cudaStream_t hipStream_t
#define cudaSuccess hipSuccess
#define __trap() do { abort(); __builtin_unreachable(); } while(0)
#define CUBLAS_STATUS_SUCCESS HIPBLAS_STATUS_SUCCESS
#define CUBLAS_STATUS_NOT_INITIALIZED HIPBLAS_STATUS_NOT_INITIALIZED
#define CUBLAS_STATUS_ALLOC_FAILED HIPBLAS_STATUS_ALLOC_FAILED
#define CUBLAS_STATUS_INVALID_VALUE HIPBLAS_STATUS_INVALID_VALUE
#define CUBLAS_STATUS_ARCH_MISMATCH HIPBLAS_STATUS_ARCH_MISMATCH
#define CUBLAS_STATUS_MAPPING_ERROR HIPBLAS_STATUS_MAPPING_ERROR
#define CUBLAS_STATUS_EXECUTION_FAILED HIPBLAS_STATUS_EXECUTION_FAILED
#define CUBLAS_STATUS_INTERNAL_ERROR HIPBLAS_STATUS_INTERNAL_ERROR
#define CUBLAS_STATUS_NOT_SUPPORTED HIPBLAS_STATUS_NOT_SUPPORTED
#define __CUDA_ARCH__ 1300
#if defined(__gfx1100__) || defined(__gfx1101__) || defined(__gfx1102__) || defined(__gfx1103__) || \
defined(__gfx1150__) || defined(__gfx1151__)
#define RDNA3
#endif
#if defined(__gfx1030__) || defined(__gfx1031__) || defined(__gfx1032__) || defined(__gfx1033__) || \
defined(__gfx1034__) || defined(__gfx1035__) || defined(__gfx1036__) || defined(__gfx1037__)
#define RDNA2
#endif
#if defined(__gfx1010__) || defined(__gfx1012__)
#define RDNA1
#endif
#ifndef __has_builtin
#define __has_builtin(x) 0
#endif
typedef int8_t int8x4_t __attribute__((ext_vector_type(4)));
typedef uint8_t uint8x4_t __attribute__((ext_vector_type(4)));
static __device__ __forceinline__ int __vsubss4(const int a, const int b) {
const int8x4_t va = reinterpret_cast<const int8x4_t&>(a);
const int8x4_t vb = reinterpret_cast<const int8x4_t&>(b);
#if __has_builtin(__builtin_elementwise_sub_sat)
const int8x4_t c = __builtin_elementwise_sub_sat(va, vb);
return reinterpret_cast<const int &>(c);
#else
int8x4_t c;
int16_t tmp;
#pragma unroll
for (int i = 0; i < 4; i++) {
tmp = va[i] - vb[i];
if(tmp > std::numeric_limits<int8_t>::max()) tmp = std::numeric_limits<int8_t>::max();
if(tmp < std::numeric_limits<int8_t>::min()) tmp = std::numeric_limits<int8_t>::min();
c[i] = tmp;
}
return reinterpret_cast<int &>(c);
#endif // __has_builtin(__builtin_elementwise_sub_sat)
}
static __device__ __forceinline__ int __vsub4(const int a, const int b) {
return __vsubss4(a, b);
}
static __device__ __forceinline__ unsigned int __vcmpeq4(unsigned int a, unsigned int b) {
const uint8x4_t& va = reinterpret_cast<const uint8x4_t&>(a);
const uint8x4_t& vb = reinterpret_cast<const uint8x4_t&>(b);
unsigned int c;
uint8x4_t& vc = reinterpret_cast<uint8x4_t&>(c);
#pragma unroll
for (int i = 0; i < 4; ++i) {
vc[i] = va[i] == vb[i] ? 0xff : 0x00;
}
return c;
}
static __device__ __forceinline__ unsigned int __vcmpne4(unsigned int a, unsigned int b) {
const uint8x4_t& va = reinterpret_cast<const uint8x4_t&>(a);
const uint8x4_t& vb = reinterpret_cast<const uint8x4_t&>(b);
unsigned int c;
uint8x4_t& vc = reinterpret_cast<uint8x4_t&>(c);
#pragma unroll
for (int i = 0; i < 4; ++i) {
vc[i] = va[i] == vb[i] ? 0x00 : 0xff;
}
return c;
}
#if defined(__HIP_PLATFORM_AMD__) && HIP_VERSION < 50600000
// __shfl_xor() for half2 was added in ROCm 5.6
static __device__ __forceinline__ half2 __shfl_xor(half2 var, int laneMask, int width) {
typedef union half2_b32 {
half2 val;
int b32;
} half2_b32_t;
half2_b32_t tmp;
tmp.val = var;
tmp.b32 = __shfl_xor(tmp.b32, laneMask, width);
return tmp.val;
}
#endif // defined(__HIP_PLATFORM_AMD__) && HIP_VERSION < 50600000
Reference in a new issue Copy permalink