高通8Gen3设备上使用QNN的HTP加速效果不理想

[WangHao311]

您好，
大佬，借贵宝地，请问一下，为什么我在高通8Gen3设备上使用QNN的HTP加速，感觉效果很差，没有达到加速效果呢？

• x86：Python下，使用cuda，模型单次推理13ms；
• android：QNN下，使用HTP，模型单次推理106ms。

设置了QnnHtpPerfInfrastructure_PowerConfig_t如下：

sample_app::StatusCode sample_app::QnnSampleApp::setBurstPerformanceMode() {
    QnnHtpPerfInfrastructure_PowerConfig_t powerConfig;
    memset(&powerConfig, 0, sizeof(powerConfig));
    powerConfig.option = QNN_HTP_PERF_INFRASTRUCTURE_POWER_CONFIGOPTION_DCVS_V3;
    powerConfig.dcvsV3Config.dcvsEnable = 0;
    powerConfig.dcvsV3Config.setDcvsEnable = 1;
    powerConfig.dcvsV3Config.contextId = m_powerConfigId;
    powerConfig.dcvsV3Config.powerMode = QNN_HTP_PERF_INFRASTRUCTURE_POWERMODE_PERFORMANCE_MODE;
    powerConfig.dcvsV3Config.setSleepLatency = 1; // True to consider Latency parameter otherwise False
    powerConfig.dcvsV3Config.setBusParams = 1; // True to consider Bus parameter otherwise False
    powerConfig.dcvsV3Config.setCoreParams = 1; // True to consider Core parameter otherwise False
    powerConfig.dcvsV3Config.sleepDisable = 0; // True to consider sleep/LPM modes, False to enable
    powerConfig.dcvsV3Config.setSleepDisable = 0; // True to consider sleep disable/enable parameter otherwise False
    powerConfig.dcvsV3Config.sleepLatency = 40; // range 40-2000 micro sec
    powerConfig.dcvsV3Config.busVoltageCornerMin = DCVS_VOLTAGE_VCORNER_MAX_VOLTAGE_CORNER;
    powerConfig.dcvsV3Config.busVoltageCornerTarget = DCVS_VOLTAGE_VCORNER_MAX_VOLTAGE_CORNER;
    powerConfig.dcvsV3Config.busVoltageCornerMax = DCVS_VOLTAGE_VCORNER_MAX_VOLTAGE_CORNER;
    powerConfig.dcvsV3Config.coreVoltageCornerMin = DCVS_VOLTAGE_VCORNER_MAX_VOLTAGE_CORNER;
    powerConfig.dcvsV3Config.coreVoltageCornerTarget = DCVS_VOLTAGE_VCORNER_MAX_VOLTAGE_CORNER;
    powerConfig.dcvsV3Config.coreVoltageCornerMax = DCVS_VOLTAGE_VCORNER_MAX_VOLTAGE_CORNER;

    const QnnHtpPerfInfrastructure_PowerConfig_t *powerConfigs[] =
            {&powerConfig, NULL};
    if (m_perfInfra) {
        m_perfInfra->setPowerConfig(m_powerConfigId, powerConfigs);
    }
    return StatusCode::SUCCESS;
}

[zhouwg]

很抱歉2025年2月初才认真回答您的这个issue。为了技术交流方便，我将用中文回答您的问题。

• 2024年7月后完全没有碰llama.cpp的东西了
• 由于中国程序员都能理解的缘故此前做github开源项目每月需要支付100-300的vpn费用，2024年7月彻底没碰llama.cpp后也完全没有碰github相关的东西了（其实不碰以后发现也不错，体会到很多时候就是一个心态问题：太在意太认真会有动力，但凡事过犹不及）

2025年1月28号因为春节前DeepSeek-R1火爆全世界的缘故，又开始致力于完善ggml-qnn（这玩意就像骑自行车，一旦会了后就不会忘记，捡起来很快），请参考这个帖子:#246。

得益于上游llama.cpp的软件架构有了巨大改进，已经在2月初花了几天时间改进2024年4月到5月份的原有代码并提交了一个改进版ggml-qnn后端实现到这个工程（看到开源社区有人也在做ggml-qnn的工作，其C++代码写的比较炫，封装的比较多比较绕，可能导致对llama.cpp源代码与QNN SDK不太熟悉的程序员看的比较晕，那个来源于此项目的实现本质上与当前简洁明了且只有一个源文件一目了然便于移植与扩展的实现一样）。如果哪天发现在上游llama.cpp社区被解封后，会提交PR到上游llama.cpp社区。

从代码角度分析，在高通手机上使用ggml-qnn后端做llama.cpp推理时，由于几个可以想到的原因，

• 在2024年上半年研究ggml-qnn后端时从小米的一个开源项目学到的东西，当时就感觉高通QNN SDK与ggml-qnn后端可能存在一些设计架构不完全兼容的问题，与多媒体领域的OpenMAX IL硬件加速的设计架构不太一样。站在2025年2月的时间点用大白话说，高通的QNN技术架构有自己的专用计算图优化框架，其设计与Intel sycl或华为cann似乎不同（这两者似乎与多媒体领域的OpenMAX IL硬件加速架构有相似之处)，或者从代码角度看就是如下的不同：
  https://github.com/zhouwg/kantv/blob/ggml-qnn-rpc/core/ggml/qnnsample/QnnSampleApp.cpp#L599
  https://github.com/zhouwg/kantv/blob/ggml-qnn-rpc/core/ggml/llamacpp/ggml-qnn.cpp#L1973
  https://github.com/zhouwg/kantv/blob/ggml-qnn-rpc/core/ggml/llamacpp/ggml-qnn.cpp#L3164

• 2024年看到了Intel官方提交的ggml-sycl后端代码后，进一步确信了以上推测

• 2025年1月28号后，看到上游llama.cpp社区中出现了华为官方提交的华为昇腾GPU加速后端ggml-cann，仔细看了他们的代码后，更加确信了2024年上半年的推测

• tensor数据与图节点计算结果在手机AP端的CPU系统内存与Hexagon NPU的设备内存之间加载与传递本身也是有性能损耗，如果在Hexagon NPU上做特定算子(比如matmul)计算加速的性能损耗高于前者，还不如全部使用AP端的CPU进行推理，但是我看到Intel官方出品的ggml-sycl后端却相反，华为官方出品的ggml-cann后端也有大约3-4倍的性能提升

• 在2024年看到了一家来自清华系的AI创业公司在其技术文档中宣称在高通高端手机上使用其闭源ggml-qnn后端加速的llama.cpp推理时性能提升了好几倍，迄今还没彻底想明白他们是怎么做的。猜测是采用了类似小米的技术：https://github.com/XiaoMi/StableDiffusionOnDevice

• 在Intel与华为等中美顶尖大厂纷纷下场推出了举世闻名的开源推理框架llama.cpp中针对自家NPU芯片的硬件加速后端后，Qualcomm为何迄今还没有亲自下场，个人猜测与上述分析有一定的关系，但是高通做了pytorch的官方QNN后端并特意在文档中声明哪个QNN后端处于开发阶段

其性能低于llama.cpp高度优化的cpu后端的推理性能。

为了便于您更好得理解上面的解释，下面贴出2024年4月5号在一位热心高通工程师在github上的指点/帮助下取得决定性进展得源代码文件(可以说ggml-qnn的秘诀就在这个文件里，理解了这个，理解了以上3个链接，再加上理解了llama.cpp的源代码(主要是其中的数据结构与backend子系统)，还有熟悉AI算子开发，ggml-qnn剩下的都是一些c/c++编程的routine work)：

//==============================================================================
//
//  Copyright (c) 2020-2024 Qualcomm Technologies, Inc.
//  All Rights Reserved.
//  Confidential and Proprietary - Qualcomm Technologies, Inc.

//  saver_output.c is generated automatically by Qualcomm's dedicated tool
//
//  this customized saver_output.c is used to troubleshooting issue in
//  PoC-S26: offload a simple f32 2x2 matrix addition operation to QNN CPU backend
//  https://github.com/zhouwg/kantv/issues/121
//
//==============================================================================

#include <stdio.h>
#include <stdlib.h>
#include <string.h>

#include "QnnInterface.h"

#include "ggml-jni.h"

#define VALIDATE(res, value) \
   do { \
      if (res == 0 || res == QNN_COMMON_ERROR_NOT_SUPPORTED) \
      { \
         res = value; \
         if (res != 0) \
         { \
            if (res == QNN_COMMON_ERROR_NOT_SUPPORTED) \
            { \
               LOGGD("WARNING! Line %d QNN feature/API not supported\n", __LINE__); \
               GGML_JNI_NOTIFY("WARNING! Line %d QNN feature/API not supported\n", __LINE__); \
            } else { \
               LOGGD("ERROR! Line %d with error value: %d\n", __LINE__, (unsigned int)error); \
            } \
         } \
      } \
   } \
   while(0)


static void qnn_saver_logcallback(const char* fmt,
                                 QnnLog_Level_t level,
                                 uint64_t timestamp,
                                 va_list argp) {

    static unsigned char s_qnn_saver_buf[JNI_BUF_LEN];

    const char * levelStr = "";
    switch (level) {
        case QNN_LOG_LEVEL_ERROR:
            levelStr = " ERROR ";
            break;
        case QNN_LOG_LEVEL_WARN:
            levelStr = "WARNING";
            break;
        case QNN_LOG_LEVEL_INFO:
            levelStr = "  INFO ";
            break;
        case QNN_LOG_LEVEL_DEBUG:
            levelStr = " DEBUG ";
            break;
        case QNN_LOG_LEVEL_VERBOSE:
            levelStr = "VERBOSE";
            break;
        case QNN_LOG_LEVEL_MAX:
            levelStr = "UNKNOWN";
            break;
    }

    double ms = (double)timestamp / 1000000.0;

    {
        int len_content = 0;
        memset(s_qnn_saver_buf, 0, JNI_BUF_LEN);
        len_content = vsnprintf(s_qnn_saver_buf, JNI_BUF_LEN, fmt, argp);
        snprintf((s_qnn_saver_buf + len_content), JNI_BUF_LEN - len_content, "\n");
        LOGGD("%8.1fms [%-7s] %s ", ms, levelStr, s_qnn_saver_buf);
        //if (level <= QNN_LOG_LEVEL_INFO)
        {
            GGML_JNI_NOTIFY("%8.1fms [%-7s] %s ", ms, levelStr, s_qnn_saver_buf);
        }
    }
}

int qnn_saver_main(int argc, char **argv) {
    LOGGI("enter %s", __func__);
    GGML_JNI_NOTIFY("enter %s", __func__);
    Qnn_ErrorHandle_t error = 0;
    QnnLog_Level_t logLevel = QNN_LOG_LEVEL_VERBOSE;
    int logging = 1;
    for (int i = 1; i < argc; i++) {
        char *arg = argv[i];
        if (!strcmp("--logging", arg) || !strcmp("-l", arg)) {
            logging = 1;
            if (i + 1 == argc) {
                printf("No log level provided, defaulting to QNN_LOG_LEVEL_ERROR\n");
                break;
            }
            char *value = argv[++i];
            if (!strcmp("error", value)) {
                logLevel = QNN_LOG_LEVEL_ERROR;
            } else if (!strcmp("warn", value)) {
                logLevel = QNN_LOG_LEVEL_WARN;
            } else if (!strcmp("info", value)) {
                logLevel = QNN_LOG_LEVEL_INFO;
            } else if (!strcmp("debug", value)) {
                logLevel = QNN_LOG_LEVEL_DEBUG;
            } else if (!strcmp("verbose", value)) {
                logLevel = QNN_LOG_LEVEL_VERBOSE;
            } else {
                printf("WARNING: Unknown log level provided: %s, defaulting to QNN_LOG_LEVEL_ERROR\n",
                       value);
            }
        } else {
            printf("Usage: %s [options]\n\n"
                   "-l <level>, --logging <level>      Enable logging, acceptable levels are: error,warn,info,debug,verbose\n",
                   argv[0]);
            return -1;
        }
    }

    LOGGD("log level %d\n", logLevel);
    FILE *fp = fopen("/sdcard/kantv/params.bin", "rb");
    if (!fp) {
        error = -1;
        LOGGI("ERROR! Could not open params.bin, ensure this file is in the current working directory when executing this program\n");
        GGML_JNI_NOTIFY("ERROR! Could not open params.bin, ensure this file is in the current working directory when executing this program\n");
        return error;
    }

    const QnnInterface_t **providerList = NULL;
    uint32_t numProviders;
    VALIDATE(error, QnnInterface_getProviders(&providerList, &numProviders));
    LOGGD("numProviders %d\n", numProviders);
    GGML_JNI_NOTIFY("numProviders %d\n", numProviders);
    for (int idx = 0; idx < numProviders; idx++) {
        LOGGD("backend name %s\n", providerList[idx]->providerName);
        GGML_JNI_NOTIFY("backend name %s\n", providerList[idx]->providerName);
    }
    QNN_INTERFACE_VER_TYPE interface = providerList[0]->QNN_INTERFACE_VER_NAME;

    Qnn_LogHandle_t loghandle = NULL;
    if (logging) {
        VALIDATE(error, interface.logCreate(qnn_saver_logcallback, logLevel, &loghandle));
    }
    //VALIDATE(error, interface.propertyHasCapability((QnnProperty_Key_t) 304)); //QNN_PROPERTY_GRAPH_SUPPORT_NULL_INPUTS
    VALIDATE(error, interface.propertyHasCapability((QnnProperty_Key_t) QNN_PROPERTY_GRAPH_SUPPORT_NULL_INPUTS));

    const QnnBackend_Config_t *backend_0_config_0[] = {NULL};
    Qnn_BackendHandle_t backend_0;
    VALIDATE(error, interface.backendCreate(loghandle, backend_0_config_0, &backend_0));

    const QnnDevice_Config_t *device_0_config_0[] = {NULL};
    Qnn_DeviceHandle_t device_0;
    VALIDATE(error, interface.deviceCreate(loghandle, device_0_config_0, &device_0));

    const QnnContext_Config_t *context_0_config_0[] = {NULL};
    Qnn_ContextHandle_t context_0;
    VALIDATE(error, interface.contextCreate(backend_0, device_0, context_0_config_0, &context_0));

    const QnnGraph_Config_t *context_0_convReluModel_config_0[] = {NULL};
    Qnn_GraphHandle_t context_0_convReluModel;
    VALIDATE(error,
             interface.graphCreate(context_0, "convReluModel", context_0_convReluModel_config_0,
                                   &context_0_convReluModel));

    //how to compose qnn graph

    //step-1:
    uint32_t context_0_convReluModel_tensor_0_dims[] = {1, 299, 299, 3};
    Qnn_QuantizeParams_t context_0_convReluModel_tensor_0_quantizeParams = {
            (Qnn_Definition_t) 2147483647/*QNN_DEFINITION_UNDEFINED*/,
            (Qnn_QuantizationEncoding_t) 2147483647/*QNN_QUANTIZATION_ENCODING_UNDEFINED*/, .scaleOffsetEncoding = {0.0, 0}};
    Qnn_ClientBuffer_t context_0_convReluModel_tensor_0_clientBuf = {NULL, 0};
    Qnn_TensorV1_t context_0_convReluModel_tensor_0_v1 = {0, "input_0",
                                                          (Qnn_TensorType_t) 0/*QNN_TENSOR_TYPE_APP_WRITE*/,
                                                          0/*QNN_TENSOR_DATA_FORMAT_FLAT_BUFFER*/,
                                                          (Qnn_DataType_t) 562/*QNN_DATATYPE_FLOAT_32*/,
                                                          context_0_convReluModel_tensor_0_quantizeParams,
                                                          4, context_0_convReluModel_tensor_0_dims,
                                                          (Qnn_TensorMemType_t) 0/*QNN_TENSORMEMTYPE_RAW*/,
                                                          context_0_convReluModel_tensor_0_clientBuf};
    Qnn_Tensor_t context_0_convReluModel_tensor_0 = {
            (Qnn_TensorVersion_t) 1, .v1 = context_0_convReluModel_tensor_0_v1};
    VALIDATE(error, interface.tensorCreateGraphTensor(context_0_convReluModel,
                                                      &context_0_convReluModel_tensor_0));




    //step-2:
    uint32_t context_0_convReluModel_tensor_1_dims[] = {3, 3, 3, 32};
    Qnn_QuantizeParams_t context_0_convReluModel_tensor_1_quantizeParams = {
            (Qnn_Definition_t) 2147483647,
            (Qnn_QuantizationEncoding_t) 2147483647, .scaleOffsetEncoding = {0.0, 0}};
    static float context_0_convReluModel_tensor_1_data[864];
    fread(context_0_convReluModel_tensor_1_data, 4, 864, fp);
    Qnn_ClientBuffer_t context_0_convReluModel_tensor_1_clientBuf = {
            (void *) context_0_convReluModel_tensor_1_data, 3456};
    Qnn_TensorV1_t context_0_convReluModel_tensor_1_v1 = {0,
                                                          "InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D_weight",
                                                          (Qnn_TensorType_t) 4/*QNN_TENSOR_TYPE_STATIC*/,
                                                          0/*QNN_TENSOR_DATA_FORMAT_FLAT_BUFFER*/,
                                                          (Qnn_DataType_t) 562/*QNN_DATATYPE_FLOAT_32*/,
                                                          context_0_convReluModel_tensor_1_quantizeParams,
                                                          4, context_0_convReluModel_tensor_1_dims,
                                                          (Qnn_TensorMemType_t) 0,
                                                          context_0_convReluModel_tensor_1_clientBuf};
    Qnn_Tensor_t context_0_convReluModel_tensor_1 = {
            (Qnn_TensorVersion_t) 1, .v1 = context_0_convReluModel_tensor_1_v1};
    VALIDATE(error, interface.tensorCreateGraphTensor(context_0_convReluModel,
                                                      &context_0_convReluModel_tensor_1));



    //step-3:
    uint32_t context_0_convReluModel_tensor_2_dims[] = {32};
    Qnn_QuantizeParams_t context_0_convReluModel_tensor_2_quantizeParams = {
            (Qnn_Definition_t) 2147483647,
            (Qnn_QuantizationEncoding_t) 2147483647, .scaleOffsetEncoding = {0.0, 0}};
    static float context_0_convReluModel_tensor_2_data[32];
    fread(context_0_convReluModel_tensor_2_data, 4, 32, fp);
    Qnn_ClientBuffer_t context_0_convReluModel_tensor_2_clientBuf = {
            (void *) context_0_convReluModel_tensor_2_data, 128};
    Qnn_TensorV1_t context_0_convReluModel_tensor_2_v1 = {0,
                                                          "InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D_bias",
                                                          (Qnn_TensorType_t) 4/*QNN_TENSOR_TYPE_STATIC*/,
                                                          0,
                                                          (Qnn_DataType_t) 562/*QNN_DATATYPE_FLOAT_32*/,
                                                          context_0_convReluModel_tensor_2_quantizeParams,
                                                          1, context_0_convReluModel_tensor_2_dims,
                                                          (Qnn_TensorMemType_t) 0,
                                                          context_0_convReluModel_tensor_2_clientBuf};
    Qnn_Tensor_t context_0_convReluModel_tensor_2 = {
            (Qnn_TensorVersion_t) 1, .v1 = context_0_convReluModel_tensor_2_v1};
    VALIDATE(error, interface.tensorCreateGraphTensor(context_0_convReluModel,
                                                      &context_0_convReluModel_tensor_2));



    //step-4:
    uint32_t context_0_convReluModel_tensor_3_dims[] = {2};
    Qnn_QuantizeParams_t context_0_convReluModel_tensor_3_quantizeParams = {
            (Qnn_Definition_t) 2147483647,
            (Qnn_QuantizationEncoding_t) 2147483647, .scaleOffsetEncoding = {0.0, 0}};
    static uint32_t context_0_convReluModel_tensor_3_data[2];
    fread(context_0_convReluModel_tensor_3_data, 4, 2, fp);
    Qnn_ClientBuffer_t context_0_convReluModel_tensor_3_clientBuf = {
            (void *) context_0_convReluModel_tensor_3_data, 8};
    Qnn_TensorV1_t context_0_convReluModel_tensor_3_v1 = {0,
                                                          "InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D_dilation",
                                                          (Qnn_TensorType_t) 4/*QNN_TENSOR_TYPE_STATIC*/, 0,
                                                          (Qnn_DataType_t) 306/*QNN_DATATYPE_UINT_32*/,
                                                          context_0_convReluModel_tensor_3_quantizeParams,
                                                          1, context_0_convReluModel_tensor_3_dims,
                                                          (Qnn_TensorMemType_t) 0/*QNN_TENSORMEMTYPE_RAW*/,
                                                          context_0_convReluModel_tensor_3_clientBuf};
    Qnn_Tensor_t context_0_convReluModel_tensor_3 = {
            (Qnn_TensorVersion_t) 1, .v1 = context_0_convReluModel_tensor_3_v1};
    VALIDATE(error, interface.tensorCreateGraphTensor(context_0_convReluModel, &context_0_convReluModel_tensor_3));




    //step-5:
    uint32_t context_0_convReluModel_tensor_4_dims[] = {2, 2};
    Qnn_QuantizeParams_t context_0_convReluModel_tensor_4_quantizeParams = {
            (Qnn_Definition_t) 2147483647,
            (Qnn_QuantizationEncoding_t) 2147483647, .scaleOffsetEncoding = {0.0, 0}};
    static uint32_t context_0_convReluModel_tensor_4_data[4];
    fread(context_0_convReluModel_tensor_4_data, 4, 4, fp);
    Qnn_ClientBuffer_t context_0_convReluModel_tensor_4_clientBuf = {
            (void *) context_0_convReluModel_tensor_4_data, 16};
    Qnn_TensorV1_t context_0_convReluModel_tensor_4_v1 = {0,
                                                          "InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D_pad_amount",
                                                          (Qnn_TensorType_t) 4/*QNN_TENSOR_TYPE_STATIC*/, 0,
                                                          (Qnn_DataType_t) 306/*QNN_DATATYPE_UINT_32*/,
                                                          context_0_convReluModel_tensor_4_quantizeParams,
                                                          2, context_0_convReluModel_tensor_4_dims,
                                                          (Qnn_TensorMemType_t) 0,
                                                          context_0_convReluModel_tensor_4_clientBuf};
    Qnn_Tensor_t context_0_convReluModel_tensor_4 = {
            (Qnn_TensorVersion_t) 1, .v1 = context_0_convReluModel_tensor_4_v1};
    VALIDATE(error, interface.tensorCreateGraphTensor(context_0_convReluModel,
                                                      &context_0_convReluModel_tensor_4));




    //step-6:
    uint32_t context_0_convReluModel_tensor_5_dims[] = {2};
    Qnn_QuantizeParams_t context_0_convReluModel_tensor_5_quantizeParams = {
            (Qnn_Definition_t) 2147483647,
            (Qnn_QuantizationEncoding_t) 2147483647, .scaleOffsetEncoding = {0.0, 0}};
    static uint32_t context_0_convReluModel_tensor_5_data[2];
    fread(context_0_convReluModel_tensor_5_data, 4, 2, fp);
    Qnn_ClientBuffer_t context_0_convReluModel_tensor_5_clientBuf = {
            (void *) context_0_convReluModel_tensor_5_data, 8};
    Qnn_TensorV1_t context_0_convReluModel_tensor_5_v1 = {0,
                                                          "InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D_stride",
                                                          (Qnn_TensorType_t) 4/*QNN_TENSOR_TYPE_STATIC*/, 0,
                                                          (Qnn_DataType_t) 306/*QNN_DATATYPE_UINT_32*/,
                                                          context_0_convReluModel_tensor_5_quantizeParams,
                                                          1, context_0_convReluModel_tensor_5_dims,
                                                          (Qnn_TensorMemType_t) 0,
                                                          context_0_convReluModel_tensor_5_clientBuf};
    Qnn_Tensor_t context_0_convReluModel_tensor_5 = {
            (Qnn_TensorVersion_t) 1, .v1 = context_0_convReluModel_tensor_5_v1};
    VALIDATE(error, interface.tensorCreateGraphTensor(context_0_convReluModel,
                                                      &context_0_convReluModel_tensor_5));




    //step-7:
    uint32_t context_0_convReluModel_tensor_6_dims[] = {1, 149, 149, 32};
    Qnn_QuantizeParams_t context_0_convReluModel_tensor_6_quantizeParams = {
            (Qnn_Definition_t) 2147483647,
            (Qnn_QuantizationEncoding_t) 2147483647, .scaleOffsetEncoding = {0.0, 0}};
    Qnn_ClientBuffer_t context_0_convReluModel_tensor_6_clientBuf = {NULL, 0};
    Qnn_TensorV1_t context_0_convReluModel_tensor_6_v1 = {0,
                                                          "InceptionV3_InceptionV3_Conv2d_1a_3x3_BatchNorm_FusedBatchNorm_0",
                                                          (Qnn_TensorType_t) 3/*QNN_TENSOR_TYPE_NATIVE*/, 0,
                                                          (Qnn_DataType_t) 562/*QNN_DATATYPE_FLOAT_32*/,
                                                          context_0_convReluModel_tensor_6_quantizeParams,
                                                          4, context_0_convReluModel_tensor_6_dims,
                                                          (Qnn_TensorMemType_t) 0,
                                                          context_0_convReluModel_tensor_6_clientBuf};
    Qnn_Tensor_t context_0_convReluModel_tensor_6 = {
            (Qnn_TensorVersion_t) 1, .v1 = context_0_convReluModel_tensor_6_v1};
    VALIDATE(error, interface.tensorCreateGraphTensor(context_0_convReluModel, &context_0_convReluModel_tensor_6));


    //step-8:
    Qnn_Param_t context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D_0_param_0 = {
            (Qnn_ParamType_t) 1/*QNN_PARAMTYPE_TENSOR*/,
            "dilation",
            .tensorParam = context_0_convReluModel_tensor_3
    };
    Qnn_Param_t context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D_0_param_1 = {
            (Qnn_ParamType_t) 1/*QNN_PARAMTYPE_TENSOR*/,
            "pad_amount",
            .tensorParam = context_0_convReluModel_tensor_4
    };
    Qnn_Param_t context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D_0_param_2 = {
            (Qnn_ParamType_t) 1/*QNN_PARAMTYPE_TENSOR*/,
            "stride",
            .tensorParam = context_0_convReluModel_tensor_5
    };
    Qnn_Param_t context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D_0_param_3 = {
            (Qnn_ParamType_t) 0/*QNN_PARAMTYPE_SCALAR*/,
            "group",
            .scalarParam = {
                    (Qnn_DataType_t) 306, .uint32Value = 1}
    };
    Qnn_Param_t context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D_0_params[] = {
            context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D_0_param_0,
            context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D_0_param_1,
            context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D_0_param_2,
            context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D_0_param_3};

    Qnn_Tensor_t context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D_0_inputs[] = {
            context_0_convReluModel_tensor_0,
            context_0_convReluModel_tensor_1,
            context_0_convReluModel_tensor_2};

    Qnn_Tensor_t context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D_0_outputs[] = {
            context_0_convReluModel_tensor_6
    };

    Qnn_OpConfig_t context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D_0 = {
            (Qnn_OpConfigVersion_t) 1,
            .v1 = {
                    "InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D",
                    "qti.aisw",
                    "Conv2d",
                    4,
                    context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D_0_params,
                    3,
                    context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D_0_inputs,
                    1,
                    context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D_0_outputs
            }
    };
    VALIDATE(error, interface.backendValidateOpConfig(backend_0, context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D_0));
    VALIDATE(error, interface.graphAddNode(context_0_convReluModel, context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Conv2D_0));




    //step-9:
    uint32_t context_0_convReluModel_tensor_7_dims[] = {1, 149, 149, 32};
    Qnn_QuantizeParams_t context_0_convReluModel_tensor_7_quantizeParams = {
            (Qnn_Definition_t) 2147483647,
            (Qnn_QuantizationEncoding_t) 2147483647, .scaleOffsetEncoding = {0.0, 0}};
    Qnn_ClientBuffer_t context_0_convReluModel_tensor_7_clientBuf = {NULL, 0};
    Qnn_TensorV1_t context_0_convReluModel_tensor_7_v1 = {0,
                                                          "InceptionV3_InceptionV3_Conv2d_1a_3x3_Relu_0",
                                                          (Qnn_TensorType_t) 1/*QNN_TENSOR_TYPE_APP_READ*/, 0,
                                                          (Qnn_DataType_t) 562/*QNN_DATATYPE_FLOAT_32*/,
                                                          context_0_convReluModel_tensor_7_quantizeParams,
                                                          4, context_0_convReluModel_tensor_7_dims,
                                                          (Qnn_TensorMemType_t) 0,
                                                          context_0_convReluModel_tensor_7_clientBuf};
    Qnn_Tensor_t context_0_convReluModel_tensor_7 = {
            (Qnn_TensorVersion_t) 1, .v1 = context_0_convReluModel_tensor_7_v1};
    VALIDATE(error, interface.tensorCreateGraphTensor(context_0_convReluModel,
                                                      &context_0_convReluModel_tensor_7));



    //step-10:
    Qnn_Param_t context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Relu_0_params[] = {};
    Qnn_Tensor_t context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Relu_0_inputs[] = {
            context_0_convReluModel_tensor_6
    };
    Qnn_Tensor_t context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Relu_0_outputs[] = {
            context_0_convReluModel_tensor_7
    };
    Qnn_OpConfig_t context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Relu_0 = {
            (Qnn_OpConfigVersion_t) 1, .v1 = {
                    "InceptionV3_InceptionV3_Conv2d_1a_3x3_Relu",
                    "qti.aisw",
                    "Relu",
                    0,
                    context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Relu_0_params,
                    1,
                    context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Relu_0_inputs,
                    1,
                    context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Relu_0_outputs
            }
    };
    VALIDATE(error, interface.backendValidateOpConfig(backend_0,context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Relu_0));
    VALIDATE(error, interface.graphAddNode(context_0_convReluModel,context_0_convReluModel_InceptionV3_InceptionV3_Conv2d_1a_3x3_Relu_0));

    //step-10:
    VALIDATE(error, interface.graphFinalize(context_0_convReluModel, NULL, NULL));

    Qnn_Tensor_t context_0_convReluModel_inputTensors_0[] = {context_0_convReluModel_tensor_0};
    Qnn_Tensor_t context_0_convReluModel_outputTensors_0[] = {context_0_convReluModel_tensor_7};
    VALIDATE(error,interface.graphExecute(context_0_convReluModel, context_0_convReluModel_inputTensors_0,
                                    1, context_0_convReluModel_outputTensors_0, 1, NULL, NULL));


    VALIDATE(error, interface.contextFree(context_0, NULL));

    VALIDATE(error, interface.deviceFree(device_0));

    VALIDATE(error, interface.backendFree(backend_0));

    if (logging) {
        VALIDATE(error, interface.logFree(loghandle));
    }

    if (fclose(fp)) error = -1;

    LOGGI("leave %s", __func__);
    GGML_JNI_NOTIFY("leave %s", __func__);
    return error == 0 || error == QNN_COMMON_ERROR_NOT_SUPPORTED ? 0 : error;
}

我不是AI技术领域的专家（但非常看好我国的90后尤其00后程序员与AI专家尤其是DeepSeek公司的众多AI专家，如果没有DeepSeek-R1在春节前后震惊世界的杰出表现，也大概率不会再回到github，进而也没有这个回答)，以上推测仅仅是根据公开技术文档与公开代码学习后的个人看法，理解不当之处请您指正。

最后，借用这个帖子，表达一个个人看法：llama.cpp这个来自欧洲天才程序员的个人项目，逐渐成为github上最有影响力之一的开源端侧推理框架，而且滋润了包括个人开源项目在内的无数下游开源项目或商业公司商业项目或高校研究机构研究项目，为什没有商业化：去年我曾建议作者在中国开一个llama.cpp官方技术微信公众号定期收取打赏年入百万应该问题不大，因为我看到公众号上一个写篇篇10万+财经文章的号主每年仅仅收取12次费用每次打赏人数过数万甚至过十万，个人对这个在文章中宣称年入几百万不差钱的号主每年打赏12元，迄今累计公众号打赏可能接近万元-------有时候看到思考深刻独到或令人拍案叫绝的好文章忍不住打赏几十几百。llama.cpp这个开源项目需要的高性能开发机器与优秀程序员的开发成本，仅仅是由github上的赞助者的赞助覆盖的吗？不得不说，开源文化在美国/欧洲的确有它的土壤，在14亿人的我国如果还很年轻或者不是公司许可的最好不要做开源项目---相当费时费力---将本职工作做好即可。

如果您认为这个回答能解答您的问题，请帮助关闭掉这个issue，谢谢您的合作。

[zhouwg]

您好，由于这个issue长时间没有关闭，且本项目的ggml-qnn相关工作已经由kantv-ai团队成员于2025年2月13号提交到上游社区，您有相关问题可以去上游社区提问，上游社区里有很多AI专家，他们或许能给出更加好的意见。

如果未来您还有需要，请重新打开此issue，谢谢理解。