Added the armlib third-party dependency.

2026-09-11 18:29:22 -07:00 · 2026-08-19 12:04:32 +02:00
parent 0b010a3077
commit fb21fd80eb
110 changed files with 122964 additions and 0 deletions
@@ -0,0 +1,186 @@
+/**
+  **************************************************************************
+  * @file     at32f435_437_conf.h
+  * @brief    at32f435_437 config header file
+  **************************************************************************
+  *                       Copyright notice & Disclaimer
+  *
+  * The software Board Support Package (BSP) that is made available to
+  * download from Artery official website is the copyrighted work of Artery.
+  * Artery authorizes customers to use, copy, and distribute the BSP
+  * software and its related documentation for the purpose of design and
+  * development in conjunction with Artery microcontrollers. Use of the
+  * software is governed by this copyright notice and the following disclaimer.
+  *
+  * THIS SOFTWARE IS PROVIDED ON "AS IS" BASIS WITHOUT WARRANTIES,
+  * GUARANTEES OR REPRESENTATIONS OF ANY KIND. ARTERY EXPRESSLY DISCLAIMS,
+  * TO THE FULLEST EXTENT PERMITTED BY LAW, ALL EXPRESS, IMPLIED OR
+  * STATUTORY OR OTHER WARRANTIES, GUARANTEES OR REPRESENTATIONS,
+  * INCLUDING BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY,
+  * FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT.
+  *
+  **************************************************************************
+  */
+
+/* define to prevent recursive inclusion -------------------------------------*/
+#ifndef __AT32F435_437_CONF_H
+#define __AT32F435_437_CONF_H
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/**
+  * @brief in the following line adjust the value of high speed external crystal (hext)
+  * used in your application
+  *
+  * tip: to avoid modifying this file each time you need to use different hext, you
+  *      can define the hext value in your toolchain compiler preprocessor.
+  *
+  */
+#if !defined  HEXT_VALUE
+#define HEXT_VALUE                       ((uint32_t)8000000) /*!< value of the high speed external crystal in hz */
+#endif
+
+/**
+  * @brief in the following line adjust the high speed external crystal (hext) startup
+  * timeout value
+  */
+#define HEXT_STARTUP_TIMEOUT             ((uint16_t)0x3000)  /*!< time out for hext start up */
+#define HICK_VALUE                       ((uint32_t)8000000) /*!< value of the high speed internal clock in hz */
+#define LEXT_VALUE                       ((uint32_t)32768)   /*!< value of the low speed external clock in hz */
+
+
+/*
+ * Do not import any at32f435_437_xxx header file in this file.
+ * This will cause developers who only want to include the at32f435_437 header file to get a bunch of unnecessary contents,
+ * which will lead to possible conflicts of variable/function names!!!
+ * ---
+ * So all the module enable definitions below have been moved to CMakeLists.txt,
+ * and all the driver header files of at32f435_437_xxx are included only when they are used.
+ * ---
+ * If you do not do this, the 'usart_init ' function in the usart driver of at32 will cause the conflicting types error by pm3's usart module .
+ */
+
+
+/* module define -------------------------------------------------------------*/
+// #define CRM_MODULE_ENABLED
+// #define TMR_MODULE_ENABLED
+// #define ERTC_MODULE_ENABLED
+// #define GPIO_MODULE_ENABLED
+// #define I2C_MODULE_ENABLED
+// #define USART_MODULE_ENABLED
+// #define PWC_MODULE_ENABLED
+// #define CAN_MODULE_ENABLED
+// #define ADC_MODULE_ENABLED
+// #define DAC_MODULE_ENABLED
+// #define SPI_MODULE_ENABLED
+// #define EDMA_MODULE_ENABLED
+// #define DMA_MODULE_ENABLED
+// #define DEBUG_MODULE_ENABLED
+// #define FLASH_MODULE_ENABLED
+// #define CRC_MODULE_ENABLED
+// #define WWDT_MODULE_ENABLED
+// #define WDT_MODULE_ENABLED
+// #define EXINT_MODULE_ENABLED
+// #define SDIO_MODULE_ENABLED
+// #define XMC_MODULE_ENABLED
+// #define USB_MODULE_ENABLED
+// #define ACC_MODULE_ENABLED
+// #define MISC_MODULE_ENABLED
+// #define QSPI_MODULE_ENABLED
+// #define DVP_MODULE_ENABLED
+// #define SCFG_MODULE_ENABLED
+// #define EMAC_MODULE_ENABLED
+
+/* includes ------------------------------------------------------------------*/
+// #ifdef CRM_MODULE_ENABLED
+// #include "at32f435_437_crm.h"
+// #endif
+// #ifdef TMR_MODULE_ENABLED
+// #include "at32f435_437_tmr.h"
+// #endif
+// #ifdef ERTC_MODULE_ENABLED
+// #include "at32f435_437_ertc.h"
+// #endif
+// #ifdef GPIO_MODULE_ENABLED
+// #include "at32f435_437_gpio.h"
+// #endif
+// #ifdef I2C_MODULE_ENABLED
+// #include "at32f435_437_i2c.h"
+// #endif
+// #ifdef USART_MODULE_ENABLED
+// #include "at32f435_437_usart.h"
+// #endif
+// #ifdef PWC_MODULE_ENABLED
+// #include "at32f435_437_pwc.h"
+// #endif
+// #ifdef CAN_MODULE_ENABLED
+// #include "at32f435_437_can.h"
+// #endif
+// #ifdef ADC_MODULE_ENABLED
+// #include "at32f435_437_adc.h"
+// #endif
+// #ifdef DAC_MODULE_ENABLED
+// #include "at32f435_437_dac.h"
+// #endif
+// #ifdef SPI_MODULE_ENABLED
+// #include "at32f435_437_spi.h"
+// #endif
+// #ifdef DMA_MODULE_ENABLED
+// #include "at32f435_437_dma.h"
+// #endif
+// #ifdef DEBUG_MODULE_ENABLED
+// #include "at32f435_437_debug.h"
+// #endif
+// #ifdef FLASH_MODULE_ENABLED
+// #include "at32f435_437_flash.h"
+// #endif
+// #ifdef CRC_MODULE_ENABLED
+// #include "at32f435_437_crc.h"
+// #endif
+// #ifdef WWDT_MODULE_ENABLED
+// #include "at32f435_437_wwdt.h"
+// #endif
+// #ifdef WDT_MODULE_ENABLED
+// #include "at32f435_437_wdt.h"
+// #endif
+// #ifdef EXINT_MODULE_ENABLED
+// #include "at32f435_437_exint.h"
+// #endif
+// #ifdef SDIO_MODULE_ENABLED
+// #include "at32f435_437_sdio.h"
+// #endif
+// #ifdef XMC_MODULE_ENABLED
+// #include "at32f435_437_xmc.h"
+// #endif
+// #ifdef ACC_MODULE_ENABLED
+// #include "at32f435_437_acc.h"
+// #endif
+// #ifdef MISC_MODULE_ENABLED
+// #include "at32f435_437_misc.h"
+// #endif
+// #ifdef EDMA_MODULE_ENABLED
+// #include "at32f435_437_edma.h"
+// #endif
+// #ifdef QSPI_MODULE_ENABLED
+// #include "at32f435_437_qspi.h"
+// #endif
+// #ifdef SCFG_MODULE_ENABLED
+// #include "at32f435_437_scfg.h"
+// #endif
+// #ifdef EMAC_MODULE_ENABLED
+// #include "at32f435_437_emac.h"
+// #endif
+// #ifdef DVP_MODULE_ENABLED
+// #include "at32f435_437_dvp.h"
+// #endif
+// #ifdef USB_MODULE_ENABLED
+// #include "at32f435_437_usb.h"
+// #endif
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif
@@ -0,0 +1,132 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_common_tables_f16.h
+ * Description:  Extern declaration for common tables
+ *
+ * @version  V1.10.0
+ * @date     08 July 2021
+ *
+ * Target Processor: Cortex-M and Cortex-A cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2021 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#ifndef _ARM_COMMON_TABLES_F16_H
+#define _ARM_COMMON_TABLES_F16_H
+
+#include "arm_math_types_f16.h"
+
+#ifdef   __cplusplus
+extern "C"
+{
+#endif
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_FFT_ALLOW_TABLES)
+
+  /* F16 */
+  #if !defined(__CC_ARM) && defined(ARM_FLOAT16_SUPPORTED)
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_F16_16)
+    extern const float16_t twiddleCoefF16_16[32];
+  #endif /* !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) */
+
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_F16_32)
+    extern const float16_t twiddleCoefF16_32[64];
+  #endif /* !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) */
+
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_F16_64)
+    extern const float16_t twiddleCoefF16_64[128];
+  #endif /* !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) */
+
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_F16_128)
+    extern const float16_t twiddleCoefF16_128[256];
+  #endif /* !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) */
+
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_F16_256)
+    extern const float16_t twiddleCoefF16_256[512];
+  #endif /* !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) */
+
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_F16_512)
+    extern const float16_t twiddleCoefF16_512[1024];
+  #endif /* !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) */
+
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_F16_1024)
+    extern const float16_t twiddleCoefF16_1024[2048];
+  #endif /* !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) */
+
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_F16_2048)
+    extern const float16_t twiddleCoefF16_2048[4096];
+  #endif /* !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) */
+
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_F16_4096)
+    extern const float16_t twiddleCoefF16_4096[8192];
+    #define twiddleCoefF16 twiddleCoefF16_4096
+  #endif /* !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) */
+  
+ 
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_RFFT_F16_32)
+  extern const float16_t twiddleCoefF16_rfft_32[32];
+  #endif
+
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_RFFT_F16_64)
+  extern const float16_t twiddleCoefF16_rfft_64[64];
+  #endif
+
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_RFFT_F16_128)
+  extern const float16_t twiddleCoefF16_rfft_128[128];
+  #endif
+
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_RFFT_F16_256)
+  extern const float16_t twiddleCoefF16_rfft_256[256];
+  #endif
+
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_RFFT_F16_512)
+  extern const float16_t twiddleCoefF16_rfft_512[512];
+  #endif
+
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_RFFT_F16_1024)
+  extern const float16_t twiddleCoefF16_rfft_1024[1024];
+  #endif
+
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_RFFT_F16_2048)
+  extern const float16_t twiddleCoefF16_rfft_2048[2048];
+  #endif
+
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_RFFT_F16_4096)
+  extern const float16_t twiddleCoefF16_rfft_4096[4096];
+  #endif
+
+  #endif /* ARMAC5 */
+    
+#endif /* !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_FFT_ALLOW_TABLES) */
+
+#if !defined(__CC_ARM) && defined(ARM_FLOAT16_SUPPORTED)
+
+#if (defined(ARM_MATH_MVEF) || defined(ARM_MATH_HELIUM)) && !defined(ARM_MATH_AUTOVECTORIZE)
+       extern const float16_t exp_tab_f16[8];
+       extern const float16_t __logf_lut_f16[8];
+#endif /* (defined(ARM_MATH_MVEF) || defined(ARM_MATH_HELIUM)) && !defined(ARM_MATH_AUTOVECTORIZE) */
+#endif 
+       
+
+#ifdef   __cplusplus
+}
+#endif
+
+#endif /*  _ARM_COMMON_TABLES_F16_H */
+
+  
@@ -0,0 +1,86 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_const_structs.h
+ * Description:  Constant structs that are initialized for user convenience.
+ *               For example, some can be given as arguments to the arm_cfft_f32() function.
+ *
+ * @version  V1.10.0
+ * @date     08 July 2021
+ *
+ * Target Processor: Cortex-M and Cortex-A cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2021 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#ifndef _ARM_CONST_STRUCTS_H
+#define _ARM_CONST_STRUCTS_H
+
+#include "arm_math_types.h"
+#include "arm_common_tables.h"
+#include "dsp/transform_functions.h"
+
+#ifdef   __cplusplus
+extern "C"
+{
+#endif
+   extern const arm_cfft_instance_f64 arm_cfft_sR_f64_len16;
+   extern const arm_cfft_instance_f64 arm_cfft_sR_f64_len32;
+   extern const arm_cfft_instance_f64 arm_cfft_sR_f64_len64;
+   extern const arm_cfft_instance_f64 arm_cfft_sR_f64_len128;
+   extern const arm_cfft_instance_f64 arm_cfft_sR_f64_len256;
+   extern const arm_cfft_instance_f64 arm_cfft_sR_f64_len512;
+   extern const arm_cfft_instance_f64 arm_cfft_sR_f64_len1024;
+   extern const arm_cfft_instance_f64 arm_cfft_sR_f64_len2048;
+   extern const arm_cfft_instance_f64 arm_cfft_sR_f64_len4096;
+
+   extern const arm_cfft_instance_f32 arm_cfft_sR_f32_len16;
+   extern const arm_cfft_instance_f32 arm_cfft_sR_f32_len32;
+   extern const arm_cfft_instance_f32 arm_cfft_sR_f32_len64;
+   extern const arm_cfft_instance_f32 arm_cfft_sR_f32_len128;
+   extern const arm_cfft_instance_f32 arm_cfft_sR_f32_len256;
+   extern const arm_cfft_instance_f32 arm_cfft_sR_f32_len512;
+   extern const arm_cfft_instance_f32 arm_cfft_sR_f32_len1024;
+   extern const arm_cfft_instance_f32 arm_cfft_sR_f32_len2048;
+   extern const arm_cfft_instance_f32 arm_cfft_sR_f32_len4096;
+
+   extern const arm_cfft_instance_q31 arm_cfft_sR_q31_len16;
+   extern const arm_cfft_instance_q31 arm_cfft_sR_q31_len32;
+   extern const arm_cfft_instance_q31 arm_cfft_sR_q31_len64;
+   extern const arm_cfft_instance_q31 arm_cfft_sR_q31_len128;
+   extern const arm_cfft_instance_q31 arm_cfft_sR_q31_len256;
+   extern const arm_cfft_instance_q31 arm_cfft_sR_q31_len512;
+   extern const arm_cfft_instance_q31 arm_cfft_sR_q31_len1024;
+   extern const arm_cfft_instance_q31 arm_cfft_sR_q31_len2048;
+   extern const arm_cfft_instance_q31 arm_cfft_sR_q31_len4096;
+
+   extern const arm_cfft_instance_q15 arm_cfft_sR_q15_len16;
+   extern const arm_cfft_instance_q15 arm_cfft_sR_q15_len32;
+   extern const arm_cfft_instance_q15 arm_cfft_sR_q15_len64;
+   extern const arm_cfft_instance_q15 arm_cfft_sR_q15_len128;
+   extern const arm_cfft_instance_q15 arm_cfft_sR_q15_len256;
+   extern const arm_cfft_instance_q15 arm_cfft_sR_q15_len512;
+   extern const arm_cfft_instance_q15 arm_cfft_sR_q15_len1024;
+   extern const arm_cfft_instance_q15 arm_cfft_sR_q15_len2048;
+   extern const arm_cfft_instance_q15 arm_cfft_sR_q15_len4096;
+
+#ifdef   __cplusplus
+}
+#endif
+
+#endif
+
@@ -0,0 +1,77 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_const_structs_f16.h
+ * Description:  Constant structs that are initialized for user convenience.
+ *               For example, some can be given as arguments to the arm_cfft_f16() function.
+ *
+ * @version  V1.10.0
+ * @date     08 July 2021
+ *
+ * Target Processor: Cortex-M and Cortex-A cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2021 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#ifndef _ARM_CONST_STRUCTS_F16_H
+#define _ARM_CONST_STRUCTS_F16_H
+
+#include "arm_math_types_f16.h"
+#include "arm_common_tables.h"
+#include "arm_common_tables_f16.h"
+#include "dsp/transform_functions_f16.h"
+
+#ifdef   __cplusplus
+extern "C"
+{
+#endif
+
+#if !defined(__CC_ARM) && defined(ARM_FLOAT16_SUPPORTED)
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || (defined(ARM_TABLE_TWIDDLECOEF_F16_16) && defined(ARM_TABLE_BITREVIDX_FLT_16))
+   extern const arm_cfft_instance_f16 arm_cfft_sR_f16_len16;
+   #endif
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || (defined(ARM_TABLE_TWIDDLECOEF_F16_32) && defined(ARM_TABLE_BITREVIDX_FLT_32))
+   extern const arm_cfft_instance_f16 arm_cfft_sR_f16_len32;
+    #endif
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || (defined(ARM_TABLE_TWIDDLECOEF_F16_64) && defined(ARM_TABLE_BITREVIDX_FLT_64))
+   extern const arm_cfft_instance_f16 arm_cfft_sR_f16_len64;
+    #endif
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || (defined(ARM_TABLE_TWIDDLECOEF_F16_128) && defined(ARM_TABLE_BITREVIDX_FLT_128))
+   extern const arm_cfft_instance_f16 arm_cfft_sR_f16_len128;
+    #endif
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || (defined(ARM_TABLE_TWIDDLECOEF_F16_256) && defined(ARM_TABLE_BITREVIDX_FLT_256))
+   extern const arm_cfft_instance_f16 arm_cfft_sR_f16_len256;
+    #endif
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || (defined(ARM_TABLE_TWIDDLECOEF_F16_512) && defined(ARM_TABLE_BITREVIDX_FLT_512))
+   extern const arm_cfft_instance_f16 arm_cfft_sR_f16_len512;
+    #endif
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || (defined(ARM_TABLE_TWIDDLECOEF_F16_1024) && defined(ARM_TABLE_BITREVIDX_FLT_1024))
+   extern const arm_cfft_instance_f16 arm_cfft_sR_f16_len1024;
+    #endif
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || (defined(ARM_TABLE_TWIDDLECOEF_F16_2048) && defined(ARM_TABLE_BITREVIDX_FLT_2048))
+   extern const arm_cfft_instance_f16 arm_cfft_sR_f16_len2048;
+    #endif
+  #if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || (defined(ARM_TABLE_TWIDDLECOEF_F16_4096) && defined(ARM_TABLE_BITREVIDX_FLT_4096))
+   extern const arm_cfft_instance_f16 arm_cfft_sR_f16_len4096;
+  #endif
+#endif
+
+#ifdef   __cplusplus
+}
+#endif
+
+#endif
@@ -0,0 +1,236 @@
+/******************************************************************************
+ * @file     arm_math.h
+ * @brief    Public header file for CMSIS DSP Library
+ * @version  V1.10.0
+ * @date     08 July 2021
+ * Target Processor: Cortex-M and Cortex-A cores
+ ******************************************************************************/
+/*
+ * Copyright (c) 2010-2021 Arm Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+/**
+   \mainpage CMSIS DSP Software Library
+   *
+   * \section intro Introduction
+   *
+   * This user manual describes the CMSIS DSP software library,
+   * a suite of common signal processing functions for use on Cortex-M and Cortex-A processor 
+   * based devices.
+   *
+   * The library is divided into a number of functions each covering a specific category:
+   * - Basic math functions
+   * - Fast math functions
+   * - Complex math functions
+   * - Filtering functions
+   * - Matrix functions
+   * - Transform functions
+   * - Motor control functions
+   * - Statistical functions
+   * - Support functions
+   * - Interpolation functions
+   * - Support Vector Machine functions (SVM)
+   * - Bayes classifier functions
+   * - Distance functions
+   * - Quaternion functions
+   *
+   * The library has generally separate functions for operating on 8-bit integers, 16-bit integers,
+   * 32-bit integer and 32-bit floating-point values.
+   *
+   * The library is providing vectorized versions of most algorthms for Helium
+   * and of most f32 algorithms for Neon.
+   *
+   * When using a vectorized version, provide a little bit of padding after the end of
+   * a buffer (3 words) because the vectorized code may read a little bit after the end
+   * of a buffer. You don't have to modify your buffers but just ensure that the
+   * end of buffer + padding is not outside of a memory region.
+   *
+   * \section using Using the Library
+   *
+   * The library is released in source form. It is strongly advised to compile the library using -Ofast to
+   * have the best performances.
+   *
+   * The library functions are declared in the public file <code>arm_math.h</code> which is placed in the <code>Include</code> folder.
+   * Simply include this file. If you don't want to include everything, you can also rely
+   * on headers in Include/dsp folder and use only what you need.
+   *
+   * \section example Examples
+   *
+   * The library ships with a number of examples which demonstrate how to use the library functions.
+   *
+   * \section toolchain Toolchain Support
+   *
+   * The library is now tested on Fast Models building with cmake.
+   * Core M0, M4, M7, M33, M55, A32 are tested.
+   *
+   *
+   * \section preprocessor Preprocessor Macros
+   *
+   * Each library project have different preprocessor macros.
+   *
+   * - ARM_MATH_BIG_ENDIAN:
+   *
+   * Define macro ARM_MATH_BIG_ENDIAN to build the library for big endian targets. By default library builds for little endian targets.
+   *
+   * - ARM_MATH_MATRIX_CHECK:
+   *
+   * Define macro ARM_MATH_MATRIX_CHECK for checking on the input and output sizes of matrices
+   *
+   * - ARM_MATH_ROUNDING:
+   *
+   * Define macro ARM_MATH_ROUNDING for rounding on support functions
+   *
+   * - ARM_MATH_LOOPUNROLL:
+   *
+   * Define macro ARM_MATH_LOOPUNROLL to enable manual loop unrolling in DSP functions
+   *
+   * - ARM_MATH_NEON:
+   *
+   * Define macro ARM_MATH_NEON to enable Neon versions of the DSP functions.
+   * It is not enabled by default when Neon is available because performances are 
+   * dependent on the compiler and target architecture.
+   *
+   * - ARM_MATH_NEON_EXPERIMENTAL:
+   *
+   * Define macro ARM_MATH_NEON_EXPERIMENTAL to enable experimental Neon versions of 
+   * of some DSP functions. Experimental Neon versions currently do not have better
+   * performances than the scalar versions.
+   *
+   * - ARM_MATH_HELIUM:
+   *
+   * It implies the flags ARM_MATH_MVEF and ARM_MATH_MVEI and ARM_MATH_MVE_FLOAT16.
+   *
+   * - ARM_MATH_HELIUM_EXPERIMENTAL:
+   *
+   * Only taken into account when ARM_MATH_MVEF, ARM_MATH_MVEI or ARM_MATH_MVE_FLOAT16 are defined.
+   * Enable some vector versions which may have worse performance than scalar
+   * depending on the core / compiler configuration.
+   *
+   * - ARM_MATH_MVEF:
+   *
+   * Select Helium versions of the f32 algorithms.
+   * It implies ARM_MATH_FLOAT16 and ARM_MATH_MVEI.
+   *
+   * - ARM_MATH_MVEI:
+   *
+   * Select Helium versions of the int and fixed point algorithms.
+   *
+   * - ARM_MATH_MVE_FLOAT16:
+   *
+   * MVE Float16 implementations of some algorithms (Requires MVE extension).
+   *
+   * - DISABLEFLOAT16:
+   *
+   * Disable float16 algorithms when __fp16 is not supported for a
+   * specific compiler / core configuration.
+   * This is only valid for scalar. When vector architecture is
+   * supporting f16 then it can't be disabled.
+   *
+   * - ARM_MATH_AUTOVECTORIZE:
+   *
+   * With Helium or Neon, disable the use of vectorized code with C intrinsics
+   * and use pure C instead. The vectorization is then done by the compiler.
+   *
+   * <hr>
+   * \section pack CMSIS-DSP in ARM::CMSIS Pack
+   *
+   * The following files relevant to CMSIS-DSP are present in the <b>ARM::CMSIS</b> Pack directories:
+   * |File/Folder                      |Content                                                                 |
+   * |---------------------------------|------------------------------------------------------------------------|
+   * |\b CMSIS\\Documentation\\DSP     | This documentation                                                     |
+   * |\b CMSIS\\DSP\\Examples          | Example projects demonstrating the usage of the library functions      |
+   * |\b CMSIS\\DSP\\Include           | DSP_Lib include files for using and building the lib
+   * |\b CMSIS\\DSP\\PrivateInclude    | DSP_Lib private include files for building the lib                                               |
+   * |\b CMSIS\\DSP\\Lib               | DSP_Lib binaries                                                       |
+   * |\b CMSIS\\DSP\\Source            | DSP_Lib source files                                                   |
+   *
+   * <hr>
+   * \section rev Revision History of CMSIS-DSP
+   * Please refer to \ref ChangeLog_pg.
+   */
+
+
+
+
+
+
+
+
+
+
+
+/**
+ * @defgroup groupExamples Examples
+ */
+
+
+
+
+
+#ifndef _ARM_MATH_H
+#define _ARM_MATH_H
+
+
+#include "arm_math_types.h"
+#include "arm_math_memory.h"
+
+#include "dsp/none.h"
+#include "dsp/utils.h"
+
+#include "dsp/basic_math_functions.h"  
+#include "dsp/interpolation_functions.h"
+#include "dsp/bayes_functions.h"
+#include "dsp/matrix_functions.h"
+#include "dsp/complex_math_functions.h"
+#include "dsp/statistics_functions.h"
+#include "dsp/controller_functions.h"
+#include "dsp/support_functions.h"
+#include "dsp/distance_functions.h"
+#include "dsp/svm_functions.h"
+#include "dsp/fast_math_functions.h"
+#include "dsp/transform_functions.h"
+#include "dsp/filtering_functions.h"
+#include "dsp/quaternion_math_functions.h"
+
+
+
+#ifdef   __cplusplus
+extern "C"
+{
+#endif
+
+
+
+
+//#define TABLE_SPACING_Q31     0x400000
+//#define TABLE_SPACING_Q15     0x80
+
+
+
+
+
+#ifdef   __cplusplus
+}
+#endif
+
+
+#endif /* _ARM_MATH_H */
+
+/**
+ *
+ * End of file.
+ */
@@ -0,0 +1,59 @@
+/******************************************************************************
+ * @file     arm_math_f16.h
+ * @brief    Public header file for f16 function of the CMSIS DSP Library
+ * @version  V1.10.0
+ * @date     08 July 2021
+ * Target Processor: Cortex-M and Cortex-A cores
+ ******************************************************************************/
+/*
+ * Copyright (c) 2010-2021 Arm Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#ifndef _ARM_MATH_F16_H
+#define _ARM_MATH_F16_H
+
+#include "arm_math.h"
+
+#ifdef   __cplusplus
+extern "C"
+{
+#endif
+
+#include "arm_math_types_f16.h"
+#include "dsp/none.h"
+#include "dsp/utils.h"
+#include "dsp/basic_math_functions_f16.h"
+#include "dsp/interpolation_functions_f16.h"
+#include "dsp/bayes_functions_f16.h"
+#include "dsp/matrix_functions_f16.h"
+#include "dsp/complex_math_functions_f16.h"
+#include "dsp/statistics_functions_f16.h"
+#include "dsp/controller_functions_f16.h"
+#include "dsp/support_functions_f16.h"
+#include "dsp/distance_functions_f16.h"
+#include "dsp/svm_functions_f16.h"
+#include "dsp/fast_math_functions_f16.h"
+#include "dsp/transform_functions_f16.h"
+#include "dsp/filtering_functions_f16.h"
+
+#ifdef   __cplusplus
+}
+#endif
+
+#endif /* _ARM_MATH_F16_H */
+
+
@@ -0,0 +1,206 @@
+/******************************************************************************
+ * @file     arm_math_memory.h
+ * @brief    Public header file for CMSIS DSP Library
+ * @version  V1.10.0
+ * @date     08 July 2021
+ * Target Processor: Cortex-M and Cortex-A cores
+ ******************************************************************************/
+/*
+ * Copyright (c) 2010-2021 Arm Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#ifndef _ARM_MATH_MEMORY_H_
+
+#define _ARM_MATH_MEMORY_H_
+
+#include "arm_math_types.h"
+
+
+#ifdef   __cplusplus
+extern "C"
+{
+#endif
+
+/**
+  @brief definition to read/write two 16 bit values.
+  @deprecated
+ */
+#if   defined ( __CC_ARM )
+  #define __SIMD32_TYPE int32_t __packed
+#elif defined ( __ARMCC_VERSION ) && ( __ARMCC_VERSION >= 6010050 )
+  #define __SIMD32_TYPE int32_t
+#elif defined ( __GNUC__ )
+  #define __SIMD32_TYPE int32_t
+#elif defined ( __ICCARM__ )
+  #define __SIMD32_TYPE int32_t __packed
+#elif defined ( __TI_ARM__ )
+  #define __SIMD32_TYPE int32_t
+#elif defined ( __CSMC__ )
+  #define __SIMD32_TYPE int32_t
+#elif defined ( __TASKING__ )
+  #define __SIMD32_TYPE __un(aligned) int32_t
+#elif defined(_MSC_VER )
+  #define __SIMD32_TYPE int32_t
+#else
+  #error Unknown compiler
+#endif
+
+#define __SIMD32(addr)        (*(__SIMD32_TYPE **) & (addr))
+#define __SIMD32_CONST(addr)  ( (__SIMD32_TYPE * )   (addr))
+#define _SIMD32_OFFSET(addr)  (*(__SIMD32_TYPE * )   (addr))
+#define __SIMD64(addr)        (*(      int64_t **) & (addr))
+
+
+/* SIMD replacement */
+
+
+/**
+  @brief         Read 2 Q15 from Q15 pointer.
+  @param[in]     pQ15      points to input value
+  @return        Q31 value
+ */
+__STATIC_FORCEINLINE q31_t read_q15x2 (
+  q15_t const * pQ15)
+{
+  q31_t val;
+
+#ifdef __ARM_FEATURE_UNALIGNED
+  memcpy (&val, pQ15, 4);
+#else
+  val = (pQ15[1] << 16) | (pQ15[0] & 0x0FFFF) ;
+#endif
+
+  return (val);
+}
+
+/**
+  @brief         Read 2 Q15 from Q15 pointer and increment pointer afterwards.
+  @param[in]     pQ15      points to input value
+  @return        Q31 value
+ */
+#define read_q15x2_ia(pQ15) read_q15x2((*(pQ15) += 2) - 2)
+
+/**
+  @brief         Read 2 Q15 from Q15 pointer and decrement pointer afterwards.
+  @param[in]     pQ15      points to input value
+  @return        Q31 value
+ */
+#define read_q15x2_da(pQ15) read_q15x2((*(pQ15) -= 2) + 2)
+
+/**
+  @brief         Write 2 Q15 to Q15 pointer and increment pointer afterwards.
+  @param[in]     pQ15      points to input value
+  @param[in]     value     Q31 value
+  @return        none
+ */
+__STATIC_FORCEINLINE void write_q15x2_ia (
+  q15_t ** pQ15,
+  q31_t    value)
+{
+  q31_t val = value;
+#ifdef __ARM_FEATURE_UNALIGNED
+  memcpy (*pQ15, &val, 4);
+#else
+  (*pQ15)[0] = (q15_t)(val & 0x0FFFF);
+  (*pQ15)[1] = (q15_t)((val >> 16) & 0x0FFFF);
+#endif
+
+ *pQ15 += 2;
+}
+
+/**
+  @brief         Write 2 Q15 to Q15 pointer.
+  @param[in]     pQ15      points to input value
+  @param[in]     value     Q31 value
+  @return        none
+ */
+__STATIC_FORCEINLINE void write_q15x2 (
+  q15_t * pQ15,
+  q31_t   value)
+{
+  q31_t val = value;
+
+#ifdef __ARM_FEATURE_UNALIGNED
+  memcpy (pQ15, &val, 4);
+#else
+  pQ15[0] = (q15_t)(val & 0x0FFFF);
+  pQ15[1] = (q15_t)(val >> 16);
+#endif
+}
+
+
+/**
+  @brief         Read 4 Q7 from Q7 pointer
+  @param[in]     pQ7       points to input value
+  @return        Q31 value
+ */
+__STATIC_FORCEINLINE q31_t read_q7x4 (
+  q7_t const * pQ7)
+{
+  q31_t val;
+
+#ifdef __ARM_FEATURE_UNALIGNED
+  memcpy (&val, pQ7, 4);
+#else
+  val =((pQ7[3] & 0x0FF) << 24)  | ((pQ7[2] & 0x0FF) << 16)  | ((pQ7[1] & 0x0FF) << 8)  | (pQ7[0] & 0x0FF);
+#endif 
+  return (val);
+}
+
+/**
+  @brief         Read 4 Q7 from Q7 pointer and increment pointer afterwards.
+  @param[in]     pQ7       points to input value
+  @return        Q31 value
+ */
+#define read_q7x4_ia(pQ7) read_q7x4((*(pQ7) += 4) - 4)
+
+/**
+  @brief         Read 4 Q7 from Q7 pointer and decrement pointer afterwards.
+  @param[in]     pQ7       points to input value
+  @return        Q31 value
+ */
+#define read_q7x4_da(pQ7) read_q7x4((*(pQ7) -= 4) + 4)
+
+/**
+  @brief         Write 4 Q7 to Q7 pointer and increment pointer afterwards.
+  @param[in]     pQ7       points to input value
+  @param[in]     value     Q31 value
+  @return        none
+ */
+__STATIC_FORCEINLINE void write_q7x4_ia (
+  q7_t ** pQ7,
+  q31_t   value)
+{
+  q31_t val = value;
+#ifdef __ARM_FEATURE_UNALIGNED
+  memcpy (*pQ7, &val, 4);
+#else
+  (*pQ7)[0] = (q7_t)(val & 0x0FF);
+  (*pQ7)[1] = (q7_t)((val >> 8) & 0x0FF);
+  (*pQ7)[2] = (q7_t)((val >> 16) & 0x0FF);
+  (*pQ7)[3] = (q7_t)((val >> 24) & 0x0FF);
+
+#endif
+  *pQ7 += 4;
+}
+
+
+#ifdef   __cplusplus
+}
+#endif
+
+#endif /*ifndef _ARM_MATH_MEMORY_H_ */
@@ -0,0 +1,163 @@
+/******************************************************************************
+ * @file     arm_math_types_f16.h
+ * @brief    Public header file for f16 function of the CMSIS DSP Library
+ * @version  V1.10.0
+ * @date     08 July 2021
+ * Target Processor: Cortex-M and Cortex-A cores
+ ******************************************************************************/
+/*
+ * Copyright (c) 2010-2021 Arm Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#ifndef _ARM_MATH_TYPES_F16_H
+#define _ARM_MATH_TYPES_F16_H
+
+#include "arm_math_types.h"
+
+#ifdef   __cplusplus
+extern "C"
+{
+#endif
+
+#if !defined( __CC_ARM )
+
+/**
+ * @brief 16-bit floating-point type definition.
+ * This is already defined in arm_mve.h
+ *
+ * This is not fully supported on ARM AC5.
+ */
+
+/*
+
+Check if the type __fp16 is available.
+If it is not available, f16 version of the kernels
+won't be built.
+
+*/
+#if !(__ARM_FEATURE_MVE & 2)
+  #if !defined(DISABLEFLOAT16)
+    #if defined(__ARM_FP16_FORMAT_IEEE) || defined(__ARM_FP16_FORMAT_ALTERNATIVE)
+      typedef __fp16 float16_t;
+      #define ARM_FLOAT16_SUPPORTED
+    #endif
+  #endif
+#else
+  /* When Vector float16, this flag is always defined and can't be disabled */
+  #define ARM_FLOAT16_SUPPORTED
+#endif
+
+#if defined(ARM_MATH_NEON) || (defined(ARM_MATH_MVEF)  && !defined(ARM_MATH_AUTOVECTORIZE)) /* floating point vector*/
+
+#if defined(ARM_MATH_MVE_FLOAT16) || defined(ARM_MATH_NEON_FLOAT16)
+
+  /**
+   * @brief 16-bit floating-point 128-bit vector data type
+   */
+  typedef __ALIGNED(2) float16x8_t f16x8_t;
+
+  /**
+   * @brief 16-bit floating-point 128-bit vector pair data type
+   */
+  typedef float16x8x2_t f16x8x2_t;
+
+  /**
+   * @brief 16-bit floating-point 128-bit vector quadruplet data type
+   */
+  typedef float16x8x4_t f16x8x4_t;
+
+  /**
+   * @brief 16-bit ubiquitous 128-bit vector data type
+   */
+  typedef union _any16x8_t
+  {
+      float16x8_t     f;
+      int16x8_t       i;
+  } any16x8_t;
+#endif
+
+#endif
+
+#if defined(ARM_MATH_NEON)
+
+
+#if defined(ARM_MATH_NEON_FLOAT16)
+  /**
+   * @brief 16-bit float 64-bit vector data type.
+   */
+  typedef  __ALIGNED(2) float16x4_t f16x4_t;
+
+  /**
+   * @brief 16-bit floating-point 128-bit vector triplet data type
+   */
+  typedef float16x8x3_t f16x8x3_t;
+
+  /**
+   * @brief 16-bit floating-point 64-bit vector pair data type
+   */
+  typedef float16x4x2_t f16x4x2_t;
+
+  /**
+   * @brief 16-bit floating-point 64-bit vector triplet data type
+   */
+  typedef float16x4x3_t f16x4x3_t;
+
+  /**
+   * @brief 16-bit floating-point 64-bit vector quadruplet data type
+   */
+  typedef float16x4x4_t f16x4x4_t;
+
+  /**
+   * @brief 16-bit ubiquitous 64-bit vector data type
+   */
+  typedef union _any16x4_t
+  {
+      float16x4_t     f;
+      int16x4_t       i;
+  } any16x4_t;
+#endif
+
+#endif
+
+
+
+#if defined(ARM_FLOAT16_SUPPORTED)
+
+#if defined(__ICCARM__)
+
+#define F16INFINITY   ((float16_t) INFINITY)
+
+#else
+
+#define F16INFINITY ((float16_t)__builtin_inf())
+
+#endif
+
+#define F16_MAX   ((float16_t)__FLT16_MAX__)
+#define F16_MIN   (-(_Float16)__FLT16_MAX__)
+
+#define F16_ABSMAX   ((float16_t)__FLT16_MAX__)
+#define F16_ABSMIN   ((float16_t)0.0f16)
+
+#endif /* ARM_FLOAT16_SUPPORTED*/
+#endif /* !defined( __CC_ARM ) */
+
+#ifdef   __cplusplus
+}
+#endif
+
+#endif /* _ARM_MATH_F16_H */
@@ -0,0 +1,231 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mve_tables.h
+ * Description:  common tables like fft twiddle factors, Bitreverse, reciprocal etc
+ *               used for MVE implementation only
+ *
+ * @version  V1.10.0
+ * @date     04 October 2021
+ *
+ * Target Processor: Cortex-M and Cortex-A cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2021 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+ #ifndef _ARM_MVE_TABLES_H
+ #define _ARM_MVE_TABLES_H
+
+#include "arm_math_types.h"
+
+#ifdef   __cplusplus
+extern "C"
+{
+#endif
+
+
+ 
+
+#if defined(ARM_MATH_MVEF) && !defined(ARM_MATH_AUTOVECTORIZE)
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_FFT_ALLOW_TABLES)
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_F32_16) || defined(ARM_TABLE_TWIDDLECOEF_F32_32)
+
+extern uint32_t rearranged_twiddle_tab_stride1_arr_16_f32[2];
+extern uint32_t rearranged_twiddle_tab_stride2_arr_16_f32[2];
+extern uint32_t rearranged_twiddle_tab_stride3_arr_16_f32[2];
+extern float32_t rearranged_twiddle_stride1_16_f32[8];
+extern float32_t rearranged_twiddle_stride2_16_f32[8];
+extern float32_t rearranged_twiddle_stride3_16_f32[8];
+#endif
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_F32_64) || defined(ARM_TABLE_TWIDDLECOEF_F32_128)
+
+extern uint32_t rearranged_twiddle_tab_stride1_arr_64_f32[3];
+extern uint32_t rearranged_twiddle_tab_stride2_arr_64_f32[3];
+extern uint32_t rearranged_twiddle_tab_stride3_arr_64_f32[3];
+extern float32_t rearranged_twiddle_stride1_64_f32[40];
+extern float32_t rearranged_twiddle_stride2_64_f32[40];
+extern float32_t rearranged_twiddle_stride3_64_f32[40];
+#endif
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_F32_256) || defined(ARM_TABLE_TWIDDLECOEF_F32_512)
+
+extern uint32_t rearranged_twiddle_tab_stride1_arr_256_f32[4];
+extern uint32_t rearranged_twiddle_tab_stride2_arr_256_f32[4];
+extern uint32_t rearranged_twiddle_tab_stride3_arr_256_f32[4];
+extern float32_t rearranged_twiddle_stride1_256_f32[168];
+extern float32_t rearranged_twiddle_stride2_256_f32[168];
+extern float32_t rearranged_twiddle_stride3_256_f32[168];
+#endif
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_F32_1024) || defined(ARM_TABLE_TWIDDLECOEF_F32_2048)
+
+extern uint32_t rearranged_twiddle_tab_stride1_arr_1024_f32[5];
+extern uint32_t rearranged_twiddle_tab_stride2_arr_1024_f32[5];
+extern uint32_t rearranged_twiddle_tab_stride3_arr_1024_f32[5];
+extern float32_t rearranged_twiddle_stride1_1024_f32[680];
+extern float32_t rearranged_twiddle_stride2_1024_f32[680];
+extern float32_t rearranged_twiddle_stride3_1024_f32[680];
+#endif
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_F32_4096) || defined(ARM_TABLE_TWIDDLECOEF_F32_8192)
+
+extern uint32_t rearranged_twiddle_tab_stride1_arr_4096_f32[6];
+extern uint32_t rearranged_twiddle_tab_stride2_arr_4096_f32[6];
+extern uint32_t rearranged_twiddle_tab_stride3_arr_4096_f32[6];
+extern float32_t rearranged_twiddle_stride1_4096_f32[2728];
+extern float32_t rearranged_twiddle_stride2_4096_f32[2728];
+extern float32_t rearranged_twiddle_stride3_4096_f32[2728];
+#endif
+
+
+#endif /* !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_FFT_ALLOW_TABLES) */
+
+#endif /* defined(ARM_MATH_MVEF) && !defined(ARM_MATH_AUTOVECTORIZE) */
+
+
+
+#if defined(ARM_MATH_MVEI)  && !defined(ARM_MATH_AUTOVECTORIZE)
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_FFT_ALLOW_TABLES)
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_Q31_16) || defined(ARM_TABLE_TWIDDLECOEF_Q31_32)
+
+extern uint32_t rearranged_twiddle_tab_stride1_arr_16_q31[2];
+extern uint32_t rearranged_twiddle_tab_stride2_arr_16_q31[2];
+extern uint32_t rearranged_twiddle_tab_stride3_arr_16_q31[2];
+extern q31_t rearranged_twiddle_stride1_16_q31[8];
+extern q31_t rearranged_twiddle_stride2_16_q31[8];
+extern q31_t rearranged_twiddle_stride3_16_q31[8];
+#endif
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_Q31_64) || defined(ARM_TABLE_TWIDDLECOEF_Q31_128)
+
+extern uint32_t rearranged_twiddle_tab_stride1_arr_64_q31[3];
+extern uint32_t rearranged_twiddle_tab_stride2_arr_64_q31[3];
+extern uint32_t rearranged_twiddle_tab_stride3_arr_64_q31[3];
+extern q31_t rearranged_twiddle_stride1_64_q31[40];
+extern q31_t rearranged_twiddle_stride2_64_q31[40];
+extern q31_t rearranged_twiddle_stride3_64_q31[40];
+#endif
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_Q31_256) || defined(ARM_TABLE_TWIDDLECOEF_Q31_512)
+
+extern uint32_t rearranged_twiddle_tab_stride1_arr_256_q31[4];
+extern uint32_t rearranged_twiddle_tab_stride2_arr_256_q31[4];
+extern uint32_t rearranged_twiddle_tab_stride3_arr_256_q31[4];
+extern q31_t rearranged_twiddle_stride1_256_q31[168];
+extern q31_t rearranged_twiddle_stride2_256_q31[168];
+extern q31_t rearranged_twiddle_stride3_256_q31[168];
+#endif
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_Q31_1024) || defined(ARM_TABLE_TWIDDLECOEF_Q31_2048)
+
+extern uint32_t rearranged_twiddle_tab_stride1_arr_1024_q31[5];
+extern uint32_t rearranged_twiddle_tab_stride2_arr_1024_q31[5];
+extern uint32_t rearranged_twiddle_tab_stride3_arr_1024_q31[5];
+extern q31_t rearranged_twiddle_stride1_1024_q31[680];
+extern q31_t rearranged_twiddle_stride2_1024_q31[680];
+extern q31_t rearranged_twiddle_stride3_1024_q31[680];
+#endif
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_Q31_4096) || defined(ARM_TABLE_TWIDDLECOEF_Q31_8192)
+
+extern uint32_t rearranged_twiddle_tab_stride1_arr_4096_q31[6];
+extern uint32_t rearranged_twiddle_tab_stride2_arr_4096_q31[6];
+extern uint32_t rearranged_twiddle_tab_stride3_arr_4096_q31[6];
+extern q31_t rearranged_twiddle_stride1_4096_q31[2728];
+extern q31_t rearranged_twiddle_stride2_4096_q31[2728];
+extern q31_t rearranged_twiddle_stride3_4096_q31[2728];
+#endif
+
+
+#endif /* !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_FFT_ALLOW_TABLES) */
+
+#endif /* defined(ARM_MATH_MVEI) */
+
+
+
+#if defined(ARM_MATH_MVEI)  && !defined(ARM_MATH_AUTOVECTORIZE)
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_FFT_ALLOW_TABLES)
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_Q15_16) || defined(ARM_TABLE_TWIDDLECOEF_Q15_32)
+
+extern uint32_t rearranged_twiddle_tab_stride1_arr_16_q15[2];
+extern uint32_t rearranged_twiddle_tab_stride2_arr_16_q15[2];
+extern uint32_t rearranged_twiddle_tab_stride3_arr_16_q15[2];
+extern q15_t rearranged_twiddle_stride1_16_q15[8];
+extern q15_t rearranged_twiddle_stride2_16_q15[8];
+extern q15_t rearranged_twiddle_stride3_16_q15[8];
+#endif
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_Q15_64) || defined(ARM_TABLE_TWIDDLECOEF_Q15_128)
+
+extern uint32_t rearranged_twiddle_tab_stride1_arr_64_q15[3];
+extern uint32_t rearranged_twiddle_tab_stride2_arr_64_q15[3];
+extern uint32_t rearranged_twiddle_tab_stride3_arr_64_q15[3];
+extern q15_t rearranged_twiddle_stride1_64_q15[40];
+extern q15_t rearranged_twiddle_stride2_64_q15[40];
+extern q15_t rearranged_twiddle_stride3_64_q15[40];
+#endif
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_Q15_256) || defined(ARM_TABLE_TWIDDLECOEF_Q15_512)
+
+extern uint32_t rearranged_twiddle_tab_stride1_arr_256_q15[4];
+extern uint32_t rearranged_twiddle_tab_stride2_arr_256_q15[4];
+extern uint32_t rearranged_twiddle_tab_stride3_arr_256_q15[4];
+extern q15_t rearranged_twiddle_stride1_256_q15[168];
+extern q15_t rearranged_twiddle_stride2_256_q15[168];
+extern q15_t rearranged_twiddle_stride3_256_q15[168];
+#endif
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_Q15_1024) || defined(ARM_TABLE_TWIDDLECOEF_Q15_2048)
+
+extern uint32_t rearranged_twiddle_tab_stride1_arr_1024_q15[5];
+extern uint32_t rearranged_twiddle_tab_stride2_arr_1024_q15[5];
+extern uint32_t rearranged_twiddle_tab_stride3_arr_1024_q15[5];
+extern q15_t rearranged_twiddle_stride1_1024_q15[680];
+extern q15_t rearranged_twiddle_stride2_1024_q15[680];
+extern q15_t rearranged_twiddle_stride3_1024_q15[680];
+#endif
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_Q15_4096) || defined(ARM_TABLE_TWIDDLECOEF_Q15_8192)
+
+extern uint32_t rearranged_twiddle_tab_stride1_arr_4096_q15[6];
+extern uint32_t rearranged_twiddle_tab_stride2_arr_4096_q15[6];
+extern uint32_t rearranged_twiddle_tab_stride3_arr_4096_q15[6];
+extern q15_t rearranged_twiddle_stride1_4096_q15[2728];
+extern q15_t rearranged_twiddle_stride2_4096_q15[2728];
+extern q15_t rearranged_twiddle_stride3_4096_q15[2728];
+#endif
+
+
+#endif /* !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_FFT_ALLOW_TABLES) */
+
+#endif /* defined(ARM_MATH_MVEI) */
+
+
+
+#ifdef   __cplusplus
+}
+#endif
+
+#endif /*_ARM_MVE_TABLES_H*/
+
@@ -0,0 +1,109 @@
+/* ----------------------------------------------------------------------
+ * Project:      CMSIS DSP Library
+ * Title:        arm_mve_tables_f16.h
+ * Description:  common tables like fft twiddle factors, Bitreverse, reciprocal etc
+ *               used for MVE implementation only
+ *
+ * @version  V1.10.0
+ * @date     04 October 2021
+ *
+ * Target Processor: Cortex-M and Cortex-A cores
+ * -------------------------------------------------------------------- */
+/*
+ * Copyright (C) 2010-2021 ARM Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+ #ifndef _ARM_MVE_TABLES_F16_H
+ #define _ARM_MVE_TABLES_F16_H
+
+#include "arm_math_types_f16.h"
+
+#ifdef   __cplusplus
+extern "C"
+{
+#endif
+
+
+ 
+
+#if defined(ARM_MATH_MVE_FLOAT16) && !defined(ARM_MATH_AUTOVECTORIZE)
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_FFT_ALLOW_TABLES)
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_F16_16) || defined(ARM_TABLE_TWIDDLECOEF_F16_32)
+
+extern uint32_t rearranged_twiddle_tab_stride1_arr_16_f16[2];
+extern uint32_t rearranged_twiddle_tab_stride2_arr_16_f16[2];
+extern uint32_t rearranged_twiddle_tab_stride3_arr_16_f16[2];
+extern float16_t rearranged_twiddle_stride1_16_f16[8];
+extern float16_t rearranged_twiddle_stride2_16_f16[8];
+extern float16_t rearranged_twiddle_stride3_16_f16[8];
+#endif
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_F16_64) || defined(ARM_TABLE_TWIDDLECOEF_F16_128)
+
+extern uint32_t rearranged_twiddle_tab_stride1_arr_64_f16[3];
+extern uint32_t rearranged_twiddle_tab_stride2_arr_64_f16[3];
+extern uint32_t rearranged_twiddle_tab_stride3_arr_64_f16[3];
+extern float16_t rearranged_twiddle_stride1_64_f16[40];
+extern float16_t rearranged_twiddle_stride2_64_f16[40];
+extern float16_t rearranged_twiddle_stride3_64_f16[40];
+#endif
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_F16_256) || defined(ARM_TABLE_TWIDDLECOEF_F16_512)
+
+extern uint32_t rearranged_twiddle_tab_stride1_arr_256_f16[4];
+extern uint32_t rearranged_twiddle_tab_stride2_arr_256_f16[4];
+extern uint32_t rearranged_twiddle_tab_stride3_arr_256_f16[4];
+extern float16_t rearranged_twiddle_stride1_256_f16[168];
+extern float16_t rearranged_twiddle_stride2_256_f16[168];
+extern float16_t rearranged_twiddle_stride3_256_f16[168];
+#endif
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_F16_1024) || defined(ARM_TABLE_TWIDDLECOEF_F16_2048)
+
+extern uint32_t rearranged_twiddle_tab_stride1_arr_1024_f16[5];
+extern uint32_t rearranged_twiddle_tab_stride2_arr_1024_f16[5];
+extern uint32_t rearranged_twiddle_tab_stride3_arr_1024_f16[5];
+extern float16_t rearranged_twiddle_stride1_1024_f16[680];
+extern float16_t rearranged_twiddle_stride2_1024_f16[680];
+extern float16_t rearranged_twiddle_stride3_1024_f16[680];
+#endif
+
+#if !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_ALL_FFT_TABLES) || defined(ARM_TABLE_TWIDDLECOEF_F16_4096) || defined(ARM_TABLE_TWIDDLECOEF_F16_8192)
+
+extern uint32_t rearranged_twiddle_tab_stride1_arr_4096_f16[6];
+extern uint32_t rearranged_twiddle_tab_stride2_arr_4096_f16[6];
+extern uint32_t rearranged_twiddle_tab_stride3_arr_4096_f16[6];
+extern float16_t rearranged_twiddle_stride1_4096_f16[2728];
+extern float16_t rearranged_twiddle_stride2_4096_f16[2728];
+extern float16_t rearranged_twiddle_stride3_4096_f16[2728];
+#endif
+
+
+#endif /* !defined(ARM_DSP_CONFIG_TABLES) || defined(ARM_FFT_ALLOW_TABLES) */
+
+#endif /* defined(ARM_MATH_MVE_FLOAT16) && !defined(ARM_MATH_AUTOVECTORIZE) */
+
+
+
+#ifdef   __cplusplus
+}
+#endif
+
+#endif /*_ARM_MVE_TABLES_F16_H*/
+
@@ -0,0 +1,373 @@
+/******************************************************************************
+ * @file     arm_vec_math.h
+ * @brief    Public header file for CMSIS DSP Library
+ * @version  V1.10.0
+ * @date     08 July 2021
+ * Target Processor: Cortex-M and Cortex-A cores
+ ******************************************************************************/
+/*
+ * Copyright (c) 2010-2021 Arm Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#ifndef _ARM_VEC_MATH_H
+#define _ARM_VEC_MATH_H
+
+#include "arm_math_types.h"
+#include "arm_common_tables.h"
+#include "arm_helium_utils.h"
+
+#ifdef   __cplusplus
+extern "C"
+{
+#endif
+
+#if (defined(ARM_MATH_MVEF) || defined(ARM_MATH_HELIUM)) && !defined(ARM_MATH_AUTOVECTORIZE)
+
+#define INV_NEWTON_INIT_F32         0x7EF127EA
+
+static const float32_t __logf_rng_f32=0.693147180f;
+
+
+/* fast inverse approximation (3x newton) */
+__STATIC_INLINE f32x4_t vrecip_medprec_f32(
+    f32x4_t x)
+{
+    q31x4_t         m;
+    f32x4_t         b;
+    any32x4_t       xinv;
+    f32x4_t         ax = vabsq(x);
+
+    xinv.f = ax;
+    m = 0x3F800000 - (xinv.i & 0x7F800000);
+    xinv.i = xinv.i + m;
+    xinv.f = 1.41176471f - 0.47058824f * xinv.f;
+    xinv.i = xinv.i + m;
+
+    b = 2.0f - xinv.f * ax;
+    xinv.f = xinv.f * b;
+
+    b = 2.0f - xinv.f * ax;
+    xinv.f = xinv.f * b;
+
+    b = 2.0f - xinv.f * ax;
+    xinv.f = xinv.f * b;
+
+    xinv.f = vdupq_m(xinv.f, INFINITY, vcmpeqq(x, 0.0f));
+    /*
+     * restore sign
+     */
+    xinv.f = vnegq_m(xinv.f, xinv.f, vcmpltq(x, 0.0f));
+
+    return xinv.f;
+}
+
+/* fast inverse approximation (4x newton) */
+__STATIC_INLINE f32x4_t vrecip_hiprec_f32(
+    f32x4_t x)
+{
+    q31x4_t         m;
+    f32x4_t         b;
+    any32x4_t       xinv;
+    f32x4_t         ax = vabsq(x);
+
+    xinv.f = ax;
+
+    m = 0x3F800000 - (xinv.i & 0x7F800000);
+    xinv.i = xinv.i + m;
+    xinv.f = 1.41176471f - 0.47058824f * xinv.f;
+    xinv.i = xinv.i + m;
+
+    b = 2.0f - xinv.f * ax;
+    xinv.f = xinv.f * b;
+
+    b = 2.0f - xinv.f * ax;
+    xinv.f = xinv.f * b;
+
+    b = 2.0f - xinv.f * ax;
+    xinv.f = xinv.f * b;
+
+    b = 2.0f - xinv.f * ax;
+    xinv.f = xinv.f * b;
+
+    xinv.f = vdupq_m(xinv.f, INFINITY, vcmpeqq(x, 0.0f));
+    /*
+     * restore sign
+     */
+    xinv.f = vnegq_m(xinv.f, xinv.f, vcmpltq(x, 0.0f));
+
+    return xinv.f;
+}
+
+__STATIC_INLINE f32x4_t vdiv_f32(
+    f32x4_t num, f32x4_t den)
+{
+    return vmulq(num, vrecip_hiprec_f32(den));
+}
+
+/**
+  @brief         Single-precision taylor dev.
+  @param[in]     x              f32 quad vector input
+  @param[in]     coeffs         f32 quad vector coeffs
+  @return        destination    f32 quad vector
+ */
+
+__STATIC_INLINE f32x4_t vtaylor_polyq_f32(
+        f32x4_t           x,
+        const float32_t * coeffs)
+{
+    f32x4_t         A = vfmasq(vdupq_n_f32(coeffs[4]), x, coeffs[0]);
+    f32x4_t         B = vfmasq(vdupq_n_f32(coeffs[6]), x, coeffs[2]);
+    f32x4_t         C = vfmasq(vdupq_n_f32(coeffs[5]), x, coeffs[1]);
+    f32x4_t         D = vfmasq(vdupq_n_f32(coeffs[7]), x, coeffs[3]);
+    f32x4_t         x2 = vmulq(x, x);
+    f32x4_t         x4 = vmulq(x2, x2);
+    f32x4_t         res = vfmaq(vfmaq_f32(A, B, x2), vfmaq_f32(C, D, x2), x4);
+
+    return res;
+}
+
+__STATIC_INLINE f32x4_t vmant_exp_f32(
+    f32x4_t     x,
+    int32x4_t * e)
+{
+    any32x4_t       r;
+    int32x4_t       n;
+
+    r.f = x;
+    n = r.i >> 23;
+    n = n - 127;
+    r.i = r.i - (n << 23);
+
+    *e = n;
+    return r.f;
+}
+
+
+__STATIC_INLINE f32x4_t vlogq_f32(f32x4_t vecIn)
+{
+    q31x4_t         vecExpUnBiased;
+    f32x4_t         vecTmpFlt0, vecTmpFlt1;
+    f32x4_t         vecAcc0, vecAcc1, vecAcc2, vecAcc3;
+    f32x4_t         vecExpUnBiasedFlt;
+
+    /*
+     * extract exponent
+     */
+    vecTmpFlt1 = vmant_exp_f32(vecIn, &vecExpUnBiased);
+
+    vecTmpFlt0 = vecTmpFlt1 * vecTmpFlt1;
+    /*
+     * a = (__logf_lut_f32[4] * r.f) + (__logf_lut_f32[0]);
+     */
+    vecAcc0 = vdupq_n_f32(__logf_lut_f32[0]);
+    vecAcc0 = vfmaq(vecAcc0, vecTmpFlt1, __logf_lut_f32[4]);
+    /*
+     * b = (__logf_lut_f32[6] * r.f) + (__logf_lut_f32[2]);
+     */
+    vecAcc1 = vdupq_n_f32(__logf_lut_f32[2]);
+    vecAcc1 = vfmaq(vecAcc1, vecTmpFlt1, __logf_lut_f32[6]);
+    /*
+     * c = (__logf_lut_f32[5] * r.f) + (__logf_lut_f32[1]);
+     */
+    vecAcc2 = vdupq_n_f32(__logf_lut_f32[1]);
+    vecAcc2 = vfmaq(vecAcc2, vecTmpFlt1, __logf_lut_f32[5]);
+    /*
+     * d = (__logf_lut_f32[7] * r.f) + (__logf_lut_f32[3]);
+     */
+    vecAcc3 = vdupq_n_f32(__logf_lut_f32[3]);
+    vecAcc3 = vfmaq(vecAcc3, vecTmpFlt1, __logf_lut_f32[7]);
+    /*
+     * a = a + b * xx;
+     */
+    vecAcc0 = vfmaq(vecAcc0, vecAcc1, vecTmpFlt0);
+    /*
+     * c = c + d * xx;
+     */
+    vecAcc2 = vfmaq(vecAcc2, vecAcc3, vecTmpFlt0);
+    /*
+     * xx = xx * xx;
+     */
+    vecTmpFlt0 = vecTmpFlt0 * vecTmpFlt0;
+    vecExpUnBiasedFlt = vcvtq_f32_s32(vecExpUnBiased);
+    /*
+     * r.f = a + c * xx;
+     */
+    vecAcc0 = vfmaq(vecAcc0, vecAcc2, vecTmpFlt0);
+    /*
+     * add exponent
+     * r.f = r.f + ((float32_t) m) * __logf_rng_f32;
+     */
+    vecAcc0 = vfmaq(vecAcc0, vecExpUnBiasedFlt, __logf_rng_f32);
+    // set log0 down to -inf
+    vecAcc0 = vdupq_m(vecAcc0, -INFINITY, vcmpeqq(vecIn, 0.0f));
+    return vecAcc0;
+}
+
+__STATIC_INLINE f32x4_t vexpq_f32(
+    f32x4_t x)
+{
+    // Perform range reduction [-log(2),log(2)]
+    int32x4_t       m = vcvtq_s32_f32(vmulq_n_f32(x, 1.4426950408f));
+    f32x4_t         val = vfmsq_f32(x, vcvtq_f32_s32(m), vdupq_n_f32(0.6931471805f));
+
+    // Polynomial Approximation
+    f32x4_t         poly = vtaylor_polyq_f32(val, exp_tab);
+
+    // Reconstruct
+    poly = (f32x4_t) (vqaddq_s32((q31x4_t) (poly), vqshlq_n_s32(m, 23)));
+
+    poly = vdupq_m(poly, 0.0f, vcmpltq_n_s32(m, -126));
+    return poly;
+}
+
+__STATIC_INLINE f32x4_t arm_vec_exponent_f32(f32x4_t x, int32_t nb)
+{
+    f32x4_t         r = x;
+    nb--;
+    while (nb > 0) {
+        r = vmulq(r, x);
+        nb--;
+    }
+    return (r);
+}
+
+__STATIC_INLINE f32x4_t vrecip_f32(f32x4_t vecIn)
+{
+    f32x4_t     vecSx, vecW, vecTmp;
+    any32x4_t   v;
+
+    vecSx = vabsq(vecIn);
+
+    v.f = vecIn;
+    v.i = vsubq(vdupq_n_s32(INV_NEWTON_INIT_F32), v.i);
+
+    vecW = vmulq(vecSx, v.f);
+
+    // v.f = v.f * (8 + w * (-28 + w * (56 + w * (-70 + w *(56 + w * (-28 + w * (8 - w)))))));
+    vecTmp = vsubq(vdupq_n_f32(8.0f), vecW);
+    vecTmp = vfmasq(vecW, vecTmp, -28.0f);
+    vecTmp = vfmasq(vecW, vecTmp, 56.0f);
+    vecTmp = vfmasq(vecW, vecTmp, -70.0f);
+    vecTmp = vfmasq(vecW, vecTmp, 56.0f);
+    vecTmp = vfmasq(vecW, vecTmp, -28.0f);
+    vecTmp = vfmasq(vecW, vecTmp, 8.0f);
+    v.f = vmulq(v.f,  vecTmp);
+
+    v.f = vdupq_m(v.f, INFINITY, vcmpeqq(vecIn, 0.0f));
+    /*
+     * restore sign
+     */
+    v.f = vnegq_m(v.f, v.f, vcmpltq(vecIn, 0.0f));
+    return v.f;
+}
+
+__STATIC_INLINE f32x4_t vtanhq_f32(
+    f32x4_t val)
+{
+    f32x4_t         x =
+        vminnmq_f32(vmaxnmq_f32(val, vdupq_n_f32(-10.f)), vdupq_n_f32(10.0f));
+    f32x4_t         exp2x = vexpq_f32(vmulq_n_f32(x, 2.f));
+    f32x4_t         num = vsubq_n_f32(exp2x, 1.f);
+    f32x4_t         den = vaddq_n_f32(exp2x, 1.f);
+    f32x4_t         tanh = vmulq_f32(num, vrecip_f32(den));
+    return tanh;
+}
+
+__STATIC_INLINE f32x4_t vpowq_f32(
+    f32x4_t val,
+    f32x4_t n)
+{
+    return vexpq_f32(vmulq_f32(n, vlogq_f32(val)));
+}
+
+#endif /* (defined(ARM_MATH_MVEF) || defined(ARM_MATH_HELIUM)) && !defined(ARM_MATH_AUTOVECTORIZE)*/
+
+#if (defined(ARM_MATH_MVEI) || defined(ARM_MATH_HELIUM)) && !defined(ARM_MATH_AUTOVECTORIZE)
+#endif /* (defined(ARM_MATH_MVEI) || defined(ARM_MATH_HELIUM)) */
+
+#if (defined(ARM_MATH_NEON) || defined(ARM_MATH_NEON_EXPERIMENTAL)) && !defined(ARM_MATH_AUTOVECTORIZE)
+
+#include "NEMath.h"
+/**
+ * @brief Vectorized integer exponentiation
+ * @param[in]    x           value
+ * @param[in]    nb          integer exponent >= 1
+ * @return x^nb
+ *
+ */
+__STATIC_INLINE  float32x4_t arm_vec_exponent_f32(float32x4_t x, int32_t nb)
+{
+    float32x4_t r = x;
+    nb --;
+    while(nb > 0)
+    {
+        r = vmulq_f32(r , x);
+        nb--;
+    }
+    return(r);
+}
+
+
+__STATIC_INLINE float32x4_t __arm_vec_sqrt_f32_neon(float32x4_t  x)
+{
+    float32x4_t x1 = vmaxq_f32(x, vdupq_n_f32(FLT_MIN));
+    float32x4_t e = vrsqrteq_f32(x1);
+    e = vmulq_f32(vrsqrtsq_f32(vmulq_f32(x1, e), e), e);
+    e = vmulq_f32(vrsqrtsq_f32(vmulq_f32(x1, e), e), e);
+    return vmulq_f32(x, e);
+}
+
+__STATIC_INLINE int16x8_t __arm_vec_sqrt_q15_neon(int16x8_t vec)
+{
+    float32x4_t tempF;
+    int32x4_t tempHI,tempLO;
+
+    tempLO = vmovl_s16(vget_low_s16(vec));
+    tempF = vcvtq_n_f32_s32(tempLO,15);
+    tempF = __arm_vec_sqrt_f32_neon(tempF);
+    tempLO = vcvtq_n_s32_f32(tempF,15);
+
+    tempHI = vmovl_s16(vget_high_s16(vec));
+    tempF = vcvtq_n_f32_s32(tempHI,15);
+    tempF = __arm_vec_sqrt_f32_neon(tempF);
+    tempHI = vcvtq_n_s32_f32(tempF,15);
+
+    return(vcombine_s16(vqmovn_s32(tempLO),vqmovn_s32(tempHI)));
+}
+
+__STATIC_INLINE int32x4_t __arm_vec_sqrt_q31_neon(int32x4_t vec)
+{
+  float32x4_t temp;
+
+  temp = vcvtq_n_f32_s32(vec,31);
+  temp = __arm_vec_sqrt_f32_neon(temp);
+  return(vcvtq_n_s32_f32(temp,31));
+}
+
+#endif /*  (defined(ARM_MATH_NEON) || defined(ARM_MATH_NEON_EXPERIMENTAL)) && !defined(ARM_MATH_AUTOVECTORIZE) */
+
+#ifdef   __cplusplus
+}
+#endif
+
+
+#endif /* _ARM_VEC_MATH_H */
+
+/**
+ *
+ * End of file.
+ */
@@ -0,0 +1,312 @@
+/******************************************************************************
+ * @file     arm_vec_math_f16.h
+ * @brief    Public header file for CMSIS DSP Library
+ * @version  V1.10.0
+ * @date     08 July 2021
+ * Target Processor: Cortex-M and Cortex-A cores
+ ******************************************************************************/
+/*
+ * Copyright (c) 2010-2021 Arm Limited or its affiliates. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#ifndef _ARM_VEC_MATH_F16_H
+#define _ARM_VEC_MATH_F16_H
+
+#include "arm_math_types_f16.h"
+#include "arm_common_tables_f16.h"
+#include "arm_helium_utils.h"
+
+#ifdef   __cplusplus
+extern "C"
+{
+#endif
+
+#if defined(ARM_FLOAT16_SUPPORTED)
+
+
+#if defined(ARM_MATH_MVE_FLOAT16) && !defined(ARM_MATH_AUTOVECTORIZE)
+
+
+static const float16_t __logf_rng_f16=0.693147180f16;
+
+/* fast inverse approximation (3x newton) */
+__STATIC_INLINE f16x8_t vrecip_medprec_f16(
+    f16x8_t x)
+{
+    q15x8_t         m;
+    f16x8_t         b;
+    any16x8_t       xinv;
+    f16x8_t         ax = vabsq(x);
+
+    xinv.f = ax;
+
+    m = 0x03c00 - (xinv.i & 0x07c00);
+    xinv.i = xinv.i + m;
+    xinv.f = 1.41176471f16 - 0.47058824f16 * xinv.f;
+    xinv.i = xinv.i + m;
+
+    b = 2.0f16 - xinv.f * ax;
+    xinv.f = xinv.f * b;
+
+    b = 2.0f16 - xinv.f * ax;
+    xinv.f = xinv.f * b;
+
+    b = 2.0f16 - xinv.f * ax;
+    xinv.f = xinv.f * b;
+
+    xinv.f = vdupq_m_n_f16(xinv.f, F16INFINITY, vcmpeqq_n_f16(x, 0.0f));
+    /*
+     * restore sign
+     */
+    xinv.f = vnegq_m(xinv.f, xinv.f, vcmpltq_n_f16(x, 0.0f));
+
+    return xinv.f;
+}
+
+/* fast inverse approximation (4x newton) */
+__STATIC_INLINE f16x8_t vrecip_hiprec_f16(
+    f16x8_t x)
+{
+    q15x8_t         m;
+    f16x8_t         b;
+    any16x8_t       xinv;
+    f16x8_t         ax = vabsq(x);
+
+    xinv.f = ax;
+
+    m = 0x03c00 - (xinv.i & 0x07c00);
+    xinv.i = xinv.i + m;
+    xinv.f = 1.41176471f16 - 0.47058824f16 * xinv.f;
+    xinv.i = xinv.i + m;
+
+    b = 2.0f16 - xinv.f * ax;
+    xinv.f = xinv.f * b;
+
+    b = 2.0f16 - xinv.f * ax;
+    xinv.f = xinv.f * b;
+
+    b = 2.0f16 - xinv.f * ax;
+    xinv.f = xinv.f * b;
+
+    b = 2.0f16 - xinv.f * ax;
+    xinv.f = xinv.f * b;
+
+    xinv.f = vdupq_m_n_f16(xinv.f, F16INFINITY, vcmpeqq_n_f16(x, 0.0f));
+    /*
+     * restore sign
+     */
+    xinv.f = vnegq_m(xinv.f, xinv.f, vcmpltq_n_f16(x, 0.0f));
+
+    return xinv.f;
+}
+
+__STATIC_INLINE f16x8_t vdiv_f16(
+    f16x8_t num, f16x8_t den)
+{
+    return vmulq(num, vrecip_hiprec_f16(den));
+}
+
+
+/**
+  @brief         Single-precision taylor dev.
+  @param[in]     x              f16  vector input
+  @param[in]     coeffs         f16  vector coeffs
+  @return        destination    f16  vector
+ */
+
+__STATIC_INLINE float16x8_t vtaylor_polyq_f16(
+        float16x8_t           x,
+        const float16_t * coeffs)
+{
+    float16x8_t         A = vfmasq(vdupq_n_f16(coeffs[4]), x, coeffs[0]);
+    float16x8_t         B = vfmasq(vdupq_n_f16(coeffs[6]), x, coeffs[2]);
+    float16x8_t         C = vfmasq(vdupq_n_f16(coeffs[5]), x, coeffs[1]);
+    float16x8_t         D = vfmasq(vdupq_n_f16(coeffs[7]), x, coeffs[3]);
+    float16x8_t         x2 = vmulq(x, x);
+    float16x8_t         x4 = vmulq(x2, x2);
+    float16x8_t         res = vfmaq(vfmaq_f16(A, B, x2), vfmaq_f16(C, D, x2), x4);
+
+    return res;
+}
+
+#define VMANT_EXP_F16(x)  \
+    any16x8_t       r;    \
+    int16x8_t       n;    \
+                          \
+    r.f = x;              \
+    n = r.i >> 10;        \
+    n = n - 15;           \
+    r.i = r.i - (n << 10);\
+                          \
+    vecExpUnBiased = n;   \
+    vecTmpFlt1 = r.f;
+
+__STATIC_INLINE float16x8_t vlogq_f16(float16x8_t vecIn)
+{
+    q15x8_t             vecExpUnBiased;
+    float16x8_t         vecTmpFlt0, vecTmpFlt1;
+    float16x8_t         vecAcc0, vecAcc1, vecAcc2, vecAcc3;
+    float16x8_t         vecExpUnBiasedFlt;
+
+    /*
+     * extract exponent
+     */
+    VMANT_EXP_F16(vecIn);
+
+    vecTmpFlt0 = vecTmpFlt1 * vecTmpFlt1;
+    /*
+     * a = (__logf_lut_f16[4] * r.f) + (__logf_lut_f16[0]);
+     */
+    vecAcc0 = vdupq_n_f16(__logf_lut_f16[0]);
+    vecAcc0 = vfmaq(vecAcc0, vecTmpFlt1, __logf_lut_f16[4]);
+    /*
+     * b = (__logf_lut_f16[6] * r.f) + (__logf_lut_f16[2]);
+     */
+    vecAcc1 = vdupq_n_f16(__logf_lut_f16[2]);
+    vecAcc1 = vfmaq(vecAcc1, vecTmpFlt1, __logf_lut_f16[6]);
+    /*
+     * c = (__logf_lut_f16[5] * r.f) + (__logf_lut_f16[1]);
+     */
+    vecAcc2 = vdupq_n_f16(__logf_lut_f16[1]);
+    vecAcc2 = vfmaq(vecAcc2, vecTmpFlt1, __logf_lut_f16[5]);
+    /*
+     * d = (__logf_lut_f16[7] * r.f) + (__logf_lut_f16[3]);
+     */
+    vecAcc3 = vdupq_n_f16(__logf_lut_f16[3]);
+    vecAcc3 = vfmaq(vecAcc3, vecTmpFlt1, __logf_lut_f16[7]);
+    /*
+     * a = a + b * xx;
+     */
+    vecAcc0 = vfmaq(vecAcc0, vecAcc1, vecTmpFlt0);
+    /*
+     * c = c + d * xx;
+     */
+    vecAcc2 = vfmaq(vecAcc2, vecAcc3, vecTmpFlt0);
+    /*
+     * xx = xx * xx;
+     */
+    vecTmpFlt0 = vecTmpFlt0 * vecTmpFlt0;
+    vecExpUnBiasedFlt = vcvtq_f16_s16(vecExpUnBiased);
+    /*
+     * r.f = a + c * xx;
+     */
+    vecAcc0 = vfmaq(vecAcc0, vecAcc2, vecTmpFlt0);
+    /*
+     * add exponent
+     * r.f = r.f + ((float32_t) m) * __logf_rng_f16;
+     */
+    vecAcc0 = vfmaq(vecAcc0, vecExpUnBiasedFlt, __logf_rng_f16);
+    // set log0 down to -inf
+    vecAcc0 = vdupq_m_n_f16(vecAcc0, -(_Float16)F16INFINITY, vcmpeqq_n_f16(vecIn, 0.0f));
+    return vecAcc0;
+}
+
+__STATIC_INLINE float16x8_t vexpq_f16(
+    float16x8_t x)
+{
+    // Perform range reduction [-log(2),log(2)]
+    int16x8_t       m = vcvtq_s16_f16(vmulq_n_f16(x, 1.4426950408f16));
+    float16x8_t     val = vfmsq_f16(x, vcvtq_f16_s16(m), vdupq_n_f16(0.6931471805f16));
+
+    // Polynomial Approximation
+    float16x8_t         poly = vtaylor_polyq_f16(val, exp_tab_f16);
+
+    // Reconstruct
+    poly = (float16x8_t) (vqaddq_s16((int16x8_t) (poly), vqshlq_n_s16(m, 10)));
+
+    poly = vdupq_m_n_f16(poly, 0.0f16, vcmpltq_n_s16(m, -14));
+    return poly;
+}
+
+__STATIC_INLINE float16x8_t arm_vec_exponent_f16(float16x8_t x, int16_t nb)
+{
+    float16x8_t         r = x;
+    nb--;
+    while (nb > 0) {
+        r = vmulq(r, x);
+        nb--;
+    }
+    return (r);
+}
+
+__STATIC_INLINE f16x8_t vpowq_f16(
+    f16x8_t val,
+    f16x8_t n)
+{
+    return vexpq_f16(vmulq_f16(n, vlogq_f16(val)));
+}
+
+#define INV_NEWTON_INIT_F16  0x7773
+
+__STATIC_INLINE f16x8_t vrecip_f16(f16x8_t vecIn)
+{
+    f16x8_t     vecSx, vecW, vecTmp;
+    any16x8_t   v;
+
+    vecSx = vabsq(vecIn);
+
+    v.f = vecIn;
+    v.i = vsubq(vdupq_n_s16(INV_NEWTON_INIT_F16), v.i);
+
+    vecW = vmulq(vecSx, v.f);
+
+    // v.f = v.f * (8 + w * (-28 + w * (56 + w * (-70 + w *(56 + w * (-28 + w * (8 - w)))))));
+    vecTmp = vsubq(vdupq_n_f16(8.0f16), vecW);
+    vecTmp = vfmasq_n_f16(vecW, vecTmp, -28.0f16);
+    vecTmp = vfmasq_n_f16(vecW, vecTmp, 56.0f16);
+    vecTmp = vfmasq_n_f16(vecW, vecTmp, -70.0f16);
+    vecTmp = vfmasq_n_f16(vecW, vecTmp, 56.0f16);
+    vecTmp = vfmasq_n_f16(vecW, vecTmp, -28.0f16);
+    vecTmp = vfmasq_n_f16(vecW, vecTmp, 8.0f16);
+    v.f = vmulq(v.f,  vecTmp);
+
+    v.f = vdupq_m_n_f16(v.f, F16INFINITY, vcmpeqq_n_f16(vecIn, 0.0f));
+    /*
+     * restore sign
+     */
+    v.f = vnegq_m(v.f, v.f, vcmpltq_n_f16(vecIn, 0.0f));
+    return v.f;
+}
+
+__STATIC_INLINE f16x8_t vtanhq_f16(
+    f16x8_t val)
+{
+    f16x8_t         x =
+        vminnmq_f16(vmaxnmq_f16(val, vdupq_n_f16(-10.f16)), vdupq_n_f16(10.0f16));
+    f16x8_t         exp2x = vexpq_f16(vmulq_n_f16(x, 2.f16));
+    f16x8_t         num = vsubq_n_f16(exp2x, 1.f16);
+    f16x8_t         den = vaddq_n_f16(exp2x, 1.f16);
+    f16x8_t         tanh = vmulq_f16(num, vrecip_f16(den));
+    return tanh;
+}
+
+#endif /* defined(ARM_MATH_MVE_FLOAT16)  && !defined(ARM_MATH_AUTOVECTORIZE)*/
+
+
+
+#ifdef   __cplusplus
+}
+#endif
+
+#endif /* ARM FLOAT16 SUPPORTED */
+
+#endif /* _ARM_VEC_MATH_F16_H */
+
+/**
+ *
+ * End of file.
+ */
@@ -0,0 +1,283 @@
+/**************************************************************************//**
+ * @file     cmsis_compiler.h
+ * @brief    CMSIS compiler generic header file
+ * @version  V5.1.0
+ * @date     09. October 2018
+ ******************************************************************************/
+/*
+ * Copyright (c) 2009-2018 Arm Limited. All rights reserved.
+ *
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the License); you may
+ * not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an AS IS BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#ifndef __CMSIS_COMPILER_H
+#define __CMSIS_COMPILER_H
+
+#include <stdint.h>
+
+/*
+ * Arm Compiler 4/5
+ */
+#if   defined ( __CC_ARM )
+  #include "cmsis_armcc.h"
+
+
+/*
+ * Arm Compiler 6.6 LTM (armclang)
+ */
+#elif defined (__ARMCC_VERSION) && (__ARMCC_VERSION >= 6010050) && (__ARMCC_VERSION < 6100100)
+  #include "cmsis_armclang_ltm.h"
+
+  /*
+ * Arm Compiler above 6.10.1 (armclang)
+ */
+#elif defined (__ARMCC_VERSION) && (__ARMCC_VERSION >= 6100100)
+  #include "cmsis_armclang.h"
+
+
+/*
+ * GNU Compiler
+ */
+#elif defined ( __GNUC__ )
+  #include "cmsis_gcc.h"
+
+
+/*
+ * IAR Compiler
+ */
+#elif defined ( __ICCARM__ )
+  #include <cmsis_iccarm.h>
+
+
+/*
+ * TI Arm Compiler
+ */
+#elif defined ( __TI_ARM__ )
+  #include <cmsis_ccs.h>
+
+  #ifndef   __ASM
+    #define __ASM                                  __asm
+  #endif
+  #ifndef   __INLINE
+    #define __INLINE                               inline
+  #endif
+  #ifndef   __STATIC_INLINE
+    #define __STATIC_INLINE                        static inline
+  #endif
+  #ifndef   __STATIC_FORCEINLINE
+    #define __STATIC_FORCEINLINE                   __STATIC_INLINE
+  #endif
+  #ifndef   __NO_RETURN
+    #define __NO_RETURN                            __attribute__((noreturn))
+  #endif
+  #ifndef   __USED
+    #define __USED                                 __attribute__((used))
+  #endif
+  #ifndef   __WEAK
+    #define __WEAK                                 __attribute__((weak))
+  #endif
+  #ifndef   __PACKED
+    #define __PACKED                               __attribute__((packed))
+  #endif
+  #ifndef   __PACKED_STRUCT
+    #define __PACKED_STRUCT                        struct __attribute__((packed))
+  #endif
+  #ifndef   __PACKED_UNION
+    #define __PACKED_UNION                         union __attribute__((packed))
+  #endif
+  #ifndef   __UNALIGNED_UINT32        /* deprecated */
+    struct __attribute__((packed)) T_UINT32 { uint32_t v; };
+    #define __UNALIGNED_UINT32(x)                  (((struct T_UINT32 *)(x))->v)
+  #endif
+  #ifndef   __UNALIGNED_UINT16_WRITE
+    __PACKED_STRUCT T_UINT16_WRITE { uint16_t v; };
+    #define __UNALIGNED_UINT16_WRITE(addr, val)    (void)((((struct T_UINT16_WRITE *)(void*)(addr))->v) = (val))
+  #endif
+  #ifndef   __UNALIGNED_UINT16_READ
+    __PACKED_STRUCT T_UINT16_READ { uint16_t v; };
+    #define __UNALIGNED_UINT16_READ(addr)          (((const struct T_UINT16_READ *)(const void *)(addr))->v)
+  #endif
+  #ifndef   __UNALIGNED_UINT32_WRITE
+    __PACKED_STRUCT T_UINT32_WRITE { uint32_t v; };
+    #define __UNALIGNED_UINT32_WRITE(addr, val)    (void)((((struct T_UINT32_WRITE *)(void *)(addr))->v) = (val))
+  #endif
+  #ifndef   __UNALIGNED_UINT32_READ
+    __PACKED_STRUCT T_UINT32_READ { uint32_t v; };
+    #define __UNALIGNED_UINT32_READ(addr)          (((const struct T_UINT32_READ *)(const void *)(addr))->v)
+  #endif
+  #ifndef   __ALIGNED
+    #define __ALIGNED(x)                           __attribute__((aligned(x)))
+  #endif
+  #ifndef   __RESTRICT
+    #define __RESTRICT                             __restrict
+  #endif
+  #ifndef   __COMPILER_BARRIER
+    #warning No compiler specific solution for __COMPILER_BARRIER. __COMPILER_BARRIER is ignored.
+    #define __COMPILER_BARRIER()                   (void)0
+  #endif
+
+
+/*
+ * TASKING Compiler
+ */
+#elif defined ( __TASKING__ )
+  /*
+   * The CMSIS functions have been implemented as intrinsics in the compiler.
+   * Please use "carm -?i" to get an up to date list of all intrinsics,
+   * Including the CMSIS ones.
+   */
+
+  #ifndef   __ASM
+    #define __ASM                                  __asm
+  #endif
+  #ifndef   __INLINE
+    #define __INLINE                               inline
+  #endif
+  #ifndef   __STATIC_INLINE
+    #define __STATIC_INLINE                        static inline
+  #endif
+  #ifndef   __STATIC_FORCEINLINE
+    #define __STATIC_FORCEINLINE                   __STATIC_INLINE
+  #endif
+  #ifndef   __NO_RETURN
+    #define __NO_RETURN                            __attribute__((noreturn))
+  #endif
+  #ifndef   __USED
+    #define __USED                                 __attribute__((used))
+  #endif
+  #ifndef   __WEAK
+    #define __WEAK                                 __attribute__((weak))
+  #endif
+  #ifndef   __PACKED
+    #define __PACKED                               __packed__
+  #endif
+  #ifndef   __PACKED_STRUCT
+    #define __PACKED_STRUCT                        struct __packed__
+  #endif
+  #ifndef   __PACKED_UNION
+    #define __PACKED_UNION                         union __packed__
+  #endif
+  #ifndef   __UNALIGNED_UINT32        /* deprecated */
+    struct __packed__ T_UINT32 { uint32_t v; };
+    #define __UNALIGNED_UINT32(x)                  (((struct T_UINT32 *)(x))->v)
+  #endif
+  #ifndef   __UNALIGNED_UINT16_WRITE
+    __PACKED_STRUCT T_UINT16_WRITE { uint16_t v; };
+    #define __UNALIGNED_UINT16_WRITE(addr, val)    (void)((((struct T_UINT16_WRITE *)(void *)(addr))->v) = (val))
+  #endif
+  #ifndef   __UNALIGNED_UINT16_READ
+    __PACKED_STRUCT T_UINT16_READ { uint16_t v; };
+    #define __UNALIGNED_UINT16_READ(addr)          (((const struct T_UINT16_READ *)(const void *)(addr))->v)
+  #endif
+  #ifndef   __UNALIGNED_UINT32_WRITE
+    __PACKED_STRUCT T_UINT32_WRITE { uint32_t v; };
+    #define __UNALIGNED_UINT32_WRITE(addr, val)    (void)((((struct T_UINT32_WRITE *)(void *)(addr))->v) = (val))
+  #endif
+  #ifndef   __UNALIGNED_UINT32_READ
+    __PACKED_STRUCT T_UINT32_READ { uint32_t v; };
+    #define __UNALIGNED_UINT32_READ(addr)          (((const struct T_UINT32_READ *)(const void *)(addr))->v)
+  #endif
+  #ifndef   __ALIGNED
+    #define __ALIGNED(x)              __align(x)
+  #endif
+  #ifndef   __RESTRICT
+    #warning No compiler specific solution for __RESTRICT. __RESTRICT is ignored.
+    #define __RESTRICT
+  #endif
+  #ifndef   __COMPILER_BARRIER
+    #warning No compiler specific solution for __COMPILER_BARRIER. __COMPILER_BARRIER is ignored.
+    #define __COMPILER_BARRIER()                   (void)0
+  #endif
+
+
+/*
+ * COSMIC Compiler
+ */
+#elif defined ( __CSMC__ )
+   #include <cmsis_csm.h>
+
+ #ifndef   __ASM
+    #define __ASM                                  _asm
+  #endif
+  #ifndef   __INLINE
+    #define __INLINE                               inline
+  #endif
+  #ifndef   __STATIC_INLINE
+    #define __STATIC_INLINE                        static inline
+  #endif
+  #ifndef   __STATIC_FORCEINLINE
+    #define __STATIC_FORCEINLINE                   __STATIC_INLINE
+  #endif
+  #ifndef   __NO_RETURN
+    // NO RETURN is automatically detected hence no warning here
+    #define __NO_RETURN
+  #endif
+  #ifndef   __USED
+    #warning No compiler specific solution for __USED. __USED is ignored.
+    #define __USED
+  #endif
+  #ifndef   __WEAK
+    #define __WEAK                                 __weak
+  #endif
+  #ifndef   __PACKED
+    #define __PACKED                               @packed
+  #endif
+  #ifndef   __PACKED_STRUCT
+    #define __PACKED_STRUCT                        @packed struct
+  #endif
+  #ifndef   __PACKED_UNION
+    #define __PACKED_UNION                         @packed union
+  #endif
+  #ifndef   __UNALIGNED_UINT32        /* deprecated */
+    @packed struct T_UINT32 { uint32_t v; };
+    #define __UNALIGNED_UINT32(x)                  (((struct T_UINT32 *)(x))->v)
+  #endif
+  #ifndef   __UNALIGNED_UINT16_WRITE
+    __PACKED_STRUCT T_UINT16_WRITE { uint16_t v; };
+    #define __UNALIGNED_UINT16_WRITE(addr, val)    (void)((((struct T_UINT16_WRITE *)(void *)(addr))->v) = (val))
+  #endif
+  #ifndef   __UNALIGNED_UINT16_READ
+    __PACKED_STRUCT T_UINT16_READ { uint16_t v; };
+    #define __UNALIGNED_UINT16_READ(addr)          (((const struct T_UINT16_READ *)(const void *)(addr))->v)
+  #endif
+  #ifndef   __UNALIGNED_UINT32_WRITE
+    __PACKED_STRUCT T_UINT32_WRITE { uint32_t v; };
+    #define __UNALIGNED_UINT32_WRITE(addr, val)    (void)((((struct T_UINT32_WRITE *)(void *)(addr))->v) = (val))
+  #endif
+  #ifndef   __UNALIGNED_UINT32_READ
+    __PACKED_STRUCT T_UINT32_READ { uint32_t v; };
+    #define __UNALIGNED_UINT32_READ(addr)          (((const struct T_UINT32_READ *)(const void *)(addr))->v)
+  #endif
+  #ifndef   __ALIGNED
+    #warning No compiler specific solution for __ALIGNED. __ALIGNED is ignored.
+    #define __ALIGNED(x)
+  #endif
+  #ifndef   __RESTRICT
+    #warning No compiler specific solution for __RESTRICT. __RESTRICT is ignored.
+    #define __RESTRICT
+  #endif
+  #ifndef   __COMPILER_BARRIER
+    #warning No compiler specific solution for __COMPILER_BARRIER. __COMPILER_BARRIER is ignored.
+    #define __COMPILER_BARRIER()                   (void)0
+  #endif
+
+
+#else
+  #error Unknown compiler.
+#endif
+
+
+#endif /* __CMSIS_COMPILER_H */
+
--- a/Show More
+++ b/Show More