diff --git a/components/bootloader_support/src/secure_boot_signatures.c b/components/bootloader_support/src/secure_boot_signatures.c
index f7e435e86..076dfb335 100644
--- a/components/bootloader_support/src/secure_boot_signatures.c
+++ b/components/bootloader_support/src/secure_boot_signatures.c
@@ -25,7 +25,6 @@
 typedef SHA_CTX sha_context;
 #else
 #include "hwcrypto/sha.h"
-typedef esp_sha_context sha_context;
 #endif
 
 typedef struct {
@@ -42,7 +41,9 @@ extern const uint8_t signature_verification_key_end[] asm("_binary_signature_ver
 
 esp_err_t esp_secure_boot_verify_signature(uint32_t src_addr, uint32_t length)
 {
-    sha_context sha;
+#ifdef BOOTLOADER_BUILD
+    SHA_CTX sha;
+#endif
     uint8_t digest[32];
     ptrdiff_t keylen;
     const uint8_t *data;
@@ -83,12 +84,8 @@ esp_err_t esp_secure_boot_verify_signature(uint32_t src_addr, uint32_t length)
     ets_sha_finish(&sha, SHA2_256, digest);
     ets_sha_disable();
 #else
-    /* Use thread-safe esp-idf SHA layer */
-    esp_sha256_init(&sha);
-    esp_sha256_start(&sha, false);
-    esp_sha256_update(&sha, data, length);
-    esp_sha256_finish(&sha, digest);
-    esp_sha256_free(&sha);
+    /* Use thread-safe esp-idf SHA function */
+    esp_sha(SHA2_256, data, length, digest);
 #endif
 
     keylen = signature_verification_key_end - signature_verification_key_start;
diff --git a/components/esp32/hwcrypto/aes.c b/components/esp32/hwcrypto/aes.c
index 169465822..8f27cd17c 100644
--- a/components/esp32/hwcrypto/aes.c
+++ b/components/esp32/hwcrypto/aes.c
@@ -28,6 +28,7 @@
 #include <string.h>
 #include "hwcrypto/aes.h"
 #include "rom/aes.h"
+#include "soc/dport_reg.h"
 #include <sys/lock.h>
 
 static _lock_t aes_lock;
@@ -36,14 +37,23 @@ void esp_aes_acquire_hardware( void )
 {
     /* newlib locks lazy initialize on ESP-IDF */
     _lock_acquire(&aes_lock);
-    ets_aes_enable();
+    /* Enable AES hardware */
+    REG_SET_BIT(DPORT_PERI_CLK_EN_REG, DPORT_PERI_EN_AES);
+    /* Clear reset on digital signature & secure boot units,
+       otherwise AES unit is held in reset also. */
+    REG_CLR_BIT(DPORT_PERI_RST_EN_REG,
+                DPORT_PERI_EN_AES
+                | DPORT_PERI_EN_DIGITAL_SIGNATURE
+                | DPORT_PERI_EN_SECUREBOOT);
 }
 
 void esp_aes_release_hardware( void )
 {
-    uint8_t zero[256/8] = { 0 };
-    ets_aes_setkey_enc(zero, AES256);
-    ets_aes_disable();
+    /* Disable AES hardware */
+    REG_SET_BIT(DPORT_PERI_RST_EN_REG, DPORT_PERI_EN_AES);
+    /* Don't return other units to reset, as this pulls
+       reset on RSA & SHA units, respectively. */
+    REG_CLR_BIT(DPORT_PERI_CLK_EN_REG, DPORT_PERI_EN_AES);
     _lock_release(&aes_lock);
 }
 
diff --git a/components/esp32/hwcrypto/sha.c b/components/esp32/hwcrypto/sha.c
index 06b00c54a..601981e2a 100644
--- a/components/esp32/hwcrypto/sha.c
+++ b/components/esp32/hwcrypto/sha.c
@@ -26,242 +26,264 @@
  */
 
 #include <string.h>
+#include <stdio.h>
 #include <sys/lock.h>
+#include <byteswap.h>
+#include <assert.h>
+
 #include "hwcrypto/sha.h"
 #include "rom/ets_sys.h"
+#include "soc/dport_reg.h"
+#include "soc/hwcrypto_reg.h"
 
-
-static _lock_t sha_lock;
-
-void esp_sha_acquire_hardware( void )
-{
-    /* newlib locks lazy initialize on ESP-IDF */
-    _lock_acquire(&sha_lock);
-    ets_sha_enable();
+inline static uint32_t SHA_LOAD_REG(esp_sha_type sha_type) {
+    return SHA_1_LOAD_REG + sha_type * 0x10;
 }
 
-void esp_sha_release_hardware( void )
-{
-    /* Want to empty internal SHA buffers where possible,
-       need to check if this is sufficient for this. */
-    SHA_CTX zero = { 0 };
-    ets_sha_init(&zero);
-    ets_sha_disable();
-    _lock_release(&sha_lock);
+inline static uint32_t SHA_BUSY_REG(esp_sha_type sha_type) {
+    return SHA_1_BUSY_REG + sha_type * 0x10;
 }
 
-/* Generic esp_shaX_update implementation */
-static void esp_sha_update( esp_sha_context *ctx, const unsigned char *input, size_t ilen, size_t block_size)
+inline static uint32_t SHA_START_REG(esp_sha_type sha_type) {
+    return SHA_1_START_REG + sha_type * 0x10;
+}
+
+inline static uint32_t SHA_CONTINUE_REG(esp_sha_type sha_type) {
+    return SHA_1_CONTINUE_REG + sha_type * 0x10;
+}
+
+/* Single lock for SHA engine memory block
+*/
+static _lock_t memory_block_lock;
+
+typedef struct {
+    _lock_t lock;
+    bool in_use;
+} sha_engine_state;
+
+/* Pointer to state of each concurrent SHA engine.
+
+   Indexes:
+   0 = SHA1
+   1 = SHA2_256
+   2 = SHA2_384 or SHA2_512
+*/
+static sha_engine_state engine_states[3];
+
+/* Index into the sha_engine_state array */
+inline static size_t sha_engine_index(esp_sha_type type) {
+    switch(type) {
+    case SHA1:
+        return 0;
+    case SHA2_256:
+        return 1;
+    default:
+        return 2;
+    }
+}
+
+/* Return state & digest length (in bytes) for a given SHA type */
+inline static size_t sha_length(esp_sha_type type) {
+    switch(type) {
+    case SHA1:
+        return 20;
+    case SHA2_256:
+        return 32;
+    case SHA2_384:
+        return 64;
+    case SHA2_512:
+        return 64;
+    default:
+        return 0;
+    }
+}
+
+/* Return block size (in bytes) for a given SHA type */
+inline static size_t block_length(esp_sha_type type) {
+    switch(type) {
+    case SHA1:
+    case SHA2_256:
+        return 64;
+    case SHA2_384:
+    case SHA2_512:
+        return 128;
+    default:
+        return 0;
+    }
+}
+
+void esp_sha_lock_memory_block(void)
 {
-    /* Feed the SHA engine one block at a time */
+    _lock_acquire(&memory_block_lock);
+}
+
+void esp_sha_unlock_memory_block(void)
+{
+    _lock_release(&memory_block_lock);
+}
+
+/* Lock to hold when changing SHA engine state,
+   allows checking of sha_engines_all_idle()
+*/
+static _lock_t state_change_lock;
+
+inline static bool sha_engines_all_idle() {
+    return !engine_states[0].in_use
+        && !engine_states[1].in_use
+        && !engine_states[2].in_use;
+}
+
+static void esp_sha_lock_engine_inner(sha_engine_state *engine);
+
+bool esp_sha_try_lock_engine(esp_sha_type sha_type)
+{
+    sha_engine_state *engine = &engine_states[sha_engine_index(sha_type)];
+    if(_lock_try_acquire(&engine->lock) != 0) {
+        /* This SHA engine is already in use */
+        return false;
+    } else {
+        esp_sha_lock_engine_inner(engine);
+        return true;
+    }
+}
+
+void esp_sha_lock_engine(esp_sha_type sha_type)
+{
+    sha_engine_state *engine = &engine_states[sha_engine_index(sha_type)];
+    _lock_acquire(&engine->lock);
+    esp_sha_lock_engine_inner(engine);
+}
+
+static void esp_sha_lock_engine_inner(sha_engine_state *engine)
+{
+    _lock_acquire(&state_change_lock);
+
+    if (sha_engines_all_idle()) {
+        /* Enable SHA hardware */
+        REG_SET_BIT(DPORT_PERI_CLK_EN_REG, DPORT_PERI_EN_SHA);
+        /* also clear reset on secure boot, otherwise SHA is held in reset */
+        REG_CLR_BIT(DPORT_PERI_RST_EN_REG,
+                    DPORT_PERI_EN_SHA
+                    | DPORT_PERI_EN_SECUREBOOT);
+        ets_sha_enable();
+    }
+
+    _lock_release(&state_change_lock);
+
+    assert( !engine->in_use && "in_use flag should be cleared" );
+    engine->in_use = true;
+}
+
+
+void esp_sha_unlock_engine(esp_sha_type sha_type)
+{
+    sha_engine_state *engine = &engine_states[sha_engine_index(sha_type)];
+
+    _lock_acquire(&state_change_lock);
+
+    assert( engine->in_use && "in_use flag should be set" );
+    engine->in_use = false;
+
+    if (sha_engines_all_idle()) {
+        /* Disable SHA hardware */
+        /* Don't assert reset on secure boot, otherwise AES is held in reset */
+        REG_SET_BIT(DPORT_PERI_RST_EN_REG, DPORT_PERI_EN_SHA);
+        REG_CLR_BIT(DPORT_PERI_CLK_EN_REG, DPORT_PERI_EN_SHA);
+    }
+
+    _lock_release(&state_change_lock);
+
+    _lock_release(&engine->lock);
+}
+
+void esp_sha_wait_idle(void)
+{
+    while(REG_READ(SHA_1_BUSY_REG) == 1) {}
+    while(REG_READ(SHA_256_BUSY_REG) == 1) {}
+    while(REG_READ(SHA_384_BUSY_REG) == 1) {}
+    while(REG_READ(SHA_512_BUSY_REG) == 1) {}
+}
+
+void esp_sha_read_digest_state(esp_sha_type sha_type, void *digest_state)
+{
+    sha_engine_state *engine = &engine_states[sha_engine_index(sha_type)];
+    assert(engine->in_use && "SHA engine should be locked" );
+
+    esp_sha_lock_memory_block();
+
+    esp_sha_wait_idle();
+
+    REG_WRITE(SHA_LOAD_REG(sha_type), 1);
+    while(REG_READ(SHA_BUSY_REG(sha_type)) == 1) { }
+
+    uint32_t *digest_state_words = (uint32_t *)digest_state;
+    uint32_t *reg_addr_buf = (uint32_t *)(SHA_TEXT_BASE);
+    if(sha_type == SHA2_384 || sha_type == SHA2_512) {
+        /* for these ciphers using 64-bit states, swap each pair of words */
+        for(int i = 0; i < sha_length(sha_type)/4; i += 2) {
+            digest_state_words[i+1] = reg_addr_buf[i];
+            digest_state_words[i]= reg_addr_buf[i+1];
+        }
+    } else {
+        memcpy(digest_state_words, reg_addr_buf, sha_length(sha_type));
+    }
+    asm volatile ("memw");
+
+    esp_sha_unlock_memory_block();
+}
+
+void esp_sha_block(esp_sha_type sha_type, const void *data_block, bool is_first_block)
+{
+    sha_engine_state *engine = &engine_states[sha_engine_index(sha_type)];
+    assert(engine->in_use && "SHA engine should be locked" );
+
+    esp_sha_lock_memory_block();
+
+    esp_sha_wait_idle();
+
+    /* Fill the data block */
+    uint32_t *reg_addr_buf = (uint32_t *)(SHA_TEXT_BASE);
+    uint32_t *data_words = (uint32_t *)data_block;
+    for (int i = 0; i < block_length(sha_type) / 4; i++) {
+        reg_addr_buf[i] = __bswap_32(data_words[i]);
+    }
+    asm volatile ("memw");
+
+    if(is_first_block) {
+        REG_WRITE(SHA_START_REG(sha_type), 1);
+    } else {
+        REG_WRITE(SHA_CONTINUE_REG(sha_type), 1);
+    }
+
+    esp_sha_unlock_memory_block();
+
+    /* Note: deliberately not waiting for this operation to complete,
+       as a performance tweak - delay waiting until the next time we need the SHA
+       unit, instead.
+    */
+}
+
+void esp_sha(esp_sha_type sha_type, const unsigned char *input, size_t ilen, unsigned char *output)
+{
+    size_t block_len = block_length(sha_type);
+
+    esp_sha_lock_engine(sha_type);
+
+    SHA_CTX ctx;
+    ets_sha_init(&ctx);
     while(ilen > 0) {
-        size_t chunk_len = (ilen > block_size) ? block_size : ilen;
-        ets_sha_update(&ctx->context, ctx->context_type, input, chunk_len * 8);
+        size_t chunk_len = (ilen > block_len) ? block_len : ilen;
+        esp_sha_lock_memory_block();
+        esp_sha_wait_idle();
+        ets_sha_update(&ctx, sha_type, input, chunk_len * 8);
+        esp_sha_unlock_memory_block();
         input += chunk_len;
         ilen -= chunk_len;
     }
+    esp_sha_lock_memory_block();
+    esp_sha_wait_idle();
+    ets_sha_finish(&ctx, sha_type, output);
+    esp_sha_unlock_memory_block();
+
+    esp_sha_unlock_engine(sha_type);
 }
-
-void esp_sha1_init( esp_sha_context *ctx )
-{
-    bzero( ctx, sizeof( esp_sha_context ) );
-}
-
-void esp_sha1_free( esp_sha_context *ctx )
-{
-    if ( ctx == NULL ) {
-        return;
-    }
-
-    bzero( ctx, sizeof( esp_sha_context ) );
-}
-
-void esp_sha1_clone( esp_sha_context *dst, const esp_sha_context *src )
-{
-    *dst = *src;
-}
-
-/*
- * SHA-1 context setup
- */
-void esp_sha1_start( esp_sha_context *ctx )
-{
-    ctx->context_type = SHA1;
-    esp_sha_acquire_hardware();
-    ets_sha_init(&ctx->context);
-}
-
-/*
- * SHA-1 process buffer
- */
-void esp_sha1_update( esp_sha_context *ctx, const unsigned char *input, size_t ilen )
-{
-    esp_sha_update(ctx, input, ilen, 64);
-}
-
-/*
- * SHA-1 final digest
- */
-void esp_sha1_finish( esp_sha_context *ctx, unsigned char output[20] )
-{
-    ets_sha_finish(&ctx->context, ctx->context_type, output);
-    esp_sha_release_hardware();
-}
-
-/* Full SHA-1 calculation */
-void esp_sha1( const unsigned char *input, size_t ilen, unsigned char output[20] )
-{
-    esp_sha_context ctx;
-
-    esp_sha1_init( &ctx );
-    esp_sha1_start( &ctx );
-    esp_sha1_update( &ctx, input, ilen );
-    esp_sha1_finish( &ctx, output );
-    esp_sha1_free( &ctx );
-}
-
-void esp_sha256_init( esp_sha_context *ctx )
-{
-    bzero( ctx, sizeof( esp_sha_context ) );
-}
-
-void esp_sha256_free( esp_sha_context *ctx )
-{
-    if ( ctx == NULL ) {
-        return;
-    }
-
-    bzero( ctx, sizeof( esp_sha_context ) );
-}
-
-void esp_sha256_clone( esp_sha_context *dst, const esp_sha_context *src )
-{
-    *dst = *src;
-}
-
-/*
- * SHA-256 context setup
- */
-void esp_sha256_start( esp_sha_context *ctx, int is224 )
-{
-    if ( is224 == 0 ) {
-        /* SHA-256 */
-        ctx->context_type = SHA2_256;
-        esp_sha_acquire_hardware();
-        ets_sha_init(&ctx->context);
-    } else {
-        /* SHA-224 is not supported! */
-        ctx->context_type = SHA_INVALID;
-    }
-}
-
-/*
- * SHA-256 process buffer
- */
-void esp_sha256_update( esp_sha_context *ctx, const unsigned char *input, size_t ilen )
-{
-    if( ctx->context_type == SHA2_256 ) {
-        esp_sha_update(ctx, input, ilen, 64);
-    }
-    /* SHA-224 is a no-op */
-}
-
-/*
- * SHA-256 final digest
- */
-void esp_sha256_finish( esp_sha_context *ctx, unsigned char output[32] )
-{
-    if ( ctx->context_type == SHA2_256 ) {
-        ets_sha_finish(&ctx->context, ctx->context_type, output);
-        esp_sha_release_hardware();
-    } else {
-        /* No hardware SHA-224 support, but mbedTLS API doesn't allow failure.
-           For now, zero the output to make it clear it's not valid. */
-        bzero( output, 28 );
-    }
-}
-
-/*
- * Full SHA-256 calculation
- */
-void esp_sha256( const unsigned char *input, size_t ilen, unsigned char output[32], int is224 )
-{
-    esp_sha_context ctx;
-
-    esp_sha256_init( &ctx );
-    esp_sha256_start( &ctx, is224 );
-    esp_sha256_update( &ctx, input, ilen );
-    esp_sha256_finish( &ctx, output );
-    esp_sha256_free( &ctx );
-}
-
-
-/////
-void esp_sha512_init( esp_sha_context *ctx )
-{
-    memset( ctx, 0, sizeof( esp_sha_context ) );
-}
-
-void esp_sha512_free( esp_sha_context *ctx )
-{
-    if ( ctx == NULL ) {
-        return;
-    }
-
-    bzero( ctx, sizeof( esp_sha_context ) );
-}
-
-void esp_sha512_clone( esp_sha_context *dst, const esp_sha_context *src )
-{
-    *dst = *src;
-}
-
-/*
- * SHA-512 context setup
- */
-void esp_sha512_start( esp_sha_context *ctx, int is384 )
-{
-    if ( is384 == 0 ) {
-        /* SHA-512 */
-        ctx->context_type = SHA2_512;
-    } else {
-        /* SHA-384 */
-        ctx->context_type = SHA2_384;
-    }
-    esp_sha_acquire_hardware();
-    ets_sha_init(&ctx->context);
-}
-
-/*
- * SHA-512 process buffer
- */
-void esp_sha512_update( esp_sha_context *ctx, const unsigned char *input, size_t ilen )
-{
-    esp_sha_update(ctx, input, ilen, 128);
-}
-
-/*
- * SHA-512 final digest
- */
-void esp_sha512_finish( esp_sha_context *ctx, unsigned char output[64] )
-{
-    ets_sha_finish(&ctx->context, ctx->context_type, output);
-    esp_sha_release_hardware();
-}
-
-/*
- * Full SHA-512 calculation
- */
-void esp_sha512( const unsigned char *input, size_t ilen, unsigned char output[64], int is384 )
-{
-    esp_sha_context ctx;
-
-    esp_sha512_init( &ctx );
-    esp_sha512_start( &ctx, is384 );
-    esp_sha512_update( &ctx, input, ilen );
-    esp_sha512_finish( &ctx, output );
-    esp_sha512_free( &ctx );
-}
-
-////
-
diff --git a/components/esp32/include/hwcrypto/sha.h b/components/esp32/include/hwcrypto/sha.h
index a165c46c1..2a0ec78ab 100644
--- a/components/esp32/include/hwcrypto/sha.h
+++ b/components/esp32/include/hwcrypto/sha.h
@@ -1,246 +1,200 @@
-/*
- *  ESP32 hardware accelerated SHA1/256/512 implementation
- *  based on mbedTLS FIPS-197 compliant version.
- *
- *  Copyright (C) 2006-2015, ARM Limited, All Rights Reserved
- *  Additions Copyright (C) 2016, Espressif Systems (Shanghai) PTE Ltd
- *  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
- *
- *  http://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.
- *
- */
+// Copyright 2015-2016 Espressif Systems (Shanghai) PTE LTD
+//
+// 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
 
+//     http://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 _ESP_SHA_H_
 #define _ESP_SHA_H_
 
 #include "rom/sha.h"
-
 #include "esp_types.h"
 
+/** @brief Low-level support functions for the hardware SHA engine
+ *
+ * @note If you're looking for a SHA API to use, try mbedtls component
+ * mbedtls/shaXX.h. That API supports hardware acceleration.
+ *
+ * The API in this header provides some building blocks for implementing a
+ * full SHA API such as the one in mbedtls, and also a basic SHA function esp_sha().
+ *
+ * Some technical details about the hardware SHA engine:
+ *
+ * - SHA accelerator engine calculates one digest at a time, per SHA
+ *   algorithm type. It initialises and maintains the digest state
+ *   internally. It is possible to read out an in-progress SHA digest
+ *   state, but it is not possible to restore a SHA digest state
+ *   into the engine.
+ *
+ * - The memory block SHA_TEXT_BASE is shared between all SHA digest
+ *   engines, so all engines must be idle before this memory block is
+ *   modified.
+ *
+ */
+
 #ifdef __cplusplus
 extern "C" {
 #endif
 
-/**
- * \brief          SHA-1 context structure
- */
-typedef struct {
-    /* both types defined in rom/sha.h */
-    SHA_CTX context;
-    enum SHA_TYPE context_type;
-} esp_sha_context;
+/* Defined in rom/sha.h */
+typedef enum SHA_TYPE esp_sha_type;
 
-/**
- * \brief Lock access to SHA hardware unit
+/** @brief Calculate SHA1 or SHA2 sum of some data, using hardware SHA engine
  *
- * SHA hardware unit can only be used by one
- * consumer at a time.
+ * @note For more versatile SHA calculations, where data doesn't need
+ * to be passed all at once, try the mbedTLS mbedtls/shaX.h APIs. The
+ * hardware-accelerated mbedTLS implementation is also faster when
+ * hashing large amounts of data.
  *
- * esp_sha_xxx API calls automatically manage locking & unlocking of
- * hardware, this function is only needed if you want to call
- * ets_sha_xxx functions directly.
- */
-void esp_sha_acquire_hardware( void );
-
-/**
- * \brief Unlock access to SHA hardware unit
+ * @note It is not necessary to lock any SHA hardware before calling
+ * this function, thread safety is managed internally.
  *
- * esp_sha_xxx API calls automatically manage locking & unlocking of
- * hardware, this function is only needed if you want to call
- * ets_sha_xxx functions directly.
- */
-void esp_sha_release_hardware( void );
-
-/**
- * \brief          Initialize SHA-1 context
+ * @note If a TLS connection is open then this function may block
+ * indefinitely waiting for a SHA engine to become available. Use the
+ * mbedTLS SHA API to avoid this problem.
  *
- * \param ctx      SHA-1 context to be initialized
- */
-void esp_sha1_init( esp_sha_context *ctx );
-
-/**
- * \brief          Clear SHA-1 context
+ * @param sha_type SHA algorithm to use.
  *
- * \param ctx      SHA-1 context to be cleared
- */
-void esp_sha1_free( esp_sha_context *ctx );
-
-/**
- * \brief          Clone (the state of) a SHA-1 context
+ * @param input Input data buffer.
  *
- * \param dst      The destination context
- * \param src      The context to be cloned
- */
-void esp_sha1_clone( esp_sha_context *dst, const esp_sha_context *src );
-
-/**
- * \brief          SHA-1 context setup
+ * @param ilen Length of input data in bytes.
  *
- * \param ctx      context to be initialized
+ * @param output Buffer for output SHA digest. Output is 20 bytes for
+ * sha_type SHA1, 32 bytes for sha_type SHA2_256, 48 bytes for
+ * sha_type SHA2_384, 64 bytes for sha_type SHA2_512.
  */
-void esp_sha1_start( esp_sha_context *ctx );
+void esp_sha(esp_sha_type sha_type, const unsigned char *input, size_t ilen, unsigned char *output);
 
-/**
- * \brief          SHA-1 process buffer
+/* @brief Begin to execute a single SHA block operation
  *
- * \param ctx      SHA-1 context
- * \param input    buffer holding the  data
- * \param ilen     length of the input data
- */
-void esp_sha1_update( esp_sha_context *ctx, const unsigned char *input, size_t ilen );
-
-/**
- * \brief          SHA-1 final digest
+ * @note This is a piece of a SHA algorithm, rather than an entire SHA
+ * algorithm.
  *
- * \param ctx      SHA-1 context
- * \param output   SHA-1 checksum result
- */
-void esp_sha1_finish( esp_sha_context *ctx, unsigned char output[20] );
-
-/**
- * \brief          Calculate SHA-1 of input buffer
+ * @note Call esp_sha_try_lock_engine() before calling this
+ * function. Do not call esp_sha_lock_memory_block() beforehand, this
+ * is done inside the function.
  *
- * \param input    buffer holding the data
- * \param ilen     length of the input data
- * \param output   SHA-1 checksum result
- */
-void esp_sha1( const unsigned char *input, size_t ilen, unsigned char output[20] );
-
-/**
- * \brief          SHA-256 context structure
- */
-
-/**
- * \brief          Initialize SHA-256 context
+ * @param sha_type SHA algorithm to use.
  *
- * \param ctx      SHA-256 context to be initialized
- */
-void esp_sha256_init( esp_sha_context *ctx );
-
-/**
- * \brief          Clear SHA-256 context
+ * @param data_block Pointer to block of data. Block size is
+ * determined by algorithm (SHA1/SHA2_256 = 64 bytes,
+ * SHA2_384/SHA2_512 = 128 bytes)
  *
- * \param ctx      SHA-256 context to be cleared
- */
-void esp_sha256_free( esp_sha_context *ctx );
-
-/**
- * \brief          Clone (the state of) a SHA-256 context
+ * @param is_first_block If this parameter is true, the SHA state will
+ * be initialised (with the initial state of the given SHA algorithm)
+ * before the block is calculated. If false, the existing state of the
+ * SHA engine will be used.
  *
- * \param dst      The destination context
- * \param src      The context to be cloned
+ * @return As a performance optimisation, this function returns before
+ * the SHA block operation is complete. Both this function and
+ * esp_sha_read_state() will automatically wait for any previous
+ * operation to complete before they begin. If using the SHA registers
+ * directly in another way, call esp_sha_wait_idle() after calling this
+ * function but before accessing the SHA registers.
  */
-void esp_sha256_clone( esp_sha_context *dst, const esp_sha_context *src );
+void esp_sha_block(esp_sha_type sha_type, const void *data_block, bool is_first_block);
 
-/**
- * \brief          SHA-256 context setup
+/** @brief Read out the current state of the SHA digest loaded in the engine.
  *
- * \param ctx      context to be initialized
- * \param is224    0 = use SHA256, 1 = use SHA224
- */
-void esp_sha256_start( esp_sha_context *ctx, int is224 );
-
-/**
- * \brief          SHA-256 process buffer
+ * @note This is a piece of a SHA algorithm, rather than an entire SHA algorithm.
  *
- * \param ctx      SHA-256 context
- * \param input    buffer holding the  data
- * \param ilen     length of the input data
- */
-void esp_sha256_update( esp_sha_context *ctx, const unsigned char *input, size_t ilen );
-
-/**
- * \brief          SHA-256 final digest
+ * @note Call esp_sha_try_lock_engine() before calling this
+ * function. Do not call esp_sha_lock_memory_block() beforehand, this
+ * is done inside the function.
  *
- * \param ctx      SHA-256 context
- * \param output   SHA-224/256 checksum result
- */
-void esp_sha256_finish( esp_sha_context *ctx, unsigned char output[32] );
-
-/**
- * \brief          Calculate SHA-256 of input buffer
+ * If the SHA suffix padding block has been executed already, the
+ * value that is read is the SHA digest (in big endian
+ * format). Otherwise, the value that is read is an interim SHA state.
  *
- * \param input    buffer holding the data
- * \param ilen     length of the input data
- * \param output   SHA-224/256 checksum result
- * \param is224    0 = use SHA256, 1 = use SHA224
- */
-void esp_sha256( const unsigned char *input, size_t ilen, unsigned char output[32], int is224 );
-
-//
-
-/**
- * \brief          SHA-512 context structure
- */
-
-/**
- * \brief          Initialize SHA-512 context
+ * @param sha_type SHA algorithm in use.
+ *
+ * @param state Pointer to a memory buffer to hold the SHA state. Size
+ * is 20 bytes (SHA1), 64 bytes (SHA2_256), or 128 bytes (SHA2_384 or
+ * SHA2_512).
  *
- * \param ctx      SHA-512 context to be initialized
  */
-void esp_sha512_init( esp_sha_context *ctx );
+void esp_sha_read_digest_state(esp_sha_type sha_type, void *digest_state);
 
 /**
- * \brief          Clear SHA-512 context
+ * @brief Obtain exclusive access to a particular SHA engine
  *
- * \param ctx      SHA-512 context to be cleared
+ * @param sha_type Type of SHA engine to use.
+ *
+ * Blocks until engine is available. Note: Can block indefinitely
+ * while a TLS connection is open, suggest using
+ * esp_sha_try_lock_engine() and failing over to software SHA.
  */
-void esp_sha512_free( esp_sha_context *ctx );
+void esp_sha_lock_engine(esp_sha_type sha_type);
 
 /**
- * \brief          Clone (the state of) a SHA-512 context
+ * @brief Try and obtain exclusive access to a particular SHA engine
  *
- * \param dst      The destination context
- * \param src      The context to be cloned
+ * @param sha_type Type of SHA engine to use.
+ *
+ * @return Returns true if the SHA engine is locked for exclusive
+ * use. Call esp_sha_unlock_sha_engine() when done.  Returns false if
+ * the SHA engine is already in use, caller should use software SHA
+ * algorithm for this digest.
  */
-void esp_sha512_clone( esp_sha_context *dst, const esp_sha_context *src );
+bool esp_sha_try_lock_engine(esp_sha_type sha_type);
 
 /**
- * \brief          SHA-512 context setup
+ * @brief Unlock an engine previously locked with esp_sha_lock_engine() or esp_sha_try_lock_engine()
  *
- * \param ctx      context to be initialized
- * \param is384    0 = use SHA512, 1 = use SHA384
+ * @param sha_type Type of engine to release.
  */
-void esp_sha512_start( esp_sha_context *ctx, int is384 );
+void esp_sha_unlock_engine(esp_sha_type sha_type);
 
 /**
- * \brief          SHA-512 process buffer
+ * @brief Acquire exclusive access to the SHA shared memory block at SHA_TEXT_BASE
  *
- * \param ctx      SHA-512 context
- * \param input    buffer holding the  data
- * \param ilen     length of the input data
+ * This memory block is shared across all the SHA algorithm types.
+ *
+ * Caller should have already locked a SHA engine before calling this function.
+ *
+ * Note that it is possible to obtain exclusive access to the memory block even
+ * while it is in use by the SHA engine. Caller should use esp_sha_wait_idle()
+ * to ensure the SHA engine is not reading from the memory block in hardware.
+ *
+ * @note You do not need to lock the memory block before calling esp_sha_block() or esp_sha_read_digest_state(), these functions handle memory block locking internally.
+ *
+ * Call esp_sha_unlock_memory_block() when done.
  */
-void esp_sha512_update( esp_sha_context *ctx, const unsigned char *input, size_t ilen );
+void esp_sha_lock_memory_block(void);
 
 /**
- * \brief          SHA-512 final digest
+ * @brief Release exclusive access to the SHA register memory block at SHA_TEXT_BASE
  *
- * \param ctx      SHA-512 context
- * \param output   SHA-384/512 checksum result
- */
-void esp_sha512_finish( esp_sha_context *ctx, unsigned char output[64] );
-
-/**
- * \brief          Calculate SHA-512 of input buffer.
+ * Caller should have already locked a SHA engine before calling this function.
  *
- * \param input    buffer holding the data
- * \param ilen     length of the input data
- * \param output   SHA-384/512 checksum result
- * \param is384    0 = use SHA512, 1 = use SHA384
+ * Call following esp_sha_lock_memory_block().
  */
-void esp_sha512( const unsigned char *input, size_t ilen, unsigned char output[64], int is384 );
+void esp_sha_unlock_memory_block(void);
 
-//
+/** @brief Wait for the SHA engine to finish any current operation
+ *
+ * @note This function does not ensure exclusive access to any SHA
+ * engine. Caller should use esp_sha_try_lock_engine() and
+ * esp_sha_lock_memory_block() as required.
+ *
+ * @note Functions declared in this header file wait for SHA engine
+ * completion automatically, so you don't need to use this API for
+ * these. However if accessing SHA registers directly, you will need
+ * to call this before accessing SHA registers if using the
+ * esp_sha_block() function.
+ *
+ * @note This function busy-waits, so wastes CPU resources.
+ * Best to delay calling until you are about to need it.
+ *
+ */
+void esp_sha_wait_idle(void);
 
 #ifdef __cplusplus
 }
diff --git a/components/esp32/include/rom/sha.h b/components/esp32/include/rom/sha.h
index 8082a394c..5dd9c9981 100644
--- a/components/esp32/include/rom/sha.h
+++ b/components/esp32/include/rom/sha.h
@@ -1,9 +1,10 @@
 /*
   ROM functions for hardware SHA support.
 
-  It is not recommended to use these functions directly,
-  use the wrapper functions in hwcrypto/sha.h instead.
-
+  It is not recommended to use these functions directly.  If using
+  them from esp-idf then use the esp_sha_lock_engine() and
+  esp_sha_lock_memory_block() functions in hwcrypto/sha.h to ensure
+  exclusive access.
  */
 // Copyright 2015-2016 Espressif Systems (Shanghai) PTE LTD
 //
@@ -38,6 +39,8 @@ enum SHA_TYPE {
     SHA2_256,
     SHA2_384,
     SHA2_512,
+
+
     SHA_INVALID = -1,
 };
 
diff --git a/components/esp32/include/soc/dport_reg.h b/components/esp32/include/soc/dport_reg.h
index 1be0fdee1..f84346717 100644
--- a/components/esp32/include/soc/dport_reg.h
+++ b/components/esp32/include/soc/dport_reg.h
@@ -94,6 +94,16 @@
 #define DPORT_PERI_RST_EN_V  0xFFFFFFFF
 #define DPORT_PERI_RST_EN_S  0
 
+/* The following bits apply to DPORT_PERI_CLK_EN_REG, DPORT_PERI_RST_EN_REG
+ */
+#define DPORT_PERI_EN_AES (1<<0)
+#define DPORT_PERI_EN_SHA (1<<1)
+#define DPORT_PERI_EN_RSA (1<<2)
+/* NB: Secure boot reset will hold SHA & AES in reset */
+#define DPORT_PERI_EN_SECUREBOOT (1<<3)
+/* NB: Digital signature reset will hold AES & RSA in reset */
+#define DPORT_PERI_EN_DIGITAL_SIGNATURE (1<<4)
+
 #define DPORT_WIFI_BB_CFG_REG          (DR_REG_DPORT_BASE + 0x024)
 /* DPORT_WIFI_BB_CFG : R/W ;bitpos:[31:0] ;default: 32'h0 ; */
 /*description: */
diff --git a/components/esp32/include/soc/hwcrypto_reg.h b/components/esp32/include/soc/hwcrypto_reg.h
index 4f38b1ba9..10e80d935 100644
--- a/components/esp32/include/soc/hwcrypto_reg.h
+++ b/components/esp32/include/soc/hwcrypto_reg.h
@@ -30,8 +30,31 @@
 #define RSA_MULT_MODE_REG             (DR_REG_RSA_BASE + 0x80c)
 #define RSA_MULT_START_REG            (DR_REG_RSA_BASE + 0x810)
 
-#define RSA_INTERRUPT_REG             (DR_REG_RSA_BASE + 0X814)
+#define RSA_INTERRUPT_REG             (DR_REG_RSA_BASE + 0x814)
 
-#define RSA_CLEAN_ADDR                (DR_REG_RSA_BASE + 0X818)
+#define RSA_CLEAN_REG                 (DR_REG_RSA_BASE + 0x818)
+
+/* SHA acceleration registers */
+#define SHA_TEXT_BASE          ((DR_REG_SHA_BASE) + 0x00)
+
+#define SHA_1_START_REG         ((DR_REG_SHA_BASE) + 0x80)
+#define SHA_1_CONTINUE_REG      ((DR_REG_SHA_BASE) + 0x84)
+#define SHA_1_LOAD_REG          ((DR_REG_SHA_BASE) + 0x88)
+#define SHA_1_BUSY_REG          ((DR_REG_SHA_BASE) + 0x8c)
+
+#define SHA_256_START_REG       ((DR_REG_SHA_BASE) + 0x90)
+#define SHA_256_CONTINUE_REG    ((DR_REG_SHA_BASE) + 0x94)
+#define SHA_256_LOAD_REG        ((DR_REG_SHA_BASE) + 0x98)
+#define SHA_256_BUSY_REG        ((DR_REG_SHA_BASE) + 0x9c)
+
+#define SHA_384_START_REG       ((DR_REG_SHA_BASE) + 0xa0)
+#define SHA_384_CONTINUE_REG    ((DR_REG_SHA_BASE) + 0xa4)
+#define SHA_384_LOAD_REG        ((DR_REG_SHA_BASE) + 0xa8)
+#define SHA_384_BUSY_REG        ((DR_REG_SHA_BASE) + 0xac)
+
+#define SHA_512_START_REG       ((DR_REG_SHA_BASE) + 0xb0)
+#define SHA_512_CONTINUE_REG    ((DR_REG_SHA_BASE) + 0xb4)
+#define SHA_512_LOAD_REG        ((DR_REG_SHA_BASE) + 0xb8)
+#define SHA_512_BUSY_REG        ((DR_REG_SHA_BASE) + 0xbc)
 
 #endif
diff --git a/components/esp32/include/soc/soc.h b/components/esp32/include/soc/soc.h
index 61ec2e1f2..5781b4c85 100755
--- a/components/esp32/include/soc/soc.h
+++ b/components/esp32/include/soc/soc.h
@@ -142,6 +142,7 @@
 
 #define DR_REG_DPORT_BASE                       0x3ff00000
 #define DR_REG_RSA_BASE                         0x3ff02000
+#define DR_REG_SHA_BASE                         0x3ff03000
 #define DR_REG_UART_BASE                        0x3ff40000
 #define DR_REG_SPI1_BASE                        0x3ff42000
 #define DR_REG_SPI0_BASE                        0x3ff43000
diff --git a/components/mbedtls/Kconfig b/components/mbedtls/Kconfig
index d6e2a2dcb..1e347582f 100644
--- a/components/mbedtls/Kconfig
+++ b/components/mbedtls/Kconfig
@@ -69,4 +69,16 @@ config MBEDTLS_MPI_INTERRUPT_NUM
         CPU interrupt number for MPI interrupt to connect to. Must be otherwise unused.
         Eventually this assignment will be handled automatically at runtime.
 
+config MBEDTLS_HARDWARE_SHA
+   bool "Enable hardware SHA acceleration"
+   default y
+   help
+       Enable hardware accelerated SHA1, SHA256, SHA384 & SHA512 in mbedTLS.
+
+       Due to a hardware limitation, hardware acceleration is only
+       guaranteed if SHA digests are calculated one at a time. If more
+       than one SHA digest is calculated at the same time, only will
+       be calculated fully in hardware and the rest will be calculated
+       (at least partially calculated) in software.
+
 endmenu
diff --git a/components/mbedtls/component.mk b/components/mbedtls/component.mk
index bd7209a92..d2cd0455f 100644
--- a/components/mbedtls/component.mk
+++ b/components/mbedtls/component.mk
@@ -5,4 +5,3 @@
 COMPONENT_ADD_INCLUDEDIRS := port/include include
 
 COMPONENT_SRCDIRS := library port
-
diff --git a/components/mbedtls/port/esp_bignum.c b/components/mbedtls/port/esp_bignum.c
index ec2e42039..2fe4d920c 100644
--- a/components/mbedtls/port/esp_bignum.c
+++ b/components/mbedtls/port/esp_bignum.c
@@ -34,6 +34,8 @@
 #include "esp_intr.h"
 #include "esp_attr.h"
 
+#include "soc/dport_reg.h"
+
 #include "freertos/FreeRTOS.h"
 #include "freertos/task.h"
 #include "freertos/semphr.h"
@@ -72,7 +74,16 @@ void esp_mpi_acquire_hardware( void )
 {
     /* newlib locks lazy initialize on ESP-IDF */
     _lock_acquire(&mpi_lock);
-    ets_bigint_enable();
+    REG_SET_BIT(DPORT_PERI_CLK_EN_REG, DPORT_PERI_EN_RSA);
+    /* also clear reset on digital signature, otherwise RSA is held in reset */
+    REG_CLR_BIT(DPORT_PERI_RST_EN_REG,
+                DPORT_PERI_EN_RSA
+                | DPORT_PERI_EN_DIGITAL_SIGNATURE);
+
+    REG_CLR_BIT(DPORT_RSA_PD_CTRL_REG, DPORT_RSA_PD);
+
+    while(REG_READ(RSA_CLEAN_REG) != 1);
+
 #ifdef CONFIG_MBEDTLS_MPI_USE_INTERRUPT
     rsa_isr_initialise();
 #endif
@@ -80,7 +91,12 @@ void esp_mpi_acquire_hardware( void )
 
 void esp_mpi_release_hardware( void )
 {
-    ets_bigint_disable();
+    REG_SET_BIT(DPORT_RSA_PD_CTRL_REG, DPORT_RSA_PD);
+
+    /* don't reset digital signature unit, as this resets AES also */
+    REG_SET_BIT(DPORT_PERI_RST_EN_REG, DPORT_PERI_EN_RSA);
+    REG_CLR_BIT(DPORT_PERI_CLK_EN_REG, DPORT_PERI_EN_RSA);
+
     _lock_release(&mpi_lock);
 }
 
diff --git a/components/mbedtls/port/esp_sha1.c b/components/mbedtls/port/esp_sha1.c
new file mode 100644
index 000000000..675159693
--- /dev/null
+++ b/components/mbedtls/port/esp_sha1.c
@@ -0,0 +1,395 @@
+/*
+ *  SHA-1 implementation with hardware ESP32 support added.
+ *  Uses mbedTLS software implementation for failover when concurrent
+ *  SHA operations are in use.
+ *
+ *  Copyright (C) 2006-2015, ARM Limited, All Rights Reserved
+ *  Additions Copyright (C) 2016, Espressif Systems (Shanghai) PTE LTD
+ *  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
+ *
+ *  http://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.
+ *
+ */
+/*
+ *  The SHA-1 standard was published by NIST in 1993.
+ *
+ *  http://www.itl.nist.gov/fipspubs/fip180-1.htm
+ */
+
+#if !defined(MBEDTLS_CONFIG_FILE)
+#include "mbedtls/config.h"
+#else
+#include MBEDTLS_CONFIG_FILE
+#endif
+
+#if defined(MBEDTLS_SHA1_C) && defined(MBEDTLS_SHA1_ALT)
+
+#include "mbedtls/sha1.h"
+
+#include <string.h>
+
+#if defined(MBEDTLS_SELF_TEST)
+#if defined(MBEDTLS_PLATFORM_C)
+#include "mbedtls/platform.h"
+#else
+#include <stdio.h>
+#define mbedtls_printf printf
+#endif /* MBEDTLS_PLATFORM_C */
+#endif /* MBEDTLS_SELF_TEST */
+
+#include "hwcrypto/sha.h"
+
+/* Implementation that should never be optimized out by the compiler */
+static void mbedtls_zeroize( void *v, size_t n ) {
+    volatile unsigned char *p = (unsigned char*)v; while( n-- ) *p++ = 0;
+}
+
+/*
+ * 32-bit integer manipulation macros (big endian)
+ */
+#ifndef GET_UINT32_BE
+#define GET_UINT32_BE(n,b,i)                            \
+{                                                       \
+    (n) = ( (uint32_t) (b)[(i)    ] << 24 )             \
+        | ( (uint32_t) (b)[(i) + 1] << 16 )             \
+        | ( (uint32_t) (b)[(i) + 2] <<  8 )             \
+        | ( (uint32_t) (b)[(i) + 3]       );            \
+}
+#endif
+
+#ifndef PUT_UINT32_BE
+#define PUT_UINT32_BE(n,b,i)                            \
+{                                                       \
+    (b)[(i)    ] = (unsigned char) ( (n) >> 24 );       \
+    (b)[(i) + 1] = (unsigned char) ( (n) >> 16 );       \
+    (b)[(i) + 2] = (unsigned char) ( (n) >>  8 );       \
+    (b)[(i) + 3] = (unsigned char) ( (n)       );       \
+}
+#endif
+
+void mbedtls_sha1_init( mbedtls_sha1_context *ctx )
+{
+    memset( ctx, 0, sizeof( mbedtls_sha1_context ) );
+}
+
+void mbedtls_sha1_free( mbedtls_sha1_context *ctx )
+{
+    if( ctx == NULL )
+        return;
+
+    if (ctx->mode == ESP_MBEDTLS_SHA1_HARDWARE) {
+        esp_sha_unlock_engine(SHA1);
+    }
+    mbedtls_zeroize( ctx, sizeof( mbedtls_sha1_context ) );
+}
+
+void mbedtls_sha1_clone( mbedtls_sha1_context *dst,
+                         const mbedtls_sha1_context *src )
+{
+    *dst = *src;
+
+    if (src->mode == ESP_MBEDTLS_SHA1_HARDWARE) {
+        /* Copy hardware digest state out to cloned state,
+           which will be a software digest.
+        */
+        esp_sha_read_digest_state(SHA1, dst->state);
+        dst->mode = ESP_MBEDTLS_SHA1_SOFTWARE;
+    }
+}
+
+
+/*
+ * SHA-1 context setup
+ */
+void mbedtls_sha1_starts( mbedtls_sha1_context *ctx )
+{
+    ctx->total[0] = 0;
+    ctx->total[1] = 0;
+
+    ctx->state[0] = 0x67452301;
+    ctx->state[1] = 0xEFCDAB89;
+    ctx->state[2] = 0x98BADCFE;
+    ctx->state[3] = 0x10325476;
+    ctx->state[4] = 0xC3D2E1F0;
+
+    if (ctx->mode == ESP_MBEDTLS_SHA1_HARDWARE) {
+        esp_sha_unlock_engine(SHA1);
+    }
+    ctx->mode = ESP_MBEDTLS_SHA1_UNUSED;
+}
+
+static void mbedtls_sha1_software_process( mbedtls_sha1_context *ctx, const unsigned char data[64] );
+
+void mbedtls_sha1_process( mbedtls_sha1_context *ctx, const unsigned char data[64] )
+{
+    bool first_block = false;
+    if (ctx->mode == ESP_MBEDTLS_SHA1_UNUSED) {
+        /* try to use hardware for this digest */
+        if (esp_sha_try_lock_engine(SHA1)) {
+            ctx->mode = ESP_MBEDTLS_SHA1_HARDWARE;
+            first_block = true;
+        } else {
+            ctx->mode = ESP_MBEDTLS_SHA1_SOFTWARE;
+        }
+    }
+
+    if (ctx->mode == ESP_MBEDTLS_SHA1_HARDWARE) {
+        esp_sha_block(SHA1, data, first_block);
+    } else {
+        mbedtls_sha1_software_process(ctx, data);
+    }
+}
+
+
+static void mbedtls_sha1_software_process( mbedtls_sha1_context *ctx, const unsigned char data[64] )
+{
+    uint32_t temp, W[16], A, B, C, D, E;
+
+    GET_UINT32_BE( W[ 0], data,  0 );
+    GET_UINT32_BE( W[ 1], data,  4 );
+    GET_UINT32_BE( W[ 2], data,  8 );
+    GET_UINT32_BE( W[ 3], data, 12 );
+    GET_UINT32_BE( W[ 4], data, 16 );
+    GET_UINT32_BE( W[ 5], data, 20 );
+    GET_UINT32_BE( W[ 6], data, 24 );
+    GET_UINT32_BE( W[ 7], data, 28 );
+    GET_UINT32_BE( W[ 8], data, 32 );
+    GET_UINT32_BE( W[ 9], data, 36 );
+    GET_UINT32_BE( W[10], data, 40 );
+    GET_UINT32_BE( W[11], data, 44 );
+    GET_UINT32_BE( W[12], data, 48 );
+    GET_UINT32_BE( W[13], data, 52 );
+    GET_UINT32_BE( W[14], data, 56 );
+    GET_UINT32_BE( W[15], data, 60 );
+
+#define S(x,n) ((x << n) | ((x & 0xFFFFFFFF) >> (32 - n)))
+
+#define R(t)                                            \
+(                                                       \
+    temp = W[( t -  3 ) & 0x0F] ^ W[( t - 8 ) & 0x0F] ^ \
+           W[( t - 14 ) & 0x0F] ^ W[  t       & 0x0F],  \
+    ( W[t & 0x0F] = S(temp,1) )                         \
+)
+
+#define P(a,b,c,d,e,x)                                  \
+{                                                       \
+    e += S(a,5) + F(b,c,d) + K + x; b = S(b,30);        \
+}
+
+    A = ctx->state[0];
+    B = ctx->state[1];
+    C = ctx->state[2];
+    D = ctx->state[3];
+    E = ctx->state[4];
+
+#define F(x,y,z) (z ^ (x & (y ^ z)))
+#define K 0x5A827999
+
+    P( A, B, C, D, E, W[0]  );
+    P( E, A, B, C, D, W[1]  );
+    P( D, E, A, B, C, W[2]  );
+    P( C, D, E, A, B, W[3]  );
+    P( B, C, D, E, A, W[4]  );
+    P( A, B, C, D, E, W[5]  );
+    P( E, A, B, C, D, W[6]  );
+    P( D, E, A, B, C, W[7]  );
+    P( C, D, E, A, B, W[8]  );
+    P( B, C, D, E, A, W[9]  );
+    P( A, B, C, D, E, W[10] );
+    P( E, A, B, C, D, W[11] );
+    P( D, E, A, B, C, W[12] );
+    P( C, D, E, A, B, W[13] );
+    P( B, C, D, E, A, W[14] );
+    P( A, B, C, D, E, W[15] );
+    P( E, A, B, C, D, R(16) );
+    P( D, E, A, B, C, R(17) );
+    P( C, D, E, A, B, R(18) );
+    P( B, C, D, E, A, R(19) );
+
+#undef K
+#undef F
+
+#define F(x,y,z) (x ^ y ^ z)
+#define K 0x6ED9EBA1
+
+    P( A, B, C, D, E, R(20) );
+    P( E, A, B, C, D, R(21) );
+    P( D, E, A, B, C, R(22) );
+    P( C, D, E, A, B, R(23) );
+    P( B, C, D, E, A, R(24) );
+    P( A, B, C, D, E, R(25) );
+    P( E, A, B, C, D, R(26) );
+    P( D, E, A, B, C, R(27) );
+    P( C, D, E, A, B, R(28) );
+    P( B, C, D, E, A, R(29) );
+    P( A, B, C, D, E, R(30) );
+    P( E, A, B, C, D, R(31) );
+    P( D, E, A, B, C, R(32) );
+    P( C, D, E, A, B, R(33) );
+    P( B, C, D, E, A, R(34) );
+    P( A, B, C, D, E, R(35) );
+    P( E, A, B, C, D, R(36) );
+    P( D, E, A, B, C, R(37) );
+    P( C, D, E, A, B, R(38) );
+    P( B, C, D, E, A, R(39) );
+
+#undef K
+#undef F
+
+#define F(x,y,z) ((x & y) | (z & (x | y)))
+#define K 0x8F1BBCDC
+
+    P( A, B, C, D, E, R(40) );
+    P( E, A, B, C, D, R(41) );
+    P( D, E, A, B, C, R(42) );
+    P( C, D, E, A, B, R(43) );
+    P( B, C, D, E, A, R(44) );
+    P( A, B, C, D, E, R(45) );
+    P( E, A, B, C, D, R(46) );
+    P( D, E, A, B, C, R(47) );
+    P( C, D, E, A, B, R(48) );
+    P( B, C, D, E, A, R(49) );
+    P( A, B, C, D, E, R(50) );
+    P( E, A, B, C, D, R(51) );
+    P( D, E, A, B, C, R(52) );
+    P( C, D, E, A, B, R(53) );
+    P( B, C, D, E, A, R(54) );
+    P( A, B, C, D, E, R(55) );
+    P( E, A, B, C, D, R(56) );
+    P( D, E, A, B, C, R(57) );
+    P( C, D, E, A, B, R(58) );
+    P( B, C, D, E, A, R(59) );
+
+#undef K
+#undef F
+
+#define F(x,y,z) (x ^ y ^ z)
+#define K 0xCA62C1D6
+
+    P( A, B, C, D, E, R(60) );
+    P( E, A, B, C, D, R(61) );
+    P( D, E, A, B, C, R(62) );
+    P( C, D, E, A, B, R(63) );
+    P( B, C, D, E, A, R(64) );
+    P( A, B, C, D, E, R(65) );
+    P( E, A, B, C, D, R(66) );
+    P( D, E, A, B, C, R(67) );
+    P( C, D, E, A, B, R(68) );
+    P( B, C, D, E, A, R(69) );
+    P( A, B, C, D, E, R(70) );
+    P( E, A, B, C, D, R(71) );
+    P( D, E, A, B, C, R(72) );
+    P( C, D, E, A, B, R(73) );
+    P( B, C, D, E, A, R(74) );
+    P( A, B, C, D, E, R(75) );
+    P( E, A, B, C, D, R(76) );
+    P( D, E, A, B, C, R(77) );
+    P( C, D, E, A, B, R(78) );
+    P( B, C, D, E, A, R(79) );
+
+#undef K
+#undef F
+
+    ctx->state[0] += A;
+    ctx->state[1] += B;
+    ctx->state[2] += C;
+    ctx->state[3] += D;
+    ctx->state[4] += E;
+}
+
+/*
+ * SHA-1 process buffer
+ */
+void mbedtls_sha1_update( mbedtls_sha1_context *ctx, const unsigned char *input, size_t ilen )
+{
+    size_t fill;
+    uint32_t left;
+
+    if( ilen == 0 )
+        return;
+
+    left = ctx->total[0] & 0x3F;
+    fill = 64 - left;
+
+    ctx->total[0] += (uint32_t) ilen;
+    ctx->total[0] &= 0xFFFFFFFF;
+
+    if( ctx->total[0] < (uint32_t) ilen )
+        ctx->total[1]++;
+
+    if( left && ilen >= fill )
+    {
+        memcpy( (void *) (ctx->buffer + left), input, fill );
+        mbedtls_sha1_process( ctx, ctx->buffer );
+        input += fill;
+        ilen  -= fill;
+        left = 0;
+    }
+
+    while( ilen >= 64 )
+    {
+        mbedtls_sha1_process( ctx, input );
+        input += 64;
+        ilen  -= 64;
+    }
+
+    if( ilen > 0 )
+        memcpy( (void *) (ctx->buffer + left), input, ilen );
+}
+
+static const unsigned char sha1_padding[64] =
+{
+ 0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
+};
+
+/*
+* SHA-1 final digest
+ */
+void mbedtls_sha1_finish( mbedtls_sha1_context *ctx, unsigned char output[20] )
+{
+    uint32_t last, padn;
+    uint32_t high, low;
+    unsigned char msglen[8];
+
+    high = ( ctx->total[0] >> 29 )
+         | ( ctx->total[1] <<  3 );
+    low  = ( ctx->total[0] <<  3 );
+
+    PUT_UINT32_BE( high, msglen, 0 );
+    PUT_UINT32_BE( low,  msglen, 4 );
+
+    last = ctx->total[0] & 0x3F;
+    padn = ( last < 56 ) ? ( 56 - last ) : ( 120 - last );
+
+    mbedtls_sha1_update( ctx, sha1_padding, padn );
+    mbedtls_sha1_update( ctx, msglen, 8 );
+
+    /* if state is in hardware, read it out */
+    if (ctx->mode == ESP_MBEDTLS_SHA1_HARDWARE) {
+        esp_sha_read_digest_state(SHA1, ctx->state);
+        esp_sha_unlock_engine(SHA1);
+        ctx->mode = ESP_MBEDTLS_SHA1_SOFTWARE;
+    }
+
+    PUT_UINT32_BE( ctx->state[0], output,  0 );
+    PUT_UINT32_BE( ctx->state[1], output,  4 );
+    PUT_UINT32_BE( ctx->state[2], output,  8 );
+    PUT_UINT32_BE( ctx->state[3], output, 12 );
+    PUT_UINT32_BE( ctx->state[4], output, 16 );
+
+}
+
+#endif /* MBEDTLS_SHA1_C && MBEDTLS_SHA1_ALT */
diff --git a/components/mbedtls/port/esp_sha256.c b/components/mbedtls/port/esp_sha256.c
new file mode 100644
index 000000000..fd136062a
--- /dev/null
+++ b/components/mbedtls/port/esp_sha256.c
@@ -0,0 +1,367 @@
+/*
+ *  SHA-256 implementation with hardware ESP32 support added.
+ *  Uses mbedTLS software implementation for failover when concurrent
+ *  SHA operations are in use.
+ *
+ *  Copyright (C) 2006-2015, ARM Limited, All Rights Reserved
+ *  Additions Copyright (C) 2016, Espressif Systems (Shanghai) PTE LTD
+ *  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
+ *
+ *  http://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.
+ *
+ */
+
+/*
+ *  The SHA-256 Secure Hash Standard was published by NIST in 2002.
+ *
+ *  http://csrc.nist.gov/publications/fips/fips180-2/fips180-2.pdf
+ */
+
+#if !defined(MBEDTLS_CONFIG_FILE)
+#include "mbedtls/config.h"
+#else
+#include MBEDTLS_CONFIG_FILE
+#endif
+
+#if defined(MBEDTLS_SHA256_C) && defined(MBEDTLS_SHA256_ALT)
+
+#include "mbedtls/sha256.h"
+
+#include <string.h>
+
+#if defined(MBEDTLS_SELF_TEST)
+#if defined(MBEDTLS_PLATFORM_C)
+#include "mbedtls/platform.h"
+#else
+#include <stdio.h>
+#define mbedtls_printf printf
+#endif /* MBEDTLS_PLATFORM_C */
+#endif /* MBEDTLS_SELF_TEST */
+
+#include "hwcrypto/sha.h"
+
+/* Implementation that should never be optimized out by the compiler */
+static void mbedtls_zeroize( void *v, size_t n ) {
+    volatile unsigned char *p = v; while( n-- ) *p++ = 0;
+}
+
+/*
+ * 32-bit integer manipulation macros (big endian)
+ */
+#ifndef GET_UINT32_BE
+#define GET_UINT32_BE(n,b,i)                            \
+do {                                                    \
+    (n) = ( (uint32_t) (b)[(i)    ] << 24 )             \
+        | ( (uint32_t) (b)[(i) + 1] << 16 )             \
+        | ( (uint32_t) (b)[(i) + 2] <<  8 )             \
+        | ( (uint32_t) (b)[(i) + 3]       );            \
+} while( 0 )
+#endif
+
+#ifndef PUT_UINT32_BE
+#define PUT_UINT32_BE(n,b,i)                            \
+do {                                                    \
+    (b)[(i)    ] = (unsigned char) ( (n) >> 24 );       \
+    (b)[(i) + 1] = (unsigned char) ( (n) >> 16 );       \
+    (b)[(i) + 2] = (unsigned char) ( (n) >>  8 );       \
+    (b)[(i) + 3] = (unsigned char) ( (n)       );       \
+} while( 0 )
+#endif
+
+void mbedtls_sha256_init( mbedtls_sha256_context *ctx )
+{
+    memset( ctx, 0, sizeof( mbedtls_sha256_context ) );
+}
+
+void mbedtls_sha256_free( mbedtls_sha256_context *ctx )
+{
+    if( ctx == NULL )
+        return;
+
+    if (ctx->mode == ESP_MBEDTLS_SHA256_HARDWARE) {
+        esp_sha_unlock_engine(SHA2_256);
+    }
+    mbedtls_zeroize( ctx, sizeof( mbedtls_sha256_context ) );
+}
+
+void mbedtls_sha256_clone( mbedtls_sha256_context *dst,
+                           const mbedtls_sha256_context *src )
+{
+    *dst = *src;
+
+    if (src->mode == ESP_MBEDTLS_SHA256_HARDWARE) {
+        /* Copy hardware digest state out to cloned state,
+           which will become a software digest.
+        */
+        esp_sha_read_digest_state(SHA2_256, dst->state);
+        dst->mode = ESP_MBEDTLS_SHA256_SOFTWARE;
+    }
+}
+
+/*
+ * SHA-256 context setup
+ */
+void mbedtls_sha256_starts( mbedtls_sha256_context *ctx, int is224 )
+{
+    ctx->total[0] = 0;
+    ctx->total[1] = 0;
+
+    if( is224 == 0 )
+    {
+        /* SHA-256 */
+        ctx->state[0] = 0x6A09E667;
+        ctx->state[1] = 0xBB67AE85;
+        ctx->state[2] = 0x3C6EF372;
+        ctx->state[3] = 0xA54FF53A;
+        ctx->state[4] = 0x510E527F;
+        ctx->state[5] = 0x9B05688C;
+        ctx->state[6] = 0x1F83D9AB;
+        ctx->state[7] = 0x5BE0CD19;
+    }
+    else
+    {
+        /* SHA-224 */
+        ctx->state[0] = 0xC1059ED8;
+        ctx->state[1] = 0x367CD507;
+        ctx->state[2] = 0x3070DD17;
+        ctx->state[3] = 0xF70E5939;
+        ctx->state[4] = 0xFFC00B31;
+        ctx->state[5] = 0x68581511;
+        ctx->state[6] = 0x64F98FA7;
+        ctx->state[7] = 0xBEFA4FA4;
+    }
+
+    ctx->is224 = is224;
+    if (ctx->mode == ESP_MBEDTLS_SHA256_HARDWARE) {
+        esp_sha_unlock_engine(SHA2_256);
+    }
+    ctx->mode = ESP_MBEDTLS_SHA256_UNUSED;
+}
+
+#if !defined(MBEDTLS_SHA256_PROCESS_ALT)
+static const uint32_t K[] =
+{
+    0x428A2F98, 0x71374491, 0xB5C0FBCF, 0xE9B5DBA5,
+    0x3956C25B, 0x59F111F1, 0x923F82A4, 0xAB1C5ED5,
+    0xD807AA98, 0x12835B01, 0x243185BE, 0x550C7DC3,
+    0x72BE5D74, 0x80DEB1FE, 0x9BDC06A7, 0xC19BF174,
+    0xE49B69C1, 0xEFBE4786, 0x0FC19DC6, 0x240CA1CC,
+    0x2DE92C6F, 0x4A7484AA, 0x5CB0A9DC, 0x76F988DA,
+    0x983E5152, 0xA831C66D, 0xB00327C8, 0xBF597FC7,
+    0xC6E00BF3, 0xD5A79147, 0x06CA6351, 0x14292967,
+    0x27B70A85, 0x2E1B2138, 0x4D2C6DFC, 0x53380D13,
+    0x650A7354, 0x766A0ABB, 0x81C2C92E, 0x92722C85,
+    0xA2BFE8A1, 0xA81A664B, 0xC24B8B70, 0xC76C51A3,
+    0xD192E819, 0xD6990624, 0xF40E3585, 0x106AA070,
+    0x19A4C116, 0x1E376C08, 0x2748774C, 0x34B0BCB5,
+    0x391C0CB3, 0x4ED8AA4A, 0x5B9CCA4F, 0x682E6FF3,
+    0x748F82EE, 0x78A5636F, 0x84C87814, 0x8CC70208,
+    0x90BEFFFA, 0xA4506CEB, 0xBEF9A3F7, 0xC67178F2,
+};
+
+#define  SHR(x,n) ((x & 0xFFFFFFFF) >> n)
+#define ROTR(x,n) (SHR(x,n) | (x << (32 - n)))
+
+#define S0(x) (ROTR(x, 7) ^ ROTR(x,18) ^  SHR(x, 3))
+#define S1(x) (ROTR(x,17) ^ ROTR(x,19) ^  SHR(x,10))
+
+#define S2(x) (ROTR(x, 2) ^ ROTR(x,13) ^ ROTR(x,22))
+#define S3(x) (ROTR(x, 6) ^ ROTR(x,11) ^ ROTR(x,25))
+
+#define F0(x,y,z) ((x & y) | (z & (x | y)))
+#define F1(x,y,z) (z ^ (x & (y ^ z)))
+
+#define R(t)                                    \
+(                                               \
+    W[t] = S1(W[t -  2]) + W[t -  7] +          \
+           S0(W[t - 15]) + W[t - 16]            \
+)
+
+#define P(a,b,c,d,e,f,g,h,x,K)                  \
+{                                               \
+    temp1 = h + S3(e) + F1(e,f,g) + K + x;      \
+    temp2 = S2(a) + F0(a,b,c);                  \
+    d += temp1; h = temp1 + temp2;              \
+}
+
+static void mbedtls_sha256_software_process( mbedtls_sha256_context *ctx, const unsigned char data[64] );
+
+void mbedtls_sha256_process( mbedtls_sha256_context *ctx, const unsigned char data[64] )
+{
+    bool first_block = false;
+
+    if (ctx->mode == ESP_MBEDTLS_SHA256_UNUSED) {
+        /* try to use hardware for this digest */
+        if (!ctx->is224 && esp_sha_try_lock_engine(SHA2_256)) {
+            ctx->mode = ESP_MBEDTLS_SHA256_HARDWARE;
+            first_block = true;
+        } else {
+            ctx->mode = ESP_MBEDTLS_SHA256_SOFTWARE;
+        }
+    }
+
+    if (ctx->mode == ESP_MBEDTLS_SHA256_HARDWARE) {
+        esp_sha_block(SHA2_256, data, first_block);
+    } else {
+        mbedtls_sha256_software_process(ctx, data);
+    }
+}
+
+
+static void mbedtls_sha256_software_process( mbedtls_sha256_context *ctx, const unsigned char data[64] )
+{
+    uint32_t temp1, temp2, W[64];
+    uint32_t A[8];
+    unsigned int i;
+
+    for( i = 0; i < 8; i++ )
+        A[i] = ctx->state[i];
+
+#if defined(MBEDTLS_SHA256_SMALLER)
+    for( i = 0; i < 64; i++ )
+    {
+        if( i < 16 )
+            GET_UINT32_BE( W[i], data, 4 * i );
+        else
+            R( i );
+
+        P( A[0], A[1], A[2], A[3], A[4], A[5], A[6], A[7], W[i], K[i] );
+
+        temp1 = A[7]; A[7] = A[6]; A[6] = A[5]; A[5] = A[4]; A[4] = A[3];
+        A[3] = A[2]; A[2] = A[1]; A[1] = A[0]; A[0] = temp1;
+    }
+#else /* MBEDTLS_SHA256_SMALLER */
+    for( i = 0; i < 16; i++ )
+        GET_UINT32_BE( W[i], data, 4 * i );
+
+    for( i = 0; i < 16; i += 8 )
+    {
+        P( A[0], A[1], A[2], A[3], A[4], A[5], A[6], A[7], W[i+0], K[i+0] );
+        P( A[7], A[0], A[1], A[2], A[3], A[4], A[5], A[6], W[i+1], K[i+1] );
+        P( A[6], A[7], A[0], A[1], A[2], A[3], A[4], A[5], W[i+2], K[i+2] );
+        P( A[5], A[6], A[7], A[0], A[1], A[2], A[3], A[4], W[i+3], K[i+3] );
+        P( A[4], A[5], A[6], A[7], A[0], A[1], A[2], A[3], W[i+4], K[i+4] );
+        P( A[3], A[4], A[5], A[6], A[7], A[0], A[1], A[2], W[i+5], K[i+5] );
+        P( A[2], A[3], A[4], A[5], A[6], A[7], A[0], A[1], W[i+6], K[i+6] );
+        P( A[1], A[2], A[3], A[4], A[5], A[6], A[7], A[0], W[i+7], K[i+7] );
+    }
+
+    for( i = 16; i < 64; i += 8 )
+    {
+        P( A[0], A[1], A[2], A[3], A[4], A[5], A[6], A[7], R(i+0), K[i+0] );
+        P( A[7], A[0], A[1], A[2], A[3], A[4], A[5], A[6], R(i+1), K[i+1] );
+        P( A[6], A[7], A[0], A[1], A[2], A[3], A[4], A[5], R(i+2), K[i+2] );
+        P( A[5], A[6], A[7], A[0], A[1], A[2], A[3], A[4], R(i+3), K[i+3] );
+        P( A[4], A[5], A[6], A[7], A[0], A[1], A[2], A[3], R(i+4), K[i+4] );
+        P( A[3], A[4], A[5], A[6], A[7], A[0], A[1], A[2], R(i+5), K[i+5] );
+        P( A[2], A[3], A[4], A[5], A[6], A[7], A[0], A[1], R(i+6), K[i+6] );
+        P( A[1], A[2], A[3], A[4], A[5], A[6], A[7], A[0], R(i+7), K[i+7] );
+    }
+#endif /* MBEDTLS_SHA256_SMALLER */
+
+    for( i = 0; i < 8; i++ )
+        ctx->state[i] += A[i];
+}
+#endif /* !MBEDTLS_SHA256_PROCESS_ALT */
+
+/*
+ * SHA-256 process buffer
+ */
+void mbedtls_sha256_update( mbedtls_sha256_context *ctx, const unsigned char *input,
+                    size_t ilen )
+{
+    size_t fill;
+    uint32_t left;
+
+    if( ilen == 0 )
+        return;
+
+    left = ctx->total[0] & 0x3F;
+    fill = 64 - left;
+
+    ctx->total[0] += (uint32_t) ilen;
+    ctx->total[0] &= 0xFFFFFFFF;
+
+    if( ctx->total[0] < (uint32_t) ilen )
+        ctx->total[1]++;
+
+    if( left && ilen >= fill )
+    {
+        memcpy( (void *) (ctx->buffer + left), input, fill );
+        mbedtls_sha256_process( ctx, ctx->buffer );
+        input += fill;
+        ilen  -= fill;
+        left = 0;
+    }
+
+    while( ilen >= 64 )
+    {
+        mbedtls_sha256_process( ctx, input );
+        input += 64;
+        ilen  -= 64;
+    }
+
+    if( ilen > 0 )
+        memcpy( (void *) (ctx->buffer + left), input, ilen );
+}
+
+static const unsigned char sha256_padding[64] =
+{
+ 0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
+};
+
+/*
+ * SHA-256 final digest
+ */
+void mbedtls_sha256_finish( mbedtls_sha256_context *ctx, unsigned char output[32] )
+{
+    uint32_t last, padn;
+    uint32_t high, low;
+    unsigned char msglen[8];
+
+    high = ( ctx->total[0] >> 29 )
+         | ( ctx->total[1] <<  3 );
+    low  = ( ctx->total[0] <<  3 );
+
+    PUT_UINT32_BE( high, msglen, 0 );
+    PUT_UINT32_BE( low,  msglen, 4 );
+
+    last = ctx->total[0] & 0x3F;
+    padn = ( last < 56 ) ? ( 56 - last ) : ( 120 - last );
+
+    mbedtls_sha256_update( ctx, sha256_padding, padn );
+    mbedtls_sha256_update( ctx, msglen, 8 );
+
+    /* if state is in hardware, read it out */
+    if (ctx->mode == ESP_MBEDTLS_SHA256_HARDWARE) {
+        esp_sha_read_digest_state(SHA2_256, ctx->state);
+        esp_sha_unlock_engine(SHA2_256);
+        ctx->mode = ESP_MBEDTLS_SHA256_SOFTWARE;
+    }
+
+    PUT_UINT32_BE( ctx->state[0], output,  0 );
+    PUT_UINT32_BE( ctx->state[1], output,  4 );
+    PUT_UINT32_BE( ctx->state[2], output,  8 );
+    PUT_UINT32_BE( ctx->state[3], output, 12 );
+    PUT_UINT32_BE( ctx->state[4], output, 16 );
+    PUT_UINT32_BE( ctx->state[5], output, 20 );
+    PUT_UINT32_BE( ctx->state[6], output, 24 );
+
+    if( ctx->is224 == 0 )
+        PUT_UINT32_BE( ctx->state[7], output, 28 );
+}
+
+#endif /* MBEDTLS_SHA256_C && MBEDTLS_SHA256_ALT */
diff --git a/components/mbedtls/port/esp_sha512.c b/components/mbedtls/port/esp_sha512.c
new file mode 100644
index 000000000..cfd0f3fdf
--- /dev/null
+++ b/components/mbedtls/port/esp_sha512.c
@@ -0,0 +1,408 @@
+/*
+ *  SHA-512 implementation with hardware ESP32 support added.
+ *  Uses mbedTLS software implementation for failover when concurrent
+ *  SHA operations are in use.
+ *
+ *  Copyright (C) 2006-2015, ARM Limited, All Rights Reserved
+ *  Additions Copyright (C) 2016, Espressif Systems (Shanghai) PTE LTD
+ *  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
+ *
+ *  http://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.
+ *
+ */
+
+/*
+ *  The SHA-512 Secure Hash Standard was published by NIST in 2002.
+ *
+ *  http://csrc.nist.gov/publications/fips/fips180-2/fips180-2.pdf
+ */
+
+#if !defined(MBEDTLS_CONFIG_FILE)
+#include "mbedtls/config.h"
+#else
+#include MBEDTLS_CONFIG_FILE
+#endif
+
+#if defined(MBEDTLS_SHA512_C) && defined(MBEDTLS_SHA512_ALT)
+
+#include "mbedtls/sha512.h"
+
+#if defined(_MSC_VER) || defined(__WATCOMC__)
+  #define UL64(x) x##ui64
+#else
+  #define UL64(x) x##ULL
+#endif
+
+#include <string.h>
+
+#if defined(MBEDTLS_SELF_TEST)
+#if defined(MBEDTLS_PLATFORM_C)
+#include "mbedtls/platform.h"
+#else
+#include <stdio.h>
+#define mbedtls_printf printf
+#endif /* MBEDTLS_PLATFORM_C */
+#endif /* MBEDTLS_SELF_TEST */
+
+#include "hwcrypto/sha.h"
+
+inline static esp_sha_type sha_type(const mbedtls_sha512_context *ctx)
+{
+    return ctx->is384 ? SHA2_384 : SHA2_512;
+}
+
+/* Implementation that should never be optimized out by the compiler */
+static void mbedtls_zeroize( void *v, size_t n ) {
+    volatile unsigned char *p = v; while( n-- ) *p++ = 0;
+}
+
+/*
+ * 64-bit integer manipulation macros (big endian)
+ */
+#ifndef GET_UINT64_BE
+#define GET_UINT64_BE(n,b,i)                            \
+{                                                       \
+    (n) = ( (uint64_t) (b)[(i)    ] << 56 )       \
+        | ( (uint64_t) (b)[(i) + 1] << 48 )       \
+        | ( (uint64_t) (b)[(i) + 2] << 40 )       \
+        | ( (uint64_t) (b)[(i) + 3] << 32 )       \
+        | ( (uint64_t) (b)[(i) + 4] << 24 )       \
+        | ( (uint64_t) (b)[(i) + 5] << 16 )       \
+        | ( (uint64_t) (b)[(i) + 6] <<  8 )       \
+        | ( (uint64_t) (b)[(i) + 7]       );      \
+}
+#endif /* GET_UINT64_BE */
+
+#ifndef PUT_UINT64_BE
+#define PUT_UINT64_BE(n,b,i)                            \
+{                                                       \
+    (b)[(i)    ] = (unsigned char) ( (n) >> 56 );       \
+    (b)[(i) + 1] = (unsigned char) ( (n) >> 48 );       \
+    (b)[(i) + 2] = (unsigned char) ( (n) >> 40 );       \
+    (b)[(i) + 3] = (unsigned char) ( (n) >> 32 );       \
+    (b)[(i) + 4] = (unsigned char) ( (n) >> 24 );       \
+    (b)[(i) + 5] = (unsigned char) ( (n) >> 16 );       \
+    (b)[(i) + 6] = (unsigned char) ( (n) >>  8 );       \
+    (b)[(i) + 7] = (unsigned char) ( (n)       );       \
+}
+#endif /* PUT_UINT64_BE */
+
+void mbedtls_sha512_init( mbedtls_sha512_context *ctx )
+{
+    memset( ctx, 0, sizeof( mbedtls_sha512_context ) );
+}
+
+void mbedtls_sha512_free( mbedtls_sha512_context *ctx )
+{
+    if( ctx == NULL )
+        return;
+
+    if (ctx->mode == ESP_MBEDTLS_SHA512_HARDWARE) {
+        esp_sha_unlock_engine(sha_type(ctx));
+    }
+    mbedtls_zeroize( ctx, sizeof( mbedtls_sha512_context ) );
+}
+
+void mbedtls_sha512_clone( mbedtls_sha512_context *dst,
+                           const mbedtls_sha512_context *src )
+{
+    *dst = *src;
+
+    if (src->mode == ESP_MBEDTLS_SHA512_HARDWARE) {
+        /* Copy hardware digest state out to cloned state,
+           which will be a software digest.
+        */
+        esp_sha_read_digest_state(sha_type(dst), dst->state);
+        dst->mode = ESP_MBEDTLS_SHA512_SOFTWARE;
+    }
+}
+
+
+/*
+ * SHA-512 context setup
+ */
+void mbedtls_sha512_starts( mbedtls_sha512_context *ctx, int is384 )
+{
+    ctx->total[0] = 0;
+    ctx->total[1] = 0;
+
+    if( is384 == 0 )
+    {
+        /* SHA-512 */
+        ctx->state[0] = UL64(0x6A09E667F3BCC908);
+        ctx->state[1] = UL64(0xBB67AE8584CAA73B);
+        ctx->state[2] = UL64(0x3C6EF372FE94F82B);
+        ctx->state[3] = UL64(0xA54FF53A5F1D36F1);
+        ctx->state[4] = UL64(0x510E527FADE682D1);
+        ctx->state[5] = UL64(0x9B05688C2B3E6C1F);
+        ctx->state[6] = UL64(0x1F83D9ABFB41BD6B);
+        ctx->state[7] = UL64(0x5BE0CD19137E2179);
+    }
+    else
+    {
+        /* SHA-384 */
+        ctx->state[0] = UL64(0xCBBB9D5DC1059ED8);
+        ctx->state[1] = UL64(0x629A292A367CD507);
+        ctx->state[2] = UL64(0x9159015A3070DD17);
+        ctx->state[3] = UL64(0x152FECD8F70E5939);
+        ctx->state[4] = UL64(0x67332667FFC00B31);
+        ctx->state[5] = UL64(0x8EB44A8768581511);
+        ctx->state[6] = UL64(0xDB0C2E0D64F98FA7);
+        ctx->state[7] = UL64(0x47B5481DBEFA4FA4);
+    }
+
+    ctx->is384 = is384;
+    if (ctx->mode == ESP_MBEDTLS_SHA512_HARDWARE) {
+        esp_sha_unlock_engine(sha_type(ctx));
+    }
+    ctx->mode = ESP_MBEDTLS_SHA512_UNUSED;
+}
+
+
+/*
+ * Round constants
+ */
+static const uint64_t K[80] =
+{
+    UL64(0x428A2F98D728AE22),  UL64(0x7137449123EF65CD),
+    UL64(0xB5C0FBCFEC4D3B2F),  UL64(0xE9B5DBA58189DBBC),
+    UL64(0x3956C25BF348B538),  UL64(0x59F111F1B605D019),
+    UL64(0x923F82A4AF194F9B),  UL64(0xAB1C5ED5DA6D8118),
+    UL64(0xD807AA98A3030242),  UL64(0x12835B0145706FBE),
+    UL64(0x243185BE4EE4B28C),  UL64(0x550C7DC3D5FFB4E2),
+    UL64(0x72BE5D74F27B896F),  UL64(0x80DEB1FE3B1696B1),
+    UL64(0x9BDC06A725C71235),  UL64(0xC19BF174CF692694),
+    UL64(0xE49B69C19EF14AD2),  UL64(0xEFBE4786384F25E3),
+    UL64(0x0FC19DC68B8CD5B5),  UL64(0x240CA1CC77AC9C65),
+    UL64(0x2DE92C6F592B0275),  UL64(0x4A7484AA6EA6E483),
+    UL64(0x5CB0A9DCBD41FBD4),  UL64(0x76F988DA831153B5),
+    UL64(0x983E5152EE66DFAB),  UL64(0xA831C66D2DB43210),
+    UL64(0xB00327C898FB213F),  UL64(0xBF597FC7BEEF0EE4),
+    UL64(0xC6E00BF33DA88FC2),  UL64(0xD5A79147930AA725),
+    UL64(0x06CA6351E003826F),  UL64(0x142929670A0E6E70),
+    UL64(0x27B70A8546D22FFC),  UL64(0x2E1B21385C26C926),
+    UL64(0x4D2C6DFC5AC42AED),  UL64(0x53380D139D95B3DF),
+    UL64(0x650A73548BAF63DE),  UL64(0x766A0ABB3C77B2A8),
+    UL64(0x81C2C92E47EDAEE6),  UL64(0x92722C851482353B),
+    UL64(0xA2BFE8A14CF10364),  UL64(0xA81A664BBC423001),
+    UL64(0xC24B8B70D0F89791),  UL64(0xC76C51A30654BE30),
+    UL64(0xD192E819D6EF5218),  UL64(0xD69906245565A910),
+    UL64(0xF40E35855771202A),  UL64(0x106AA07032BBD1B8),
+    UL64(0x19A4C116B8D2D0C8),  UL64(0x1E376C085141AB53),
+    UL64(0x2748774CDF8EEB99),  UL64(0x34B0BCB5E19B48A8),
+    UL64(0x391C0CB3C5C95A63),  UL64(0x4ED8AA4AE3418ACB),
+    UL64(0x5B9CCA4F7763E373),  UL64(0x682E6FF3D6B2B8A3),
+    UL64(0x748F82EE5DEFB2FC),  UL64(0x78A5636F43172F60),
+    UL64(0x84C87814A1F0AB72),  UL64(0x8CC702081A6439EC),
+    UL64(0x90BEFFFA23631E28),  UL64(0xA4506CEBDE82BDE9),
+    UL64(0xBEF9A3F7B2C67915),  UL64(0xC67178F2E372532B),
+    UL64(0xCA273ECEEA26619C),  UL64(0xD186B8C721C0C207),
+    UL64(0xEADA7DD6CDE0EB1E),  UL64(0xF57D4F7FEE6ED178),
+    UL64(0x06F067AA72176FBA),  UL64(0x0A637DC5A2C898A6),
+    UL64(0x113F9804BEF90DAE),  UL64(0x1B710B35131C471B),
+    UL64(0x28DB77F523047D84),  UL64(0x32CAAB7B40C72493),
+    UL64(0x3C9EBE0A15C9BEBC),  UL64(0x431D67C49C100D4C),
+    UL64(0x4CC5D4BECB3E42B6),  UL64(0x597F299CFC657E2A),
+    UL64(0x5FCB6FAB3AD6FAEC),  UL64(0x6C44198C4A475817)
+};
+
+static void mbedtls_sha512_software_process( mbedtls_sha512_context *ctx, const unsigned char data[128] );
+
+void mbedtls_sha512_process( mbedtls_sha512_context *ctx, const unsigned char data[128] )
+{
+    bool first_block = false;
+
+    if (ctx->mode == ESP_MBEDTLS_SHA512_UNUSED) {
+        /* try to use hardware for this digest */
+        if (esp_sha_try_lock_engine(sha_type(ctx))) {
+            ctx->mode = ESP_MBEDTLS_SHA512_HARDWARE;
+            first_block = true;
+        } else {
+            ctx->mode = ESP_MBEDTLS_SHA512_SOFTWARE;
+        }
+    }
+
+    if (ctx->mode == ESP_MBEDTLS_SHA512_HARDWARE) {
+        esp_sha_block(sha_type(ctx), data, first_block);
+    } else {
+        mbedtls_sha512_software_process(ctx, data);
+    }
+}
+
+
+
+static void mbedtls_sha512_software_process( mbedtls_sha512_context *ctx, const unsigned char data[128] )
+{
+    int i;
+    uint64_t temp1, temp2, W[80];
+    uint64_t A, B, C, D, E, F, G, H;
+
+#define  SHR(x,n) (x >> n)
+#define ROTR(x,n) (SHR(x,n) | (x << (64 - n)))
+
+#define S0(x) (ROTR(x, 1) ^ ROTR(x, 8) ^  SHR(x, 7))
+#define S1(x) (ROTR(x,19) ^ ROTR(x,61) ^  SHR(x, 6))
+
+#define S2(x) (ROTR(x,28) ^ ROTR(x,34) ^ ROTR(x,39))
+#define S3(x) (ROTR(x,14) ^ ROTR(x,18) ^ ROTR(x,41))
+
+#define F0(x,y,z) ((x & y) | (z & (x | y)))
+#define F1(x,y,z) (z ^ (x & (y ^ z)))
+
+#define P(a,b,c,d,e,f,g,h,x,K)                  \
+{                                               \
+    temp1 = h + S3(e) + F1(e,f,g) + K + x;      \
+    temp2 = S2(a) + F0(a,b,c);                  \
+    d += temp1; h = temp1 + temp2;              \
+}
+
+    for( i = 0; i < 16; i++ )
+    {
+        GET_UINT64_BE( W[i], data, i << 3 );
+    }
+
+    for( ; i < 80; i++ )
+    {
+        W[i] = S1(W[i -  2]) + W[i -  7] +
+               S0(W[i - 15]) + W[i - 16];
+    }
+
+    A = ctx->state[0];
+    B = ctx->state[1];
+    C = ctx->state[2];
+    D = ctx->state[3];
+    E = ctx->state[4];
+    F = ctx->state[5];
+    G = ctx->state[6];
+    H = ctx->state[7];
+    i = 0;
+
+    do
+    {
+        P( A, B, C, D, E, F, G, H, W[i], K[i] ); i++;
+        P( H, A, B, C, D, E, F, G, W[i], K[i] ); i++;
+        P( G, H, A, B, C, D, E, F, W[i], K[i] ); i++;
+        P( F, G, H, A, B, C, D, E, W[i], K[i] ); i++;
+        P( E, F, G, H, A, B, C, D, W[i], K[i] ); i++;
+        P( D, E, F, G, H, A, B, C, W[i], K[i] ); i++;
+        P( C, D, E, F, G, H, A, B, W[i], K[i] ); i++;
+        P( B, C, D, E, F, G, H, A, W[i], K[i] ); i++;
+    }
+    while( i < 80 );
+
+    ctx->state[0] += A;
+    ctx->state[1] += B;
+    ctx->state[2] += C;
+    ctx->state[3] += D;
+    ctx->state[4] += E;
+    ctx->state[5] += F;
+    ctx->state[6] += G;
+    ctx->state[7] += H;
+}
+
+/*
+ * SHA-512 process buffer
+ */
+void mbedtls_sha512_update( mbedtls_sha512_context *ctx, const unsigned char *input,
+                    size_t ilen )
+{
+    size_t fill;
+    unsigned int left;
+
+    if( ilen == 0 )
+        return;
+
+    left = (unsigned int) (ctx->total[0] & 0x7F);
+    fill = 128 - left;
+
+    ctx->total[0] += (uint64_t) ilen;
+
+    if( ctx->total[0] < (uint64_t) ilen )
+        ctx->total[1]++;
+
+    if( left && ilen >= fill )
+    {
+        memcpy( (void *) (ctx->buffer + left), input, fill );
+        mbedtls_sha512_process( ctx, ctx->buffer );
+        input += fill;
+        ilen  -= fill;
+        left = 0;
+    }
+
+    while( ilen >= 128 )
+    {
+        mbedtls_sha512_process( ctx, input );
+        input += 128;
+        ilen  -= 128;
+    }
+
+    if( ilen > 0 )
+        memcpy( (void *) (ctx->buffer + left), input, ilen );
+}
+
+static const unsigned char sha512_padding[128] =
+{
+ 0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
+};
+
+/*
+ * SHA-512 final digest
+ */
+void mbedtls_sha512_finish( mbedtls_sha512_context *ctx, unsigned char output[64] )
+{
+    size_t last, padn;
+    uint64_t high, low;
+    unsigned char msglen[16];
+
+    high = ( ctx->total[0] >> 61 )
+         | ( ctx->total[1] <<  3 );
+    low  = ( ctx->total[0] <<  3 );
+
+    PUT_UINT64_BE( high, msglen, 0 );
+    PUT_UINT64_BE( low,  msglen, 8 );
+
+    last = (size_t)( ctx->total[0] & 0x7F );
+    padn = ( last < 112 ) ? ( 112 - last ) : ( 240 - last );
+
+    mbedtls_sha512_update( ctx, sha512_padding, padn );
+    mbedtls_sha512_update( ctx, msglen, 16 );
+
+    /* if state is in hardware, read it out */
+    if (ctx->mode == ESP_MBEDTLS_SHA512_HARDWARE) {
+        esp_sha_read_digest_state(sha_type(ctx), ctx->state);
+        esp_sha_unlock_engine(sha_type(ctx));
+        ctx->mode = ESP_MBEDTLS_SHA512_SOFTWARE;
+    }
+
+    PUT_UINT64_BE( ctx->state[0], output,  0 );
+    PUT_UINT64_BE( ctx->state[1], output,  8 );
+    PUT_UINT64_BE( ctx->state[2], output, 16 );
+    PUT_UINT64_BE( ctx->state[3], output, 24 );
+    PUT_UINT64_BE( ctx->state[4], output, 32 );
+    PUT_UINT64_BE( ctx->state[5], output, 40 );
+
+    if( ctx->is384 == 0 )
+    {
+        PUT_UINT64_BE( ctx->state[6], output, 48 );
+        PUT_UINT64_BE( ctx->state[7], output, 56 );
+    }
+}
+
+#endif /* MBEDTLS_SHA512_C && MBEDTLS_SHA512_ALT */
diff --git a/components/mbedtls/port/include/mbedtls/esp_config.h b/components/mbedtls/port/include/mbedtls/esp_config.h
index db87c6ef3..cd9f87e69 100644
--- a/components/mbedtls/port/include/mbedtls/esp_config.h
+++ b/components/mbedtls/port/include/mbedtls/esp_config.h
@@ -243,11 +243,14 @@
 #define MBEDTLS_AES_ALT
 #endif
 
-/* Currently hardware SHA does not work with TLS handshake,
-   due to concurrency issue. Internal TW#7111. */
-//#define MBEDTLS_SHA1_ALT
-//#define MBEDTLS_SHA256_ALT
-//#define MBEDTLS_SHA512_ALT
+/* MBEDTLS_SHAxx_ALT to enable hardware SHA support
+   with software fallback.
+*/
+#ifdef CONFIG_MBEDTLS_HARDWARE_SHA
+#define MBEDTLS_SHA1_ALT
+#define MBEDTLS_SHA256_ALT
+#define MBEDTLS_SHA512_ALT
+#endif
 
 /* The following MPI (bignum) functions have ESP32 hardware support,
    Uncommenting these macros will use the hardware-accelerated
diff --git a/components/mbedtls/port/include/sha1_alt.h b/components/mbedtls/port/include/sha1_alt.h
index f5e69b3f9..fbe740c7e 100644
--- a/components/mbedtls/port/include/sha1_alt.h
+++ b/components/mbedtls/port/include/sha1_alt.h
@@ -1,16 +1,25 @@
-// Copyright 2015-2016 Espressif Systems (Shanghai) PTE LTD
-//
-// 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
-
-//     http://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.
+/*
+ *  SHA-1 implementation with hardware ESP32 support added.
+ *  Uses mbedTLS software implementation for failover when concurrent
+ *  SHA operations are in use.
+ *
+ *  Copyright (C) 2006-2015, ARM Limited, All Rights Reserved
+ *  Additions Copyright (C) 2016, Espressif Systems (Shanghai) PTE LTD
+ *  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
+ *
+ *  http://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 _SHA1_ALT_H_
 #define _SHA1_ALT_H_
 
@@ -20,17 +29,73 @@ extern "C" {
 
 #if defined(MBEDTLS_SHA1_ALT)
 
-#include "hwcrypto/sha.h"
+typedef enum {
+    ESP_MBEDTLS_SHA1_UNUSED, /* first block hasn't been processed yet */
+    ESP_MBEDTLS_SHA1_HARDWARE, /* using hardware SHA engine */
+    ESP_MBEDTLS_SHA1_SOFTWARE, /* using software SHA */
+} esp_mbedtls_sha1_mode;
 
-typedef esp_sha_context mbedtls_sha1_context;
+/**
+ * \brief          SHA-1 context structure
+ */
+typedef struct
+{
+    uint32_t total[2];          /*!< number of bytes processed  */
+    uint32_t state[5];          /*!< intermediate digest state  */
+    unsigned char buffer[64];   /*!< data block being processed */
+    esp_mbedtls_sha1_mode mode;
+}
+mbedtls_sha1_context;
 
-#define mbedtls_sha1_init       esp_sha1_init
-#define mbedtls_sha1_starts     esp_sha1_start
-#define mbedtls_sha1_clone      esp_sha1_clone
-#define mbedtls_sha1_update     esp_sha1_update
-#define mbedtls_sha1_finish     esp_sha1_finish
-#define mbedtls_sha1_free       esp_sha1_free
-#define mbedtls_sha1_process(...)
+/**
+ * \brief          Initialize SHA-1 context
+ *
+ * \param ctx      SHA-1 context to be initialized
+ */
+void mbedtls_sha1_init( mbedtls_sha1_context *ctx );
+
+/**
+ * \brief          Clear SHA-1 context
+ *
+ * \param ctx      SHA-1 context to be cleared
+ */
+void mbedtls_sha1_free( mbedtls_sha1_context *ctx );
+
+/**
+ * \brief          Clone (the state of) a SHA-1 context
+ *
+ * \param dst      The destination context
+ * \param src      The context to be cloned
+ */
+void mbedtls_sha1_clone( mbedtls_sha1_context *dst,
+                         const mbedtls_sha1_context *src );
+
+/**
+ * \brief          SHA-1 context setup
+ *
+ * \param ctx      context to be initialized
+ */
+void mbedtls_sha1_starts( mbedtls_sha1_context *ctx );
+
+/**
+ * \brief          SHA-1 process buffer
+ *
+ * \param ctx      SHA-1 context
+ * \param input    buffer holding the  data
+ * \param ilen     length of the input data
+ */
+void mbedtls_sha1_update( mbedtls_sha1_context *ctx, const unsigned char *input, size_t ilen );
+
+/**
+ * \brief          SHA-1 final digest
+ *
+ * \param ctx      SHA-1 context
+ * \param output   SHA-1 checksum result
+ */
+void mbedtls_sha1_finish( mbedtls_sha1_context *ctx, unsigned char output[20] );
+
+/* Internal use */
+void mbedtls_sha1_process( mbedtls_sha1_context *ctx, const unsigned char data[64] );
 
 #endif
 
diff --git a/components/mbedtls/port/include/sha256_alt.h b/components/mbedtls/port/include/sha256_alt.h
index 143d8c75e..cc87333aa 100644
--- a/components/mbedtls/port/include/sha256_alt.h
+++ b/components/mbedtls/port/include/sha256_alt.h
@@ -1,16 +1,25 @@
-// Copyright 2015-2016 Espressif Systems (Shanghai) PTE LTD
-//
-// 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
-
-//     http://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.
+/*
+ *  SHA-256 implementation with hardware ESP32 support added.
+ *  Uses mbedTLS software implementation for failover when concurrent
+ *  SHA operations are in use.
+ *
+ *  Copyright (C) 2006-2015, ARM Limited, All Rights Reserved
+ *  Additions Copyright (C) 2016, Espressif Systems (Shanghai) PTE LTD
+ *  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
+ *
+ *  http://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 _SHA256_ALT_H_
 #define _SHA256_ALT_H_
 
@@ -20,17 +29,76 @@ extern "C" {
 
 #if defined(MBEDTLS_SHA256_ALT)
 
-#include "hwcrypto/sha.h"
+typedef enum {
+    ESP_MBEDTLS_SHA256_UNUSED, /* first block hasn't been processed yet */
+    ESP_MBEDTLS_SHA256_HARDWARE, /* using hardware SHA engine */
+    ESP_MBEDTLS_SHA256_SOFTWARE, /* using software SHA */
+} esp_mbedtls_sha256_mode;
 
-typedef esp_sha_context mbedtls_sha256_context;
+/**
+ * \brief          SHA-256 context structure
+ */
+typedef struct
+{
+    uint32_t total[2];          /*!< number of bytes processed  */
+    uint32_t state[8];          /*!< intermediate digest state  */
+    unsigned char buffer[64];   /*!< data block being processed */
+    int is224;                  /*!< 0 => SHA-256, else SHA-224 */
+    esp_mbedtls_sha256_mode mode;
+}
+mbedtls_sha256_context;
 
-#define mbedtls_sha256_init     esp_sha256_init
-#define mbedtls_sha256_clone    esp_sha256_clone
-#define mbedtls_sha256_starts   esp_sha256_start
-#define mbedtls_sha256_update   esp_sha256_update
-#define mbedtls_sha256_finish   esp_sha256_finish
-#define mbedtls_sha256_free     esp_sha256_free
-#define mbedtls_sha256_process(...)
+/**
+ * \brief          Initialize SHA-256 context
+ *
+ * \param ctx      SHA-256 context to be initialized
+ */
+void mbedtls_sha256_init( mbedtls_sha256_context *ctx );
+
+/**
+ * \brief          Clear SHA-256 context
+ *
+ * \param ctx      SHA-256 context to be cleared
+ */
+void mbedtls_sha256_free( mbedtls_sha256_context *ctx );
+
+/**
+ * \brief          Clone (the state of) a SHA-256 context
+ *
+ * \param dst      The destination context
+ * \param src      The context to be cloned
+ */
+void mbedtls_sha256_clone( mbedtls_sha256_context *dst,
+                           const mbedtls_sha256_context *src );
+
+/**
+ * \brief          SHA-256 context setup
+ *
+ * \param ctx      context to be initialized
+ * \param is224    0 = use SHA256, 1 = use SHA224
+ */
+void mbedtls_sha256_starts( mbedtls_sha256_context *ctx, int is224 );
+
+/**
+ * \brief          SHA-256 process buffer
+ *
+ * \param ctx      SHA-256 context
+ * \param input    buffer holding the  data
+ * \param ilen     length of the input data
+ */
+void mbedtls_sha256_update( mbedtls_sha256_context *ctx, const unsigned char *input,
+                    size_t ilen );
+
+/**
+ * \brief          SHA-256 final digest
+ *
+ * \param ctx      SHA-256 context
+ * \param output   SHA-224/256 checksum result
+ */
+void mbedtls_sha256_finish( mbedtls_sha256_context *ctx, unsigned char output[32] );
+
+/* Internal use */
+void mbedtls_sha256_process( mbedtls_sha256_context *ctx, const unsigned char data[64] );
 
 #endif
 
diff --git a/components/mbedtls/port/include/sha512_alt.h b/components/mbedtls/port/include/sha512_alt.h
index 8044b4275..70ae24e16 100644
--- a/components/mbedtls/port/include/sha512_alt.h
+++ b/components/mbedtls/port/include/sha512_alt.h
@@ -1,16 +1,25 @@
-// Copyright 2015-2016 Espressif Systems (Shanghai) PTE LTD
-//
-// 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
-
-//     http://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.
+/*
+ *  SHA-512 implementation with hardware ESP32 support added.
+ *  Uses mbedTLS software implementation for failover when concurrent
+ *  SHA operations are in use.
+ *
+ *  Copyright (C) 2006-2015, ARM Limited, All Rights Reserved
+ *  Additions Copyright (C) 2016, Espressif Systems (Shanghai) PTE LTD
+ *  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
+ *
+ *  http://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 _SHA512_ALT_H_
 #define _SHA512_ALT_H_
 
@@ -19,17 +28,77 @@ extern "C" {
 #endif
 
 #if defined(MBEDTLS_SHA512_ALT)
-#include "hwcrypto/sha.h"
 
-typedef esp_sha_context  mbedtls_sha512_context;
+typedef enum {
+    ESP_MBEDTLS_SHA512_UNUSED, /* first block hasn't been processed yet */
+    ESP_MBEDTLS_SHA512_HARDWARE, /* using hardware SHA engine */
+    ESP_MBEDTLS_SHA512_SOFTWARE, /* using software SHA */
+} esp_mbedtls_sha512_mode;
 
-#define mbedtls_sha512_init     esp_sha512_init
-#define mbedtls_sha512_clone    esp_sha512_clone
-#define mbedtls_sha512_starts   esp_sha512_start
-#define mbedtls_sha512_update   esp_sha512_update
-#define mbedtls_sha512_finish   esp_sha512_finish
-#define mbedtls_sha512_free     esp_sha512_free
-#define mbedtls_sha512_process(...)
+/**
+ * \brief          SHA-512 context structure
+ */
+typedef struct
+{
+    uint64_t total[2];          /*!< number of bytes processed  */
+    uint64_t state[8];          /*!< intermediate digest state  */
+    unsigned char buffer[128];  /*!< data block being processed */
+    int is384;                  /*!< 0 => SHA-512, else SHA-384 */
+    esp_mbedtls_sha512_mode mode;
+}
+mbedtls_sha512_context;
+
+/**
+ * \brief          Initialize SHA-512 context
+ *
+ * \param ctx      SHA-512 context to be initialized
+ */
+void mbedtls_sha512_init( mbedtls_sha512_context *ctx );
+
+/**
+ * \brief          Clear SHA-512 context
+ *
+ * \param ctx      SHA-512 context to be cleared
+ */
+void mbedtls_sha512_free( mbedtls_sha512_context *ctx );
+
+/**
+ * \brief          Clone (the state of) a SHA-512 context
+ *
+ * \param dst      The destination context
+ * \param src      The context to be cloned
+ */
+void mbedtls_sha512_clone( mbedtls_sha512_context *dst,
+                           const mbedtls_sha512_context *src );
+
+/**
+ * \brief          SHA-512 context setup
+ *
+ * \param ctx      context to be initialized
+ * \param is384    0 = use SHA512, 1 = use SHA384
+ */
+void mbedtls_sha512_starts( mbedtls_sha512_context *ctx, int is384 );
+
+/**
+ * \brief          SHA-512 process buffer
+ *
+ * \param ctx      SHA-512 context
+ * \param input    buffer holding the  data
+ * \param ilen     length of the input data
+ */
+void mbedtls_sha512_update( mbedtls_sha512_context *ctx, const unsigned char *input,
+                    size_t ilen );
+
+/**
+ * \brief          SHA-512 final digest
+ *
+ * \param ctx      SHA-512 context
+ * \param output   SHA-384/512 checksum result
+ */
+void mbedtls_sha512_finish( mbedtls_sha512_context *ctx, unsigned char output[64] );
+
+/* Internal use */
+void mbedtls_sha512_process( mbedtls_sha512_context *ctx, const unsigned char data[128] );
 
 #endif
 
diff --git a/docs/build_system.rst b/docs/build_system.rst
index 4ba89f865..ad263da7a 100644
--- a/docs/build_system.rst
+++ b/docs/build_system.rst
@@ -244,6 +244,7 @@ The following variables can be set inside ``component.mk`` to control the build
   settings. Component-specific additions can be made via ``CXXFLAGS
   +=``. It is also possible (although not recommended) to override
   this variable completely for a component.
+- ``FLAGS_basename`` allows you to set compilation flags to apply to a single source file only. For example, this can useful for disabling warnings in a single upstream source file. The ``basename`` portion is the directory (relative to ``COMPONENT_PATH``) and the base filename (without extension) of the source file. For example, if a file inside ``COMPONENT_PATH`` is ``library/alpha/widget.c`` then you can set variable ``FLAGS_library/alpha/widget := -DTEST`` to pass the TEST macro when compiling this source file only.
 
 Component Configuration
 -----------------------
diff --git a/make/component_wrapper.mk b/make/component_wrapper.mk
index c6ea52a2b..63f687d93 100644
--- a/make/component_wrapper.mk
+++ b/make/component_wrapper.mk
@@ -155,17 +155,17 @@ endif
 # This pattern is generated for each COMPONENT_SRCDIR to compile the files in it.
 define GenerateCompileTargets
 # $(1) - directory containing source files, relative to $(COMPONENT_PATH)
-$(1)/%.o: $$(COMPONENT_PATH)/$(1)/%.c | $(1)
+$(1)/%.o: $$(COMPONENT_PATH)/$(1)/%.c $(COMMON_MAKEFILES) $(COMPONENT_MAKEFILE) | $(1)
 	$$(summary) CC $$@
-	$$(CC) $$(CFLAGS) $$(CPPFLAGS) $$(addprefix -I ,$$(COMPONENT_INCLUDES)) $$(addprefix -I ,$$(COMPONENT_EXTRA_INCLUDES)) -I$(1) -c $$< -o $$@
+	$$(CC) $$(CFLAGS) $$(CPPFLAGS) $$(FLAGS_$$(subst .o,,$$(@))) $$(addprefix -I ,$$(COMPONENT_INCLUDES)) $$(addprefix -I ,$$(COMPONENT_EXTRA_INCLUDES)) -I$(1) -c $$< -o $$@
 
-$(1)/%.o: $$(COMPONENT_PATH)/$(1)/%.cpp | $(1)
+$(1)/%.o: $$(COMPONENT_PATH)/$(1)/%.cpp $(COMMON_MAKEFILES) $(COMPONENT_MAKEFILE) | $(1)
 	$$(summary) CXX $$@
-	$$(CXX) $$(CXXFLAGS) $$(CPPFLAGS) $$(addprefix -I,$$(COMPONENT_INCLUDES)) $$(addprefix -I,$$(COMPONENT_EXTRA_INCLUDES)) -I$(1) -c $$< -o $$@
+	$$(CXX) $$(CXXFLAGS) $$(CPPFLAGS) $$(FLAGS_$$(subst .o,,$$(@))) $$(addprefix -I,$$(COMPONENT_INCLUDES)) $$(addprefix -I,$$(COMPONENT_EXTRA_INCLUDES)) -I$(1) -c $$< -o $$@
 
-$(1)/%.o: $$(COMPONENT_PATH)/$(1)/%.S | $(1)
+$(1)/%.o: $$(COMPONENT_PATH)/$(1)/%.S $(COMMON_MAKEFILES) $(COMPONENT_MAKEFILE) | $(1)
 	$$(summary) AS $$@
-	$$(CC) $$(CPPFLAGS) $$(addprefix -I ,$$(COMPONENT_INCLUDES)) $$(addprefix -I ,$$(COMPONENT_EXTRA_INCLUDES)) -I$(1) -c $$< -o $$@
+	$$(CC) $$(CPPFLAGS) $$(FLAGS_$$(subst .o,,$$(@))) $$(addprefix -I ,$$(COMPONENT_INCLUDES)) $$(addprefix -I ,$$(COMPONENT_EXTRA_INCLUDES)) -I$(1) -c $$< -o $$@
 
 # CWD is build dir, create the build subdirectory if it doesn't exist
 $(1):