Merge remote-tracking branch 'origin/master' into msc_stall

22 files changed, 140 insertions, 107 deletions

diff --git a/.github/pull_request_template.md b/.github/pull_request_template.md
deleted file mode 100644
index aa148eb79..000000000
--- a/.github/pull_request_template.md
+++ /dev/null
@@ -1,5 +0,0 @@
-**Describe the PR**
-A clear and concise description of what this PR solve.
-
-**Additional context**
-If applicable, add any other context about the PR and/or screenshots here.
diff --git a/.github/workflows/build.yml b/.github/workflows/build.yml
index e0c1d0a14..352875a9d 100644
--- a/.github/workflows/build.yml
+++ b/.github/workflows/build.yml
@@ -7,6 +7,7 @@ on:
   pull_request:
   release:
     types: [ published ]
+
 concurrency:
   group: ${{ github.workflow }}-${{ github.ref }}
   cancel-in-progress: true
@@ -15,9 +16,9 @@ env:
   HIL_JSON: test/hil/tinyusb.json
 
 jobs:
-  # Check if code paths changed (skip builds if doc-only)
+  # Check if the code changes and we need to run ci build
+  # Cannot use paths filter in the on-event since we want this workflow to run even when there are no code changes, to register the commit chain
   check-paths:
-    if: github.event_name == 'pull_request' || github.event_name == 'push'
     runs-on: ubuntu-latest
     permissions:
       contents: read
@@ -46,13 +47,8 @@ jobs:
               - '.github/workflows/ci_set_matrix.py'
 
   set-matrix:
-    needs: [check-paths]
-    if: |
-      always() && (
-        github.event_name == 'release' ||
-        github.event_name == 'workflow_dispatch' ||
-        needs.check-paths.outputs.code_changed == 'true'
-      )
+    needs: [ check-paths ]
+    if: needs.check-paths.outputs.code_changed == 'true'
     runs-on: ubuntu-latest
     outputs:
       json: ${{ steps.set-matrix-json.outputs.matrix }}
@@ -197,7 +193,7 @@ jobs:
   # Build Make/CMake on Windows/MacOS
   # ---------------------------------------
   build-os:
-    needs: [check-paths]
+    needs: [ check-paths ]
     if: needs.check-paths.outputs.code_changed == 'true'
     uses: ./.github/workflows/build_util.yml
     strategy:
@@ -215,8 +211,9 @@ jobs:
   # Zephyr
   # ---------------------------------------
   zephyr:
-    needs: [check-paths]
-    if: needs.check-paths.outputs.code_changed == 'true'
+    needs: [ check-paths ]
+    # skip zephyr build due to failed build, fix later
+    if: false && needs.check-paths.outputs.code_changed == 'true'
     runs-on: ubuntu-latest
     steps:
       - name: Checkout TinyUSB
@@ -238,10 +235,8 @@ jobs:
   # Run on PR only (hil-tinyusb), hil-hfp only run on non-forked PR
   # ---------------------------------------
   hil-build:
-    needs: [check-paths, set-matrix]
-    if: |
-      github.repository_owner == 'hathach' &&
-      (github.event_name == 'workflow_dispatch' || needs.check-paths.outputs.code_changed == 'true')
+    needs: set-matrix
+    if: github.repository_owner == 'hathach'
     uses: ./.github/workflows/build_util.yml
     strategy:
       fail-fast: false
@@ -260,10 +255,7 @@ jobs:
   # self-hosted on local VM, for attached hardware checkout HIL_JSON
   # ---------------------------------------
   hil-tinyusb:
-    needs: [check-paths, hil-build]
-    if: |
-      github.repository_owner == 'hathach' &&
-      (github.event_name == 'release' || github.event_name == 'workflow_dispatch' || needs.check-paths.outputs.code_changed == 'true')
+    needs: hil-build
     runs-on: [ self-hosted, X64, hathach, hardware-in-the-loop ]
     steps:
       - name: Get Skip Boards from previous run
@@ -303,11 +295,10 @@ jobs:
   # Since IAR Token secret is not passed to forked PR, only build non-forked PR
   # ---------------------------------------
   hil-hfp:
-    needs: [check-paths]
+    needs: [ check-paths ]
     if: |
       github.repository_owner == 'hathach' &&
-      github.event.pull_request.head.repo.fork == false &&
-      (github.event_name == 'release' || github.event_name == 'workflow_dispatch' || needs.check-paths.outputs.code_changed == 'true')
+      !(github.event_name == 'pull_request' && github.event.pull_request.head.repo.fork == true)
     runs-on: [ self-hosted, Linux, X64, hifiphile ]
     env:
       IAR_LMS_BEARER_TOKEN: ${{ secrets.IAR_LMS_BEARER_TOKEN }}
@@ -364,11 +355,9 @@ jobs:
     secrets: inherit
 
   membrowse-comment:
-    needs: [check-paths, membrowse]
-    if: >
-      always() &&
-      github.event_name == 'pull_request' &&
-      needs.check-paths.outputs.code_changed == 'true'
+    needs: membrowse
+    # skip membrowse comment since it is too verbal
+    if: false && github.event_name == 'pull_request'
     runs-on: ubuntu-latest
     permissions:
       contents: read
diff --git a/.github/workflows/ci_set_matrix.py b/.github/workflows/ci_set_matrix.py
index 9ab08601d..05a48b6c9 100755
--- a/.github/workflows/ci_set_matrix.py
+++ b/.github/workflows/ci_set_matrix.py
@@ -15,8 +15,8 @@ toolchain_list = [
 
 # family: [supported toolchain]
 family_list = {
-    "at32f45x at32f402_405 at32f403a_407 at32f413 at32f415 at32f423 at32f425 at32f435_437 broadcom_32bit da1469x": [
-        "arm-gcc"],
+    "at32f45x at32f402_405 at32f403a_407 at32f413 at32f415 at32f423 at32f425 at32f435_437": ["arm-gcc"],
+    "broadcom_32bit da1469x": ["arm-gcc"],
     "broadcom_64bit": ["aarch64-gcc"],
     "ch32v10x ch32v20x ch32v30x fomu gd32vf103 hpmicro": ["riscv-gcc"],
     "imxrt": ["arm-gcc", "arm-clang"],
@@ -38,9 +38,9 @@ family_list = {
     "stm32h7": ["arm-gcc", "arm-clang", "arm-iar"],
     "stm32h7rs stm32l0 stm32l4": ["arm-gcc", "arm-clang", "arm-iar"],
     "stm32n6": ["arm-gcc"],
-    "stm32u0 stm32wb stm32wba": ["arm-gcc", "arm-clang", "arm-iar"],
-    "stm32u5": ["arm-gcc", "arm-clang", "arm-iar"],
-    "-bespressif_s2_devkitc": ["esp-idf"],
+    "stm32u0 stm32u5": ["arm-gcc", "arm-clang", "arm-iar"],
+    "stm32wb stm32wba": ["arm-gcc", "arm-clang", "arm-iar"],
+    # "-bespressif_s2_devkitc": ["esp-idf"],
     # S3, P4 will be built by hil test
     # "-bespressif_s3_devkitm": ["esp-idf"],
     # "-bespressif_p4_function_ev": ["esp-idf"],
diff --git a/.idea/cmake.xml b/.idea/cmake.xml
index cc73ca8fc..5f9e1acd2 100644
--- a/.idea/cmake.xml
+++ b/.idea/cmake.xml
@@ -131,6 +131,7 @@
       <configuration PROFILE_NAME="stm32h7s3nucleo" ENABLED="false" CONFIG_NAME="Debug" GENERATION_OPTIONS="-DBOARD=stm32h7s3nucleo -DLOG=1" />
       <configuration PROFILE_NAME="stm32l0538disco" ENABLED="false" CONFIG_NAME="Debug" GENERATION_OPTIONS="-DBOARD=stm32l0538disco -DLOG=0 -DLOGGER=RTT" />
       <configuration PROFILE_NAME="stm32l476disco" ENABLED="false" CONFIG_NAME="Debug" GENERATION_OPTIONS="-DBOARD=stm32l476disco -DLOG=1 -DLOGGER=RTT" />
+      <configuration PROFILE_NAME="stm32u083cdk" ENABLED="false" CONFIG_NAME="Debug" GENERATION_OPTIONS="-DBOARD=stm32u083cdk -DLOG=1 -DLOGGER=RTT" />
       <configuration PROFILE_NAME="stm32u575nucleo" ENABLED="false" CONFIG_NAME="Debug" GENERATION_OPTIONS="-DBOARD=stm32u575nucleo -DLOG=1 -DLOGGER=RTT" />
       <configuration PROFILE_NAME="stm32u5a5nucleo" ENABLED="false" CONFIG_NAME="Debug" GENERATION_OPTIONS="-DBOARD=stm32u5a5nucleo -DLOG=1 -DLOGGER=RTT" />
       <configuration PROFILE_NAME="stm32wb55nucleo" ENABLED="false" GENERATION_OPTIONS="-DBOARD=stm32wb55nucleo" />
diff --git a/README.rst b/README.rst
index 0eb1e84b9..a3863c375 100644
--- a/README.rst
+++ b/README.rst
@@ -101,10 +101,11 @@ If you have a special requirement, ``usbd_app_driver_get_cb()`` can be used to w
 Host Stack
 ----------
 
-- Human Interface Device (HID): Keyboard, Mouse, Generic
-- Mass Storage Class (MSC)
 - Communication Device Class: CDC-ACM
 - Vendor serial over USB: FTDI, CP210x, CH34x, PL2303
+- Human Interface Device (HID): Keyboard, Mouse, Generic
+- Mass Storage Class (MSC)
+- Musical Instrument Digital Interface (MIDI)
 - Hub with multiple-level support
 
 Similar to the Device Stack, if you have a special requirement, ``usbh_app_driver_get_cb()`` can be used to write your own class driver without modifying the stack.
diff --git a/docs/getting_started.rst b/docs/getting_started.rst
index 0c3fcec80..8442305d4 100644
--- a/docs/getting_started.rst
+++ b/docs/getting_started.rst
@@ -34,7 +34,10 @@ Get the Code
    $ python tools/get_deps.py -b stm32h743eval  # or python tools/get_deps.py stm32h7
 
 .. note::
-   For rp2040 `pico-sdk <https://github.com/raspberrypi/pico-sdk>`_ or `esp-idf <https://github.com/espressif/esp-idf>`_ for Espressif targets are required; install them per vendor instructions.
+   Some MCU families require additional SDKs, please follow their instructions to install and set it up
+
+   * **rp2040**: Requires `pico-sdk <https://github.com/raspberrypi/pico-sdk>`_
+   * **Espressif (esp32)**: Requires `esp-idf <https://github.com/espressif/esp-idf>`_. Only a few examples support the ESP-IDF build system. Look for ones with `src/CMakeLists.txt` that contain `idf_component_register()`, such as `cdc_msc_freertos`.
 
 Simple Device Example
 ---------------------
diff --git a/examples/device/cdc_dual_ports/src/tusb_config.h b/examples/device/cdc_dual_ports/src/tusb_config.h
index 0da4032a7..710c01ee2 100644
--- a/examples/device/cdc_dual_ports/src/tusb_config.h
+++ b/examples/device/cdc_dual_ports/src/tusb_config.h
@@ -104,6 +104,8 @@
 #define CFG_TUD_CDC_TX_BUFSIZE   (TUD_OPT_HIGH_SPEED ? 512 : 64)
 
 // CDC Endpoint transfer buffer size, more is faster
+// Leave it as default size (512 for HS, 64 for FS) unless your host application
+// is able to send ZLP (Zero Length Packet) to terminate transfer !
 #define CFG_TUD_CDC_EP_BUFSIZE   (TUD_OPT_HIGH_SPEED ? 512 : 64)
 
 #ifdef __cplusplus
diff --git a/examples/device/cdc_msc/src/tusb_config.h b/examples/device/cdc_msc/src/tusb_config.h
index fdb2ddf18..3f2f05f20 100644
--- a/examples/device/cdc_msc/src/tusb_config.h
+++ b/examples/device/cdc_msc/src/tusb_config.h
@@ -104,6 +104,8 @@
 #define CFG_TUD_CDC_TX_BUFSIZE   (TUD_OPT_HIGH_SPEED ? 512 : 64)
 
 // CDC Endpoint transfer buffer size, more is faster
+// Leave it as default size (512 for HS, 64 for FS) unless your host application
+// is able to send ZLP (Zero Length Packet) to terminate transfer !
 #define CFG_TUD_CDC_EP_BUFSIZE   (TUD_OPT_HIGH_SPEED ? 512 : 64)
 
 // MSC Buffer size of Device Mass storage
diff --git a/examples/device/cdc_msc_freertos/src/tusb_config.h b/examples/device/cdc_msc_freertos/src/tusb_config.h
index 6b1937a8d..8277b1604 100644
--- a/examples/device/cdc_msc_freertos/src/tusb_config.h
+++ b/examples/device/cdc_msc_freertos/src/tusb_config.h
@@ -111,6 +111,8 @@
 #define CFG_TUD_CDC_TX_BUFSIZE   (TUD_OPT_HIGH_SPEED ? 512 : 64)
 
 // CDC Endpoint transfer buffer size, more is faster
+// Leave it as default size (512 for HS, 64 for FS) unless your host application
+// is able to send ZLP (Zero Length Packet) to terminate transfer !
 #define CFG_TUD_CDC_EP_BUFSIZE   (TUD_OPT_HIGH_SPEED ? 512 : 64)
 
 // MSC Buffer size of Device Mass storage
diff --git a/examples/device/cdc_uac2/src/tusb_config.h b/examples/device/cdc_uac2/src/tusb_config.h
index b7ece8b7c..5eb2e8f74 100644
--- a/examples/device/cdc_uac2/src/tusb_config.h
+++ b/examples/device/cdc_uac2/src/tusb_config.h
@@ -160,6 +160,8 @@ extern "C" {
 #define CFG_TUD_CDC_TX_BUFSIZE   (TUD_OPT_HIGH_SPEED ? 512 : 64)
 
 // CDC Endpoint transfer buffer size, more is faster
+// Leave it as default size (512 for HS, 64 for FS) unless your host application
+// is able to send ZLP (Zero Length Packet) to terminate transfer !
 #define CFG_TUD_CDC_EP_BUFSIZE   (TUD_OPT_HIGH_SPEED ? 512 : 64)
 
 #ifdef __cplusplus
diff --git a/hw/bsp/family_support.cmake b/hw/bsp/family_support.cmake
index baa8422fe..ad9a4f94d 100644
--- a/hw/bsp/family_support.cmake
+++ b/hw/bsp/family_support.cmake
@@ -238,6 +238,7 @@ function(family_add_bloaty TARGET)
     COMMAND ${BLOATY_EXE} ${OPTION_LIST} $<TARGET_FILE:${TARGET}>
     VERBATIM)
 
+  set_property(TARGET ${TARGET}-bloaty PROPERTY FOLDER ${TARGET})
   # post build
   #  add_custom_command(TARGET ${TARGET} POST_BUILD
   #    COMMAND ${BLOATY_EXE} --csv ${OPTION_LIST} $<TARGET_FILE:${TARGET}> > ${CMAKE_CURRENT_BINARY_DIR}/${TARGET}_bloaty.csv
@@ -258,6 +259,8 @@ function(family_add_linkermap TARGET)
     VERBATIM
     )
 
+  set_property(TARGET ${TARGET}-linkermap PROPERTY FOLDER ${TARGET})
+
   # post build
   add_custom_command(TARGET ${TARGET} POST_BUILD
     COMMAND python ${LINKERMAP_PY} ${OPTION_LIST} $<TARGET_FILE:${TARGET}>.map
@@ -508,6 +511,8 @@ exit"
     COMMAND ${JLINKEXE} -device ${JLINK_DEVICE} ${OPTION_LIST} -if ${JLINK_IF} -JTAGConf -1,-1 -speed auto -CommandFile $<TARGET_FILE_DIR:${BINARY_TARGET}>/${BINARY_TARGET}.jlink
     VERBATIM
     )
+
+  set_property(TARGET ${NAME_TARGET}-jlink PROPERTY FOLDER ${TARGET})
 endfunction()
 
 
@@ -521,6 +526,8 @@ function(family_flash_stlink TARGET)
     DEPENDS ${TARGET}
     COMMAND ${STM32_PROGRAMMER_CLI} --connect port=swd --write $<TARGET_FILE:${TARGET}> --go
     )
+
+  set_property(TARGET ${TARGET}-stlink PROPERTY FOLDER ${TARGET})
 endfunction()
 
 
@@ -534,6 +541,8 @@ function(family_flash_stflash TARGET)
     DEPENDS ${TARGET}
     COMMAND ${ST_FLASH} write $<TARGET_FILE_DIR:${TARGET}>/${TARGET}.bin 0x8000000
     )
+
+  set_property(TARGET ${TARGET}-stflash PROPERTY FOLDER ${TARGET})
 endfunction()
 
 
@@ -560,6 +569,8 @@ function(family_flash_openocd TARGET)
     COMMAND ${OPENOCD} -c "tcl_port disabled; gdb_port disabled" ${OPTION_LIST} -c "init; halt; program $<TARGET_FILE:${TARGET}>" -c reset ${OPTION_LIST2} -c exit
     VERBATIM
     )
+
+  set_property(TARGET ${TARGET}-openocd PROPERTY FOLDER ${TARGET})
 endfunction()
 
 
@@ -621,6 +632,8 @@ function(family_flash_wlink_rs TARGET)
     DEPENDS ${TARGET}
     COMMAND ${WLINK_RS} flash $<TARGET_FILE:${TARGET}>
     )
+
+  set_property(TARGET ${TARGET}-wlink-rs PROPERTY FOLDER ${TARGET})
 endfunction()
 
 
@@ -634,6 +647,8 @@ function(family_flash_pyocd TARGET)
     DEPENDS ${TARGET}
     COMMAND ${PYOCD} flash -t ${PYOCD_TARGET} $<TARGET_FILE:${TARGET}>
     )
+
+  set_property(TARGET ${TARGET}-pyocd PROPERTY FOLDER ${TARGET})
 endfunction()
 
 
@@ -643,6 +658,7 @@ function(family_flash_uf2 TARGET FAMILY_ID)
     DEPENDS ${TARGET}
     COMMAND python ${UF2CONV_PY} -f ${FAMILY_ID} --deploy $<TARGET_FILE_DIR:${TARGET}>/${TARGET}.uf2
     )
+  set_property(TARGET ${TARGET}-uf2 PROPERTY FOLDER ${TARGET})
 endfunction()
 
 
@@ -657,6 +673,8 @@ function(family_flash_teensy TARGET)
     COMMAND ${CMAKE_OBJCOPY} -Oihex $<TARGET_FILE:${TARGET}> $<TARGET_FILE_DIR:${TARGET}>/${TARGET}.hex
     COMMAND ${TEENSY_CLI} --mcu=${TEENSY_MCU} -w -s $<TARGET_FILE_DIR:${TARGET}>/${TARGET}.hex
     )
+
+  set_property(TARGET ${TARGET}-teensy PROPERTY FOLDER ${TARGET})
 endfunction()
 
 
@@ -675,6 +693,8 @@ function(family_flash_nxplink TARGET)
     DEPENDS ${TARGET}
     COMMAND ${LINKSERVER_PATH} flash ${NXPLINK_DEVICE} load $<TARGET_FILE:${TARGET}>
     )
+
+  set_property(TARGET ${TARGET}-nxplink PROPERTY FOLDER ${TARGET})
 endfunction()
 
 
@@ -688,6 +708,8 @@ function(family_flash_dfu_util TARGET OPTION)
     COMMAND ${DFU_UTIL} -R -d ${DFU_UTIL_VID_PID} -a 0 -D $<TARGET_FILE_DIR:${TARGET}>/${TARGET}.bin
     VERBATIM
     )
+
+  set_property(TARGET ${TARGET}-dfu-util PROPERTY FOLDER ${TARGET})
 endfunction()
 
 function(family_flash_msp430flasher TARGET)
@@ -703,6 +725,8 @@ function(family_flash_msp430flasher TARGET)
     COMMAND ${CMAKE_COMMAND} -E env LD_LIBRARY_PATH=${MSP430FLASHER_PARENT_DIR}
             ${MSP430FLASHER} -w $<TARGET_FILE_DIR:${TARGET}>/${TARGET}.hex -z [VCC]
     )
+
+  set_property(TARGET ${TARGET}-msp430flasher PROPERTY FOLDER ${TARGET})
 endfunction()
 
 function(family_flash_uniflash TARGET)
@@ -717,6 +741,8 @@ function(family_flash_uniflash TARGET)
     COMMAND ${DSLITE} ${UNIFLASH_OPTION} -f $<TARGET_FILE_DIR:${TARGET}>/${TARGET}.hex
     VERBATIM
     )
+
+  set_property(TARGET ${TARGET}-uniflash PROPERTY FOLDER ${TARGET})
 endfunction()
 
 #----------------------------------
diff --git a/src/class/mtp/mtp_device.h b/src/class/mtp/mtp_device.h
index a33f1dc08..6cce7efbb 100644
--- a/src/class/mtp/mtp_device.h
+++ b/src/class/mtp/mtp_device.h
@@ -18,7 +18,7 @@
  * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
  * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
  * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
- * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN0
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
  * THE SOFTWARE.
  *
  * This file is part of the TinyUSB stack.
diff --git a/src/class/video/video_device.h b/src/class/video/video_device.h
index f14555e4f..2750bb2fb 100644
--- a/src/class/video/video_device.h
+++ b/src/class/video/video_device.h
@@ -99,8 +99,7 @@ int tud_video_commit_cb(uint_fast8_t ctl_idx, uint_fast8_t stm_idx,
  * @param[in]   stm_idx       Destination streaming interface index
  * @param[out]  payload_buf   Payload storage buffer (target buffer for requested data)
  * @param[in]   payload_size  Size of payload_buf (requested data size)
- * @param[in]   offset        Current byte offset relative to given bufsize from tud_video_n_frame_xfer (framesize)
- * @return video_error_code_t */
+ * @param[in]   offset        Current byte offset relative to given bufsize from tud_video_n_frame_xfer (framesize)  */
 void tud_video_prepare_payload_cb(uint_fast8_t ctl_idx, uint_fast8_t stm_idx, tud_video_payload_request_t* request);
 
 //--------------------------------------------------------------------+
diff --git a/src/common/tusb_verify.h b/src/common/tusb_verify.h
index c9e06361c..bd00b9d11 100644
--- a/src/common/tusb_verify.h
+++ b/src/common/tusb_verify.h
@@ -73,8 +73,13 @@
   #define TU_MESS_FAILED() do {} while (0)
 #endif
 
+// Custom defined application function
+#ifdef CFG_TUSB_DEBUG_BREAKPOINT
+  extern void CFG_TUSB_DEBUG_BREAKPOINT(void);
+  #define TU_BREAKPOINT() CFG_TUSB_DEBUG_BREAKPOINT()
+
 // Halt CPU (breakpoint) when hitting error, only apply for Cortex M3, M4, M7, M33. M55
-#if defined(__ARM_ARCH_7M__) || defined (__ARM_ARCH_7EM__) || defined(__ARM_ARCH_8M_MAIN__) || defined(__ARM_ARCH_8_1M_MAIN__) || \
+#elif defined(__ARM_ARCH_7M__) || defined (__ARM_ARCH_7EM__) || defined(__ARM_ARCH_8M_MAIN__) || defined(__ARM_ARCH_8_1M_MAIN__) || \
     defined(__ARM7M__) || defined (__ARM7EM__) || defined(__ARM8M_MAINLINE__) || defined(__ARM8EM_MAINLINE__)
   #define TU_BREAKPOINT() do {                                                                              \
     volatile uint32_t* ARM_CM_DHCSR =  ((volatile uint32_t*) 0xE000EDF0UL); /* Cortex M CoreDebug->DHCSR */ \
diff --git a/src/device/usbd.c b/src/device/usbd.c
index 1e21c667a..cca4169d7 100644
--- a/src/device/usbd.c
+++ b/src/device/usbd.c
@@ -666,8 +666,14 @@ void tud_task_ext(uint32_t timeout_ms, bool in_isr) {
     return;
   }
 
-  // Loop until there is no more events in the queue
-  while (1) {
+  // Loop until there are no more events in the queue or CFG_TUD_TASK_EVENTS_PER_RUN is reached
+  for (unsigned epr = 0;; epr++) {
+#if CFG_TUD_TASK_EVENTS_PER_RUN > 0
+    if (epr >= CFG_TUD_TASK_EVENTS_PER_RUN) {
+      TU_LOG_USBD("USBD event limit (" TU_XSTRING(CFG_TUD_TASK_EVENTS_PER_RUN) ") reached\r\n");
+      break;
+    }
+#endif
     dcd_event_t event;
     if (!osal_queue_receive(_usbd_q, &event, timeout_ms)) {
       return;
diff --git a/src/device/usbd.h b/src/device/usbd.h
index bd5a3c395..4016a45a4 100644
--- a/src/device/usbd.h
+++ b/src/device/usbd.h
@@ -819,7 +819,7 @@ bool tud_vendor_control_xfer_cb(uint8_t rhport, uint8_t stage, tusb_control_requ
   /* Interface */ \
   9, TUSB_DESC_INTERFACE, _itfnum, 0, 0, TUD_DFU_APP_CLASS, TUD_DFU_APP_SUBCLASS, DFU_PROTOCOL_RT, _stridx, \
   /* Function */ \
-  9, DFU_DESC_FUNCTIONAL, _attr, U16_TO_U8S_LE(_timeout), U16_TO_U8S_LE(_xfer_size), U16_TO_U8S_LE(0x0101)
+  9, DFU_DESC_FUNCTIONAL, _attr, U16_TO_U8S_LE(_timeout), U16_TO_U8S_LE(_xfer_size), U16_TO_U8S_LE(0x0110)
 
 //--------------------------------------------------------------------+
 // DFU Descriptor Templates
@@ -833,7 +833,7 @@ bool tud_vendor_control_xfer_cb(uint8_t rhport, uint8_t stage, tusb_control_requ
 #define TUD_DFU_DESCRIPTOR(_itfnum, _alt_count, _stridx, _attr, _timeout, _xfer_size) \
   TU_XSTRCAT(TUD_DFU_ALT_,_alt_count)(_itfnum, 0, _stridx), \
   /* Function */ \
-  9, DFU_DESC_FUNCTIONAL, _attr, U16_TO_U8S_LE(_timeout), U16_TO_U8S_LE(_xfer_size), U16_TO_U8S_LE(0x0101)
+  9, DFU_DESC_FUNCTIONAL, _attr, U16_TO_U8S_LE(_timeout), U16_TO_U8S_LE(_xfer_size), U16_TO_U8S_LE(0x0110)
 
 #define TUD_DFU_ALT(_itfnum, _alt, _stridx) \
   /* Interface */ \
diff --git a/src/host/usbh.c b/src/host/usbh.c
index a725b7c8b..41f41dcfb 100644
--- a/src/host/usbh.c
+++ b/src/host/usbh.c
@@ -599,8 +599,14 @@ void tuh_task_ext(uint32_t timeout_ms, bool in_isr) {
     return;
   }
 
-  // Loop until there is no more events in the queue
-  while (1) {
+  // Loop until there are no more events in the queue or CFG_TUH_TASK_EVENTS_PER_RUN is reached
+  for (unsigned epr = 0;; epr++) {
+#if CFG_TUH_TASK_EVENTS_PER_RUN > 0
+    if (epr >= CFG_TUH_TASK_EVENTS_PER_RUN) {
+      TU_LOG_USBH("USBH event limit (" TU_XSTRING(CFG_TUH_TASK_EVENTS_PER_RUN) ") reached\r\n");
+      break;
+    }
+#endif
     hcd_event_t event;
     if (!osal_queue_receive(_usbh_q, &event, timeout_ms)) { return; }
 
diff --git a/src/portable/mentor/musb/dcd_musb.c b/src/portable/mentor/musb/dcd_musb.c
index f0c5aa722..d329285e9 100644
--- a/src/portable/mentor/musb/dcd_musb.c
+++ b/src/portable/mentor/musb/dcd_musb.c
@@ -32,12 +32,6 @@
 #define MUSB_DEBUG 2
 #define MUSB_REGS(rhport)   ((musb_regs_t*) MUSB_BASES[rhport])
 
-#if __GNUC__ > 8 && defined(__ARM_FEATURE_UNALIGNED)
-/* GCC warns that an address may be unaligned, even though
- * the target CPU has the capability for unaligned memory access. */
-_Pragma("GCC diagnostic ignored \"-Waddress-of-packed-member\"");
-#endif
-
 #include "musb_type.h"
 #include "device/dcd.h"
 
@@ -73,7 +67,10 @@ typedef struct TU_ATTR_PACKED
 
 typedef struct
 {
-  tusb_control_request_t setup_packet;
+  union {
+    tusb_control_request_t setup_packet;
+    uint32_t setup_buffer[2];
+  };
   uint16_t     remaining_ctrl; /* The number of bytes remaining in data stage of control transfer. */
   int8_t       status_out;
   pipe_state_t pipe0;
@@ -174,9 +171,8 @@ static void process_setup_packet(uint8_t rhport) {
   musb_regs_t* musb_regs = MUSB_REGS(rhport);
 
   // Read setup packet
-  uint32_t *p = (void*)&_dcd.setup_packet;
-  p[0] = musb_regs->fifo[0];
-  p[1] = musb_regs->fifo[0];
+  _dcd.setup_buffer[0] = musb_regs->fifo[0];
+  _dcd.setup_buffer[1] = musb_regs->fifo[0];
 
   _dcd.pipe0.buf       = NULL;
   _dcd.pipe0.length    = 0;
@@ -193,14 +189,13 @@ static void process_setup_packet(uint8_t rhport) {
   }
 }
 
-static bool handle_xfer_in(uint8_t rhport, uint_fast8_t ep_addr)
-{
+static bool handle_xfer_in(uint8_t rhport, uint_fast8_t ep_addr) {
   unsigned epnum = tu_edpt_number(ep_addr);
   unsigned epnum_minus1 = epnum - 1;
   pipe_state_t  *pipe = &_dcd.pipe[tu_edpt_dir(ep_addr)][epnum_minus1];
   const unsigned rem  = pipe->remaining;
 
-  if (!rem) {
+  if (rem == 0 && pipe->length > 0) {
     pipe->buf = NULL;
     return true;
   }
diff --git a/src/portable/st/stm32_fsdev/fsdev_common.h b/src/portable/st/stm32_fsdev/fsdev_common.h
index c53e345b0..b749a92ff 100644
--- a/src/portable/st/stm32_fsdev/fsdev_common.h
+++ b/src/portable/st/stm32_fsdev/fsdev_common.h
@@ -63,23 +63,21 @@ TU_VERIFY_STATIC(FSDEV_BTABLE_BASE % 8 == 0, "BTABLE base must be aligned to 8 b
 
 // CFG_TUSB_FSDEV_PMA_SIZE is PMA buffer size in bytes.
 // - 512-byte devices, access with a stride of two words (use every other 16-bit address)
-// - 1024-byte devices, access with a stride of one word (use every 16-bit address)
+// - 1024-byte devices, access with a stride of one word (use every 16-bit address) or 32-bit address
 // - 2048-byte devices, access with 32-bit address
-
-// For purposes of accessing the packet
-#if CFG_TUSB_FSDEV_PMA_SIZE == 512
-  // 1x16 bit / word access scheme
-  #define FSDEV_PMA_STRIDE  2
-  #define pma_access_scheme TU_ATTR_ALIGNED(4)
-#elif CFG_TUSB_FSDEV_PMA_SIZE == 1024
-  // 2x16 bit / word access scheme
-  #define FSDEV_PMA_STRIDE 1
-  #define pma_access_scheme
-#elif CFG_TUSB_FSDEV_PMA_SIZE == 2048
+#if CFG_TUSB_FSDEV_PMA_SIZE == 2048 || TU_CHECK_MCU(OPT_MCU_STM32U0)
   // 32 bit access scheme
   #define FSDEV_BUS_32BIT
   #define FSDEV_PMA_STRIDE 1
   #define pma_access_scheme
+#elif CFG_TUSB_FSDEV_PMA_SIZE == 1024
+  // 2x16 bit / word access scheme
+  #define FSDEV_PMA_STRIDE 1
+  #define pma_access_scheme
+#elif CFG_TUSB_FSDEV_PMA_SIZE == 512
+  // 1x16 bit / word access scheme
+  #define FSDEV_PMA_STRIDE  2
+  #define pma_access_scheme TU_ATTR_ALIGNED(4)
 #endif
 
 // The fsdev_bus_t type can be used for both register and PMA access necessities
diff --git a/src/tusb_option.h b/src/tusb_option.h
index abf5e0608..8e270e5f2 100644
--- a/src/tusb_option.h
+++ b/src/tusb_option.h
@@ -340,14 +340,14 @@
 #if defined(TUP_USBIP_FSDEV)
   #define CFG_TUD_EDPT_DEDICATED_HWFIFO 1
 
-  #if CFG_TUSB_FSDEV_PMA_SIZE == 512
-    #define CFG_TUSB_FIFO_HWFIFO_DATA_STRIDE 2 // 16-bit data
+  #if CFG_TUSB_FSDEV_PMA_SIZE == 2048 || TU_CHECK_MCU(OPT_MCU_STM32U0)
+    #define CFG_TUSB_FIFO_HWFIFO_DATA_STRIDE 4 // 32-bit data
     #define CFG_TUSB_FIFO_HWFIFO_ADDR_STRIDE 4 // 32-bit address increase
   #elif CFG_TUSB_FSDEV_PMA_SIZE == 1024
     #define CFG_TUSB_FIFO_HWFIFO_DATA_STRIDE 2 // 16-bit data
     #define CFG_TUSB_FIFO_HWFIFO_ADDR_STRIDE 2 // 16-bit address increase
-  #elif CFG_TUSB_FSDEV_PMA_SIZE == 2048
-    #define CFG_TUSB_FIFO_HWFIFO_DATA_STRIDE 4 // 32-bit data
+  #elif CFG_TUSB_FSDEV_PMA_SIZE == 512
+    #define CFG_TUSB_FIFO_HWFIFO_DATA_STRIDE 2 // 16-bit data
     #define CFG_TUSB_FIFO_HWFIFO_ADDR_STRIDE 4 // 32-bit address increase
   #endif
 #endif
@@ -560,6 +560,11 @@
   #define CFG_TUD_INTERFACE_MAX   16
 #endif
 
+// max events processed in one tud_task_ext() call, 0 for unlimited
+#ifndef CFG_TUD_TASK_EVENTS_PER_RUN
+  #define CFG_TUD_TASK_EVENTS_PER_RUN  16
+#endif
+
 // default to max hardware endpoint, but can be smaller to save RAM
 #ifndef CFG_TUD_ENDPPOINT_MAX
   #define CFG_TUD_ENDPPOINT_MAX   TUP_DCD_ENDPOINT_MAX
@@ -679,6 +684,11 @@
   #define CFG_TUH_MEM_DCACHE_LINE_SIZE CFG_TUSB_MEM_DCACHE_LINE_SIZE
 #endif
 
+// max events processed in one tuh_task_ext() call, 0 for unlimited
+#ifndef CFG_TUH_TASK_EVENTS_PER_RUN
+  #define CFG_TUH_TASK_EVENTS_PER_RUN  16
+#endif
+
 //------------- CLASS -------------//
 
 #ifndef CFG_TUH_HUB
diff --git a/tools/codespell/ignore-words.txt b/tools/codespell/ignore-words.txt
index 957cbd86b..5b6e2e98b 100644
--- a/tools/codespell/ignore-words.txt
+++ b/tools/codespell/ignore-words.txt
@@ -1,14 +1,15 @@
-synopsys
-sie
-tre
-thre
-hsi
-fro
-dout
-mot
-te
 attch
+busses
+dout
 endianess
+fro
+hsi
+inout
+mot
 pris
-busses
 ser
+sie
+synopsys
+te
+thre
+tre
diff --git a/tools/get_deps.py b/tools/get_deps.py
index a75f0c169..954b2ece7 100755
--- a/tools/get_deps.py
+++ b/tools/get_deps.py
@@ -330,18 +330,16 @@ def main():
     parser.add_argument('-b', '--board', action='append', default=[], help='Boards to fetch')
     parser.add_argument('-D', '--define', action='append', default=[], help='Have no effect')
     parser.add_argument('-f1', '--build-flags-on', action='append', default=[], help='Have no effect')
-    parser.add_argument('--print', action='store_true', help='Print commit hash only')
     args = parser.parse_args()
 
     families = args.families
     boards = args.board
-    print_only = args.print
 
     status = 0
-    deps = list(deps_mandatory.keys())
+    deps = []
 
     if 'all' in families:
-        deps += deps_optional.keys()
+        deps.extend(deps_optional.keys())
     else:
         families = list(families)
         if boards is not None:
@@ -349,24 +347,16 @@ def main():
                 f = find_family(b)
                 if f is not None:
                     families.append(f)
-
         for f in families:
             for d in deps_optional:
                 if d not in deps and f in deps_optional[d][2].split():
                     deps.append(d)
+        if len(deps) == 0:
+            print('WARN: no additional dependencies found for given boards or families')
 
-    if print_only:
-        pvalue = {}
-        # print only without arguments, always add CMSIS_5
-        if len(families) == 0 and len(boards) == 0:
-            deps.append('lib/CMSIS_5')
-        for d in deps:
-            commit = deps_all[d][1]
-            pvalue[d] = commit
-        print(pvalue)
-    else:
-        with Pool() as pool:
-            status = sum(pool.map(get_a_dep, deps))
+    deps.extend(deps_mandatory.keys())
+    with Pool() as pool:
+        status = sum(pool.map(get_a_dep, deps))
     return status