Updates the SPI Flash driver used for user storage MTD.
Moves ESP32 and ESP32S3 to use common driver.
Updates KConfig options to keep backwards compatibility.
Signed-off-by: Filipe Cavalcanti <filipe.cavalcanti@espressif.com>
Curently, the driver code for HW accelerated AES is not usable
since it's not registered within esp32_crypto. This commit fixes
it as well as a few bugs.
Signed-off-by: Vlad Pruteanu <pruteanuvlad1611@yahoo.com>
Fix exception issue when PM and DEBUG_ASSERTIONS enabled for esp32 and esp32s3.
This error happens due to PM system workflow, when system gets into sleep-modes with
removing PM_NORMAL count, system is raising an exception for not having
enough number of PM_NORMAL state count if DEBUG_ASSERTIONS enabled.
Signed-off-by: Eren Terzioglu <eren.terzioglu@espressif.com>
Adds Kconfig knobs for the ESP32-S3 RX FIFO full threshold and timeout and wires them into the UART driver.
Clears RX FIFO overflow interrupts by resetting the hardware FIFO so reception resumes cleanly.
Signed-off-by: Thiago Finelon <thiago.sfinelon@gmail.com>
Nuttx currently has 2 types of sleep interfaces:
1. Signal-scheduled sleep: nxsig_sleep() / nxsig_usleep() / nxsig_nanosleep()
Weaknesses:
a. Signal-dependent: The signal-scheduled sleep method is bound to the signal framework, while some driver sleep operations do not depend on signals.
b. Timespec conversion: Signal-scheduled sleep involves timespec conversion, which has a significant impact on performance.
2. Busy sleep: up_mdelay() / up_udelay()
Weaknesses:
a. Does not actively trigger scheduling, occupy the CPU loading.
3. New interfaces: Scheduled sleep: nxsched_sleep() / nxsched_usleep() / nxsched_msleep() / nxsched_ticksleep()
Strengths:
a. Does not depend on the signal framework.
b. Tick-based, without additional computational overhead.
Currently, the Nuttx driver framework extensively uses nxsig_* interfaces. However, the driver does not need to rely on signals or timespec conversion.
Therefore, a new set of APIs is added to reduce dependencies on other modules.
(This PR also aims to make signals optional, further reducing the code size of Nuttx.)
Signed-off-by: chao an <anchao.archer@bytedance.com>
Updates MCUBoot version and default address for virtual E-Fuse, depending
if MCUBoot is enabled.
Signed-off-by: Filipe Cavalcanti <filipe.cavalcanti@espressif.com>
Make sure interrupt stack is colored before IRQ is enabled.
Currently, after calling irq_initialize in nx_start to
enable interrupts, there is still a period of execution
path before the interrupt stack is colored.
Signed-off-by: guoshengyuan1 <guoshengyuan1@xiaomi.com>
Incorrect FSM status checks in the suspend path could cause
deadlocks when power management was enabled. This patch fixes
the condition to allow proper UART suspend/resume handling.
Signed-off-by: Thiago Finelon <thiago.sfinelon@gmail.com>
After irq_dispatch finished, the interrupt stack will be checked to
determine whether overflow occurs.
The relevant configuration reuses the configuration of stack overflow
detection during context switching.
Signed-off-by: guoshengyuan1 <guoshengyuan1@xiaomi.com>
Many different architectures re-implemented the exact same code for
`up_*delay` because it was originally written as architecture dependent
code. Busy-waiting can be done regardless of architecture, so this
commit moves that duplicated implementation to a common file with weak
definitions so that individual architectures (see tc32) are still able
to override the definition if needed/desired.
Default implementation is not included if ARCH_TIMER is enabled, since
it is more accurate and provides its own weak definitions to override.
Signed-off-by: Matteo Golin <matteo.golin@gmail.com>
Many times, context switching does not occur when the thread stack
memory is almost exhausted, so we need to mofify up_check_tcbstack to
accurately detect it.
Co-authored-by: Chengdong Wang <wangchengdong@lixiang.com>
Signed-off-by: guoshengyuan1 <guoshengyuan1@xiaomi.com>
Complete the missing scheduling information in some architectures
In these architectures (riscv, avr, tricor) context switching
can occur in both up_switch_context and xx_doirq
Co-authored-by: yinshengkai <yinshengkai@xiaomi.com>
Signed-off-by: guoshengyuan1 <guoshengyuan1@xiaomi.com>
Fix bug in temperature sensor driver where direct casting of lower half
structure pointer could lead to incorrect memory access.
Signed-off-by: Huang Qi <huangqi3@xiaomi.com>
These archs only align the size of the stack, forgeting to do the
stack start addr alignment, this patch fixes it.
Signed-off-by: Chengdong Wang <wangchengdong@lixiang.com>
Fixes low and inconsistent bandwidth issues.
Adds new configuration options for buffer management.
Moves code around to make the implementation cleaner and easier to debug.
Signed-off-by: Filipe Cavalcanti <filipe.cavalcanti@espressif.com>
Updates E-Fuse driver for ESP32, now sharing a common implementation across Xtensa devices.
Signed-off-by: Filipe Cavalcanti <filipe.cavalcanti@espressif.com>
Updates E-Fuse driver for ESP32S2, sharing a common implementation for Xtensa devices.
Signed-off-by: Filipe Cavalcanti <filipe.cavalcanti@espressif.com>
Updates E-Fuse driver for ESP32S3, now sharing a common implementation across
Xtensa devices.
Signed-off-by: Filipe Cavalcanti <filipe.cavalcanti@espressif.com>
Adds SYS_STARTUP_FN which calls constructors and init functions on
common source code. Requires compatibility changes on linker script.
Signed-off-by: Filipe Cavalcanti <filipe.cavalcanti@espressif.com>
Introduce EXT1 wakeup functionality to the ESP32-S3 power management
subsystem. With this feature, the ESP32-S3 can wake up from PM_STANDBY
or PM_SLEEP when one or more RTC GPIOs trigger the EXT1 condition.
Changes include:
- Added EXT1 wakeup configuration and preparation logic in rtc code.
- Integrated EXT1 handling in pmstandby() and pmsleep().
- New board-level Kconfig options under ESP32-S3 DevKit menu:
- CONFIG_PM_EXT1_WAKEUP: enable EXT1 wakeup support
- CONFIG_PM_EXT1_WAKEUP_RTC_GPIO<n>: select RTC GPIOs as wake sources
- CONFIG_PM_EXT1_WAKEUP_TRIGGER_MODE: choose wakeup on HIGH or LOW level
- Fixing Kconfig style
- Fixing comments standard
- Adding PM documentation for esp32s3-devkit
Impact:
- No build impact unless CONFIG_PM_EXT1_WAKEUP is enabled.
- When enabled, ESP32-S3 can wake from low-power states via RTC GPIOs.
- No new public API introduced; controlled via board Kconfig.
Tested on ESP32-S3 DevKit:
- Configured GPIO4 and GPIO5 as EXT1 wake sources.
- Verified wakeup from light sleep and deep sleep on external signals.
- Wakeup reason correctly reported as EXT1.
Signed-off-by: Thiago Finelon <thiago.sfinelon@gmail.com>
Set the affinity of the task before activating it. There is no parameter
or other interface in "kthread_create" to set the affinity mask,
like in "pthread_create".
Signed-off-by: Jukka Laitinen <jukka.laitinen@tii.ae>
Prior to this commit, it wasn't possible to load ELF modules from
the external PSRAM. There were two main issues about it: 1) copying
data using the instruction bus was being used instead of the data
bus (this, per si, isn't a problem, but requires special attention
regarding data alignment), and 2) the cache was not being properly
cleaned and flushed to properly access the loaded data using the
instruction bus.
Signed-off-by: Tiago Medicci Serrano <tiago.medicci@espressif.com>
If both SPI Flash support (`CONFIG_ESP32S3_SPIFLASH`) and PSRAM
(`CONFIG_ESP32S3_SPIRAM`) are enabled, the PSRAM can only be
assigned to the user's heap (`CONFIG_ESP32S3_SPIRAM_USER_HEAP`).
Additionaly, `CONFIG_ESP32S3_SPI_FLASH_SUPPORT_PSRAM_STACK` must be
set because the system will end up allocating tasks' stacks from
the external PSRAM. This has an impact when dealing with SPI flash
operations because the cache must be disabled and the running task
should not rely on any data from the PSRAM. To accomplish that, It
offloads the SPI flash operation to a work queue (which, by
definition, allocates its heap from the kernel heap).
The same (assigning the PSRAM to the user's heap) is valid when the
Wi-Fi is enabled because the lower-half driver requires data being
allocated from the internal memory (which can only be achieved by
allocating from the kernel heap when both the kernel and user heaps
exists).
Signed-off-by: Tiago Medicci Serrano <tiago.medicci@espressif.com>
Use this command to extract archives.
Not all architectures are modified, only those commands I know
or could be logically deducted from the rest were added.
Signed-off-by: Stepan Pressl <pressl.stepan@gmail.com>
After the signals have been delivered, the local irqs need to be
disabled until the context switch. But just calling
leave_critical_section(regs[REG_PS]) will enable them if they were
enabled in the context.
Any interrupt on this cpu before xtensa_context_restore() would set
the rtcb->xcp.regs to NULL, causing a crash in context restore.
Fix this by calling leave_critical_section(up_irq_save()) instead.
Signed-off-by: Jukka Laitinen <jukka.laitinen@tii.ae>
return early due to signal interruption (EINTR). This makes the
main thread think the async operation is done, but the background
worker thread is still running—risking access to freed memory,
race conditions, crashes or undefined behavior.
Using nxsem_wait_uninterruptible():the main thread waits until the
worker thread finishes, preventing these issues.
Signed-off-by: nuttxs <zhaoqing.zhang@sony.com>
