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>
The interrupt handler accesses the device as well as the driver's private
data. Thus, must take the big kernel lock in SMP mode to protect them.
Signed-off-by: Jukka Laitinen <jukka.laitinen@tii.ae>
In SMP mode one CPU can be executing the MMC interrupt while another CPU
disables (e.g. via watchdog timeout). As it is disabled the other CPU
assumes it's safe to start configuring the device after this.
This causes a leak in the driver's private data as well as a mutual
exclusion leak on the device itself.
Fix this by aborting any triggered interrupt by checking whether it's
even enabled.
Signed-off-by: Jukka Laitinen <jukka.laitinen@tii.ae>
Update board reset with BOOTROM functions calls
- normal reboot
- reboot to bootloader
normal reboot and reboot bootloader now possible from nsh
Signed-off-by: Serg Podtynnyi <serg@podtynnyi.com>
Currently the irq handler checks many reserved bits, which is
a waste of resources:
1. pending_rx_ep bit 0 is reserved (always 0)
2. pending_rx_ep and pending_tx_ep have only bits 1, 2, 3 and 4
defined, no need to scan MPFS_USB_NENDPOINTS (9) bits as the
rest are reserved
Fix this by checking only the relevant bits.
Signed-off-by: Eero Nurkkala <eero.nurkkala@offcode.fi>
Signed-off-by: Jukka Laitinen <jukka.laitinen@tii.ae>
As emmcsd driver does not support separate WRCOMPLETE interrupt the
SDIOWAIT_WRCOMPLETE event shall not be waited.
The SDIOWAIT_TRANSFERDONE event indicates that both "transfer done"
and "write complete" events are completed.
mpfs_sendcmd() shall not wait for data lines ready before
sending commands to the card. This prevents mmcsd driver
to poll card busy status prior to next activity after
a write operation.
This fixes the same issue for other targets, which was already fixed for
xtensa in commit 50d94863.
After the signals have been delivered, the local irqs need to be
disabled until the context switch. But just calling
leave_critical_section(regs[xx]) will enable them if they were
enabled in the context.
Signed-off-by: Jukka Laitinen <jukka.laitinen@tii.ae>
1. When the Clang compiler turns on "-fsanitize=kernel-address", inlining, global variables, and stack detection are enabled by default and must be turned off manually.
2. -mllvm is the parameter passing method of Clang, and --param is the parameter passing method of GCC
After the modification, KASan compilation and operation will be supported for Clang 18 and above
Signed-off-by: wangmingrong1 <wangmingrong1@xiaomi.com>
When the vaddr parameter passed to map_region() is not aligned to the page directory,
it causes incorrect address mapping for later regions.
For example, in the sv32 case, `PGPOOL` started at `0x80a00000` with a size of `1024*4096B`,
leading to page table errors for the range `0x80c00000~0x80e00000`.
This patch fixes the issue by ensuring map_region() correctly handles unaligned vaddr cases.
Signed-off-by: wushenhui <wushenhui@xiaomi.com>
This commit fixes an issue when `CONFIG_DEBUG_SYSCALL_INFO=y`: the
`cmd` variable doesn't exist (instead, `regs[REG_A0]` represents
the syscall command directly. Also, it fixes the parameter for
`up_dump_register`, as `tcb` is a pointer here. By applying these
fixes, debugging syscall info is now possible again.
Previously, if an event was generated in hardware after taking spin
lock it was not correctly accounted for in current reading cycle.
Now, we check for events and compensate count accordingly.
Signed-off-by: Martin Vajnar <martin.vajnar@gmail.com>
Decouples the NuttX build from the MCUBoot common source on RISC-V Espressif
devices. Allows using different branches for each.
Signed-off-by: Filipe Cavalcanti <filipe.cavalcanti@espressif.com>
This helps in debugging loaded elf files in CONFIG_BUILD_KERNEL. If a user space exception occurs,
one would beed the process name in order to debug the correct process/elf file.
Only dumping the pid and name of the crashed task/thread doesn't help, since different processes
may have helper threads with the same name.
Signed-off-by: Jukka Laitinen <jukka.laitinen@tii.ae>
There is no more linking error for MPFS after the flagging is corrected in
drivers/timers/arch_alarm.c
Signed-off-by: Jukka Laitinen <jukka.laitinen@tii.ae>
Use the flag CONFIG_ARCH_HAVE_PERF_EVENTS to detect whether the architecture specific code
provides the up_perf_* functions. Now it is mixed with CONFIG_ARCH_PERF_EVENTS, which should
select just whether the perf events (perf_*) are enabled for the configuration.
- drivers/timers/arch_alarm.c: Don't compile the up_perf_* functions here if the
CONFIG_ARCH_HAVE_PERF_EVENTS is defined
- arch/*/*_perf.c: Change CONFIG_ARCH_PERF_EVENTS -> CONFIG_ARCH_HAVE_PERF_EVENTS to
select whether architecture specific up_perf_* functions are provided
Signed-off-by: Jukka Laitinen <jukka.laitinen@tii.ae>
Add up_perf_ functions for MPFS, which don't rely on alarm/oneshot interface.
Also add optimized up_udelay and up_ndelay functions, which use the MTIMER
directly to measure time; making them accurate and more multithreading friendly.
Signed-off-by: Jukka Laitinen <jukka.laitinen@tii.ae>
Even when enabled, the PCNT counter doesn't accumulate into the 32-bit value.
Instead, a value in range [PCNT_LOW_LIMIT, PCNT_HIGH_LIMIT] is always returned.
This is due to interrupt events associated with limit overflows are disabled on the periphery,
therefore the ISR responsible for the accumulation never gets called.
Fixed by enabling the associated interrupt events.
Signed-off-by: michal matias <mich4l.matias@gmail.com>
The version and the git repository of Espressif's MCUboot port can
be changed by setting the `MCUBOOT_VERSION` and `MCUBOOT_URL`
environment variables before running the `make bootloader` command.
Signed-off-by: Tiago Medicci Serrano <tiago.medicci@espressif.com>
To fix a fpu illegal instruction exception due to fcsr have invalid
rounding-mode, restore fcsr/f0~f31 with a random value from stack because fpu
context is not save into stack when mstatus.fs is clean(not dirty).
The number of save/restore fpu context operations is not equal.
In fact, CONFIG_ARCH_LAZYFPU is a performance optimization mechanism aimed at
minimizing the saving and restoring of FPU registers. In some cases, it
may only need to be saved once but must be restored multiple times.
Therefore, this optimization mechanism dictates that saving cannot be
done on the stack; it must be stored at a fixed location(e.g. task
control block buffer), as the stack requires push/pop operations to be
matched.
In case of that CONFIG_ARCH_LAZYFPU is off, save/restore FPU registers
always into stack to avoid the above issues.
Signed-off-by: Jinliang Li <lijinliang1@lixiang.com>
Add missing bootloader_enable_qio_mode() function in flash_qio_mode.c when
CONFIG_ESPRESSIF_FLASH_MODE_QIO option is enabled.
Signed-off-by: Leo Chung <gewalalb@gmail.com>
Add missing bootloader_enable_qio_mode() function in flash_qio_mode.c when
CONFIG_ESPRESSIF_FLASH_MODE_QIO option is enabled.
Signed-off-by: Leo Chung <gewalalb@gmail.com>
Add missing bootloader_enable_qio_mode() function in flash_qio_mode.c when
CONFIG_ESPRESSIF_FLASH_MODE_QIO option is enabled.
Signed-off-by: Leo Chung <gewalalb@gmail.com>
This adds new fields to the metadata section used by MCUBoot.
The openocd-esp32 project requires these fields to properly map the
flash segments and enable using SW breakpoints and flash through
openocd-esp32.
Signed-off-by: Tiago Medicci Serrano <tiago.medicci@espressif.com>
