Current NuttX SBI assumes empty PMP settings but that is not always
true, for example some bootloaders may have PMP entries locked before
handling over to NuttX. This patch revises it by not using hardcoded
PMP region number.
Signed-off-by: Yanfeng Liu <yfliu2008@qq.com>
Reproduce:
configure:
./tools/configure.sh qemu-armv7a:nsh -j8
open kasan:
CONFIG_MM_KASAN=y
run:
qemu-system-arm -cpu cortex-a7 -nographic -machine virt,virtualization=off,gic-version=2 -net none -chardev stdio,id=con,mux=on -serial chardev:con -mon chardev=con,mode=readline -kernel ./nuttx
error:
run hanged
Rootcause:
In Qemu armv7a, the ram-end reserved for MMU PGTABLE:
And the up_allocate_heap() directly use the ram-end for heap in arch/arm/src/common/arm_allocateheap.c
*heap_size = CONFIG_RAM_END - g_idle_topstack;
Then they are conflict.
Usually, we won't use the heap end, so can't find the error.
BUT, the KASAN will use the heap end for shadow, so found the error.
Resolve:
up_allocate_heap() consider of MMU, re-define the RAM_END at chip.h
Signed-off-by: ligd <liguiding1@xiaomi.com>
This commit adds support for TX DMA transfers for USART peripheral. Code
refactor in sam_serial.h was also required in order to have correct
defines for all possible cases (both RX and TX DMA used, just one used,
none used).
Signed-off-by: Michal Lenc <michallenc@seznam.cz>
When an application is being loaded `up_addrenv_create ` calls
`create_region` to create the address environment. Only the first
entry is mapped when the region is created. Virtual memory that is
not mapped will trigger an exception when accessed. Other memory
pages are allocated and mapped on-demand. This enables setting
larger heap and stack for the process without compromising the
overall system memory.
Current `CONFIG_PAGING` refers to an experimental implementation
to enable embedded MCUs with some limited RAM space to execute
large programs from some non-random access media.
On-demand paging should be implemented for the kernel mode with
address environment implementation enabled.
Previously we need turn off the PMP locks in K230 vendor u-boot to use
NuttX, this complicates the setup process. This patch supports running
NuttX with original vendor u-boot so that to reduce setup complexity.
It also enables empty NSH prompt string in AMP master config as the apps
side support is ready.
Signed-off-by: Yanfeng Liu <yfliu2008@qq.com>
cpu0 thread0: cpu1:
sched_yield()
nxsched_set_priority()
nxsched_running_setpriority()
nxsched_reprioritize_rtr()
nxsched_add_readytorun()
up_cpu_pause()
IRQ enter
arm64_pause_handler()
enter_critical_section() begin
up_cpu_paused() pick thread0
arm64_restorestate() set thread0 tcb->xcp.regs to CURRENT_REGS
up_switch_context()
thread0 -> thread1
arm64_syscall()
case SYS_switch_context
change thread0 tcb->xcp.regs
restore_critical_section()
enter_critical_section() done
leave_critical_section()
IRQ leave with restore CURRENT_REGS
ERROR !!!
Reason:
As descript above, cpu0 swith task: thread0 -> thread1, and the
syscall() execute slowly, this time cpu1 pick thread0 to run at
up_cpu_paused(). Then cpu0 syscall execute, cpu1 IRQ leave error.
Resolve:
Move arm64_restorestate() after enter_critical_section() done
This is a continued fix with:
https://github.com/apache/nuttx/pull/6833
Signed-off-by: ligd <liguiding1@xiaomi.com>
Use private naming to avoid conflicts with user applications
In file included from libuv/src/unix/internal.h:25,
from libuv/src/unix/udp.c:23:
libuv/src/uv-common.h:57: warning: "container_of" redefined
57 | #define container_of(ptr, type, member) \
|
In file included from nuttx/include/nuttx/list.h:47,
from nuttx/include/nuttx/tls.h:40,
from nuttx/include/nuttx/sched.h:48,
from nuttx/include/nuttx/arch.h:87,
from nuttx/include/nuttx/userspace.h:35,
from nuttx/include/nuttx/mm/mm.h:30,
from nuttx/include/nuttx/kmalloc.h:34,
from nuttx/include/nuttx/lib/lib.h:31,
from nuttx/include/stdio.h:35,
from apps/system/libuv/libuv/include/uv.h:59,
from libuv/src/unix/udp.c:22:
nuttx/include/nuttx/nuttx.h:48: note: this is the location of the previous definition
48 | #define container_of(ptr, type, member) \
|
Signed-off-by: chao an <anchao@lixiang.com>
change optimization to -Os for CONFIG_DEBUG_FULLOPT to be compatible with other architectures
and add an option to select CONFIG_DEBUG_CUSTOMOPT
Signed-off-by: p-szafonimateusz <p-szafonimateusz@xiaomi.com>
enable FPU and implement up_fpucmp to pass ostest
With aggressive optimization enabled (-O2/-O3), ostest FPU test will fail.
This is because the compiler will generate additional vector
instructions between subsequent up_fpucmp() calls (loop vectorization
somewhere in usleep() call), which will consequently overwrite
the expected FPU context (XMM registers).
Signed-off-by: p-szafonimateusz <p-szafonimateusz@xiaomi.com>
1. Fix issue of system blocking due to disable dcache.
2. Support Ext-SRAM-Cache mmu mapping in SMP mode.
Signed-off-by: chenwen@espressif.com <chenwen@espressif.com>
__revoke_low_memory() is called in intel64_lowsetup()
fixes b4b96a6435 (PR #11758) in which the multiboot2 header was accessed
after revoking the low memory which caused page fault.
Signed-off-by: p-szafonimateusz <p-szafonimateusz@xiaomi.com>
BSS nulling can now be optimized by the compiler, so it is necessary
to enable SSE instructions early in __nxstart
Signed-off-by: p-szafonimateusz <p-szafonimateusz@xiaomi.com>
The stack pointer must be aligned to 16 bytes, otherwise the system crash on the first unaligned data access with vector instruction.
The problem is only observable with optimization enabled, when vector instructions are generated.
Signed-off-by: p-szafonimateusz <p-szafonimateusz@xiaomi.com>
Signed-off-by: Bowen Wang <wangbowen6@xiaomi.com>
this is a follow up to the change from 2335b69120 which missed
updating stack frame length for this memset
Signed-off-by: p-szafonimateusz <p-szafonimateusz@xiaomi.com>
During removal of F1 related stuff, code that configures FLASH
latency was removed, which rendered some of the F3 line unbootable.
It was done by mistake, since previous removed block was
'#ifdef VALUE_LINE', and block with FLASH code was '#ifndef VALUE_LINE'
and so it should not have been removed.
Signed-off-by: Michał Łyszczek <michal.lyszczek@bofc.pl>
