Virtio RNG support (CONFIG_DRIVERS_VIRTIO_RNG=y) selects
CONFIG_ARCH_HAVE_RNG. On the other hand, if CONFIG_DEV_URANDOM=y,
it defaults to CONFIG_DEV_URANDOM_ARCH if CONFIG_ARCH_HAVE_RNG=y.
DEV_URANDOM_ARCH definition states that the implementation of the
/dev/urandom should be provided by the architecture-specifig logic,
including the function devurandom_register(). In this case, the
/dev/urandom may refer to the same driver as /dev/random that is
provided by the Virtio RNG driver, which is implemented by this
commit.
Notes:
1. This commit do the same thing as TCP did: https://github.com/apache/nuttx/pull/10627
2. UDP uses `iob_navail(false)` but TCP uses `iob_navail(true)`, this is because of a problem related to TCP recv window (https://github.com/apache/nuttx/pull/4142), so we don't need to change UDP now.
Signed-off-by: Zhe Weng <wengzhe@xiaomi.com>
Summary:
There are differences in different server implementations, so it is necessary to check whether the returned nwqid satisfies the requested number.
Signed-off-by: chenrun1 <chenrun1@xiaomi.com>
By default in SMP, obtaining this_task requires disabling interrupts, obtaining the current CPU index, accessing a global variable, and re-enabling interrupts. Storing this_task in percpu makes retrieval faster.
Signed-off-by: liwenxiang1 <liwenxiang1@xiaomi.com>
1. Change the CONFIG_RPMSG_VIRTIO to CONFIG_RPMSG_VIRTIO_LITE
2. fix the error when modify the sim_rpmsg_virtio to port to
rpsmg_virtio_lite;
Signed-off-by: Bowen Wang <wangbowen6@xiaomi.com>
CONFIG_DEV_ZERO is no longer needed in some defconfig files due to
- https://github.com/apache/nuttx/pull/15223
This PR removes CONFIG_DEV_ZERO from the affected defconfig files:
```bash
make distclean ; tools/refresh.sh --silent canmv230:knsh
make distclean ; tools/refresh.sh --silent canmv230:master
make distclean ; tools/refresh.sh --silent canmv230:nsbi
...
```
The failed builds above were extracted from [nuttx-prometheus-to-mastodon.json](https://lupyuen.org/articles/mastodon.html#appendix-post-nuttx-builds-to-mastodon)
when accept conn receives TCP_RESET, only accept conn itself should handle
the event, and there is no need to notify the listening conn. otherwise,
the server's poll will return with POLLERR and POLLHUP. this may cause
some applications to execute incorrect logic in the future.
Signed-off-by: zhanghongyu <zhanghongyu@xiaomi.com>
/dev/zero is a very commonly used config, for example use
/dev/zero to test the filesystem, mtd or block devices performance.
So let's default enable it when not enable DEFAULT_SMALL
Signed-off-by: Bowen Wang <wangbowen6@xiaomi.com>
* Add a migration warning to the update release system and the elf
programs documentation pages. This is just to add a papertrail from
where the documentation originate in case of error during the migration
process.
* Improve the building nuttx with app out of source tree
* Add new guide pages
* Building uclibc++
* Custom app directories
* Debugging ELF loadable modules
* Multiple NSH sessions
* NSH network link management
* RAM & ROM disks
* Reading CAN messages
* Remove device drivers with NSH
1. remove up_interrupt_context() check, which should be safe in interrupt context
2. remove sem instance check will be handle in nxsem_trywait()
Signed-off-by: chao an <anchao@lixiang.com>
Add a migration warning to the update release system and the elf
programs documentation pages. This is just to add a papertrail from
where the documentation originate in case of error during the migration
process. Also fix a bit of formatting in the pages
Because rpmsg_virtio_lite is a better name to inform user that the
rpmsg virtio (original name) is a lite implementation of rpmsg virtio.
Signed-off-by: Bowen Wang <wangbowen6@xiaomi.com>
Register addresses are defined. Bit defintions are still needed.
Bitmaps for OCTOSPI peripheral register
Incremental updates to qspi files
Redefined QSPI/OCTOSPI registers for STM32H5
Fixed register definitions for the STM32H5. However, more work still needs to be done. Some bits that shared registers on the STM32H7 have different register locations on the H5. This still needs to be accounted for.
Fixed more differences vs STM32H7 qspi
Added ifdef for including stm32_dma.h in stm32_qspi.c. Added stm32_qspi.c to Make.defs.
Register fixes. SPI activity but not able to format device yet.
Fixed DCYC mask
Set HCLK frequency to correct value. Undid ccrconfig debug.
Tested Interrupt Mode (single SPI). Added alternate bytes to meminfo and cmdinfo structures. Updated Kconfig variables for STM32H5.
Fixed base register in hardware/stm32_qspi.h. Updated qspi_dumpregs.
The base register was previously set to STM32_QUADSPI_BASE, changed to the correctly named STM32_OCTOSPI1_BASE. However, these defines for the OCTOSPI registers are not even used. Instead qspi_putreg and qspi_getreg utilize the priv->base value to access OCTOSPI registers.
Removed altbytes code, left as before. Moved QSPI clock selection to stm32h5xx_rcc.c
Changed STM32H5_QUADSPI to STM32H5_QSPI1
Added hook to define QSPI_CLK_FREQUENCY as STM32_QSPI_FREQUENCY from board.h
Removed changes to nuttx qspi.h
style fixes
When CONFIG_LIBC_ARCH_ATOMIC is defined, the atomic_fetch_x
interface conflicts with the atomic_fetch_x interface in libcxx.
Therefore, it has been changed to a inline function.
Signed-off-by: zhangyuan29 <zhangyuan29@xiaomi.com>
chip/efm32_leserial.c: In function 'efm32_restoreuartint':
Error: chip/efm32_leserial.c:329:37: error: 'len' undeclared (first use in this function); did you mean 'ien'?
329 | efm32_restoreuartint_nolock(priv, len);
| ^~~
| ien
Signed-off-by: hujun5 <hujun5@xiaomi.com>
There were two simple errors when running ci tests on pull #15351. These errors were unrelated to that pull. After investigating I found two simple errors. 1. The len variable was likely a typo that should have been the function argument, ien. 2. The spinlock_t lock was defined in a section dependent on CONFIG_SERIAL_TERMIOS. However, the spinlock_t lock was used in code that did not require the SERIAL_TERMIOS config. Therefore errors resulted.
imxrt_edma_s missing lock variable
CI build tests noted that the variable g_edma was referencing a lock variable that did not exist in the structure. This change adds that missing lock as a spinlock_t type.
Remove unused flags variable from imxrt_dmach_stop
Removed extra =
Otherwise the free holder list will leak, causing either a crash due to
holder->htcb = NULL, or the free holder list becomes (erroneously) empty
even though most of the holder entries are free.
Before the MPU initialization, the up_update_task(this_cpu()) function is called at a time when hardware cache coherency is not yet enabled.
In certain critical scenarios, Core 1 reads a zero value for tcb from the global variable g_assignedtask and stores this zero value into the tpidr
register. This results in subsequent interrupt handlers reading a zero tcb, causing an exception.
Signed-off-by: hujun5 <hujun5@xiaomi.com>
the same variable type in linux is char, in order to avoid modifying
the third-party library code when porting the third-party library code.
Signed-off-by: zhanghongyu <zhanghongyu@xiaomi.com>
