boards/risc-v: add NTFC test for ksmp64

[imeghar2408-max]

Assisted-by: GitHub Copilot

Summary

Add NTFC (NuttX Test Framework Configuration) support for the existing RISC-V ksmp64 kernel SMP configuration (rv-virt:ksmp64).

This enables automated CI test coverage for the SMP kernel build, exercising ostest and SMP-specific test cases across 4 harts under QEMU.

Testing

Build

$ ./tools/configure.sh rv-virt:ksmp64
$ make -j$(nproc)

Verified key configuration options:

CONFIG_BUILD_KERNEL=y
CONFIG_ARCH_USE_S_MODE=y
CONFIG_SMP=y
CONFIG_SMP_NCPUS=4
CONFIG_SYSTEM_NSH=y
CONFIG_TESTING_OSTEST=y
CONFIG_TESTING_SMP=y

Kernel application build/import

$ make export -j$(nproc)
$ cd ../apps
$ ./tools/mkimport.sh -z -x ../nuttx/nuttx-export-*.tar.gz
$ make import -j$(nproc)

QEMU SMP boot

$ qemu-system-riscv64 -semihosting -M virt,aclint=on -cpu rv64 -smp 4 -kernel ./nuttx -nographic

Boots cleanly to NSH with all 4 harts recognized by OpenSBI.

NTFC test run

$ NTFCDIR=/tmp/nuttx-ntfc ./tools/testbuild.sh -C -R /tmp/nuttx-ksmp64-testlist

933 tests collected, 3 skipped, exit code 0 — no failures.

Static checks

$ git diff --check
$ ./tools/checkpatch.sh -c -u -m -g "$BASE..HEAD"
$ ./tools/checkpatch.sh -c

All checks pass.

Fixes #20031

[github-actions]

MemBrowse Memory Report

No memory changes detected for:

• arduino-mega2560
• esp32-devkitc
• hifive1-revb
• mirtoo
• qemu-armv8a
• qemu-intel64
• rx65n-rsk2mb
• s698pm-dkit
• stm32-nucleo-f103rb

[imeghar2408-max]

CI is currently failing during the NTFC boot wait for the ksmp64 test. The kernel builds successfully and boots locally with 4 HARTs, but CI uses ntfc==0.0.1, whose boot wait appears to be limited to ~5s. Newer NTFC changes add configurable boot_timeout and exec_cwd.

What would be the preferred approach here — update/release NTFC first and then use those options in this PR, or handle it another way?