fix compile error

[hujun260]

Summary

fix compile error

riscv-none-elf-ld: warning: /home/hujun5/downloads1/vela_sim/nuttx/nuttx has a LOAD segment with RWX permissions
riscv-none-elf-ld: /home/hujun5/downloads1/vela_sim/nuttx/staging/libarch.a(riscv_modifyreg32.o): in function spin_lock_irqsave_wo_note': /home/hujun5/downloads1/vela_sim/nuttx/include/nuttx/spinlock.h:534:(.text.modifyreg32+0x1e): undefined reference to this_cpu'
riscv-none-elf-ld: /home/hujun5/downloads1/vela_sim/nuttx/staging/libarch.a(riscv_modifyreg32.o): in function spin_unlock_irqrestore_wo_note': /home/hujun5/downloads1/vela_sim/nuttx/include/nuttx/spinlock.h:701:(.text.modifyreg32+0x82): undefined reference to this_cpu'
riscv-none-elf-ld: /home/hujun5/downloads1/vela_sim/nuttx/staging/libarch.a(bl602_serial.o): in function spin_lock_irqsave_wo_note': /home/hujun5/downloads1/vela_sim/nuttx/include/nuttx/spinlock.h:534:(.text.up_putc+0x12): undefined reference to this_cpu'
riscv-none-elf-ld: /home/hujun5/downloads1/vela_sim/nuttx/staging/libarch.a(bl602_serial.o): in function spin_unlock_irqrestore_wo_note': /home/hujun5/downloads1/vela_sim/nuttx/include/nuttx/spinlock.h:701:(.text.up_putc+0x84): undefined reference to this_cpu'
riscv-none-elf-ld: /home/hujun5/downloads1/vela_sim/nuttx/staging/libsched.a(irq_attach.o): in function spin_lock_irqsave_wo_note': /home/hujun5/downloads1/vela_sim/nuttx/include/nuttx/spinlock.h:534:(.text.irq_attach+0x26): undefined reference to this_cpu'
riscv-none-elf-ld: /home/hujun5/downloads1/vela_sim/nuttx/staging/libsched.a(irq_attach.o):/home/hujun5/downloads1/vela_sim/nuttx/include/nuttx/spinlock.h:701: more undefined references to `this_cpu' follow

Impact

none

Testing

ci

[nuttxpr]

[Experimental Bot, please feedback here]

The PR summary needs to be more descriptive. Here's a breakdown of the issues and how to improve the PR:

Issues:

• Vague Summary: "fix compile error" doesn't explain what compile error or how it's being fixed.
• Missing Information: The "Impact" section is overly simplistic. While the user impact might be minimal, there are likely impacts on the build (at least for the RISC-V architecture).
• Insufficient Testing: "ci" isn't enough detail. You need to provide:
  • Specific build hosts and target configurations used for testing.
  • Relevant snippets of logs from before and after the change, showing the error being resolved.

Improved PR Summary:

## Summary

This PR fixes a compile-time error encountered on the RISC-V architecture due to missing `this_cpu` symbol references within spinlock functions. The issue arises from the use of `spin_lock_irqsave_wo_note` and `spin_unlock_irqrestore_wo_note` in specific code paths where the `this_cpu` variable is not defined. 

This PR resolves the issue by [briefly explain your solution, e.g., "ensuring the `this_cpu` definition is available in the affected files/modules" or "refactoring the code to avoid direct use of `this_cpu` in those specific instances."]. 

**Related Issue:** [If applicable, link to a NuttX issue]

## Impact

* **Is new feature added? Is existing feature changed?** No.
* **Impact on user (will user need to adapt to change)?** NO 
* **Impact on build (will build process change)?** YES. This fixes a build error specifically affecting the RISC-V architecture.
* **Impact on hardware (will arch(s) / board(s) / driver(s) change)?** NO
* **Impact on documentation (is update required / provided)?** NO 
* **Impact on security (any sort of implications)?** NO 
* **Impact on compatibility (backward/forward/interoperability)?** NO 
* **Anything else to consider?**  N/A

## Testing

I confirm that changes are verified on the following setups:

* **Build Host:** [e.g., Ubuntu 20.04, GCC 10.2.1]
* **Target:**  [e.g., RISC-V simulator, QEMU, specific RISC-V board:config]

**Testing logs before change:**

[Paste relevant log snippet showing the compile errors]

**Testing logs after change:**

[Paste relevant log snippet showing successful compilation]

Key Points:

• Be clear, concise, and provide specific details.
• Focus on what the change is, why it's needed, and how it was implemented.
• Provide concrete evidence of testing to demonstrate the fix.

[GUIDINGLI]

define the up cupu index to zero when not in SMP mode

[hujun260]

up_cpu_index may not zero when in AMP mode

[xiaoxiang781216]

why not revert #14161? or even #14190 since #14200 fix the dead lock.