Support multi-threading in the Patmos Platform

[EhsanKhodadad]

No description provided.

[Copilot]

Pull request overview

This PR aims to add multi-threaded runtime support for the Patmos platform by introducing Patmos implementations of the LF threading/synchronization abstractions when LF_SINGLE_THREADED is not defined.

Changes:

• Added a Patmos “threaded” branch in lf_patmos_support.c with implementations for threads, mutexes, and condition variables.
• Added Patmos platform typedefs for lf_thread_t, lf_mutex_t, and lf_cond_t under !LF_SINGLE_THREADED.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 9 comments.

File	Description
low_level_platform/impl/src/lf_patmos_support.c	Adds Patmos-side threaded implementations (threads/mutex/cond) and related helpers under #else of LF_SINGLE_THREADED.
low_level_platform/api/platform/lf_patmos_support.h	Introduces threaded Patmos typedefs and includes <pthread.h> when !LF_SINGLE_THREADED.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[edwardalee]

Looks to me like the Copilot suggestions need to be addressed.

[Copilot]

Pull request overview

Copilot reviewed 3 out of 3 changed files in this pull request and generated 5 comments.

[edwardalee]

As you address issues raised by Copilot, please record a brief comment on how the issue was addressed and mark the comment "Resolved".

[Copilot]

Pull request overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated 6 comments.

[EhsanKhodadad]

As you address issues raised by Copilot, please record a brief comment on how the issue was addressed and mark the comment "Resolved".

@edwardalee I addressed all Copilot's issues. I feel it is now ready for your review.

[Copilot]

Pull request overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated 1 comment.

[edwardalee]

Looks OK to me, except it looks like the call to _lf_patmos_global_lock_acquire should be implemented in the Patmos-specific implementation of lf_disable_interrupts_nested rather than conditionally called in lf_atomic_irq.c‎.

[EhsanKhodadad]

Looks OK to me, except it looks like the call to _lf_patmos_global_lock_acquire should be implemented in the Patmos-specific implementation of lf_disable_interrupts_nested rather than conditionally called in lf_atomic_irq.c‎.

Thank you for your review, @edwardalee, and sorry for my delayed answer.

I added _lf_patmos_global_lock_release function here because, as discussed in #582 (comment), lf_disable_interrupts_nested() only disables interrupts on the current core and does not provide mutual exclusion across cores.

Can we implement these functions in a separate file, named lf_atomic_patmos.c?

[edwardalee]

Looks OK to me, except it looks like the call to _lf_patmos_global_lock_acquire should be implemented in the Patmos-specific implementation of lf_disable_interrupts_nested rather than conditionally called in lf_atomic_irq.c‎.

Thank you for your review, @edwardalee, and sorry for my delayed answer.

I added _lf_patmos_global_lock_release function here because, as discussed in #582 (comment), lf_disable_interrupts_nested() only disables interrupts on the current core and does not provide mutual exclusion across cores.

Can we implement these functions in a separate file, named lf_atomic_patmos.c?

Is lf_disable_interrupts_nested used anywhere else? If not, then with careful documentation, I think my suggestion could still make sense and it would avoid putting Patmos-specific code in the platform-independent file. Alternatively, maybe there is a better name for the function lf_disable_interrupts_nested that abstracts the functionality?

[EhsanKhodadad]

Looks OK to me, except it looks like the call to _lf_patmos_global_lock_acquire should be implemented in the Patmos-specific implementation of lf_disable_interrupts_nested rather than conditionally called in lf_atomic_irq.c‎.

Thank you for your review, @edwardalee, and sorry for my delayed answer.
I added _lf_patmos_global_lock_release function here because, as discussed in #582 (comment), lf_disable_interrupts_nested() only disables interrupts on the current core and does not provide mutual exclusion across cores.
Can we implement these functions in a separate file, named lf_atomic_patmos.c?

Is lf_disable_interrupts_nested used anywhere else? If not, then with careful documentation, I think my suggestion could still make sense and it would avoid putting Patmos-specific code in the platform-independent file. Alternatively, maybe there is a better name for the function lf_disable_interrupts_nested that abstracts the functionality?

Thank you for your feedback, @edwardalee , I edited my PR in 6dc4221, so we have no changes in lf_atomic_irq.c in this PR.