nuttxpr commented on PR #15380:
URL: https://github.com/apache/nuttx/pull/15380#issuecomment-2564948663

   [**\[Experimental Bot, please feedback 
here\]**](https://github.com/search?q=repo%3Aapache%2Fnuttx+13552&type=issues)
   
   
   
   No, this PR description does not fully meet the NuttX requirements.  Here's 
why and how to improve it:
   
   **Missing Information in Summary:**
   
   * **What functional part of the code is being changed?**  Be specific.  
e.g., "The TCP/IP stack, specifically the `accept()` function implementation."
   * **How does the change exactly work?** Explain the code changes you made to 
prevent the listening socket from being notified.  For example, "The change 
prevents propagation of the TCP_RESET event from the accepted socket to the 
listening socket by [explain the mechanism, e.g., modifying a flag, changing an 
if condition, etc.]."
   
   **Missing Information in Impact:**
   
   Fill out *all* impact sections. Even if the answer is "NO," explicitly state 
it.  This PR likely has implications for:
   
   * **Is new feature added? Is existing feature changed?** This is a bug fix, 
so "Existing feature changed" should be marked. Briefly describe the original 
faulty behavior.
   * **Impact on user:** Potentially YES. Explain that the incorrect 
POLLERR/POLLHUP behavior is fixed, which might affect applications relying on 
(or misinterpreting) the previous behavior.  Explain how users who *were* 
relying on the bug should adapt.
   * **Impact on documentation:**  If the previous behavior was documented 
(even implicitly),  "YES" is needed. Briefly explain what documentation needs 
updating.  If no documentation changes are needed, explicitly state "NO - The 
change corrects undocumented, erroneous behavior."
   * **Impact on compatibility:** Potentially YES if applications were relying 
on the bug. State "Potentially YES. Applications incorrectly relying on the 
previous POLLERR/POLLHUP behavior on the listening socket after a TCP_RESET on 
the accepted socket will need to be adjusted."  If you're confident there's no 
compatibility impact, explain why.
   
   
   **Missing Information in Testing:**
   
   * **Build Host(s):** Be specific. Example: "Linux Ubuntu 20.04, x86_64, GCC 
9.4.0"
   * **Target(s):** Be *very* specific. Example: "STM32F746NGH6 (Nucleo-F746ZG 
board), NuttX configuration: `nsh`"
   * **Testing logs before change:** Include actual logs demonstrating the 
problem.  Show that the listening socket receives POLLERR/POLLHUP when a client 
connection resets.
   * **Testing logs after change:** Include logs demonstrating that the 
listening socket *no longer* receives POLLERR/POLLHUP under the same reset 
conditions.  Show that the accepted socket handles the reset gracefully.
   
   
   
   **Example of an Improved Summary:**
   
   The TCP/IP stack, specifically the `accept()` function implementation, is 
changed to correct the handling of TCP_RESET events. Previously, when a client 
connection reset after being accepted, the listening socket was also notified 
of the event, resulting in POLLERR and POLLHUP being returned by `poll()`. This 
change isolates the TCP_RESET event to the accepted socket, preventing this 
incorrect notification to the listening socket.  This fixes unexpected behavior 
that could lead to incorrect application logic.  Fixes [NuttX Issue # if 
applicable].
   
   
   By providing this more complete information, your PR will be much easier to 
review and understand, increasing its chances of being merged quickly.
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: commits-unsubscr...@nuttx.apache.org

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

Reply via email to