Hi NuttXers! I've just submitted a proposal for making NuttX's pipe (FIFO) POSIX-compliant regarding `open` for write-only when there are no readers.
Nowadays, opening a FIFO for write-only (with O_NONBLOCK cleared) won't block even if there are no readers. Only opening for read-only (with no writers) would block. This is not a POSIX-compliant behavior. According to https://pubs.opengroup.org/onlinepubs/9699919799/functions/open.html: When opening a FIFO with O_RDONLY or O_WRONLY set: > - If O_NONBLOCK is set, an open() for reading-only shall return > without delay. An open() for writing-only shall return an error if no > process currently has the file open for reading. > - If O_NONBLOCK is clear, an open() for reading-only shall block the > calling thread until a thread opens the file for writing. An open() for > writing-only shall block the calling thread until a thread opens the file > for reading. > > It would be a *possible* breaking change if any application was written specifically to expect an `open` for write-only (and `O_NONBLOCK` cleared, of course) to be non-blocking even if there are no readers. Being able to port POSIX-compliant applications that expect `open` to block when opening for write-only with no readers would enhance NuttX's objective of "to achieve a high degree of standards compliance. The primary governing standards are POSIX and ANSI standards". That is the case for [RTP Tools]( https://github.com/apache/nuttx-apps/pull/1651), for instance. So, please consider the following changes on PRs: 1. https://github.com/apache/nuttx/pull/8985 2. https://github.com/apache/nuttx-apps/pull/1707 Let me know if you think it isn't advisable to accept the proposed changes! Best regards, -- Tiago Medicci Serrano Embedded Software Engineer MSc Electronics/Microelectronics m: +55 (19) 981403886 <+55+(19)+981403886> e: tiago.medi...@gmail.com a: Campinas, Brazil Follow me: <https://www.linkedin.com/in/tiago-serrano-924458b6> <https://github.com/tmedicci>
