Hi all, Way back, Espressif introduced the esp-hal-3rdparty repository for NuttX, and here we are today. A few weeks ago, Tiago mentioned that Espressif would follow whatever direction the community suggested, but now they are proposing a specific path.
If we follow this approach, where does that leave us regarding other vendors? My concern is that NuttX might become too tightly coupled with Espressif. That said, I understand the points made and won't push this further. Best regards, *--Felipe Moura de Oliveira* Linkedin <https://www.linkedin.com/in/felipe-oliveira-75a651a0> On Fri, 2 Oct 2026 at 22:13 Tomek CEDRO <[email protected]> wrote: > On Sat, Oct 3, 2026 at 12:43 AM Gregory Nutt <[email protected]> wrote: > > The last remaining question here is whether it is okay to Apache to > > accept such donation. If yes then we can create > > apache/nuttx-hal-espressif repo and work on it just as any other > > apache/nuttx-* repo. I guess this hal needs to be a separate repo > > because Espressif still want to have local clone for validation / > > development. > > > > That is an option if and if the HAL is granted to the ASF. Otherwise, > what you suggesting might be prohibited by the ASF. I would have to do > more research to understand that fully. > > Yes exactly Tiago underlined that several times that they will > "donate" their HAL to ASF so it becomes apache/nuttx-hal-espressif and > the whole community is more involved.. this is very honorable step > from Espressif and they again seem to pave the way for HAL integration > in NuttX in the least painful possible way.. at the same time taking > care of software supply chain :-) > > Most of the raised issues should be solved that way: > 1. Code ownership and supply chain security. > 2. Community involvement in faster PR processing. > 3. We should have more control and impact on the HAL quality and > availability. > 4. HAL is under the NuttX umbrella just as proposed "Zephyr Umbrella > Way" (also proposed by Renesas). > > I guess ASF should not have problem with accepting such aleady Apache > 2.0 licensed code. Maybe only ICLAs should be obtained from all > existing authors? > > -- > CeDeROM, SQ7MHZ, http://www.tomek.cedro.info >
