Hi Eduard, Yes, the modification appears correct.
Also I noticed that the "nxsem_wait_uninterruptible(&priv->txsem)" is already inside a "#ifdef CONFIG_STM32H7_SPI_DMA", so your modification will make the logic symmetric. You just need to create a fork, do your modification and submit a PR. BR, Alan On 10/16/20, Eduard Niesner <niesneredu...@gmail.com> wrote: > I attached it as .txt. > > *Please note:* I made the code build and the SPI seems to work as expected > - I am communicating with an at45db flash over SPI and mounted smartFS on > it and the communication seems to work. I am not sure if the changes that I > did are enough or if the entire spi_interrupt function should be surrounded > by the #ifdef CONFIG_STM32H7_SPI_DMA condition as well (and also where it > is called from). > It would be good if someone with more experience that understands the > impact of the change could look into it. > > PS: If the changes look ok, I will register on github and create the pull > request. Do I need to get any approvals to create branches or pull > requests on nuttx incubator? > > Edi > > > > On Fri, Oct 16, 2020 at 7:28 PM Alan Carvalho de Assis <acas...@gmail.com> > wrote: > >> Hi Eduard, >> >> Unfortunately the mailing list is refusing patches with extension >> .patch, we need to rename it to .txt to get it here. >> >> BTW, you can submit a Pull Request directly to >> https://github.com/apache/incubator-nuttx and we could review it. >> >> BR, >> >> Alan >> >> On 10/16/20, Eduard Niesner <niesneredu...@gmail.com> wrote: >> > Hi all, >> > >> > I am not familiar with the code from >> nuttx\arch\arm\src\stm32h7\stm32_spi.c >> > but I believe that there is an issue. >> > "txresult" is defined and used only if the CONFIG_STM32H7_SPI_DMA is >> > defined. >> > But in the spi_interrupt function, the txresult is used regardless of >> > whether CONFIG_STM32H7_SPI_DMA is defined or not. >> > >> > This generates a build issue when you configure SPI without >> > CONFIG_STM32H7_SPI_DMA. >> > >> > Is there anyone that knows more about this? >> > I implemented a fix and it seems to be working - but since I am not >> > familiar with the code I am not sure if this is the right thing to do. >> > I attached a patch. >> > >> > Thanks, >> > Edi >> > >> >
