Re: [I] [BUG] Potential pointer arithmetic issue? [nuttx]
acassis commented on issue #16386:
URL: https://github.com/apache/nuttx/issues/16386#issuecomment-5021632212
> > [@emomaxd](https://github.com/emomaxd) if you can, please submit your
port to mainline, it could help more people to experiment and it ends up
evolving the support faster
>
> [@acassis](https://github.com/acassis), sorry for the late reply. I
noticed your comment again while reviewing my previous issues.
>
> You were also right about the cache state being involved in the original
problem; I documented the result
[here](https://github.com/apache/nuttx/issues/16384#issuecomment-4111318286).
>
> The TI AM67A port is still under development in the [T3 Gemstone
fork](https://github.com/t3gemstone/nuttx). I am no longer involved in that
project, so the decision to upstream it belongs to its current maintainers.
>
> Separately, I now have initial RK3588S support booting on a 4 GiB Orange
Pi 5 Pro. It reaches NSH with a polling UART2 console, the ARM generic timer
works, PSCI reset works, and OSTest completes successfully on the board.
>
> Peripheral drivers, SMP, and device-tree consumption are not implemented
yet. My preference is to submit the tested minimal port first and add those
capabilities in focused follow-up PRs.
>
> Would you consider that a reasonable scope for the initial PR, or would
you prefer to see additional peripheral support first?
Understood, I will ping the project.
Cool, I think RK3588S will be very useful. I think I have a small portable
game player powered by this processor.
It is better to submit the minimal port, it will be easier to review.
("Release early, release often")
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
Re: [I] [BUG] Potential pointer arithmetic issue? [nuttx]
emomaxd commented on issue #16386: URL: https://github.com/apache/nuttx/issues/16386#issuecomment-5021139762 > [@emomaxd](https://github.com/emomaxd) if you can, please submit your port to mainline, it could help more people to experiment and it ends up evolving the support faster @acassis, sorry for the late reply. I noticed your comment again while reviewing my previous issues. You were also right about the cache state being involved in the original problem; I documented the result [here](https://github.com/apache/nuttx/issues/16384#issuecomment-4111318286). The TI AM67A port is still under development in the [T3 Gemstone fork](https://github.com/t3gemstone/nuttx). I am no longer involved in that project, so the decision to upstream it belongs to its current maintainers. Separately, I now have initial RK3588S support booting on a 4 GiB Orange Pi 5 Pro. It reaches NSH with a polling UART2 console, the ARM generic timer works, PSCI reset works, and OSTest completes successfully on the board. Peripheral drivers, SMP, and device-tree consumption are not implemented yet. My preference is to submit the tested minimal port first and add those capabilities in focused follow-up PRs. Would you consider that a reasonable scope for the initial PR, or would you prefer to see additional peripheral support first? -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
Re: [I] [BUG] Potential pointer arithmetic issue? [nuttx]
acassis commented on issue #16386: URL: https://github.com/apache/nuttx/issues/16386#issuecomment-2984159211 @emomaxd if you can, please submit your port to mainline, it could help more people to experiment and it ends up evolving the support faster -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
Re: [I] [BUG] Potential pointer arithmetic issue? [nuttx]
emomaxd closed issue #16386: [BUG] Potential pointer arithmetic issue? URL: https://github.com/apache/nuttx/issues/16386 -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
Re: [I] [BUG] Potential pointer arithmetic issue? [nuttx]
linguini1 commented on issue #16386: URL: https://github.com/apache/nuttx/issues/16386#issuecomment-2982059805 > @linguini1 I don't work on this anymore, so I can't help with that. Can the issue be closed then? -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
Re: [I] [BUG] Potential pointer arithmetic issue? [nuttx]
emomaxd commented on issue #16386: URL: https://github.com/apache/nuttx/issues/16386#issuecomment-2982058318 @linguini1 I don't work on this anymore, so I can't help with that. -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
Re: [I] [BUG] Potential pointer arithmetic issue? [nuttx]
linguini1 commented on issue #16386: URL: https://github.com/apache/nuttx/issues/16386#issuecomment-2977973662 @emomaxd Any updates on this? -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
Re: [I] [BUG] Potential pointer arithmetic issue? [nuttx]
emomaxd commented on issue #16386: URL: https://github.com/apache/nuttx/issues/16386#issuecomment-2888465254 Thanks for your concern, I’ll let you know on Monday. -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
Re: [I] [BUG] Potential pointer arithmetic issue? [nuttx]
linguini1 commented on issue #16386: URL: https://github.com/apache/nuttx/issues/16386#issuecomment-2888415655 I am also not too sure here. As others pointed out, `volatile uint32_t *stk = (void *)*g_idle_topstack;` is an odd expression. Can you confirm that `volatile uint32_t *stk = (void *)g_idle_topstack;` also didn't work? -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
Re: [I] [BUG] Potential pointer arithmetic issue? [nuttx]
acassis commented on issue #16386: URL: https://github.com/apache/nuttx/issues/16386#issuecomment-2888400647 @emomaxd thank you for explanation. Could it be something related to cache been enabled? Then when you try to do a right boolean operation the right value is read. @linguini1 since you did a NuttX port to Raspberry Pi 4 recently, do you have some suggestion that could help @emomaxd here? -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
Re: [I] [BUG] Potential pointer arithmetic issue? [nuttx]
emomaxd commented on issue #16386: URL: https://github.com/apache/nuttx/issues/16386#issuecomment-2888393948 @acassis, My goal is to get NuttX running correctly on the R5F core of the TI AM67A SoC—nothing more. Currently, it crashes at nx_start(), as shown in the attached image. Some modifications in the code may look unusual (size, stk, ptr) , but they are only there for debugging purposes and to highlight the issue—they’re not part of the intended logic. The original upstream code doesn’t work on the R5F core. To fix this, I added an & before g_idle_topstack, which made it run correctly. I've also tested with different compilers, but the issue persists. At this point, I suspect a possible hardware-related issue, but I’m not sure how to approach it or rule it out. -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
Re: [I] [BUG] Potential pointer arithmetic issue? [nuttx]
acassis commented on issue #16386: URL: https://github.com/apache/nuttx/issues/16386#issuecomment-2888388724 @emomaxd could you please explain what you want to accomplish not how you want to do it? Maybe if we understand your goal we could help better. Also since you are using a custom board that you cannot reveal to us (I get it is from TI), please try to duplicate this issue in a common board like the stm32f4discover. -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
