Re: fcntl(fd, F_SETFL, O_NONBLOCK) on Local Socket not working

2022-11-03 Thread Xiang Xiao
In the old design, the driver/socket needed to return -ENOTTY to activate the default action(in file_vioctl) even if they handle O_NONBLOCK internally. This approach confuses many people and introduces the error you hit several times. So https://github.com/apache/incubator-nuttx/pull/6976 fixes thi

Re: ARM -m startup does not check for idle thread stack space

2022-11-03 Thread Alan C. Assis
Hi Carlos, I tested and confirmed the issue: $ patch -p1 < ~/Desktop/0001-Changes-to-force-fill-bss-section-causing-sp-past-th.patch patching file arch/arm/src/armv7-m/arm_vectors.c $ ./tools/configure.sh lm3s6965-ek:qemu-flat $ make -j8 $ qemu-system-arm -semihosting \ -M lm3s6965ev

Re: fcntl(fd, F_SETFL, O_NONBLOCK) on Local Socket not working

2022-11-03 Thread Jernej Turnsek
I think I have a fix to this problem. If I change the initialization value of the variable ret in function local_ioctl (file local_sockif.c) from OK to -ENOTTY (like the fix in netdev_ioctl), it starts to work in my case. I think it should not be a problem for other cases. What do you think? On Th

Re: fcntl(fd, F_SETFL, O_NONBLOCK) on Local Socket not working

2022-11-03 Thread Jernej Turnsek
My branch is not the latest one, but I have checked the code which PR is fixing and I have the latest one. I think the problem is with the local_ioctl function, where there is no code to properly set s_flags in case of FIONBIO. On Thu, Nov 3, 2022 at 7:04 PM Xiang Xiao wrote: > Does your branch

write-only interrupt registers

2022-11-03 Thread TimH
In the midst (still!) of sorting out the SAMA5D2 TSD and ADC support. It is pretty much all working now, with quite a few errors addressed along the way. Guess I'm the only one actually using this processor for real LOL. The last thing I am addressing is a conflict between the interrupts used for

Re: fcntl(fd, F_SETFL, O_NONBLOCK) on Local Socket not working

2022-11-03 Thread Xiang Xiao
Does your branch contain this patch? https://github.com/apache/incubator-nuttx/pull/5933 On Fri, Nov 4, 2022 at 1:11 AM Jernej Turnsek wrote: > Hi, > > I am trying to set O_NONBLOCK flag with the help of fcntl on Local Socket > structure (AF_UNIX) and I am not getting the non blocking functional

fcntl(fd, F_SETFL, O_NONBLOCK) on Local Socket not working

2022-11-03 Thread Jernej Turnsek
Hi, I am trying to set O_NONBLOCK flag with the help of fcntl on Local Socket structure (AF_UNIX) and I am not getting the non blocking functionality. If setting the SOCK_NONBLOCK flag while creating the socket, it is working fine. I am using some Linux based code, where this functionality is work

Re: libpq and libscpi: how to start

2022-11-03 Thread Alan C. Assis
Hi Sebastien, Good question, AFAIK there is not yet a command to just download the necessary tar balls and keep them in a downloaded area to use later. If I remember correctly some build systems like Buildroot and Yocto has this option. BR, Alan On 11/3/22, Sebastien Lorquet wrote: > Hi, > >

Re: libpq and libscpi: how to start

2022-11-03 Thread Sebastien Lorquet
Hi, littlefs too. This raises an important question for me: how do I handle such a build in a self-contained, "airgapped" machine? Are there provisions to use a pre-downloaded tarball for external projects? Sebastien Le 28/10/2022 à 18:30, Alan C. Assis a écrit : Hi Sebastien, AFAIK there

Fwd: ARM -m startup does not check for idle thread stack space

2022-11-03 Thread Carlos Sanchez
Hi all, I have detected the armv{6,7,8}-m/arm_vectors.c assigns the initial stack pointer value (using the first entry at the exception vectors table) just to _ebss + CONFIG_IDLETHREAD_STACKSIZE, without further checking. This can yield to an unusable binary if .data or .bss sections grow too much