Re: [GitHub] [incubator-nuttx] xiaoxiang781216 commented on a change in pull request #1487: libc: Avoid ctype function to evaluate the argument more than once

Yes, you are right: function version is still 20Bytes bigger than macro version. Do you have a better solution to fix this problem? Just to remind people that my original size concern was not with the table approach.  I think the size if about a wash. Applications that do a lot of string man

Re: Some questions related PR#1478

2. Is there a way to flush a fat file system? Because I think the unexpected behaviour I've seen was related to debugging and fat cache flush. I tried to use fsync on the folder (which was proposed in a linux related discussion), but open on a folder returns errno 21 (is a directory). fsync o

Re: [GitHub] [incubator-nuttx] xiaoxiang781216 commented on a change in pull request #1487: libc: Avoid ctype function to evaluate the argument more than once

  My size concern is only in the PROTECTED build mode.  In that case, the ROM table would be duplicated:  Once in user-space and once in kernel space.  That is the wasteful part because the OS does NOT do any significant string processing and is a case where there would be an unnecessary

Re: Some questions related PR#1478

Hmmm... I don't agree. The Coding Standard encourages empty "while loops": https://cwiki.apache.org/confluence/display/NUTTX/Coding+Standard#while Why should this not applicable for "for loops"? I think there are many already in the tree. Actually, I think that is an error in the coding docum

Re: [GitHub] [incubator-nuttx] xiaoxiang781216 commented on a change in pull request #1487: libc: Avoid ctype function to evaluate the argument more than once

On 7/31/2020 8:44 AM, Xiang Xiao wrote: Macro version still evaluate the argument more than once regardless you name it kisdigit or isdigit. I am fine with any following combination: 1.lookup table for both userspace and kernel 2.normal function for both userspace and kernel 3.lookup table for u

Re: Some questions related PR#1478

My opinion: If a loop does not require a body, I suggest to use braces, rather than a semicolon, because it is less prone to programmer error. I would be okay with either: while (condition) { } or while (condition) {} And of the two, I think the second one expresses the intent best.

Re: Some questions related PR#1478

Before we change the coding standard, I suggest that we get more feedback from the dev community. My opinion: If a loop does not require a body, I suggest to use braces, rather than a semicolon, because it is less prone to programmer error. ... I should not be so opposed to coding standard ch

Re: Setting MAC address on STM32.

Thank you both. I followed the example of same70-xplained, and essentially I do the exact same thing. However, as I get it, the final step would be to call stm32_macaddress() to set the address to the MAC. This function takes as an argument a struct stm32_ethmac_s which is private to the stm32

Re: Setting MAC address on STM32.

Yes I saw this, but it does not seem a very good option to me. It requires you to set up the hardware in the application, and does not seem very efficient. Where do you initialize the IP address?  You should use the same socket descriptor to call the ioctl to set the MAC address at that locati

Re: Setting MAC address on STM32.

On the other hand, the MAC address is a hard-wired address, accessed in a hardware specific way. In my opinion the application shouldn't mess with it and it is best to be set once during the hardware initialization. The code follows the same model and uses the same IOCTL commands as does Linux

Re: Setting MAC address on STM32.

Also as Mr. Greg explained the application needs to setup the MAC address when the connection was lost and recovered. I didn't mean to imply that.  It simply has be bring the network down and then bring it back up again when the link is again detected by the PHY. The MAC address is just a

Re: stm32f7 CONFIG_RAM understanding

External SRAM? On 8/4/2020 1:08 PM, Oleg Evseev wrote: Hi, I wonder why stm32f7 boards has such RAM parameters in defconfigs: CONFIG_RAM_SIZE=245760 CONFIG_RAM_START=0x2001 Where do these numbers come from while memory map is this? изображение.png --- With regards, Oleg

Roadmap?

One of the things that I think we are missing is a Roadmap to guide and prioritize new feature development.  Other RTOS' (like Zephyr) do have such published roadmaps and are responsive to needs and requests of users and sponsors.  I have even seen web pages where the Zephyr team has laid out w

Re: USB host mass storage class automount.

just a short question: I'd like to have my USB stick attached to USB high speed host automounted as vfat. Would it be acceptable to call mount/umount directly in the usbhost_storage.c where the blockdriver is registered/unregistered? For sure switchable with Kconfig options. I tried to setup a

Re: USB host mass storage class automount.

But: I haven't yet found a way to determine if I enumerated really a mass storage class device. So with only activating mass storage it seems rather clean, but if I have multiple class drivers active, it gets complicated. Can you give a hint how to retrieve the type of driver bound to the cur

Re: USB host mass storage class automount.

Would it be acceptable that the automount handler is called from usbhost_storage.c, from a modularity point of view? Because that way it would be only a notification towards the work queue, using a mechanism that is already in place. I don't like the idea to have another kthread that does onl

Re: USB host mass storage class automount.

Would it be acceptable that the automount handler is called from usbhost_storage.c, from a modularity point of view? Because that way it would be only a notification towards the work queue, using a mechanism that is already in place. I don't like the idea to have another kthread that does onl

Re: Roadmap?

I am interested in industrial devices communicating by fieldbuses. I use the NuttX port of FreeModbus stack and I would like to see more port of communication stacks like this. I prefer CANopen and Profinet, There are several open source communication stack for these protocols. Where would we

Re: Roadmap?

I made a wiki page with the raw list of things that people are asking for in this thread: https://cwiki.apache.org/confluence/display/NUTTX/Roadmap Check it out, add anything I missed, and correct errors I made. Maybe the next step would be to try to complete this list, then maybe prioritize

Re: Roadmap?

Jira is another option available to us.

Re: Roadmap?

We probably should also include all Github Issues that are listed as enhancements. If we use a Github Project Board like Brennan suggested this would be pretty easy to link them there. Jira will link to github issues as well, as I recall.  We went through all of this in the early days of t

10.0 Release?

We finally released 9.1.0 on 2020-07-22.  I think the target date for that one was 2020-07-01, wasn't it.  The next release should then be sometime next month, right? 2020-09-01?  Or 2020-09-22? I don't know what the policy of the NuttX Incubator is, but prior to Apache, I used the major revis

Re: 10.0 Release?

Nathan started this here: https://cwiki.apache.org/confluence/display/NUTTX/NuttX+9.2 I would say let's fill that out, we can change it to 10.0 later. I will kick off getting the tracking board and the PR changes going in next week, that I thought worked quite well for realtime and async coll

Re: cpp cxx help - No thread API

Guys, I think we should move your bitbucket libcxx to github.com/nuttx to make it more official. What do you think? Other (better) option should including official support to NuttX on llvm libcxx. i would NEVER recommend github.com/nuttx for any significant use.  There are many people with

Re: 10.0 Release?

On 8/14/2020 3:39 PM, Abdelatif Guettouche wrote: We've already agreed on the version numbering. It's pretty much the same as it was + the patch number. So if there is incompatible change to the APIs, we need to bump the major number to 10. of course, there are degreses of incompatibility.  7

Trending

apache/incubator-nuttx is a GitHub trending project: https://github.com/trending/c

Re: Color ANSI support in nsh

This is my suggestion, it should be enabled by default, but disabled by default if CONFIG_DEFAULT_SMALL is selected. What do you think? I would prefer to see it just disabled by default.

Re: Color ANSI support in nsh

I would suggest that you look at how this is handled in standard terminals. This works in the form of escape sequence which give a code that the terminal can interpret as being a color. This would make it available to applications in a standard form. Good point.  The GNU part of GNU/Linux d

Re: Color ANSI support in nsh

A basic terminfo/termcaps  could be very trivial A more general solution would be handy but what is there really in the embedded world that doesn't support vt100? There more code there is, the more interesting ways it can go wrong :-) I have sometimes used cat /dev/ttyS0 on the host as a termi

Keyboard FeatherWing

Adam, Are you still working a the Giant Board?  This would be a good companion board: https://www.tindie.com/products/arturo182/keyboard-featherwing-qwerty-keyboard-26-lcd/ There is a "Solder Party" channel forum on discord.com.  Groguard discusses using the Giant Boards with the keyboard Fe

NuttX project on Hackaday

Matias' bike computer project is featured on Hackaday: https://hackaday.com/2020/08/20/bike-computer-powers-on-long-after-your-legs-give-out/ Also interesting, but off topic, ESP32 Altair Emulator: https://hackaday.com/2020/08/20/esp32-altair-emulator-gets-split-personality/

Re: defining a BLE GATT server

You're right, maybe I'm caught on the callback interface because that is how it is implemented now, but in the end there's always a packet you process. I still wonder if also having L2CAP in kernel would be better since it deals with multiple connection handling, packet fragmentation and so on

Re: defining a BLE GATT server

Also, I understand that Xiaomi intends to replace the entire Bluetooth stack with a newer BT 5.0 stack.  I don't know if that is still a plan of record or not, but you probably should coordinate Bluetooth architectural changes with Xiao Xiang in any event.

Re: defining a BLE GATT server

Thanks for the heads up. Would be good to get Xiao Xiang's input on this then. Maybe he already encountered this issue. You should also be aware of the wireless architectural principles of https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=139629397 Some of the things you hav

Re: defining a BLE GATT server

On 8/24/2020 2:20 PM, Brennan Ashton wrote: On Mon, Aug 24, 2020 at 12:34 PM Gregory Nutt wrote: Some of the things you have speculated about would be contrary to these principles. The basic wireless rules are: 1. Network sockets are used to communicate with complex wireless networks (like

Re: defining a BLE GATT server

I have been puzzling through the Goobledegook GATT server (GGK, https://github.com/moovel/gatt-server fork). It uses the Bluez D-Bus interface. It is GPLv3 so not a candidate for porting. And also not much use for understanding APIs since the OS interface is hidden inside of the Bluez D-Bus int

Re: LVGL moving to Kconfig, how we could integrated better with them

Of course it would be cool if you could simply do "make menuconfig" on NuttX and have LVGL options appear there, but I believe that would couple LVGL and NuttX build systems which may not be the best idea. I'm not sure if that can be done cleanly (it would always produce one .config and config.

Re: 答复: [External Mail]Re: defining a BLE GATT server

I will add my two cents as well.  I have already responded in comments on PR #1651, but I will copy my reponse here form continuity: I am in 100% agreement with Brennan and Matias. I am very opposed to this change because it violates all of the architectural principles that have been establish

Re: LPWORK Queue and USB Host Serial

A short question concerning the use of LPWORK queue in FT232R and CDCACM USB Host drivers: Both block one LPWORK task per opened instance permanently. The FT232R rxdata_work USB transfer returns after each 16ms (USBHOST_FT232R_LATENCY) but since there are always at least two status bytes tra

Re: LPWORK Queue and USB Host Serial

On 8/26/2020 7:28 AM, Gregory Nutt wrote: A short question concerning the use of LPWORK queue in FT232R and CDCACM USB Host drivers: Both block one LPWORK task per opened instance permanently. The FT232R rxdata_work USB transfer returns after each 16ms (USBHOST_FT232R_LATENCY) but since

Re: LVGL moving to Kconfig, how we could integrated better with them

If you can make LVGL optionally NOT generate its own config.h and .config we could indeed just include the Kconfig and Make.defs (as it is done now). I think generating the config file would be part of a "toplevel" LVGL build system which we would completely bypass. Couldn't it be done in the

Re: defining a BLE GATT server

erstand it shouldn't invalidate the LL in principle. But for any change on the host layer I would like to know if there are plans or work in progress. On Mon, Aug 24, 2020, at 11:47, Gregory Nutt wrote: Also, I understand that Xiaomi intends to replace the entire Bluetooth stack with a newer BT 5.0

Re: defining a BLE GATT server

I fully agree with this. AF_BLUETOOTH is already in place, but it only implements "BTPROTO_L2CAP" all of the GATT and advertising/scanning related functionality is implemented over "BTPROTO_HCI" #define BTPROTO_L2CAP 0 #

Re: 答复: [External Mail]Re: defining a BLE GATT server

As mentioned in the title, this is just an experimental project that needs to discuss further the implementation of the bluetooth stack. The implementation of bt stack in userspace does not conflict with the current architecture, this is just a choice. On the contrary, the complete BLE+MESH s

Re: 答复: [External Mail]Re: defining a BLE GATT server

What you should do is to coordinate and and cooperate with the NuttX team to get help from the community.  If you chose to go your own way and not discussion or cooperation with the team, then you are on your own.  That code will not come into the repository. If you want to do things correc

Re: 答复: 答复: [External Mail]Re: defining a BLE GATT server

Mr.Greg, I really don't understand why you are so angry, and there are so many verbal attacks. I just share a feasibility. What I expect is discussion and respect, not devaluation. If you don't like it, just say it like others. Yes, I am angry.  I am very pissed off that you completely igno

Re: 答复: [External Mail]Re: defining a BLE GATT server

Lets bring this back to a constructive place. An Chao I really appreciate the detailed information that you wrote, while I don't think it will be appropriate to include directly upstream it is still valuable to know how you are trying to use this and it helps us make an informed decision movin

Re: 答复: [External Mail]Re: defining a BLE GATT server

On 8/26/2020 10:56 AM, Gregory Nutt wrote: Lets bring this back to a constructive place. An Chao I really appreciate the detailed information that you wrote, while I don't think it will be appropriate to include directly upstream it is still valuable to know how you are trying to use thi

Re: 答复: [External Mail]Re: defining a BLE GATT server

I'm agreeing with you Greg, anything that comes into Apps or the OS must use the socket API. Then we are in agreement.  I am a mother hen and will defend the OS architecture to the teeth (hmm.. hens.. teeth... sorry about the mixed metaphor). I don't get involved in most code changes.  I

Re: 答复: 答复: [External Mail]Re: defining a BLE GATT server

Let me clarify one thing: The approach we take is a pure userspace solution, it isn't related to NuttX kernel and no plan to upstream it. That is good to know. Thank you.  I will sleep better tonight. The major difference is where to put L2CAP/HCI code(userspace v.s. kernelspace) and it isn

Re: nuttx.org certificate error

I'm getting the following error on firefox: Websites prove their identity via certificates. Firefox does not trust this site because it uses a certificate that is not valid for nuttx.org. The certificate is only valid for the following names: shortener.secureserver.net, www.shortener.secure

Cygwin Build Broken?

Has anyone built NuttX under Cygwin recently.  I haven't and just did so today for the first time in a long time.  It does not build.  I get this error: $ tools/configure.sh -c nucleo-l476rg:nsh $ make -j V=1 Results in: arm-none-eabi-ld --entry=__start -nostartfiles -nodefaultlibs -

Re: nuttx.org certificate error

Are we sure GoDaddy supports SSL on parked domains? My offer still stands to sort the DNS bit out. I think probably not.  There are no options at all when creating the re-direction.  (There is an option for adding DNSSEC to the accoutn but that requires an additional payment which I won't b

Re: nuttx.org certificate error

It seems that Firefox does not display the error if it autocompletes the url with http://. For some reason yesterday it went via https://. If you use https:// explicitly the failure is still there. That explains why I never see the problem. Does apache allow for custom domain? Then it would

Re: nuttx.org certificate error

I vote that we end the re-direction.  It is time to break that tie. When that tie is broken, this Issue becomes more important: https://github.com/apache/incubator-nuttx/issues/1498

Re: nuttx.org certificate error

Would you be opening to setting an A record in the DNS? I could then make sure the redirect is working properly while we work through updating all the links. I am not going to give you my login.  I can add an A record if you like.  There is already an A record that points to a IP address tha

Re: nuttx.org certificate error

I think that in the end it might be worth to completely switch to nuttx.apache.org then. If Apache wants this to be the frontface of project websites, it only adds confusion to have nuttx.org there. I have also been mirroring the incubator-nuttx repositories in the old Bitbucket area.  Pro

Re: Cygwin Build Broken?

On 8/30/2020 10:28 AM, Gregory Nutt wrote: Has anyone built NuttX under Cygwin recently.  I haven't and just did so today for the first time in a long time.  It does not build.  I get this error: $ tools/configure.sh -c nucleo-l476rg:nsh $ make -j V=1 Results in: arm-none

Re: stm32 uart init bug?

Perhaps: Bit 10PCE: Parity control enableThis bit selects the hardware parity control (generation and detection). When the parity control is enabled, the computedparity is inserted at the MSB position (9th bit if M=1; 8th bit if M=0) and the parity is checked on the received data. This bit is

Re: stm32 uart init bug?

Sometimes the best results (on small systems) are obtained if boot loader leaves the CPU as it found it out of reset. If you think in terms of an OOD life cycle: The "Destructor" for each block reset the HW to the reset state. With this approach, the SoC out of reset, and executing at __start

Re: passing extra linker flags from command line?

I see that it is possible to use: make EXTRAFLAGS= to pass flags to be used during compilation, but there does not seem to be a way to pass extra link-time flags. There is EXTRA_LIBS and EXTRA_LIBPATHS but these are used internally and if I do: make EXTRA_LIBS= EXTRA_LIBPATHS= these over

Re: Keyboard FeatherWing

I actually just got one of these FeatherWing boards. I was thinking about using it with the nRF52 Bluefruit LE - nRF52832. Seemed like a cool way to add a UX for interacting with the bluetooth stuff. Cool! That could be useful if you need a complete handheld UI. There are already an ILI93

Re: Keyboard FeatherWing

But yeah all the touch screen and backlight logic is behind the STMP811 touch/gpio expander. https://www.solder.party/docs/keyboard-featherwing/schematics_keyboard_featherwing.png That shouldn't be too bad.  I wrote the STMPE811 driver.  It is a multi-function chip and does GPIO, ADC, Touch,

Re: Next release?

I've been assiging various PRs from both repos to 10.0 milestone. Please also see if you think not assigned ones are to be surely included and assign them. I've not done this with issues as many are old ones for which no PR was proposed. Regarding bumping release: I think we had many major cha

[OT] Can one person run an open source project alone?

I ran across this and thought it interesting.  I decided to share: https://stackoverflow.blog/2020/09/09/open-source-governance-benevolent-dictator-or-decision-by-committee/ Several things resonate.

Re: PR's without adequate information

An alternative is to use a workflow. For example this action does something in that line: https://github.com/marketplace/actions/pull-request-ticket-check-action I think including the a PR comment check in the PR checks is a great idea if it is feasible:  If a section is blank, the check fa

Re: PR's without adequate information

That's why we have committers. It's been said, but I'll repeat: As committers, I know it's mundane, but we really need to make sure the commit and PR messages will be helpful to someone who looks at them in the future. But it would help the committer by at least assuring that there is someth

Re: Is the only option for using the network monitor....

On 9/17/2020 1:59 PM, David Sidrane wrote: Is the only option for using the network monitor Hi, My goal is to have a run time option to use a static IP or DHCP. (Curretly it is compile time) If either are used I want to "auto up" the eth0. Once the interface comes up , if configured for

Re: What gets patched onto previous releases?

I completely understand that it takes work to patch and regression test previous releases, but I just want to understand our policy and threshold for what gets backported. There should be an established policy.  I agree on that. In the old BDFL model, I only updated releases via patches and

Re: Driver Headers

drivers/ should be holding internal to the driver interfaces, such register definitions, low level communication interfaces readreg, writereg, etc... that might be shared across implementations. include/nuttx//.h should hold the interfaces/types required to interact with the driver, registrati

Re: Driver Headers

drivers/ should be holding internal to the driver interfaces, such register definitions, low level communication interfaces readreg, writereg, etc... that might be shared across implementations. include/nuttx//.h should hold the interfaces/types required to interact with the driver, registratio

Re: Driver Headers

drivers/ should be holding internal to the driver interfaces, such register definitions, low level communication interfaces readreg, writereg, etc... that might be shared across implementations. include/nuttx//.h should hold the interfaces/types required to interact with the driver, registra

Re: Nuttx for Nucleo-F429ZI

In most board bring-up logic, the procfs is auto-mounted on power up.  Like this from the STM32F4 Discovery:     #ifdef CONFIG_FS_PROCFS   /* Mount the procfs file system */   ret = mount(NULL, STM32_PROCFS_MOUNTPOINT, "procfs", 0, NULL);   if (ret < 0)     {   serr("ERR

Re: NAND R/W support

I'm trying to understand what level of NAND memory support currently exists in NuttX. The intention is to read and write files on the NAND memory. Device level support is in place, but probably somewhat out of date.  The obstacle to use NAND is that there is no NAND-compatibile file system o

Re: Stack size alignment on arm

In the past (9 or so years ago) there was an issue with floating point that required the stack to be 8 byte aligned. It was too long ago, so I do not remember the details, j (There is a note in the release notes, "ARM EABI: Fix stack alignment required for passing floating point values.") But I

Re: [DISCUSS] NuttX 10.0.0 Release schedule

I'm still seeing some important changes outstanding and in flux and no real stability window what do people think about shifting the schedule by 4 days. So the branch would be created on 10/7 (My evening time GMT-7) instead of 10/3. Having to deal with backports right off the bat is not really

IP Clearance

Well, we are over 10 months pushing 11 months as an Apache podling.  In my assessment, we have made zero progress on IP clearance.  I am certainly appreciative of the tool-related work of Adam Feurer and Matias N.  They have certainly assured that we have proper tool support to do the job. Bu

Re: IP Clearance

For some context most project graduate at 1 1/2 t 2 years, but some take longer, it just depends on the project. The IP clearance is important and you would need to show progress on this before you could graduate. Well, I believe that we would be in the worst case in this because this is a

Re: IP Clearance

For some context most project graduate at 1 1/2 t 2 years, but some take longer, it just depends on the project. The IP clearance is important and you would need to show progress on this before you could graduate. Well, I believe that we would be in the worst case in this because this is

Re: IP Clearance

I do not think there is zero progress as at least I’ve signed a SGA to change all the files from Xiaomi to Apache License. Zero files have been officially cleared.  That allowed us to change the Xiaomi/Pinecone copyrighted headers to Apache headers, but we still have no idea who all contribut

Re: Cannot build H7 configuration using Ubuntu for Windows 10

/bin/sh: 1: tools/mkconfig: not found tools/Makefile.unix:247: recipe for target 'include/nuttx/config.h' failed make: *** [include/nuttx/config.h] Error 127 Your problem is not that kconfig-frontends is not being built, but rathter that tools/mkconfig is not being built.  That should happen

Re: Cannot build H7 configuration using Ubuntu for Windows 10

I used 'make V=1' and I saw that mkconfig is being built successfully. root@edi-laptop:/mnt/d/Work/Development/Git/GitHub/nuttx# make V=1 make -C tools -f Makefile.host mkconfig.exe make[1]: Entering directory '/mnt/d/Work/Development/Git/GitHub/nuttx/tools' No, the binary tools/mkconfig is

Re: Cannot build H7 configuration using Ubuntu for Windows 10

I used 'make V=1' and I saw that mkconfig is being built successfully. root@edi-laptop:/mnt/d/Work/Development/Git/GitHub/nuttx# make V=1 make -C tools -f Makefile.host mkconfig.exe make[1]: Entering directory '/mnt/d/Work/Development/Git/GitHub/nuttx/tools' No, the binary tools/mkconfig

Re: Cannot build H7 configuration using Ubuntu for Windows 10

It looks like the extension got messed up. If that's the case renaming mkconfig.exe to mkconfig should at least build. Either that, or configure.sh and/or sethost.sh set up for the wrong host.  When you select Ubuntu with the -u option, sethost.sh should set the host to Windows, but the env

Re: Cannot build H7 configuration using Ubuntu for Windows 10

Yes, renaming the mkconfig*.exe* to mkconfig did the trick. I had the same problem with mkversion*.exe* - renamed it to mkversion and the build worked fine. If I get some time tomorrow, I will try to find the commit that messed up the extension. Got it!  Okay, this was easy once you know it i

Re: Cannot build H7 configuration using Ubuntu for Windows 10

I created issue: https://github.com/apache/incubator-nuttx/issues/1985 On 10/13/2020 4:15 PM, Gregory Nutt wrote: Yes, renaming the mkconfig*.exe* to mkconfig did the trick. I had the same problem with mkversion*.exe* - renamed it to mkversion and the build worked fine. If I get some time

Re: Cannot build H7 configuration using Ubuntu for Windows 10

You are right. I used your changes and build using -u and I still have the same issue (the board does not bootup). This is what I can see on the serial port: ADE This is debug output from arch/arm/src/stm32h7/start.c.  The hang occurs somewhere after line 412 ( showprogress('E');), and prob

Re: interrupt priorities on nRF52

On 10/28/2020 10:41 AM, Nathan Hartman wrote: On Tue, Oct 27, 2020 at 10:21 AM Alan Carvalho de Assis wrote: Hi Nathan, I totally agree! Zero Jitter Interrupt is a better name. We don't have zero latency, we have zero jitter. BR, Alan The latency is pretty close to zero too, because it

Re: interrupt priorities on nRF52

We might want to look at the docs and see if what Greg just explained a few minutes ago is documented there. Also worth remembering that this is only an ARMv7-M and ARMv8-M feature because it depends on (1) having true prioritized interrupts (ARMv6-M does not) and (2) the ability to execute

Re: interrupt priorities on nRF52

I'll try to experiment with this, but as I mentioned I'm not sure if it will be useful for this case since I need to interact with the OS. You would perform the time critical portion in the interrupt handler then schedule the OS interaction via PendSV.

Fwd: Fwd: [PATCH 1/1] stm32l4: correct build of stm32l4_can.c to respect L4 variant.

Forwarded Message Subject: Fwd: [PATCH 1/1] stm32l4: correct build of stm32l4_can.c to respect L4 variant. Date: Wed, 4 Nov 2020 00:26:36 +0100 From: Pavel Pisa Organization: PiKRON Ltd. To: Gregory Nutt Hello Greg, I hope that you are doing well. I am happy that

Re: Fwd: Fwd: [PATCH 1/1] stm32l4: correct build of stm32l4_can.c to respect L4 variant.

1 On 11/3/2020 6:40 PM, Gregory Nutt wrote: Hi, Pavel, I know longer deal with patches and changes directly.  These are handled only by the development team and usually only by PRs to github/apache/inclubator NuttX. I am putting dev@nuttx.apache.org on distribution.  Because of list securi

Re: UPD send delay

Probably worth taking a look at https://cwiki.apache.org/confluence/display/NUTTX/TCP+Network+Performance That deals specifically with TCP but discussion related to the Ethernet driver also applies.  The NuttX (TCP) is capable of performing at network speeds; the performance bottleneck is alwa

Re: UPD send delay

I've already tested network throughput with 'udpblaster' tool and was happily wondering by NuttX capability to flood full Ethernet bandwidth in both directions simultaneously. That was one of the reasons why we've chosen NuttX for our project. "If it's so strong in flooding, it should have low l

Re: UPD send delay

There are two threads involved: 1. The user thread that calls the UDP sendto() interface: * Lock the network. * Call netdev_txnotify_dev() to inform the driver that TX data is available.  The driver should schedule the TX poll on LP work queue. * If CONFIG_NET_UDP_WRITE_BUFFERS is en

Re: Remove Device Nodes

Calling unlink() on a character driver should work just as Alan describes (provided that the character driver supports the unlink() method).  In the specific case that you ask about about, the /dev/mtdblock1 driver is a block driver. The explanation is still correct, except that it is the bloc

Re: Remove Device Nodes

The /dev/mtd1 is, I assume, a character driver that was instantiated via bchdev_register().  It should be unlinked first before the block driver.  The bchdev_unlink() method will teardown the /dev/mtd1 instance /*when it can*/. ... And it cannot do that until all of the open references to th

Re: select()/pselect() function

fd = open("/dev/ttyS1", O_RDWR | O_NOCTTY); ... s_rc = select(fd + 1, &rset, NULL, NULL, &tv); The first argument should be 1, not fd + 1

Re: select()/pselect() function

says "nfds This argument should be set to the highest-numbered file descriptor in any of the three sets, plus 1. " In the portserial.c nfds was set to fd + 1 Daniel Pereira de Carvalho Em sex., 20 de nov. de 2020 às 14:04, Gregory Nutt escreveu: fd = open("/dev/ttyS1

Re: select()/pselect() function

From POSIX: "If the timeout interval expires without the specified condition being true for any of the specified file descriptors, the objects pointed to by the readfds, writefds, and errorfds arguments shall have all bits set to 0." If your select() timeouts on first loop iteration for any

  1   2   3   4   5   6   7   8   9   10   >