IW61x/IW41x 支持 LPC5528 我们正在寻找 IW61x / IW41x 系列的 WiFi/BLE 模块来连接 LPC55S28 微控制器（通过 SDIo / 或 SPI .. UART 太慢）。 NXP MCUXpresso SDK 是否有机会支持此类选项？ 产品：WiFi IW416 产品：WiFi IW6XX 回复：IW61x/IW41x 对 LPC5528 的支持 不幸的是，这些似乎是我们唯一的选择。如果您想了解有关 NCP 应用程序的更多信息，请参阅 RW61x 文档中的 UM12133。 很抱歉由此给您带来不便。 问候， 丹尼尔。 回复：IW61x/IW41x 对 LPC5528 的支持 RW610 / RW612 怎么样？我们也考虑使用这个 SoC，但我们需要连接一些大容量存储器，如 eMMC/SD 卡。SDIO 接口仅为从属接口？我是对的嗎？ 还有其他可以使用的选项吗？（标准 FLEXCOM SPI 受限于 30MHz，因此对我们来说太慢了）

Getting access to the GPIO ports on the iMX7ULP-EVK - how hard can it be? Hi I have tried to get access to the GPIO ports on the iMX7ULP-EVK, but it s causing me some troubles. It should be possible to either set a value or read a value from a port in the embedded Linux system by creating a node e.g.: echo 611 > /sys/class/gpio/export  Where 611 corresponds to PTF3? I'm a bit uncertain about this number, but it is coming from  cat /sys/kernel/debug/pinctrl/40ac0000.pinctrl/gpio-ranges which corresponds to the four groups of ports (PTC, PTD, PTE and PTF) and for the last one, PTF it says: 40b1000.gpio GPIOS [608-627] PINS [96 - 115] PTF3 is the fourth counting from zero, that means is should be 611 - is this right? At least I can create a node the command at the top, and if I try to make a node outside the ranges from the list, I get an error. When I have made a node, I get access to the features like "active-low", "direction" etc.. However I cannot change the direction from in to out for port 611 but I can for port 610. Could this be related to the Device Tree Source? If I read out the value of the port 611 it is zero and remains zero even if apply 3.3V to the PTF3 pin on the evaluation board (accessible on the connector for the Arduino - and yes, I have taken R155 into consideration, since the resistor needs to be mounted to enable the pin-header. I just applied the 3.3v on the pads for the resistor). Could the missing ability to read from the status of the pad be related to missing configuration of the Device Tree Source? Questions about GPIO in Device Tree Source I have managed to compile the Device Tree Source from the files embedded in the yocto project, and if I change the model name (IMX7ULP-EVK) to something else, I can see that is has changed in the .dtb file (by decompiling the file afterwards). The file is working afterwards since the processor can boot and use the MIPI interface. I then tried to add some code for the for the GPIO: In imx7ulp-evk.dts inside "&iomuxc1" &pinctrl { pinctrl_gpio_leds: gpio_leds_grp { fsl,pins = < IMX7ULP_PAD_PTF14__PTF14 0x22 /* Configure PTF14 as GPIO */ IMX7ULP_PAD_PTF15__PTF15 0x22 /* Configure PTF15 as GPIO */ >; }; };  In the imx7ulp-evkb.dts I place the following code: gpio_leds { compatible = "gpio-leds"; pinctrl-names = "default"; pinctrl-0 = <&pinctrl_gpio_leds>; led1 { label = "LED1"; gpios = <&gpio_ptf 14 GPIO_ACTIVE_HIGH>; /* PTF14 */ default-state = "off"; }; led2 { label = "LED2"; gpios = <&gpio_ptf 15 GPIO_ACTIVE_HIGH>; /* PTF15 */ default-state = "off"; }; }; I still have an error to fix before I manage to compile the additional code above, but it could be nice of there were an example how to get the GPIO ports to work from start to finish (I haven't been to find the information) just to confirm that I am on the right track? I how does this go together with the pin mux that is configured for the m4 processor? Hope some can bring some clarity to have to get the GPIOs working for an embedded linxux. i.MX7ULP Linux Yocto Project Re: Getting access to the GPIO ports on the iMX7ULP-EVK - how hard can it be? Hi Aldo Thanks for your response. I'm not sure what you mean with "peripherals being used on the M4 should be disabled on the Cortex-A" since my intention keep as many functions as possible working on the evk, and only change the functions needed to make some GPIOs free and available. I managed to compile a new .dtb file that enables the gpio pins. The definition of the leds remained the same as in the original post, but the pin-mux had to be changed slightly: &pinctrl { pinctrl_gpio_leds: gpio_leds_grp { fsl,pins = < IMX7ULP_PAD_PTF14__PTF14 0x14 /* Configure PTF14 as GPIO */ IMX7ULP_PAD_PTF15__PTF15 0x14 /* Configure PTF15 as GPIO */ >; }; }; If the GPIOs have been configured correctly, that will visible in the linux kernel cat /sys/kernel/debug/gpio providing a output similar to this:   Now the output of the pin can be changed by: echo 1 > /sys/class/leds/LED1/brightness /* turn on led */ echo 0 > /sys/class/leds/LED1/brightness /* turn off led */   This also answered the question in relation to the number of the GPIO ports, which were respectively 622 and 623. Also when a GPIO port has been defined as a an LED in the Device Tree Source, then it is not possible to access the GPIO as a node, but instead /sys/class/leds/LED1/ has to be used to access the properties. Re: Getting access to the GPIO ports on the iMX7ULP-EVK - how hard can it be? Hello, What error are you getting when trying to compile? PTA & PTB are used on the M4, and do remember that peripherals being used on the M4 should be disabled on the Cortex-A or it will crash when booting into Linux. Also, make sure to not use them on user space that can cuase the same as above, it is better to have everything configured on device tree to avoid this kind of issues. Best regards/Saludos, Aldo.

UM10204（Rev.7.0）图15（I2C总线规格和用户手册） 我认为 UM10204（I2C 总线规范和用户手册 Rev.7.0）中存在错误。即3.1.11节的图15。 我假设 I2C 总线上有两个 10 位地址目标，target1 的地址为 {2'b01, 8'h55}，target2 的地址为 {2'b01, 8'hAA}。在图15的过程中，首先指定target1的地址({2'b01,8'h55})，在重复START条件之后，指定target2的地址({2'b01,8'hAA})且R/W=1。此时 I2C 总线上就会发生冲突，因为 target1 在没有等待地址的第二个 BYTE({8'hAA}) 的情况下就开始输出第二次 R/W 之后的数据。 我认为在重复的 START 条件之后，所有目标都必须等待地址的第二个字节。我错了吗？ 回复：UM10204（Rev.7.0）的图15（I2C总线规格和用户手册） 非常感谢您的快速回复！

NXP IW611 & Apple Wireless Accessory Configuration (WAC) feature Dear NXP Team Will IW611 support Apple Wireless Accessory Configuration (WAC) feature? If it support, IW611 can connect  WAC device without entering password. https://developer.apple.com/documentation/bundleresources/entitlements/com_apple_external-accessory_wireless-configuration BR Wonhoki Re: NXP IW611 & Apple Wireless Accessory Configuration (WAC) feature Dear Christine_Li Thanks for update. Please close this case!! BR Wonho Ki Re: NXP IW611 & Apple Wireless Accessory Configuration (WAC) feature Hi, @Wonhoki  Thanks for your quick feedback and giving me a kudo to my reply. BTW, do you have any other concerns on this case? If no, would you mind me close this case? Best regards, Christine. Re: NXP IW611 & Apple Wireless Accessory Configuration (WAC) feature Hi, @Wonhoki  After I checked with our internal team, I confirmed that IW611 doesn't support Apple Wireless Accessory Configuration (WAC) feature currently. I apologize for the inconvenience to you. Best regards, Christine. Re: NXP IW611 & Apple Wireless Accessory Configuration (WAC) feature Hi, @Wonhoki  Thanks for creating case to us. I checked the Release Note of IW611, it doesn't mention IW611 supports this Apple Wireless Accessory Configuration(WAC) feature. I worry IW611 may not support this special feature. But to double confirm, I will check internally and give a confirmed answer. It will need some time, please be patient. Sorry for the inconvenience to you. Best regards, Christine.

Support of Camera module Hello,  For our application, we have choosed IMX8M-plus processor and the camera sensor from the omnivision (OSO2H10),  we are interested to know, If IMX8M-plus supports the firmware for OSO2H10 camera sensor? OSO2H10 is a raw sensor without any ISP, we have chosen IMX8M-plus to use the ISP within the processor, do you support the tunning of ISP for this camera module? Kindly share the list of camera modules/sensor supported for IMX8M-plus. Thanks,  i.MX 8 Family | i.MX 8QuadMax (8QM) | 8QuadPlus Re: Support of Camera module Thanks for the response and the links! Understood, using the reference of OV5640, I can develop the driver for OSO2H10. I can use available API for ISP tunning. Regards, Re: Support of Camera module Hello, The OSO2H10 is not support be default on MX8Mplus but it support the OV5640 and should be similar the driver for it you can migrate to your camera with the ISP usage please check: https://www.nxp.com/docs/en/user-guide/iMX8MP_CAMERA_DISPLAY_GUIDE.pdf https://www.mouser.com/datasheet/2/302/AN13736-3073771.pdf?srsltid=AfmBOoobASIHe4SOtrDvchLUQZebrqyxVnrzqtdGbq-ijvDr65FejhXf Regards

iMX8MP - WebGPU has been disabled via blocklist or the command line From chromium v117 running on iMX8MP I see the following item in the chrome://gpu page: WebGPU has been disabled via blocklist or the command line. Disabled Features: webgpu Is this expected? I see here some possible reasons, but I don't know which ones apply to the iMX8MP hardware architecture. Should I pass --enable-unsafe-webgpu command line parameter to chromium? Or is this feature unsupported on the iMX8MP hardware? i.MX 8M | i.MX 8M Mini | i.MX 8M Nano Re: iMX8MP - WebGPU has been disabled via blocklist or the command line That;s correct! Re: iMX8MP - WebGPU has been disabled via blocklist or the command line Hi @Bio_TICFSL  do you mean that even with the flags suggested by your colleague here, the implementation is full software? Re: iMX8MP - WebGPU has been disabled via blocklist or the command line Hello, Currently WebGPU are not supported by MX8MP in latest BSP, but is pure software. Regards

T2081: U-boot not responding after entering key to stop the autoboot Hi Team, One of our custom board uses T2081QDS SOC with 4GB of DDR. Currently we are able to port u-boot on the custom board and stop the autoboot at u-boot by pressing any key. But we are not able to execute any commands on the u-boot console. We can say it is receiving only the key to stop the autoboot but no other commands from the user. Can you please guide me with the solution or areas to check to resolve this issue?   Thanks, Noufal T2080  #t2081QD QorIQ T2 Devices Re: T2081: U-boot not responding after entering key to stop the autoboot Hi @LFGP  There is no issue with the RAM as seen in the boot log. PFA complete boot log.  We are also suspecting Processor UART-RX line. Debug is going on. Thanks,  Noufal Re: T2081: U-boot not responding after entering key to stop the autoboot dear @noufalp90 , let it continue with the boot process, don't hit any key, and share, on this way,  a complete log file. I want to check if there some message about RAM or DTS. Re: T2081: U-boot not responding after entering key to stop the autoboot Hi @LFGP , I have disabled almost all other peripherals. But still not working. U-boot stuck after entering key to stop autoboot.  PFA new boot log. Thanks, Noufal Re: T2081: U-boot not responding after entering key to stop the autoboot Yes, We are booting from NOR. We dont have any SD card. You meant to say removing peripheral from the u-boot config and try ? Thanks Noufal Re: T2081: U-boot not responding after entering key to stop the autoboot I see you are booting from NOR, right? , so, as a test please remove the SD card (if you have one) and any peripherical device.  Re: T2081: U-boot not responding after entering key to stop the autoboot Hi, Thank you for your suggestion. I have downloaded u-boot from https://github.com/nxp-qoriq/u-boot/tree/git.denx.de/u-boot-2016.09.y/ since the SDK document mentioned that updated version of the u-boot is 2016.09. Still we are stuck in the console. Thank you. Noufal Re: T2081: U-boot not responding after entering key to stop the autoboot dear @noufalp90 , it is necessary to re-build the image, be sure you are using the Uboot adequate version regarding the SDK.   

Open SDA software for MIMXRT1010 Hello We designed a board which is almost a clone of the MIMXRT1010-EVK. In order to setup our custom board for programming, we followed the instructions in AN13206.pdf for setting up the OpenSDA probe. The first step is to program the OpenSDA bootloader. It is working and the board anumerate as a Maintenance drive. We are able to go to the second step and load the OpenSDA firmware for our microcontroller. The board disconnect and reenumerate as a RT1010-EVK drive. We are able to program the IMXRT1010 microcontroller. However if we unplug the board and plug it in again it enumerate again as a Maintenance drive. We did the same steps with an original MIMXRT1010-EVK and the board enurate as a RT1010-EVK drive even after several plug/unplug. Could you please tell what is making our board enumerating in maintenance mode after power loss. Thanks in advance. Jérémy i.MXRT Re: Open SDA software for MIMXRT1010 Hi all I discovered that the EVK on board usb switch is in the reverse direction on my pcb. Actually, its main purpose is to delay the target (iMXRT1010) power on and not to protect the debugger's usb. It was not obvious at a first sight. the target might drawn too much current during the probe enumeration. Thanks to all for your help. Kind regards, jterrien Re: Open SDA software for MIMXRT1010 Hi Diego Thanks for your reply. So I did some progress. To answer to your question: * Yes the schematics of our board is almost a clone of the MIMXRT-1010-EVK. We removed only useless sections for us like sound code, ... * I am able to program the probe as a JLINK probe and program and debug the target IMXRT1010 under MCUxpresso * If I plug and unplug the board it enumerates as a JLINK probe and I can reprogram the target * If I switch back to open SDA, after programming the probe it enumerates as MIMRXRT1010 drive but if i unplug and then plus again the board, it enumerates as a Maintenance drive. Some conclusions:     => I re-red the AN13206. It says that in JLINK mode the prode does not drive the power converter of the target    => I went to the conclusion that my problem could come from the activation of the target so I powered the probe (with OpenSDA firmware) but by disconnecting the target power. Surprisingly the probe enumarate as a MIMRXRT-1010 drive as it should do.     Conclusion: The power drown from the target MCU during probe enumeration yield to making the probe not booting in OpenSDA mode but keeps it in maintenance mode. I do not know currently why the official board is not affected by this problem and mine is. If you have any ideas of the source of the problem I would appreciate your insight. Thanks in advance for your valuable help. Kind regards, Jérémy Re: Open SDA software for MIMXRT1010 Hi @jterrien  Thank you for reaching out.  I just have a couple of questions. I do not know if the OpenSDA FW you loaded gets lost after reset, or if the LPC43xx is entering on DFU mode. Is your on-board debugger circuitry ( for the LPC43xx ) the same than the one we have on the RT1010 EVK? Are you installing the FreeLink debug FW or the LPC Link 2 FW? It sounds like that you are following  chapter 4 Switching back to OpenSDA DapLink firmware (factory default firmware) steps from the AN.  Could you try with Segger FW ? Diego

Is it correct that only CM7_0 can be used on the S32K314? Hello, I have a question after reviewing the S32K314 datasheets (S32K3XXRM, S32K3XXDS). [Figure 1, in S32K3XXRS Figure 8] According to the diagram in [Figure 1], there are references to both CM7_0 and CM7_1 in the S32K314. [Figure 2, in S32K3XXRM Table 26] However, in [Figure 2], it states that CM7_1 cannot be used on the S32K314. Which one is correct? #S32K314 #S32K3XX #S32K3 #CM7_0 #CM7_1 Re: Is it correct that only CM7_0 can be used on the S32K314? Hi @malove, That is correct, it is not available. The "1" in S32K314 refers to 1 x M7 core, while "4" (such as S32K344) refers to 1 x M7 lockstep core. Best regards, Julián Re: Is it correct that only CM7_0 can be used on the S32K314?  @Julián_AragónM  Thank you for your reply. Oh, I have an additional question. So the S32K314 series does not have the Lockstep core function?  Re: Is it correct that only CM7_0 can be used on the S32K314? Hi @malove You can see in [Figure 1], CM7_1 is disabled for S32K314: Best regards, Julián

S32 license missing After updating the system, open the S32 design studio shows license missing, quickly solve ah!

LPC55S16: Secure Boot - skboot_authenticate() Hello, I am currently trying to implement Secure Boot functionality on a LPC55XpressoS16 eval board. My plan was to use the Secure ROM API function skboot_status_t skboot_authenticate(const uint8_t *imageStartAddr, secure_bool_t *isSignVerified) for the image verification. I managed to access it using the necessary function pointers to the ROM API table. The image itself is signed according to the SPSDK documentation (Link to documentation). I also tried the whole thing using an image that was signed via the NXP Secure Provisioning GUI tool. My problem now is that the function always returns kStatus_SKBOOT_InvalidArgument. I am actually pretty certain that the imageStartAddr parameter is correct (0x00000000) and that this is not the cause of this. UM11295 states that this is exactly the same function that the already by NXP implemented secure boot functionality in ROM uses. When I then enabled that secure boot in addition to my own, the one by NXP passed and verified the image (the device booted normally and jumped to the reset handler) but my function again returned "invalid argument". Even though allegedly the same function is used. Is there something I am overlooking or is there some additional documentation for that function (I already checked the appeneded Rom_API.pdf of the UM11295)? Re: LPC55S16: Secure Boot - skboot_authenticate() After double-checking my solution I have found that the kb_init function is mandatory, and I was passing a too small user buffer to it. You have to submit at least 4096 word-aligned bytes to this function for it to succeed. With this I was able to verify a SB2.1 image container using skboot_authenticate. My second question still holds: does the ROM API support validating a Master Boot Image from the application? One such image which the ROM bootloader happily starts up makes skboot_authenticate yield kStatus_SKBOOT_InvalidArgument. Re: LPC55S16: Secure Boot - skboot_authenticate() I am encountering the same issue, but on the IMXRT685S EVK. The device is able to succesfully boot a signed Master Boot Image with the correct keys loaded into the RKTH shadow registers. However, calling skboot_authenticate on the same memory region, both in RAM and on external flash, result in kStatus_SKBOOT_InvalidArgument. Taking inspiration from SBL, calling kb_init results in a non-defined return code "10". Verifying that it works using the Secure Provisioning Tool is not helpful, as the EVK probe presents multiple HID devices, which blhost as used by the Secure Provisioning Tool does not handle correctly. The wording in UM11147 is not 100% clear on how this function should work. I have also tried verifying a SB1.2 image but that also returns kStatus_SKBOOT_InvalidArgument. The ROM uses this function during the Secure boot flow to authenticate 1st boot image, and it also uses it to verify authenticity of the SB 2.1 files. Can you verify that the MIMXRT685S skboot_authenticate should be able to verify correct Master Boot Images from a booted first application? If you have an engineer that can hop on a call to clear this up, that would be great. Re: LPC55S16: Secure Boot - skboot_authenticate() Hello @jesm86  About secure boot, recommend you use MCUXpresso secure provisioning tool: https://www.nxp.com/design/software/development-software/mcuxpresso-software-and-tools-/mcuxpresso-secure-provisioning-tool:MCUXPRESSO-SECURE-PROVISIONING   Refer to User Guider -> 7.3 LPC55Sxx/LPC55xx/LPC553x device workflow. BR Alice

Gstreamer vpudec Memory leaking on i.MX 8M Plus Dear NXP team, We’ve encountered an issue on our device, which displays six camera feeds (800x600@60fps) on a Full HD screen. Our application uses a GStreamer-based pipeline [2] implemented in C++. When running this pipeline, we received error [6] after approximately 1.5 hours of uptime. The device is based on the i.MX 8M PLUS and runs Mickledore (kernel 6.1.36). Since Variscite has not modified GStreamer, I'm reaching out for your input. The kernel settings remain unchanged; by default, CmaTotal is set to 983040 kB. From what I understand, the VPU’s DMA usage reallocates CMA memory dynamically. We've observed a gradual decline in available CMA memory [3], and after approximately 531,723 seconds (around six days), the pipeline failed to allocate a contiguous block of 2040 pages, triggering error [4] and causing our application to crash. Initially, I suspected the issue might be related to the GStreamer vpudec plugin’s use of VPU memory (use-vpu-memory=true by default), so I disabled it. This seemed to temporarily stabilize cmaFree, but the available memory still gradually decreased and fluctuated over time. [5] We are displaying a static mosaic layout and rarely recreate the pipeline. Since the issue only occurs after several days of uptime, simply increasing CMA memory would likely delay the problem rather than resolve it. I’ve been monitoring /sys/kernel/debug/dma_buf/bufinfo regularly. On one device, after 16 hours and 27 minutes, I saw: "Total 176 objects, 377,487,360 bytes", and after 18 hours and 41 minutes: "Total 184 objects, 392,232,960 bytes". The hex value 01843200 also increased from 83 to 91, along with overall memory usage. [7] I’d appreciate your thoughts on this issue, and any recommendations you might have for a long-term solution or further diagnostics I should consider. Regards, Szilard B [1] dmesg_cma_error_device1.txt, dmesg_cma_error_device2.txt, dmesg_cma_error_device3.txt [2] gst-launch-1.0 \ rtspsrc location="rtsp://admin:[email protected]/profile2/media.smp" latency=200 ! rtph265depay ! h265parse config-interval=-1 ! watchdog timeout=2000 ! vpudec disable-reorder=true ! videomixer.sink_0 \ rtspsrc location="rtsp://admin:[email protected]/profile2/media.smp" latency=200 ! rtph265depay ! h265parse config-interval=-1 ! watchdog timeout=2000 ! vpudec disable-reorder=true ! videomixer.sink_1 \ rtspsrc location="rtsp://admin:[email protected]/profile2/media.smp" latency=200 ! rtph265depay ! h265parse config-interval=-1 ! watchdog timeout=2000 ! vpudec disable-reorder=true ! videomixer.sink_2 \ rtspsrc location="rtsp://admin:[email protected]/profile2/media.smp" latency=200 ! rtph265depay ! h265parse config-interval=-1 ! watchdog timeout=2000 ! vpudec disable-reorder=true ! videomixer.sink_3 \ rtspsrc location="rtsp://admin:[email protected]/profile2/media.smp" latency=200 ! rtph265depay ! h265parse config-interval=-1 ! watchdog timeout=2000 ! vpudec disable-reorder=true ! videomixer.sink_4 \ rtspsrc location="rtsp://admin:[email protected]/profile2/media.smp" latency=200 ! rtph265depay ! h265parse config-interval=-1 ! watchdog timeout=2000 ! vpudec disable-reorder=true ! videomixer.sink_5 \ imxcompositor_g2d name=videomixer \ sink_0::xpos=0    sink_0::ypos=0 sink_0::width=600 sink_0::height=480   \ sink_1::xpos=640  sink_1::ypos=0 sink_1::width=600 sink_1::height=480   \ sink_2::xpos=1280 sink_2::ypos=0 sink_2::width=600 sink_2::height=480   \ sink_3::xpos=0    sink_3::ypos=560 sink_3::width=600 sink_3::height=480 \ sink_4::xpos=640  sink_4::ypos=560 sink_4::width=600 sink_4::height=480 \ sink_5::xpos=1280 sink_5::ypos=560 sink_5::width=600 sink_5::height=480 \ ! waylandsink sync=false enable-last-sample=false qos=false enable-tile=true [3] cma-20250709-155500.log [4] cma: cma_alloc: linux,cma: alloc failed, req-size: 2040 pages, ret: -12 [5] cma-our-software-use-vpu-memory-false.log [6]  1:25:48.1 / 99:99:99. (gst-launch-1.0:123309): GStreamer-CRITICAL **: 12:56:03.228: gst_buffer_get_memory_range: assertion 'GST_IS_BUFFER (buffer)' failed (gst-launch-1.0:123309): GStreamer-CRITICAL **: 12:56:03.228: gst_buffer_peek_memory: assertion 'GST_IS_BUFFER (buffer)' failed (gst-launch-1.0:123309): GStreamer-CRITICAL **: 12:56:03.228: gst_mini_object_unref: assertion 'GST_MINI_OBJECT_REFCOUNT_VALUE (mini_object) > 0' failed ERROR: from element /GstPipeline:pipeline0/GstWatchdog:watchdog5: Watchdog triggered Additional debug info: /usr/src/debug/gstreamer1.0-plugins-bad/1.22.0.imx-r0/gst/debugutils/gstwatchdog.c(195): gst_watchdog_trigger (): /GstPipeline:pipeline0/GstWatchdog:watchdog5: Watchdog triggered Execution ended after 1:25:50.443886000 Setting pipeline to NULL ... Total showed frames (128758), playing for (1:25:50.444041500), fps (24.999). Freeing pipeline ... [7] device2_3x2-mosaic_16h27min.txt, device2_3x2-mosaic_18h41m.txt Re: Gstreamer vpudec Memory leaking on i.MX 8M Plus Hello @szilard_b_cib  I hope you are doing very well. Unfortunately, I could not replicate by my side. Are you using the BSP from Variscite? Have you tried with our BSP? I my case the CmaFree is always the same. Best regards, Salas.

Not able to configure P4_6 pin as input interrupt on mcxn947 #define CINT_GPIO GPIO4 #define CINT_PORT PORT4 #define CINT_PIN 6U #define CINT_IRQ GPIO40_IRQn #define CINT_IRQ_HANDLER GPIO40_IRQHandler CLOCK_EnableClock(kCLOCK_Gpio4); CLOCK_EnableClock(kCLOCK_Port4); GPIO_SetPinInterruptConfig(CINT_GPIO, CINT_PIN, kGPIO_InterruptFallingEdge); NVIC_EnableIRQ(CINT_IRQ); gpio_pin_config_t sw1_config = { kGPIO_DigitalInput, 0, }; /* Sets the input pin configuration */ GPIO_PinInit(CINT_GPIO, CINT_PIN, &sw1_config); PRINTF("Init CINT\r\n"); void CINT_IRQ_HANDLER(void) { GPIO_GpioClearInterruptFlags(CINT_GPIO, 1U << CINT_PIN); g_touch = true; SDK_ISR_EXIT_BARRIER; } Does GPIO00_IRQn maps to port 0 pins 0-15 and GPIO01_IRQn to 16-31? FRDM-Training MCXN Re: Not able to configure P4_6 pin as input interrupt on mcxn947 您好@EdwinHz， 感谢您的确认。 我在将 P4_6 引脚设置为输入中断时遇到了问题。当我为端口 0 引脚进行配置时，同样的代码工作正常，但对于 P4_6 似乎不起作用。 MCXN947 的所有引脚都能配置为 GPIO 输入中断吗？ 如能提供任何帮助，将不胜感激。 Re: Not able to configure P4_6 pin as input interrupt on mcxn947 嗨，@sstcor、 没错。每个 GPIO 有两个中断向量；每个端口的 32 个引脚各占一半。

how void avoid multiple definition of `file__404_html'; when i compiling the S32K366 programm wiht LWIP，there are errors. It appears that the following codes caused the errors. #define HTTPD_FSDATA_FILE "fsdata.c"  ........ #include HTTPD_FSDATA_FILE how to void the errors. the S32DS 3.6 broblem message as below： Description Resource Path Location Type multiple definition of `file__404_html'; ./stacks/tcpip/lwip/src/apps/http/fs.o:C:\Users\Administrator\workspaceS32DS.3.6.1\k366lan8720\k366lan8720_M7_0_0\Debug_FLASH/../stacks/tcpip/lwip/src/apps/http/fsdata.c:324: first defined here k366lan8720_M7_0_0 line 324, external location: d:\nxp\s32ds.3.6.1\s32ds\build_tools\gcc_v10.2\gcc-10.2-arm32-eabi\arm-none-eabi\bin\real-ld.exe: .\stacks\tcpip\lwip\src\apps\http\fsdata.o:C:\Users\Administrator\workspaceS32DS.3.6.1\k366lan8720\k366lan8720_M7_0_0\stacks\tcpip\lwip\src\apps\http\fsdata.c C/C++ Problem Description Resource Path Location Type multiple definition of `file__404_html'; ./stacks/tcpip/lwip/src/apps/http/fs.o:C:\Users\Administrator\workspaceS32DS.3.6.1\k366lan8720\k366lan8720_M7_0_0\Debug_FLASH/../stacks/tcpip/lwip/src/apps/http/fsdata.c:324: first defined here k366lan8720_M7_0_0 line 324, external location: d:\nxp\s32ds.3.6.1\s32ds\build_tools\gcc_v10.2\gcc-10.2-arm32-eabi\arm-none-eabi\bin\real-ld.exe: .\stacks\tcpip\lwip\src\apps\http\fsdata.o:C:\Users\Administrator\workspaceS32DS.3.6.1\k366lan8720\k366lan8720_M7_0_0\stacks\tcpip\lwip\src\apps\http\fsdata.c C/C++ Problem multiple definition of `file__img_sics_gif'; ./stacks/tcpip/lwip/src/apps/http/fs.o:C:\Users\Administrator\workspaceS32DS.3.6.1\k366lan8720\k366lan8720_M7_0_0\Debug_FLASH/../stacks/tcpip/lwip/src/apps/http/fsdata.c:316: first defined here k366lan8720_M7_0_0 line 316, external location: d:\nxp\s32ds.3.6.1\s32ds\build_tools\gcc_v10.2\gcc-10.2-arm32-eabi\arm-none-eabi\bin\real-ld.exe: .\stacks\tcpip\lwip\src\apps\http\fsdata.o:C:\Users\Administrator\workspaceS32DS.3.6.1\k366lan8720\k366lan8720_M7_0_0\stacks\tcpip\lwip\src\apps\http\fsdata.c C/C++ Problem multiple definition of `file__index_html'; ./stacks/tcpip/lwip/src/apps/http/fs.o:C:\Users\Administrator\workspaceS32DS.3.6.1\k366lan8720\k366lan8720_M7_0_0\Debug_FLASH/../stacks/tcpip/lwip/src/apps/http/fsdata.c:332: first defined here k366lan8720_M7_0_0 line 332, external location: d:\nxp\s32ds.3.6.1\s32ds\build_tools\gcc_v10.2\gcc-10.2-arm32-eabi\arm-none-eabi\bin\real-ld.exe: .\stacks\tcpip\lwip\src\apps\http\fsdata.o:C:\Users\Administrator\workspaceS32DS.3.6.1\k366lan8720\k366lan8720_M7_0_0\stacks\tcpip\lwip\src\apps\http\fsdata.c C/C++ Problem

S32K396 中的以太网问题 你好， 我正在使用 MCPTR2AK396 评估板，并成功测试了 Gmac_IP_internalloopback_s32K396 示例项目的以太网功能。现在，我想实现一个 PING 示例，在不使用 Free RTOS 的情况下，通过以太网与外部主机 PC 通信。你能否提供一个独立组网 (SA) 的示例项目来演示如何实现这一目标。 Re: Ethernet issue in S32K396 你好@bINDU3123 与 TCP/IP 协议栈一起，还有一个名为 lwip_baremetal_s32k396.它可以演示基本功能，例如从主机 PC 对板执行 ping 操作，这可能会有所帮助。 BR、VaneB

Gstreamer vpudec i.MX 8M Plus のメモリリーク NXPチームの皆様 弊社のデバイスで、フル HD 画面に 6 つのカメラ フィード (800x600@60fps) が表示される問題が発生しました。私たちのアプリケーションはC++で実装されたGStreamerベースのパイプライン[2]を使用します。このパイプラインを実行すると、約1.5時間の稼働後にエラー[6]が発生しました。このデバイスは i.MX 8M PLUS をベースにしており、Mickledore (カーネル 6.1.36) を実行します。Variscite は GStreamer を変更していないため、皆さんのご意見を伺っています。 カーネル設定は変更されません。デフォルトでは、CmaTotal は 983040 kB に設定されています。私の理解では、VPU の DMA 使用により CMA メモリが動的に再割り当てされます。利用可能なCMAメモリ[3]が徐々に減少し、約531,723秒（約6日間）後、パイプラインが2040ページの連続したブロックを割り当てることができず、エラー[4]が発生し、アプリケーションがクラッシュしました。 当初、この問題は GStreamer vpudec プラグインによる VPU メモリの使用 (デフォルトでは use-vpu-memory=true) に関連しているのではないかと疑ったSO、無効にしました。これにより、cmaFree は一時的に安定したように見えましたが、使用可能なメモリは依然として徐々に減少し、時間の経過とともに変動しました。[5] 静的なモザイク レイアウトを表示しており、パイプラインを再作成することはほとんどありません。この問題は数日間の稼働後にのみ発生するため、CMA メモリを増やすだけでは問題は解決されず、むしろ遅延する可能性があります。 私は /sys/kernel/debug/dma_buf/bufinfo を定期的に監視しています。あるデバイスでは、16 時間 27 分後に次の画面が表示されました。 「合計176オブジェクト、377,487,360バイト」 、 そして18時間41分後: 「合計 184 個のオブジェクト、392,232,960 バイト」 。 16 進数値 01843200 も、全体的なメモリ使用量とともに 83 から 91 に増加しました。[7] この問題についてのご意見、また長期的な解決策や私が検討すべきさらなる診断について何かアドバイスがあればいただければ幸いです。 よろしくお願いいたします。 シラードB [1] dmesg_cma_error_device1.txt、dmesg_cma_error_device2.txt、dmesg_cma_error_device3.txt [2] gst-launch-1.0\ rtspsrc 場所="rtsp://admin:[email protected]/profile2/プレスリリース、製品ニュース.smp"レイテンシ=200!rtph265depay !h265parse 設定間隔 = -1 !ウォッチドッグタイムアウト=2000!vpudec の disable-reorder=true !ビデオミキサー.sink_0\ rtspsrc 場所="rtsp://admin:[email protected]/profile2/プレスリリース、製品ニュース.smp"レイテンシ=200!rtph265depay !h265parse 設定間隔 = -1 !ウォッチドッグタイムアウト=2000!vpudec の disable-reorder=true !ビデオミキサー.sink_1\ rtspsrc 場所="rtsp://admin:[email protected]/profile2/プレスリリース、製品ニュース.smp"レイテンシ=200!rtph265depay !h265parse 設定間隔 = -1 !ウォッチドッグタイムアウト=2000!vpudec の disable-reorder=true !ビデオミキサー.sink_2\ rtspsrc 場所="rtsp://admin:[email protected]/profile2/media.smp"レイテンシ=200!rtph265depay !h265parse 設定間隔 = -1 !ウォッチドッグタイムアウト=2000!vpudec の disable-reorder=true !ビデオミキサー.sink_3\ rtspsrc 場所="rtsp://admin:[email protected]/profile2/プレスリリース、製品ニュース.smp"レイテンシ=200!rtph265depay !h265parse 設定間隔 = -1 !ウォッチドッグタイムアウト=2000!vpudec の disable-reorder=true !ビデオミキサー.sink_4\ rtspsrc 場所="rtsp://admin:[email protected]/profile2/プレスリリース、製品ニュース.smp"レイテンシ=200!rtph265depay !h265parse 設定間隔 = -1 !ウォッチドッグタイムアウト=2000!vpudec の disable-reorder=true !ビデオミキサー.sink_5\ imxcompositor_g2d 名前=ビデオミキサー \ sink_0::xpos=0 sink_0::ypos=0 sink_0::width=600 sink_0::height=480 \ sink_1::xpos=640 sink_1::ypos=0 sink_1::width=600 sink_1::height=480 \ sink_2::xpos=1280 sink_2::ypos=0 sink_2::width=600 sink_2::height=480 \ sink_3::xpos=0 sink_3::ypos=560 sink_3::width=600 sink_3::height=480 \ sink_4::xpos=640 sink_4::ypos=560 sink_4::width=600 sink_4::height=480 \ sink_5::xpos=1280 sink_5::ypos=560 sink_5::width=600 sink_5::height=480 \ !waylandsink sync=false enable-last-sample=false qos=false enable-tile=true [3] cma-20250709-155500.log [4] cma: cma_alloc: Linux,cma: alloc 失敗、要求サイズ: 2040 ページ、戻り値: -12 [5] cma-our-ソフトウェア-use-vpu-memory-false.log [6] 1:25:48.1 / 99:99:99。 (gst-launch-1.0:123309):GStreamer-CRITICAL **: 12:56:03.228:gst_buffer_get_memory_range: アサーション 'GST_IS_BUFFER (buffer)' が失敗しました (gst-launch-1.0:123309): GStreamer-CRITICAL **: 12:56:03.228:gst_buffer_peek_memory: アサーション 'GST_IS_BUFFER (buffer)' が失敗しました (gst-launch-1.0:123309): GStreamer-CRITICAL **: 12:56:03.228:gst_mini_object_unref: アサーション 'GST_MINI_OBJECT_REFCOUNT_VALUE (mini_object) > 0' が失敗しました エラー: 要素 /GstPipeline:pipeline0/GstWatchdog:watchdog5 から: ウォッチドッグがトリガーされました 追加のデバッグ情報: /usr/src/debug/gstreamer1.0-plugins-bad/1.22.0.imx-r0/gst/debugutils/gstwatchdog.c(195):gst_watchdog_trigger(): /GstPipeline:pipeline0/GstWatchdog:watchdog5: ウォッチドッグが起動 実行は1:25:50.443886000後に終了しました パイプラインを NULL に設定しています... 合計表示フレーム数（128758）、再生時間（1:25:50.444041500）、fps（24.999）。 パイプラインを解放します... [7] device2_3x2-mosaic_16h27min.txt、デバイス2_3x2-モザイク_18h41m.txt Re: Gstreamer vpudec Memory leaking on i.MX 8M Plus こんにちは@szilard_b_cib お元気でお過ごしのことと思います。 残念ながら、私の側では再現できませんでした。Variscite の BSP を使用していますか?弊社の BSP を試しましたか? 私 の CASE、CmaFree は常に同じです。 よろしくお願いいたします。 サラス。

i.MX RT700でAClockウィジェットが壊れる症状 サポートチーム様 デバイス: i.MX RT700 AClock ウィジェットを使用して画面を構成すると、画面が壊れます。 i.MX RT500 / i.MX RT700 の基本的な例として、AClock Widget を使用しないプロジェクトです。 現象をご確認ください。 テスト環境は以下の通りです。 EVキット：MIMXRT700-EVK SCH-89280 改訂 A3 700-89280 レヴァ1 EVキット: ZC143AC72MIPI 700-94475 改訂A SCH-94475 改訂A GUI ガイダー : GUI ガイダー-1.9.1-GA LVGLバージョン: v9 SDKバージョン: MCUX SDK 25.03.00 ツールチェーンバージョン: MCUX IDEs 24.12 | ARMGC 13.2.1 Re: AClock Widget broken Symptoms in i.MX RT700 こんにちは@Dolbae 、 この問題を指摘していただきありがとうございます。 当チームでは正式にリリースされた製品に対してのみサポートを提供できますが、現時点では RT700 が正式にリリースされていないため、この問題のフォローアップのみをお手伝いできます。 社内の Gui-Guider チームに提出しましたので、新しい展開があればお知らせします。 よろしくお願いします、 ギャビン

iMX8MMのSJC_DISABLE、JTAG_SMODE、JTAG_HEOヒューズのプログラミング こんにちは、 AN4581 アプリケーション ノートのセクション 5.7 では、デバイスを完全に保護するために SJC_DISABLE、JTAG_SMODE、および JTAG_HEO ヒューズをプログラムすることが推奨されていますが、IMX8MMRM と IMX8MMSRM の両方を調べても、これらのヒューズの正確な位置 (バンク、ワード、ビット) を見つけることができませんでした。この情報はどこでCAN見つかりますか? ご回答をお待ちしています。 Re: Programming SJC_DISABLE, JTAG_SMODE and JTAG_HEO fuses on iMX8MM こんにちは@igorpadykovさん、 この情報も送っていただけると幸いです。 前もって感謝します DJ Re: Programming SJC_DISABLE, JTAG_SMODE and JTAG_HEO fuses on iMX8MM こんにちは、 この情報も送っていただくことは可能でしょうか？ また、これは imx8m nano と mini では違いますか? Re: Programming SJC_DISABLE, JTAG_SMODE and JTAG_HEO fuses on iMX8MM 私たちもこの情報を知りたいです。 Re: Programming SJC_DISABLE, JTAG_SMODE and JTAG_HEO fuses on iMX8MM こんにちは@igorpadykovさん、 imx8m miniのヒューズに関する情報が必要です。それを私に提供していただけますか？セキュリティ上の理由から、JTAG を無効にする必要があります。 よろしくお願いいたします。 ジュリアン Re: Programming SJC_DISABLE, JTAG_SMODE and JTAG_HEO fuses on iMX8MM @igorpadykov JTAG_HEO ヒューズの情報も送っていただけますか? ありがとう！ Re: Programming SJC_DISABLE, JTAG_SMODE and JTAG_HEO fuses on iMX8MM これも私と共有していただけますか？なぜこれが公開アプリケーション ノート、リファレンス マニュアル、またはセキュリティ リファレンス マニュアルに投稿されないのでしょうか?隠蔽によるセキュリティはセキュリティではありません。 Re: Programming SJC_DISABLE, JTAG_SMODE and JTAG_HEO fuses on iMX8MM このメールを公表情報としていただけますか？なぜこの情報はセキュリティリファレンスマニュアルに記載されていないのですか?i.MX8M Nano の場合、これはセキュリティ リファレンス マニュアルに記載されています。しかし、Mini が同じヒューズを使用しているかどうかはわかりません。 また、AN4581 では、興味深いヒューズ DIR_BT_DIS もリストされています。しかし、それに関する参考資料はCAN見つかりません。 Re: Programming SJC_DISABLE, JTAG_SMODE and JTAG_HEO fuses on iMX8MM また、imx8m-mini の JTAG を無効にしたいのですが、リファレンス・マニュアル、Rev. 2、08/2019、第 6 章.2 Fusemap にそれに関する情報が見つかりませんでした。私に正しい方向を指示していただけますか。 ご回答をお待ちしています。 Re: Programming SJC_DISABLE, JTAG_SMODE and JTAG_HEO fuses on iMX8MM こんにちは rodrigo_travess 追加情報はメールで送信されました。 よろしくお願いします イゴール

Graphic shaking phenomenon during i.MX RT 700 operation Dear Support team,  Device : i.MX RT700 I have observed a graphic phenomenon related to the LCD. As the clock hand moves, the BACKGROUND image and the remaining images move together. You can see video clip.      This phenomenon can be seen in the basic example provided by GUI GUIDER. However, this issue does not appear on the i.MX RT 500. In my opinion, this phenomenon requires optimization.   The test environment is as follows. EVKIT : MIMXRT700-EVK SCH-89280 REV A3 700-89280 REVA1 EVKIT : ZC143AC72MIPI 700-94475 REV A SCH-94475 REV A  GUI-GUIDER  : GUI Guider-1.9.1-GA LVGL version : v9 SDK version  : MCUX SDK 25.03.00 toolchain version : MCUX IDE 24.12 | ARMGC 13.2.1  Re: Graphic shaking phenomenon during i.MX RT 700 operation It appears to be an issue related to the LCD driver or SmartWatch optimization.   After reviewing the code, I confirmed that the GPU is not the cause since LV_VGLITE (PXP) is not being used. To investigate whether the slowdown was due to the MIPI-DSI fetching the buffer, I tested by moving the buffer to SRAM, but the same issue persisted. Additionally, I configured the buffer with double buffering in partial refresh mode, but the issue remained.   However, when switching to full refresh mode, normal operation was observed. In conclusion, it seems the issue arises from the MIPI DSI driver and optimization used in the Smart Watch demo not being fully refined yet.   You can resolve the issue by modifying the following code:   File: <source/lvgl_support.c> Line 51: #define DEMO_DISPLAY_USE_PARTIAL_REFRESH 0     Set this to use LV_DISPLAY_RENDER_MODE_FULL mode for proper functionality. Re: Graphic shaking phenomenon during i.MX RT 700 operation Please refer to this post: https://community.nxp.com/t5/GUI-Guider/AClock-Widget-broken-Symptoms-in-i-MX-RT700/m-p/2133591/highlight/false#M1551

Ls1043A 的 NSP 正式支持哪些管理程序 亲爱的恩智浦团队 我想知道 Ls1043A 正式支持哪些 Hypervisor。 QNX/COQOS/Jailhouse/Xen/ 其他？ 他们的 ASIL 如何？ 您能告诉我吗？ 感谢并致意 卡鲁纳卡 Re: Which hypervisors are officially supported by NSP for Ls1043A 1.请参阅 https://www.nxp.com.cn/docs/en/application-note/AN13138.pdf 2.您可以联系营销团队获取业绩数据。 3.是的，你可以。 Re: Which hypervisors are officially supported by NSP for Ls1043A 嗨，yipingwang、 谢谢您的回答。我还有几个问题。 1.)请问恩智浦是否在 Ls1043A 平台上支持 XEN？ 2）与 Jailhouse 的合作如何？ 3）在 Linux SDK 的情况下，如果我们使用 KVM，是否可以使用 KVM 进行生产？ 我们正在寻找可以用于生产的虚拟机监控器。 感谢并致意 卡鲁纳卡 Re: Which hypervisors are officially supported by NSP for Ls1043A 实时边缘软件支持 Jailhouse。 Linux LSDK 支持 KVM。