Multi Source Translation Content

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

Multi Source Translation Content

Discussions

Sort by:
MX95上のWindows こんにちは、なぜNXPのWindows IoTウェブ上でMX95に関する情報が全く見られないのですか? https://www.nxp.com/design/design-center/software/embedded-software/i-mx-software/windows-10-11-iot-enterprise-for-i-mx-applications-processors:IMXWIN10IOT    Windows Re: Windows on MX95 現在、i.MX95はNXPの公開Windows 10/11 IoT Enterprise BSPページでのサポートデバイス/プラットフォームにリストアップされておらず、公開のi.MX95 Windows BSPも提供されていません。
View full article
LS1043A 最高核心温度 您好,我的 LS1043AXE7QQB(额定温度 105°C)核心(热二极管读数)在 25°C 的环境温度下正常运行时温度为 65°C。我们对系统进行了生产测试,在环境温度为 70 摄氏度的条件下运行 16 小时(每个系统一次)。在此测试过程中,核心温度升至 121 摄氏度。我们目前没有遇到任何问题,我们将改进散热管理以减少此类问题。CPU 短时间内运行温度超过 105 度会有什么影响?16 小时可接受的最高核心温度是多少? Re: LS1043A maximum core temperature 你好, 对于 QorIQ Layerscape 系列(LS1043A、LS1046A、LX2160A——均采用相同的热额定值框架): 参数 Value 源代码 最大工作温度 105 °C 数据手册 绝对最大值/存储温度 150°C (150°C下累计≤1008小时) 数据表表格——绝对最大额定值       105°C Tj(max) 指的是内部热二极管测量的硅结温度——这正是您的核心温度读数所反映的。 2. 高于105°C运行的影响 来自多个针对同一处理器系列的支持案例的官方立场: “在高于 Tj(max) = 105°C 的温度下运行会降低设备的可靠性,因此不被支持。 ” “长时间将设备暴露在绝对最大额定功率条件下可能会影响其可靠性或造成永久性损坏。 ” “该设备的工作温度不得高于 105°C。” 从半导体物理学的角度来看,超过 Tj(max) 会加速几种众所周知的损耗机制: 电迁移——金属互连线的迁移,随温度呈指数级加速 TDDB (时变介质击穿)——栅极氧化层退化 NBTI (负偏压温度不稳定性)——PMOS器件的阈值电压漂移 热载流子注入——随时间推移影响晶体管特性。 所有这些都遵循阿伦尼乌斯动力学:降解速率大约每高于合格限值 10°C 就翻一番。在 121°C(比额定温度 105°C 高 16°C)时,降解机制的运行速度约为合格速率的 3-4 倍。这不会导致立即/灾难性故障——设备仍可运行——但会加速消耗其使用寿命可靠性预算。 测试期间未观察到故障是意料之中的:在 121°C 时,远低于 150°C 的存储温度,在 150°C 时会立即发生物理损坏。这种影响表现为可靠性悄然下降,而不是瞬间损坏。   此致
View full article
PaxデバイスAPDU NFCとMifare Ultralight C こんにちは、 私はPaxデバイスで開発中で、Mifare ULTRALIGHT C(MF0ICU2)を管理したいと考えています。 PaxデバイスはカスタムのPayDroid OSを使用しており、標準的なAndroidではありません。そのため、私はNFCのAndroidライブラリを持っていません。 IPCCをライブラリ経由で公開するNeptuneLiteAPIというものをインストールする必要があります。 まず、命名規則と標準規格について少し混乱しています。APIが話している内容: ISO14443 AまたはB 1、2、または3 M1カード 通常、Ultralight Cは14443 Type Aですが、このオプションではカードを読み取れず、「M1カード」オプションで読み取れます。M1カードとは何ですか?クラシック1kのことですか? ドキュメントにはAPDUを使い、この種の呼び出しを実行するコマンドを公開する必要があると書かれています カードの読み書き(M1機能付き)に成功し、ライブラリはカード対応の互換性モードを使うデフォルトの読み書き方法を公開します 今、認証を設定し、APDUコマンドを使ってチップにコマンドを送ろうとしています。 以下は私が使えるコードの簡略化です iPicc = getDal().getPicc(piccType); iPicc.open(); while(){ PiccCardInfo cardInfo = iPicc.detect(EDetectMode.ONLY_M); if(cardInfo!=null){ //Read : works byte[] page0=iPicc.m1Read((byte) 0); // Write : works iPicc.m1Write((byte)(13),hexStringToByteArray("AAAAAAAA000000000000000000000000")); } 執筆に関しては3つの方法があります。 byte[] :  isoCommand(byte cid, byte[] send) ApduRespInfo : isoCommandByApdu(バイト cid, ApduSendInfo apduSendInfo) byte[] cmdExchange(byte[] dataIn, int recvLen) これらの方法はすべてAPDUコマンドを求めます。 もし私の理解が正しければ、Mifare Ultralight CはAPDUを尊重していないのに、APDUしか方法がないということですか? あるウェブサイトで、「FF」クラスを使うとチップに直接指示を送れるとわかりました。 しかし、どのコマンドを送信すればいいのか分からず、いくつか試してみた。 エラーが発生しています(残念ながら、これらは標準的なものではなく、NeptuneLiteAPiです)。 センサーエリアに特定のカードが存在しない、またはカードが有効化されていないPICC#3(検索カードエラーではありません) PICC#97(予期せぬエラー) でも不思議なことに、カードの直前で検出され読み書きに成功しているので、カードはしっかり「コネクテッド」されているのです。 直接命令、APDU命令、CRC計算など、いくつかのコマンドを試しましたが、どれも機能しませんでした。 //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("30")); // PICC#97(unexpected error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("3004")); // PICC#97(unexpected error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("3000")); // PICC#97(unexpected error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("3000A802")); // Avec CRC => No specific card in sensing area, or card is not activated PICC#3(not search card error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("300002A8")); // Avec CRC inversé => No specific card in sensing area, or card is not activated PICC#3(not search card error) //byte[] cmd2 = iPicc.cmdExchange(CryptageUtils.hexStringToByteArray("3000A802"),17); // Card is not activated. PICC#3(not search card error) // byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("30000000")); // No specific card in sensing area, or card is not activated PICC#3(not search card error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("ff300000")); // No specific card in sensing area, or card is not activated PICC#3(not search card error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("ff30000F")); // No specific card in sensing area, or card is not activated PICC#3(not search card error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("FFCA00000004")); // No specific card in sensing area, or card is not activated PICC#3(not search card error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("3000000004")); // No specific card in sensing area, or card is not activated PICC#3(not search card error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("3001")); // No specific card in sensing area, or card is not activated PICC#3(not search card error) もしどなたか助けてくれたり、正しい道に導いていただけるなら、本当にありがたいです。 よろしくお願いいたします。 Re: Pax device APDU NFC & Mifare Ultralight C こんにちは@thibaut @Antoine1 @Erableto 、 この問題の解決策は見つかりましたか?私も同じ問題に直面しました。Mifare Ultralight EV1ではカウンターが読み取れません。主にcmdExchangeで試しましたが、NeptuneLiteライブラリの他の方法も試しましたが、PICC#3や#97のようなエラーが必ず出ます。 Re: Pax device APDU NFC & Mifare Ultralight C こんにちは@thibautと@Erableto 、 質問の答えは見つかりましたか? なぜなら、私たちも同じ問題を抱えているからです。 😊 よろしくお願いします。 Re: Pax device APDU NFC & Mifare Ultralight C こんにちは、ティボー。解決できましたか?私もあなたと同じことをしていますが、Mifare Classic 1kを使っていて、同じ問題を抱えています。私も助けを求めています。 進展があれば随時ご報告します。一緒にやり遂げられるといいな。 よろしくお願いします。
View full article
S32K3 SAI TDM problem Hi, NXP expert I'd like to ask about the S32K322 chip. What is the maximum supported block size for the SAI interface TDM? Is it frame size × width? That is, 16 × 32 bits = 512 bits, or 64 bytes? Thanks! Re: S32K3 SAI TDM问题 Hi @Chenxu1  You must also consider the maximum supported bit clock (BCLK) rate of 12.288 MHz. For example: TDM8, 16-bit, 48 kHz → BCLK = 8 × 16 × 48,000 = 6.144 MHz < 12.288 MHz (Supported) TDM16, 16-bit, 48 kHz → BCLK = 16 × 16 × 48,000 = 12.288 MHz (Supported at the limit) TDM16, 32-bit, 48 kHz → BCLK = 16 × 32 × 48,000 = 24.576 MHz > 12.288 MHz (Not supported) Therefore, a configuration of 16 words × 32 bits = 512 bits (64 bytes) represents the theoretical maximum frame size supported by the hardware. However, the practical configuration is also limited by the maximum supported BCLK rate. BR, VaneB Re: S32K3 SAI TDM问题 Hi,thanks for your reply. I mean, does the TDM DMA buffer have a length limit? 1024bytes? Re: S32K3 SAI TDM问题 Hi @Chenxu1  DMA transfers are limited by the DMA Major Loop Count implementation. Due to the CITER[14:0] the number of bytes transferred in a single DMA transaction must be less than 32,767 bytes.
View full article
CodeWarrior MPC82xx 调试版许可证 我需要使用 CodeWarrior for Power Architecture 8.8.x 调试 PowerQuicc MPC82xx 处理器。在NXP网站上查找后,似乎唯一的选择是价格昂贵的CW-DS-NETAPPS许可证。有没有更便宜的替代方案?我只需要能够通过我已有的TAP pod进行调试/JTAG操作即可。 Re: CodeWarrior License for MPC82xx Debug 您也可以从以下链接下载评估版(免费试用版),但试用期为 30 天。 每个客户账户(电子邮件地址)均可通过以下链接下载评估版。 请选择“下载评估版”。 https://www.nxp.com/design/software/development-software/codewarrior-development-tools/codewarrior-network-applications/codewarrior-development-suites-for-networked-applications:CW-DS-NETAPPS 谢谢!
View full article
MPC82xxデバッグ用CodeWarriorライセンス PowerQuicc MPC82xxプロセッサをCodeWarrior for Power Architecture 8.8.xでデバッグする必要があります。NXPのサイトを見回したところ、唯一の選択肢は非常に高価なCW-DS-NETAPPSライセンスのようです。これに使えるもっと安価なものはありますか?本当に必要なのは、すでに持っているTAPポッドを使ってデバッグやJTAGができるようになればいいだけです。 Re: CodeWarrior License for MPC82xx Debug また、以下のリンクから評価版(無料トライアル)をダウンロードすることもできますが、30日間の制限があります すべてのお客様のアカウント(メールアドレス)は、以下のリンクから評価版をダウンロードできます。 「評価版をダウンロード」を選択してください。 https://www.nxp.com/design/software/development-software/codewarrior-development-tools/codewarrior-network-applications/codewarrior-development-suites-for-networked-applications:CW-DS-NETAPPS よろしくお願いします。
View full article
Pax device APDU NFC & Mifare Ultralight C Hello, I'm developing on Pax device and I want to manage Mifare ULTRALIGHT C (MF0ICU2) Pax devices use a custom PayDroid OS and not a standard android, consequently, I don't have NFC android libraries.  I need to install something call NeptuneLiteAPI that expose IPCC via a library. First, I'm a little bit lost with naming convention & standard. Api is speaking about : ISO14443 A or B 1,2 or 3 M1 card Normally, Ultralight C is 14443 Type A, but I can't read my card with this option, I succeed to read with "M1 Card" option. What is a M1 card ? it it Classic 1k? Documentation mention that I need to use APDU and expose some command to execute this kind of call I succeed to read and write the card (with M1 function), the library expose default read/write method that use compatibility mode compatible with my card Now, i'm trying to configure authentication and to send some command to the chip, via the APDU command. Here is a simplication of the code I can use iPicc = getDal().getPicc(piccType); iPicc.open(); while(){ PiccCardInfo cardInfo = iPicc.detect(EDetectMode.ONLY_M); if(cardInfo!=null){ //Read : works byte[] page0=iPicc.m1Read((byte) 0); // Write : works iPicc.m1Write((byte)(13),hexStringToByteArray("AAAAAAAA000000000000000000000000")); } For writing, i have 3 methods: byte[] :  isoCommand(byte cid, byte[] send) ApduRespInfo : isoCommandByApdu(byte cid, ApduSendInfo apduSendInfo) byte[] cmdExchange(byte[] dataIn, int recvLen) All these method ask for an APDU command. if I understood well, Mifare Ultralight C does't respect APDU but APDU is the only way? And on some website, I found that using instruction class "FF", I can sent direct instruction to chip. But I'm lost in the command I need to send, and I try several. I have the error (unfortunatly, they are the one NeptuneLiteAPi and not a standard one). No specific card in sensing area, or card is not activated PICC#3(not search card error) PICC#97(unexpected error) But it's strange, I well detect the card just before and succeed to read/write, so the card is well "connected" I tried several commands (direct instruction, APDU instruction, CRC computation), nothing works. //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("30")); // PICC#97(unexpected error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("3004")); // PICC#97(unexpected error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("3000")); // PICC#97(unexpected error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("3000A802")); // Avec CRC => No specific card in sensing area, or card is not activated PICC#3(not search card error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("300002A8")); // Avec CRC inversé => No specific card in sensing area, or card is not activated PICC#3(not search card error) //byte[] cmd2 = iPicc.cmdExchange(CryptageUtils.hexStringToByteArray("3000A802"),17); // Card is not activated. PICC#3(not search card error) // byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("30000000")); // No specific card in sensing area, or card is not activated PICC#3(not search card error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("ff300000")); // No specific card in sensing area, or card is not activated PICC#3(not search card error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("ff30000F")); // No specific card in sensing area, or card is not activated PICC#3(not search card error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("FFCA00000004")); // No specific card in sensing area, or card is not activated PICC#3(not search card error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("3000000004")); // No specific card in sensing area, or card is not activated PICC#3(not search card error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("3001")); // No specific card in sensing area, or card is not activated PICC#3(not search card error)  If anyone can well me or put me on the right way, I would really appreciate. Regards. Re: Pax device APDU NFC & Mifare Ultralight C Hello @thibaut @Antoine1 @Erableto, did you ever find a solution to this issue? I stumbled upon the same problem: I cannot read counters on a Mifare Ultralight EV1; I tried with cmdExchange mainly, but also with all other methods provided by NeptuneLite library, but I always get errors like PICC#3 or #97. Re: Pax device APDU NFC & Mifare Ultralight C Hello @thibaut and @Erableto, did you find the answer to your question? Because we have the same problem on our side. 😊 Regards. Re: Pax device APDU NFC & Mifare Ultralight C Hi Thibaut. Did you figure it out? I'm doing the same as you but with a Mifare Classic 1k and I have the same problems. I'm looking for help too. I will keep you posted with any progress I manage to make. I hope we can get this done together. Regards.
View full article
Pax 设备 APDU NFC 和 Mifare Ultralight C 你好, 我正在开发 Pax 设备,并且想要管理Mifare ULTRALIGHT C (MF0ICU2) 卡。 Pax 设备使用定制的 PayDroid 操作系统,而不是标准的安卓系统,因此,我没有 NFC 安卓库。 我需要安装一个名为 NeptuneLiteAPI 的库,该库可以公开 IPCC。 首先,我对命名规则和标准有点困惑。API 指的是: ISO14443 A 或 B 1、2 或 3 M1卡 通常情况下,Ultralight C 是 14443 Type A,但我无法使用此选项读取我的卡,使用“M1 Card”选项可以成功读取。M1卡是什么?它是经典的1000K卡吗? 文档提到我需要使用 APDU 并公开一些命令来执行此类调用。 我成功地对卡进行了读写操作(使用 M1 函数),该库提供的默认读写方法使用了与我的卡兼容的兼容模式。 现在,我正在尝试配置身份验证,并通过 APDU 命令向芯片发送一些命令。 以下是我可以使用的代码的简化版本 iPicc = getDal().getPicc(piccType); iPicc.open(); while(){ PiccCardInfo cardInfo = iPicc.detect(EDetectMode.ONLY_M); if(cardInfo!=null){ //Read : works byte[] page0=iPicc.m1Read((byte) 0); // Write : works iPicc.m1Write((byte)(13),hexStringToByteArray("AAAAAAAA000000000000000000000000")); } 写作方面,我有三种方法: 字节[] : isoCommand(字节 cid, 字节[] send) ApduRespInfo : isoCommandByApdu(字节 cid, ApduSendInfo apduSendInfo) 字节[] cmdExchange(字节[] dataIn, int recvLen) 所有这些方法都需要 APDU 命令。 如果我理解正确的话,Mifare Ultralight C 不支持 APDU,但 APDU 却是唯一的传输方式? 我在一些网站上发现,使用指令类“FF”,可以直接向芯片发送指令。 但我不知道该发送什么命令,我尝试了好几个。 我遇到了错误(不幸的是,它们是 NeptuneLiteAPi 而不是标准版本)。 感应区域内未找到特定卡,或卡未激活 PICC#3(非搜索卡错误) PICC#97(意外错误) 但很奇怪,我之前明明能检测到这张卡,而且读写操作也成功了,所以这张卡应该是“连接正常”的。 我尝试了几个命令(直接指令、APDU 指令、CRC 计算),但都无效。 //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("30")); // PICC#97(unexpected error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("3004")); // PICC#97(unexpected error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("3000")); // PICC#97(unexpected error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("3000A802")); // Avec CRC => No specific card in sensing area, or card is not activated PICC#3(not search card error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("300002A8")); // Avec CRC inversé => No specific card in sensing area, or card is not activated PICC#3(not search card error) //byte[] cmd2 = iPicc.cmdExchange(CryptageUtils.hexStringToByteArray("3000A802"),17); // Card is not activated. PICC#3(not search card error) // byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("30000000")); // No specific card in sensing area, or card is not activated PICC#3(not search card error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("ff300000")); // No specific card in sensing area, or card is not activated PICC#3(not search card error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("ff30000F")); // No specific card in sensing area, or card is not activated PICC#3(not search card error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("FFCA00000004")); // No specific card in sensing area, or card is not activated PICC#3(not search card error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("3000000004")); // No specific card in sensing area, or card is not activated PICC#3(not search card error) //byte[] cmd2 = iPicc.isoCommand(cardInfo.getCID(), CryptageUtils.hexStringToByteArray("3001")); // No specific card in sensing area, or card is not activated PICC#3(not search card error) 如果有人能指点我一下或者给我指明正确的方向,我将不胜感激。 问候。 Re: Pax device APDU NFC & Mifare Ultralight C 你好@thibaut @Antoine1 @Erableto , 你找到解决这个问题的方法了吗?我也遇到了同样的问题:我无法读取 Mifare Ultralight EV1 上的计数器;我主要尝试使用 cmdExchange,但也尝试了 NeptuneLite 库提供的所有其他方法,但我总是收到类似 PICC#3 或 #97 的错误。 Re: Pax device APDU NFC & Mifare Ultralight C 你好@thibaut和@Erableto , 你找到问题的答案了吗? 因为我们这边也存在同样的问题。 😊 此致问候 Re: Pax device APDU NFC & Mifare Ultralight C 你好,Thibaut。你弄明白了吗?我和你做的一样,但我用的是 Mifare Classic 1k 卡,也遇到了同样的问题。我也需要帮助。 我会及时向您汇报任何进展。我希望我们能一起完成这件事。 此致问候
View full article
FRDM i.MX 95 Pro Hands-On: NPU Inference with eIQ Neutron Introduction This hands-on video explains how to take a trained AI model and deploy it for NPU-accelerated inference using eIQ Neutron on the FRDM i.MX 95 PRO development board. YOLOv8n is used as the example model. The video covers the complete workflow, including host setup, model export, calibration, INT8 quantization, Neutron compilation, board deployment, benchmarking, and troubleshooting. By the End of This Training, You Will Be Able To Understand the main components of the eIQ Neutron-S hardware and software stack. Download and export a YOLOv8n model to TFLite. Quantize the model to INT8 for Neutron NPU execution. Compile the quantized model for the i.MX 95 target. Review compiler statistics and determine which operations are delegated to the NPU. Deploy the model, delegate, driver, and firmware to the board. Run CPU and NPU benchmarks. Compare inference latency and calculate the NPU speed-up. Hardware and Prerequisites Before starting the training, verify that you have the required development board, host system, network connection, model, calibration data, and software tools. Development Board FRDM i.MX 95 PRO development board. NXP BSP LF6.18.20 2.0.0. eIQ Neutron runtime compatible with Neutron SDK 3.2.1 or later. TensorFlow Lite 2.19.0 benchmark application. Ethernet or USB-CDC connectivity. SSH and SCP access to the board. Root access on the target system. Linux Host Computer Ubuntu 20.04 or Ubuntu 22.04 is recommended. Python 3.8 or later. Python virtual environment support. SSH and SCP command-line tools. Sufficient storage for the SDK, models, calibration images, and generated artifacts.     Watch the Complete Training Video Watch the video from beginning to end and execute the commands in the presented order. Each stage generates an artifact required by the next stage. (function() { var wrapper = document.getElementById('lia-vid-6406503147112w960h540r369'); var videoEl = wrapper ? wrapper.querySelector('video-js') : null; if (videoEl) { if (window.videojs) { window.videojs(videoEl).ready(function() { this.on('loadedmetadata', function() { this.el().querySelectorAll('.vjs-load-progress div[data-start]').forEach(function(bar) { bar.setAttribute('role', 'presentation'); bar.setAttribute('aria-hidden', 'true'); }); }); }); } }})(); (view in My Videos)     Command Cheat Sheet The following commands are organized in execution order. Replace placeholders such as with the values for your environment. 1. Download and Configure the Neutron SDK # Extract the SDK archive unzip eiq-neutron-sdk-linux-3.2.1.zip # Enter the SDK directory cd eiq-neutron-sdk-linux-3.2.1 # Add the SDK tools to PATH for the current terminal export PATH="$PWD/bin:$PATH" # Optional: make the PATH configuration persistent echo 'export PATH="/path/to/eiq-neutron-sdk-linux-3.2.1/bin:$PATH"' >> ~/.bashrc # Reload the shell configuration source ~/.bashrc # Verify the SDK tools tflite-profiler --help tflite-quantizer --help neutron-compiler --help   2. Create the Python Environment # Create a Python virtual environment python3 -m venv .venv # Activate the virtual environment source .venv/bin/activate # Upgrade pip python3 -m pip install --upgrade pip # Install the required packages pip install generate-parameter-library-py \ ultralytics \ hf \ ai-edge-litert \ litert_torch \ numpy \ opencv-python   3. Download the YOLOv8n Model # Download the trained YOLOv8n weights hf download Ultralytics/YOLOv8 yolov8n.pt --local-dir . # Verify the downloaded model ls -lh yolov8n.pt   4. Export YOLOv8n to TFLite # Export the PyTorch model to TFLite yolo export model=yolov8n.pt format=tflite # Locate the generated TFLite model find . -name "*.tflite" -type f # Update the following commands if your exported # model uses a different file name or directory.   5. Download the COCO Calibration Dataset # Create a directory for the calibration dataset mkdir -p coco cd coco # Download the COCO training images wget --no-check-certificate \ http://images.cocodataset.org/zips/train2017.zip # Extract the images unzip train2017.zip # Return to the project directory cd .. # Verify the image directory find ./coco/train2017 -type f -name "*.jpg" | head   6. Create the Calibration Conversion Script Save the following script as jpg2bin.py . It converts the JPEG calibration images into raw binary tensors expected by the profiler. #!/usr/bin/env python3 import argparse import glob import os import random import cv2 import numpy as np def letterbox(image, new_shape=640, color=(114, 114, 114)): height, width = image.shape[:2] ratio = min(new_shape / height, new_shape / width) resized_height = int(round(height * ratio)) resized_width = int(round(width * ratio)) resized = cv2.resize( image, (resized_width, resized_height), interpolation=cv2.INTER_LINEAR, ) canvas = np.full( (new_shape, new_shape, 3), color, dtype=np.uint8, ) top = (new_shape - resized_height) // 2 left = (new_shape - resized_width) // 2 canvas[ top:top + resized_height, left:left + resized_width, ] = resized return canvas def main(): parser = argparse.ArgumentParser( description="Convert JPEG images to YOLOv8n calibration tensors." ) parser.add_argument( "--images", default="./coco/train2017", help="Directory containing the calibration JPEG images.", ) parser.add_argument( "--out", default="./calib_yolov8n", help="Output directory for the binary tensors.", ) parser.add_argument( "--num", type=int, default=1000, help="Maximum number of calibration images.", ) parser.add_argument( "--size", type=int, default=640, help="Model input width and height.", ) parser.add_argument( "--shuffle", action="store_true", help="Shuffle the image list before selecting images.", ) args = parser.parse_args() os.makedirs(args.out, exist_ok=True) files = sorted( glob.glob(os.path.join(args.images, "*.jpg")) ) if args.shuffle: random.seed(0) random.shuffle(files) files = files[:args.num] if not files: raise RuntimeError( f"No JPEG images were found in {args.images}" ) for index, file_path in enumerate(files): image = cv2.imread(file_path) if image is None: print(f"Skipping unreadable image: {file_path}") continue # Convert BGR to RGB. image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB) # Letterbox resize to 640 x 640. image = letterbox(image, args.size) # Convert to float32 and normalize to [0, 1]. tensor = image.astype(np.float32) / 255.0 # Add batch dimension. # Final layout: 1 x 640 x 640 x 3, NHWC. tensor = np.expand_dims(tensor, axis=0) output_path = os.path.join( args.out, f"calib_{index:05d}.bin", ) tensor.tofile(output_path) print( f"Calibration tensors written to: {args.out}" ) if __name__ == "__main__": main() # Make the script executable chmod +x jpg2bin.py # Generate up to 1000 calibration tensors python3 jpg2bin.py \ --images ./coco/train2017 \ --out ./calib_yolov8n \ --num 1000 \ --size 640 \ --shuffle # Count the generated binary files find ./calib_yolov8n -type f -name "*.bin" | wc -l # Check the size of a generated tensor ls -lh ./calib_yolov8n | head   7. Inspect the TFLite Input and Output Tensors python3 - <<'PY' from ai_edge_litert.interpreter import Interpreter model_path = "yolov8n.tflite" interpreter = Interpreter(model_path=model_path) interpreter.allocate_tensors() print("Input details:") for tensor in interpreter.get_input_details(): print(tensor) print("\nOutput details:") for tensor in interpreter.get_output_details(): print(tensor) PY Record the exact input tensor name reported by the model. Replace serving_default_args_0 in the profiling command if your model uses a different name.   8. Profile the Floating-Point TFLite Model # Replace serving_default_args_0 if the model # reports a different input tensor name. tflite-profiler \ --input yolov8n.tflite \ --dataset serving_default_args_0,calib_yolov8n \ --output yolov8n_profile.csv # Verify that the profiling CSV was generated ls -lh yolov8n_profile.csv # Preview the beginning of the profile head yolov8n_profile.csv   9. Quantize the Model to INT8 # Generate the INT8 TFLite model tflite-quantizer \ --input yolov8n.tflite \ --profile yolov8n_profile.csv \ --output yolov8n_quant.tflite # Compare the model file sizes ls -lh yolov8n.tflite yolov8n_quant.tflite   10. Verify the Quantized Model python3 - <<'PY' from ai_edge_litert.interpreter import Interpreter model_path = "yolov8n_quant.tflite" interpreter = Interpreter(model_path=model_path) interpreter.allocate_tensors() print("Quantized input details:") for tensor in interpreter.get_input_details(): print("Name:", tensor["name"]) print("Shape:", tensor["shape"]) print("Data type:", tensor["dtype"]) print("Quantization:", tensor["quantization"]) print() print("Quantized output details:") for tensor in interpreter.get_output_details(): print("Name:", tensor["name"]) print("Shape:", tensor["shape"]) print("Data type:", tensor["dtype"]) print("Quantization:", tensor["quantization"]) print() PY   11. Compile the Model for the Neutron NPU # Compile the INT8 model for the i.MX 95 Neutron NPU neutron-compiler \ --input yolov8n_quant.tflite \ --output yolov8n_neutron.tflite \ --target imx95 # Verify the generated model ls -lh yolov8n_neutron.tflite Generate Compiler Statistics # Generate compiler and delegation statistics neutron-compiler \ --input yolov8n_quant.tflite \ --output yolov8n_neutron_stats.tflite \ --target imx95 \ --dump-statistics # Review the compiler output and record: # 1. Number of delegated nodes # 2. Number of CPU fallback nodes # 3. Unsupported operators # 4. Generated NeutronGraph partitions Create a Profiling Build # Compile a profiling-enabled model neutron-compiler \ --input yolov8n_quant.tflite \ --output yolov8n_neutron_prof.tflite \ --target imx95 \ --use-profiling # Verify all generated models ls -lh yolov8n_quant.tflite \ yolov8n_neutron.tflite \ yolov8n_neutron_prof.tflite   12. Configure the Board IP Address # Replace the example address with the board IP address export BOARD_IP="192.168.1.100" # Verify network connectivity ping -c 4 "$BOARD_IP" # Test the SSH connection ssh root@"$BOARD_IP"   13. Create the Deployment Directory # Create the working directory remotely ssh root@"$BOARD_IP" \ "mkdir -p /root/neutron_demo"   14. Copy Runtime Artifacts to the Board The driver, delegate, and firmware may already be included in the BSP. Copy them only when the required or matching versions are not already installed. # Copy the Neutron driver scp libNeutronDriver.so \ root@"$BOARD_IP":/lib/ # Copy the Neutron delegate scp libneutron_delegate.so \ root@"$BOARD_IP":/usr/lib/ # Copy the Neutron firmware scp NeutronFirmware.elf \ root@"$BOARD_IP":/lib/firmware/ # Copy the plain INT8 model for the CPU baseline scp yolov8n_quant.tflite \ root@"$BOARD_IP":/root/neutron_demo/ # Copy the Neutron-compiled model scp yolov8n_neutron.tflite \ root@"$BOARD_IP":/root/neutron_demo/ # Optional: copy the profiling-enabled model scp yolov8n_neutron_prof.tflite \ root@"$BOARD_IP":/root/neutron_demo/   15. Verify the Files on the Board # Connect to the board ssh root@"$BOARD_IP" # Enter the deployment directory cd /root/neutron_demo # Verify the models ls -lh /root/neutron_demo/ # Check the driver ls -lh /lib/libNeutronDriver.so # Check the delegate ls -lh /usr/lib/libneutron_delegate.so # Check the firmware ls -lh /lib/firmware/NeutronFirmware.elf # Refresh the shared-library cache ldconfig # Verify that the delegate is visible to the loader ldconfig -p | grep -i neutron   16. Check the BSP and Runtime Environment # Display the Linux distribution and BSP information cat /etc/os-release # Display the kernel version uname -a # Search for Neutron-related kernel messages dmesg | grep -i neutron # Search for firmware-related messages dmesg | grep -i firmware # Locate the TensorFlow Lite benchmark application find /usr/bin -name "benchmark_model*" -type f 2>/dev/null   17. Run the CPU Baseline cd /root/neutron_demo /usr/bin/tensorflow-lite-2.19.0/examples/benchmark_model \ --graph=yolov8n_quant.tflite \ --num_threads=4 \ --num_runs=50 Record the average inference latency, minimum latency, maximum latency, initialization time, and memory information reported by the benchmark.   18. Run the NPU Benchmark cd /root/neutron_demo /usr/bin/tensorflow-lite-2.19.0/examples/benchmark_model \ --graph=yolov8n_neutron.tflite \ --external_delegate_path=/usr/lib/libneutron_delegate.so \ --num_runs=50 Confirm that the benchmark output reports that the external delegate was loaded and that one or more model partitions were delegated.   19. Save Benchmark Results # Save the CPU benchmark output /usr/bin/tensorflow-lite-2.19.0/examples/benchmark_model \ --graph=yolov8n_quant.tflite \ --num_threads=4 \ --num_runs=50 \ 2>&1 | tee cpu_benchmark.txt # Save the NPU benchmark output /usr/bin/tensorflow-lite-2.19.0/examples/benchmark_model \ --graph=yolov8n_neutron.tflite \ --external_delegate_path=/usr/lib/libneutron_delegate.so \ --num_runs=50 \ 2>&1 | tee npu_benchmark.txt # Review the saved results cat cpu_benchmark.txt cat npu_benchmark.txt   20. Calculate the NPU Speed-Up Use the average latency reported by each benchmark: Speed-up = CPU average latency / NPU average latency Example: CPU average latency = 100 ms NPU average latency = 10 ms Speed-up = 100 / 10 Speed-up = 10x   Required Deployment Artifacts Artifact Example File Purpose Neutron driver libNeutronDriver.so Communicates with the Neutron firmware and passes model buffers, weights, kernels, and input or output data. Neutron delegate libneutron_delegate.so Connects TensorFlow Lite to the Neutron runtime and delegates supported NeutronGraph operations to the NPU. Neutron firmware NeutronFirmware.elf Runs on the dedicated RISC-V core and executes the generated Neutron microcode. Quantized TFLite model yolov8n_quant.tflite Provides the CPU baseline and serves as the input to the Neutron compiler. Neutron-compiled model yolov8n_neutron.tflite Contains NeutronGraph operations generated for NPU execution.   Troubleshooting Use the following table to identify common host, model, compilation, deployment, and runtime problems. Area Problem Possible Cause Recommended Solution Host setup SDK commands are not found. The SDK bin directory is not included in PATH . Run export PATH="$PWD/bin:$PATH" from the SDK directory. Add the absolute SDK path to ~/.bashrc if the configuration must persist. Python environment A required Python module cannot be imported. The virtual environment is not active or the package was installed in a different environment. Run source .venv/bin/activate . Verify the active interpreter with which python3 , and reinstall the missing dependency with pip install PACKAGE_NAME . Model export The expected yolov8n.tflite file cannot be found. The exporter created a subdirectory or used a different filename. Run find . -name "*.tflite" -type f . Update the following commands to use the actual exported model path. Calibration No calibration binary files are generated. The image directory is incorrect or does not contain JPEG files. Verify the directory with find ./coco/train2017 -name "*.jpg" | head . Pass the correct directory through the --images argument. Calibration The calibration tensor has the wrong shape or size. The resize, batch dimension, layout, or data type is incorrect. Confirm an NHWC tensor with shape 1 x 640 x 640 x 3 , RGB channel order, float32 data, and values normalized to [0, 1] . Calibration The quantized model produces inaccurate predictions. The calibration preprocessing differs from inference preprocessing. Verify letterbox resizing, padding color, RGB conversion, normalization, tensor layout, input size, and data type. Use representative images that match the intended application. Profiler The profiler reports an invalid or unknown input tensor. The tensor name in --dataset does not match the model input name. Inspect the model input details with the TFLite interpreter. Replace serving_default_args_0 with the exact name reported by the model. Profiler The profiler cannot read the calibration files. The dataset path is incorrect, the folder is empty, or the binary format does not match the input tensor. Verify the folder path, count the files, and confirm the expected number of bytes for each tensor. Quantization The quantized model still uses floating-point tensors. The model was not fully quantized or the generated profile is incomplete. Re-run profiling with valid calibration data. Inspect the input, output, and intermediate tensor data types before compiling the model. Compilation neutron-compiler fails. The input model is invalid, not INT8, or contains unsupported quantization parameters. Confirm that the input is the quantized TFLite model. Inspect its tensor types and quantization parameters, then run the compiler again with statistics enabled.   Conclusion Congratulations! You have completed the end-to-end workflow for deploying and running an AI model on the eIQ Neutron-S NPU integrated into the FRDM i.MX 95 PRO. Throughout this training, you learned how to prepare the Linux host environment, export YOLOv8n to TFLite, build a representative calibration dataset, quantize the model to INT8, and compile it for the i.MX 95 target. You also learned how to transfer the required runtime artifacts to the board and compare CPU and NPU inference performance. This workflow is not limited to YOLOv8n. You can use the same process as a starting point for deploying your own computer vision models. However, each model must be evaluated individually because input formats, preprocessing requirements, quantization behavior, and operator support may differ. Recommended Next Steps Validate the accuracy of the INT8 model against the original floating-point model. Review the compiler statistics to identify unsupported operations and CPU fallback sections. Measure end-to-end application performance, including preprocessing, inference, and post-processing. Test the workflow with your own model and a calibration dataset that represents your real application. Use a profiling-enabled build to identify the most expensive layers and potential performance bottlenecks. FRDM-IMX9 FRDM-Training Hands-On Training i.MX Application Processors
View full article
Title: S32K344 + FS26: external watchdog during HSE FW installation by the SBAF Hello, We need to install the HSE firmware (FULL_MEM 0.2.55) in the field on S32K344 boards that use an FS26 SBC, without a debugger. Our software writes the pink image in code flash, sets IVT+0x2C to its address, programs the HSE FW feature flag in UTEST and triggers a destructive reset. The SBAF then installs the firmware. The FS26 runs a window watchdog (128 ms, challenger, reaction RSTB + FS0B). As far as we understand, the FS26 RSTB is connected to the MCU RESET_B, so after the destructive reset the FS26 goes back to INIT_FS and our software must close INIT_FS (first watchdog refresh) within the INIT_FS timeout (about 256 ms). While the SBAF installs the firmware, none of our code runs and nobody serves the FS26. In another thread NXP mentioned that the installation takes about 0.8-1 s, which is longer than the INIT_FS timeout. The attached diagram shows what we expect (A) and what we are afraid of (B): the FS26 resets the MCU during the installation, the SBAF starts again because the UTEST flag and the IVT pointer are still set, and the board ends up in a reset loop. Our questions: 1. Is this scenario real? Does the installation by the SBAF actually take longer than the FS26 INIT_FS timeout? 2. If the installation is interrupted by an external reset (RESET_B pulled low by the FS26), is it safe? Does the SBAF restart it from scratch at the next reset, or can the HSE be left partially installed? 3. What is the recommended way to install the HSE firmware on a board with an FS26 ? Installing with the FS26 in debug mode? Is there any other option, in the FS26 configuration or in the SBAF/IVT, so that the installation can complete in the field? (Diagram attached: FS26_HSE_install_timing.png. Times not to scale.) Thank you in advance. Best regards, Fabio Re: Title: S32K344 + FS26: external watchdog during HSE FW installation by the SBAF Hi @FabioG  1. The HSE firmware installation process can take approximately 1 second to complete. After the reset, the host application must wait until HSE_STATUS_INIT_OK is set before issuing any HSE service requests or proceeding with initialization. 2. Power instability, unexpected resets during HSE installation/update, failure to wait for HSE_STATUS_INIT_OK, execution of BIST before HSE_STATUS_INIT_OK is reached, or unsupported clock configurations may corrupt or erase the HSE FW. Also it is recommended to use a functional reset rather than a power-on reset during the installation process and to ensure that the power supply remains stable throughout the operation. 3. Using the FS26 in Debug Mode is recommended during development and programming activities, as the watchdog can be configured to provide either an 8-second timeout window or an unlimited open window. This helps avoid automatic resets and reduces the risk of interrupting HSE operations. BR, VaneB Re: Title: S32K344 + FS26: external watchdog during HSE FW installation by the SBAF Hi @VaneB   I have another question:  First start after the installation. After the SBAF has installed the HSE firmware, our bootloader starts and, during its normal start-up, performs a software functional reset to start the application, possibly before HSE_STATUS_INIT_OK is set. Is a functional reset at that moment safe, or is the installation still in progress until HSE_STATUS_INIT_OK is set for the first time? Fabio Re: Title: S32K344 + FS26: external watchdog during HSE FW installation by the SBAF Hi @FabioG  The BIST test should only be started after SBAF has entered the WFI state and the HSE_STATUS_INIT_OK bit is set. Regarding the wait time for HSE_STATUS_INIT_OK to be asserted, there is no fixed time. The initialization time depends on several factors, including the core frequency and the HSE firmware version. Therefore, measuring the HSE boot time on your target platform is the best way to determine an appropriate timeout value. Finally, it is not recommended to assume that a functional reset is safe before HSE_STATUS_INIT_OK is asserted. The HSE installation/initialization may still be in progress at that stage. As a result, the system should treat the startup process as incomplete until HSE_STATUS_INIT_OK is reached. Re: Title: S32K344 + FS26: external watchdog during HSE FW installation by the SBAF Hi VaneB, thank you for the answer. A short follow-up on the BIST point (2). Our software runs the BIST at every power-on, early in the start-up. Following your advice, on devices with the HSE firmware installed we could wait for HSE_STATUS_INIT_OK before starting the BIST. If HSE_STATUS_INIT_OK is not set within a short time (how much?), we would skip the BIST for that power cycle. Would this be enough to avoid any risk for the HSE firmware, or is there anything else we should take care of? Thank you. Best regards, Fabio
View full article
タイトル: S32K344 + FS26: SBAFによるHSE FWインストール中の外部ウォッチドッグ こんにちは、 デバッガを使用せずに、FS26 SBCを使用するS32K344ボードにHSEファームウェア(FULL_MEM 0.2.55)を現場でインストールする必要があります。私たちのソフトウェアはコードフラッシュでピンク画像を書き込み、IVT+0x2Cをそのアドレスに設定し、UTESTでHSE FW機能フラグをプログラムして破壊的なリセットをトリガーします。SBAFはその後ファームウェアをインストールします。 FS26はウィンドウウォッチドッグを実行します(128ms、チャレンジャー、反応RSTB + FS0B)。私たちの理解では、FS26のRSTBはMCU RESET_Bに接続されており、破壊的なリセット後はFS26がINIT_FSに戻り、ソフトウェアはINIT_FSタイムアウト(約256ms)内にINIT_FSを閉じ(最初のウォッチドッグリフレッシュ)しなければなりません。 SBAFがファームウェアをインストールしている間は、私たちのコードは一切実行されず、誰もFS26にサービスを提供しません。別のThreadでNXPはインストールに約0.8〜1秒かかると言っていましたが、これはINIT_FSタイムアウトよりも長い時間です。添付の図は、私たちが期待すること(A)と恐れていること(B)を示しています:FS26は設置中にMCUをリセットし、UTESTフラグとIVTポインタがまだ設定されているためSBAFが再起動し、基板はリセットループに陥ります。 私たちの質問: 1. このシナリオは現実のものですか?SBAFによるインストールは、FS26のINIT_FSタイムアウトよりも実際に時間がかかるのでしょうか? 2. インストール中に外部リセット(FS26によってRESET_Bがローにプルダウンされる)によって中断された場合、安全ですか?SBAFは次のリセット時に一から再起動しますか?それともHSEは部分的にインストールしたままにしてもいいのでしょうか? 3. FS26搭載のボードにHSEファームウェアをインストールする推奨方法は何でしょうか?FS26をデバッグモードでインストールしますか?現場で設置を完了させるために、FS26構成やSBAF/IVTで他に選択肢はありますか? (添付の図:FS26_HSE_install_timing.png)(時間は縮尺通りではありません。) 事前に感謝いたします。 よろしくお願いします、 ファビオ Re: Title: S32K344 + FS26: external watchdog during HSE FW installation by the SBAF こんにちは、 @FabioGさん 1.HSEファームウェアのインストールプロセスは約1秒で完了することがあります。リセット後、ホストアプリケーションはHSE_STATUS_INIT_OKが設定されるまでHSEサービスリクエストを発行したり初期化を進めたりするのを待たなければなりません。 2.電源の不安定性、HSE のインストール/アップデート中の予期しないリセット、HSE_STATUS_INIT_OK を待たないこと、HSE_STATUS_INIT_OK に到達する前に BIST が実行されること、またはサポートされていないクロック構成により、HSE ファームウェアが破損または消去される可能性があります。 また、インストールプロセス中は、電源投入リセットではなく機能リセットを使用することを推奨します。さらに、動作中は電源供給が安定していることを確認してください。 3. 開発およびプログラミング活動中は、監視ドッグが8秒のタイムアウトウィンドウまたは無制限の開ウィンドウのいずれかを提供できるように設定できるため、デバッグモードでFS26を使用することが推奨されます。これにより、自動リセットを回避し、HSE(安全衛生環境)業務の中断リスクを軽減できます。 BR、VaneB Re: Title: S32K344 + FS26: external watchdog during HSE FW installation by the SBAF こんにちは、 @VaneBさん もう一つ質問があります。 インストール後、最初に起動してください。SBAFがHSEファームウェアをインストールした後、ブートローダーが起動し、通常の起動時にソフトウェア機能リセットを行いアプリケーションを起動します。場合によってはHSE_STATUS_INIT_OK設定前に行われます。その時点で機能リセットを実行しても安全でしょうか、それともHSE_STATUS_INIT_OKが初めて設定されるまでインストールは継続中でしょうか? ファビオ Re: Title: S32K344 + FS26: external watchdog during HSE FW installation by the SBAF こんにちは、VaneBさん。 ご回答ありがとうございます。BISTの点(2)について簡単に補足します。 私たちのソフトウェアは、起動初期の電源オン時にBISTを実行します。あなたのアドバイスに従い、HSEファームウェアがインストールされているデバイスならHSE_STATUS_INIT_OKを待ってからBISTを起動できます。HSE_STATUS_INIT_OKが短時間(どのくらいの時間?)以内に設定されない場合、その電源サイクルではBISTをスキップします。 これでHSEファームウェアへのリスクを回避できるでしょうか?それとも他に注意すべき点はありますか? よろしくお願いします。 よろしくお願いします、 ファビオ Re: Title: S32K344 + FS26: external watchdog during HSE FW installation by the SBAF こんにちは、 @FabioGさん BISTテストは、SBAFがWFI状態に入り、HSE_STATUS_INIT_OKビットが設定された後にのみ開始する必要があります。 HSE_STATUS_INIT_OKがアサートされるまでの待機時間については、決まった時間はありません。初期化時間は、コア周波数やHSEファームウェアのバージョンなど、いくつかの要因によって異なります。したがって、ターゲットプラットフォームでのHSEブート時間を測定することが、適切なタイムアウト値を決定する最良の方法です。 最後に、HSE_STATUS_INIT_OKがアサートされる前に機能リセットが安全であると想定することは推奨されません。その段階では、HSE(環境・安全・衛生)システムの設置/初期化がまだ進行中である可能性があります。その結果、システムはHSE_STATUS_INIT_OKに到達するまで、起動プロセスが未完了であるとみなすべきである。
View full article
标题:S32K344 + FS26:SBAF 在 HSE 固件安装期间的外部看门狗 你好, 我们需要在没有调试器的情况下,在现场将 HSE 固件(FULL_MEM 0.2.55)安装到使用 FS26 SBC 的 S32K344 板上。我们的软件将粉色图像写入代码闪存,将 IVT+0x2C 设置为其地址,在 UTEST 中对 HSE FW 功能标志进行编程,并触发信号破坏性复位。然后 SBAF 安装固件。 FS26 运行窗口监视程序(128 毫秒,挑战者,反应 RSTB + FS0B)。据我们了解,FS26 RSTB 连接到 MCU RESET_B,因此在破坏性复位后,FS26 会返回到 INIT_FS,我们的软件必须在 INIT_FS 超时(约 256 毫秒)内关闭 INIT_FS(第一次看门狗刷新)。 在 SBAF 安装固件期间,我们的任何代码都不会运行,也没有人为 FS26 提供服务。在另一个帖子中,NXP提到安装大约需要0.8-1秒,这比INIT_FS超时时间要长。附图显示了我们期望的结果 (A) 和我们担心的结果 (B):FS26 在安装过程中 RESET MCU,由于 UTEST 标志和 IVT 指针仍然设置,SBAF 再次启动,最终板进入 RESET 循环。 我们的疑问: 1. 这种情况是真的吗?SBAF 的安装时间是否真的比 FS26 INIT_FS 超时时间更长? 2. 如果安装过程被外部复位(FS26 将 RESET_B 拉低)中断,是否安全?SBAF 会在下次 RESET 时从头开始重新启动吗?还是可以保留部分安装的 HSE? 3. 在带有 FS26 的板上安装 HSE 固件的推荐方法是什么?是否在调试模式下安装 FS26?在 FS26 配置或 SBAF/IVT 中,是否有其他选项可以让安装能够在现场完成? (附图:FS26_HSE_install_timing.png。时间未按比例绘制。) 提前谢谢您。 此致, 法比奥 Re: Title: S32K344 + FS26: external watchdog during HSE FW installation by the SBAF 嗨@FabioG 1.HSE固件安装过程大约需要1秒钟即可完成。重置后,主机应用程序必须等到 HSE_STATUS_INIT_OK 设置后才能发出任何 HSE 服务请求或继续进行初始化。 2.电源不稳定、HSE 安装/更新期间意外 RESET、未等待 HSE_STATUS_INIT_OK、在达到 HSE_STATUS_INIT_OK 之前执行 BIST 或不支持的时钟配置可能会损坏或擦除 HSE 固件。 此外,建议在安装过程中使用功能复位而不是上电复位,并确保电源在整个运行过程中保持稳定。 3.建议在开发和编程活动期间使用 FS26 的调试模式,因为可以将看门狗配置为提供 8 秒超时窗口或无限期打开窗口。这有助于避免自动RESET,并降低中断 HSE 操作的风险。 BR,VaneB Re: Title: S32K344 + FS26: external watchdog during HSE FW installation by the SBAF 嗨 VaneB, 谢谢你的回答。对 BIST 点 (2) 的简要后续说明。 我们的软件会在每次上电时,在启动初期运行 BIST。按照您的建议,对于安装了 HSE 固件的设备,我们可以在启动 BIST 之前等待 HSE_STATUS_INIT_OK。如果在短时间内(多久?)未设置 HSE_STATUS_INIT_OK,我们将跳过该电源循环的 BIST。 这样是否足以避免 HSE 固件的任何风险,还是还有其他需要注意的地方? 谢谢! 此致, 法比奥 Re: Title: S32K344 + FS26: external watchdog during HSE FW installation by the SBAF 嗨@FabioG 只有在 SBAF 进入 WFI 状态且 HSE_STATUS_INIT_OK 位设置后,才应启动 BIST 测试。 关于 HSE_STATUS_INIT_OK 钳位的等待时间,没有固定的时间。初始化时间取决于几个因素,包括核心频率和 HSE 固件版本。因此,测量目标平台上的 HSE 启动时间是确定合适的超时值的最佳方法。 最后,不建议在 HSE_STATUS_INIT_OK 钳位之前假定功能复位是安全的。届时 HSE 安装/初始化工作可能仍在进行中。因此,系统应将启动过程视为未完成,直到达到 HSE_STATUS_INIT_OK 状态。 Re: Title: S32K344 + FS26: external watchdog during HSE FW installation by the SBAF 嗨@VaneB 我还有个问题: 安装完成后首次启动。SBAF 安装 HSE 固件后,我们的引导加载程序启动,并在正常启动期间执行软件功能 RESET 以启动应用程序,这可能发生在 HSE_STATUS_INIT_OK 设置之前。此时进行功能 RESET 是否安全?或者说,在 HSE_STATUS_INIT_OK 首次被设置之前,安装仍在进行中? 法比奥
View full article
求助:S32K312 上的 HSE 固件刷新 您好,NXP团队: 我目前正在研究S32K312上的HSE固件。我已经有了HSE固件的BIN文件,但我不知道如何正确地将其刷写到S32K312上。 我尝试使用 PEmicro 烧录 BIN 文件 ,但失败了。我还尝试 在 S32 Design Studio 中使用 IVT 配置 ,但当我选择 S32K312 时 ,它显示“不支持的处理器”。 我已经查阅了所有可用的示例和帖子,但我仍然卡在 HSE 固件刷写这一步。 请问能否提供在S32K312上刷入HSE BIN固件并使HSE正常运行的正确且最简短的步骤? 请具体指导我如何以及在哪里刷入 BIN 文件,以及刷入后需要做什么。 (我已将 Bin 文件打包在 Zip 文件中),请查收。 @nxp 谢谢,此致敬礼! 索赫尔·贾卡蒂 Re: Help Required: HSE Firmware Flashing on S32K312 嗨@Mohmedsohel 您是否已经查看过S32K3 MCU 通用 HSE 演示示例中提供的 S32K344_HSE_FW_INSTALL 演示? 另外,您提到您尝试使用 S32 配置工具中的 IVT 工具。但是,该工具不支持 S32K3 设备。对于 S32K3,可以直接在 S32DS 项目的“项目设置”→“启动代码”→“startup_cm7.s”下修改 IVT。 BR,VaneB
View full article
Fault Injection to cache location using an interface provide in emcem Hello team, I have addition question to below query (what values to select as arguments for eMcem_setupinjectionChannel - NXP Community😞 I need to inject a 2-bit fault in Cache location but I'm not sure what it means as mentioned in the reference manual and we use NXP s32K394 microcontroller: I need to call below function 2 times to inject the faults in cache and get the 2-bit error in cache like below:  If I'm doing wrongly, could you please share the example code how to do this injection, and currently we are using SAF package in our project, so if there are already fault injection code exists for the these, please share which file I need to refer to inject all kind of faults using (EIM) for memory test interface which is provided in SAF module. Re: Fault Injection to cache location using an interface provide in emcem Hello @Rudved_3366, I haven’t found an example of this cache injection, but here it is in a nutshell: Before injecting the CM7_0 I-cache tag double-bit error, verify that the relevant I-cache error-bank entries (IEBRn) are available and unlocked. I tested it on S32K344, which I have on the bench, I injected:  if (eMcem_SetupInjectionChannel(EMCEM_EIM_CH_3, 1, 2 ) != (Std_ReturnType)E_OK) { while (1U); } if (eMcem_InjectFault(EMCEM_EIM_CH_3) != (Std_ReturnType)E_OK) { while (1U); } Then, in the handler, disable the EIM. void BusFault_Handler(void) { if(IP_EIM->EIMCR & EIM_EIMCR_GEIEN_MASK) { IP_EIM->EIMCR = EIM_EIMCR_GEIEN(0); } } NOTE that the recovery code (such as the vector table referenced by VTOR, the exception handler, and whatever the handler calls) must not be in a cacheable region, because we do not want the error to be injected again during recovery. Then, read the IEBR registers again and clear them by writing 0 to them. Call an API to invalidate the cache. Regards, Daniel Re: Fault Injection to cache location using an interface provide in emcem Hello @danielmartynek ,  can you please check above query also which is related to fault injection to cache address.
View full article
s32ds 许可证已过期 你好, 代码:4F70-F0A3-101C-3C96 我谨此申请续订 S32DS 2018R1 许可证。 感谢您的支持。 顺祝商祺! Re: s32ds license Expired 您好, 您的S32DS许可证已延期。
View full article
MCX W72 Product Lifecycle and EOL Inquiry Hello NXP Team, We are evaluating the MCX W72 / FRDM-MCXW72 for a new design and would like to request information regarding Current lifecycle status Expected longevity/availability period Any planned EOL or NRND status Existing PCN or EOL notifications Recommended successor devices, if applicable Please share any available lifecycle documentation. FRDM-Training Re: MCX W72 Product Lifecycle and EOL Inquiry Thank you for the clarification and detailed information regarding the MCX W72 product longevity and lifecycle status. Re: MCX W72 Product Lifecycle and EOL Inquiry Hello, hope you are doing well. The MCX W72 is included in NXP’s Product Longevity Program until 2040, and is the recommended product for new designs. This product follow NXP's standard Product Lifecycle: it is fully active today, and any future status change would be communicated through NXP's standard discontinuance notification process, all applicable NXP End of Life and Product Change Notification policies continue to apply. Best regards, Sofia.
View full article
s32dsライセンスの有効期限が切れました こんにちは、 コード: 4F70-F0A3-101C-3C96 S32DS 2018R1のライセンス更新をお願いしたく存じます。 再開まで今しばらくお待ちください。 よろしくお願いいたします。 Re: s32ds license Expired こんにちは、 お客様のS32DSライセンスが延長されました。
View full article
RW610 PowerTableUtil.exe ユーティリティが見つかりません こんにちは。UM12154ドキュメントで参照されているPowerTableUtil.exeファイルを探しています。サンプルとなる電力表のスプレッドシートは見つけることができましたが、ツール自体は見つかりませんでした。どうやらRW610の製品ページにあるはずなのに、そこで見つかりません。どんなご支援でも大変ありがたく思います 製品: WiFi RW6XX Re: RW610 PowerTableUtil.exe missing utility こんにちは、 @allenandrew1357 さん、 PowerTableUtil.exeツールは、デバイスの公式製品ページの「Secure ソフトウェア」セクションでご覧いただけます。 このセクションにアクセスできるかどうか、確認を手伝ってもらえますか? このページの指示に従ってSecure Access Rightsを通じてアクセスを申請する必要があります: Secure Access Rights | NXP Semiconductors BR ハビブ Re: RW610 PowerTableUtil.exe missing utility セキュリティセクションを見ましたが、まだアクセス権を申請していないようです。リクエストを送信しました。承認され次第、またご連絡します。ありがとう! Re: RW610 PowerTableUtil.exe missing utility 安全なアクセス権を申請しました。それらは今朝承認されました。それでも、Secure ソフトウェアのセクションをクリックしても、利用可能な結果は一切ありません。ページには「表示する情報がありません」と表示されます。これからどうすればいいか、何かアドバイスはありますか?そちらの環境にファイルは存在しますか? Re: RW610 PowerTableUtil.exe missing utility わかりましたが、私はすでにセキュアアクセス権を持っています。添付のスクリーンショットをご覧ください。RW610のページが「認可アクセス権のある製品リスト」セクションに含まれていることも確認しました。それでも、そこには安全なソフトウェアパッケージが0つも存在しているようです。心配なのは、PowerTableUtil.exeファイルがソフトウェアセクションで実際にリンクされていないことです。同じ問題を詳述した別のフォーラム投稿がありましたが、解決はNXP社員向けのプライベートDMで行われました。もう少し詳しく調べていただけませんか? Re: RW610 PowerTableUtil.exe missing utility こんにちは、@allenandrew1357 さん。 このプロセスの詳細については、セキュアアクセス権FAQをご参照ください。 BR ハビブ Re: RW610 PowerTableUtil.exe missing utility こんにちは、 @allenandrew1357 さん、 このファイルへのアクセスを得るために、代理店や営業担当者への連絡をご協力いただけますか? ご迷惑をおかけして申し訳ございません。 BR ハビブ
View full article
cst creates invalid RSA-PSS signatures with old OpenSSL When using cst with OpenSSL < 3.1 the salt used in the PSS signatures is too long. Processors like the i.MX91 expect the salt to have the same length as the digest, but before OpenSSL 3.1 the default was to make the salt as big as possible. But even with newer OpenSSL versions using the default is not good as it will accept shorter salts during signature verification. The fix is to call EVP_PKEY_CTX_set_rsa_pss_saltlen with second parameter set to RSA_PSS_SALTLEN_DIGEST. Note that OpenSSL 3.0 is EOL since last month. Security Re: cst creates invalid RSA-PSS signatures with old OpenSSL Hello, Thank you for reaching out, I will report this to the appropiate team. Best regards/Salduos, Aldo.
View full article
[lf_v2026.04]uboot-imx 补丁修复了 fsl_lpspi 中的多块传输问题 你好, 在 i.MX93 上使用 lf_v2025.04 的 uboot-imx 中,通过 fsl_lpspi 进行 SPI 传输 当一个完整的 FIFO 块紧跟在另一个块之后时,例如使用 SPI TPM 时,就会发生故障: => tpm2 获取功能 0x6 0x100 $loadaddr 20 lpspi_xfer_single:接收超时! 根本原因:spi_xfer_single() 在每个块之后都会额外排队一个 TCR。 该命令占用 TX FIFO 条目,并且只有在延迟一段时间后才会加载。 如果下一个数据块在此之前开始,则一个数据字将不被接受。 RX 循环会一直等待直到超时。 附件中的系列补丁修复了这个问题(也适用于 lf_v2026.04): 将 FSR TXCOUNT/RXCOUNT 掩码加宽 1/3(适用于 FIFO 更深的器件) 2/3 修复多块传输(实际修复) 3/3 将 spi_xfer_single() 改为静态函数(功能无变化) 在带有 SPI TPM 2.0(GPIO 片选)的 i.MX93 板上进行了测试。 注意:新代码最终仅在片选信号被复用为 GPIO 时才会取消置位。 但我们并未对此进行全面核实。 如果能将此功能纳入未来的 uboot-imx 版本中就太好了。 谢谢,并致以最诚挚的问候! 第谷·基希纳 -- emlix GmbH 总部:德国哥廷根市,柏林大街12号,邮编37073 电话:+49 (0)551 30664-0,电子邮件:[email protected] 哥廷根地方法院,登记号 HR B 3160 董事总经理:Heike Jordan、Uwe Kracke 博士 增值税号:DE 205 198 055 柏林办事处:Panoramastr.德国柏林 10178 号 1 号 波恩办事处:Bachstr.德国波恩 53115 6 号 慕尼黑办事处:Am Knie 16, 81241 München, 德国 http://www.emlix.com emlix——您的嵌入式Linux合作伙伴 PS: @xiaoningwang似乎是原作者,或许您可以看看。 Re: [lf_v2026.04] uboot-imx patches to fix multi-block transfers in fsl_lpspi 我已经把补丁发给开发团队了,谢谢你的建议。
View full article