Vinit Vyas

I can share my experience based in India. As per my understanding : 1: There will be high demand for VLSI background students or working professionals in the next five years due to semiconductor initiative. 2: Once Indian semiconductor companies start manufacturing devices . There will be increased demand for embedded system background folks in validation. 3: Semiconductor devices will be used for building embedded system platforms. It will increase the demand of hardware designers and firmware developers. 4: For each silicon device there will be multiple hardware platforms. For each hardware platform, there will be multiple embedded application software opportunities. So demand for embedded application software folks will always increase. 5: Will AI impact embedded software jobs ? I think it will help embedded software folks. Code generating tools exist for more than a decade and that didn't impact embedded software developers. Answer to Will firmware and embedded system hardware engineers be in higher demand than application software engineers? by Sanjay Kumar https://lnkd.in/gNbkCTCs #embedded #embedkari #opentowork

5

Can you combine FPGA cross-talk with cross-FPGA covert channels for a new security attack in cloud-based FPGAs? Find out from our paper! 🚨

94

SV DATA TYPES FAQs SV Data types is a very important topic mainly because a lot of new powerful data types were introduced in SV which were not present in Verilog. To give you a few examples, SV introduces new data types like: Structure Union Enum Typedef Mailbox Semaphore Queues Associative arrays Dynamic arrays & so on... Understanding SV data types is key to mastering SV. Why not dedicate one day learning about them? (NOTE: Solutions for these are already available in "System Verilog Tutorial Series" & System Verilog Essential series" @SwitiSpeaksOfficial. Please go through these series in detail) Link for SV data types FAQs in COMMENT #sv #systemverilog #verification #rtldesign #switispeaks #vlsi #semiconductors

13

1 Comment

How to start verifying a simple RTL design (Ex. Memory Controller) in UVM --> 1st Part: The very step is to understand the specification thoroughly with each of the below details: a. Working principle b. I/O ports and Signal level description c. Submodule details d. Is there any dependency of one signal with another, etc. e. Design detail understanding The next step is to build the Verification Plan and Feature / Testcase plan. List down the features that are needed to be verified. Some of them are below: ✓ Check whether clock and reset are working correctly. ✓ Check the correctness of the data by writing some value and read the same value of a single register. ✓ Repeat the process for all register. ✓ Check the default value of the register by reading the data. ✓ Check the controllers ability to handle multiple write and read transaction and Back to back writes. ✓ Check the correctness of WO register by reading from them. ✓ Check the correctness of RO register by writing on them. ✓ Check the Address translation and Address decoding issues being handled correctly. ✓ Check the invalid address access by RW from invalid or out of range addresses. ✓ Check the accessibility of locked condition of register. ✓ Check the controllers capability of accessing a single register from multiple address of different submap. ✓ Check the enablement and Quircky register accessibility. Above is a indicative list. Apart from the one mentioned above one also needs to add below cases: a. Stress testing. b. Use callbacks to inject errors. c. Multiple regression. d. Corner cases. The next step in order to verify the features verification planning is needed for which we need to code the following UVC's / objects along with Interface. a. Transaction b. Driver c. Monitor d. Sequencer e. Agent f. Sequence item g. Configuration h. Env i. Test #vlsi #asic #electronics #engineering

311

Domain Synchronisation in UVM: The UVM runtime phases has been classified as reset, configure, main and shutdown and each one of them serves different useful purpose. But one of the key benefits of using Runtime phases in UVM is Domain Synchronisation. Let's say we have two different environment in independent domain: agnt_1 and agnt_2. Now my first requirement is to bring two different domain in sync with each other so that the reset phase of domain 1 and the reset phase of domain 2 are in sync with each other and so on and for this we need uvm domain concept. Code snippet: class uvm_domain extends uvm_phase; uvm_domain dom_1; uvm_domain dom_2; agnt_1.set_domain(dom_1); agnt_2.set_domain(dom_2); dom_1.sync(dom_2); But, the most advantageous part of using uvm domain is that we can synchronise any particular phase of a domain with a different phase of another domain E.g: pre-reset phase of domain 1 can be synchronised with the pre-shutdown phase of domain 2 using the domain sync concept.That means we can migrate from any no of phases in a UVM domain. There are other different bunch of useful methods for uvm domain which are get_uvm_schedule which will be helpful in knowing when all the run time phases will be executed in UVM Domain. But the use of uvm domain needs to be handled with extreme care otherwise it can lead potential race condition. #vlsi #asic #electricalengineering #semiconductorindustry #linkedin

144

1 Comment

📢 Understanding Power, Performance, and Area (PPA) Analysis in VLSI: Key to Efficient Chip Design(Part-7) 🛤 Design Complexity and Iterations: As chip designs become more complex, the time and effort required for PPA optimization increase. The design iterations necessary to achieve optimal PPA metrics can be time-consuming and may delay the time-to-market for new chips and products. 🛤 Heterogeneous Integration: The integration of different types of components, such as CPUs, GPUs, and AI accelerators, on a single chip to create heterogeneous SoCs, introduces challenges in balancing power, performance, and area for each component. Coordinating and optimizing these diverse elements become a bottleneck in the overall PPA analysis. 🛤 Constraints and Trade-offs: Designers often face conflicting requirements and trade-offs when optimizing PPA metrics. For example, maximizing performance may increase power consumption or area. Achieving the desired balance requires careful analysis and sometimes necessitates compromise in one or more PPA aspects. PPA analysis are crucial aspects of VLSI chip design. Achieving an optimal balance between these factors ensures the creation of energy-efficient, high-performance, and cost-effective chips. As technology continues to evolve, PPA analysis will remain at the heart of innovative chip design, empowering us to build the next generation of electronic devices that power our modern world.

3

Broadcom's 10-1 Stock Split: What Investors Need to Know Broadcom, a leading semiconductor and infrastructure software company, recently announced a 10-1 stock split. This move has significant implications for investors. Let's break down what it means. What is a Stock Split? A stock split is when a company divides its existing shares into multiple new shares. In a 10-1 stock split, each share is split into ten. If you had one share worth $1,000, after the split, you'd have ten shares worth $100 each. The total value of your investment remains the same, but you have more shares at a lower price. Why Broadcom is Splitting Its Stock 1. Accessibility: Lower share prices make the stock more affordable to a broader range of investors. 2. Liquidity: More shares and lower prices can lead to increased trading volume and better liquidity. 3. Positive Perception: Stock splits can signal company confidence and attract more investor interest. 4. Employee Benefits: Easier for employees to participate in stock option plans. Impact on Investors 1. Affordability: More affordable shares can attract new investors. 2. Portfolio Flexibility: Investors will have more shares to manage, providing more flexibility. 3. Market Reaction: Splits often lead to short-term stock price boosts due to increased investor interest. 4. Dividend Adjustments: Dividends per share will be adjusted to reflect the split, keeping total payouts consistent. Broadcom's 10-1 stock split is aimed at making its stock more accessible and attractive, potentially boosting liquidity and market interest. For existing shareholders, this means more shares without changing the total value of their holdings, and for new investors, a lower price per share offers a more accessible entry point. While the split doesn't change the company's intrinsic value, it provides more trading flexibility and can positively affect market perception. Always consider your own investment goals and market conditions when making investment decisions. Making Complex Simple - Tom Hoff , Dominic Viadero, CPA

4

Interrupts which are asynchronous by nature are common in hardware designs and to handle those interrupts using UVM require some definite steps to deal with it. There are multiple ways to handle the interrupts in UVM. Among them the most popular are discussed below: [1] Using grab() and lock() inbuilt methods: ✓ lock() There are three sequences each consisting of two sequence_item and are feed through sequencer arbitration. The first two sequence items of each individual sequences are executed first and once the lock() sequence arrives, all the items within lock are executed and then again the rest of other items of other sequences. task body(); fork `uvm_do_with(seq_1,{pri == 1;}) `uvm_do_with(seq_2,{pri == 2;}) begin lock(); `uvm_do_with(seq_4,{pri == 3;}) unlock(); end Once the unlock call is executed, the control over the lock sequence is over resulting in execution of other sequence items of sequences. ✓ grab() grab() also work in the same way as lock() with only difference is that if there is any grab() call waiting at the sequencer arbitration it will bypass the arbitration queue resulting in executing the grab sequence first and then the other sequences. ✓ Conclusion: lock() method is used to process priority based interrupt while grab() method is used to execute non-ignored or non-maskable interrupt which are neither disabled nor ignored. [2] Using ISR: ✓ Steps: a. Create an interrupt object and an handler. b. Create the ISR c. Register the ISR with the interrupt object. d. Trigger the interrupt. [3] Using Event triggering: Judicious use of uvm_event and triggered methods are useful in handling interrupt mechanism in UVM. [4] Polling: In this case, the driver continuously checks the status of the interrupt to detect if an interrupt has occurred and it detects this with the help of loop. If interrupt is detected it initiates the ISR. [5] Arbitration: If multiple sequence items are appearing at Sequencer and it needs to handle all of them, then the sequencer uses inbuilt arbitration mechanism like arb FIFO, arb weighted, etc and appropriate priority can be assigned. #vlsi #asic #electronics #electricalengineering

121

3 Comments

A talk, “Compiler-assisted Kernel-based P4 Pipeline Offloading Using Intel IPU” was presented at the 2024 P4 Workshop by Debashis Chatterjee and Neha Singh from Intel Corporation. The video and slides from their talk can now be viewed on-demand. Overview: Introduction to P4-TC, a bridge for growing the network programmability ecosystem. The P4-TC datapath is defined using P4 and it utilizes a Linux kernel native P4 implementation. Benefits to this approach - some mundane developer knowledge is automated into the compiler and it utilizes common interfaces for either software or hardware datapaths. Learn more! View video on-demand | https://lnkd.in/g-hkXY28  View slides | https://lnkd.in/gVn-NnDN #P4 #P4Workshop #P4lang #OpenSource

20

1 Comment

𝙄𝙛 𝙮𝙤𝙪 𝙬𝙖𝙣𝙩 𝙩𝙤 𝙬𝙤𝙧𝙠 𝙞𝙣 𝘾𝙤𝙧𝙚 𝙀𝙢𝙗𝙚𝙙𝙙𝙚𝙙, 𝙞𝙩 𝙨𝙩𝙖𝙧𝙩𝙨 𝙬𝙞𝙩𝙝 𝘾 . 𝑪 𝒓𝒆𝒎𝒂𝒊𝒏𝒔 𝒂𝒍𝒐𝒏𝒆 𝒖𝒏𝒕𝒊𝒍 𝒊𝒕 𝒃𝒆𝒄𝒐𝒎𝒆𝒔 𝑪++.  There is no provision of Friendship in C but C++ has . Our one moth Firmware based core agenda will start with Interview Questions in Dec. Early joiners can take advantage of C preparation in advance. Here is another version of Agenda. December batch will get new topics like HAL, FreeRTOS etc in live sessions and recording of old topics such as Assembly basics, Thumb programming. #microcontroller #microcontrollers #embedded #embeddedsystem #embeddedlinux #embeddedc #freertos #firmware #firmwareengineer #devicedriver #softwareengineers #electronics #embeddedsystems #embeddedengineer #embedkari #opentowork #freshers

46

1 Comment

𝙋𝙤𝙨𝙞𝙩𝙞𝙫𝙚 𝙚𝙣𝙚𝙧𝙜𝙮 𝙠𝙣𝙤𝙬𝙨 𝙣𝙤 𝙗𝙤𝙪𝙣𝙙𝙖𝙧𝙞𝙚𝙨. 𝙄𝙛 𝙚𝙫𝙚𝙧𝙮𝙤𝙣𝙚 𝙬𝙚𝙧𝙚 𝙩𝙤 𝙨𝙥𝙧𝙚𝙖𝙙 𝙥𝙤𝙨𝙞𝙩𝙞𝙫𝙚 𝙚𝙣𝙚𝙧𝙜𝙮 𝙤𝙣 𝙩𝙝𝙚 𝙄𝙣𝙩𝙚𝙧𝙣𝙚𝙩, 𝙩𝙝𝙚 𝙬𝙤𝙧𝙡𝙙 𝙬𝙤𝙪𝙡𝙙 𝙗𝙚 𝙖 𝙢𝙪𝙘𝙝 𝙗𝙚𝙩𝙩𝙚𝙧 𝙥𝙡𝙖𝙘𝙚. 𝙇𝙪 𝙒𝙚𝙞 Just finished live session on Inline function for our morning 8AM batch. Most of the discussion was around functions in C. We have C++ in the evening slot. There two different Embedded software development slots also in the evening. Soon starting a series of workshops for Dec batch early joiners .. 𝑰𝒇 𝒔𝒐𝒎𝒆𝒐𝒏𝒆 𝒑𝒍𝒂𝒏𝒏𝒊𝒏𝒈 𝒕𝒐 𝒋𝒐𝒊𝒏 𝒕𝒉𝒆 𝒐𝒏𝒆 𝒎𝒐𝒏𝒕𝒉 𝒂𝒄𝒕𝒊𝒗𝒊𝒕𝒊𝒆𝒔 𝒘𝒊𝒕𝒉𝒊𝒏 𝒎𝒂𝒊𝒏(𝒇𝒊𝒓𝒎𝒘𝒂𝒓𝒆 𝒘𝒐𝒓𝒌𝒔𝒉𝒐𝒑) , Reboot and set your frequency for Dec activities. Do a startup now with Embedded C using 8bit MCU if never tried . #microcontroller #microcontrollers #embedded #embeddedsystem #embeddedlinux #embeddedc #freertos #firmware #firmwareengineer #devicedriver #softwareengineers #electronics #embeddedsystems #embeddedengineer #embedkari #opentowork #freshers

6

2 Comments

𝗦𝘆𝘀𝘁𝗲𝗺𝗩𝗲𝗿𝗶𝗹𝗼𝗴 𝗥𝗮𝗻𝗱𝗼𝗺 𝗦𝘁𝗮𝗯𝗶𝗹𝗶𝘁𝘆 - A concept every SoC/ASIC DV engineer should understand. In this deep dive you’ll learn: - 𝗛𝗶𝗲𝗿𝗮𝗿𝗰𝗵𝗶𝗰𝗮𝗹 𝗦𝗲𝗲𝗱𝗶𝗻𝗴: How threads and objects generate random numbers - 𝗢𝗯𝗷𝗲𝗰𝘁 & 𝗧𝗵𝗿𝗲𝗮𝗱 𝗦𝘁𝗮𝗯𝗶𝗹𝗶𝘁𝘆: How ordering of objects and threads in your code affect randomization - What you should keep in mind as you develop your testbench, so you can recreate regression failures - What does the simulator do with the +𝚗𝚝𝚋_𝚛𝚊𝚗𝚍𝚘𝚖_𝚜𝚎𝚎𝚍 and -𝚜𝚟_𝚜𝚎𝚎𝚍 command line arguments Read the article here: https://lnkd.in/gsX4XZpC ~~~ ✍️ Join 1500+ readers in my upcoming newsletter: https://lnkd.in/g945QtaS #verilog #systemverilog #asic

198

5 Comments

Locking/grabbing a sequencer in UVM ***************************************** Sometimes, locking or grabbing a sequencer in UVM can be confusing or misleading. The key idea here is that the locking/grabbing applies to the requests (or sequence items) generated by the sequence, rather than locking or grabbing the sequencer itself. This distinction means that other sequences can still be started on the same locked sequencer and won’t be blocked if they aren’t generating any requests. A lock request goes through the same arbitration process as any other request. A lock is granted only once all earlier requests are completed, and no other locks or grabs are currently blocking the sequence. A grab request, on the other hand, is prioritized at the front of the arbitration queue. It will be arbitrated before any other requests and granted as soon as no other grabs or locks are blocking the sequence. To clarify that more here is an example :

52

2 Comments

India's PCB manufacturing started growing at a steady state owing to the demand for embedded devices. Its critical to maintain superior quality PCB fabrication during mass production. The surface treatment process of PCB has a significant impact on the performance, reliability, and durability of the circuit board. Here are some common PCB surface treatment processes and their applicable scenarios: 1. Lead Free HASL: Applicable scenarios: Lead free tin spraying is an environmentally friendly surface treatment method suitable for most general applications. It provides good welding performance and is suitable for welding through hole components. 2. ENIG, Electroless Nickel Immersion Gold： Applicable scenarios: Electroplated gold is commonly used for high reliability, high performance, and high-frequency applications, such as communication equipment, aerospace, and medical devices. It provides a smooth, corrosion-resistant surface that helps improve welding performance. 3. Electroplated Tin: Applicable scenarios: Electroplated tin plating is an economical surface treatment method suitable for general applications, especially consumer electronics products. It provides good welding performance, but relatively less corrosion resistance than other processing methods. 4. Hot Air Solder Leveling : Applicable scenarios: HASL is a common surface treatment method that is suitable for general applications and has a low cost. However, it may not be suitable for small-sized components as it may leave uneven surfaces on them. 5. Electroplated Gold: Applicable scenarios: Sinking gold is a precious metal surface treatment method commonly used in high-end applications such as aerospace, medical instruments, and precision instruments. Sinking gold provides high conductivity and excellent corrosion resistance. 6.Organic Solderability Preservatives： Applicable scenario: OSP is an organic solder protectant suitable for SMT (Surface Mount Technology) components. It provides good welding performance in the short term, but relatively poor corrosion resistance, suitable for consumer electronic products used in the short term. 7. ENEPIG, Electroless Nickel Electroless Palladium Immersion Gold： Applicable scenarios: ENEPIG is a high-performance surface treatment method suitable for high-density, high-frequency, and high reliability applications, such as advanced electronic and communication devices. #pcb #embedded #electronics

6

A pretty comprehensive intro to GPIOs in a system recommended browsing for all SoC architects, System designers, analog engineers and anyone interested in electronics in general GPIOs form the backbone of most chips. Small microcontrollers practically have few other digital ports, purely analog chips still have one or two GPIO And large SoCs still have a bunch (as many as a hundred), just for test and debug and future proofing, also to add some unknown functions on board at a future date, even though large SoCs do nearly their entire data payload transfer through MIPI, PCIe or other specialised IOs

55

3 Comments

I’m thrilled to announce that my paper, “Averting Hacks of PCIe Transport using CMA/SPDM,” has been published! 🚀 In this paper, I dive into how secure connections are built to protect data, using symmetric and asymmetric flows. I also discuss how next-gen ICVIP systems are evolving to verify SPDM software for tighter security 🔐. We break down how Diffie-Hellman Key Exchange (DHE) 🔑 generates secure keys and how the Digital Signature Algorithm (DSA) ✍️ is used to authenticate data through digital signatures. The real game-changer? Elliptical curve cryptography 🧠! It's not only faster but also way more efficient than traditional methods for establishing secure connections. 🚀 I’d love for you to check it out here and share any valuable feedback or insights! #PCIe #SIEMENS #Cybersecurity #Encryption #CMA #SPDM #EllipticalCurveCryptography #DHE #DSA #ICVIP #NextGenTech #SecureConnections #TechInnovation

52

8 Comments

User space in Kernel: 📔 What is User Space in Linux? User space refers to the part of the operating system where user applications run. It is separate from the kernel space, which is reserved for the kernel and core OS operations. User space applications interact with the hardware and kernel through system calls and APIs. 🔑 Key Features of User Space 1️⃣ Isolation: Each user-space application has its own virtual address space, preventing it from directly accessing the memory or resources of other processes or the kernel. 2️⃣ Limited Privileges: User space has restricted access to hardware and kernel operations to ensure security and stability. 3️⃣ System Call Interface: User space communicates with the kernel using system calls like read, write, open, ioctl, etc. ✅ Key Components in User Space 1️⃣ User Applications: Programs like editors (e.g., vim), web browsers, and custom scripts run in user space. 2️⃣ System Libraries: Libraries like glibc provide APIs for user applications to interact with the kernel. 3️⃣ System Calls: The interface that bridges user space and kernel space. Examples: open(), read(), write(), mmap(). 🤝 Interaction Between User Space and Kernel Space 1️⃣ System Calls: User applications invoke kernel services through system calls. Example: write(fd, buf, count) makes a system call to transfer data to a file descriptor. 2️⃣ File Descriptors: Used by user applications to interact with files, devices, and sockets, which are managed by the kernel. 3️⃣ Memory Mapping: Using mmap, user space can map kernel memory (e.g., device memory) into its virtual address space. 4️⃣ Device Interaction: Procfs and sysfs expose kernel information to user space via files in /proc and /sys. Device drivers provide interfaces through /dev files, which can be accessed by user applications. 🧠 Communication Mechanisms in User Space 1️⃣ Inter-Process Communication (IPC): Mechanisms like pipes, sockets, message queues, shared memory, and signals facilitate communication between user-space processes. 2️⃣ Procfs and Sysfs: User space can read/write to /proc and /sys files to get or configure kernel information. 3️⃣ IOCTL: Provides a way to send control commands to device drivers from user space. 👉 Memory Management in User Space 1️⃣ Virtual Address Space: User space applications use virtual addresses, which the kernel maps to physical memory. 2️⃣ Dynamic Memory: Functions like malloc allocate memory in user space. Memory can also be explicitly managed using mmap. 3️⃣ Segmentation: User space virtual memory is divided into regions such as text, data, heap, and stack. #userspace #systemdesign #embedded #linux #learning

203

2 Comments

If a student had learned C , Microprocessor and digital electronics as a part of engineering, S/he can learn developing firmware development within one month. The average student may take 3 months. Students from other similar branches(but without Microprocessor and digital electronics) may take six months . You have mentioned Firmware , so you need to learn digital electronics, Microcontroller, Embedded C and Assembly. I provide live sessions in the following phases for Firmware development: Phase 1 : Learn C from scratch. It will be based on normal desktop C Phase 2: Learn Embedded C with 8 bit MCU 8051 Phase 3: Learn ARM baremetal drivers using STM32. These three phases need minimum two months. Those who prefer embedded software career , we do additional phases Phase 4: Linux Application development Phase 5: Embedded C++ We provide online material for basic electronics, RTOS and Operating system as well. If you are really interested to get exposure in practical Embedded software Experience. The first responsibility goes to you only. Embedded is a vast field and so learning also depends on your current level of Embedded system. Here is a list of checklist you should look first: 1: A 64-bit Windows 10 system with good harddisk (1TB) and RAM(8G) should be sufficient for your journey till Embedded Linux. If you have lower configuration , You can start with that also. 2: Hardware Knowledge - You should have educational background of analog & digital electronics. These days most of the board's use USB power or mobile charger equivalent power supply. It is not mandatory to learn power supply design to start Embedded software journey. Similarly you can get MCU evaluation boards starting from 300 Rs . 3: Make sure you learn C at Desktop level first without any dependancy on board. 4: Start with father of MCU AKA 8051 based product to clear Embedded C concept. If you have budget buy Nuvoton MS51 which has in-built programming option. You just need to connect with USB cable Install Keil IDE and clear Embedded C concepts with simulator and/or board. 5: Buy Arduino Uno R3 or latest version along with few other interfaces like LCD, seven segment display etc. Install Arduino IDE to play with that and later Atmel studio to work at C level. 6: Buy STM32 based Nucleo-64 board based in your budget and start learning ARM from board & ARM website documents. 7: Once you have solid background of baremetal Embedded programming. You can learn freeRTOS with STM32 board. 8: If you want to learn IoT , Then buy ESP32 board which will cover both bluetooth & WiFi. Another simpler option for only WiFi can be Node MCU. 9: At advance level , You can opt for Beaglebone black for computing or Raspberry PI for IoT/AI. All above boards have very good community support. If you don't have this kind of budget , Buy at least STM32 Nucleo64 board to understand Embedded C with ARM Baremetal and RTOS https://lnkd.in/gWcrcWXJ #embedded #embedkari #opentowork

16