P. Bernardi, Gabriele Filipponi, T. Foscale, Giorgio Insinga
{"title":"Low cost external serial interface watchdog for SoCs and FPGAs automatic characterization tests","authors":"P. Bernardi, Gabriele Filipponi, T. Foscale, Giorgio Insinga","doi":"10.1109/LATS58125.2023.10154486","DOIUrl":"https://doi.org/10.1109/LATS58125.2023.10154486","url":null,"abstract":"Manufacturers must characterize their design deeply when designing and producing devices like FPGAs and SoCs. Information collected through simulation and physical experiments is the primary data source for manufacturers that can then decide the optimal working ranges of multiple critical parameters such as operating voltage, frequency, temperatures, etc. With complex devices such as SoCs, and FPGAs with integrated PLLs and voltage regulators, each combination of voltage and frequency can be checked by communicating the desired parameters to the DUT, running a functional test, and observing the results. However once the ATE sends the desired parameters to the DUT through SPI or other serial interfaces, the DUT may freeze and stop to accept new commands entirely. This is particularly problematic for targeted characterization that may include a minimal number of boards and DUTs and where the ATE may simply be a simple laptop without any automatic DUT reset capabilities. This paper presents an external serial communication watchdog designed using an ESP32-based board. Our watchdog can detect the communications coming from the ATE, monitor the answers from the DUT, and restart it through power cycling in case of freezing.","PeriodicalId":145157,"journal":{"name":"2023 IEEE 24th Latin American Test Symposium (LATS)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133909291","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Analyzing Side-Channel Attack Vulnerabilities at RTL","authors":"X. Lai, M. Jenihhin","doi":"10.1109/LATS58125.2023.10154497","DOIUrl":"https://doi.org/10.1109/LATS58125.2023.10154497","url":null,"abstract":"How much the timing side-channel attacks' vulner-ability can be mitigated at the early RTL design phase? This paper summarizes an Invited Talk presenting basic approaches for analyzing timing side-channel attack vulnerability in a generic information flow (illustrated on an RSA algorithm implementation) and in a common security architecture. The latter assumes, as an example, a Physical Unclonable Function (PUF) based chip-level authentication structure with potentially vulnerable ECC-powered Fuzzy Extractors.","PeriodicalId":145157,"journal":{"name":"2023 IEEE 24th Latin American Test Symposium (LATS)","volume":"141 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115882070","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Organizing and Sponsorship","authors":"","doi":"10.1109/lats58125.2023.10154499","DOIUrl":"https://doi.org/10.1109/lats58125.2023.10154499","url":null,"abstract":"","PeriodicalId":145157,"journal":{"name":"2023 IEEE 24th Latin American Test Symposium (LATS)","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124445494","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
T. Copetti, A. Castelnuovo, T. Gemmeke, L. Bolzani
{"title":"Evaluating a New RRAM Manufacturing Test Strategy","authors":"T. Copetti, A. Castelnuovo, T. Gemmeke, L. Bolzani","doi":"10.1109/LATS58125.2023.10154503","DOIUrl":"https://doi.org/10.1109/LATS58125.2023.10154503","url":null,"abstract":"Memristive devices represent a promising candidate to complement the CMOS technology due to their ability to overcome CMOS device technology limits as well as to implement not only memory but also computing elements. These characteristics make the implementation of emerging applications with strict constraints in terms of performance, area and power consumption possible. However, the adoption of these novel devices depends on being able to guarantee their quality after manufacturing. High-volume manufacturing tests pose significant challenges in terms of fault detection capability, test time as well as implementation overheads. In this context, this paper proposes a new implementation Design-for-Testability (DfT) strategy to test Resistive Random Access Memories (RRAMs). A case study based on a RRAM was implemented using the 28nm TSMC technology library and validated through electrical simulations. The new DfT strategy was evaluated with respect to its ability to detect unique and conventional faults. Finally, the paper presents a discussion about the introduced overheads and compares the proposed approach with state-of-the-art related to RRAM manufacturing test strategies.","PeriodicalId":145157,"journal":{"name":"2023 IEEE 24th Latin American Test Symposium (LATS)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114378850","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Salvatore Pappalardo, A. Ruospo, Ian O’Connor, B. Deveautour, Ernesto Sánchez, A. Bosio
{"title":"A Fault Injection Framework for AI Hardware Accelerators","authors":"Salvatore Pappalardo, A. Ruospo, Ian O’Connor, B. Deveautour, Ernesto Sánchez, A. Bosio","doi":"10.1109/LATS58125.2023.10154505","DOIUrl":"https://doi.org/10.1109/LATS58125.2023.10154505","url":null,"abstract":"Deep Neural Networks (DNNs) have proven to give very good results for many complex tasks and applications, such as object recognition in images/videos and natural language processing. Some relevant applications of DNNs are defined by real-time safety-critical systems, which typically require the adoption of DNN accelerators that are usually implemented as systolic arrays. Assessing their reliability is not trivial and may depend on several factors such as the size of the array and the data precision. In this paper, we present a cross-layer framework for systolic array DNN accelerators described at RTL level allowing to inject faults at channel granularity for convolutional layers. The basic idea is to simulate the execution of the Channel Under Test (ChUT) at RTL level. Faulty outputs collected from the RTL simulation are then used at software level to complete the execution of the DNN and thus determine the impact of the injected faults at application level. Interestingly, the software execution is more than 100 times faster than the corresponding hardware simulation.","PeriodicalId":145157,"journal":{"name":"2023 IEEE 24th Latin American Test Symposium (LATS)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117182786","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pooja Choudhary, Lava Bhargava, M. Fujita, Virendra Singh
{"title":"LUT-based Arithmetic Circuit Approximation with Formal Guarantee on Worst Case Relative Error","authors":"Pooja Choudhary, Lava Bhargava, M. Fujita, Virendra Singh","doi":"10.1109/LATS58125.2023.10154494","DOIUrl":"https://doi.org/10.1109/LATS58125.2023.10154494","url":null,"abstract":"We are presenting an automatic approach to produce approximate circuit with formal error guarantees on worst-case relative error (WCRE). The key concept is based on LUTs, SAT -based error evaluation, and property-checking techniques. These approximated circuits are employed to improve scalability and automate the designs for arithmetic circuits. The proposed 8 bit approximate multiplier shows an 83.33 % and 25.3 % decrease in power consumption and delay as w.r.t. exact multiplier. We demonstrated that the use of an approximate multiplier in FIR filter degrades SNR by 1.2 dB.","PeriodicalId":145157,"journal":{"name":"2023 IEEE 24th Latin American Test Symposium (LATS)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122728418","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Holistic IJTAG-based External and Internal Fault Monitoring in UAVs","authors":"Foisal Ahmed, M. Jenihhin","doi":"10.1109/LATS58125.2023.10154489","DOIUrl":"https://doi.org/10.1109/LATS58125.2023.10154489","url":null,"abstract":"Cyber-Physical Systems (CPSs), such as Unmanned Aerial Vehicles (UAVs), use System-on-Chip (SoC) based computing platforms to perform multiple complex tasks in safety-critical applications that require a highly dependable operation. Due to continuous technological manufacturing miniaturization SoCs face a wide spectrum of chip-level reliability issues such as aging, soft and hard errors during the operational lifetime of a UAV. In addition, external (off-chip) faults in the sensors, actuators, and motors are another cause of UAV failures. While existing works examine either on-chip faults (internal) or sensors/actuators faults (external) separately, this research proposes a UAV health monitoring infrastructure considering both external and internal faults holistically. The proposed method relies on the IEEE 1687 standard (IJTAG) and employs on-chip embedded instruments as health monitors to instantly access external and internal sensor data. Experimental results for functional simulation of a real-life case-study design demonstrate both types of fault detection by serving only three clock cycles and the localization process using 16 and 30 clock cycles for the case of single and double faults, respectively.","PeriodicalId":145157,"journal":{"name":"2023 IEEE 24th Latin American Test Symposium (LATS)","volume":"248 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114550553","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}