16/32-Bit Architecture XC2300A Derivatives 16/32-Bit Single-Chip Microcontroller with 32-Bit Performance XC2000 Family / Base Line Errata Sheet V1.6 2013-06 Microcontrollers Edition 2013-06 Published by Infineon Technologies AG 81726 Munich, Germany © 2013 Infineon Technologies AG All Rights Reserved. Legal Disclaimer The information given in this document shall in no event be regarded as a guarantee of conditions or characteristics. With respect to any examples or hints given herein, any typical values stated herein and/or any information regarding the application of the device, Infineon Technologies hereby disclaims any and all warranties and liabilities of any kind, including without limitation, warranties of non-infringement of intellectual property rights of any third party. Information For further information on technology, delivery terms and conditions and prices, please contact the nearest Infineon Technologies Office (www.infineon.com). Warnings Due to technical requirements, components may contain dangerous substances. For information on the types in question, please contact the nearest Infineon Technologies Office. Infineon Technologies components may be used in life-support devices or systems only with the express written approval of Infineon Technologies, if a failure of such components can reasonably be expected to cause the failure of that life-support device or system or to affect the safety or effectiveness of that device or system. Life support devices or systems are intended to be implanted in the human body or to support and/or maintain and sustain and/or protect human life. If they fail, it is reasonable to assume that the health of the user or other persons may be endangered. 16/32-Bit Architecture XC2300A Derivatives 16/32-Bit Single-Chip Microcontroller with 32-Bit Performance XC2000 Family / Base Line Errata Sheet V1.6 2013-06 Microcontrollers XC2300A Derivatives XC2000 Family / Base Line Table of Contents 1 History List / Change Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 2 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 3 Current Documentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 4 4.1 4.2 4.3 4.4 Errata Device Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Functional Deviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Deviations from Electrical and Timing Specification . . . . . . . . . . . . . . . . . 11 Application Hints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Documentation Updates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 5 5.1 5.2 5.3 5.4 5.5 Short Errata Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Errata Removed in this errata sheet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Functional Deviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Deviations from Electrical and Timing Specification . . . . . . . . . . . . . . . . . Application Hints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Documentation Updates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 15 16 18 19 21 6 6.1 Detailed Errata Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Functional Deviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . BROM_TC.006 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . BSL_CAN_X.001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DPRAM_X.002 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . EBC_X.007 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ECC_X.002 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ESR_X.001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ESR_X.004 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . GPT12E_X.002 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . OCDS_X.003 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PAD_X.001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PARITY_X.001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RESET_X.003 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RESET_X.004 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SCU_X.012 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . StartUp_X.003 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . USIC_AI.004 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . USIC_AI.005 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . USIC_AI.016 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WDT_X.002 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Deviations from Electrical and Timing Specification . . . . . . . . . . . . . . FLASH_X.P001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SWD_X.P002 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 22 22 22 23 27 28 29 29 30 32 32 37 37 38 38 39 40 40 41 41 44 44 44 6.2 Errata Sheet 4 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line 6.3 6.4 Application Hints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ADC_AI.H002 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CAPCOM12_X.H001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CC6_X.H001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ECC_X.H001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . GPT12_AI.H001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . GPT12E_X.H002 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . INT_X.H002 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . INT_X.H004 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . LXBUS_X.H001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MultiCAN_AI.H005 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MultiCAN_AI.H006 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MultiCAN_AI.H007 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MultiCAN_AI.H008 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MultiCAN_TC.H002 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MultiCAN_TC.H003 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MultiCAN_TC.H004 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . OCDS_X.H003 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PVC_X.H001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RESET_X.H003 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RTC_X.H003 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SCU_X.H009 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SWD_X.H001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . USIC_AI.H001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . USIC_AI.H002 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . USIC_AI.H003 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Documentation Updates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . EBC_X.D001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ECC_X.D002 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RESET_X.D001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . USIC_X.D001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . USIC_X.D002 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Errata Sheet 5 45 45 45 47 47 47 48 49 50 50 51 51 51 52 52 53 53 54 54 56 56 56 57 57 57 58 59 59 59 60 60 60 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line History List / Change Summary 1 History List / Change Summary Table 1 History List Version Date Remark1) 1.0 26.06.2008 new Errata Sheet 1.1 05.09.2008 1.2 30.01.2009 Documentation Updates are added into the Errata Sheet 1.3 12.10.2009 Errata No. 01542AERRA 1.4 23.09.2010 Errata No. 01869AERRA 1.5 22.02.2013 Errata No. 02398AERRA 1.6 18.06.2013 Errata No. 02627AERRA. Removed EES-AA, ES-AA Marking/Step references. Added ES-AB, AB Marking/Step references. DPRAM_X.002 is fixed in AB step. 1) Errata changes to the previous Errata Sheet are marked in Chapter 5 ”Short Errata Description”. Trademarks C166TM, TriCoreTM and DAVETM are trademarks of Infineon Technologies AG. We Listen to Your Comments Is there any information in this document that you feel is wrong, unclear or missing? Your feedback will help us to continuously improve the quality of this document. Please send your proposal (including a reference to this document) to: [email protected] Errata Sheet 6 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line General 2 General This Errata Sheet describes the deviations of the XC2300A Derivatives from the current user documentation. Each erratum identifier follows the pattern Module_Arch.TypeNumber: • • • • Module: subsystem, peripheral, or function affected by the erratum Arch: microcontroller architecture where the erratum was initially detected. – AI: Architecture Independent – TC: TriCore – X: XC166 / XE166 / XC2000 Family Type: category of deviation – [none]: Functional Deviation – P: Parametric Deviation – H: Application Hint – D: Documentation Update Number: ascending sequential number within the three previous fields. As this sequence is used over several derivatives, including already solved deviations, gaps inside this enumeration can occur. This Errata Sheet applies to all temperature and frequency versions and to all memory size variants of this device, unless explicitly noted otherwise. Note: This device is equipped with a C166S V2 Core. Some of the errata have workarounds which are possibly supported by the tool vendors. Some corresponding compiler switches need possibly to be set. Please see the respective documentation of your compiler. For effects of issues related to the on-chip debug system, see also the documentation of the debug tool vendor. Some errata of this Errata Sheet do not refer to all of the XC2300A Derivatives, please look to the overview: Table 2 for Functional Deviations Table 3 for Deviations from Electrical and Timing Specification Table 4 for Application Hints Table 5 for Documentation Updates Errata Sheet 7 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Current Documentation 3 Current Documentation The Infineon XC2000 Family comprises device types from the XC2200 group, the XC2300 group and the XC2700 group. The XC23xxA device types belong to the XC2300 group. Device XC23xxA Marking/Step AA, ES-AB, AB Package PG-LQFP-64, PG-LQFP-100, PG-LQFP-144 This Errata Sheet refers to the following documentation: • • • • • XC2300A Derivatives User’s Manual XC233xA Data Sheet XC236xA Data Sheet XC238xA Data Sheet Documentation Addendum (if applicable) Make sure you always use the corresponding documentation for this device available in category 'Documents' at www.infineon.com/xc2300. The specific test conditions for EES and ES are documented in a separate Status Sheet. Note: Devices marked with EES or ES are engineering samples which may not be completely tested in all functional and electrical characteristics, therefore they should be used for evaluation only. Errata Sheet 8 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Errata Device Overview 4 Errata Device Overview This chapter gives an overview of the dependencies of individual errata to devices and steps. An X in the column of the sales codes shows that this erratum is valid. 4.1 Functional Deviations Table 2 shows the dependencies of functional deviations in the derivatives. Errata Device Overview: Functional Deviations XC23xxA Table 2 AA ES-AB AB1) Functional Deviation BROM_TC.006 X X BSL_CAN_X.001 X X DPRAM_X.002 X EBC_X.007 X X ECC_X.002 X X ESR_X.001 X X ESR_X.004 X X GPT12E_X.002 X X OCDS_X.003 X X PAD_X.001 X X PARITY_X.001 X X RESET_X.003 X X RESET_X.004 X X SCU_X.012 X X StartUp_X.003 X X Errata Sheet 9 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Errata Device Overview Errata Device Overview: Functional Deviations (cont’d) XC23xxA Table 2 AA ES-AB AB1) Functional Deviation USIC_AI.004 X X USIC_AI.005 X X USIC_AI.016 X X WDT_X.002 X X 1) From AA step to AB step, 1 erratum has been fixed. Errata Sheet 10 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Errata Device Overview 4.2 Deviations from Electrical and Timing Specification Table 3 shows the dependencies of deviations from the electrical and timing specification in the derivatives. XC23xxA Errata Device Overview: Deviations from Electrical and Timing Specification AA AC/DC/ADC Deviation ES-AB, AB Table 3 FLASH_X.P001 X X SWD_X.P002 X X Errata Sheet 11 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Errata Device Overview 4.3 Application Hints Table 4 shows the dependencies of application hints in the derivatives. XC23xxA Errata Device Overview: Application Hints AA Hint ES-AB, AB Table 4 ADC_AI.H002 X X CAPCOM12_X.H001 X X CC6_X.H001 X X ECC_X.H001 X X GPT12_AI.H001 X X GPT12E_X.H002 X X INT_X.H002 X X INT_X.H004 X X LXBUS_X.H001 X X MultiCAN_AI.H005 X X MultiCAN_AI.H006 X X MultiCAN_AI.H007 X X MultiCAN_AI.H008 X X MultiCAN_TC.H002 X X MultiCAN_TC.H003 X X MultiCAN_TC.H004 X X OCDS_X.H003 X X PVC_X.H001 X X RESET_X.H003 X X RTC_X.H003 X X Errata Sheet 12 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Errata Device Overview XC23xxA Errata Device Overview: Application Hints (cont’d) AA Hint ES-AB, AB Table 4 SCU_X.H009 X X SWD_X.H001 X X USIC_AI.H001 X X USIC_AI.H002 X X USIC_AI.H003 X X Errata Sheet 13 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Errata Device Overview 4.4 Documentation Updates Table 5 shows the dependencies oft documentation updates in the derivatives. XC23xxA Errata Device Overview: Documentation Updates AA Documentation Updates ES-AB, AB Table 5 EBC_X.D001 X X ECC_X.D002 X X RESET_X.D001 X X USIC_X.D001 X X USIC_X.D002 X X Errata Sheet 14 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Short Errata Description 5 Short Errata Description This chapter gives an overview on the deviations and application hints. Changes to the last Errata Sheet are shown in the column “Chg”. 5.1 Errata Removed in this errata sheet Table 6 shows a short description of the errata removed from this version if errata shete. Table 6 Errata removed in this Errata Sheet Errata Short Description ECC_X.001 SBRAM ECC Disabled in FSM Applicable only to EES-AA, ES-AA Marking/Steps. EES-AA, ES-AA reference removed from this ES HARR_X.001 Not Requested HARR after User Reset Applicable only to EES-AA, ES-AA Marking/Steps. EES-AA, ES-AA reference removed from this ES Startup_X.001 RAM Initialization before ECC/Parity Activation Applicable only to EES-AA, ES-AA Marking/Steps. EES-AA, ES-AA reference removed from this ES Errata Sheet Chg 15 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Short Errata Description 5.2 Functional Deviations Table 7 shows a short description of the functional deviations. Table 7 Functional Deviations Functional Deviation Short Description BROM_TC.006 Baud Rate Detection for CAN Bootstrap Loader 22 BSL_CAN_X.001 Quartz Crystal Settling Time after PORST too Long for CAN Bootstrap Loader 22 DPRAM_X.002 Undefined Data read from Dual-Port RAM (DPRAM) 23 EBC_X.007 Bus Arbitration not Properly Working 27 ECC_X.002 Incorrect ECC Error Indication for DPRAM 28 ESR_X.001 ESREXSTAT1 and ESREXSTAT2 Status Bits are Cleared by Hardware 29 ESR_X.004 Wrong Value of SCU_RSTCONx Registers after ESRy Application Reset 29 GPT12E_X.002 Effects of GPT Module Microarchitecture 30 OCDS_X.003 Peripheral Debug Mode Settings cleared by Reset 32 PAD_X.001 Additional Edges in the Input Signal 32 PARITY_X.001 PMTSR Register Initialization 37 RESET_X.003 P2.[2:0] and P10.[12:0] Switch to Input 37 RESET_X.004 Sticky “Register Access Trap” forces device into power-save mode after reset. 38 SCU_X.012 Wake-Up Timer RUNCON Command 38 StartUp_X.003 Debug Interface Configuration from Flash can Fail Upon Power-On 39 USIC_AI.004 Receive shifter baudrate limitation 40 USIC_AI.005 Only 7 data bits are generated in IIC mode when TBUF is loaded in SDA hold time 40 Errata Sheet 16 Chg Pg V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Short Errata Description Table 7 Functional Deviations (cont’d) Functional Deviation Short Description USIC_AI.016 Transmit parameters are updated during FIFO buffer bypass 41 WDT_X.002 Clearing the Internal Flag which Stores Preceding WDT Reset Request 41 Errata Sheet 17 Chg Pg V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Short Errata Description 5.3 Deviations from Electrical and Timing Specification Table 8 shows a short description of the electrical- and timing deviations from the specification. Table 8 Deviations from Electrical and Timing Specification AC/DC/ADC Deviation Short Description FLASH_X.P001 Test Condition for Flash parameter NER in Data Sheets 44 SWD_X.P002 Supply Watchdog (SWD) Supervision Level in Data Sheet. 44 Errata Sheet 18 Chg Pg V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Short Errata Description 5.4 Application Hints Table 9 shows a short description of the application hints. Table 9 Application Hints Hint ADC_AI.H002 Short Description Minimizing Power Consumption of an ADC Module Chg Pg 45 CAPCOM12_X.H001 Enabling or Disabling Single Event Operation 45 CC6_X.H001 Modifications of Bit MODEN in Register CCU6x_KSCFG 47 ECC_X.H001 ECC Error Indication Permanently Set 47 GPT12_AI.H001 Modification of Block Prescalers BPS1 and BPS2 47 GPT12E_X.H002 Reading of Concatenated Timers 48 INT_X.H002 Increased Latency for Hardware Traps 49 INT_X.H004 SCU Interrupts Enabled After Reset 50 LXBUS_X.H001 Do Not Access Reserved Locations on the LXBus 50 MultiCAN_AI.H005 TxD Pulse upon short disable request 51 MultiCAN_AI.H006 Time stamp influenced by resynchronization 51 MultiCAN_AI.H007 Alert Interrupt Behavior in case of BusOff 51 MultiCAN_AI.H008 Effect of CANDIS on SUSACK 51 MultiCAN_TC.H002 Double Synchronization of receive input 52 MultiCAN_TC.H003 Message may be discarded before transmission in STT mode 53 MultiCAN_TC.H004 Double remote request 53 OCDS_X.H003 Debug Interface Configuration by User Software 54 PVC_X.H001 PVC Threshold Level 2 54 Errata Sheet 19 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Short Errata Description Table 9 Application Hints (cont’d) Hint Short Description RESET_X.H003 How to Trigger a PORST after an Internal Failure 56 SCU_X.H009 WUCR.TTSTAT can be set after a PowerUp 56 SWD_X.H001 Application Influence on the SWD 57 USIC_AI.H001 FIFO RAM Parity Error Handling 57 USIC_AI.H002 Configuration of USIC Port Pins 57 USIC_AI.H003 PSR.RXIDLE Cleared by Software 58 Errata Sheet 20 Chg Pg V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Short Errata Description 5.5 Documentation Updates Table 10 gives a short description of the documentation updates. Table 10 Documentation Updates Documentation Updates Short Description EBC_X.D001 Visibility of Internal LXBus Cycles on External Address Bus 59 ECC_X.D002 Initialization of the Read-Control Logic 59 RESET_X.D001 Reset Types of Trap Registers 60 USIC_X.D001 Connection from CCU6x to USICx 60 USIC_X.D002 USIC1 Channel 0 Connection DX0C and DX0D 60 Errata Sheet 21 Chg Pg V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description 6 Detailed Errata Description This chapter provides a detailed description for each erratum. If applicable a workaround is suggested. 6.1 Functional Deviations BROM_TC.006 Baud Rate Detection for CAN Bootstrap Loader In a specific corner case, the baud rate detected during reception of the initialization frame for the CAN bootstrap loader may be incorrect. The probability for this sporadic problem is relatively low, and it decreases with decreasing CAN baud rate and increasing module clock frequency. Workaround: If communication fails, the host should repeat the CAN bootstrap loader initialization procedure after a reset of the device. BSL_CAN_X.001 Quartz Crystal Settling Time after PORST too Long for CAN Bootstrap Loader The startup configuration of the CAN bootstrap loader when called immediately after PORST limits the settling time of the external oscillation to 0.5 ms. For typical quartz crystal this settling time is too short. The CAN bootstrap loader generates a time-out and goes into Startup Error State. Workaround • • For low performance CAN applications a ceramic resonator with settling time less than 0.5 ms can be used. An alternative is the Internal Start from on-chip Flash memory as startup mode after PORST. Then switch the system clock to external source and trigger a software reset with CAN bootstrap loader mode selected. Now the Errata Sheet 22 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description device starts with a CAN bootstrap loader without limitation of the oscillator settling time. DPRAM_X.002 Undefined Data read from Dual-Port RAM (DPRAM) Under special conditions, undefined data may be read from up to two word locations L1, L2 in the DPRAM (address range F600H .. FDFFH). The problem (affecting L1 and/or L2) is due to a redundancy circuit that may not be properly disabled during the internal power-up reset sequence. The problem may only occur when an operand that was written via port B is (later) read via port A of the DPRAM. This is only the case when dual read transfers are performed on both port A and B in parallel, which can only happen in the two scenarios described below. Note: Field experience has shown that the probability for this problem is relatively low, as it further depends on several other conditions, e.g.: - For an initial power-on reset without residual voltage, the problem is less likely to occur at room temperature and above. - The problem is more likely to occur when a residual voltage at VDDI1 was present at the last power-off/on cycle. - If the problem does occur, then it is “latched” at power up. If it hasn’t occurred in the current power cycle, then the problem only has the possibility of occurring at the next power-off/on cycle. - The addresses of L1 and L2 may vary from device to device and from one power-off/on cycle to the next, but will not change for a given device until the next power-off/on cycle; i.e. if L1 and L2 are not read via port A in one of the critical scenarios described below, the problem has no effect on the application. Scenario 1 (Context Switch) In this scenario, the problem may occur under the following sequence of conditions (all conditions must be met): 1. L1 and/or L2 are located in the DPRAM in an area that is used to store the context of a global register bank, and Errata Sheet 23 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description 2. An explicit software write operation (i.e. no context switch) to L1 and/or L2 is performed, and 3. The Context Pointer (register CP) is modified (e.g. by MOV/POP/SCXT CP, ... ) in one of the following instructions, leading to a context store (old context) and a context load (new context) of the global bank into/from DPRAM, and 4. One of the General Purpose Registers Rx with an odd register number x (x = 1, 3, .., 15) is loaded from locations L1 or L2. In this case, the contents of Rx may be incorrect. This scenario may typically occur at the beginning of interrupt routines when a GPR (e.g. R15 used as user stack pointer) is saved into the new GPR context in DPRAM, as shown in the following Figure 1. Errata Sheet 24 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description 1 1 MOV 0F804h, R15 … ; n≥ 0 other instructions 2 SCXT CP,#0F7E6h Store R15 via Port B 0F804 Load R15 via Port A New context 2.b Global register bank R15 R14 Load new context 0F7E6 R1 R0 Old context 2.a Store old context Register File DPRAM Figure 1 Context Switch with Pointer Passing in R15 Scenario 2 (CoXXX Instructions) In this scenario, the problem may occur when one of the CoXXX instructions CoABS, CoADD*1), CoCMP, CoLOAD*, CoMAC*, CoMAX, CoMIN, CoMUL*, CoSUB* reads L1 or L2 via the addressing mode [IDXi*2)], [RWm*]. Note: All other CoXXX instructions (CoASHR, CoMOV, CoNEG, CoNOP, CoRND, CoSHL/R, CoSTORE) and addressing modes are not affected. 1) * stands for all variants of the respective CoXXX instruction 2) * stands for all variants of the [IDXi*], [RWm*] addressing mode Errata Sheet 25 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description Analysis of Development Tools/Summary of Supplier Feedback • TASKING VX-toolset for C166 (latest version v3.1r1)1) – Scenario 1 (Context Switch): In the most basic form (without __registerbank(...) function qualifier), the compiler uses push/pop instructions to save/restore registers in the interrupt frame, i.e. the critical scenario does not occur. – When the __registerbank(...) function qualifier is used to define a global register bank, the user stack-pointer (R15) is copied and the context-pointer is set to the new register bank. This is rated as critical only if R15 is used to access the user stack within the interrupt routine. Exception: In the routine for the reset vector, the user stack-pointer is not copied (makes no sense for the reset vector). This is rated as uncritical. – Scenario 2 (CoXXX Instructions): The compiler currently does not use the [IDXi*] addressing-mode. In setjmp.src from the C-library and macmull.src from the runtime-library CoSTORE instructions are used. This is rated as uncritical. • “Classic” TASKING C166/ST10 toolset (latest version v8.9r1) – Scenario 1 (Context Switch): In all cases where the user stack-pointer is copied, register R0 (even register number) is used. This is rated as uncritical. – Scenario 2 (CoXXX Instructions): The compiler currently does not use the [IDXi*] addressing-mode. In the runtime library modules: cp*w.asm and udil.asm / umol.asm, CoXXX instructions may be used, but none of them belongs to the category of critical instructions in the scope of this problem. • Keil C166 Development Tools – Scenario 1 (Context Switch): Without the using attribute, the critical scenario will not occur. With the using attribute, in all cases where the user stack-pointer is copied, register R0 (even register number) is used. This is rated as uncritical. 1) Analysis also applicable to derived free toolsets VX-toolset Lite Edition, resp. XE166 toolset Errata Sheet 26 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description – Scenario 2 (CoXXX Instructions): The compiler (including the intrinsics) currently does not use the [IDXi*] addressing-mode. Workaround (Scenario 1) • • • • On assembler level, do not perform a context switch of the global register bank in case locations in the DPRAM corresponding to the location of a register Rx with an odd register number are written by software (see example for critical scenario in Figure 1). With the compiler of the TASKING VX-toolset, do not use the __registerbank(...) function qualifier to define a global register bank. With the compiler of the “Classic” TASKING C166/ST10 toolset, no workaround for this scenario is required (see analysis above). With the compiler of the Keil C166 toolset, no workaround for this scenario is required (see analysis above). Workaround (Scenario 2) • • On assembler level, do not use the CoXXX instructions CoABS, CoADD*, CoCMP, CoLOAD*, CoMAC*, CoMAX, CoMIN, CoMUL*, CoSUB* with the addressing mode [IDXi*], [RWm*]. Instead, use e.g. the other addressing modes of these instructions. For compilers of the Keil and TASKING toolsets, no workaround is required (see analysis above). Usage of 3rd Party Software When using operating systems and/or other 3rd party software or libraries, please check with your supplier. EBC_X.007 Bus Arbitration not Properly Working Due to a mismatch of pad propagation delays and internal operation cycle time the arbitration feature of the External Bus Controller (EBC) can only be used with severe restrictions. It is recommended not to use this feature, as future members of the XC2000/XE166 Family will no longer support bus arbitration. Errata Sheet 27 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description Workaround The usable conditions also depend on the application system and can only be defined for a specific use case. ECC_X.002 Incorrect ECC Error Indication for DPRAM Under certain conditions, the ECC error flag for the dual port memory (DPRAM) may indicate an error when none exists. Conditions under which ECC error is incorrectly indicated: • • • ECC memory protection has been selected for DPRAM (SCU_MCHKCON.SELDP = 0). The ECC check is enabled for DPRAM (SCU_ECCCON.DPEN = 1). A concurrent read and write access is made to the same byte or word in the DPRAM (possible due to instruction pipeline). Under the above conditions, an ECC error may be indicated for DPRAM (flag SCU_ECCSTAT.DP = 1), although there is no error in DPRAM. It should be noted that despite the incorrect ECC error indication, the data are delivered error free, and the ECC logic still functions correctly (correcting data single bit errors, if present). There is no data corruption in this case. The ECC bits that were written are generated correctly, as well as check and correction is not affected for that was read. This problem is limited to the DPRAM. All other SRAM memories cannot perform concurrent read and write accesses, and therefore cannot have this issue. Workaround Use parity protection for DPRAM (SCU_MCHKCON.SELDP = 1). Single-bit errors will be correctly indicated by parity logic, but will not be corrected. Errata Sheet 28 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description ESR_X.001 ESREXSTAT1 and ESREXSTAT2 Status Bits are Cleared by Hardware The status bits of register ESREXSTAT1 and ESREXSTAT2 are set as the trigger signal was active. The status bits are not stored and are only set as long as the external trigger inputs are active. Workaround None. ESR_X.004 Wrong Value of SCU_RSTCONx Registers after ESRy Application Reset SCU_RSTCONx registers are reset only by Power-On, but they may be wrongly affected after a second application reset requested by an ESRy pin. This may lead to the SCU_RSTCONx register values being set to zero, which could unexpectedly disable reset sources within the user application. The conditions which lead to this behavior are: 1. First, an application reset by SW (software), CPU (Central Processing Unit), MP (Memory), WDT (Watchdog Timer) or ESRy (External Service Request y) occurs. 2. Following this, an application reset on an ESRy pin occurs. 3. If the above mentioned ESRy reset occurs during a critical time window of the SSW (startup software), then it’s possible that the application will operate with the wrong SCU_RSTCONx register value. The critical time window occurs when the SSW is writing the SCU_RSTCONx registers, and at the same time, the ESRy reset request is processed by the reset circuitry. The width of this critical window fcritical window is less than 13 cycles. Errata Sheet 29 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description Application Reset t critical window Reset by ESRy pin Start of SSW SSW Write RSTCON Start of SSW Application Application Runs Software End of SSW Application Runs ESR_X.004 Fig. 1 Figure 2 Critical application reset sequence Workaround • • Initialize SCU_RSTCONx registers by user software after any reset, or assure that a second application reset request with an ESR pin does not occur during the critical time window. GPT12E_X.002 Effects of GPT Module Microarchitecture The present GPT module implementation provides some enhanced features (e.g. block prescalers BPS1, BPS2) while still maintaining timing and functional compatibility with the original implementation in the C166 Family of microcontrollers. Both the GPT1 and GPT2 blocks use a finite state machine to control the actions within each block. Since multiple interactions are possible between the timers (T2 .. T6) and register CAPREL, these elements are processed sequentially within each block in different states. However, all actions are normally completed within one basic clock cycle. The GPT2 state machine has 4 states (2 states when BPS2 = 01B) and processes T6 before T5. The GPT1 state machine has 8 states (4 states when BPS1 = 01B) and processes the timers in the order T3 - T2 (all actions except capture) - T4 - T2 (capture). In the following, two effects of the internal module microarchitecture that may require special consideration in an application are described in more detail. Errata Sheet 30 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description 1.) Reading T3 by Software with T2/T4 in Reload Mode When T2 or T4 are used to reload T3 on overflow/underflow, and T3 is read by software on the fly, the following unexpected values may be read from T3: • • when T3 is counting up, 0000H or 0001H may be read from T3 directly after an overflow, although the reload value in T2/T4 is higher (0001H may be read in particular if BPS1 = 01B and T3I = 000B), when T3 is counting down, FFFFH or FFFEH may be read from T3 directly after an underflow, although the reload value in T2/T4 is lower (FFFEH may be read in particular if BPS1 = 01B and T3I = 000B). Note: All timings derived from T3 in this configuration (e.g. distance between interrupt requests, PWM waveform on T3OUT, etc.) are accurate except for the specific case described under 2.) below. Workaround: • • When T3 counts up, and value_x < reload value is read from T3, value_x should be replaced with the reload value for further calculations. When T3 counts down, and value_x > reload value is read from T3, value_x should be replaced with the reload value for further calculations. Alternatively, if the intention is to identify the overflow/underflow of T3, the T3 interrupt request may be used. 2.) Reload of T3 from T2 with setting BPS1 = 01B and T3I = 000B When T2 is used to reload T3 in the configuration with BPS1 = 01B and T3I = 000B (i.e. fastest configuration/highest resolution of T3), the reload of T3 is performed with a delay of one basic clock cycle. Workaround 1: To compensate the delay and achieve correct timing, • • increment the reload value in T2 by 1 when T3 is configured to count up, decrement the reload value in T2 by 1 when T3 is configured to count down. Errata Sheet 31 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description Workaround 2: Alternatively, use T4 instead of T2 as reload register for T3. In this configuration the reload of T3 is not delayed, i.e. the effect described above does not occur with T4. OCDS_X.003 Peripheral Debug Mode Settings cleared by Reset The behavior (run/stop) of the peripheral modules in debug mode is defined in bitfield SUMCFG in the KSCCFG registers. The intended behavior is, that after an application reset has occurred during a debug session, a peripheral reenters the mode defined for debug mode. For some peripherals, the debug mode setting in SUMCFG is erroneously set to normal mode upon any reset (instead upon a debug reset only). It remains in this state until SUMCFG is written by software or the debug system. Some peripherals will not re-enter the state defined for debug mode after an application reset: GPT12, CAPCOM2, and MultiCAN will resume normal operation like after reset, i.e. they are inactive until they are initialized by software. In case the RTC has been running before entry into debug mode, and it was configured in SUMCFG to stop in debug mode, it will resume operation as before entry into debug mode instead. All other peripheral modules, i.e. ADC, CCU6 and USIC, will correctly re-enter the state defined for debug mode after an application reset in debug mode. For Flash and CPU, bitfield SUMCFG must be configured to normal mode anyway, since they are required for debugging. Workaround None. PAD_X.001 Additional Edges in the Input Signal The digital input- and I/O-pins are designed using Schmitt trigger input structures with hysteresis. Even with this structure, it is possible that very slow Errata Sheet 32 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description rising edges may generate spikes, resulting in unexpected additional edges at the input signal. The next picture Figure 3 is an example for a slow input signal, with spikes shown on the slow rising input signal. tr. 0.9 VDDP 0.7 VDDP Input Signal 0.3 VDDP 0.1 VDDP Internal Signal Unexpected Edges PAD_X.001 Fig. 1 Figure 3 Example for a Slow Input Signal The first rising edge in Figure 3 of the internal signal is always valid. The edges which are marked with “Unexpected Edges” must be ignored. Measurements have shown that a spike can be generated under the following conditions. Table 11 Conditions for Additional Edges in the Input Signal Parameter Symbol VDDP Junction Temperature TJ System frequency fsys Rising Slope tr Digital supply voltage Errata Sheet Typ. Value Unit Note 4.5 to 5.5 V full range °C all MHz >1 µs 33 Upper Voltage Range V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description The reaction to this spike generation strongly depends on the application (hardware, software, internal and external noise). Although it is not possible to define how the application will react in all cases, it is possible to categorize how applications are typically affected, as shown below. Applications which can be affected by a spike: • • • CCU6, CAPCOM2 and GPT inputs Port inputs Interrupts input Applications which should not be affected by a spike due to faster rising slope tr which is necessary for the application, due the interface protocol or due multiple sampling of the hardware: • • • USIC CAN Others Workaround for Input Capture Conditions 1. Workaround for all Affected Applications Use rising edges with faster rising slope tr than defined in Table 11. 2. Workaround for CCU6, CAPCOM2 and GPT Inputs No generic solution is available for these applications. Add or switch to a software solution. For example, the software could check whether the measured signal values are in the expected range. 3. Workaround for Port Inputs 1. The captured time interval value should be checked whether it is in a reasonable range. Errata Sheet 34 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description 2. The input pin should be read several times, and a majority decision may be made to decide whether the edge was correct or erratic. Only if 1) and 2) indicate that the edge was reliable, the captured value should be used for further calculations. Otherwise, a substituted (extrapolated) value might be used. 4. Workaround for Interrupt Inputs 4.1 Falling Edge Detection Approach 1. Measure the time interval since the last interrupt (shown as in tinterval in Figure 4 below) and check that it is in the expected time range. In the example, an erratic edge would cause the measured time intervalt erratic interval to be approximately 50% of the expected value. 2. The state of the input pin that caused the interrupt could be read several times in the interrupt service routine and a majority decision made to check if the input pin really is at a low level to determine whether this is a genuine falling edge interrupt or whether the interrupt was triggered by a spike generated by a slow rising edge. 0.7 VDDP Input Signal 0.3 VDDP Internal Signal Unexpected Edges Unexpected Edges terratic interval tinterval PAD_X.001 Fig. 2 Figure 4 Falling Edge Detection Approach Only if 1) and/or 2) indicate that the edge was reliable, should the rest of the interrupt service routine be executed. Errata Sheet 35 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description 4.2 Rising Edge Detection Approach In case of rising edge detection multiple interrupts would be generated when the spike occurs. The time interval since the last interrupt can be measured – if it is very small compared to the expected value, this would indicate a spike and the interrupt should be ignored. 0.7 VDDP Input Signal 0.3 VDDP Internal Signal terratic interval tinterval Figure 5 PAD_X.001 Fig. 3 Rising Edge Detection Approach If the time between the rising signal edge and the rising edge caused by a spike terratic is less than the ISR service time, a workaround would be to clear the interrupt request (IR) flag before the return from interrupt is done. The clearing of the IR flag will avoid a further erratic interrupt. The preferred solution for interrupt handling is to use the rising edge detection. 4.3 Rising Edge and Falling Edge Detection Approach The rising edge detection workaround works also if both edges are used as trigger for interrupt and the following conditions are valid: • • terratic interval << tinterval r and terratic interval << tinterval f Errata Sheet 36 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description 0.7 VDDP Input Signal 0.3 VDDP Internal Signal Figure 6 terratic interval tinterval r tinterval f PAD_X.001 Fig. 4 Rising Edge and Falling Edge Detection Approach PARITY_X.001 PMTSR Register Initialization The PMTSR register content after start-up is 0100H, meaning the parity logic for SBRAM is not in standard mode of operation. Workaround If parity will be used as Memory Control mechanism for SBRAM, it must be enabled by initializing the PMTSR register with 8000H. RESET_X.003 P2.[2:0] and P10.[12:0] Switch to Input During the execution of an Application Reset and Debug Reset the pins P2.[2:0] and P10.[12:0] are intermediately switched to input. These pins return to their previous mode approximately 35 system clock cycles after the application reset counter has expired (approx. 0.6 µs with default reset delay at 80 MHz). If such a pin is used as output, make sure that this short interruption does not lead to critical system conditions. Workaround External pull devices can be added to have a defined level on these pins during Application and Debug Reset. Errata Sheet 37 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description RESET_X.004 Sticky “Register Access Trap” forces device into powersave mode after reset. The system control unit (SCU) provides trap generation, to respond to certain system level events or faults. Certain trap sources maintain sticky trap flags which are only cleared explicitly by software, or by a power-on reset. These sticky trap flags are contained in the SCU register DMPMIT. In case the “Register Access Trap” flag (DMPMIT.RAT) becomes set, but is not cleared before a debug, internal application, or application reset occurs, then the microcontroller will reset, but will fail to start-up correctly. The microcontroller start-up software will detect that the sticky trap flag is set, and will force the device into power-save mode with DMP_1 shut down and DMP_M powered. Workaround In response to the trap event, software must explicitly clear the sticky trap flag using the SCU register DMPMITCLR, before executing a debug, internal application, or application reset. Note that this workaround does not address unexpected debug, internal application, or application resets which occur between the sticky trap event and the clearing of the sticky flags by software. To keep this exposure period as short as possible, it is recommended to clear the flag early in the trap routine. Note: Register DMPMITCLR is protected by the register security mechanism after execution of the EINIT instruction and must be unlocked before accessing. SCU_X.012 Wake-Up Timer RUNCON Command The Wake-Up Timer can be started and stopped by the WUCR.RUNCON bit field. Under the precondition that the Wake-Up Timer is configured to stop when reaching zero (WUCR.ASP=1B) and if two Wake-Up Timer commands are executed successively (e.g. “start” is directly followed by “stop”) then the second command will be ignored and will not change the state of the Wake-Up Timer. Errata Sheet 38 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description Workaround After executing the first command wait at least 4 Wake-Up Timer cycles (fWUT) before writing again to the WUCR.RUNCON bit field and requesting the second command. StartUp_X.003 Debug Interface Configuration from Flash can Fail Upon Power-On This erratum only affects devices with a Date Code before DC1242 (i.e. digit value < 1242). XC2000/XE166 devices allow the user to select between a number of debug interface options including type (JTAG/DAP) and pin assignment. The primary selection is done by configuration pins upon power-on, where one of the supported options is to install the debug interface according to the value taken from dedicated locations in user Flash (C001F0H..C001F3H). This option is selected by configuration pin values (HWCFG) xxxxx111B (code start from internal Flash) or x1100000B (code start from external memory). The other configurations directly selecting a debug mode work correctly. The start-up procedure reads the dedicated locations in Flash too early - before Flash redundancy is installed - which can lead to an unrecoverable read error and terminate the boot process if the block from C001F0H to C001FFH is programmed by the user. A limited number of devices are affected – a rough estimation is below 1% from the production - and the (mis) behavior is constant. That means any device is either always error free or always failing if no programming of the block from C001F0H to C001FFH is done after the last power-on. Note, that only the two mentioned modes upon power-on and only the read from dedicated locations during start-up are affected but not in general Flash and debug interface functionality. Workaround 1. Do not program the page from C00180H to C001FFH. This provides an erased, error free flash read during start-up (without installed flash Errata Sheet 39 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description redundancy) of DBGPRR register value which allows start-up from internal/external memory and JTAG position A. 2. If these start-up configurations are used during development, a device that does not start-up in the desired debug configuration should be replaced by another device. 3. Alternatively, select a debug interface not from Flash data but directly using configuration pins - refer to the User’s Manual. With this it is not possible to start from external memory, nor is JTAG position A available. USIC_AI.004 Receive shifter baudrate limitation If the frame length of SCTRH.FLE does not match the frame length of the master, then the baudrate of the SSC slave receiver is limited to fsys/2 instead of fsys. Workaround None. USIC_AI.005 Only 7 data bits are generated in IIC mode when TBUF is loaded in SDA hold time When the delay time counter is used to delay the data line SDA (HDEL > 0), and the empty transmit buffer TBUF was loaded between the end of the acknowledge bit and the expiration of programmed delay time HDEL, only 7 data bits are transmitted. With setting HDEL=0 the delay time will be tHDEL = 4 x 1/fSYS + delay (approximately 60ns @ 80MHz). Workaround • • Do not use the delay time counter, i.e use only HDEL=0 (default), or write TBUF before the end of the last transmission (end of the acknowledge bit) is reached. Errata Sheet 40 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description USIC_AI.016 Transmit parameters are updated during FIFO buffer bypass Transmit Control Information (TCI) can be transferred from the bypass structure to the USIC channel when a bypass data is loaded into TBUF. Depending on the setting of TCSR register bit fields, different transmit parameters are updated by TCI: • • • • • When SELMD = 1, PCR.CTR[20:16] is updated by BYPCR.SELO (applicable only in SSC mode) When WLEMD = 1, SCTR.WLE and TCSR.EOF are updated by BYPCR.BWLE When FLEMD = 1, SCTR.FLE[4:0] is updated by BYPCR.BWLE When HPCMD = 1, SCTR.HPCDIR and SCTR.DSM are updated by BHPC When all of the xxMD bits are 0, no transmit parameters will be updated However in the current device, independent of the xxMD bits setting, the following are always updated by the TCI generated by the bypass structure, when TBUF is loaded with a bypass data: • • • WLE, HPCDIR and DSM bits in SCTR register EOF and SOF bits in TCSR register PCR.CTR[20:16] (applicable only in SSC mode) Workaround The application must take into consideration the above behaviour when using FIFO buffer bypass. WDT_X.002 Clearing the Internal Flag which Stores Preceding WDT Reset Request The information that the WDT has already been exceeded once is stored in an internal flag. In contrary to the documentation, that this flag can be cleared by writing a 1B to bit WDTCS.CLRIRF at any time, clearing of the internal flag is only possible, when the WDT is in Prewarning Mode. Errata Sheet 41 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description Workaround 1 Applications following the proposal of Application Note AP16103 (section `Using ESR pins to trigger a PORST reset`) to trigger a Power-on Reset upon a WDT event will find the internal flag cleared upon the Power-on Reset and thus will have no issue with this limitation. Workaround 2 In case the WDT triggers a User Reset upon a WDT overflow, the internal flag will not be cleared by the reset itself. Any further overflow of the WDT will lead to a permanent reset of the device. Applications which intentionally let the WDT exceed once, e.g. in conjunction with an initial self test, might want to have the internal flag cleared to prevent a permanent reset upon a real WDT overflow. If the internal flag shall be cleared by software, this must be done as a reaction on a WDT overflow in the time frame the WDT is in Prewarning Mode before the permanent User Reset will be triggered. The CPU is notified upon the WDT entering Prewarning Mode by issuing an interrupt request. The application can react on this request and clear the internal flag now by writing a 1B to bit WDTCS.CLRIRF e.g. within an ISR. Workaround 3 Some applications may not want to use or rely on the interrupt logic in conjunction with a WDT overflow event. The proposed remedy in this case is, to initiate a Power Reset to clear the internal flag by changing the settings of the active Supply Watchdog (SWD) as follows: 1. Disable SFR protection. 2. Write the inverted value of bit LxALEV to register SWDCON0, where x stands for the number of the comparator which currently would trigger a Power Reset. In doing so, a Power Reset for VDDI_1 and VDDI_M will be activated clearing the internal flag. The application may store information on preceding WDT events in the Standby-SRAM. This can be done any time after the WDT reset without timing limitations or the need to use the interrupt logic. Errata Sheet 42 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description Note: Although the supply for the DPRAM, DSRAM and PSRAM will be switched off during the active reset phase, it depends on the external buffer capacitance at the VDDI_1 pins, the actual system clock frequency and the environmental conditions, whether the content of these RAMs will be preserved in this case or not. However, the Standby-RAM itself is not cleared upon this reset. Errata Sheet 43 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description 6.2 Deviations from Electrical and Timing Specification FLASH_X.P001 Test Condition for Flash parameter NER in Data Sheets The Flash endurance parameter NER `Number of erase cycles` for 15000 cycles is documented with a wrong Test Condition. The Test Condition states today: `tRET≥ 5 years; Valid for up to 64 user selected sectors (data storage)`. In fact the amount of Flash memory validated for this cycling rate is more limited and the Test Condition must therefore state the following: • tRET≥ 5 years; Valid for Flash module 3 (up to 64 kbytes) Note: The related use case for this parameter is data storage with high cycling rate in general and EEPROM emulation in particular. For these applications concurrent operation of data storage to and program execution from Flash is assumed. Refer also to parameter NPP. SWD_X.P002 Supply Watchdog (SWD) Supervision Level in Data Sheet. The Supply Watchdog (SWD) Supervision Level VSWD tolerance boundaries for 5.5 V are changed from ± 0.15 V to ± 0.30 V. Errata Sheet 44 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description 6.3 Application Hints ADC_AI.H002 Minimizing Power Consumption of an ADC Module For a given number of A/D conversions during a defined period of time, the total energy (power over time) required by the ADC analog part during these conversions via supply VDDPA is approximately proportional to the converter active time. Recommendation for Minimum Power Consumption: In order to minimize the contribution of A/D conversions to the total power consumption, it is recommended 1. to select the internal operating frequency of the analog part (fADCI) near the maximum value specified in the Data Sheet, and 2. to switch the ADC to a power saving state (via ANON) while no conversions are performed. Note that a certain wake-up time is required before the next set of conversions when the power saving state is left. Note: The selected internal operating frequency of the analog part that determines the conversion time will also influence the sample time tS. The sample time tS can individually be adapted for the analog input channels via bit field STC. CAPCOM12_X.H001 Enabling or Disabling Single Event Operation The single event operation mode of the CAPCOM1/2 unit eliminates the need for software to react after the first compare match when only one event is required within a certain time frame. The enable bit SEEy for a channel CCy is cleared by hardware after the compare event, thus disabling further events for this channel. One Channel in Single Event Operation As the Single Event Enable registers CC1_SEE, CC2_SEE are not located in the bit-addressable SFR address range, they can only be modified by instructions Errata Sheet 45 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description operating on data type WORD. This is no problem when only one channel of a CAPCOM unit is used in single event mode. Two or more Channels in Single Event Operation When two or more channels of a CAPCOM unit are independently operating in single event mode, usually an OR instruction is used to enable one or more compare events in register CCn_SEE, while an AND instruction may be used to disable events before they have occurred. In these cases, the timing relation of the channels must be considered, otherwise the following typical problem may occur: • • • • In the Memory stage, software reads register CCn_SEE with bit SEEy = 1B (event for channel CCy has not yet occurred) Meanwhile, event for CCy occurs, and bit SEEy is cleared to 0B by hardware In the Write-Back stage, software writes CCn_SEE with bit SEEx = 1B (intended event for CCx enabled via OR instruction) and bit SEEy = 1B or, as inverse procedure, software writes CCn_SEE with bit SEEx = 0B (intended event for CCx disabled via AND instruction) and bit SEEy = 1B In these cases, another unintended event for channel CCy is enabled. To avoid this effect, one of the following solutions - depending on the characteristics of the application - is recommended to enable or disable further compare events for CAPCOM channels concurrently operating in single event mode: • • • • Modify register CCn_SEE only when it is ensured that no compare event in single event mode can occur, i.e. when CCn_SEE = 0x0000, or Modify register CCn_SEE only when it is ensured that there is a sufficient time distance to the events of all channels operating in single event mode, such that none of the bits in CCn_SEE can change in the meantime, or Use single event operation for one channel only (i.e. only one bit SEMx may be = 1B), and/or Use one of the standard compare modes, and emulate single event operation for a channel CCs by disabling further compare events in bit field MODs (in register CCn_Mz) in the corresponding interrupt service routine. Writing to register CCn_Mz is uncritical, as this register is not modified by hardware. Errata Sheet 46 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description CC6_X.H001 Modifications of Bit MODEN in Register CCU6x_KSCFG For each module, setting bit MODEN = 0 immediately switches off the module clock. Care must be taken that the module clock is only switched off when the module is in a defined state (e.g. stop mode) in order to avoid undesired effects in an application. In addition, for a CCU6 module in particular, if bit MODEN is changed to 0 while the internal functional blocks have not reached their defined stop conditions, and later MODEN is set to 1 and the mode is not set to run mode, this leads to a lock situation where the module clock is not switched on again. ECC_X.H001 ECC Error Indication Permanently Set The ECC error flag of the ECCSTAT register for the DPRAM, DSRAM, PSRAM and SBRAM can not be cleared, if a memory location with an ECC error is selected and the ECC is enabled. The memory can be selected by an active or by the latest read or write access. Workaround Select a memory location without ECC error in the respective memory (e.g. make a read to another address) and then clear the ECC error flag. Be aware that the new selected address may also have an ECC error. GPT12_AI.H001 Modification of Block Prescalers BPS1 and BPS2 The block prescalers BPS1 and BPS2, controlled via bit fields T3CON.BSP1 and T6CON.BPS2, determine the basic clock for the GPT1 and GPT2 block, respectively. After reset, when initializing a block prescaler BPSx to a value different from its default value (00B), it must be initialized first before any mode involving external trigger signals is configured for the associated GPTx block. These modes include counter, incremental interface, capture, and reload mode. Otherwise, unintended count/capture/reload events may occur. Errata Sheet 47 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description In case a block prescaler BPSx needs to be modified during operation of the GPTx block, disable related interrupts before modification of BPSx, and afterwards clear the corresponding service request flags and re-initialize those registers (T2, T3, T4 in block GPT1, and T5, T6, CAPREL in block GPT2) that might be affected by an unintended count/capture/reload event. GPT12E_X.H002 Reading of Concatenated Timers For measuring longer time periods, a core timer (T3 or T6) may be concatenated with an auxiliary timer (T2/T4 or T5) of the same timer block. In this case, the core timer contains the low part, and the auxiliary timer contains the high part of the extended timer value. When reading the low and high parts of concatenated timers, care must be taken to obtain consistent values in particular after a timer overflow/underflow (e.g. one part may already have considered an overflow, while the other has not). This is a general issue when reading multi-word results with consecutive instructions, and not necessarily unique to the GPT module microarchitecture. The following algorithm may be used to read concatenated GPT timers, represented by Timer_high (for auxiliary timer, here: T2) and Timer_low (for core timer, here: T3). In this example, the high part is read twice, and reading of the low part is repeated if two different values were read for the high part. • • • • read Timer_high_temp = T2 read Timer_low = T3 wait two basic clock cycles (to allow increment/decrement of auxiliary timer in case of core timer overflow/underflow) - see Table 12 below read Timer_high = T2 – if Timer_high is not equal to Timer_high_temp: read Timer_low = T3 After execution of this algorithm, Timer_high and Timer_low represent a consistent time stamp of the concatenated timers. The equivalent number of system clock cycles corresponding to two basic clock cycles is shown in the following Table 12: Errata Sheet 48 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description Table 12 Equivalent Number of System Clock Cycles Required to Wait for Two Basic Clock Cycles Setting of BPS1 BPS1 = 01 BPS1 = 00 BPS1 = 11 BPS1 = 10 Required Number of System Clocks 8 Setting of BPS2 BPS2 = 01 BPS2 = 00 BPS2 = 11 BPS2 = 10 Required Number of System Clocks 4 16 8 32 16 64 32 In case the required timer resolution can be achieved with different combinations of the Block Prescaler BPS1/BPS2 and the Individual Prescalers TxI, the variant with the smallest value for the Block Prescaler may be chosen to minimize the waiting time. E.g. in order to run T6 at fSYS/512, select BPS2 = 00B, T6I = 111B, and insert 8 NOPs (or other instructions) to ensure the required waiting time before reading Timer_high the second time. INT_X.H002 Increased Latency for Hardware Traps When a condition for a HW trap occurs (i.e. one of the bits in register TFR is set to 1B), the next valid instruction that reaches the Memory stage is replaced with the corresponding TRAP instruction. In some special situations described in the following, a valid instruction may not immediately be available at the Memory stage, resulting in an increased delay in the reaction to the trap request: 1. When the CPU is in break mode, e.g. single-stepping over such instructions as SBRK or BSET TFR.x (where x = one of the trap flags in register TFR) will have no (immediate) effect until the next instruction enters the Memory stage of the pipeline (i.e. until a further single-step is performed). 2. When the pipeline is running empty due to (mispredicted) branches and a relatively slow program memory (with many wait states), servicing of the trap is delayed by the time for the next access to this program memory, even if vector table and trap handler are located in a faster memory. However, the situation when the pipeline/prefetcher are completely empty is quite rare due to the advanced prefetch mechanism of the C166S V2 core. Errata Sheet 49 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description INT_X.H004 SCU Interrupts Enabled After Reset Following a reset, the SCU interrupts are enabled by default (register SCU_INTDIS = 0000H). This may lead to interrupt requests being triggered in the SCU immediately, even before user software has begun to execute. In the SCU, multiple interrupt sources are `ORed` to a common interrupt node of the CPU interrupt controller. Due to the “ORing” of multiple interrupt sources, only one interrupt request to the interrupt controller will be generated if multiple sources at the input of this OR gate are active at the same time. If user software enables an interrupt in the interrupt controller (SCU_xIC) which shares the same node as the SCU interrupt request active after reset, it may lead to the effect of suppressing the intended interrupt source. So, for all SCU interrupt sources which will not be used, make sure to disable the interrupt source (SCU_INTDIS) and clear any pending request flags (SCU_xIC.IR) before enabling interrupts in interrupt controller. LXBUS_X.H001 Do Not Access Reserved Locations on the LXBus Some of the on-chip peripherals are connected via the LXBus. The EBC controls this access by using CS7 to define the LXBus area. The memory map lists several sections occupied by the on-chip LXBus peripherals MultiCAN and the USIC modules. The reserved sections within the address range 20’0000H ... 20’FFFFH shown in the memory map are designated to additional peripherals for future derivatives. The sizes of the reserved sections depend on the chosen device type. These reserved sections must not be accessed by user software nor by debuggers. Access to these sections may lead to a CPU lock situation caused by a bus lock and also makes the software incompatible with other derivatives. The error mode can only be left by a reset. Errata Sheet 50 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description MultiCAN_AI.H005 TxD Pulse upon short disable request If a CAN disable request is set and then canceled in a very short time (one bit time or less) then a dominant transmit pulse may be generated by MultiCAN module, even if the CAN bus is in the idle state. Example for setup of the CAN disable request: MCAN_KSCCFG.MODEN = 0 and then MCAN_KSCCFG.MODEN = 1 Workaround Set all INIT bits to 1 before requesting module disable. MultiCAN_AI.H006 Time stamp influenced by resynchronization The time stamp measurement feature is not based on an absolute time measurement, but on actual CAN bit times which are subject to the CAN resynchronization during CAN bus operation.The time stamp value merely indicates the number of elapsed actual bit times. Those actual bit times can be shorter or longer than nominal bit time length due to the CAN resynchronization events. Workaround None. MultiCAN_AI.H007 Alert Interrupt Behavior in case of Bus-Off The MultiCAN module shows the following behavior in case of a bus-off status: TEC=0x60 or REC=0x60 EWRN Figure 7 Errata Sheet REC=0x1, TEC=0x1 BOFF INIT REC=0x60, TEC=0x1 EWRN+BOFF INIT REC=0x0, TEC=0x0 ALERT INIT Alert Interrupt Behavior in case of Bus-Off 51 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description When the threshold for error warning (EWRN) is reached (default value of Error Warning Level EWRN = 0x60), then the EWRN interrupt is issued. The bus-off (BOFF) status is reached if TEC > 255 according to CAN specification, changing the MultiCAN module with REC and TEC to the same value 0x1, setting the INIT bit to 1B, and issuing the BOFF interrupt. The bus-off recovery phase starts automatically. Every time an idle time is seen, REC is incremented. If REC = 0x60, a combined status EWRN+BOFF is reached. The corresponding interrupt can also be seen as a pre-warning interrupt, that the bus-off recovery phase will be finished soon. When the bus-off recovery phase has finished (128 times idle time have been seen on the bus), EWRN and BOFF are cleared, the ALERT interrupt bit is set and the INIT bit is still set. MultiCAN_AI.H008 Effect of CANDIS on SUSACK When a CAN node is disabled by setting bit NCR.CANDIS = 1B, the node waits for the bus idle state and then sets bit NSR.SUSACK = 1B. However, SUSACK has no effect on applications, as its original intention is to have an indication that the suspend mode of the node is reached during debugging. MultiCAN_TC.H002 Double Synchronization of receive input The MultiCAN module has a double synchronization stage on the CAN receive inputs. This double synchronization delays the receive data by 2 module clock cycles. If the MultiCAN is operating at a low module clock frequency and high CAN baudrate, this delay may become significant and has to be taken into account when calculating the overall physical delay on the CAN bus (transceiver delay etc.). Errata Sheet 52 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description MultiCAN_TC.H003 Message may be discarded before transmission in STT mode If MOFCRn.STT=1 (Single Transmit Trial enabled), bit TXRQ is cleared (TXRQ=0) as soon as the message object has been selected for transmission and, in case of error, no retransmission takes places. Therefore, if the error occurs between the selection for transmission and the real start of frame transmission, the message is actually never sent. Workaround In case the transmission shall be guaranteed, it is not suitable to use the STT mode. In this case, MOFCRn.STT shall be 0. MultiCAN_TC.H004 Double remote request Assume the following scenario: A first remote frame (dedicated to a message object) has been received. It performs a transmit setup (TXRQ is set) with clearing NEWDAT. MultiCAN starts to send the receiver message object (data frame), but loses arbitration against a second remote request received by the same message object as the first one (NEWDAT will be set). When the appropriate message object (data frame) triggered by the first remote frame wins the arbitration, it will be sent out and NEWDAT is not reset. This leads to an additional data frame, that will be sent by this message object (clearing NEWDAT). There will, however, not be more data frames than there are corresponding remote requests. Errata Sheet 53 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description C AN Bus r e m o te re q u e s t r e m o te re q u e s t d a ta d a ta lo s s o f a rb itra tio n M u ltiC A N s e tu p d a ta o b je c t c le a r s e t NEW DAT by H W c le a r NEW DAT by H W Figure 8 s e tu p d a ta s e tu p o b je c t d a ta o b je c t c le a r NEW DAT by H W Loss of Arbitration OCDS_X.H003 Debug Interface Configuration by User Software If the debug interface must be (re)configured, the sequence of actions to follow is: 1. 2. 3. 4. activate internal test-logic reset by installing SCU_DBGPRR.TRSTGT=0 disable debug interface by installing SCU_DBGPRR.DBGEN=0 install desired debug interface configuration in SCU_DBGPRR[11:0] activate pull-devices (if internal will be used) by installing Px_IOCRy accordingly 5. enable debug interface by installing SCU_DBGPRR.DBGEN=1 6. release internal test-logic reset by installing SCU_DBGPRR.TRSTGT=1 These steps must be performed as separate, sequential write operations. PVC_X.H001 PVC Threshold Level 2 The Power Validation Circuits (PVCM, PVC1) compare the supply voltage of the respective domain (DMP_M, DMP_1) with programmable levels (LEV1V and LEV2V in register SCU_PVCMCON0 or SCU_PVC1CON0). The default value of LEV1V is used to generate a reset request in the case of low core voltage. Errata Sheet 54 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description LEV2V can generate an interrupt request at a higher voltage, to be used as a warning. Due to variations of the tolerance of both the Embedded Voltage Regulators (EVR) and the PVC levels, this interrupt can be triggered inadvertently, even though the core voltage is within the normal range. It is, therefore, recommended not to use this warning level. LEV2V can be disabled by executing the following sequence: 1. Disable the PVC level threshold 2 interrupt request SCU_PVCMCON0.L2INTEN and SCU_ PVC1CON0.L2INTEN. 2. Disable the PVC interrupt request flag source SCU_INTDIS.PVCMI2 and SCU_INTDIS.PVC1I2. 3. Clear the PVC interrupt request flag source SCU_DMPMITCLR.PVCMI2 and SCU_ DMPMITCLR.PVC1I2. 4. Clear the PVC interrupt request flag by writing to SCU_INTCLR.PVCMI2 and SCU_INTCLR.PVC1I2. 5. Clear the selected SCU request flag (default is SCU_1IC.IR). The Power Validation Circuits (PVCM) compare the supply voltage of the respective domain (DMP_M) with programmable levels (LEV1V and LEV2V in register SCU_PVCMCON0). The default value of LEV1V is used to generate a reset request in the case of low core voltage. LEV2V can generate an interrupt request at a higher voltage, to be used as a warning. Due to variations of the tolerance of both the Embedded Voltage Regulators (EVR) and the PVC levels, this interrupt can be triggered inadvertently, even though the core voltage is within the normal range. It is, therefore, recommended not to use this warning level. LEV2V can be disabled by executing the following sequence: 1. Disable the PVC level threshold 2 interrupt request SCU_PVCMCON0.L2INTEN. 2. Disable the PVC interrupt request flag source SCU_INTDIS.PVCMI2. 3. Clear the PVC interrupt request flag source SCU_DMPMITCLR.PVCMI2. 4. Clear the PVC interrupt request flag by writing to SCU_INTCLR.PVCMI2. 5. Clear the selected SCU request flag (default is SCU_1IC.IR). Errata Sheet 55 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description RESET_X.H003 How to Trigger a PORST after an Internal Failure There is no internal User Reset that restores the complete device including the power system like a Power-On Reset. In some applications it is possible to connect ESR1 or ESR2 with the PORST pin and set the used ESR pin as Reset output. With this a WDT or Software Reset can trigger a Power-On Reset. A detailed description is in the Application Note AP16103. RTC_X.H003 Changing the RTC Configuration The count input clock fRTC for the Real Time Clock module (RTC) can be selected via bit field RTCCLKSEL in register RTCCLKCON. Whenever the system clock is less than 4 times faster than the RTC count input clock (fSYS < fRTC × 4), Asynchronous Mode must be selected (bit RTCCM = 1B in register RTCCLKCON). To assure data consistency in the count registers T14, RTCL, RTCH, the RTC module must be temporarily switched off by setting bit MODEN = 0B in register RTC_KSCCFG before register RTCCLKCON is modified, i.e. whenever • • changing the operating mode (Synchronous/Asynchronous) Mode in bit RTCCM, or changing the RTC count source in bit field RTCCLKSEL. SCU_X.H009 WUCR.TTSTAT can be set after a Power-Up After power-up the wake-up clock fWU is selected for the Wake-Up Timer (WUT). In this case, the trim interrupt trigger cannot be used, because the WUT trim trigger status bit (WUCR.TTSTAT) might become set erroneously. This happens sporadically and is, therefore, difficult to find in the development phase of an application. If the trim interrupt trigger is enabled this may lead to unintended SCU interrupts that may also block other interrupt sources (see INT_X.H004). This can be avoided by executing the following sequence: 1. Disable the trim interrupt source SCU_INTDIS.WUTI 2. Clear the trim interrupt request flag by writing to INTCLR.WUTI 3. Clear the selected SCU request flag (default is SCU_1IC.IR) Errata Sheet 56 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description SWD_X.H001 Application Influence on the SWD The internal Supply Watchdog (SWD) monitors the external supply voltage of the pad I/O domain VDDPB which is connected to the device. Therefore, adjustable threshold levels are defined over the complete supply voltage range. These limits are also influenced by system environment and may deviate due to external influences slightly from the values given in the Datasheet. Independent of the SWD is the internal start up and operation protected by the PVC, which monitor the core voltage. USIC_AI.H001 FIFO RAM Parity Error Handling A false RAM parity error may be signalled by the USIC module, which may optionally lead to a trap request (if enabled) for the USIC RAM, under the following conditions: • • • a receive FIFO buffer is configured for the USIC module, and after the last power-up, less data elements than configured in bit field SIZE have been received in the FIFO buffer, and the last data element is read from the receiver buffer output register OUTRL (i.e. the buffer is empty after this read access). Once the number of received data elements is greater than or equal to the receive buffer size configured in bit field SIZE, the effect described above can no longer occur. To avoid false parity errors, it is recommended to initialize the USIC RAM before using the receive buffer FIFO. This can be achieved by configuring a 64-entry transmit FIFO and writing 64 times the value 0x0 to the FIFO input register IN00 to fill the whole FIFO RAM with 0x0. USIC_AI.H002 Configuration of USIC Port Pins Setting up alternate output functions of USIC port pins through Pn.IOCRy registers before enabling the USIC protocol (CCR.MODE = 0001B, 0010B, 0011B or 0100B) might lead to unintended spikes on these port pins. To avoid Errata Sheet 57 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description the unintended spikes, either of the following two sequences can be used to enable the protocol: • • Sequence 1: – Write the initial output value to the port pin through Pn_OMR – Enable the output driver for the general purpose output through Pn_IOCRx – Enable USIC protocol through CCR.MODE – Select the USIC alternate output function through Pn_IOCRx Sequence 2: – Enable USIC protocol through CCR.MODE – Enable the output driver for the USIC alternate output function through Pn_IOCRx Similarly, after the protocol is established, switching off the USIC channel by reseting CCR.MODE directly might cause undesired transitions on the output pin. The following sequence is recommended: • • • Write the passive output value to the port pin through Pn_OMR Enable the output driver for the general purpose output through Pn_IOCRx Disable USIC protocol through CCR.MODE USIC_AI.H003 PSR.RXIDLE Cleared by Software If PSR.RXIDLE is cleared by software, the USIC is not able to receive until the receive line is detected IDLE again (see User’s Manual chapter Idle Time). For UART based busses with higher traffic e.g. LIN it is possible that sometimes the next frame starts sending before PSR.RXIDLE is set 1B by hardware again. This generates an error. A solution is, that the PSR.RXIDLE bit is not cleared in software. Errata Sheet 58 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description 6.4 Documentation Updates EBC_X.D001 Visibility of Internal LXBus Cycles on External Address Bus EBC chapter “Access Control to LXBus Modules” receives the following correction: In the first paragraph the term “read mode” is replaced by “tri-state mode”. The following is added: Despite the above mentioned measures, accesses to internal LXBus modules are to some extent reflected on the non-multiplexed address pins A[23:0] of the external bus. 1. During an internal LXBus access, the external address bus is tri-stated. The switch to tri-state mode occurs in the same cycle as the internal LXBus access. This may induce residual voltage which can lead to undefined logic levels on the address bus pins. Those in turn can cause unwanted switching activity on attached device input stages. Therefore attached devices should be equipped with an input hysteresis filter to avoid unwanted switching activity. 2. After an internal LXBus access is completed the address of the location accessed last on the LXBus becomes visible on the external address bus, unless an external bus cycle immediately follows the LXBus cycle. Due to this behavior, switching activity on the address bus can be observed even if no external access is active. Note: A functional impact due to this behavior is not expected because external bus control signals are held inactive during the internal LXBus access. ECC_X.D002 Initialization of the Read-Control Logic In chapter `ECC Error Handling` (8.14.3) of the User's Manual version 2.0 the following note should be added. Note: The state of the read-control logic must get cleared after initialization of a RAM with ECC enabled. This is achieved with a read operation from one Errata Sheet 59 V1.6, 2013-06 XC2300A Derivatives XC2000 Family / Base Line Detailed Errata Description location in each of the RAMs which was initialized before. This read operation must get executed before enabling the corresponding ECC trap. RESET_X.D001 Reset Types of Trap Registers The reset type of SCU registers TRAPDIS, TRAPSET, TRAPNP and TRAPNP1 is an Application Reset. In the next revision of the user’s manual the reset type of this registers will be changed from a Power-on Reset to an Application Reset. USIC_X.D001 Connection from CCU6x to USICx In the User’s Manual V2.0 chapter 20 “Capture/Compare Unit 6 (CCU6)” the connection between CCU6x and USICx is correct described but not in chapter 21 “Universal Serial Interface Channel”. The correct description for chapter 21 “Universal Serial Interface Channel” is CCU6x_COUT63 to USICx_DX2F. USIC_X.D002 USIC1 Channel 0 Connection DX0C and DX0D The connectivity description for USIC1 channel 0 DX0C and DX0D in the User’s Manual V2.0 chapter “Universal Serial Interface Channel”, table “I/O Connections of USIC1” is not correct. The correct I/O connection for USIC1 channel 0 DX0C is P10.12 and for DX0D is P10.13. Errata Sheet 60 V1.6, 2013-06 w w w . i n f i n e o n . c o m 02627AERRA Published by Infineon Technologies AG