k60 bootloader

/*********************************************************************** * $File Name: AN2295.bootloader.changelog.txt$ * Project: AN2295 Developper's HC08 Bootloader change log * $Version: 9.1.67.0$ * $Date: Sep-12-2011$ * $Last Modified By: r30323$ * Company: Freescale Semiconductor * Security: General Business * ************************************************************************/ 01-Jul-2003: Single-wire option implemented into QT/QY and JK/JL bootloaders. Free locations in interrupt vector tables used too. [r30323] 01-Aug-2003: SR12 port done, but not tested yet at all! [r30323] 16-Oct-2003: LJ12 bootloader successfully tested & prepared for release [r30323] LJ24 and ext. xtal options were not tested yet. 20-Oct-2003: corrected MR8 BootLoader bug (part of application vector table has been protected by FLBPR) corrected RAM allocation for exec. buffer [r30323] 20-Oct-2003: release 4.0 [r30323] 19-Nov-2003: AP8/16/32/64 version created (but not yet tested) [r30323] 27-Jan-2004: QT/QY + JK/JL families only: ONEBIT now correctly initializes to the default value (Bootloading will not be affected by erased private area) [r30323] 28-Jan-2003: SR12 bootloader corrected and tested! Works OK now. [r30323] 02-Feb-2004: AP8/16/32/64 version tested by 3rd party. Reported working. [r30323] Major update - version 2 of protocol implemented (specially for HCS08) Created port for HCS08(GB/GT)(32/60), successfully tested. [r30323] 12-Feb-2004: in AP8/16/32/64 version corrected FLS_END allocation also tried to squeeze bootloader user table to minimum (80B) instead of 512B (erase page size) with following warning: This should work OK except one scenario: if only portion of FLASH is being re-programmed (e.g. S19 without interrupt vectors = no int. vect. translation occurs) AND and these memory location are in the same (last) page as bootloader user table, the table will be erased and not restored Same backported to MR & LJ families [r30323] 18-Feb-2004: Release/revision 5.0 31-Mar-2004: AZ/AS32 version created, to be tested [r30323] 08-Apr-2004: AZ/AS32 version tested, AB32 port generated based on this. Not tested yet. [r30323] 09-Apr-2004: GR16 version created (as part of GR4/8 tree), minor modifications of GR4/8. Not tested. QB4/8 version created, not tested yet. [r30323] 13-Apr-2004: LJ version had incorrect SCBR setting in Ext. Xtal mode (4.9152MHz). [r30323] 14-Apr-2004: LB8 version created & tested. The location of ERARNGE & PRGRNGE in ROM routines jump table is different from other HC08s. [r30323] 16-Apr-2004: S08GT16 family added. SCIAPI documentation (sci.h in JK/JL, QY/QY, LB) updated. [r30323] 27-Apr-2004: AB/AS/AZ32/60 family unified. For Large HC08s (with more than 1 Flash block), the FC protocol version 3 has been created. It's combination of IDENT format of S08 (version 2) and vector relocation for regular HC08 (version 1). All three (AB/AS/AZ) were ported to protocol version 3. [r30323] All bootloader code for AB/AS/AZ family is related to A-suffixed MCUs (the ones with new flash type). [r30323] 29-Apr-2004: GZ8/16 family added. Not yet tested. GZ60 added as a part of AB/AS/AZ32/60 tree. [r30323] 07-May-2004: developed EEPROM programming routines (right now only for AZ60A family, to be extended for all other EEPROM enabled HC08s). [r30323] 10-May-2004: removed GZ60 from ABASAZ tree (now too different from EEPROM enabled ABASAZ HC08s). [r30323] 12-May-2004: corrected & tested a complete ABASAZ tree with EEPROM programming features. [r30323] 03-Nov-2004: Release/revision 6.0, Freescale introduced [r30323] 01-Dec-2004: GR8A version introduced (8MHz clock XTAL needed), to be tested. [r30323] 05-Jun-2005: S08Rx (RD/RG/RE) family introduced and tested. RC family not included, SCI not present. [r30323] 10-Jun-2005: JK/JL8 family added and preliminary tested. [r30323] 14-Jun-2005: Minor typo in master PC software. [r30323] 20-Jun-2005: GR8A family tested by 3rd party. Reported working. AP family cleaned up a little. [r30323] 30-Jun-2005: MWCW 3.0 to MWCW 3.1 migration. [r30323] 25-Jul-2005: Ported to QC family. Several versions exists: 'no-trim' that depends on factory trim @0xFFC0. Since trim value is different on different ICG frequency settings, bootloader and/or application may not communicate correctly with another trim value. 'no-trim' bootloader starts at 3.2MHz BUS clock and expects factory trim @0xFFC0 to be valid for this freq. There're also two auto-trim bootloader versions: 'low-trim' that works at 3.2MHz BUS clock, 'hi-trim' that operates at 6.4MHz BUS clock (5V only). [r30323] QC auto-trim back ported to QB family ('low-trim' only since QB has no high-speed 6.4MHz BUS clock option). All auto-trim version use own trim value that is stored on first byte of bootloader private table. It can be accessed from application by form of simple pointer, that is stored on unused memory location 0xFFF8. See example: #define TRIMAPIADDR 0xfff8 #define TRIMAPI (*((unsigned int *)(TRIMAPIADDR))) #define TRIMValue (*((unsigned char *)TRIMAPI)) QY8 chip not yet supported since it also has several internal oscillator frequencies selection. This needs to be backported to QY/QT bootloader tree in future. [r30323] 22-Aug-2005: QY/QT tree got Bump COP inside Read subroutine. This will allow to use this SCIAPI routine also in appliations that have COP enabled. TO BE TESTED. [r30323] 05-Sep-2005: GZ60 family corrected no. of Flash blocks and memory allocations. GZ32 & GZ48 introduced. To be tested. [r30323] 26-Sep-2005: LJ/LK24 family had incorrect RAM mapping. Also FLBPR register is in FLASH instead(!) LJ/LK24 corrections tested by 3rd party. Reported working. GR family included MOV #$1C,$0033 instruction in order to correctly initialize an unused timer channel. Added another optional configuration for GR8A family with COP enabled and LVI set to 3.0V. [r30323] 26-Sep-2005: Release/revision 7.0 25-Oct-2005: AS/AZ32/60A families had incorrect interrupt vectors allocations. Corrected. [r30323] 18-Jan-2006: S08AW32/48/64 family created and briefly tested. [r30323] 22-Feb-2006: It has been reported that some QT/QY masksets are susceptible to unwanted erasure that is described in: http://www.freescale.com/files/microcontrollers/doc/errata/MSE908QY4_3L69J.pdf No workaround so far (either use newer mask set or relocated QT/QY bootloader one erase page lower in memory to avoid code at 0xFFB0-0xFFBF memory location. [r30323] 22-Feb-2006: ----------------------------------------------------------------------------------------------------------------- MAJOR CHANGE: PIN RESET as a valid source of bootloader start has been removed from all targets since this very often caused bootloader lockup - usually when a capacitor was placed on RESET pin, * ILOP instruction at the end of bootloading caused MCU reset, * the capacitor has been discharged by internal transistor, * then after reset * capacito