* StopClick version 1.0 for SetCPU1.6 users * Written by Ismo Suihko, August 4, 1990 (isuihko@ujocs.joensuu.fi) * Released to the Public Domain * * Routines _GetMMUType and TestFlags are borrowed from SetCPU1.6, * which is written by Dave Haynie. * Also include file setcpu.i is borrowed from SetCPU sources. * * Stops clicking of empty drives by patching 1.3 KS V34.5 ROM (check * it with command 'version') image loaded in 2-bit RAM with * 'SetCPU FASTROM'. Disk drive heads will after the patch be stepped * (silently) in a negative rather than positive direction during the * regular system checks for a disk change. * Version 1.6 of SetCPU isn't doing that patch anymore. * StopClick also fixes trackdisk read/write bug (just like NoClick did). * This patching is possible, because ROM image isn't protected at all. * StopClick does nothing, if there isn't MMU and the given address * doesn't seem to contain right version of the ROM image, so it * is safe to run this even if you sometimes boot on 68000 mode. * * Based on the NoClick program, which is written by Norman Iscove. * It didn't work on my system, because of SetPatch and TDpatch13 * (TDpatch13 comes with CrossDos (a great product!)). * * USAGE: * First check the address of the KERNEL by using command 'SetCPU VERBOSE' * If you have A2620 or A2630 with 2 MB 32-bit RAM, then output may * contain this line * KERNEL: (PADDR: $ 3a0000) (VADDR: $ fc0000) (SIZE: 256K) * ^^^^^^ * Then give that address as a parameter to StopClick, like * StopClick 3a0000 ;(you can use numbers 0-9 and letters A-F and a-f) * Don't put '$' before the address, because it is used by shell. * Put StopClick to a startup-sequence after SetCPU FASTROM when you have * checked what the address of the KERNEL is. If you add there more memory, * you have to check the address again and do modifications if necessary. * * e.g. * SetCPU FASTROM * stopclick 3a0000 ; KERNEL is at address $3a0000 * * There aren't any checks if there is a FASTROM image REALLY IN USE. * We only check, if there are signs of a 1.3 KS V34.2 ROM image.. * If FASTROM has been actice once, but then disactivated, the image could * still be in the 32-bit RAM, so it is found from there, and patches are * made but without any effect (and no harm either). * * Return codes: * 0 complete success, fixes made now or they had been made earlier * 5 no MMU (Memory Management Unit) * 10 no KS ROM image found, or wrong version of ROM * 15 KS ROM image found, but modifications couldn't be done?? * 99 invalid parameter * We get return code 5 always if there isn't a MMU. * * BUGS: * No known bugs. * Maybe the address of KERNEL could be found by this program somehow... * * Sometimes clicking will end only after a couple of seconds, just like * with NoClick. Don't ask me why it happens so, as I don't know. * And there may be some odd floppy disk drives which don't like to step * in a negative direction. I haven't yet heard anyone who owns such a drive. * * History: * V0.9 Jun 23, 1990 works, do we need more? * V1.0 Aug 04, 1990 some more checking, now we check if there is a MMU, * earlier we checked only if there was a '020 or better. * program is now 480 bytes long * V?.? ??? ??, ???? will find the address of the ROM image by itself * * To compile first get the setcpu.i include file from the SetCPU1.6 sources, * comment out (add ';'s) the last two lines containing the text * ; machine mc68020 * ; mc68881 * and then compile with commands * Assem StopClick.a -o StopClick.o -i INC: * Blink StopClick.o to StopClick * Makefile is also provided for easier compilation and linking. * exec.library routines LVODisable EQU -120 LVOEnable EQU -126 LVOCloseLibrary EQU -414 LVOOpenLibrary EQU -552 * dos.library routines LVOWrite EQU -48 LVOOutput EQU -60 * some macros, etc. include "setcpu.i" movem.l d0/a0,-(sp) save information of command line parameters bsr _GetMMUType tst.l d0 is there a MMU? 68851 or 68030? bne.s scan * no MMU addq.l #8,sp forget parameters before quitting moveq #5,d0 no MMU bra exit scan movem.l (sp)+,d0/a0 * Now check, if there is a valid hexadesimal address given subq.l #1,d0 take out end of line (null) beq.s scan_err is there any parameters at all subq.l #1,d0 counter (length of the parameter) moveq #0,d1 result moveq #0,d2 character from the command line scan_loop move.b (a0)+,d2 get a character from the command line cmpi.b #' ',d2 check for the end of hex number beq.s scan_ready is it a space cmpi.b #';',d2 beq.s scan_ready is it a start of a comment cmpi.b #9,d2 beq.s scan_ready is it a tab moveq #'0',d3 cmpi.b #'9',d2 ble.s scan_number bclr.l #5,d2 [a-f] -> [A-F] cmpi.b #'A',d2 blt.s scan_err less than 'A' cmpi.b #'F',d2 bgt.s scan_err greater than 'F' moveq #'A'-10,d3 scan_number sub.b d3,d2 blt.s scan_err less than '0' asl.l #4,d1 multiply by $10 add.l d2,d1 dbra d0,scan_loop scan_ready tst.w d1 should be of form xxxx0000 beq.s check scan_err moveq #99,d0 bra.s exit check movea.l d1,a2 address of possible ROM image * Now check if the given address may contain ROM image with exec of V34.2 cmpi.w #' 3',$1c(a2) bne.s bad_rom cmpi.l #'4.2 ',$1e(a2) bne.s bad_rom * add offset of trackdisk.device, should be of version 34.1 adda.l #$29564,a2 cmpi.l #'34.1',$36(a2) beq.s patch bad_rom moveq #10,d0 no ROM image found bra.s exit patch moveq #2,d7 2 patches * disable interrupts CALLSYS Disable * first we check, if we need to patch, then we check was the patch successful * Stop the clicks * fix 'bchg #$01,$41(a3)' to 'bset #$01,$41(a3)' lea $104(a2),a0 cmpi.w #$086b,(a0) bne.s 1$ move.w #$08eb,(a0) 1$ cmpi.w #$08eb,(a0) bne.s 2$ subq #1,d7 2$ * Repair trackdisk read/write bug originally at $feaf9c (1.3) * fix 'cmp.l $8000,d0' to 'cmpi.l #$8000,d0' lea $1a38(a2),a0 cmpi.w #$b0b9,(a0) bne.s 3$ move.w #$0c80,(a0) 3$ cmpi.w #$0c80,(a0) bne.s 4$ subq #1,d7 4$ * enable interrupts CALLSYS Enable move.l d7,d0 if d7 = 0, patching was succesful beq.s exit moveq #15,d0 patches failed?? exit tst.b d0 beq.s real_exit * print the usage text move.l d0,-(sp) push the return code lea dosname(pc),a1 moveq #0,d0 CALLSYS OpenLibrary tst.l d0 beq.s 2$ move.l d0,a5 exg a5,a6 jsr LVOOutput(a6) move.l d0,d1 beq.s 1$ lea usage(pc),a0 move.l a0,d2 moveq #usagelen,d3 jsr LVOWrite(a6) 1$ exg a5,a6 move.l a5,a1 CALLSYS CloseLibrary 2$ move.l (sp)+,d0 pop the return code real_exit rts * The following routines are borroved from SetCPU1.6 sources. * I optimized the code a slightly to make my program smaller than 488 * bytes, because otherwise it would have taken 1+2 disk blocks using OFS. * I removed also the 68040 check, because at this context we don't need * to know whether there is exactly '030 or '040 as both have MMU. ;====================================================================== ; ; This routine checks CPU flags early in ExecBase for extended ; CPUs that test as a 68020 under 1.3. If these flags are set, ; the actual CPU/MMU type test can be skipped. ; ;====================================================================== TestFlags: moveq.l #0,d0 * btst.b #AFB_68040,ATNFLGS(a6) ; Does the OS think an '040 is here? * beq.s NoEarly40 * move.l #68040,d0 * rts *NoEarly40: btst.b #AFB_68030,ATNFLGS(a6) ; Does the OS think an '030 is here? beq.s NoEarly30 move.l #68030,d0 ; Sure does... NoEarly30: rts ;====================================================================== ; ; This function returns 0L if the system contains no MMU, ; 68851L if the system does contain an 68851, or the CPU number ; for CPUs with integral CPUs. ; ; This routine seems to lock up on at least some CSA 68020 ; boards, though it runs just fine on those from Ronin and ; Commodore, as well as all 68030 boards it's been tested on. ; ; ULONG GetMMUType() ; ;====================================================================== _GetMMUType: move.l 4,a6 ; Get ExecBase bsr TestFlags ; Check extended CPU types tst.l d0 beq.s MMURealTest rts ; For any other machine, a real test must be done. The test will ; try an MMU instruction. The instruction will fail unless we're ; on a "bogus MMU" system, where the FPU responds as an MMU. MMURealTest: movem.l a3/a4/a5,-(sp) ; Save this stuff suba.l a1,a1 CALLSYS FindTask ; Call FindTask(0L) move.l d0,a3 move.l TC_TRAPCODE(a3),a4 ; Change the exception vector lea MMUTraps(pc),a0 move.l a0,TC_TRAPCODE(a3) moveq.l #-1,d0 ; Try to detect undecode FPU subq.l #4,sp ; Get a local variable PMOVE_ tc,(sp) ; Let's try an MMU instruction addq.l #4,sp ; Return that local move.l a4,TC_TRAPCODE(a3) ; Reset exception stuff movem.l (sp)+,a3/a4/a5 ; and return the registers rts ; This is the exception code. No matter what machine we're on, ; we get an exception. If the MMU's in place, we should get a ; privilige violation; if not, an F-Line emulation exception. MMUTraps: move.l (sp)+,d0 ; Get Amiga supplied exception # cmpi #11,d0 ; Is it an F-Line? beq.s MMUNope ; If so, go to the fail routine move.l #68851,d0 ; We have MMU addq.l #4,2(sp) ; Skip the MMU instruction rte MMUNope: moveq.l #0,d0 ; It dinna woik, addq.l #4,2(sp) ; Skip the MMU instruction rte * some strings dosname dc.b 'dos.library',0 usage dc.b 'StopClick v1.0 by IS. ' dc.b 'Usage: StopClick address-of-1.3-KS-FASTROM-image',10 usagelen EQU *-usage END