![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to ChangeLog CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [UNIX Amiga Emulator] / uae / src|
Return to ChangeLog CVS log | Up to [UNIX Amiga Emulator] / uae / src |
1.1.1.8 ! root 1: 990216 picasso96.c: Some optimizations. ! 2: cfgfile.c: Save target specific options. Allow simple/smart choice ! 3: for centering options. Delete mount_data junk, substitute immediately. ! 4: Whitespace handling fixes. Fix config_description typo. ! 5: m68k flag handling redone. ! 6: 990212 Implement rotate instructions without loops. Some other m68k ! 7: optimizations. ! 8: Some configure.in fixes. ! 9: 990122 gtkui.c: Fixes for gtk 1.1.13. ! 10: 990113 genp2c.c: Don't include custom.h. ! 11: 990109 Some fixes in cfgfile.c. Make svga.c compile again. 1.1.1.7 root 12: 981229 Miscellaneous fixes in the new config file stuff. 13: In config files, allow 68ec020 as CPU type, and get rid of 24 bit 14: addressing option. 15: Make the get_disp_ea non-inline. 16: SGI sound support (Ari Heikkinen). 17: More updates by Brian: 18: audio.c: Handle frequency, too, when changing sound prefs. Do some 19: work in update_audio even if not generating sound output. 20: filesys.c: Request 50 buffers in our parampacket. 21: Assorted right control key fixes. 22: Some updates in od-win32 (not nearly done...) 23: 981228 Config file and option handling redone; based on work by Brian. 24: Rename UAEn: devices to DHn: (Brian). 25: filesys.c: Implement ACTION_CHANGE_MODE, ACTION_PARENT_FH; add support 26: for variable hardfile blocksizes (Brian). 27: Some Picasso96 updates (Brian). 28: 981226 Win32 fsusage.c fixes; zfile bug fix; gui_fps addition (Brian). 29: 981221 Fix gtkui to work with newer versions of gtk. 30: Fix gencpu so that FPU emulation doesn't crash when unswapped word reads 31: are available. 32: 981129 gencpu.c: Delete all the queue_amode stuff again. 33: xwin.c: Allow switching between DGA and window mode (Marcus Sundberg). 34: DGA seems broken in X 3.3.3, however :( 35: 981011 Teach cpuopti.c about epilogues that increment esp by popping. 36: 980930 Add BSD socket code (Mathias Ortmann). 37: 980923 filesys.c: Implement FH_FROM_LOCK (Brian King) 38: 980823 xwin.c: French keyboard patch (Rapha�l Poss). 39: 980810 Fix RDTSC detection (Rapha�l Poss). 40: newcpu.c: Patch by Brian King to display what kind of CPU table is 41: built. 42: 980716 More AGA patches: use new macro GET_PLANES to calculate number of 43: bitplanes from BPLCON0; adapt the copper code; add BPLPT[HL][78] and 44: BPLDAT[78] (Toni Wilen) 1.1.1.6 root 45: 980715 Change sd-uss/sound.c not to keep a file open in setup_sound, to 46: support reconfiguration. 47: 980713 Fix m68k detection in configure.in. Clear execbase after allocating 48: chipmem (Holger Jakob) 49: od-dos update, tui.c patches (Gustavo Goedert) 50: Get rid of --enable-sound, always enable sound. 51: Make svga.c compilable. 52: 980712 md-ppc-gcc patches (Holger Jakob) 53: 980707 gtk ui: Better CPU speed selection. 54: Move initializations from add_filesys_unit to set_filesys_unit. 55: 980706 AGA patches by Toni Wilen: Call calcdiw when changing fmode; fix 56: references to bplcon variables to go through dp_for_drawing; ifdef out 57: some code that's too hard to fix for now. 58: "Fix" the AGA color code. In pfield_linetoscr_aga, add lframe_end and 59: diw_end parameters and use them. Add some fill_line code for AGA. Make 60: sure no code touches acolors if [AGA_CHIPSET != 0] 61: New function finish_line_aga. 62: 980705 drawing.c/custom.c: Get rid of line_changed, replace with a single 63: variable thisline_changed. Pass it to hsync_record_line_state and 64: evaluate it there. Get rid of BORDER_PREV and BORDER_NEXT states; 65: add LINE_BLACK/LINE_REMEMBERED_AS_BLACK to implement scanlines option. 66: Remove code in pfield_draw_line that checks line_changed. 1.1.1.5 root 67: 980702 filesys.c: New function move_filesys_unit. 68: Win32 GUI fixes. 69: 980701 gtk UI beautification and code cleanup (Michael Krause) 70: Beginning to make audio code run-time configurable (mostly by Michael 71: Krause) 72: Fix off-by-one errors in memory check handlers (Jake Hamby) 73: configure: Add AM_CYGWIN32, AM_EXEEXT and AM_MINGW32 and fix some Win32 74: configuration problems. 75: genp2c: Fix symbol name braindamage. 76: md-i386-{gcc,watcom}.c: Don't do machdep_init when [_WIN32]. 77: Many Win32 compilation fixes. 78: filesys.c: Attempt to fix the mount code chaos. 1.1.1.4 root 79: 980629 MOVEP bug fixes (Philippe Gerin) 1.1.1.5 root 80: Comment out code using readdir_r again. 81: 980628 gencpu.c: Generate an end label only when it's needed. 82: configure.in: Check for windows.h, ddraw.h, sys/utime.h; check sizeof 83: __int64; check whether a small DirectX program compiles; check for 84: Watcom C; don't do GNOME_FILEUTILS_CHECKS when compiling on Win32 85: without GNU C; use AC_EGREP_CPP in some places instead of compiling a 86: file by hand; check for GAS .p2align feature; check for USE_UNDERSCORE 87: with a small test program; add code to set up od-win32/md-i386-watcom 88: directories when appropriate. 89: genp2c.c: Use GAS .p2align feature; don't generate .type directive. 90: Rename md-*/memory.h to md-*/maccess.h. 91: More fixes in Win32 code. 92: Add wcc.sh and wrc.sh, two scripts to make it possible to use the 93: Watcom C compiler and the resource compiler in a Cygwin32 build 94: environment. 95: 980626 Change zfile.c not to use fopen_del. 96: 980614 Delete ihf_sem stuff. 97: picasso_refresh now nonstatic and called by check_picasso after a 98: notice_screen_contents_lost call. 99: More functions to access filesystem mount information. 100: More Win32 code. 101: 980613 Use an AC_TRY_LINK block to check for libpthread. 102: 980611 Rename lockscr/unlockscr to gfx_{lock,unlock}_picasso and add new 103: functions lockscr/unlockscr which are now called by drawing.c. 104: xwin.c: Clear picasso_invalid_lines in graphics_subinit and when 105: doing an XPutImage for a refresh. 106: New function reset_all_systems, which is a bit misnamed at the moment 107: because it only resets the filesystem. Called from m68k_go when a 108: reset is needed. 109: In filesys_prepare_reset, examine unit_pipe instead of self to 110: determine whether a unit has a thread that needs to be killed. Move 111: the reset_state and reset_sync_sem variable to UnitInfo so they are 112: available at all times. 113: Don't need to call finish_drawing_frame if we do a reset, it was 114: done a couple of lines earlier. 115: In uae_reset, do nothing if any reset or quit operations are in 116: progress. 117: Make DMACONR static. 118: 980609 Add memory size fields in struct uae_prefs. New variables 119: allocated_xxxmem for every block of memory that holds the actual 120: amount currently allocated, copied from the preferences at allocation 121: time. 122: Create a new structure "uaedev_mount_info" to hold filesystem mount 123: information. Add a pointer to one of those in struct uae_options, 124: and copy it from currprefs to a filesystem private place when 125: starting the filesystem threads to a work area. 126: Terminate filesystem threads during a reset, and restart the whole 127: filesystem code afterwards so that prefs changes are noticed. 128: 980608 If check_prefs_changed_gfx returns nonzero, or if we switch from/to 129: picasso mode, call notice_new_xcolors and notice_screen_contents_lost. 130: Delete the frame_do_semup crap. 1.1.1.4 root 131: 980606 Create config.h link in configure, not in Makefile. 1.1.1.5 root 132: 980604 xwin.c: No longer need picasso_max[wh]. Improve picasso invalid lines 133: code a bit more. Fix some dithering bugs. 134: gtkui.c (add_checkbox_widget, add_radio_widget): New helper functions. 135: Patches by Marcus Sundberg: 136: Make X11 picasso clut emulation mode work in DGA 137: gfxutil.c (setup_dither): Free redvals. 138: 980602 Split out init_aspect_maps from reset_drawing. 139: Call check_prefs_changed_gfx from vsync_handle_redraw, and initialize 140: row and aspect maps if it returns nonzero. 141: Use an enum instead of integer values to communicate between 142: hsync_handler and hsync_record_line_state. 143: gtkui.c: Use gtk_window_set_title (Marcus Sundberg) 144: xwin.h: Add lockscr/unlockscr prototypes. 145: 980601 Move check_prefs_changed_xxx into vsync_handle_redraw. 146: Make ersatz.c a little more crash-proof. 147: 980531 New graphics specs 'a' and 'p'. Support them in the X11 version. 148: 980530 sysdeps.h, posixemu.c: Change (char *) to (const char *) in the 149: arguments of some of the posixemu functions. 150: sysdeps.h: Delete getopt declaration. Include "getopt.h" in main.c. 151: aclocal.m4: add fsusage.c support from Gnome. 152: fsusage.[ch]: New files from fileutils, slightly modified to fit 153: into UAE and to add Win32 support. Adapt do_info filesys.c to use 154: it, and delete all the old cruft. 155: Some updates for the Win32 port. 156: 980529 Patches for X11 version by Marcus Sundberg (Picasso invalid lines 157: optimization, XShm availability check, cleanup, etc.) 158: 980528 Add field emergmem to struct vidbuf_description, and change 159: pfield_draw_line to use it. 160: 980516 Patches by Samuel Devulder and Holger Jakob: update Amiga port, 161: add support for compiling cpuemu.c in seveal steps again, add 162: cheatsearch in the debugger. 163: 980515 GGI patches (Christian Schmitt). 164: sd-mme: New directory (Marcus Sundberg). 165: Changes in the X11 version to support mixed DGA/normal binaries, 166: hotkeys, and mouse grabbing (Marcus Sundberg) 167: xwin.c: Copy a bit of code from Win32 port that constrains 168: lastmx/lastmy to real coordinates when ievent_alive. 169: 980510 gencpu.c: Slightly better flag generation code. 170: picasso96.c (wgfx_flushline): Don't do anything if Picasso screen 171: is invisible. 172: 980508 Don't disable GGI if we can't find svgalib. 173: newcpu.[ch], gencpu.c, md-i386-gcc/memory.h: Implement unswapped 174: word reads for opcodes. 175: 980505 drawing.c: add missing #ifdef X86_ASSEMBLY. 176: 980504 More BSD fixes from Krister. 177: xwin.c: Call XAllocColorCells for dithered modes. Don't set image_mem 1.1.1.3 root 178: in get_image, instead arrange for it to store the address in a pointer 179: chosen by the caller. 1.1.1.5 root 180: picasso96.h: Add a clut array to picasso_vidinfo. 181: picasso96.c (do_fillrect, do_blit): Use picasso_vidinfo.pixbytes 182: instead of picasso96_state.BytesPerPixel where appropriate. 183: (do_fillrect, do_blit, wgfx_do_flushline): Handle the case when the 184: Amiga-side buffer is a 256 color palette mode and the native buffer 185: is truecolor by looking up colors in the new picasso_vidinfo.clut. 186: xwin.c: Adapt to support emulating a palette display in truecolor. 187: Make tui.c compile. 188: 980501 picasso96.c: Make sure there's an unlockscr for every lockscr call. 189: In SetColorArray, detect when no colors changed. 190: configure.in, aclocal.m4, Makefile.in: Use AM_PATH_GTK from gtk 191: distribution. 192: New field rgbformat in picasso_vidinfo. Set it in xwin.c and svga.c. 193: custom.c (mousehack_helper): Use Picasso screen offsets correctly. 194: 980427 Call decide_line in DIWSTOP. Implement DIWHIGH register. 195: 980426 filesys.c (get_nname): Don't allow Unix "." and ".." directories. 196: gencpu.c: Approximate the number of cycles per instruction. Make 197: generated functions return that number, and use it in newcpu.c in 198: the do_cycles call. 199: custom.c, drawing.c: init_hardware_for_drawing_frame must be 200: called from init_drawing_frame. 201: Move read_processor_time specific stuff to new file md-xxx/rpt.h. 202: md-i386-gcc/support.c: Try to detect whether the CPU has RDTSC. 203: newcpu.c: m68k_speed == 0 means approximate A500 speed; 204: m68k-speed == -1 means use frame rate hack when available. Some 205: changes to support this. 206: 980425 custom.c: New copper code. Unfortunately not much faster than the 207: old one. 208: Get rid of some unused variables throughout. 209: filesys.c: Use readdir_r when available. 210: picasso96.c: In BlitPattern, correctly handle the pattern offsets. 211: missing.c: Get rid of getopt. Add GNU getopt instead. 212: In SPRxPTH, SPRxPTL, call decide_sprites instead of decide_line. 213: 980421 custom.c: Add an extra argument, hpos, to plenty of functions, and 214: use it instead of calling current_hpos. 215: 980420 BeOS update (Christian Bauer). 216: When enabling additional planes in post_decide_line, update 217: thisline_decision.bplcon0. 218: Tweak copper timings. 219: 980419 Extend post_decide_line to handle cases where some planes are 220: enabled much later in the line than the others. 221: In pfield_doline_unaligned_h, don't increment dataptr. 222: Split custom.c, move drawing code to new file drawing.c. New header 223: file drawing.h for declarations of shared stuff. 224: osemu.h: Delete junk. 1.1.1.3 root 225: 980418 gtkui.c: Don't crash if -G option is given. 1.1.1.5 root 226: audio.c, custom.c: Don't use the generic event handling mechanism 227: for audio emulation, the overhead is too high. update_audio: New 228: function, handles audio updates "on demand" or every hsync (Idea by 229: Petter Schau). 230: 980417 Some cleanups in custom.c to make splitting it up easier. 1.1.1.3 root 231: 980416 custom.c: Don't use bplhires in do_modulos. 1.1.1.5 root 232: 980415 readdisk.c: Support FFS (Jim Cooper). 233: BSD patches by Krister Walfridsson. 234: configure.in, Makefile.in: More fixes for build != source directory. 235: Arrange to build p2c_x86.o when using md-i386-gcc. 236: sysdeps.h: Include limits.h. 237: Remove silly code that restricted possible version numbers. 238: 980320 configure.in: Check for vprintf. 239: Make write_log a varargs function. Provide a write_log_standard 240: function that is used by most systems. 241: Refuse to compile with non-ANSI compilers. 242: gencpu.c: Rearrange some of the generated code to reduce register 243: pressure. 1.1.1.3 root 244: 245: Release 0.7.0 246: 980401 gengenblitter.c: Complete rewrite. Much smaller, much faster, much 247: better results. 248: Check for -lXi to support newer versions of gtk+. 249: 980316 gencpu.c (read_counts): Two silly bugs fixed. 250: main.c (save_options): Don't write out 'a' if cpu_level < 2 251: configure.in: Changes to try to be compatible with Solaris /bin/sh. 252: 980315 New UAE-Control version from Tauno & Cloanto. 253: Make configure set UNALIGNED_PROFITABLE for x86. 254: Get rid of "-o", add save command to tui.c and gtkui.c. 255: Fix some bugs in save_options. 256: 980314 Many improvements in the gtk UI. Made CPU type changeable at run-time. 257: Make DGA && ! VIDMODE configuration compile. Kludge around the fact 258: that DGA misses colormap changes. Call XWarpPointer to try to avoid 259: the click-moves-framebuffer bug. 260: Make sure p2c.[ch] don't get deleted. 261: 980313 Move all CPU_LEVEL handling from build68k.c to gencpu.c. In gencpu.c, 262: ignore all parameters and don't write to standard output, instead 263: hardcode all filenames and generate them all at once. Arrange to 264: generate code for all CPU emulator variants and to select them at 265: run-time. 266: Get rid of copper cheat mode, re-use "-C" option for CPU selection. 267: 980312 In COPxLCL, mask out LSB (Axel Wefers). 268: newcpu.h: Always compile in both get_disp_ea versions with a different 269: name. Delete USE_POINTER option; always use pointer. Don't fill 270: prefetch in m68k_setpc, do it in callers where necessary. Compile in 271: two versions of get_ixxxx, with different names for prefetch and 272: non-prefetch. Change gencpu.c to use these appropriately. 273: Don't generate 16 small CPU emulation source files. With a decent 274: machine it's no problem to compile all the code in one file. Also turn 275: off -O3 for gcc, -O2 does fine since all the necessary __inline__s are 276: there, and the compiler takes less memory that way. 277: Don't include <linux/joystick.h>, duplicate some definitions from 278: there instead. Makes configure.in simpler, too. 279: Fix Makefile.in/configure.in and a couple of other places to support 280: building in a separate directory (sort of; not complete yet). 281: Remove p2c.[ch] and genp2c when doing make clean. 282: Delete code to support Tcl/Tk and XForms GUIs. 283: 971215 More Amiga and BeOS patches from Sam and Christian. 284: 971205 Move some more code from custom.c to audio.c. Optimize sample handlers 285: a bit. 286: Make RDTSC asm statement volatile. 287: 971203 Latest GGI patch from Christian Schmitt. 288: Fix configure to enable zfile decompression for GGI target. 289: Fix mouse button mapping in ggi.c. 290: 971128 Fix double definition of gfxvidinfo in custom.c. 291: Fix impossible register constraints in asm statement in newcpu.c 292: Sprinkle casts and rename some variables to make C++ compilers happy. 293: Add BeOS patches by Christian Bauer. 294: 971125 More Amiga patches from Sam. 295: 971120 Get rid of DONT_WANT_SOUND. 296: 971119 Support USS sound not only on Linux (Douglas Carmichael) 297: Update Win32 and DOS ports. 298: Add Polish documentation by Tomasz Sienicki. 299: 971116 Add a missing return statement in filesys.c (Brian King) 300: 971112 Fix silly error in xwin.c. 301: 971108 Add some more decompressors to zfile.c. Set FLOPPY_SPEED to 4. 302: 971107 Make ggi.c compile again, fix a few bugs and add Picasso support. 303: 971106 Fill prefetch queue after executing an encrypted ROM illegal insn. 304: 971105 Initialize the sprites during customreset now that init_hardware_frame 305: no longer does it. 306: 971104 Implement action_set_file_size as required by the specs. 307: SetPanning/picasso96_state: XOffset and YOffset are signed. 308: 971103 xwin.c: Offer 640x512 and 320x256 resolutions in windowed Picasso mode. 309: Add a few more memory banks in memory.c (A3000 stuff - nonfunctional) 310: Don't map kickmem at address 0 if using replacement Kickstart. 311: Must fill prefetch buffer when calling calltraps or ersatz functions. 312: Write out hardfile parameters in write_filesys_config(). 313: xwin.c: Support 15 bit displays. 314: 971101 Add latest patches by Samuel Devulder. Improve hardfile support in 315: tui.c. 316: 971031 Offer "-U" only ifdef PICASSO96. 317: Add support for mounting multiple hardfiles in any order. 318: Don't reset sprite state in init_hardware_frame. Instead, allow SPRxPT 319: functions to set the state to SPR_restart in the vertical blanking 320: area. 321: 971030 Fix 8MB Chip RAM. 322: filesys_dev_storeinfo: New function and calltrap. Use it to store 323: per-unit data (which now includes geometry information) in the 324: parampacket. Modify filesys.asm to use it instead of trying to find 325: the values by itself. 326: 971028 "AMIROMTYPE1" has 11 characters, not 0. 327: 971027 Implement encrypted ROM features. Comment out non-working InvertRect 328: optimization. 329: 971026 Fix some problems in PlanarToDirect and CopyColorIndexMappingA2U. 330: 971022 Remove fixed paths in genp2c.c (oops...) 331: Some more XF86VidMode hacks. 332: Include target.h in sysdeps.h. Define PICASSO96_SUPPORTED for X11, 333: SVGAlib and Win32. Don't compile in Picasso support if this macro isn't 334: defined. 335: 971021 xwin.c: Picasso code wants to use bit_unit, not bitdepth. Also correct 336: byteorder (it's B8G8R8A8 from the Amiga's point of view). 337: Fix Picasso InvertRect function. 338: Half-hearted attempt at 24 bit support in the X11 version, with some 339: awful code in custom.c. 340: Make the X11/VidMode version try to resize the screen for Picasso mode 341: switches. 342: 971020 Delete option -5, make -4 take an argument. Disable z3fastmem and 343: Picasso emulation when using a 24 bit address space. 344: 971019 record_color_change: No need to remember a color table if the current 345: line hasn't been decided yet. 346: Add XF86VidMode extension support. 347: Integrate Stefan's Zorro III support, add support for up to 64MB fast. 348: xwin.c: Refresh the whole Picasso screen if there are Expose events. 349: Fix module length calculation for *Tracker modules. 350: Use a 32 bit address space by default if CPU_LEVEL > 1. New option -5. 351: 971018 Apparently it's not possible to turn on bitplane DMA in the middle of 352: the line, so call post_decide_line only when the number of planes is 353: modified, not in DMACON. 354: genp2c.c: New file, generates planar to chunky conversion routines. 355: Remove those which are now generated from custom.c. 356: 971017 Make row_map an array, not a pointer to one. 357: Audio periods may not be 0, so initialize them to 65535 in customreset. 358: Implement "S" command in the debugger. Add a bunch of module detection 359: routines. 360: 971015 od-linux/sound.c: If FRAME_RATE_HACK, reduce vsynctime a bit to give 361: the sound emulation more room to breathe. 362: 971013 Merge with Brian's latest Picasso code. 363: 971012 Fix a few more portability problems that showed up when compiling with 364: Watcom C. Also remove superfluous includes from some files. 365: Add md-i386-watcom directory. Update od-win32 with latest code. 366: Remove fast_memcmp routine. 367: Include penguin.h before xwin.h everywhere. Declare a few semaphores 368: for multithreaded setups in xwin.h. Move inhibit_frame stuff to xwin.h. 369: 971011 gencpu.c: Don't generate labels without following statement. Don't 370: directly generate assembly statements if X86_ASSEMBLY, instead generate 371: code that uses macros defined in machdep/m68k.h. 372: 971008 xwin.c, svga.c, custom.c, uaelib.c: Add picasso96 code. Add Picasso 373: selection to configure. 374: Move common variables (mouse stuff, gfxvidinfo, xcolors) from graphics 375: files to custom.c. 376: memory.c: custom chip area isn't supposed to be mapped at 0x200000. Use 377: dummy bank for now. 378: od-linux/joystick.c: Open different devices for different joysticks. 379: From Win32 sources: set vsyncmintime in m68k_go. 380: 971005 genblitter.c: blitzero is initialized to 1, not to 0, so we need to 381: clear, not set it. 382: 970928 Add another user interface for X11 using libgtk+. 383: 970925 audio.c: For stereo sound, double all lookup table values. 384: Added (preliminary) win32.c. 385: 970923 DISK_check_change: Should call gui_filename when ejecting. 386: 970922 Fix od-linux/joystick.c compilation problem with recent kernels. 387: Merge results of the UAE meeting: 388: New serial code (Stefan Reinauer) 389: Expansion "test card" support (Stefan Reinauer) 390: New GGI code (Christian Schmitt), also configure/Makefile changes 391: to make it compile. 392: Fix 68020/FPU code. 393: Call do_mousehack before every read of the mouse position to get more 394: accurate values. 395: Add sd-file directory; add support for it in configure. 396: Add stereo support in audio.c, od-linux/sound.[ch], and main.c. 397: Fix copper problem: call decide_line in calc_copcomp_true; and if the 398: copper wouldn't wait normally (position reached), but the hardware is 399: fetching data in Hires 4 plane mode, delay until ddf stop. 400: t-ascii.h: Asciiart target understands "-x" parameter. 401: 970823 gencpu.c [X86_ASSEMBLY]: Turn input constraints for scratch into output 402: constraints. 403: 970626 Fix 24 bit address space logic (Mathias Ortmann). 404: Raise exception 3 for Bcc.L insns if CPU_LEVEL < 1. Reported by Stefan 405: Ropke. 406: Remove USER_PROGRAMS_BEHAVE for normal emulator. Remove hacks to 407: generate expanded CPU functions for specific opcodes. Replace with code 408: that reads an instruction count file and sorts gencpu output by 409: frequency of the called opcodes. 410: 970625 Calltraps (at least those with call_m68k calls in them) were broken by 411: the CPU emulator changes. Now use Axxx opcodes as calltraps. 412: Include pOS port by Samuel Devulder. 413: 970624 Solaris sound patches by Manfred Thole. 414: gencpu.c: The generated code shouldn't just check for odd addresses, 415: it should also call Exception... 416: 970622 gencpu.c: don't emit return statements; make code jump to a function 417: end label instead. Make genamode() check for odd addresses, also reduce 418: code duplication. Interpret getv parameter so that data is fetched if 419: getv == 1, and the address is checked if getv != 0. Remove exception 3 420: checks from get_long() etc. Get rid of buserr. 421: cpuopti.c: Make a few more checks whether optimization is possible. Add 422: code to undo GCC's epilogue scheduling. 423: newcpu.c: Split all the specialflag handling into new function. 424: Implement -w option as in Win32 port. tui.c: add it to misc menu. 425: 907621 Fix yesterday's CPU emulator changes. Make cpuopti aware of functions 426: with two RET insns. Make sure that NO_EXCEPTION_3 and 427: NO_PREFETCH_BUFFER are on when emulating a 68020. 428: Add usage information for src/dst operands to table68k. 429: Implement CIA A OVL bit properly. 430: 970620 New NeXTwin.m from Ian. Fix a number of prefetch bugs in gencpu.c. 431: newcpu.h: m68k_incpc(), get_i{byte,word,long}: New function/macros. 432: Also add oldpc parameter to m68k_do_[bj]sr(). gencpu.c: use all this. 433: CPU functions now don't use nextiword; they keep the PC constant and 434: increment it with the right value at the end. 435: If chipmem_size > 2MB, disable fastmem. 436: In configure.in and od-linux/joystick.c, use ljoystick.h as name for 437: the replacement file for a broken <linux/joystick.h> 438: configure.in: Check for post-gcc-2.7.2 option "-mno-schedule-prologue" 439: and use it for compiling the optimized CPU files if it exists. 440: 970618 Rename hardfile.device to uaehf.device to avoid name clashes. Suggested 441: by George Nicolaidis. 442: Reimplement CPU prefetch by making gencpu aware of it so it generates 443: code that fetches data from the right places automatically. Also reduce 444: prefetch length to 4 bytes. 445: 970612 memory.h: Include osdep/memory.h. missing.c: implement xmalloc(). 446: Throughout: use xmalloc instead of malloc. 447: Apply changes by Sam Devulder. 448: 970608 Fix ncurses.c so that it links. Fix configure so that it can be enabled 449: Get rid of spixstate by implementing a new sprite drawing method. 450: 970607 do_copper_cheat(): New function, Implement inaccurate, line-based 451: copper as an option (-C). calc_copcomp_true(): Fix possible hr 4 plane 452: bug (split the wait into several pieces in this mode). 453: Allow 0 bytes for fast and bogomem (to override .uaerc) 454: 970606 Fix some bugs in transdisk, and make it compile with SAS C (Joanne Dow) 455: Stupid compilation fixes: Get rid of "==" in shell scripts, check that 456: B300 and friends are defined in serial.c before using them. 457: Implement blitter prefetch. Fix silly typo in main.c that prevented 458: immediate blits. 459: init_ham_decoding(): new function, handles case where the first few 460: pixels of a HAM display are outside the window. 461: 970605 disk.c: Get rid of need_read. Save cylinder/track number for the data 462: in the MFM buffer for each drive. Force rereads on disk insert and 463: write operations. In drive_fill_bigbuf(), do nothing if saved data 464: matches current. Fix disk change logic: give the program time to read 465: the change bit a couple of times before inserting a new disk. 466: custom.c: Fix really subtle bug: when deciding a line as border, 467: plfstrt/plflinelen were not set up properly, which could confuse 468: post_decide_line(), and eventually there would be lines decided as 469: playfield with plfstrt == 0, plflinelen == -32 ==> Bang. A related 470: bug was that setting DDFSTRT to 0 would result in both plfstrt and 471: plfstop having value 0. 472: gfxutil.c: Fix bug that could cause color values to go out of bounds. 473: xwin.c, svga.c: When dithering, color values were not doubled properly. 474: 970604 filesys.c: when deleting an a_ino, remove it from hash table (Mathias 475: Ortmann). In do_find(), if creating a file failed and we didn't have 476: an a_ino before, call delete_aino on the one we created. 477: More configure changes: Add $X_PRE_LIBS and $X_EXTRA_LIBS in some 478: places. See if MIT-SHM support for X will compile and link. Always 479: compile it in if it will. Turn DONT_WANT_SHM and LOW_BANDWIDTH into 480: run-time options (-T, -L). Use single line buffer when dithering and 481: in low-bandwidth mode (but in all cases there's a full buffer for the 482: X image). 483: memory.c: Don't map in custom regs at 0x200000 if chipmem_size > 2MB. 484: Split up init_sound(): new function setup_sound(). Make "-B" accept 485: minimum and maximum values as in Gustavo's latest patch. Add code to 486: tui.c to support changing buffer size, number of bits and frequency. 487: 970603 If using compiler, define NO_EXCEPTION_3 and NO_PREFETCH_BUFFER. 488: (bugreport by Raphael Poss). 489: GGI support from Christian Schmitt/Stefan Reinauer. More serial patches 490: from Stefan. 3 button joystick patch from Mathias. 491: Add linemem field to gfxvidinfo, for systems that don't want to keep 492: a buffer of the full screen. Adjust custom.c to use it instead of 493: bufmem if it's not NULL (mostly by Mathias). svga.c, xwin.c: Use single 494: line buffer in some cases. xwin.c: If USE_DGA_EXTENSION, undef 495: LOW_BANDWIDTH. Move test of DONT_WANT_SHM after inclusion of config.h 496: Make configure.in a bit more structured. 497: Move colormodes array to main.c. 498: 970602 gfxlib.c: Start implementing layers workalike. 499: md-i386-gcc/support.c: Don't mix sleep and alarm. (Petter Schau) 500: 970601 Remove may_run_compiled. Print error and abort if m68k_go is called 501: twice. 502: Fix yesterdays bugfix in svga.c. 503: Change FRAME_RATE_HACK so it tries to average the times a bit. 504: Re-implement a call_m68k function in autoconf.c. The new version 505: works by doing some heavy setjmp/longjmp magic and switching stacks. 506: With this, it's even possible to call things that Wait(), so use 507: the new functionality in gfxlib.c to call LockLayer() in WritePixel(). 508: 970531 Turn some inline functions in memory.h into macros. Get rid of 509: get_along and get_aword. Implement CIA A OVL bit. 510: main.c: Add missing close_sound() call. 511: Remove INTEL_FLAG_OPT, use X86_ASSEMBLY for everything. 512: tui.c/svgancui.c: Implement scrollable menus. Add code to select 513: video mode from a given list. Add "graphics_setup" call to graphics 514: specific files. Use it in svga.c to build a video mode menu. 515: Fix bug in svga.c that would crash non-linear modes. 516: disk.c needs to include gensound.h and sounddep/sound.h. 517: Fix bug in fuzzy_memset. In pfield_init_linetoscr, always adjust ddf 518: values if lores hack is being used. 519: 970530 Apply some patches by Samuel Devulder. 520: 970529 filesys.c: In "Houston, we have a problem" case, do not only delete 521: the oldest ExKey, but also use it instead of random memory for the 522: new one. Attempt to reduce the number of lost exkeys by allocating 523: them in action_examine_next instead of action_examine_object. 524: Initialize the "uniq" field of unit->rootnode. 525: Attempt to speed up sprite drawing loop. 526: Implement sprite<->sprite collisions. 527: Get rid of -Wtraditional. 528: 970524 New config option NO_PREFETCH_BUFFER. Emulate 16 byte prefetch if 529: unset. 530: For exception 3, the stack frame seems to be: 531: (saved SR).w, (faulting address).l, (faulting instruction).w, 0.w, 532: (pc at fault time).l. Make everything that could generate this fault 533: handle it properly. That got Cybernoid II working! 534: 970523 Makefile.in: Make sure $(DEBUGFLAGS) is used for CPU emulation files. 535: memory.c: Don't map chipmem at 0x200000, use custom chip bank for now. 536: 970507 configure.in: Mutate ac_cv_c_inline=no to ac_cv_c_inline= 537: Check for tcgetattr, <sys/termios.h>, cfmakeraw, <sys/ioctl.h> 538: Check for sem_init in -lposix4 (Marc Zimmermann) 539: serial.c, cia.c, cia.h, serial.c: changes (Stefan Reinauer) 540: Make disk gap 360 (Michael Krause) 541: 970404 In configure, check for features.h before including it in test program. 542: 970401 New option -4. Bugreport by Mirko Viviani. 543: 970324 Petter Schau says the exception frame for Bus Error and Address Error 544: is 14 bytes, not 6, on the 68000. Change Exception() accordingly. 545: It's *** OUCH *** day! Make hardfile beginio return the correct value. 546: Also load execbase before calling ReplyMsg(). 547: It's *** OUCH *** day again! The filesystem asm code was an odd number 548: of bytes, leaving the hardfile code unaligned. 549: 970318 custom.c: Rename spron to sprst, and use new variable spron to indicate 550: for which sprites the sprite pointer registers have been written. Don't 551: do sprite DMA if spron==0. (The problem was that some programs write 552: SPRxCTL and then SPRxPT, which used to set the sprite state to 553: SPR_restart, but it must be SPR_waiting_start.) 554: Turn a few m68k_reset()s into uae_reset()s. 555: New variable nr_joysticks. Delete joystickpresent. New arg "nr" for 556: getjoystate() and read_joystick(). 557: New extra keycode AK_RCTRL. Handle multiple joysticks & different key 558: replacements in keybuf.c/custom.c. Make tui.c handle all the settings. 559: 970310 od-linux/joystick.c now compiles again without the joystick header. 560: Fix some problems in configure; also check whether [n]curses.h defines 561: attr_t. Keep count of warnings and errors during configuration and 562: print a hopefully helpful message at the end. 563: disk.c sanitized by Mathias Ortmann. 564: Change the replacement getopt around a bit. 565: execlib.c, timerdev.c: Files removed. This will never be compatible 566: enough. Also remove related code (several SPCFLAGs, etc.) 567: Make main.c a bit nicer by introducing additional defines for the t-* 568: files. Also canonicalize the .uaerc file on all systems, eliminating 569: lines starting with '#' in the process. 570: Remove code in autoconf.c to set up 8MB chip; move it to filesys.asm. 571: Try doing the copper end-of-line hack in copper_memory_cycles(). 572: Disable the copper speedup hack. 573: 970309 Add a small, dumb hash table to struct _unit to speed up repeated 574: aino lookups. Delete aino in action_delete_object(). Move children in 575: action_rename_object(). In action_examine_next(), call new function 576: lookup_child_aino_for_exnext() which behaves more or less like 577: lookup_child_aino(), except it uses new function get_aname() to 578: translate filenames. 579: More portability fixes. New links set up in configure: sounddep, 580: target.h. All machine-dependent and joystick code removed 581: from os.c, renamed to audio.c. os.h renamed to audio.h, joystick stuff 582: moved to joystick.h. New directory targets/. 583: missing.c: New file. 584: 970308 Rewrite filesystem. Now it keeps one unique "a_inode" structure for 585: every object in the filesystem. This contains Amiga name, native 586: name, Amiga permissions, lock counts and several links for building 587: a tree structure. Keys are now made only for files which need a fd 588: and a file position. 589: Treat invalid locking modes als shared. 590: Comment out the code using readdir_r because that function appears 591: to be broken in glibc-2.0.1. 592: Clear RES2 in all packets before handling them. 593: Give mousehack the same task priority as the input.device. 594: Fix (?) some silly bugs in disk.c. 595: 970307 Include some Acorn port code from Peter Teichmann. 596: Yet Another Serial Diff from Stefan. 597: 970306 filesys.c: Don't just crash if lookup_key() has no success. 598: action_find_write() is supposed to create the file if it does not 599: exist. 600: 970305 More DOS patches. Make the Makefile parallel-make safe. 601: svgancui.c: Do cyclic search if character is typed in file selector. 602: my_alphasort(): New function (work around glibc bug) 603: custom.c: Fix error in yesterdays change. 604: filesys.h: New file. 605: 970304 New serial patch from Stefan and Christian. 606: Remove the assembler code to generate inputevents. Instead, add a new 607: Amiga C program ("mousehack") which does the trick. New function 608: mousehack_helper() in custom.c, activated by calltrap 0xF0FF70. 609: Mouse default is now always normal_mouse, not follow_mouse. 610: Do disk change checks only every 1/2 second. disk.c: Reverse 611: nodskchange to dskchange, and make sure there has been a step with the 612: drive empty before calling drive_insert() from DISK_check_change() (but 613: this _still_ doesn't work. Arrrghh). 614: 970303 filesys.c: Initialize _all_ of struct _unit in startup_handler(). 615: Remove message_lock and friends and use new variables cmds_sent, 616: cmds_acked and cmds_completed to determine when to send dummy messages. 617: Changed examinekey handling so that "dir all" on a large tree doesn't 618: crash anymore (it used to reuse the top-level exkeys which were still 619: in use after a wraparound). Don't allocate examinekeys if we're 620: examining files. 621: Check for readdir_r in configure. Use a wrapper my_readdir() around 622: either readdir or readdir_r. 623: Call filesys_reset() also from filesys_diagentry(). 624: Get rid of most of uaelib.c. Now there isn't a library, just a calltrap 625: at 0xF0FF60. Changed uae_pragmas.h accordingly and recompiled uaectrl. 626: Added a bunch of "volatile"s in various places in filesys.c and 627: penguin.h. 628: Make the filesystem reset-safe in the threaded case (paranoia changes). 629: Check disk changes only every 1/10 second in vsync_handler(). 630: Add assembler code to generate InputEvents once every vblank to 631: filesys.asm. (Doesn't work yet.) 632: 970302 filesys.c: Get rid of DosPacket type, pass pointers into Amiga memory 633: instead. Use GET_PCK_* and PUT_PCK_* macros to access fields. 634: Start making filesystem code thread safe & multithreaded. Move static 635: variables from some functions into struct _unit, likewise the key 636: and exkey variables. Add code to filesys.asm to queue commands which 637: did not complete and to reply to them later. Do multithreading if 638: compiled with SUPPORT_THREADS. See source for details. 639: Change hardfile and uaelib code not to call 68k routines. Make emulator 640: traps more efficient (no need to save regs, don't need SPCFLAG_EMULTRAP 641: anymore). 642: 970301 Fixed ncurses.c so it works with currprefs. 643: filesys_reset(): New function, called from customreset(). 644: Changed source layout. Break up machdep into machdep, osdep and 645: threaddep. 646: penguin.h: Now supports arbitrary types in pipes. 647: Changed filesys.c so that no 68k code is called from native C code. 648: filesys.asm, filesys.sh: New files. filesys.asm contains some 649: startup code and the main loop for the filesystem. Several locks 650: are allocated by the assembler code for use by the C code. 651: DOS_AllocMem(), DOS_FreeMem(), make_dev(), build_parmpacket(), 652: filesys_init(): Functions removed. 653: 970228 ncurses.c: Fix problem in b/w modes. 654: finish_drawing_frame(): Don't need to draw lines in multipenguin mode. 655: Move all options into new struct uae_prefs. New variables currprefs, 656: changed_prefs. 657: svgancui.c, tui_filereq(): If user presses a key, start search at 658: current position, not at the beginning. 659: DISK_check_change(): new function, compares currprefs with 660: changed_prefs and handles diskchanges if they differ. Called from 661: vsync_handler(). 662: DISK_validate_filename(): new function, checks whether a file exists. 663: Modify svga.c and tui.c so that F11 now brings up a limited UI. 664: 970227 Reimplement Mathias Ortmann's frame rate hack: New option HAVE_RDTSC 665: in config.h. New typedef frame_time_t. New variables vsynctime and 666: vsyncmintime. New function machdep_init() in md-*/support.c which 667: sets up vsynctime in the md-i386-gcc and md-i386-linux cases. Use 668: all of those in events.h, do_cycles() to determine whether to spend 669: more time on the CPU emulation. 670: Add missing "int i;" to xwin.c, init_colors(). 671: Unroll some loops in sample_*handler(). 672: Improve quitting/resetting behaviour: Move quit_program from GUI 673: specific files to main.c. New functions uae_reset() and uae_quit(), 674: quit_program is now handled in vsync_handler() and m68k_go(). 675: New variable sample_evtime, used in customreset() and set up in 676: init_sound(). 677: 970226 Solaris sound update from Manfred Thole. 678: keybuf.c: Decided that it doesn't need locking after all. 679: Add a bunch of Gustavo's latest diffs. (I hate CR characters!) 680: Move gfxvidinfo from all the target-specific files to custom.c 681: More Win32 stuff: New function notice_new_xcolors(). 682: 970225 configure.in: In shell string comparisons, always put spaces around = 683: Add warning about standard SVGAlib being incompatible with 684: LinuxThreads. 685: os.c, SOLARIS_SOUND: Remove reference to obsolete variable smplcnt. 686: Two new variables sound8_base and sound16_base. 687: sysdeps.h: Split off some stuff into md-amiga/amiga-kludges.h. Also 688: make the main types now uae_u32 and variants instead of ULONG, etc. 689: Convert some files to use the new types. 690: cia.c: use int/long instead of ULONG/UWORD/etc for most static vars. 691: xdebug.c: Some cleanup. 692: md-i386-linux/penguin.h: Fix bug. Add "chunks" parameter to 693: init_comm_pipe. Use it to decide in write routine whether to signal 694: the other thread if it's waiting. Add no_buffer parameter to 695: write_comm_pipe_int. 696: Start adding Mathias Ortmann's Win32 patches. 697: md-i386-win32: New directory 698: include/sysconfig.h.win32: New file 699: custom.c: is_lastline: new variable, set in hsync_handler(). 700: sysdeps.h: Make it possible for system specific code to say that 701: POSIX/ANSI stdio/malloc functions aren't present and are defined 702: by a system-specific file instead. In filesys.c, don't reference 703: errno if DONT_HAVE_POSIX is defined. 704: readcpu.h: Kludge around the fact that Messysoft C apparently doesn't 705: handle enums properly. 706: newcpu.c/sysdeps.h: Preprocessor magic for 64 bit constants to get 707: around a problem with MS C. 708: md-*/penguin.h: Add dummy definitions for semaphores. 709: keybuf.c: Use semaphores to synchronize access. 710: os.c, win32 sound stuff: add memcpy avoidance code. 711: 970221 X86.S: Don't use .type ifdef USE_UNDERSCORE. 712: Move thread support configuration to configure from config.h. Change 713: pipe implementation to use POSIX semaphore functions. 714: 970217 From Samuel Devulder: Fix off-by-one error in LINE_TO_SCR, EHB case. 715: 970215 customreset(): Initialize current_colors.color_regs to invalid values 716: [reported by Samuel Devulder]. config.h: Uncomment ECS_DENISE [reported 717: by Mathias Ortmann]. Add ami-win.c and custom.c patch from Samuel 718: Devulder. 719: 970213 disk.c: Set FLOPPY_SPEED to 3 720: custom.c: SPRxCTL_1(): if spron was SPR_restart set it to 721: SPR_waiting_start. do_sprites(): don't do that. 722: SPRxPT{H,L}: Call decide_line(). 723: 970210 custom.c, X86.S: Do LORES_HACK for dual playfield screens. Plug memory 724: leak in custom_reset(). Fix pfield_draw_line() not to do so many 725: multiplications. Make line_changed a char array. In remember_ctable(), 726: remember comparison results also if negative. 727: In 8 and 16 bit graphics modes, use the high parts of the color values 728: to duplicate the actual values. This saves a lot of code in the 729: hdoubling routines. Add a special fill_line in the X86_ASSEMBLY case. 730: Check for glibc-2 and libpthread in configure. 731: options.h: In fuzzy_memset(), do rounding slightly more conservative. 732: Add tons of new keycodes to svga.c 733: In record_color_change(), also check if vertically outside DIW. 734: New variable hdiwstate, to represent the horizontal DIW state machine 735: (thanks Petter!) Handled in reset_decisions(), finish_decisions() and 736: decide_diw(). 737: 970209 Change ncurses.c not to do so many multiplications/divisions. Add bold 738: character support. Add color support. Make configure recognize 739: --without-svgalib and enable ncurses.c in that case. 740: 970208 X86.S, custom.c: Do LORES_HACK for EHB screens also. 741: New function write_log(); convert some fprintf(stderr, ...) calls. 742: 970207 Argh, Linux/i386/ELF check in configure got broken by latest sh-utils. 743: In disk.c, drive_reset(), clear nodskchange flag. 744: 970206 Sprite fix from Mathias Ortmann: _really_ get rid of SPR_vtop, and 745: replace all uses by SPR_stop. 1.1.1.2 root 746: 970204 Disk change fix from Mathias Ortmann. 747: 970203 New file ncurses.c, from Samuel Devulder. 748: Disabled Exec stuff completely. Now it isn't even compiled in anymore. 749: 970129 Changes from Samuel Devulder for Amiga and Christian Bauer for BeBox. 750: 970122 Reorder the set_planes_unaligned functions so that better code is 751: generated. Remove the or_planes_unaligned functions, add a parameter 752: do_or to the set_planes functions and rely on the compiler to optimize 753: the additional if away. 754: post_decide_line() can now turn DMA off early. 755: Make struct flag_struct 64 bits large for i386 systems, so we can use 756: long move insns without clobbering the X flag. 757: "make clean" should delete machdep/support.o. 758: Maybe we do need SPR_vtop after all - undo 970120 change. Instead, 759: comment out the code that clears the pos and ctl words at vsync time. 760: In xwin.c, restrict window height to 300 or 600 depending on linedbl 761: mode. 762: 970121 Get rid of SPCFLAG_BLIT, call do_blitter() directly. 763: In finish_decisions(), don't check for sprite draws in the border. 764: Reimplement the SMART_UPDATE define so that the memcmpy() overhead can 765: be avoided by not setting it (but this usually loses). 766: Detect 7 lores planes in BPLCON0() and change to 6 to avoid memory 767: clobbers. 768: In pfield_draw_line(), do memcpy linedoubling after handling the line 769: to be doubled, not when we see a LINE_AS_PREVIOUS. Get rid of prev_ypos 770: parameter. 771: Add special code for bpldelay1 == bpldelay2 to the UNALIGNED_PROFITABLE 772: code. 773: In pfield_init_linetoscr(), when delays 1 and 2 differ, clear the areas 774: which are covered only by half of the planes at the start and end of 775: the DDF area. 776: Replace FASTER_CPU and LINUX_SOUND_SLOW_MACHINE with M68K_SPEED. 777: Replace fast_unaligned_memset macros with fuzzy_memset; use this also 778: in pfield_init_linetoscr(). 779: Use real (PAL) vblank end in do_sprites(). 780: 970120 Mask out high byte in DDF{STRT,STOP} 781: Oops, forgot about long frames while calculating the bounds of all the 782: maxvpos*2-sized arrays. Increment them all by two. 783: Try to detect empty sprites before recording them for the drawing code, 784: but be careful not to hurt attached sprites. 785: From Paul Kienitz: hack for keyboard code in cia.c to work better with 786: newer Kickstarts; small cleanup in cpuopti.c. 787: Undo yesterday's change in copper_memory_cycles(), instead call 788: decide_line() each iteration and test whether copper_cycle_time == -1 789: to determine whether we're vertically in the DIW. 790: calculate_copper_cycle_time() could get astronomical plane counts due 791: to missing masking operation. 792: No need to put the copper to sleep for one memory cycle after a WAIT, 793: merge that with the two cycles for the next instruction. 794: Get rid of SPR_vtop, no longer needed (it was unnatural anyway) 795: 970119 Implemented alternate pfield_doline functions that access the apixels 796: array at unaligned addresses; conditional on UNALIGNED_PROFITABLE. 797: Move LORES_HACK from config.h to custom.c. 798: Fix loop exit test for pfield_linetoscr_hdouble_asm8(). Clear the 799: right parts of apixels in pfield_init_linetoscr() when LORES_HACK is 800: enabled. 801: Get rid of bordercol array: the decision-making code should handle 802: lines in the border correctly by setting line_changed. 803: Disable bitplane DMA if more than 4 hires bitplanes are set in BPLCON0. 804: Fix some more copper timing inaccuracies: used to test 805: thisline_decision.which without calling decide_line() before. 806: 970118 Now remember BPLCON1 changes during data fetch. New functions 807: decide_delay(), pfield_adjust_delay(); new fields in struct drawinfo 808: and new variables next_delay_change, delay_changes. 809: Forgot to set copstate to COP_morewait in one case in do_copper(). 810: Add a kludge that checks for copper position 0xDC and changes it to 811: 0xDE. 812: Restore equality test for vertical sprite positions. Disable sprite 813: DMA before the end of the vertical blank area. 814: Mask not only the positions, but also comparison values in 815: calc_copcomp_true(). 816: 970117 Change custom_bput to put the byte value in each half of the word that 817: is put. 818: 970116 disk.c: Set the gap between sector 11 and 0 to 380 words. Andromeda's 819: "Seven Seas" demo expects at least 320. Make the KILLEHB bit 820: conditional on AGA_CHIPSET. Decide BPLCON1 (scrolling delays) in 821: decide_diw(). Fix off-by-one errors in the very_broken_program case 822: of decide_plane(). Fix aspect calculation when horizontal resolutions 823: differ. Simplify copper horizontal comparison logic slightly. 824: Fix divide by zero errors when reading from empty floppies. 825: 970115 pfield_draw_line(): Check for frame_redraw_necessary also for border 826: lines. New global variable pfield_last_drawn_line, used to implement 827: clever Y centering. Make the KILLEHB test conditional on ECS_AGNUS. 828: Move joystick and sound initialization before gui startup. Adapt tui.c 829: so it understands the gfx_requested_* options. 830: We can't always avoid clearing pixdata.apixels if DIW is larger than 831: DDF. There may be sprites in the area outside DDF, but inside the DIW, 832: so we can't just adjust linetoscr_diw_{start,end} blindly in 833: pfield_init_linetoscr() [Sanity Arte demo]. 834: 970114 Bebox diffs from Christian Bauer. 835: Support for Symmetric Multi Penguins under Linux (start second thread 836: with clone and use some pipes for communication - rather simpleminded 837: for now). Rework do_flush_line() and do_flush_screen() so that X 838: libraries are only called from the first thread. 839: Eliminate valid field from struct decision (set which to -2 instead of 840: valid = 0) 841: Fix definition for gui_filename() in xdebug.c to use const char * for 842: the second parameter. Remove CRs from disk.c (now how did they get in 843: there??). Likewise for cpuopti.c. 844: 970113 When post_decide_line() detects mid-line DDF change, must adjust 845: bitplane pointers in decide_plane() also. Make plfstrt the earlier a 846: position in this case, and adjust bitplane pointers/modulos 847: accordingly. 848: current_hpos() isn't what GCC calls a constant function. 849: Replace line_drawinfo[current_change_set] with new variable 850: curr_drawinfo throughout. Likewise for new variables prev_drawinfo, 851: {curr,prev}_{color_{change,table}s,sprite_positions}; all initialized 852: in init_drawing_frame (). Turn line_data into a 2dim. array, do some 853: CSE/SR on line_data[lineno] in pfield_doline_{l,h}() and 854: decide_plane(). Make some custom registers unsigned int instead of 855: UWORD. 856: calc_copcomp_true(): Masking with 0xF000 isn't likely to give 0xC0000. 857: Comment out bplcon{3,4} from struct decision. 858: 970112 filesys.c, action_set_protect(): if __unix is defined, check whether 859: user wants to change directory protection, and if so, force rwx to be 860: the new protection. 861: Force sane window sizes in xwin.c, and set width/height from mode specs 862: in svga.c. 863: Implement LORES_HACK compile-time option. 864: Whee, there _are_ broken programs that enable bitplane DMA in the 865: middle of the data fetch area, draw half a picture on the first line 866: and the full picture on all subsequent lines. New function 867: post_decide_line(), called from BPLCON0() and DMACON() which tries to 868: detect this case and turns the decision from border to playfield if 869: necessary. 870: New arrays, sprvstart and sprvstop, calculated in SPRxPOS and SPRxCTL, 871: and used by do_sprites. Handle sprites with vstart == vstop correctly. 872: Sprite vpos comparisons are now done with <= , not ==. 873: _Now_ the X86-specific linetoscr routines do their masking correctly. 874: configure: If we have GCC, #define GCCCONSTFUNC as __attribute__ 875: ((const)) 876: 970111 Eliminate can_draw from struct decision. 877: Change vertical DIW code; use enum of two states (waiting for start and 878: for stop) which is updated in decide_line(). Lines during the wait for 879: stop phase are decided as border. No data is fetched outside the 880: vertical DIW. 881: Optimize copper memory cycle calculation a little: the new variable 882: copper_cycle_time is set to a time >= 0 if this time is constant 883: (i.e. 2). Calculated whenever DMA, BPLCON0, or DIW state changes. 884: blitter.c, genblitter.c, custom.c: For BLTBDAT, follow the HRM which 885: says that immediate data is shifted as soon as it is written. This 886: seems to be wrong for the A channel, though. Also, don't change prevb 887: during blitter loops if DMA B is off (Sanity Arte). Turn FAST_BLITTER 888: stuff into run-time option "-n", and eliminate the ancient blitting 889: code (former FAST_BLITTER == 0). 890: Big OOPS: Copper move checked copcon, but there was no COPCON function 891: so it was always zero -> copper-controlled blitter never worked. As a 892: speed optimization, recognize 00010000 copper commands and put the 893: copper to sleep until the blit is finished. 894: Slow down the blitter in decide_line() if bitplane DMA steals blitter/ 895: copper/CPU cycles (but this code is commented out for now). 896: 970110 New function pfield_init_linetoscr(). Move calculation of diw_end and 897: diw_start values there from x86-specific code. Also use DDF values to 898: determine where data in pixdata.apixels is valid. The x86-specific 899: linetoscr routines now only do the alignment masking (and they now do 900: it correctly, too). 901: Got rid of the display-related contents of struct draw_info, make 902: static variables and them initialize from pfield_draw_line() only when 903: necessary in function pfield_expand_dp_bplcon (). Get rid of the four 904: saved_* arrays, remember line_drawinfo instead now that it's smaller. 905: Color compare avoidance didn't work because color_src_match didn't get 906: set when a compare was successful. Fixed. 907: Reorder parts of the decision-making code so that less redundant checks 908: are performed. 909: Add clever x centering. Change "-O" option parameter format. 910: In md-i386-linux/support.c, memcmpy(), do memcmp() first, then switch 911: to a plain memcpy routine once we find the first difference. 912: Patches from Stefan Reinauer/Christian Schmitt for 913: xdebug.c/configure.in 914: New uae-ui.tk from Volker Zota. 915: Add --enable-profiling to configure.in. 916: Check for period==0 in AUDxPER() was accidentally commented out - 917: revert. 918: 970109 Some tweaks in X86.S, plus a new option X86_PPRO_OPT to use movzbl 919: instead of xorl/movb when possible. 920: Fix maxline calculation in finish_drawing_frame(). 921: Make next_lineno always a valid array index. Decrease minfirstline to 922: 21 (NTSC vblank end). New option -O to specify display parameters, 923: supersedes -d and -C which will be removed eventually. Differentiate 924: between line doubling and aspect correction. Put aspect correction code 925: in custom.c. 926: Remove some more unused stuff from custom.c. 927: Fix Makefile problems when cpuopti isn't used. 928: 970108 use_lores isn't initialized before custom_init(), setup lores_frame and 929: friends in customreset(). Optimize spixstate clearing. Optimize border 930: drawing with copper color changes (use fill_line instead of linetoscr) 931: Fix the sprite DMA logic, it now works with the Interference demo. 932: Silly bug: Sprite 7 never got displayed (introduced sometime last week) 933: Implement USER_PROGRAMS_BEHAVE for normal (non-compiler) CPU emulation. 934: Add most of Gustavo's last patch (new dir md-i386-dos) 935: 970107 Finish rewriting most of the drawing code in custom.c/X86.S. See 936: README.PROGRAMMERS and source for details. Improvements include better 937: copper precision, better interlace support, hopefully much less bugs 938: and not too many new ones. 939: Remove calc_adjustment() and gfxvidinfo.x_adjust; replace with code in 940: custom.c that looks at gfxvidinfo.follow_diwstop (new field) 941: Some operations (like EHB playfields) now have a higher penalty than 942: before, the assumption being that these are rare and the normal case 943: runs faster if it doesn't have to care too much about exceptions. 944: Fix sign mismatch bug in BPL1MOD() and BPL2MOD(). 945: 970104 Move x_ShowEA() from xdebug.c in newcpu.c, rename as ShowEA(). 946: 970102 Add check for XK_W to decode_es() in xwin.c. 947: New uae-ui.tk from Volker Zota. 948: 961218 Add a patch by Bernie Solomon to overcome a problem with K&R compilers. 949: Change X86.S in a way that may make it either faster or slower. 950: Fix ADF_EXT1. 951: 961217 #ifdef XK_dead_*, since Solaris doesn't have it. 952: 961202 Fixed joystick.h problem in configure. X11/DGA support from Kai 953: Kollmorgen. 954: 961117 Some fixes in svgancui.c. 955: 961112 Gustavo's latest patch, with GUS support for DOS by Michael Sontheimer. 956: Add menu browsing to tui.c/svgancui.c 957: 961107 Fix m68k_areg and m68k_dreg macros (put args in parentheses) 958: Initialize gfxvidinfo.maxlinetoscr in xwin.c (found by Sam Devulder) 959: 961104 svgancui.c tui_refresh() routine tweaked to work with the non-standard 960: ncurses that came with my new SuSE-Linux CD. 961: 961031 Added Christian Schmitt's xdebug.c (disabled for this release) 962: 961024 Module search function from Michael Krause. 963: 961019 Add Samuel Devulder's Amiga patches. 964: 961018 Don't use -lm on NeXTStep. 965: 961017 configure checks for -lforms. 966: 961016 Ian Stephenson's new NeXTwin.m. Make CLXDAT return 0 (no collisions) 967: transdisk and transrom patches from Rodney Hester. Don't warn about 968: 256K Kickstarts. Check diskfile sizes and try to determine whether it's 969: a normal or a HD diskfile. No HD drive identification yet (is it 970: different?) 971: 961013 Corrected lots of bugs in the return address stack. Fix some errors in 972: the compiler memory management. Use the "locked" flag for hash entries 973: differently, and don't enforce it on a cache flush. Compile JMP to a 974: constant address. 975: 961012 Remove regs from cache before sign extending them for ADDA, MOVEA and 976: similar things. Compile NOPs to nothing. Implement variable shifts/ 977: rotates. Most shifts clear the V flag: Handle this in the compiler by 978: some extra magic sequences after shifts or in the CC_AFTER_RO case in 979: flush_cc_cache(). 980: Implement a stack containing return addresses and their hash entries, 981: and compile JSR (to constant addresses)/BSR/RTS instructions inline. 982: 961011 compiler.c: Several bugs introduced in the last two days fixed. 983: New function compile_note_modify() which is used when an ea is 984: modified. This replaces calls to storeea with eaino_s == eaino_d. 985: Bah, the 386 doesn't set flags on a NOT. Use CC_TEST_REG again. 986: Do flag life analysis until no more changes are found. 987: Make sure all "data" parts of EAs are locked exclusively (esp. for 988: LEA, PEA) 989: 961010 compiler.c: Finish yesterdays changes. Get rid of reg/offset 990: bookkeeping in m68k_compile_block(), keep the information in struct 991: ea_info instead. Rewrite all instructions to use the new fetch and 992: store routines. 993: Don't force a byteorder conversion in sync_reg() if the reg doesn't need 994: to be written back. Don't do a SAHF or equivalent in compile_condbranch 995: if we are doing a BRA or BF. 996: Don't use CC_TEST_REG for NEG and NOT (this was inefficient for NOT and 997: wrong for NEG). 998: 961009 compiler.c: Two types of locks for registers; shared and exclusive. 999: Implement better EA fetching and storing routines. For pdi addressing 1000: modes, lock the necessary registers shared before calling prepareea. 1001: Unlock those registers again in generate_possible_exit(). 1002: 961008 MOVEM instructions in the compiler to an absolute address could have 1003: caused random memory corruption. 1004: 961007 Spanish keyboard support for X by Ruben Martinez. Some configure fixes 1005: (don't assume current dir is in path, don't assume ncurses.h exists if 1006: libncurses does). 1007: If we're calling 68k code from an emulator function, don't use compiled 1008: code. 1009: Add a compilation option USER_PROGRAMS_BEHAVE (for compiler). Make 1010: compiler hash table larger. 1011: 961004 Fix typo in configure.in. 1012: Modify the compiler so it doesn't ever throw away code, unless it's 1013: necessary for a cache flush. 1014: Didn't quite squash the compiler flag bug yesterday. Try again. 1015: In filesys.c, do_find(), add a parameter fallback that indicates that 1016: the open should be retried with O_RDONLY if it failed with O_RDWR. 1017: Use this in find_input(). This fixes problems with r/o files on r/w 1018: volumes. 1019: 961003 Play around with the x86 specific memory access functions. Problem: 1020: __attribute__((regparam)) isn't reliable, and it seems that that's a 1021: rather hard to fix bug in GCC. Add some alternatives to 1022: md-i386-linux/memory.h; these aren't used yet though. 1023: Huh, the compiler was completely wrong about the flag needs of 1024: conditional instructions. I wonder how it could work at all... 1025: Remove a and d arrays from struct regstruct; replace with array regs 1026: with 16 elements. This allows better code for get_disp_ea() 1027: 961002 BeBox diffs from Christian Bauer. Move all code from main() to new 1028: function real_main(). 1029: 961001 Use do_get_mem_*() functions for nextiword(), nextilong() and new 1030: function nextibyte(). 1031: 960930 genblitter optimizations: Don't generate expressions in genblitter, use 1032: a second program called gengenblitter to generate a table of 1033: expressions for genblitter to use. gengenblitter knows about a lot of 1034: rules to simplify expressions. Some other tweaks to make GCC output 1035: better assembly for blitfunc.c. 1036: 960927 Turn off LED() in xwin.c (may fix SunOS joystick emulation problems) 1037: Call XFlush() and XSync() when exiting. 1038: Add some DOS patches. 1039: Add hooks in filesys.c to support filesystem units in a GUI. Use them 1040: in tui.c. Bugfixes in svgancui.c. Change the way tui_refresh() works. 1041: New option "-A". 1042: 960925 For read-only filesystems (CD...), use O_RDONLY in action_find_input. 1043: 960923 Save options file even if rename fails. 1044: Mirror address space every 16MB like we used to (got turned off by 1045: accident). 1046: Move some checks out of options.h into configure. Set up md-* 1047: directories with machine-dependent stuff in them; configure creates 1048: a link called machdep to the appropriate one. 1049: 960918 Solaris sound update from Manfred Thole. 1050: 960917 Add a timer.device replacement (buggy). 1051: 960916 Tristan Cavelier's french keyboard patch. 1052: Apply ReadTheSpecsBeforeYouWrite(tm) technology to AddPort(). 1053: Implement RemPort(), GetCC(). 1054: 960915 Tighten up some of the return values in Exec. 1055: 960914 EXEC: Implement RemTask(), ObtainSemaphoreList() and 1056: ReleaseSemaphoreList(). Set up more fields of the ExecBase properly 1057: and Enqueue it in the library list. Make InitResident call AddLibrary 1058: when appropriate (otherwise, intuition is never put into the library 1059: list). Fix major braino in Obtain/ReleaseSemaphore() (tried to Enqueue 1060: tasks into two lists). Make OpenLibrary() honor the version parameter. 1061: 960913 EXEC: Idle task now runs in user mode (abuse 0xFF0D as second STOP 1062: instruction). Make EXEC_IntTrap perform the RTE before trying to 1063: schedule. 1064: 960912 Solaris sound didn't compile. So much for making blind changes without 1065: being able to test them. 1066: 960911 As soon as you compile it under Solaris, it bombs... Now check that the 1067: hardfile exists before determining the size in main.c. 1068: execlib fixes: Set master DMA enable bit at startup, fix interrupt 1069: code to survive an interrupt that occurs before the proper SetIntVector 1070: call. Implement RawDoFmt(), CloseDevice(), CloseLibrary(), 1071: FindResident(). 1.1 root 1072: 960905 Lower 8 bits of POTGOR should be 0. 1073: 960904 Implement MULU, MULS, ADDX and SUBX in compiler. 1074: 960903 ncurses interface works. DOS people: please use the new portable tui.c 1075: and turn dos-ui.c into a frontend for conio.h, similar to svgancui.c 1076: In xui.c, unlink the temporary files ASAP. 1077: Two new command-line options, "-o" and "-G". 1078: Get rid of sz_unknown. 1079: compiler.h, compiler.c: New files. 1080: filesys.c, action_read(): Added a hack to test whether the first four 1081: bytes read from a file a file contain 0x000003F3 (HUNK_HEADER), in 1082: which case there's probably a LoadSeg() call going on and we ought to 1083: flush the cache. Also, use the fact that memory is now linear. 1084: 960902 Struggle a bit with dos-ui.c, rename it to tui.c (text-based ui), add 1085: new file svgancui.c (ncurses ui) and try to make these two understand 1086: each other. 1087: 960829 New transdisk from Stefan Ropke (can write images) 1088: Apply Patrick Hanevolds BeBox patches. 1089: 96082x many changes to Exec. Still not booting (DMACON seems not to be set up 1090: properly, and sometimes Signal is called with task==0). 1091: 960821 Solaris sound from Manfred Thole. 1092: 960820 In filesys.c, get_fileinfo(), return volume name for top-level 1093: directory. 1094: 960819 Reimplement WritePixel(). Layer locking is still missing. 1095: 960818 Reverse meaning of -x flag for X11. More Exec code. 1096: Several changes to the main CPU loop and the way call traps work. 1097: 960817 Implement some simple Exec functions (Lists, AllocMem & friends) 1098: 960816 filesys.c fixes from David Varley. 1099: 960815 Implement graphics library region functions. 1100: 960814 Modify memory layout: All memory is linear now; get_real_address() 1101: returns a (UBYTE *). 1102: 960810 Add Gustavo's latest DOS code. 1103: In configure, include <sys/param.h> earlier for the sake of some *BSD. 1104: Add three options for sound code (bits, frequency, buffer size). 1105: Make -S option take a parameter; remove SPEED_UP_SIDMON_MODULES. 1106: New version of transrom (return 0, duh). 1107: 960805 Fixed the blitter line mode emulation (this time, really! I hope.) 1108: Thanks to Fabio Ciucci for a detailed description of the correct 1109: behaviour. 1110: 960730 Spent the last three days trying to make the filesystem emulation 1111: bootable by adding an expansion device for it. A fair bit of trial and 1112: error, but now it works! Even the hardfile now works with Kick 1.3, but 1113: you have to mount it manually after booting. 1114: hardfile.c, filesys.c, expansion.c and autoconf.c now are an 1115: entangled mess which badly needs to be cleaned up. 1116: We can now mount more than one filesystem even with Kick 1.3: we 1117: recognize them by param 2 of the startup packet, which has the same 1118: value as the dn_Startup field of the devicenode. 1119: Make Kickstart replacement work again (didn't like the CPU emulator 1120: changes) 1121: Include preliminary serial support from Stefan Reinauer. 1122: 960728 Fixed the scancodes for <> and #' keys in svga.c 1123: Apparently, CIA timer latches ought to be 0xFFFF after a reset 1124: (prevents Eagleplayer from crashing occasionally). #ifdeffed out a 1.1.1.2 root 1125: slightly weird call to INTREQ from RethinkICRB() 1.1 root 1126: 960727 Sound code rewritten again. Now uses events (five new ones). All 1127: modules seem to play correctly now. It's a bit slower, though. 1128: AF sound should work again, patches from Marcus Sundberg. 1129: 960723 ... and more CPU emulation bugs fixed by Herman. 1130: Added uae.library from Tauno Taipaleenmaki, plus support files in the 1131: "amiga" directory. 1132: Added some files for the XFree86/OS2 port from Krister Bergman. 1133: Reorganized all the directories. 1134: 960717 Add latest patch from Herman. 1135: More FAST_BLITTER options: 4 and 5. These finish the blit immediately. 1136: 960713 Call XAutoRepeatXXX() only if we know we want to change the state. 1137: 960709 Work around a few AIX bugs (ughhh... what a broken OS). New test for 1138: sys/select.h. 1139: In xui.c, don't try to communicate with a dead process. 1140: 960704 More bugfixes from Herman. 1141: Get rid of dummy_xlate(), use default_xlate() to abort with error. 1142: In svga.c, set correct_aspect to 0 if using lores. 1143: Change the Bcc exception 3 handling code a bit. 1144: Add new program "cpuopti" that takes gcc-generated assembly files and 1145: optimizes away unnecessary pushl/popl instructions on the x86. 1146: 960701 68020 exception handling & various bugfixes from Herman ten Brugge. 1147: In ersatz.c, use 0xFF0D instead of 0xF00D as trap opcode. 1148: Do MMU "emulation" (no-op MMU instructions, probably many cases decoded 1149: incorrectly). 1150: 960627 New file LIESMICH, with minimal German instructions. Somebody want to 1151: translate the docs to French, Spanish, Italian, all the Scandinavian 1152: languages, Hebrew, Polish, Dutch, Russian, Welsh, Gaelic and something 1153: [email protected] understands? (No offense meant - I'm sure there are 1154: brillant minds using AOL. There must be.) 1155: 960626 In zfile.c, make sure we delete the temporary file even if file 1156: decompression fails. 1157: Reimplemented sound support, this time following the HRM closely. Seems 1158: to work fine most of the time, but occasionally loses drum beats in 1159: TFMX modules. do_sound() is now unconditional, it calls 1160: flush_sound_buffer(), which is system-dependent. This breaks everything 1161: except Linux sound for now :( 1162: Add Herman ten Brugge's 68020/68881 emulation code. 1163: FreeBSD chokes if sys/mount.h is included without sys/types.h. Fix the 1164: configure script. 1165: 960622 More work on AGA code. 1166: Removed 2-bit display routines from custom.c. Ian: please use dithering 1167: code. 1168: Make slowmem and chipmem sizes run-time configurable. 1169: Mask out the high bits in BLT?PTH(). 1170: Add exception 3 checks to Bcc/BSR instructions. Seems to be not quite 1171: correct yet. 1172: Completely rewrote the disk emulation. We now create a big track buffer 1173: full of MFM data, used in a cyclic fashion by DSKBYTR() and friends. 1174: The current position in the buffer is determined by the cycle counter. 1175: DMA reads are now done in one piece. Two new event types: ev_diskblk 1176: and ev_diskindex. 1177: ARRGGHH! The RSI demo has _completely_ broken disk loading routines. 1178: They start disk DMA on a block while the blitter is still working on 1179: the same memory region. 1180: In zfile.c, unlink() the temporary file directly after opening it so 1181: it will be freed by the OS as soon as we close it (or crash... :) 1182: 960621 Merge pfield_sprite_h() and pfield_sprite_l() into one function. 1183: Split off new files blitter.c and blitter.h from custom.c. 1184: Fixed null-pointer dereference in zfile.c when filename has no 1185: extension. 1186: Add new field maxline to gfxvidinfo that prevents writing past the 1187: screen buffer if it has less lines than expected. 1188: Modify the 256 color allocation routine to set nb_cols to the same 1189: value first for r,g and b; allocate everything as usual and fill the 1190: lost colors afterwards with those that have the greatest difference. 1191: 960620 For some extremely weird reason, it is faster to simply call the 1192: assembly-optimized pfield_linetoscr_full routines _twice_ to double the 1193: line than calling the new (also assembly-optimized) 1194: pfield_linetoscr_full_double routines, at least when writing directly 1195: into the framebuffer. If someone has a good theory why, let me know 1196: about it. 1197: Drawing into memory is faster with the _double routines, but not 1198: noticeably (about 1%). 1199: Consequence: Removed the new pfield_linetoscr_full_double assembly 1200: routines again. 1201: Add a "--disable-gui" option to configure. 1202: Add Samuel Devulder's dithering code. Provide support for mono and 1203: 16 color X servers. 1204: Put an assembly version of DitherLine() in X86.S. 1205: New color mode (3) available, tells UAE to use 256 colors and dither 1206: the image. This is meant for people who have only a 256 color X server 1207: available, but it works with the SVGA version, too. It's much slower 1208: than normal 256 color mode, though. 1209: Use the color allocation algorithm from the dither code in 1210: alloc_colors256(), too. 1211: 960619 New files: zfile.c and zfile.h, from Samuel Devulder. These provide 1212: transparent decompression of various compressed file formats. Enabled 1213: by configure only for Unix systems. 1214: Eliminate dont_want_aspect. New option "-C", sets correct_aspect. Make 1215: video mode 3 640x480, video mode 4 800x600. Honor correct_aspect in 1216: 320x400, 640x480 and 800x600 mode with SVGAlib. 1217: Add fields maxlinetoscr and x_adjust to gfxvidinfo. 1218: Rename pfield_slow_linetoscr() to pfield_do_linetoscr(). Use new fields 1219: in gfxvidinfo to enable writing directly to the framebuffer in 1220: resolutions with <800 horizontal pixels. 1221: New function pfield_do_linetoscr_full(), called when sure that full 1222: line update is being done. 1223: Move some of the usage() information into some target_specific_usage() 1224: functions, defined in the various system-specific files. 1225: Support lower resolutions with LOW_BANDWIDTH, too. 1226: memcpy() within screen memory is dog slow. Prevent some common cases: 1227: 1. Use fill_line() to double a line in the border. 1228: 2. Add a parameter to pfield_do_linetoscr_full() to indicate whether 1229: the line has to be doubled. New functions 1230: pfield_linetoscr_*_double_slow() that simply do a memcpy(), plus 1231: optimized ones in X86.S that are much faster by avoiding the copy. 1232: Smart update code could get confused by lores modes when 1233: prev_max_diwstop changed from one frame to the next. New variable 1234: frame_redraw_necessary, which acts as a counter. It's initialized to 1235: 2 in interlace mode to handle that case properly, too. 1236: Move sprite priority checks into pfield_sprite_?(), move HAM 1237: calculations into new function decode_ham6() to simplify the various 1238: linetoscr functions. 1239: Fixed a bug that would cause sprites to get deformed in some cases 1240: when they overlapped (never seen it). 1241: 960618 New file: dos-snd.c, from Gustavo. 1242: 1243: 960617 Release 0.6.0 1244: 960617 Fix memory clobber in X86.S when diwfirstword==0. Eliminate double test 1245: for HAVE_SYS_MOUNT_H in sysdeps.h. 1246: 960616 Handle programs that don't wait for the blitter by finishing a pending 1247: blit as soon as a blitter register is written to (regardless of DMACON, 1248: to be safe). 1249: Call gui_led only if the LED changed. 1250: 960615 Fixed dual playfield sprite priorities. Optimized dual playfield code 1251: in X86.S. More work on AGA code. Get joystick status only in 1252: vsync_handler to speed things up. 1253: Make blitter fill more intelligent with table lookups instead of loop. 1254: 960614 DOS update from Gustavo. Pass real bitplane pointer to 1255: bpl_data_differs(), don't calculate it again. 1256: 960610 Fixed silly INTEL_FLAG_OPT bug: The new __asm__ expressions use 1257: regflags, which is of course _regflags on a.out... 1258: More BeBox fixes from Christian. 1259: Make the CIA bank bigger (A0..BF) and relax the CIA A/B selection 1260: checks to make broken software (Kingsoft's Emerald Mine, reportedly) 1261: happy. 1262: Default framerate is now 1. 1263: 960603 New BeBox code from Christian. Fix test for ELF in configure.in. 1264: 960529 genblitter.c: generate blitfunc as a function taking ULONGs and 1265: returning an ULONG. This makes custom.o about 4500 bytes smaller on the 1266: x86 by removing worthless size prefixes, and should increase 1267: performance (better for the cache, and the P6 reportedly doesn't like 1268: 16 bit instructions anyway). 1269: memory.c: Load Kickstart in one block and swap bytes after that if 1270: necessary (eliminates the need for the special bebox load routine) 1271: custom.c: Some preliminary work for AGA support. 1272: 960528 Saw a real awful piece of code that detects the right mouse button by 1273: reading a value from POT0DAT, counting up to 25, reading from POT0DAT 1274: again, and deciding the button is pressed if the value changed. Urgl! 1275: Added POT0DAT() in custom.c ... 1276: Gustavo's latest patches. 1277: 960525 Do index pulse handling in CIA_hsync_handler(). Make sure bit 12 in 1278: INTREQ gets set on an index pulse. Don't call DISK_Index() for now, 1279: since this breaks the Workbench. 1280: The CIA code should _probably_ not be resetting the bits in INTREQ. 1281: Disable this as an experiment, let's see what breaks. 1282: 960524 Z flag handling for ABCD/SBCD/ADDX/SUBX was wrong (must be clear if not 1283: zero, was set if zero). Some other flag bugs in these functions fixed. 1284: 960523 gfxlib.c update from Christian Schmitt. Some new Mac code from Ernesto, 1285: still not all of it, though. Italian keyboard from Alessandro Bissacco. 1286: 960522 Fix the CHK instruction. Implement middle mouse button. 1287: 960515 Fix some Nextstep problems in configure.in. 1288: 960513 New Makefile target "streifenfrei" to remove all output from configure. 1289: Use autoconf 2.7; 2.9 is buggy. 1290: 960512 Remove BIG_ALIGNMENT hack. Remove superfluous pfield_fetchdata() call 1291: in pfield_doline_slow_l() that resulted in garbage on the screen. 1292: Move blttbl array from custom.h to genblitter.c 1293: Use new lookup tables imm8_table and areg_byteinc in CPU emulation 1294: instead of ?: expressions. 1295: GCC appears to generate better code if the opcode is passed as ULONG to 1296: the CPU emulation functions instead of as UWORD. 1297: Modify some of the flag emulation __asm__ statements so that GCC uses 1298: less registers for them. 1299: 960510 Cast malloc() results to char *, for C++ compilers. 1300: 960509 Use test for GCC and __i386__ in options.h to decide whether to use 1301: intel flag optimization. 1302: Use test for wish4.0 to decide whether to use the GUI. 1303: 960507 New bebox.cpp from Christian. AF sound patches from Marcus Sundberg. 1304: 960505 Fix the configuration stuff so it works. Fix some bugs in gfxlib.c 1305: 960504 Use GNU autoconf scripts to generate Makefile and sysconfig.h. Add new 1306: files sysdeps.h that includes the right system header files. Remove 1307: amiga.h and Makefile. 1308: 960502 More Next code from Ian (new Uae.app directory, 2 bit drawing for 1309: custom.c). 1310: 960501 Allow the sound driver to set the frequency to a value that is slightly 1311: different from the requested one. 1312: 960430 Added AmigaOS port from Olaf Barthel. 1313: 960427 Fixed single playfield sprite priorities in X86.S. 1314: 960426 Added new NeXTwin.m from Ian. 1315: Turn off the blitter delay in line mode. 1316: Added a CPU optimization from Olivier Raoul: gencpu now reads a file 1317: "insncount" that can be generated by UAE during execution and generates 1318: specific functions for the most common opcodes. Not documented yet. 1319: 960414 Reimplemented the fast disk emulation in a more sane way. Also it's 1320: only done for reads now. (Plus, it no longer works :-/ Need to work 1321: some more on this). 1322: In blitter emulation, defer not only the setting of the bltfinish bit, 1323: but actually defer the whole blit for the number of cycles it will 1324: take. This makes programs the clear the screen while it is still being 1325: displayed look much better. 1326: Changed supervisor mode handling around a bit (was messy). 1327: New DOS code from Gustavo. 1328: 960412 Removed smart update method 2. Added Alessandro Bissacco's blitter 1329: optimization patch. 1330: Fixed some sprite problems: during hysnc, clear sprpos and sprctl, and 1331: don't do_sprites() in line 0. 1332: 960409 Be a little more intelligent about when to call pfield_linetoscr() from 1333: pfield_doline_slow(). It's only necessary if a color register changed. 1334: If it can be deferred to the end of the line, use the the _full 1335: variants. 1336: Some CIA TOD fixes. 1337: Better transdisk, accepts first/last track parameters. 1338: 960405 New function memcmpy(), inline assembler for the i386. Compares two 1339: memory blocks and copies them at the same time. Used in 1340: bpl_data_differs. 1341: 960404 French keyboard from Tristan Cavelier. Parallel port emulation from 1342: Stefan Reinauer. 1343: Turn bpl?pt and r_bpl?pt into arrays, simplify some code using them. 1344: 960401 Some parts of Ernesto's new Mac sources. 1345: 960331 filesys.c: Close filedescriptors in free_key(). get_path() sometimes 1346: read uninitialized bytes. 1347: Minor cleanups all over the place. 1348: unchain_mnn() leaves sibling intact, but clobbers next. customreset() 1349: didn't care about that and crashed. 1350: Use the __asm__("intel_flag_lookup") magic not only for DOS. 1351: Support 320x300 for X (without LOW_BANDWIDTH). 1352: Add new function fast_memcmp(), defined as memcmp if __i386__ is 1353: undefined, otherwise a slightly optimized inline assembly function. 1354: 960329 Some bug fixes in gencpu.c 1355: 960328 Stupid programs write odd playfield modulos into the hw registers... 1356: Eek! Sometimes, linedescr[] was indexed with a negative value... Add a 1357: vpos >= minfirstline check to pfield_may_need_update(). I really need 1358: to rewrite this in Oberon... 1359: Don't turn off SHM when LOW_BANDWIDTH is selected. 1360: Keep the nextpos and linepos variables from pfield_doline_slow_?() as 1361: globals and initialize them in pfield_may_need_udpate(). 1362: Modified alloc_colors256() some more. 1363: Try to make sprite DMA follow the Amiga DMA slots more closely by 1364: adding a currhp parameter to do_sprites() and calling it at the end 1365: of a scanline or in pfield_may_need_update(). 1366: 960327 Added Gustavo's latest DOS code, as far as possible. Modified X86.S to 1367: use a macro SYM() to create symbol names either with underscore or 1368: without. 1369: Fixed some very silly SVGAlib lores bugs. Move screen 8 pixels up in 1370: low resolution. 1371: Use diwstop instead of diwstart to caclulate the DIW for 320x200 modes. 1372: Reset it not only in calcdiw(), but also in init_frame(). 1373: Speed up 256 color allocation a little bit. 1374: 960326 New function flush_block() in xwin.c and svga.c. Called from 1375: do_flush_line() and (new function) do_flush_screen(). Modify struct 1376: vidbuf_description, new member maxblocklines that records the maximum 1377: number of scanlines that do_flush_line() should collect before calling 1378: flush_block(). If it is zero, then flush_line() is used as before. 1379: This makes the special case stuff for SHM unnecessary, so it's removed. 1380: Fixed a stupid bug in events code: Put ev_hsync in front of ev_copper, 1381: so that do_copper gets called with hpos == 0 instead of hpos == 227 at 1382: the beginning of the line. 1383: New function init_frame() that sets up some variables, called from 1384: customreset() and vsync_handler(). 1385: Move smart update code to pfield_doline() from pfield_doline_h() and 1386: ..._l(). Clean up these two with a macro DO_ONE_PLANE, removing 1387: duplicate code. 1388: Implement big blits and ECS chip ids. New configuration options 1389: ECS_AGNUS and ECS_DENISE. 1390: Implement MOVEP. 1391: 960325 Before calling pfield_may_need_update(), make sure the value that was 1392: written to a register actually differs from the previous one. 1393: New expansion code from Stefan Reinauer. 1394: Modify struct mem_notify_node to contain a pointer to a valid bit 1395: instead of the valid bit itself, so that the playfield code has to 1396: check only one flag to decide whether to draw the line (SU method 2) 1397: 960324 The chip_flags approach has several problems. A flag may cover an area 1398: in memory that spans two scanlines, or several scanlines may use the 1399: same area in memory. 1400: Provide two other, different methods for the smart update: 1401: 1. Save all the bitplane data for each line and compare the chipmem 1402: with the saved data to decide whether an update is necessary. 1403: 2. Use a chain of memory_notify_node structs instead of a single bit 1404: for flags. Each scanline owns a set of these structs and chains them 1405: into the right place, and the chipmem_?put routines walk the chains 1406: and set the modified bit for each element. 1407: Unfortunately, both methods are slower than the old one. Especially the 1408: first one has some advantages though: it speeds up programs that clear 1409: the screen each update and write the same contents back, and it 1410: can speed up double-buffering programs at full framerate, since it does 1411: not have to check whether the bitplane pointers have changed. 1412: Remember the first and last drawn line for each frame, and pass them to 1413: flush_screen. Use this in xwin.c for SHM. Modify SHM logic: if smart 1414: update is selected, call XShmPutImage() from flush_line() instead of 1415: from flush_screen(). But gather blocks of consecutive lines to avoid 1416: calling XShmPutImage() too often. 1417: 960323 "Smart update" method, based very loosely on Alessandro Bissacco's 1418: work, implemented: 1419: Add an array chip_flags that records any change made to specific words 1420: in chip memory. 1421: Move all the hardware registers relevant to playfield display into a 1422: new structure bplinfo. Remember the contents of this structure in 1423: struct line_description for every line on the screen. Avoid redrawing 1424: in pfield_doline if the screen memory has not been modified for the 1425: line to be drawn and if the information from the last frame matches 1426: that in the current frame. 1427: 960322 Duplicate all functions responsible for drawing (pfield_doline, 1428: pfield_doline_slow, pfield_sprite, pfield_*word_*res) and append _h 1429: or _l (ah, the golden C++ days, when we had templates... which never 1430: worked). Modify the *_l functions to support drawing in low resolution 1431: (i.e. 320x200). Change the options -d and -H to support selecting color 1432: mode and resolution, thereby getting rid of some SVGALIB_* options. 1433: Get rid of the old pfield_*word_*res functions. 1434: Collect information about the minimum diwstart value during the 1435: previous frame and use that in case 320x200 was selected to fit the 1436: image on the screen. 1437: Get rid of HAVE_JOYSTICK (breaks the "-J" option) 1438: 960321 Give up in op_illg() if opcode is 0x4e7b and the exception handler for 1439: it is NULL. In that case, it's a 68020 Kickstart. 1440: 960319 New Next code from Ian. SunOS target was still screwed (two CFLAGS). 1441: Turn -x option into a "hair cursor" option. Support multiple mounts, 1442: but warn if more than one is seen (still does not work with Kick 1.3). 1443: Integrated Samuel Devulder's new LOW_BANDWIDTH patch. 1444: 960318 Integrated the start of Markus Gietzen's gfx library replacement. 1445: Changed the interface between custom.c and the graphics code. There has 1446: to be a buffer holding the data of all the lines on the screen. For 1447: linear SVGAlib modes, this is the video memory. Eliminated -x option, 1448: prepare_line(), DrawPixel() and the X_?BIT_SCREEN options, replace with 1449: 3 versions of pfield_linetoscr for different datatypes. Implement 1450: proper handling of the borders above and below the DIW in custom.c 1451: instead of in flush_screen(). 1452: Move common code (color allocation) from all the graphics files to new 1453: file gfxutil.c. 1454: Swedish keyboard from Per Olofsson. 1455: 1456: 960315 Release 0.5.0 1457: 960315 Newest DOS port from Gustavo. New sgi-gcc target. 1458: 960314 Add -lXext to osf target. Use "rb" for fopen() in main.c 1459: 960310 Play a bit with the __asm__ constraints. Implement CHK. 1460: 960307 Fix some stupid Makefile errors. Call parse_cmdline() if no init file 1461: is found. 1462: First round of NextStep patches; new file NeXTwin.m 1463: Stupid bug in custom.c: The #define is called NO_FAST_DISK, not 1464: NO_FAST_FLOPPY. 1465: Fix one of the problems the Deform demo had with the blitter: The 1466: line texture wasn't rotated to the right position in blitter_init(). 1467: 960306 Use some black GCC __asm__ magic to calculate flags more quickly 1468: (config INTEL_FLAG_OPT) 1469: HAM was broken by pfield_linetoscr changes. Fixed. 1470: LOW_BANDWIDTH sometimes got confused when the display window changed. 1471: 960305 BCD instructions work well enough for Locomotive Basic on CPE. 1472: New function parse_cmdline_and_init_file() that reads ~/.uaerc or 1473: ./.uaerc and adds the options found in those files to those in argv[]. 1474: Patch from Denis Sablic to make disk image filenames runtime 1475: configurable. Extend it for the ROM filename. 1476: Remove some of the run-time configurable stuff from config.h. 1477: More X keysym fixes (I finally got a proper .Xmodmap ...) 1478: for SVGAlib: Delete key, and PgUp/PgDn as Amiga keys. 1479: The sound driver can now handle 22050 Hz and/or 8 bit samples. Both of 1480: these are the default with LINUX_SOUND_SLOW_MACHINE. 1481: Also use the SNDCTL_DSP_GETBLKSIZE ioctl and write the buffer when 1482: there is enough data. 1483: Abuse keybuf.c to provide a fake joystick emulation with the numeric 1484: pad (enable with -J) 1485: 960304 bebox.cpp from Christian. 1486: Rename pfield_linetoscr_full to pfield_linetoscr_full8. 1487: More x86 assembly: pfield_linetoscr_full16. 1488: 960303 SunOS target from Dirk Vangestel. Put some not-so-superfluous includes 1489: back in. Added the DOS specific patches to filesys.c. 1490: Started x86 assembly optimizations. New file X86.S, includes an 1491: optimized version of pfield_linetoscr (for 8 bit screens only). 1492: Don't use SA_RESTART if it isn't defined. 1493: prepare_line() wasn't called often enough (again...), leading to memory 1494: corruption and "hit the reset button" bugs when using SVGAlib. 1495: 960226 Get rid of bool, true and false in amiga.h and everywhere else. 1496: Copy yesterdays code to AF_SOUND code, too. 1497: Include first round of Christian Bauer's BeBox patches. Add AIX fix 1498: from Samuel Mackrill. 1499: Add #include <string.h> and #include <stdlib.h> in some more places. 1500: Modify gencpu.c and readcpu.c so that immediate types j and J (0..15 1501: and 1..8) can be merged. Replace HAVE_ONE_GIG... option with a new 1502: CPU_EMU_SIZE option that can be set to a value in the range 0..5, 1503: where 0 is minimum size and 5 is equivalent to HAVE_ONE_GIG... 1504: 960225 Remove multiplications in sound code, use a big lookup table. 1505: 960222 Go back to having only one memory bank table of type addrbank instead 1506: of several function pointer arrays. Eliminate the HAVE_BROKEN_SOFTWARE 1507: define by extending the memory bank table to 65536 entries. 1508: Remove remaining DUALCPU code. Allow BTST src,imm instructions in 1509: table68k. 1510: 960220 Back out filesys.c change from Stefan Reinauer: don't call V36 specific 1511: functions in expansion.library. 1512: Better fix for the filesys.c assign problems by Oliver Moehrke. 1513: Make new playfield code conditional on EXPERIMENTAL_SCREEN_UPDATE. 1514: Improve the 64bit variant of said code. Try to get alignment for 1515: apixels by putting it into a union with a long. 1516: Kill the child in gui_exit(). Make graphics_init() return a value and 1517: check it in main.c to see whether graphics were initialized. If not, 1518: call gui_exit() anyway. 1519: Use new 256 color allocation routine from svga.c in xwin.c, too. 1520: Make LOW_BANDWIDTH compile _and_ work (wow...). 1521: 960219 Fix bug in memory.c: When loading 256K Kickstart, copy 256K, not 128K. 1522: 960218 Finished the rest of the UI: reset, debug and quit buttons as well as 1523: drive LEDs. New target linux-gui; all others use dummy file nogui.c 1524: for now. 1525: 960217 The GUI now supports inserting and ejecting diskfiles and displays 1526: their names. Two pipes are used now, one for reading and one for 1527: writing. Properly wait for the process to start up. 1528: Use sigaction() instead of signal() in debug.c. 1529: More fixes for the new display routines: Need to do a final call to 1530: pfield_*word_*res with the data shifted right from the last fetched 1531: word. 1532: 960216 Copper & sprite bugfixes from Marco Nelissen. Better X keyboard support 1533: from Bruno Coste. 1534: Move my_strdup() from filesys.c to options.h 1535: New files xui.c, uae-ui (shell script) and uae-ui.tk (wish script). 1536: These create a GUI for the X version of the emulator. So far, the 1537: power LED works. 1538: 960215 Don't rely on SVGAlib headers defining any keycodes at all. 1539: Include Thorsten Frueauf's NetBSD patch. Apparently, the devname 1540: variable in hardfile.c caused a namespace collision, so rename it to 1541: uaedevname. Fastmem fixes, debugger help, and ConfigDev allocation in 1542: hardfile and filesys from Stefan Reinauer. 1543: xwin.c: Give the window a title. 1544: Half-implemented the DIVU.L, DIVS.L, MULU.L and MULS.L instructions. 1545: 960213 AF sound system support from Marcus Sundberg. Win95 keyboard support 1546: for SVGAlib from Thierry Lescot. Added most of Gustavo Goedert's MS-DOS 1547: port. 1548: When using the old pfield_*word_*res functions, the gen_pfield_tables() 1549: function was undefined. 1550: Hmmm, /dev/brain must have been unmounted when I wrote the new pfield_* 1551: functions. Now they make sure they access the apixels array only with 1552: addresses that are multiples of 16 bytes. 1553: 960212 Fix copy&paste bug in cia.c found by Marco Nelissen: Need to call 1554: RethinkICRB() instead of RethinkICRA() for CIA B timers. 1555: Added Stefan Reinauers autoconfig area/fastmem support. Added run-time 1556: options for use_slow_mem and use_fast_mem. 1557: Oops, the pfield_*word_*res() optimizations break completely on HPs and 1558: Sparcs. Now use this optimization only if we are using GCC 2.7. 1559: HPUX fixes in the Makefile. New target hpux-gcc. 1560: The code for clearing the areas above and below the DIW was trying to 1561: be a little too clever selecting the color. Just use the current 1562: background color. 1563: Added channel attachment to the sound code. Untested. 1564: 960211 Found some flag setting bugs in the rotate/shift instructions and in 1565: addx/subx (which I thought I had already fixed) by reimplementing 1566: DUALCPU mode to work with Ed's e68k. 1567: Argh! The copper is completely disabled during DDF in Hires 4 bitplane 1568: mode - so copper positions 0x30 and 0x34 are about 640 pixels apart. 1569: The Kickstart uses FF39FFFE to wait for the end of line 256 instead of 1570: FFDFFFFE as in all the other modes. This is implemented now, but 1571: probably not 100% correct. 1572: Added check whether SVGAlib doesn't support our mode. 1573: If using GCC >= 2.7, define __attribute__((aligned(16))) as 1574: BIG_ALIGNMENT in config.h. Use it for the apixels array to prevent 1575: potential problems on the Alpha. 1576: Tried to make the color selection routine produce better results in 256 1577: color mode. 1578: 960210 readdisk didn't work at all for large files - fixed. 1579: In pfield_doline(), clear the apixels array if playfield DMA is off. 1580: In SPRxCTL(), reset sprite mode when sprpos==sprctl==0 1581: Rewrite pfield_orword_hires() and friends to use lookup tables. No 1582: noticeable improvement on my machine, but might be different on other 1583: machines. Two versions of this code: one using 64bit longs on the Alpha. 1584: New targets linux-debug and linux-prof. 1585: Use 1MB slow mem. 1586: -funroll-loops is a little excessive. Remove it. 1587: Add support for linear addressing modes for SVGAlib version. Speedup 1588: is 20-30% on my old 486 VLB. Now, would someone please fix SVGAlib to 1589: support that for the S3-864 as well? 1590: Set SVGAlib keyboard mode to DONT_CATCH_CTRLC. 1591: Change custom.c, xwin.c and svga.c to keep track of the display window 1592: and background color and to clear the areas above and below the display 1593: window if necessary to remove garbage images from previous frames. 1594: Set the LOF bit in the Kickstart replacement init routine. Also 1595: initialize CIAs to Kick 1.3-like values. 1596: 960209 Some more -DHAVE_xxx_H options in the Makefile and filesys.c. DEC Alpha 1597: fixes for filesys.c from Ed and Marcus. HPUX fixes. 1598: New field in table68k for CPU type, parsed by readcpu and compared to 1599: CPU_LEVEL which is defined in config.h. New instructions MOVE2C, MOVEC2 1600: and CAS. Also a new field for priviledge level that is used by gencpu to 1601: generate tests for regs.s and Exception(8) calls if necessary. 1602: Move parsing of table68k into a new file build68k.c. Generate a file 1603: called cpudefs.c containing the information from table68k in 1604: machine-readable form. Change readcpu.c to use the array found in 1605: cpudefs.c. 1606: Don't generate the d8(An,Dn) addressing mode in-line. Generate a call to 1607: get_disp_ea(), which is an inline function in newcpu.h if UAE is 1608: configured for 68000 emulation; for 68020 emulation, it is defined in 1609: newcpu.c. 1610: Debugger could crash on illegals because there was no mnemonic defined 1611: in lookuptab[]. 1612: Return zero for SERDATR. 1613: 960207 It seems that the copper stops completely if an attempt to write to a 1614: specific register fails because of COPDANG. Apparently, the Kickstarts 1615: use this to stop the copper with a 0x00000000 longword. This leaves the 1616: question why they don't do 0xFFFFFFFE like everybody else. 1617: Enable 512K slow memory by default. 1618: Integrate Stefan Reinauers battclock patch that makes it work with 1619: newer Kickstarts. 1620: In cia.c, do the little index pulse hack even if dskdmaen==1, because 1621: DMA is only enabled when dskdmaen==2 (Helps the "Mad Decision" demo). 1622: Add solaris entry to Makefile (GCC, Openwin, -DHAVE_STATFS_H) 1623: Define HAVE_STRINGS_H for SGIs, change filesys.c accordingly. 1624: 960206 Build a smaller table instead of cputbl.c. Use readcpu and the smaller 1625: table to build the cpufunctbl array at run-time. 1626: Nuked the other half of the old CPU emulator. Reimplemented debugger to 1627: use the information generated by readcpu. 1628: Define HAVE_STATFS_H for SGIs in the Makefile, and use that as a test 1629: in filesys.c instead of #ifndef __linux. 1630: Hack filesys.c, get_path() to support the assign command. 1631: Add -fstrength-reduce and -funroll-loops in Makefile to make people 1632: nervous. 1633: Rewrite Sam Devulder's LOW_BANDWIDTH patch and add it to xwin.c. 1634: 960205 Forgot some PC-relative bit instructions in table68k. 1635: Nuked half of the old CPU emulator. 1636: Split gencpu.c into two parts, new one is called readcpu.c and just 1637: parses the table68k file. Link it to the main program. 1638: 1639: 960205 Release 0.4 1640: 960203 filesys.c, action_read(): Slightly more efficient code (translate Amiga 1641: address to real pointer). 1642: Moved some common code in the generate_* functions in gencpu.c to a 1643: separate function. 1644: 960202 Added an experimental fast disk option. Currently turned off by 1645: default (it's not such a big win). 1646: Attached sprite fixes (overlapping att. sprites looked bad, Katakis). 1647: Add sleep(1) before resetting the console to text mode when using 1648: SVGAlib: this might fix some screen corruption problems. 1649: Add sprite/playfield priority checking to the most important case 1650: (single playfield, no HAM). 1651: In filesys.c, do_find(): open() returns -1 on error, not zero. 1652: Return ERROR_OBJECT_WRONG_TYPE if do_find() is called for a directory 1653: (fixes Champions of Krynn harddisk installation). 1654: 960201 Don't abort if sound driver not present, just set produce_sound to 0. 1655: New files keybuf.c and keybuf.h to record keypresses in the right 1656: order and without losing any. In cia.c, force 15 scanlines between 1657: keypresses, just to be sure. 1658: unixfs.device _does_ work with Kick 1.3: just don't trust what Kick 1.3 1659: sends in the startup packet. For now, disable more than one mount per 1660: command line. 1661: Started integrating Ernesto's new Mac sources. 1662: Remove superfluous includes from some files. 1663: 960131 Added Ed's unixfs.device (great stuff). 1664: Adding ULONGs to pointers is a bad idea on the Alpha if the ULONG value 1665: really is signed. Add some casts to LONG in (pc_p + src) expressions 1666: in genpu.c. 1667: If DMACON is written and copper DMA is enabled, do a COPJMP1 at once. 1668: Helps the "Interference" demo. 1669: 960129 More SGI fixes from Ed. Bugfixes and transdisk improvements from Marcus 1670: Sundberg. 1671: Remove EXTRA_DEFINES from Makefile. Breaks some systems. 1672: Move common sprite code from pfield_doline() and pfield_doline_slow() 1673: to new function pfield_sprite(). The same sprite may appear more than 1674: once on the same line, so don't shift out the bits of sprdata[] and 1675: sprdatb[] while displaying it (Turrican I). 1676: In xwin.c and svga.c, barf if LINUX_SVGALIB doesn't match the file 1677: being compiled. 1678: Make all .o files depend on config.h in the Makefile. 1679: No need to exit if sound driver unavailable, but -S given. 1680: Small debugger fix: Missing space in output. 1681: Fix for the sprite logic: Specifically, use a state variable indicating 1682: whether the sprite has been restarted after a VSYNC. Fixes most 1683: Turrican problems. 1684: 960124 Added Denis Sablic's patch for sound run-time option. 1685: Added Ed Hanway's patch for better Makefile, X mouse cursor blanking 1686: and more SGI compilation fixes. 1687: 960123 Include options.h everywhere. 1688: Handle 8 bit GrayScale visuals like PseudoColor. 1689: Remove C++ leftovers from joystick code. 1690: 960122 When using the joystick driver, the button test must come after 1691: handle_events() in vsync_handler(). 1692: 960118 Removed all the remaining C++ comments. Changed all inline keywords to 1693: __inline__. Define __inline__ if not using gcc. 1694: Make proper prototypes for everything. Compile with maximum warnings + 1695: -ansi + -pedantic. 1696: Remove CIA_cycle(), obsolete. 1697: Reimplemented the STOP optimization in newcpu.c. Removed DualCPU 1698: support in CPU emulator. 1699: Real nasty bug in pfield_doline() fixed: sprxpos could be evaluated as 1700: negative, with not-so-amusing results. (Need to rewrite this in 1701: Oberon to get array bounds checking :-) 1702: 960117 Heroic effort: Rewrote the thing in C. This might help fix some 1703: problems with users being unable to compile it. 1704: Fixed a problem in hsync_handler(): Only call flush_line() for lines 1705: in the display window, i.e. when we did a prepare_line() before. 1706: Better code for relative branches: Don't use setpc(getpc()+x) calls, 1707: increment regs.pc_p instead. 1708: 960116 Reimplemented the function to load the Kickstart ROM. Use stdio instead 1709: of fstreams since this apparently does not work on the Mac. Detect 256K 1710: Kickstarts. Detect corrupt ROM images (calculate checksum). 1711: Added Ernesto Corvi's Mac port. Changed it around a bit, so it 1712: probably won't compile. 1713: 960115 Reinstate config.h options for X screen depth, so that DrawPixel() can 1714: be inlined in custom.cc for speed. xlinebuffer is now incremented in 1715: each call to DrawPixel() (for both X and SVGAlib) to get rid of some 1716: address calculations. 1717: 960114 Fixed X generic pixel drawing routine for SHM. 1718: Still trying to fix the harddisk emulation. 1719: uae.device no longer breaks the debugger (can step through uae.device 1720: functions now) 1721: Bugs affecting performance: SPCFLAG_STOP never got reset, and DSKLEN() 1722: would set SPCFLAG_DISK even if DMA was being turned off. 1723: Made slow memory a run-time option. 1724: Defer interrupts by one CPU instruction to give programs a chance to 1725: read INTREQR ("Seeing is Believing" and "Substance" demos) 1726: Added ScrollLock hack for X, too. 1727: 960113 SVGAlib version compiles again. Fixed SVGAlib mouse bug. 1728: Fixed SHM bug: Maximum scanline is 313, not 312. 1729: Sometimes, disk.cc missed a side change and would read the wrong data. 1730: Fixed. Apparently, this was the worst compatibility problem. 1731: Implemented trace mode. 1732: 960112 Changed layout of class amigamemory a little so that gcc can generate 1733: better addressing modes. 1734: Finally wrote functions in gencpu to generate MOVEMs. 1735: 960109 Integrated Ed Hanway's patches for better X support and run-time 1736: configuration of some options. 1737: Got rid of the direct VGA memory access. (Need to do this differently). 1738: Changed the method of drawing lines: custom.cc now tells the graphics 1739: code the line number and whether it needs to be doubleed before drawing 1740: it. 1741: Added Andre Beck's MIT-SHM patch. 1742: Remove warnings for newcpu.cc. 1743: 960108 Fixed exceptions in op_illg(): Need to decrement PC. 1744: 960107 Added an "uae.device" resident module at 0xF00000. This emulates a hard 1745: disk (fixed size 8MB for now). 1746: 960106 Moved some common code from pfield_doline() and pfield_doline_slow() to 1747: a separate function. This fixes a potential HAM bug (two static vars 1748: for the same value). 1749: Sound support for Linux. Works only with graphics off and the CPU 1750: slowed down. 1751: Better SVGAlib keyboard support. 1752: 960105 Added AvailMem(), AllocMem(), AllocAbs() and FreeMem() dummies. 1753: The Hardwired demo times the multiplication instructions and prints 1754: "This demo don't like Axel" if they are too fast. Apparently, Axel has 1755: a 68040. Added a WANT_SLOW_MULTIPLY option to config.h. 1756: Fixed the fast blitter emulation (seems to work now). 1757: 960104 Fixed all the ChangeLog entries from 95 that said 96 (oops??!) 1758: pfield_may_need_update() should check whether bitplane DMA is on. 1759: Added ersatz.cc and ersatz.h. The purpose of these files is to 1760: implement one or two Kickstart functions that are commonly called from 1761: bootblocks. This should help support some games and demos that only use 1762: the Kickstart as an initial track loader. So far, it's only good enough 1763: for one program. 1764: 951223 More intelligent event handling in the CPU emulator. Slightly faster. 1765: 951222 Optimize CPU emulation by inlining cctrue(). Also, the real PC no 1766: longer needs to be incremented each instruction. The real PC value 1767: now has to be fetched by m68k_getpc(). 1768: Added direct screen access for SVGAlib, but it didn't help much. I'll 1769: probably remove it again. 1770: The gencpu executable is 2M smaller if it allocates memory 1771: dynamically. 1772: 951216 custom_bput() enhanced a little. Now remembers the value that was 1773: written in the other half of the register. 1774: Apparently, the USEx bits of BLTCON0 are ignored in line draw mode. 1775: (Silents-Demo) 1776: 1777: 951212 Release 0.3 1778: 951212 Included Alessandro's battclock patch. 1779: Fixed mouse autocalibration for wrong-aspect screens. 1780: 951210 Joystick support for Linux. 1781: Better color matching for 256 color modes. 1782: Rewrote most of the memory management in plain C, mainly to avoid the 1783: various template bugs in some compilers, but also because gcc can't 1784: handle the regparm attribute for virtual functions yet. 1785: 951209 Added Alessandro's patches for automatic mouse calibration. 1786: Tried to optimize the blitter loop a bit. Doesn't work. 1787: 951205 Added Hannu's patches for diskfile write support. 1788: 951201 More portability fixes, especially for weird compilers that workstation 1789: vendors ship. 1790: 951125 More keys for X, some keyboard support for SVGAlib (slightly 1791: dangerous); beginnings of US keyboard support. 1792: 951124 256 color mode support for SVGAlib. FAST! 1793: 951123 Tiny tweak to central loops in playfield code results in much better 1794: assembly. Apparently, GCC does not feel too good about optimizing 1795: reference variables. 1796: 951113 Bug fix in the events code (nextevent must not be changed while events 1797: are being processed). 1798: 951112 The playfield can now be drawn with the right aspect (i.e., every line 1799: twice). Slower, but looks nicer and supports interlaced mode 1800: (flickerfree!). The first 29 lines (blanking interval) are omitted 1801: now to fit everything in 800x600. 1802: Oops, disk drive identification did not work. Fixed, all four drives 1803: appear now. 1804: Oops, CIA_reset() was there all the time but never got called. 1805: Optimized the planar->chunky conversion in pfield_doline() a little. 1806: This really is time critical. 1807: 951111 If colors change on a line, we might still be able to use the full 1808: line update, if the change occurred outside the DIW. Have to be a 1809: little careful, though. 1810: Playfield code now uses real pointers again, but makes sure it can 1811: read enough bytes to complete the current line without segfaulting. 1812: Oops, my "better estimate" (see below) was a mistake. Restored the 1813: original 4 hw cycles per instruction relation. This slows down the 1814: emulator while booting (more screen refreshs), but speeds up many 1815: demos that otherwise spend most of their time waiting for a specific 1816: raster position. 1817: Playfield modulos must be added immediately when DDF stops. (WOC demo) 1818: Portability fixes in disk.cc (null pointers). 1819: Implemented average frame rate calculation ('c' in the debugger). 1820: Debugger fixes. 1821: Sprite fixes (height was wrong, wrong priorities in fast update) 1822: Added an AutoConfig expansion RAM device. The OS seems to recognize 1823: it, but breaks sometime later (seems to be the same bug that kills 1824: the emulator when there's RAM at 0xC00000). 1825: SVGAlib now has mouse support, too. 1826: HAM mode implemented. 1827: 951110 Implemented fast complete line playfield update. Switches back if 1828: necessary. Fixed attached sprites bug. 1829: Copper wait no longer slows down the CPU emulation. (there were 1830: lots of new copper bugs because of this, but I hope I fixed all 1831: of them). 1832: Changed cia?t? from UWORD to ULONG, since there were overflows in 1833: the assertion (because of 65535+1 == 0). 1834: 951108 The DIVS remainder could have the wrong sign. 1835: New HAVE_BROKEN_SOFTWARE define in config.h that makes the emulator 1836: mask out the high byte of all addresses. 1837: Wrote genblitter.cc and replaced the awful code in blitter_blit() 1838: and blitter_line() with calls to the generated function. 1839: 951107 Implemented ?BCD. These don't really work yet. 1840: 1841: 951106 Release 0.2 1842: 951105 Fixed a bug in the new CIA timer code. 1843: Fixed flag settings for some cases of ADDX/SUBX 1844: 951104 Changed the main loop a bit - call all the *cycle() functions only if 1845: it is really necessary. I hope everything still works - at least it is 1846: twice as fast now :-D 1847: Improved mouse emulation - avoid overflows, permit "wraparound". 1848: Added debugger 'g' command. 1849: Removed some useless casts from the CPU emulator. 1850: Added a configuration option to prevent checks for odd addresses. Only 1851: very bad programs need this check. 1852: 951103 The sprite emulation sometimes did not work too well, especially with 1853: dual playfield mode. Hopefully fixed. 1854: 951102 The 10 cycle counter is more like a 5 cycle counter in this 1855: implementation. (Fine, that makes floppy loading even faster :-) 1856: Implemented mouse movement. Totally out of sync with the X pointer, 1857: but at least something. 1858: Found out about the XAutoRepeatOff() function. Real useful. 1859: DualCPU mode works again. 1860: Corrected flag settings for CMPA.W. 1861: Fixed DIV instructions: Don't change dest in case of overflow. 1862: 951031 The PC is now converted to a real pointer. This is safe (much safer 1863: than converting bitplane pointers, since the PC always has to be valid) 1864: and should speed things up a little. 1865: 951029 Some primitive event handling: 2 mouse buttons and 1 joystick button 1866: are emulated, as well as parts of the keyboard (only german mapping). 1867: Fixes: Delay values and sprite widths are always LoRes. Playfield 2 1868: uses colors 9..15. 1869: Implemented a frame rate configuration option that can speed up the 1870: emulation quite a bit. 1871: 951028 A-ha. The 68000 has one strange feature after all. MOVEM.W to Dn sign 1872: extends the value to a longword. 1873: Katakis expects that accessing odd addresses causes exception 3, so 1874: we have to emulate this. *Sigh* 1875: Do at least something with the index pulse bit. It's more or less a 1876: toggle now. 1877: Implemented sprites & dual playfields, both with small glitches. 1878: 951027 When LOF is set, the frame is one line longer. 1879: 951026 Implemented a preliminary diskfile type for special formats. Turrican 1880: boots up correctly (crashes if I want to play a game, though). 1881: Implemented EHB and added a few missing masking operations in the color 1882: handling. 1883: Fixed the absw addressing mode - need to sign extend here as well. 1884: 951025 Fixed another extremely stupid bug, this time in the playfield code. 1885: Each pixel was calculated four times... 1886: Increased the size of the apixels array - this was being overrun. 1887: 951024 Fixed an extremely stupid bug in flag settings for byte and word 1888: operations. 1889: The "sectors until gap" field must be in the range 1..11 instead of 1890: 0..10, or very bad things will happen if it is used to calculate a 1891: blit size. 1892: 951021 Don't use real pointers to fetch bitplane data - too many segfaults. 1893: Delay the copper after executing an instruction, or wait sequences like 1894: FFDF FFFE 0200 FFFE won't work. 1895: Some programs expect that the CPU is completely turned off if the 1896: blitter nasty bit is set. So be it. 1897: The RSI Megademo chokes when disk DMA fetches two syncs. Leave one out. 1898: The LOF bit wasn't set correctly. 1899: 951020 Blitter fixes: Load fill carry from FCI each line; keep the masked 1900: value of bltadat in bltolda 1901: Give the CPU twice as many "cycles" - this still is not even remotely 1902: correct, but a better estimate. 1903: 951017 Added SVGAlib support. Unfortunately, my graphics card isn't fully 1904: supported by the current SVGAlib version. 1905: 951015 More small optimizations to memory.h and custom.cc. 1906: Added code for 256 color screens. The type of screen has to be 1907: selected at compile time. 1908: 951014 Modified the memory management to be (hopefully) a little more 1909: efficient. 1910: Changed gencpu.cc to generate functions and a big function pointer 1911: array instead of the old case labels. The generated C++ files now 1912: compile a lot faster, and the code should be better as well. 1913: 951013 Implemented pseudo POTGOR to keep the RMB from being detected as 1914: pressed all the time. 1915: GenerateDecTab() generated TRAP incorrectly. 1916: Made some more instructions (MOVE.B A0,D0 etc) illegal in gencpu.cc. 1917: Fixed X colors. Fixed data fetching for HiRes screens. 1918: 951011 EXT.WL instructions set the flags. 1919: Implemented second CPU emulator, plus "DualCPU" option. The new CPU 1920: emulator is generated automatically from a file called table68k by 1921: a program called gencpu. 1922: 951002 Removed version information from the file headers. 1923: Removed some sillinesses in custom.cc. 1924: DSKLEN bug fixed (mask out 0xC000). 1925: Playfield code rewritten. Now supports horizontal scrolling. Implemented 1926: some primitive color support. 1927: 950927 Fixed predecrement/postincrement addressing with byte size for the 1928: stack pointer. 1929: 950926 Fixes to the disk code: Terminate disk DMA, flip disk side, read more 1930: than just two sectors during disk DMA, fix checksum calculation. 1931: Fixed copper comparison bugs. 1932: 1933: 950901 Release 0.1 1934: 950826 bugfix in GenerateDecTab(): LEA.W instructions were generated. 1935: 950312 New file disk.cc, for (guess what) disk support. Doesn't work yet. 1936: 950311 Multiplication instructions were generating 16 bit results. 1937: Sign extension was wrong in some places. 1938: Removed bugs in the blitter (one word width: ALWM & AFWM and line mode) 1939: The Kickstart logo looks perfect now. 1940: Playfield routine should check BPLCON0 instead of BLTCON0 1941: 950309 ASR D8,D0 is not a valid instruction. 1942: Bitplane pointers were not incremented. 1943: Copper comparison didn't work. COP2LC and bitplane pointers were written 1944: incorrectly. Fixed. 1945: Something that bears a striking resemblance to the familiar Kickstart 1946: logo now appears in my X window! 1947: 950306 ROR used to set the X flag. Bug fixes in GenerateDecTab. 1948: More bug fixes in several 68000 instructions. 1949: Improved blitter. ?????? KS1.3 doesn't seem to set BLTDMOD on a 1950: line draw. What is this?? 1951: Fixed stupid bugs in playfield code. 1952: 950305 fixed bugs in opcode generation and interrupt handling. Implemented 1953: STOP. Fixed flag settings: Result wasn't masked. Ooops!! 1954: ADDQ #imm,An doesn't set flags. 1955: One-shot timer mode didn't work. 1956: Implemented blitter, doesn't work. 1957: Some primitive and slow playfield support, only to make the things that 1958: hopefully will soon appear visible. 1959: 950304 CIA timers ought to work OK. Copper is emulated. 1960: Emulator loops infinitely when trying to initialize the console.device 1961: 950303 experiments with X, figured out how to open a window and draw pixels. 1962: No graphics emulation yet, though. 1963: 950301 Nearly all instructions work. 1964: Fixed supervisor mode stack. 1965: The memory manager now handles CIA chips. 1966: 950228 Bug fixes in 68000 code, exceptions and more instructions implemented 1967: Memory management extended to handle custom chips as well. 1968: Debugger improved. 1969: Kickstart passes memory & CPU type test, builds up some system 1970: structures. 1971: 950227 MC68000 & little debugger for tracing. It can execute the first couple 1972: of KS1.3 instructions, but doesn't get past the memory test. 1973: 950225 Work begun
This archive runs on limited infrastructure. Preserving old code on modern bandwidth. Automated agents are requested to crawl responsibly.