![[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 |
1.1.1.19! root 1: 9B1125 Panning support for Picasso96 (in part from WinUAE). ! 2: More state save fixes from Toni, and a small disk.c update as well ! 3: which corrects track/gap lengths. ! 4: More evolution of the AHI code. ! 5: 9B1124 Merged a few more small pieces from WinUAE (larger pieces in case of ! 6: the Picasso code). ! 7: CLXCON2 was never called or statesaved. Fixed by Toni. ! 8: Sound sample frequency now always based on PAL timings. ! 9: 9B1123 Goodness! Massive memory corruption bug in picasso96.c - passing ! 10: last color changed to DX_SetPalette instead of count. ! 11: Fixes for the AHI code. ! 12: Changed prefetch emulation to be more similar to the one MAME uses. ! 13: Trace mode fixes (Bernd Roesch & myself). ! 14: 9B1122 Redo the ExNext code in the filesystem to make sure we always match ! 15: opendir with closedir and to fix some renaming problems. ! 16: Fix hd changing code in gtkui. 1.1.1.18 root 17: 9B1121 Mini-merge of noncontroversial WinUAE filesys.c code. 18: Add harddisk page to gtkui. Also change gui behaviour to wait for 19: the user to unpause before starting the emulator. 20: Add Toni's drive-led display. 21: Fixed a rather amazing source of crashes (we could fail to notice 22: the end of the frame in some cases). 23: Default to scanline mode. 24: Some tweaks to the SDL driver and the Win32 gui, mostly buffer-size 25: related. 26: Moved C source files back from tools directory (but still build them 27: there) to preserve my CVS history. 28: 9B1119 Move all the build tools (gencpu, cpuopti etc.) into their own 29: directory which is configured separately. This was necessary to 30: avoid picking up SDL CFLAGS (which can #define main to _SDL_main, and 31: that breaks if you don't link with SDL), and to make cross-compiling 32: at least a little more likely. 33: Remove sanity check for SDL header file - rely on sdl-config getting 34: things right (not that we have much choice). 35: Move thread support code past sound code in configure, and make sure 36: that SDL sound enables threads. 37: Move sername into options structure. 38: Check for existance of isinf/isnan before using them. 39: Rather large WinUAE merge - this tree compiles again under Windows. 40: Most of the additional features are missing though. 41: 9B1118 Restore code to mirror banks for 24 bit address space which got lost 42: when all the JIT memory changes were put in. 43: Speed up the CPU time calibration code. 44: Fix the broken file selector handling in the gtk ui. Add kickstart 45: selector. 46: Don't call XCloseDisplay, it apparently leads to crashes. 47: Try to solve the "missing write_log" problem in a sane way. 48: SDL joystick support. 49: When configure detects win32, use fsdb_win32 rather than fsdb_unix. 50: 9B1117 Proper masking for DDFSTOP; proper handling of AUDLEN == 0, changed 51: delayoffset code, gencpu BFINS fix, and lots of disk updates (all 52: Toni). 53: Some additional regs for move2c/movec2 (also by Toni IIRC) 54: 9B1115 COLOR_READ typo fix (Toni). 55: 9B1024 Filesystem reset leak fixes (Brian) 56: Allocate enough memory for custom state saving. 57: bploff renamed to bpl_off. 58: AGA copper cycle diagram (Toni) 59: Disable disk_data_used optimization. 60: 9B0926 Delayed palette setting for Picasso modes. Also fixed problem with 61: palette setting in CLUT emulation - we need to redraw the entire 62: screen if that happens. 63: 9B0921 Miscellaneous small bugfixes. Delete some obsolete code. 64: 9B0907 Beginnings of SDL support. Exterminated some penguins. 65: 9B0825 Some xwin.c fixes from Maik Qualmann. 1.1.1.17 root 66: 9B0730 Some sprite state machine tweaks. Fix sprite vs. DIW comparison. 67: HAM linebuffer was too small by half - oops. 68: Make absolutely, positively sure that when hsync_handler calls 69: sync_copper_with_cpu, we don't take any invalid shortcuts. 1.1.1.18 root 70: Memory management changes for JIT (Bernie [2], mostly). 1.1.1.17 root 71: 9B0729 Silly mistake - need to call flush_plane_data unless plfleft is 72: -1, even if no planes were enabled. 73: 9B0728 Small hack to reduce the time we lose with unnecessary 74: disk_doupdate_read calls. 75: Avoid calling flush_plane_data if we have no bitplanes enabled. 76: Tweak the SPCFLAG_INT handling to make "Delirium" demo work. 77: Limit rate of DDF warning messages. 78: Adjust display parameters so that things are shifted to the left 79: even more. 80: Fix (hopefully) CIA & blitter saving code. 81: 9B0727 Copper code reorganized; speedup code rewritten. Can now be 82: enabled with a GUI setting. 83: Fixed the delayoffset stuff. 84: 9B0725 AGA sprites, and some patches to sprite/playfield collision code, 85: mostly by Toni. 86: Fixed a small sprite restore problem; added missing code to map 87: fastmem banks when restoring them. 88: Some more state saving updates from Toni. 89: 9B0724 Fix a nasty bug in the copper SKIP instruction introduced by recent 90: changes. 91: Update the FAST_COPPER code. Not quite enabled yet. 92: More work on the memory configuration stuff. 93: Fix CPU type writeout in cfgfile.c (taken from WinUAE source). 94: Move the WOC demo BPLCON0 fix to compute_toscr_delay (Toni). 95: Added Toni's new disk patches, which include HD disk support by 96: Adil Temel. 97: Rewrote the memory and audio save/restore code. 98: Mask DDFSTOP with 0xFE, not 0xFC. 99: 9B0723 Yet another new and correct copper state machine (this time for sure) 100: Fix some AGA problems in DDF speedup code. 101: In gencpu.c, branch insns return a hardcoded cycle count; this had 102: not been updated for the CYCLE_UNIT stuff. Fixed. 103: xwin.c: If mouse is grabbed, we don't need mousehack. 104: Don't clear audio state if produce_sound == 0; this causes problems 105: if we want to turn the sound back on. 106: If an audio channel had a period of PERIOD_MAX, allow AUDxPER to 107: set its evtime back to a reasonable value. 108: Massive cleanup of memory initialization code; it will now use changed 109: prefs on reset. Added some memory options to gtkui.c. 110: Added Toni's state saving code, then added gtkui options and code to 111: initiate saving/restoring. 112: 9B0722 filesys.c: Fix writeout of hardfile params (Bernie [2]) 113: Add an A1000 Kickstart disk loader (Gerald Schnabel). Got rid of 114: the automount_uaedev code; we can detect the Kickstart version and 115: do the right thing. 116: Deleted a few unused variables (Toni) 117: Avoid aborting in decode_ham if Amiga code turns on the HAM bit but 118: has too few bitplanes (found by Toni). 119: Make the dummy mem bank return other values than 0; helps some 120: stupid memory detection code. 121: Implemented a little exec task lister for the debugger. 122: Added new debugger commands from Mark Cox. Improved his copper list 123: debugger by adding recorded timing information to it. 124: MOVE16 fixes from Gwenole Beauchesne. 125: xBCD fixes by Maik Qualmann, with some additional flag fixes by myself 126: (ABCD & SBCD behave exactly like on a 68000 now; dunno about NBCD). 127: 9B0520 AGA bscan2 fixes and disk.c kickstart speedup fixes from Toni. 128: Remove bogus changes I made to one of Toni's earlier disk.c updates. 129: Call RethinkICRA/B from INTREQ (suggested by Toni). 130: Properly save mixed_stereo setting. 131: NEGX uses X and Z (Bernd Roesch). 132: Danish keyboard support (Jesper Bojesen). 1.1.1.16 root 133: 9B0311 Repair "emulated but no output" sound setting. 134: 9B0310 Fix massive bug introduced in do_cycles_slow with the rescaling 135: changes. 136: CLXCON logic updated for AGA (Toni). Change meaning of collision_level 137: and add support for it in cfgfile.c. 138: 9B0309 Fix problem where a fsdb file could prevent deletion of otherwise 139: empty directory discovered by Torsten Enderling. 140: Rescale return values of CPU emulator functions. 141: 9B0225 Undo parts of the audio changes that made the "accurate" setting the 142: default. 143: 9B0221 Make fsdb_unix compile again. 144: 9B0219 Fix ddf for cases where start isn't aligned to fetchunit. 145: Fix some problems with the sound code going into infinite loops or 146: crashing. 147: Try to re-add Toni's ddf delay fixes properly. 148: Added a fix from Krister Walfridsson for X11 SHM code. 149: Rename OBJECT_NOT_FOUND to OBJECT_NOT_AROUND (not sure why, but the 150: win32 port does this; Fellow picked it up as well...) 151: Give in and allocate ainos with xcalloc. 152: Import fsdb_win32 code from Fellow. 153: Some tidyup and fixes in fsdb code. 154: Lose the lastdiv10 variable, and update CIA timers when reading them 155: (mostly by Bernie [2] Meyer) 156: Move reset_uaedevices and get_new_device to filesys.c. Add extra 157: parameter to get_new_device. 158: 9B0213 Fix mouse pointer jumping bug in X11 code (Ralf Hoffmann) 159: Scaled event handling by a factor of 512. This allows more accurate 160: sound output frequencies, as well as more fine grained CPU speed 161: settings. 162: Implemented totally untested emulator side AHI support. 163: 9B0211 Partially reverted changes made to the audio event code in 0.8.0. 164: Implemented "mixed" stereo. 165: 9B0210 Fixed invisible sprite bug - sprite_width got a new meaning, and not 166: everything had been updated. 167: 9B0121 cycles => get_cycles () (from Bernie [2] Meyer's JIT patchkit) 168: 9B0120 Final bugfix in speedup code (plfleft got set too early). 169: 9B0118 Fixed SPEEDUP code for AGA. Speedup now starts when fetch_state is 170: fetched_plane0. 171: 9B0115 First batch of AGA playfield changes (Toni & myself). 172: 9B0110 Small fix for the m68k_run macro (Bernie [2] Meyer) 173: 9B0109 Use ntscmode from prefs structure (Dwayne C. Litzenberger) 174: Some 040 FPU fixes (Bernie [2] Meyer) 175: Delay DSKREADY a while; fix other small disk bugs (Toni Wilen) 176: 9A0902 F0FF10 is now a puts () calltrap. 177: 9A0830 Add some missing initializations in filesys.c, thanks to debugging by 178: Harald Frank. 179: 9A0827 Fix a crash with framerate != 1. 180: Don't allow programs to see left & right or top & bottom simultaneously 181: with the keyboard joystick replacements. (Gee, games on the CPC and on 182: the Amiga actually have something in common - some of them crash when 183: the joystick is pushed in too many directions at the same time...) 1.1.1.15 root 184: 9A0825 Put in the new sprite code. This has existed for many months but 185: I've only now found the final bugs. 186: New version of Bernd Meyer's special_mem code. Also add his 187: set_special/unset_special code, the noflags changes for gencpu, 188: and a fix for cpuopti's string matching. 189: Tiny clarification in table68k. 190: Add 68040 support from Basilisk. 191: readcpu.h is now included from newcpu.h. 192: 9A0824 Rather massive overhaul of the DDF code. Many other timing fixes 193: (copper state machine mostly) (Sam Jordan and myself) 194: Get rid of stupid optimizations in color recording. 195: Synchronize CPU and copper in custom_wget. 196: 9A0816 Introduce variable "special_mem", set it in each memory handler 197: that doesn't work on real memory (Bernd Meyer). 1.1.1.14 root 198: 9A0815 Fix a really stupid bug in cfgfile_strval. 199: Nonfunctional support for switching between windowed/fullscreen modes 200: in gtkui. 201: Add (semi-broken) support in the debugger to create a trace logfile. 202: Take out most port-specific flags handling; x86 version kept with 203: slight name changes to indicate that this is code is suitable for 204: other ports as well. Took out HAVE_GAS while I was in configure. 205: 9A0811 Fix the line doubling slowdown. 206: 9A0809 Software autofire. 207: Remove one line from Makefile.in in the hope it will fix the build 208: errors people are seeing. 209: 9A0807 Tighten up some minor aspects of spcflags handling. 210: Fix an AGA bug in customreset. 211: Set default CPU speed to approximate a A500. Tweak gencpu so that the 212: timings for this are more accurate. 213: Ignore T0 bit if not emulating a 68020. 214: Add an option to set the number of emulated floppy drives. 215: Improve the mfmcode function a bit. Faster and more accurate. 216: More SCSI patches from Patrick. 217: Add gui_data structure from WinUAE. Turn on gui leds again. 218: 9A0725 Fix incorrect define of m68k_run for non-x86 systems (could have 219: sworn I already fixed that). 1.1.1.13 root 220: 9A0723 Some SCSI patches by Patrick Ohly (which I got two months ago... oops) 221: Filesystem fixes from Brian. 222: AGA updates from Toni. 223: 9A0627 Apydia loader fixes from Toni. 224: 9A0611 Fix bug in filesys.c causing in-use ainos to get discarded. Thanks 225: to Brian for the analysis. 226: 9A0608 Added some disk.c changes from Toni and a crude hack of my own to 227: speed up disk accesses done by the Kickstart. 228: 9A0330 Change blitter not to translate memory addresses to native. Had to 229: disable 32 bit blits. 230: Floppy code changes (Toni Wilen) 231: Delete GGI port. 232: Fix toggle_inhibit_frame (Brian). 233: 9A0228 Some WARPUP fixes, and changes to accomodate broken Amiga gzip (Sam 234: Jordan). 235: 9A0221 Tweak POTGO a bit. Seems to work with BC Kid now. 236: 9A0220 Fix off-by-one errors in expansion.c (Toni Wilen). 237: Remove AGA checks in DDFSTRT/DDFSTOP (Toni Wilen). 238: Turn off/restrict some warning messages. 1.1.1.12 root 239: 9A0209 Fix some problems where the blitter nasty flag wouldn't ever get reset. 240: cia.c: Set all parallel port related bits to 1 when reading. 241: 9A0208 gencpu.c: Fix saved PC for division by zero exception (found by Sam 242: Jordan). 243: 9A0131 update_copper: Mask until_hpos properly. 244: Delay values are masked to 3 bit in hires mode (Sam Jordan). 245: 9A0102 Only delay horizontal copper wait if vertical position matches exactly. 246: Tweak vblank end position for sprites. 247: Calculate playfield borders more accurately for case where we have 248: sprites. 249: 9A0101 Reintroduce ev_copper; use it sparingly for longer waits within a 250: single line (but _not_ to time reads). 251: Fix a dumb bug in yesterday's changes that screwed up scrolling. 252: 991231 drawing.c: Cruftectomy. genp2c.c: Gone. X86.S: mostly gone. 1.1.1.11 root 253: 991229 Always use NTSC vblank_lastline for sprite purposes. 254: Make sprite logic recognize vertical stop even if vertical start has 255: not been reached (found by Sam Jordan). 256: Prevent move2c/movec2 instructions from incrementing the PC if they 257: raise an illegal instruction exception (found by Toni Wilen). 258: 991212 ncurses port compiles again. 259: Disable serial interrupts in custom.c; they break things. 260: 991208 Add a missing decide_line call in update_copper_1. 261: 991206 Fix copper emulation to use state machine figured out by Sam Jordan. 262: 991205 Fix event table corruption in disksync_handler. 263: 991113 gencpu.c: Fix a problem with BTST. 264: 991110 fpp.c: Initial set of changes merging code from Lauri Pesonen. 265: 991024 Add missing type in m68k_run1. 266: Divisions by zero clear overflow bit. 1.1.1.10 root 267: 991023 Fix compilation problem for ! HAVE_STEREO_SUPPORT. 268: 991022 Rewrite disk read code to transfer some DMA words every raster line. 269: fpp_movem_index1 and fpp_movem_index2 were reversed (Brian King, Sam 270: Jordan). 271: Fix accuracy of fpp fix insns (Sam Jordan). 272: Allow ports to define their own FP 68881/host conversion functions (Sam 273: Jordan). 274: Get rid of all the HAVE_GET_WORD_UNSWAPPED code. 275: Fix prefetch for insn opcodes (Sam Jordan). 276: Fix call insn syntax in X86.S. 277: Tidy up sound interpolation support. 278: Readd old command line arg parsing. 279: Change "long unsigned int" to "unsigned long" in a couple of places. 280: Use Sam's version of the filesys.c fixes (action_parent_fh, 281: action_change_mode). 282: Fix a place in disk.c where a mfm buffer was zeroed instead of 0xaa'ed 283: (Sam Jordan). 284: 991006 CPU emulator fixes by Christian Bauer 285: tui.c compilation fixes by Christof Petig. 286: Fix operand in one insn in X86.S (Brian). 287: zfile.c fix for filenames with embedded spaces (Ingo Ruhnke). 288: linux-ppc configury patch (Holger Jakob). 289: s,static __inline__,STATIC_INLINE, (Sam Jordan). 290: cfgfile.c: Don't use "t" modifier in fopen (Sam Jordan). 291: serial.c: Don't close file if fd < 0 (Sam Jordan). 292: filesys.c: Open files on readonly devices with O_RDONLY (Sam Jordan). 293: Fix bugs in parent_fh and change_mode actions (Sam Jordan). 294: WARPUP patch (Sam Jordan). 1.1.1.9 root 295: 990621 Initial version of Richard Hoffmann's sound interpolation patches. 296: Fix a crash in action_set_comment. 297: De-inline 68k trace code. 298: 990508 Fix rdtsc asm statement to show that edx is clobbered. 299: gtkui.c: Add pause button (Patrick Ohly) 300: configure.in: Check for -mpreferred-stack-boundary compiler option 301: and use it if found to set a sane value. 302: 990429 X11 patches by Patrick Ohly. 303: Filesystem code extended to keep a simple database with information 304: about each file that is not representable on the host FS, such as 305: script/pure bits or filenotes. 306: 990417 Bits and pieces merged from the Win32 port; mostly options stuff. Got 307: rid of getopt; added chipset options to prefs structure. 308: 990411 Patches by Patrick Ohly: SCSI support and timehack tool. 309: 990410 Applied easy parts of Toni's latest AGA patches. 310: 990315 Attempts to fix Picasso crashes in SVGAlib version. 1.1.1.8 root 311: 990216 picasso96.c: Some optimizations. 312: cfgfile.c: Save target specific options. Allow simple/smart choice 313: for centering options. Delete mount_data junk, substitute immediately. 314: Whitespace handling fixes. Fix config_description typo. 315: m68k flag handling redone. 316: 990212 Implement rotate instructions without loops. Some other m68k 317: optimizations. 318: Some configure.in fixes. 319: 990122 gtkui.c: Fixes for gtk 1.1.13. 320: 990113 genp2c.c: Don't include custom.h. 321: 990109 Some fixes in cfgfile.c. Make svga.c compile again. 1.1.1.7 root 322: 981229 Miscellaneous fixes in the new config file stuff. 323: In config files, allow 68ec020 as CPU type, and get rid of 24 bit 324: addressing option. 325: Make the get_disp_ea non-inline. 326: SGI sound support (Ari Heikkinen). 327: More updates by Brian: 328: audio.c: Handle frequency, too, when changing sound prefs. Do some 329: work in update_audio even if not generating sound output. 330: filesys.c: Request 50 buffers in our parampacket. 331: Assorted right control key fixes. 332: Some updates in od-win32 (not nearly done...) 333: 981228 Config file and option handling redone; based on work by Brian. 334: Rename UAEn: devices to DHn: (Brian). 335: filesys.c: Implement ACTION_CHANGE_MODE, ACTION_PARENT_FH; add support 336: for variable hardfile blocksizes (Brian). 337: Some Picasso96 updates (Brian). 338: 981226 Win32 fsusage.c fixes; zfile bug fix; gui_fps addition (Brian). 339: 981221 Fix gtkui to work with newer versions of gtk. 340: Fix gencpu so that FPU emulation doesn't crash when unswapped word reads 341: are available. 342: 981129 gencpu.c: Delete all the queue_amode stuff again. 343: xwin.c: Allow switching between DGA and window mode (Marcus Sundberg). 344: DGA seems broken in X 3.3.3, however :( 345: 981011 Teach cpuopti.c about epilogues that increment esp by popping. 346: 980930 Add BSD socket code (Mathias Ortmann). 347: 980923 filesys.c: Implement FH_FROM_LOCK (Brian King) 348: 980823 xwin.c: French keyboard patch (Rapha�l Poss). 349: 980810 Fix RDTSC detection (Rapha�l Poss). 350: newcpu.c: Patch by Brian King to display what kind of CPU table is 351: built. 352: 980716 More AGA patches: use new macro GET_PLANES to calculate number of 353: bitplanes from BPLCON0; adapt the copper code; add BPLPT[HL][78] and 354: BPLDAT[78] (Toni Wilen) 1.1.1.6 root 355: 980715 Change sd-uss/sound.c not to keep a file open in setup_sound, to 356: support reconfiguration. 357: 980713 Fix m68k detection in configure.in. Clear execbase after allocating 358: chipmem (Holger Jakob) 359: od-dos update, tui.c patches (Gustavo Goedert) 360: Get rid of --enable-sound, always enable sound. 361: Make svga.c compilable. 362: 980712 md-ppc-gcc patches (Holger Jakob) 363: 980707 gtk ui: Better CPU speed selection. 364: Move initializations from add_filesys_unit to set_filesys_unit. 365: 980706 AGA patches by Toni Wilen: Call calcdiw when changing fmode; fix 366: references to bplcon variables to go through dp_for_drawing; ifdef out 367: some code that's too hard to fix for now. 368: "Fix" the AGA color code. In pfield_linetoscr_aga, add lframe_end and 369: diw_end parameters and use them. Add some fill_line code for AGA. Make 370: sure no code touches acolors if [AGA_CHIPSET != 0] 371: New function finish_line_aga. 372: 980705 drawing.c/custom.c: Get rid of line_changed, replace with a single 373: variable thisline_changed. Pass it to hsync_record_line_state and 374: evaluate it there. Get rid of BORDER_PREV and BORDER_NEXT states; 375: add LINE_BLACK/LINE_REMEMBERED_AS_BLACK to implement scanlines option. 376: Remove code in pfield_draw_line that checks line_changed. 1.1.1.5 root 377: 980702 filesys.c: New function move_filesys_unit. 378: Win32 GUI fixes. 379: 980701 gtk UI beautification and code cleanup (Michael Krause) 380: Beginning to make audio code run-time configurable (mostly by Michael 381: Krause) 382: Fix off-by-one errors in memory check handlers (Jake Hamby) 383: configure: Add AM_CYGWIN32, AM_EXEEXT and AM_MINGW32 and fix some Win32 384: configuration problems. 385: genp2c: Fix symbol name braindamage. 386: md-i386-{gcc,watcom}.c: Don't do machdep_init when [_WIN32]. 387: Many Win32 compilation fixes. 388: filesys.c: Attempt to fix the mount code chaos. 1.1.1.4 root 389: 980629 MOVEP bug fixes (Philippe Gerin) 1.1.1.5 root 390: Comment out code using readdir_r again. 391: 980628 gencpu.c: Generate an end label only when it's needed. 392: configure.in: Check for windows.h, ddraw.h, sys/utime.h; check sizeof 393: __int64; check whether a small DirectX program compiles; check for 394: Watcom C; don't do GNOME_FILEUTILS_CHECKS when compiling on Win32 395: without GNU C; use AC_EGREP_CPP in some places instead of compiling a 396: file by hand; check for GAS .p2align feature; check for USE_UNDERSCORE 397: with a small test program; add code to set up od-win32/md-i386-watcom 398: directories when appropriate. 399: genp2c.c: Use GAS .p2align feature; don't generate .type directive. 400: Rename md-*/memory.h to md-*/maccess.h. 401: More fixes in Win32 code. 402: Add wcc.sh and wrc.sh, two scripts to make it possible to use the 403: Watcom C compiler and the resource compiler in a Cygwin32 build 404: environment. 405: 980626 Change zfile.c not to use fopen_del. 406: 980614 Delete ihf_sem stuff. 407: picasso_refresh now nonstatic and called by check_picasso after a 408: notice_screen_contents_lost call. 409: More functions to access filesystem mount information. 410: More Win32 code. 411: 980613 Use an AC_TRY_LINK block to check for libpthread. 412: 980611 Rename lockscr/unlockscr to gfx_{lock,unlock}_picasso and add new 413: functions lockscr/unlockscr which are now called by drawing.c. 414: xwin.c: Clear picasso_invalid_lines in graphics_subinit and when 415: doing an XPutImage for a refresh. 416: New function reset_all_systems, which is a bit misnamed at the moment 417: because it only resets the filesystem. Called from m68k_go when a 418: reset is needed. 419: In filesys_prepare_reset, examine unit_pipe instead of self to 420: determine whether a unit has a thread that needs to be killed. Move 421: the reset_state and reset_sync_sem variable to UnitInfo so they are 422: available at all times. 423: Don't need to call finish_drawing_frame if we do a reset, it was 424: done a couple of lines earlier. 425: In uae_reset, do nothing if any reset or quit operations are in 426: progress. 427: Make DMACONR static. 428: 980609 Add memory size fields in struct uae_prefs. New variables 429: allocated_xxxmem for every block of memory that holds the actual 430: amount currently allocated, copied from the preferences at allocation 431: time. 432: Create a new structure "uaedev_mount_info" to hold filesystem mount 433: information. Add a pointer to one of those in struct uae_options, 434: and copy it from currprefs to a filesystem private place when 435: starting the filesystem threads to a work area. 436: Terminate filesystem threads during a reset, and restart the whole 437: filesystem code afterwards so that prefs changes are noticed. 438: 980608 If check_prefs_changed_gfx returns nonzero, or if we switch from/to 439: picasso mode, call notice_new_xcolors and notice_screen_contents_lost. 440: Delete the frame_do_semup crap. 1.1.1.4 root 441: 980606 Create config.h link in configure, not in Makefile. 1.1.1.5 root 442: 980604 xwin.c: No longer need picasso_max[wh]. Improve picasso invalid lines 443: code a bit more. Fix some dithering bugs. 444: gtkui.c (add_checkbox_widget, add_radio_widget): New helper functions. 445: Patches by Marcus Sundberg: 446: Make X11 picasso clut emulation mode work in DGA 447: gfxutil.c (setup_dither): Free redvals. 448: 980602 Split out init_aspect_maps from reset_drawing. 449: Call check_prefs_changed_gfx from vsync_handle_redraw, and initialize 450: row and aspect maps if it returns nonzero. 451: Use an enum instead of integer values to communicate between 452: hsync_handler and hsync_record_line_state. 453: gtkui.c: Use gtk_window_set_title (Marcus Sundberg) 454: xwin.h: Add lockscr/unlockscr prototypes. 455: 980601 Move check_prefs_changed_xxx into vsync_handle_redraw. 456: Make ersatz.c a little more crash-proof. 457: 980531 New graphics specs 'a' and 'p'. Support them in the X11 version. 458: 980530 sysdeps.h, posixemu.c: Change (char *) to (const char *) in the 459: arguments of some of the posixemu functions. 460: sysdeps.h: Delete getopt declaration. Include "getopt.h" in main.c. 461: aclocal.m4: add fsusage.c support from Gnome. 462: fsusage.[ch]: New files from fileutils, slightly modified to fit 463: into UAE and to add Win32 support. Adapt do_info filesys.c to use 464: it, and delete all the old cruft. 465: Some updates for the Win32 port. 466: 980529 Patches for X11 version by Marcus Sundberg (Picasso invalid lines 467: optimization, XShm availability check, cleanup, etc.) 468: 980528 Add field emergmem to struct vidbuf_description, and change 469: pfield_draw_line to use it. 470: 980516 Patches by Samuel Devulder and Holger Jakob: update Amiga port, 471: add support for compiling cpuemu.c in seveal steps again, add 472: cheatsearch in the debugger. 473: 980515 GGI patches (Christian Schmitt). 474: sd-mme: New directory (Marcus Sundberg). 475: Changes in the X11 version to support mixed DGA/normal binaries, 476: hotkeys, and mouse grabbing (Marcus Sundberg) 477: xwin.c: Copy a bit of code from Win32 port that constrains 478: lastmx/lastmy to real coordinates when ievent_alive. 479: 980510 gencpu.c: Slightly better flag generation code. 480: picasso96.c (wgfx_flushline): Don't do anything if Picasso screen 481: is invisible. 482: 980508 Don't disable GGI if we can't find svgalib. 483: newcpu.[ch], gencpu.c, md-i386-gcc/memory.h: Implement unswapped 484: word reads for opcodes. 485: 980505 drawing.c: add missing #ifdef X86_ASSEMBLY. 486: 980504 More BSD fixes from Krister. 487: xwin.c: Call XAllocColorCells for dithered modes. Don't set image_mem 1.1.1.3 root 488: in get_image, instead arrange for it to store the address in a pointer 489: chosen by the caller. 1.1.1.5 root 490: picasso96.h: Add a clut array to picasso_vidinfo. 491: picasso96.c (do_fillrect, do_blit): Use picasso_vidinfo.pixbytes 492: instead of picasso96_state.BytesPerPixel where appropriate. 493: (do_fillrect, do_blit, wgfx_do_flushline): Handle the case when the 494: Amiga-side buffer is a 256 color palette mode and the native buffer 495: is truecolor by looking up colors in the new picasso_vidinfo.clut. 496: xwin.c: Adapt to support emulating a palette display in truecolor. 497: Make tui.c compile. 498: 980501 picasso96.c: Make sure there's an unlockscr for every lockscr call. 499: In SetColorArray, detect when no colors changed. 500: configure.in, aclocal.m4, Makefile.in: Use AM_PATH_GTK from gtk 501: distribution. 502: New field rgbformat in picasso_vidinfo. Set it in xwin.c and svga.c. 503: custom.c (mousehack_helper): Use Picasso screen offsets correctly. 504: 980427 Call decide_line in DIWSTOP. Implement DIWHIGH register. 505: 980426 filesys.c (get_nname): Don't allow Unix "." and ".." directories. 506: gencpu.c: Approximate the number of cycles per instruction. Make 507: generated functions return that number, and use it in newcpu.c in 508: the do_cycles call. 509: custom.c, drawing.c: init_hardware_for_drawing_frame must be 510: called from init_drawing_frame. 511: Move read_processor_time specific stuff to new file md-xxx/rpt.h. 512: md-i386-gcc/support.c: Try to detect whether the CPU has RDTSC. 513: newcpu.c: m68k_speed == 0 means approximate A500 speed; 514: m68k-speed == -1 means use frame rate hack when available. Some 515: changes to support this. 516: 980425 custom.c: New copper code. Unfortunately not much faster than the 517: old one. 518: Get rid of some unused variables throughout. 519: filesys.c: Use readdir_r when available. 520: picasso96.c: In BlitPattern, correctly handle the pattern offsets. 521: missing.c: Get rid of getopt. Add GNU getopt instead. 522: In SPRxPTH, SPRxPTL, call decide_sprites instead of decide_line. 523: 980421 custom.c: Add an extra argument, hpos, to plenty of functions, and 524: use it instead of calling current_hpos. 525: 980420 BeOS update (Christian Bauer). 526: When enabling additional planes in post_decide_line, update 527: thisline_decision.bplcon0. 528: Tweak copper timings. 529: 980419 Extend post_decide_line to handle cases where some planes are 530: enabled much later in the line than the others. 531: In pfield_doline_unaligned_h, don't increment dataptr. 532: Split custom.c, move drawing code to new file drawing.c. New header 533: file drawing.h for declarations of shared stuff. 534: osemu.h: Delete junk. 1.1.1.3 root 535: 980418 gtkui.c: Don't crash if -G option is given. 1.1.1.5 root 536: audio.c, custom.c: Don't use the generic event handling mechanism 537: for audio emulation, the overhead is too high. update_audio: New 538: function, handles audio updates "on demand" or every hsync (Idea by 539: Petter Schau). 540: 980417 Some cleanups in custom.c to make splitting it up easier. 1.1.1.3 root 541: 980416 custom.c: Don't use bplhires in do_modulos. 1.1.1.5 root 542: 980415 readdisk.c: Support FFS (Jim Cooper). 543: BSD patches by Krister Walfridsson. 544: configure.in, Makefile.in: More fixes for build != source directory. 545: Arrange to build p2c_x86.o when using md-i386-gcc. 546: sysdeps.h: Include limits.h. 547: Remove silly code that restricted possible version numbers. 548: 980320 configure.in: Check for vprintf. 549: Make write_log a varargs function. Provide a write_log_standard 550: function that is used by most systems. 551: Refuse to compile with non-ANSI compilers. 552: gencpu.c: Rearrange some of the generated code to reduce register 553: pressure. 1.1.1.3 root 554: 555: Release 0.7.0 556: 980401 gengenblitter.c: Complete rewrite. Much smaller, much faster, much 557: better results. 558: Check for -lXi to support newer versions of gtk+. 559: 980316 gencpu.c (read_counts): Two silly bugs fixed. 560: main.c (save_options): Don't write out 'a' if cpu_level < 2 561: configure.in: Changes to try to be compatible with Solaris /bin/sh. 562: 980315 New UAE-Control version from Tauno & Cloanto. 563: Make configure set UNALIGNED_PROFITABLE for x86. 564: Get rid of "-o", add save command to tui.c and gtkui.c. 565: Fix some bugs in save_options. 566: 980314 Many improvements in the gtk UI. Made CPU type changeable at run-time. 567: Make DGA && ! VIDMODE configuration compile. Kludge around the fact 568: that DGA misses colormap changes. Call XWarpPointer to try to avoid 569: the click-moves-framebuffer bug. 570: Make sure p2c.[ch] don't get deleted. 571: 980313 Move all CPU_LEVEL handling from build68k.c to gencpu.c. In gencpu.c, 572: ignore all parameters and don't write to standard output, instead 573: hardcode all filenames and generate them all at once. Arrange to 574: generate code for all CPU emulator variants and to select them at 575: run-time. 576: Get rid of copper cheat mode, re-use "-C" option for CPU selection. 577: 980312 In COPxLCL, mask out LSB (Axel Wefers). 578: newcpu.h: Always compile in both get_disp_ea versions with a different 579: name. Delete USE_POINTER option; always use pointer. Don't fill 580: prefetch in m68k_setpc, do it in callers where necessary. Compile in 581: two versions of get_ixxxx, with different names for prefetch and 582: non-prefetch. Change gencpu.c to use these appropriately. 583: Don't generate 16 small CPU emulation source files. With a decent 584: machine it's no problem to compile all the code in one file. Also turn 585: off -O3 for gcc, -O2 does fine since all the necessary __inline__s are 586: there, and the compiler takes less memory that way. 587: Don't include <linux/joystick.h>, duplicate some definitions from 588: there instead. Makes configure.in simpler, too. 589: Fix Makefile.in/configure.in and a couple of other places to support 590: building in a separate directory (sort of; not complete yet). 591: Remove p2c.[ch] and genp2c when doing make clean. 592: Delete code to support Tcl/Tk and XForms GUIs. 593: 971215 More Amiga and BeOS patches from Sam and Christian. 594: 971205 Move some more code from custom.c to audio.c. Optimize sample handlers 595: a bit. 596: Make RDTSC asm statement volatile. 597: 971203 Latest GGI patch from Christian Schmitt. 598: Fix configure to enable zfile decompression for GGI target. 599: Fix mouse button mapping in ggi.c. 600: 971128 Fix double definition of gfxvidinfo in custom.c. 601: Fix impossible register constraints in asm statement in newcpu.c 602: Sprinkle casts and rename some variables to make C++ compilers happy. 603: Add BeOS patches by Christian Bauer. 604: 971125 More Amiga patches from Sam. 605: 971120 Get rid of DONT_WANT_SOUND. 606: 971119 Support USS sound not only on Linux (Douglas Carmichael) 607: Update Win32 and DOS ports. 608: Add Polish documentation by Tomasz Sienicki. 609: 971116 Add a missing return statement in filesys.c (Brian King) 610: 971112 Fix silly error in xwin.c. 611: 971108 Add some more decompressors to zfile.c. Set FLOPPY_SPEED to 4. 612: 971107 Make ggi.c compile again, fix a few bugs and add Picasso support. 613: 971106 Fill prefetch queue after executing an encrypted ROM illegal insn. 614: 971105 Initialize the sprites during customreset now that init_hardware_frame 615: no longer does it. 616: 971104 Implement action_set_file_size as required by the specs. 617: SetPanning/picasso96_state: XOffset and YOffset are signed. 618: 971103 xwin.c: Offer 640x512 and 320x256 resolutions in windowed Picasso mode. 619: Add a few more memory banks in memory.c (A3000 stuff - nonfunctional) 620: Don't map kickmem at address 0 if using replacement Kickstart. 621: Must fill prefetch buffer when calling calltraps or ersatz functions. 622: Write out hardfile parameters in write_filesys_config(). 623: xwin.c: Support 15 bit displays. 624: 971101 Add latest patches by Samuel Devulder. Improve hardfile support in 625: tui.c. 626: 971031 Offer "-U" only ifdef PICASSO96. 627: Add support for mounting multiple hardfiles in any order. 628: Don't reset sprite state in init_hardware_frame. Instead, allow SPRxPT 629: functions to set the state to SPR_restart in the vertical blanking 630: area. 631: 971030 Fix 8MB Chip RAM. 632: filesys_dev_storeinfo: New function and calltrap. Use it to store 633: per-unit data (which now includes geometry information) in the 634: parampacket. Modify filesys.asm to use it instead of trying to find 635: the values by itself. 636: 971028 "AMIROMTYPE1" has 11 characters, not 0. 637: 971027 Implement encrypted ROM features. Comment out non-working InvertRect 638: optimization. 639: 971026 Fix some problems in PlanarToDirect and CopyColorIndexMappingA2U. 640: 971022 Remove fixed paths in genp2c.c (oops...) 641: Some more XF86VidMode hacks. 642: Include target.h in sysdeps.h. Define PICASSO96_SUPPORTED for X11, 643: SVGAlib and Win32. Don't compile in Picasso support if this macro isn't 644: defined. 645: 971021 xwin.c: Picasso code wants to use bit_unit, not bitdepth. Also correct 646: byteorder (it's B8G8R8A8 from the Amiga's point of view). 647: Fix Picasso InvertRect function. 648: Half-hearted attempt at 24 bit support in the X11 version, with some 649: awful code in custom.c. 650: Make the X11/VidMode version try to resize the screen for Picasso mode 651: switches. 652: 971020 Delete option -5, make -4 take an argument. Disable z3fastmem and 653: Picasso emulation when using a 24 bit address space. 654: 971019 record_color_change: No need to remember a color table if the current 655: line hasn't been decided yet. 656: Add XF86VidMode extension support. 657: Integrate Stefan's Zorro III support, add support for up to 64MB fast. 658: xwin.c: Refresh the whole Picasso screen if there are Expose events. 659: Fix module length calculation for *Tracker modules. 660: Use a 32 bit address space by default if CPU_LEVEL > 1. New option -5. 661: 971018 Apparently it's not possible to turn on bitplane DMA in the middle of 662: the line, so call post_decide_line only when the number of planes is 663: modified, not in DMACON. 664: genp2c.c: New file, generates planar to chunky conversion routines. 665: Remove those which are now generated from custom.c. 666: 971017 Make row_map an array, not a pointer to one. 667: Audio periods may not be 0, so initialize them to 65535 in customreset. 668: Implement "S" command in the debugger. Add a bunch of module detection 669: routines. 670: 971015 od-linux/sound.c: If FRAME_RATE_HACK, reduce vsynctime a bit to give 671: the sound emulation more room to breathe. 672: 971013 Merge with Brian's latest Picasso code. 673: 971012 Fix a few more portability problems that showed up when compiling with 674: Watcom C. Also remove superfluous includes from some files. 675: Add md-i386-watcom directory. Update od-win32 with latest code. 676: Remove fast_memcmp routine. 677: Include penguin.h before xwin.h everywhere. Declare a few semaphores 678: for multithreaded setups in xwin.h. Move inhibit_frame stuff to xwin.h. 679: 971011 gencpu.c: Don't generate labels without following statement. Don't 680: directly generate assembly statements if X86_ASSEMBLY, instead generate 681: code that uses macros defined in machdep/m68k.h. 682: 971008 xwin.c, svga.c, custom.c, uaelib.c: Add picasso96 code. Add Picasso 683: selection to configure. 684: Move common variables (mouse stuff, gfxvidinfo, xcolors) from graphics 685: files to custom.c. 686: memory.c: custom chip area isn't supposed to be mapped at 0x200000. Use 687: dummy bank for now. 688: od-linux/joystick.c: Open different devices for different joysticks. 689: From Win32 sources: set vsyncmintime in m68k_go. 690: 971005 genblitter.c: blitzero is initialized to 1, not to 0, so we need to 691: clear, not set it. 692: 970928 Add another user interface for X11 using libgtk+. 693: 970925 audio.c: For stereo sound, double all lookup table values. 694: Added (preliminary) win32.c. 695: 970923 DISK_check_change: Should call gui_filename when ejecting. 696: 970922 Fix od-linux/joystick.c compilation problem with recent kernels. 697: Merge results of the UAE meeting: 698: New serial code (Stefan Reinauer) 699: Expansion "test card" support (Stefan Reinauer) 700: New GGI code (Christian Schmitt), also configure/Makefile changes 701: to make it compile. 702: Fix 68020/FPU code. 703: Call do_mousehack before every read of the mouse position to get more 704: accurate values. 705: Add sd-file directory; add support for it in configure. 706: Add stereo support in audio.c, od-linux/sound.[ch], and main.c. 707: Fix copper problem: call decide_line in calc_copcomp_true; and if the 708: copper wouldn't wait normally (position reached), but the hardware is 709: fetching data in Hires 4 plane mode, delay until ddf stop. 710: t-ascii.h: Asciiart target understands "-x" parameter. 711: 970823 gencpu.c [X86_ASSEMBLY]: Turn input constraints for scratch into output 712: constraints. 713: 970626 Fix 24 bit address space logic (Mathias Ortmann). 714: Raise exception 3 for Bcc.L insns if CPU_LEVEL < 1. Reported by Stefan 715: Ropke. 716: Remove USER_PROGRAMS_BEHAVE for normal emulator. Remove hacks to 717: generate expanded CPU functions for specific opcodes. Replace with code 718: that reads an instruction count file and sorts gencpu output by 719: frequency of the called opcodes. 720: 970625 Calltraps (at least those with call_m68k calls in them) were broken by 721: the CPU emulator changes. Now use Axxx opcodes as calltraps. 722: Include pOS port by Samuel Devulder. 723: 970624 Solaris sound patches by Manfred Thole. 724: gencpu.c: The generated code shouldn't just check for odd addresses, 725: it should also call Exception... 726: 970622 gencpu.c: don't emit return statements; make code jump to a function 727: end label instead. Make genamode() check for odd addresses, also reduce 728: code duplication. Interpret getv parameter so that data is fetched if 729: getv == 1, and the address is checked if getv != 0. Remove exception 3 730: checks from get_long() etc. Get rid of buserr. 731: cpuopti.c: Make a few more checks whether optimization is possible. Add 732: code to undo GCC's epilogue scheduling. 733: newcpu.c: Split all the specialflag handling into new function. 734: Implement -w option as in Win32 port. tui.c: add it to misc menu. 735: 907621 Fix yesterday's CPU emulator changes. Make cpuopti aware of functions 736: with two RET insns. Make sure that NO_EXCEPTION_3 and 737: NO_PREFETCH_BUFFER are on when emulating a 68020. 738: Add usage information for src/dst operands to table68k. 739: Implement CIA A OVL bit properly. 740: 970620 New NeXTwin.m from Ian. Fix a number of prefetch bugs in gencpu.c. 741: newcpu.h: m68k_incpc(), get_i{byte,word,long}: New function/macros. 742: Also add oldpc parameter to m68k_do_[bj]sr(). gencpu.c: use all this. 743: CPU functions now don't use nextiword; they keep the PC constant and 744: increment it with the right value at the end. 745: If chipmem_size > 2MB, disable fastmem. 746: In configure.in and od-linux/joystick.c, use ljoystick.h as name for 747: the replacement file for a broken <linux/joystick.h> 748: configure.in: Check for post-gcc-2.7.2 option "-mno-schedule-prologue" 749: and use it for compiling the optimized CPU files if it exists. 750: 970618 Rename hardfile.device to uaehf.device to avoid name clashes. Suggested 751: by George Nicolaidis. 752: Reimplement CPU prefetch by making gencpu aware of it so it generates 753: code that fetches data from the right places automatically. Also reduce 754: prefetch length to 4 bytes. 755: 970612 memory.h: Include osdep/memory.h. missing.c: implement xmalloc(). 756: Throughout: use xmalloc instead of malloc. 757: Apply changes by Sam Devulder. 758: 970608 Fix ncurses.c so that it links. Fix configure so that it can be enabled 759: Get rid of spixstate by implementing a new sprite drawing method. 760: 970607 do_copper_cheat(): New function, Implement inaccurate, line-based 761: copper as an option (-C). calc_copcomp_true(): Fix possible hr 4 plane 762: bug (split the wait into several pieces in this mode). 763: Allow 0 bytes for fast and bogomem (to override .uaerc) 764: 970606 Fix some bugs in transdisk, and make it compile with SAS C (Joanne Dow) 765: Stupid compilation fixes: Get rid of "==" in shell scripts, check that 766: B300 and friends are defined in serial.c before using them. 767: Implement blitter prefetch. Fix silly typo in main.c that prevented 768: immediate blits. 769: init_ham_decoding(): new function, handles case where the first few 770: pixels of a HAM display are outside the window. 771: 970605 disk.c: Get rid of need_read. Save cylinder/track number for the data 772: in the MFM buffer for each drive. Force rereads on disk insert and 773: write operations. In drive_fill_bigbuf(), do nothing if saved data 774: matches current. Fix disk change logic: give the program time to read 775: the change bit a couple of times before inserting a new disk. 776: custom.c: Fix really subtle bug: when deciding a line as border, 777: plfstrt/plflinelen were not set up properly, which could confuse 778: post_decide_line(), and eventually there would be lines decided as 779: playfield with plfstrt == 0, plflinelen == -32 ==> Bang. A related 780: bug was that setting DDFSTRT to 0 would result in both plfstrt and 781: plfstop having value 0. 782: gfxutil.c: Fix bug that could cause color values to go out of bounds. 783: xwin.c, svga.c: When dithering, color values were not doubled properly. 784: 970604 filesys.c: when deleting an a_ino, remove it from hash table (Mathias 785: Ortmann). In do_find(), if creating a file failed and we didn't have 786: an a_ino before, call delete_aino on the one we created. 787: More configure changes: Add $X_PRE_LIBS and $X_EXTRA_LIBS in some 788: places. See if MIT-SHM support for X will compile and link. Always 789: compile it in if it will. Turn DONT_WANT_SHM and LOW_BANDWIDTH into 790: run-time options (-T, -L). Use single line buffer when dithering and 791: in low-bandwidth mode (but in all cases there's a full buffer for the 792: X image). 793: memory.c: Don't map in custom regs at 0x200000 if chipmem_size > 2MB. 794: Split up init_sound(): new function setup_sound(). Make "-B" accept 795: minimum and maximum values as in Gustavo's latest patch. Add code to 796: tui.c to support changing buffer size, number of bits and frequency. 797: 970603 If using compiler, define NO_EXCEPTION_3 and NO_PREFETCH_BUFFER. 798: (bugreport by Raphael Poss). 799: GGI support from Christian Schmitt/Stefan Reinauer. More serial patches 800: from Stefan. 3 button joystick patch from Mathias. 801: Add linemem field to gfxvidinfo, for systems that don't want to keep 802: a buffer of the full screen. Adjust custom.c to use it instead of 803: bufmem if it's not NULL (mostly by Mathias). svga.c, xwin.c: Use single 804: line buffer in some cases. xwin.c: If USE_DGA_EXTENSION, undef 805: LOW_BANDWIDTH. Move test of DONT_WANT_SHM after inclusion of config.h 806: Make configure.in a bit more structured. 807: Move colormodes array to main.c. 808: 970602 gfxlib.c: Start implementing layers workalike. 809: md-i386-gcc/support.c: Don't mix sleep and alarm. (Petter Schau) 810: 970601 Remove may_run_compiled. Print error and abort if m68k_go is called 811: twice. 812: Fix yesterdays bugfix in svga.c. 813: Change FRAME_RATE_HACK so it tries to average the times a bit. 814: Re-implement a call_m68k function in autoconf.c. The new version 815: works by doing some heavy setjmp/longjmp magic and switching stacks. 816: With this, it's even possible to call things that Wait(), so use 817: the new functionality in gfxlib.c to call LockLayer() in WritePixel(). 818: 970531 Turn some inline functions in memory.h into macros. Get rid of 819: get_along and get_aword. Implement CIA A OVL bit. 820: main.c: Add missing close_sound() call. 821: Remove INTEL_FLAG_OPT, use X86_ASSEMBLY for everything. 822: tui.c/svgancui.c: Implement scrollable menus. Add code to select 823: video mode from a given list. Add "graphics_setup" call to graphics 824: specific files. Use it in svga.c to build a video mode menu. 825: Fix bug in svga.c that would crash non-linear modes. 826: disk.c needs to include gensound.h and sounddep/sound.h. 827: Fix bug in fuzzy_memset. In pfield_init_linetoscr, always adjust ddf 828: values if lores hack is being used. 829: 970530 Apply some patches by Samuel Devulder. 830: 970529 filesys.c: In "Houston, we have a problem" case, do not only delete 831: the oldest ExKey, but also use it instead of random memory for the 832: new one. Attempt to reduce the number of lost exkeys by allocating 833: them in action_examine_next instead of action_examine_object. 834: Initialize the "uniq" field of unit->rootnode. 835: Attempt to speed up sprite drawing loop. 836: Implement sprite<->sprite collisions. 837: Get rid of -Wtraditional. 838: 970524 New config option NO_PREFETCH_BUFFER. Emulate 16 byte prefetch if 839: unset. 840: For exception 3, the stack frame seems to be: 841: (saved SR).w, (faulting address).l, (faulting instruction).w, 0.w, 842: (pc at fault time).l. Make everything that could generate this fault 843: handle it properly. That got Cybernoid II working! 844: 970523 Makefile.in: Make sure $(DEBUGFLAGS) is used for CPU emulation files. 845: memory.c: Don't map chipmem at 0x200000, use custom chip bank for now. 846: 970507 configure.in: Mutate ac_cv_c_inline=no to ac_cv_c_inline= 847: Check for tcgetattr, <sys/termios.h>, cfmakeraw, <sys/ioctl.h> 848: Check for sem_init in -lposix4 (Marc Zimmermann) 849: serial.c, cia.c, cia.h, serial.c: changes (Stefan Reinauer) 850: Make disk gap 360 (Michael Krause) 851: 970404 In configure, check for features.h before including it in test program. 852: 970401 New option -4. Bugreport by Mirko Viviani. 853: 970324 Petter Schau says the exception frame for Bus Error and Address Error 854: is 14 bytes, not 6, on the 68000. Change Exception() accordingly. 855: It's *** OUCH *** day! Make hardfile beginio return the correct value. 856: Also load execbase before calling ReplyMsg(). 857: It's *** OUCH *** day again! The filesystem asm code was an odd number 858: of bytes, leaving the hardfile code unaligned. 859: 970318 custom.c: Rename spron to sprst, and use new variable spron to indicate 860: for which sprites the sprite pointer registers have been written. Don't 861: do sprite DMA if spron==0. (The problem was that some programs write 862: SPRxCTL and then SPRxPT, which used to set the sprite state to 863: SPR_restart, but it must be SPR_waiting_start.) 864: Turn a few m68k_reset()s into uae_reset()s. 865: New variable nr_joysticks. Delete joystickpresent. New arg "nr" for 866: getjoystate() and read_joystick(). 867: New extra keycode AK_RCTRL. Handle multiple joysticks & different key 868: replacements in keybuf.c/custom.c. Make tui.c handle all the settings. 869: 970310 od-linux/joystick.c now compiles again without the joystick header. 870: Fix some problems in configure; also check whether [n]curses.h defines 871: attr_t. Keep count of warnings and errors during configuration and 872: print a hopefully helpful message at the end. 873: disk.c sanitized by Mathias Ortmann. 874: Change the replacement getopt around a bit. 875: execlib.c, timerdev.c: Files removed. This will never be compatible 876: enough. Also remove related code (several SPCFLAGs, etc.) 877: Make main.c a bit nicer by introducing additional defines for the t-* 878: files. Also canonicalize the .uaerc file on all systems, eliminating 879: lines starting with '#' in the process. 880: Remove code in autoconf.c to set up 8MB chip; move it to filesys.asm. 881: Try doing the copper end-of-line hack in copper_memory_cycles(). 882: Disable the copper speedup hack. 883: 970309 Add a small, dumb hash table to struct _unit to speed up repeated 884: aino lookups. Delete aino in action_delete_object(). Move children in 885: action_rename_object(). In action_examine_next(), call new function 886: lookup_child_aino_for_exnext() which behaves more or less like 887: lookup_child_aino(), except it uses new function get_aname() to 888: translate filenames. 889: More portability fixes. New links set up in configure: sounddep, 890: target.h. All machine-dependent and joystick code removed 891: from os.c, renamed to audio.c. os.h renamed to audio.h, joystick stuff 892: moved to joystick.h. New directory targets/. 893: missing.c: New file. 894: 970308 Rewrite filesystem. Now it keeps one unique "a_inode" structure for 895: every object in the filesystem. This contains Amiga name, native 896: name, Amiga permissions, lock counts and several links for building 897: a tree structure. Keys are now made only for files which need a fd 898: and a file position. 899: Treat invalid locking modes als shared. 900: Comment out the code using readdir_r because that function appears 901: to be broken in glibc-2.0.1. 902: Clear RES2 in all packets before handling them. 903: Give mousehack the same task priority as the input.device. 904: Fix (?) some silly bugs in disk.c. 905: 970307 Include some Acorn port code from Peter Teichmann. 906: Yet Another Serial Diff from Stefan. 907: 970306 filesys.c: Don't just crash if lookup_key() has no success. 908: action_find_write() is supposed to create the file if it does not 909: exist. 910: 970305 More DOS patches. Make the Makefile parallel-make safe. 911: svgancui.c: Do cyclic search if character is typed in file selector. 912: my_alphasort(): New function (work around glibc bug) 913: custom.c: Fix error in yesterdays change. 914: filesys.h: New file. 915: 970304 New serial patch from Stefan and Christian. 916: Remove the assembler code to generate inputevents. Instead, add a new 917: Amiga C program ("mousehack") which does the trick. New function 918: mousehack_helper() in custom.c, activated by calltrap 0xF0FF70. 919: Mouse default is now always normal_mouse, not follow_mouse. 920: Do disk change checks only every 1/2 second. disk.c: Reverse 921: nodskchange to dskchange, and make sure there has been a step with the 922: drive empty before calling drive_insert() from DISK_check_change() (but 923: this _still_ doesn't work. Arrrghh). 924: 970303 filesys.c: Initialize _all_ of struct _unit in startup_handler(). 925: Remove message_lock and friends and use new variables cmds_sent, 926: cmds_acked and cmds_completed to determine when to send dummy messages. 927: Changed examinekey handling so that "dir all" on a large tree doesn't 928: crash anymore (it used to reuse the top-level exkeys which were still 929: in use after a wraparound). Don't allocate examinekeys if we're 930: examining files. 931: Check for readdir_r in configure. Use a wrapper my_readdir() around 932: either readdir or readdir_r. 933: Call filesys_reset() also from filesys_diagentry(). 934: Get rid of most of uaelib.c. Now there isn't a library, just a calltrap 935: at 0xF0FF60. Changed uae_pragmas.h accordingly and recompiled uaectrl. 936: Added a bunch of "volatile"s in various places in filesys.c and 937: penguin.h. 938: Make the filesystem reset-safe in the threaded case (paranoia changes). 939: Check disk changes only every 1/10 second in vsync_handler(). 940: Add assembler code to generate InputEvents once every vblank to 941: filesys.asm. (Doesn't work yet.) 942: 970302 filesys.c: Get rid of DosPacket type, pass pointers into Amiga memory 943: instead. Use GET_PCK_* and PUT_PCK_* macros to access fields. 944: Start making filesystem code thread safe & multithreaded. Move static 945: variables from some functions into struct _unit, likewise the key 946: and exkey variables. Add code to filesys.asm to queue commands which 947: did not complete and to reply to them later. Do multithreading if 948: compiled with SUPPORT_THREADS. See source for details. 949: Change hardfile and uaelib code not to call 68k routines. Make emulator 950: traps more efficient (no need to save regs, don't need SPCFLAG_EMULTRAP 951: anymore). 952: 970301 Fixed ncurses.c so it works with currprefs. 953: filesys_reset(): New function, called from customreset(). 954: Changed source layout. Break up machdep into machdep, osdep and 955: threaddep. 956: penguin.h: Now supports arbitrary types in pipes. 957: Changed filesys.c so that no 68k code is called from native C code. 958: filesys.asm, filesys.sh: New files. filesys.asm contains some 959: startup code and the main loop for the filesystem. Several locks 960: are allocated by the assembler code for use by the C code. 961: DOS_AllocMem(), DOS_FreeMem(), make_dev(), build_parmpacket(), 962: filesys_init(): Functions removed. 963: 970228 ncurses.c: Fix problem in b/w modes. 964: finish_drawing_frame(): Don't need to draw lines in multipenguin mode. 965: Move all options into new struct uae_prefs. New variables currprefs, 966: changed_prefs. 967: svgancui.c, tui_filereq(): If user presses a key, start search at 968: current position, not at the beginning. 969: DISK_check_change(): new function, compares currprefs with 970: changed_prefs and handles diskchanges if they differ. Called from 971: vsync_handler(). 972: DISK_validate_filename(): new function, checks whether a file exists. 973: Modify svga.c and tui.c so that F11 now brings up a limited UI. 974: 970227 Reimplement Mathias Ortmann's frame rate hack: New option HAVE_RDTSC 975: in config.h. New typedef frame_time_t. New variables vsynctime and 976: vsyncmintime. New function machdep_init() in md-*/support.c which 977: sets up vsynctime in the md-i386-gcc and md-i386-linux cases. Use 978: all of those in events.h, do_cycles() to determine whether to spend 979: more time on the CPU emulation. 980: Add missing "int i;" to xwin.c, init_colors(). 981: Unroll some loops in sample_*handler(). 982: Improve quitting/resetting behaviour: Move quit_program from GUI 983: specific files to main.c. New functions uae_reset() and uae_quit(), 984: quit_program is now handled in vsync_handler() and m68k_go(). 985: New variable sample_evtime, used in customreset() and set up in 986: init_sound(). 987: 970226 Solaris sound update from Manfred Thole. 988: keybuf.c: Decided that it doesn't need locking after all. 989: Add a bunch of Gustavo's latest diffs. (I hate CR characters!) 990: Move gfxvidinfo from all the target-specific files to custom.c 991: More Win32 stuff: New function notice_new_xcolors(). 992: 970225 configure.in: In shell string comparisons, always put spaces around = 993: Add warning about standard SVGAlib being incompatible with 994: LinuxThreads. 995: os.c, SOLARIS_SOUND: Remove reference to obsolete variable smplcnt. 996: Two new variables sound8_base and sound16_base. 997: sysdeps.h: Split off some stuff into md-amiga/amiga-kludges.h. Also 998: make the main types now uae_u32 and variants instead of ULONG, etc. 999: Convert some files to use the new types. 1000: cia.c: use int/long instead of ULONG/UWORD/etc for most static vars. 1001: xdebug.c: Some cleanup. 1002: md-i386-linux/penguin.h: Fix bug. Add "chunks" parameter to 1003: init_comm_pipe. Use it to decide in write routine whether to signal 1004: the other thread if it's waiting. Add no_buffer parameter to 1005: write_comm_pipe_int. 1006: Start adding Mathias Ortmann's Win32 patches. 1007: md-i386-win32: New directory 1008: include/sysconfig.h.win32: New file 1009: custom.c: is_lastline: new variable, set in hsync_handler(). 1010: sysdeps.h: Make it possible for system specific code to say that 1011: POSIX/ANSI stdio/malloc functions aren't present and are defined 1012: by a system-specific file instead. In filesys.c, don't reference 1013: errno if DONT_HAVE_POSIX is defined. 1014: readcpu.h: Kludge around the fact that Messysoft C apparently doesn't 1015: handle enums properly. 1016: newcpu.c/sysdeps.h: Preprocessor magic for 64 bit constants to get 1017: around a problem with MS C. 1018: md-*/penguin.h: Add dummy definitions for semaphores. 1019: keybuf.c: Use semaphores to synchronize access. 1020: os.c, win32 sound stuff: add memcpy avoidance code. 1021: 970221 X86.S: Don't use .type ifdef USE_UNDERSCORE. 1022: Move thread support configuration to configure from config.h. Change 1023: pipe implementation to use POSIX semaphore functions. 1024: 970217 From Samuel Devulder: Fix off-by-one error in LINE_TO_SCR, EHB case. 1025: 970215 customreset(): Initialize current_colors.color_regs to invalid values 1026: [reported by Samuel Devulder]. config.h: Uncomment ECS_DENISE [reported 1027: by Mathias Ortmann]. Add ami-win.c and custom.c patch from Samuel 1028: Devulder. 1029: 970213 disk.c: Set FLOPPY_SPEED to 3 1030: custom.c: SPRxCTL_1(): if spron was SPR_restart set it to 1031: SPR_waiting_start. do_sprites(): don't do that. 1032: SPRxPT{H,L}: Call decide_line(). 1033: 970210 custom.c, X86.S: Do LORES_HACK for dual playfield screens. Plug memory 1034: leak in custom_reset(). Fix pfield_draw_line() not to do so many 1035: multiplications. Make line_changed a char array. In remember_ctable(), 1036: remember comparison results also if negative. 1037: In 8 and 16 bit graphics modes, use the high parts of the color values 1038: to duplicate the actual values. This saves a lot of code in the 1039: hdoubling routines. Add a special fill_line in the X86_ASSEMBLY case. 1040: Check for glibc-2 and libpthread in configure. 1041: options.h: In fuzzy_memset(), do rounding slightly more conservative. 1042: Add tons of new keycodes to svga.c 1043: In record_color_change(), also check if vertically outside DIW. 1044: New variable hdiwstate, to represent the horizontal DIW state machine 1045: (thanks Petter!) Handled in reset_decisions(), finish_decisions() and 1046: decide_diw(). 1047: 970209 Change ncurses.c not to do so many multiplications/divisions. Add bold 1048: character support. Add color support. Make configure recognize 1049: --without-svgalib and enable ncurses.c in that case. 1050: 970208 X86.S, custom.c: Do LORES_HACK for EHB screens also. 1051: New function write_log(); convert some fprintf(stderr, ...) calls. 1052: 970207 Argh, Linux/i386/ELF check in configure got broken by latest sh-utils. 1053: In disk.c, drive_reset(), clear nodskchange flag. 1054: 970206 Sprite fix from Mathias Ortmann: _really_ get rid of SPR_vtop, and 1055: replace all uses by SPR_stop. 1.1.1.2 root 1056: 970204 Disk change fix from Mathias Ortmann. 1057: 970203 New file ncurses.c, from Samuel Devulder. 1058: Disabled Exec stuff completely. Now it isn't even compiled in anymore. 1059: 970129 Changes from Samuel Devulder for Amiga and Christian Bauer for BeBox. 1060: 970122 Reorder the set_planes_unaligned functions so that better code is 1061: generated. Remove the or_planes_unaligned functions, add a parameter 1062: do_or to the set_planes functions and rely on the compiler to optimize 1063: the additional if away. 1064: post_decide_line() can now turn DMA off early. 1065: Make struct flag_struct 64 bits large for i386 systems, so we can use 1066: long move insns without clobbering the X flag. 1067: "make clean" should delete machdep/support.o. 1068: Maybe we do need SPR_vtop after all - undo 970120 change. Instead, 1069: comment out the code that clears the pos and ctl words at vsync time. 1070: In xwin.c, restrict window height to 300 or 600 depending on linedbl 1071: mode. 1072: 970121 Get rid of SPCFLAG_BLIT, call do_blitter() directly. 1073: In finish_decisions(), don't check for sprite draws in the border. 1074: Reimplement the SMART_UPDATE define so that the memcmpy() overhead can 1075: be avoided by not setting it (but this usually loses). 1076: Detect 7 lores planes in BPLCON0() and change to 6 to avoid memory 1077: clobbers. 1078: In pfield_draw_line(), do memcpy linedoubling after handling the line 1079: to be doubled, not when we see a LINE_AS_PREVIOUS. Get rid of prev_ypos 1080: parameter. 1081: Add special code for bpldelay1 == bpldelay2 to the UNALIGNED_PROFITABLE 1082: code. 1083: In pfield_init_linetoscr(), when delays 1 and 2 differ, clear the areas 1084: which are covered only by half of the planes at the start and end of 1085: the DDF area. 1086: Replace FASTER_CPU and LINUX_SOUND_SLOW_MACHINE with M68K_SPEED. 1087: Replace fast_unaligned_memset macros with fuzzy_memset; use this also 1088: in pfield_init_linetoscr(). 1089: Use real (PAL) vblank end in do_sprites(). 1090: 970120 Mask out high byte in DDF{STRT,STOP} 1091: Oops, forgot about long frames while calculating the bounds of all the 1092: maxvpos*2-sized arrays. Increment them all by two. 1093: Try to detect empty sprites before recording them for the drawing code, 1094: but be careful not to hurt attached sprites. 1095: From Paul Kienitz: hack for keyboard code in cia.c to work better with 1096: newer Kickstarts; small cleanup in cpuopti.c. 1097: Undo yesterday's change in copper_memory_cycles(), instead call 1098: decide_line() each iteration and test whether copper_cycle_time == -1 1099: to determine whether we're vertically in the DIW. 1100: calculate_copper_cycle_time() could get astronomical plane counts due 1101: to missing masking operation. 1102: No need to put the copper to sleep for one memory cycle after a WAIT, 1103: merge that with the two cycles for the next instruction. 1104: Get rid of SPR_vtop, no longer needed (it was unnatural anyway) 1105: 970119 Implemented alternate pfield_doline functions that access the apixels 1106: array at unaligned addresses; conditional on UNALIGNED_PROFITABLE. 1107: Move LORES_HACK from config.h to custom.c. 1108: Fix loop exit test for pfield_linetoscr_hdouble_asm8(). Clear the 1109: right parts of apixels in pfield_init_linetoscr() when LORES_HACK is 1110: enabled. 1111: Get rid of bordercol array: the decision-making code should handle 1112: lines in the border correctly by setting line_changed. 1113: Disable bitplane DMA if more than 4 hires bitplanes are set in BPLCON0. 1114: Fix some more copper timing inaccuracies: used to test 1115: thisline_decision.which without calling decide_line() before. 1116: 970118 Now remember BPLCON1 changes during data fetch. New functions 1117: decide_delay(), pfield_adjust_delay(); new fields in struct drawinfo 1118: and new variables next_delay_change, delay_changes. 1119: Forgot to set copstate to COP_morewait in one case in do_copper(). 1120: Add a kludge that checks for copper position 0xDC and changes it to 1121: 0xDE. 1122: Restore equality test for vertical sprite positions. Disable sprite 1123: DMA before the end of the vertical blank area. 1124: Mask not only the positions, but also comparison values in 1125: calc_copcomp_true(). 1126: 970117 Change custom_bput to put the byte value in each half of the word that 1127: is put. 1128: 970116 disk.c: Set the gap between sector 11 and 0 to 380 words. Andromeda's 1129: "Seven Seas" demo expects at least 320. Make the KILLEHB bit 1130: conditional on AGA_CHIPSET. Decide BPLCON1 (scrolling delays) in 1131: decide_diw(). Fix off-by-one errors in the very_broken_program case 1132: of decide_plane(). Fix aspect calculation when horizontal resolutions 1133: differ. Simplify copper horizontal comparison logic slightly. 1134: Fix divide by zero errors when reading from empty floppies. 1135: 970115 pfield_draw_line(): Check for frame_redraw_necessary also for border 1136: lines. New global variable pfield_last_drawn_line, used to implement 1137: clever Y centering. Make the KILLEHB test conditional on ECS_AGNUS. 1138: Move joystick and sound initialization before gui startup. Adapt tui.c 1139: so it understands the gfx_requested_* options. 1140: We can't always avoid clearing pixdata.apixels if DIW is larger than 1141: DDF. There may be sprites in the area outside DDF, but inside the DIW, 1142: so we can't just adjust linetoscr_diw_{start,end} blindly in 1143: pfield_init_linetoscr() [Sanity Arte demo]. 1144: 970114 Bebox diffs from Christian Bauer. 1145: Support for Symmetric Multi Penguins under Linux (start second thread 1146: with clone and use some pipes for communication - rather simpleminded 1147: for now). Rework do_flush_line() and do_flush_screen() so that X 1148: libraries are only called from the first thread. 1149: Eliminate valid field from struct decision (set which to -2 instead of 1150: valid = 0) 1151: Fix definition for gui_filename() in xdebug.c to use const char * for 1152: the second parameter. Remove CRs from disk.c (now how did they get in 1153: there??). Likewise for cpuopti.c. 1154: 970113 When post_decide_line() detects mid-line DDF change, must adjust 1155: bitplane pointers in decide_plane() also. Make plfstrt the earlier a 1156: position in this case, and adjust bitplane pointers/modulos 1157: accordingly. 1158: current_hpos() isn't what GCC calls a constant function. 1159: Replace line_drawinfo[current_change_set] with new variable 1160: curr_drawinfo throughout. Likewise for new variables prev_drawinfo, 1161: {curr,prev}_{color_{change,table}s,sprite_positions}; all initialized 1162: in init_drawing_frame (). Turn line_data into a 2dim. array, do some 1163: CSE/SR on line_data[lineno] in pfield_doline_{l,h}() and 1164: decide_plane(). Make some custom registers unsigned int instead of 1165: UWORD. 1166: calc_copcomp_true(): Masking with 0xF000 isn't likely to give 0xC0000. 1167: Comment out bplcon{3,4} from struct decision. 1168: 970112 filesys.c, action_set_protect(): if __unix is defined, check whether 1169: user wants to change directory protection, and if so, force rwx to be 1170: the new protection. 1171: Force sane window sizes in xwin.c, and set width/height from mode specs 1172: in svga.c. 1173: Implement LORES_HACK compile-time option. 1174: Whee, there _are_ broken programs that enable bitplane DMA in the 1175: middle of the data fetch area, draw half a picture on the first line 1176: and the full picture on all subsequent lines. New function 1177: post_decide_line(), called from BPLCON0() and DMACON() which tries to 1178: detect this case and turns the decision from border to playfield if 1179: necessary. 1180: New arrays, sprvstart and sprvstop, calculated in SPRxPOS and SPRxCTL, 1181: and used by do_sprites. Handle sprites with vstart == vstop correctly. 1182: Sprite vpos comparisons are now done with <= , not ==. 1183: _Now_ the X86-specific linetoscr routines do their masking correctly. 1184: configure: If we have GCC, #define GCCCONSTFUNC as __attribute__ 1185: ((const)) 1186: 970111 Eliminate can_draw from struct decision. 1187: Change vertical DIW code; use enum of two states (waiting for start and 1188: for stop) which is updated in decide_line(). Lines during the wait for 1189: stop phase are decided as border. No data is fetched outside the 1190: vertical DIW. 1191: Optimize copper memory cycle calculation a little: the new variable 1192: copper_cycle_time is set to a time >= 0 if this time is constant 1193: (i.e. 2). Calculated whenever DMA, BPLCON0, or DIW state changes. 1194: blitter.c, genblitter.c, custom.c: For BLTBDAT, follow the HRM which 1195: says that immediate data is shifted as soon as it is written. This 1196: seems to be wrong for the A channel, though. Also, don't change prevb 1197: during blitter loops if DMA B is off (Sanity Arte). Turn FAST_BLITTER 1198: stuff into run-time option "-n", and eliminate the ancient blitting 1199: code (former FAST_BLITTER == 0). 1200: Big OOPS: Copper move checked copcon, but there was no COPCON function 1201: so it was always zero -> copper-controlled blitter never worked. As a 1202: speed optimization, recognize 00010000 copper commands and put the 1203: copper to sleep until the blit is finished. 1204: Slow down the blitter in decide_line() if bitplane DMA steals blitter/ 1205: copper/CPU cycles (but this code is commented out for now). 1206: 970110 New function pfield_init_linetoscr(). Move calculation of diw_end and 1207: diw_start values there from x86-specific code. Also use DDF values to 1208: determine where data in pixdata.apixels is valid. The x86-specific 1209: linetoscr routines now only do the alignment masking (and they now do 1210: it correctly, too). 1211: Got rid of the display-related contents of struct draw_info, make 1212: static variables and them initialize from pfield_draw_line() only when 1213: necessary in function pfield_expand_dp_bplcon (). Get rid of the four 1214: saved_* arrays, remember line_drawinfo instead now that it's smaller. 1215: Color compare avoidance didn't work because color_src_match didn't get 1216: set when a compare was successful. Fixed. 1217: Reorder parts of the decision-making code so that less redundant checks 1218: are performed. 1219: Add clever x centering. Change "-O" option parameter format. 1220: In md-i386-linux/support.c, memcmpy(), do memcmp() first, then switch 1221: to a plain memcpy routine once we find the first difference. 1222: Patches from Stefan Reinauer/Christian Schmitt for 1223: xdebug.c/configure.in 1224: New uae-ui.tk from Volker Zota. 1225: Add --enable-profiling to configure.in. 1226: Check for period==0 in AUDxPER() was accidentally commented out - 1227: revert. 1228: 970109 Some tweaks in X86.S, plus a new option X86_PPRO_OPT to use movzbl 1229: instead of xorl/movb when possible. 1230: Fix maxline calculation in finish_drawing_frame(). 1231: Make next_lineno always a valid array index. Decrease minfirstline to 1232: 21 (NTSC vblank end). New option -O to specify display parameters, 1233: supersedes -d and -C which will be removed eventually. Differentiate 1234: between line doubling and aspect correction. Put aspect correction code 1235: in custom.c. 1236: Remove some more unused stuff from custom.c. 1237: Fix Makefile problems when cpuopti isn't used. 1238: 970108 use_lores isn't initialized before custom_init(), setup lores_frame and 1239: friends in customreset(). Optimize spixstate clearing. Optimize border 1240: drawing with copper color changes (use fill_line instead of linetoscr) 1241: Fix the sprite DMA logic, it now works with the Interference demo. 1242: Silly bug: Sprite 7 never got displayed (introduced sometime last week) 1243: Implement USER_PROGRAMS_BEHAVE for normal (non-compiler) CPU emulation. 1244: Add most of Gustavo's last patch (new dir md-i386-dos) 1245: 970107 Finish rewriting most of the drawing code in custom.c/X86.S. See 1246: README.PROGRAMMERS and source for details. Improvements include better 1247: copper precision, better interlace support, hopefully much less bugs 1248: and not too many new ones. 1249: Remove calc_adjustment() and gfxvidinfo.x_adjust; replace with code in 1250: custom.c that looks at gfxvidinfo.follow_diwstop (new field) 1251: Some operations (like EHB playfields) now have a higher penalty than 1252: before, the assumption being that these are rare and the normal case 1253: runs faster if it doesn't have to care too much about exceptions. 1254: Fix sign mismatch bug in BPL1MOD() and BPL2MOD(). 1255: 970104 Move x_ShowEA() from xdebug.c in newcpu.c, rename as ShowEA(). 1256: 970102 Add check for XK_W to decode_es() in xwin.c. 1257: New uae-ui.tk from Volker Zota. 1258: 961218 Add a patch by Bernie Solomon to overcome a problem with K&R compilers. 1259: Change X86.S in a way that may make it either faster or slower. 1260: Fix ADF_EXT1. 1261: 961217 #ifdef XK_dead_*, since Solaris doesn't have it. 1262: 961202 Fixed joystick.h problem in configure. X11/DGA support from Kai 1263: Kollmorgen. 1264: 961117 Some fixes in svgancui.c. 1265: 961112 Gustavo's latest patch, with GUS support for DOS by Michael Sontheimer. 1266: Add menu browsing to tui.c/svgancui.c 1267: 961107 Fix m68k_areg and m68k_dreg macros (put args in parentheses) 1268: Initialize gfxvidinfo.maxlinetoscr in xwin.c (found by Sam Devulder) 1269: 961104 svgancui.c tui_refresh() routine tweaked to work with the non-standard 1270: ncurses that came with my new SuSE-Linux CD. 1271: 961031 Added Christian Schmitt's xdebug.c (disabled for this release) 1272: 961024 Module search function from Michael Krause. 1273: 961019 Add Samuel Devulder's Amiga patches. 1274: 961018 Don't use -lm on NeXTStep. 1275: 961017 configure checks for -lforms. 1276: 961016 Ian Stephenson's new NeXTwin.m. Make CLXDAT return 0 (no collisions) 1277: transdisk and transrom patches from Rodney Hester. Don't warn about 1278: 256K Kickstarts. Check diskfile sizes and try to determine whether it's 1279: a normal or a HD diskfile. No HD drive identification yet (is it 1280: different?) 1281: 961013 Corrected lots of bugs in the return address stack. Fix some errors in 1282: the compiler memory management. Use the "locked" flag for hash entries 1283: differently, and don't enforce it on a cache flush. Compile JMP to a 1284: constant address. 1285: 961012 Remove regs from cache before sign extending them for ADDA, MOVEA and 1286: similar things. Compile NOPs to nothing. Implement variable shifts/ 1287: rotates. Most shifts clear the V flag: Handle this in the compiler by 1288: some extra magic sequences after shifts or in the CC_AFTER_RO case in 1289: flush_cc_cache(). 1290: Implement a stack containing return addresses and their hash entries, 1291: and compile JSR (to constant addresses)/BSR/RTS instructions inline. 1292: 961011 compiler.c: Several bugs introduced in the last two days fixed. 1293: New function compile_note_modify() which is used when an ea is 1294: modified. This replaces calls to storeea with eaino_s == eaino_d. 1295: Bah, the 386 doesn't set flags on a NOT. Use CC_TEST_REG again. 1296: Do flag life analysis until no more changes are found. 1297: Make sure all "data" parts of EAs are locked exclusively (esp. for 1298: LEA, PEA) 1299: 961010 compiler.c: Finish yesterdays changes. Get rid of reg/offset 1300: bookkeeping in m68k_compile_block(), keep the information in struct 1301: ea_info instead. Rewrite all instructions to use the new fetch and 1302: store routines. 1303: Don't force a byteorder conversion in sync_reg() if the reg doesn't need 1304: to be written back. Don't do a SAHF or equivalent in compile_condbranch 1305: if we are doing a BRA or BF. 1306: Don't use CC_TEST_REG for NEG and NOT (this was inefficient for NOT and 1307: wrong for NEG). 1308: 961009 compiler.c: Two types of locks for registers; shared and exclusive. 1309: Implement better EA fetching and storing routines. For pdi addressing 1310: modes, lock the necessary registers shared before calling prepareea. 1311: Unlock those registers again in generate_possible_exit(). 1312: 961008 MOVEM instructions in the compiler to an absolute address could have 1313: caused random memory corruption. 1314: 961007 Spanish keyboard support for X by Ruben Martinez. Some configure fixes 1315: (don't assume current dir is in path, don't assume ncurses.h exists if 1316: libncurses does). 1317: If we're calling 68k code from an emulator function, don't use compiled 1318: code. 1319: Add a compilation option USER_PROGRAMS_BEHAVE (for compiler). Make 1320: compiler hash table larger. 1321: 961004 Fix typo in configure.in. 1322: Modify the compiler so it doesn't ever throw away code, unless it's 1323: necessary for a cache flush. 1324: Didn't quite squash the compiler flag bug yesterday. Try again. 1325: In filesys.c, do_find(), add a parameter fallback that indicates that 1326: the open should be retried with O_RDONLY if it failed with O_RDWR. 1327: Use this in find_input(). This fixes problems with r/o files on r/w 1328: volumes. 1329: 961003 Play around with the x86 specific memory access functions. Problem: 1330: __attribute__((regparam)) isn't reliable, and it seems that that's a 1331: rather hard to fix bug in GCC. Add some alternatives to 1332: md-i386-linux/memory.h; these aren't used yet though. 1333: Huh, the compiler was completely wrong about the flag needs of 1334: conditional instructions. I wonder how it could work at all... 1335: Remove a and d arrays from struct regstruct; replace with array regs 1336: with 16 elements. This allows better code for get_disp_ea() 1337: 961002 BeBox diffs from Christian Bauer. Move all code from main() to new 1338: function real_main(). 1339: 961001 Use do_get_mem_*() functions for nextiword(), nextilong() and new 1340: function nextibyte(). 1341: 960930 genblitter optimizations: Don't generate expressions in genblitter, use 1342: a second program called gengenblitter to generate a table of 1343: expressions for genblitter to use. gengenblitter knows about a lot of 1344: rules to simplify expressions. Some other tweaks to make GCC output 1345: better assembly for blitfunc.c. 1346: 960927 Turn off LED() in xwin.c (may fix SunOS joystick emulation problems) 1347: Call XFlush() and XSync() when exiting. 1348: Add some DOS patches. 1349: Add hooks in filesys.c to support filesystem units in a GUI. Use them 1350: in tui.c. Bugfixes in svgancui.c. Change the way tui_refresh() works. 1351: New option "-A". 1352: 960925 For read-only filesystems (CD...), use O_RDONLY in action_find_input. 1353: 960923 Save options file even if rename fails. 1354: Mirror address space every 16MB like we used to (got turned off by 1355: accident). 1356: Move some checks out of options.h into configure. Set up md-* 1357: directories with machine-dependent stuff in them; configure creates 1358: a link called machdep to the appropriate one. 1359: 960918 Solaris sound update from Manfred Thole. 1360: 960917 Add a timer.device replacement (buggy). 1361: 960916 Tristan Cavelier's french keyboard patch. 1362: Apply ReadTheSpecsBeforeYouWrite(tm) technology to AddPort(). 1363: Implement RemPort(), GetCC(). 1364: 960915 Tighten up some of the return values in Exec. 1365: 960914 EXEC: Implement RemTask(), ObtainSemaphoreList() and 1366: ReleaseSemaphoreList(). Set up more fields of the ExecBase properly 1367: and Enqueue it in the library list. Make InitResident call AddLibrary 1368: when appropriate (otherwise, intuition is never put into the library 1369: list). Fix major braino in Obtain/ReleaseSemaphore() (tried to Enqueue 1370: tasks into two lists). Make OpenLibrary() honor the version parameter. 1371: 960913 EXEC: Idle task now runs in user mode (abuse 0xFF0D as second STOP 1372: instruction). Make EXEC_IntTrap perform the RTE before trying to 1373: schedule. 1374: 960912 Solaris sound didn't compile. So much for making blind changes without 1375: being able to test them. 1376: 960911 As soon as you compile it under Solaris, it bombs... Now check that the 1377: hardfile exists before determining the size in main.c. 1378: execlib fixes: Set master DMA enable bit at startup, fix interrupt 1379: code to survive an interrupt that occurs before the proper SetIntVector 1380: call. Implement RawDoFmt(), CloseDevice(), CloseLibrary(), 1381: FindResident(). 1.1 root 1382: 960905 Lower 8 bits of POTGOR should be 0. 1383: 960904 Implement MULU, MULS, ADDX and SUBX in compiler. 1384: 960903 ncurses interface works. DOS people: please use the new portable tui.c 1385: and turn dos-ui.c into a frontend for conio.h, similar to svgancui.c 1386: In xui.c, unlink the temporary files ASAP. 1387: Two new command-line options, "-o" and "-G". 1388: Get rid of sz_unknown. 1389: compiler.h, compiler.c: New files. 1390: filesys.c, action_read(): Added a hack to test whether the first four 1391: bytes read from a file a file contain 0x000003F3 (HUNK_HEADER), in 1392: which case there's probably a LoadSeg() call going on and we ought to 1393: flush the cache. Also, use the fact that memory is now linear. 1394: 960902 Struggle a bit with dos-ui.c, rename it to tui.c (text-based ui), add 1395: new file svgancui.c (ncurses ui) and try to make these two understand 1396: each other. 1397: 960829 New transdisk from Stefan Ropke (can write images) 1398: Apply Patrick Hanevolds BeBox patches. 1399: 96082x many changes to Exec. Still not booting (DMACON seems not to be set up 1400: properly, and sometimes Signal is called with task==0). 1401: 960821 Solaris sound from Manfred Thole. 1402: 960820 In filesys.c, get_fileinfo(), return volume name for top-level 1403: directory. 1404: 960819 Reimplement WritePixel(). Layer locking is still missing. 1405: 960818 Reverse meaning of -x flag for X11. More Exec code. 1406: Several changes to the main CPU loop and the way call traps work. 1407: 960817 Implement some simple Exec functions (Lists, AllocMem & friends) 1408: 960816 filesys.c fixes from David Varley. 1409: 960815 Implement graphics library region functions. 1410: 960814 Modify memory layout: All memory is linear now; get_real_address() 1411: returns a (UBYTE *). 1412: 960810 Add Gustavo's latest DOS code. 1413: In configure, include <sys/param.h> earlier for the sake of some *BSD. 1414: Add three options for sound code (bits, frequency, buffer size). 1415: Make -S option take a parameter; remove SPEED_UP_SIDMON_MODULES. 1416: New version of transrom (return 0, duh). 1417: 960805 Fixed the blitter line mode emulation (this time, really! I hope.) 1418: Thanks to Fabio Ciucci for a detailed description of the correct 1419: behaviour. 1420: 960730 Spent the last three days trying to make the filesystem emulation 1421: bootable by adding an expansion device for it. A fair bit of trial and 1422: error, but now it works! Even the hardfile now works with Kick 1.3, but 1423: you have to mount it manually after booting. 1424: hardfile.c, filesys.c, expansion.c and autoconf.c now are an 1425: entangled mess which badly needs to be cleaned up. 1426: We can now mount more than one filesystem even with Kick 1.3: we 1427: recognize them by param 2 of the startup packet, which has the same 1428: value as the dn_Startup field of the devicenode. 1429: Make Kickstart replacement work again (didn't like the CPU emulator 1430: changes) 1431: Include preliminary serial support from Stefan Reinauer. 1432: 960728 Fixed the scancodes for <> and #' keys in svga.c 1433: Apparently, CIA timer latches ought to be 0xFFFF after a reset 1434: (prevents Eagleplayer from crashing occasionally). #ifdeffed out a 1.1.1.2 root 1435: slightly weird call to INTREQ from RethinkICRB() 1.1 root 1436: 960727 Sound code rewritten again. Now uses events (five new ones). All 1437: modules seem to play correctly now. It's a bit slower, though. 1438: AF sound should work again, patches from Marcus Sundberg. 1439: 960723 ... and more CPU emulation bugs fixed by Herman. 1440: Added uae.library from Tauno Taipaleenmaki, plus support files in the 1441: "amiga" directory. 1442: Added some files for the XFree86/OS2 port from Krister Bergman. 1443: Reorganized all the directories. 1444: 960717 Add latest patch from Herman. 1445: More FAST_BLITTER options: 4 and 5. These finish the blit immediately. 1446: 960713 Call XAutoRepeatXXX() only if we know we want to change the state. 1447: 960709 Work around a few AIX bugs (ughhh... what a broken OS). New test for 1448: sys/select.h. 1449: In xui.c, don't try to communicate with a dead process. 1450: 960704 More bugfixes from Herman. 1451: Get rid of dummy_xlate(), use default_xlate() to abort with error. 1452: In svga.c, set correct_aspect to 0 if using lores. 1453: Change the Bcc exception 3 handling code a bit. 1454: Add new program "cpuopti" that takes gcc-generated assembly files and 1455: optimizes away unnecessary pushl/popl instructions on the x86. 1456: 960701 68020 exception handling & various bugfixes from Herman ten Brugge. 1457: In ersatz.c, use 0xFF0D instead of 0xF00D as trap opcode. 1458: Do MMU "emulation" (no-op MMU instructions, probably many cases decoded 1459: incorrectly). 1460: 960627 New file LIESMICH, with minimal German instructions. Somebody want to 1461: translate the docs to French, Spanish, Italian, all the Scandinavian 1462: languages, Hebrew, Polish, Dutch, Russian, Welsh, Gaelic and something 1463: [email protected] understands? (No offense meant - I'm sure there are 1464: brillant minds using AOL. There must be.) 1465: 960626 In zfile.c, make sure we delete the temporary file even if file 1466: decompression fails. 1467: Reimplemented sound support, this time following the HRM closely. Seems 1468: to work fine most of the time, but occasionally loses drum beats in 1469: TFMX modules. do_sound() is now unconditional, it calls 1470: flush_sound_buffer(), which is system-dependent. This breaks everything 1471: except Linux sound for now :( 1472: Add Herman ten Brugge's 68020/68881 emulation code. 1473: FreeBSD chokes if sys/mount.h is included without sys/types.h. Fix the 1474: configure script. 1475: 960622 More work on AGA code. 1476: Removed 2-bit display routines from custom.c. Ian: please use dithering 1477: code. 1478: Make slowmem and chipmem sizes run-time configurable. 1479: Mask out the high bits in BLT?PTH(). 1480: Add exception 3 checks to Bcc/BSR instructions. Seems to be not quite 1481: correct yet. 1482: Completely rewrote the disk emulation. We now create a big track buffer 1483: full of MFM data, used in a cyclic fashion by DSKBYTR() and friends. 1484: The current position in the buffer is determined by the cycle counter. 1485: DMA reads are now done in one piece. Two new event types: ev_diskblk 1486: and ev_diskindex. 1487: ARRGGHH! The RSI demo has _completely_ broken disk loading routines. 1488: They start disk DMA on a block while the blitter is still working on 1489: the same memory region. 1490: In zfile.c, unlink() the temporary file directly after opening it so 1491: it will be freed by the OS as soon as we close it (or crash... :) 1492: 960621 Merge pfield_sprite_h() and pfield_sprite_l() into one function. 1493: Split off new files blitter.c and blitter.h from custom.c. 1494: Fixed null-pointer dereference in zfile.c when filename has no 1495: extension. 1496: Add new field maxline to gfxvidinfo that prevents writing past the 1497: screen buffer if it has less lines than expected. 1498: Modify the 256 color allocation routine to set nb_cols to the same 1499: value first for r,g and b; allocate everything as usual and fill the 1500: lost colors afterwards with those that have the greatest difference. 1501: 960620 For some extremely weird reason, it is faster to simply call the 1502: assembly-optimized pfield_linetoscr_full routines _twice_ to double the 1503: line than calling the new (also assembly-optimized) 1504: pfield_linetoscr_full_double routines, at least when writing directly 1505: into the framebuffer. If someone has a good theory why, let me know 1506: about it. 1507: Drawing into memory is faster with the _double routines, but not 1508: noticeably (about 1%). 1509: Consequence: Removed the new pfield_linetoscr_full_double assembly 1510: routines again. 1511: Add a "--disable-gui" option to configure. 1512: Add Samuel Devulder's dithering code. Provide support for mono and 1513: 16 color X servers. 1514: Put an assembly version of DitherLine() in X86.S. 1515: New color mode (3) available, tells UAE to use 256 colors and dither 1516: the image. This is meant for people who have only a 256 color X server 1517: available, but it works with the SVGA version, too. It's much slower 1518: than normal 256 color mode, though. 1519: Use the color allocation algorithm from the dither code in 1520: alloc_colors256(), too. 1521: 960619 New files: zfile.c and zfile.h, from Samuel Devulder. These provide 1522: transparent decompression of various compressed file formats. Enabled 1523: by configure only for Unix systems. 1524: Eliminate dont_want_aspect. New option "-C", sets correct_aspect. Make 1525: video mode 3 640x480, video mode 4 800x600. Honor correct_aspect in 1526: 320x400, 640x480 and 800x600 mode with SVGAlib. 1527: Add fields maxlinetoscr and x_adjust to gfxvidinfo. 1528: Rename pfield_slow_linetoscr() to pfield_do_linetoscr(). Use new fields 1529: in gfxvidinfo to enable writing directly to the framebuffer in 1530: resolutions with <800 horizontal pixels. 1531: New function pfield_do_linetoscr_full(), called when sure that full 1532: line update is being done. 1533: Move some of the usage() information into some target_specific_usage() 1534: functions, defined in the various system-specific files. 1535: Support lower resolutions with LOW_BANDWIDTH, too. 1536: memcpy() within screen memory is dog slow. Prevent some common cases: 1537: 1. Use fill_line() to double a line in the border. 1538: 2. Add a parameter to pfield_do_linetoscr_full() to indicate whether 1539: the line has to be doubled. New functions 1540: pfield_linetoscr_*_double_slow() that simply do a memcpy(), plus 1541: optimized ones in X86.S that are much faster by avoiding the copy. 1542: Smart update code could get confused by lores modes when 1543: prev_max_diwstop changed from one frame to the next. New variable 1544: frame_redraw_necessary, which acts as a counter. It's initialized to 1545: 2 in interlace mode to handle that case properly, too. 1546: Move sprite priority checks into pfield_sprite_?(), move HAM 1547: calculations into new function decode_ham6() to simplify the various 1548: linetoscr functions. 1549: Fixed a bug that would cause sprites to get deformed in some cases 1550: when they overlapped (never seen it). 1551: 960618 New file: dos-snd.c, from Gustavo. 1552: 1553: 960617 Release 0.6.0 1554: 960617 Fix memory clobber in X86.S when diwfirstword==0. Eliminate double test 1555: for HAVE_SYS_MOUNT_H in sysdeps.h. 1556: 960616 Handle programs that don't wait for the blitter by finishing a pending 1557: blit as soon as a blitter register is written to (regardless of DMACON, 1558: to be safe). 1559: Call gui_led only if the LED changed. 1560: 960615 Fixed dual playfield sprite priorities. Optimized dual playfield code 1561: in X86.S. More work on AGA code. Get joystick status only in 1562: vsync_handler to speed things up. 1563: Make blitter fill more intelligent with table lookups instead of loop. 1564: 960614 DOS update from Gustavo. Pass real bitplane pointer to 1565: bpl_data_differs(), don't calculate it again. 1566: 960610 Fixed silly INTEL_FLAG_OPT bug: The new __asm__ expressions use 1567: regflags, which is of course _regflags on a.out... 1568: More BeBox fixes from Christian. 1569: Make the CIA bank bigger (A0..BF) and relax the CIA A/B selection 1570: checks to make broken software (Kingsoft's Emerald Mine, reportedly) 1571: happy. 1572: Default framerate is now 1. 1573: 960603 New BeBox code from Christian. Fix test for ELF in configure.in. 1574: 960529 genblitter.c: generate blitfunc as a function taking ULONGs and 1575: returning an ULONG. This makes custom.o about 4500 bytes smaller on the 1576: x86 by removing worthless size prefixes, and should increase 1577: performance (better for the cache, and the P6 reportedly doesn't like 1578: 16 bit instructions anyway). 1579: memory.c: Load Kickstart in one block and swap bytes after that if 1580: necessary (eliminates the need for the special bebox load routine) 1581: custom.c: Some preliminary work for AGA support. 1582: 960528 Saw a real awful piece of code that detects the right mouse button by 1583: reading a value from POT0DAT, counting up to 25, reading from POT0DAT 1584: again, and deciding the button is pressed if the value changed. Urgl! 1585: Added POT0DAT() in custom.c ... 1586: Gustavo's latest patches. 1587: 960525 Do index pulse handling in CIA_hsync_handler(). Make sure bit 12 in 1588: INTREQ gets set on an index pulse. Don't call DISK_Index() for now, 1589: since this breaks the Workbench. 1590: The CIA code should _probably_ not be resetting the bits in INTREQ. 1591: Disable this as an experiment, let's see what breaks. 1592: 960524 Z flag handling for ABCD/SBCD/ADDX/SUBX was wrong (must be clear if not 1593: zero, was set if zero). Some other flag bugs in these functions fixed. 1594: 960523 gfxlib.c update from Christian Schmitt. Some new Mac code from Ernesto, 1595: still not all of it, though. Italian keyboard from Alessandro Bissacco. 1596: 960522 Fix the CHK instruction. Implement middle mouse button. 1597: 960515 Fix some Nextstep problems in configure.in. 1598: 960513 New Makefile target "streifenfrei" to remove all output from configure. 1599: Use autoconf 2.7; 2.9 is buggy. 1600: 960512 Remove BIG_ALIGNMENT hack. Remove superfluous pfield_fetchdata() call 1601: in pfield_doline_slow_l() that resulted in garbage on the screen. 1602: Move blttbl array from custom.h to genblitter.c 1603: Use new lookup tables imm8_table and areg_byteinc in CPU emulation 1604: instead of ?: expressions. 1605: GCC appears to generate better code if the opcode is passed as ULONG to 1606: the CPU emulation functions instead of as UWORD. 1607: Modify some of the flag emulation __asm__ statements so that GCC uses 1608: less registers for them. 1609: 960510 Cast malloc() results to char *, for C++ compilers. 1610: 960509 Use test for GCC and __i386__ in options.h to decide whether to use 1611: intel flag optimization. 1612: Use test for wish4.0 to decide whether to use the GUI. 1613: 960507 New bebox.cpp from Christian. AF sound patches from Marcus Sundberg. 1614: 960505 Fix the configuration stuff so it works. Fix some bugs in gfxlib.c 1615: 960504 Use GNU autoconf scripts to generate Makefile and sysconfig.h. Add new 1616: files sysdeps.h that includes the right system header files. Remove 1617: amiga.h and Makefile. 1618: 960502 More Next code from Ian (new Uae.app directory, 2 bit drawing for 1619: custom.c). 1620: 960501 Allow the sound driver to set the frequency to a value that is slightly 1621: different from the requested one. 1622: 960430 Added AmigaOS port from Olaf Barthel. 1623: 960427 Fixed single playfield sprite priorities in X86.S. 1624: 960426 Added new NeXTwin.m from Ian. 1625: Turn off the blitter delay in line mode. 1626: Added a CPU optimization from Olivier Raoul: gencpu now reads a file 1627: "insncount" that can be generated by UAE during execution and generates 1628: specific functions for the most common opcodes. Not documented yet. 1629: 960414 Reimplemented the fast disk emulation in a more sane way. Also it's 1630: only done for reads now. (Plus, it no longer works :-/ Need to work 1631: some more on this). 1632: In blitter emulation, defer not only the setting of the bltfinish bit, 1633: but actually defer the whole blit for the number of cycles it will 1634: take. This makes programs the clear the screen while it is still being 1635: displayed look much better. 1636: Changed supervisor mode handling around a bit (was messy). 1637: New DOS code from Gustavo. 1638: 960412 Removed smart update method 2. Added Alessandro Bissacco's blitter 1639: optimization patch. 1640: Fixed some sprite problems: during hysnc, clear sprpos and sprctl, and 1641: don't do_sprites() in line 0. 1642: 960409 Be a little more intelligent about when to call pfield_linetoscr() from 1643: pfield_doline_slow(). It's only necessary if a color register changed. 1644: If it can be deferred to the end of the line, use the the _full 1645: variants. 1646: Some CIA TOD fixes. 1647: Better transdisk, accepts first/last track parameters. 1648: 960405 New function memcmpy(), inline assembler for the i386. Compares two 1649: memory blocks and copies them at the same time. Used in 1650: bpl_data_differs. 1651: 960404 French keyboard from Tristan Cavelier. Parallel port emulation from 1652: Stefan Reinauer. 1653: Turn bpl?pt and r_bpl?pt into arrays, simplify some code using them. 1654: 960401 Some parts of Ernesto's new Mac sources. 1655: 960331 filesys.c: Close filedescriptors in free_key(). get_path() sometimes 1656: read uninitialized bytes. 1657: Minor cleanups all over the place. 1658: unchain_mnn() leaves sibling intact, but clobbers next. customreset() 1659: didn't care about that and crashed. 1660: Use the __asm__("intel_flag_lookup") magic not only for DOS. 1661: Support 320x300 for X (without LOW_BANDWIDTH). 1662: Add new function fast_memcmp(), defined as memcmp if __i386__ is 1663: undefined, otherwise a slightly optimized inline assembly function. 1664: 960329 Some bug fixes in gencpu.c 1665: 960328 Stupid programs write odd playfield modulos into the hw registers... 1666: Eek! Sometimes, linedescr[] was indexed with a negative value... Add a 1667: vpos >= minfirstline check to pfield_may_need_update(). I really need 1668: to rewrite this in Oberon... 1669: Don't turn off SHM when LOW_BANDWIDTH is selected. 1670: Keep the nextpos and linepos variables from pfield_doline_slow_?() as 1671: globals and initialize them in pfield_may_need_udpate(). 1672: Modified alloc_colors256() some more. 1673: Try to make sprite DMA follow the Amiga DMA slots more closely by 1674: adding a currhp parameter to do_sprites() and calling it at the end 1675: of a scanline or in pfield_may_need_update(). 1676: 960327 Added Gustavo's latest DOS code, as far as possible. Modified X86.S to 1677: use a macro SYM() to create symbol names either with underscore or 1678: without. 1679: Fixed some very silly SVGAlib lores bugs. Move screen 8 pixels up in 1680: low resolution. 1681: Use diwstop instead of diwstart to caclulate the DIW for 320x200 modes. 1682: Reset it not only in calcdiw(), but also in init_frame(). 1683: Speed up 256 color allocation a little bit. 1684: 960326 New function flush_block() in xwin.c and svga.c. Called from 1685: do_flush_line() and (new function) do_flush_screen(). Modify struct 1686: vidbuf_description, new member maxblocklines that records the maximum 1687: number of scanlines that do_flush_line() should collect before calling 1688: flush_block(). If it is zero, then flush_line() is used as before. 1689: This makes the special case stuff for SHM unnecessary, so it's removed. 1690: Fixed a stupid bug in events code: Put ev_hsync in front of ev_copper, 1691: so that do_copper gets called with hpos == 0 instead of hpos == 227 at 1692: the beginning of the line. 1693: New function init_frame() that sets up some variables, called from 1694: customreset() and vsync_handler(). 1695: Move smart update code to pfield_doline() from pfield_doline_h() and 1696: ..._l(). Clean up these two with a macro DO_ONE_PLANE, removing 1697: duplicate code. 1698: Implement big blits and ECS chip ids. New configuration options 1699: ECS_AGNUS and ECS_DENISE. 1700: Implement MOVEP. 1701: 960325 Before calling pfield_may_need_update(), make sure the value that was 1702: written to a register actually differs from the previous one. 1703: New expansion code from Stefan Reinauer. 1704: Modify struct mem_notify_node to contain a pointer to a valid bit 1705: instead of the valid bit itself, so that the playfield code has to 1706: check only one flag to decide whether to draw the line (SU method 2) 1707: 960324 The chip_flags approach has several problems. A flag may cover an area 1708: in memory that spans two scanlines, or several scanlines may use the 1709: same area in memory. 1710: Provide two other, different methods for the smart update: 1711: 1. Save all the bitplane data for each line and compare the chipmem 1712: with the saved data to decide whether an update is necessary. 1713: 2. Use a chain of memory_notify_node structs instead of a single bit 1714: for flags. Each scanline owns a set of these structs and chains them 1715: into the right place, and the chipmem_?put routines walk the chains 1716: and set the modified bit for each element. 1717: Unfortunately, both methods are slower than the old one. Especially the 1718: first one has some advantages though: it speeds up programs that clear 1719: the screen each update and write the same contents back, and it 1720: can speed up double-buffering programs at full framerate, since it does 1721: not have to check whether the bitplane pointers have changed. 1722: Remember the first and last drawn line for each frame, and pass them to 1723: flush_screen. Use this in xwin.c for SHM. Modify SHM logic: if smart 1724: update is selected, call XShmPutImage() from flush_line() instead of 1725: from flush_screen(). But gather blocks of consecutive lines to avoid 1726: calling XShmPutImage() too often. 1727: 960323 "Smart update" method, based very loosely on Alessandro Bissacco's 1728: work, implemented: 1729: Add an array chip_flags that records any change made to specific words 1730: in chip memory. 1731: Move all the hardware registers relevant to playfield display into a 1732: new structure bplinfo. Remember the contents of this structure in 1733: struct line_description for every line on the screen. Avoid redrawing 1734: in pfield_doline if the screen memory has not been modified for the 1735: line to be drawn and if the information from the last frame matches 1736: that in the current frame. 1737: 960322 Duplicate all functions responsible for drawing (pfield_doline, 1738: pfield_doline_slow, pfield_sprite, pfield_*word_*res) and append _h 1739: or _l (ah, the golden C++ days, when we had templates... which never 1740: worked). Modify the *_l functions to support drawing in low resolution 1741: (i.e. 320x200). Change the options -d and -H to support selecting color 1742: mode and resolution, thereby getting rid of some SVGALIB_* options. 1743: Get rid of the old pfield_*word_*res functions. 1744: Collect information about the minimum diwstart value during the 1745: previous frame and use that in case 320x200 was selected to fit the 1746: image on the screen. 1747: Get rid of HAVE_JOYSTICK (breaks the "-J" option) 1748: 960321 Give up in op_illg() if opcode is 0x4e7b and the exception handler for 1749: it is NULL. In that case, it's a 68020 Kickstart. 1750: 960319 New Next code from Ian. SunOS target was still screwed (two CFLAGS). 1751: Turn -x option into a "hair cursor" option. Support multiple mounts, 1752: but warn if more than one is seen (still does not work with Kick 1.3). 1753: Integrated Samuel Devulder's new LOW_BANDWIDTH patch. 1754: 960318 Integrated the start of Markus Gietzen's gfx library replacement. 1755: Changed the interface between custom.c and the graphics code. There has 1756: to be a buffer holding the data of all the lines on the screen. For 1757: linear SVGAlib modes, this is the video memory. Eliminated -x option, 1758: prepare_line(), DrawPixel() and the X_?BIT_SCREEN options, replace with 1759: 3 versions of pfield_linetoscr for different datatypes. Implement 1760: proper handling of the borders above and below the DIW in custom.c 1761: instead of in flush_screen(). 1762: Move common code (color allocation) from all the graphics files to new 1763: file gfxutil.c. 1764: Swedish keyboard from Per Olofsson. 1765: 1766: 960315 Release 0.5.0 1767: 960315 Newest DOS port from Gustavo. New sgi-gcc target. 1768: 960314 Add -lXext to osf target. Use "rb" for fopen() in main.c 1769: 960310 Play a bit with the __asm__ constraints. Implement CHK. 1770: 960307 Fix some stupid Makefile errors. Call parse_cmdline() if no init file 1771: is found. 1772: First round of NextStep patches; new file NeXTwin.m 1773: Stupid bug in custom.c: The #define is called NO_FAST_DISK, not 1774: NO_FAST_FLOPPY. 1775: Fix one of the problems the Deform demo had with the blitter: The 1776: line texture wasn't rotated to the right position in blitter_init(). 1777: 960306 Use some black GCC __asm__ magic to calculate flags more quickly 1778: (config INTEL_FLAG_OPT) 1779: HAM was broken by pfield_linetoscr changes. Fixed. 1780: LOW_BANDWIDTH sometimes got confused when the display window changed. 1781: 960305 BCD instructions work well enough for Locomotive Basic on CPE. 1782: New function parse_cmdline_and_init_file() that reads ~/.uaerc or 1783: ./.uaerc and adds the options found in those files to those in argv[]. 1784: Patch from Denis Sablic to make disk image filenames runtime 1785: configurable. Extend it for the ROM filename. 1786: Remove some of the run-time configurable stuff from config.h. 1787: More X keysym fixes (I finally got a proper .Xmodmap ...) 1788: for SVGAlib: Delete key, and PgUp/PgDn as Amiga keys. 1789: The sound driver can now handle 22050 Hz and/or 8 bit samples. Both of 1790: these are the default with LINUX_SOUND_SLOW_MACHINE. 1791: Also use the SNDCTL_DSP_GETBLKSIZE ioctl and write the buffer when 1792: there is enough data. 1793: Abuse keybuf.c to provide a fake joystick emulation with the numeric 1794: pad (enable with -J) 1795: 960304 bebox.cpp from Christian. 1796: Rename pfield_linetoscr_full to pfield_linetoscr_full8. 1797: More x86 assembly: pfield_linetoscr_full16. 1798: 960303 SunOS target from Dirk Vangestel. Put some not-so-superfluous includes 1799: back in. Added the DOS specific patches to filesys.c. 1800: Started x86 assembly optimizations. New file X86.S, includes an 1801: optimized version of pfield_linetoscr (for 8 bit screens only). 1802: Don't use SA_RESTART if it isn't defined. 1803: prepare_line() wasn't called often enough (again...), leading to memory 1804: corruption and "hit the reset button" bugs when using SVGAlib. 1805: 960226 Get rid of bool, true and false in amiga.h and everywhere else. 1806: Copy yesterdays code to AF_SOUND code, too. 1807: Include first round of Christian Bauer's BeBox patches. Add AIX fix 1808: from Samuel Mackrill. 1809: Add #include <string.h> and #include <stdlib.h> in some more places. 1810: Modify gencpu.c and readcpu.c so that immediate types j and J (0..15 1811: and 1..8) can be merged. Replace HAVE_ONE_GIG... option with a new 1812: CPU_EMU_SIZE option that can be set to a value in the range 0..5, 1813: where 0 is minimum size and 5 is equivalent to HAVE_ONE_GIG... 1814: 960225 Remove multiplications in sound code, use a big lookup table. 1815: 960222 Go back to having only one memory bank table of type addrbank instead 1816: of several function pointer arrays. Eliminate the HAVE_BROKEN_SOFTWARE 1817: define by extending the memory bank table to 65536 entries. 1818: Remove remaining DUALCPU code. Allow BTST src,imm instructions in 1819: table68k. 1820: 960220 Back out filesys.c change from Stefan Reinauer: don't call V36 specific 1821: functions in expansion.library. 1822: Better fix for the filesys.c assign problems by Oliver Moehrke. 1823: Make new playfield code conditional on EXPERIMENTAL_SCREEN_UPDATE. 1824: Improve the 64bit variant of said code. Try to get alignment for 1825: apixels by putting it into a union with a long. 1826: Kill the child in gui_exit(). Make graphics_init() return a value and 1827: check it in main.c to see whether graphics were initialized. If not, 1828: call gui_exit() anyway. 1829: Use new 256 color allocation routine from svga.c in xwin.c, too. 1830: Make LOW_BANDWIDTH compile _and_ work (wow...). 1831: 960219 Fix bug in memory.c: When loading 256K Kickstart, copy 256K, not 128K. 1832: 960218 Finished the rest of the UI: reset, debug and quit buttons as well as 1833: drive LEDs. New target linux-gui; all others use dummy file nogui.c 1834: for now. 1835: 960217 The GUI now supports inserting and ejecting diskfiles and displays 1836: their names. Two pipes are used now, one for reading and one for 1837: writing. Properly wait for the process to start up. 1838: Use sigaction() instead of signal() in debug.c. 1839: More fixes for the new display routines: Need to do a final call to 1840: pfield_*word_*res with the data shifted right from the last fetched 1841: word. 1842: 960216 Copper & sprite bugfixes from Marco Nelissen. Better X keyboard support 1843: from Bruno Coste. 1844: Move my_strdup() from filesys.c to options.h 1845: New files xui.c, uae-ui (shell script) and uae-ui.tk (wish script). 1846: These create a GUI for the X version of the emulator. So far, the 1847: power LED works. 1848: 960215 Don't rely on SVGAlib headers defining any keycodes at all. 1849: Include Thorsten Frueauf's NetBSD patch. Apparently, the devname 1850: variable in hardfile.c caused a namespace collision, so rename it to 1851: uaedevname. Fastmem fixes, debugger help, and ConfigDev allocation in 1852: hardfile and filesys from Stefan Reinauer. 1853: xwin.c: Give the window a title. 1854: Half-implemented the DIVU.L, DIVS.L, MULU.L and MULS.L instructions. 1855: 960213 AF sound system support from Marcus Sundberg. Win95 keyboard support 1856: for SVGAlib from Thierry Lescot. Added most of Gustavo Goedert's MS-DOS 1857: port. 1858: When using the old pfield_*word_*res functions, the gen_pfield_tables() 1859: function was undefined. 1860: Hmmm, /dev/brain must have been unmounted when I wrote the new pfield_* 1861: functions. Now they make sure they access the apixels array only with 1862: addresses that are multiples of 16 bytes. 1863: 960212 Fix copy&paste bug in cia.c found by Marco Nelissen: Need to call 1864: RethinkICRB() instead of RethinkICRA() for CIA B timers. 1865: Added Stefan Reinauers autoconfig area/fastmem support. Added run-time 1866: options for use_slow_mem and use_fast_mem. 1867: Oops, the pfield_*word_*res() optimizations break completely on HPs and 1868: Sparcs. Now use this optimization only if we are using GCC 2.7. 1869: HPUX fixes in the Makefile. New target hpux-gcc. 1870: The code for clearing the areas above and below the DIW was trying to 1871: be a little too clever selecting the color. Just use the current 1872: background color. 1873: Added channel attachment to the sound code. Untested. 1874: 960211 Found some flag setting bugs in the rotate/shift instructions and in 1875: addx/subx (which I thought I had already fixed) by reimplementing 1876: DUALCPU mode to work with Ed's e68k. 1877: Argh! The copper is completely disabled during DDF in Hires 4 bitplane 1878: mode - so copper positions 0x30 and 0x34 are about 640 pixels apart. 1879: The Kickstart uses FF39FFFE to wait for the end of line 256 instead of 1880: FFDFFFFE as in all the other modes. This is implemented now, but 1881: probably not 100% correct. 1882: Added check whether SVGAlib doesn't support our mode. 1883: If using GCC >= 2.7, define __attribute__((aligned(16))) as 1884: BIG_ALIGNMENT in config.h. Use it for the apixels array to prevent 1885: potential problems on the Alpha. 1886: Tried to make the color selection routine produce better results in 256 1887: color mode. 1888: 960210 readdisk didn't work at all for large files - fixed. 1889: In pfield_doline(), clear the apixels array if playfield DMA is off. 1890: In SPRxCTL(), reset sprite mode when sprpos==sprctl==0 1891: Rewrite pfield_orword_hires() and friends to use lookup tables. No 1892: noticeable improvement on my machine, but might be different on other 1893: machines. Two versions of this code: one using 64bit longs on the Alpha. 1894: New targets linux-debug and linux-prof. 1895: Use 1MB slow mem. 1896: -funroll-loops is a little excessive. Remove it. 1897: Add support for linear addressing modes for SVGAlib version. Speedup 1898: is 20-30% on my old 486 VLB. Now, would someone please fix SVGAlib to 1899: support that for the S3-864 as well? 1900: Set SVGAlib keyboard mode to DONT_CATCH_CTRLC. 1901: Change custom.c, xwin.c and svga.c to keep track of the display window 1902: and background color and to clear the areas above and below the display 1903: window if necessary to remove garbage images from previous frames. 1904: Set the LOF bit in the Kickstart replacement init routine. Also 1905: initialize CIAs to Kick 1.3-like values. 1906: 960209 Some more -DHAVE_xxx_H options in the Makefile and filesys.c. DEC Alpha 1907: fixes for filesys.c from Ed and Marcus. HPUX fixes. 1908: New field in table68k for CPU type, parsed by readcpu and compared to 1909: CPU_LEVEL which is defined in config.h. New instructions MOVE2C, MOVEC2 1910: and CAS. Also a new field for priviledge level that is used by gencpu to 1911: generate tests for regs.s and Exception(8) calls if necessary. 1912: Move parsing of table68k into a new file build68k.c. Generate a file 1913: called cpudefs.c containing the information from table68k in 1914: machine-readable form. Change readcpu.c to use the array found in 1915: cpudefs.c. 1916: Don't generate the d8(An,Dn) addressing mode in-line. Generate a call to 1917: get_disp_ea(), which is an inline function in newcpu.h if UAE is 1918: configured for 68000 emulation; for 68020 emulation, it is defined in 1919: newcpu.c. 1920: Debugger could crash on illegals because there was no mnemonic defined 1921: in lookuptab[]. 1922: Return zero for SERDATR. 1923: 960207 It seems that the copper stops completely if an attempt to write to a 1924: specific register fails because of COPDANG. Apparently, the Kickstarts 1925: use this to stop the copper with a 0x00000000 longword. This leaves the 1926: question why they don't do 0xFFFFFFFE like everybody else. 1927: Enable 512K slow memory by default. 1928: Integrate Stefan Reinauers battclock patch that makes it work with 1929: newer Kickstarts. 1930: In cia.c, do the little index pulse hack even if dskdmaen==1, because 1931: DMA is only enabled when dskdmaen==2 (Helps the "Mad Decision" demo). 1932: Add solaris entry to Makefile (GCC, Openwin, -DHAVE_STATFS_H) 1933: Define HAVE_STRINGS_H for SGIs, change filesys.c accordingly. 1934: 960206 Build a smaller table instead of cputbl.c. Use readcpu and the smaller 1935: table to build the cpufunctbl array at run-time. 1936: Nuked the other half of the old CPU emulator. Reimplemented debugger to 1937: use the information generated by readcpu. 1938: Define HAVE_STATFS_H for SGIs in the Makefile, and use that as a test 1939: in filesys.c instead of #ifndef __linux. 1940: Hack filesys.c, get_path() to support the assign command. 1941: Add -fstrength-reduce and -funroll-loops in Makefile to make people 1942: nervous. 1943: Rewrite Sam Devulder's LOW_BANDWIDTH patch and add it to xwin.c. 1944: 960205 Forgot some PC-relative bit instructions in table68k. 1945: Nuked half of the old CPU emulator. 1946: Split gencpu.c into two parts, new one is called readcpu.c and just 1947: parses the table68k file. Link it to the main program. 1948: 1949: 960205 Release 0.4 1950: 960203 filesys.c, action_read(): Slightly more efficient code (translate Amiga 1951: address to real pointer). 1952: Moved some common code in the generate_* functions in gencpu.c to a 1953: separate function. 1954: 960202 Added an experimental fast disk option. Currently turned off by 1955: default (it's not such a big win). 1956: Attached sprite fixes (overlapping att. sprites looked bad, Katakis). 1957: Add sleep(1) before resetting the console to text mode when using 1958: SVGAlib: this might fix some screen corruption problems. 1959: Add sprite/playfield priority checking to the most important case 1960: (single playfield, no HAM). 1961: In filesys.c, do_find(): open() returns -1 on error, not zero. 1962: Return ERROR_OBJECT_WRONG_TYPE if do_find() is called for a directory 1963: (fixes Champions of Krynn harddisk installation). 1964: 960201 Don't abort if sound driver not present, just set produce_sound to 0. 1965: New files keybuf.c and keybuf.h to record keypresses in the right 1966: order and without losing any. In cia.c, force 15 scanlines between 1967: keypresses, just to be sure. 1968: unixfs.device _does_ work with Kick 1.3: just don't trust what Kick 1.3 1969: sends in the startup packet. For now, disable more than one mount per 1970: command line. 1971: Started integrating Ernesto's new Mac sources. 1972: Remove superfluous includes from some files. 1973: 960131 Added Ed's unixfs.device (great stuff). 1974: Adding ULONGs to pointers is a bad idea on the Alpha if the ULONG value 1975: really is signed. Add some casts to LONG in (pc_p + src) expressions 1976: in genpu.c. 1977: If DMACON is written and copper DMA is enabled, do a COPJMP1 at once. 1978: Helps the "Interference" demo. 1979: 960129 More SGI fixes from Ed. Bugfixes and transdisk improvements from Marcus 1980: Sundberg. 1981: Remove EXTRA_DEFINES from Makefile. Breaks some systems. 1982: Move common sprite code from pfield_doline() and pfield_doline_slow() 1983: to new function pfield_sprite(). The same sprite may appear more than 1984: once on the same line, so don't shift out the bits of sprdata[] and 1985: sprdatb[] while displaying it (Turrican I). 1986: In xwin.c and svga.c, barf if LINUX_SVGALIB doesn't match the file 1987: being compiled. 1988: Make all .o files depend on config.h in the Makefile. 1989: No need to exit if sound driver unavailable, but -S given. 1990: Small debugger fix: Missing space in output. 1991: Fix for the sprite logic: Specifically, use a state variable indicating 1992: whether the sprite has been restarted after a VSYNC. Fixes most 1993: Turrican problems. 1994: 960124 Added Denis Sablic's patch for sound run-time option. 1995: Added Ed Hanway's patch for better Makefile, X mouse cursor blanking 1996: and more SGI compilation fixes. 1997: 960123 Include options.h everywhere. 1998: Handle 8 bit GrayScale visuals like PseudoColor. 1999: Remove C++ leftovers from joystick code. 2000: 960122 When using the joystick driver, the button test must come after 2001: handle_events() in vsync_handler(). 2002: 960118 Removed all the remaining C++ comments. Changed all inline keywords to 2003: __inline__. Define __inline__ if not using gcc. 2004: Make proper prototypes for everything. Compile with maximum warnings + 2005: -ansi + -pedantic. 2006: Remove CIA_cycle(), obsolete. 2007: Reimplemented the STOP optimization in newcpu.c. Removed DualCPU 2008: support in CPU emulator. 2009: Real nasty bug in pfield_doline() fixed: sprxpos could be evaluated as 2010: negative, with not-so-amusing results. (Need to rewrite this in 2011: Oberon to get array bounds checking :-) 2012: 960117 Heroic effort: Rewrote the thing in C. This might help fix some 2013: problems with users being unable to compile it. 2014: Fixed a problem in hsync_handler(): Only call flush_line() for lines 2015: in the display window, i.e. when we did a prepare_line() before. 2016: Better code for relative branches: Don't use setpc(getpc()+x) calls, 2017: increment regs.pc_p instead. 2018: 960116 Reimplemented the function to load the Kickstart ROM. Use stdio instead 2019: of fstreams since this apparently does not work on the Mac. Detect 256K 2020: Kickstarts. Detect corrupt ROM images (calculate checksum). 2021: Added Ernesto Corvi's Mac port. Changed it around a bit, so it 2022: probably won't compile. 2023: 960115 Reinstate config.h options for X screen depth, so that DrawPixel() can 2024: be inlined in custom.cc for speed. xlinebuffer is now incremented in 2025: each call to DrawPixel() (for both X and SVGAlib) to get rid of some 2026: address calculations. 2027: 960114 Fixed X generic pixel drawing routine for SHM. 2028: Still trying to fix the harddisk emulation. 2029: uae.device no longer breaks the debugger (can step through uae.device 2030: functions now) 2031: Bugs affecting performance: SPCFLAG_STOP never got reset, and DSKLEN() 2032: would set SPCFLAG_DISK even if DMA was being turned off. 2033: Made slow memory a run-time option. 2034: Defer interrupts by one CPU instruction to give programs a chance to 2035: read INTREQR ("Seeing is Believing" and "Substance" demos) 2036: Added ScrollLock hack for X, too. 2037: 960113 SVGAlib version compiles again. Fixed SVGAlib mouse bug. 2038: Fixed SHM bug: Maximum scanline is 313, not 312. 2039: Sometimes, disk.cc missed a side change and would read the wrong data. 2040: Fixed. Apparently, this was the worst compatibility problem. 2041: Implemented trace mode. 2042: 960112 Changed layout of class amigamemory a little so that gcc can generate 2043: better addressing modes. 2044: Finally wrote functions in gencpu to generate MOVEMs. 2045: 960109 Integrated Ed Hanway's patches for better X support and run-time 2046: configuration of some options. 2047: Got rid of the direct VGA memory access. (Need to do this differently). 2048: Changed the method of drawing lines: custom.cc now tells the graphics 2049: code the line number and whether it needs to be doubleed before drawing 2050: it. 2051: Added Andre Beck's MIT-SHM patch. 2052: Remove warnings for newcpu.cc. 2053: 960108 Fixed exceptions in op_illg(): Need to decrement PC. 2054: 960107 Added an "uae.device" resident module at 0xF00000. This emulates a hard 2055: disk (fixed size 8MB for now). 2056: 960106 Moved some common code from pfield_doline() and pfield_doline_slow() to 2057: a separate function. This fixes a potential HAM bug (two static vars 2058: for the same value). 2059: Sound support for Linux. Works only with graphics off and the CPU 2060: slowed down. 2061: Better SVGAlib keyboard support. 2062: 960105 Added AvailMem(), AllocMem(), AllocAbs() and FreeMem() dummies. 2063: The Hardwired demo times the multiplication instructions and prints 2064: "This demo don't like Axel" if they are too fast. Apparently, Axel has 2065: a 68040. Added a WANT_SLOW_MULTIPLY option to config.h. 2066: Fixed the fast blitter emulation (seems to work now). 2067: 960104 Fixed all the ChangeLog entries from 95 that said 96 (oops??!) 2068: pfield_may_need_update() should check whether bitplane DMA is on. 2069: Added ersatz.cc and ersatz.h. The purpose of these files is to 2070: implement one or two Kickstart functions that are commonly called from 2071: bootblocks. This should help support some games and demos that only use 2072: the Kickstart as an initial track loader. So far, it's only good enough 2073: for one program. 2074: 951223 More intelligent event handling in the CPU emulator. Slightly faster. 2075: 951222 Optimize CPU emulation by inlining cctrue(). Also, the real PC no 2076: longer needs to be incremented each instruction. The real PC value 2077: now has to be fetched by m68k_getpc(). 2078: Added direct screen access for SVGAlib, but it didn't help much. I'll 2079: probably remove it again. 2080: The gencpu executable is 2M smaller if it allocates memory 2081: dynamically. 2082: 951216 custom_bput() enhanced a little. Now remembers the value that was 2083: written in the other half of the register. 2084: Apparently, the USEx bits of BLTCON0 are ignored in line draw mode. 2085: (Silents-Demo) 2086: 2087: 951212 Release 0.3 2088: 951212 Included Alessandro's battclock patch. 2089: Fixed mouse autocalibration for wrong-aspect screens. 2090: 951210 Joystick support for Linux. 2091: Better color matching for 256 color modes. 2092: Rewrote most of the memory management in plain C, mainly to avoid the 2093: various template bugs in some compilers, but also because gcc can't 2094: handle the regparm attribute for virtual functions yet. 2095: 951209 Added Alessandro's patches for automatic mouse calibration. 2096: Tried to optimize the blitter loop a bit. Doesn't work. 2097: 951205 Added Hannu's patches for diskfile write support. 2098: 951201 More portability fixes, especially for weird compilers that workstation 2099: vendors ship. 2100: 951125 More keys for X, some keyboard support for SVGAlib (slightly 2101: dangerous); beginnings of US keyboard support. 2102: 951124 256 color mode support for SVGAlib. FAST! 2103: 951123 Tiny tweak to central loops in playfield code results in much better 2104: assembly. Apparently, GCC does not feel too good about optimizing 2105: reference variables. 2106: 951113 Bug fix in the events code (nextevent must not be changed while events 2107: are being processed). 2108: 951112 The playfield can now be drawn with the right aspect (i.e., every line 2109: twice). Slower, but looks nicer and supports interlaced mode 2110: (flickerfree!). The first 29 lines (blanking interval) are omitted 2111: now to fit everything in 800x600. 2112: Oops, disk drive identification did not work. Fixed, all four drives 2113: appear now. 2114: Oops, CIA_reset() was there all the time but never got called. 2115: Optimized the planar->chunky conversion in pfield_doline() a little. 2116: This really is time critical. 2117: 951111 If colors change on a line, we might still be able to use the full 2118: line update, if the change occurred outside the DIW. Have to be a 2119: little careful, though. 2120: Playfield code now uses real pointers again, but makes sure it can 2121: read enough bytes to complete the current line without segfaulting. 2122: Oops, my "better estimate" (see below) was a mistake. Restored the 2123: original 4 hw cycles per instruction relation. This slows down the 2124: emulator while booting (more screen refreshs), but speeds up many 2125: demos that otherwise spend most of their time waiting for a specific 2126: raster position. 2127: Playfield modulos must be added immediately when DDF stops. (WOC demo) 2128: Portability fixes in disk.cc (null pointers). 2129: Implemented average frame rate calculation ('c' in the debugger). 2130: Debugger fixes. 2131: Sprite fixes (height was wrong, wrong priorities in fast update) 2132: Added an AutoConfig expansion RAM device. The OS seems to recognize 2133: it, but breaks sometime later (seems to be the same bug that kills 2134: the emulator when there's RAM at 0xC00000). 2135: SVGAlib now has mouse support, too. 2136: HAM mode implemented. 2137: 951110 Implemented fast complete line playfield update. Switches back if 2138: necessary. Fixed attached sprites bug. 2139: Copper wait no longer slows down the CPU emulation. (there were 2140: lots of new copper bugs because of this, but I hope I fixed all 2141: of them). 2142: Changed cia?t? from UWORD to ULONG, since there were overflows in 2143: the assertion (because of 65535+1 == 0). 2144: 951108 The DIVS remainder could have the wrong sign. 2145: New HAVE_BROKEN_SOFTWARE define in config.h that makes the emulator 2146: mask out the high byte of all addresses. 2147: Wrote genblitter.cc and replaced the awful code in blitter_blit() 2148: and blitter_line() with calls to the generated function. 2149: 951107 Implemented ?BCD. These don't really work yet. 2150: 2151: 951106 Release 0.2 2152: 951105 Fixed a bug in the new CIA timer code. 2153: Fixed flag settings for some cases of ADDX/SUBX 2154: 951104 Changed the main loop a bit - call all the *cycle() functions only if 2155: it is really necessary. I hope everything still works - at least it is 2156: twice as fast now :-D 2157: Improved mouse emulation - avoid overflows, permit "wraparound". 2158: Added debugger 'g' command. 2159: Removed some useless casts from the CPU emulator. 2160: Added a configuration option to prevent checks for odd addresses. Only 2161: very bad programs need this check. 2162: 951103 The sprite emulation sometimes did not work too well, especially with 2163: dual playfield mode. Hopefully fixed. 2164: 951102 The 10 cycle counter is more like a 5 cycle counter in this 2165: implementation. (Fine, that makes floppy loading even faster :-) 2166: Implemented mouse movement. Totally out of sync with the X pointer, 2167: but at least something. 2168: Found out about the XAutoRepeatOff() function. Real useful. 2169: DualCPU mode works again. 2170: Corrected flag settings for CMPA.W. 2171: Fixed DIV instructions: Don't change dest in case of overflow. 2172: 951031 The PC is now converted to a real pointer. This is safe (much safer 2173: than converting bitplane pointers, since the PC always has to be valid) 2174: and should speed things up a little. 2175: 951029 Some primitive event handling: 2 mouse buttons and 1 joystick button 2176: are emulated, as well as parts of the keyboard (only german mapping). 2177: Fixes: Delay values and sprite widths are always LoRes. Playfield 2 2178: uses colors 9..15. 2179: Implemented a frame rate configuration option that can speed up the 2180: emulation quite a bit. 2181: 951028 A-ha. The 68000 has one strange feature after all. MOVEM.W to Dn sign 2182: extends the value to a longword. 2183: Katakis expects that accessing odd addresses causes exception 3, so 2184: we have to emulate this. *Sigh* 2185: Do at least something with the index pulse bit. It's more or less a 2186: toggle now. 2187: Implemented sprites & dual playfields, both with small glitches. 2188: 951027 When LOF is set, the frame is one line longer. 2189: 951026 Implemented a preliminary diskfile type for special formats. Turrican 2190: boots up correctly (crashes if I want to play a game, though). 2191: Implemented EHB and added a few missing masking operations in the color 2192: handling. 2193: Fixed the absw addressing mode - need to sign extend here as well. 2194: 951025 Fixed another extremely stupid bug, this time in the playfield code. 2195: Each pixel was calculated four times... 2196: Increased the size of the apixels array - this was being overrun. 2197: 951024 Fixed an extremely stupid bug in flag settings for byte and word 2198: operations. 2199: The "sectors until gap" field must be in the range 1..11 instead of 2200: 0..10, or very bad things will happen if it is used to calculate a 2201: blit size. 2202: 951021 Don't use real pointers to fetch bitplane data - too many segfaults. 2203: Delay the copper after executing an instruction, or wait sequences like 2204: FFDF FFFE 0200 FFFE won't work. 2205: Some programs expect that the CPU is completely turned off if the 2206: blitter nasty bit is set. So be it. 2207: The RSI Megademo chokes when disk DMA fetches two syncs. Leave one out. 2208: The LOF bit wasn't set correctly. 2209: 951020 Blitter fixes: Load fill carry from FCI each line; keep the masked 2210: value of bltadat in bltolda 2211: Give the CPU twice as many "cycles" - this still is not even remotely 2212: correct, but a better estimate. 2213: 951017 Added SVGAlib support. Unfortunately, my graphics card isn't fully 2214: supported by the current SVGAlib version. 2215: 951015 More small optimizations to memory.h and custom.cc. 2216: Added code for 256 color screens. The type of screen has to be 2217: selected at compile time. 2218: 951014 Modified the memory management to be (hopefully) a little more 2219: efficient. 2220: Changed gencpu.cc to generate functions and a big function pointer 2221: array instead of the old case labels. The generated C++ files now 2222: compile a lot faster, and the code should be better as well. 2223: 951013 Implemented pseudo POTGOR to keep the RMB from being detected as 2224: pressed all the time. 2225: GenerateDecTab() generated TRAP incorrectly. 2226: Made some more instructions (MOVE.B A0,D0 etc) illegal in gencpu.cc. 2227: Fixed X colors. Fixed data fetching for HiRes screens. 2228: 951011 EXT.WL instructions set the flags. 2229: Implemented second CPU emulator, plus "DualCPU" option. The new CPU 2230: emulator is generated automatically from a file called table68k by 2231: a program called gencpu. 2232: 951002 Removed version information from the file headers. 2233: Removed some sillinesses in custom.cc. 2234: DSKLEN bug fixed (mask out 0xC000). 2235: Playfield code rewritten. Now supports horizontal scrolling. Implemented 2236: some primitive color support. 2237: 950927 Fixed predecrement/postincrement addressing with byte size for the 2238: stack pointer. 2239: 950926 Fixes to the disk code: Terminate disk DMA, flip disk side, read more 2240: than just two sectors during disk DMA, fix checksum calculation. 2241: Fixed copper comparison bugs. 2242: 2243: 950901 Release 0.1 2244: 950826 bugfix in GenerateDecTab(): LEA.W instructions were generated. 2245: 950312 New file disk.cc, for (guess what) disk support. Doesn't work yet. 2246: 950311 Multiplication instructions were generating 16 bit results. 2247: Sign extension was wrong in some places. 2248: Removed bugs in the blitter (one word width: ALWM & AFWM and line mode) 2249: The Kickstart logo looks perfect now. 2250: Playfield routine should check BPLCON0 instead of BLTCON0 2251: 950309 ASR D8,D0 is not a valid instruction. 2252: Bitplane pointers were not incremented. 2253: Copper comparison didn't work. COP2LC and bitplane pointers were written 2254: incorrectly. Fixed. 2255: Something that bears a striking resemblance to the familiar Kickstart 2256: logo now appears in my X window! 2257: 950306 ROR used to set the X flag. Bug fixes in GenerateDecTab. 2258: More bug fixes in several 68000 instructions. 2259: Improved blitter. ?????? KS1.3 doesn't seem to set BLTDMOD on a 2260: line draw. What is this?? 2261: Fixed stupid bugs in playfield code. 2262: 950305 fixed bugs in opcode generation and interrupt handling. Implemented 2263: STOP. Fixed flag settings: Result wasn't masked. Ooops!! 2264: ADDQ #imm,An doesn't set flags. 2265: One-shot timer mode didn't work. 2266: Implemented blitter, doesn't work. 2267: Some primitive and slow playfield support, only to make the things that 2268: hopefully will soon appear visible. 2269: 950304 CIA timers ought to work OK. Copper is emulated. 2270: Emulator loops infinitely when trying to initialize the console.device 2271: 950303 experiments with X, figured out how to open a window and draw pixels. 2272: No graphics emulation yet, though. 2273: 950301 Nearly all instructions work. 2274: Fixed supervisor mode stack. 2275: The memory manager now handles CIA chips. 2276: 950228 Bug fixes in 68000 code, exceptions and more instructions implemented 2277: Memory management extended to handle custom chips as well. 2278: Debugger improved. 2279: Kickstart passes memory & CPU type test, builds up some system 2280: structures. 2281: 950227 MC68000 & little debugger for tracing. It can execute the first couple 2282: of KS1.3 instructions, but doesn't get past the memory test. 2283: 950225 Work begun
This archive runs on limited infrastructure. Preserving old code on modern bandwidth. Automated agents are requested to crawl responsibly.