This repository has been archived by the owner on Aug 20, 2023. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 26
/
ChangeLog
1559 lines (780 loc) · 54.7 KB
/
ChangeLog
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
2012-02-04 Thomas K <[email protected]>
* Bugfix: creating src/python/Makefile by configure missed
2012-02-04 Thomas K <[email protected]>
* Bugfix makefiles: pysimulationmember.h wasn't packed in dist
2012-02-04 Thomas K <[email protected]>
* Bugfix delivery-check script: dist check wasn't done, if build doxygen doc wasn't selected
2012-01-29 Petr Hluzin <[email protected]>
* Fixed indentation which I broke by my previous commit.
2012-01-22 Petr Hluzin <[email protected]>
* Do not reject programs compiled with `#include <avr/signature.h>'
2012-01-22 Petr Hluzin <[email protected]>
* Fixed indentation.
2012-01-15 Petr Hluzin <[email protected]>
* Fix: If program clears the interrupt flag, then interrupt should not fire.
2012-01-11 Petr Hluzin <[email protected]>
* Add BREAK-instruction, causes simulavr to halt.
2012-01-04 Petr Hluzin <[email protected]>
* Fixed ICALL instruction timing.
2012-01-04 Thomas K <[email protected]>
* Add tool tab-check.py: a script to count tabs in source files. It should help to decrease changes only because of tabs (because developers can use different tab spacing and start to indent code against his tab space setting to make code more readable, but the next developer maybe will change this again because of another tab spacing)
2011-12-29 Thomas K <[email protected]>
* Bump version to 1.0rc1.
2011-12-29 Thomas K <[email protected]>
* Add header files for at90canbase and atmega8 to package headers to fix buggy package, which is created by "make dist"
2011-12-29 Thomas K <[email protected]>
* Bugfix for delivery-check.conf.sample: one line wasn't commented out, so it was interpreted as shell command
2011-12-29 Thomas K <[email protected]>
* Introduce a bash shell script to run a check, if one or more new commits are ready to push to savannah. It clones the git repo to a new directory, runs there the complete build sequence (bootstrap, configure, make), build, if configured, doxygen and sphinx docu, verilog and tcl module (python module is build in every case) and creates also dist package (+ checks this package by running build sequence) See sample config file delivery-check.conf.sample and run "./delivery-check -h" for command line options.
2011-12-21 Petr Hluzin <[email protected]>
* Fixed build failures under gcc-4.4.5
2011-12-14 Petr Hluzin <[email protected]>
* Fixed build failures under gcc-4.4.5
2011-12-04 Petr Hluzin <[email protected]>
* Fixed whitespace broken by my previous commit.
2011-12-04 Petr Hluzin <[email protected]>
* Reworked std::multimap in SystemClock::Step() to binary heap -> 2.5x speedup. The multimap contains one AvrDevice* (if single-core simulation) and possibly serialrx/serialtx (if UI is used). Allocations and deallocations in the container took ~50% of CPU time. After rework the simulation speed improved from 3.28 MIPS to 8.2 MIPS on my PC on busy-loop test using MSVC. Similar speedup when GCC is used. (See https://savannah.nongnu.org/support/?106678 for performance details.) Still, the MinHeap::RemoveMinimumAndInsert() costs 5.5% exclusive time (6.7 ns/cycle) and SystemClock::Step() 11.5% exclusive time (possibly includes inlined stuff). Maybe we can improve that further.
2011-11-27 Petr Hluzin <[email protected]>
* (IrqFunktor creation made slightly more readable.)
2011-11-27 Petr Hluzin <[email protected]>
* Added comment about how to compile without BFD library on Linux. The ./configure cannot find my /usr/include/bfd.h and /usr/i686-pc-linux-gnu/lib/libbfd.a and I do not know how to persuade it. I thought autoconf was supposed to help people.
2011-11-20 Petr Hluzin <[email protected]>
* Simplifying interrupt handling. The old `newIrqPc' mechanism: When irqSystem->GetNewPc() returned the new PC value, then the next Step() invocation arrived on if(noDirectIrqJump == 0) and executed `else' branch. Therefore one more ordinary instruction was executed and interrupt hander was invoked in next Step() invocation. This looks similar to SEI handling, however datasheet talks about delaying interrupt invocations for SEI instruction only, on the other hand `newIrqPc+noDirectIrqJump' mechanism adds 1 cycle delay per _interrupt invocation_ (and I do not see any basis for that in datasheets).
2011-11-19 Petr Hluzin <[email protected]>
* Implemented `--file' option for MSVC compiler. Now it is possible to load program memory from ELF file when compiling using Visual Studio. Previously the only way was to use GDB and "load" command. Can be also used as an alternative for BFD library. However you cannot load symbols from ELF file, yet.
2011-11-19 Petr Hluzin <[email protected]>
* Fixed typo. I am an idiot. No wonder I could not Google "the BDF library".
2011-11-19 Petr Hluzin <[email protected]>
* Assignment to HWStack::m_ThreadList does not make sense, prevent SWIG from doing it. It keeps state of detection of a switch, which is difficult to do right. It might make sense to replace the list of threads in ThreadList::m_threads, though. Altered to produce better diagnostics if a compiler tries to use an assignment operator.
2011-11-19 Petr Hluzin <[email protected]>
* Better diagnostic for avr_op_ILLEGAL instruction. If it is executed then show the opcode bytes (in avr-objdump style). Also show byte-based address and word-based index.
2011-11-19 Petr Hluzin <[email protected]>
* Removed almost unused class MemoryOffsets. The class was used only on two places, well, one place; there is no prospect for having it used more in future.
2011-11-19 Petr Hluzin <[email protected]>
* Added #include <cstdio>, needed for `NULL'. Geez, compiler or something should be doing the boring stuff like this. Reported by Yann Dirson [1] and and Sebas Vila-Marta [2]. [1] https://savannah.nongnu.org/bugs/?34270 [2] https://savannah.nongnu.org/patch/?7640
2011-11-19 Petr Hluzin <[email protected]>
* Removed unused member `Sram'. The last use (in HWStackSram) was removed in previous commit. The previous commit was the last remaining part of patch #7079. https://savannah.nongnu.org/patch/index.php?7079
2011-11-19 Petr Hluzin <[email protected]>
* Use `core' field to access memory. This makes it easier to search, makes HWStackSram easier to understand and makes the way RAM is accessed unified with DecodedInstruction code. The transform is ok because mem->myOffset was always 0, no need to add any offset.
2011-11-15 Petr Hluzin <[email protected]>
* Fix: If SEI instruction enabled interrupts, we allowed them immediately. When SEI enables interrupts, datasheet guarantees the next instruction is still executed with interrupts disabled. This is useful for SEI - SLEEP sequence. ("When using the SEI instruction to enable interrupts, the instruction following SEI will be executed before any pending interrupts") Also GCC relies on that when growing stack in function prologues. (SEI instruction is shortcut for "BSET I" instruction.)
2011-11-14 Petr Hluzin <[email protected]>
* Whitespace broken by previous commit, equivalent change to locals
2011-11-13 Petr Hluzin <[email protected]>
* Improved code documentation.
2011-11-13 Petr Hluzin <[email protected]>
* Improved code documentation.
2011-11-13 Petr Hluzin <[email protected]>
* Fixed: Simulation was terminating on a EEPROM operation. If EEPROM or Flash operation was in progress (i.e. virtual Hardware::CpuCycle() returned nonzero) then: 1) "CPU-waitstate" was printed zero or more times (correct) 2) `cpuCycles' eventually became 0 (correct) 3) Hardware::CpuCycle() returned nonzero (correct) 4) "CPU-Hold by IO-Hardware " was printed (correct) 5) instruction was not executed (correct) 6) `cpuCycles' was decremented causing it to be <0 (bug) 7) AvrDevice::Step() returned -1 and simulation was terminated.
2011-11-13 Petr Hluzin <[email protected]>
* Cleanup, equivalent changes, preparation for next commit. Since `hwCycleList' is std::vector<Hardware*> and the operator[] is inline-able then this code is as fast as previous and faster to understand. A part of condition was always false.
2011-11-13 Petr Hluzin <[email protected]>
* Improved code documentation.
2011-11-13 Petr Hluzin <[email protected]>
* HWAcomp::CpuCycle() is unused, removing. Actually it *is* called but its body is the same as in predecessor Hardware:CpuCycle(). The function is not going to be extended anyway.
2011-11-13 Petr Hluzin <[email protected]>
* Whitespace made consistent with the rest of the file.
2011-10-09 Petr Hluzin <[email protected]>
* Fixed a test when adding a breakpoint. Flash->GetSize() returns bytes, no need to divide.
2011-09-25 Petr Hluzin <[email protected]>
* Added diagnostic for compiling Debug builds with Python
2011-09-25 Petr Hluzin <[email protected]>
* Fixed build error "reference cannot be declared `mutable'" on gcc-4.6 Reported by Sebastian and again by Yann Dirson. http://lists.gnu.org/archive/html/simulavr-devel/2011-05/msg00007.html https://savannah.nongnu.org/bugs/?34270#comment4
2011-09-25 Petr Hluzin <[email protected]>
* Allow building with VS2010 - stdint.h clash. VS2010 finally does ship with stdint.h. Unfortunately VS compiler finds our compatibility header for older VS. This allows switching compilers without the pain of renaming the file.
2011-09-25 Petr Hluzin <[email protected]>
* simulavr can now compile as a Python library under VS2008
2011-09-25 Petr Hluzin <[email protected]>
* Release configuration did not build under VS2008. Reported by Сергей Смирнов, I did not hit into the wall until now. http://lists.nongnu.org/archive/html/simulavr-devel/2011-05/msg00005.html
2011-09-25 Petr Hluzin <[email protected]>
* Keep GetCurrentTime macro undefined on Windows. (part 2) Addition to commit 8b70ab38b37ded5cfbab04a90587b9d3cf38ef55: #include <winsock2.h> includes winbase.h which defines some backward-compatibility macro #define GetCurrentTime() GetTickCount(). This clashes with SystemClock::GetCurrentTime() and thus prevents including mysocket.h and systemclock.h in the same file on MinGW/Visual Studio. Preparation for SWIG/Python on Windows.
2011-09-25 Petr Hluzin <[email protected]>
* Calls to malloc() redirect to _malloc_dbg(), MSVC only On MS Visual Studio call debug-enhanced version. This change does not enable reporting memory leaks (there are a lot). Does not affect operator new. http://msdn.microsoft.com/en-us/library/10t349zs.aspx
2011-09-15 Petr Hluzin <[email protected]>
* Keep GetCurrentTime macro undefined on Windows. #include <winsock2.h> includes winbase.h which defines some backward-compatibility macro #define GetCurrentTime() GetTickCount(). This clashes with SystemClock::GetCurrentTime() and thus prevents including mysocket.h and systemclock.h in the same file on MinGW/Visual Studio. Preparation for SWIG/Python on Windows.
2011-09-15 Petr Hluzin <[email protected]>
* Use for-loop when the iteration count is limited. Also more readable and shorter. No functional changes.
2011-09-15 Petr Hluzin <[email protected]>
* SWIG failed on __declspec(). (Preparation for compiling with SWIG/Python in visual Studio.) If SWIG is executed without _MSC_VER macro then it tries to use class GdbServerSocketUnix (instead of GdbServerSocketMingW), so I added -D_MSC_VER on SWIG command-line (not committed yet). However SWIG parser cannot cope with __declspec() used by MS compiler. Maybe we should use #ifdef WIN32 instead of MINGW and _MSC_VER.
2011-09-14 Petr Hluzin <[email protected]>
* Fixed bug #34287: commit f658676ad64c4e10f9dff8da808869ee26b0a2e3 breaks python simulation The commit changed behavior when cpuCycles > 0, the old code relied on `int bpFlag = 0;'.
2011-08-15 Petr Hluzin <[email protected]>
* Fixed SCK glitch in SPI in certain situations. If SCK pin was in non-alternate mode (=SPI off) with HWPort::alternatePort==0, pin enabled (DDRxy=1) and was outputting "active" (=non-idle) clock level, then executing lines SCK.SetUseAlternatePortIfDdrSet(1); // `alternatePort' is undefined here SCK.SetAlternatePort(spcr & CPOL); caused a sampling clock edge and confused a remote SPI device. (This happened with dearduinoed adafruitPCD8544-Nokia-5110-LCD library using my LcdNokia3310 class with assigned pins PB5 - SCLK, PB3 - SDIN, PB0 - DC, PB2 - SCE#, PB4 - RESET#. I must not forget to set PB5 to idle before doing SPCR = 0x58 next time.)
2011-08-14 Petr Hluzin <[email protected]>
* More missing indentation. Added documentation.
2011-08-14 Petr Hluzin <[email protected]>
* Thread switching code incorrectly considered any SP write as a switch. Well, it seems I accidentally omitted 'virtual' when porting from my private tree. So it always returned -1 and the AvrFlash::LooksLikeContextSwitch() never filtered out the setup at reset and at prologues/epilogues.
2011-08-11 Petr Hluzin <[email protected]>
* Added some documentation.
2011-08-11 Petr Hluzin <[email protected]>
* Initialization of some member vars was missing. Also shortened scope of some locals. No functional changes.
2011-08-10 Petr Hluzin <[email protected]>
* Removed incorrect comment. I was wrong - the case is reachable. The 'if' uses '%' operator and 'switch' uses '/'.
2011-08-10 Petr Hluzin <[email protected]>
* Introduced indentation; no functional changes. In `switch ((clkcnt/clkdiv)&1)' the block `case 1:' seems to be unreachable because of the condition `if (!(clkcnt%clkdiv))'. If it is really unreachable then SPI master does not work.
2011-07-24 Petr Hluzin <[email protected]>
* Fixed compile errors (gcc 4.6) Reported by Sebastian: http://lists.gnu.org/archive/html/simulavr-devel/2011-05/msg00006.html
2011-07-23 Petr Hluzin <[email protected]>
* Added reporting of AVR threads to GDB. Note: the detection added in previous patch fails on libraries which use a temporary stack while switching threads (AvrX, Dr. Tak's Real Time Kernel - DTRTK). Also GDB fails on libraries with nontrivial stack layout (FreeRTOS).
2011-07-06 Petr Hluzin <[email protected]>
* Added tracking of threads running on AVR. Added detection of stack switching code and bookkeeping of the threads. Works for BeRTOS. Does not work for AvrX, Dr. Tak's Real Time Kernel and other multi-threading libraries which use a special temporary stack. Also GDB (7.2) kinda sucks in unwinding the stack, so we pretend state as it was in last function entry (which breaks FreeRTOS).
2011-07-06 Petr Hluzin <[email protected]>
* +List of last few 'call' and 'jump' executed. (For debugging.) Keep list of several last code jump to aid in debugging simulavr itself. This is specially useful when tracing is not enabled (too verbose).
2011-06-12 Petr Hluzin <[email protected]>
* Merge branch 'master' of git.sv.gnu.org:/srv/git/simulavr
2011-06-12 Petr Hluzin <[email protected]>
* More docs; detect invalid instructions. (From my old repo.) Detect if program attempts to execute LD or ST instruction which modifies X,Y, or Z register. Instruction datasheet 0856H–AVR–07/09 says "The result of these combinations is undefined". (This is also from my old SVN repository.)
2011-06-12 Petr Hluzin <[email protected]>
* Minor improvements from my old CVS working copy. Added assertions, documentation. Better names, word "address" used consistently with GDB, better scopes of local vars. Removed unused or trivially used member variables.
2011-05-01 Thomas K <[email protected]>
* Fix some bugs, which broke "make check"
2011-04-26 Petr Hluzin <[email protected]>
* Build failures on Debian. When building on "Debian GNU/Linux testing" users got compile errors due to undeclared stderr, printf and typeid. Patch provided by bug-reporter. Thanks! http://savannah.nongnu.org/bugs/?33148
2011-04-17 Petr Hluzin <[email protected]>
* Added comments and assertion. Removed endianity code. Endianity of flash is fixed, endianity of host is irrelevant.
2011-04-17 Petr Hluzin <[email protected]>
* Improved documentation (made shorter). If a documentation uses lot of words to tell little information then people will learn to avoid reading the documentation. Specifically avoid writing something what is obvious from the function's name and arguments. It is OK if doxygen docs do not make a sense without a name of its class/function (they are used together, anyway). Also "word" is a term for a CPU's natural size of data - which is 8 bits for AVRs. Also there is always exactly one AvrFlash per AvrDevice (and this is obvious), so "start of THIS memory block" is redundant. Also Flash always starts at 0, therefore offset is also the address.
2011-04-17 Petr Hluzin <[email protected]>
* Added comment about missing ESPM instruction, Removed unused member.
2011-04-17 Petr Hluzin <[email protected]>
* Unified line endings, I think. TortoiseGit says 24 lines were changed and TortoiseMerge indicates no leading whitespace was changed. So it must be newlines.
2011-04-17 Petr Hluzin <[email protected]>
* Incorrect jump if destination is >= 128 KiB. First 16b of JMP instruction encode 6 higher bits of destination address. A local var name made consistent with avr_op_CALL.
2011-04-17 Petr Hluzin <[email protected]>
* Added documentation, removed unused members, assertions. Conditions that are always true (and where false indicates problem) turned to assertions. Some variables are constant for all MCU types. Some ugly allocations turned to prettier C++ ones.
2011-04-17 Petr Hluzin <[email protected]>
* Added --gdb-debug option as a long alias for -G
2011-04-17 Petr Hluzin <[email protected]>
* Documentation, removal unused code. Better names for #include guards. Consistent formatting (done only on few places). Members allPins, GetPin() .... are not called. Remove? (Called from TCL/Python?)
2011-04-17 Petr Hluzin <[email protected]>
* Removing widespread case of nasty multiple inheritance with member variable. Member "trace_on" is only used on AvrDevice class.
2011-03-20 Petr Hluzin <[email protected]>
* Ignore Visual Studio's index files. They are 17 and 32 MB jumbos.
2011-03-16 Petr Hluzin <[email protected]>
* Failed to build on VS2008.
2011-03-06 Petr Hluzin <[email protected]>
* Suppressed printing "Going to gdb..." by default. Similar messages are printed only when 'global_verbose_on' is on. Also more informative and less jargon message "Waiting for connection from GDB on port XXX" is already printed later.
2011-03-06 Petr Hluzin <[email protected]>
* Reordered members of classes - makes instances smaller and consistent with other classes.
2011-03-06 Petr Hluzin <[email protected]>
* Added assertions. If there is a bug in caller (or any code for that matter): Fail early, fail loudly.
2011-02-27 Petr Hluzin <[email protected]>
* More understandable --help output. Users do not know or care what `gdb-server' is. We made up the word.
2011-02-27 Petr Hluzin <[email protected]>
* Fixed array index out of bounds crashes. avr_op_BRBC::Trace() and avr_op_BRBS::Trace() crash used this->bitmask in range 1..128 to index array with 8 entries. Bug #32592 reported and patch proposed by Patricio. Thanks Patricio! (I added detection of bogus bitmasks.)
2011-02-27 Petr Hluzin <[email protected]>
* Fixed more build failures on Cygwin. I accidentally left the Windows version of Socket::Poll() visible to Cygwin compiler.
2011-02-27 Petr Hluzin <[email protected]>
* Fixed build failures on Cygwin. Ancient GCC 3.4.4 selected bad constructor for std::vector<const Hardware*> debugInterruptTable: std::vector(InputIterator _First, InputIterator _Last); instead of std::vector(size_type _Count, const Type& _Val);
2011-02-27 Petr Hluzin <[email protected]>
* Added some assertions.
2011-02-27 Petr Hluzin <[email protected]>
* Merged similar parts of MinGW and Posix code. Makes it easier for code-navigation tools.
2011-02-27 Petr Hluzin <[email protected]>
* Class Socket did inherit from `sockstream' class even though the code never used it. I need to cleanup network code to implement for asynchronous sockets.
2011-02-27 Petr Hluzin <[email protected]>
* Do not burn 100% CPU when waiting for GDB commands. I think value-trace displays and "UIs" will not be able to communicate while simulavr is waiting for GDB. On the other side the code indicates that these clients cannot be used together with GDB anyway. Also GDB is still burning 100% waiting until GDB makes TCP connection. (This was the case even with original ioctlsocket(_socket, ...) code.)
2011-02-27 Petr Hluzin <[email protected]>
* Merge branch 'master' of git.sv.gnu.org:/srv/git/simulavr
2011-02-27 Petr Hluzin <[email protected]>
* Added GPIORx registers to atmega1284abase.h and atmega1284abase.h -based devices.
2011-02-27 Petr Hluzin <[email protected]>
* Added GPIORx registers to atmega1284abase.h and atmega1284abase.h -based devices.
2011-02-27 Petr Hluzin <[email protected]>
* Marking more registers as not-simulated.
2011-02-27 Petr Hluzin <[email protected]>
* Better diagnostic if user reads/writes a register we currently do not simulate. Accessing a register reserved by Atmel is a bug in user program. Accessing a register which simulavr currently does not simulate is a bummer, but not bug in user's program. Distinguish these two. Also USART1 registers were unreachable for ATmega164A/164PA/324A/324PA/644A/644PA/1284/1284P.
2011-02-27 Petr Hluzin <[email protected]>
* Some USART registers for ATmega48/88/168/328 were on wrong addresses. Corrected the addresses according to datasheet 8271 revision C. Added diagnostic for any user code relying on the wrong addresses in simulavr.
2011-02-27 Petr Hluzin <[email protected]>
* Added a dummy register type which writes diagnostic when accessed. This is needed to distinguish accesses to registers not present on silicon (InvalidMem) and present on silicon but not-yet-simulated registers.
2011-02-27 Petr Hluzin <[email protected]>
* Support for Added support for ATmega164A/164PA/324A/324PA/644A/644PA/1284/1284P was not being compiled.
2011-02-27 Petr Hluzin <[email protected]>
* Added support for ATmega164A/164PA/324A/324PA/644A/644PA/1284/1284P. As usually, functionality has been verified using datasheets, not by existing real code.
2011-02-27 Petr Hluzin <[email protected]>
* Removed extra whitespace and obvious comments.
2011-02-27 Petr Hluzin <[email protected]>
* Added copies of atmega668base.{c,h} files. This will make tracking of future differences easier.
2011-02-27 Petr Hluzin <[email protected]>
* Weird classes marked as unused. Their purpose is either unclear (to me). At least class PinChange has been obsoleted by ExternalIRQPort class. Should we remove these 2 files. (Their constructors can be made private, yet the code compiles. No friend declarations. No references from other files.)
2011-02-27 Petr Hluzin <[email protected]>
* Better code documentation. Better variable names.
2011-02-27 Petr Hluzin <[email protected]>
* Better developer diagnostic if device constructor creates multiple instances of e.g. HWUsart and forgets to specify a different ID.
2011-02-27 Petr Hluzin <[email protected]>
* Better debugging: detect if a device constructor uses the same interrupt index for multiple peripherals. Added code for dumping interrupt table in format suitable for verifying against Atmel datasheets.
2011-02-20 Petr Hluzin <[email protected]>
* ATmega8 capability was not compiled. It is hard to impress users with the set of supported MCUs if we do not actually compile them in, isn't it? (I verified Cygwin build and for some reason it now list ATmega8 and some much more devices. Dunno why.) MSVC build did not have the problem.
2011-02-20 Petr Hluzin <[email protected]>
* Fix initialization order of static variables across compilation units. Constructor of static AVRFactoryEntryMaker_at90s4433 called AvrFactory::reg() when 'devmap' was not initialized yet. This caused assertions/crash with MSVC. Now it works nice with MSVC and Cygwin.
2011-02-20 Petr Hluzin <[email protected]>
* My failed attempts to fix order of static initialization under MSVC compiler. Note: Order of initialization of static variables in different CPP files is not defined by C++ standard. It works with GCC by pure luck (or clever representation of uninitialized std::map). This will document for future generations my vain attempts to put the initialization into PE section in array before initializers in .crt$xcu. The ultimate fix goes in the next commit.
2011-02-20 Petr Hluzin <[email protected]>
* SWIG did does not like GCC attributes.
2011-02-19 Petr Hluzin <[email protected]>
* Add compiler attributes for better diagnostics. Silences warning C4715 'not all control paths return a value' in prescalermux.cpp.
2011-02-19 Petr Hluzin <[email protected]>
* +documentation
2011-02-19 Petr Hluzin <[email protected]>
* Append newlines for messages/warnings/errors. C++ library did buffer them indefinitely (bad) and then print bunch of messages on single line (also bad).
2011-02-19 Petr Hluzin <[email protected]>
* Adding support for compiling under MS Visual Studio 2008. See discussion [1] and this proposed patch/implementation [2]. Compiles and links under VS2008. When launching it asserts during static initialization in AvrFactory::reg() under Windows. Compiles fine under Cygwin, as always.
2011-02-19 Petr Hluzin <[email protected]>
* GDB 'disassemble' command displayed garbage. Recently reported by David Madden on mailing-list.
2011-02-19 Petr Hluzin <[email protected]>
* the remaining part of patch #7029: better diagnostics if user does not supply '--device' option
2011-02-19 Petr Hluzin <[email protected]>
* better comment, from patch #7063
2011-02-19 Petr Hluzin <[email protected]>
* signedness warning, unused variables
2011-02-19 Petr Hluzin <[email protected]>
* typo, whitespace changes, removed comment. (Vectors have to be appropriately sized for [] operation. But it is not a bug - they do not supposed to grow on assignment via [].)
2011-02-19 Petr Hluzin <[email protected]>
* Committing the remaining part of patch #7035
2011-02-19 Petr Hluzin <[email protected]>
* Removed #include <unistd.h> - the code does not use any of those functions listed at: http://pubs.opengroup.org/onlinepubs/007908799/xsh/unistd.h.html (Also Visual Studio does not ship the header.)
2011-02-19 Petr Hluzin <[email protected]>
* Fixed uninitialized variable for SP. Removed (an other) unused variable.
2011-02-19 Petr Hluzin <[email protected]>
* StringToUnsignedChar() incorrectly reported success when paring large numbers. Value 'LONG_MAX' on does not fit into 'unsigned char'. Correct testing for overflow by C specs is difficult, I doubt I fixed it all.
2011-02-19 Petr Hluzin <[email protected]>
* Avoid non-constant length of an array. (This also adds bounds checking.)
2011-02-19 Petr Hluzin <[email protected]>
* Added missing #include. Also moved 'using namespace' detect headers depending on it (they should not).
2011-02-19 Petr Hluzin <[email protected]>
* Uninitialized variables
2011-02-19 Petr Hluzin <[email protected]>
* Variadic macros converted to C99 style
2010-12-19 Petr Hluzin <invalid@invalid>
* Merge branch 'master' of git://git.savannah.nongnu.org/simulavr
2010-11-21 John McCullough <>
* AT90CAN support by John McCullough
2010-10-24 Petr Hluzin <invalid@invalid>
* documentation: some clarifications and phrasing
2010-10-24 Petr Hluzin <invalid@invalid>
* Fixed interrupt vectors for ATmega32.
2010-10-18 Petr Hluzin <invalid@invalid>
* Compilation errors introduced by my clumsy edits. Reported by Ivca Safranko.
2010-10-17 Petr Hluzn <invalid@invalid>
* +documentation
2010-10-17 Petr Hluzn <invalid@invalid>
* ATmega8 fixes: * interrupt table uses RJMP which is 2 bytes * there are 128 pages ("Table 89: No. of Words in a Page and no. of Pages in the Flash") * unnecessary cast
2010-10-17 Petr Hluzn <invalid@invalid>
* Add ATmega8 support. Written by Ivca Safranko. Many thanks. (Whitespace changes by me.)
2010-10-17 Petr Hluzn <invalid@invalid>
* minor: reworked overused 'stack11bit' parameter into child constructors. No functional changes.
2010-10-17 Petr Hluzn <invalid@invalid>
* ATmega32's "EEPROM Ready" interrupt is 17, not 15. Checked against ATmega32 datasheet revision P. (Note datasheet lists reset vector as 1, not 0 as we do.)
2010-09-05 unknown <[email protected]>
* fix reading a freed memory
2010-07-19 Thomas K <[email protected]>
* Remove unneccessary method wrapper from python SWIG file.
2010-07-19 Thomas K <[email protected]>
* Apply patch: Extended python interface: SetAnalog
2010-06-19 Thomas K <[email protected]>
* Bugfix for make install, bug reported by Sebastian on june 14th 2010
2010-06-15 Thomas K <[email protected]>
* Ignore file .settings for VC (for eclipse)
2010-06-15 Thomas K <[email protected]>
* Bugfix: wrong output for registers in decoder trace
2010-06-02 Thomas K <[email protected]>
* Hide eclipse project files from VCS
2010-04-19 Thomas K <[email protected]>
* Improve web site content and sphinx documentation
2010-04-19 Thomas K <[email protected]>
* Add multicore example to examples/python
2010-04-13 Thomas K <[email protected]>
* Bugfix: add missed config files for sphinx docs
2010-03-30 Thomas K <[email protected]>
* Integrate sources for creating simulavr web site
2010-03-24 Thomas K <[email protected]>
* Remove deprecation notice for TCL interface in sphinx documentation
2010-03-24 Thomas K <[email protected]>
* Update branches.txt
2010-03-24 Thomas K <[email protected]>
* Remove cvs-git-update.sh script
2010-03-24 Thomas K <[email protected]>
* README for documentation and update reST files to reflect change to simulavr
2010-03-24 Thomas K <[email protected]>
* Set version to 1.0rc0
2010-03-24 Thomas K <[email protected]>
* Revert "Change name of application from simulavr to avrs"
2010-03-24 Thomas K <[email protected]>
* Revert "Change examples to play with renamed program name / python interface"
2010-03-19 Thomas K <[email protected]>
* First version of new documentation with Sphinx, content is taken from texinfo documentation
2010-03-19 Thomas K <[email protected]>
* Bugfix II: make check on windows/msys platform
2010-03-18 Thomas K <[email protected]>
* Bugfix: fix 2 showstopper on "make check" on windows/msys platform
2010-03-16 Thomas K <[email protected]>
* Change examples to play with renamed program name / python interface
2010-03-16 Thomas K <[email protected]>
* make: reorder header files
2010-03-16 Thomas K <[email protected]>
* Bugfix enable-tcl.m4: this macro is wrong, but will not be fixed because TCL interface deprecated
2010-03-16 Thomas K <[email protected]>
* Change name of application from simulavr to avrs
2010-03-08 Thomas K <[email protected]>
* Changes accordingly patch 7084 (Petr Hluzin): Diagnostic when running without a program
2010-03-08 Thomas K <[email protected]>
* CHanges accordingly patch 7083 (Petr Hluzin): write out current PC on illegal IO access and on illegal instructions
2010-03-07 Thomas K <[email protected]>
* Changes accordingly patch 7079 (Petr Hluzin): finish changes in decoder
2010-03-06 Thomas K <[email protected]>
* CHanges accordingly patch 7079 (Petr Hluzin): change instructions till P
2010-03-04 Thomas K <[email protected]>
* Changes accordingly patch 7079 (Petr Hluzin): change instructions till J
2010-03-03 Thomas K <[email protected]>
* Changes accordingly patch 7079 (Petr Hluzin): introduce EIND register for devices, which use this
2010-03-03 Thomas K <[email protected]>
* Changes accordingly patch 7079 (Petr Hluzin): change instructions till from A to C
2010-03-03 Thomas K <[email protected]>
* Fix compiler warning in main.cpp: wrong formatting in format string
2010-03-03 Thomas K <[email protected]>
* Changes accordingly patch 7079 (Petr Hluzin): new methods to get access to registers and IO space
2010-03-01 Thomas K <[email protected]>
* Add HWStack definitions to python interface
2010-03-01 Onno Kortmann <[email protected]>
* Another TODO item.
2010-03-01 Onno Kortmann <[email protected]>
* Fixes for tracing of SPL/SPH and trace the three level stack pointer
2010-02-26 Thomas K <[email protected]>
* Merge branch 'master' of upload repo with new changes from my repo
2010-02-26 Thomas K <[email protected]>
* Rewrite HWStack, part III: Reimplementation of ThreeLevelStack
2010-02-26 Thomas K <[email protected]>
* Rewrite HWStack, part II: give core access to stack for bytes and addresses
2010-02-25 Onno Kortmann <[email protected]>
* Fix program counter so that it stays on the 'right PC' for long commands
2010-02-25 Onno Kortmann <[email protected]>
* CORE.PCb trace variable giving twice the PC for easier debugging
2010-02-25 Onno Kortmann <[email protected]>
* Make traceval_direct return registered TraceValues
2010-02-25 Onno Kortmann <[email protected]>
* Fix in avr.v to support Icarus Verilog 0.9.x
2010-02-25 Onno Kortmann <[email protected]>
* PCs in sync in verilog and in AVR core.
2010-02-25 Onno Kortmann <[email protected]>
* Set time in AVR library code
2010-02-25 Onno Kortmann <[email protected]>
* AVR VPI interface: start, stop and set-time implementation
2010-02-25 Onno Kortmann <[email protected]>
* Implement SetCurrentTime() to allow driving AVRs from outside software
2010-02-25 Onno Kortmann <[email protected]>
* Export of SetDumpTraceArgs as $avr_dumparg() to Verilog
2010-02-25 Onno Kortmann <[email protected]>
* VPI minor fix for avr_trace
2010-02-25 Onno Kortmann <[email protected]>
* TODO items and minor documentation fixes
2010-02-25 Onno Kortmann <[email protected]>
* Change documented semantics of TraceValue::cycle() call
2010-02-25 Onno Kortmann <[email protected]>
* In traceval.*, make all error messages unique
2010-02-25 Onno Kortmann <[email protected]>
* Hex output for invalid memory accesses
2010-02-25 Onno Kortmann <[email protected]>
* Verilog header location fix for (K)ubuntu 9.10
2010-02-24 Thomas K <[email protected]>
* Rewrite HWStack, part I, inspired from patch 7079 (Petr Hluzin)
2010-02-23 Thomas K <[email protected]>
* Change accordingly patch 7064 (Petr Hluzin): diagnostics if wrong gdb connects
2010-02-23 Thomas K <[email protected]>
* Changes accordingly patch 7063 (Petr Hluzin): local variables' fixes
2010-02-22 Onno Kortmann <[email protected]>
* Include fix for traceval.h to enable compiling on (K)ubuntu 9.10.
2010-02-17 Onno Kortmann <[email protected]>
* Check for /usr/bin/time in configure
2010-02-17 Onno Kortmann <[email protected]>
* Changes according to patch 7063 (Petr Hluzin): single device clock and unitialized variable fixes
2010-02-17 Onno Kortmann <[email protected]>
* Display freq in MHz and Hz to make all people happy
2010-02-17 Onno Kortmann <[email protected]>
* Changes according to patch 7034 (Petr Hluzin): display frequency in fool-proof way
2010-02-17 Onno Kortmann <[email protected]>
* Fix for the dependencies to remake avr.vpi properly
2010-02-16 Onno Kortmann <[email protected]>
* Fix of template files for AVR factory for newer version of python-cheetah
2010-02-16 Onno Kortmann <[email protected]>
* Remove casting of pointers before equality check
2010-01-15 Thomas K <[email protected]>
* automake: Change m4-macros to give the possibility to build python interface with Python3
2010-01-09 Thomas K <[email protected]>
* Changes accordingly patch 7036 and 7037 (Petr Hluzin): allow bigger packet sizes than 400 chars
2010-01-07 Thomas K <[email protected]>
* Changes accordingly patch 7035 (Petr Hluzin): dependence on host byte order (endianity)
2010-01-06 Thomas K <[email protected]>
* Change accordingly patch 7033 (Petr Hluzin): sizeof value for accept call
2010-01-06 Thomas K <[email protected]>
* Changes accordingly patch 7032 (Petr Hluzin): uninitialized vars in HWUart
2010-01-05 Thomas K <[email protected]>
* Changes accordingly patch 7031 (Petr Hluzin): fix minor memory leaks
2010-01-05 Thomas K <[email protected]>
* Change accordingly patch 7029 (Petr Hluzin): accept canonical µC names
2009-12-14 Thomas K <[email protected]>
* Update texinfo documentation following on changes in examples
2009-12-13 Thomas K <[email protected]>
* automake: add test for Itcl and change Makefile.am in examples to respect this
2009-12-13 Thomas K <[email protected]>
* Bugfix: fix changed classes for simulavr.tcl to get tcl examples working again
2009-12-11 Thomas K <[email protected]>
* Bugfix zum Change: Free all allocated space from AvrDevice in destructor
2009-12-11 Thomas K <[email protected]>
* Add a method to AvrDevice to get a memory cell instance (RWMemoryMember)
2009-12-11 Thomas K <[email protected]>
* Free all allocated space from AvrDevice in destructor
2009-12-11 Thomas K <[email protected]>
* Remove old sources for external irq, remove unused HWMcucr class
2009-12-11 Thomas K <[email protected]>
* Implement rewrite of external interrupts also for AT90S8515 and AT90S4433
2009-12-10 Thomas K <[email protected]>
* Bugfix: make dist wasn't working and some files are missed in dist package
2009-12-09 Thomas K <[email protected]>
* Change external interrupt implementation on ATmegaX8 and ATmega128 to new implementation
2009-12-09 Thomas K <[email protected]>
* Bugfix external interrupts: class destructor and a bug in ExternalIRQPort
2009-12-09 Thomas K <[email protected]>
* Add more tests for external interrupts
2009-12-09 Thomas K <[email protected]>
* Bugfix: makefile template in regress/extinttest missed definitions from Makefile
2009-12-09 Thomas K <[email protected]>
* Add Makefile in regress/extinttest to autoconf configuration
2009-12-09 Thomas K <[email protected]>
* Rename python modul timer_testall to regress_unittest
2009-12-09 Thomas K <[email protected]>
* Add regression test for external interrupt INT0 and INT1
2009-12-08 Thomas K <[email protected]>
* Bugfix ATtiny2313: wrong stack ceil
2009-12-08 Thomas K <[email protected]>
* Bugfix ATmega16/32: GICR/GIFR with wrong addresses
2009-12-08 Thomas K <[email protected]>
* Bugfix ATmega48/88 ...: wrong pins for INT0/1
2009-12-08 Thomas K <[email protected]>
* Extend IRQ system to support level interrupts
2009-12-08 Thomas K <[email protected]>
* HWTimer: reset interrupt flag on writing 1 to flag register
2009-12-08 Thomas K <[email protected]>
* ATtiny2313: add reset function for GPIO registers
2009-12-07 Thomas K <[email protected]>
* Rewrite external IRQ functionality: fill methods with functionality
2009-12-05 Thomas K <[email protected]>
* Bugfix: HWPort::CalcOutputs was wrong
2009-12-04 Thomas K <[email protected]>
* Add empty framework for rewrite of external interrupt handling
2009-12-04 Thomas K <[email protected]>
* HWPort: new features PIN toggle PORT and variable port size
2009-12-03 Thomas K <[email protected]>
* Bugfix: remove access to removed global variable and correct a python example
2009-12-03 Thomas K <[email protected]>
* Cleanup code in pin.cpp/h and net.cpp/h
2009-12-02 Thomas K <[email protected]>
* Rewrite of classes Pin and Net, MirrorNet is now obsolet
2009-12-02 Thomas K <[email protected]>
* automake: provide setup.py for install python module by using python's own install functionality
2009-11-23 Thomas K <[email protected]>
* automake: bugfix for install python module on Windows/MinGW
2009-11-22 Thomas K <[email protected]>
* automake: complete install target for make to get a full installation
2009-11-16 Thomas K <[email protected]>
* Bugfix HWEeprom: wrong time calculation on write eeprom
2009-11-11 Thomas K <[email protected]>
* Merge branch 'extra-dev' of /media/exchange/simulavr into extra-dev
2009-11-11 Thomas K <[email protected]>
* Add methods RunTimeRange and ResetClock to SystemClock class
2009-11-06 Thomas K <[email protected]>
* Update tcl interface because of removing trace.cpp
2009-11-06 Thomas K <[email protected]>
* Remove dependencies to trace.h and delete trace.cpp/trace.h
2009-11-06 Thomas K <[email protected]>
* Switch trace usage to sysConHandler, trace.cpp isn't used now
2009-11-06 Thomas K <[email protected]>
* Implement functionality for trace support in SystemConsoleHandler class
2009-11-06 Thomas K <[email protected]>
* Remove unused code from trace.cpp, this removes also option -M from main.cpp
2009-11-04 Thomas K <[email protected]>
* AvrDevice: save IRam and ERam size and implement 2 access methods to get this values back
2009-11-04 Thomas K <[email protected]>
* HWEeprom: initialize eeprom memory to 0xff (as in a "fresh" device)
2009-11-02 Thomas K <[email protected]>
* Implement ATtiny2313