forked from danmar/cppcheck
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathcheckbufferoverrun.cpp
More file actions
1071 lines (960 loc) · 45.4 KB
/
checkbufferoverrun.cpp
File metadata and controls
1071 lines (960 loc) · 45.4 KB
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
/*
* Cppcheck - A tool for static C/C++ code analysis
* Copyright (C) 2007-2022 Cppcheck team.
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
//---------------------------------------------------------------------------
// Buffer overrun..
//---------------------------------------------------------------------------
#include "checkbufferoverrun.h"
#include "astutils.h"
#include "errorlogger.h"
#include "library.h"
#include "mathlib.h"
#include "settings.h"
#include "symboldatabase.h"
#include "token.h"
#include "tokenize.h"
#include "utils.h"
#include "valueflow.h"
#include <algorithm>
#include <cstdlib>
#include <functional>
#include <iterator>
#include <memory>
#include <numeric> // std::accumulate
#include <sstream>
#include <tinyxml2.h>
//---------------------------------------------------------------------------
// Register this check class (by creating a static instance of it)
namespace {
CheckBufferOverrun instance;
}
//---------------------------------------------------------------------------
// CWE ids used:
static const CWE CWE131(131U); // Incorrect Calculation of Buffer Size
static const CWE CWE170(170U); // Improper Null Termination
static const CWE CWE_ARGUMENT_SIZE(398U); // Indicator of Poor Code Quality
static const CWE CWE_ARRAY_INDEX_THEN_CHECK(398U); // Indicator of Poor Code Quality
static const CWE CWE682(682U); // Incorrect Calculation
static const CWE CWE758(758U); // Reliance on Undefined, Unspecified, or Implementation-Defined Behavior
static const CWE CWE_POINTER_ARITHMETIC_OVERFLOW(758U); // Reliance on Undefined, Unspecified, or Implementation-Defined Behavior
static const CWE CWE_BUFFER_UNDERRUN(786U); // Access of Memory Location Before Start of Buffer
static const CWE CWE_BUFFER_OVERRUN(788U); // Access of Memory Location After End of Buffer
//---------------------------------------------------------------------------
static const ValueFlow::Value *getBufferSizeValue(const Token *tok)
{
const std::list<ValueFlow::Value> &tokenValues = tok->values();
const auto it = std::find_if(tokenValues.begin(), tokenValues.end(), std::mem_fn(&ValueFlow::Value::isBufferSizeValue));
return it == tokenValues.end() ? nullptr : &*it;
}
static int getMinFormatStringOutputLength(const std::vector<const Token*> ¶meters, nonneg int formatStringArgNr)
{
if (formatStringArgNr <= 0 || formatStringArgNr > parameters.size())
return 0;
if (parameters[formatStringArgNr - 1]->tokType() != Token::eString)
return 0;
const std::string &formatString = parameters[formatStringArgNr - 1]->str();
bool percentCharFound = false;
int outputStringSize = 0;
bool handleNextParameter = false;
std::string digits_string;
bool i_d_x_f_found = false;
int parameterLength = 0;
int inputArgNr = formatStringArgNr;
for (int i = 1; i + 1 < formatString.length(); ++i) {
if (formatString[i] == '\\') {
if (i < formatString.length() - 1 && formatString[i + 1] == '0')
break;
++outputStringSize;
++i;
continue;
}
if (percentCharFound) {
switch (formatString[i]) {
case 'f':
case 'x':
case 'X':
case 'i':
i_d_x_f_found = true;
handleNextParameter = true;
parameterLength = 1; // TODO
break;
case 'c':
case 'e':
case 'E':
case 'g':
case 'o':
case 'u':
case 'p':
case 'n':
handleNextParameter = true;
parameterLength = 1; // TODO
break;
case 'd':
i_d_x_f_found = true;
parameterLength = 1;
if (inputArgNr < parameters.size() && parameters[inputArgNr]->hasKnownIntValue())
parameterLength = MathLib::toString(parameters[inputArgNr]->getKnownIntValue()).length();
handleNextParameter = true;
break;
case 's':
parameterLength = 0;
if (inputArgNr < parameters.size() && parameters[inputArgNr]->tokType() == Token::eString)
parameterLength = Token::getStrLength(parameters[inputArgNr]);
handleNextParameter = true;
break;
}
}
if (formatString[i] == '%')
percentCharFound = !percentCharFound;
else if (percentCharFound) {
digits_string.append(1, formatString[i]);
}
if (!percentCharFound)
outputStringSize++;
if (handleNextParameter) {
int tempDigits = std::abs(std::atoi(digits_string.c_str()));
if (i_d_x_f_found)
tempDigits = std::max(tempDigits, 1);
if (digits_string.find('.') != std::string::npos) {
const std::string endStr = digits_string.substr(digits_string.find('.') + 1);
const int maxLen = std::max(std::abs(std::atoi(endStr.c_str())), 1);
if (formatString[i] == 's') {
// For strings, the length after the dot "%.2s" will limit
// the length of the string.
if (parameterLength > maxLen)
parameterLength = maxLen;
} else {
// For integers, the length after the dot "%.2d" can
// increase required length
if (tempDigits < maxLen)
tempDigits = maxLen;
}
}
if (tempDigits < parameterLength)
outputStringSize += parameterLength;
else
outputStringSize += tempDigits;
parameterLength = 0;
digits_string.clear();
i_d_x_f_found = false;
percentCharFound = false;
handleNextParameter = false;
++inputArgNr;
}
}
return outputStringSize;
}
//---------------------------------------------------------------------------
static bool getDimensionsEtc(const Token * const arrayToken, const Settings *settings, std::vector<Dimension> * const dimensions, ErrorPath * const errorPath, bool * const mightBeLarger, MathLib::bigint* path)
{
const Token *array = arrayToken;
while (Token::Match(array, ".|::"))
array = array->astOperand2();
if (array->variable() && array->variable()->isArray() && !array->variable()->dimensions().empty()) {
*dimensions = array->variable()->dimensions();
if (dimensions->size() >= 1 && ((*dimensions)[0].num <= 1 || !(*dimensions)[0].tok)) {
visitAstNodes(arrayToken,
[&](const Token *child) {
if (child->originalName() == "->") {
*mightBeLarger = true;
return ChildrenToVisit::none;
}
return ChildrenToVisit::op1_and_op2;
});
}
} else if (const Token *stringLiteral = array->getValueTokenMinStrSize(settings)) {
Dimension dim;
dim.tok = nullptr;
dim.num = Token::getStrArraySize(stringLiteral);
dim.known = array->hasKnownValue();
dimensions->emplace_back(dim);
} else if (array->valueType() && array->valueType()->pointer >= 1 && (array->valueType()->isIntegral() || array->valueType()->isFloat())) {
const ValueFlow::Value *value = getBufferSizeValue(array);
if (!value)
return false;
if (path)
*path = value->path;
*errorPath = value->errorPath;
Dimension dim;
dim.known = value->isKnown();
dim.tok = nullptr;
const int typeSize = array->valueType()->typeSize(*settings, array->valueType()->pointer > 1);
if (typeSize == 0)
return false;
dim.num = value->intvalue / typeSize;
dimensions->emplace_back(dim);
}
return !dimensions->empty();
}
static ValueFlow::Value makeSizeValue(MathLib::bigint size, MathLib::bigint path)
{
ValueFlow::Value v(size);
v.path = path;
return v;
}
static std::vector<ValueFlow::Value> getOverrunIndexValues(const Token* tok,
const Token* arrayToken,
const std::vector<Dimension>& dimensions,
const std::vector<const Token*>& indexTokens,
MathLib::bigint path)
{
const Token *array = arrayToken;
while (Token::Match(array, ".|::"))
array = array->astOperand2();
bool isArrayIndex = tok->str() == "[";
if (isArrayIndex) {
const Token* parent = tok;
while (Token::simpleMatch(parent, "["))
parent = parent->astParent();
if (!parent || parent->isUnaryOp("&"))
isArrayIndex = false;
}
bool overflow = false;
std::vector<ValueFlow::Value> indexValues;
for (int i = 0; i < dimensions.size() && i < indexTokens.size(); ++i) {
MathLib::bigint size = dimensions[i].num;
if (!isArrayIndex)
size++;
const bool zeroArray = array->variable() && array->variable()->isArray() && dimensions[i].num == 0;
std::vector<ValueFlow::Value> values = !zeroArray
? ValueFlow::isOutOfBounds(makeSizeValue(size, path), indexTokens[i])
: std::vector<ValueFlow::Value>{};
if (values.empty()) {
if (indexTokens[i]->hasKnownIntValue())
indexValues.push_back(indexTokens[i]->values().front());
else
indexValues.push_back(ValueFlow::Value::unknown());
continue;
}
overflow = true;
indexValues.push_back(values.front());
}
if (overflow)
return indexValues;
return {};
}
void CheckBufferOverrun::arrayIndex()
{
for (const Token *tok = mTokenizer->tokens(); tok; tok = tok->next()) {
if (tok->str() != "[")
continue;
const Token *array = tok->astOperand1();
while (Token::Match(array, ".|::"))
array = array->astOperand2();
if (!array || ((!array->variable() || array->variable()->nameToken() == array) && array->tokType() != Token::eString))
continue;
if (!array->scope()->isExecutable()) {
// LHS in non-executable scope => This is just a definition
const Token *parent = tok;
while (parent && !Token::simpleMatch(parent->astParent(), "="))
parent = parent->astParent();
if (!parent || parent == parent->astParent()->astOperand1())
continue;
}
if (astIsContainer(array))
continue;
std::vector<const Token *> indexTokens;
for (const Token *tok2 = tok; tok2 && tok2->str() == "["; tok2 = tok2->link()->next()) {
if (!tok2->astOperand2()) {
indexTokens.clear();
break;
}
indexTokens.emplace_back(tok2->astOperand2());
}
if (indexTokens.empty())
continue;
std::vector<Dimension> dimensions;
ErrorPath errorPath;
bool mightBeLarger = false;
MathLib::bigint path = 0;
if (!getDimensionsEtc(tok->astOperand1(), mSettings, &dimensions, &errorPath, &mightBeLarger, &path))
continue;
// Positive index
if (!mightBeLarger) { // TODO check arrays with dim 1 also
const std::vector<ValueFlow::Value>& indexValues =
getOverrunIndexValues(tok, tok->astOperand1(), dimensions, indexTokens, path);
if (!indexValues.empty())
arrayIndexError(tok, dimensions, indexValues);
}
// Negative index
bool neg = false;
std::vector<ValueFlow::Value> negativeIndexes;
for (const Token * indexToken : indexTokens) {
const ValueFlow::Value *negativeValue = indexToken->getValueLE(-1, mSettings);
if (negativeValue) {
negativeIndexes.emplace_back(*negativeValue);
neg = true;
} else {
negativeIndexes.emplace_back(ValueFlow::Value::unknown());
}
}
if (neg) {
negativeIndexError(tok, dimensions, negativeIndexes);
}
}
}
static std::string stringifyIndexes(const std::string& array, const std::vector<ValueFlow::Value>& indexValues)
{
if (indexValues.size() == 1)
return MathLib::toString(indexValues[0].intvalue);
std::ostringstream ret;
ret << array;
for (const ValueFlow::Value& index : indexValues) {
ret << "[";
if (index.isNonValue())
ret << "*";
else
ret << index.intvalue;
ret << "]";
}
return ret.str();
}
static std::string arrayIndexMessage(const Token* tok,
const std::vector<Dimension>& dimensions,
const std::vector<ValueFlow::Value>& indexValues,
const Token* condition)
{
auto add_dim = [](const std::string &s, const Dimension &dim) {
return s + "[" + MathLib::toString(dim.num) + "]";
};
const std::string array = std::accumulate(dimensions.begin(), dimensions.end(), tok->astOperand1()->expressionString(), add_dim);
std::ostringstream errmsg;
if (condition)
errmsg << ValueFlow::eitherTheConditionIsRedundant(condition)
<< " or the array '" + array + "' is accessed at index " << stringifyIndexes(tok->astOperand1()->expressionString(), indexValues) << ", which is out of bounds.";
else
errmsg << "Array '" << array << "' accessed at index " << stringifyIndexes(tok->astOperand1()->expressionString(), indexValues) << ", which is out of bounds.";
return errmsg.str();
}
void CheckBufferOverrun::arrayIndexError(const Token* tok,
const std::vector<Dimension>& dimensions,
const std::vector<ValueFlow::Value>& indexes)
{
if (!tok) {
reportError(tok, Severity::error, "arrayIndexOutOfBounds", "Array 'arr[16]' accessed at index 16, which is out of bounds.", CWE_BUFFER_OVERRUN, Certainty::normal);
reportError(tok, Severity::warning, "arrayIndexOutOfBoundsCond", "Array 'arr[16]' accessed at index 16, which is out of bounds.", CWE_BUFFER_OVERRUN, Certainty::normal);
return;
}
const Token *condition = nullptr;
const ValueFlow::Value *index = nullptr;
for (const ValueFlow::Value& indexValue : indexes) {
if (!indexValue.errorSeverity() && !mSettings->severity.isEnabled(Severity::warning))
return;
if (indexValue.condition)
condition = indexValue.condition;
if (!index || !indexValue.errorPath.empty())
index = &indexValue;
}
reportError(getErrorPath(tok, index, "Array index out of bounds"),
index->errorSeverity() ? Severity::error : Severity::warning,
index->condition ? "arrayIndexOutOfBoundsCond" : "arrayIndexOutOfBounds",
arrayIndexMessage(tok, dimensions, indexes, condition),
CWE_BUFFER_OVERRUN,
index->isInconclusive() ? Certainty::inconclusive : Certainty::normal);
}
void CheckBufferOverrun::negativeIndexError(const Token* tok,
const std::vector<Dimension>& dimensions,
const std::vector<ValueFlow::Value>& indexes)
{
if (!tok) {
reportError(tok, Severity::error, "negativeIndex", "Negative array index", CWE_BUFFER_UNDERRUN, Certainty::normal);
return;
}
const Token *condition = nullptr;
const ValueFlow::Value *negativeValue = nullptr;
for (const ValueFlow::Value& indexValue : indexes) {
if (!indexValue.errorSeverity() && !mSettings->severity.isEnabled(Severity::warning))
return;
if (indexValue.condition)
condition = indexValue.condition;
if (!negativeValue || !indexValue.errorPath.empty())
negativeValue = &indexValue;
}
reportError(getErrorPath(tok, negativeValue, "Negative array index"),
negativeValue->errorSeverity() ? Severity::error : Severity::warning,
"negativeIndex",
arrayIndexMessage(tok, dimensions, indexes, condition),
CWE_BUFFER_UNDERRUN,
negativeValue->isInconclusive() ? Certainty::inconclusive : Certainty::normal);
}
//---------------------------------------------------------------------------
void CheckBufferOverrun::pointerArithmetic()
{
if (!mSettings->severity.isEnabled(Severity::portability))
return;
for (const Token *tok = mTokenizer->tokens(); tok; tok = tok->next()) {
if (!Token::Match(tok, "+|-"))
continue;
if (!tok->valueType() || tok->valueType()->pointer == 0)
continue;
if (!tok->isBinaryOp())
continue;
if (!tok->astOperand1()->valueType() || !tok->astOperand2()->valueType())
continue;
const Token *arrayToken, *indexToken;
if (tok->astOperand1()->valueType()->pointer > 0) {
arrayToken = tok->astOperand1();
indexToken = tok->astOperand2();
} else {
arrayToken = tok->astOperand2();
indexToken = tok->astOperand1();
}
if (!indexToken || !indexToken->valueType() || indexToken->valueType()->pointer > 0 || !indexToken->valueType()->isIntegral())
continue;
std::vector<Dimension> dimensions;
ErrorPath errorPath;
bool mightBeLarger = false;
MathLib::bigint path = 0;
if (!getDimensionsEtc(arrayToken, mSettings, &dimensions, &errorPath, &mightBeLarger, &path))
continue;
if (tok->str() == "+") {
// Positive index
if (!mightBeLarger) { // TODO check arrays with dim 1 also
const std::vector<const Token *> indexTokens{indexToken};
const std::vector<ValueFlow::Value>& indexValues =
getOverrunIndexValues(tok, arrayToken, dimensions, indexTokens, path);
if (!indexValues.empty())
pointerArithmeticError(tok, indexToken, &indexValues.front());
}
if (const ValueFlow::Value *neg = indexToken->getValueLE(-1, mSettings))
pointerArithmeticError(tok, indexToken, neg);
} else if (tok->str() == "-") {
if (arrayToken->variable() && arrayToken->variable()->isArgument())
continue;
const Token *array = arrayToken;
while (Token::Match(array, ".|::"))
array = array->astOperand2();
if (array->variable() && array->variable()->isArray()) {
const ValueFlow::Value *v = indexToken->getValueGE(1, mSettings);
if (v)
pointerArithmeticError(tok, indexToken, v);
}
}
}
}
void CheckBufferOverrun::pointerArithmeticError(const Token *tok, const Token *indexToken, const ValueFlow::Value *indexValue)
{
if (!tok) {
reportError(tok, Severity::portability, "pointerOutOfBounds", "Pointer arithmetic overflow.", CWE_POINTER_ARITHMETIC_OVERFLOW, Certainty::normal);
reportError(tok, Severity::portability, "pointerOutOfBoundsCond", "Pointer arithmetic overflow.", CWE_POINTER_ARITHMETIC_OVERFLOW, Certainty::normal);
return;
}
std::string errmsg;
if (indexValue->condition)
errmsg = "Undefined behaviour, when '" + indexToken->expressionString() + "' is " + MathLib::toString(indexValue->intvalue) + " the pointer arithmetic '" + tok->expressionString() + "' is out of bounds.";
else
errmsg = "Undefined behaviour, pointer arithmetic '" + tok->expressionString() + "' is out of bounds.";
reportError(getErrorPath(tok, indexValue, "Pointer arithmetic overflow"),
Severity::portability,
indexValue->condition ? "pointerOutOfBoundsCond" : "pointerOutOfBounds",
errmsg,
CWE_POINTER_ARITHMETIC_OVERFLOW,
indexValue->isInconclusive() ? Certainty::inconclusive : Certainty::normal);
}
//---------------------------------------------------------------------------
ValueFlow::Value CheckBufferOverrun::getBufferSize(const Token *bufTok) const
{
if (!bufTok->valueType())
return ValueFlow::Value(-1);
const Variable *var = bufTok->variable();
if (!var || var->dimensions().empty()) {
const ValueFlow::Value *value = getBufferSizeValue(bufTok);
if (value)
return *value;
}
if (!var)
return ValueFlow::Value(-1);
MathLib::bigint dim = std::accumulate(var->dimensions().begin(), var->dimensions().end(), 1LL, [](MathLib::bigint i1, const Dimension &dim) {
return i1 * dim.num;
});
ValueFlow::Value v;
v.setKnown();
v.valueType = ValueFlow::Value::ValueType::BUFFER_SIZE;
if (var->isPointerArray())
v.intvalue = dim * mSettings->sizeof_pointer;
else if (var->isPointer())
return ValueFlow::Value(-1);
else {
const MathLib::bigint typeSize = bufTok->valueType()->typeSize(*mSettings);
v.intvalue = dim * typeSize;
}
return v;
}
//---------------------------------------------------------------------------
static bool checkBufferSize(const Token *ftok, const Library::ArgumentChecks::MinSize &minsize, const std::vector<const Token *> &args, const MathLib::bigint bufferSize, const Settings *settings)
{
const Token * const arg = (minsize.arg > 0 && minsize.arg - 1 < args.size()) ? args[minsize.arg - 1] : nullptr;
const Token * const arg2 = (minsize.arg2 > 0 && minsize.arg2 - 1 < args.size()) ? args[minsize.arg2 - 1] : nullptr;
switch (minsize.type) {
case Library::ArgumentChecks::MinSize::Type::STRLEN:
if (settings->library.isargformatstr(ftok, minsize.arg)) {
return getMinFormatStringOutputLength(args, minsize.arg) < bufferSize;
} else if (arg) {
const Token *strtoken = arg->getValueTokenMaxStrLength();
if (strtoken)
return Token::getStrLength(strtoken) < bufferSize;
}
break;
case Library::ArgumentChecks::MinSize::Type::ARGVALUE:
if (arg && arg->hasKnownIntValue())
return arg->getKnownIntValue() <= bufferSize;
break;
case Library::ArgumentChecks::MinSize::Type::SIZEOF:
// TODO
break;
case Library::ArgumentChecks::MinSize::Type::MUL:
if (arg && arg2 && arg->hasKnownIntValue() && arg2->hasKnownIntValue())
return (arg->getKnownIntValue() * arg2->getKnownIntValue()) <= bufferSize;
break;
case Library::ArgumentChecks::MinSize::Type::VALUE:
return minsize.value <= bufferSize;
case Library::ArgumentChecks::MinSize::Type::NONE:
break;
}
return true;
}
void CheckBufferOverrun::bufferOverflow()
{
const SymbolDatabase *symbolDatabase = mTokenizer->getSymbolDatabase();
for (const Scope * scope : symbolDatabase->functionScopes) {
for (const Token *tok = scope->bodyStart; tok != scope->bodyEnd; tok = tok->next()) {
if (!Token::Match(tok, "%name% (") || Token::simpleMatch(tok, ") {"))
continue;
if (!mSettings->library.hasminsize(tok))
continue;
const std::vector<const Token *> args = getArguments(tok);
for (int argnr = 0; argnr < args.size(); ++argnr) {
if (!args[argnr]->valueType() || args[argnr]->valueType()->pointer == 0)
continue;
const std::vector<Library::ArgumentChecks::MinSize> *minsizes = mSettings->library.argminsizes(tok, argnr + 1);
if (!minsizes || minsizes->empty())
continue;
// Get buffer size..
const Token *argtok = args[argnr];
while (argtok && argtok->isCast())
argtok = argtok->astOperand2() ? argtok->astOperand2() : argtok->astOperand1();
while (Token::Match(argtok, ".|::"))
argtok = argtok->astOperand2();
if (!argtok || !argtok->variable())
continue;
if (argtok->valueType() && argtok->valueType()->pointer == 0)
continue;
// TODO: strcpy(buf+10, "hello");
const ValueFlow::Value bufferSize = getBufferSize(argtok);
if (bufferSize.intvalue <= 0)
continue;
// buffer size == 1 => do not warn for dynamic memory
if (bufferSize.intvalue == 1) {
const Token *tok2 = argtok;
while (Token::simpleMatch(tok2->astParent(), "."))
tok2 = tok2->astParent();
while (Token::Match(tok2, "[|."))
tok2 = tok2->astOperand1();
const Variable *var = tok2 ? tok2->variable() : nullptr;
if (var) {
if (var->isPointer())
continue;
if (var->isArgument() && var->isReference())
continue;
}
}
const bool error = std::none_of(minsizes->begin(), minsizes->end(), [=](const Library::ArgumentChecks::MinSize &minsize) {
return checkBufferSize(tok, minsize, args, bufferSize.intvalue, mSettings);
});
if (error)
bufferOverflowError(args[argnr], &bufferSize, (bufferSize.intvalue == 1) ? Certainty::inconclusive : Certainty::normal);
}
}
}
}
void CheckBufferOverrun::bufferOverflowError(const Token *tok, const ValueFlow::Value *value, const Certainty::CertaintyLevel &certainty)
{
reportError(getErrorPath(tok, value, "Buffer overrun"), Severity::error, "bufferAccessOutOfBounds", "Buffer is accessed out of bounds: " + (tok ? tok->expressionString() : "buf"), CWE_BUFFER_OVERRUN, certainty);
}
//---------------------------------------------------------------------------
void CheckBufferOverrun::arrayIndexThenCheck()
{
if (!mSettings->severity.isEnabled(Severity::portability))
return;
const SymbolDatabase *symbolDatabase = mTokenizer->getSymbolDatabase();
for (const Scope * const scope : symbolDatabase->functionScopes) {
for (const Token *tok = scope->bodyStart; tok && tok != scope->bodyEnd; tok = tok->next()) {
if (Token::simpleMatch(tok, "sizeof (")) {
tok = tok->linkAt(1);
continue;
}
if (Token::Match(tok, "%name% [ %var% ]")) {
tok = tok->next();
const int indexID = tok->next()->varId();
const std::string& indexName(tok->strAt(1));
// Iterate AST upwards
const Token* tok2 = tok;
const Token* tok3 = tok2;
while (tok2->astParent() && tok2->tokType() != Token::eLogicalOp && tok2->str() != "?") {
tok3 = tok2;
tok2 = tok2->astParent();
}
// Ensure that we ended at a logical operator and that we came from its left side
if (tok2->tokType() != Token::eLogicalOp || tok2->astOperand1() != tok3)
continue;
// check if array index is ok
// statement can be closed in parentheses, so "(| " is using
if (Token::Match(tok2, "&& (| %varid% <|<=", indexID))
arrayIndexThenCheckError(tok, indexName);
else if (Token::Match(tok2, "&& (| %any% >|>= %varid% !!+", indexID))
arrayIndexThenCheckError(tok, indexName);
}
}
}
}
void CheckBufferOverrun::arrayIndexThenCheckError(const Token *tok, const std::string &indexName)
{
reportError(tok, Severity::style, "arrayIndexThenCheck",
"$symbol:" + indexName + "\n"
"Array index '$symbol' is used before limits check.\n"
"Defensive programming: The variable '$symbol' is used as an array index before it "
"is checked that is within limits. This can mean that the array might be accessed out of bounds. "
"Reorder conditions such as '(a[i] && i < 10)' to '(i < 10 && a[i])'. That way the array will "
"not be accessed if the index is out of limits.", CWE_ARRAY_INDEX_THEN_CHECK, Certainty::normal);
}
//---------------------------------------------------------------------------
void CheckBufferOverrun::stringNotZeroTerminated()
{
// this is currently 'inconclusive'. See TestBufferOverrun::terminateStrncpy3
if (!mSettings->severity.isEnabled(Severity::warning) || !mSettings->certainty.isEnabled(Certainty::inconclusive))
return;
const SymbolDatabase *symbolDatabase = mTokenizer->getSymbolDatabase();
for (const Scope * const scope : symbolDatabase->functionScopes) {
for (const Token *tok = scope->bodyStart; tok && tok != scope->bodyEnd; tok = tok->next()) {
if (!Token::simpleMatch(tok, "strncpy ("))
continue;
const std::vector<const Token *> args = getArguments(tok);
if (args.size() != 3)
continue;
const Token *sizeToken = args[2];
if (!sizeToken->hasKnownIntValue())
continue;
const ValueFlow::Value &bufferSize = getBufferSize(args[0]);
if (bufferSize.intvalue < 0 || sizeToken->getKnownIntValue() < bufferSize.intvalue)
continue;
const Token *srcValue = args[1]->getValueTokenMaxStrLength();
if (srcValue && Token::getStrLength(srcValue) < sizeToken->getKnownIntValue())
continue;
// Is the buffer zero terminated after the call?
bool isZeroTerminated = false;
for (const Token *tok2 = tok->next()->link(); tok2 != scope->bodyEnd; tok2 = tok2->next()) {
if (!Token::simpleMatch(tok2, "] ="))
continue;
const Token *rhs = tok2->next()->astOperand2();
if (!rhs || !rhs->hasKnownIntValue() || rhs->getKnownIntValue() != 0)
continue;
if (isSameExpression(mTokenizer->isCPP(), false, args[0], tok2->link()->astOperand1(), mSettings->library, false, false))
isZeroTerminated = true;
}
if (isZeroTerminated)
continue;
// TODO: Locate unsafe string usage..
terminateStrncpyError(tok, args[0]->expressionString());
}
}
}
void CheckBufferOverrun::terminateStrncpyError(const Token *tok, const std::string &varname)
{
const std::string shortMessage = "The buffer '$symbol' may not be null-terminated after the call to strncpy().";
reportError(tok, Severity::warning, "terminateStrncpy",
"$symbol:" + varname + '\n' +
shortMessage + '\n' +
shortMessage + ' ' +
"If the source string's size fits or exceeds the given size, strncpy() does not add a "
"zero at the end of the buffer. This causes bugs later in the code if the code "
"assumes buffer is null-terminated.", CWE170, Certainty::inconclusive);
}
//---------------------------------------------------------------------------
void CheckBufferOverrun::argumentSize()
{
// Check '%type% x[10]' arguments
if (!mSettings->severity.isEnabled(Severity::warning))
return;
const SymbolDatabase *symbolDatabase = mTokenizer->getSymbolDatabase();
for (const Scope * const scope : symbolDatabase->functionScopes) {
for (const Token *tok = scope->bodyStart; tok != scope->bodyEnd; tok = tok->next()) {
if (!tok->function() || !Token::Match(tok, "%name% ("))
continue;
// If argument is '%type% a[num]' then check bounds against num
const Function *callfunc = tok->function();
const std::vector<const Token *> callargs = getArguments(tok);
for (nonneg int paramIndex = 0; paramIndex < callargs.size() && paramIndex < callfunc->argCount(); ++paramIndex) {
const Variable* const argument = callfunc->getArgumentVar(paramIndex);
if (!argument || !argument->nameToken() || !argument->isArray())
continue;
if (!argument->valueType() || !callargs[paramIndex]->valueType())
continue;
if (argument->valueType()->type != callargs[paramIndex]->valueType()->type)
continue;
const Token * calldata = callargs[paramIndex];
while (Token::Match(calldata, "::|."))
calldata = calldata->astOperand2();
if (!calldata->variable() || !calldata->variable()->isArray())
continue;
if (calldata->variable()->dimensions().size() != argument->dimensions().size())
continue;
bool err = false;
for (int d = 0; d < argument->dimensions().size(); ++d) {
const auto& dim1 = calldata->variable()->dimensions()[d];
const auto& dim2 = argument->dimensions()[d];
if (!dim1.known || !dim2.known)
break;
if (dim1.num < dim2.num)
err = true;
}
if (err)
argumentSizeError(tok, tok->str(), paramIndex, callargs[paramIndex]->expressionString(), calldata->variable(), argument);
}
}
}
}
void CheckBufferOverrun::argumentSizeError(const Token *tok, const std::string &functionName, nonneg int paramIndex, const std::string ¶mExpression, const Variable *paramVar, const Variable *functionArg)
{
const std::string strParamNum = std::to_string(paramIndex + 1) + getOrdinalText(paramIndex + 1);
ErrorPath errorPath;
errorPath.emplace_back(tok, "Function '" + functionName + "' is called");
if (functionArg)
errorPath.emplace_back(functionArg->nameToken(), "Declaration of " + strParamNum + " function argument.");
if (paramVar)
errorPath.emplace_back(paramVar->nameToken(), "Passing buffer '" + paramVar->name() + "' to function that is declared here");
errorPath.emplace_back(tok, "");
reportError(errorPath, Severity::warning, "argumentSize",
"$symbol:" + functionName + '\n' +
"Buffer '" + paramExpression + "' is too small, the function '" + functionName + "' expects a bigger buffer in " + strParamNum + " argument", CWE_ARGUMENT_SIZE, Certainty::normal);
}
//---------------------------------------------------------------------------
// CTU..
//---------------------------------------------------------------------------
std::string CheckBufferOverrun::MyFileInfo::toString() const
{
std::string xml;
if (!unsafeArrayIndex.empty())
xml = " <array-index>\n" + CTU::toString(unsafeArrayIndex) + " </array-index>\n";
if (!unsafePointerArith.empty())
xml += " <pointer-arith>\n" + CTU::toString(unsafePointerArith) + " </pointer-arith>\n";
return xml;
}
bool CheckBufferOverrun::isCtuUnsafeBufferUsage(const Check *check, const Token *argtok, MathLib::bigint *offset, int type)
{
const CheckBufferOverrun *c = dynamic_cast<const CheckBufferOverrun *>(check);
if (!c)
return false;
if (!argtok->valueType() || argtok->valueType()->typeSize(*c->mSettings) == 0)
return false;
const Token *indexTok = nullptr;
if (type == 1 && Token::Match(argtok, "%name% [") && argtok->astParent() == argtok->next() && !Token::simpleMatch(argtok->linkAt(1), "] ["))
indexTok = argtok->next()->astOperand2();
else if (type == 2 && Token::simpleMatch(argtok->astParent(), "+"))
indexTok = (argtok == argtok->astParent()->astOperand1()) ?
argtok->astParent()->astOperand2() :
argtok->astParent()->astOperand1();
if (!indexTok)
return false;
if (!indexTok->hasKnownIntValue())
return false;
if (!offset)
return false;
*offset = indexTok->getKnownIntValue() * argtok->valueType()->typeSize(*c->mSettings);
return true;
}
bool CheckBufferOverrun::isCtuUnsafeArrayIndex(const Check *check, const Token *argtok, MathLib::bigint *offset)
{
return CheckBufferOverrun::isCtuUnsafeBufferUsage(check, argtok, offset, 1);
}
bool CheckBufferOverrun::isCtuUnsafePointerArith(const Check *check, const Token *argtok, MathLib::bigint *offset)
{
return CheckBufferOverrun::isCtuUnsafeBufferUsage(check, argtok, offset, 2);
}
/** @brief Parse current TU and extract file info */
Check::FileInfo *CheckBufferOverrun::getFileInfo(const Tokenizer *tokenizer, const Settings *settings) const
{
CheckBufferOverrun checkBufferOverrun(tokenizer, settings, nullptr);
MyFileInfo *fileInfo = new MyFileInfo;
fileInfo->unsafeArrayIndex = CTU::getUnsafeUsage(tokenizer, settings, &checkBufferOverrun, isCtuUnsafeArrayIndex);
fileInfo->unsafePointerArith = CTU::getUnsafeUsage(tokenizer, settings, &checkBufferOverrun, isCtuUnsafePointerArith);
if (fileInfo->unsafeArrayIndex.empty() && fileInfo->unsafePointerArith.empty()) {
delete fileInfo;
return nullptr;
}
return fileInfo;
}
Check::FileInfo * CheckBufferOverrun::loadFileInfoFromXml(const tinyxml2::XMLElement *xmlElement) const
{
const std::string arrayIndex("array-index");
const std::string pointerArith("pointer-arith");
MyFileInfo *fileInfo = new MyFileInfo;
for (const tinyxml2::XMLElement *e = xmlElement->FirstChildElement(); e; e = e->NextSiblingElement()) {
if (e->Name() == arrayIndex)
fileInfo->unsafeArrayIndex = CTU::loadUnsafeUsageListFromXml(e);
else if (e->Name() == pointerArith)
fileInfo->unsafePointerArith = CTU::loadUnsafeUsageListFromXml(e);
}
if (fileInfo->unsafeArrayIndex.empty() && fileInfo->unsafePointerArith.empty()) {
delete fileInfo;
return nullptr;
}
return fileInfo;
}
/** @brief Analyse all file infos for all TU */
bool CheckBufferOverrun::analyseWholeProgram(const CTU::FileInfo *ctu, const std::list<Check::FileInfo*> &fileInfo, const Settings& settings, ErrorLogger &errorLogger)
{
if (!ctu)
return false;
bool foundErrors = false;
(void)settings; // This argument is unused
const std::map<std::string, std::list<const CTU::FileInfo::CallBase *>> callsMap = ctu->getCallsMap();
for (Check::FileInfo *fi1 : fileInfo) {
const MyFileInfo *fi = dynamic_cast<MyFileInfo*>(fi1);
if (!fi)
continue;
for (const CTU::FileInfo::UnsafeUsage &unsafeUsage : fi->unsafeArrayIndex)
foundErrors |= analyseWholeProgram1(callsMap, unsafeUsage, 1, errorLogger);
for (const CTU::FileInfo::UnsafeUsage &unsafeUsage : fi->unsafePointerArith)
foundErrors |= analyseWholeProgram1(callsMap, unsafeUsage, 2, errorLogger);
}
return foundErrors;
}
bool CheckBufferOverrun::analyseWholeProgram1(const std::map<std::string, std::list<const CTU::FileInfo::CallBase *>> &callsMap, const CTU::FileInfo::UnsafeUsage &unsafeUsage, int type, ErrorLogger &errorLogger)
{
const CTU::FileInfo::FunctionCall *functionCall = nullptr;
const std::list<ErrorMessage::FileLocation> &locationList =
CTU::FileInfo::getErrorPath(CTU::FileInfo::InvalidValueType::bufferOverflow,
unsafeUsage,
callsMap,
"Using argument ARG",
&functionCall,
false);
if (locationList.empty())
return false;
const char *errorId = nullptr;
std::string errmsg;
CWE cwe(0);
if (type == 1) {
errorId = "ctuArrayIndex";
if (unsafeUsage.value > 0)
errmsg = "Array index out of bounds; '" + unsafeUsage.myArgumentName + "' buffer size is " + MathLib::toString(functionCall->callArgValue) + " and it is accessed at offset " + MathLib::toString(unsafeUsage.value) + ".";
else
errmsg = "Array index out of bounds; buffer '" + unsafeUsage.myArgumentName + "' is accessed at offset " + MathLib::toString(unsafeUsage.value) + ".";
cwe = (unsafeUsage.value > 0) ? CWE_BUFFER_OVERRUN : CWE_BUFFER_UNDERRUN;
} else {
errorId = "ctuPointerArith";
errmsg = "Pointer arithmetic overflow; '" + unsafeUsage.myArgumentName + "' buffer size is " + MathLib::toString(functionCall->callArgValue);
cwe = CWE_POINTER_ARITHMETIC_OVERFLOW;
}
const ErrorMessage errorMessage(locationList,
emptyString,
Severity::error,
errmsg,
errorId,
cwe, Certainty::normal);
errorLogger.reportErr(errorMessage);
return true;
}
void CheckBufferOverrun::objectIndex()
{
const SymbolDatabase *symbolDatabase = mTokenizer->getSymbolDatabase();
for (const Scope *functionScope : symbolDatabase->functionScopes) {
for (const Token *tok = functionScope->bodyStart; tok != functionScope->bodyEnd; tok = tok->next()) {
if (!Token::simpleMatch(tok, "["))
continue;
const Token *obj = tok->astOperand1();
const Token *idx = tok->astOperand2();
if (!idx || !obj)
continue;
if (idx->hasKnownIntValue()) {
if (idx->getKnownIntValue() == 0)
continue;
}
if (idx->hasKnownIntValue() && idx->getKnownIntValue() == 0)
continue;
std::vector<ValueFlow::Value> values = getLifetimeObjValues(obj, false, -1);
for (const ValueFlow::Value& v:values) {