projects / thirdparty / gcc.git / blame
| Commit | Line | Data |
|---|---|---|
| c63539ff ML |
1 | .. |
| 2 | Copyright 1988-2022 Free Software Foundation, Inc. | |
| 3 | This is part of the GCC manual. | |
| 4 | For copying conditions, see the copyright.rst file. | |
| 5 | ||
| 6 | .. index:: bug criteria | |
| 7 | ||
| 8 | .. _bug-criteria: | |
| 9 | ||
| 10 | Have You Found a Bug? | |
| 11 | ********************* | |
| 12 | ||
| 13 | If you are not sure whether you have found a bug, here are some guidelines: | |
| 14 | ||
| 15 | .. index:: fatal signal, core dump | |
| 16 | ||
| 17 | * If the compiler gets a fatal signal, for any input whatever, that is a | |
| 18 | compiler bug. Reliable compilers never crash. | |
| 19 | ||
| 20 | .. index:: invalid assembly code, assembly code, invalid | |
| 21 | ||
| 22 | * If the compiler produces invalid assembly code, for any input whatever | |
| 23 | (except an ``asm`` statement), that is a compiler bug, unless the | |
| 24 | compiler reports errors (not just warnings) which would ordinarily | |
| 25 | prevent the assembler from being run. | |
| 26 | ||
| 27 | .. index:: undefined behavior, undefined function value, increment operators | |
| 28 | ||
| 29 | * If the compiler produces valid assembly code that does not correctly | |
| 30 | execute the input source code, that is a compiler bug. | |
| 31 | ||
| 32 | However, you must double-check to make sure, because you may have a | |
| 33 | program whose behavior is undefined, which happened by chance to give | |
| 34 | the desired results with another C or C++ compiler. | |
| 35 | ||
| 36 | For example, in many nonoptimizing compilers, you can write :samp:`x;` | |
| 37 | at the end of a function instead of :samp:`return x;`, with the same | |
| 38 | results. But the value of the function is undefined if ``return`` | |
| 39 | is omitted; it is not a bug when GCC produces different results. | |
| 40 | ||
| 41 | Problems often result from expressions with two increment operators, | |
| 42 | as in ``f (*p++, *p++)``. Your previous compiler might have | |
| 43 | interpreted that expression the way you intended; GCC might | |
| 44 | interpret it another way. Neither compiler is wrong. The bug is | |
| 45 | in your code. | |
| 46 | ||
| 47 | After you have localized the error to a single source line, it should | |
| 48 | be easy to check for these things. If your program is correct and | |
| 49 | well defined, you have found a compiler bug. | |
| 50 | ||
| 51 | * If the compiler produces an error message for valid input, that is a | |
| 52 | compiler bug. | |
| 53 | ||
| 54 | .. index:: invalid input | |
| 55 | ||
| 56 | * If the compiler does not produce an error message for invalid input, | |
| 57 | that is a compiler bug. However, you should note that your idea of | |
| 58 | 'invalid input' might be someone else's idea of 'an extension' or | |
| 59 | 'support for traditional practice'. | |
| 60 | ||
| 61 | * If you are an experienced user of one of the languages GCC supports, your | |
| 3ed1b4ce | 62 | suggestions for improvement of GCC are welcome in any case. |