Mercurial > hg > CbC > CbC_gcc
comparison gcc/doc/portability.texi @ 0:a06113de4d67
first commit
| author | kent <kent@cr.ie.u-ryukyu.ac.jp> |
|---|---|
| date | Fri, 17 Jul 2009 14:47:48 +0900 |
| parents | |
| children | 04ced10e8804 |
comparison
equal
deleted
inserted
replaced
| -1:000000000000 | 0:a06113de4d67 |
|---|---|
| 1 @c Copyright (C) 1988, 1989, 1992, 1993, 1994, 1995, 1996, 1997, 1998, | |
| 2 @c 1999, 2000, 2001, 2002, 2004 Free Software Foundation, Inc. | |
| 3 @c This is part of the GCC manual. | |
| 4 @c For copying conditions, see the file gcc.texi. | |
| 5 | |
| 6 @node Portability | |
| 7 @chapter GCC and Portability | |
| 8 @cindex portability | |
| 9 @cindex GCC and portability | |
| 10 | |
| 11 GCC itself aims to be portable to any machine where @code{int} is at least | |
| 12 a 32-bit type. It aims to target machines with a flat (non-segmented) byte | |
| 13 addressed data address space (the code address space can be separate). | |
| 14 Target ABIs may have 8, 16, 32 or 64-bit @code{int} type. @code{char} | |
| 15 can be wider than 8 bits. | |
| 16 | |
| 17 GCC gets most of the information about the target machine from a machine | |
| 18 description which gives an algebraic formula for each of the machine's | |
| 19 instructions. This is a very clean way to describe the target. But when | |
| 20 the compiler needs information that is difficult to express in this | |
| 21 fashion, ad-hoc parameters have been defined for machine descriptions. | |
| 22 The purpose of portability is to reduce the total work needed on the | |
| 23 compiler; it was not of interest for its own sake. | |
| 24 | |
| 25 @cindex endianness | |
| 26 @cindex autoincrement addressing, availability | |
| 27 @findex abort | |
| 28 GCC does not contain machine dependent code, but it does contain code | |
| 29 that depends on machine parameters such as endianness (whether the most | |
| 30 significant byte has the highest or lowest address of the bytes in a word) | |
| 31 and the availability of autoincrement addressing. In the RTL-generation | |
| 32 pass, it is often necessary to have multiple strategies for generating code | |
| 33 for a particular kind of syntax tree, strategies that are usable for different | |
| 34 combinations of parameters. Often, not all possible cases have been | |
| 35 addressed, but only the common ones or only the ones that have been | |
| 36 encountered. As a result, a new target may require additional | |
| 37 strategies. You will know | |
| 38 if this happens because the compiler will call @code{abort}. Fortunately, | |
| 39 the new strategies can be added in a machine-independent fashion, and will | |
| 40 affect only the target machines that need them. |