| Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 1 | Copyright 2004 Linus Torvalds | 
|  | 2 | Copyright 2004 Pavel Machek <pavel@suse.cz> | 
| Bob Copeland | e833195 | 2006-06-23 02:06:09 -0700 | [diff] [blame] | 3 | Copyright 2006 Bob Copeland <me@bobcopeland.com> | 
| Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 4 |  | 
|  | 5 | Using sparse for typechecking | 
|  | 6 | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ | 
|  | 7 |  | 
|  | 8 | "__bitwise" is a type attribute, so you have to do something like this: | 
|  | 9 |  | 
|  | 10 | typedef int __bitwise pm_request_t; | 
|  | 11 |  | 
|  | 12 | enum pm_request { | 
|  | 13 | PM_SUSPEND = (__force pm_request_t) 1, | 
|  | 14 | PM_RESUME = (__force pm_request_t) 2 | 
|  | 15 | }; | 
|  | 16 |  | 
|  | 17 | which makes PM_SUSPEND and PM_RESUME "bitwise" integers (the "__force" is | 
|  | 18 | there because sparse will complain about casting to/from a bitwise type, | 
|  | 19 | but in this case we really _do_ want to force the conversion). And because | 
|  | 20 | the enum values are all the same type, now "enum pm_request" will be that | 
|  | 21 | type too. | 
|  | 22 |  | 
|  | 23 | And with gcc, all the __bitwise/__force stuff goes away, and it all ends | 
|  | 24 | up looking just like integers to gcc. | 
|  | 25 |  | 
|  | 26 | Quite frankly, you don't need the enum there. The above all really just | 
|  | 27 | boils down to one special "int __bitwise" type. | 
|  | 28 |  | 
|  | 29 | So the simpler way is to just do | 
|  | 30 |  | 
|  | 31 | typedef int __bitwise pm_request_t; | 
|  | 32 |  | 
|  | 33 | #define PM_SUSPEND ((__force pm_request_t) 1) | 
|  | 34 | #define PM_RESUME ((__force pm_request_t) 2) | 
|  | 35 |  | 
|  | 36 | and you now have all the infrastructure needed for strict typechecking. | 
|  | 37 |  | 
|  | 38 | One small note: the constant integer "0" is special. You can use a | 
|  | 39 | constant zero as a bitwise integer type without sparse ever complaining. | 
|  | 40 | This is because "bitwise" (as the name implies) was designed for making | 
|  | 41 | sure that bitwise types don't get mixed up (little-endian vs big-endian | 
|  | 42 | vs cpu-endian vs whatever), and there the constant "0" really _is_ | 
|  | 43 | special. | 
|  | 44 |  | 
| Sam Ravnborg | 20375bf | 2009-04-10 13:18:08 +0200 | [diff] [blame] | 45 | __bitwise__ - to be used for relatively compact stuff (gfp_t, etc.) that | 
|  | 46 | is mostly warning-free and is supposed to stay that way.  Warnings will | 
|  | 47 | be generated without __CHECK_ENDIAN__. | 
|  | 48 |  | 
|  | 49 | __bitwise - noisy stuff; in particular, __le*/__be* are that.  We really | 
|  | 50 | don't want to drown in noise unless we'd explicitly asked for it. | 
|  | 51 |  | 
|  | 52 |  | 
| Bob Copeland | e833195 | 2006-06-23 02:06:09 -0700 | [diff] [blame] | 53 | Getting sparse | 
|  | 54 | ~~~~~~~~~~~~~~ | 
| Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 55 |  | 
| Dave Jones | a55028f | 2007-03-08 19:45:26 -0500 | [diff] [blame] | 56 | You can get latest released versions from the Sparse homepage at | 
|  | 57 | http://www.kernel.org/pub/linux/kernel/people/josh/sparse/ | 
| Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 58 |  | 
| Dave Jones | a55028f | 2007-03-08 19:45:26 -0500 | [diff] [blame] | 59 | Alternatively, you can get snapshots of the latest development version | 
|  | 60 | of sparse using git to clone.. | 
| Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 61 |  | 
| Dave Jones | a55028f | 2007-03-08 19:45:26 -0500 | [diff] [blame] | 62 | git://git.kernel.org/pub/scm/linux/kernel/git/josh/sparse.git | 
|  | 63 |  | 
|  | 64 | DaveJ has hourly generated tarballs of the git tree available at.. | 
| Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 65 |  | 
| Bob Copeland | e833195 | 2006-06-23 02:06:09 -0700 | [diff] [blame] | 66 | http://www.codemonkey.org.uk/projects/git-snapshots/sparse/ | 
| Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 67 |  | 
|  | 68 |  | 
|  | 69 | Once you have it, just do | 
|  | 70 |  | 
|  | 71 | make | 
|  | 72 | make install | 
|  | 73 |  | 
| Bob Copeland | e833195 | 2006-06-23 02:06:09 -0700 | [diff] [blame] | 74 | as a regular user, and it will install sparse in your ~/bin directory. | 
|  | 75 |  | 
|  | 76 | Using sparse | 
|  | 77 | ~~~~~~~~~~~~ | 
|  | 78 |  | 
|  | 79 | Do a kernel make with "make C=1" to run sparse on all the C files that get | 
|  | 80 | recompiled, or use "make C=2" to run sparse on the files whether they need to | 
|  | 81 | be recompiled or not.  The latter is a fast way to check the whole tree if you | 
|  | 82 | have already built it. | 
|  | 83 |  | 
| Geert Uytterhoeven | a887a07 | 2008-06-20 15:45:12 +0200 | [diff] [blame] | 84 | The optional make variable CF can be used to pass arguments to sparse.  The | 
|  | 85 | build system passes -Wbitwise to sparse automatically.  To perform endianness | 
|  | 86 | checks, you may define __CHECK_ENDIAN__: | 
| Bob Copeland | e833195 | 2006-06-23 02:06:09 -0700 | [diff] [blame] | 87 |  | 
| Geert Uytterhoeven | a887a07 | 2008-06-20 15:45:12 +0200 | [diff] [blame] | 88 | make C=2 CF="-D__CHECK_ENDIAN__" | 
| Bob Copeland | e833195 | 2006-06-23 02:06:09 -0700 | [diff] [blame] | 89 |  | 
|  | 90 | These checks are disabled by default as they generate a host of warnings. |