Commit graph

1605 commits

Author SHA1 Message Date
Bill Currie
e8c17c68e6 Fix a segfault when the switch expression errors. 2012-05-04 18:13:11 +09:00
Bill Currie
5df25133b3 Implement code movement for unconditional jumps.
That is, when the destination of the jump is reachable via only the jump.
2012-05-04 18:00:05 +09:00
Bill Currie
33bfac0508 Process all basic blocks for jump threading.
I'd copied the for loop from the dead block removal code, but jump
threading doesn't need to look at the following block...
2012-05-04 14:10:30 +09:00
Bill Currie
f169a7732d Factor out label unuse.
I expect to need this more often in the future.
2012-05-04 14:01:39 +09:00
Bill Currie
75aa28cfac Mark all subsequent blocks as reachable after if/goto merge.
If an if/goto merge is done in the first dead block pass, no blocks after
the merge have their rechable flag set because they've never been tested.
2012-05-04 10:58:18 +09:00
Bill Currie
8ddd58f951 Fix the dropping of the block after the merged if/goto.
The naive implementation of the if/goto merging was letting the old target
of the if get dropped because the block would lose its label and thus be
judged unreachable because the preceeding goto block was still in the list.
Instead, when the if/goto are "merged", mark the goto block as unreachable,
the following block as reachable, and break out of the analysis loop to
force the removal of the goto block. Since the dead block removal function
loops until no action is taken, all other dead blocks will be removed.
2012-05-04 10:07:55 +09:00
Bill Currie
e866619de6 Output basic block flow diagrams to files.
The output can be controlled via --block-dot (not yet documented). The
files a named <sourcefile>.<function>.<stage>.dot. Currently, stage will be
one of "initial" (after expression to statement conversion), "thread"
(after jump threading), "dead" (after dead block removal), "final" (final
state before actual code emission).
2012-05-04 09:45:51 +09:00
Bill Currie
3da44ace52 Merge if and goto blocks when if only skips over the goto. 2012-05-03 22:21:32 +09:00
Bill Currie
6afdfb5fac Unuse label expressions that are no longer necessary. 2012-05-03 22:20:00 +09:00
Bill Currie
402a578bf8 Add some more helper functions.
Things were getting messy with the strcmps.
2012-05-03 22:17:23 +09:00
Bill Currie
88bed3644e Rename some helper functions. 2012-05-03 22:11:52 +09:00
Bill Currie
159f1bafea Add a little more info to statement flow diagrams.
The relative block number and the label user count are printed now.
2012-05-03 22:09:49 +09:00
Bill Currie
6900907129 Remove dead labels when jump threading.
This lets the dead block removal do a better job.
2012-05-03 19:32:44 +09:00
Bill Currie
ab73a267cd Do not unconditionally remove labels from blocks.
Labels can be shared between multiple flow-control instructions, so use the
label's used counter to determine when to remove the label. This was
causing problems with the jump threading.
2012-05-03 19:22:57 +09:00
Bill Currie
c79620f04c Make statement blocks more identifiable. 2012-05-03 19:10:44 +09:00
Bill Currie
43b5edf46b Implement jump threading.
First real optimization :)
2012-05-03 17:42:58 +09:00
Bill Currie
78a9ba2557 Make expression alias chains not-a-bug.
The common cause seems to be casting a cast (very common, and I'm not sure
just realiasing the expression would be right). It does't cause any harm
(particularly, it doesn't trigger alias def chains), so I won't worry about
it.
2012-05-03 13:57:31 +09:00
Bill Currie
006882407d Fix the source of alias def chains.
The actual bug might still be elsewhere, but at least now I know the alias
chains were coming from accessing .return and .param_N, which are unions
(not directly usable by the progs engine). Emitting a reference to a union
(or struct) would create an alias def, but an alias expression was created
in the expression tree to simplify return/param access. The double layer
(sometimes 3 or 4) alias isn't really neaded, so rather than layering the
aliases, just re-alias the alaised def.
2012-05-03 13:28:16 +09:00
Bill Currie
ec98e3e206 Mark alias def chains as a bug. 2012-05-03 13:27:30 +09:00
Bill Currie
5f676c367f Add a non-error diagnostic that can't be silenced.
It is inteded for flagging buggy conditions in the compiler, particularly
after having fixed the original bug (in case something comes back from the
dead).
2012-05-03 13:24:24 +09:00
Bill Currie
3306039da2 Print the destination type for alias expression bubbles. 2012-05-03 11:55:19 +09:00
Bill Currie
e267e0a664 Allow control of progdefs.h CRC writing.
CRC writing defaults to off for non-v6 progs, but on for v6 progs. The
--progdefs option forces CRC writing.
2012-05-03 00:22:24 +09:00
Bill Currie
44fcd76475 Add some missing command line option docs. 2012-05-03 00:03:13 +09:00
Bill Currie
15de69d868 Use dstring for progdefs.h writing.
And finally get qfcc into the thing :P
2012-05-02 23:39:49 +09:00
Bill Currie
04b9b3a11d Fix progdefs.h writing. 2012-05-02 22:44:45 +09:00
Bill Currie
ce278b2fbf Ensure .zero does not get a vector field for v6 progs.
v6 progs expects .zero to be only 1 word. The code actually tried to keep
vector out of .zero, but it seems I'd rearranged the structure defintion
without updating the code that kills the vector field. Problem spotted by
divVerent.
2012-05-02 22:29:17 +09:00
Bill Currie
e06ee34287 Allow quat * vec in ruamoko. 2012-04-26 12:00:27 +09:00
Bill Currie
bc1b483525 Nuke the rcsid stuff.
It's pretty useless in git.
2012-04-22 10:56:32 +09:00
Bill Currie
eefa89e72e Remove the AM_CONDITIONALs from tools.
Now that the tools directories aren't entered when those tools aren't being
built, there's no reason to use AM_CONDITIONAL in there.
2012-02-10 02:50:42 +09:00
Bill Currie
c8163fc0de Add an option to prevent the use of default paths.
Despair has things locked down such that running qfcc during a build fails
due to lack of read access to /usr/local/lib. This is actually a good
thing as accidentally hitting old includes/libs (when a file gets deleted
in the tree) hides bugs. Thus, --no-default-paths to turn off default
search paths.
2012-01-04 13:53:04 +09:00
Jeff Teunissen
6ead583195 Ruamoko: Implement infinity.
The special token __INFINITY__, like __FILE__ and friends, will expand to
a floating-point expression containing a value the C compiler considers
infinite. Obviously, this assumes that the system has relatively modern
float hardware -- but if it doesn't, having Ruamoko be able to represent
float infinity is the least of your problems. :)
2011-12-14 12:20:10 -05:00
Bill Currie
ea02f3cd44 Fix a missed byteswap when loading qfo files. 2011-11-15 17:36:55 +09:00
Bill Currie
a71acc9ae5 Move the essential init code into Sys_Init(). 2011-09-11 14:56:47 +09:00
Bill Currie
e42c5a4272 More tests and fix vector/quaternion global dumps. 2011-08-11 14:58:13 +09:00
Bill Currie
c6e0e094d8 Fix an unterminated array.
That was sneaky. It took building qfcc on an arm to trigger that bug.
2011-07-30 15:01:16 +09:00
Bill Currie
d910c14935 Clean up some doxygen warnings. 2011-07-10 19:12:07 +09:00
Bill Currie
617e70c8e3 Clean up a bunch of bison unused warnings. 2011-06-21 19:52:47 +09:00
Bill Currie
0f7390dd60 Clean up all the "set but not used" warnings.
gcc on my system is failing to treat this specific warning as an error :/
2011-06-19 10:48:02 +09:00
Bill Currie
c7612dcd99 Make statement dumps a little more informative. 2011-04-10 09:08:46 +09:00
Bill Currie
fd6e341e06 Clean up cast_expr() a little bit. 2011-04-10 00:29:27 +09:00
Bill Currie
57ed603658 Extract values from constant defs. 2011-04-09 12:42:41 +09:00
Bill Currie
cc714864a8 Support converting to/from unsigned values. 2011-04-09 11:32:33 +09:00
Bill Currie
55cc0f9206 Bring back the unsigned type (PROGS version bump)
This is only low-level support (the unsigned keyword still does not work),
but sufficient to make switch statements using jump tables work.
2011-04-09 10:07:47 +09:00
Bill Currie
b522853765 Ensure the target of def_op relocs gets updated.
These occur only when switch statements use jump tables, thus why the
problem went unnoticed :/
2011-04-08 17:09:40 +09:00
Bill Currie
237f11c472 Fix cast expressions.
Casting between ints and floats needs special treatment to get the
conversion operator, but other casts need to be aliases.
2011-04-08 13:55:26 +09:00
Bill Currie
c4a398850a Fix defs as constants for vectors and quaternions. 2011-04-04 21:31:59 +09:00
Bill Currie
38b25e0c38 Do not double-allocate space for immediates.
Caused by changed semantics in def creation.
2011-04-03 13:11:50 +09:00
Bill Currie
92ba110a87 Ensure a field def is emitted for @this. 2011-04-03 13:03:48 +09:00
Bill Currie
7f67e6eb17 Refactor the def handling code a little. 2011-04-03 13:03:22 +09:00
Bill Currie
6d0b13dbf0 Resolve types whenever possible. 2011-04-03 13:02:04 +09:00
Bill Currie
2c9c1bd968 Ensure defs for @self and @this are generated. 2011-04-03 13:00:48 +09:00
Bill Currie
2de7733b6a Fix initialized variable type handling.
Same problem as initialized constants :P
2011-03-30 21:21:38 +09:00
Bill Currie
8e18c76bde Make initialized variables defs rather than immediate constants. 2011-03-30 19:58:09 +09:00
Bill Currie
7dbd2ec172 Allow field aliasing.
Initializing a field variable to another field will set the new field to
point to the same location. No type checking is done.

eg:
.SEL thinkMethod = think;
2011-03-30 09:33:47 +09:00
Bill Currie
550fff5467 Allow assignments between any field types. 2011-03-30 08:14:17 +09:00
Bill Currie
75c3653be4 Implement some missed relocs. 2011-03-30 08:13:13 +09:00
Bill Currie
1911cc1472 Mingw build fixes. 2011-03-27 08:26:42 +09:00
Bill Currie
896791b209 Correct the spelling of "int" and clean up the mess. 2011-03-25 16:53:04 +09:00
Bill Currie
1617f0765e Support the default type without segfaulting. 2011-03-25 09:59:48 +09:00
Bill Currie
a7cfe3b215 Preserve the current class across protocol defs.
It seems that protocol defs are allowed in class implementations, so don't
let the protocol def kill the current class for the class implementation.
2011-03-24 11:55:17 +09:00
Bill Currie
71af121665 Detect missing @end tags.
Give a warning when @end is forgotten in class implementations.
2011-03-24 11:44:22 +09:00
Bill Currie
8d3508cf20 Allocate space for temp defs using size rather than type.
Statement operands throw away the high level type information, so store
type size in the operand and use this size for allocating space for temps
rather than using the low-level type.
2011-03-23 21:32:14 +09:00
Bill Currie
9b62c8c126 Rename ty_type_e to ty_meta_e and type_t.ty to type_t.meta 2011-03-22 16:06:47 +09:00
Bill Currie
d4be914019 Fix initialized vector variables.
The whole initialized variable system is wonky, but it will do for now.
2011-03-22 15:10:06 +09:00
Bill Currie
6a49861b49 Fix the bogus source file for "redefined" functions. 2011-03-22 13:45:44 +09:00
Bill Currie
7dbd5f656c Fix traditional vector component access. 2011-03-22 13:05:42 +09:00
Bill Currie
f1d9570dc8 Fix constant value type handling.
Rename immediate.[ch] to value.[ch] and clean up the mess.

Add convert_value() to convert the type of a value (only scalar types).
2011-03-22 12:24:39 +09:00
Bill Currie
1219df0872 Fix field immediates.
Field immediates need to be treated the same way as pointer immediates when
merging values.
2011-03-21 18:11:46 +09:00
Bill Currie
46093c0937 Use the correct meta type for spacial types.
Although vector and quaternion types have symbol tables, they are not
really structs, so set the meta type to "none", allowing the types to be
encoded correctly.
2011-03-21 17:26:49 +09:00
Bill Currie
cefe7e2e94 Use the correct source of defs for field defs. 2011-03-21 17:07:40 +09:00
Bill Currie
0c0f4360da Make field defs "nosave".
The engine treats fielddefs with DEF_SAVEGLOBAL set as bogus.
2011-03-21 17:06:45 +09:00
Bill Currie
f5ce9e2004 Fix a merge induced error in overloaded functions. 2011-03-20 19:04:18 +09:00
Bill Currie
1ada2a96cf Merge branch 'master' into qfcc-codegen
Conflicts:
	tools/qfcc/source/expr.c
	tools/qfcc/source/function.c
	tools/qwaq/test.r
2011-03-20 15:58:31 +09:00
Bill Currie
a919e5f619 Make class_check_ivars() use class_add_ivars() to add the ivars.
This avoids implementation ivar blocks trashing the ivar offsets in
derived classes.
2011-03-20 13:35:59 +09:00
Bill Currie
8c8a6a4d63 The function return type is the address of the type descriptor... 2011-03-17 20:31:20 +09:00
Bill Currie
01144d23cf Add a new gcc warning option and fix the two bugs it found. 2011-03-17 19:58:56 +09:00
Bill Currie
5aa0b34570 Add the cast statement to the statement block.
*sob*
2011-03-10 20:43:53 +09:00
Bill Currie
f663f587fe Fix the missing local defs for builtin functions. 2011-03-10 20:16:35 +09:00
Bill Currie
899af3119b Use the correct type when generating a move expression. 2011-03-10 19:30:21 +09:00
Bill Currie
2129eaaf20 Dereference moves need movepi rather than movei. 2011-03-10 19:29:23 +09:00
Bill Currie
cbbbf87e20 Resurrect the stats output. 2011-03-09 15:51:17 +09:00
Bill Currie
8368f57fbf Close a potential buffer overflow. 2011-03-09 11:09:46 +09:00
Bill Currie
4d90eba2ad Remove some unused globals. 2011-03-09 11:08:30 +09:00
Bill Currie
2464a89d37 Preserve relocs across def churning.
Freeing then re-allocating a def to change its storage from external is
really not the right way to do it, but for now this fixes the loss of the
relocs. With this, the menus seem to work :)
2011-03-09 10:52:29 +09:00
Bill Currie
39278ba8cc Explicitly select between direct and indirect moves. 2011-03-09 10:30:57 +09:00
Bill Currie
b95cdc33d9 Set instace_size to the correct size. 2011-03-08 22:45:34 +09:00
Bill Currie
55eed1ae07 Make module dumping a little more informative. 2011-03-08 22:43:53 +09:00
Bill Currie
7c770b75c5 Do not creat field defs for local storage.
This prevents field params and local vars from messing up the real field
defs.
2011-03-08 19:56:02 +09:00
Bill Currie
f66e20db17 Set the dereferenced type in the field value. 2011-03-08 19:29:36 +09:00
Bill Currie
7198be5726 Treat entities a bit like structures in field_expr ().
Look in the entity field symbol table for the field before looking in the
normal symbol table. This allows entity fields to be accessed even when
the current scope has symbol of the same name. However, checking the
normal symbol table where there is no such field allows for field
variables when I get around to implementing them.
2011-03-08 19:28:11 +09:00
Bill Currie
505bde308e Fix the handling of @system defs.
When treating a def as external, actually say it is external.
2011-03-07 20:53:27 +09:00
Bill Currie
8dbf8a97ea Be more informative with ext/def internal errors. 2011-03-07 20:52:31 +09:00
Bill Currie
6312ccba81 Fix storage class blocks.
@extern { defs }; etc
2011-03-07 20:51:51 +09:00
Bill Currie
4db8ebb6af Support qcc style local function variables. 2011-03-07 20:04:05 +09:00
Bill Currie
853cc53d0f FIx the handling of storage class. 2011-03-07 17:49:01 +09:00
Bill Currie
7649549de6 Handle external field declarations. 2011-03-07 17:23:52 +09:00
Bill Currie
49e446b637 Make field defs global.
It turns out that field defs need to be global after all, so use separate
field def tables in the linker.
2011-03-07 14:45:50 +09:00
Bill Currie
041c5d53b8 Check for the vector being a constant before trying to get its value. 2011-03-07 14:04:36 +09:00
Bill Currie
c069bafad8 Support casting from enum to float. 2011-03-07 14:04:05 +09:00
Bill Currie
5a78758781 Support return in void functions in v6 code. 2011-03-07 13:52:28 +09:00