bceb98e5a4
Create structures for new qfo format.
...
This breaks a lot of existing code, so that code has been commented out
for now.
2011-02-17 15:02:09 +09:00
9c9a71f1af
Allow defs to specify use of offset relocations.
...
Access to struct fields in near data can be done using only one operand,
but offset relocs need to be used. However, as not all defs want offset
relocs, a flag has been added to the def struct.
2011-02-15 22:55:15 +09:00
d04865884f
Set the method's function pointer.
2011-02-15 22:27:13 +09:00
3721b2edca
Method lists are made of methods, not integers :P
2011-02-15 22:26:44 +09:00
c068138d91
More addresses in hex.
2011-02-15 22:26:09 +09:00
814e36d3c8
Correct an error message and tidy some code.
2011-02-15 22:25:40 +09:00
4336fc2c73
Move method function defs into the far data space.
...
As they are never referenced directly by instructions, there's no need for
them to be in the near data space, taking precious def locations.
2011-02-15 15:33:18 +09:00
5252d72856
Correct the type for _OBJ_SELECTOR_TABLE.
2011-02-15 15:33:12 +09:00
b3da6ef3d4
Allow zero size arrays to match sized arrays of the same type.
...
This allows externally declared arrays to not specify a size.
2011-02-15 12:08:03 +09:00
b2d4aee61e
Reimplement selector expressions to work with far data.
...
They currently execute a lea instruction every method call, but the
optimizer should take care of that once its written.
2011-02-15 10:48:05 +09:00
6bd63f811f
Preserve any relocs when converting a def from extern to otherwise.
2011-02-15 10:47:16 +09:00
62456dc1e9
Yet another test.
2011-02-15 09:48:10 +09:00
4e86b28b59
Use the def type rather than expression type to select the init method.
...
Local defs may be initialized with non-constant expressions, which
includes function calls. Both function calls and aggregate initializers
use block expressions.
2011-02-15 09:38:44 +09:00
19f6faf0ad
Fix aggregate initializers.
...
Initializing arrays and structs seems to be working.
2011-02-15 09:30:37 +09:00
d162838299
Resurrect constant_expr().
2011-02-15 09:28:27 +09:00
bc882ddc3c
Fix the parser side of aggregate initializers.
2011-02-14 23:31:37 +09:00
09af401f0c
Reset current_func at the end of each function.
2011-02-14 23:10:46 +09:00
5fa349927f
Preserve the ivars inheritance chain across build_struct().
...
build_struct() unlinks the structure symbol table from its parent, but
that breaks the ivar inheritance chain.
2011-02-14 23:10:46 +09:00
5707888fcf
Fix some error-recovery segfaults.
2011-02-14 23:10:46 +09:00
f1fccb19dd
Allow same-type re-declarations of uninitialized variables.
...
This is especially important for declaring variables that were previously
declared external.
2011-02-14 23:10:46 +09:00
b3d5789377
More tests.
2011-02-14 23:10:46 +09:00
23b01be214
Handle invalid initializers gracefully.
2011-02-14 23:10:46 +09:00
4c63a0bad6
Generate a warning when creating a var with no type.
...
This makes the internal error in new_def() non-functional, but the warning
will catch the bug and there's no harm in leaving the test in new_def().
2011-02-14 23:10:46 +09:00
ee866ae3a6
Enable the small struct optimization.
...
For now, anyway. It would probably be much better to handle this sort of
thing in the optimizer stage.
2011-02-14 23:10:45 +09:00
bcaf3687c5
Fix loading structs into params via an offset pointer.
...
For certain values of "fix" :/. The code is ugly, but it does the right
thing: calculate the effect address and use the resulting pointer in a
move instruction.
2011-02-14 23:10:45 +09:00
352be1f8bd
Handle null string pointers.
2011-02-14 23:10:45 +09:00
e10d561459
Treat small structs as integers (disabled).
...
This is more to stash away a potential optimization. I need to fix the
real problem first.
2011-02-14 23:10:45 +09:00
6b29568fa9
Set the source file and line when copying an expression.
...
This fixes the crazy line numbers in the debug info.
2011-02-14 23:10:45 +09:00
fc18ebd923
Add the source line number to most node labels.
2011-02-14 23:10:45 +09:00
3e5989463f
Update the def_tail when removing the last def from the space.
2011-02-14 23:10:45 +09:00
11485556e5
Quote expression names to avoid problems with null pointers.
2011-02-14 23:10:45 +09:00
c139099294
Put another bandaid on the type creation system.
...
It seems that building types is very fragile: all too easy to fix things
in one place and break something somewhere else. I guess the type system
will need a comlete rewrite at some stage.
2011-02-14 23:10:45 +09:00
e3325b199e
More tests.
2011-02-14 23:10:45 +09:00
e0d61705b4
And fix build_struct for built-in struct types.
...
The previous fix broke builtin structs (eg, @param). Calling find_type()
for built-in structs has proven to be a bad idea :)
2011-02-13 21:07:22 +09:00
a19b7ec8f9
Make struct decls actually work.
...
find_type() does type comparisons based on tag name for struct/union/enum
but when it returns the previous tag (eg, from "struct foo;") the returned
type has a null symtab pointer, so setting the symtab pointer in the type
before calling find_type() only throws away the symtab. Thus, since
find_type() doesn't check the symtab pointer when comparing the types, set
the symtab pointer after calling find_type().
2011-02-13 20:56:12 +09:00
acced90101
Create the symbol for the struct as early as possible.
2011-02-13 20:55:33 +09:00
b545f72f93
Report errors for incomplete types and missing fields.
2011-02-13 20:23:08 +09:00
c336aaeb80
Handle incomplete types gracefully.
2011-02-13 19:26:20 +09:00
ef32d1fb6d
Save away some of my tests.
2011-02-13 19:02:39 +09:00
976ff8a6bb
Allow cast expressions to be lvalues.
...
These really need to be alias expressions rather than cast expressions
so the two can be separated as cast expressions should not be lvalues.
2011-02-13 17:54:15 +09:00
fa742ceed6
Handle @self and @this correctly.
2011-02-13 17:52:19 +09:00
547f361bd3
Allow (void) qc-style functions.
2011-02-13 17:37:11 +09:00
d75c35ba30
Allow qc style function params.
...
Due to ambiguities in the grammar, qc-style function params and c-style
function params had to be completely separated. This means that qc-style
functions can not use pointers and must use qc-style function declarations
for parameters, and c-style functions must use c-style param declarations.
While this rule is tedious for converting the Ruamoko library, it does
actually make for a more consistent language.
2011-02-13 17:07:14 +09:00
8a314e9e87
Class pointers need to be directly addressable.
2011-02-13 16:05:50 +09:00
0b5565396b
Add a class type to symbols and rework the CLASS_NAME handling.
...
This fixes the problem with [Array alloc] producing a warning about Class
not repsonding to -alloc.
2011-02-13 16:05:09 +09:00
3f690eacb1
Fix another double insertion of a symbol.
2011-02-13 15:28:49 +09:00
127a1c8772
Fix the creation of enum values.
2011-02-13 15:19:55 +09:00
f24de89a2a
Handle type-modified constants.
...
The fix for pointers broke constants (particularly, enums). Oops.
2011-02-13 14:13:32 +09:00
fa6b8d773c
Correctly emit class and category symtab references.
2011-02-13 13:21:00 +09:00
c4f364fd37
Emit pointers properly.
...
Rename ReuseConstant to emit_value and use emit_value to emit any non-short
constant. This fixes the null pointer in the call to __obj_exec_class.
2011-02-13 10:09:42 +09:00
1bcf434b20
Move _OBJ_MODULE to the far data space.
2011-02-12 23:00:04 +09:00
d42a379924
Dump the contents of a pointer def.
2011-02-12 22:59:29 +09:00
eede83dea1
Avoid crashing when a null class pointer is found.
2011-02-12 22:58:50 +09:00
3505de6868
Point the near and far data spaced into the merged data space.
...
This lets the relocations work on the final data without having to adjust
their locations.
2011-02-12 22:35:44 +09:00
181596f5bf
Make relocation records store the space of the relocation.
...
The space is meaningless for op_* relocations as they are always in the
code space, but def_* relocations need to know which space holds the
location to be adjusted.
2011-02-12 22:34:38 +09:00
cfdc2e5d6b
Complete the documentation for relocations.
...
This covers only in-memory storage of relocation records, not object file
relocations.
2011-02-12 15:55:58 +09:00
8c01497936
Give __obj_exec_class a sterner beating.
...
When linking a def into a defspace, ensure it can be removed. This fixes
the missing __obj_exec_class def in qwaq.
2011-02-11 22:29:57 +09:00
75d90d2f00
Propogate the specifiers through external_decl_list.
2011-02-11 22:29:57 +09:00
0efc759f2d
Catch attempts to create a def with a null type.
2011-02-11 22:29:57 +09:00
5f3c1872b9
Put class and category reference/name symbols in the symbol table.
...
This fixes the undefined defs in qwaq.
2011-02-11 22:29:57 +09:00
179c1f7058
Support move statements (structure copy).
2011-02-11 22:29:57 +09:00
e19f7e3521
Sort out the parameters for QC style functions :)
2011-02-11 22:29:57 +09:00
370f0b97ed
Produce cleaner output for a = b op c.
2011-02-11 22:29:57 +09:00
d93d8d7d46
Remove the dirty hack used for accessing params and the return value.
...
Instead of using the equivalent of *(float*)&.return, now use the
equivalent of (float).return. No conversion is done in the "cast".
NOTE: this sort of cast should be separated from normal casts.
2011-02-11 22:29:38 +09:00
f5412c2b20
Remove a fixme.
2011-02-11 22:29:38 +09:00
c185fbb782
Set the type of the super structure to the correct type.
2011-02-11 22:29:38 +09:00
7d2a95ee40
Mark a few problem spots.
2011-02-11 22:29:37 +09:00
a1d4d4bae1
Handle block expressions in is_indirect().
2011-02-11 22:29:30 +09:00
b01f960db9
Disable the subgraphs for now.
...
They tend to make large trees difficult to read. I might re-enable them
when I've gotten a better handle on dot.
2011-02-10 15:11:08 +09:00
c3a991e555
Support nil in initializers.
2011-02-10 14:51:52 +09:00
adeef4a4eb
Copy named expressions.
...
Since expressions can be modified, making DAGs this early causes
corruption of the expression tree.
2011-02-10 14:43:49 +09:00
6f625e426f
Use the correct type when deferencing an offset pointer.
2011-02-10 14:26:12 +09:00
1099ca086c
Dump the offending expression and statement for a bad instruction.
2011-02-10 14:25:09 +09:00
4a41d97afc
Angle brackets and html :P
2011-02-10 14:24:31 +09:00
01a7af17e5
Beat __obj_exec_class into submission.
2011-02-10 00:15:19 +09:00
006c16807d
Add the statement for based loads.
...
Oops :P
2011-02-09 22:40:16 +09:00
a43946abf3
Give short values special treatment.
...
As a short value fits directly into a statement, it needs a fake def that
holds its value in the def's offset in order for the value to be written to
the statement.
2011-02-09 22:13:43 +09:00
c7adde8709
Fix some case&paste errors.
2011-02-09 22:04:54 +09:00
1b8edc07c9
Adjust the offsets of ivars in derived classes.
...
As ivar structs are not independent, their starting offset is not
necessarily zero (almost never will be for derived classes).
2011-02-09 22:02:52 +09:00
d66a441867
Write out the far data space as well as the near.
...
This causes the Objective-QC structs to be emitted now that they havve been
moved to the far data space. Relocs are broken, though.
2011-02-09 21:45:29 +09:00
244d969ae3
Ensure any tail bytes in the strings data are zeroed.
2011-02-09 18:32:37 +09:00
2d5d8eda5a
Save the caller's address when creating new relocations.
2011-02-09 17:16:49 +09:00
29c1f7ac03
Use the correct type object in emit_struct().
2011-02-09 17:15:29 +09:00
59a683d39e
Protect the name passed into emit_struct().
...
The name passed into emit_struct is sometimes created via va(), but the
struct creation code also uses va. This makes a mess of names.
2011-02-09 17:12:47 +09:00
3e1b034aaf
Use EMIT_DEF correctly.
2011-02-09 15:58:32 +09:00
e75295b00c
Make normal C style functions generate code.
2011-02-09 10:25:23 +09:00
68cdc904f5
Give temporary variables unique names.
2011-02-09 10:04:55 +09:00
681ded9375
Fix debug line number information.
2011-02-09 09:59:43 +09:00
5eff00fdeb
Fix debug information.
2011-02-09 08:47:01 +09:00
71c05bd899
Update for the new syntax.
...
This one compiles to progs even :) (though it looks broken).
2011-02-08 22:39:42 +09:00
81125cff3a
Creation of .ctor now mostly works.
...
There seems to be problems with the actual code generation using incorrect
addresses, but a very simple test file compiles successfully.
2011-02-08 22:38:34 +09:00
1647b1e61d
Check for the correct type in emit_symtab_cat_def_cnt().
2011-02-08 22:31:09 +09:00
9b52c590f8
Do not try to emit anything when there are no selector references.
2011-02-08 22:28:32 +09:00
da14c95bb8
Initialize current_symtab in class_init() when necessary.
...
If an objective-qc symbol is the first token to be read, class_init() is
called before the parser gets a chance to perform the initialization
action. However, class_init() setting current_symtab if the first
objective-qc symbol is seen in a function body would be bad.
2011-02-08 22:26:35 +09:00
c01f742b7b
Fix some Class vs ClassPtr confusion.
2011-02-08 18:25:23 +09:00
0d5ab6600d
Store the source expression in a statement.
...
This makes problem reporting more informative.
2011-02-08 18:18:34 +09:00
faab931ffc
Fix assignment of id to class pointers.
2011-02-08 17:12:01 +09:00
6643c5dd98
Fix messages to super.
2011-02-08 17:09:27 +09:00
505e141633
Ensure the class symbol is in the symbol table.
2011-02-08 17:08:58 +09:00
93a5808659
Support direct struct/union field access.
2011-02-08 14:48:26 +09:00
3c849b970b
Handle union access now that they're detected properly.
2011-02-08 14:45:48 +09:00
204a267b6a
Print the def name for a pointer expr when known.
2011-02-08 14:43:57 +09:00
3b080fcbb0
Fix an uninitialized variable.
2011-02-08 13:27:37 +09:00
2594be8b1b
Fix a type-o ;)
2011-02-08 13:26:49 +09:00
8c6fc6fb03
Fix a double insertion of a symbol.
2011-02-08 08:44:08 +09:00
daa88d8f1c
Fix a type clash warning I'd failed to notice.
2011-02-08 08:23:25 +09:00
35d53d0d77
Fix typedefs and class names once and for all.
...
I suspect Holub did them that way as a teaching exercise :P
2011-02-08 08:20:32 +09:00
7e2173f268
Remove some old code.
2011-02-08 08:16:36 +09:00
92799650c0
Implement struct field access via a pointer.
2011-02-07 22:15:10 +09:00
7a1a9e3f2d
Better handling of the current storage for function scopes.
2011-02-07 22:11:22 +09:00
ff70177041
Fix type handling for local variables.
2011-02-07 21:38:58 +09:00
952a02eb23
Make print_type() output a newline.
2011-02-07 21:31:37 +09:00
a28ac636a7
Make local declarations support C syntax.
2011-02-07 21:29:04 +09:00
da4ec3d47c
Allow typedefs and classes in structs.
2011-02-07 20:55:13 +09:00
aefd2a27d5
Make type handling in the grammar more consistent.
2011-02-07 20:54:49 +09:00
3c64ae7b66
Fix overzealous type merging.
...
Use the type name instead of the symbol table for comparing
struct/union/enum types.
2011-02-07 20:48:01 +09:00
265a244ba2
Set the symbol type for structs etc as soon as the tag is created.
...
Tags (struct/union/enum) are always types.
2011-02-07 15:30:49 +09:00
65a07ada1c
Implement offset dereferences.
2011-02-07 14:56:29 +09:00
d2b464053c
Fix assignment sub-expressions.
2011-02-07 14:43:07 +09:00
6fa5bbe473
Support constant folding on enums.
2011-02-07 11:31:23 +09:00
d549e5c6b5
Allow enums to be tested.
2011-02-07 11:30:56 +09:00
ef2ad46f7a
Convert complex types to a suitable low-level type.
...
This takes care of moving structures etc around.
2011-02-07 10:55:09 +09:00
b629c12b31
Do not change the type of the return operand.
...
Doing so made a mess of returning expressions. It was probably a holdover
from before getting instructions with void operand types working.
2011-02-07 10:21:54 +09:00
829fb33e92
Fix converting nil to a pointer type.
2011-02-07 10:10:48 +09:00
b8127f7543
Do not segfault when printing a pointer value with a bogus type
2011-02-07 09:57:26 +09:00
28ba4b5712
Correct a comment.
2011-02-07 09:56:33 +09:00
d564e6f017
Clean out the old debug prints from emit_function()
2011-02-07 08:46:12 +09:00
76034861c9
Implement direct ivar access in methods.
...
A warning needs to be generated for access to ivars in a class method, but
it seems to work (unable to check due to other errors).
2011-02-07 08:32:52 +09:00
3025e04ac5
Partially resurrect field_expr()
...
For now, only entity.field and ivar access via a pointer have been
implemented. The rest will come later as needed.
2011-02-07 08:30:28 +09:00
1d108a8cec
Support emitting symbolic constants.
2011-02-06 20:08:54 +09:00
e4ac92b5f2
Auto-cast between enums and the default type.
2011-02-06 20:08:16 +09:00
6b7a81f5a7
Run find_type() early on enums and structs.
2011-02-06 20:07:19 +09:00
d54a96e68a
Make enum values visible.
...
Put them in the current symbol table as well as the enum symbol table. The
latter is really just for keeping track of what values are in the enum.
2011-02-06 20:06:09 +09:00
2860a8b28b
enum too is a scalar.
2011-02-06 20:01:55 +09:00
f2c272f0b8
Make adding a symbol to multiple tables an internal error.
2011-02-06 20:00:14 +09:00
d7cdf8077b
Rename id_ivars to object_ivars.
2011-02-06 15:41:15 +09:00
c3c693e9c7
Correct construction of type_id.
2011-02-06 15:36:21 +09:00
e27f3fd7c3
Copy the method's params into the method symbol.
2011-02-06 15:14:42 +09:00
00bba1eb3a
Fix creation of uninitialized global variables.
2011-02-06 14:37:25 +09:00
59d3dc6ae0
Check early for assignment of a class pointer to id.
2011-02-06 14:35:17 +09:00
2f7b892452
Fix passing messages to classes.
...
Rename class_pointer_def() to class_pointer_symbol() and use it when a
class name is used as the receiver of a message.
2011-02-06 14:34:40 +09:00
67957a14df
Bring back convert_name(), but siimplified.
...
It is now mainly for converting __FILE__ etc, handling expression symbols,
and checking for undefined identifiers.
2011-02-06 14:29:14 +09:00
d03fdc5e67
Correctly calculate the size of the ivars of a class.
...
Need to include the size of the super class ivars.
2011-02-05 00:04:35 +09:00
690ae1cb57
Fix send_message().
2011-02-04 23:49:25 +09:00
29fbc3b073
Struct and ivar declarations now seem to work.
...
However, not bit-fields.
2011-02-04 23:48:58 +09:00
1f6abb2cce
Minor cleanups to message_expr().
...
Fix for class names no longer being pointers to the class.
2011-02-04 23:22:51 +09:00
eff3957487
Resurrect the address expression code for arrays etc.
...
This fixes selector finding.
2011-02-04 23:20:37 +09:00
c542e1a453
Possibly fix normal methods.
2011-02-04 22:31:23 +09:00
d9446c5eb0
Fix builtin methods.
2011-02-04 22:31:22 +09:00
d0ecbb38b8
Copy the spec bits and storage class for QC functions.
2011-02-04 15:25:21 +09:00
f721144ca5
Rewrite emit_ivars to use emit_structure.
...
The old code was broken anyway :) (it was segfaulting).
2011-02-04 15:11:47 +09:00
acb3c0da11
Fix class initialization.
2011-02-04 14:29:47 +09:00
04b86fcdb7
Move the Objective-QC structs from type.c to class.c
2011-02-04 12:07:32 +09:00
7b22bfba63
Better Objective-QC keyword detection.
...
Use a flag in the keyword struct rather than checking for "@" in the
keyword. This catches such keywords as "id" and "Protocol" as well.
2011-02-04 08:38:58 +09:00
993bee4f59
Do not count unimplemented methods.
...
The method emitter skips unimplemented methods, so counting them when
figuring out how many methods to emit leads to a segfault.
2011-02-04 08:26:16 +09:00
e04ed72afa
Increment the index when emitting array elements.
...
Oops :)
2011-02-04 08:25:27 +09:00
d7853682af
Fix conversion of nil to a real type.
2011-02-03 21:37:15 +09:00
f44a26e268
Fix type chaining when dealing with decls.
2011-02-03 21:32:58 +09:00
54c8a10fe4
Don't trip over incomplete structs (again:P)
2011-02-03 21:31:14 +09:00
12d8658ece
Create a typedef for obj_module_t.
2011-02-03 18:05:46 +09:00
b07392b818
Allow ivars to use C syntax declarations.
2011-02-03 18:04:41 +09:00
dc6296e9b1
Allow ivars to use typedefs and other classes.
2011-02-03 18:03:37 +09:00
2203903bb2
Temporarily splice the class's ivar chain into the symbol table.
...
The root class's ivar symbol table needs to be connected to the global
symbol table while building the current class's ivar symbol table. This
allows access to other classes and typedefs.
2011-02-03 18:02:49 +09:00
79df3d2cca
Put newly created classes into the symbol table.
2011-02-03 17:28:49 +09:00
564b7b9968
Use abstract_decl instead of type wherever a generic type is needed.
2011-02-03 10:53:59 +09:00
a208aed6ef
Fix a segfault when encoding incomplete structs.
2011-02-03 10:52:42 +09:00
22ccd41164
Give struct types a name.
...
Use the struct/union's name (minus "tag ") for the type name.
2011-02-03 10:52:12 +09:00
d12cdca758
Functions and enums now seem to be working properly.
...
find_type now operates recursively (depth-first) so built up typechains
work as expected.
@overload is treated as a specifier (directly as a storage class, similaar
to typedef).
2011-02-03 10:48:12 +09:00
52799a4e1c
Must merge specifiers :P
2011-02-02 22:59:08 +09:00
4420a7125e
Ensure structs get added to the symbol table.
2011-02-02 22:53:53 +09:00
a4eaa2a89b
Allow typedefed types and classes in parameters.
2011-02-02 21:58:53 +09:00
aeecdc84dc
Get typedef working.
2011-02-02 21:58:11 +09:00
cca829aa6b
Remove a stray semicolon from the grammar.
2011-02-02 21:57:41 +09:00
400abe7fce
Allow freeing of typechains.
2011-02-02 15:15:24 +09:00
1fe031e8cb
Build C style var/func/abs decl types from the outside in.
...
QuakeC style are built from the inside out.
2011-02-02 15:15:04 +09:00
7fb8c8247e
Don't do a find_type on the parsed function.
...
This is needed for C style declarations, but breaks QuakeC style. The fix
is to do the find_type once the return type has been filled in.
2011-02-02 15:08:59 +09:00
1452da676d
Allow types to be built from the outside in.
...
This is required for C style decl processing.
2011-02-02 15:07:44 +09:00
ac91e9376b
The big type-system rewrite. Horribly broken :)
...
While broken, it is good enough to compile modulo.r.
This gives qfcc C's type system with QC specifics bolted on top (rather
than trying to cram C's type system into QC's). Pointers and arrays now
use C syntax and semantics (though I think there's a lot of breakage in
there at the moment). Functions themselves are still dual mode: both QC
style and C style are available.
2011-02-01 21:23:03 +09:00
d11d128c96
Avoid segfaulting on errors.
2011-02-01 21:18:39 +09:00
80ce179bc2
Scan the param list for problems with void.
2011-02-01 21:18:08 +09:00
c53001800a
Create a struct for representing specifiers.
...
The specifiers are yet to come (next few commits), but this will be
necessary when they do.
2011-02-01 21:15:51 +09:00
917c5f08b0
Make a var representing the default type.
...
The default type for vars is integer for ruamoko and float for v6 qc.
2011-02-01 21:12:02 +09:00
54c14eb899
Add "return" to pascal.
...
I'm still not sure I'll keep it, but it's useful for testing.
2011-01-30 22:16:59 +09:00
e7424e1496
Don't segfault on bare return statements.
...
I will probably need to do something about the required operand for v6
progs, but this will do for now.
2011-01-30 21:43:56 +09:00
440a13b649
Untwist params' knickers for pascal, too.
2011-01-30 21:42:10 +09:00
861ab93274
Don't type-check against the wrong type :P
2011-01-29 09:56:08 +09:00
afc6605739
Don't set the space for external defs.
2011-01-29 09:55:14 +09:00
d80a3413e5
Fix a segfault when freeing locations.
...
When chasing linked lists with a double pointer, must check the first
dereference and not the pointer itself.
2011-01-29 09:54:47 +09:00
e1226bd3f1
Quaternions aren't vectors :P
2011-01-28 21:57:04 +09:00
2cce903cc9
Set the default storage class for local vars.
2011-01-28 21:56:17 +09:00
bc7d42d176
Immediate constants now work, though still not shared.
2011-01-28 21:39:01 +09:00
3c95667b79
Print data contents for most simple types.
2011-01-28 21:21:29 +09:00
4476074021
Print more addresses as hex.
2011-01-28 18:03:26 +09:00
4837e0bd18
Fix most of the FIXMEs in qfcc.c.
...
Functions and defs seem to be relocated correctly. Constants are very
broken, however.
2011-01-28 15:17:30 +09:00
834417b8c8
Add "debug" diagnotic printing, and use it.
...
Debug diagnostics are silent for verbosity levels less than 1.
2011-01-28 13:28:45 +09:00
a37bdd9fb5
Free removed sblocks.
2011-01-28 13:23:20 +09:00
80424f6ad5
Ensure functions are always properly terminated.
...
Control that reaches the end of a function must see a return statement.
2011-01-28 13:14:30 +09:00
9159daf99d
Do simple unreachable block analysis and dead block removal.
2011-01-28 11:45:04 +09:00
90a2738459
Add the label to the block to which it refers.
2011-01-28 11:41:53 +09:00
c158b14286
Remove the labels list from the pr struct.
...
I'm not sure they were ever needed.
2011-01-28 11:40:25 +09:00
4738c767a2
Fix branch label handling.
...
Conditional branches and goto now go to the correct location.
2011-01-27 21:11:32 +09:00
95b17d9fd4
Rename add_statement_ref to add_statement_def_ref.
2011-01-27 21:10:37 +09:00
bd78b8a7ee
Enable more end-of-compilation code.
2011-01-27 15:44:12 +09:00
cdea1d07aa
Produce code for progs.src mode.
...
Separate compilation will take fixing object files.
The generated code is broken due to various relocation fixups being
broken, and float immediates seem to be badly broken.
Debug information is broken too.
2011-01-27 15:44:11 +09:00
c7bbd40df6
Give special treatment to opcodes that have void operand types.
2011-01-27 15:44:11 +09:00
71b5decbe8
Get assignments to .return working.
...
I really do not like the mechanism I currently use for handling pointer
derefences vs pointer assignments, but this will have to do for now until
I can get qfcc producing code again.
2011-01-27 15:44:11 +09:00
56106892c8
The & operator produces a pointer, not void.
2011-01-27 15:44:10 +09:00
1c3cc571f1
Pascal variables need defs too.
2011-01-27 15:44:10 +09:00
f532fbc6df
Convert all integer operands to float for v6 progs.
...
Converting only opc is no longer enough.
2011-01-27 15:43:59 +09:00
1ed1efbdfb
Implement the ROTL macro correctly.
...
Bit rotation requires unsigned ints.
2011-01-27 15:43:50 +09:00
a6e65488ec
Remove the "test" parameter from test_expr.
...
The parameter became redundant at some stage.
2011-01-27 08:31:51 +09:00
f7d60397fb
Set the type for zer_def used by labels.
2011-01-26 22:43:59 +09:00
36c99a8da1
Get function def creation working again.
2011-01-26 14:48:22 +09:00
5231e8e9f8
Handle .zero, .return and .param* properly.
2011-01-26 08:01:24 +09:00
4966120e3b
Untwist param's knickers.
2011-01-25 20:21:41 +09:00
6ce3c5a59a
Get statments being emitted to the code segment.
...
The first function seems to work fine, but there's a problems with the
scope of params causing params to get their knickers in a twist (tangled
linked list).
2011-01-25 15:46:48 +09:00
fd3b594ca5
Move label relocs to the relevant statement block.
2011-01-25 15:45:31 +09:00
849d013f5e
Allocate space for function parameters.
2011-01-25 15:43:57 +09:00
f45919df09
Need alias defs after all.
2011-01-25 15:42:26 +09:00
a648f043de
Give functions their own defspace.
...
The defspace is propogated through the function's sub-scopes.
2011-01-25 12:34:45 +09:00
27c5139612
Give defspaces some backing memory.
...
I'm not sure I want all defspaces to have backing memory, but this will do
for now.
2011-01-25 12:16:13 +09:00
34ffce50fa
Ensure current_symtab gets set at the start of parsing.
...
current_symtab must be set even if the first thing encountered is a syntax
error.
2011-01-25 12:16:13 +09:00
69b889756d
Fix a misplaced !
2011-01-25 12:16:13 +09:00
726f293d0d
Allocate space for defs and handle basic initializations.
...
Block initializers (arrays and structs) are still broken.
2011-01-25 12:16:13 +09:00
30701980bc
Get QC cfunctions and defs working.
2011-01-25 12:16:13 +09:00
ade2e92ea4
Fix the immediate breakage in immediate handling.
2011-01-25 09:08:11 +09:00
7e914b8c53
Fix copyright headers.
...
Add forgotten templates and fill in the information in places I had
forgotten. I hope I got the dates right (timezone issues).
2011-01-25 09:01:37 +09:00
49b7c2d527
Document make_symbol
2011-01-25 09:01:37 +09:00
0624408317
Move the diagnostic functions into their own file.
2011-01-24 21:54:57 +09:00
d618e51dc8
Move save_string and make_string into strpool
...
Might not be the perfect place, but at least they're strongly related.
2011-01-24 21:32:48 +09:00
483cbeba03
Fix a few FIXMEs.
2011-01-24 21:13:37 +09:00
afc5b2827d
Simplify def handling, use far data.
...
Since there is now a proper symbol table, defs are now just references to
memory locations and the symbol table takes care of duplicates.
Also, start using far data for ObjQC structures.
The qfo functions have been stubbed out until I figure out what to do with
object files in the new scheme.
2011-01-24 15:41:43 +09:00
ce02fe8660
Fix the order of paramenters in the tree output.
...
Also get call nodes looking a little better.
2011-01-23 12:59:30 +09:00
a79b6f0a6c
Partially implement casts.
2011-01-23 11:53:08 +09:00
bf626c9e00
Fix some simple fixmes
2011-01-23 11:13:56 +09:00
20b14af024
Add "is_math" type check and use for unary +/-.
2011-01-23 11:12:58 +09:00
341f370662
Rename is_math() to is_math_op().
2011-01-23 11:12:05 +09:00
dea6771d92
Create type_nil for passing nil into ...
2011-01-22 16:41:31 +09:00
4ac80c2932
Get qfcc running again.
...
It segs when it tries to link, but that's because a real object file has
not been created yet. However, I can get my tree and flow diagrams :)
2011-01-22 15:52:57 +09:00
55d3d17807
Add a missing break.
...
"return" falling through to "goto" was bad news :)
2011-01-22 15:52:01 +09:00
d5f669af7a
Nuke temp reference counting.
...
The whole reason for this crazy developement branch :)
2011-01-22 11:40:53 +09:00
3b462e16cf
Produce flow diagrams of the basic blocks.
2011-01-22 11:40:53 +09:00
246778b5de
Use function pointer tables instead of switch.
...
This seems to be better for complicated code. switch seems to be better
suited for relatively simple code.
2011-01-22 11:40:53 +09:00
2f09efe547
Break out print_expr into its own file.
2011-01-21 20:31:40 +09:00
bcb2667b73
Possibly get vector calls working.
2011-01-21 19:40:46 +09:00
52b561f7cb
Nuke bind expressions.
...
Since I'm planning on implementing CSE and other optimizations, they're
rather redundant (and I'm having trouble getting them to work).
2011-01-21 19:07:58 +09:00
0d01930056
Link a bool to its block rather than the block's head.
...
This makes the flow make a little more sense.
2011-01-21 18:50:17 +09:00
977c48d728
Make operator type overrides clear.
2011-01-21 15:57:53 +09:00
aec9f256a7
Get calls mostly working.
...
Bind seems to be broken, but parameters seem to be being set correctly.
2011-01-21 15:54:53 +09:00
5c5fa67e43
Minor tweaks of the expression tree output.
2011-01-21 15:34:42 +09:00
Bill Currie