Bill Currie
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
Bill Currie
9159daf99d
Do simple unreachable block analysis and dead block removal.
2011-01-28 11:45:04 +09:00
Bill Currie
90a2738459
Add the label to the block to which it refers.
2011-01-28 11:41:53 +09:00
Bill Currie
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
Bill Currie
4738c767a2
Fix branch label handling.
...
Conditional branches and goto now go to the correct location.
2011-01-27 21:11:32 +09:00
Bill Currie
95b17d9fd4
Rename add_statement_ref to add_statement_def_ref.
2011-01-27 21:10:37 +09:00
Bill Currie
bd78b8a7ee
Enable more end-of-compilation code.
2011-01-27 15:44:12 +09:00
Bill Currie
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
Bill Currie
c7bbd40df6
Give special treatment to opcodes that have void operand types.
2011-01-27 15:44:11 +09:00
Bill Currie
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
Bill Currie
56106892c8
The & operator produces a pointer, not void.
2011-01-27 15:44:10 +09:00
Bill Currie
1c3cc571f1
Pascal variables need defs too.
2011-01-27 15:44:10 +09:00
Bill Currie
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
Bill Currie
1ed1efbdfb
Implement the ROTL macro correctly.
...
Bit rotation requires unsigned ints.
2011-01-27 15:43:50 +09:00
Bill Currie
a6e65488ec
Remove the "test" parameter from test_expr.
...
The parameter became redundant at some stage.
2011-01-27 08:31:51 +09:00
Bill Currie
f7d60397fb
Set the type for zer_def used by labels.
2011-01-26 22:43:59 +09:00
Bill Currie
36c99a8da1
Get function def creation working again.
2011-01-26 14:48:22 +09:00
Bill Currie
5231e8e9f8
Handle .zero, .return and .param* properly.
2011-01-26 08:01:24 +09:00
Bill Currie
4966120e3b
Untwist param's knickers.
2011-01-25 20:21:41 +09:00
Bill Currie
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
Bill Currie
fd3b594ca5
Move label relocs to the relevant statement block.
2011-01-25 15:45:31 +09:00
Bill Currie
849d013f5e
Allocate space for function parameters.
2011-01-25 15:43:57 +09:00
Bill Currie
f45919df09
Need alias defs after all.
2011-01-25 15:42:26 +09:00
Bill Currie
a648f043de
Give functions their own defspace.
...
The defspace is propogated through the function's sub-scopes.
2011-01-25 12:34:45 +09:00
Bill Currie
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
Bill Currie
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
Bill Currie
69b889756d
Fix a misplaced !
2011-01-25 12:16:13 +09:00
Bill Currie
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
Bill Currie
30701980bc
Get QC cfunctions and defs working.
2011-01-25 12:16:13 +09:00
Bill Currie
ade2e92ea4
Fix the immediate breakage in immediate handling.
2011-01-25 09:08:11 +09:00
Bill Currie
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
Bill Currie
49b7c2d527
Document make_symbol
2011-01-25 09:01:37 +09:00
Bill Currie
0624408317
Move the diagnostic functions into their own file.
2011-01-24 21:54:57 +09:00
Bill Currie
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
Bill Currie
483cbeba03
Fix a few FIXMEs.
2011-01-24 21:13:37 +09:00
Bill Currie
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
Bill Currie
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
Bill Currie
a79b6f0a6c
Partially implement casts.
2011-01-23 11:53:08 +09:00
Bill Currie
bf626c9e00
Fix some simple fixmes
2011-01-23 11:13:56 +09:00
Bill Currie
20b14af024
Add "is_math" type check and use for unary +/-.
2011-01-23 11:12:58 +09:00
Bill Currie
341f370662
Rename is_math() to is_math_op().
2011-01-23 11:12:05 +09:00
Bill Currie
dea6771d92
Create type_nil for passing nil into ...
2011-01-22 16:41:31 +09:00
Bill Currie
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
Bill Currie
55d3d17807
Add a missing break.
...
"return" falling through to "goto" was bad news :)
2011-01-22 15:52:01 +09:00
Bill Currie
d5f669af7a
Nuke temp reference counting.
...
The whole reason for this crazy developement branch :)
2011-01-22 11:40:53 +09:00
Bill Currie
3b462e16cf
Produce flow diagrams of the basic blocks.
2011-01-22 11:40:53 +09:00
Bill Currie
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
Bill Currie
2f09efe547
Break out print_expr into its own file.
2011-01-21 20:31:40 +09:00
Bill Currie
bcb2667b73
Possibly get vector calls working.
2011-01-21 19:40:46 +09:00
Bill Currie
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
Bill Currie
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
Bill Currie
977c48d728
Make operator type overrides clear.
2011-01-21 15:57:53 +09:00
Bill Currie
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
Bill Currie
5c5fa67e43
Minor tweaks of the expression tree output.
2011-01-21 15:34:42 +09:00
Bill Currie
96fae582e3
Implement bind expressions.
2011-01-21 11:27:40 +09:00
Bill Currie
4a24393d8d
Temp expressions now use operands instead of defs.
2011-01-21 11:26:43 +09:00
Bill Currie
2998e578fb
Get very basic deref assignments working.
...
More complicated expressions will not work yet.
2011-01-21 10:59:16 +09:00
Bill Currie
ff6e2584da
Handle block sub-expressions that produce results.
2011-01-21 10:57:47 +09:00
Bill Currie
16e013ff5d
Allow the operand to override the symbol's type.
2011-01-21 10:56:32 +09:00
Bill Currie
3f2793ef72
Get a lot of statements being generated.
2011-01-20 23:26:13 +09:00
Bill Currie
e45ea4b2ae
Move the statement printing near the top of the file.
2011-01-20 18:07:24 +09:00
Bill Currie
35b7278967
Re-disable the next-edge constraints.
...
This gives a flatter layout.
2011-01-20 15:36:10 +09:00
Bill Currie
08e387ebac
Slight improvement to branch instruction nodes/labels.
2011-01-20 15:33:13 +09:00
Bill Currie
f8cc347dce
Print expressions as dot files.
...
While the resulting images can be a little confusing, they're much easier
to read than my cryptic postfix notation.
2011-01-20 15:23:19 +09:00
Bill Currie
0594609b3a
Get the label from the right expression.
2011-01-19 22:25:43 +09:00
Bill Currie
5deda9c9b9
Clean out ex_def as ex_symbol replaces it.
2011-01-19 22:25:04 +09:00
Bill Currie
b5424bca7d
Start work on getting expressions converted to internal statements.
2011-01-19 15:47:45 +09:00
Bill Currie
cac2e04b18
Avoid namespace issues with dstatements.
2011-01-19 11:35:34 +09:00
Bill Currie
f2ef24d0cc
Make symbols use the new value struct for constants.
2011-01-19 11:09:54 +09:00
Bill Currie
e08efe036b
Move the constant expression values into their own struct.
2011-01-19 08:43:24 +09:00
Bill Currie
ff6e1b4d6c
Get functions attempting to generate code.
...
However, the statements are only printed as expressions.
2011-01-19 08:38:09 +09:00
Bill Currie
87ce9f5333
Get basic function creation working.
...
Even more FIXMEs, but the pascal parser can now create functions. The
function records even get emitted.
2011-01-18 12:47:13 +09:00
Bill Currie
8fb1b49a9d
Add support for function symbols.
2011-01-18 10:46:42 +09:00
Bill Currie
b186173d18
Fix a segfault in structure finalization.
2011-01-18 08:59:44 +09:00
Bill Currie
21c0c960ae
Initialize data before types.
...
This is needed for string data. It might be better to init saved strings
separately, but this works for now.
2011-01-18 08:58:43 +09:00
Bill Currie
3fe7a7250a
Handle unnamed structures in make_structure.
...
Symbols may not have null names.
2011-01-18 08:56:47 +09:00
Bill Currie
b76a5c5284
Add my QuakePascal test program.
...
Trivial as it is...
2011-01-17 22:44:58 +09:00
Bill Currie
ea3895805b
Rewrite much to use symbols and symtabs. Gut emit.c. Massive breakage.
...
That which isn't rewritten is horribly broken. However, this does include a
nice mechanism for building QC structs for emitting data.
emit.c has been gutted in the spirit of "throw one away".
There is much work to be done to get even variables emitted, let alone
code. Things should be a little more fun from here on.
2011-01-17 22:34:41 +09:00
Bill Currie
85de101aa3
Break out the defspace code.
...
This also makes some improvements to the free block handling code.
2011-01-17 22:34:41 +09:00
Bill Currie
d3a98af783
Make internal_error globally available.
2011-01-17 22:34:41 +09:00
Bill Currie
537b930ba6
Add new_symbol_type to ease creation of typed symbols.
...
This is inteded for code generation functions that need to create
variables and structures.
2011-01-17 12:04:41 +09:00
Bill Currie
ffea505678
Cleanup QC's constant handling based on QP.
2011-01-14 12:10:28 +09:00
Bill Currie
c38ea0b1bc
Correct the spelling of "nil".
2011-01-14 12:07:40 +09:00
Bill Currie
fb8b3d1211
Rework the lexer and parser to use symbols instead of defs.
...
Currently parses the test program just fine, but code generation is very
very broken.
2011-01-13 14:58:16 +09:00
Bill Currie
8f71986306
Add symbol expressions (like def expressions).
...
They will eventually replace def expressions. Currently quite broken.
2011-01-13 14:54:24 +09:00
Bill Currie
008f4e1929
Don't return .return in advanced mode.
2011-01-13 14:50:25 +09:00
Bill Currie
5b8725ca66
Rewrite the symbol (table) management in light of actual usage.
2011-01-13 14:48:38 +09:00
Bill Currie
ac14db7b1b
Create a function to ease param list building.
...
This is especially useful for QuakePascal, but it will mean I'll be able to
get rid of the parameter reversals in Ruamoko later on.
2011-01-13 14:45:53 +09:00
Bill Currie
258c896e4d
Create a function to build if/else blocks.
2011-01-13 14:33:16 +09:00
Bill Currie
638f4445cc
More return code cleanup.
...
Don't use the true void return in traditional mode.
Prefer the true void return over "done" at the end of functions.
Don't emit the above if the last emitted statement is a return and there is
no label at the end of the function.
2011-01-13 00:46:46 +09:00
Bill Currie
3c9991364b
Add a "void return" instruction.
...
I got fed up with always having to explicty return something.
2011-01-13 00:29:56 +09:00
Bill Currie
9ce9b70484
Add variadic functions.
...
Already extending QuakePascal, and many basics aren't there yet :P
2011-01-12 23:39:59 +09:00
Bill Currie
02b3dde83b
Make QuakePascal programs runnable.
...
Generate a mini-main function (".main", to avoid namespace pollution) that
calls the function named by "program", and make qwaq check for ".main" as
well as "main". If both are present, ".main" will take priority.
2011-01-12 23:39:22 +09:00
Bill Currie
332b7da127
Minor header cleanup.
2011-01-12 23:34:15 +09:00
Bill Currie
96ba4605fa
Fix the micro-optimization that avoided emitting x = x
...
The check was broken by alias defs. If a def is aliased, the alias must be
checked.
2011-01-12 23:33:51 +09:00
Bill Currie
c787923c99
Forgot to remove that print_expr.
2011-01-11 12:11:54 +09:00
Bill Currie
08bf7e24de
Produce nicer code for post-op.
...
The temp notice is still there for arrays (that's caused by issues in
constfold.c, not easily fixilbe), but it's harmless and correct (and nice)
code is much more important.
2011-01-11 12:07:48 +09:00
Bill Currie
ad787ae98a
Set the expression type to the correct value for addresses.
2011-01-11 12:06:40 +09:00
Bill Currie
21d7c78eb7
Allow deep copying of expression trees.
...
Certain expression types (eg labels) are created fresh.
2011-01-11 12:05:29 +09:00
Bill Currie
c73ac30e3f
Turns out decrementing the users was very very bad.
2011-01-11 10:56:36 +09:00
Bill Currie
4af5dad424
More math identities.
...
I forgot about 0 for multiplication and division. Detects division by zero.
2011-01-11 08:49:43 +09:00
Bill Currie
5b43fc1de3
Improve interaction between temps and const_folding.
...
Unfortunately, this seems to overdo the decrementing, so it could be
dangerous, but the ruamoko directory builds properly.
2011-01-11 08:47:52 +09:00
Bill Currie
abe471110c
Make inc/dec_users return the the expression.
...
Just for convenience.
2011-01-11 08:44:43 +09:00
Bill Currie
18ae178eaf
Make new_vector_expr and new_quaternion_expr const-correct.
2011-01-11 08:43:34 +09:00
Bill Currie
4da2657bfe
Fix most of the notice spam (hacky).
...
The implementation and declaration of OP_DONE disagree (not sure what to do
about it yet), so give it special treatment when emitting.
2011-01-11 08:07:04 +09:00
Bill Currie
5f70e27613
Quaternions use quaternion_val, not vector_val :P
2011-01-10 13:23:12 +09:00
Bill Currie
cd38bd71fa
Swap the order of scale and base for calculating the array index.
...
fold_constants doesn't like dags very much, but it can do the right thing
if the expressions are in the right order. This is merely a workaround
until I get to work on the code generator itself.
2011-01-10 12:26:11 +09:00
Bill Currie
d327c860cb
Get messages working again.
2011-01-10 12:26:11 +09:00
Bill Currie
aac8185e4e
Print some addresses in hex.
2011-01-10 12:25:31 +09:00
Bill Currie
6ebee4ad0e
Get the type of Method right.
...
It's a direct structure rather than a pointer.
2011-01-10 12:25:31 +09:00
Bill Currie
ba27db6b93
Use some math identities to eliminate operations on non-constants.
2011-01-10 12:25:31 +09:00
Bill Currie
75ec6bf244
Clean out some unnecessary types from the progs engine and clean up the mess.
...
This is a nasty commit, sorry, but 99% of the commit is interdependent.
2011-01-10 12:25:31 +09:00
Bill Currie
c69ac1f56f
Don't segfault when build_builtin_function detects an error.
2011-01-10 12:25:31 +09:00
Bill Currie
806d9110a2
Support both (* *) and // style comments.
2011-01-10 12:25:31 +09:00
Bill Currie
a17e73c863
Generate code.
...
It's now time to work on the revamp of the symbols, trees, etc.
2011-01-10 12:25:31 +09:00
Bill Currie
e34f7caeb0
Use op instead of integer_val for ADDOP etc.
2011-01-10 12:25:31 +09:00
Bill Currie
edaa6235e7
Get functions/procedures emitted.
...
No actual code is emitted, though.
2011-01-10 12:25:31 +09:00
Bill Currie
74016573b4
Need frame grabbing after all.
2011-01-10 12:25:31 +09:00
Bill Currie
129a36064f
Fix a typo that broke procedure declarations.
2011-01-10 12:25:31 +09:00
Bill Currie
7971b0868e
Break out the begin_function code into its own function.
2011-01-10 12:25:31 +09:00
Bill Currie
c5d0acf0fa
Allow arrays to have ranges (not fully implmented)
2011-01-10 12:25:31 +09:00
Bill Currie
25c36a51a8
-f and -F were reversed in the usage.
2011-01-10 12:25:31 +09:00
Bill Currie
fed82466cc
The first stages of Quake-Pascal.
...
At present, this parses a subset of the pascal language with a few type
extensions. There is already work on the type system.
2011-01-10 12:25:31 +09:00
Bill Currie
ae4ca52661
Make function.h more independent.
2011-01-10 12:25:31 +09:00
Bill Currie
66ac56063f
Break out the frame macro code from qc-lex.l
2011-01-10 12:25:31 +09:00
Bill Currie
a09f02db1a
Break out the debug and string functions from qc-lex.l.
2011-01-10 12:25:30 +09:00
Bill Currie
695f71eeb6
Get clost to the look I wanted for symtab_flat_copy's graph.
2011-01-10 12:25:30 +09:00
Bill Currie
782a4c9a8d
New symbol table code.
2011-01-10 12:25:30 +09:00
Bill Currie
fbdbcfada4
Remove deprecated @argc and @argv
2011-01-05 13:10:57 +09:00
Bill Currie
a27ea057c4
Add a test for calling class methods.
2011-01-03 18:11:46 +09:00
Bill Currie
02f78e46f8
More expression docs.
2011-01-03 16:25:59 +09:00
Bill Currie
5627e43465
Use the method's selector rather than its def name in error messages.
2011-01-03 16:25:29 +09:00
Bill Currie
a1f36a9944
Make new_function take a def rather than a name.
...
Moves into new_function some code that should have been there in the first
place.
2011-01-03 16:25:29 +09:00
Bill Currie
4c69a97ac7
Allow normal declarations within a class implementation.
...
I had always been under the assumption that such was illegal in
Objective-C, but after working on Forge, I now know otherwise.
2011-01-03 16:25:29 +09:00
Bill Currie
dc5bafb8fb
Minimal error recovery.
...
No more bailing on the first syntax error. While the recovery is not
perfect (things can get right twisted), qfcc will now continue parsing
after a syntax error.
2011-01-03 16:25:29 +09:00
Bill Currie
a4635d79e8
Slightly more informative error locations.
...
Inside a class or protocol definition is not "at top level" ;)
2011-01-03 16:25:29 +09:00
Bill Currie
530e8fae34
Rework class_type_t to be more useful.
...
This allows current_class to refer to procols as well (for error
reporting).
2011-01-03 16:25:29 +09:00
Bill Currie
f738ef5dba
Remove redundant RETURN statement rule.
...
The same effect can be achieved with one rule using opt_expr.
2011-01-03 16:19:28 +09:00
Bill Currie
58042719e5
A little more documentation for expressions.
2011-01-03 16:19:28 +09:00
Bill Currie
1a767849ab
Sort the options in the switch.
...
Should make finding them a little easier.
2010-12-24 17:35:33 +09:00
Bill Currie
d6b3d4891a
Finally, help output for qfprogs.
...
Nothing great, but better than nothing.
2010-12-24 17:33:57 +09:00
Bill Currie
f4d8b82115
Rearrange the options into alpha order.
2010-12-24 17:08:48 +09:00
Bill Currie
28740bb57f
Add unsigned integer constant support.
2010-12-23 19:32:28 +09:00
Bill Currie
6c631c6d5d
Start documenting qfcc.
2010-12-19 11:35:18 +09:00
Bill Currie
52e844ee48
Produce slightly better v6 modulo code for complex inputs.
...
I'm having problems with temp defs, but I imagine this code is rarely used
so I won't worry about it for now. It will, however, make a good test case
for when I rework emit.c after I do some reading.
2010-12-19 11:23:33 +09:00
Bill Currie
b90833d86f
v6 modulo test
2010-12-19 11:19:23 +09:00
Bill Currie
fa08e4afb7
Print users info when dumping statements.
2010-12-19 11:16:04 +09:00
Bill Currie
f5706f9c44
Make control of expression dumping less of a hassle.
2010-12-19 11:15:13 +09:00
Bill Currie
d981968208
v6 modulo implementation now produces good code.
...
At least for simple inputs, anyway. That needs to be tested next.
2010-12-18 22:21:55 +09:00
Bill Currie
00e50c0ad3
Rewrite the v6 modulo implementation.
...
While trying to understand bind expressions, I discovered that the v6
modulo code was incorrect. This version is correct but uses one too many
temps. That's next.
2010-12-18 22:21:55 +09:00