fd2b7ee6f9
Use more type checking helper functions
2020-02-19 21:41:46 +09:00
a6003ed08a
Walk qfo alias chain for type size and alignment
...
While the basic type is stored in the alias type record, it's no good
for size or alignment as it will give incorrect results for complex
types.
2020-02-19 21:41:46 +09:00
a0914e1ec8
Fix the typedef test case to actually work
2020-02-19 21:41:46 +09:00
6a70d2e362
Give alias types a unique encoding
...
The encoding is used as the def name and it needs to be different than
the alias target or the linker throws it away as an external def.
2020-02-19 21:41:46 +09:00
6bc803c72d
Use correct encoding for alias types
...
I got confused which field was which.
2020-02-19 21:41:46 +09:00
6bcc2c49ab
Use helper functions for type checks
...
They hide the evil details of aliased types. More to come :/
2020-02-19 21:41:46 +09:00
9610788dea
Fix some more type aliasing issues
...
Getting there... (I knew this would be a big job)
2020-02-19 21:41:46 +09:00
d50a27a045
Race down the alias chain before checking types
...
This takes care of some of the type aliasing issues.
2020-02-19 21:41:46 +09:00
a5aba6c8ac
Implement type aliasing
...
The separate types are in the file, but there are multiple issues
2020-02-19 21:41:46 +09:00
2f18364364
Start work on encoding typedef chains
2020-02-19 21:41:46 +09:00
adb7f5d601
Quieten the test script build rules
2020-02-19 21:41:46 +09:00
b00c866c4e
Allow casting between string and pointer types
2020-02-19 21:41:46 +09:00
7a315b4a89
Fix storage class for for-loop declarations
...
Getting "i redeclared" when i was declared in a for loop in two
different functions was a tad unexpected.
2020-02-19 21:41:46 +09:00
67e183c8b6
Put the guts of qwaq's reflection code into ruamoko
2020-02-19 21:41:46 +09:00
4c40928112
Remove what appears to be a redundant check
...
It was long wrong anyway as it checked past the end of the function's
parameters, which caused a segfault when calling varargs functions with
no formal parameters.
2020-02-19 02:53:38 +09:00
a65d6bce09
Fix a warning that got through
...
I forgot to compile test in optimized...
2020-02-19 02:43:27 +09:00
bd6dcafdc8
Replace system defines/includes with qfcc's
...
Right now, it probably works only with modern gcc.
2020-02-19 02:35:09 +09:00
2d52da9c0d
Fix segfault in unlimited params
2020-02-19 02:35:09 +09:00
c61d0b6ff0
Allow unlimited parameters in function declarations
...
However, definitions are still limited to 8 parameters. This allows
processing of C headers for type information.
2020-02-19 02:35:09 +09:00
ee228504aa
Fix self-referenced enum declarations
...
eg:
typedef enum foo {
bar = 1,
baz = bar,
} foo;
2020-02-19 02:35:09 +09:00
7a399c956b
Encode function parameter alignment
...
The encoding is 3:5 giving 3 bits for alignment (log2) and 5 bits for
size, with alignment in the 3 most significant bits. This keeps the
format backwards compatible as until doubles were added, all types were
aligned to 1 word which gets encoded as 0, and the size is unaffected.
2020-02-16 17:10:43 +09:00
1bc08c59f6
Add tests for %%
...
double fails due to qfcc aligning double param locals, but the engine
not doing so.
2020-02-16 17:02:38 +09:00
1a9510834a
Add a missed opcode conversion for %%
2020-02-16 12:10:09 +09:00
9d2d33fa50
Implement %% (true modulo) support in qfcc
...
However, it's not quite working yet
2020-02-16 11:57:58 +09:00
4269c8cb07
Rename the mod instruction to rem
...
Because % really implements remainder rather than true modulo, and I
plan on adding %% to implement true modulo.
2020-02-16 11:04:30 +09:00
db9996023f
Add some tests for double comparison
...
More testing the engine than the compiler, but hey :)
2020-02-15 23:49:12 +09:00
14acfad7c4
Fix incorrect placement of far data
...
All the care in aligning things was undone by not updating the
calculations of the pointers.
2020-02-15 23:49:12 +09:00
1985b6d4fd
Avoid creating a struct temp for ivar struct return
...
This fixed the uninitialized temp warning in HUD.r. The problem was
caused by the flow analyzer not being able to detect that the struct
temp was being initialized by the move statement due to the address of
the temp being in a pointer temp. While it would be good to use a
constant pointer for the address of the struct temp or improving the
flow analyzer to track actual data, avoiding the temp in the first place
results in nicer code as it removes a move statement.
2020-02-15 23:49:12 +09:00
9c996df7b4
Add a test case for the uninit temp in HUD.r
2020-02-15 23:49:12 +09:00
7bfa0f7a92
Allow pragmas to have arguments
...
It does mean only one pragma per line, but that's not such a big deal.
2020-02-15 23:49:12 +09:00
91f5023681
Promote bugs to internal errors
...
Mostly so I can catch them in test cases
2020-02-15 23:49:12 +09:00
344d429134
Test array initializer double demotions
...
Turns out array inits are very strict about types (bug?).
2020-02-15 23:49:12 +09:00
6ce99afa5b
Catch double demotion in global initializers
...
Local initializers are handled by regular assignments
2020-02-15 23:49:12 +09:00
c5ce18591f
Catch and warn demotion of double in assignments
2020-02-15 23:49:12 +09:00
08ca59d0df
Add tests for double demotion
2020-02-15 23:49:12 +09:00
4bf37b274b
Ensure double is not in zero or param structs for v6
2020-02-15 23:49:12 +09:00
be30a0eb19
Fix missing alignment init on zero and param types
2020-02-15 23:49:12 +09:00
3e651b43f8
Handle aliased values when emitting statements
...
With this, cast address initializers work. I have to wonder if the alias
value short-circuit was legacy from long before the rewrite, as it was
quite trivial to handle in the back-end.
2020-02-15 23:49:12 +09:00
e4eb793fb3
Treat aliased values as constant
...
One step closer to cast address initializers working.
2020-02-15 23:49:12 +09:00
ce9902baed
Don't short-circuit aliased values
...
Not sure why I thought it was a good idea as it turns out this is why
cast pointer initializers were being lost.
2020-02-15 23:49:12 +09:00
3257e7145b
Add failing global init test too
2020-02-15 23:49:12 +09:00
4caa875442
Finish up alignment tests and add address cast
...
It turns out that initializing a local int with a pointer cast doesn't
work.
2020-02-15 23:49:12 +09:00
a4a57b6ffd
Implement aligned allocations
2020-02-15 23:49:12 +09:00
293f10211a
Start on alignment test
...
Currently fails (deliberately, WIP)
2020-02-15 23:49:12 +09:00
0542daacdf
Create more double related tests
...
Including catching warnings :) (yay -Werror)
2020-02-15 23:49:12 +09:00
7e09a94469
Fix "casts" between signed and unsigned int
2020-02-15 23:49:12 +09:00
5d8d805b60
Fix test for single overload functions
...
All functions are stored in the overload functions table, even those
that are never explicitly overloaded, but only explicitly overloaded
functions (those with @overload) use the type-qualified naming.
2020-02-15 23:49:12 +09:00
533fb8acc9
Implement double constants
2020-02-15 23:49:12 +09:00
2cd62fe01b
Fix several double-related bug
...
float is promoted to double through ... for non-v6 code.
PR_Sprintf has custom param access via P_*, messed up doubles.
2020-02-15 23:49:12 +09:00
eb7f825158
Test for full-float %
2020-02-15 23:49:12 +09:00
8920c59515
Find @override functions even when there's only one
2020-02-15 23:49:12 +09:00
df7c08a010
Add support for doubles to Ruamoko
...
Only as scalars, I still need to think about what to do for vectors and
quaternions due to param size issues. Also, doubles are not yet
guaranteed to be correctly aligned.
2020-02-15 23:49:12 +09:00
13b608f40c
Don't truncat float % float
...
This allows full usage, eg, x % pi, but otherwise maintains
compatibility with integer %
2020-02-15 23:49:12 +09:00
16f8dca72e
Align local and far data spaces
...
I plan on adding doubles, and so it's necessary to ensure that attempts
to align doubles in local or far data spaces remain aligned after final
linking.
2020-02-15 23:49:12 +09:00
197f856a30
Fix incorrect scalar/quaternion division
...
It's just not possible.
2020-02-15 23:49:12 +09:00
9248e8cf01
Update for doxygen 1.8.16
2020-02-11 15:22:42 +09:00
9d09d65d9a
Clean up some export stuff
...
While it's not there yet, the eventual goal is to support multiple
models in the one blend file, so per-object settings is a requirement.
2019-08-20 16:52:43 +09:00
c14e5623ad
Get the mdl frames script property working
...
Maybe late 2.78+, blender added the ability to use any ID type in
PointerProperty, which is just what I wanted for the frames script.
2019-08-16 22:34:45 +09:00
839d90981d
Clean up a few last things for blender 2.80
...
Many many thanks to Aleksander Marhall for the initial work.
2019-08-16 22:32:37 +09:00
008b5a4d2c
Import/Export fixes after some Blender 2.8 API changes
2019-08-16 21:38:29 +09:00
11e5f39fdc
Fix for the export error when there is no Shape Keys available
2019-08-16 21:38:29 +09:00
36da80c0d0
Export proper frame names
2019-08-16 21:38:29 +09:00
33d114c09e
Set playback/animation range to start from frame 1
2019-08-16 21:38:29 +09:00
ff5f9f1831
Add "empty" skin during export for models with no materials or texture nodes
2019-08-16 21:38:29 +09:00
3913febace
MDL properties in object panel are back
2019-08-16 21:38:29 +09:00
31fa7a81ef
Fix for single skin being marked as skingroup
2019-08-16 21:38:29 +09:00
94c3c8c2cb
Skingroup image export order based on texture node position (height) in shader editor. Nodes with higher position are exported first.
2019-08-16 21:38:29 +09:00
4f29f57e77
Proper node positioning after import
2019-08-16 21:38:29 +09:00
a13c382969
Skingroup images import order fix
2019-08-16 21:38:29 +09:00
4d1f05186b
Skingroups export added
2019-08-16 21:38:29 +09:00
17e76149a1
Adjust playback range on import, to fit mdl's animation length
2019-08-16 21:38:29 +09:00
66fe862a87
Import skingroups as separate materials with skins as textures
2019-08-16 21:38:29 +09:00
ae86790e6c
Initial Blender 2.8 changes - it works, now it's time for improvements
2019-08-16 21:38:29 +09:00
8112c4e8b6
Moved MDL settings from Object tab to export options/presets
2019-08-16 21:38:29 +09:00
90a707f7f9
Export animation range selected with start/end playback range in Timeline Editor
2019-08-16 21:38:29 +09:00
c3fa78ef4d
Include test for 2d vector expressions
2019-07-06 14:49:28 +09:00
8caf2eb584
Mark some new functions as pure
2019-06-27 21:37:48 +09:00
a5ee58cebb
Support 2d vector expressions
...
[x, y] expands to [x, y, 0] (for now, might add a 2d vector type).
2019-06-18 11:54:45 +09:00
83fac13a0c
Fix debug line numbers for vector expressions
2019-06-18 11:53:58 +09:00
b37c331e76
Catch taking size of null type
...
This should help catch similar errors in the future.
2019-06-18 10:39:17 +09:00
0f1f477e64
Set up temp aliases correctly
...
Fixes vector expressions as sub-expresses. I really don't know why I did
the temp alias setup that way.
2019-06-18 10:38:19 +09:00
fc50376297
Fix a minor error check mistake
2019-06-18 08:54:18 +09:00
fe73547f43
Update alias type sameness check
...
This one seems to be fairly robust. Fixes alias being used to cast
pointers (maybe a better way, but this works for now).
2019-06-18 08:53:05 +09:00
f7825fe7cf
Print types properly in pointer value expressions
2019-06-18 00:22:24 +09:00
b996fb7aa4
Make operand->type actual type instead of low-level
...
And clean up the resulting mess. This fixes struct copy, but uncovers
another bug :/
2019-06-17 23:38:34 +09:00
e0c8285f07
Extend nested struct test to cover struct copy
...
Fails :P
2019-06-17 22:57:40 +09:00
dd52b7fea1
Merge aliased alias expressions
...
This comes up when accessing struct fields nested in another struct.
Fixes the nested struct fields test.
2019-06-17 22:48:42 +09:00
e9c24dbf1c
Test case for accessing nested struct fields
...
Currently fails.
2019-06-17 22:47:44 +09:00
ec128ffeee
Leave a FIXME for daglabel_t.live
2019-06-16 19:24:19 +09:00
d6d3027411
Mark the correct operand as live
...
This fixes vecexpr (and possibly other cases).
2019-06-16 19:21:02 +09:00
6e21c3ae2e
Treat func statements similarly to flow statements
...
func statements need their operands marked live like flow statements do
because usage is more indirect.
2019-06-16 19:20:21 +09:00
db4a7a139e
Use the alias code when making vars live
...
Not sure the live forcing flag is needed anymore (need to test).
2019-06-16 19:17:45 +09:00
2977c145d0
Clean up dag live alias code a little
...
Mainly, this makes it possible to reuse the alias code.
2019-06-16 19:17:01 +09:00
82b334e919
Number dot dumps to help identify order
2019-06-16 19:12:18 +09:00
c40f4194e9
Use tempop_visit_all for flow and dags
...
Fixes t3 of vecexpr, but t2 is broken (lost first assignment).
2019-06-16 16:56:39 +09:00
3c4903245a
Fix some curly space
2019-06-16 16:55:54 +09:00
bc271d8a02
Add tempop_visit_all
...
Works the same as def_visit_all, but for temp operands.
2019-06-16 16:52:49 +09:00
fa69aeef0f
Improve handling of temp aliases
...
This makes all tests pass when not optimizing. More work needs to be
done in dags.
2019-06-12 00:37:02 +09:00
b18a744288
Use offset alias expressions
...
They are not quite working yet as the operand generator does not use the
offset yet.
2019-06-10 23:55:16 +09:00
cc27949a34
Ensure pointer values always have a type
...
The dags generator was creating a pointer value with no type which
caused print_statement to segfault.
2019-06-10 23:52:39 +09:00
ee1f5f9478
Add support for binary alias expressions
...
Doesn't quite work yet.
2019-06-10 23:48:58 +09:00
913b9f52e0
Add an offset alias expression
...
This should make dealing with def elements (vector etc) easier.
2019-06-10 18:13:28 +09:00
efdfc7436f
Move ev_types into types.[ch]
...
No wonder I couldn't find it the other day...
2019-06-10 18:04:25 +09:00
08ab42fc15
Handle vector expressions as sub-expressions
...
Now my little game builds again :)
2019-06-10 15:50:35 +09:00
49ed4310fd
Fix assigning int to enum or enum to int
...
Or float, for v6 progs.
2019-06-10 08:46:40 +09:00
83fb588727
Support vector/quaternion division by float
...
Implemented via v*(1/f) or q*(1/f) to give CSE a chance to optimize the
division if necessary as otherwise the engine would have to divide every
time.
2019-06-10 08:44:36 +09:00
7dc1a0640a
Delay creating actual enum symbols
...
In order to keep enumerator type and enum type the same, the values need
to have their type set after the enum type is finalized, and then the
appropriate symbols created in the parent scope. This fixes the infinite
recursion when assigning an enum value to its own type.
2019-06-10 07:44:23 +09:00
f81484a068
Test compiling enums
...
It turns out the enumerator type and enum type wind up with different
instances of the same type (due to the way type chaining works). This
results in infinite recursion in assign_expr and check_types_compatible.
2019-06-10 07:17:41 +09:00
ef4ad52239
Make binary_expr fold constants
...
This is where constant folding should have happened all along. While
unary_expr should fold constants too, it seems to already try to do so
and it's a bit much of a mess to clean up right now.
2019-06-10 00:36:13 +09:00
bf0543f7fb
Be paranoid about temp var sizes
...
Prevents array index errors.
2019-06-10 00:19:48 +09:00
28a2b96800
Implement automatic array sizing from initializer
2019-06-09 23:11:38 +09:00
00f12263ca
Fix casting arrays to pointers
2019-06-09 22:55:38 +09:00
3196bcc851
Allow fold_constants to handle some math identities
...
Such as x*0 -> 0, x*1 -> x, x+0 ->x, etc.
2019-06-09 22:34:23 +09:00
c98d1ff08c
Document some problems with nil
...
While he's a pretty cool character in HZD, he seems to have some
identity problems when in a compiler.
2019-06-09 22:29:51 +09:00
439cd15620
Do type promotion for short
...
fold_constants used to do this when it was overly aggressive. Fixes the
ICE when accessing an array in a struct via a pointer.
2019-06-09 22:27:53 +09:00
227ac46ffe
Clean up array expression processing
...
And make sure constants are folded since fold_constants is no longer
recursive.
2019-06-09 22:25:55 +09:00
f70801fa52
Add a test for accessing an array in a struct
...
That's accessed via a pointer. Whee. However, at this stage, the problem
is really a promotion issue for short -> integer.
2019-06-09 21:44:58 +09:00
e46a44d9c6
Update TODO
2019-06-09 21:37:46 +09:00
6253623e9b
Remove some weird dereference code
...
It appears to have been an attempt at optimizing dereferences, but it
instead just utterly mangled them.
2019-06-09 21:37:46 +09:00
f49303e774
Emit code for address expressions
...
It helps if the necessary code is actually emitted in the first place.
2019-06-09 21:37:46 +09:00
8cef85e5de
Cast scalar types when necessary on assignment
...
This fixes type promotion errors that broke some tests.
2019-06-09 20:12:50 +09:00
e849c2d1ce
Mark block expr result as rvalue when used as one
...
At least for return. There may be other cases that need it, but this
fixes comma-expr.r (minus int->float promotion issues).
2019-06-09 19:29:21 +09:00
d9c9686288
Make initialized globals non-constant
...
This is for modern code. Traditional code still treats initialized
globals as constant and nosave. This will make a bit of a mess of
modern code that expects traditional behavior.
2019-06-09 19:23:23 +09:00
b7b6a89b91
Simplify fold_constants
...
While this does break automatic type promotion, it does stop
fold_constants recursing through complex expressions: only the top level
expression needs to be folded, and then only if both sides are actually
constant.
2019-06-09 19:20:35 +09:00
2ee36e8b99
Rename qfot_type_t.ty to meta
...
When I did up a ruamoko implementation of a type encoding dumper, I
named the field meta, which feels much better than ty.
2019-06-09 18:10:07 +09:00
f093516962
Move qfo type encoding structs
...
This makes them available to the gamecode VM, and thus the debugger.
2019-06-09 18:07:58 +09:00
7fce68649a
Implement comma expressions
...
Doesn't quite work (attempt to suppress warning for return a = 3, 5;
failed).
2019-06-09 16:56:20 +09:00
689d1ad3ec
Remove some debug prints
2019-06-09 16:45:01 +09:00
33c2eb430f
Stop bison complaining about %expect
...
And make sure warnings get fixed.
2019-06-09 16:19:27 +09:00
1db600548e
Fix for decl test scope issue
2019-06-09 13:55:07 +09:00
ec89149b29
Build for loop decl inits correctly
...
The multiple expressions are chained together and evaluate in left to
right order.
2019-06-09 13:54:03 +09:00
d29ea55826
Add a test for for loop declarations
...
I think the compiler segfaulting indicates a test failure.
2019-06-09 13:02:36 +09:00
625e767684
Update qf dependency stuff for recent automake
...
Not sure just what version of automake broke things, but I do remember
having a bad time getting the dependencies to work in the first place.
At least now they should be more reliable (until automake changes
things).
2019-06-09 13:00:55 +09:00
5bd3c8c985
Allow variable declarations in for loops
...
Matches C, C++ and C#
2019-06-08 21:52:17 +09:00
be4c3a82a2
Detect redefinitions of structs and unions
...
While the redefinition was being detected, it was misreported as the tag
being wrong, and on the wrong line, too.
2019-06-08 19:25:25 +09:00
f2102187b9
Clean up emit_protocol_list
...
Now it uses emit_structure, which will make it easier should things go
wrong.
2019-06-08 19:23:49 +09:00
6acbdcb3e5
Correct a typo
2019-06-08 19:23:10 +09:00
df0316bb13
Clean up some comment FIXMEs
2019-06-08 19:22:29 +09:00
e217bdd55f
Assign vector expressions to a temp for return
...
Rather than direct to .return, allowing return of vector expressions
with function calls without trashing the vector.
2019-06-08 13:36:03 +09:00
ebab6574b9
Allow taking the address of a temporary variable
2019-06-08 13:35:47 +09:00
e27da9c7db
Avoid folding constants in assign_expr
...
This prevents fold_constants from complaining about attempts to take
the address of a vector.
2019-06-08 13:31:30 +09:00
398e1cf83a
Use output file as base for dot files
...
Ensures the output goes not go into the source directory when using a
separate build tree.
2019-06-08 13:30:06 +09:00
3327127e36
Fix a typo in a comment
2019-06-08 13:28:29 +09:00
4ef4a7e955
Start work on simple names in symbol tables
...
I don't remember what the goal was (stopped working on it eight months
ago), but some possibilities include:
- better handling of nil (have trouble with assigning into struts)
- automatic forward declarations ala C# and jai (I was watching vids
about jai at the time)
- something for pascal
- simply that the default symbol type should not be var (in which case,
goal accomplished)
2019-06-06 07:01:44 +09:00
7e31704ebe
Be more informative for pass by value size errors
2019-06-06 06:46:38 +09:00
a5da8da1dc
Be more informative for symbol type internal errors
2019-06-06 06:45:48 +09:00
d00a2ef82d
Convert constant vector expressions in binary expressions
...
Fixes not being able to compare with eg [1, 2, 3]
2019-06-06 06:44:19 +09:00
319ecfbbfa
Create a usable dot for vector expressions
...
It leaves a lot to be desired, but it beats "FIXME".
2019-06-06 06:42:31 +09:00
607ec3124e
More dead code removal.
...
I have no idea why this hung around. It appears to be old array/struct
initialization.
2018-10-15 22:12:40 +09:00
ff5c1cd4c2
Clean up redundant type parameter to initialize_def.
2018-10-15 22:11:20 +09:00
1746ce3af5
Make int type name reflect its keyword.
...
Makes qc error messages make more sense.
2018-10-15 20:54:05 +09:00
302934a1e4
Remove some dead code.
2018-10-15 20:52:46 +09:00
2a2754e581
More ways to break qfcc :/
2018-10-15 19:48:19 +09:00
6c53be928b
Support returning vector expressions.
...
Currently fails due to a scheduling problem when the vector expression
contains functions.
2018-10-13 23:32:53 +09:00
aaab18c9a7
Nuke a dead field.
...
It looks like ex_temp_t.expr was never really used, and I have no idea
what I had intended (not surprising after ~17 years).
2018-10-13 12:07:50 +09:00
2f2edae43b
Make error, warning and notice give compiler file/line too.
...
But only when verbosity > 1 (-vv).
2018-10-13 09:31:00 +09:00
4828934e50
Fix a equality check typo.
...
Fixes bad code generation when assigning to vector.x in a function with no
parameters and the vector is the first local var.
2018-10-12 23:39:05 +09:00
78e0a8dc52
Support assigning non-constant vector expressions.
2018-10-12 22:05:17 +09:00
e03dfa29b4
Clean up the module .ctor construction a little.
2018-10-12 21:09:36 +09:00
f56d25cda6
Don't double print messages.
2018-10-12 20:46:45 +09:00
668896aa82
More diagnostic improvements.
2018-10-12 16:01:11 +09:00
71b3d30aa1
Make the ice a little more informative.
2018-10-12 14:51:39 +09:00
f58c2fef5a
Switch QF's quaterions from wxyz to xyzw.
...
After messing with SIMD stuff for a little, I think I now understand why
the industry went with xyzw instead of the mathematical wxyz. Anyway, this
will make for less pain in the future (assuming I got everything).
2018-10-12 14:33:55 +09:00
b7aa5bb8fe
Fix an oops messing up edict size.
2018-10-12 14:33:26 +09:00
f8606ebab8
Fix a object file segfault when spaces have no data.
...
ie, don't attempt to copy memory from a null pointer (just copy the null
pointer).
2018-10-12 14:00:34 +09:00
3bd9f96b07
Make internal error messages a little more helpful.
...
They now include the file and line of the compiler as well.
2018-10-12 13:57:59 +09:00
3656cb966c
Fix some bison deprecation warnings.
2018-10-12 09:32:49 +09:00
3bc664811e
Fix lost error count tracking.
2018-10-11 22:03:14 +09:00
a096babb1f
Duplicate memory when creating a qfo from progs.
...
Fixes some double-free bugs when linking after compiling.
2018-10-11 21:37:01 +09:00
f673ab9c3d
Fix misreporting of failed progs load.
2018-10-11 13:39:48 +09:00
82809eb856
Simplify the call to PR_LoadProgsFile.
...
I've decided that setting pr.max_edicts and pr.zone_size as part of the
local progs initialization rather than in PR_LoadProgsFile makes more
sense. For one, it is unlikely for the limits to change every time progs is
reloaded. Also, they seem to be a property of the VM rather than the progs.
However, there is nothing stopping the caller from updating max_edicts and
zone_size every call.
2018-10-11 10:06:48 +09:00
673b45be05
Add diagnostic hooks.
2018-10-09 12:46:13 +09:00
d95c01be1a
More defspace tests.
2018-10-09 12:45:26 +09:00
34bcf7faab
Do a pure/const/noreturn/format attribute pass.
...
I always wanted these, but as gcc now provides warnings for functions that
could do with such attributes, finding all the functions is much easier.
2018-10-09 12:42:21 +09:00
05fd2d8cdd
Start work on a unit test for defspace.
2018-09-09 22:37:09 +09:00
3a4667318c
Fix the qfcc test autodependencies.
2018-09-09 22:36:20 +09:00
d1cb15664f
Fix some spelling issues in a comment.
2018-09-09 22:34:24 +09:00
71569b45c9
Pre-format diagnostic messages.
...
This will make it easier to hook diagnostics enabling unit tests.
2018-09-09 18:09:32 +09:00
113e0d82ba
Fix a signed comparison error.
...
Not sure how this fix got lost.
2018-09-09 16:48:19 +09:00
aebd9288cd
Force thread count to 1 when pthreads is unavailable.
...
Don't want the thread count being misreported.
2018-09-09 13:41:22 +09:00
fa1514798b
Print the number of threads used by qfvis.
2018-09-09 13:41:00 +09:00
06ab36de3d
Slight cleanup of winding allocation.
...
It seems gcc doesn't care if the & is present when calculating field
offsets, but it not being there bothered me very much and might as well use
our "standard" macro anyway.
2018-09-09 13:38:32 +09:00
59e85b7d5e
Fix a pile of dead assignments.
...
Some were actual bugs!
2018-09-08 22:23:57 +09:00
59b04dbcca
Make many qfo int fields unsigned.
...
I don't remember why I thought signed was a good idea, but unsigned does
make more sense for most fields.
2018-09-08 19:45:19 +09:00
f75ff2d269
Fix some null argument issues.
2018-09-07 22:30:44 +09:00
8f028b8a0d
Fix some double frees.
2018-09-07 20:03:19 +09:00
8fd5be0ee0
Fix a pile of sizeof goofs.
...
While scan-build wasn't what I was looking for, it has proven useful
anyway: many of the sizeof errors were just noise, but a few were actual
bugs (allocating too much or too little memory).
2018-09-07 20:00:57 +09:00
b37b0180cb
Actually do a test for triangle.r.
...
If the kahan triangle area method breaks, I did something wrong with qfcc's
handling of parentheses (ie, floating point math is not truly associative).
2018-08-23 20:52:32 +09:00
1b9149ef31
Fix an argument passing error in test-harness.
...
It failed to pass the last argument to the ruamoko program being run.
2018-08-23 20:22:29 +09:00
2fcda44ab0
Kill dag leaf nodes on assignment.
...
Fixes the failing swap test caused by a's original value being used (via t)
after being written.
2018-08-23 20:07:22 +09:00
cb366dd825
Better flow dot dump "scripting".
...
Not sure I want to keep the statements in the live dump, but the system
is more flexible now.
2018-08-23 20:07:11 +09:00
cdbdf3f0eb
Make print_operand global.
2018-08-23 20:05:16 +09:00
8b8b42bde2
Fix a typo in a comment.
2018-08-23 20:04:51 +09:00
f250065003
Add tests for swapping vars and triangle area.
...
Triangle area was meant just to check Kahan's equation, but it found a
problem with swapping vars. swap.r currently fails.
2018-08-21 16:27:22 +09:00
0f279cd3f0
Add a comment about fake statement numbers.
...
I had forgotten how they worked (not surprising after 6 years).
2018-08-21 15:57:51 +09:00
1b84ea747c
Dump statements during flow analysis.
2018-08-21 15:56:39 +09:00
5ba13d759b
Fix a FIXME.
...
Set of everything got added shortly after that FIXME was added, but I
forgot to fix it.
2018-08-21 15:56:03 +09:00
fbdf560a7a
Rename print_extra_live to print_flow_vars.
2018-08-21 15:55:11 +09:00
5dbb7b6d4d
Handle all modelgen commands.
...
Just in case someone ever adds models.qc to progs.src.
2018-08-20 00:11:02 +09:00
027e76b4c9
Add a doc block to parse_cpp_line.
...
It took me too long to figure it out :P
2018-08-20 00:09:20 +09:00
4f58429137
Fix an unhealthy pile of gcc 8 warnings.
...
While some of the warnings were merely annoying, some where actual bugs or
unearthed bugs in related code.
2018-08-20 00:05:00 +09:00
81202d7b29
Add a shutdown builtin to qwaq.
2018-08-19 19:44:45 +09:00
b795beb520
Add support for \? to qc strings.
...
And fix up the man page a bit.
2018-08-19 19:31:52 +09:00
0de0eb2fc7
Fix a bunch of bit-rot.
2018-06-08 13:21:30 +09:00
f522e58d53
Fix some blender warnings.
2018-06-08 12:45:57 +09:00
2329fb1885
Fix some warnings and bitrot.
2016-10-25 00:10:30 +09:00
932647bad3
Fix a segfault with switch on an undefined symbol.
...
Forgot to check if the test expression was not an error.
2016-01-10 13:41:49 +09:00
bf40f0126e
Convert vector expressions when passing parameters.
...
Calling a function with [vector stuff] caused an ICE.
2016-01-08 21:16:37 +09:00
1061256c4f
Add a missing expression type to dot_expr.
2016-01-08 21:15:47 +09:00
431074d58f
Update label destination when moving labels.
...
This fixes a segfault when optimizing the empty-body test. The label was
getting moved, but the statement block to which it pointed was not updated
and thus it pointed to dead data.
2016-01-07 21:24:29 +09:00
d54ee6c8c3
Do not promote debug messages to warnings.
...
I think I copied the notice code for debug and forgot to edit the
promotion. Oops.
2016-01-07 20:08:07 +09:00
6e65de251c
Put qwaq's epoch back to 0.
...
4Gs causes a few problems at this stage. I really need to look into putting
long and double into qc.
2016-01-04 20:42:57 +09:00
735fcf68d5
Clean up usage of va_copy.
...
AC_TYPE_VA_LIST is no longer necessary, and the code is easier to read.
2016-01-03 21:16:23 +09:00
a54f3d9035
Shutdown some subsystems on exit.
...
This fixes the problem with X keyboard repeat not being restored.
2016-01-03 17:39:03 +09:00
c71eccfb10
Remove MAX_THREADS.
...
This fixes a buffer overflow with more than 4 threads.
2015-08-14 10:57:51 +09:00
2cd59c962c
Rename qfcc's basename to file_basename.
...
Avoids naming conflicts with JohnnyonFlame's toolchain.
2014-01-31 12:22:11 +09:00
3efb0c538f
Separate file search from loading.
...
QFS_LoadFile (and its wrappers) now take a file handle rather than a
path. This will make vpath usage a little cleaner to implement.
2014-01-28 16:22:05 +09:00
6c0f271783
Fix some more automake warnings.
2013-11-24 13:26:12 +09:00
f5501fbf24
Fix a pile of automake deprecation warnings.
...
s/INCLUDES/AM_CPPFLAGS/g
I <3 sed :)
2013-11-24 13:11:50 +09:00
6996e7295a
Add a getLine method to Script.
...
It returns the rest of the line (minus // style comments) as the token. I
needed it in another project but this is my central repository for
script.py.
2013-11-24 11:21:40 +09:00
eeaf2bc43f
Return None instead of False from parse_entity.
...
Don't remember how I found that one, but None makes much more sense.
2013-11-24 11:19:50 +09:00
c88d1987a1
Add warnings for empty blocks in if/else/do.
...
Saw a discussion of such in #qc and that gcc implemented it. I realized it
would be pretty easy to detect and very useful (I've made such mistakes at
times).
2013-09-30 18:09:04 +09:00
db460155e9
Fix multiple function calls in expressions.
2013-09-27 23:15:57 +09:00
70d18ecfa1
Add a test for functions in binary expressions.
...
As expected, the binary_expr() cleanup broke multiple function calls in a
single expression.
2013-09-27 23:15:57 +09:00
4bc40b3917
Rewrite binary_expr().
...
It is now in its own file and uses table lookups to check for valid type
and operator combinations, and also the resulting type of the expression.
This probably breaks multiple function calls in the one expression.
2013-09-27 23:15:57 +09:00
ec98954dfc
Remove the small struct optimization.
...
It might cause trouble for the binary_expr() cleanup.
2013-09-27 23:15:57 +09:00
3295370328
Expose and use field_expr().
...
This may cause problems later on, but it is needed for the binary_expr()
cleanup.
2013-09-27 23:15:57 +09:00
ca9693d9cd
Expose convert_from_bool().
2013-09-27 23:15:57 +09:00
b66cbeeb57
Run fold_constants on comparison operands.
...
This is a bit of a workaround to ensure the operands have their types
setup correctly. Really, binary_expr needs to handle expression types
properly.
This fixes the bogus error for comparing the result of pointer subtraction
with an integer.
2013-09-27 23:15:57 +09:00
c13f2c2c23
Fix the support for [s, v] for quaterions.
2013-09-27 23:15:57 +09:00
7f50333b2c
Add support for constant [] vector expressions.
2013-09-27 23:15:57 +09:00
7a7a685105
Add support for actual vector expressions.
...
Currently, they can represent either vectors or quaternions, and the
quaternions can be in either [s, v] form or [w, x, y, z] form.
Many things will not actual work yet as the vector expression needs to be
converted into the appropriate form for assigning the elements to the
components of the "vector" type.
2013-09-27 23:15:57 +09:00
2572811bf4
Add support for quaternion.w/x/y/z.
...
It's sometimes more useful to have direct access to each individual
component of the imaginary part of the quaternion, and then for
consistency, alias w and s.
2013-09-27 23:15:57 +09:00
a6cdc8735a
Start implementing vec = [x,y,z].
...
This is a nice feature found in fteqcc (also a bit of a challenge from
Spike). Getting bison to accept the new expression required rewriting the
state expression grammar, so this is mostly for the state expression. A
test to ensure the state expression doesn't break is included.
2013-09-27 23:15:56 +09:00
35de42aeac
Catch bogus comparison types.
...
This fixes an ICE when trying to compare a function with a float.
2013-09-27 23:15:56 +09:00
03fb47d7ba
Fix some sizeof goofs.
...
Yay for continually improving gcc warnings :)
2013-09-26 17:45:30 +09:00
16d899ace5
Remove the string:string concatentation TODO item.
...
I think it may have been for compatibility with a certain qcc variant (no
idea which one, though). While the shift/reduce conflict is fixable using
"%prec IFX" on the const:string rule, the colon breaks test?"a":"b".
Putting parentheses around "a" allows such a construct, requiring them
breaks comatibility with C. I think this feature just isn't worth that.
2013-06-26 23:04:35 +09:00
257597fd38
Update ifstring in TODO
...
It seems to be done, but...
2013-06-26 23:04:35 +09:00
cef2136050
Implement "not" logic for while and do-while.
...
This goes towards complementing the "if not" logic extension. I need to
check if fteqcc supports "not" with "while" (the version I have access to
at the moment does not), and also whether it would be good to support
"not" with "for", and if so, what form the syntax should take.
2013-06-26 23:04:35 +09:00
45a09e195e
Add more TODO items (strings stuff)
2013-06-26 23:04:35 +09:00
6fa63f6c54
Implement the basics of fteqcc's -Fifstring.
...
It's -Cifstring in qfcc, but otherwise the intent is the same. More testing
and investigation into what fteqcc does is required.
2013-06-26 23:02:55 +09:00
8d34d33055
Implement "if not (foo)" syntax.
...
It is syntactic sugar for if (!(foo)), but is useful for avoiding
inconsistencies between such things as if (string) and if (!string), even
though qcc can't parse if not (string). It also makes for easier to read
code when the logic in the condition is complex.
2013-06-26 23:02:55 +09:00
77337e2402
Separate "" and nil.
...
It turns out this is required for compatibility with qcc (and C, really).
Once string to boolean conversions are sorted out completely (not that
simple as qcc is inconsistent with if (string) vs if (!string)), Qgets can
be implemented :)
2013-06-26 23:02:55 +09:00
fd1ea9e00e
Add an option to control the enum switch warning.
2013-06-26 09:29:37 +09:00
29df4ac7ee
Fix case label lookup.
...
It looks like I had forgotten that the compare function is supposed to
return true/false (unlike memcmp's sorting ability). Also, avoid the
pointers in the value struct as they can change without notice.
Using enums in switches now works nicely, including warnings for unused
enum values.
2013-06-26 09:29:31 +09:00
8ac2c3a04d
Check for unused enum values in switch statements.
...
Unfortunately, it turns out the value lookup is broken (including for
duplicate cases).
2013-06-26 09:25:45 +09:00
35bc981402
Allow use of a jump table when switching on an enum.
2013-06-26 09:25:45 +09:00
64f76bd762
Use the right value when getting a function's defref.
...
Either I had gotten confused while writing the code and mixed up line and
offset, or I had changed offset to line at one stage but missed a place.
This fixes the segfault when compiling chewed-alias.r and return-ivar.r
2013-06-24 09:16:55 +09:00
125ef1f0ff
Move the whole separator test/creation into a function.
...
This will make the next stage easier. (except that seems to be slower)
2013-03-19 20:39:01 +09:00
f2452eb3c3
Rewrite the inner-loop of FindSeparators.
...
For the most part, it's just refactoring the code so the plane creation and
testing are in separate functions, but there is one important difference:
the plane test now checks only the two points on either side of the point
used to create the plane.
Because the portal winding is guaranteed to be convex and planar, if both
points are on the plane, all points are, and if neither point is behind the
plane, no points are.a
This shaved about 5 seconds off the level 4 run using 4 threads (~198s to
~193s) and about 12s from the single threaded run (~682s to ~670s (hmm,
gained some time in recent changes)).
2013-03-19 17:00:00 +09:00
d7c1bc8d02
Correct a comment.
...
I had gotten confused between figuring out the windings and writing the
comments, I guess.
2013-03-19 16:23:47 +09:00
8938870e46
Make the default output a little nicer.
2013-03-19 13:07:44 +09:00
dff0b89a6c
Detect the number of CPUs available.
...
Now qfvis will default to multi-threaded on multi-core machines.
2013-03-19 12:05:50 +09:00
88e5adcec6
Make the base vis multi-threaded.
...
Now multi-threaded qfvis is on par with tyrutils vis (differences usually
<1s, sometimes more, sometimes less).
2013-03-19 11:42:09 +09:00
32b6d15931
Use a sorted queue for portals.
...
qsort is used to sort the queue by nummightsee. At ~4ms for 20k portals, I
think it's affordable. Using a queue rather than scanning the portal list
each time loses the dynamic sorting when mightsee gets updated, but it
seemed to shave off 4s anyway (~207s to ~203s (maybe, yay random times)).
Another step towards threaded base-vis.
2013-03-18 21:14:12 +09:00
7e40981dcd
Move the LeafThread setup to its own generic function.
...
This is for threading base-vis.
2013-03-18 21:11:46 +09:00
cb096c601d
Use a per-portal rwlock for portal updates.
...
This should make qfvis scale a little better with cpu count.
2013-03-18 15:03:11 +09:00
c824e668ed
Rework some of the pthread stuff.
...
Init/uninit is now separate from portal vising.
The global lock has a better name and is now a rwlock.
Use a separate lock for the stats.
2013-03-18 14:26:52 +09:00
134381f79b
Reduce the locking in the portal completion code.
...
It doesn't seem to make much difference, but the less room for contention,
the better.
2013-03-18 13:45:19 +09:00
ffb6d628bd
Simplify the pthreads detection macros.
2013-03-18 13:31:35 +09:00
1c20a49dba
Use the recursive set allocator for mightsee.
...
This completely removes the lock used to protect the set allocation code
while keeping the use of the set api clean.
2013-03-18 13:30:50 +09:00
a28ec8aa82
Revert "Allocate stack blocks and mightsee in one block."
...
This reverts commit 1ea79e8626
2013-03-18 12:47:59 +09:00
ad247fa12d
Rename some variables and remove some comments.
...
The old variable names were confusing ("target" winding comes from
"portal"?), and the comments were from when I really didn't understand
concepts like separating planes. While they weren't wrong, they were quite
inadequate and I want to write new ones.
2013-03-17 21:52:08 +09:00
ccc432a7ea
Give the fields of pstack_t clearer names.
...
And some comments.
2013-03-17 19:18:38 +09:00
1ea79e8626
Allocate stack blocks and mightsee in one block.
...
This bypasses set_new, but completely removes the use of the global lock
from within RecursiveClusterFlow. This seems to give a small speedup: 203
seconds threaded.
2013-03-17 16:37:27 +09:00
1d262f7dea
Clean up FindSeparators a little bit.
...
This was testing an idea I had to remove the plane flips. It seems to have
been good for the initial plane orientation, but was a slight slowdown for
the pass-portal test. However, this makes the code a little easier to work
with for my idea on improving the algorithm itself.
2013-03-17 10:16:47 +09:00
5dba419233
Cache stack blocks and working mightsee sets.
...
Since the stack structure in the thread data is a linked list, move the
stack blocks off the program stack and into malloced memory. More
importantly, when the stack block is allocated, the mightsee working set is
allocated too, and as neither are freed, this greatly reduces contention
for the lock. Also, because the memory is kept, single threaded time for
gmsp3v2 dropped from 695s to 670s. Threaded is now about 207s (down from
350).
2013-03-16 22:58:59 +09:00
2ea143283c
Rewrite mightsee_more to manipulate the sets directly.
...
While using set operators was clearer, it was rather expensive (about 25s
for gmsp3v2). qfvis now completes the map in about 695s (single threaded).
About 15s faster than tyr for the same conditions (1 thread, level 4).
2013-03-16 21:51:41 +09:00
195bdcb92f
Rework FindSeparators to make use of the winding direction.
...
This is the second part of the separator search optimization from tyrutils
vis. With this, qfvis is getting close to tyrutils vis when
running single threaded (qfvis is suffering some nasty thread contention
and thus can't get below about 350 seconds with 4 threads). 808s vs 707s.
2013-03-15 22:05:01 +09:00
9b10304c2f
Make CopyWinding const-correct.
2013-03-15 19:25:24 +09:00
5a2ee06787
Reverse the winding for backside portals.
...
This is part 1 of another optimization from tyrutils vis. It seems that
just reversing the winding gives a tiny speedup.
2013-03-15 19:22:57 +09:00
46d41ad9ac
Split up separator finding and winding clipping.
...
Interesting, it makes very little (maybe faster) difference to find all the
separators for levels 3 and 4. This might be due to the higher levels using
most of the planes to fully clip source away. Anyway, it makes the code a
little clearer (one function, one task).
2013-03-15 16:00:39 +09:00
f80ae52828
Make vis's ClipWinding const-correct.
2013-03-15 15:28:25 +09:00
77c858060d
Add a bunch more statistics.
...
Now I know why sphere culling was a loss: 78% of all tested target portals
were trimmed by ClipToSeparators (50% eventually clipped away entirely).
2013-03-14 19:43:46 +09:00
97da7fe31d
Document some fields.
2013-03-14 19:43:46 +09:00
8032d1d4d1
Split out the mightsee intersection/subset tests.
...
Having the code in separate functions makes the flow in the main loop a
little easier to follow.
2013-03-14 19:43:46 +09:00
eec87bd61b
Remove thread from stack_t.
...
It really wasn't gaining anything and made reading the code a little
harder.
2013-03-14 19:43:46 +09:00
057a5cc624
Make BasePortalVis another 17% faster.
...
I had forgotten to skip the refined tests when the sphere was entirely on
the relevant side of the plane. Now BasePortalVis for gmsp3v2 takes 11s on
my machine (it was 13 with the previous optimization and 15.9 before that).
Also, write some comments describing how BasePortalVis works.
2013-03-14 14:01:26 +09:00
5d6df082f2
Move the vis stats vars into thread data.
...
This should make the stats more reliable when running multi-threaded
(chains is still random, but it seems there are set access issues).
2013-03-14 12:52:40 +09:00
0cae04d71a
Alter test_sphere to return -1, 0 or 1.
...
Representing the side of the plane on which the sphere lies is much more
useful as more complicated tests can be done using just the one call.
-1: the sphere is entirely on the back side of the plane
0: the sphere is intersecting the plane
1: the sphere is entirely on the front side of the plane
2013-03-14 09:05:26 +09:00
b9d71218f6
Use sphere culling in the base vis.
...
Base vis was done first for testing. Optimized base vis is down to ~12.4s
from ~16s (29% faster?).
2013-03-13 21:32:18 +09:00
fe55bb678e
Set the default level to 4.
...
It was supposed to be 2, but for some reason I neglected to set it when I
set up the options parsing. However, level 4 is the standard for production
maps, and it happens to be faster than level 2 (at least for gmsp3v2.bsp)
2013-03-09 15:01:54 +09:00
3232c22795
Implement mightsee updates from tyrutils.
...
It doesn't seem to make much difference, but I suspect that's my set code
not being optimal for the job.
2013-03-09 14:32:51 +09:00
d1e65257b6
Implement the cached separators idea from tyrutils.
...
I think the reason I didn't think of that when I tried to improve qfvis's
performance many years ago is I just simply did not understand
ClipToSeparators. However, the difference caching the separators makes is
phenomenal. Before the change, single threaded qfvis would get stuck on one
particular portal for at least a day (I gave up waiting), but now even a
debug build will complete gmsp3v2.bsp in less than 12 minutes (4 threads on
my quad-core). And that's at level 2! Getting stuck for a day was at level
0.
2013-03-08 22:20:29 +09:00
ca0b03687f
Change the naming of ALLOC's free-list.
...
Rather than prefixing free_ to the supplied name, suffix _freelist to the
supplied name. The biggest advantage of this is it allows the free-list to
be a structure member. It also cleans up the name-space a little.
2013-03-08 22:16:31 +09:00
dbdfdb6d28
Add support for PRT2 portal files.
...
These seem to be identical to PRT1-AM but with a different count order in
the header. Taken from tyrutils-0.5.
2013-03-07 18:51:32 +09:00
0649fd7022
Correct some verbosity levels.
2013-03-07 18:51:23 +09:00
4e87072d93
Fix a bunch of whitespace.
2013-03-07 14:31:00 +09:00
4ccb4a4667
Make the test and vis set pointers const.
2013-03-07 14:28:15 +09:00
281c5f185d
Correct the "can see anything new" check.
...
I had the super-set and sub-set parameters backwards, but that's because I
had misinterpreted the old code.
2013-03-07 14:24:30 +09:00
81e4d3072c
Plug a bunch of memory leaks.
...
Not that they really matter, but it makes reading valgrind's output a
little easier.
2013-03-07 13:42:19 +09:00
9b4beaf3f8
Use a set for portalsee.
...
This removes the hard coding of the size of portalsee, which looks to have
been too small for some maps.
2013-03-07 11:35:08 +09:00
299ff8f575
Use set functions for qfvis.
...
While noticeably slower than the previous expanded set manipulation code,
this is much easier to read. I can worry about optimizing the set code when
I get qfvis behaving better.
2013-03-07 11:06:55 +09:00
5950803462
Implement the sun lighting calculations.
...
The sun(s) won't work right now because the vectors aren't properly set up.
2013-03-07 09:11:23 +09:00
4d4bcaf3a3
Make the Test functions const-correct.
2013-03-07 09:11:23 +09:00
0e82b37001
Parse the sun information from the worldspawn entity.
2013-03-07 09:11:23 +09:00
f1a18742f1
Cache the worldspawn entity.
...
It is an error (for now) if there is not exactly one worldspawn entity.
2013-03-07 09:11:23 +09:00
4f0a1df3c1
Implement sky tracing.
...
Again, this is from Jardrup's light tool. Sun data is not yet parsed, nor
are any sun lighting calculations done.
2013-03-07 09:11:23 +09:00
7cd1894a6d
Add the solid-sky option from Jardrup's light tool.
...
However, actual sky tracing is not yet implemented.
2013-03-07 09:11:23 +09:00
32127d3ec7
Kill a sprintf.
...
While 16 chars is enough for any 32-bit int, if int ever changes to 64
bit... (need 21). Just avoid the issue with good old va :)
2013-03-07 09:11:23 +09:00
Bill Currie
khreathor