mesa.git
14 years agoMerge remote branch 'cworth/master'
Eric Anholt [Thu, 24 Jun 2010 22:13:03 +0000 (15:13 -0700)]
Merge remote branch 'cworth/master'

Conflicts:
ast_to_hir.cpp
ir.cpp

This brings in the talloc-based memory management work, so that the
compiler (almost) no longer leaks memory.

14 years agoMake inlined function variables auto, not in/out.
Eric Anholt [Thu, 24 Jun 2010 22:03:05 +0000 (15:03 -0700)]
Make inlined function variables auto, not in/out.

14 years agoAttach a pointer to variable names in LIR dumping.
Eric Anholt [Thu, 24 Jun 2010 16:07:38 +0000 (09:07 -0700)]
Attach a pointer to variable names in LIR dumping.

Since variable names are not unique, and we like to make lots of
__retvals and assignment_tmps and a,b,c,d this helps in debugging.

14 years agoQuiet unused arg warning for ir_constant cloning.
Eric Anholt [Wed, 23 Jun 2010 23:43:08 +0000 (16:43 -0700)]
Quiet unused arg warning for ir_constant cloning.

14 years agoMove ir_constant cloning alongside the other cloning functions.
Eric Anholt [Wed, 23 Jun 2010 23:42:37 +0000 (16:42 -0700)]
Move ir_constant cloning alongside the other cloning functions.

14 years agoDon't forget to add the declaration of our temporary variable for assigns.
Eric Anholt [Thu, 24 Jun 2010 16:06:12 +0000 (09:06 -0700)]
Don't forget to add the declaration of our temporary variable for assigns.

Otherwise, dead code elimination gets confused since it relies on
seeing decls.

14 years agoir_function_inlining: Re-add the "s/return/retval =/" functionality.
Eric Anholt [Thu, 24 Jun 2010 15:59:57 +0000 (08:59 -0700)]
ir_function_inlining: Re-add the "s/return/retval =/" functionality.

I ripped it out with the cloning changes yesterday, and should have
tested and noticed that there were now returns all over.

14 years agoFix variable remapping in function cloning.
Eric Anholt [Thu, 24 Jun 2010 20:31:34 +0000 (13:31 -0700)]
Fix variable remapping in function cloning.

It's (ht, data, key) not (ht, key, data).

14 years agoglsl2 main: Switch from realloc to talloc_realloc to construct program source.
Carl Worth [Thu, 24 Jun 2010 02:09:56 +0000 (19:09 -0700)]
glsl2 main: Switch from realloc to talloc_realloc to construct program source.

This closes 1 leak in the glsl-orangebook-ch06-bump.frag test leaving
4 to go, (all of which are inside hash_table.c).

14 years agoglsl_type: Add a talloc-based new
Carl Worth [Thu, 24 Jun 2010 02:04:45 +0000 (19:04 -0700)]
glsl_type: Add a talloc-based new

And hook it up at the two sites it's called.

Note that with this change we still don't use glsl_type* objects as
talloc contexts, (see things like get_array_instance that accept both
a talloc 'ctx' as well as a glsl_type*). The reason for this is that
the code is still using many instance of glsl_type objects not created
with new.

This closes 3 leaks in the glsl-orangebook-ch06-bump.frag test:

total heap usage: 55,623 allocs, 55,618

Leaving only 5 leaks to go.

14 years agoClose memory leaks in glsl_type (constructor and get_array_instance)
Carl Worth [Sat, 19 Jun 2010 00:52:59 +0000 (17:52 -0700)]
Close memory leaks in glsl_type (constructor and get_array_instance)

Add a talloc ctx to both get_array_instance and the glsl_type
constructor in order to be able to call talloc_size instead of
malloc.

This fix now makes glsl-orangebook-ch06-bump.frag 99.99% leak free:

total heap usage: 55,623 allocs, 55,615

Only 8 missing frees now.

14 years agoClose memory leak in lexer.
Carl Worth [Sat, 19 Jun 2010 00:43:40 +0000 (17:43 -0700)]
Close memory leak in lexer.

Simply call talloc_strdup rather than strdup, (using the talloc_parent
of our 'state' object, (known here as yyextra).

This fix now makes glsl-orangebook-ch06-bump.frag 99.97% leak free:

total heap usage: 55,623 allocs, 55,609 frees

Only 14 missing frees now.

14 years agomain: Close memory leak of shader string from load_text_file.
Carl Worth [Sat, 19 Jun 2010 00:37:02 +0000 (17:37 -0700)]
main: Close memory leak of shader string from load_text_file.

Could have just added a call to free() to main, but since we're using
talloc everywhere else, we might as well just use it here too. So pass
a new 'ctx' argument to load_text_file.

This removes a single memory leak from all invocations of the
standalone glsl compiler.

14 years agos_symbol: Close memory leak of symbol name.
Carl Worth [Thu, 24 Jun 2010 01:30:55 +0000 (18:30 -0700)]
s_symbol: Close memory leak of symbol name.

Easily done now that s_expression is allocated with talloc. Simply
switch from new to talloc_strdup and the job is done.

This closes the great majority (11263) of the remaining leaks in the
glsl-orangebook-ch06-bump.frag test:

total heap usage: 55,623 allocs, 55,546 frees
(was 44,283 frees)

This test is now 99.86% leak-free.

14 years agoClose memory leak in ir_call::get_error_instruction.
Carl Worth [Thu, 24 Jun 2010 01:25:04 +0000 (18:25 -0700)]
Close memory leak in ir_call::get_error_instruction.

By propagating a 'ctx' parameter through these calls.

This fix happens to have no impact on glsl-orangebook-ch06-bump.frag,
(since it doesn't trigger any errors).

14 years agoClose memory leaks from generate_constructor_intro
Carl Worth [Thu, 24 Jun 2010 01:19:46 +0000 (18:19 -0700)]
Close memory leaks from generate_constructor_intro

By simply propagating a 'ctx' parameter through these function
calls. (We do this because these function are otherwise only receiving
an exec_list, which is not a valid talloc context.)

This closes 1611 leaks in the glsl-orangebook-ch06-bump.frag test:

total heap usage: 55,623 allocs, 44,283 frees
(was 42,672 frees)

14 years agoexec_node: Add new talloc-based new()
Carl Worth [Thu, 24 Jun 2010 01:11:51 +0000 (18:11 -0700)]
exec_node: Add new talloc-based new()

And fix all callers to use the tallbac-based new for exec_node
construction. We make ready use of talloc_parent in order to get
valid, (and appropriate) talloc owners for everything we construct
without having to add new 'ctx' parameters up and down all the call
trees.

This closes the majority of the memory leaks in the
glsl-orangebook-ch06-bump.frag test:

total heap usage: 55,623 allocs, 42,672 frees
(was 14,533 frees)

Now 76.7% leak-free. Woo-hoo!

14 years agoast_node: Add new talloc-based new()
Carl Worth [Thu, 24 Jun 2010 00:12:11 +0000 (17:12 -0700)]
ast_node: Add new talloc-based new()

And use the talloc-based new for all of the ast objects created by the
parser.  This closes a lot of memory leaks, and will allow us to use
these ast objects as talloc parents in the future, (for things like
exec_nodes, etc.).

This closes 164 leaks in the glsl-orangebook-ch06-bump.frag test:

total heap usage: 55,623 allocs, 14,553 frees
(was 14,389 frees)

14 years agoexec_node: Remove destructor from exec_node and all descendants.
Carl Worth [Wed, 23 Jun 2010 23:27:18 +0000 (16:27 -0700)]
exec_node: Remove destructor from exec_node and all descendants.

Two of these destructors are non-empty, (s_symbol and s_list), so this
commit could potentially introduce memory leaks, (though, no additional
leaks are found in glsl-orangebook-ch06-bump.frag at least---perhaps
the current code is never calling delete on these classes?).

Going forward, we will switch to talloc for exec_node so we won't need
explicit destrcutors to free up any memory used.

14 years agoglsl_symbol_table: Add new talloc-based new()
Carl Worth [Wed, 23 Jun 2010 22:47:04 +0000 (15:47 -0700)]
glsl_symbol_table: Add new talloc-based new()

We take advantage of overloading of the new operator (with an
additional parameter!) to make this look as "C++ like" as possible.

This closes 507 memory leaks when compiling glsl-orangebook-ch06-bump.frag
when measured with:

valgrind ./glsl glsl-orangebook-ch06-bump.frag

as seen here:

total heap usage: 55,623 allocs, 14,389 frees
(was 13,882 frees before)

14 years agoglsl2 main: Use talloc to allocate _mesa_glsl_parse_state
Carl Worth [Wed, 23 Jun 2010 22:43:38 +0000 (15:43 -0700)]
glsl2 main: Use talloc to allocate _mesa_glsl_parse_state

This is a short-lived object. It exists only for the duration of the
compile_shader() function, (as opposed to the shader and whole_program
which live longer).

The state is created with the same talloc parent as the shader, so
that other allocation can be done with talloc_parent(state) as the
owner in order to attach to a long-lived object.

14 years agoglsl2 main: Use talloc to allocate whole_program struct.
Carl Worth [Wed, 23 Jun 2010 20:34:05 +0000 (13:34 -0700)]
glsl2 main: Use talloc to allocate whole_program struct.

This way, whole_program can be our top-level talloc context object,
allowing us to free the lot with a single talloc_free in the end.

14 years agoast_node: Remove empty destructor.
Carl Worth [Wed, 23 Jun 2010 23:16:32 +0000 (16:16 -0700)]
ast_node: Remove empty destructor.

This wasn't serving any purpose. So delete it.

14 years agopreprocessor: Initialize a potentially uninitialized variable.
Carl Worth [Tue, 22 Jun 2010 22:51:34 +0000 (15:51 -0700)]
preprocessor: Initialize a potentially uninitialized variable.

My current reading of the relevant static functions suggests that last
is never used without being uninitialized, (we only use it if the
expansion function returned non-NULL and the expansion functions
always set it before returning non-NULL).

Apparently gcc isn't coming to the same conclusion. Initializing this
to NULL nicely quites gcc and will guarantee a nice, early segfault if
my anaylsis turns out to be wrong.

14 years agopreprocessor: Remove dead code _token_list_length
Carl Worth [Tue, 22 Jun 2010 22:50:38 +0000 (15:50 -0700)]
preprocessor: Remove dead code _token_list_length

As gcc noticed, this function is not currently being used. Good-bye.

14 years agopreprocessor: Remove dead code _string_list_append_list
Carl Worth [Tue, 22 Jun 2010 22:34:59 +0000 (15:34 -0700)]
preprocessor: Remove dead code _string_list_append_list

As gcc noticed, this function is not currently being used. Good-bye.

14 years agoconfigure: Ensure that config.h can be safely included multiple times.
Carl Worth [Thu, 17 Jun 2010 22:25:07 +0000 (15:25 -0700)]
configure: Ensure that config.h can be safely included multiple times.

Use AH_TOP and AH_BOTTOM macros so that the standard include guard
mechanisms are emitted by autoheader into the generated config.h file.

14 years agoconfigure: Remove some dead code.
Carl Worth [Thu, 17 Jun 2010 21:08:36 +0000 (14:08 -0700)]
configure: Remove some dead code.

This block of code is useless because a (nearly-equivalent) assignment
is made immediately after. The only difference is the omission of
-Wunreadchable-code in the assignment being used. Presumably, that was
intended to be -Wunreachable-code (without the first 'd'), but since
this hasn't been being used we just drop it.

14 years agoFix typos of "variable" as "varaible"
Carl Worth [Thu, 17 Jun 2010 07:37:39 +0000 (00:37 -0700)]
Fix typos of "variable" as "varaible"

One of these was just in a comment. But ther other was in an enum
tag, (which is apparently not being used anywhere yet).

14 years agoir_variable: Add some missing initialization to the constructor.
Carl Worth [Thu, 17 Jun 2010 07:35:46 +0000 (00:35 -0700)]
ir_variable: Add some missing initialization to the constructor.

Thanks to valgrind for noticing this problem.

14 years agoFix double usage of the post-inc/dec's temporary pre-inc/dec copy.
Eric Anholt [Wed, 23 Jun 2010 21:57:47 +0000 (14:57 -0700)]
Fix double usage of the post-inc/dec's temporary pre-inc/dec copy.

Fixes CorrectSwizzle3.frag.

14 years agoAvoid using the RHS of an assignment twice.
Eric Anholt [Wed, 23 Jun 2010 21:51:14 +0000 (14:51 -0700)]
Avoid using the RHS of an assignment twice.

This would fix double-evaluation of assignment RHS expressions,
including possible side effects.

14 years agoget_lvalue_copy doesn't need all the checking of do_assignment().
Eric Anholt [Wed, 23 Jun 2010 21:43:50 +0000 (14:43 -0700)]
get_lvalue_copy doesn't need all the checking of do_assignment().

14 years agoast_to_hir: Clone LHS derefs of assignment expressions.
Eric Anholt [Wed, 23 Jun 2010 19:40:17 +0000 (12:40 -0700)]
ast_to_hir: Clone LHS derefs of assignment expressions.

14 years agoir_function_inlining: Allow inlining of loops and conditionals.
Eric Anholt [Wed, 23 Jun 2010 19:19:07 +0000 (12:19 -0700)]
ir_function_inlining: Allow inlining of loops and conditionals.

The new cloning code handles them.

14 years agoAdd a virtual clone() method to ir_instruction.
Eric Anholt [Wed, 23 Jun 2010 18:37:12 +0000 (11:37 -0700)]
Add a virtual clone() method to ir_instruction.

This will be used by function inlining, the linker, and avoiding double
usage of the LHS deref chains in ++, *=, and similar operations.

14 years agoAvoid using the same ir_constant 0.0 multiple times in mat constructors.
Eric Anholt [Wed, 23 Jun 2010 21:33:30 +0000 (14:33 -0700)]
Avoid using the same ir_constant 0.0 multiple times in mat constructors.

14 years agoglcpp: Support line continuations within preprocessor directives.
Kenneth Graunke [Wed, 23 Jun 2010 19:31:09 +0000 (12:31 -0700)]
glcpp: Support line continuations within preprocessor directives.

Fixes CorrectPreprocess5.frag.

14 years agoglcpp: Make standalone binary use preprocess().
Kenneth Graunke [Wed, 23 Jun 2010 21:00:27 +0000 (14:00 -0700)]
glcpp: Make standalone binary use preprocess().

This prevents the two code paths from getting out of sync.  Also, future
work will need the shader source as a string anyway.

Unfortunately, this copies and pastes load_text_file from main.cpp, with
small changes (support for reading from stdin, talloc).

14 years agoglcpp: Fix "dangerous trailing context" warning.
Kenneth Graunke [Mon, 21 Jun 2010 22:15:34 +0000 (15:15 -0700)]
glcpp: Fix "dangerous trailing context" warning.

Flex couldn't be sure whether "def" and "ndef" were part of the #ifdef
and #ifndef patterns or the trailing context of the #if pattern.

14 years agoRemove comment support from the main lexer.
Kenneth Graunke [Mon, 21 Jun 2010 22:12:34 +0000 (15:12 -0700)]
Remove comment support from the main lexer.

Now handled by the preprocessor.

14 years agoglcpp: Recognize plain "//" as a comment.
Kenneth Graunke [Mon, 21 Jun 2010 22:11:01 +0000 (15:11 -0700)]
glcpp: Recognize plain "//" as a comment.

Found in glsl-orangebook-ch06-bump.(frag|vert).  This was resulting in
the comments getting passed through to the main compiler's lexer.

14 years agoGenerate errors for empty constructors instead of asserting
Ian Romanick [Wed, 23 Jun 2010 20:58:34 +0000 (13:58 -0700)]
Generate errors for empty constructors instead of asserting

This causes the following tests to pass:

     glslparsertest/glsl2/constructor-10.vert

14 years agolinker: Update some comments, remove a couple FINISHMEs
Ian Romanick [Wed, 23 Jun 2010 19:18:21 +0000 (12:18 -0700)]
linker: Update some comments, remove a couple FINISHMEs

14 years agolinker: Use InfoLog in assign_attribute_locations
Ian Romanick [Wed, 23 Jun 2010 19:14:02 +0000 (12:14 -0700)]
linker: Use InfoLog in assign_attribute_locations

Since the program is now passed in, refactor the parameter list to the
function as well.

14 years agolinker: Refactor parameters to cross_validate_uniforms
Ian Romanick [Wed, 23 Jun 2010 19:09:14 +0000 (12:09 -0700)]
linker: Refactor parameters to cross_validate_uniforms

14 years agolinker: Write errors to info log instead of stdout
Ian Romanick [Wed, 23 Jun 2010 19:07:22 +0000 (12:07 -0700)]
linker: Write errors to info log instead of stdout

14 years agolinker: Initial implementation of varying slot allocation
Ian Romanick [Wed, 23 Jun 2010 18:23:01 +0000 (11:23 -0700)]
linker: Initial implementation of varying slot allocation

14 years agolinker: Limit attribute allocation to MAX_VERTEX_ATTRIBS
Ian Romanick [Wed, 23 Jun 2010 00:41:37 +0000 (17:41 -0700)]
linker: Limit attribute allocation to MAX_VERTEX_ATTRIBS

14 years agolinker: Support matrix and array vertex inputs
Ian Romanick [Wed, 23 Jun 2010 00:29:19 +0000 (17:29 -0700)]
linker: Support matrix and array vertex inputs

14 years agolinker: Initial implementation of attribute slot allocation
Ian Romanick [Tue, 22 Jun 2010 00:18:31 +0000 (17:18 -0700)]
linker: Initial implementation of attribute slot allocation

14 years agoImport some bits Mesa's mtypes.h
Ian Romanick [Tue, 22 Jun 2010 20:41:50 +0000 (13:41 -0700)]
Import some bits Mesa's mtypes.h

14 years agoImport gl_program_parameter and gl_program_parameter_list types from Mesa
Ian Romanick [Tue, 22 Jun 2010 00:08:51 +0000 (17:08 -0700)]
Import gl_program_parameter and gl_program_parameter_list types from Mesa

14 years agolinker: Store the par-linked per-stage shaders in the glsl_program
Ian Romanick [Mon, 21 Jun 2010 23:16:05 +0000 (16:16 -0700)]
linker: Store the par-linked per-stage shaders in the glsl_program

14 years agolinker: Initial implementation of uniform slot allocation
Ian Romanick [Mon, 21 Jun 2010 23:10:42 +0000 (16:10 -0700)]
linker: Initial implementation of uniform slot allocation

14 years agoImport gl_uniform and gl_uniform_list types from Mesa
Ian Romanick [Mon, 21 Jun 2010 23:06:07 +0000 (16:06 -0700)]
Import gl_uniform and gl_uniform_list types from Mesa

14 years agoir_variable: Add query to get number of slots used by a variable
Ian Romanick [Mon, 21 Jun 2010 23:05:29 +0000 (16:05 -0700)]
ir_variable: Add query to get number of slots used by a variable

14 years agoglsl_type: Add method to get number of slots used by a type
Ian Romanick [Mon, 21 Jun 2010 23:05:00 +0000 (16:05 -0700)]
glsl_type: Add method to get number of slots used by a type

14 years agolinker: Document what remains to be done
Ian Romanick [Mon, 21 Jun 2010 19:03:24 +0000 (12:03 -0700)]
linker: Document what remains to be done

14 years agoir_variable: Set locations for shader built-in variables
Ian Romanick [Mon, 21 Jun 2010 18:42:57 +0000 (11:42 -0700)]
ir_variable: Set locations for shader built-in variables

14 years agoir_variable: Track the location of uniforms, varings, attributes, etc.
Ian Romanick [Mon, 21 Jun 2010 18:42:02 +0000 (11:42 -0700)]
ir_variable: Track the location of uniforms, varings, attributes, etc.

14 years agolinker: Initial implementation of interstage input / output validation
Ian Romanick [Sat, 19 Jun 2010 02:02:10 +0000 (19:02 -0700)]
linker: Initial implementation of interstage input / output validation

14 years agoir_variable: Add method to get string representing interpolation qualifier
Ian Romanick [Sat, 19 Jun 2010 02:00:28 +0000 (19:00 -0700)]
ir_variable: Add method to get string representing interpolation qualifier

14 years agolinker: Initial implementation of interstage uniform validation
Ian Romanick [Sat, 19 Jun 2010 00:13:42 +0000 (17:13 -0700)]
linker: Initial implementation of interstage uniform validation

14 years agoir_constant: Add method to determine if two constants have the same value
Ian Romanick [Fri, 18 Jun 2010 02:50:36 +0000 (19:50 -0700)]
ir_constant: Add method to determine if two constants have the same value

14 years agoFix multiple usage of an rvalue in constructor component splitting.
Eric Anholt [Tue, 22 Jun 2010 21:22:42 +0000 (14:22 -0700)]
Fix multiple usage of an rvalue in constructor component splitting.

Store the thing in a variable and deref it for each swizzle if we have to.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
14 years agoir_validate: Use callback functionality in ir_hierarchical_visitor
Ian Romanick [Wed, 23 Jun 2010 01:41:50 +0000 (18:41 -0700)]
ir_validate: Use callback functionality in ir_hierarchical_visitor

14 years agoir_hierarchical_visitor: Add generic callback functionality
Ian Romanick [Wed, 23 Jun 2010 01:37:12 +0000 (18:37 -0700)]
ir_hierarchical_visitor: Add generic callback functionality

14 years agoir_validate: New pass for checking our invariants.
Eric Anholt [Tue, 22 Jun 2010 19:07:21 +0000 (12:07 -0700)]
ir_validate: New pass for checking our invariants.

14 years agoir: Give ir_instruction a print visitor helper.
Eric Anholt [Tue, 22 Jun 2010 19:09:21 +0000 (12:09 -0700)]
ir: Give ir_instruction a print visitor helper.

This avoids spamming each file with includes of ir_print_visitor.h
because someone was doing debugging at some point, and is less typing
when doing debugging.

14 years agoHook up texturing in the hierarchical visitor.
Eric Anholt [Tue, 22 Jun 2010 17:45:15 +0000 (10:45 -0700)]
Hook up texturing in the hierarchical visitor.

14 years agoInclude stdio.h and stdlib.h everywhere, and don't cook our own #define NULL.
Eric Anholt [Tue, 22 Jun 2010 17:38:52 +0000 (10:38 -0700)]
Include stdio.h and stdlib.h everywhere, and don't cook our own #define NULL.

14 years agoAdd missing build products to gitignore
Ian Romanick [Tue, 22 Jun 2010 19:24:42 +0000 (12:24 -0700)]
Add missing build products to gitignore

14 years agoglcpp: Initialize error state.
Kenneth Graunke [Mon, 21 Jun 2010 20:55:12 +0000 (13:55 -0700)]
glcpp: Initialize error state.

14 years agoglcpp: Accept #elif without an expression if the expression doesn't matter.
Kenneth Graunke [Mon, 21 Jun 2010 19:20:22 +0000 (12:20 -0700)]
glcpp: Accept #elif without an expression if the expression doesn't matter.

Issue a warning.  nVidia's compiler seems to accept this; apparently
GCC < 4.4 did as well: http://gcc.gnu.org/gcc-4.4/porting_to.html

14 years agoglcpp: Print to the main compiler's infolog, not stdout.
Kenneth Graunke [Mon, 21 Jun 2010 18:47:55 +0000 (11:47 -0700)]
glcpp: Print to the main compiler's infolog, not stdout.

14 years agoUse yy_scan_string and stop caring about shader->SourceLen.
Kenneth Graunke [Mon, 21 Jun 2010 18:43:42 +0000 (11:43 -0700)]
Use yy_scan_string and stop caring about shader->SourceLen.

We had to call strlen on the preprocessed source, which seemed a bit
pointless; also, we updated shader->SourceLen but not shader->Source,
which was even more confusing.  Just leave both untouched.

14 years agoglcpp: Rework handling of "defined" operator.
Kenneth Graunke [Sat, 19 Jun 2010 06:06:54 +0000 (23:06 -0700)]
glcpp: Rework handling of "defined" operator.

It's now done in the grammar, and as a result, can easily handle
parenthesis.  defined ( identifier ) is now supported.

Fixes glcpp/tests/065-if-defined-parens.c.

14 years agoglcpp/tests: Add a test for #if defined (identifier)
Kenneth Graunke [Sat, 19 Jun 2010 06:04:44 +0000 (23:04 -0700)]
glcpp/tests: Add a test for #if defined (identifier)

Previously we only tested #if defined identifier (without parenthesis).

14 years agoglcpp: Ignore junk tokens at end of #ifdef/#ifndef (with warning)
Kenneth Graunke [Sat, 19 Jun 2010 03:08:15 +0000 (20:08 -0700)]
glcpp: Ignore junk tokens at end of #ifdef/#ifndef (with warning)

Both nVidia's compiler and GCC accept this.
Fixes CorrectPreprocess11.frag.

14 years agoglcpp: Add glcpp_warning for printing warnings to the info log.
Kenneth Graunke [Sat, 19 Jun 2010 02:54:25 +0000 (19:54 -0700)]
glcpp: Add glcpp_warning for printing warnings to the info log.

14 years agoglcpp: Add boolean 'error' flag.
Kenneth Graunke [Mon, 21 Jun 2010 19:39:49 +0000 (12:39 -0700)]
glcpp: Add boolean 'error' flag.

We used to check if the info log is non-empty, but when we print
warnings, this will no longer be valid.

14 years agoglcpp: Rename "errors" to "info_log."
Kenneth Graunke [Sat, 19 Jun 2010 02:52:36 +0000 (19:52 -0700)]
glcpp: Rename "errors" to "info_log."

Eventually, we'll want to be be able to print out warnings as well.

14 years agoglcpp: Don't include newlines as part of #version-passthrough.
Kenneth Graunke [Fri, 18 Jun 2010 22:41:00 +0000 (15:41 -0700)]
glcpp: Don't include newlines as part of #version-passthrough.

Fixes glcpp/tests/064-version.c.

14 years agoglcpp/tests: Add a test for #version followed by #define.
Kenneth Graunke [Fri, 18 Jun 2010 22:34:50 +0000 (15:34 -0700)]
glcpp/tests: Add a test for #version followed by #define.

This isn't really a C file, but...that's probably okay.

14 years agoglcpp: Fix line and column numbering.
Kenneth Graunke [Fri, 18 Jun 2010 22:23:50 +0000 (15:23 -0700)]
glcpp: Fix line and column numbering.

Lines were off by one, and column numbering was completely daft.

14 years agoglcpp: Remove calls to exit().
Kenneth Graunke [Thu, 17 Jun 2010 19:58:54 +0000 (12:58 -0700)]
glcpp: Remove calls to exit().

Calling exit() would be really bad once integrated into mesa.  Even in
the standalone binary, we want to print the error log first.

Since each case already flags an error, compilation will still fail,
but it may go on (with something fudged) and generate more errors.

14 years agoglcpp: Add line locations to "Unterminated #if" error message.
Kenneth Graunke [Thu, 17 Jun 2010 19:41:46 +0000 (12:41 -0700)]
glcpp: Add line locations to "Unterminated #if" error message.

14 years agoglcpp: Add line locations to various mismatched #if error messages.
Kenneth Graunke [Thu, 17 Jun 2010 19:30:57 +0000 (12:30 -0700)]
glcpp: Add line locations to various mismatched #if error messages.

14 years agoglcpp: Add line locations to "reserved name" error messages.
Kenneth Graunke [Thu, 17 Jun 2010 19:21:53 +0000 (12:21 -0700)]
glcpp: Add line locations to "reserved name" error messages.

14 years agoglcpp: Print locations in error messages where possible.
Kenneth Graunke [Thu, 17 Jun 2010 00:31:50 +0000 (17:31 -0700)]
glcpp: Print locations in error messages where possible.

14 years agoglcpp: Introduce new glcpp_error function.
Kenneth Graunke [Thu, 17 Jun 2010 19:03:25 +0000 (12:03 -0700)]
glcpp: Introduce new glcpp_error function.

14 years agoglcpp: Set locations on tokens.
Kenneth Graunke [Wed, 16 Jun 2010 23:58:31 +0000 (16:58 -0700)]
glcpp: Set locations on tokens.

14 years agoglcpp: Set line locations in the lexer.
Kenneth Graunke [Thu, 17 Jun 2010 00:41:12 +0000 (17:41 -0700)]
glcpp: Set line locations in the lexer.

14 years agoglcpp: Add plumbing to support line locations.
Kenneth Graunke [Wed, 16 Jun 2010 23:35:57 +0000 (16:35 -0700)]
glcpp: Add plumbing to support line locations.

14 years agoglcpp: Add %error-verbose.
Kenneth Graunke [Thu, 17 Jun 2010 20:07:13 +0000 (13:07 -0700)]
glcpp: Add %error-verbose.

14 years agoglcpp: Actually support #ifdef and #ifndef.
Kenneth Graunke [Thu, 17 Jun 2010 21:36:34 +0000 (14:36 -0700)]
glcpp: Actually support #ifdef and #ifndef.

Strangely, the lexer never created these tokens, even though the parser
already had code to handle them.

14 years agoglcpp: Build a reentrant parser.
Kenneth Graunke [Wed, 16 Jun 2010 23:26:28 +0000 (16:26 -0700)]
glcpp: Build a reentrant parser.

14 years agoglcpp: Print errors on stdout instead of stderr (non-standalone version).
Kenneth Graunke [Wed, 16 Jun 2010 20:19:27 +0000 (13:19 -0700)]
glcpp: Print errors on stdout instead of stderr (non-standalone version).

Otherwise, piglit marks tests as "warn" when the shader was (correctly)
failing.

14 years agoglcpp/tests: Add extra newline at the end of expected output.
Kenneth Graunke [Thu, 17 Jun 2010 21:49:40 +0000 (14:49 -0700)]
glcpp/tests: Add extra newline at the end of expected output.

This newline at EOF is harmless and generated by the previous commit.