7 * have a client-provided error-handling callback for the context, and
8 call it, rather than asserting/crashing etc, to make the API resilient and helpful
10 * probably should turn off signal handlers and backtracing, leaving that to
13 * enums and ABI: give enums specific numbers, in ranges, to make it
14 possible to maintain a logical ordering whilst preserving ABI.
16 * expose the statements in the API? (mostly so they can be stringified?)
18 * support more arithmetic ops and comparison modes
20 * access to a function by address::
22 extern gcc_jit_function *
23 gcc_jit_context_get_function (ctxt,
24 void *); /* need type information */
26 so you can access "static" fns in your code.
28 * ability to turn a function into a function pointer::
30 gcc_jit_function_as_rvalue ()
32 * expressing branch probabilies (like __builtin_expect)::
34 extern gcc_jit_rvalue *
35 gcc_jit_rvalue_likely (gcc_jit_rvalue *rvalue,
41 gcc_jit_block_set_likelihood (gcc_jit_block *block,
44 be better? (for expressing how hot the current location is)
46 * add a SONAME to the library (and potentially version the symbols?)
48 * do we need alternative forms of division (floor vs rounding)?
50 * are we missing any ops?
54 * gcc_jit_context_new_unary_op: various checks needed
56 * gcc_jit_context_new_binary_op: various checks needed
58 * gcc_jit_context_new_comparison: must be numeric or pointer types
60 * gcc_jit_context_new_array_access: "index" must be of numeric type.
62 * gcc_jit_lvalue_access_field: must be field of correct struct
64 * gcc_jit_rvalue_access_field: must be field of correct struct
66 * gcc_jit_block_add_assignment_op: check the types
68 * Implement more kinds of casts e.g. pointers
72 * fixing all the state issues: make it work repeatedly with optimization
75 * make the dirty dirty hacks less egregious...
77 * test under valgrind; fix memory leaks
79 * re-architect gcc so we don't have to reinitialize everything every time
84 * get DejaGnu to build and run C++ testcases
86 * measure code coverage in testing of libgccjit.so
90 * try porting llvmpipe to gcc
94 * Detect and issue warnings/errors about uses of uninitialized variables
96 * Warn about unused objects in a context (e.g. rvalues/lvalues)? (e.g.
97 for gcc_jit_context_new_call vs gcc_jit_block_add_eval)
101 * Currently each function has a single stmt_list, which is built in
102 postprocessing by walking the list of blocks. Presumably we could
103 have each block have its own stmt_list, avoiding the need for this
104 traversal, and having the block structure show up within tree dumps.
105 Alternatively, could we skip tree and go straight to gimple?
107 * ability to give contexts names, for ease of debugging?
112 * "switch" and "case" ?
114 * sizeof (should this be an API hook?) do we even need it? presumably
115 client code can just do the sizeof() in its own code.
117 * do we need unary plus?