duktape

Commit Graph

Author	SHA1	Message	Date
Sami Vaarala	05370e419b	Fix incorrect assert	8 years ago
Sami Vaarala	735f9024c4	Value stack rework cleanups * Wrap checks to duk_require_stack() and variants. * Wrap check to value stack grow. * Add internal helper duk_set_top_and_wipe(); for now it's just two duk_set_top() calls but can be optimized later.	8 years ago
Sami Vaarala	a9fa2362ec	Rework value stack grow/shrink, lazy shrink * Make value stack and call stack limits configurable via DUK_USE_xxx options. Also make value stack grow/shrink constants configurable. * Rewrite value stack grow/shrink check primitives for better hot/cold path handling. * Use a proportional spare for grow and shrink sizes so that applications needing a large value stack have fewer value stack resizes. * Grow value stack allocation when entering a call or when explicitly requested via e.g. duk_require_stack(). * Never shrink the value stack when entering a call, so that the unwind path is guaranteed to have value stack to handle a protected call return. This guarantee is only needed for protected call but is now applied to all calls for simplicity. * Don't perform a value stack shrink check at all in function return anymore. It would be OK from protected call semantics perspective to do a shrink attempt without throwing if it fails. * Perform a value stack shrink check in mark-and-sweep only for now. When emergency GC is running, shrink to a minimal size respecting current value stack reserve.	8 years ago
Sami Vaarala	d164bd44ce	Fix duk_{check,require}_stack_top() reserve	8 years ago
Sami Vaarala	ea217a30ec	duk_unpack_array_like() fastpath	8 years ago
Sami Vaarala	2d8af4dc86	Simplify .call/.apply handling	8 years ago
Sami Vaarala	22b72646a2	Don't shrink value stack for a call	8 years ago
Sami Vaarala	4c2a198978	Direct call handling of .call() and .apply() Handle .call() and .apply() directly without them being visible on the call stack.	8 years ago
Sami Vaarala	4bc11b8ea6	Fix Reflect.construct() extra argument handling	8 years ago
Sami Vaarala	adcc4e5bed	Use %NativeFunctionPrototype% for native functions Not applied to built-in native functions which must inherit directly from Function.prototype.	8 years ago
Sami Vaarala	8d5ec2bff3	Handle missing and empty _Formals in dump format The distinction matters when loading back: if _Formals was omitted, we don't want to load back an empty _Formals array for a function template because it then breaks function instance .length computation.	8 years ago
Sami Vaarala	e6fcf1e1c9	Fix a harmless MSVC cast warning	8 years ago
Sami Vaarala	eaebc35e71	Move large and rare opcodes into NOINLINE helpers Move handling of the following opcodes into NOINLINE helpers to make the main dispatch function smaller and a bit better performing: * INITSET and INITGET * TRYCATCH * ENDTRY * ENDCATCH * ENDFIN * INITENUM and NEXTENUM	8 years ago
Sami Vaarala	e9f20f62f5	Fix out-of-sync opcode debug printing	8 years ago
Sami Vaarala	a589a843c3	Remove unnecessary ENDFIN value push	8 years ago
Sami Vaarala	e302835c37	Fix out-of-date executor comments (act stability)	8 years ago
Sami Vaarala	7f95695d4a	Simplify ENDFIN handling ENDFIN only needs reg_catch base index (relative to valstack_bottom). Assertions reference 'cat' catcher but that's not needed at runtime.	8 years ago
Sami Vaarala	3ede444c8c	Add duk_hboundfunc internal type	8 years ago
Sami Vaarala	be5a4d2231	Avoid duk_hbufobj validity assert trigger duk_hbufobj validity assertions require that if the buf pointer is NULL, the buffer length and offset are zero. This is violated by the shared slice code because it sets ->length before allocating a buffer, and since an allocation may trigger side effects, the assert may trigger in mark-and-sweep. I don't think this causes issues besides an assert failure.	8 years ago
Sami Vaarala	1581417c05	Add missing require stack for bound func handling	8 years ago
Sami Vaarala	b98bfd60d3	Fix some explicit NULL init bugs	8 years ago
Sami Vaarala	c21ae1a040	Add catcher/activation freelist initial draft	8 years ago
Sami Vaarala	4c52c793c6	Fix DUK_ASSERT_HTHREAD_VALID() assert The assert cannot require thr->valstack != NULL (like DUK_ASSERT_CTX_VALID) or torture tests will trigger the assert.	8 years ago
Sami Vaarala	55645c0f89	Skip ROM objects using REACHABLE in mark-and-sweep With ROM objects having REACHABLE always set, mark-and-sweep doesn't need a specific ROM object check when deciding whether to mark an object or not: the REACHABLE check causes the object not to be marked.	8 years ago
Sami Vaarala	97724d5d45	Add duk_hproxy internal struct	8 years ago
Sami Vaarala	4cc08436ec	Rework duk_activations into a linked list Remove thr->callstack as a monolithic array and replace it with a linked list of duk_activations. thr->callstack_curr is the current call (or NULL if no call is in progress), and act->parent chains to a previous call or NULL. thr->callstack_top is kept because it's needed by some internals at present; it may be removed in the future.	8 years ago
Sami Vaarala	21ea24b636	Linked list for catchstack, attach to activation Remove explicit catchstack. Make catchers a single linked list of duk_catcher structs, attached to a duk_activation.	8 years ago
Sami Vaarala	7284a5b192	Bump DUK_VERSION to 2.1.99	8 years ago
Sami Vaarala	56dae760cc	Bump DUK_VERSION for 2.1.0 release	8 years ago
Sami Vaarala	4bba87d7e8	Fix potentially stale tv_key in deleteProperty	8 years ago
Sami Vaarala	0bb978cb3e	Allow Annex B octal escapes in RegExp charclass	8 years ago
Sami Vaarala	3739645ce9	Add duk_get_xxx_default() API calls	8 years ago
Sami Vaarala	02e8adcce8	Source reorg for duk_api_public.h.in * Merge duk_api_public.h.in to duktape.h.in. * Adjust define order so that duk_config.h sees DUK_VERSION which allows user config fixups to react to it.	8 years ago
Sami Vaarala	5b04a73ed5	Add duk_opt_xxx() API calls	8 years ago
Sami Vaarala	49f87b6422	Allow mark-and-sweep during finalizer processing Tweak mark-and-sweep so that if finalizers are present (heap->finalize_list is not NULL), rescue decisions are postponed (free decisions are not). In concrete terms this means that objects normally rescued keep their FINALIZED flag so that their finalizer won't be called again if the object turns out to be unreachable in a later run.	8 years ago
Sami Vaarala	1876bb785a	Fix RegExp group SyntaxError handling	8 years ago
Sami Vaarala	c6f9ee1550	Fix Symbol object .valueOf()	8 years ago
Sami Vaarala	6ef6ece6a3	Allow duk_push_heapptr() for obj pending finalizer	8 years ago
Sami Vaarala	111c0eac97	Check finalize_list also for temproots This wasn't necessary before: finalize_list only contained unreachable objects so nothing could point to them while we marked heap_allocated. But when duk_push_heapptr() is allowed to push unreachable pointers (which are pending finalization), it's possible for an object in heap_allocated to point to an object on finalize_list, which also means that the latter object can get a TEMPROOT flag.	8 years ago
Sami Vaarala	f8384d9520	Shared assert helper for 'in heap_allocated'	8 years ago
Sami Vaarala	6aade4f7bf	Bump refcounts on finalize_list insert	8 years ago
Sami Vaarala	cc02912201	Forced stack reallocs in finalizer torture	9 years ago
Sami Vaarala	71a05b79ce	Fix duk_api_inspect() stale 'tv' pointer A pointer to the value stack was obtained before duk_push_bare_object() and used after the push. If value stack resize happens as a side effect of the push (mark-and-sweep, finalizers, etc) the 'tv' pointer could be stale. Found using torture tests.	8 years ago
Sami Vaarala	e11753082f	Use DUK_COMPILE_xxx flags are internally Avoids unnecessary translation.	8 years ago
Sami Vaarala	cb1c8957d4	Add shebang comment parsing (on first line only)	8 years ago
Sami Vaarala	55dbf32393	Rework side effect protections and finalization Several small reworks in refzero, finalizer, mark-and-sweep, and error handling to avoid side effect issues.	8 years ago
Sami Vaarala	a94be8e0bd	Make global conditional to DUK_USE_GLOBAL_BINDING	8 years ago
Sami Vaarala	045c3cbf0c	Make HTML syntax support conditional	8 years ago
Sami Vaarala	9ab8d5be16	Shared helper for single line comments in lexer	8 years ago
Michael Drake	79b93bf580	Add ES6 SingleLineHTMLCloseComment support ("-->").	8 years ago

... 4 5 6 7 8 ...

563 Commits (bb6e4911b5470557db7f809fc98975f7929eb999)