emscripten - Emscripten: An LLVM-to-JavaScript Compiler

about summary refs log tree commit diff

path: root/src/parseTools.js

Age	Commit message (Collapse)	Author
2013-06-19	keep \|0 on function calls, allow other bitwise ops on heap accesses etc.	Alon Zakai

2013-06-10	coerce output of integer modulus, since x%0 is NaN, which is not an integer	Alon Zakai

2013-06-01	limit a warning message	Alon Zakai

2013-05-30	handle illegal mathops in processMathop; valid for <64 bits in some cases	Alon Zakai

2013-05-23	fix argument counting regression from 5efd506 ; fixes #1202	Alon Zakai

2013-05-22	handle dynamic indexes properly, even out of bounds, for array gep lookups; ↵	Alon Zakai
	fixes #865
2013-05-20	correctly count the number of normal vars even when there are i64s before a ↵	Alon Zakai
	varargs
2013-05-16	make some more asm things closure-friendly	Alon Zakai

2013-05-10	show all dangerous pointer casts only in VERBOSE=1	Alon Zakai

2013-05-09	do not attempt aligned reads/writes on illegal values	Alon Zakai

2013-05-09	FORCE_ALIGNED_MEMORY option	Alon Zakai

2013-05-04	make memory management sane: 0 is null, then static including globals, then ↵	Alon Zakai
	stack, then dynamic/sbrk. deprecate shared libs (BUILD_AS_SHARED_LIB)
2013-05-03	assert on no fp80s etc. in non-x86 target	Alon Zakai

2013-05-02	use 64-bit alignment for double writes in le32	Alon Zakai

2013-05-01	remove unneeded corrections in unrolled memcpy	Alon Zakai

2013-05-01	fix alignment detection in asm for makeCopyValues, strip corrections before ↵	Alon Zakai
	calculating
2013-04-29	do not get confused by local vars that alias names of library functions, and ↵	Alon Zakai
	are used as function pointers
2013-04-29	todo	Alon Zakai

2013-04-29	tweak handling of unaligned doubles a little	Alon Zakai

2013-04-29	read/write 64-bit aligned doubles directly in le32, when aligned properly	Alon Zakai

2013-04-29	do not hope for 64-bit alignment with doubles in DOUBLE_MODE 0 since we will ↵	Alon Zakai
	read only 32 bits anyhow
2013-04-26	refer to wiki for function pointer cast issues	Alon Zakai

2013-04-26	improve isFunctionType	Alon Zakai

2013-04-26	check bitcasts in eatLLVMIdent, so that call-bitcast is checked	Alon Zakai

2013-04-25	more docs for function pointer casts	Alon Zakai

2013-04-25	clearer warnings on incompatible function pointer casts	Alon Zakai

2013-04-23	emit coercion on i64 math call return values	Alon Zakai

2013-04-23	clear errors on calling dead functions	Alon Zakai

2013-04-22	show clear error message instead of 987 when too many setjmp calls	Alon Zakai

2013-04-14	use compiled i64 div and rem	Alon Zakai

2013-04-14	use compiled compiler-rt code for i64 multiplication	Alon Zakai

2013-04-14	add handwritten asm i64Subtract	Alon Zakai

2013-04-13	find structural returns of >64 bits	Alon Zakai

2013-04-05	test various chunk sizes for memory allocation	Alon Zakai

2013-04-04	pre-apply {{{ FI_* }}} as well	Alon Zakai

2013-04-04	pre-apply simple postSets	Alon Zakai

2013-04-04	do not emit the memory initialization into itself	Alon Zakai

2013-04-04	refactor slab writing in makePointer	Alon Zakai

2013-04-04	make sure allocate() types have the same length as the slab	Alon Zakai

2013-04-04	fix s_x_x	Alon Zakai

2013-04-04	always flatten out ta2 allocate()ions	Alon Zakai

2013-04-04	write out all the initial global allocations as one big ALLOC_NONE	Alon Zakai

2013-04-04	refactor chunkifying in makePointer	Alon Zakai

2013-04-04	compress in makePointer only in non-ta2, in preparation for consolidating ↵	Alon Zakai
	all ta2 allocations
2013-03-31	differentiate better the ambiguous case of llvm call (where the type can be ↵	Alon Zakai
	either just the return type, or the entire function type): ** and above are certainly just the return type
2013-03-21	asmify CHECK_HEAP_ALIGN calls	Alon Zakai

2013-03-19	handle negative zero; fixes #921	Alon Zakai

2013-03-15	asmify tolower	Alon Zakai

2013-03-13	remove unneeded 64-bit shifts code	Alon Zakai

2013-03-12	infrastructure for implementing i64 math in asm.js, and implement i64Add and ↵	Alon Zakai
	llvm_uadd_with_overflow_i64 that way