1889 Commits

This Branch

This Branch

All Branches

Author	SHA1	Message	Date
Lioncash	fe7c21b687	frontend/ir/cond: Remove unused invert() function ... This is no longer used by anything in the codebase, so it can be removed.	2019-04-16 21:37:12 -04:00
Merry	c8446bda2b	Merge pull request #480 from lioncash/info ... common/fp/info: Make half-precision info struct functions return correctly sized types	2019-04-15 12:14:30 +01:00
Lioncash	2c2fdb435c	common/fp/info: Make half-precision info struct functions return correctly sized types ... While initially done to potentially prevent creating bugs due to C++ having a silly type-promotion mechanism involving types < sizeof(int) and unsignedness, given that the bulk of these functions' usages are on exit paths, these can return the correct type to avoid the need to cast at every usage point.	2019-04-15 04:55:00 -04:00
Merry	973c87445e	Merge pull request #479 from lioncash/rsqrts ... A64: Handle half-precision variants of FRSQRTS	2019-04-15 08:28:30 +01:00
Lioncash	28e0f4fe59	A64: Implement FRSQRTS' half-precision vector variant	2019-04-14 21:12:54 -04:00
Lioncash	6f7a370cfb	A64: Implement FRSQRTS' half-precision scalar variant ... With the necessary machinery in place, we can now handle the half-precision variant.	2019-04-14 21:12:54 -04:00
Lioncash	793b3b38d4	frontend/ir_emitter: Add half-precision opcode variant of FPVectorRSqrtStepFused	2019-04-14 21:12:54 -04:00
Lioncash	db4d134726	frontend/ir_emitter: Add half-precision opcode variant of FPRSqrtStepFused	2019-04-14 21:12:49 -04:00
Lioncash	cff4394717	common/fp/op/FPRSqrtStepFused: Add half-precision specialization for FPRSqrtStepFused	2019-04-14 20:51:58 -04:00
Merry	eea732febf	Merge pull request #478 from lioncash/stepfused ... A64: Handle half-precision variants of FRECPE and FRECPS	2019-04-14 12:40:18 +01:00
Merry	0c62dba214	Merge pull request #477 from lioncash/rsqrt ... A64: Handle half-precision variants of FRSQRTE	2019-04-14 11:21:05 +01:00
Lioncash	b67cc72213	A64: Implement half-precision vector variant of FRECPE	2019-04-14 06:14:20 -04:00
Lioncash	4b3f5b8a30	A64: Implement half-precision scalar variant of FRECPE	2019-04-14 06:14:20 -04:00
Lioncash	4519e4641c	A64: Implement half-precision vector variant of FRECPS	2019-04-14 06:14:19 -04:00
Lioncash	abcf1a6e09	A64: Implement half-precision scalar variant of FRECPS	2019-04-14 06:14:19 -04:00
Lioncash	c55db96819	frontend/ir_emitter: Add half-precision opcode for FPVectorRecipEstimate	2019-04-14 06:14:19 -04:00
Lioncash	1615ba0adc	frontend/ir_emitter: Add half-precision opcode for FPRecipEstimate	2019-04-14 06:14:19 -04:00
Lioncash	e93161e25b	common/fp/op: Add half-precision specialization for FPRecipEstimate	2019-04-14 06:14:19 -04:00
Lioncash	4ae0a27ea4	frontend/ir_emitter: Add half-precision opcode for FPVectorRecipStepFused	2019-04-14 06:14:19 -04:00
Lioncash	065143b395	frontend/ir_emitter: Add half-precision opcode for FPRecipStepFused	2019-04-14 06:14:18 -04:00
Lioncash	408bbfb766	common/fp/op: Add half-precision specialization for FPRecipStepFused	2019-04-14 06:13:43 -04:00
Lioncash	e35352b69e	A64: Implement half-precision variant of FRSQRTE's vector variant	2019-04-14 06:12:20 -04:00
Lioncash	0f677744bb	A64: Implement half-precision variant of FRSQRTE's scalar variant	2019-04-14 06:12:20 -04:00
Lioncash	5edbb415c5	frontend/ir_emitter: Add half-precision opcode variant for FPVectorRSqrtEstimate	2019-04-14 06:12:20 -04:00
Lioncash	f1e556632c	frontend/ir_emitter: Add half-precision opcode variant for FPRSqrtEstimate	2019-04-14 06:11:45 -04:00
Lioncash	9c9087ea47	common/fp/op/FPRSqrtEstimate: Add half-precision specialization for FPRSqrtEstimate	2019-04-14 06:10:27 -04:00
Merry	5ae8e4e043	Merge pull request #476 from lioncash/frint ... A64: Handle half-precision variants of FRINT* instructions	2019-04-14 10:57:09 +01:00
Lioncash	f8fb10289a	A64: Enable half-precision vector FRINT* variants	2019-04-13 18:05:00 -04:00
Lioncash	7c5cc15f4e	A64: Enable half-precision variants of floating-point FRINT* variants ... With all the backing machinery in place, we can remove the fallback check for half-precision.	2019-04-13 17:54:36 -04:00
Lioncash	95a96b32c5	frontend/microinstruction: Add FPVectorRoundInt types to ReadsFromAndWritesToFPSRCumulativeExceptionBits() ... All variants were previously missing from this.	2019-04-13 17:51:13 -04:00
Lioncash	722daae0d4	frontend/ir_emitter: Add half-precision variant of FPVectorRoundInt	2019-04-13 17:49:04 -04:00
Lioncash	8243705134	frontend/ir_emitter: Add half-precision variant of FPRoundInt	2019-04-13 17:44:37 -04:00
Lioncash	590388b9ec	fp/op/FPRoundInt: Add half-precision specialization of FPRoundInt	2019-04-13 17:37:45 -04:00
Merry	b34f42575d	Merge pull request #475 from lioncash/muladd ... A64: Enable half-precision variants of floating-point multiply-add instructions	2019-04-13 14:30:03 +01:00
Merry	2218912292	A64/location_descriptor: Ensure FZ16 is included in the FPCR mask	2019-04-13 08:05:04 -04:00
Merry	9faf1bdfd4	Merge pull request #474 from lioncash/bracing ... load_store_*: Make bracing consistent and variables const where applicable	2019-04-13 10:53:21 +01:00
Merry	5f351a669c	Merge pull request #473 from lioncash/sqshlu ... A64: Implement SQSHLU	2019-04-13 08:31:17 +01:00
Lioncash	a935f35553	A64: Implement FMLA/FMLS' half-precision vector indexed variants	2019-04-13 02:19:06 -04:00
Lioncash	caaa04cba1	A64: Implement FMLA/FMLS' half-precision scalar indexed variants	2019-04-13 02:11:44 -04:00
Lioncash	894169dc7a	A64: Implement half-precision vector variants of FMLA/FMLS	2019-04-13 01:48:22 -04:00
Lioncash	278c7ae744	ir/frontend: Add half-precision opcode for FPVectorMulAdd	2019-04-13 01:42:35 -04:00
Lioncash	0eb6412c4d	A64: Enable half-precision floating point variants of FP data-processing three register instructions ... This handles half-precision floating point for: - FMADD - FMSUB - FNMADD - FNMSUB	2019-04-13 01:35:32 -04:00
Lioncash	13b41525cb	frontend/ir_emitter: Add half-precision opcode for FPMulAdd	2019-04-13 00:18:09 -04:00
Lioncash	2ae0c4807a	fp/op/FPMulAdd: Add half-precision floating-point specialization	2019-04-13 00:07:34 -04:00
Lioncash	94ac9becda	load_store_*: Make bracing consistent and variables const where applicable ... Makes bracing consistent, and variables const where applicable to be consistent with the rest of the codebase. In most bracing cases, they'd need to be added to conditionals that would involve checking stack pointer alignment in the future anyways.	2019-04-12 23:51:36 -04:00
Lioncash	be3fc0f9ff	A64: Implement SQSHLU's scalar variant	2019-04-12 19:49:33 -04:00
Lioncash	508c77c3ee	A64: Implement SQSHLU's vector variant ... The vector shift by immediate category is now fully implemented.	2019-04-12 19:45:20 -04:00
Lioncash	31867874de	frontend/ir_emitter: Add opcodes for signed saturated left shifts with unsigned saturation	2019-04-12 19:44:23 -04:00
Merry	5e77d110cc	Merge pull request #472 from lioncash/exception ... general: Mark hash functions as noexcept	2019-04-12 21:33:47 +01:00
Merry	ed8c218076	Merge pull request #471 from lioncash/sqrdmulh ... A64: Implement SQRDMULH's scalar vector variant	2019-04-12 20:29:22 +01:00