RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-19 02:52:53 +02:00
Code Issues Projects Releases Wiki Activity
13b9387d93
llvm-mirror/lib/Target/AArch64/AArch64.td

608 lines
28 KiB
TableGen
Raw Normal View History

AArch64/ARM64: move ARM64 into AArch64's place This commit starts with a "git mv ARM64 AArch64" and continues out from there, renaming the C++ classes, intrinsics, and other target-local objects for consistency. "ARM64" test directories are also moved, and tests that began their life in ARM64 use an arm64 triple, those from AArch64 use an aarch64 triple. Both should be equivalent though. This finishes the AArch64 merge, and everyone should feel free to continue committing as normal now. llvm-svn: 209577
2014-05-24 14:50:23 +02:00
//=- AArch64.td - Describe the AArch64 Target Machine --------*- tablegen -*-=//
ARM64: initial backend import This adds a second implementation of the AArch64 architecture to LLVM, accessible in parallel via the "arm64" triple. The plan over the coming weeks & months is to merge the two into a single backend, during which time thorough code review should naturally occur. Everything will be easier with the target in-tree though, hence this commit. llvm-svn: 205090
2014-03-29 11:18:08 +01:00
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
//
//===----------------------------------------------------------------------===//
//===----------------------------------------------------------------------===//
Test commit after access grant. Modified comment by adding a period. llvm-svn: 272808
2016-06-15 19:24:52 +02:00
// Target-independent interfaces which we are implementing.
ARM64: initial backend import This adds a second implementation of the AArch64 architecture to LLVM, accessible in parallel via the "arm64" triple. The plan over the coming weeks & months is to merge the two into a single backend, during which time thorough code review should naturally occur. Everything will be easier with the target in-tree though, hence this commit. llvm-svn: 205090
2014-03-29 11:18:08 +01:00
//===----------------------------------------------------------------------===//
include "llvm/Target/Target.td"
//===----------------------------------------------------------------------===//
AArch64/ARM64: move ARM64 into AArch64's place This commit starts with a "git mv ARM64 AArch64" and continues out from there, renaming the C++ classes, intrinsics, and other target-local objects for consistency. "ARM64" test directories are also moved, and tests that began their life in ARM64 use an arm64 triple, those from AArch64 use an aarch64 triple. Both should be equivalent though. This finishes the AArch64 merge, and everyone should feel free to continue committing as normal now. llvm-svn: 209577
2014-05-24 14:50:23 +02:00
// AArch64 Subtarget features.
ARM64: initial backend import This adds a second implementation of the AArch64 architecture to LLVM, accessible in parallel via the "arm64" triple. The plan over the coming weeks & months is to merge the two into a single backend, during which time thorough code review should naturally occur. Everything will be easier with the target in-tree though, hence this commit. llvm-svn: 205090
2014-03-29 11:18:08 +01:00
//
[ARM64] Port over missing subtarget features, and CPU definitions from AArch64. llvm-svn: 206198
2014-04-14 19:38:00 +02:00
def FeatureFPARMv8 : SubtargetFeature<"fp-armv8", "HasFPARMv8", "true",
[ARM64] Ports the Cortex-A53 Machine Model description from AArch64. Summary: This port includes the rudimentary latencies that were provided for the Cortex-A53 Machine Model in the AArch64 backend. It also changes the SchedAlias for COPY in the Cyclone model to an explicit WriteRes mapping to avoid conflicts in other subtargets. Differential Revision: http://reviews.llvm.org/D3427 Patch by Dave Estes <cestes@codeaurora.org>! llvm-svn: 206652
2014-04-18 23:22:04 +02:00
"Enable ARMv8 FP">;
[ARM64] Port over missing subtarget features, and CPU definitions from AArch64. llvm-svn: 206198
2014-04-14 19:38:00 +02:00
def FeatureNEON : SubtargetFeature<"neon", "HasNEON", "true",
"Enable Advanced SIMD instructions", [FeatureFPARMv8]>;
[ARM][AArch64] Armv8.4-A Enablement Initial patch adding assembly support for Armv8.4-A. Besides adding v8.4 as a supported architecture to the usual places, this also adds target features for the different crypto algorithms. Armv8.4-A introduced new crypto algorithms, made them optional, and allows different combinations: - none of the v8.4 crypto functions are supported, which is independent of the implementation of the Armv8.0 SHA1 and SHA2 instructions. - the v8.4 SHA512 and SHA3 support is implemented, in this case the Armv8.0 SHA1 and SHA2 instructions must also be implemented. - the v8.4 SM3 and SM4 support is implemented, which is independent of the implementation of the Armv8.0 SHA1 and SHA2 instructions. - all of the v8.4 crypto functions are supported, in this case the Armv8.0 SHA1 and SHA2 instructions must also be implemented. The v8.4 crypto instructions are added to AArch64 only, and not AArch32, and are made optional extensions to Armv8.2-A. The user-facing Clang options will map on these new target features, their naming will be compatible with GCC and added in follow-up patches. The Armv8.4-A instruction sets can be downloaded here: https://developer.arm.com/products/architecture/a-profile/exploration-tools Differential Revision: https://reviews.llvm.org/D48625 llvm-svn: 335953
2018-06-29 10:43:19 +02:00
def FeatureSM4 : SubtargetFeature<
"sm4", "HasSM4", "true",
"Enable SM3 and SM4 support", [FeatureNEON]>;
def FeatureSHA2 : SubtargetFeature<
"sha2", "HasSHA2", "true",
"Enable SHA1 and SHA256 support", [FeatureNEON]>;
def FeatureSHA3 : SubtargetFeature<
"sha3", "HasSHA3", "true",
"Enable SHA512 and SHA3 support", [FeatureNEON, FeatureSHA2]>;
def FeatureAES : SubtargetFeature<
"aes", "HasAES", "true",
"Enable AES support", [FeatureNEON]>;
// Crypto has been split up and any combination is now valid (see the
// crypto defintions above). Also, crypto is now context sensitive:
// it has a different meaning for e.g. Armv8.4 than it has for Armv8.2.
// Therefore, we rely on Clang, the user interacing tool, to pass on the
// appropriate crypto options. But here in the backend, crypto has very little
// meaning anymore. We kept the Crypto defintion here for backward
// compatibility, and now imply features SHA2 and AES, which was the
// "traditional" meaning of Crypto.
[ARM64] Port over missing subtarget features, and CPU definitions from AArch64. llvm-svn: 206198
2014-04-14 19:38:00 +02:00
def FeatureCrypto : SubtargetFeature<"crypto", "HasCrypto", "true",
[ARM][AArch64] Armv8.4-A Enablement Initial patch adding assembly support for Armv8.4-A. Besides adding v8.4 as a supported architecture to the usual places, this also adds target features for the different crypto algorithms. Armv8.4-A introduced new crypto algorithms, made them optional, and allows different combinations: - none of the v8.4 crypto functions are supported, which is independent of the implementation of the Armv8.0 SHA1 and SHA2 instructions. - the v8.4 SHA512 and SHA3 support is implemented, in this case the Armv8.0 SHA1 and SHA2 instructions must also be implemented. - the v8.4 SM3 and SM4 support is implemented, which is independent of the implementation of the Armv8.0 SHA1 and SHA2 instructions. - all of the v8.4 crypto functions are supported, in this case the Armv8.0 SHA1 and SHA2 instructions must also be implemented. The v8.4 crypto instructions are added to AArch64 only, and not AArch32, and are made optional extensions to Armv8.2-A. The user-facing Clang options will map on these new target features, their naming will be compatible with GCC and added in follow-up patches. The Armv8.4-A instruction sets can be downloaded here: https://developer.arm.com/products/architecture/a-profile/exploration-tools Differential Revision: https://reviews.llvm.org/D48625 llvm-svn: 335953
2018-06-29 10:43:19 +02:00
"Enable cryptographic instructions", [FeatureNEON, FeatureSHA2, FeatureAES]>;
[ARM64] Port over missing subtarget features, and CPU definitions from AArch64. llvm-svn: 206198
2014-04-14 19:38:00 +02:00
[ARM64] Support crc predicate on ARM64. According to the specification, CRC is an optional extension of the architecture. llvm-svn: 207214
2014-04-25 11:25:42 +02:00
def FeatureCRC : SubtargetFeature<"crc", "HasCRC", "true",
"Enable ARMv8 CRC-32 checksum instructions">;
RAS extensions are part of ARMv8.2-A. This change enables them by introducing a new instruction to ARM and AArch64 targets and several system registers. Patch by: Roger Ferrer Ibanez and Oliver Stannard Differential Revision: http://reviews.llvm.org/D20282 llvm-svn: 271670
2016-06-03 16:03:27 +02:00
def FeatureRAS : SubtargetFeature<"ras", "HasRAS", "true",
"Enable ARMv8 Reliability, Availability and Serviceability Extensions">;
[AArch64] Refactor LSE support as feature separate from V8.1a support. Summary: This is preparation for ThunderX processors that have Large System Extension (LSE) atomic instructions, but not the other instructions introduced by V8.1a. This will mimic changes to GCC as described here: https://gcc.gnu.org/ml/gcc-patches/2015-06/msg00388.html LSE instructions are: LD/ST<op>, CAS*, SWP Reviewers: t.p.northover, echristo, jmolloy, rengolin Subscribers: aemerson, mehdi_amini Differential Revision: https://reviews.llvm.org/D26621 llvm-svn: 288279
2016-11-30 23:25:24 +01:00
def FeatureLSE : SubtargetFeature<"lse", "HasLSE", "true",
"Enable ARMv8.1 Large System Extension (LSE) atomic instructions">;
[AArch64] Falkor supports Rounding Double Multiply Add/Subtract instructions. Falkor only partially implements the ARMv8.1a extensions, so this patch refactors the support for the SQRDML[A|S]H instruction into a separate feature. Differential Revision: https://reviews.llvm.org/D28681 llvm-svn: 292142
2017-01-16 17:28:43 +01:00
def FeatureRDM : SubtargetFeature<"rdm", "HasRDM", "true",
"Enable ARMv8.1 Rounding Double Multiply Add/Subtract instructions">;
[AArch64] Lower READCYCLECOUNTER using MRS PMCCTNR_EL0. This matches the ARM behavior. In both cases, the register is part of the optional Performance Monitors extension, so, add the feature, and enable it for the A-class processors we support. Differential Revision: http://reviews.llvm.org/D12425 llvm-svn: 246555
2015-09-01 18:23:45 +02:00
def FeaturePerfMon : SubtargetFeature<"perfmon", "HasPerfMon", "true",
"Enable ARMv8 PMUv3 Performance Monitors extension">;
[AArch64] Add subtarget features for ARMv8.2-A This adds subtarget features for ARMv8.2-A, which builds on (and requires the features from) ARMv8.1-A. Most assembler-visible features of ARMv8.2-A are system instructions, and are all required parts of the architecture, so just depend on the HasV8_2aOps subtarget feature. There is also one large, optional feature, which adds 16-bit floating point versions of all existing floating-point instructions (VFP and SIMD), this is represented by the FeatureFullFP16 subtarget feature. Differential Revision: http://reviews.llvm.org/D15013 llvm-svn: 254154
2015-11-26 16:23:32 +01:00
def FeatureFullFP16 : SubtargetFeature<"fullfp16", "HasFullFP16", "true",
"Full FP16", [FeatureFPARMv8]>;
[ARM/AArch64] Support FP16 +fp16fml instructions Add +fp16fml feature for new FP16 instructions, which are a mandatory part of FP16 from v8.4-A and an optional part of FP16 from v8.2-A. It doesn't seem to be possible to model this in LLVM, but the relationship between the options is handled by the related clang patch. In keeping with what I think is the usual practice, the fp16fml extension is accepted regardless of base architecture version. Builds on/replaces Sjoerd Meijer's patch to add these instructions at https://reviews.llvm.org/D49839. Differential Revision: https://reviews.llvm.org/D50228 llvm-svn: 340013
2018-08-17 13:29:49 +02:00
def FeatureFP16FML : SubtargetFeature<"fp16fml", "HasFP16FML", "true",
"Enable FP16 FML instructions", [FeatureFullFP16]>;
[AArch64] Add ARMv8.2-A Statistical Profiling Extension The Statistical Profiling Extension is an optional extension to ARMv8.2-A. Since it is an optional extension, I have added the FeatureSPE subtarget feature to control it. The assembler-visible parts of this extension are the new "psb csync" instruction, which is equivalent to "hint #17", and a number of system registers. Differential Revision: http://reviews.llvm.org/D15021 llvm-svn: 254401
2015-12-01 11:48:51 +01:00
def FeatureSPE : SubtargetFeature<"spe", "HasSPE", "true",
"Enable Statistical Profiling extension">;
[AArch64] Add an SVE target feature to the backend and TargetParser. The feature will be used properly once assembler/disassembler support begins to land. llvm-svn: 307917
2017-07-13 17:19:56 +02:00
def FeatureSVE : SubtargetFeature<"sve", "HasSVE", "true",
"Enable Scalable Vector Extension (SVE) instructions">;
ARM64: initial backend import This adds a second implementation of the AArch64 architecture to LLVM, accessible in parallel via the "arm64" triple. The plan over the coming weeks & months is to merge the two into a single backend, during which time thorough code review should naturally occur. Everything will be easier with the target in-tree though, hence this commit. llvm-svn: 205090
2014-03-29 11:18:08 +01:00
/// Cyclone has register move instructions which are "free".
def FeatureZCRegMove : SubtargetFeature<"zcm", "HasZeroCycleRegMove", "true",
Fix typo. llvm-svn: 208006
2014-05-05 23:50:57 +02:00
"Has zero-cycle register moves">;
ARM64: initial backend import This adds a second implementation of the AArch64 architecture to LLVM, accessible in parallel via the "arm64" triple. The plan over the coming weeks & months is to merge the two into a single backend, during which time thorough code review should naturally occur. Everything will be easier with the target in-tree though, hence this commit. llvm-svn: 205090
2014-03-29 11:18:08 +01:00
/// Cyclone has instructions which zero registers for "free".
def FeatureZCZeroing : SubtargetFeature<"zcz", "HasZeroCycleZeroing", "true",
"Has zero-cycle zeroing instructions">;
AArch64: work around how Cyclone handles "movi.2d vD, #0". For Cylone, the instruction "movi.2d vD, #0" is executed incorrectly in some rare circumstances. Work around the issue conservatively by avoiding the instruction entirely. This patch changes CodeGen so that problematic instructions are never generated, and the AsmParser so that an equivalent instruction is used (with a warning). llvm-svn: 320965
2017-12-18 11:36:00 +01:00
/// ... but the floating-point version doesn't quite work in rare cases on older
/// CPUs.
def FeatureZCZeroingFPWorkaround : SubtargetFeature<"zcz-fp-workaround",
"HasZeroCycleZeroingFPWorkaround", "true",
"The zero-cycle floating-point zeroing instruction has a bug">;
[AArch64] Define subtarget feature strict-align. This commit defines subtarget feature strict-align and uses it instead of cl::opt -aarch64-strict-align to decide whether strict alignment should be forced. rdar://problem/21529937 llvm-svn: 243516
2015-07-29 16:17:26 +02:00
def FeatureStrictAlign : SubtargetFeature<"strict-align",
"StrictAlign", "true",
"Disallow all unaligned memory "
"access">;
[AArch64] Support reserving x1-7 registers. Summary: Reserving registers x1-7 is used to support CONFIG_ARM64_LSE_ATOMICS in Linux kernel. This change adds support for reserving registers x1 through x7. Reviewers: javed.absar, phosek, srhines, nickdesaulniers, efriedma Reviewed By: nickdesaulniers, efriedma Subscribers: niravd, jfb, manojgupta, nickdesaulniers, jyknight, efriedma, kristof.beyls, llvm-commits Differential Revision: https://reviews.llvm.org/D48580 llvm-svn: 341706
2018-09-07 22:58:57 +02:00
foreach i = {1-7,18,20} in
def FeatureReserveX#i : SubtargetFeature<"reserve-x"#i, "ReserveXRegister["#i#"]", "true",
"Reserve X"#i#", making it unavailable "
"as a GPR">;
[AArch64] Support reserving x20 register Register x20 is a callee-saved register which may be used for other purposes in certain contexts, for example to hold special variables within the kernel. This change adds support for reserving this register both to frontend and backend to make this register usable for these purposes. Differential Revision: https://reviews.llvm.org/D46552 llvm-svn: 334531
2018-06-12 22:00:50 +02:00
[AArch64] Support adding X[8-15,18] registers as CSRs. Summary: Specifying X[8-15,18] registers as callee-saved is used to support CONFIG_ARM64_LSE_ATOMICS in Linux kernel. As part of this patch we: - use custom CSR list/mask when user specifies custom CSRs - update Machine Register Info's list of CSRs with additional custom CSRs in LowerCall and LowerFormalArguments. Reviewers: srhines, nickdesaulniers, efriedma, javed.absar Reviewed By: nickdesaulniers Subscribers: kristof.beyls, jfb, llvm-commits Differential Revision: https://reviews.llvm.org/D52216 llvm-svn: 342824
2018-09-23 00:17:50 +02:00
foreach i = {8-15,18} in
def FeatureCallSavedX#i : SubtargetFeature<"call-saved-x"#i,
"CustomCallSavedXRegs["#i#"]", "true", "Make X"#i#" callee saved.">;
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
def FeatureUseAA : SubtargetFeature<"use-aa", "UseAA", "true",
"Use alias analysis during codegen">;
def FeatureBalanceFPOps : SubtargetFeature<"balance-fp-ops", "BalanceFPOps",
"true",
"balance mix of odd and even D-registers for fp multiply(-accumulate) ops">;
def FeaturePredictableSelectIsExpensive : SubtargetFeature<
"predictable-select-expensive", "PredictableSelectIsExpensive", "true",
"Prefer likely predicted branches over selects">;
def FeatureCustomCheapAsMoveHandling : SubtargetFeature<"custom-cheap-as-move",
"CustomAsCheapAsMove", "true",
"Use custom code for TargetInstrInfo::isAsCheapAsAMove()">;
[AArch64] Add new target feature to handle cheap as move for Exynos This feature enables special handling of cheap as move in the existing custom handling specifically for Exynos processors. Differential revision: https://reviews.llvm.org/D42387 llvm-svn: 323774
2018-01-30 16:40:22 +01:00
def FeatureExynosCheapAsMoveHandling : SubtargetFeature<"exynos-cheap-as-move",
"ExynosAsCheapAsMove", "true",
"Use Exynos specific code in TargetInstrInfo::isAsCheapAsAMove()",
[FeatureCustomCheapAsMoveHandling]>;
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
def FeaturePostRAScheduler : SubtargetFeature<"use-postra-scheduler",
"UsePostRAScheduler", "true", "Schedule again after register allocation">;
def FeatureSlowMisaligned128Store : SubtargetFeature<"slow-misaligned-128store",
"Misaligned128StoreIsSlow", "true", "Misaligned 128 bit stores are slow">;
[AArch64] Rename 'no-quad-ldst-pairs' to 'slow-paired-128' In order to follow the pattern of the existing 'slow-misaligned-128store' option, rename the option 'no-quad-ldst-pairs' to 'slow-paired-128'. llvm-svn: 292954
2017-01-24 18:34:31 +01:00
def FeatureSlowPaired128 : SubtargetFeature<"slow-paired-128",
"Paired128IsSlow", "true", "Paired 128 bit loads and stores are slow">;
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
[AArch64][Falkor] Avoid generating STRQro* instructions Summary: STRQro* instructions are slower than the alternative ADD/STRQui expanded instructions on Falkor, so avoid generating them unless we're optimizing for code size. Reviewers: t.p.northover, mcrosier Subscribers: aemerson, rengolin, javed.absar, kristof.beyls, llvm-commits Differential Revision: https://reviews.llvm.org/D37020 llvm-svn: 311931
2017-08-28 22:48:43 +02:00
def FeatureSlowSTRQro : SubtargetFeature<"slow-strqro-store", "STRQroIsSlow",
"true", "STR of Q register with register offset is slow">;
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
def FeatureAlternateSExtLoadCVTF32Pattern : SubtargetFeature<
"alternate-sextload-cvt-f32-pattern", "UseAlternateSExtLoadCVTF32Pattern",
"true", "Use alternative pattern for sextload convert to f32">;
AArch64: Macrofusion: Split features, add missing combinations. AArch64InstrInfo::shouldScheduleAdjacent() determines whether two instruction can benefit from macroop fusion on apple CPUs. The list turned out to be incomplete: - the "rr" variants of the instructions were missing - even the "rs" variants can have shift value == 0 and behave like the "rr" variants This also splits the MacropFusion target feature into ArithmeticBccFusion and ArithmeticCbzFusion. Differential Revision: https://reviews.llvm.org/D25142 llvm-svn: 283243
2016-10-04 21:28:21 +02:00
def FeatureArithmeticBccFusion : SubtargetFeature<
"arith-bcc-fusion", "HasArithmeticBccFusion", "true",
"CPU fuses arithmetic+bcc operations">;
def FeatureArithmeticCbzFusion : SubtargetFeature<
"arith-cbz-fusion", "HasArithmeticCbzFusion", "true",
"CPU fuses arithmetic + cbz/cbnz operations">;
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
[AArch64] Add new target feature to fuse address generation with load or store This feature enables the fusion of the address generation and a corresponding load or store together. Differential revision: https://reviews.llvm.org/D42393 llvm-svn: 323782
2018-01-30 17:28:01 +01:00
def FeatureFuseAddress : SubtargetFeature<
"fuse-address", "HasFuseAddress", "true",
"CPU fuses address generation and memory operations">;
[AArch64] Add new subtarget feature to fuse AES crypto operations This feature enables the fusion of such operations on Cortex A57, as recommended in its Software Optimisation Guide, section 4.13, and on Exynos M1. Differential revision: https://reviews.llvm.org/D28491 llvm-svn: 293738
2017-02-01 03:54:39 +01:00
def FeatureFuseAES : SubtargetFeature<
"fuse-aes", "HasFuseAES", "true",
"CPU fuses AES crypto operations">;
AArch64: Add FuseCryptoEOR fusion rules There's some additional rules available on newer apple CPUs. rdar://41235346 llvm-svn: 342590
2018-09-19 22:50:51 +02:00
def FeatureFuseCryptoEOR : SubtargetFeature<
"fuse-crypto-eor", "HasFuseCryptoEOR", "true",
"CPU fuses AES/PMULL and EOR operations">;
[PATCH] [AArch64] Add new target feature to fuse conditional select This feature enables the fusion of the comparison and the conditional select instructions together. Differential revision: https://reviews.llvm.org/D42392 llvm-svn: 325939
2018-02-23 20:27:43 +01:00
def FeatureFuseCCSelect : SubtargetFeature<
"fuse-csel", "HasFuseCCSelect", "true",
"CPU fuses conditional select operations">;
[AArch64] Add new target feature to fuse literal generation This feature enables the fusion of such operations on Cortex A57, as recommended in its Software Optimisation Guide, sections 4.14 and 4.15. Differential revision: https://reviews.llvm.org/D28698 llvm-svn: 293739
2017-02-01 03:54:42 +01:00
def FeatureFuseLiterals : SubtargetFeature<
"fuse-literals", "HasFuseLiterals", "true",
"CPU fuses literal generation operations">;
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
def FeatureDisableLatencySchedHeuristic : SubtargetFeature<
"disable-latency-sched-heuristic", "DisableLatencySchedHeuristic", "true",
"Disable latency scheduling heuristic">;
[AArch64] Assembler support for v8.3 RCpc Added assembler and disassembler support for the new Release Consistent processor consistent instructions, introduced with ARM v8.3-A for AArch64. Differential Revision: https://reviews.llvm.org/D36522 llvm-svn: 310575
2017-08-10 11:52:55 +02:00
def FeatureRCPC : SubtargetFeature<"rcpc", "HasRCPC", "true",
"Enable support for RCPC extension">;
[AArch64] Optionally use the Newton series for reciprocal estimation Add support for estimating the square root or its reciprocal and division or reciprocal using the combiner generic Newton series. Differential revision: https://reviews.llvm.org/D25291 llvm-svn: 284986
2016-10-24 18:14:58 +02:00
def FeatureUseRSqrt : SubtargetFeature<
"use-reciprocal-square-root", "UseRSqrt", "true",
"Use the reciprocal square root approximation">;
[AArch64] [Assembler] option to disable negative immediate conversions Summary: Similar to the ARM target in https://reviews.llvm.org/rL298380, this patch adds identical infrastructure for disabling negative immediate conversions, and converts the existing aliases to the new infrastucture. Reviewers: rengolin, javed.absar, olista01, SjoerdMeijer, samparker Reviewed By: samparker Subscribers: samparker, aemerson, llvm-commits Differential Revision: https://reviews.llvm.org/D31243 llvm-svn: 298908
2017-03-28 12:02:56 +02:00
[AArch64] Assembler support for the ARMv8.2a dot product instructions Dot product is an optional ARMv8.2a extension, see also the public architecture specification here: https://developer.arm.com/products/architecture/a-profile/exploration-tools. This patch adds AArch64 assembler support for these dot product instructions. Differential Revision: https://reviews.llvm.org/D36515 llvm-svn: 310480
2017-08-09 16:59:54 +02:00
def FeatureDotProd : SubtargetFeature<
"dotprod", "HasDotProd", "true",
"Enable dot product support">;
[AArch64] [Assembler] option to disable negative immediate conversions Summary: Similar to the ARM target in https://reviews.llvm.org/rL298380, this patch adds identical infrastructure for disabling negative immediate conversions, and converts the existing aliases to the new infrastucture. Reviewers: rengolin, javed.absar, olista01, SjoerdMeijer, samparker Reviewed By: samparker Subscribers: samparker, aemerson, llvm-commits Differential Revision: https://reviews.llvm.org/D31243 llvm-svn: 298908
2017-03-28 12:02:56 +02:00
def FeatureNoNegativeImmediates : SubtargetFeature<"no-neg-immediates",
"NegativeImmediates", "false",
"Convert immediates and instructions "
"to their negated or complemented "
"equivalent when the immediate does "
"not fit in the encoding.">;
[AArch64] Add new subtarget feature to fold LSL into address mode. Summary: This feature enables folding of logical shift operations of up to 3 places into addressing mode on Kryo and Falkor that have a fastpath LSL. Reviewers: mcrosier, rengolin, t.p.northover Subscribers: junbuml, gberry, llvm-commits, aemerson Differential Revision: https://reviews.llvm.org/D31113 llvm-svn: 299240
2017-03-31 20:16:53 +02:00
def FeatureLSLFast : SubtargetFeature<
"lsl-fast", "HasLSLFast", "true",
"CPU has a fastpath logical shift of up to 3 places">;
[AArch64] Test patch Fix formatting by adding a missing blank line to test new network setup. llvm-svn: 320760
2017-12-15 00:06:18 +01:00
[AArch64] Enable aggressive FMA on T99 and provide AArch64 options for others. This patch enables aggressive FMA by default on T99, and provides a -mllvm option to enable the same on other AArch64 micro-arch's (-mllvm -aarch64-enable-aggressive-fma). Test case demonstrating the effects on T99 is included. Patch by: steleman (Stefan Teleman) Differential Revision: https://reviews.llvm.org/D40696 llvm-svn: 323474
2018-01-25 22:55:39 +01:00
def FeatureAggressiveFMA :
SubtargetFeature<"aggressive-fma",
"HasAggressiveFMA",
"true",
"Enable Aggressive FMA for floating-point.">;
[AArch64][v8.5A] Add PSTATE manipulation instructions XAFlag and AXFlag These new instructions manipluate the NZCV bits, to convert between the regular Arm floating-point comare format and an alternative format. Patch by Pablo Barrio! Differential revision: https://reviews.llvm.org/D52473 llvm-svn: 343187
2018-09-27 11:11:27 +02:00
def FeatureAltFPCmp : SubtargetFeature<"altnzcv", "HasAlternativeNZCV", "true",
"Enable alternative NZCV format for floating point comparisons">;
[AArch64][v8.5A] Add FRINT[32,64][Z,X] instructions These are some new variants of the "Floating-point Round to Integral" family of instructions, which round to the nearest floating-point value which fits in a 32- or 64-bit integer. Patch by Pablo Barrio! Differential revision: https://reviews.llvm.org/D52475 llvm-svn: 343209
2018-09-27 15:32:06 +02:00
def FeatureFRInt3264 : SubtargetFeature<"fptoint", "HasFRInt3264", "true",
"Enable FRInt[32|64][Z|X] instructions that round a floating-point number to "
"an integer (in FP format) forcing it to fit into a 32- or 64-bit int" >;
[AArch64][v8.5A] Add speculation barrier to AArch64 instruction set This is a new barrier which limits speculative execution of the instructions following it. Patch by Pablo Barrio! Differential revision: https://reviews.llvm.org/D52476 llvm-svn: 343211
2018-09-27 15:39:06 +02:00
def FeatureSpecCtrl : SubtargetFeature<"specctrl", "HasSpecCtrl", "true",
"Enable speculation control barrier" >;
[AArch64][v8.5A] Add prediction invalidation instructions to AArch64 This adds new system instructions which act as barriers to speculative execution based on earlier execution within a particular execution context. Patch by Pablo Barrio! Differential revision: https://reviews.llvm.org/D52479 llvm-svn: 343214
2018-09-27 15:47:40 +02:00
def FeaturePredCtrl : SubtargetFeature<"predctrl", "HasPredCtrl", "true",
"Enable execution and data prediction invalidation instructions" >;
[AArch64] Rename v8.1a from "extension" to "architecture" v8.1a is renamed to architecture, accordingly to approaches in ARM backend. Excess generic cpu is removed. Intended use: "generic" cpu with "v8.1a" subtarget feature Reviewers: jmolloy Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D8766 llvm-svn: 233810
2015-04-01 16:49:29 +02:00
//===----------------------------------------------------------------------===//
// Architectures.
//
def HasV8_1aOps : SubtargetFeature<"v8.1a", "HasV8_1aOps", "true",
[AArch64] Falkor supports Rounding Double Multiply Add/Subtract instructions. Falkor only partially implements the ARMv8.1a extensions, so this patch refactors the support for the SQRDML[A|S]H instruction into a separate feature. Differential Revision: https://reviews.llvm.org/D28681 llvm-svn: 292142
2017-01-16 17:28:43 +01:00
"Support ARM v8.1a instructions", [FeatureCRC, FeatureLSE, FeatureRDM]>;
[AArch64] Rename v8.1a from "extension" to "architecture" v8.1a is renamed to architecture, accordingly to approaches in ARM backend. Excess generic cpu is removed. Intended use: "generic" cpu with "v8.1a" subtarget feature Reviewers: jmolloy Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D8766 llvm-svn: 233810
2015-04-01 16:49:29 +02:00
[AArch64] Add subtarget features for ARMv8.2-A This adds subtarget features for ARMv8.2-A, which builds on (and requires the features from) ARMv8.1-A. Most assembler-visible features of ARMv8.2-A are system instructions, and are all required parts of the architecture, so just depend on the HasV8_2aOps subtarget feature. There is also one large, optional feature, which adds 16-bit floating point versions of all existing floating-point instructions (VFP and SIMD), this is represented by the FeatureFullFP16 subtarget feature. Differential Revision: http://reviews.llvm.org/D15013 llvm-svn: 254154
2015-11-26 16:23:32 +01:00
def HasV8_2aOps : SubtargetFeature<"v8.2a", "HasV8_2aOps", "true",
RAS extensions are part of ARMv8.2-A. This change enables them by introducing a new instruction to ARM and AArch64 targets and several system registers. Patch by: Roger Ferrer Ibanez and Oliver Stannard Differential Revision: http://reviews.llvm.org/D20282 llvm-svn: 271670
2016-06-03 16:03:27 +02:00
"Support ARM v8.2a instructions", [HasV8_1aOps, FeatureRAS]>;
[AArch64] Add subtarget features for ARMv8.2-A This adds subtarget features for ARMv8.2-A, which builds on (and requires the features from) ARMv8.1-A. Most assembler-visible features of ARMv8.2-A are system instructions, and are all required parts of the architecture, so just depend on the HasV8_2aOps subtarget feature. There is also one large, optional feature, which adds 16-bit floating point versions of all existing floating-point instructions (VFP and SIMD), this is represented by the FeatureFullFP16 subtarget feature. Differential Revision: http://reviews.llvm.org/D15013 llvm-svn: 254154
2015-11-26 16:23:32 +01:00
[ARM][AArch64] ARMv8.3-A enablement The beta ARMv8.3 ISA specifications have been released for AArch64 and AArch32, these can be found at: https://developer.arm.com/products/architecture/a-profile/exploration-tools An introduction to this architecture update can be found at: https://community.arm.com/processors/b/blog/posts/armv8-a-architecture-2016-additions This patch is the first in a series which will add ARM v8.3-A support in LLVM and Clang. It adds the necessary changes that create targets for both the ARM and AArch64 backends. Differential Revision: https://reviews.llvm.org/D36514 llvm-svn: 310561
2017-08-10 11:41:00 +02:00
def HasV8_3aOps : SubtargetFeature<"v8.3a", "HasV8_3aOps", "true",
[AArch64] Assembler support for v8.3 RCpc Added assembler and disassembler support for the new Release Consistent processor consistent instructions, introduced with ARM v8.3-A for AArch64. Differential Revision: https://reviews.llvm.org/D36522 llvm-svn: 310575
2017-08-10 11:52:55 +02:00
"Support ARM v8.3a instructions", [HasV8_2aOps, FeatureRCPC]>;
[ARM][AArch64] ARMv8.3-A enablement The beta ARMv8.3 ISA specifications have been released for AArch64 and AArch32, these can be found at: https://developer.arm.com/products/architecture/a-profile/exploration-tools An introduction to this architecture update can be found at: https://community.arm.com/processors/b/blog/posts/armv8-a-architecture-2016-additions This patch is the first in a series which will add ARM v8.3-A support in LLVM and Clang. It adds the necessary changes that create targets for both the ARM and AArch64 backends. Differential Revision: https://reviews.llvm.org/D36514 llvm-svn: 310561
2017-08-10 11:41:00 +02:00
[ARM][AArch64] Armv8.4-A Enablement Initial patch adding assembly support for Armv8.4-A. Besides adding v8.4 as a supported architecture to the usual places, this also adds target features for the different crypto algorithms. Armv8.4-A introduced new crypto algorithms, made them optional, and allows different combinations: - none of the v8.4 crypto functions are supported, which is independent of the implementation of the Armv8.0 SHA1 and SHA2 instructions. - the v8.4 SHA512 and SHA3 support is implemented, in this case the Armv8.0 SHA1 and SHA2 instructions must also be implemented. - the v8.4 SM3 and SM4 support is implemented, which is independent of the implementation of the Armv8.0 SHA1 and SHA2 instructions. - all of the v8.4 crypto functions are supported, in this case the Armv8.0 SHA1 and SHA2 instructions must also be implemented. The v8.4 crypto instructions are added to AArch64 only, and not AArch32, and are made optional extensions to Armv8.2-A. The user-facing Clang options will map on these new target features, their naming will be compatible with GCC and added in follow-up patches. The Armv8.4-A instruction sets can be downloaded here: https://developer.arm.com/products/architecture/a-profile/exploration-tools Differential Revision: https://reviews.llvm.org/D48625 llvm-svn: 335953
2018-06-29 10:43:19 +02:00
def HasV8_4aOps : SubtargetFeature<"v8.4a", "HasV8_4aOps", "true",
"Support ARM v8.4a instructions", [HasV8_3aOps, FeatureDotProd]>;
[AArch64][v8.5A] Add FRINT[32,64][Z,X] instructions These are some new variants of the "Floating-point Round to Integral" family of instructions, which round to the nearest floating-point value which fits in a 32- or 64-bit integer. Patch by Pablo Barrio! Differential revision: https://reviews.llvm.org/D52475 llvm-svn: 343209
2018-09-27 15:32:06 +02:00
def HasV8_5aOps : SubtargetFeature<
"v8.5a", "HasV8_5aOps", "true", "Support ARM v8.5a instructions",
[AArch64][v8.5A] Add prediction invalidation instructions to AArch64 This adds new system instructions which act as barriers to speculative execution based on earlier execution within a particular execution context. Patch by Pablo Barrio! Differential revision: https://reviews.llvm.org/D52479 llvm-svn: 343214
2018-09-27 15:47:40 +02:00
[HasV8_4aOps, FeatureAltFPCmp, FeatureFRInt3264, FeatureSpecCtrl,
FeaturePredCtrl]
[AArch64][v8.5A] Add FRINT[32,64][Z,X] instructions These are some new variants of the "Floating-point Round to Integral" family of instructions, which round to the nearest floating-point value which fits in a 32- or 64-bit integer. Patch by Pablo Barrio! Differential revision: https://reviews.llvm.org/D52475 llvm-svn: 343209
2018-09-27 15:32:06 +02:00
>;
[ARM/AArch64][v8.5A] Add Armv8.5-A target This patch allows targeting Armv8.5-A, adding the architecture to tablegen and setting the options to be identical to Armv8.4-A for the time being. Subsequent patches will add support for the different features included in the Armv8.5-A Reference Manual. Patch by Pablo Barrio! Differential revision: https://reviews.llvm.org/D52470 llvm-svn: 343102
2018-09-26 14:48:21 +02:00
ARM64: initial backend import This adds a second implementation of the AArch64 architecture to LLVM, accessible in parallel via the "arm64" triple. The plan over the coming weeks & months is to merge the two into a single backend, during which time thorough code review should naturally occur. Everything will be easier with the target in-tree though, hence this commit. llvm-svn: 205090
2014-03-29 11:18:08 +01:00
//===----------------------------------------------------------------------===//
// Register File Description
//===----------------------------------------------------------------------===//
AArch64/ARM64: move ARM64 into AArch64's place This commit starts with a "git mv ARM64 AArch64" and continues out from there, renaming the C++ classes, intrinsics, and other target-local objects for consistency. "ARM64" test directories are also moved, and tests that began their life in ARM64 use an arm64 triple, those from AArch64 use an aarch64 triple. Both should be equivalent though. This finishes the AArch64 merge, and everyone should feel free to continue committing as normal now. llvm-svn: 209577
2014-05-24 14:50:23 +02:00
include "AArch64RegisterInfo.td"
Re-commit: [globalisel] Tablegen-erate current Register Bank Information Summary: Adds a RegisterBank tablegen class that can be used to declare the register banks and an associated tablegen pass to generate the necessary code. Changes since first commit attempt: * Added missing guards * Added more missing guards * Found and fixed a use-after-free bug involving Twine locals Reviewers: t.p.northover, ab, rovka, qcolombet Reviewed By: qcolombet Subscribers: aditya_nandakumar, rengolin, kristof.beyls, vkalintiris, mgorny, dberris, llvm-commits, rovka Differential Revision: https://reviews.llvm.org/D27338 llvm-svn: 292478
2017-01-19 12:15:55 +01:00
include "AArch64RegisterBanks.td"
AArch64/ARM64: move ARM64 into AArch64's place This commit starts with a "git mv ARM64 AArch64" and continues out from there, renaming the C++ classes, intrinsics, and other target-local objects for consistency. "ARM64" test directories are also moved, and tests that began their life in ARM64 use an arm64 triple, those from AArch64 use an aarch64 triple. Both should be equivalent though. This finishes the AArch64 merge, and everyone should feel free to continue committing as normal now. llvm-svn: 209577
2014-05-24 14:50:23 +02:00
include "AArch64CallingConvention.td"
ARM64: initial backend import This adds a second implementation of the AArch64 architecture to LLVM, accessible in parallel via the "arm64" triple. The plan over the coming weeks & months is to merge the two into a single backend, during which time thorough code review should naturally occur. Everything will be easier with the target in-tree though, hence this commit. llvm-svn: 205090
2014-03-29 11:18:08 +01:00
//===----------------------------------------------------------------------===//
// Instruction Descriptions
//===----------------------------------------------------------------------===//
AArch64/ARM64: move ARM64 into AArch64's place This commit starts with a "git mv ARM64 AArch64" and continues out from there, renaming the C++ classes, intrinsics, and other target-local objects for consistency. "ARM64" test directories are also moved, and tests that began their life in ARM64 use an arm64 triple, those from AArch64 use an aarch64 triple. Both should be equivalent though. This finishes the AArch64 merge, and everyone should feel free to continue committing as normal now. llvm-svn: 209577
2014-05-24 14:50:23 +02:00
include "AArch64Schedule.td"
include "AArch64InstrInfo.td"
ARM64: initial backend import This adds a second implementation of the AArch64 architecture to LLVM, accessible in parallel via the "arm64" triple. The plan over the coming weeks & months is to merge the two into a single backend, during which time thorough code review should naturally occur. Everything will be easier with the target in-tree though, hence this commit. llvm-svn: 205090
2014-03-29 11:18:08 +01:00
AArch64/ARM64: move ARM64 into AArch64's place This commit starts with a "git mv ARM64 AArch64" and continues out from there, renaming the C++ classes, intrinsics, and other target-local objects for consistency. "ARM64" test directories are also moved, and tests that began their life in ARM64 use an arm64 triple, those from AArch64 use an aarch64 triple. Both should be equivalent though. This finishes the AArch64 merge, and everyone should feel free to continue committing as normal now. llvm-svn: 209577
2014-05-24 14:50:23 +02:00
def AArch64InstrInfo : InstrInfo;
ARM64: initial backend import This adds a second implementation of the AArch64 architecture to LLVM, accessible in parallel via the "arm64" triple. The plan over the coming weeks & months is to merge the two into a single backend, during which time thorough code review should naturally occur. Everything will be easier with the target in-tree though, hence this commit. llvm-svn: 205090
2014-03-29 11:18:08 +01:00
AArch64: TableGenerate system instruction operands. The way the named arguments for various system instructions are handled at the moment has a few problems: - Large-scale duplication between AArch64BaseInfo.h and AArch64BaseInfo.cpp - That weird Mapping class that I have no idea what I was on when I thought it was a good idea. - Searches are performed linearly through the entire list. - We print absolutely all registers in upper-case, even though some are canonically mixed case (SPSel for example). - The ARM ARM specifies sysregs in terms of 5 fields, but those are relegated to comments in our implementation, with a slightly opaque hex value indicating the canonical encoding LLVM will use. This adds a new TableGen backend to produce efficiently searchable tables, and switches AArch64 over to using that infrastructure. llvm-svn: 274576
2016-07-05 23:23:04 +02:00
//===----------------------------------------------------------------------===//
// Named operands for MRS/MSR/TLBI/...
//===----------------------------------------------------------------------===//
include "AArch64SystemOperands.td"
ARM64: initial backend import This adds a second implementation of the AArch64 architecture to LLVM, accessible in parallel via the "arm64" triple. The plan over the coming weeks & months is to merge the two into a single backend, during which time thorough code review should naturally occur. Everything will be easier with the target in-tree though, hence this commit. llvm-svn: 205090
2014-03-29 11:18:08 +01:00
//===----------------------------------------------------------------------===//
AArch64/ARM64: move ARM64 into AArch64's place This commit starts with a "git mv ARM64 AArch64" and continues out from there, renaming the C++ classes, intrinsics, and other target-local objects for consistency. "ARM64" test directories are also moved, and tests that began their life in ARM64 use an arm64 triple, those from AArch64 use an aarch64 triple. Both should be equivalent though. This finishes the AArch64 merge, and everyone should feel free to continue committing as normal now. llvm-svn: 209577
2014-05-24 14:50:23 +02:00
// AArch64 Processors supported.
ARM64: initial backend import This adds a second implementation of the AArch64 architecture to LLVM, accessible in parallel via the "arm64" triple. The plan over the coming weeks & months is to merge the two into a single backend, during which time thorough code review should naturally occur. Everything will be easier with the target in-tree though, hence this commit. llvm-svn: 205090
2014-03-29 11:18:08 +01:00
//
AArch64/ARM64: move ARM64 into AArch64's place This commit starts with a "git mv ARM64 AArch64" and continues out from there, renaming the C++ classes, intrinsics, and other target-local objects for consistency. "ARM64" test directories are also moved, and tests that began their life in ARM64 use an arm64 triple, those from AArch64 use an aarch64 triple. Both should be equivalent though. This finishes the AArch64 merge, and everyone should feel free to continue committing as normal now. llvm-svn: 209577
2014-05-24 14:50:23 +02:00
include "AArch64SchedA53.td"
[AArch64] Basic Sched Model for Cortex-A57. Patch by Dave Estes<cestes@codeaurora.org> Differential Revision: http://reviews.llvm.org/D4008 llvm-svn: 210705
2014-06-11 23:06:56 +02:00
include "AArch64SchedA57.td"
AArch64/ARM64: move ARM64 into AArch64's place This commit starts with a "git mv ARM64 AArch64" and continues out from there, renaming the C++ classes, intrinsics, and other target-local objects for consistency. "ARM64" test directories are also moved, and tests that began their life in ARM64 use an arm64 triple, those from AArch64 use an aarch64 triple. Both should be equivalent though. This finishes the AArch64 merge, and everyone should feel free to continue committing as normal now. llvm-svn: 209577
2014-05-24 14:50:23 +02:00
include "AArch64SchedCyclone.td"
[AArch64] Add a basic SchedMachineModel for Falkor. Differential Revision: https://reviews.llvm.org/D26972 llvm-svn: 288194
2016-11-29 21:00:27 +01:00
include "AArch64SchedFalkor.td"
[AArch64] Add support for Qualcomm Kryo CPU. Machine model description by Dave Estes <cestes@codeaurora.org>. llvm-svn: 260686
2016-02-12 16:51:51 +01:00
include "AArch64SchedKryo.td"
[AArch64] Change the filename of the Exynos M1 scheduling defs After request by Matthias Braun in https://reviews.llvm.org/D42387. llvm-svn: 323686
2018-01-29 21:22:24 +01:00
include "AArch64SchedExynosM1.td"
[AArch64] Add pipeline model for Exynos M3 Add the scheduling and cost model for Exynos M3. Differential revision: https://reviews.llvm.org/D42387 llvm-svn: 323773
2018-01-30 16:40:16 +01:00
include "AArch64SchedExynosM3.td"
[AArch64] Add Cavium ThunderX support This set of patches adds support for Cavium ThunderX ARM64 processors: * ThunderX * ThunderX T81 * ThunderX T83 * ThunderX T88 Patch by Stefan Teleman Differential Revision: https://reviews.llvm.org/D28891 llvm-svn: 295475
2017-02-17 19:34:24 +01:00
include "AArch64SchedThunderX.td"
[AArch64] Vulcan is now ThunderXT99 Broadcom Vulcan is now Cavium ThunderX2T99. LLVM Bugzilla: http://bugs.llvm.org/show_bug.cgi?id=32113 Minor fixes for the alignments of loops and functions for ThunderX T81/T83/T88 (better performance). Patch was tested with SpecCPU2006. Patch by Stefan Teleman Differential Revision: https://reviews.llvm.org/D30510 llvm-svn: 297190
2017-03-07 20:42:40 +01:00
include "AArch64SchedThunderX2T99.td"
ARM64: initial backend import This adds a second implementation of the AArch64 architecture to LLVM, accessible in parallel via the "arm64" triple. The plan over the coming weeks & months is to merge the two into a single backend, during which time thorough code review should naturally occur. Everything will be easier with the target in-tree though, hence this commit. llvm-svn: 205090
2014-03-29 11:18:08 +01:00
[AArch64]: Add support for Cortex-A35 Adds support for the new Cortex-A35 ARMv8-A core. llvm-svn: 254503
2015-12-02 12:53:44 +01:00
def ProcA35 : SubtargetFeature<"a35", "ARMProcFamily", "CortexA35",
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
"Cortex-A35 ARM processors", [
[AArch64]: Add support for Cortex-A35 Adds support for the new Cortex-A35 ARMv8-A core. llvm-svn: 254503
2015-12-02 12:53:44 +01:00
FeatureCRC,
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
FeatureCrypto,
FeatureFPARMv8,
FeatureNEON,
FeaturePerfMon
]>;
[AArch64]: Add support for Cortex-A35 Adds support for the new Cortex-A35 ARMv8-A core. llvm-svn: 254503
2015-12-02 12:53:44 +01:00
[ARM64] Port over missing subtarget features, and CPU definitions from AArch64. llvm-svn: 206198
2014-04-14 19:38:00 +02:00
def ProcA53 : SubtargetFeature<"a53", "ARMProcFamily", "CortexA53",
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
"Cortex-A53 ARM processors", [
FeatureBalanceFPOps,
[AArch64] Lower READCYCLECOUNTER using MRS PMCCTNR_EL0. This matches the ARM behavior. In both cases, the register is part of the optional Performance Monitors extension, so, add the feature, and enable it for the A-class processors we support. Differential Revision: http://reviews.llvm.org/D12425 llvm-svn: 246555
2015-09-01 18:23:45 +02:00
FeatureCRC,
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
FeatureCrypto,
FeatureCustomCheapAsMoveHandling,
FeatureFPARMv8,
[AArch64] Enable FeatureFuseAES on Cortex-A53. It improves performance on Cortex-A53. llvm-svn: 304307
2017-05-31 17:50:03 +02:00
FeatureFuseAES,
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
FeatureNEON,
FeaturePerfMon,
FeaturePostRAScheduler,
FeatureUseAA
]>;
[ARM64] Port over missing subtarget features, and CPU definitions from AArch64. llvm-svn: 206198
2014-04-14 19:38:00 +02:00
[ARM][AArch64] Cortex-A75 and Cortex-A55 support This patch introduces support for Cortex-A75 and Cortex-A55, Arm's latest big.LITTLE A-class cores. They implement the ARMv8.2-A architecture, including the cryptography and RAS extensions, plus the optional dot product extension. They also implement the RCpc AArch64 extension from ARMv8.3-A. Cortex-A75: https://developer.arm.com/products/processors/cortex-a/cortex-a75 Cortex-A55: https://developer.arm.com/products/processors/cortex-a/cortex-a55 Differential Revision: https://reviews.llvm.org/D36667 llvm-svn: 311316
2017-08-21 10:43:06 +02:00
def ProcA55 : SubtargetFeature<"a55", "ARMProcFamily", "CortexA55",
"Cortex-A55 ARM processors", [
[AArch64] Add V8_2aOps feature to Cortex-A55 and 75 Add the missing hardware features the ProcA55 and ProcA75 feature. These are already enabled via the target parser, but I had missed them in the backend. Differential Revision: https://reviews.llvm.org/D37974 llvm-svn: 313535
2017-09-18 16:46:14 +02:00
HasV8_2aOps,
[ARM][AArch64] Cortex-A75 and Cortex-A55 support This patch introduces support for Cortex-A75 and Cortex-A55, Arm's latest big.LITTLE A-class cores. They implement the ARMv8.2-A architecture, including the cryptography and RAS extensions, plus the optional dot product extension. They also implement the RCpc AArch64 extension from ARMv8.3-A. Cortex-A75: https://developer.arm.com/products/processors/cortex-a/cortex-a75 Cortex-A55: https://developer.arm.com/products/processors/cortex-a/cortex-a55 Differential Revision: https://reviews.llvm.org/D36667 llvm-svn: 311316
2017-08-21 10:43:06 +02:00
FeatureCrypto,
FeatureFPARMv8,
FeatureFuseAES,
FeatureNEON,
FeatureFullFP16,
FeatureDotProd,
FeatureRCPC,
FeaturePerfMon
]>;
[ARM64] Port over missing subtarget features, and CPU definitions from AArch64. llvm-svn: 206198
2014-04-14 19:38:00 +02:00
def ProcA57 : SubtargetFeature<"a57", "ARMProcFamily", "CortexA57",
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
"Cortex-A57 ARM processors", [
FeatureBalanceFPOps,
[AArch64] Lower READCYCLECOUNTER using MRS PMCCTNR_EL0. This matches the ARM behavior. In both cases, the register is part of the optional Performance Monitors extension, so, add the feature, and enable it for the A-class processors we support. Differential Revision: http://reviews.llvm.org/D12425 llvm-svn: 246555
2015-09-01 18:23:45 +02:00
FeatureCRC,
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
FeatureCrypto,
FeatureCustomCheapAsMoveHandling,
FeatureFPARMv8,
[AArch64] Add new subtarget feature to fuse AES crypto operations This feature enables the fusion of such operations on Cortex A57, as recommended in its Software Optimisation Guide, section 4.13, and on Exynos M1. Differential revision: https://reviews.llvm.org/D28491 llvm-svn: 293738
2017-02-01 03:54:39 +01:00
FeatureFuseAES,
[AArch64] Add new target feature to fuse literal generation This feature enables the fusion of such operations on Cortex A57, as recommended in its Software Optimisation Guide, sections 4.14 and 4.15. Differential revision: https://reviews.llvm.org/D28698 llvm-svn: 293739
2017-02-01 03:54:42 +01:00
FeatureFuseLiterals,
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
FeatureNEON,
FeaturePerfMon,
FeaturePostRAScheduler,
FeaturePredictableSelectIsExpensive
]>;
[ARM64] Port over missing subtarget features, and CPU definitions from AArch64. llvm-svn: 206198
2014-04-14 19:38:00 +02:00
[AArch64] Restore codegen for AArch64 Cortex-A72/A73 after NFCI Summary: Code generation for Cortex-A72/Cortex-A73 was accidentally changed by r271555, which was a NFCI. The isCortexA57() predicate was not true for Cortex-A72/Cortex-A73 before r271555 (since it was checking the CPU string). Because Cortex-A72/Cortex-A73 inherit all features from Cortex-A57, all decisions previously guarded by isCortexA57() are now taken. This change restores the behaviour before r271555 by adding separate ProcA72/ProcA73, which have the required features to preserve code generation. Reviewers: kristof.beyls, aadg, mcrosier, rengolin Subscribers: mcrosier, llvm-commits, aemerson, t.p.northover, MatzeB, rengolin Differential Revision: http://reviews.llvm.org/D21182 llvm-svn: 273277
2016-06-21 17:53:54 +02:00
def ProcA72 : SubtargetFeature<"a72", "ARMProcFamily", "CortexA72",
"Cortex-A72 ARM processors", [
FeatureCRC,
FeatureCrypto,
FeatureFPARMv8,
[AArch64] Enable FeatureFuseAES on Cortex-A72. This patch enables fusing dependent AESE/AESMC and AESD/AESIMC instruction pairs on Cortex-A72, as recommended in the Software Optimization Guide, section 4.10. llvm-svn: 303073
2017-05-15 17:15:22 +02:00
FeatureFuseAES,
[AArch64] Restore codegen for AArch64 Cortex-A72/A73 after NFCI Summary: Code generation for Cortex-A72/Cortex-A73 was accidentally changed by r271555, which was a NFCI. The isCortexA57() predicate was not true for Cortex-A72/Cortex-A73 before r271555 (since it was checking the CPU string). Because Cortex-A72/Cortex-A73 inherit all features from Cortex-A57, all decisions previously guarded by isCortexA57() are now taken. This change restores the behaviour before r271555 by adding separate ProcA72/ProcA73, which have the required features to preserve code generation. Reviewers: kristof.beyls, aadg, mcrosier, rengolin Subscribers: mcrosier, llvm-commits, aemerson, t.p.northover, MatzeB, rengolin Differential Revision: http://reviews.llvm.org/D21182 llvm-svn: 273277
2016-06-21 17:53:54 +02:00
FeatureNEON,
FeaturePerfMon
]>;
def ProcA73 : SubtargetFeature<"a73", "ARMProcFamily", "CortexA73",
"Cortex-A73 ARM processors", [
FeatureCRC,
FeatureCrypto,
FeatureFPARMv8,
[AArch64] Enable FeatureFuseAES on Cortex-A73. It improves performance on Cortex-A73. llvm-svn: 304304
2017-05-31 17:25:25 +02:00
FeatureFuseAES,
[AArch64] Restore codegen for AArch64 Cortex-A72/A73 after NFCI Summary: Code generation for Cortex-A72/Cortex-A73 was accidentally changed by r271555, which was a NFCI. The isCortexA57() predicate was not true for Cortex-A72/Cortex-A73 before r271555 (since it was checking the CPU string). Because Cortex-A72/Cortex-A73 inherit all features from Cortex-A57, all decisions previously guarded by isCortexA57() are now taken. This change restores the behaviour before r271555 by adding separate ProcA72/ProcA73, which have the required features to preserve code generation. Reviewers: kristof.beyls, aadg, mcrosier, rengolin Subscribers: mcrosier, llvm-commits, aemerson, t.p.northover, MatzeB, rengolin Differential Revision: http://reviews.llvm.org/D21182 llvm-svn: 273277
2016-06-21 17:53:54 +02:00
FeatureNEON,
FeaturePerfMon
]>;
[ARM][AArch64] Cortex-A75 and Cortex-A55 support This patch introduces support for Cortex-A75 and Cortex-A55, Arm's latest big.LITTLE A-class cores. They implement the ARMv8.2-A architecture, including the cryptography and RAS extensions, plus the optional dot product extension. They also implement the RCpc AArch64 extension from ARMv8.3-A. Cortex-A75: https://developer.arm.com/products/processors/cortex-a/cortex-a75 Cortex-A55: https://developer.arm.com/products/processors/cortex-a/cortex-a55 Differential Revision: https://reviews.llvm.org/D36667 llvm-svn: 311316
2017-08-21 10:43:06 +02:00
def ProcA75 : SubtargetFeature<"a75", "ARMProcFamily", "CortexA75",
"Cortex-A75 ARM processors", [
[AArch64] Add V8_2aOps feature to Cortex-A55 and 75 Add the missing hardware features the ProcA55 and ProcA75 feature. These are already enabled via the target parser, but I had missed them in the backend. Differential Revision: https://reviews.llvm.org/D37974 llvm-svn: 313535
2017-09-18 16:46:14 +02:00
HasV8_2aOps,
[ARM][AArch64] Cortex-A75 and Cortex-A55 support This patch introduces support for Cortex-A75 and Cortex-A55, Arm's latest big.LITTLE A-class cores. They implement the ARMv8.2-A architecture, including the cryptography and RAS extensions, plus the optional dot product extension. They also implement the RCpc AArch64 extension from ARMv8.3-A. Cortex-A75: https://developer.arm.com/products/processors/cortex-a/cortex-a75 Cortex-A55: https://developer.arm.com/products/processors/cortex-a/cortex-a55 Differential Revision: https://reviews.llvm.org/D36667 llvm-svn: 311316
2017-08-21 10:43:06 +02:00
FeatureCrypto,
FeatureFPARMv8,
FeatureFuseAES,
FeatureNEON,
FeatureFullFP16,
FeatureDotProd,
FeatureRCPC,
FeaturePerfMon
]>;
AArch64: Enable AES instruction fusion on Cyclone. Note that cyclone itself doesn't fuse, but newer apple chips do and we are using cyclone as the default when targeting apple OSes. The current code also does not capture all fusion patterns of apple CPUs yet; I am still looking for ways to refactor the code nicely to extend it. llvm-svn: 316036
2017-10-17 23:46:15 +02:00
// Note that cyclone does not fuse AES instructions, but newer apple chips do
// perform the fusion and cyclone is used by default when targetting apple OSes.
[ARM64] Port over missing subtarget features, and CPU definitions from AArch64. llvm-svn: 206198
2014-04-14 19:38:00 +02:00
def ProcCyclone : SubtargetFeature<"cyclone", "ARMProcFamily", "Cyclone",
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
"Cyclone", [
FeatureAlternateSExtLoadCVTF32Pattern,
AArch64: Enable AES instruction fusion on Cyclone. Note that cyclone itself doesn't fuse, but newer apple chips do and we are using cyclone as the default when targeting apple OSes. The current code also does not capture all fusion patterns of apple CPUs yet; I am still looking for ways to refactor the code nicely to extend it. llvm-svn: 316036
2017-10-17 23:46:15 +02:00
FeatureArithmeticBccFusion,
FeatureArithmeticCbzFusion,
[ARM64] Port over missing subtarget features, and CPU definitions from AArch64. llvm-svn: 206198
2014-04-14 19:38:00 +02:00
FeatureCrypto,
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
FeatureDisableLatencySchedHeuristic,
FeatureFPARMv8,
AArch64: Enable AES instruction fusion on Cyclone. Note that cyclone itself doesn't fuse, but newer apple chips do and we are using cyclone as the default when targeting apple OSes. The current code also does not capture all fusion patterns of apple CPUs yet; I am still looking for ways to refactor the code nicely to extend it. llvm-svn: 316036
2017-10-17 23:46:15 +02:00
FeatureFuseAES,
AArch64: Add FuseCryptoEOR fusion rules There's some additional rules available on newer apple CPUs. rdar://41235346 llvm-svn: 342590
2018-09-19 22:50:51 +02:00
FeatureFuseCryptoEOR,
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
FeatureNEON,
[AArch64] Lower READCYCLECOUNTER using MRS PMCCTNR_EL0. This matches the ARM behavior. In both cases, the register is part of the optional Performance Monitors extension, so, add the feature, and enable it for the A-class processors we support. Differential Revision: http://reviews.llvm.org/D12425 llvm-svn: 246555
2015-09-01 18:23:45 +02:00
FeaturePerfMon,
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
FeatureZCRegMove,
AArch64: work around how Cyclone handles "movi.2d vD, #0". For Cylone, the instruction "movi.2d vD, #0" is executed incorrectly in some rare circumstances. Work around the issue conservatively by avoiding the instruction entirely. This patch changes CodeGen so that problematic instructions are never generated, and the AsmParser so that an equivalent instruction is used (with a warning). llvm-svn: 320965
2017-12-18 11:36:00 +01:00
FeatureZCZeroing,
FeatureZCZeroingFPWorkaround
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
]>;
[ARM64] Port over missing subtarget features, and CPU definitions from AArch64. llvm-svn: 206198
2014-04-14 19:38:00 +02:00
[AArch64] Add support for Samsung Exynos-M1 Adds core tuning support for new Samsung Exynos-M1 core (ARMv8-A). Differential Revision: http://reviews.llvm.org/D15663 llvm-svn: 256828
2016-01-05 13:51:59 +01:00
def ProcExynosM1 : SubtargetFeature<"exynosm1", "ARMProcFamily", "ExynosM1",
[AArch64] Create feature set for Samsung Exynos-M2 Since Exynos-M2 improved the FP square root unit a bit over the one in Exynos-M1, it does not benefit from using the Newton series for such operations. llvm-svn: 285246
2016-10-27 00:06:20 +02:00
"Samsung Exynos-M1 processors",
[AArch64] Rename 'no-quad-ldst-pairs' to 'slow-paired-128' In order to follow the pattern of the existing 'slow-misaligned-128store' option, rename the option 'no-quad-ldst-pairs' to 'slow-paired-128'. llvm-svn: 292954
2017-01-24 18:34:31 +01:00
[FeatureSlowPaired128,
[AArch64] Create feature set for Samsung Exynos-M2 Since Exynos-M2 improved the FP square root unit a bit over the one in Exynos-M1, it does not benefit from using the Newton series for such operations. llvm-svn: 285246
2016-10-27 00:06:20 +02:00
FeatureCRC,
FeatureCrypto,
[AArch64] Add new target feature to handle cheap as move for Exynos This feature enables special handling of cheap as move in the existing custom handling specifically for Exynos processors. Differential revision: https://reviews.llvm.org/D42387 llvm-svn: 323774
2018-01-30 16:40:22 +01:00
FeatureExynosCheapAsMoveHandling,
[AArch64] Create feature set for Samsung Exynos-M2 Since Exynos-M2 improved the FP square root unit a bit over the one in Exynos-M1, it does not benefit from using the Newton series for such operations. llvm-svn: 285246
2016-10-27 00:06:20 +02:00
FeatureFPARMv8,
[AArch64] Add new subtarget feature to fuse AES crypto operations This feature enables the fusion of such operations on Cortex A57, as recommended in its Software Optimisation Guide, section 4.13, and on Exynos M1. Differential revision: https://reviews.llvm.org/D28491 llvm-svn: 293738
2017-02-01 03:54:39 +01:00
FeatureFuseAES,
[AArch64] Create feature set for Samsung Exynos-M2 Since Exynos-M2 improved the FP square root unit a bit over the one in Exynos-M1, it does not benefit from using the Newton series for such operations. llvm-svn: 285246
2016-10-27 00:06:20 +02:00
FeatureNEON,
FeaturePerfMon,
FeaturePostRAScheduler,
[AArch64] Add FeatureSlowMisaligned128Store to Exynos M1 and M2 This feature now gates such stores after r289845. Thus the Exynos processors now need this feature. llvm-svn: 289898
2016-12-16 01:18:00 +01:00
FeatureSlowMisaligned128Store,
[AArch64] Create feature set for Samsung Exynos-M2 Since Exynos-M2 improved the FP square root unit a bit over the one in Exynos-M1, it does not benefit from using the Newton series for such operations. llvm-svn: 285246
2016-10-27 00:06:20 +02:00
FeatureUseRSqrt,
FeatureZCZeroing]>;
def ProcExynosM2 : SubtargetFeature<"exynosm2", "ARMProcFamily", "ExynosM1",
[AArch64] Create a separate feature set for Exynos M3 Distinguish the features from Exynos M2. llvm-svn: 323139
2018-01-22 20:03:26 +01:00
"Samsung Exynos-M2 processors",
[AArch64] Rename 'no-quad-ldst-pairs' to 'slow-paired-128' In order to follow the pattern of the existing 'slow-misaligned-128store' option, rename the option 'no-quad-ldst-pairs' to 'slow-paired-128'. llvm-svn: 292954
2017-01-24 18:34:31 +01:00
[FeatureSlowPaired128,
[AArch64] Create feature set for Samsung Exynos-M2 Since Exynos-M2 improved the FP square root unit a bit over the one in Exynos-M1, it does not benefit from using the Newton series for such operations. llvm-svn: 285246
2016-10-27 00:06:20 +02:00
FeatureCRC,
FeatureCrypto,
[AArch64] Add new target feature to handle cheap as move for Exynos This feature enables special handling of cheap as move in the existing custom handling specifically for Exynos processors. Differential revision: https://reviews.llvm.org/D42387 llvm-svn: 323774
2018-01-30 16:40:22 +01:00
FeatureExynosCheapAsMoveHandling,
[AArch64] Create feature set for Samsung Exynos-M2 Since Exynos-M2 improved the FP square root unit a bit over the one in Exynos-M1, it does not benefit from using the Newton series for such operations. llvm-svn: 285246
2016-10-27 00:06:20 +02:00
FeatureFPARMv8,
[AArch64] Adjust the feature set for Exynos M2 Add fusion of AES operations. llvm-svn: 308388
2017-07-19 00:51:25 +02:00
FeatureFuseAES,
[AArch64] Create feature set for Samsung Exynos-M2 Since Exynos-M2 improved the FP square root unit a bit over the one in Exynos-M1, it does not benefit from using the Newton series for such operations. llvm-svn: 285246
2016-10-27 00:06:20 +02:00
FeatureNEON,
FeaturePerfMon,
FeaturePostRAScheduler,
[AArch64] Add FeatureSlowMisaligned128Store to Exynos M1 and M2 This feature now gates such stores after r289845. Thus the Exynos processors now need this feature. llvm-svn: 289898
2016-12-16 01:18:00 +01:00
FeatureSlowMisaligned128Store,
[AArch64] Create feature set for Samsung Exynos-M2 Since Exynos-M2 improved the FP square root unit a bit over the one in Exynos-M1, it does not benefit from using the Newton series for such operations. llvm-svn: 285246
2016-10-27 00:06:20 +02:00
FeatureZCZeroing]>;
[AArch64] Add support for Samsung Exynos-M1 Adds core tuning support for new Samsung Exynos-M1 core (ARMv8-A). Differential Revision: http://reviews.llvm.org/D15663 llvm-svn: 256828
2016-01-05 13:51:59 +01:00
[AArch64] Add pipeline model for Exynos M3 Add the scheduling and cost model for Exynos M3. Differential revision: https://reviews.llvm.org/D42387 llvm-svn: 323773
2018-01-30 16:40:16 +01:00
def ProcExynosM3 : SubtargetFeature<"exynosm3", "ARMProcFamily", "ExynosM3",
[AArch64] Create a separate feature set for Exynos M3 Distinguish the features from Exynos M2. llvm-svn: 323139
2018-01-22 20:03:26 +01:00
"Samsung Exynos-M3 processors",
[FeatureCRC,
FeatureCrypto,
[AArch64] Add new target feature to handle cheap as move for Exynos This feature enables special handling of cheap as move in the existing custom handling specifically for Exynos processors. Differential revision: https://reviews.llvm.org/D42387 llvm-svn: 323774
2018-01-30 16:40:22 +01:00
FeatureExynosCheapAsMoveHandling,
[AArch64] Create a separate feature set for Exynos M3 Distinguish the features from Exynos M2. llvm-svn: 323139
2018-01-22 20:03:26 +01:00
FeatureFPARMv8,
[AArch64] Add new target feature to fuse address generation with load or store This feature enables the fusion of the address generation and a corresponding load or store together. Differential revision: https://reviews.llvm.org/D42393 llvm-svn: 323782
2018-01-30 17:28:01 +01:00
FeatureFuseAddress,
[AArch64] Create a separate feature set for Exynos M3 Distinguish the features from Exynos M2. llvm-svn: 323139
2018-01-22 20:03:26 +01:00
FeatureFuseAES,
[PATCH] [AArch64] Add new target feature to fuse conditional select This feature enables the fusion of the comparison and the conditional select instructions together. Differential revision: https://reviews.llvm.org/D42392 llvm-svn: 325939
2018-02-23 20:27:43 +01:00
FeatureFuseCCSelect,
[AArch64] Create a separate feature set for Exynos M3 Distinguish the features from Exynos M2. llvm-svn: 323139
2018-01-22 20:03:26 +01:00
FeatureFuseLiterals,
[AArch64] Add pipeline model for Exynos M3 Add the scheduling and cost model for Exynos M3. Differential revision: https://reviews.llvm.org/D42387 llvm-svn: 323773
2018-01-30 16:40:16 +01:00
FeatureLSLFast,
[AArch64] Create a separate feature set for Exynos M3 Distinguish the features from Exynos M2. llvm-svn: 323139
2018-01-22 20:03:26 +01:00
FeatureNEON,
FeaturePerfMon,
FeaturePostRAScheduler,
[AArch64] Add pipeline model for Exynos M3 Add the scheduling and cost model for Exynos M3. Differential revision: https://reviews.llvm.org/D42387 llvm-svn: 323773
2018-01-30 16:40:16 +01:00
FeaturePredictableSelectIsExpensive,
[AArch64] Create a separate feature set for Exynos M3 Distinguish the features from Exynos M2. llvm-svn: 323139
2018-01-22 20:03:26 +01:00
FeatureZCZeroing]>;
[AArch64] Add support for Qualcomm Kryo CPU. Machine model description by Dave Estes <cestes@codeaurora.org>. llvm-svn: 260686
2016-02-12 16:51:51 +01:00
def ProcKryo : SubtargetFeature<"kryo", "ARMProcFamily", "Kryo",
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
"Qualcomm Kryo processors", [
[AArch64] Add support for Qualcomm Kryo CPU. Machine model description by Dave Estes <cestes@codeaurora.org>. llvm-svn: 260686
2016-02-12 16:51:51 +01:00
FeatureCRC,
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
FeatureCrypto,
FeatureCustomCheapAsMoveHandling,
FeatureFPARMv8,
FeatureNEON,
FeaturePerfMon,
FeaturePostRAScheduler,
[Kryo] Enable ZCZeroing feature This feature uses immediate #0 to zero a register. Differential Revision: http://reviews.llvm.org/D19985 llvm-svn: 275143
2016-07-12 04:04:01 +02:00
FeaturePredictableSelectIsExpensive,
[AArch64] Add new subtarget feature to fold LSL into address mode. Summary: This feature enables folding of logical shift operations of up to 3 places into addressing mode on Kryo and Falkor that have a fastpath LSL. Reviewers: mcrosier, rengolin, t.p.northover Subscribers: junbuml, gberry, llvm-commits, aemerson Differential Revision: https://reviews.llvm.org/D31113 llvm-svn: 299240
2017-03-31 20:16:53 +02:00
FeatureZCZeroing,
FeatureLSLFast
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
]>;
[AArch64] Add support for Qualcomm's Falkor CPU. Differential Revision: https://reviews.llvm.org/D26673 llvm-svn: 287036
2016-11-15 22:34:12 +01:00
def ProcFalkor : SubtargetFeature<"falkor", "ARMProcFamily", "Falkor",
"Qualcomm Falkor processors", [
FeatureCRC,
FeatureCrypto,
[AArch64] Update the feature set for Qualcomm's Falkor CPU. llvm-svn: 291010
2017-01-04 22:26:23 +01:00
FeatureCustomCheapAsMoveHandling,
[AArch64] Add support for Qualcomm's Falkor CPU. Differential Revision: https://reviews.llvm.org/D26673 llvm-svn: 287036
2016-11-15 22:34:12 +01:00
FeatureFPARMv8,
FeatureNEON,
[AArch64] Update the feature set for Qualcomm's Falkor CPU. llvm-svn: 291010
2017-01-04 22:26:23 +01:00
FeaturePerfMon,
FeaturePostRAScheduler,
FeaturePredictableSelectIsExpensive,
[AArch64] Falkor supports Rounding Double Multiply Add/Subtract instructions. Falkor only partially implements the ARMv8.1a extensions, so this patch refactors the support for the SQRDML[A|S]H instruction into a separate feature. Differential Revision: https://reviews.llvm.org/D28681 llvm-svn: 292142
2017-01-16 17:28:43 +01:00
FeatureRDM,
[AArch64] Add new subtarget feature to fold LSL into address mode. Summary: This feature enables folding of logical shift operations of up to 3 places into addressing mode on Kryo and Falkor that have a fastpath LSL. Reviewers: mcrosier, rengolin, t.p.northover Subscribers: junbuml, gberry, llvm-commits, aemerson Differential Revision: https://reviews.llvm.org/D31113 llvm-svn: 299240
2017-03-31 20:16:53 +02:00
FeatureZCZeroing,
[AArch64][Falkor] Avoid generating STRQro* instructions Summary: STRQro* instructions are slower than the alternative ADD/STRQui expanded instructions on Falkor, so avoid generating them unless we're optimizing for code size. Reviewers: t.p.northover, mcrosier Subscribers: aemerson, rengolin, javed.absar, kristof.beyls, llvm-commits Differential Revision: https://reviews.llvm.org/D37020 llvm-svn: 311931
2017-08-28 22:48:43 +02:00
FeatureLSLFast,
FeatureSlowSTRQro
[AArch64] Add support for Qualcomm's Falkor CPU. Differential Revision: https://reviews.llvm.org/D26673 llvm-svn: 287036
2016-11-15 22:34:12 +01:00
]>;
[AArch64] Add basic support for Qualcomm's Saphira CPU. llvm-svn: 314105
2017-09-25 16:05:00 +02:00
def ProcSaphira : SubtargetFeature<"saphira", "ARMProcFamily", "Saphira",
"Qualcomm Saphira processors", [
FeatureCrypto,
FeatureCustomCheapAsMoveHandling,
FeatureFPARMv8,
FeatureNEON,
FeatureSPE,
FeaturePerfMon,
FeaturePostRAScheduler,
FeaturePredictableSelectIsExpensive,
FeatureZCZeroing,
FeatureLSLFast,
HasV8_3aOps]>;
[AArch64] Vulcan is now ThunderXT99 Broadcom Vulcan is now Cavium ThunderX2T99. LLVM Bugzilla: http://bugs.llvm.org/show_bug.cgi?id=32113 Minor fixes for the alignments of loops and functions for ThunderX T81/T83/T88 (better performance). Patch was tested with SpecCPU2006. Patch by Stefan Teleman Differential Revision: https://reviews.llvm.org/D30510 llvm-svn: 297190
2017-03-07 20:42:40 +01:00
def ProcThunderX2T99 : SubtargetFeature<"thunderx2t99", "ARMProcFamily",
"ThunderX2T99",
"Cavium ThunderX2 processors", [
[AArch64] Enable aggressive FMA on T99 and provide AArch64 options for others. This patch enables aggressive FMA by default on T99, and provides a -mllvm option to enable the same on other AArch64 micro-arch's (-mllvm -aarch64-enable-aggressive-fma). Test case demonstrating the effects on T99 is included. Patch by: steleman (Stefan Teleman) Differential Revision: https://reviews.llvm.org/D40696 llvm-svn: 323474
2018-01-25 22:55:39 +01:00
FeatureAggressiveFMA,
[AArch64] Vulcan is now ThunderXT99 Broadcom Vulcan is now Cavium ThunderX2T99. LLVM Bugzilla: http://bugs.llvm.org/show_bug.cgi?id=32113 Minor fixes for the alignments of loops and functions for ThunderX T81/T83/T88 (better performance). Patch was tested with SpecCPU2006. Patch by Stefan Teleman Differential Revision: https://reviews.llvm.org/D30510 llvm-svn: 297190
2017-03-07 20:42:40 +01:00
FeatureCRC,
FeatureCrypto,
FeatureFPARMv8,
FeatureArithmeticBccFusion,
FeatureNEON,
FeaturePostRAScheduler,
FeaturePredictableSelectIsExpensive,
FeatureLSE,
HasV8_1aOps]>;
[AARCH64] Add support for Broadcom Vulcan Adding core tuning support for new Broadcom Vulcan core (ARMv8.1A). Differential Revision: http://reviews.llvm.org/D21500 llvm-svn: 273148
2016-06-20 13:13:31 +02:00
[AArch64] Add Cavium ThunderX support This set of patches adds support for Cavium ThunderX ARM64 processors: * ThunderX * ThunderX T81 * ThunderX T83 * ThunderX T88 Patch by Stefan Teleman Differential Revision: https://reviews.llvm.org/D28891 llvm-svn: 295475
2017-02-17 19:34:24 +01:00
def ProcThunderX : SubtargetFeature<"thunderx", "ARMProcFamily", "ThunderX",
"Cavium ThunderX processors", [
FeatureCRC,
FeatureCrypto,
FeatureFPARMv8,
FeaturePerfMon,
FeaturePostRAScheduler,
FeaturePredictableSelectIsExpensive,
FeatureNEON]>;
def ProcThunderXT88 : SubtargetFeature<"thunderxt88", "ARMProcFamily",
"ThunderXT88",
"Cavium ThunderX processors", [
FeatureCRC,
FeatureCrypto,
FeatureFPARMv8,
FeaturePerfMon,
FeaturePostRAScheduler,
FeaturePredictableSelectIsExpensive,
FeatureNEON]>;
def ProcThunderXT81 : SubtargetFeature<"thunderxt81", "ARMProcFamily",
"ThunderXT81",
"Cavium ThunderX processors", [
FeatureCRC,
FeatureCrypto,
FeatureFPARMv8,
FeaturePerfMon,
FeaturePostRAScheduler,
FeaturePredictableSelectIsExpensive,
FeatureNEON]>;
def ProcThunderXT83 : SubtargetFeature<"thunderxt83", "ARMProcFamily",
"ThunderXT83",
"Cavium ThunderX processors", [
FeatureCRC,
FeatureCrypto,
FeatureFPARMv8,
FeaturePerfMon,
FeaturePostRAScheduler,
FeaturePredictableSelectIsExpensive,
FeatureNEON]>;
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
def : ProcessorModel<"generic", NoSchedModel, [
FeatureFPARMv8,
[AArch64] Enable FeatureFuseAES for the generic processor model. Summary: Scheduling AESE/AESMC and AESD/AESIMC instruction pairs back-to-back gives a double digit speedup on benchmarks using those instructions on Cortex-A processors. In GCC, this optimization is part of the generic processor model as well. This change should not have a major performance impact on processors that do not optimize AES instruction pairs, although I only had access to Cortex-A processors for benchmarking. Reviewers: rengolin, kristof.beyls, javed.absar, evandro, silviu.baranga, MatzeB, mcrosier, joelkevinjones, joel_k_jones, bmakam, t.p.northover Reviewed By: evandro Subscribers: sbaranga, aemerson, llvm-commits Differential Revision: https://reviews.llvm.org/D33836 llvm-svn: 305457
2017-06-15 11:31:23 +02:00
FeatureFuseAES,
AArch64: Do not test for CPUs, use SubtargetFeatures Testing for specific CPUs has a number of problems, better use subtarget features: - When some tweak is added for a specific CPU it is often desirable for the next version of that CPU as well, yet we often forget to add it. - It is hard to keep track of checks scattered around the target code; Declaring all target specifics together with the CPU in the tablegen file is a clear representation. - Subtarget features can be tweaked from the command line. To discourage people from using CPU checks in the future I removed the isCortexXX(), isCyclone(), ... functions. I added an getProcFamily() function for exceptional circumstances but made it clear in the comment that usage is discouraged. Reformat feature list in AArch64.td to have 1 feature per line in alphabetical order to simplify merging and sorting for out of tree tweaks. No functional change intended. Differential Revision: http://reviews.llvm.org/D20762 llvm-svn: 271555
2016-06-02 20:03:53 +02:00
FeatureNEON,
FeaturePerfMon,
FeaturePostRAScheduler
]>;
[ARM64] Port over missing subtarget features, and CPU definitions from AArch64. llvm-svn: 206198
2014-04-14 19:38:00 +02:00
[ARM][AArch64] Cortex-A75 and Cortex-A55 support This patch introduces support for Cortex-A75 and Cortex-A55, Arm's latest big.LITTLE A-class cores. They implement the ARMv8.2-A architecture, including the cryptography and RAS extensions, plus the optional dot product extension. They also implement the RCpc AArch64 extension from ARMv8.3-A. Cortex-A75: https://developer.arm.com/products/processors/cortex-a/cortex-a75 Cortex-A55: https://developer.arm.com/products/processors/cortex-a/cortex-a55 Differential Revision: https://reviews.llvm.org/D36667 llvm-svn: 311316
2017-08-21 10:43:06 +02:00
// FIXME: Cortex-A35 and Cortex-A55 are currently modeled as a Cortex-A53.
[AArch64]: Add support for Cortex-A35 Adds support for the new Cortex-A35 ARMv8-A core. llvm-svn: 254503
2015-12-02 12:53:44 +01:00
def : ProcessorModel<"cortex-a35", CortexA53Model, [ProcA35]>;
[ARM64] Ports the Cortex-A53 Machine Model description from AArch64. Summary: This port includes the rudimentary latencies that were provided for the Cortex-A53 Machine Model in the AArch64 backend. It also changes the SchedAlias for COPY in the Cyclone model to an explicit WriteRes mapping to avoid conflicts in other subtargets. Differential Revision: http://reviews.llvm.org/D3427 Patch by Dave Estes <cestes@codeaurora.org>! llvm-svn: 206652
2014-04-18 23:22:04 +02:00
def : ProcessorModel<"cortex-a53", CortexA53Model, [ProcA53]>;
[ARM][AArch64] Cortex-A75 and Cortex-A55 support This patch introduces support for Cortex-A75 and Cortex-A55, Arm's latest big.LITTLE A-class cores. They implement the ARMv8.2-A architecture, including the cryptography and RAS extensions, plus the optional dot product extension. They also implement the RCpc AArch64 extension from ARMv8.3-A. Cortex-A75: https://developer.arm.com/products/processors/cortex-a/cortex-a75 Cortex-A55: https://developer.arm.com/products/processors/cortex-a/cortex-a55 Differential Revision: https://reviews.llvm.org/D36667 llvm-svn: 311316
2017-08-21 10:43:06 +02:00
def : ProcessorModel<"cortex-a55", CortexA53Model, [ProcA55]>;
[AArch64] Basic Sched Model for Cortex-A57. Patch by Dave Estes<cestes@codeaurora.org> Differential Revision: http://reviews.llvm.org/D4008 llvm-svn: 210705
2014-06-11 23:06:56 +02:00
def : ProcessorModel<"cortex-a57", CortexA57Model, [ProcA57]>;
[ARM][AArch64] Cortex-A75 and Cortex-A55 support This patch introduces support for Cortex-A75 and Cortex-A55, Arm's latest big.LITTLE A-class cores. They implement the ARMv8.2-A architecture, including the cryptography and RAS extensions, plus the optional dot product extension. They also implement the RCpc AArch64 extension from ARMv8.3-A. Cortex-A75: https://developer.arm.com/products/processors/cortex-a/cortex-a75 Cortex-A55: https://developer.arm.com/products/processors/cortex-a/cortex-a55 Differential Revision: https://reviews.llvm.org/D36667 llvm-svn: 311316
2017-08-21 10:43:06 +02:00
// FIXME: Cortex-A72, Cortex-A73 and Cortex-A75 are currently modeled as a Cortex-A57.
[AArch64] Restore codegen for AArch64 Cortex-A72/A73 after NFCI Summary: Code generation for Cortex-A72/Cortex-A73 was accidentally changed by r271555, which was a NFCI. The isCortexA57() predicate was not true for Cortex-A72/Cortex-A73 before r271555 (since it was checking the CPU string). Because Cortex-A72/Cortex-A73 inherit all features from Cortex-A57, all decisions previously guarded by isCortexA57() are now taken. This change restores the behaviour before r271555 by adding separate ProcA72/ProcA73, which have the required features to preserve code generation. Reviewers: kristof.beyls, aadg, mcrosier, rengolin Subscribers: mcrosier, llvm-commits, aemerson, t.p.northover, MatzeB, rengolin Differential Revision: http://reviews.llvm.org/D21182 llvm-svn: 273277
2016-06-21 17:53:54 +02:00
def : ProcessorModel<"cortex-a72", CortexA57Model, [ProcA72]>;
def : ProcessorModel<"cortex-a73", CortexA57Model, [ProcA73]>;
[ARM][AArch64] Cortex-A75 and Cortex-A55 support This patch introduces support for Cortex-A75 and Cortex-A55, Arm's latest big.LITTLE A-class cores. They implement the ARMv8.2-A architecture, including the cryptography and RAS extensions, plus the optional dot product extension. They also implement the RCpc AArch64 extension from ARMv8.3-A. Cortex-A75: https://developer.arm.com/products/processors/cortex-a/cortex-a75 Cortex-A55: https://developer.arm.com/products/processors/cortex-a/cortex-a55 Differential Revision: https://reviews.llvm.org/D36667 llvm-svn: 311316
2017-08-21 10:43:06 +02:00
def : ProcessorModel<"cortex-a75", CortexA57Model, [ProcA75]>;
[ARM64] Port over missing subtarget features, and CPU definitions from AArch64. llvm-svn: 206198
2014-04-14 19:38:00 +02:00
def : ProcessorModel<"cyclone", CycloneModel, [ProcCyclone]>;
[AArch64] Add the scheduling model for Exynos-M1 Summary: Add the core scheduling model for the Samsung Exynos-M1 (ARMv8-A). Reviewers: jmolloy, rengolin, christof, MinSeongKIM, t.p.northover Subscribers: aemerson, rengolin, MatzeB Differential Revision: http://reviews.llvm.org/D16644 llvm-svn: 259958
2016-02-06 01:01:41 +01:00
def : ProcessorModel<"exynos-m1", ExynosM1Model, [ProcExynosM1]>;
[AArch64] Create feature set for Samsung Exynos-M2 Since Exynos-M2 improved the FP square root unit a bit over the one in Exynos-M1, it does not benefit from using the Newton series for such operations. llvm-svn: 285246
2016-10-27 00:06:20 +02:00
def : ProcessorModel<"exynos-m2", ExynosM1Model, [ProcExynosM2]>;
[AArch64] Add pipeline model for Exynos M3 Add the scheduling and cost model for Exynos M3. Differential revision: https://reviews.llvm.org/D42387 llvm-svn: 323773
2018-01-30 16:40:16 +01:00
def : ProcessorModel<"exynos-m3", ExynosM3Model, [ProcExynosM3]>;
[AArch64, ARM] Add support for Samsung Exynos M4 Create a separate feature set for Exynos M4 and add test cases. llvm-svn: 334115
2018-06-06 20:56:00 +02:00
def : ProcessorModel<"exynos-m4", ExynosM3Model, [ProcExynosM3]>;
[AArch64] Add a basic SchedMachineModel for Falkor. Differential Revision: https://reviews.llvm.org/D26972 llvm-svn: 288194
2016-11-29 21:00:27 +01:00
def : ProcessorModel<"falkor", FalkorModel, [ProcFalkor]>;
[AArch64] Add basic support for Qualcomm's Saphira CPU. llvm-svn: 314105
2017-09-25 16:05:00 +02:00
def : ProcessorModel<"saphira", FalkorModel, [ProcSaphira]>;
[AArch64] Add support for Qualcomm Kryo CPU. Machine model description by Dave Estes <cestes@codeaurora.org>. llvm-svn: 260686
2016-02-12 16:51:51 +01:00
def : ProcessorModel<"kryo", KryoModel, [ProcKryo]>;
[AArch64] Add Cavium ThunderX support This set of patches adds support for Cavium ThunderX ARM64 processors: * ThunderX * ThunderX T81 * ThunderX T83 * ThunderX T88 Patch by Stefan Teleman Differential Revision: https://reviews.llvm.org/D28891 llvm-svn: 295475
2017-02-17 19:34:24 +01:00
// Cavium ThunderX/ThunderX T8X Processors
def : ProcessorModel<"thunderx", ThunderXT8XModel, [ProcThunderX]>;
def : ProcessorModel<"thunderxt88", ThunderXT8XModel, [ProcThunderXT88]>;
def : ProcessorModel<"thunderxt81", ThunderXT8XModel, [ProcThunderXT81]>;
def : ProcessorModel<"thunderxt83", ThunderXT8XModel, [ProcThunderXT83]>;
[AArch64] Vulcan is now ThunderXT99 Broadcom Vulcan is now Cavium ThunderX2T99. LLVM Bugzilla: http://bugs.llvm.org/show_bug.cgi?id=32113 Minor fixes for the alignments of loops and functions for ThunderX T81/T83/T88 (better performance). Patch was tested with SpecCPU2006. Patch by Stefan Teleman Differential Revision: https://reviews.llvm.org/D30510 llvm-svn: 297190
2017-03-07 20:42:40 +01:00
// Cavium ThunderX2T9X Processors. Formerly Broadcom Vulcan.
def : ProcessorModel<"thunderx2t99", ThunderX2T99Model, [ProcThunderX2T99]>;
ARM64: initial backend import This adds a second implementation of the AArch64 architecture to LLVM, accessible in parallel via the "arm64" triple. The plan over the coming weeks & months is to merge the two into a single backend, during which time thorough code review should naturally occur. Everything will be easier with the target in-tree though, hence this commit. llvm-svn: 205090
2014-03-29 11:18:08 +01:00
//===----------------------------------------------------------------------===//
// Assembly parser
//===----------------------------------------------------------------------===//
def GenericAsmParserVariant : AsmParserVariant {
int Variant = 0;
string Name = "generic";
[AsmParser] Backends can parameterize ASM tokenization. llvm-svn: 252439
2015-11-09 01:31:07 +01:00
string BreakCharacters = ".";
[AArch64][SVE] Asm: Add parsing of merging/zeroing suffix for SVE predicate vector operands Summary: Parsing of the '/m' (merging) or '/z' (zeroing) suffix of a predicate operand. Patch [2/3] in a series to add predicated ADD/SUB instructions for SVE. Reviewers: rengolin, mcrosier, evandro, fhahn, echristo, MatzeB, t.p.northover Reviewed By: fhahn Subscribers: t.p.northover, MatzeB, aemerson, javed.absar, tschuett, llvm-commits, kristof.beyls Differential Revision: https://reviews.llvm.org/D41442 llvm-svn: 322070
2018-01-09 12:17:06 +01:00
string TokenizingCharacters = "[]*!/";
ARM64: initial backend import This adds a second implementation of the AArch64 architecture to LLVM, accessible in parallel via the "arm64" triple. The plan over the coming weeks & months is to merge the two into a single backend, during which time thorough code review should naturally occur. Everything will be easier with the target in-tree though, hence this commit. llvm-svn: 205090
2014-03-29 11:18:08 +01:00
}
def AppleAsmParserVariant : AsmParserVariant {
int Variant = 1;
string Name = "apple-neon";
[AsmParser] Backends can parameterize ASM tokenization. llvm-svn: 252439
2015-11-09 01:31:07 +01:00
string BreakCharacters = ".";
[AArch64][SVE] Asm: Add parsing of merging/zeroing suffix for SVE predicate vector operands Summary: Parsing of the '/m' (merging) or '/z' (zeroing) suffix of a predicate operand. Patch [2/3] in a series to add predicated ADD/SUB instructions for SVE. Reviewers: rengolin, mcrosier, evandro, fhahn, echristo, MatzeB, t.p.northover Reviewed By: fhahn Subscribers: t.p.northover, MatzeB, aemerson, javed.absar, tschuett, llvm-commits, kristof.beyls Differential Revision: https://reviews.llvm.org/D41442 llvm-svn: 322070
2018-01-09 12:17:06 +01:00
string TokenizingCharacters = "[]*!/";
ARM64: initial backend import This adds a second implementation of the AArch64 architecture to LLVM, accessible in parallel via the "arm64" triple. The plan over the coming weeks & months is to merge the two into a single backend, during which time thorough code review should naturally occur. Everything will be easier with the target in-tree though, hence this commit. llvm-svn: 205090
2014-03-29 11:18:08 +01:00
}
//===----------------------------------------------------------------------===//
// Assembly printer
//===----------------------------------------------------------------------===//
AArch64/ARM64: move ARM64 into AArch64's place This commit starts with a "git mv ARM64 AArch64" and continues out from there, renaming the C++ classes, intrinsics, and other target-local objects for consistency. "ARM64" test directories are also moved, and tests that began their life in ARM64 use an arm64 triple, those from AArch64 use an aarch64 triple. Both should be equivalent though. This finishes the AArch64 merge, and everyone should feel free to continue committing as normal now. llvm-svn: 209577
2014-05-24 14:50:23 +02:00
// AArch64 Uses the MC printer for asm output, so make sure the TableGen
ARM64: initial backend import This adds a second implementation of the AArch64 architecture to LLVM, accessible in parallel via the "arm64" triple. The plan over the coming weeks & months is to merge the two into a single backend, during which time thorough code review should naturally occur. Everything will be easier with the target in-tree though, hence this commit. llvm-svn: 205090
2014-03-29 11:18:08 +01:00
// AsmWriter bits get associated with the correct class.
def GenericAsmWriter : AsmWriter {
string AsmWriterClassName = "InstPrinter";
[MCInstPrinter] Enable MCInstPrinter to change its behavior based on the per-function subtarget. Currently, code-gen passes the default or generic subtarget to the constructors of MCInstPrinter subclasses (see LLVMTargetMachine::addPassesToEmitFile), which enables some targets (AArch64, ARM, and X86) to change their instprinter's behavior based on the subtarget feature bits. Since the backend can now use different subtargets for each function, instprinter has to be changed to use the per-function subtarget rather than the default subtarget. This patch takes the first step towards enabling instprinter to change its behavior based on the per-function subtarget. It adds a bit "PassSubtarget" to AsmWriter which tells table-gen to pass a reference to MCSubtargetInfo to the various print methods table-gen auto-generates. I will follow up with changes to instprinters of AArch64, ARM, and X86. llvm-svn: 233411
2015-03-27 21:36:02 +01:00
int PassSubtarget = 1;
ARM64: initial backend import This adds a second implementation of the AArch64 architecture to LLVM, accessible in parallel via the "arm64" triple. The plan over the coming weeks & months is to merge the two into a single backend, during which time thorough code review should naturally occur. Everything will be easier with the target in-tree though, hence this commit. llvm-svn: 205090
2014-03-29 11:18:08 +01:00
int Variant = 0;
bit isMCAsmWriter = 1;
}
def AppleAsmWriter : AsmWriter {
let AsmWriterClassName = "AppleInstPrinter";
[MCInstPrinter] Enable MCInstPrinter to change its behavior based on the per-function subtarget. Currently, code-gen passes the default or generic subtarget to the constructors of MCInstPrinter subclasses (see LLVMTargetMachine::addPassesToEmitFile), which enables some targets (AArch64, ARM, and X86) to change their instprinter's behavior based on the subtarget feature bits. Since the backend can now use different subtargets for each function, instprinter has to be changed to use the per-function subtarget rather than the default subtarget. This patch takes the first step towards enabling instprinter to change its behavior based on the per-function subtarget. It adds a bit "PassSubtarget" to AsmWriter which tells table-gen to pass a reference to MCSubtargetInfo to the various print methods table-gen auto-generates. I will follow up with changes to instprinters of AArch64, ARM, and X86. llvm-svn: 233411
2015-03-27 21:36:02 +01:00
int PassSubtarget = 1;
ARM64: initial backend import This adds a second implementation of the AArch64 architecture to LLVM, accessible in parallel via the "arm64" triple. The plan over the coming weeks & months is to merge the two into a single backend, during which time thorough code review should naturally occur. Everything will be easier with the target in-tree though, hence this commit. llvm-svn: 205090
2014-03-29 11:18:08 +01:00
int Variant = 1;
int isMCAsmWriter = 1;
}
//===----------------------------------------------------------------------===//
// Target Declaration
//===----------------------------------------------------------------------===//
AArch64/ARM64: move ARM64 into AArch64's place This commit starts with a "git mv ARM64 AArch64" and continues out from there, renaming the C++ classes, intrinsics, and other target-local objects for consistency. "ARM64" test directories are also moved, and tests that began their life in ARM64 use an arm64 triple, those from AArch64 use an aarch64 triple. Both should be equivalent though. This finishes the AArch64 merge, and everyone should feel free to continue committing as normal now. llvm-svn: 209577
2014-05-24 14:50:23 +02:00
def AArch64 : Target {
let InstructionSet = AArch64InstrInfo;
ARM64: initial backend import This adds a second implementation of the AArch64 architecture to LLVM, accessible in parallel via the "arm64" triple. The plan over the coming weeks & months is to merge the two into a single backend, during which time thorough code review should naturally occur. Everything will be easier with the target in-tree though, hence this commit. llvm-svn: 205090
2014-03-29 11:18:08 +01:00
let AssemblyParserVariants = [GenericAsmParserVariant, AppleAsmParserVariant];
let AssemblyWriters = [GenericAsmWriter, AppleAsmWriter];
[MachineOperand][Target] MachineOperand::isRenamable semantics changes Summary: Add a target option AllowRegisterRenaming that is used to opt in to post-register-allocation renaming of registers. This is set to 0 by default, which causes the hasExtraSrcRegAllocReq/hasExtraDstRegAllocReq fields of all opcodes to be set to 1, causing MachineOperand::isRenamable to always return false. Set the AllowRegisterRenaming flag to 1 for all in-tree targets that have lit tests that were effected by enabling COPY forwarding in MachineCopyPropagation (AArch64, AMDGPU, ARM, Hexagon, Mips, PowerPC, RISCV, Sparc, SystemZ and X86). Add some more comments describing the semantics of the MachineOperand::isRenamable function and how it is set and maintained. Change isRenamable to check the operand's opcode hasExtraSrcRegAllocReq/hasExtraDstRegAllocReq bit directly instead of relying on it being consistently reflected in the IsRenamable bit setting. Clear the IsRenamable bit when changing an operand's register value. Remove target code that was clearing the IsRenamable bit when changing registers/opcodes now that this is done conservatively by default. Change setting of hasExtraSrcRegAllocReq in AMDGPU target to be done in one place covering all opcodes that have constant pipe read limit restrictions. Reviewers: qcolombet, MatzeB Subscribers: aemerson, arsenm, jyknight, mcrosier, sdardis, nhaehnle, javed.absar, tpr, arichardson, kristof.beyls, kbarton, fedor.sergeev, asb, rbar, johnrusso, simoncook, jordy.potman.lists, apazos, sabuasal, niosHD, escha, nemanjai, llvm-commits Differential Revision: https://reviews.llvm.org/D43042 llvm-svn: 325931
2018-02-23 19:25:08 +01:00
let AllowRegisterRenaming = 1;
ARM64: initial backend import This adds a second implementation of the AArch64 architecture to LLVM, accessible in parallel via the "arm64" triple. The plan over the coming weeks & months is to merge the two into a single backend, during which time thorough code review should naturally occur. Everything will be easier with the target in-tree though, hence this commit. llvm-svn: 205090
2014-03-29 11:18:08 +01:00
}
Reference in New Issue Copy Permalink