aboutsummaryrefslogtreecommitdiff
path: root/lib/Target
AgeCommit message (Collapse)Author
2012-01-27Better diagnostic for malformed .org assembly directive.Jim Grosbach
Provide source line number information. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@149101 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-26Keep source information, if available, around for ARM Fixups.Jim Grosbach
Adjust an example MachObjectWriter diagnostic to use the information to issue a better message. Before: LLVM ERROR: unknown ARM fixup kind! After: x.s:6:5: error: unsupported relocation on symbol beq bar ^ rdar://9800182 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@149093 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-26Handle call-clobbered ymm registers on Win64.Jakob Stoklund Olesen
The Win64 calling convention has xmm6-15 as callee-saved while still clobbering all ymm registers. Add a YMM_HI_6_15 pseudo-register that aliases the clobbered part of the ymm registers, and mark that as call-clobbered. This allows live xmm registers across calls. This hack wouldn't be necessary with RegisterMask operands representing the call clobbers, but they are not quite operational yet. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@149088 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-26Tidy up. Fix mismatched return types for error handling.Jim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@149062 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-26Add support for the R_ARM_TARGET1 relocation, which should be given to ↵James Molloy
relocations applied to all C++ constructors and destructors. This enables the linker to match concrete relocation types (absolute or relative) with whatever library or C++ support code is being linked against. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@149057 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-26Fix for the following bug in AVX codegen for double-to-int conversions:Victor Umansky
. "fptosi" and "fptoui" IR instructions are defined with round-to-zero rounding mode. . Currently for AVX mode for <4xdouble> and <8xdouble> the "VCVTPD2DQ.128" and "VCVTPD2DQ.256" instructions are selected (for .fp_to_sint. DAG node operation ) by AVX codegen. However they use round-to-nearest-even rounding mode. . Consequently, the conversion produces incorrect numbers. The fix is to replace selection of VCVTPD2DQ instructions with VCVTTPD2DQ instructions. The latter use truncate (i.e. round-to-zero) rounding mode. As .fp_to_sint. DAG node operation is used only for lowering of "fptosi" and "fptoui" IR instructions, the fix in X86InstrSSE.td definition file doesn.t have an impact on other LLVM flows. The patch includes changes in the .td file, LIT test for the changes and a fix in a legacy LIT test (which produced asm code conflicting with LLVN IR spec). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@149056 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-26Add HasXOP predicate check covering a bunch of XOP intrinsic patterns.Craig Topper
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@149054 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-26Fix AVX vs SSE patterns ordering issue for VPCMPESTRM and VPCMPISTRM.Craig Topper
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@149053 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-26Remove some more patterns by custom lowering intrinsics to target specific ↵Craig Topper
nodes. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@149052 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-25Properly emit ctors / dtors with priorities into desired sectionsAnton Korobeynikov
and let linker handle the rest. This finally fixes PR5329 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148990 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-25ARM assemly parsing and validation of IT instruction.Jim Grosbach
"Although a Thumb2 instruction, the IT mnemonic shall be permitted in ARM mode, and the condition verified to match the condition code(s) on the following instruction(s)." PR11853 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148969 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-25fix a bug I introduced in r148929, this is not a splat!Chris Lattner
Thanks to Eli for noticing. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148947 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-25Custom lower PSIGN and PSHUFB intrinsics to their corresponding target ↵Craig Topper
specific nodes so we can remove the isel patterns. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148933 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-25use ConstantVector::getSplat in a few places.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148929 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-25Custom lower phadd and phsub intrinsics to target specific nodes. Remove the ↵Craig Topper
patterns that are no longer necessary. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148927 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-25Remove AVX 256-bit unaligned load intrinsics. 128-bit versions had been ↵Craig Topper
removed a while ago. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148922 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-25Mark 64-bit register RA_64 unused too.Akira Hatanaka
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148918 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-25Modify MipsFrameLowering::emitPrologue and emitEpilogue.Akira Hatanaka
- Use MipsAnalyzeImmediate to expand immediates that do not fit in 16-bit. - Change the types of variables so that they are sufficiently large to handle 64-bit pointers. - Emit instructions to set register $28 in a function prologue after instructions which store callee-saved registers have been emitted. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148917 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-25Modify MipsRegisterInfo::eliminateFrameIndex to use MipsAnalyzeImmediate toAkira Hatanaka
expand offsets that do not fit in the 16-bit immediate field of load and store instructions. Also change the types of variables so that they are sufficiently large to handle 64-bit pointers. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148916 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-25Merge intrinsic pattern and no pattern versions of VCVTSD2SI intruction ↵Craig Topper
definitions. Matches non-AVX version of same instructions. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148914 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-25MipsAnalyzeImmediate.h: Fix to add DataTypes.h for msvc.NAKAMURA Takumi
inttypes.h is not supplied in msvc. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148912 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-25Target/Mips: Unbreak CMake build.NAKAMURA Takumi
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148909 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-25Lower 64-bit immediates using MipsAnalyzeImmediate that has just been added. Akira Hatanaka
Add a test case to show fewer instructions are needed to load an immediate with the new way of loading immediates. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148908 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-25Add class MipsAnalyzeImmediate which comes up with an instruction sequence toAkira Hatanaka
load an immediate. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148900 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-25NEON VLD4(all lanes) assembly parsing and encoding.Jim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148884 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24Tidy up. Rename VLD4DUP patterns for consistency.Jim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148883 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24NEON VLD3(all lanes) assembly parsing and encoding.Jim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148882 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24Sign-extend 32-bit integer arguments when they are passed in 64-bit registers,Akira Hatanaka
which is what N32/64 does. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148875 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24Pass CCState by reference.Akira Hatanaka
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148871 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24Pattern for f32 to i64 conversion.Akira Hatanaka
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148869 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24Intel Syntax: Extend special hand coded logic, to recognize special ↵Devang Patel
instructions, for intel syntax. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148864 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-2464-bit sign extension in register instructions.Akira Hatanaka
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148862 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24NEON VST4(one lane) assembly parsing and encoding.Jim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148836 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24Widen the instruction encoder that TblGen emits to a 64 bits, which should ↵Owen Anderson
accomodate every target I can think of offhand. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148833 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24NEON VLD4(one lane) assembly parsing and encoding.Jim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148832 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24NEON Two-operand assembly aliases for VSRA.Jim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148821 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24NEON Two-operand assembly aliases for VSLI.Jim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148819 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24NEON Two-operand assembly aliases for VSRI.Jim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148818 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24NEON add correct predicates for some asm aliases.Jim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148815 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24C++, CBE, and TLOF support for ConstantDataSequentialChris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148805 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24ZERO_EXTEND operation is optimized for AVX.Elena Demikhovsky
v8i16 -> v8i32, v4i32 -> v4i64 - used vpunpck* instructions. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148803 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24Use correct register class for am2offset register operands.Anton Korobeynikov
This pacifies machine verifier git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148782 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24Add comments near load pattern fragments indicating that all integer vector ↵Craig Topper
loads are promoted to v2i64 or v4i64 so that no one tries to reintroduce pattern fragments for other types. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148771 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24NEON VST4(multiple 4 element structures) assembly parsing.Jim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148764 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24NEON VLD4(multiple 4 element structures) assembly parsing.Jim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148762 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24Tidy up. Remove some vertical space for readability.Jim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148761 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24Revert r148686 (and r148694, a fix to it) due to a serious layeringChandler Carruth
violation -- MC cannot depend on CodeGen. Specifically, the MCTargetDesc component of each target is actually a subcomponent of the MC library. As such, it cannot depend on the target-independent code generator, because MC itself cannot depend on the target-independent code generator. This change moved a flag from the ARM MCTargetDesc file ARMMCAsmInfo.cpp to the CodeGen layer in ARMException.cpp, leaving behind an 'extern' to refer back to it. That layering order isn't viable givin the constraints outlined above. Commandline flags are designed to be static specifically to avoid these types of bugs. Fixing this is likely going to require some non-trivial refactoring. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148759 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24Fix typo.Jim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148757 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-24NEON VST3(single element from one lane) assembly parsing.Jim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148755 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-23Fix typo. Devang Patel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@148751 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-09-23 12:10:39 +0000