aboutsummaryrefslogtreecommitdiff
path: root/lib/Target
AgeCommit message (Collapse)Author
2012-09-04This patch optimizes shuffle instruction - generates 2 instructions instead ↵Elena Demikhovsky
of 4. Since this specific shuffle is widely used in many workloads we have ~10% performance on them. shufflevector <8 x float> %A, <8 x float> %B, <8 x i32> <i32 0, i32 8, i32 2, i32 10, i32 4, i32 12, i32 6, i32 14> vmovaps (%rdx), %ymm0 vshufps $8, %ymm0, %ymm0, %ymm0 vmovaps (%rcx), %ymm1 vshufps $8, %ymm0, %ymm1, %ymm1 vunpcklps %ymm0, %ymm1, %ymm0 vmovaps (%rcx), %ymm0 vmovsldup (%rdx), %ymm1 vblendps $85, %ymm0, %ymm1, %ymm0 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163134 91177308-0d34-0410-b5e6-96231b3b80d8
2012-09-03[ms-inline asm] Asm operands can map to one or more MCOperands. Therefore, addChad Rosier
the NumMCOperands argument to the GetMCInstOperandNum() function that is set to the number of MCOperands this asm operand mapped to. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163124 91177308-0d34-0410-b5e6-96231b3b80d8
2012-09-03[ms-inline asm] Add a comment.Chad Rosier
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163123 91177308-0d34-0410-b5e6-96231b3b80d8
2012-09-03[ms-inline asm] Add an interface to the GetMCInstOperandNum() function in theChad Rosier
MCTargetAsmParser class. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163122 91177308-0d34-0410-b5e6-96231b3b80d8
2012-09-03Remove always true checks. Noticed by Adhemerval Zanella.Roman Divacky
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163117 91177308-0d34-0410-b5e6-96231b3b80d8
2012-09-03Add braces to the case statement.Chad Rosier
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163116 91177308-0d34-0410-b5e6-96231b3b80d8
2012-09-03Removed unused argument.Chad Rosier
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163104 91177308-0d34-0410-b5e6-96231b3b80d8
2012-09-03some peepholes that should match horizontal add/sub operations.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163103 91177308-0d34-0410-b5e6-96231b3b80d8
2012-09-03[ms-inline asm] Expose the Kind and Opcode variables from theChad Rosier
MatchInstructionImpl() function. These values are used by the ConvertToMCInst() function to index into the ConversionTable. The values are also needed to call the GetMCInstOperandNum() function. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163101 91177308-0d34-0410-b5e6-96231b3b80d8
2012-09-03Move ErrorLoc decl into the scope where it's actually used.Chad Rosier
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163100 91177308-0d34-0410-b5e6-96231b3b80d8
2012-09-02Not all targets have efficient ISel code generation for select instructions.Nadav Rotem
For example, the ARM target does not have efficient ISel handling for vector selects with scalar conditions. This patch adds a TLI hook which allows the different targets to report which selects are supported well and which selects should be converted to CF duting codegen prepare. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163093 91177308-0d34-0410-b5e6-96231b3b80d8
2012-09-01Limit domain conversion to cases where it won't break dep chains.Tim Northover
NEON domain conversion was too heavy-handed with its widened registers, which could have stripped existing instructions of their dependency, leaving them vulnerable to scheduling errors. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163070 91177308-0d34-0410-b5e6-96231b3b80d8
2012-09-01Fix Thumb2 fixup kind in the integrated-as.Logan Chien
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163063 91177308-0d34-0410-b5e6-96231b3b80d8
2012-09-01TyposCraig Topper
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163053 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-31SelectionDAG: when constructing VZEXT_LOAD from other loads, make sure itsManman Ren
output chain is correctly setup. As an example, if the original load must happen before later stores, we need to make sure the constructed VZEXT_LOAD is constrained to be before the stores. rdar://11457792 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163036 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-31Mark FMA4 instructions as commutable and add them to the folding tables.Craig Topper
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163035 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-31Remove an unused argument. The MCInst opcode is set in the ConvertToMCInst()Chad Rosier
function nowadays. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163030 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-31Add selection of RegOp2MemOpTable3 to canFoldMemoryOperandCraig Topper
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163029 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-31Fix PR12359Michael Liao
- In addition to undefined, if V2 is zero vector, skip 2nd PSHUFB and POR as well as PSHUFB will zero elements with negative indices. Patch by Sriram Murali <sriram.murali@intel.com> git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163018 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-31The instruction DINS may be transformed into DINSU or DEXTM dependingJack Carter
on the size of the extraction and its position in the 64 bit word. This patch allows support of the dext transformations with mips64 direct object output. 0 <= msb < 32 0 <= lsb < 32 0 <= pos < 32 1 <= size <= 32 DINS The field is entirely contained in the right-most word of the doubleword 32 <= msb < 64 0 <= lsb < 32 0 <= pos < 32 2 <= size <= 64 DINSM The field straddles the words of the doubleword 32 <= msb < 64 32 <= lsb < 64 32 <= pos < 64 1 <= size <= 32 DINSU The field is entirely contained in the left-most word of the doubleword git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163010 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-31Add a comment to explain what's really going on.Chad Rosier
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163005 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-31The ConvertToMCInst() function can't fail, so remove the now dead ↵Chad Rosier
Match_ConversionFail enum. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163002 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-31Mark FMA3 instructions as commutable so that the operands to the multiply ↵Craig Topper
part can be commuted. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@163001 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-31Add support for converting llvm.fma to fma4 instructions.Craig Topper
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162999 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-31Clean up AddedComplexity further after adding UseSSExMichael Liao
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162973 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-31Fix a couple of typos in EmitAtomic.Jakob Stoklund Olesen
Thumb2 instructions are mostly constrained to rGPR, not tGPR which is for Thumb1. rdar://problem/12203728 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162968 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-31X86: Fix encoding of 'movd %xmm0, %rax'Jim Grosbach
The assembly string for the VMOVPQIto64rr instruction incorrectly lacked the 'v' prefix, resulting in mis-assembly of the vanilla movd instruction. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162963 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-31With the fix in r162954/162955 every cvt function returns true. Thus, haveChad Rosier
the ConvertToMCInst() return void, rather then a bool. Update all the cvt functions as well. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162961 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-30Fix for r162954. Return the Error.Chad Rosier
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162955 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-30Move a check to the validateInstruction() function where it more properly ↵Chad Rosier
belongs. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162954 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-30Typo.Chad Rosier
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162952 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-30Introduce 'UseSSEx' to force SSE legacy encodingMichael Liao
- Add 'UseSSEx' to force SSE legacy insn not being selected when AVX is enabled. As the penalty of inter-mixing SSE and AVX instructions, we need prevent SSE legacy insn from being generated except explicitly specified through some intrinsics. For patterns supported by both SSE and AVX, so far, we force AVX insn will be tried first relying on AddedComplexity or position in td file. It's error-prone and introduces bugs accidentally. 'UseSSEx' is disabled when AVX is turned on. For SSE insns inherited by AVX, we need this predicate to force VEX encoding or SSE legacy encoding only. For insns not inherited by AVX, we still use the previous predicates, i.e. 'HasSSEx'. So far, these insns fall into the following categories: * SSE insns with MMX operands * SSE insns with GPR/MEM operands only (xFENCE, PREFETCH, CLFLUSH, CRC, and etc.) * SSE4A insns. * MMX insns. * x87 insns added by SSE. 2 test cases are modified: - test/CodeGen/X86/fast-isel-x86-64.ll AVX code generation is different from SSE one. 'vcvtsi2sdq' cannot be selected by fast-isel due to complicated pattern and fast-isel fallback to materialize it from constant pool. - test/CodeGen/X86/widen_load-1.ll AVX code generation is different from SSE one after fixing SSE/AVX inter-mixing. Exec-domain fixing prefers 'vmovapd' instead of 'vmovaps'. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162919 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-30PPCISelLowering.cpp: Fix r162725.NAKAMURA Takumi
[Tobias von Koch] What's happening here is that the CR6SET/CR6UNSET is breaking the chain of register copies glued to the function call (BL_SVR4 node). The scheduler then moves other instructions in between those and the function call, which isn't good! Right. That's the case where there is no chain of register copies before the call, so InFlag == 0... Attached is a new revision of the patch which should fix this for good. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162916 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-30PPCISelLowering.cpp: Whitespace.NAKAMURA Takumi
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162915 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-30Add support for moving pure S-register to NEON pipeline if desiredTim Northover
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162898 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-30Only perform DAG combine on FMAs of legal types.Craig Topper
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162892 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-30Fix PR13727Michael Liao
- The root cause is that target constant materialization in X86 fast-isel creates a PC-rel addressing which may overflow 32-bit range in non-Small code model if .rodata section is allocated too far away from code segment in MCJIT, which uses Large code model so far. - Follow the similar logic to fix non-Small code model in fast-isel by skipping non-Small code model. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162881 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-29Rename hasVolatileMemoryRef() to hasOrderedMemoryRef().Jakob Stoklund Olesen
Ordered memory operations are more constrained than volatile loads and stores because they must be ordered with respect to all other memory operations. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162861 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-29Reserve space for the mandatory traceback fields on PPC64.Hal Finkel
We need to reserve space for the mandatory traceback fields, though leaving them as zero is appropriate for now. Although the ABI calls for these fields to be filled in fully, no compiler on Linux currently does this, and GDB does not read these fields. GDB uses the first word of zeroes during exception handling to find the end of the function and the size field, allowing it to compute the beginning of the function. DWARF information is used for everything else. We need the extra 8 bytes of pad so the size field is found in the right place. As a comparison, GCC fills in a few of the fields -- language, number of saved registers -- but ignores the rest. IBM's proprietary OSes do make use of the full traceback table facility. Patch by Bill Schmidt. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162854 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-29Refactor setExecutionDomain to be clearer about what it's doing and more robust.Tim Northover
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162844 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-29Make helper function static.Benjamin Kramer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162843 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-29Make MemoryBuiltins aware of TargetLibraryInfo.Benjamin Kramer
This disables malloc-specific optimization when -fno-builtin (or -ffreestanding) is specified. This has been a problem for a long time but became more severe with the recent memory builtin improvements. Since the memory builtin functions are used everywhere, this required passing TLI in many places. This means that functions that now have an optional TLI argument, like RecursivelyDeleteTriviallyDeadFunctions, won't remove dead mallocs anymore if the TLI argument is missing. I've updated most passes to do the right thing. Fixes PR13694 and probably others. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162841 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-29Convert FMA4 patterns to use target specific nodes instead of intrinsics to ↵Craig Topper
align with FMA3. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162829 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-29Cleanup sloppy code. Jakob's review.Andrew Trick
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162825 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-29[arm-fast-isel] Add support for ARM PIC.Jush Lu
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162823 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-29Fix ARM vector copies of overlapping register tuples.Andrew Trick
I have tested the fix, but have not been successfull in generating a robust unit test. This can only be exposed through particular register assignments. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162821 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-29cleanupAndrew Trick
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162820 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-28Typo.Chad Rosier
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162807 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-28Add comments on the literal value used.Michael Liao
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162805 91177308-0d34-0410-b5e6-96231b3b80d8
2012-08-28The instruction DEXT may be transformed into DEXTU or DEXTM dependingJack Carter
on the size of the extraction and its position in the 64 bit word. This patch allows support of the dext transformations with mips64 direct object output. 0 <= msb < 32 0 <= lsb < 32 0 <= pos < 32 1 <= size <= 32 DINS The field is entirely contained in the right-most word of the doubleword 32 <= msb < 64 0 <= lsb < 32 0 <= pos < 32 2 <= size <= 64 DINSM The field straddles the words of the doubleword 32 <= msb < 64 32 <= lsb < 64 32 <= pos < 64 1 <= size <= 32 DINSU The field is entirely contained in the left-most word of the doubleword git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@162782 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-09-23 03:22:27 +0000