aboutsummaryrefslogtreecommitdiff
path: root/lib/Target/ARM/ARMISelLowering.cpp
AgeCommit message (Collapse)Author
2011-09-13Tidy up a few 80 column violations.Jim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139636 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-09Thumb unconditional branches are allowed in IT blocks, and therefore should ↵Owen Anderson
have a predicate operand, unlike conditional branches. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139415 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-09Thumb2 assembly parsing and encoding for LDREX/LDREXB/LDREXD/LDREXH.Jim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139381 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-06Add codegen support for vector select (in the IR this means a selectDuncan Sands
with a vector condition); such selects become VSELECT codegen nodes. This patch also removes VSETCC codegen nodes, unifying them with SETCC nodes (codegen was actually often using SETCC for vector SETCC already). This ensures that various DAG combiner optimizations kick in for vector comparisons. Passes dragonegg bootstrap with no testsuite regressions (nightly testsuite as well as "make check-all"). Patch mostly by Nadav Rotem. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139159 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-06Fix fall outs from my recent change on how carry bit is modeled during isel.Evan Cheng
Now the 'S' instructions, e.g. ADDS, treat S bit as optional operand as well. Also fix isel hook to correctly set the optional operand. rdar://10073745 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139157 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-01Null-initialize to shut up -Wuninitialized warnings.Eli Friedman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@138974 91177308-0d34-0410-b5e6-96231b3b80d8
2011-08-3164-bit atomic cmpxchg for ARM.Eli Friedman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@138868 91177308-0d34-0410-b5e6-96231b3b80d8
2011-08-31Some 64-bit atomic operations on ARM. 64-bit cmpxchg coming next.Eli Friedman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@138845 91177308-0d34-0410-b5e6-96231b3b80d8
2011-08-30Follow up to r138791.Evan Cheng
Add a instruction flag: hasPostISelHook which tells the pre-RA scheduler to call a target hook to adjust the instruction. For ARM, this is used to adjust instructions which may be setting the 's' flag. ADC, SBC, RSB, and RSC instructions have implicit def of CPSR (required since it now uses CPSR physical register dependency rather than "glue"). If the carry flag is used, then the target hook will *fill in* the optional operand with CPSR. Otherwise, the hook will remove the CPSR implicit def from the MachineInstr. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@138810 91177308-0d34-0410-b5e6-96231b3b80d8
2011-08-30Change ARM / Thumb2 addc / adde and subc / sube modeling to use physicalEvan Cheng
register dependency (rather than glue them together). This is general goodness as it gives scheduler more freedom. However it is motivated by a nasty bug in isel. When a i64 sub is expanded to subc + sube. libcall #1 \ \ subc \ / \ \ / \ \ / libcall #2 sube If the libcalls are not serialized (i.e. both have chains which are dag entry), legalizer can serialize them in arbitrary orders. If it's unlucky, it can force libcall #2 before libcall #1 in the above case. subc | libcall #2 | libcall #1 | sube However since subc and sube are "glued" together, this ends up being a cycle when the scheduler combine subc and sube as a single scheduling unit. The right solution is to fix LegalizeType too chains the libcalls together. However, LegalizeType is not processing nodes in order so that's harder than it should be. For now, the move to physical register dependency will do. rdar://10019576 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@138791 91177308-0d34-0410-b5e6-96231b3b80d8
2011-08-29Expand ATOMIC_LOAD and ATOMIC_STORE for architectures I don't know well ↵Eli Friedman
enough to fix properly. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@138751 91177308-0d34-0410-b5e6-96231b3b80d8
2011-08-27Silence GCC warnings and make an array const.Benjamin Kramer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@138706 91177308-0d34-0410-b5e6-96231b3b80d8
2011-08-26Atomic load/store on ARM/Thumb.Eli Friedman
I don't really like the patterns, but I'm having trouble coming up with a better way to handle them. I plan on making other targets use the same legalization ARM-without-memory-barriers is using... it's not especially efficient, but if anyone cares, it's not that hard to fix for a given target if there's some better lowering. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@138621 91177308-0d34-0410-b5e6-96231b3b80d8
2011-08-12ARM expansion of pre-indexed store pseudos should maintain memoperands.Jim Grosbach
Partial fix for rdar://9945172. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@137513 91177308-0d34-0410-b5e6-96231b3b80d8
2011-08-12Silence a bunch (but not all) "variable written but not read" warningsDuncan Sands
when building with assertions disabled. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@137460 91177308-0d34-0410-b5e6-96231b3b80d8
2011-08-11ARM STRH assembly parsing and encoding.Jim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@137353 91177308-0d34-0410-b5e6-96231b3b80d8
2011-08-09ARM fix typo in pre-indexed store lowering.Jim Grosbach
rdar://9915869 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@137148 91177308-0d34-0410-b5e6-96231b3b80d8
2011-08-05ARM refactor indexed store instructions.Jim Grosbach
Refactor STR[B] pre and post indexed instructions to use addressing modes for memory operands, which is necessary for assembly parsing and is more consistent with the rest of the memory instruction definitions. Make some incremental progress on refactoring away the mega-operand addrmode2 along the way, which is nice. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@136978 91177308-0d34-0410-b5e6-96231b3b80d8
2011-08-03New approach to r136737: insert the necessary fences for atomic ops in ↵Eli Friedman
platform-independent code, since a bunch of platforms (ARM, Mips, PPC, Alpha are the relevant targets here) need to do essentially the same thing. I think this completes the basic CodeGen for atomicrmw and cmpxchg. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@136813 91177308-0d34-0410-b5e6-96231b3b80d8
2011-08-02ARM backend support for atomicrmw and cmpxchg with non-monotonic ordering. ↵Eli Friedman
Not especially pretty, but seems to work well enough. If this looks okay, I'll put together similar patches for Mips, PPC, and Alpha. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@136737 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-29Add support for the 'Q' constraint.Eric Christopher
Fixes rdar://9866494 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@136523 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-27Code generation for 'fence' instruction.Eli Friedman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@136283 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-27ARM extend instructions simplification.Jim Grosbach
Refactor the SXTB, SXTH, SXTB16, UXTB, UXTH, and UXTB16 instructions to not have an 'r' and an 'r_rot' version, but just a single version with a rotate that can be zero. Use plain Pat<>'s for the ISel of the non-rotated version. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@136225 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-21Split up the ARM so_reg ComplexPattern into so_reg_reg and so_reg_imm, ↵Owen Anderson
allowing us to distinguish the encodings that use shifted registers from those that use shifted immediates. This is necessary to allow the fixed-length decoder to distinguish things like BICS vs LDRH. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@135693 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-20Sink ARMMCExpr and ARMAddressingModes into MC layer. First step to separate ↵Evan Cheng
ARM MC code from target. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@135636 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-18land David Blaikie's patch to de-constify Type, with a few tweaks.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@135375 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-13Improve codegen for select's:Evan Cheng
if (x != 0) x = 1 if (x == 1) x = 1 Previous codegen looks like this: mov r1, r0 cmp r1, #1 mov r0, #0 moveq r0, #1 The naive lowering select between two different values. It should recognize the test is equality test so it's more a conditional move rather than a select: cmp r0, #1 movne r0, #0 rdar://9758317 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@135017 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-08Add an intrinsic and codegen support for fused multiply-accumulate. The intentCameron Zwarich
is to use this for architectures that have a native FMA instruction. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@134742 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-08Add more info to FIXME.Jim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@134729 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-01ARMv7M vs. ARMv7E-M support.Jim Grosbach
The DSP instructions in the Thumb2 instruction set are an optional extension in the Cortex-M* archtitecture. When present, the implementation is considered an "ARMv7E-M implementation," and when not, an "ARMv7-M implementation." Add a subtarget feature hook for the v7e-m instructions and hook it up. The cortex-m3 cpu is an example of a v7m implementation, while the cortex-m4 is a v7e-m implementation. rdar://9572992 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@134261 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-01Add support for the 'j' immediate constraint. This is conditionalized onEric Christopher
supporting the instruction that the constraint is for 'movw'. Part of rdar://9119939 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@134222 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-01Add support for the ARM 't' register constraint. And another testcaseEric Christopher
for the 'x' register constraint. Part of rdar://9119939 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@134220 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-01We'll return a null RC by default if we can't match.Eric Christopher
Part of rdar://9119939 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@134217 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-01Add support for the 'x' constraint.Eric Christopher
Part of rdar://9307836 and rdar://9119939 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@134215 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-30Capitalize the unsigned part of the initializer.Eric Christopher
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@134211 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-30Rename Pair to RCPair lacking any better naming ideas.Eric Christopher
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@134210 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-30Add support for the 'h' constraint.Eric Christopher
Part of rdar://9119939 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@134203 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-30Add a convenience typedef for std::pair<unsigned, const TargetRegisterClass*>.Eric Christopher
No functional change. Part of rdar://9119939 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@134198 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-29Remove getRegClassForInlineAsmConstraint from the ARM port.Eric Christopher
Part of rdar://9643582 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@134095 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-28- Rename TargetInstrDesc, TargetOperandInfo to MCInstrDesc and MCOperandInfo andEvan Cheng
sink them into MC layer. - Added MCInstrInfo, which captures the tablegen generated static data. Chang TargetInstrInfo so it's based off MCInstrInfo. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@134021 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-28Remove warning: 'c0' may be used uninitialized in this function.Chad Rosier
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@134014 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-24The Neon VCVT (between floating-point and fixed-point, Advanced SIMD) Chad Rosier
instructions can be used to match combinations of multiply/divide and VCVT (between floating-point and integer, Advanced SIMD). Basically the VCVT immediate operand that specifies the number of fraction bits corresponds to a floating-point multiply or divide by the corresponding power of 2. For example, VCVT (floating-point to fixed-point, Advanced SIMD) can replace a combination of VMUL and VCVT (floating-point to integer) as follows: Example (assume d17 = <float 8.000000e+00, float 8.000000e+00>): vmul.f32 d16, d17, d16 vcvt.s32.f32 d16, d16 becomes: vcvt.s32.f32 d16, d16, #3 Similarly, VCVT (fixed-point to floating-point, Advanced SIMD) can replace a combinations of VCVT (integer to floating-point) and VDIV as follows: Example (assume d17 = <float 8.000000e+00, float 8.000000e+00>): vcvt.f32.s32 d16, d16 vdiv.f32 d16, d17, d16 becomes: vcvt.f32.s32 d16, d16, #3 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@133813 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-21Handle the memory-ness of all U+ ARM constraints.Eric Christopher
Noticed on inspection. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@133553 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-18Remove unused but set variables.Benjamin Kramer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@133347 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-16Mark ldrexd/strexd w/ volatile memory by defaultBruno Cardoso Lopes
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@133175 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-16Revision r128665 added an optimization to make use of NEON multiplierChad Rosier
accumulator forwarding. Specifically (from SVN log entry): Distribute (A + B) * C to (A * C) + (B * C) to make use of NEON multiplier accumulator forwarding: vadd d3, d0, d1 vmul d3, d3, d2 => vmul d3, d0, d2 vmla d3, d1, d2 Make sure it catches cases where operand 1 is add/fadd/sub/fsub, which was intended in the original revision. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@133127 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-15A minor simplification: no functional change.Bob Wilson
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@133047 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-15PerformBFICombine - (bfi A, (and B, Mask1), Mask2) -> (bfi A, B, Mask2) iffEvan Cheng
the bits being cleared by the AND are not demanded by the BFI. The previous BFI dag combine rule was actually incorrect (or used to be correct until BFI representation changed). rdar://9609030 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@133034 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-14Add an optimization that looks for a specific pair-wise add pattern and ↵Tanya Lattner
generates a vpaddl instruction instead of scalarizing the add. Includes a test case. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@133027 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-14Add one more argument to the prefetch intrinsic to indicate whether it's a dataBruno Cardoso Lopes
or instruction cache access. Update the targets to match it and also teach autoupgrade. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@132976 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2026-02-11 09:39:33 +0000