aboutsummaryrefslogtreecommitdiff
path: root/lib/CodeGen/SelectionDAG
AgeCommit message (Collapse)Author
2013-02-20DAGCombiner: Fold pointless truncate, bitcast, buildvector seriesArnold Schwaighofer
(2xi32) (truncate ((2xi64) bitcast (buildvector i32 a, i32 x, i32 b, i32 y))) can be folded into a (2xi32) (buildvector i32 a, i32 b). Such a DAG would cause uneccessary vdup instructions followed by vmovn instructions. We generate this code on ARM NEON for a setcc olt, 2xf64, 2xf64. For example, in the vectorized version of the code below. double A[N]; double B[N]; void test_double_compare_to_double() { int i; for(i=0;i<N;i++) A[i] = (double)(A[i] < B[i]); } radar://13191881 Fixes bug 15283. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175670 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-20Fix PR15267Michael Liao
- When extloading from a vector with non-byte-addressable element, e.g. <4 x i1>, the current logic breaks. Extend the current logic to fix the case where the element type is not byte-addressable by loading all bytes, bit-extracting/packing each element. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175642 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-20Move the SplatByte helper to APInt and generalize it a bit.Benjamin Kramer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175621 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-20Fix #includes, so we include only what we really need.Jakub Staszak
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175581 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-18[ms-inline asm] Remove a redundant call to the setHasMSInlineAsm function.Chad Rosier
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175456 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-18[ms-inline asm] Fix undefined behavior to reset hasMSInlineAsm in advance of ↵NAKAMURA Takumi
SelectAllBasicBlocks(). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175422 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-16LegalizeDAG.cpp doesn't need DenseMap.Jakub Staszak
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175365 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-16[ms-inline asm] Do not omit the frame pointer if we have ms-inline assembly.Chad Rosier
If the frame pointer is omitted, and any stack changes occur in the inline assembly, e.g.: "pusha", then any C local variable or C argument references will be incorrect. I pass no judgement on anyone who would do such a thing. ;) rdar://13218191 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175334 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-15Use the 'target-features' and 'target-cpu' attributes to reset the subtarget ↵Bill Wendling
features. If two functions require different features (e.g., `-mno-sse' vs. `-msse') then we want to honor that, especially during LTO. We can do that by resetting the subtarget's features depending upon the 'target-feature' attribute. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175314 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-15enable SDISel sincos optimization for GNU environmentsPaul Redmond
- add sincos to runtime library if target triple environment is GNU - added canCombineSinCosLibcall() which checks that sincos is in the RTL and if the environment is GNU then unsafe fpmath is enabled (required to preserve errno) - extended sincos-opt lit test Reviewed by: Hal Finkel git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175283 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-14Dont merge consecutive loads/stores into vectors when noimplicitfloat is used.Nadav Rotem
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175190 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-14Add some legality checks for SETCC before introducing it in the DAG combiner ↵Owen Anderson
post-operand legalization. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175149 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-12Add static cast to unsigned char whenever a character classification ↵Guy Benyei
function is called with a signed char argument, in order to avoid assertions in Windows Debug configuration. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175006 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-12PR14562 - Truncation of left shift became undefPaul Redmond
DAGCombiner::ReduceLoadWidth was converting (trunc i32 (shl i64 v, 32)) into (shl i32 v, 32) into undef. To prevent this, check the shift count against the final result size. Patch by: Kevin Schoedel Reviewed by: Nadav Rotem git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@174972 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-12Check type for legality before forming a select from loads.Pete Cooper
Sorry for the lack of a test case. I tried writing one for i386 as i know selects are illegal on this target, but they are actually considered legal by isel and expanded later. I can't see any targets to trigger this, but checking for the legality of a node before forming it is general goodness. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@174934 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-11Currently, codegen may spent some time in SDISel passes even if an entireEvan Cheng
function is successfully handled by fast-isel. That's because function arguments are *always* handled by SDISel. Introduce FastLowerArguments to allow each target to provide hook to handle formal argument lowering. As a proof-of-concept, add ARMFastIsel::FastLowerArguments to handle functions with 4 or fewer scalar integer (i8, i16, or i32) arguments. It completely eliminates the need for SDISel for trivial functions. rdar://13163905 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@174855 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-11Remove unnecessary code.Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@174854 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-08DAGCombiner: Constant folding around pre-increment loads/storesHal Finkel
Previously, even when a pre-increment load or store was generated, we often needed to keep a copy of the original base register for use with other offsets. If all of these offsets are constants (including the offset which was combined into the addressing mode), then this is clearly unnecessary. This change adjusts these other offsets to use the new incremented address. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@174746 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-08Revert 172027 and 174336. Remove diagnostics about over-aligned stack objects.Bob Wilson
Aside from the question of whether we report a warning or an error when we can't satisfy a requested stack object alignment, the current implementation of this is not good. We're not providing any source location in the diagnostics and the current warning is not connected to any warning group so you can't control it. We could improve the source location somewhat, but we can do a much better job if this check is implemented in the front-end, so let's do that instead. <rdar://problem/13127907> git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@174741 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-06Tweak check to avoid integer overflow (for insanely large alignments)Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@174482 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-05Reapply r174343, with a fix for a scary DAG combine bug where it failed to ↵Owen Anderson
differentiate between the alignment of the base point of a load, and the overall alignment of the load. This caused infinite loops in DAG combine with the original application of this patch. ORIGINAL COMMIT LOG: When the target-independent DAGCombiner inferred a higher alignment for a load, it would replace the load with one with the higher alignment. However, it did not place the new load in the worklist, which prevented later DAG combines in the same phase (for example, target-specific combines) from ever seeing it. This patch corrects that oversight, and updates some tests whose output changed due to slightly different DAGCombine outputs. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@174431 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-05Revert r174343, "When the target-independent DAGCombiner inferred a higher ↵NAKAMURA Takumi
alignment for a load," It caused hangups in compiling clang/lib/Parse/ParseDecl.cpp and clang/lib/Driver/Tools.cpp in stage2 on some hosts. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@174374 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-05When the target-independent DAGCombiner inferred a higher alignment for a load,Owen Anderson
it would replace the load with one with the higher alignment. However, it did not place the new load in the worklist, which prevented later DAG combines in the same phase (for example, target-specific combines) from ever seeing it. This patch corrects that oversight, and updates some tests whose output changed due to slightly different DAGCombine outputs. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@174343 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-04SelectionDAG: Teach FoldConstantArithmetic how to deal with vectors.Benjamin Kramer
This required disabling a PowerPC optimization that did the following: input: x = BUILD_VECTOR <i32 16, i32 16, i32 16, i32 16> lowered to: tmp = BUILD_VECTOR <i32 8, i32 8, i32 8, i32 8> x = ADD tmp, tmp The add now gets folded immediately and we're back at the BUILD_VECTOR we started from. I don't see a way to fix this currently so I left it disabled for now. Fix some trivially foldable X86 tests too. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@174325 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-02rdar://13126763Shuxin Yang
Fix a bug in DAGCombine. The symptom is mistakenly optimizing expression "x + x*x" into "x * 3.0". git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@174239 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-01Fix errant fallthrough in the generation of the lifetime markers.Nadav Rotem
Found by Alexander Kornienko. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@174207 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-31When lowering memcpys to loads and stores, make sure we don't promote alignmentsLang Hames
past the natural stack alignment. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@174085 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-29Add a special handling case for untyped CopyFromReg node in GetCostForDef() ↵Weiming Zhao
of ScheduleDAGRRList git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@173833 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-29Teach SDISel to combine fsin / fcos into a fsincos node if the followingEvan Cheng
conditions are met: 1. They share the same operand and are in the same BB. 2. Both outputs are used. 3. The target has a native instruction that maps to ISD::FSINCOS node or the target provides a sincos library call. Implemented the generic optimization in sdisel and enabled it for Mac OSX. Also added an additional optimization for x86_64 Mac OSX by using an alternative entry point __sincos_stret which returns the two results in xmm0 / xmm1. rdar://13087969 PR13204 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@173755 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-27Legalizer: Reword comment again, per Duncan's suggestion.Benjamin Kramer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@173625 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-27Legalizer: Add an assert and tweak a comment to clarify the assumptions this ↵Benjamin Kramer
code makes. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@173620 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-27When the legalizer is splitting vector shifts, the result may not have the ↵Benjamin Kramer
right shift amount type. Fix that by adding a cast to the shift expander. This came up with vector shifts on sse-less X86 CPUs. <2 x i64> = shl <2 x i64> <2 x i64> -> i64,i64 = shl i64 i64; shl i64 i64 -> i32,i32,i32,i32 = shl_parts i32 i32 i64; shl_parts i32 i32 i64 Now we cast the last two i64s to the right type. Fixes the crash in PR14668. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@173615 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-25This patch aims to reduce compile time in LegalizeTypes by using SmallDenseMap,Preston Gurd
with an initial number of elements, instead of DenseMap, which has zero initial elements, in order to avoid the copying of elements when the size changes and to avoid allocating space every time LegalizeTypes is run. This patch will not affect the memory footprint, because DenseMap will increase the element size to 64 when the first element is added. Patch by Wan Xiaofei. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@173448 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-22Make APFloat constructor require explicit semantics.Tim Northover
Previously we tried to infer it from the bit width size, with an added IsIEEE argument for the PPC/IEEE 128-bit case, which had a default value. This default value allowed bugs to creep in, where it was inappropriate. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@173138 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-20Revert 172708.Nadav Rotem
The optimization handles esoteric cases but adds a lot of complexity both to the X86 backend and to other backends. This optimization disables an important canonicalization of chains of SEXT nodes and makes SEXT and ZEXT asymmetrical. Disabling the canonicalization of consecutive SEXT nodes into a single node disables other DAG optimizations that assume that there is only one SEXT node. The AVX mask optimizations is one example. Additionally this optimization does not update the cost model. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@172968 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-18Use AttributeSet accessor methods instead of Attribute accessor methods.Bill Wendling
Further encapsulation of the Attribute object. Don't allow direct access to the Attribute object as an aggregate. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@172853 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-18Remove unused parameter. Also use the AttributeSet query methods instead of ↵Bill Wendling
the Attribute query methods. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@172852 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-17Optimization for the following SIGN_EXTEND pairs:Elena Demikhovsky
v8i8 -> v8i64, v8i8 -> v8i32, v4i8 -> v4i64, v4i16 -> v4i64 for AVX and AVX2. Bug 14865. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@172708 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-14This patch addresses an incorrect transformation in the DAG combiner.Bill Schmidt
The included test case is derived from one of the GCC compatibility tests. The problem arises after the selection DAG has been converted to type-legalized form. The combiner first sees a 64-bit load that can be converted into a pre-increment form. The original load feeds into a SRL that isolates the upper 32 bits of the loaded doubleword. This looks like an opportunity for DAGCombiner::ReduceLoadWidth() to replace the 64-bit load with a 32-bit load. However, this transformation is not valid, as the replacement load is not a pre-increment load. The pre-increment load produces an extra result, which feeds a subsequent add instruction. The replacement load only has one result value, and this value is propagated to all uses of the pre- increment load, including the add. Because the add is looking for the second result value as its operand, it ends up attempting to add a constant to a token chain, resulting in a crash. So the patch simply disables this transformation for any load with more than two result values. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@172480 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-12When lowering an inreg sext first shift left, then right arithmetically.Benjamin Kramer
Shifting right two times will only yield zero. Should fix SingleSource/UnitTests/SignlessTypes/factor. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@172322 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-11PPC: Implement efficient lowering of sign_extend_inreg.Nadav Rotem
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@172269 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-11Remove some accidentaly duplicated code. This needs urgent cleanup :(Benjamin Kramer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@172248 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-11Split TargetLowering into a CodeGen and a SelectionDAG part.Benjamin Kramer
This fixes some of the cycles between libCodeGen and libSelectionDAG. It's still a complete mess but as long as the edges consist of virtual call it doesn't cause breakage. BasicTTI did static calls and thus broke some build configurations. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@172246 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-11For inline asm:Eric Christopher
- recognize string "{memory}" in the MI generation - mark as mayload/maystore when there's a memory clobber constraint. PR14859. Patch by Krzysztof Parzyszek git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@172228 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-10PR14896: Handle memcpy from constant string where the memcpy size is larger ↵Evan Cheng
than the string size. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@172124 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-10Remove unneeded includes from FunctionLoweringInfo.h.Jakub Staszak
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@172123 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-10Stack Alignment: throw error if we can't satisfy the minimal alignmentManman Ren
requirement when creating stack objects in MachineFrameInfo. Add CreateStackObjectWithMinAlign to throw error when the minimal alignment can't be achieved and to clamp the alignment when the preferred alignment can't be achieved. Same is true for CreateVariableSizedObject. Will not emit error in CreateSpillStackObject or CreateStackObject. As long as callers of CreateStackObject do not assume the object will be aligned at the requested alignment, we should not have miscompile since later optimizations which look at the object's alignment will have the correct information. rdar://12713765 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@172027 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-09Fix a DAG combine bug visitBRCOND() is transforming br(xor(x, y)) to br(x != y).Evan Cheng
It cahced XOR's operands before calling visitXOR() but failed to update the operands when visitXOR changed the XOR node. rdar://12968664 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171999 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-09Refactor to expose RTLIB calls to targets.Tim Northover
fp128 is almost but not quite completely illegal as a type on AArch64. As a result it needs to have a register class (for argument passing mainly), but all operations need to be lowered to runtime calls. Currently there's no way for targets to do this (without duplicating code), as the relevant functions are hidden in SelectionDAG. This patch changes that. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171971 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-08Add fp128 rtlib function names to LLVMTim Northover
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171867 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-09-03 23:56:15 +0000