aboutsummaryrefslogtreecommitdiff
path: root/lib/Target/ARM/ARMISelLowering.cpp
AgeCommit message (Collapse)Author
2011-03-29Add intrinsics @llvm.arm.neon.vmulls and @llvm.arm.neon.vmullu.* back. FrontendsEvan Cheng
was lowering them to sext / uxt + mul instructions. Unfortunately the optimization passes may hoist the extensions out of the loop and separate them. When that happens, the long multiplication instructions can be broken into several scalar instructions, causing significant performance issue. Note the vmla and vmls intrinsics are not added back. Frontend will codegen them as intrinsics vmull* + add / sub. Also note the isel optimizations for catching mul + sext / zext are not changed either. First part of rdar://8832507, rdar://9203134 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@128502 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-29Add Neon SINT_TO_FP and UINT_TO_FP lowering from v4i16 to v4f32. FixesCameron Zwarich
<rdar://problem/8875309> and <rdar://problem/9057191>. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@128492 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-29Optimizing (zext A + zext B) * C, to (VMULL A, C) + (VMULL B, C) during ↵Evan Cheng
isel lowering to fold the zero-extend's and take advantage of no-stall back to back vmul + vmla: vmull q0, d4, d6 vmlal q0, d5, d6 is faster than vaddl q0, d4, d5 vmovl q1, d6 vmul q0, q0, q1 This allows us to vmull + vmlal for: f = vmull_u8( vget_high_u8(s), c); f = vmlal_u8(f, vget_low_u8(s), c); rdar://9197392 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@128444 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-26Fix the bfi handling for or (and a mask) (and b mask). We need the twoEric Christopher
masks to match inversely for the code as is to work. For the example given we actually want: bfi r0, r2, #1, #1 not #0, however, given the way the pattern is written it's not possible at the moment. Fixes rdar://9177502 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@128320 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-21Re-apply r127953 with fixes: eliminate empty return block if it has no ↵Evan Cheng
predecessors; update dominator tree if cfg is modified. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127981 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-19Revert r127953, "SimplifyCFG has stopped duplicating returns into predecessorsDaniel Dunbar
to canonicalize IR", it broke a lot of things. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127954 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-19SimplifyCFG has stopped duplicating returns into predecessors to canonicalize IREvan Cheng
to have single return block (at least getting there) for optimizations. This is general goodness but it would prevent some tailcall optimizations. One specific case is code like this: int f1(void); int f2(void); int f3(void); int f4(void); int f5(void); int f6(void); int foo(int x) { switch(x) { case 1: return f1(); case 2: return f2(); case 3: return f3(); case 4: return f4(); case 5: return f5(); case 6: return f6(); } } => LBB0_2: ## %sw.bb callq _f1 popq %rbp ret LBB0_3: ## %sw.bb1 callq _f2 popq %rbp ret LBB0_4: ## %sw.bb3 callq _f3 popq %rbp ret This patch teaches codegenprep to duplicate returns when the return value is a phi and where the phi operands are produced by tail calls followed by an unconditional branch: sw.bb7: ; preds = %entry %call8 = tail call i32 @f5() nounwind br label %return sw.bb9: ; preds = %entry %call10 = tail call i32 @f6() nounwind br label %return return: %retval.0 = phi i32 [ %call10, %sw.bb9 ], [ %call8, %sw.bb7 ], ... [ 0, %entry ] ret i32 %retval.0 This allows codegen to generate better code like this: LBB0_2: ## %sw.bb jmp _f1 ## TAILCALL LBB0_3: ## %sw.bb1 jmp _f2 ## TAILCALL LBB0_4: ## %sw.bb3 jmp _f3 ## TAILCALL rdar://9147433 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127953 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-15The VTBL (and VTBX) instructions are rather permissive concerning the masks theyBill Wendling
accept. If a value in the mask is out of range, it uses the value 0, for VTBL, or leaves the value unchanged, for VTBX. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127700 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-15Some minor cleanups based on feedback.Bill Wendling
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127694 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-14Generate a VTBL instruction instead of a series of loads and stores when weBill Wendling
can. As Nate pointed out, VTBL isn't super performant, but it *has* to be better than this: _shuf: @ BB#0: @ %entry push {r4, r7, lr} add r7, sp, #4 sub sp, #12 mov r4, sp bic r4, r4, #7 mov sp, r4 mov r2, sp vmov d16, r0, r1 orr r0, r2, #6 orr r3, r2, #7 vst1.8 {d16[0]}, [r3] vst1.8 {d16[5]}, [r0] subs r4, r7, #4 orr r0, r2, #5 vst1.8 {d16[4]}, [r0] orr r0, r2, #4 vst1.8 {d16[4]}, [r0] orr r0, r2, #3 vst1.8 {d16[0]}, [r0] orr r0, r2, #2 vst1.8 {d16[2]}, [r0] orr r0, r2, #1 vst1.8 {d16[1]}, [r0] vst1.8 {d16[3]}, [r2] vldr.64 d16, [sp] vmov r0, r1, d16 mov sp, r4 pop {r4, r7, pc} The "illegal" testcase in vext.ll is no longer illegal. <rdar://problem/9078775> git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127630 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-14Indentation.Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127595 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-08Fix a compiler crash where a Glue value had multiple uses. Radar 9049552.Bob Wilson
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127198 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-08Fix comment typos.Bob Wilson
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127197 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-07Move getRegPressureLimit() from TargetLoweringInfo to TargetRegisterInfo.Cameron Zwarich
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127175 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-05Remove unused conditional negate operations.Bob Wilson
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127090 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-28Fix a typo which cause dag combine crash. rdar://9059537.Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@126661 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-28Support for byval parameters on ARM. Will be enabled by a forthcomingStuart Hastings
patch to the front-end. Radar 7662569. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@126655 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-23More fcopysign correctness and performance fix.Evan Cheng
The previous codegen for the slow path (when values are in VFP / NEON registers) was incorrect if the source is NaN. The new codegen uses NEON vbsl instruction to copy the sign bit. e.g. vmov.i32 d1, #0x80000000 vbsl d1, d2, d0 If NEON is not available, it uses integer instructions to copy the sign bit. rdar://9034702 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@126295 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-21Revert r124611 - "Keep track of incoming argument's location while emitting ↵Devang Patel
LiveIns." In other words, do not keep track of argument's location. The debugger (gdb) is not prepared to see line table entries for arguments. For the debugger, "second" line table entry marks beginning of function body. This requires some coordination with debugger to get this working. - The debugger needs to be aware of prolog_end attribute attached with line table entries. - The compiler needs to accurately mark prolog_end in line table entries (at -O0 and at -O1+) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@126155 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-11Implement sdiv & udiv for <4 x i16> and <8 x i8> NEON vector types.Nate Begeman
This avoids moving each element to the integer register file and calling __divsi3 etc. on it. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@125402 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-11Fix buggy fcopysign lowering.Evan Cheng
This define float @foo(float %x, float %y) nounwind readnone { entry: %0 = tail call float @copysignf(float %x, float %y) nounwind readnone ret float %0 } Was compiled to: vmov s0, r1 bic r0, r0, #-2147483648 vmov s1, r0 vcmpe.f32 s0, #0 vmrs apsr_nzcv, fpscr it lt vneglt.f32 s1, s1 vmov r0, s1 bx lr This fails to copy the sign of -0.0f because it's lost during the float to int conversion. Also, it's sub-optimal when the inputs are in GPR registers. Now it uses integer and + or operations when it's profitable. And it's correct! lsrs r1, r1, #31 bfi r0, r1, #31, #1 bx lr rdar://8984306 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@125357 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-07Fix an obvious typo which caused an isel assertion. rdar://8964854.Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@125023 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-07Add codegen support for using post-increment NEON load/store instructions.Bob Wilson
The vld1-lane, vld1-dup and vst1-lane instructions do not yet support using post-increment versions, but all the rest of the NEON load/store instructions should be handled now. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@125014 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-02Given a pair of floating point load and store, if there are no other uses ofEvan Cheng
the load, then it may be legal to transform the load and store to integer load and store of the same width. This is done if the target specified the transformation as profitable. e.g. On arm, this can transform: vldr.32 s0, [] vstr.32 s0, [] to ldr r12, [] str r12, [] rdar://8944252 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124708 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-31Keep track of incoming argument's location while emitting LiveIns.Devang Patel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124611 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-24Provide correct registers for EH stuff on ARMAnton Korobeynikov
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124151 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-21Last round of fixes for movw + movt global address codegen.Evan Cheng
1. Fixed ARM pc adjustment. 2. Fixed dynamic-no-pic codegen 3. CSE of pc-relative load of global addresses. It's now enabled by default for Darwin. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123991 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-20Sorry, several patches in one.Evan Cheng
TargetInstrInfo: Change produceSameValue() to take MachineRegisterInfo as an optional argument. When in SSA form, targets can use it to make more aggressive equality analysis. Machine LICM: 1. Eliminate isLoadFromConstantMemory, use MI.isInvariantLoad instead. 2. Fix a bug which prevent CSE of instructions which are not re-materializable. 3. Use improved form of produceSameValue. ARM: 1. Teach ARM produceSameValue to look pass some PIC labels. 2. Look for operands from different loads of different constant pool entries which have same values. 3. Re-implement PIC GA materialization using movw + movt. Combine the pair with a "add pc" or "ldr [pc]" to form pseudo instructions. This makes it possible to re-materialize the instruction, allow machine LICM to hoist the set of instructions out of the loop and make it possible to CSE them. It's a bit hacky, but it significantly improve code quality. 4. Some minor bug fixes as well. With the fixes, using movw + movt to materialize GAs significantly outperform the load from constantpool method. 186.crafty and 255.vortex improved > 20%, 254.gap and 176.gcc ~10%. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123905 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-19For ARM subtargets with useNEONForSinglePrecisionFP, double count usesAndrew Trick
of the floating point types less than 64-bits. It's somewhat of a temporary hack but forces more accurate modeling of register pressure and results in fewer spills. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123811 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-19whitespaceAndrew Trick
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123810 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-19Don't forget to emit the load from indirect symbol when using movw + movt to ↵Evan Cheng
materialize GA indirect symbols. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123809 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-17Materialize GA addresses with movw + movt pairs for Darwin in PIC mode. e.g.Evan Cheng
movw r0, :lower16:(L_foo$non_lazy_ptr-(LPC0_0+4)) movt r0, :upper16:(L_foo$non_lazy_ptr-(LPC0_0+4)) LPC0_0: add r0, pc, r0 It's not yet enabled by default as some tests are failing. I suspect bugs in down stream tools. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123619 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-14Fix 80-cols.Eric Christopher
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123494 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-10Rename TargetFrameInfo into TargetFrameLowering. Also, put couple of FIXMEs ↵Anton Korobeynikov
and fixes here and there. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123170 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-10Simplify a bunch of isVirtualRegister() and isPhysicalRegister() logic.Jakob Stoklund Olesen
These functions not longer assert when passed 0, but simply return false instead. No functional change intended. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123155 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-08Recognize inline asm 'rev /bin/bash, ' as a bswap intrinsic call.Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123048 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-07Add an explanatory message for an assertion.Bob Wilson
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123042 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-07Eliminate variable only used in debug builds.Matt Beaumont-Gay
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123040 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-07Lower some BUILD_VECTORS using VEXT+shuffle.Bob Wilson
Patch by Tim Northover. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123035 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-07Add ARM patterns to match EXTRACT_SUBVECTOR nodes.Bob Wilson
Also fix an off-by-one in SelectionDAGBuilder that was preventing shuffle vectors from being translated to EXTRACT_SUBVECTOR. Patch by Tim Northover. The test changes are needed to keep those spill-q tests from testing aligned spills and restores. If the only aligned stack objects are spill slots, we no longer realign the stack frame. Prior to this patch, an EXTRACT_SUBVECTOR was legalized by loading from the stack, which created an aligned frame index. Now, however, there is nothing except the spill slot in the stack frame, so I added an aligned alloca. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122995 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-06Re-implement r122936 with proper target hooks. Now getMaxStoresPerMemcpyEvan Cheng
etc. takes an option OptSize. If OptSize is true, it would return the inline limit for functions with attribute OptSize. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122952 91177308-0d34-0410-b5e6-96231b3b80d8
2010-12-23Radar 8803471: Fix expansion of ARM BCCi64 pseudo instructions.Bob Wilson
If the basic block containing the BCCi64 (or BCCZi64) instruction ends with an unconditional branch, that branch needs to be deleted before appending the expansion of the BCCi64 to the end of the block. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122521 91177308-0d34-0410-b5e6-96231b3b80d8
2010-12-21Add ARM-specific DAG combining to cast i64 vector element load/stores to f64.Bob Wilson
Type legalization splits up i64 values into pairs of i32 values, which leads to poor quality code when inserting or extracting i64 vector elements. If the vector element is loaded or stored, it can be treated as an f64 value and loaded or stored directly from a VPR register. Use the pre-legalization DAG combiner to cast those vector elements to f64 types so that the type legalizer won't mess them up. Radar 8755338. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122319 91177308-0d34-0410-b5e6-96231b3b80d8
2010-12-21rename MVT::Flag to MVT::Glue. "Flag" is a terrible name forChris Lattner
something that just glues two nodes together, even if it is sometimes used for flags. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122310 91177308-0d34-0410-b5e6-96231b3b80d8
2010-12-18Add some missing entries in ARMTargetLowering::getTargetNodeName.Bob Wilson
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122111 91177308-0d34-0410-b5e6-96231b3b80d8
2010-12-15Don't handle -arm-long-calls in fast isel for now.Eric Christopher
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@121919 91177308-0d34-0410-b5e6-96231b3b80d8
2010-12-14bfi A, (and B, C1), C2) -> bfi A, B, C2 iff C1 & C2 == C1. rdar://8458663Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@121746 91177308-0d34-0410-b5e6-96231b3b80d8
2010-12-13Generalize BFI isel lowering a bit.Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@121714 91177308-0d34-0410-b5e6-96231b3b80d8
2010-12-11(or (and (shl A, #shamt), mask), B) => ARMbfi B, A, ~mask where lsb(mask) == ↵Evan Cheng
#shamt. rdar://8752056 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@121606 91177308-0d34-0410-b5e6-96231b3b80d8
2010-12-07PR5207: Change APInt methods trunc(), sext(), zext(), sextOrTrunc() andJay Foad
zextOrTrunc(), and APSInt methods extend(), extOrTrunc() and new method trunc(), to be const and to return a new value instead of modifying the object in place. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@121120 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2026-02-12 12:53:52 +0000