aboutsummaryrefslogtreecommitdiff
path: root/test/CodeGen
AgeCommit message (Collapse)Author
2013-04-21Legalize vector truncates by parts rather than just splitting.Jim Grosbach
Rather than just splitting the input type and hoping for the best, apply a bit more cleverness. Just splitting the types until the source is legal often leads to an illegal result time, which is then widened and a scalarization step is introduced which leads to truly horrible code generation. With the loop vectorizer, these sorts of operations are much more common, and so it's worth extra effort to do them well. Add a legalization hook for the operands of a TRUNCATE node, which will be encountered after the result type has been legalized, but if the operand type is still illegal. If simple splitting of both types ends up with the result type of each half still being legal, just do that (v16i16 -> v16i8 on ARM, for example). If, however, that would result in an illegal result type (v8i32 -> v8i8 on ARM, for example), we can get more clever with power-two vectors. Specifically, split the input type, but also widen the result element size, then concatenate the halves and truncate again. For example on ARM, To perform a "%res = v8i8 trunc v8i32 %in" we transform to: %inlo = v4i32 extract_subvector %in, 0 %inhi = v4i32 extract_subvector %in, 4 %lo16 = v4i16 trunc v4i32 %inlo %hi16 = v4i16 trunc v4i32 %inhi %in16 = v8i16 concat_vectors v4i16 %lo16, v4i16 %hi16 %res = v8i8 trunc v8i16 %in16 This allows instruction selection to generate three VMOVN instructions instead of a sequences of moves, stores and loads. Update the ARMTargetTransformInfo to take this improved legalization into account. Consider the simplified IR: define <16 x i8> @test1(<16 x i32>* %ap) { %a = load <16 x i32>* %ap %tmp = trunc <16 x i32> %a to <16 x i8> ret <16 x i8> %tmp } define <8 x i8> @test2(<8 x i32>* %ap) { %a = load <8 x i32>* %ap %tmp = trunc <8 x i32> %a to <8 x i8> ret <8 x i8> %tmp } Previously, we would generate the truly hideous: .syntax unified .section __TEXT,__text,regular,pure_instructions .globl _test1 .align 2 _test1: @ @test1 @ BB#0: push {r7} mov r7, sp sub sp, sp, #20 bic sp, sp, #7 add r1, r0, #48 add r2, r0, #32 vld1.64 {d24, d25}, [r0:128] vld1.64 {d16, d17}, [r1:128] vld1.64 {d18, d19}, [r2:128] add r1, r0, #16 vmovn.i32 d22, q8 vld1.64 {d16, d17}, [r1:128] vmovn.i32 d20, q9 vmovn.i32 d18, q12 vmov.u16 r0, d22[3] strb r0, [sp, #15] vmov.u16 r0, d22[2] strb r0, [sp, #14] vmov.u16 r0, d22[1] strb r0, [sp, #13] vmov.u16 r0, d22[0] vmovn.i32 d16, q8 strb r0, [sp, #12] vmov.u16 r0, d20[3] strb r0, [sp, #11] vmov.u16 r0, d20[2] strb r0, [sp, #10] vmov.u16 r0, d20[1] strb r0, [sp, #9] vmov.u16 r0, d20[0] strb r0, [sp, #8] vmov.u16 r0, d18[3] strb r0, [sp, #3] vmov.u16 r0, d18[2] strb r0, [sp, #2] vmov.u16 r0, d18[1] strb r0, [sp, #1] vmov.u16 r0, d18[0] strb r0, [sp] vmov.u16 r0, d16[3] strb r0, [sp, #7] vmov.u16 r0, d16[2] strb r0, [sp, #6] vmov.u16 r0, d16[1] strb r0, [sp, #5] vmov.u16 r0, d16[0] strb r0, [sp, #4] vldmia sp, {d16, d17} vmov r0, r1, d16 vmov r2, r3, d17 mov sp, r7 pop {r7} bx lr .globl _test2 .align 2 _test2: @ @test2 @ BB#0: push {r7} mov r7, sp sub sp, sp, #12 bic sp, sp, #7 vld1.64 {d16, d17}, [r0:128] add r0, r0, #16 vld1.64 {d20, d21}, [r0:128] vmovn.i32 d18, q8 vmov.u16 r0, d18[3] vmovn.i32 d16, q10 strb r0, [sp, #3] vmov.u16 r0, d18[2] strb r0, [sp, #2] vmov.u16 r0, d18[1] strb r0, [sp, #1] vmov.u16 r0, d18[0] strb r0, [sp] vmov.u16 r0, d16[3] strb r0, [sp, #7] vmov.u16 r0, d16[2] strb r0, [sp, #6] vmov.u16 r0, d16[1] strb r0, [sp, #5] vmov.u16 r0, d16[0] strb r0, [sp, #4] ldm sp, {r0, r1} mov sp, r7 pop {r7} bx lr Now, however, we generate the much more straightforward: .syntax unified .section __TEXT,__text,regular,pure_instructions .globl _test1 .align 2 _test1: @ @test1 @ BB#0: add r1, r0, #48 add r2, r0, #32 vld1.64 {d20, d21}, [r0:128] vld1.64 {d16, d17}, [r1:128] add r1, r0, #16 vld1.64 {d18, d19}, [r2:128] vld1.64 {d22, d23}, [r1:128] vmovn.i32 d17, q8 vmovn.i32 d16, q9 vmovn.i32 d18, q10 vmovn.i32 d19, q11 vmovn.i16 d17, q8 vmovn.i16 d16, q9 vmov r0, r1, d16 vmov r2, r3, d17 bx lr .globl _test2 .align 2 _test2: @ @test2 @ BB#0: vld1.64 {d16, d17}, [r0:128] add r0, r0, #16 vld1.64 {d18, d19}, [r0:128] vmovn.i32 d16, q8 vmovn.i32 d17, q9 vmovn.i16 d16, q8 vmov r0, r1, d16 bx lr git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179989 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-21ARM: Split out cost model vcvt testcases.Jim Grosbach
They had a separate RUN line already, so may as well be in a separate file. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179988 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-21Passing arguments to varags functions under the SPARC v9 ABI.Jakob Stoklund Olesen
Arguments after the fixed arguments never use the floating point registers. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179987 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-21Fix the SETHIimm pattern for 64-bit code.Jakob Stoklund Olesen
Don't ignore the high 32 bits of the immediate. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179985 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-21ARM: fix part of test which actually needed an asserts buildTim Northover
This should fix a buildbot failure that occurred after r179977. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179978 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-21ARM: Use ldrd/strd to spill 64-bit pairs when available.Tim Northover
This allows common sp-offsets to be part of the instruction and is probably faster on modern CPUs too. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179977 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-21Remove tbaa metadata.Bill Wendling
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179970 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-20Compile varargs functions for SPARCv9.Jakob Stoklund Olesen
With a little help from the frontend, it looks like the standard va_* intrinsics can do the job. Also clean up an old bitcast hack in LowerVAARG that dealt with unaligned double loads. Load SDNodes can specify an alignment now. Still missing: Calling varargs functions with float arguments. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179961 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-20ARM: don't add FrameIndex offset for LDMIA (has no immediate)Tim Northover
Previously, when spilling 64-bit paired registers, an LDMIA with both a FrameIndex and an offset was produced. This kind of instruction shouldn't exist, and the extra operand was being confused with the predicate, causing aborts later on. This removes the invalid 0-offset from the instruction being produced. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179956 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-20Minor renaming of tests (for consistency with an in-development patch)Stephen Lin
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179954 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-20Don't litter .s files in test directory.Benjamin Kramer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179937 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-20Add CodeGen support for functions that always return arguments via a new ↵Stephen Lin
parameter attribute 'returned', which is taken advantage of in target-independent tail call opportunity detection and in ARM call lowering (when placed on an integral first parameter). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179925 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-20Allow tail call opportunity detection through nested and/or multiple ↵Stephen Lin
iterations of extractelement/insertelement indirection git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179924 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-19[mips] Instruction selection patterns for DSP-ASE vector shifts.Akira Hatanaka
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179906 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-19Fix PPC optimizeCompareInstr swapped-sub argument handlingHal Finkel
When matching a compare with a subtract where the arguments of the compare are swapped w.r.t. the arguments of the subtract, we need to negate the predicates (or CR bit indices) of the users. This, however, is not the same as inverting the predicate (negating LT -> GT, but inverting LT -> GE, for example). The ARM backend seems to do this correctly, but when I adapted the code for the PPC backend, I introduced an error in this logic. Comparison optimization is now enabled again by default. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179899 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-19Do not mangle in MS-way the globals with magic \001 in the name.Anton Korobeynikov
Based on the patch by David Nadlinger! git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179889 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-19Make test slightly more readable.Bill Wendling
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179888 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-19Add a testcase to make sure we generate the proper compact unwind section ↵Bill Wendling
for a function that cannot produce a compact unwind encoding. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179887 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-19Revert "PR14606: debug info imported_module support"Eric Christopher
This reverts commit r179836 as it seems to have caused test failures. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179840 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-19PR14606: debug info imported_module supportDavid Blaikie
Adding another CU-wide list, in this case of imported_modules (since they should be relatively rare, it seemed better to add a list where each element had a "context" value, rather than add a (usually empty) list to every scope). This takes care of DW_TAG_imported_module, but to fully address PR14606 we'll need to expand this to cover DW_TAG_imported_declaration too. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179836 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-19R600: Add pattern for the BFI_INT instructionTom Stellard
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179830 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-19R600: Reorganize lit tests and document how they should be organizedTom Stellard
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179828 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-18Disable PPC comparison optimization by defaultHal Finkel
This seems to cause a stage-2 LLVM compile failure (by crashing TableGen); do I'm disabling this for now. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179807 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-18Implement optimizeCompareInstr for PPCHal Finkel
Many PPC instructions have a so-called 'record form' which stores to a specific condition register the result of comparing the result of the instruction with zero (always as a signed comparison). For integer operations on PPC64, this is always a 64-bit comparison. This implementation is derived from the implementation in the ARM backend; there are some differences because PPC condition registers are allocatable virtual registers (although the record forms always use a specific one), and we look for a matching subtraction instruction after the compare (but before the first use) in addition to before it. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179802 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-18X86: Add an SSE2 lowering for 64 bit compares when pcmpgtq (SSE4.2) isn't ↵Benjamin Kramer
available. This pattern started popping up in vectorized min/max reductions. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179797 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-18Allow misaligned stores in x86 fast-isel.Derek Schuff
In X86FastISel::X86SelectStore(), improperly aligned stores are rejected and handled by the DAG-based ISel. However, X86FastISel::X86SelectLoad() makes no such requirement. There doesn't appear to be an x86 architectural correctness issue with allowing potentially unaligned store instructions. This patch removes this restriction. Patch by Jim Stichnot. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179774 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-18Fix for PR14824, An ARM Load/Store Optimization bugHao Liu
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179751 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-17This patch teaches x86 fast-isel to generate the native div/idiv instructionsEli Bendersky
for the sdiv/srem/udiv/urem bitcode instructions. This is done for the i8, i16, and i32 types, as well as i64 for the x86_64 target. Patch by Jim Stichnoth git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179715 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-17R600: Make Export Instruction not duplicableVincent Lejeune
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179686 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-16[XCore] Extend test to check positve offsets are folded into addresses.Richard Osborne
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179621 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-16[XCore] Give test more generic name.Richard Osborne
I intend to extend the test with more offset folding checks git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179620 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-16[XCore] Convert a couple of tests to FileCheck.Richard Osborne
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179619 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-16Implement ARM unwind opcode assembler.Logan Chien
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179591 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-16Add 64-bit multiply and divide instructions for SPARC v9.Jakob Stoklund Olesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179582 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-15R600/SI: Emit config values in register value pairs.Tom Stellard
Instead of emitting config values in a predefined order, the code emitter will now emit a 32-bit register index followed by the 32-bit config value. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179546 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-15R600/SI: Emit configuration value in the .AMDGPU.config ELF sectionTom Stellard
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179545 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-15R600: Emit ELF formatted code rather than raw ISA.Tom Stellard
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179544 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-15Avoid outputting temporary test file into source tree.Tim Northover
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179532 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-15Fix PPC64 CR spill location for callee-saved registersHal Finkel
This fixes an ABI bug for non-Darwin PPC64. For the callee-saved condition registers, the spill location is specified relative to the stack pointer (SP + 8). However, this is not relative to the SP after the new stack frame is established, but instead relative to the caller's stack pointer (it is stored into the linkage area of the parent's stack frame). So, like with the link register, we don't directly spill the CRs with other callee-saved registers, but just mark them to be spilled during prologue generation. In practice, this reverts r179457 for PPC64 (but leaves it in place for PPC32). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179500 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-14Use i32 for all SPARC shift amounts, even in 64-bit mode.Jakob Stoklund Olesen
Test case by llvm-stress. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179477 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-14Add support for the abs64 SPARC v9 code model.Jakob Stoklund Olesen
For when 16 TB just isn't enough. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179474 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-14Add support for the SPARC v9 abs44 code model.Jakob Stoklund Olesen
This is the default model for non-PIC 64-bit code. It supports text+data+bss linked anywhere in the low 16 TB of the address space. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179473 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-14Also put target flags on SPARC constant pool references.Jakob Stoklund Olesen
Constant pool entries are accessed exactly the same way as global variables. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179471 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-14Fix patterns for 64-bit pointers.Jakob Stoklund Olesen
This fixes the pic32 code model for SPARC v9. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179469 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-13Define SPARC code models.Jakob Stoklund Olesen
Currently, only abs32 and pic32 are implemented. Add a test case for abs32 with 64-bit code. 64-bit PIC code is currently broken. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179463 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-13Spill and restore PPC CR registers using the FP when we have oneHal Finkel
For functions that need to spill CRs, and have dynamic stack allocations, the value of the SP during the restore is not what it was during the save, and so we need to use the FP in these cases (as for all of the other spills and restores, but the CR restore has a special code path because its reserved slot, like the link register, is specified directly relative to the adjusted SP). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179457 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-13Further generalize this scheduler test.Andrew Trick
The order of copies depends on queue order, which is not very stable. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179456 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-13Fix a dislexic regex.Andrew Trick
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179455 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-13Add a missing REQUIRES: assertsAndrew Trick
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179453 91177308-0d34-0410-b5e6-96231b3b80d8
2013-04-13MI-Sched: schedule physreg copies.Andrew Trick
The register allocator expects minimal physreg live ranges. Schedule physreg copies accordingly. This is slightly tricky when they occur in the middle of the scheduling region. For now, this is handled by rescheduling the copy when its associated instruction is scheduled. Eventually we may instead bundle them, but only if we can preserve the bundles as parallel copies during regalloc. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179449 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-18-g5258 (git 2.32.0) at 2025-04-23 05:18:24 +0000