aboutsummaryrefslogtreecommitdiff
path: root/test/CodeGen
AgeCommit message (Collapse)Author
2008-01-18UnXFAIL this test.Tanya Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_22@46169 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-17When a live virtual register is being clobbered by an implicit def, it is ↵Evan Cheng
spilled and the spill is its kill. However, if the local allocator has determined the register has not been modified (possible when its value was reloaded), it would not issue a restore. In that case, mark the last use of the virtual register as kill. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46111 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-16Fixes a nasty dag combiner bug that causes a bunch of tests to fail at -O0.Evan Cheng
It's not safe to use the two value CombineTo variant to combine away a dead load. e.g. v1, chain2 = load chain1, loc v2, chain3 = load chain2, loc v3 = add v2, c Now we replace use of v1 with undef, use of chain2 with chain1. ReplaceAllUsesWith() will iterate through uses of the first load and update operands: v1, chain2 = load chain1, loc v2, chain3 = load chain1, loc v3 = add v2, c Now the second load is the same as the first load, SelectionDAG cse will ensure the use of second load is replaced with the first load. v1, chain2 = load chain1, loc v3 = add v1, c Then v1 is replaced with undef and bad things happen. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46099 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-16Trampoline support for x86-64. This looks likeDuncan Sands
it should work, but I have no machine to test it on. Committed because it will at least cause no harm, and maybe someone can test it for me! git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46098 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-16add testcase for regressionChris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46073 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-16make sure to use a cpu that has sse.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46060 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-16My previous commit had an incomplete message, it should have been:Chris Lattner
make the 'fp return in ST(0)' optimization smart enough to look through token factor nodes. THis allows us to compile testcases like CodeGen/X86/fp-stack-retcopy.ll into: _carg: subl $12, %esp call L_foo$stub fstpl (%esp) fldl (%esp) addl $12, %esp ret instead of: _carg: subl $28, %esp call L_foo$stub fstpl 16(%esp) movsd 16(%esp), %xmm0 movsd %xmm0, 8(%esp) fldl 8(%esp) addl $28, %esp ret Still not optimal, but much better and this is a trivial patch. Fixing the rest requires invasive surgery that is is not llvm 2.2 material. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46054 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-15verify x86 generates ud2 for llvm.trapChris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46023 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-15new testcase for llvm.trap.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@46020 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-11More CellSPU refinements:Scott Michel
- struct_2.ll: Completely unaligned load/store testing - call_indirect.ll, struct_1.ll: Add test lines to exercise X-form [$reg($reg)] addressing At this point, loads and stores should be under control (he says in an optimistic tone of voice.) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45882 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-11Disable for now.Dale Johannesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45881 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-11More CellSPU refinement and progress:Scott Michel
- Cleaned up custom load/store logic, common code is now shared [see note below], cleaned up address modes - More test cases: various intrinsics, structure element access (load/store test), updated target data strings, indirect function calls. Note: This patch contains a refactoring of the LoadSDNode and StoreSDNode structures: they now share a common base class, LSBaseSDNode, that provides an interface to their common functionality. There is some hackery to access the proper operand depending on the derived class; otherwise, to do a proper job would require finding and rearranging the SDOperands sent to StoreSDNode's constructor. The current refactor errs on the side of being conservatively and backwardly compatible while providing functionality that reduces redundant code for targets where loads and stores are custom-lowered. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45851 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-10Output sinl for a long double FSIN node, not sin.Duncan Sands
Likewise fix up a bunch of other libcalls. While there I remove NEG_F32 and NEG_F64 since they are not used anywhere. This fixes 9 Ada ACATS failures. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45833 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-10Codegen improvement has reduced one spill.Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45814 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-10new testcase for PR1845Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45795 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-09Special copy SUnit's do not have SDNode's.Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45787 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-09Fix sse2.psrl.w and sse2.psrl.q definitions.Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45772 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-08Make load->store deletion a bit smarter. This allows us to compile this:Chris Lattner
void test(long long *P) { *P ^= 1; } into just: _test: movl 4(%esp), %eax xorl $1, (%eax) ret instead of code like this: _test: movl 4(%esp), %ecx xorl $1, (%ecx) movl 4(%ecx), %edx movl %edx, 4(%ecx) ret git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45762 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-08Crashes llc when using Chris's new legalization logic.Duncan Sands
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45758 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-08remove darwin/i386 t-tChris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45743 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-08Finally implement correct ordered comparisons for PPC, even thoughChris Lattner
the code generated is not wonderful. This turns a miscompilation into a code quality bug (noted in the ppc readme). This fixes PR642, which is over 2 years old (!). Nate, please review this. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45742 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-07Update test to catch recent x86 insert regression and improvementsNate Begeman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45705 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-07Setting GlobalDirective in TargetAsmInfo by default rather thanGordon Henriksen
providing a misleading facility. It's used once in the MIPS backend and hardcoded as "\t.globl\t" everywhere else. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45676 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-07With this patch, the LowerGC transformation becomes theGordon Henriksen
ShadowStackCollector, which additionally has reduced overhead with no sacrifice in portability. Considering a function @fun with 8 loop-local roots, ShadowStackCollector introduces the following overhead (x86): ; shadowstack prologue movl L_llvm_gc_root_chain$non_lazy_ptr, %eax movl (%eax), %ecx movl $___gc_fun, 20(%esp) movl $0, 24(%esp) movl $0, 28(%esp) movl $0, 32(%esp) movl $0, 36(%esp) movl $0, 40(%esp) movl $0, 44(%esp) movl $0, 48(%esp) movl $0, 52(%esp) movl %ecx, 16(%esp) leal 16(%esp), %ecx movl %ecx, (%eax) ; shadowstack loop overhead (none) ; shadowstack epilogue movl 48(%esp), %edx movl %edx, (%ecx) ; shadowstack metadata .align 3 ___gc_fun: # __gc_fun .long 8 .space 4 In comparison to LowerGC: ; lowergc prologue movl L_llvm_gc_root_chain$non_lazy_ptr, %eax movl (%eax), %ecx movl %ecx, 48(%esp) movl $8, 52(%esp) movl $0, 60(%esp) movl $0, 56(%esp) movl $0, 68(%esp) movl $0, 64(%esp) movl $0, 76(%esp) movl $0, 72(%esp) movl $0, 84(%esp) movl $0, 80(%esp) movl $0, 92(%esp) movl $0, 88(%esp) movl $0, 100(%esp) movl $0, 96(%esp) movl $0, 108(%esp) movl $0, 104(%esp) movl $0, 116(%esp) movl $0, 112(%esp) ; lowergc loop overhead leal 44(%esp), %eax movl %eax, 56(%esp) leal 40(%esp), %eax movl %eax, 64(%esp) leal 36(%esp), %eax movl %eax, 72(%esp) leal 32(%esp), %eax movl %eax, 80(%esp) leal 28(%esp), %eax movl %eax, 88(%esp) leal 24(%esp), %eax movl %eax, 96(%esp) leal 20(%esp), %eax movl %eax, 104(%esp) leal 16(%esp), %eax movl %eax, 112(%esp) ; lowergc epilogue movl 48(%esp), %edx movl %edx, (%ecx) ; lowergc metadata (none) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45670 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-02fix this to use a valid triple.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45509 91177308-0d34-0410-b5e6-96231b3b80d8
2008-01-02verify that aligned common support doesn't break.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45495 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-31Fix PR1833 - eh.exception and eh.selector return twoDuncan Sands
values, which means doing extra legalization work. It would be easier to get this kind of thing right if there was some documentation... git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45472 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-29One readme entry is done, one is really easy (Evan, want to investigateChris Lattner
eliminating the llvm.x86.sse2.loadl.pd intrinsic?), one shuffle optzn may be done (if shufps is better than pinsw, Evan, please review), and we already know about LICM of simple instructions. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45407 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-29upgrade this testChris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45406 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-29Fold comparisons against a constant nan, and optimize ORD/UNORD Chris Lattner
comparisons with a constant. This allows us to compile isnan to: _foo: fcmpu cr7, f1, f1 mfcr r2 rlwinm r3, r2, 0, 31, 31 blr instead of: LCPI1_0: ; float .space 4 _foo: lis r2, ha16(LCPI1_0) lfs f0, lo16(LCPI1_0)(r2) fcmpu cr7, f1, f0 mfcr r2 rlwinm r3, r2, 0, 31, 31 blr git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45405 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-29this xform is implemented.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45404 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-29Codegen:Chris Lattner
as: _bar: pushl %esi subl $8, %esp movl 16(%esp), %esi call L_foo$stub fstps (%esi) addl $8, %esp popl %esi #FP_REG_KILL ret instead of: _bar: pushl %esi subl $8, %esp movl 16(%esp), %esi call L_foo$stub fstpl (%esi) cvtsd2ss (%esi), %xmm0 movss %xmm0, (%esi) addl $8, %esp popl %esi #FP_REG_KILL ret git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45401 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-29avoid going through a stack slot to convert from fpstack to xmm regChris Lattner
if we are just going to store it back anyway. This improves things like: double foo(); void bar(double *P) { *P = foo(); } git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45399 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-26one fewer uncond branch with my codegenprepare hack for single-mbb backedges.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45360 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-26Tests for changes made in r45356, where IPO optimizations would dropGordon Henriksen
collector algorithms. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45357 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-25GC poses hazards to the inliner. Consider:Gordon Henriksen
define void @f() { ... call i32 @g() ... } define void @g() { ... } The hazards are: - @f and @g have GC, but they differ GC. Inlining is invalid. This may never occur. - @f has no GC, but @g does. g's GC must be propagated to @f. The other scenarios are safe: - @f and @g have the same GC. - @f and @g have no GC. - @g has no GC. This patch adds inliner checks for the former two scenarios. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45351 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-25Noting and enforcing that GC intrinsics are valid only within aGordon Henriksen
function with GC. This will catch the error when the inliner inlines a function with GC into a caller with no GC. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45350 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-25Adjusting verification of "llvm.gc*" intrinsic prototypes to matchGordon Henriksen
LangRef. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45349 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-20Remove xfail. This is fixed.Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45254 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-20More working CellSPU tests:Scott Michel
- vec_const.ll: Vector constant loads - immed64.ll: i64, f64 constant loads git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45242 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-19CellSPU testcase, extract_elt.ll: extract vector element.Scott Michel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45219 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-19More working CellSPU test cases:Scott Michel
- call.ll: Function call - ctpop.ll: Count population - dp_farith.ll: DP arithmetic - eqv.ll: Equivalence primitives - fcmp.ll: SP comparisons - fdiv.ll: SP division - fneg-fabs.ll: SP negation, aboslute value - int2fp.ll: Integer -> SP conversion - rotate_ops.ll: Rotation primitives - select_bits.ll: (a & c) | (b & ~c) bit selection - shift_ops.ll: Shift primitives - sp_farith.ll: SP arithmentic git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45217 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-19Two more test cases: or_ops.ll (arithmetic or operations) and vecinsert.llScott Michel
(vector insertions) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45216 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-19Add new immed16.ll test case, fix CellSPU errata to make test case work.Scott Michel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45196 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-18Fix PR1872: SrcValue and SrcValueOffset should not be used to compute load / ↵Evan Cheng
store node id. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45167 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-18FIX for PR1799: When a load is unfolded from an instruction, check if it is ↵Evan Cheng
a new node. If not, do not create a new SUnit. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45157 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-17i32 immediate constant test case for CellSPUScott Michel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45134 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-17- Restore some i8 functionality in CellSPUScott Michel
- New test case: nand.ll git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45130 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-17Make invokes of inline asm legal. Teach codegenDuncan Sands
how to lower them (with no attempt made to be efficient, since they should only occur for unoptimized code). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45108 91177308-0d34-0410-b5e6-96231b3b80d8
2007-12-15Make better use of instructions that clear high bits; fix various 2-wide ↵Evan Cheng
shuffle bugs. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@45058 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-12-25 19:20:18 +0000