aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2005-08-16Fix Transforms/LoopStrengthReduce/2005-08-15-AddRecIV.llChris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22797 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-16testcase that crashes lsr, distilled from 175.vprChris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22796 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-15Turn loop strength reduction on by default.Chris Lattner
Only run createLowerConstantExpressionsPass for the simple isel. The DAG isel has no need for it. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22794 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-15Teach LLVM to know how many times a loop executes when constructed withChris Lattner
a < expression, e.g.: for (i = m; i < n; ++i) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22793 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-15Broke 80 column rule.Jim Laskey
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22792 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-15Changed code gen for int to f32 to use rounding. This makes FP resultsJim Laskey
consistent with gcc. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22791 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-15isIntImmediate is a good Idea. Add a flavor that checks bounds while it is ↵Andrew Lenharth
at it git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22790 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-14Fix last night's PPC32 regressions byNate Begeman
1. Not selecting the false value of a select_cc in the false arm, which isn't legal for nested selects. 2. Actually returning the node we created and Legalized in the FP_TO_UINT Expander. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22789 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-14Fix last night's X86 regressions by putting code for SSE in the if(SSE)Nate Begeman
block. nur. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22788 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-14only build .a on alphaAndrew Lenharth
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22787 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-14Fix FP_TO_UINT with Scalar SSE2 now that the legalizer can handle it. WeNate Begeman
now generate the relatively good code sequences: unsigned short foo(float a) { return a; } _foo: movss 4(%esp), %xmm0 cvttss2si %xmm0, %eax movzwl %ax, %eax ret and unsigned bar(float a) { return a; } _bar: movss .CPI_bar_0, %xmm0 movss 4(%esp), %xmm1 movapd %xmm1, %xmm2 subss %xmm0, %xmm2 cvttss2si %xmm2, %eax xorl $-2147483648, %eax cvttss2si %xmm1, %ecx ucomiss %xmm0, %xmm1 cmovb %ecx, %eax ret git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22786 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-14Teach the legalizer how to legalize FP_TO_UINT.Nate Begeman
Teach the legalizer to promote FP_TO_UINT to FP_TO_SINT if the wider FP_TO_UINT is also illegal. This allows us on PPC to codegen unsigned short foo(float a) { return a; } as: _foo: .LBB_foo_0: ; entry fctiwz f0, f1 stfd f0, -8(r1) lwz r2, -4(r1) rlwinm r3, r2, 0, 16, 31 blr instead of: _foo: .LBB_foo_0: ; entry fctiwz f0, f1 stfd f0, -8(r1) lwz r2, -4(r1) lis r3, ha16(.CPI_foo_0) lfs f0, lo16(.CPI_foo_0)(r3) fcmpu cr0, f1, f0 blt .LBB_foo_2 ; entry .LBB_foo_1: ; entry fsubs f0, f1, f0 fctiwz f0, f0 stfd f0, -16(r1) lwz r2, -12(r1) xoris r2, r2, 32768 .LBB_foo_2: ; entry rlwinm r3, r2, 0, 16, 31 blr git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22785 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-14Make FP_TO_UINT Illegal. This allows us to generate significantly betterNate Begeman
codegen for FP_TO_UINT by using the legalizer's SELECT variant. Implement a codegen improvement for SELECT_CC, selecting the false node in the MBB that feeds the phi node. This allows us to codegen: void foo(int *a, int b, int c) { int d = (a < b) ? 5 : 9; *a = d; } as: _foo: li r2, 5 cmpw cr0, r4, r3 bgt .LBB_foo_2 ; entry .LBB_foo_1: ; entry li r2, 9 .LBB_foo_2: ; entry stw r2, 0(r3) blr insted of: _foo: li r2, 5 li r5, 9 cmpw cr0, r4, r3 bgt .LBB_foo_2 ; entry .LBB_foo_1: ; entry or r2, r5, r5 .LBB_foo_2: ; entry stw r2, 0(r3) blr git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22784 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-13Testing a variable before it is defined doesn't work so well. It is a ↵Andrew Lenharth
fairly small thing, so just let everyone build the .a file git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22783 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-13Ooops, don't forget to clear this. The real inner loop is now:Chris Lattner
.LBB_foo_3: ; no_exit.1 lfd f2, 0(r9) lfd f3, 8(r9) fmul f4, f1, f2 fmadd f4, f0, f3, f4 stfd f4, 8(r9) fmul f3, f1, f3 fmsub f2, f0, f2, f3 stfd f2, 0(r9) addi r9, r9, 16 addi r8, r8, 1 cmpw cr0, r8, r4 ble .LBB_foo_3 ; no_exit.1 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22782 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-13Recursively scan scev expressions for common subexpressions. This allows usChris Lattner
to handle nested loops much better, for example, by being able to tell that these two expressions: {( 8 + ( 16 * ( 1 + %Tmp11 + %Tmp12)) + %c_),+,( 16 * %Tmp 12)}<loopentry.1> {(( 16 * ( 1 + %Tmp11 + %Tmp12)) + %c_),+,( 16 * %Tmp12)}<loopentry.1> Have the following common part that can be shared: {(( 16 * ( 1 + %Tmp11 + %Tmp12)) + %c_),+,( 16 * %Tmp12)}<loopentry.1> This allows us to codegen an important inner loop in 168.wupwise as: .LBB_foo_4: ; no_exit.1 lfd f2, 16(r9) fmul f3, f0, f2 fmul f2, f1, f2 fadd f4, f3, f2 stfd f4, 8(r9) fsub f2, f3, f2 stfd f2, 16(r9) addi r8, r8, 1 addi r9, r9, 16 cmpw cr0, r8, r4 ble .LBB_foo_4 ; no_exit.1 instead of: .LBB_foo_3: ; no_exit.1 lfdx f2, r6, r9 add r10, r6, r9 lfd f3, 8(r10) fmul f4, f1, f2 fmadd f4, f0, f3, f4 stfd f4, 8(r10) fmul f3, f1, f3 fmsub f2, f0, f2, f3 stfdx f2, r6, r9 addi r9, r9, 16 addi r8, r8, 1 cmpw cr0, r8, r4 ble .LBB_foo_3 ; no_exit.1 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22781 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-13Remove an unncessary argument to SimplifySelectCC and add an additionalNate Begeman
assert when creating a select_cc node. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22780 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-13Fix the fabs regression on x86 by abstracting the select_cc optimizationNate Begeman
out into SimplifySelectCC. This allows both ISD::SELECT and ISD::SELECT_CC to use the same set of simplifying folds. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22779 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-13Remove support for 64b PPC, it's been broken for a long time. It'll beNate Begeman
back once a DAG->DAG ISel exists. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22778 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-13Fix oversized GOT problem with gcc-4 on alphaAndrew Lenharth
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22777 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-13Teach SplitCriticalEdge to update LoopInfo if it is alive. This fixesChris Lattner
a problem in LoopStrengthReduction, where it would split critical edges then confused itself with outdated loop information. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22776 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-13remove dead code. The exit block list is computed on demand, thus does notChris Lattner
need to be updated. This code is a relic from when it did. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22775 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-12implement a couple of simple shift foldings.Chris Lattner
e.g. (X & 7) >> 3 -> 0 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22774 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-12Fix for 2005-08-12-rlwimi-crash.ll. Make allowance for masks being shifted toJim Laskey
zero. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22773 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-12Added test cases to guarantee use of ORC and ANDC.Jim Laskey
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22772 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-121. This changes handles the cases of (~x)&y and x&(~y) yielding ANDC, andJim Laskey
(~x)|y and x|(~y) yielding ORC. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22771 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-12testcase that crashed the ppc backend, distilled from craftyChris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22770 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-12When splitting critical edges, make sure not to leave the new block in theChris Lattner
middle of the loop. This turns a critical loop in gzip into this: .LBB_test_1: ; loopentry or r27, r28, r28 add r28, r3, r27 lhz r28, 3(r28) add r26, r4, r27 lhz r26, 3(r26) cmpw cr0, r28, r26 bne .LBB_test_8 ; loopentry.loopexit_crit_edge .LBB_test_2: ; shortcirc_next.0 add r28, r3, r27 lhz r28, 5(r28) add r26, r4, r27 lhz r26, 5(r26) cmpw cr0, r28, r26 bne .LBB_test_7 ; shortcirc_next.0.loopexit_crit_edge .LBB_test_3: ; shortcirc_next.1 add r28, r3, r27 lhz r28, 7(r28) add r26, r4, r27 lhz r26, 7(r26) cmpw cr0, r28, r26 bne .LBB_test_6 ; shortcirc_next.1.loopexit_crit_edge .LBB_test_4: ; shortcirc_next.2 add r28, r3, r27 lhz r26, 9(r28) add r28, r4, r27 lhz r25, 9(r28) addi r28, r27, 8 cmpw cr7, r26, r25 mfcr r26, 1 rlwinm r26, r26, 31, 31, 31 add r25, r8, r27 cmpw cr7, r25, r7 mfcr r25, 1 rlwinm r25, r25, 29, 31, 31 and. r26, r26, r25 bne .LBB_test_1 ; loopentry instead of this: .LBB_test_1: ; loopentry or r27, r28, r28 add r28, r3, r27 lhz r28, 3(r28) add r26, r4, r27 lhz r26, 3(r26) cmpw cr0, r28, r26 beq .LBB_test_3 ; shortcirc_next.0 .LBB_test_2: ; loopentry.loopexit_crit_edge add r2, r30, r27 add r8, r29, r27 b .LBB_test_9 ; loopexit .LBB_test_3: ; shortcirc_next.0 add r28, r3, r27 lhz r28, 5(r28) add r26, r4, r27 lhz r26, 5(r26) cmpw cr0, r28, r26 beq .LBB_test_5 ; shortcirc_next.1 .LBB_test_4: ; shortcirc_next.0.loopexit_crit_edge add r2, r11, r27 add r8, r12, r27 b .LBB_test_9 ; loopexit .LBB_test_5: ; shortcirc_next.1 add r28, r3, r27 lhz r28, 7(r28) add r26, r4, r27 lhz r26, 7(r26) cmpw cr0, r28, r26 beq .LBB_test_7 ; shortcirc_next.2 .LBB_test_6: ; shortcirc_next.1.loopexit_crit_edge add r2, r9, r27 add r8, r10, r27 b .LBB_test_9 ; loopexit .LBB_test_7: ; shortcirc_next.2 add r28, r3, r27 lhz r26, 9(r28) add r28, r4, r27 lhz r25, 9(r28) addi r28, r27, 8 cmpw cr7, r26, r25 mfcr r26, 1 rlwinm r26, r26, 31, 31, 31 add r25, r8, r27 cmpw cr7, r25, r7 mfcr r25, 1 rlwinm r25, r25, 29, 31, 31 and. r26, r26, r25 bne .LBB_test_1 ; loopentry Next up, improve the code for the loop. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22769 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-12Add a helper methodChris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22768 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-12add a helper methodChris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22767 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-12Fix a FIXME: if we are inserting code for a PHI argument, split the criticalChris Lattner
edge so that the code is not always executed for both operands. This prevents LSR from inserting code into loops whose exit blocks contain PHI uses of IV expressions (which are outside of loops). On gzip, for example, we turn this ugly code: .LBB_test_1: ; loopentry add r27, r3, r28 lhz r27, 3(r27) add r26, r4, r28 lhz r26, 3(r26) add r25, r30, r28 ;; Only live if exiting the loop add r24, r29, r28 ;; Only live if exiting the loop cmpw cr0, r27, r26 bne .LBB_test_5 ; loopexit into this: .LBB_test_1: ; loopentry or r27, r28, r28 add r28, r3, r27 lhz r28, 3(r28) add r26, r4, r27 lhz r26, 3(r26) cmpw cr0, r28, r26 beq .LBB_test_3 ; shortcirc_next.0 .LBB_test_2: ; loopentry.loopexit_crit_edge add r2, r30, r27 add r8, r29, r27 b .LBB_test_9 ; loopexit .LBB_test_2: ; shortcirc_next.0 ... blt .LBB_test_1 into this: .LBB_test_1: ; loopentry or r27, r28, r28 add r28, r3, r27 lhz r28, 3(r28) add r26, r4, r27 lhz r26, 3(r26) cmpw cr0, r28, r26 beq .LBB_test_3 ; shortcirc_next.0 .LBB_test_2: ; loopentry.loopexit_crit_edge add r2, r30, r27 add r8, r29, r27 b .LBB_t_3: ; shortcirc_next.0 .LBB_test_3: ; shortcirc_next.0 ... blt .LBB_test_1 Next step: get the block out of the loop so that the loop is all fall-throughs again. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22766 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-12Change break critical edges to not remove, then insert, PHI node entries.Chris Lattner
Instead, just update the BB in-place. This is both faster, and it prevents split-critical-edges from shuffling the PHI argument list unneccesarily. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22765 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-12match gcc's use of tabs, makes diffs easierAndrew Lenharth
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22764 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-12.section cleanup, patch from Nicholas RileyAndrew Lenharth
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22763 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-11First rev of Xcode 2.1 projectChris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22762 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-111. Added the function isOpcWithIntImmediate to simplify testing of operand withJim Laskey
specified opcode and an integer constant right operand. 2. Modified ISD::SHL, ISD::SRL, ISD::SRA to use rlwinm when applied after a mask. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22761 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-11Tidied up the use of dyn_cast<ConstantSDNode> by using isIntImmediate more.Chris Lattner
Patch by Jim Laskey. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22760 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-11Use a more efficient method of creating integer and float virtual registersChris Lattner
(avoids an extra level of indirection in MakeReg). defined MakeIntReg using RegMap->createVirtualRegister(PPC32::GPRCRegisterClass) defined MakeFPReg using RegMap->createVirtualRegister(PPC32::FPRCRegisterClass) s/MakeReg(MVT::i32)/MakeIntReg/ s/MakeReg(MVT::f64)/MakeFPReg/ Patch by Jim Laskey! git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22759 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-11Add a select_cc optimization for recognizing abs(int). This speeds up anNate Begeman
integer MPEG encoding loop by a factor of two. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22758 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-11Some SELECT_CC cleanups:Nate Begeman
1. move assertions for node creation to getNode() 2. legalize the values returned in ExpandOp immediately 3. Move select_cc optimizations from SELECT's getNode() to SELECT_CC's, allowing them to be cleaned up significantly. This paves the way to pick up additional optimizations on SELECT_CC, such as sum-of-absolute-differences. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22757 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-10Make SELECT illegal on PPC32, switch to using SELECT_CC, which more closelyNate Begeman
reflects what the hardware is capable of. This significantly simplifies the CC handling logic throughout the ISel. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22756 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-10Add new node, SELECT_CC. This node is for targets that don't nativelyNate Begeman
implement SELECT. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22755 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-10Changes for PPC32ISelPattern.cppChris Lattner
1. Clean up how SelectIntImmediateExpr handles use counts. 2. "Subtract from" was not clearing hi 16 bits. Patch by Jim Laskey git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22754 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-10Fix an oversight that may be causing PR617.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22753 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-10now that we handle non-constant strides, this testcase passesChris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22752 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-10remove some trickiness that broke yacr2 and some other programs last nightChris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22751 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-10Changed the XOR case to use the isOprNot predicate.Chris Lattner
Patch by Jim Laskey! git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22750 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-101. Refactored handling of integer immediate values for add, or, xor and sub.Chris Lattner
New routine: ISel::SelectIntImmediateExpr 2. Now checking use counts of large constants. If use count is > 2 then drop thru so that the constant gets loaded into a register. Source: int %test1(int %a) { entry: %tmp.1 = add int %a, 123456789 ; <int> [#uses=1] %tmp.2 = or int %tmp.1, 123456789 ; <int> [#uses=1] %tmp.3 = xor int %tmp.2, 123456789 ; <int> [#uses=1] %tmp.4 = sub int %tmp.3, -123456789 ; <int> [#uses=1] ret int %tmp.4 } Did Emit: .machine ppc970 .text .align 2 .globl _test1 _test1: .LBB_test1_0: ; entry addi r2, r3, -13035 addis r2, r2, 1884 ori r2, r2, 52501 oris r2, r2, 1883 xori r2, r2, 52501 xoris r2, r2, 1883 addi r2, r2, 52501 addis r3, r2, 1883 blr Now Emits: .machine ppc970 .text .align 2 .globl _test1 _test1: .LBB_test1_0: ; entry lis r2, 1883 ori r2, r2, 52501 add r3, r3, r2 or r3, r3, r2 xor r3, r3, r2 add r3, r3, r2 blr Patch by Jim Laskey! git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22749 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-10sorry!! this is temporary; for some reason the nasty constmul code seems toDuraid Madina
be an infinite loop when using g++-4.0.1*, this kills the ia64 nightly tester. A proper fix shall be forthcoming!!! thanks for not killing me. :) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22748 91177308-0d34-0410-b5e6-96231b3b80d8
2005-08-10Fix a bug compiling: select (i32 < i32), f32, f32Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22747 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-09-28 08:53:00 +0000