RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-23 19:23:23 +01:00
Code Issues Projects Releases Wiki Activity
c6275bc0ce
llvm-mirror/test/CodeGen/PowerPC/vsx-spill.ll

108 lines
3.9 KiB
LLVM
Raw Normal View History

[PowerPC] Change Test Options [NFC] Patch by amyk llvm-svn: 340639
2018-08-24 21:24:20 +02:00
; RUN: llc -verify-machineinstrs -mcpu=pwr7 -mattr=+vsx \
; RUN: -ppc-vsr-nums-as-vr -ppc-asm-full-reg-names < %s | FileCheck %s
; RUN: llc -verify-machineinstrs -mcpu=pwr7 -mattr=+vsx \
; RUN: -ppc-vsr-nums-as-vr -ppc-asm-full-reg-names < %s | FileCheck \
[Power9] Exploit D-Form VSX Scalar memory ops that target full VSX register set This patch corresponds to review: The newly added VSX D-Form (register + offset) memory ops target the upper half of the VSX register set. The existing ones target the lower half. In order to unify these and have the ability to target all the VSX registers using D-Form operations, this patch defines Pseudo-ops for the loads/stores which are expanded post-RA. The expansion then choses the correct opcode based on the register that was allocated for the operation. llvm-svn: 283212
2016-10-04 13:25:52 +02:00
; RUN: -check-prefix=CHECK-REG %s
[PowerPC] Change Test Options [NFC] Patch by amyk llvm-svn: 340639
2018-08-24 21:24:20 +02:00
; RUN: llc -verify-machineinstrs -mcpu=pwr7 -mattr=+vsx -fast-isel -O0 \
; RUN: -ppc-vsr-nums-as-vr -ppc-asm-full-reg-names < %s | FileCheck %s
; RUN: llc -verify-machineinstrs -mcpu=pwr7 -mattr=+vsx -fast-isel -O0 \
; RUN: -ppc-vsr-nums-as-vr -ppc-asm-full-reg-names < %s | \
[Power9] Exploit D-Form VSX Scalar memory ops that target full VSX register set This patch corresponds to review: The newly added VSX D-Form (register + offset) memory ops target the upper half of the VSX register set. The existing ones target the lower half. In order to unify these and have the ability to target all the VSX registers using D-Form operations, this patch defines Pseudo-ops for the loads/stores which are expanded post-RA. The expansion then choses the correct opcode based on the register that was allocated for the operation. llvm-svn: 283212
2016-10-04 13:25:52 +02:00
; RUN: FileCheck -check-prefix=CHECK-FISL %s
[PowerPC] Change Test Options [NFC] Patch by amyk llvm-svn: 340639
2018-08-24 21:24:20 +02:00
; RUN: llc -verify-machineinstrs -mcpu=pwr9 -ppc-vsr-nums-as-vr \
; RUN: -ppc-asm-full-reg-names < %s | FileCheck -check-prefix=CHECK-P9-REG %s
; RUN: llc -verify-machineinstrs -mcpu=pwr9 -fast-isel -O0 \
; RUN: -ppc-vsr-nums-as-vr -ppc-asm-full-reg-names < %s | FileCheck \
[Power9] Exploit D-Form VSX Scalar memory ops that target full VSX register set This patch corresponds to review: The newly added VSX D-Form (register + offset) memory ops target the upper half of the VSX register set. The existing ones target the lower half. In order to unify these and have the ability to target all the VSX registers using D-Form operations, this patch defines Pseudo-ops for the loads/stores which are expanded post-RA. The expansion then choses the correct opcode based on the register that was allocated for the operation. llvm-svn: 283212
2016-10-04 13:25:52 +02:00
; RUN: -check-prefix=CHECK-P9-FISL %s
[PowerPC] Add subregister classes for f64 VSX values We had stored both f64 values and v2f64, etc. values in the VSX registers. This worked, but was suboptimal because we would always spill 16-byte values even through we almost always had scalar 8-byte values. This resulted in an increase in stack-size use, extra memory bandwidth, etc. To fix this, I've added 64-bit subregisters of the Altivec registers, and combined those with the existing scalar floating-point registers to form a class of VSX scalar floating-point registers. The ABI code has also been enhanced to use this register class and some other necessary improvements have been made. llvm-svn: 205075
2014-03-29 06:29:01 +01:00
target datalayout = "E-m:e-i64:64-n32:64"
target triple = "powerpc64-unknown-linux-gnu"
define double @foo1(double %a) nounwind {
entry:
call void asm sideeffect "", "~{f0},~{f1},~{f2},~{f3},~{f4},~{f5},~{f6},~{f7},~{f8},~{f9},~{f10},~{f11},~{f12},~{f13},~{f14},~{f15},~{f16},~{f17},~{f18},~{f19},~{f20},~{f21},~{f22},~{f23},~{f24},~{f25},~{f26},~{f27},~{f28},~{f29},~{f30},~{f31}"() nounwind
br label %return
[PowerPC]Update Power VSX test cases to also test fast-isel Update of some of the VSX test cases for Power to check fast-isel codegen as well as the regular codegen. http://reviews.llvm.org/D6357 llvm-svn: 223509
2014-12-05 21:32:05 +01:00
; CHECK-REG: @foo1
[PowerPC] Change Test Options [NFC] Patch by amyk llvm-svn: 340639
2018-08-24 21:24:20 +02:00
; CHECK-REG: xxlor v2, f1, f1
; CHECK-REG: xxlor f1, v2, v2
[PowerPC]Update Power VSX test cases to also test fast-isel Update of some of the VSX test cases for Power to check fast-isel codegen as well as the regular codegen. http://reviews.llvm.org/D6357 llvm-svn: 223509
2014-12-05 21:32:05 +01:00
; CHECK-REG: blr
; CHECK-FISL: @foo1
[PowerPC] Utilize DQ-Form instructions for spill/restore and fix FrameIndex elimination to only use `lis/addi` if necessary. Currently we produce a bunch of unnecessary code when emitting the prologue/epilogue for spills/restores. Namely, if the load from stack slot/store to stack slot instruction is an X-Form instruction, we will always produce an LIS/ORI sequence for the stack offset. Furthermore, we have not exploited the P9 vector D-Form loads/stores for this purpose. This patch address both issues. Specifying the D-Form load as the instruction to use for stack spills/reloads should be safe because: 1. The stack should be aligned according to the ABI 2. If the stack isn't aligned, PPCRegisterInfo::eliminateFrameIndex() will check for the offset being a multiple of 16 and will convert it to an X-Form instruction if it isn't. Differential Revision : https://reviews.llvm.org/D38758 llvm-svn: 315500
2017-10-11 22:20:58 +02:00
; CHECK-FISL-NOT: lis
; CHECK-FISL-NOT: ori
[PowerPC] Change Test Options [NFC] Patch by amyk llvm-svn: 340639
2018-08-24 21:24:20 +02:00
; CHECK-FISL: li r3, -152
[PowerPC] Utilize DQ-Form instructions for spill/restore and fix FrameIndex elimination to only use `lis/addi` if necessary. Currently we produce a bunch of unnecessary code when emitting the prologue/epilogue for spills/restores. Namely, if the load from stack slot/store to stack slot instruction is an X-Form instruction, we will always produce an LIS/ORI sequence for the stack offset. Furthermore, we have not exploited the P9 vector D-Form loads/stores for this purpose. This patch address both issues. Specifying the D-Form load as the instruction to use for stack spills/reloads should be safe because: 1. The stack should be aligned according to the ABI 2. If the stack isn't aligned, PPCRegisterInfo::eliminateFrameIndex() will check for the offset being a multiple of 16 and will convert it to an X-Form instruction if it isn't. Differential Revision : https://reviews.llvm.org/D38758 llvm-svn: 315500
2017-10-11 22:20:58 +02:00
; CHECK-FISL-NOT: lis
; CHECK-FISL-NOT: ori
[PowerPC] Change Test Options [NFC] Patch by amyk llvm-svn: 340639
2018-08-24 21:24:20 +02:00
; CHECK-FISL: stxsdx f1, r1, r3
[PowerPC]Update Power VSX test cases to also test fast-isel Update of some of the VSX test cases for Power to check fast-isel codegen as well as the regular codegen. http://reviews.llvm.org/D6357 llvm-svn: 223509
2014-12-05 21:32:05 +01:00
; CHECK-FISL: blr
[PowerPC] Add subregister classes for f64 VSX values We had stored both f64 values and v2f64, etc. values in the VSX registers. This worked, but was suboptimal because we would always spill 16-byte values even through we almost always had scalar 8-byte values. This resulted in an increase in stack-size use, extra memory bandwidth, etc. To fix this, I've added 64-bit subregisters of the Altivec registers, and combined those with the existing scalar floating-point registers to form a class of VSX scalar floating-point registers. The ABI code has also been enhanced to use this register class and some other necessary improvements have been made. llvm-svn: 205075
2014-03-29 06:29:01 +01:00
[Power9] Exploit D-Form VSX Scalar memory ops that target full VSX register set This patch corresponds to review: The newly added VSX D-Form (register + offset) memory ops target the upper half of the VSX register set. The existing ones target the lower half. In order to unify these and have the ability to target all the VSX registers using D-Form operations, this patch defines Pseudo-ops for the loads/stores which are expanded post-RA. The expansion then choses the correct opcode based on the register that was allocated for the operation. llvm-svn: 283212
2016-10-04 13:25:52 +02:00
; CHECK-P9-REG: @foo1
[PowerPC] Emit xscpsgndp instead of xxlor when copying floating point scalar registers for P9 This patch will address using the xscpsgndp instruction to copy floating point scalar registers instead of the xxlor (specifically XXLORf) instruction that is currently used. Additionally, this patch of utilizing xscpsgndp will apply to P9, while pre-P9 will still use xxlor. Patch by amyk Differential Revision: https://reviews.llvm.org/D50004 llvm-svn: 340643
2018-08-24 22:00:24 +02:00
; CHECK-P9-REG: xscpsgndp v2, f1, f1
; CHECK-P9-REG: xscpsgndp f1, v2, v2
[Power9] Exploit D-Form VSX Scalar memory ops that target full VSX register set This patch corresponds to review: The newly added VSX D-Form (register + offset) memory ops target the upper half of the VSX register set. The existing ones target the lower half. In order to unify these and have the ability to target all the VSX registers using D-Form operations, this patch defines Pseudo-ops for the loads/stores which are expanded post-RA. The expansion then choses the correct opcode based on the register that was allocated for the operation. llvm-svn: 283212
2016-10-04 13:25:52 +02:00
; CHECK-P9-REG: blr
; CHECK-P9-FISL: @foo1
[PowerPC] Change Test Options [NFC] Patch by amyk llvm-svn: 340639
2018-08-24 21:24:20 +02:00
; CHECK-P9-FISL: stfd f31, -8(r1)
[Power9] Exploit D-Form VSX Scalar memory ops that target full VSX register set This patch corresponds to review: The newly added VSX D-Form (register + offset) memory ops target the upper half of the VSX register set. The existing ones target the lower half. In order to unify these and have the ability to target all the VSX registers using D-Form operations, this patch defines Pseudo-ops for the loads/stores which are expanded post-RA. The expansion then choses the correct opcode based on the register that was allocated for the operation. llvm-svn: 283212
2016-10-04 13:25:52 +02:00
; CHECK-P9-FISL: blr
[PowerPC] Add subregister classes for f64 VSX values We had stored both f64 values and v2f64, etc. values in the VSX registers. This worked, but was suboptimal because we would always spill 16-byte values even through we almost always had scalar 8-byte values. This resulted in an increase in stack-size use, extra memory bandwidth, etc. To fix this, I've added 64-bit subregisters of the Altivec registers, and combined those with the existing scalar floating-point registers to form a class of VSX scalar floating-point registers. The ABI code has also been enhanced to use this register class and some other necessary improvements have been made. llvm-svn: 205075
2014-03-29 06:29:01 +01:00
return: ; preds = %entry
ret double %a
}
define double @foo2(double %a) nounwind {
entry:
%b = fadd double %a, %a
call void asm sideeffect "", "~{f0},~{f1},~{f2},~{f3},~{f4},~{f5},~{f6},~{f7},~{f8},~{f9},~{f10},~{f11},~{f12},~{f13},~{f14},~{f15},~{f16},~{f17},~{f18},~{f19},~{f20},~{f21},~{f22},~{f23},~{f24},~{f25},~{f26},~{f27},~{f28},~{f29},~{f30},~{f31}"() nounwind
br label %return
[PowerPC]Update Power VSX test cases to also test fast-isel Update of some of the VSX test cases for Power to check fast-isel codegen as well as the regular codegen. http://reviews.llvm.org/D6357 llvm-svn: 223509
2014-12-05 21:32:05 +01:00
; CHECK-REG: @foo2
[PowerPC] Change Test Options [NFC] Patch by amyk llvm-svn: 340639
2018-08-24 21:24:20 +02:00
; CHECK-REG: {{xxlor|xsadddp}} v2, f1, f1
; CHECK-REG: {{xxlor|xsadddp}} f1, f0, f0
[PowerPC]Update Power VSX test cases to also test fast-isel Update of some of the VSX test cases for Power to check fast-isel codegen as well as the regular codegen. http://reviews.llvm.org/D6357 llvm-svn: 223509
2014-12-05 21:32:05 +01:00
; CHECK-REG: blr
; CHECK-FISL: @foo2
Relax fast register allocator related test cases; NFC - Relex hard coded registers and stack frame sizes - Some test cleanups - Change phi-dbg.ll to match on mir output after phi elimination instead of going through the whole codegen pipeline. This is in preparation for https://reviews.llvm.org/D52010 I'm committing all the test changes upfront that work before and after independently. llvm-svn: 345532
2018-10-29 21:10:42 +01:00
; CHECK-FISL: xsadddp [[REG0:f[0-9]+]], f1, f1
; CHECK-FISL: stxsdx [[REG0]], r1, r3
[PowerPC] Change Test Options [NFC] Patch by amyk llvm-svn: 340639
2018-08-24 21:24:20 +02:00
; CHECK-FISL: lxsdx f1, r1, r3
[PowerPC]Update Power VSX test cases to also test fast-isel Update of some of the VSX test cases for Power to check fast-isel codegen as well as the regular codegen. http://reviews.llvm.org/D6357 llvm-svn: 223509
2014-12-05 21:32:05 +01:00
; CHECK-FISL: blr
[PowerPC] Add subregister classes for f64 VSX values We had stored both f64 values and v2f64, etc. values in the VSX registers. This worked, but was suboptimal because we would always spill 16-byte values even through we almost always had scalar 8-byte values. This resulted in an increase in stack-size use, extra memory bandwidth, etc. To fix this, I've added 64-bit subregisters of the Altivec registers, and combined those with the existing scalar floating-point registers to form a class of VSX scalar floating-point registers. The ABI code has also been enhanced to use this register class and some other necessary improvements have been made. llvm-svn: 205075
2014-03-29 06:29:01 +01:00
[Power9] Exploit D-Form VSX Scalar memory ops that target full VSX register set This patch corresponds to review: The newly added VSX D-Form (register + offset) memory ops target the upper half of the VSX register set. The existing ones target the lower half. In order to unify these and have the ability to target all the VSX registers using D-Form operations, this patch defines Pseudo-ops for the loads/stores which are expanded post-RA. The expansion then choses the correct opcode based on the register that was allocated for the operation. llvm-svn: 283212
2016-10-04 13:25:52 +02:00
; CHECK-P9-REG: @foo2
[PowerPC] Emit xscpsgndp instead of xxlor when copying floating point scalar registers for P9 This patch will address using the xscpsgndp instruction to copy floating point scalar registers instead of the xxlor (specifically XXLORf) instruction that is currently used. Additionally, this patch of utilizing xscpsgndp will apply to P9, while pre-P9 will still use xxlor. Patch by amyk Differential Revision: https://reviews.llvm.org/D50004 llvm-svn: 340643
2018-08-24 22:00:24 +02:00
; CHECK-P9-REG: {{xscpsgndp|xsadddp}} v2, f1, f1
; CHECK-P9-REG: {{xscpsgndp|xsadddp}} f1, v2, v2
[Power9] Exploit D-Form VSX Scalar memory ops that target full VSX register set This patch corresponds to review: The newly added VSX D-Form (register + offset) memory ops target the upper half of the VSX register set. The existing ones target the lower half. In order to unify these and have the ability to target all the VSX registers using D-Form operations, this patch defines Pseudo-ops for the loads/stores which are expanded post-RA. The expansion then choses the correct opcode based on the register that was allocated for the operation. llvm-svn: 283212
2016-10-04 13:25:52 +02:00
; CHECK-P9-REG: blr
; CHECK-P9-FISL: @foo2
Relax fast register allocator related test cases; NFC - Relex hard coded registers and stack frame sizes - Some test cleanups - Change phi-dbg.ll to match on mir output after phi elimination instead of going through the whole codegen pipeline. This is in preparation for https://reviews.llvm.org/D52010 I'm committing all the test changes upfront that work before and after independently. llvm-svn: 345532
2018-10-29 21:10:42 +01:00
; CHECK-P9-FISL: xsadddp [[REG0:f[0-9]+]], f1, f1
; CHECK-P9-FISL: stfd [[REG0]], -152(r1)
[PowerPC] Change Test Options [NFC] Patch by amyk llvm-svn: 340639
2018-08-24 21:24:20 +02:00
; CHECK-P9-FISL: lfd f1, -152(r1)
[Power9] Exploit D-Form VSX Scalar memory ops that target full VSX register set This patch corresponds to review: The newly added VSX D-Form (register + offset) memory ops target the upper half of the VSX register set. The existing ones target the lower half. In order to unify these and have the ability to target all the VSX registers using D-Form operations, this patch defines Pseudo-ops for the loads/stores which are expanded post-RA. The expansion then choses the correct opcode based on the register that was allocated for the operation. llvm-svn: 283212
2016-10-04 13:25:52 +02:00
; CHECK-P9-FISL: blr
[PowerPC] Add subregister classes for f64 VSX values We had stored both f64 values and v2f64, etc. values in the VSX registers. This worked, but was suboptimal because we would always spill 16-byte values even through we almost always had scalar 8-byte values. This resulted in an increase in stack-size use, extra memory bandwidth, etc. To fix this, I've added 64-bit subregisters of the Altivec registers, and combined those with the existing scalar floating-point registers to form a class of VSX scalar floating-point registers. The ABI code has also been enhanced to use this register class and some other necessary improvements have been made. llvm-svn: 205075
2014-03-29 06:29:01 +01:00
return: ; preds = %entry
ret double %b
}
define double @foo3(double %a) nounwind {
entry:
call void asm sideeffect "", "~{f0},~{f1},~{f2},~{f3},~{f4},~{f5},~{f6},~{f7},~{f8},~{f9},~{f10},~{f11},~{f12},~{f13},~{f14},~{f15},~{f16},~{f17},~{f18},~{f19},~{f20},~{f21},~{f22},~{f23},~{f24},~{f25},~{f26},~{f27},~{f28},~{f29},~{f30},~{f31},~{v0},~{v1},~{v2},~{v3},~{v4},~{v5},~{v6},~{v7},~{v8},~{v9},~{v10},~{v11},~{v12},~{v13},~{v14},~{v15},~{v16},~{v17},~{v18},~{v19},~{v20},~{v21},~{v22},~{v23},~{v24},~{v25},~{v26},~{v27},~{v28},~{v29},~{v30},~{v31}"() nounwind
br label %return
; CHECK: @foo3
[PowerPC] Change Test Options [NFC] Patch by amyk llvm-svn: 340639
2018-08-24 21:24:20 +02:00
; CHECK: stxsdx f1, r1, r3
; CHECK: lxsdx f0, r1, r3
; CHECK: xsadddp f1, f0, f0
[PowerPC] Add subregister classes for f64 VSX values We had stored both f64 values and v2f64, etc. values in the VSX registers. This worked, but was suboptimal because we would always spill 16-byte values even through we almost always had scalar 8-byte values. This resulted in an increase in stack-size use, extra memory bandwidth, etc. To fix this, I've added 64-bit subregisters of the Altivec registers, and combined those with the existing scalar floating-point registers to form a class of VSX scalar floating-point registers. The ABI code has also been enhanced to use this register class and some other necessary improvements have been made. llvm-svn: 205075
2014-03-29 06:29:01 +01:00
; CHECK: blr
[Power9] Exploit D-Form VSX Scalar memory ops that target full VSX register set This patch corresponds to review: The newly added VSX D-Form (register + offset) memory ops target the upper half of the VSX register set. The existing ones target the lower half. In order to unify these and have the ability to target all the VSX registers using D-Form operations, this patch defines Pseudo-ops for the loads/stores which are expanded post-RA. The expansion then choses the correct opcode based on the register that was allocated for the operation. llvm-svn: 283212
2016-10-04 13:25:52 +02:00
; CHECK-P9-REG-LABEL: foo3
[PowerPC] Change Test Options [NFC] Patch by amyk llvm-svn: 340639
2018-08-24 21:24:20 +02:00
; CHECK-P9-REG: stdu r1, -400(r1)
[Power9] Enable the Out-of-Order scheduling model for P9 hw When switched to the MI scheduler for P9, the hardware is modeled as out of order. However, inside the MI Scheduler algorithm, we still use the in-order scheduling model as the MicroOpBufferSize isn't set. The MI scheduler take it as the hw cannot buffer the op. So, only when all the available instructions issued, the pending instruction could be scheduled. That is not true for our P9 hw in fact. This patch is trying to enable the Out-of-Order scheduling model. The buffer size 44 is picked from the P9 hw spec, and the perf test indicate that, its value won't hurt the cpu2017. With this patch, there are 3 specs improved over 3% and 1 spec deg over 3%. The detail is as follows: x264_r: +6.95% cactuBSSN_r: +6.94% lbm_r: +4.11% xz_r: -3.85% And the GEOMEAN for all the C/C++ spec in spec2017 is about 0.18% improved. Reviewer: Nemanjai Differential Revision: https://reviews.llvm.org/D55810 llvm-svn: 350285
2019-01-03 06:04:18 +01:00
; CHECK-P9-REG-DAG: lfd f30, 384(r1)
; CHECK-P9-REG-DAG: xsadddp f1, f0, f0
[Power9] Exploit D-Form VSX Scalar memory ops that target full VSX register set This patch corresponds to review: The newly added VSX D-Form (register + offset) memory ops target the upper half of the VSX register set. The existing ones target the lower half. In order to unify these and have the ability to target all the VSX registers using D-Form operations, this patch defines Pseudo-ops for the loads/stores which are expanded post-RA. The expansion then choses the correct opcode based on the register that was allocated for the operation. llvm-svn: 283212
2016-10-04 13:25:52 +02:00
; CHECK-P9-FISL-LABEL: foo3
[PowerPC] Change Test Options [NFC] Patch by amyk llvm-svn: 340639
2018-08-24 21:24:20 +02:00
; CHECK-P9-FISL: stdu r1, -400(r1)
; CHECK-P9-FISL: lfd f0, 56(r1)
; CHECK-P9-FISL: xsadddp f1, f0, f0
[PowerPC] Add subregister classes for f64 VSX values We had stored both f64 values and v2f64, etc. values in the VSX registers. This worked, but was suboptimal because we would always spill 16-byte values even through we almost always had scalar 8-byte values. This resulted in an increase in stack-size use, extra memory bandwidth, etc. To fix this, I've added 64-bit subregisters of the Altivec registers, and combined those with the existing scalar floating-point registers to form a class of VSX scalar floating-point registers. The ABI code has also been enhanced to use this register class and some other necessary improvements have been made. llvm-svn: 205075
2014-03-29 06:29:01 +01:00
return: ; preds = %entry
%b = fadd double %a, %a
ret double %b
}
Reference in New Issue Copy Permalink