Skip to content

[RISCV] Add +optimized-nfN-segment-load-store #114414

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Nov 3, 2024
Merged

[RISCV] Add +optimized-nfN-segment-load-store #114414

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
16f895c
[RISCV] Add +optimized-nf{3,4}-segment-load-store
lukel97 Oct 31, 2024
3984711
Add other NFs, including NF2. Add NF2 to the generic processor model
lukel97 Oct 31, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 7 additions & 0 deletions llvm/lib/Target/RISCV/RISCVFeatures.td
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1386,6 +1386,13 @@ def TuneOptimizedZeroStrideLoad
"true", "Optimized (perform fewer memory operations)"
"zero-stride vector load">;

foreach nf = {3-4} in
def TuneOptimizedNF#nf#SegmentLoadStore :
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

A random thought: There are 250 features for RISCV now, we may need to increase MAX_SUBTARGET_FEATURES soon.

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Isn't the limit 5*64 or 320?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes, but that won't last long, especially if we add some fusions (like XiangShan).

SubtargetFeature<"optimized-nf"#nf#"-segment-load-store",
"HasOptimizedNF"#nf#"SegmentLoadStore",
"true", "vlseg"#nf#"eN.v and vsseg"#nf#"eN.v are"
"implemented as a wide memory op and shuffle">;

def Experimental
: SubtargetFeature<"experimental", "HasExperimental",
"true", "Experimental intrinsics">;
Expand Down
4 changes: 3 additions & 1 deletion llvm/lib/Target/RISCV/RISCVProcessors.td
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -472,7 +472,9 @@ def SPACEMIT_X60 : RISCVProcessorModel<"spacemit-x60",
FeatureStdExtZvfh,
FeatureStdExtZvkt,
FeatureStdExtZvl256b]),
[TuneDLenFactor2]>;
[TuneDLenFactor2,
TuneOptimizedNF3SegmentLoadStore,
TuneOptimizedNF4SegmentLoadStore]>;

def RP2350_HAZARD3 : RISCVProcessorModel<"rp2350-hazard3",
NoSchedModel,
Expand Down
7 changes: 5 additions & 2 deletions llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -738,8 +738,11 @@ InstructionCost RISCVTTIImpl::getInterleavedMemoryOpCost(
AddressSpace, DL)) {

// Most available hardware today optimizes NF=2 as as one wide memory op
// + Factor * LMUL shuffle ops.
if (Factor == 2) {
// + Factor * LMUL shuffle ops. Some processors may also optimize NF=3
// and NF=4.
if (Factor == 2 ||
(Factor == 3 && ST->hasOptimizedNF3SegmentLoadStore()) ||
(Factor == 4 && ST->hasOptimizedNF4SegmentLoadStore())) {
InstructionCost Cost =
getMemoryOpCost(Opcode, VTy, Alignment, AddressSpace, CostKind);
MVT SubVecVT = getTLI()->getValueType(DL, SubVecTy).getSimpleVT();
Expand Down
70 changes: 47 additions & 23 deletions llvm/test/Transforms/LoopVectorize/RISCV/interleaved-cost.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,7 @@
; REQUIRES: asserts
; RUN: opt -passes=loop-vectorize -mtriple=riscv64 -mattr=+v -debug-only=loop-vectorize -disable-output < %s 2>&1 | FileCheck %s
; RUN: opt -passes=loop-vectorize -mtriple=riscv64 -mattr=+v -debug-only=loop-vectorize -disable-output < %s 2>&1 | FileCheck %s --check-prefixes=CHECK,NO-OPT
; RUN: opt -passes=loop-vectorize -mtriple=riscv64 -mattr=+v,+optimized-nf3-segment-load-store -debug-only=loop-vectorize -disable-output < %s 2>&1 | FileCheck %s --check-prefixes=CHECK,OPT-NF3
; RUN: opt -passes=loop-vectorize -mtriple=riscv64 -mattr=+v,+optimized-nf4-segment-load-store -debug-only=loop-vectorize -disable-output < %s 2>&1 | FileCheck %s --check-prefixes=CHECK,OPT-NF4

%i8.2 = type {i8, i8}
define void @i8_factor_2(ptr %data, i64 %n) {
Expand Down Expand Up @@ -48,17 +50,28 @@ for.end:
define void @i8_factor_3(ptr %data, i64 %n) {
entry:
br label %for.body
; CHECK-LABEL: Checking a loop in 'i8_factor_3'
; CHECK: Cost of 6 for VF 2: INTERLEAVE-GROUP with factor 3 at %l0, ir<%p0>
; CHECK: Cost of 6 for VF 2: INTERLEAVE-GROUP with factor 3 at <badref>, ir<%p0>
; CHECK: Cost of 12 for VF 4: INTERLEAVE-GROUP with factor 3 at %l0, ir<%p0>
; CHECK: Cost of 12 for VF 4: INTERLEAVE-GROUP with factor 3 at <badref>, ir<%p0>
; CHECK: Cost of 24 for VF 8: INTERLEAVE-GROUP with factor 3 at %l0, ir<%p0>
; CHECK: Cost of 24 for VF 8: INTERLEAVE-GROUP with factor 3 at <badref>, ir<%p0>
; CHECK: Cost of 48 for VF 16: INTERLEAVE-GROUP with factor 3 at %l0, ir<%p0>
; CHECK: Cost of 48 for VF 16: INTERLEAVE-GROUP with factor 3 at <badref>, ir<%p0>
; CHECK: Cost of 96 for VF 32: INTERLEAVE-GROUP with factor 3 at %l0, ir<%p0>
; CHECK: Cost of 96 for VF 32: INTERLEAVE-GROUP with factor 3 at <badref>, ir<%p0>
; OPT-NF3-LABEL: Checking a loop in 'i8_factor_3'
; OPT-NF3: Cost of 4 for VF 2: INTERLEAVE-GROUP with factor 3 at %l0, ir<%p0>
; OPT-NF3: Cost of 4 for VF 2: INTERLEAVE-GROUP with factor 3 at <badref>, ir<%p0>
; OPT-NF3: Cost of 4 for VF 4: INTERLEAVE-GROUP with factor 3 at %l0, ir<%p0>
; OPT-NF3: Cost of 4 for VF 4: INTERLEAVE-GROUP with factor 3 at <badref>, ir<%p0>
; OPT-NF3: Cost of 5 for VF 8: INTERLEAVE-GROUP with factor 3 at %l0, ir<%p0>
; OPT-NF3: Cost of 5 for VF 8: INTERLEAVE-GROUP with factor 3 at <badref>, ir<%p0>
; OPT-NF3: Cost of 7 for VF 16: INTERLEAVE-GROUP with factor 3 at %l0, ir<%p0>
; OPT-NF3: Cost of 7 for VF 16: INTERLEAVE-GROUP with factor 3 at <badref>, ir<%p0>
; OPT-NF3: Cost of 14 for VF 32: INTERLEAVE-GROUP with factor 3 at %l0, ir<%p0>
; OPT-NF3: Cost of 14 for VF 32: INTERLEAVE-GROUP with factor 3 at <badref>, ir<%p0>
; NO-OPT-LABEL: Checking a loop in 'i8_factor_3'
; NO-OPT: Cost of 6 for VF 2: INTERLEAVE-GROUP with factor 3 at %l0, ir<%p0>
; NO-OPT: Cost of 6 for VF 2: INTERLEAVE-GROUP with factor 3 at <badref>, ir<%p0>
; NO-OPT: Cost of 12 for VF 4: INTERLEAVE-GROUP with factor 3 at %l0, ir<%p0>
; NO-OPT: Cost of 12 for VF 4: INTERLEAVE-GROUP with factor 3 at <badref>, ir<%p0>
; NO-OPT: Cost of 24 for VF 8: INTERLEAVE-GROUP with factor 3 at %l0, ir<%p0>
; NO-OPT: Cost of 24 for VF 8: INTERLEAVE-GROUP with factor 3 at <badref>, ir<%p0>
; NO-OPT: Cost of 48 for VF 16: INTERLEAVE-GROUP with factor 3 at %l0, ir<%p0>
; NO-OPT: Cost of 48 for VF 16: INTERLEAVE-GROUP with factor 3 at <badref>, ir<%p0>
; NO-OPT: Cost of 96 for VF 32: INTERLEAVE-GROUP with factor 3 at %l0, ir<%p0>
; NO-OPT: Cost of 96 for VF 32: INTERLEAVE-GROUP with factor 3 at <badref>, ir<%p0>
for.body:
%i = phi i64 [ 0, %entry ], [ %i.next, %for.body ]
%p0 = getelementptr inbounds %i8.3, ptr %data, i64 %i, i32 0
Expand All @@ -85,17 +98,28 @@ for.end:
define void @i8_factor_4(ptr %data, i64 %n) {
entry:
br label %for.body
; CHECK-LABEL: Checking a loop in 'i8_factor_4'
; CHECK: Cost of 8 for VF 2: INTERLEAVE-GROUP with factor 4 at %l0, ir<%p0>
; CHECK: Cost of 8 for VF 2: INTERLEAVE-GROUP with factor 4 at <badref>, ir<%p0>
; CHECK: Cost of 16 for VF 4: INTERLEAVE-GROUP with factor 4 at %l0, ir<%p0>
; CHECK: Cost of 16 for VF 4: INTERLEAVE-GROUP with factor 4 at <badref>, ir<%p0>
; CHECK: Cost of 32 for VF 8: INTERLEAVE-GROUP with factor 4 at %l0, ir<%p0>
; CHECK: Cost of 32 for VF 8: INTERLEAVE-GROUP with factor 4 at <badref>, ir<%p0>
; CHECK: Cost of 64 for VF 16: INTERLEAVE-GROUP with factor 4 at %l0, ir<%p0>
; CHECK: Cost of 64 for VF 16: INTERLEAVE-GROUP with factor 4 at <badref>, ir<%p0>
; CHECK: Cost of 128 for VF 32: INTERLEAVE-GROUP with factor 4 at %l0, ir<%p0>
; CHECK: Cost of 128 for VF 32: INTERLEAVE-GROUP with factor 4 at <badref>, ir<%p0>
; OPT-NF4-LABEL: Checking a loop in 'i8_factor_4'
; OPT-NF4: Cost of 5 for VF 2: INTERLEAVE-GROUP with factor 4 at %l0, ir<%p0>
; OPT-NF4: Cost of 5 for VF 2: INTERLEAVE-GROUP with factor 4 at <badref>, ir<%p0>
; OPT-NF4: Cost of 5 for VF 4: INTERLEAVE-GROUP with factor 4 at %l0, ir<%p0>
; OPT-NF4: Cost of 5 for VF 4: INTERLEAVE-GROUP with factor 4 at <badref>, ir<%p0>
; OPT-NF4: Cost of 6 for VF 8: INTERLEAVE-GROUP with factor 4 at %l0, ir<%p0>
; OPT-NF4: Cost of 6 for VF 8: INTERLEAVE-GROUP with factor 4 at <badref>, ir<%p0>
; OPT-NF4: Cost of 8 for VF 16: INTERLEAVE-GROUP with factor 4 at %l0, ir<%p0>
; OPT-NF4: Cost of 8 for VF 16: INTERLEAVE-GROUP with factor 4 at <badref>, ir<%p0>
; OPT-NF4: Cost of 16 for VF 32: INTERLEAVE-GROUP with factor 4 at %l0, ir<%p0>
; OPT-NF4: Cost of 16 for VF 32: INTERLEAVE-GROUP with factor 4 at <badref>, ir<%p0>
; NO-OPT-LABEL: Checking a loop in 'i8_factor_4'
; NO-OPT: Cost of 8 for VF 2: INTERLEAVE-GROUP with factor 4 at %l0, ir<%p0>
; NO-OPT: Cost of 8 for VF 2: INTERLEAVE-GROUP with factor 4 at <badref>, ir<%p0>
; NO-OPT: Cost of 16 for VF 4: INTERLEAVE-GROUP with factor 4 at %l0, ir<%p0>
; NO-OPT: Cost of 16 for VF 4: INTERLEAVE-GROUP with factor 4 at <badref>, ir<%p0>
; NO-OPT: Cost of 32 for VF 8: INTERLEAVE-GROUP with factor 4 at %l0, ir<%p0>
; NO-OPT: Cost of 32 for VF 8: INTERLEAVE-GROUP with factor 4 at <badref>, ir<%p0>
; NO-OPT: Cost of 64 for VF 16: INTERLEAVE-GROUP with factor 4 at %l0, ir<%p0>
; NO-OPT: Cost of 64 for VF 16: INTERLEAVE-GROUP with factor 4 at <badref>, ir<%p0>
; NO-OPT: Cost of 128 for VF 32: INTERLEAVE-GROUP with factor 4 at %l0, ir<%p0>
; NO-OPT: Cost of 128 for VF 32: INTERLEAVE-GROUP with factor 4 at <badref>, ir<%p0>
for.body:
%i = phi i64 [ 0, %entry ], [ %i.next, %for.body ]
%p0 = getelementptr inbounds %i8.4, ptr %data, i64 %i, i32 0
Expand Down
Loading