Skip to content

[AArch64] Dont inline streaming fn into non-streaming caller #150595

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 4 commits into from
Aug 1, 2025
Merged

[AArch64] Dont inline streaming fn into non-streaming caller #150595

Show file tree
Hide file tree
Changes from 3 commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
3251573
Precommit test
sdesmalen-arm Jul 24, 2025
4cfce25
[AArch64] Dont inline streaming fn into non-streaming caller
sdesmalen-arm Jul 24, 2025
443e282
Fix tests
sdesmalen-arm Jul 31, 2025
4ca6a64
Update comments in test
sdesmalen-arm Jul 31, 2025
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/AArch64/AArch64TargetTransformInfo.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -270,6 +270,13 @@ bool AArch64TTIImpl::areInlineCompatible(const Function *Caller,
const Function *Callee) const {
SMECallAttrs CallAttrs(*Caller, *Callee);

// Never inline a function explicitly marked as being streaming,
// into a non-streaming function. Assume it was marked as streaming
// for a reason.
if (CallAttrs.caller().hasNonStreamingInterfaceAndBody() &&
CallAttrs.callee().hasStreamingInterfaceOrBody())
return false;

// When inlining, we should consider the body of the function, not the
// interface.
if (CallAttrs.callee().hasStreamingBody()) {
Expand Down
18 changes: 9 additions & 9 deletions llvm/test/Transforms/Inline/AArch64/sme-pstatesm-attrs-low-threshold.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,8 +9,8 @@ declare void @streaming_compatible_f() #0 "aarch64_pstate_sm_compatible"

; Function @streaming_callee doesn't contain any operations that may use ZA
; state and therefore can be legally inlined into a normal function.
define void @streaming_callee() #0 "aarch64_pstate_sm_enabled" {
; CHECK-LABEL: define void @streaming_callee
define void @non_streaming_callee() #0 {
; CHECK-LABEL: define void @non_streaming_callee
; CHECK-SAME: () #[[ATTR1:[0-9]+]] {
; CHECK-NEXT: call void @streaming_compatible_f()
; CHECK-NEXT: call void @streaming_compatible_f()
Expand All @@ -22,25 +22,25 @@ define void @streaming_callee() #0 "aarch64_pstate_sm_enabled" {
}

; Inline call to @streaming_callee to remove a streaming mode change.
define void @non_streaming_caller_inline() #0 {
; CHECK-LABEL: define void @non_streaming_caller_inline
define void @streaming_caller_inline() #0 "aarch64_pstate_sm_enabled" {
; CHECK-LABEL: define void @streaming_caller_inline
; CHECK-SAME: () #[[ATTR2:[0-9]+]] {
; CHECK-NEXT: call void @streaming_compatible_f()
; CHECK-NEXT: call void @streaming_compatible_f()
; CHECK-NEXT: ret void
;
call void @streaming_callee()
call void @non_streaming_callee()
ret void
}

; Don't inline call to @streaming_callee when the inline-threshold is set to 1, because it does not eliminate a streaming-mode change.
define void @streaming_caller_dont_inline() #0 "aarch64_pstate_sm_enabled" {
; CHECK-LABEL: define void @streaming_caller_dont_inline
define void @non_streaming_caller_dont_inline() #0 {
; CHECK-LABEL: define void @non_streaming_caller_dont_inline
; CHECK-SAME: () #[[ATTR1]] {
; CHECK-NEXT: call void @streaming_callee()
; CHECK-NEXT: call void @non_streaming_callee()
; CHECK-NEXT: ret void
;
call void @streaming_callee()
call void @non_streaming_callee()
ret void
}

Expand Down
Loading