f803e463f9
This PR implements parts of https://github.com/llvm/llvm-project/issues/162376 - **Broader equivalence than constant index deltas**: - Add Base-delta and Stride-delta matching for Add and GEP forms using ScalarEvolution deltas. - Reuse enabled for both constant and variable deltas when an available IR value dominates the user. - **Dominance-aware dictionary instead of linear scans**: - Tuple-keyed candidate dictionary grouped by basic block. - Walk the immediate-dominator chain to find the nearest dominating basis quickly and deterministically. - **Simple cost model and best-rewrite selection**: - Score candidate expressions and rewrites; select the highest-profit rewrite per instruction. - Skip rewriting when expressions are already foldable or high-efficiency. - **Path compression for better ILP**: - Compress chains of rewrites to a deeper dominating basis when a constant delta exists along the path, reducing dependent bumps on critical paths. - **Dependency-aware rewrite ordering**: - Build a dependency graph (basis, stride, variable delta producers) and rewrite in topological order. - This dependency graph will be needed by the next PR that adds partial strength reduction. - **Correctness enhencment** - Fix a correctness issue that reusing instructions with the same SCEV may introduce poison. --------- Co-authored-by: Kazu Hirata <kazu@google.com>
33 lines
785 B
LLVM
33 lines
785 B
LLVM
; RUN: opt < %s -passes="slsr" -S | FileCheck %s
|
|
|
|
target datalayout = "e-i64:64-v16:16-v32:32-n16:32:64"
|
|
|
|
%struct.B = type { i16 }
|
|
%struct.A = type { %struct.B, %struct.B }
|
|
|
|
define i32 @pick(i32 %0, ptr %addr) {
|
|
; `d` can be optimized by 2 approaches
|
|
; 1. a = 1 + 1 * %0
|
|
; d = 1 + 8 * %0
|
|
; = a + 7 * %0
|
|
; 2. c = (8 * %0) + 3
|
|
; d = (8 * %0) + 1
|
|
; = c - 2
|
|
; Pick candidate (2) as it can save 1 instruction from (7 * %0)
|
|
;
|
|
; CHECK-LABEL: pick
|
|
; CHECK: [[A:%.*]] = add i32 %0, 1
|
|
; CHECK: [[B:%.*]] = shl i32 %0, 3
|
|
; CHECK: [[C:%.*]] = add i32 [[B]], 3
|
|
; CHECK: store i32 [[C]], ptr %addr
|
|
; CHECK: [[D:%.*]] = add i32 [[C]], -2
|
|
; CHECK: ret i32 %d
|
|
|
|
%a = add i32 %0, 1
|
|
%b = shl i32 %0, 3
|
|
%c = add i32 %b, 3
|
|
store i32 %c, ptr %addr
|
|
%d = add i32 %b, 1
|
|
ret i32 %d
|
|
}
|