Skip to content

[mlir][ArmSME] Reword in-memory tile warning (NFC) #92415

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 1 commit into from
May 17, 2024
Merged

[mlir][ArmSME] Reword in-memory tile warning (NFC) #92415

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
4 changes: 2 additions & 2 deletions mlir/lib/Conversion/ArmSMEToLLVM/ArmSMEToLLVM.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -248,8 +248,8 @@ struct ConvertArmSMESpillsAndFillsToLLVM : public ConvertToLLVMPattern {
return failure();

tileOp->emitWarning(
"failed to allocate SME virtual tile to operation, all tile "
"operations will go through memory, expect degraded performance");
"failed to allocate SME virtual tile to operation, tile value will go "
"through memory, expect degraded performance");

// Step 1. Create an alloca for the tile at the top of the function (if one
// does not already exist).
Expand Down
4 changes: 2 additions & 2 deletions mlir/test/Conversion/ArmSMEToLLVM/tile-spills-and-fills.mlir
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -54,7 +54,7 @@
func.func @use_too_many_tiles() {
%0 = arm_sme.zero : vector<[4]x[4]xi32>
%1 = arm_sme.zero : vector<[4]x[4]xi32>
// expected-warning @below {{failed to allocate SME virtual tile to operation, all tile operations will go through memory, expect degraded performance}}
// expected-warning @below {{failed to allocate SME virtual tile to operation, tile value will go through memory, expect degraded performance}}
%2 = arm_sme.zero : vector<[8]x[8]xi16>
"test.some_use"(%0) : (vector<[4]x[4]xi32>) -> ()
"test.some_use"(%1) : (vector<[4]x[4]xi32>) -> ()
Expand Down Expand Up @@ -138,7 +138,7 @@ func.func @very_excessive_spills(%useAllTiles : vector<[16]x[16]xi8>, %memref: m
%c0 = arith.constant 0 : index
%tile = arm_sme.get_tile : vector<[4]x[4]xf32>
%mask = vector.constant_mask [4] : vector<[4]xi1>
// expected-warning @below {{failed to allocate SME virtual tile to operation, all tile operations will go through memory, expect degraded performance}}
// expected-warning @below {{failed to allocate SME virtual tile to operation, tile value will go through memory, expect degraded performance}}
%loadSlice = arm_sme.load_tile_slice %memref[%c0, %c0], %mask, %tile, %c0 : memref<?x?xf32>, vector<[4]xi1>, vector<[4]x[4]xf32>
"test.some_use"(%useAllTiles) : (vector<[16]x[16]xi8>) -> ()
"test.some_use"(%loadSlice) : (vector<[4]x[4]xf32>) -> ()
Expand Down
12 changes: 6 additions & 6 deletions mlir/test/Integration/Dialect/Linalg/CPU/ArmSME/use-too-many-tiles.mlir
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,29 +13,29 @@
/// performance (hence the warning).
func.func @use_too_many_tiles(%a: memref<?x?xi16>, %b: memref<?x?xi16>, %c: memref<?x?xi16>) {
%c0 = arith.constant 0 : index
// expected-warning @below {{failed to allocate SME virtual tile to operation, all tile operations will go through memory, expect degraded performance}}
// expected-warning @below {{failed to allocate SME virtual tile to operation, tile value will go through memory, expect degraded performance}}
%tile_a = arith.constant dense<0> : vector<[8]x[8]xi16>
// expected-warning @below {{failed to allocate SME virtual tile to operation, all tile operations will go through memory, expect degraded performance}}
// expected-warning @below {{failed to allocate SME virtual tile to operation, tile value will go through memory, expect degraded performance}}
%tile_b = arith.constant dense<1> : vector<[8]x[8]xi16>
// expected-warning @below {{failed to allocate SME virtual tile to operation, all tile operations will go through memory, expect degraded performance}}
// expected-warning @below {{failed to allocate SME virtual tile to operation, tile value will go through memory, expect degraded performance}}
%tile_c = arm_sme.tile_load %a[%c0, %c0] : memref<?x?xi16>, vector<[8]x[8]xi16>
%tile_d = arm_sme.tile_load %b[%c0, %c0] : memref<?x?xi16>, vector<[8]x[8]xi16>
%tile_e = arm_sme.tile_load %c[%c0, %c0] : memref<?x?xi16>, vector<[8]x[8]xi16>

// CHECK-LABEL: tile_a:
// CHECK-COUNT-8: ( 0, 0, 0, 0, 0, 0, 0, 0
vector.print str "tile_a:\n"
// expected-warning @below {{failed to allocate SME virtual tile to operation, all tile operations will go through memory, expect degraded performance}}
// expected-warning @below {{failed to allocate SME virtual tile to operation, tile value will go through memory, expect degraded performance}}
vector.print %tile_a : vector<[8]x[8]xi16>
// CHECK-LABEL: tile_b:
// CHECK-COUNT-8: ( 1, 1, 1, 1, 1, 1, 1, 1
vector.print str "tile_b:\n"
// expected-warning @below {{failed to allocate SME virtual tile to operation, all tile operations will go through memory, expect degraded performance}}
// expected-warning @below {{failed to allocate SME virtual tile to operation, tile value will go through memory, expect degraded performance}}
vector.print %tile_b : vector<[8]x[8]xi16>
// CHECK-LABEL: tile_c:
// CHECK-COUNT-8: ( 2, 2, 2, 2, 2, 2, 2, 2
vector.print str "tile_c:\n"
// expected-warning @below {{failed to allocate SME virtual tile to operation, all tile operations will go through memory, expect degraded performance}}
// expected-warning @below {{failed to allocate SME virtual tile to operation, tile value will go through memory, expect degraded performance}}
vector.print %tile_c : vector<[8]x[8]xi16>
// CHECK-LABEL: tile_d:
// CHECK-COUNT-8: ( 3, 3, 3, 3, 3, 3, 3, 3
Expand Down
Loading