[AArch64, ARM] Enable GlobalMerge with -O3 rather than -O1.

The pass used to be enabled by default with CodeGenOpt::Less (-O1).
This is too aggressive, considering the pass indiscriminately merges
all globals together.

Currently, performance doesn't always improve, and, on code that uses
few globals (e.g., the odd file- or function- static), more often than
not is degraded by the optimization.  Lengthy discussion can be found
on llvmdev (AArch64-focused;  ARM has similar problems):
  http://lists.cs.uiuc.edu/pipermail/llvmdev/2015-February/082800.html
Also, it makes tooling and debuggers less useful when dealing with
globals and data sections.

GlobalMerge needs to better identify those cases that benefit, and this
will be done separately.  In the meantime, move the pass to run with
-O3 rather than -O1, on both ARM and AArch64.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233024 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
Ahmed Bougacha 2015-03-23 21:17:36 +00:00
parent 6a4686f3f1
commit c9ad3ab624
13 changed files with 27 additions and 26 deletions

View File

@ -245,7 +245,7 @@ bool AArch64PassConfig::addPreISel() {
// FIXME: On AArch64, this depends on the type.
// Basically, the addressable offsets are up to 4095 * Ty.getSizeInBytes().
// and the offset has to be a multiple of the related size in bytes.
if (TM->getOptLevel() != CodeGenOpt::None)
if (TM->getOptLevel() == CodeGenOpt::Aggressive)
addPass(createGlobalMergePass(TM, 4095));
if (TM->getOptLevel() != CodeGenOpt::None)
addPass(createAArch64AddressTypePromotionPass());

View File

@ -327,7 +327,7 @@ void ARMPassConfig::addIRPasses() {
}
bool ARMPassConfig::addPreISel() {
if (TM->getOptLevel() != CodeGenOpt::None)
if (TM->getOptLevel() == CodeGenOpt::Aggressive)
// FIXME: This is using the thumb1 only constant value for
// maximal global offset for merging globals. We may want
// to look into using the old value for non-thumb1 code of

View File

@ -1,11 +1,11 @@
; RUN: llc %s -mtriple=aarch64-none-linux-gnu -enable-global-merge -o - | FileCheck %s
; RUN: llc %s -mtriple=aarch64-none-linux-gnu -enable-global-merge -global-merge-on-external -o - | FileCheck %s
; RUN: llc %s -mtriple=aarch64-none-linux-gnu -O3 -enable-global-merge -o - | FileCheck %s
; RUN: llc %s -mtriple=aarch64-none-linux-gnu -O3 -enable-global-merge -global-merge-on-external -o - | FileCheck %s
; RUN: llc %s -mtriple=aarch64-linux-gnuabi -enable-global-merge -o - | FileCheck %s
; RUN: llc %s -mtriple=aarch64-linux-gnuabi -enable-global-merge -global-merge-on-external -o - | FileCheck %s
; RUN: llc %s -mtriple=aarch64-linux-gnuabi -O3 -enable-global-merge -o - | FileCheck %s
; RUN: llc %s -mtriple=aarch64-linux-gnuabi -O3 -enable-global-merge -global-merge-on-external -o - | FileCheck %s
; RUN: llc %s -mtriple=aarch64-apple-ios -enable-global-merge -o - | FileCheck %s --check-prefix=CHECK-APPLE-IOS
; RUN: llc %s -mtriple=aarch64-apple-ios -enable-global-merge -global-merge-on-external -o - | FileCheck %s --check-prefix=CHECK-APPLE-IOS
; RUN: llc %s -mtriple=aarch64-apple-ios -O3 -enable-global-merge -o - | FileCheck %s --check-prefix=CHECK-APPLE-IOS
; RUN: llc %s -mtriple=aarch64-apple-ios -O3 -enable-global-merge -global-merge-on-external -o - | FileCheck %s --check-prefix=CHECK-APPLE-IOS
@m = internal global i32 0, align 4
@n = internal global i32 0, align 4

View File

@ -1,6 +1,6 @@
; RUN: llc %s -mtriple=aarch64-none-linux-gnu -enable-global-merge -global-merge-on-external -o - | FileCheck %s
; RUN: llc %s -mtriple=aarch64-linux-gnuabi -enable-global-merge -global-merge-on-external -o - | FileCheck %s
; RUN: llc %s -mtriple=aarch64-apple-ios -enable-global-merge -global-merge-on-external -o - | FileCheck %s --check-prefix=CHECK-APPLE-IOS
; RUN: llc %s -mtriple=aarch64-none-linux-gnu -O3 -enable-global-merge -global-merge-on-external -o - | FileCheck %s
; RUN: llc %s -mtriple=aarch64-linux-gnuabi -O3 -enable-global-merge -global-merge-on-external -o - | FileCheck %s
; RUN: llc %s -mtriple=aarch64-apple-ios -O3 -enable-global-merge -global-merge-on-external -o - | FileCheck %s --check-prefix=CHECK-APPLE-IOS
@x = global i32 0, align 4
@y = global i32 0, align 4

View File

@ -1,6 +1,6 @@
; RUN: llc %s -mtriple=aarch64-none-linux-gnu -enable-global-merge -global-merge-on-external -o - | FileCheck %s
; RUN: llc %s -mtriple=aarch64-linux-gnuabi -enable-global-merge -global-merge-on-external -o - | FileCheck %s
; RUN: llc %s -mtriple=aarch64-apple-ios -enable-global-merge -global-merge-on-external -o - | FileCheck %s --check-prefix=CHECK-APPLE-IOS
; RUN: llc %s -mtriple=aarch64-none-linux-gnu -O3 -enable-global-merge -global-merge-on-external -o - | FileCheck %s
; RUN: llc %s -mtriple=aarch64-linux-gnuabi -O3 -enable-global-merge -global-merge-on-external -o - | FileCheck %s
; RUN: llc %s -mtriple=aarch64-apple-ios -O3 -enable-global-merge -global-merge-on-external -o - | FileCheck %s --check-prefix=CHECK-APPLE-IOS
@x = global [1000 x i32] zeroinitializer, align 1
@y = global [1000 x i32] zeroinitializer, align 1

View File

@ -1,4 +1,4 @@
; RUN: llc %s -mtriple=aarch64-linux-gnuabi -enable-global-merge -o - | FileCheck %s
; RUN: llc %s -mtriple=aarch64-linux-gnuabi -O3 -enable-global-merge -o - | FileCheck %s
target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-n32:64-S128"
target triple = "arm64-apple-ios7.0.0"

View File

@ -1,6 +1,6 @@
; RUN: llc %s -mtriple=armv7-linux-gnueabi -filetype=obj -o - | \
; RUN: llc %s -mtriple=armv7-linux-gnueabi -O3 -filetype=obj -o - | \
; RUN: llvm-readobj -s -t | FileCheck -check-prefix=OBJ %s
; RUN: llc %s -mtriple=armv7-linux-gnueabi -o - | \
; RUN: llc %s -mtriple=armv7-linux-gnueabi -O3 -o - | \
; RUN: FileCheck -check-prefix=ASM %s

View File

@ -1,4 +1,4 @@
; RUN: llc -filetype=obj < %s | llvm-dwarfdump -debug-dump=info - | FileCheck %s
; RUN: llc -O3 -filetype=obj < %s | llvm-dwarfdump -debug-dump=info - | FileCheck %s
target datalayout = "e-p:32:32:32-i1:8:32-i8:8:32-i16:16:32-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:32:64-v128:32:128-a0:0:32-n32"
target triple = "thumbv7-apple-darwin10"

View File

@ -1,4 +1,4 @@
; RUN: llc < %s -mtriple=thumbv7-apple-darwin10 | FileCheck %s
; RUN: llc < %s -mtriple=thumbv7-apple-darwin10 -O3 | FileCheck %s
; CHECK: .zerofill __DATA,__bss,__MergedGlobals,16,2
@prev = external global [0 x i16]

View File

@ -1,4 +1,4 @@
; RUN: llc -filetype=obj < %s | llvm-dwarfdump -debug-dump=info - | FileCheck %s
; RUN: llc -O3 -filetype=obj < %s | llvm-dwarfdump -debug-dump=info - | FileCheck %s
; Check debug info output for merged global.
; DW_AT_location

View File

@ -1,9 +1,10 @@
; RUN: llc %s -O0 -o - | FileCheck -check-prefix=NO-MERGE %s
; RUN: llc %s -O0 -o - -enable-global-merge=false | FileCheck -check-prefix=NO-MERGE %s
; RUN: llc %s -O0 -o - -enable-global-merge=true | FileCheck -check-prefix=NO-MERGE %s
; RUN: llc %s -O1 -o - | FileCheck -check-prefix=MERGE %s
; RUN: llc %s -O1 -o - | FileCheck -check-prefix=NO-MERGE %s
; RUN: llc %s -O1 -o - -enable-global-merge=false | FileCheck -check-prefix=NO-MERGE %s
; RUN: llc %s -O1 -o - -enable-global-merge=true | FileCheck -check-prefix=MERGE %s
; RUN: llc %s -O1 -o - -enable-global-merge=true | FileCheck -check-prefix=NO-MERGE %s
; RUN: llc %s -O3 -o - | FileCheck -check-prefix=MERGE %s
; RUN: llc %s -O3 -o - -enable-global-merge=false | FileCheck -check-prefix=NO-MERGE %s
; RUN: llc %s -O3 -o - -enable-global-merge=true | FileCheck -check-prefix=MERGE %s
; MERGE-NOT: .zerofill __DATA,__bss,_bar,20,2
; MERGE-NOT: .zerofill __DATA,__bss,_baz,20,2

View File

@ -1,4 +1,4 @@
; RUN: llc < %s -mtriple=thumb-apple-darwin | FileCheck %s
; RUN: llc < %s -mtriple=thumb-apple-darwin -O3 | FileCheck %s
; Test the GlobalMerge pass. Check that the pass does not crash when using
; multiple address spaces.

View File

@ -1,4 +1,4 @@
; RUN: llc < %s -mtriple=thumb-apple-darwin -global-merge-on-const=true | FileCheck %s
; RUN: llc < %s -mtriple=thumb-apple-darwin -O3 -global-merge-on-const=true | FileCheck %s
; Test the ARMGlobalMerge pass. Use -march=thumb because it has a small
; value for the maximum offset (127).