maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/CodeGen/Thumb2/thumb2-jtb.ll

136 lines
3.8 KiB
LLVM
Raw Normal View History

ARM: yet another round of ARM test clean ups llvm-svn: 205586
2014-04-04 07:47:24 +08:00
; RUN: llc -mtriple=thumb-eabi -mcpu=arm1156t2-s -mattr=+thumb2 -arm-adjust-jump-tables=0 %s -o - | FileCheck %s
[Thumb-1] Synthesize TBB/TBH instructions to make use of compressed jump tables [Reapplying r284580 and r285917 with fix and testing to ensure emitted jump tables for Thumb-1 have 4-byte alignment] The TBB and TBH instructions in Thumb-2 allow jump tables to be compressed into sequences of bytes or shorts respectively. These instructions do not exist in Thumb-1, however it is possible to synthesize them out of a sequence of other instructions. It turns out this sequence is so short that it's almost never a lose for performance and is ALWAYS a significant win for code size. TBB example: Before: lsls r0, r0, #2 After: add r0, pc adr r1, .LJTI0_0 ldrb r0, [r0, #6] ldr r0, [r0, r1] lsls r0, r0, #1 mov pc, r0 add pc, r0 => No change in prologue code size or dynamic instruction count. Jump table shrunk by a factor of 4. The only case that can increase dynamic instruction count is the TBH case: Before: lsls r0, r4, #2 After: lsls r4, r4, #1 adr r1, .LJTI0_0 add r4, pc ldr r0, [r0, r1] ldrh r4, [r4, #6] mov pc, r0 lsls r4, r4, #1 add pc, r4 => 1 more instruction in prologue. Jump table shrunk by a factor of 2. So there is an argument that this should be disabled when optimizing for performance (and a TBH needs to be generated). I'm not so sure about that in practice, because on small cores with Thumb-1 performance is often tied to code size. But I'm willing to turn it off when optimizing for performance if people want (also note that TBHs are fairly rare in practice!) llvm-svn: 285690
2016-11-01 21:37:41 +08:00
; RUN: llc -mtriple=thumbv6-eabi -mcpu=cortex-m0 -arm-adjust-jump-tables=0 %s -o - | FileCheck %s
tbb / tbh instructions only branch forward, not backwards. llvm-svn: 77522
2009-07-30 07:20:20 +08:00
; Do not use tbb / tbh if any destination is before the jumptable.
; rdar://7102917
Flip the new block-placement pass to be on by default. This is mostly to test the waters. I'd like to get results from FNT build bots and other bots running on non-x86 platforms. This feature has been pretty heavily tested over the last few months by me, and it fixes several of the execution time regressions caused by the inlining work by preventing inlining decisions from radically impacting block layout. I've seen very large improvements in yacr2 and ackermann benchmarks, along with the expected noise across all of the benchmark suite whenever code layout changes. I've analyzed all of the regressions and fixed them, or found them to be impossible to fix. See my email to llvmdev for more details. I'd like for this to be in 3.1 as it complements the inliner changes, but if any failures are showing up or anyone has concerns, it is just a flag flip and so can be easily turned off. I'm switching it on tonight to try and get at least one run through various folks' performance suites in case SPEC or something else has serious issues with it. I'll watch bots and revert if anything shows up. llvm-svn: 154816
2012-04-16 21:49:17 +08:00
define i16 @main__getopt_internal_2E_exit_2E_ce(i32, i1 %b) nounwind {
FileCheck'ize test, and add a bit to test for r157221. llvm-svn: 157222
2012-05-22 07:50:00 +08:00
; CHECK: main__getopt_internal_2E_exit_2E_ce
; CHECK-NOT: tbb
; CHECK-NOT: tbh
; 32-bit jump tables use explicit branches, not data regions, so make sure
; we don't annotate this region.
; CHECK-NOT: data_region
Flip the new block-placement pass to be on by default. This is mostly to test the waters. I'd like to get results from FNT build bots and other bots running on non-x86 platforms. This feature has been pretty heavily tested over the last few months by me, and it fixes several of the execution time regressions caused by the inlining work by preventing inlining decisions from radically impacting block layout. I've seen very large improvements in yacr2 and ackermann benchmarks, along with the expected noise across all of the benchmark suite whenever code layout changes. I've analyzed all of the regressions and fixed them, or found them to be impossible to fix. See my email to llvmdev for more details. I'd like for this to be in 3.1 as it complements the inliner changes, but if any failures are showing up or anyone has concerns, it is just a flag flip and so can be easily turned off. I'm switching it on tonight to try and get at least one run through various folks' performance suites in case SPEC or something else has serious issues with it. I'll watch bots and revert if anything shows up. llvm-svn: 154816
2012-04-16 21:49:17 +08:00
entry:
br i1 %b, label %codeRepl127.exitStub, label %newFuncRoot
tbb / tbh instructions only branch forward, not backwards. llvm-svn: 77522
2009-07-30 07:20:20 +08:00
codeRepl127.exitStub: ; preds = %_getopt_internal.exit.ce
Flip the new block-placement pass to be on by default. This is mostly to test the waters. I'd like to get results from FNT build bots and other bots running on non-x86 platforms. This feature has been pretty heavily tested over the last few months by me, and it fixes several of the execution time regressions caused by the inlining work by preventing inlining decisions from radically impacting block layout. I've seen very large improvements in yacr2 and ackermann benchmarks, along with the expected noise across all of the benchmark suite whenever code layout changes. I've analyzed all of the regressions and fixed them, or found them to be impossible to fix. See my email to llvmdev for more details. I'd like for this to be in 3.1 as it complements the inliner changes, but if any failures are showing up or anyone has concerns, it is just a flag flip and so can be easily turned off. I'm switching it on tonight to try and get at least one run through various folks' performance suites in case SPEC or something else has serious issues with it. I'll watch bots and revert if anything shows up. llvm-svn: 154816
2012-04-16 21:49:17 +08:00
; Add an explicit edge back to before the jump table to ensure this block
; is placed first.
br i1 %b, label %newFuncRoot, label %codeRepl127.exitStub.exit
codeRepl127.exitStub.exit:
tbb / tbh instructions only branch forward, not backwards. llvm-svn: 77522
2009-07-30 07:20:20 +08:00
ret i16 0
parse_options.exit.loopexit.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 1
bb1.i.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 2
bb90.i.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 3
codeRepl104.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 4
codeRepl113.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 5
codeRepl51.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 6
codeRepl70.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 7
codeRepl119.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 8
codeRepl93.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 9
codeRepl101.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 10
codeRepl120.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 11
codeRepl89.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 12
codeRepl45.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 13
codeRepl58.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 14
codeRepl46.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 15
codeRepl50.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 16
codeRepl52.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 17
codeRepl53.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 18
codeRepl61.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 19
codeRepl85.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 20
codeRepl97.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 21
codeRepl79.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 22
codeRepl102.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 23
codeRepl54.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 24
codeRepl57.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 25
codeRepl103.exitStub: ; preds = %_getopt_internal.exit.ce
ret i16 26
Generalize MergeBlockIntoPredecessor. Replace uses of MergeBasicBlockIntoOnlyPred. Summary: Two utils methods have essentially the same functionality. This is an attempt to merge them into one. 1. lib/Transforms/Utils/Local.cpp : MergeBasicBlockIntoOnlyPred 2. lib/Transforms/Utils/BasicBlockUtils.cpp : MergeBlockIntoPredecessor Prior to the patch: 1. MergeBasicBlockIntoOnlyPred Updates either DomTree or DeferredDominance Moves all instructions from Pred to BB, deletes Pred Asserts BB has single predecessor If address was taken, replace the block address with constant 1 (?) 2. MergeBlockIntoPredecessor Updates DomTree, LoopInfo and MemoryDependenceResults Moves all instruction from BB to Pred, deletes BB Returns if doesn't have a single predecessor Returns if BB's address was taken After the patch: Method 2. MergeBlockIntoPredecessor is attempting to become the new default: Updates DomTree or DeferredDominance, and LoopInfo and MemoryDependenceResults Moves all instruction from BB to Pred, deletes BB Returns if doesn't have a single predecessor Returns if BB's address was taken Uses of MergeBasicBlockIntoOnlyPred that need to be replaced: 1. lib/Transforms/Scalar/LoopSimplifyCFG.cpp Updated in this patch. No challenges. 2. lib/CodeGen/CodeGenPrepare.cpp Updated in this patch. i. eliminateFallThrough is straightforward, but I added using a temporary array to avoid the iterator invalidation. ii. eliminateMostlyEmptyBlock(s) methods also now use a temporary array for blocks Some interesting aspects: - Since Pred is not deleted (BB is), the entry block does not need updating. - The entry block was being updated with the deleted block in eliminateMostlyEmptyBlock. Added assert to make obvious that BB=SinglePred. - isMergingEmptyBlockProfitable assumes BB is the one to be deleted. - eliminateMostlyEmptyBlock(BB) does not delete BB on one path, it deletes its unique predecessor instead. - adding some test owner as subscribers for the interesting tests modified: test/CodeGen/X86/avx-cmp.ll test/CodeGen/AMDGPU/nested-loop-conditions.ll test/CodeGen/AMDGPU/si-annotate-cf.ll test/CodeGen/X86/hoist-spill.ll test/CodeGen/X86/2006-11-17-IllegalMove.ll 3. lib/Transforms/Scalar/JumpThreading.cpp Not covered in this patch. It is the only use case using the DeferredDominance. I would defer to Brian Rzycki to make this replacement. Reviewers: chandlerc, spatel, davide, brzycki, bkramer, javed.absar Subscribers: qcolombet, sanjoy, nemanjai, nhaehnle, jlebar, tpr, kbarton, RKSimon, wmi, arsenm, llvm-commits Differential Revision: https://reviews.llvm.org/D48202 llvm-svn: 335183
2018-06-21 06:01:04 +08:00
newFuncRoot:
br label %_getopt_internal.exit.ce
tbb / tbh instructions only branch forward, not backwards. llvm-svn: 77522
2009-07-30 07:20:20 +08:00
_getopt_internal.exit.ce: ; preds = %newFuncRoot
switch i32 %0, label %codeRepl127.exitStub [
i32 -1, label %parse_options.exit.loopexit.exitStub
i32 0, label %bb1.i.exitStub
i32 63, label %bb90.i.exitStub
i32 66, label %codeRepl104.exitStub
i32 67, label %codeRepl113.exitStub
i32 71, label %codeRepl51.exitStub
i32 77, label %codeRepl70.exitStub
i32 78, label %codeRepl119.exitStub
i32 80, label %codeRepl93.exitStub
i32 81, label %codeRepl101.exitStub
i32 82, label %codeRepl120.exitStub
i32 88, label %codeRepl89.exitStub
i32 97, label %codeRepl45.exitStub
i32 98, label %codeRepl58.exitStub
i32 99, label %codeRepl46.exitStub
i32 100, label %codeRepl50.exitStub
i32 104, label %codeRepl52.exitStub
i32 108, label %codeRepl53.exitStub
i32 109, label %codeRepl61.exitStub
i32 110, label %codeRepl85.exitStub
i32 111, label %codeRepl97.exitStub
i32 113, label %codeRepl79.exitStub
i32 114, label %codeRepl102.exitStub
i32 115, label %codeRepl54.exitStub
i32 116, label %codeRepl57.exitStub
i32 118, label %codeRepl103.exitStub
]
}