| /llvm-project-15.0.7/cross-project-tests/debuginfo-tests/dexter/ |
| H A D | README.md | 161 result optimized away:
163 step 7 (Variable is optimized away and not available.)
164 step 8 (Variable is optimized away and not available.)
165 step 11 (Variable is optimized away and not available.)
166 step 14 (Variable is optimized away and not available.)
172 result optimized away:
194 result optimized away:
215 result optimized away:
244 result optimized away:
246 step 7 (Variable is optimized away and not available.)
[all …]
|
| /llvm-project-15.0.7/llvm/test/ThinLTO/X86/Inputs/ |
| H A D | weak_resolution.ll | 4 ; Alias are not optimized
7 ; Alias are not optimized
10 ; Function with an alias are not optimized
16 ; Function with an alias are not optimized
|
| /llvm-project-15.0.7/llvm/test/tools/gold/X86/Inputs/ |
| H A D | thinlto_weak_resolution.ll | 4 ; Alias are not optimized
7 ; Alias are not optimized
10 ; Function with an alias are not optimized
16 ; Function with an alias are not optimized
|
| /llvm-project-15.0.7/llvm/test/Transforms/EarlyCSE/X86/ |
| H A D | preserve_memoryssa.ll | 8 ; optimized accesses. Current MemorySSA verify will accept these.
13 ; and then AA can see that the last load is a Use(LoE). Hence not optimized as
38 ; undef they are NoAlias. The Use can be optimized further to LoE. We can
54 ; We now find the second load (Use(2) can be optimized further to LoE)
80 ; optimized to LoE. Hence after EarlyCSE, there is an access who claims is
81 ; optimized and it can be optimized further.
87 ; a correct update, but with accesses that claim to be optimized and can be
88 ; optimized further if we were to re-run MemorySSA on the IR.
|
| /llvm-project-15.0.7/lldb/test/API/tools/lldb-vscode/optimized/ |
| H A D | main.cpp | 10 int optimized = argc > 1 ? std::stoi(argv[1]) : 0; in main() local
12 printf("argc: %d, optimized: %d\n", argc, optimized); in main()
|
| /llvm-project-15.0.7/llvm/test/CodeGen/ARM/ |
| H A D | cortex-a57-misched-vfma.ll | 32 ; > VMLAS not-optimized latency to VMOVRS = 9
71 ; > VMLAfd not-optimized latency to VMOVRRD = 9
109 ; > VMLSS not-optimized latency to VMOVRS = 9
148 ; > VMLSfd not-optimized latency to VMOVRRD = 9
169 ; > VMLAS not-optimized latency to VMOVRS = 9
188 ; > VMLAS not-optimized latency to VMOVRS = 9
|
| H A D | ldst-f32-2-i32.ll | 3 ; Check if the f32 load / store pair are optimized to i32 load / store.
53 ; it can be optimized to i32 load / store
|
| /llvm-project-15.0.7/llvm/test/TableGen/ |
| H A D | ContextlessPredicates.td | 1 …sel -I %p/../../include -I %p/Common -optimize-match-table=false %s -o %T/context-non-optimized.cpp
2 // RUN: FileCheck %s --check-prefixes=CHECK_NOPT -input-file=%T/context-non-optimized.cpp
3 …al-isel -I %p/../../include -I %p/Common -optimize-match-table=true %s -o %T/context-optimized.cpp
4 // RUN: FileCheck %s --check-prefixes=CHECK_OPT -input-file=%T/context-optimized.cpp
|
| /llvm-project-15.0.7/llvm/docs/HistoricalNotes/ |
| H A D | 2003-06-25-Reoptimizer1.txt | 87 When we finally generate an optimized trace we first copy the code
89 original code, the instrumented code, and the optimized trace. The
90 optimized trace does not have instrumentation. The original code and
92 cache, where the optimized traces are kept.
98 generating the final optimized trace; we copy the same basic blocks
121 the beginning of the original code to the beginning of the optimized
132 We generate instrumentation traces and optimized traces into separate
|
| /llvm-project-15.0.7/llvm/test/Transforms/InstCombine/ |
| H A D | sprintf-void.ll | 10 ; Check that a sprintf call, that would otherwise be optimized, but with
11 ; optimized out return type, doesn't crash the optimizer.
|
| H A D | 2008-07-16-fsub.ll | 5 ; fsub of self can't be optimized away.
|
| H A D | apint-and-compare.ll | 3 ; Should be optimized to one and.
|
| /llvm-project-15.0.7/bolt/test/X86/ |
| H A D | jump-table-icp.test | 25 BOLT-INFO: ICP percentage of indirect calls that can be optimized = 0.0%
26 BOLT-INFO: ICP percentage of indirect calls that are optimized = 0.0%
27 BOLT-INFO: ICP percentage of jump table calls that can be optimized = 17.7%
28 BOLT-INFO: ICP percentage of jump table calls that are optimized = 100.0%
|
| /llvm-project-15.0.7/llvm/docs/ |
| H A D | Packaging.rst | 23 LLVM runs much more quickly when it's optimized and assertions are removed.
26 in user code. We recommend allowing users to install both optimized and debug
40 ``--enable-optimized``
|
| H A D | MemorySSA.rst | 262 ``MemoryUse``\ s keep a single operand, which is their defining or optimized
264 Traditionally ``MemorySSA`` optimized ``MemoryUse``\ s at build-time, up to a
266 Specifically, the operand of every ``MemoryUse`` was optimized to point to the
285 The second operand is the optimized access, if there was a previous call on the
291 A walk of the uses for any MemoryDef can find the accesses that were optimized
297 MemoryDef *Def; // find who's optimized or defining for this MemoryDef
302 // User who is optimized to Def
304 // User who's defining access is Def; optimized to something else or not optimized.
323 // As a user can come up twice, as an optimized access and defining
473 optimized" (it can provide a more acccurate result than if it were recomputed
[all …]
|
| /llvm-project-15.0.7/llvm/test/Transforms/GlobalOpt/ |
| H A D | 2009-11-16-MallocSingleStoreToGlobalVar.ll | 1 ; Test ensures that non-optimizable array mallocs are not optimized; specifically
4 ; element global. The global object @TOP in this test should not be optimized.
|
| /llvm-project-15.0.7/llvm/utils/valgrind/ |
| H A D | x86_64-pc-linux-gnu.supp | 77 suppress optimized strcasecmp, to be fixed in valgrind 3.6.1
83 suppress optimized strcasecmp, to be fixed in valgrind 3.6.1
|
| /llvm-project-15.0.7/llvm/test/tools/gold/X86/ |
| H A D | thinlto_weak_resolution.ll | 50 ; Alias are resolved in prevailing module, but not optimized in
57 ; not optimized in non-prevailing module (illegal to have an
67 ; not optimized in non-prevailing module (illegal to have an
|
| /llvm-project-15.0.7/llvm/test/CodeGen/AMDGPU/ |
| H A D | uniform-loop-inside-nonuniform.ll | 22 ; Prevent the loop from being optimized out
49 ; Prevent the loop from being optimized out
|
| /llvm-project-15.0.7/llvm/ |
| H A D | llvm.spec.in | 34 --enable-optimized \
62 - Make the build be optimized+assertions
|
| /llvm-project-15.0.7/polly/test/ScopInfo/ |
| H A D | non-precise-inv-load-1.ll | 5 ; optimized version. For an input of c = 127 the original accessed address
6 ; would be &I[-1] = &GI[128 -1] = &GI[127] but in our optimized version
|
| /llvm-project-15.0.7/llvm/test/Transforms/EarlyCSE/AArch64/ |
| H A D | intrinsics.ll | 8 ; Check that @llvm.aarch64.neon.ld2 is optimized away by Early CSE.
40 ; Check that the first @llvm.aarch64.neon.st2 is optimized away by Early CSE.
74 ; Check that the first @llvm.aarch64.neon.ld2 is optimized away by Early CSE.
106 ; Check that the store prevents @llvm.aarch64.neon.ld2 from being optimized
140 ; Check that @llvm.aarch64.neon.ld3 is not optimized away by Early CSE due
173 ; Check that @llvm.aarch64.neon.st3 is not optimized away by Early CSE due to
|
| /llvm-project-15.0.7/libc/AOR_v20.02/ |
| H A D | README.md | 1 This directory contains Arm's contribution of their optimized routines as
|
| /llvm-project-15.0.7/llvm/test/Transforms/MemCpyOpt/ |
| H A D | capturing-func.ll | 112 ; A potential use of the capture occurs in a later block, can't be optimized.
135 ; This case can be optimized, but would require a scan across multiple blocks
159 ; This case should not be optimized, because dest is captured before the call.
180 ; This case should not be optimized, because @g is captured before the call
|
| H A D | nonlocal-memcpy-memcpy.ll | 4 ; Test whether memcpy-memcpy dependence is optimized across
20 ; A simple partial redundancy. Test that the second memcpy is optimized
51 ; memcpy inside the triangle is optimized to copy directly from the original
99 ; optimized to copy ; directly from the original source rather than from the
|