Home
last modified time | relevance | path

Searched +full:fail +full:- +full:fast (Results 1 – 25 of 213) sorted by relevance

123456789

/llvm-project/.github/workflows/
H A Dlibcxx-build-and-test.yaml1 # This file defines pre-commit CI for libc++, libc++abi, and libunwind (on Github).
4 # when a job fails early in the pipeline. This is why the jobs are marked as `continue-on-error: false`.
11 # Therefore, we "fail-fast" for any failures during stages 1 & 2, meaning any job failing cancels all other running jobs,
12 # under the assumption that if the "smoke tests" fail, then the other configurations will likely fail in the same way.
13 # However, stage 3 does not fail fast, a
[all...]
H A Dlibclang-python-tests.yml10 - 'main'
12 - 'clang/bindings/python/**'
13 - 'clang/tools/libclang/**'
14 - 'clang/CMakeList.txt'
15 - '.github/workflows/libclang-python-tests.yml'
16 - '.github/workflows/llvm-project-test
[all...]
/llvm-project/llvm/test/CodeGen/X86/
H A Dfast-isel-branch_weights.ll1 ; RUN: llc < %s -mtriple=x86_64-apple-darwin10 | FileCheck %s
2 ; RUN: llc < %s -fast-isel -fast-isel-abort=1 -mtriple=x86_64-apple-darwin10 | FileCheck %s
6 ; CHECK-LABEL: branch_weights_test
7 ; CHECK-LABEL: success
8 ; CHECK-LABEL: fail
10 br i1 %1, label %fail, label %success, !prof !0
12 fail:
13 ret i64 -1
H A Dfast-isel-double-half-convertion.ll1 ; RUN: llc -fast-isel -fast-isel-abort=1 -mtriple=x86_64-unknown-unknown -mattr=+f16c < %s
5 ; In the future, we might want to teach fast-isel how to expand a double-to-half
6 ; conversion into a double-to-float conversion immediately followed by a
7 ; float-to-half conversion. For now, fast-isel is expected to fail.
H A Dtailcc-fastisel.ll1 ; RUN: llc < %s -mtriple=x86_64-apple-darwin -fast-isel -fast-isel-abort=1 | FileCheck %s
6 fail: ; preds = %entry
H A Dtailcall-fastisel.ll1 ; RUN: llc < %s -mtriple=x86_64-apple-darwin -tailcallopt -fast-isel -fast-isel-abort=1 | FileCheck…
6 fail: ; preds = %entry
H A Dfdiv.ll2 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -enable-unsafe-fp-math | FileCheck %s
6 ; CHECK-LABEL: exact:
8 ; CHECK-NEXT: mulsd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
9 ; CHECK-NEXT: retq
16 ; CHECK-LABEL: inexact:
18 ; CHECK-NEXT: mulsd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
19 ; CHECK-NEXT: retq
26 ; CHECK-LABEL: funky:
28 ; CHECK-NEXT: xorpd %xmm1, %xmm1
29 ; CHECK-NEXT: divsd %xmm1, %xmm0
[all …]
H A Dfast-isel-abort-warm.ll1 ; RUN: llc -fast-isel -o - %s -fast-isel-report-on-fallback -pass-remarks-missed=isel 2>&1 | FileCh…
4 ; Thus, this test may fail because inline asm gets supported in FastISel.
6 target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
7 target triple = "x86_64-apple-macosx"
26 ; CHECK-NOT: remark: <unknown>:0:0: FastISel missed
27 ; CHECK-NOT: warning: Instruction selection used fallback path for test_instruction_not_fallback
H A Dinline-asm-error.ll1 ; RUN: not llc -mtriple=i686-- -regalloc=fast -optimize-regalloc=0 < %s 2> %t1
2 ; RUN: not llc -mtriple=i686-- -regalloc=basic < %s 2> %t2
3 ; RUN: not llc -mtriple=i686-- -regalloc=greedy < %s 2> %t3
8 ; The register allocator must fail on this function.
H A Dcmpxchg-clobber-flags.ll2 ; RUN: llc -mtriple=i386-linux-gnu -verify-machineinstrs %s -o - | FileCheck %s --check-prefixe
[all...]
/llvm-project/llvm/test/CodeGen/AArch64/
H A Dfast-isel-branch_weights.ll1 ; RUN: llc -mtriple=arm64-apple-darwin -aarch64-enable-atomic-cfg-tidy=0 …
2 ; RUN: llc -mtriple=arm64-apple-darwin -aarch64-enable-atomic-cfg-tidy=0 -fast-isel -fast-isel-abor…
6 ; CHECK-LABEL: branch_weights_test
7 ; CHECK-LABEL: success
8 ; CHECK-LABEL: fail
10 br i1 %1, label %fail, label %success, !prof !0
12 fail:
13 ret i64 -1
H A Dfast-isel-tail-call.ll1 ; RUN: llc -fast-isel -pass-remarks-missed=isel -pass-remarks-missed=isel \
2 ; RUN: -mtriple arm64-- < %s 2> %t | FileCheck %s
3 ; RUN: cat %t | FileCheck %s --check-prefix MISSED
10 ; that follow a tail-call and would have been dropped by pure SDAGISel.
12 ; Here, the %struct extractvalue should fail FastISel.
16 ; CHECK-LABEL: test:
18 ; CHECK-NEXT: .Lfunc_end0:
/llvm-project/llvm/include/llvm/ADT/
H A DEpochTracker.h1 //===- llvm/ADT/EpochTracker.h - ADT epoch tracking --------------*- C++ -*-==//
5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
7 //===----------------------------------------------------------------------===//
11 /// These can be used to write iterators that are fail-fast when LLVM is built
14 //===----------------------------------------------------------------------===//
19 #include "llvm/Config/abi-breaking.h"
29 /// ("handles") pointing into themselves fail-fast. When building without
47 /// The destructor calls incrementEpoch to make use-after-free bugs
57 /// make an iterator-invalidating modification.
67 : EpochAddress(&Parent->Epoch), EpochAtCreation(Parent->Epoch) {} in HandleBase()
/llvm-project/llvm/docs/
H A DHowToBuildOnARM.rst11 This document is *NOT* tailored to help you cross-compile LLVM/Clang
13 out more about cross-compiling, please check :doc:`HowToCrossCompileLLVM`.
22 Pandaboard, have become hard-float platforms. There are a number of
27 fail due to insufficient memory. It's also a lot quicker to only build
28 the relevant back-ends (ARM and AArch64), since it's very unlikely that
29 you'll use an ARM board to cross-compile to other arches. If you're
30 running Compiler-RT tests, also include the x86 back-end, or some tests
31 will fail.
33 .. code-block:: bash
35 cmake $LLVM_SRC_DIR -DCMAKE_BUILD_TYPE=Release \
[all …]
/llvm-project/compiler-rt/test/sanitizer_common/TestCases/Linux/
H A Dsymbolize_stack_fp.cpp2 /// Test that we can fall back to the fast unwinder.
3 … %clangxx -O0 -g1 -fno-exceptions -fno-unwind-tables -fno-asynchronous-unwind-tables -fno-omit-fra…
4 // RUN: llvm-readelf -S %t | FileCheck %s --check-prefix=SEC
7 // On android %t is a wrapper python script so llvm-readelf will fail.
10 /// Fast unwinder does not work with Thumb code
13 /// No .eh_frame && -g => .debug_frame
29 return A<N - 1>().RecursiveTemplateFunction(t); in RecursiveTemplateFunction()
/llvm-project/llvm/test/CodeGen/ARM/
H A Dfast-isel-load-store-verify.ll1 …UN: llc < %s -O0 -verify-machineinstrs -fast-isel-abort=1 -relocation-model=dynamic-no-pic -mtripl…
2 …: llc < %s -O0 -verify-machineinstrs -fast-isel-abort=1 -relocation-model=dynamic-no-pic -mtriple=…
4 ; FIXME Add tests for thumbv7, they currently fail MI verification because
8 ; and that they pass MI verification (wasn't the case until 2013-06-08).
/llvm-project/llvm/test/CodeGen/NVPTX/
H A Dfcos-no-fast-math.ll1 ; RUN: not --crash llc < %s -mtriple=nvptx -mcpu=sm_20 2>&1 | FileCheck %s
3 ; Check that we fail to select fcos without fast-math enabled
14 attributes #0 = { "unsafe-fp-math" = "false" }
H A Dfsin-no-fast-math.ll1 ; RUN: not --crash llc < %s -mtriple=nvptx -mcpu=sm_20 2>&1 | FileCheck %s
3 ; Check that we fail to select fsin without fast-math enabled
14 attributes #0 = { "unsafe-fp-math" = "false" }
/llvm-project/libc/test/src/__support/
H A Dstr_to_float_comparison_test.cpp1 //===-- strtofloatingpoint comparison test --------
[all...]
/llvm-project/flang/lib/Parser/
H A Dinstrumented-parser.cpp1 //===-- lib/Parser/instrumented-parser.cpp --------------------------------===//
5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
7 //===----------------------------------------------------------------------===//
9 #include "flang/Parser/instrumented-parser.h"
32 auto tagIter{posIter->second.perTag.find(tag)}; in Fails()
33 if (tagIter == posIter->second.perTag.end()) { in Fails()
36 auto &entry{tagIter->second}; in Fails()
38 return false; // don't fail fast, we want to generate messages in Fails()
73 o << " " << (entry.pass ? "pass" : "fail") << " " << entry.count << '\n'; in Dump()
/llvm-project/cross-project-tests/debuginfo-tests/dexter-tests/
H A Ddbg-arg.c2 // UNSUPPORTED: system-windows
5 // RUN: %clang -std=gnu11 -m64 -mllvm -fast-isel=false -g %s -o %t
6 // RUN: %dexter --fail-lt 1.0 -w \
7 // RUN: --binary %t --debugger 'lldb' -- %s
28 if (mutex->waiters) { in foobar()
/llvm-project/compiler-rt/lib/sanitizer_common/
H A Dsanitizer_flags.inc1 //===-- sanitizer_flags.h ---------------------------------------*- C++ -*-===//
5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
7 //===----------------------------------------------------------------------===//
11 //===----------------------------------------------------------------------===//
19 // Default value must be a compile-time constant.
33 "stack traces (addr2line will only be used if llvm-symbolizer binary is "
38 "If available, use the fast frame-pointer-based unwinder on "
41 "If available, use the fast frame-pointer-based unwinder on fatal "
44 // and fast-unwider is also unreliable with mixing arm and thumb code [2].
49 "If available, use the fast frame-pointer-based unwinder on "
[all …]
/llvm-project/compiler-rt/test/asan/TestCases/
H A Dstrdup_oob_test.cpp1 // RUN: %clangxx_asan -O0 %s -o %t && not %run %t 2>&1 | FileCheck %s
2 // RUN: %clangxx_asan -O1 %s -o %t && not %run %t 2>&1 | FileCheck %s
3 // RUN: %clangxx_asan -O2 %s -o %t && not %run %t 2>&1 | FileCheck %s
4 // RUN: %clangxx_asan -O3 %s -o %t && not %run %t 2>&1 | FileCheck %s
7 // RUN: %clangxx_asan -O3 -xc %s -o %t && not %run %t 2>&1 | FileCheck %s
10 // REQUIRES: (arm-target-arch || armhf-target-arch), fast-unwinder-works
12 // FIXME: We fail to intercept strdup with the dynamic WinASan RTL, so it's not
14 // XFAIL: win32-dynamic-asan
23 // CHECK: AddressSanitizer: heap-buffer-overflow in main()
24 // CHECK: #0 {{.*}}main {{.*}}strdup_oob_test.cpp:[[@LINE-2]] in main()
[all …]
/llvm-project/compiler-rt/test/BlocksRuntime/
H A Dtestfilerunner.m4 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
31 unsigned size = endp - arg + 1;
34 result[size-1] = 0;
41 result[size-1] = 0;
51 - (NSString *)description {
53 if (shouldFail) [result appendString:@"fail"];
60 - (__strong char *)radar {
64 - (bool) compileUnlessExists:(bool)skip {
115 - (bool) compileWithExpectedFailure {
183 - (bool) run {
[all …]
/llvm-project/clang/docs/
H A DThinLTO.rst13 runtime performance through whole-program analysis and cross-module
16 in time or memory, and also prevents fast incremental compiles.
22 locations for later cross-module function importing. Fast and efficient
23 whole-program analysis is then performed on the combined summary index.
29 as the distinction between the fast serial thin link step and the backends
34 <http://blog.llvm.org/2016/06/thinlto-scalable-and-incremental-lto.html>`_.
43 ----------
53 -------
59 - **gold (via the gold-plugin)**:
63 - **ld64**:
[all …]

123456789