Georgios Pinitas | cfa2bba | 2019-06-27 17:00:52 +0100 | [diff] [blame] | 1 | /* |
| 2 | * Copyright (c) 2019 ARM Limited. |
| 3 | * |
| 4 | * SPDX-License-Identifier: MIT |
| 5 | * |
| 6 | * Permission is hereby granted, free of charge, to any person obtaining a copy |
| 7 | * of this software and associated documentation files (the "Software"), to |
| 8 | * deal in the Software without restriction, including without limitation the |
| 9 | * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or |
| 10 | * sell copies of the Software, and to permit persons to whom the Software is |
| 11 | * furnished to do so, subject to the following conditions: |
| 12 | * |
| 13 | * The above copyright notice and this permission notice shall be included in all |
| 14 | * copies or substantial portions of the Software. |
| 15 | * |
| 16 | * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR |
| 17 | * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, |
| 18 | * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE |
| 19 | * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER |
| 20 | * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, |
| 21 | * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE |
| 22 | * SOFTWARE. |
| 23 | */ |
| 24 | #pragma once |
| 25 | |
| 26 | #ifndef NO_MULTI_THREADING |
| 27 | |
| 28 | #include <atomic> |
| 29 | |
| 30 | namespace arm_gemm { |
| 31 | |
| 32 | class barrier { |
| 33 | private: |
| 34 | unsigned int m_threads; |
| 35 | |
| 36 | std::atomic<unsigned int> m_waiters; |
| 37 | std::atomic<unsigned int> m_leavers; |
| 38 | |
| 39 | public: |
| 40 | barrier(unsigned int threads) : m_threads(threads), m_waiters(0), m_leavers(0) { } |
| 41 | |
| 42 | /* This isn't safe if any thread is waiting... */ |
| 43 | void set_nthreads(unsigned int nthreads) { |
| 44 | m_threads = nthreads; |
| 45 | } |
| 46 | |
| 47 | void arrive_and_wait() { |
| 48 | m_waiters++; |
| 49 | |
| 50 | while (m_waiters != m_threads) { |
| 51 | ; /* spin */ |
| 52 | } |
| 53 | |
| 54 | unsigned int v = m_leavers.fetch_add(1); |
| 55 | |
| 56 | if (v == (m_threads - 1)) { |
| 57 | m_waiters -= m_threads; |
| 58 | m_leavers = 0; |
| 59 | } else { |
| 60 | while (m_leavers > 0) { |
| 61 | ; /* spin */ |
| 62 | } |
| 63 | } |
| 64 | } |
| 65 | }; |
| 66 | |
| 67 | } // namespace arm_gemm |
| 68 | |
| 69 | #else |
| 70 | |
| 71 | namespace arm_gemm { |
| 72 | |
| 73 | class barrier { |
| 74 | public: |
| 75 | barrier(unsigned int) { } |
| 76 | |
| 77 | void arrive_and_wait() { } |
| 78 | void set_nthreads(unsigned int ) { } |
| 79 | }; |
| 80 | |
| 81 | } // namespace arm_gemm |
| 82 | |
| 83 | #endif |