awsy.rst (8897B)
1 ==== 2 AWSY 3 ==== 4 5 Are We Slim Yet project (commonly known as AWSY) tracks memory usage across builds. 6 7 On treeherder, the AWSY builds are listed in subgroups of `SY`. 8 9 AWSY tests consist of three types: TP5*, TP6, and Base Memory Usage. 10 11 *\*TP5 tests are out of date and no longer maintained. These tests are scheduled to be removed:* https://bugzilla.mozilla.org/show_bug.cgi?id=1712406 12 13 Awsy tests 14 ---------- 15 16 17 .. dropdown:: base (FF) 18 :class-container: anchor-id-base-Awsy-tests 19 20 * Command to Run Locally 21 22 .. code-block:: 23 24 ./mach awsy-test --base 25 26 **Owner**: :mccr8 and Perftest Team 27 28 * **Test Task**: 29 * test-linux1804-64-clang-trunk-qr/opt 30 * awsy-base: None 31 * awsy-base-dmd: None 32 33 * test-linux1804-64-qr/opt 34 * awsy-base: None 35 * awsy-base-dmd: None 36 37 * test-linux1804-64-shippable-qr/opt 38 * awsy-base: trunk 39 * awsy-base-dmd: None 40 41 * test-macosx1470-64-shippable/opt 42 * awsy-base: trunk 43 * awsy-base-dmd: None 44 45 * test-windows11-64-24h2-shippable/opt 46 * awsy-base: trunk 47 * awsy-base-dmd: None 48 49 * test-windows11-64-24h2/opt 50 * awsy-base: None 51 * awsy-base-dmd: None 52 53 54 .. dropdown:: dmd (FF) 55 :class-container: anchor-id-dmd-Awsy-tests 56 57 * Command to Run Locally 58 59 .. code-block:: 60 61 ./mach awsy-test --dmd 62 63 **Owner**: :mccr8 and Perftest Team 64 65 * **Test Task**: 66 * test-linux1804-64-clang-trunk-qr/opt 67 * awsy-base-dmd: None 68 * awsy-dmd: None 69 70 * test-linux1804-64-qr/opt 71 * awsy-base-dmd: None 72 * awsy-dmd: None 73 74 * test-linux1804-64-shippable-qr/opt 75 * awsy-base-dmd: None 76 * awsy-dmd: None 77 78 * test-macosx1470-64-shippable/opt 79 * awsy-base-dmd: None 80 * awsy-dmd: None 81 82 * test-windows11-64-24h2-shippable/opt 83 * awsy-base-dmd: None 84 * awsy-dmd: None 85 86 * test-windows11-64-24h2/opt 87 * awsy-base-dmd: None 88 * awsy-dmd: None 89 90 91 .. dropdown:: tp5 (FF) 92 :class-container: anchor-id-tp5-Awsy-tests 93 94 * Command to Run Locally 95 96 .. code-block:: 97 98 ./mach awsy-test --tp5 99 100 **Owner**: :mccr8 and Perftest Team 101 102 * **Test Task**: 103 * test-linux1804-64-clang-trunk-qr/opt 104 105 * test-linux1804-64-qr/opt 106 107 * test-linux1804-64-shippable-qr/opt 108 109 * test-macosx1470-64-shippable/opt 110 111 * test-windows11-64-24h2-shippable/opt 112 113 * test-windows11-64-24h2/opt 114 115 116 .. dropdown:: tp6 (FF) 117 :class-container: anchor-id-tp6-Awsy-tests 118 119 * Command to Run Locally 120 121 .. code-block:: 122 123 ./mach awsy-test 124 125 **Owner**: :mccr8 and Perftest Team 126 127 * **Test Task**: 128 * test-linux1804-64-clang-trunk-qr/opt 129 * awsy-tp6: None 130 131 * test-linux1804-64-qr/opt 132 * awsy-tp6: None 133 134 * test-linux1804-64-shippable-qr/opt 135 * awsy-tp6: trunk 136 137 * test-macosx1470-64-shippable/opt 138 * awsy-tp6: trunk 139 140 * test-windows11-64-24h2-shippable/opt 141 * awsy-tp6: trunk 142 143 * test-windows11-64-24h2/opt 144 * awsy-tp6: None 145 146 147 148 149 Running AWSY Locally 150 ********************* 151 152 Running tests locally is most likely only useful for debugging what is going on in a test, 153 as the test output is only reported as raw JSON. The CLI is documented via: 154 155 .. code-block:: bash 156 157 ./mach awsy-test --help 158 159 *Currently all tests will download TP5 even if it is not used, see:* https://bugzilla.mozilla.org/show_bug.cgi?id=1683920 160 161 TP5 tests 162 ========= 163 164 .. code-block:: bash 165 166 ./mach awsy-test --tp5 167 168 TP6 tests 169 ========= 170 171 .. code-block:: bash 172 173 ./mach awsy-test 174 175 Base Memory Usage tests 176 ======================== 177 178 .. code-block:: bash 179 180 ./mach awsy-test --base 181 182 Running AWSY on Try 183 ******************** 184 185 AWSY runs can be generated through the mach try fuzzy finder: 186 187 .. code-block:: bash 188 189 ./mach try fuzzy 190 191 A query for "awsy" will return all AWSY tests. The default test is TP5, TP6 and Base test names will contain `tp6` and `base`, respectively. 192 193 The following documents all tests we currently run for AWSY. 194 *The following content was migrated from* https://wiki.mozilla.org/AWSY/Tests *and will be updated to clarify TP5/TP6 tests vs Base tests:* https://bugzilla.mozilla.org/show_bug.cgi?id=1714600 195 196 197 TP5/TP6 Tests 198 ************** 199 200 The following tests exist for both TP5 and TP6. Running `./mach awsy-test` by default will run TP6 tests*. 201 To run TP5 tests, add the `--tp5` flag: `./mach awsy-test --tp5` 202 203 *\*TP5 tests are out of date and no longer maintained. These tests are scheduled to be removed:* https://bugzilla.mozilla.org/show_bug.cgi?id=1712406 204 205 206 Explicit Memory 207 ================ 208 209 * This is memory explicitly reported by a memory reporter. It includes all the memory allocated via explicit calls to heap allocation functions (such as malloc and new), and some (only that covered by memory reporters) of the memory allocated via explicit calls to non-heap allocation functions (such as mmap and VirtualAlloc). 210 211 **Possible regression causes** 212 213 * A regression in this usually means a new feature is using or retaining more memory and should be looked at. These are easier to diagnose because we can compare memory reports. 214 215 See the `about:memory` `mdn page <https://developer.mozilla.org/docs/Mozilla/Performance/about:memory#Explicit_Allocations>`__ for more details. 216 217 218 Heap Unclassified 219 ================== 220 221 *to do: add test definition* 222 223 224 Images 225 ======= 226 227 * This is a subset of the "explicit" measurement that focuses on memory used to render images. 228 229 **Possible regression causes** 230 231 * A regression in this can indicate leaks or poor memory usage in the image subsystem. In the past this was a persistent problem. 232 233 234 JS 235 ==== 236 237 *to do: add test definition* 238 239 240 Resident Memory 241 ================ 242 243 * This is a higher level measurement provided by the operating system. We sum the "resident" memory (`RSS <https://en.wikipedia.org/wiki/Resident_set_size>`_) with the `resident-unique <https://en.wikipedia.org/wiki/Unique_set_size>`_ memory of the content processes. It's pretty noisy and large so it's not very useful in detecting smaller regressions. 244 245 **Possible regression causes** 246 247 * Regressions in this often track regressions in explicit and heap unclassified. If we see a regression in resident, but not in other reports this can indicate we are leaking untracked memory (perhaps through shared memory, graphics allocations, file handles, etc). 248 249 250 Base Content Tests 251 ******************* 252 253 * An updated test focused on supporting Fission. This measures the base overhead of an empty content process. It tracks resident unique, heap unclassified, JS, and explicit memory metrics as well as storing full memory reports as artifacts. The median value for each metric is used from across all content processes. It has much lower thresholds for alerting and is recorded in `Perfherder <https://wiki.mozilla.org/EngineeringProductivity/Projects/Perfherder>`_. 254 255 256 Base Content Explicit 257 ====================== 258 259 **Possible regression causes** 260 261 A change has caused more JavaScript to load at startup or into blank pages. 262 263 * **Common solution**: lazily load any new modules you rely on 264 * **Common solution**: Split your code out to only load what is minimally needed initially. You modified the JS engine and it's using more memory 265 * **Common solution**: Attempt to reduce your object size for the common case, these tend to add up! You implemented a new feature in JavaScript 266 * **Common solution**: Write the majority (or all of it) in compiled code (C++/Rust). This will reduce overhead and generally improve performance. 267 268 269 Base Content Heap Unclassified 270 =============================== 271 272 * The "heap-unclassified" value represents heap-allocated memory that is not measured by any memory reporter. This is typically 10-20% of "explicit". 273 274 275 **Possible regression causes** 276 277 * A regression in this can indicate that we're leaking memory or that additional memory reporters should be added. 278 * An improvement can indicate that leaks have been fixed or that we added new memory reporters. 279 280 See the `about:memory` `mdn page <https://developer.mozilla.org/docs/Mozilla/Performance/about:memory#Explicit_Allocations>`__ for more details. 281 282 283 Base Content JS 284 ================ 285 286 * This is the "js-main-runtime/" value in `about:memory` which is all the memory attributed to the javascript engine. 287 288 **Possible regression causes** 289 290 * A regression in this number can indicate leaks in the JS engine, optimizations that take performance into consideration at the expense of more memory, or problems with the garbage collector. 291 292 293 Base Content Resident Unique Memory 294 ==================================== 295 296 *to do: add test definition* 297 298 299 Other references 300 ----------------- 301 302 `Are We Slim Yet MDN web docs <https://developer.mozilla.org/en-US/docs/Mozilla/Performance/AWSY>`_