228.63 - 4.23% browser_ml_summarizer_perf.js SUM-ONNX-COMMUNITY-QWEN2.5-0.5B-INSTRUCT_BIG-total-memory-usage / browser_ml_autofill_perf.js AUTOFILL-model-run-latency + 24 more (Linux, OSX, Windows) regression on Tue February 11 2025
Categories
(Core :: Machine Learning, defect)
Tracking
()
Tracking | Status | |
---|---|---|
firefox137 | --- | affected |
People
(Reporter: intermittent-bug-filer, Unassigned)
References
(Regression)
Details
(Keywords: perf, perf-alert, regression)
Perfherder has detected a mozperftest performance regression from push e9c59acf56eba979d7dd81d912d2ba568970b42b. As author of one of the patches included in that push, we need your help to address this regression.
Regressions:
Ratio | Test | Platform | Options | Absolute values (old vs new) |
---|---|---|---|---|
229% | browser_ml_summarizer_perf.js SUM-ONNX-COMMUNITY-QWEN2.5-0.5B-INSTRUCT_BIG-total-memory-usage | windows11-64-2009-hw-ref-shippable | 900.21 -> 2,958.33 | |
224% | browser_ml_summarizer_perf.js SUM-ONNX-COMMUNITY-QWEN2.5-0.5B-INSTRUCT_BIG-total-memory-usage | macosx1015-64-shippable-qr | 895.00 -> 2,901.42 | |
222% | browser_ml_summarizer_perf.js SUM-ONNX-COMMUNITY-QWEN2.5-0.5B-INSTRUCT_BIG-total-memory-usage | linux1804-64-shippable | 896.00 -> 2,889.54 | |
198% | browser_ml_summarizer_perf.js SUM-ONNX-COMMUNITY-QWEN2.5-0.5B-INSTRUCT_BIG-total-memory-usage | windows11-64-shippable-qr | 900.46 -> 2,686.75 | |
152% | browser_ml_summarizer_perf.js SUM-XENOVA-DISTILBART-CNN-12-6_BIG-total-memory-usage | windows11-64-shippable-qr | 543.00 -> 1,366.67 | |
131% | browser_ml_summarizer_perf.js SUM-XENOVA-DISTILBART-CNN-12-6_BIG-total-memory-usage | windows11-64-2009-hw-ref-shippable | 543.00 -> 1,256.96 | |
125% | browser_ml_summarizer_perf.js SUM-XENOVA-DISTILBART-CNN-12-6_BIG-total-memory-usage | macosx1015-64-shippable-qr | 541.00 -> 1,217.29 | |
111% | browser_ml_summarizer_perf.js SUM-XENOVA-DISTILBART-CNN-12-6_BIG-total-memory-usage | linux1804-64-shippable | 541.00 -> 1,142.17 | |
51% | browser_ml_smart_tab_perf.js SMART-TAB-TOPIC-total-memory-usage | windows11-64-shippable-qr | 414.00 -> 623.96 | |
50% | browser_ml_smart_tab_perf.js SMART-TAB-TOPIC-cold-start-total-memory-usage | windows11-64-shippable-qr | 413.67 -> 621.83 | |
... | ... | ... | ... | ... |
16% | browser_ml_smart_tab_perf.js SMART-TAB-TOPIC-total-memory-usage | linux1804-64-shippable | 412.04 -> 477.38 | |
15% | browser_ml_smart_tab_perf.js SMART-TAB-TOPIC-cold-start-total-memory-usage | linux1804-64-shippable | 412.17 -> 474.75 | |
12% | browser_ml_autofill_perf.js AUTOFILL-total-memory-usage | windows11-64-shippable-qr | 197.00 -> 221.00 | |
12% | browser_ml_autofill_perf.js AUTOFILL-total-memory-usage | windows11-64-2009-hw-ref-shippable | 197.00 -> 220.62 | |
4% | browser_ml_autofill_perf.js AUTOFILL-model-run-latency | linux1804-64-shippable | 18.03 -> 18.79 |
Improvements:
Ratio | Test | Platform | Options | Absolute values (old vs new) |
---|---|---|---|---|
37% | browser_ml_smart_tab_perf.js SMART-TAB-EMBEDDING-total-memory-usage | linux1804-64-shippable | 349.00 -> 219.54 | |
36% | browser_ml_smart_tab_perf.js SMART-TAB-EMBEDDING-cold-start-total-memory-usage | linux1804-64-shippable | 349.00 -> 222.33 | |
30% | browser_ml_engine_multi_perf.js total-memory-usage | linux1804-64-shippable | 1,410.00 -> 986.58 | |
29% | browser_ml_engine_multi_perf.js total-memory-usage | macosx1015-64-shippable-qr | 1,409.00 -> 999.83 | |
28% | browser_ml_smart_tab_perf.js SMART-TAB-EMBEDDING-total-memory-usage | macosx1015-64-shippable-qr | 349.00 -> 250.71 | |
... | ... | ... | ... | ... |
7% | browser_ml_engine_perf.js EXAMPLE-cold-start-total-memory-usage | macosx1015-64-shippable-qr | 362.00 -> 335.00 |
Details of the alert can be found in the alert summary, including links to graphs and comparisons for each of the affected tests. Please follow our guide to handling regression bugs and let us know your plans within 3 business days, or the patch(es) may be backed out in accordance with our regression policy.
If you need the profiling jobs you can trigger them yourself from treeherder job view or ask a sheriff to do that for you.
You can run all of these tests on try with ./mach try perf --alert 43852
The following documentation link provides more information about this command.
For more information on performance sheriffing please see our FAQ.
If you have any questions, please do not hesitate to reach out to afinder@mozilla.com.
Updated•8 days ago
|
Comment 1•8 days ago
|
||
baseline change
Description
•