Chromium Code Reviews
|
|
Chromium Code Reviews
Issue
1550833002:
PerformanceTests/Layout/large-table-background-change* to Paint (Closed)
DescriptionPerformanceTests/Layout/large-table-background-change* to Paint
The large-table-background-change* tests are actually testing
performance of paint invalidation and paint.
Justifications for the new benchmark (https://docs.google.com/document/d/1bBKyYCW3VlUUPDpQE4xvrMFdA6tovQMZoqO9KCcmqqQ/edit#):
1. Demonstrate an example regression or improvement.
CL: https://codereview.chromium.org/1549693002/
Try results:
https://codereview.chromium.org/1542263004/
https://codereview.chromium.org/1543263003/
https://codereview.chromium.org/1543333002/
2. Monitor the benchmark or metric until alerting is stable.
The tests has been running under blink_perf.layout for some days.
3. Benchmarks should run for less than 5 minutes on linux perf trybots.
The new benchmark run for a few seconds.
4. Whenever possible, avoid creating new benchmarks.
The new benchmark includes some tests that previously under blink_perf.layout.
This CL doesn't add new tests or new metrics.
5. Run benchmarks on perf trybots on all platforms before submitting.
Done.
BUG=34736
CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_10_10_perf_bisect;tryserver.chromium.perf:win_perf_bisect
Committed: https://crrev.com/cae008ceb3508fb4607a841bfd9d99492c304b05
Cr-Commit-Position: refs/heads/master@{#367344}
Patch Set 1 #
Dependent Patchsets: Messages
Total messages: 26 (11 generated)
Description was changed from ========== PerformanceTests/Layout/large-table-background-change* to Paint The large-table-background-change* tests are actually testing performance of paint invalidation and paint. BUG=34736 ========== to ========== PerformanceTests/Layout/large-table-background-change* to Paint The large-table-background-change* tests are actually testing performance of paint invalidation and paint. BUG=34736 CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_10_10_perf_bisect;tryserver.chromium.perf:win_perf_bisect ==========
wangxianzhu@chromium.org changed reviewers: + chrishtr@chromium.org, jchaffraix@chromium.org
The CQ bit was checked by wangxianzhu@chromium.org to run a CQ dry run
Dry run: CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/patch-status/1550833002/1 View timeline at https://chromium-cq-status.appspot.com/patch-timeline/1550833002/1
The CQ bit was unchecked by commit-bot@chromium.org
Dry run: This issue passed the CQ dry run.
I am not totally convinced it makes sense to split those 2 tests out as changing the border width does trigger a layout (not the border color which is what is covered by the tests). Layout also always implies some amount of paint & paint invalidation so they are also included into the measured time. That said, I could see value in knowing which team is automatically responsible for those test so lgtm. Please add some relevant paint OWNERS to the performance dashboard.
The CQ bit was checked by wangxianzhu@chromium.org
CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/patch-status/1550833002/1 View timeline at https://chromium-cq-status.appspot.com/patch-timeline/1550833002/1
wangxianzhu@chromium.org changed reviewers: + dtu@chromium.org, nednguyen@google.com
The CQ bit was unchecked by commit-bot@chromium.org
Try jobs failed on following builders: chromium_presubmit on tryserver.chromium.linux (JOB_FAILED, http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)
On 2016/01/04 17:25:14, commit-bot: I haz the power wrote: > Try jobs failed on following builders: > chromium_presubmit on tryserver.chromium.linux (JOB_FAILED, > http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...) What are these extra benchmarks do? Please add pydoc for the benchmark explaining in detail what it's about. Also we want to see justification for adding the benchmark as outlined in https://docs.google.com/document/d/1bBKyYCW3VlUUPDpQE4xvrMFdA6tovQMZoqO9KCcmq...
On 2016/01/04 17:27:49, nednguyen (ooo till 1-3) wrote: > On 2016/01/04 17:25:14, commit-bot: I haz the power wrote: > > Try jobs failed on following builders: > > chromium_presubmit on tryserver.chromium.linux (JOB_FAILED, > > > http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...) > > What are these extra benchmarks do? Please add pydoc for the benchmark > explaining in detail what it's about. Tried to add the following pydoc: """ Runs tests under third_party/WebKit/PerformanceTests/Paint to measure paint performance. """ but it seems not to add any useful information. Every benchmark in benchmarks/blink_perf.py runs tests under a sub directory of third_party/WebKit/PerformanceTests to measure performance indicated by the name of the sub directory. > Also we want to see justification for > adding the benchmark as outlined in > https://docs.google.com/document/d/1bBKyYCW3VlUUPDpQE4xvrMFdA6tovQMZoqO9KCcmq... This CL just categorize two tests from blink_perf.layout to blink_perf.paint. 1. Demonstrate an example regression or improvement. CL: https://codereview.chromium.org/1549693002/ Try results: https://codereview.chromium.org/1542263004/ https://codereview.chromium.org/1543263003/ https://codereview.chromium.org/1543333002/ 2. Monitor the benchmark or metric until alerting is stable. The tests has been running under blink_perf.layout for some days. 3. Benchmarks should run for less than 5 minutes on linux perf trybots. The new benchmark run for a few seconds. 4. Whenever possible, avoid creating new benchmarks. The new benchmark includes some tests that previously under blink_perf.layout. This CL doesn't add new tests or new metrics. 5. Run benchmarks on perf trybots on all platforms before submitting. Done.
On 2016/01/04 18:10:09, Xianzhu wrote: > On 2016/01/04 17:27:49, nednguyen (ooo till 1-3) wrote: > > On 2016/01/04 17:25:14, commit-bot: I haz the power wrote: > > > Try jobs failed on following builders: > > > chromium_presubmit on tryserver.chromium.linux (JOB_FAILED, > > > > > > http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...) > > > > What are these extra benchmarks do? Please add pydoc for the benchmark > > explaining in detail what it's about. > > Tried to add the following pydoc: > """ > Runs tests under third_party/WebKit/PerformanceTests/Paint to measure > paint performance. > """ > but it seems not to add any useful information. Every benchmark in > benchmarks/blink_perf.py runs tests under a sub directory of > third_party/WebKit/PerformanceTests to measure performance indicated by the name > of the sub directory. > > > Also we want to see justification for > > adding the benchmark as outlined in > > > https://docs.google.com/document/d/1bBKyYCW3VlUUPDpQE4xvrMFdA6tovQMZoqO9KCcmq... > > This CL just categorize two tests from blink_perf.layout to blink_perf.paint. > > 1. Demonstrate an example regression or improvement. > CL: https://codereview.chromium.org/1549693002/ > Try results: > https://codereview.chromium.org/1542263004/ > https://codereview.chromium.org/1543263003/ > https://codereview.chromium.org/1543333002/ > > 2. Monitor the benchmark or metric until alerting is stable. > The tests has been running under blink_perf.layout for some days. > > 3. Benchmarks should run for less than 5 minutes on linux perf trybots. > The new benchmark run for a few seconds. > > 4. Whenever possible, avoid creating new benchmarks. > The new benchmark includes some tests that previously under blink_perf.layout. > This CL doesn't add new tests or new metrics. > > 5. Run benchmarks on perf trybots on all platforms before submitting. > Done. lgtm Can you put 1. --> 5. in the description for archiving purpose?
On 2016/01/04 18:10:09, Xianzhu wrote: > On 2016/01/04 17:27:49, nednguyen (ooo till 1-3) wrote: > > On 2016/01/04 17:25:14, commit-bot: I haz the power wrote: > > > Try jobs failed on following builders: > > > chromium_presubmit on tryserver.chromium.linux (JOB_FAILED, > > > > > > http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...) > > > > What are these extra benchmarks do? Please add pydoc for the benchmark > > explaining in detail what it's about. > > Tried to add the following pydoc: > """ > Runs tests under third_party/WebKit/PerformanceTests/Paint to measure > paint performance. > """ > but it seems not to add any useful information. Every benchmark in > benchmarks/blink_perf.py runs tests under a sub directory of > third_party/WebKit/PerformanceTests to measure performance indicated by the name > of the sub directory. > > > Also we want to see justification for > > adding the benchmark as outlined in > > > https://docs.google.com/document/d/1bBKyYCW3VlUUPDpQE4xvrMFdA6tovQMZoqO9KCcmq... > > This CL just categorize two tests from blink_perf.layout to blink_perf.paint. > > 1. Demonstrate an example regression or improvement. > CL: https://codereview.chromium.org/1549693002/ > Try results: > https://codereview.chromium.org/1542263004/ > https://codereview.chromium.org/1543263003/ > https://codereview.chromium.org/1543333002/ > > 2. Monitor the benchmark or metric until alerting is stable. > The tests has been running under blink_perf.layout for some days. > > 3. Benchmarks should run for less than 5 minutes on linux perf trybots. > The new benchmark run for a few seconds. > > 4. Whenever possible, avoid creating new benchmarks. > The new benchmark includes some tests that previously under blink_perf.layout. > This CL doesn't add new tests or new metrics. > > 5. Run benchmarks on perf trybots on all platforms before submitting. > Done. lgtm Can you put 1. --> 5. in the description for archiving purpose?
Description was changed from ========== PerformanceTests/Layout/large-table-background-change* to Paint The large-table-background-change* tests are actually testing performance of paint invalidation and paint. BUG=34736 CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_10_10_perf_bisect;tryserver.chromium.perf:win_perf_bisect ========== to ========== PerformanceTests/Layout/large-table-background-change* to Paint The large-table-background-change* tests are actually testing performance of paint invalidation and paint. Justifications for the new benchmark (https://docs.google.com/document/d/1bBKyYCW3VlUUPDpQE4xvrMFdA6tovQMZoqO9KCcmq...): 1. Demonstrate an example regression or improvement. CL: https://codereview.chromium.org/1549693002/ Try results: https://codereview.chromium.org/1542263004/ https://codereview.chromium.org/1543263003/ https://codereview.chromium.org/1543333002/ 2. Monitor the benchmark or metric until alerting is stable. The tests has been running under blink_perf.layout for some days. 3. Benchmarks should run for less than 5 minutes on linux perf trybots. The new benchmark run for a few seconds. 4. Whenever possible, avoid creating new benchmarks. The new benchmark includes some tests that previously under blink_perf.layout. This CL doesn't add new tests or new metrics. 5. Run benchmarks on perf trybots on all platforms before submitting. Done. BUG=34736 CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_10_10_perf_bisect;tryserver.chromium.perf:win_perf_bisect ==========
On 2016/01/04 18:11:28, nednguyen (ooo till 1-3) wrote: > On 2016/01/04 18:10:09, Xianzhu wrote: > > On 2016/01/04 17:27:49, nednguyen (ooo till 1-3) wrote: > > > On 2016/01/04 17:25:14, commit-bot: I haz the power wrote: > > > > Try jobs failed on following builders: > > > > chromium_presubmit on tryserver.chromium.linux (JOB_FAILED, > > > > > > > > > > http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...) > > > > > > What are these extra benchmarks do? Please add pydoc for the benchmark > > > explaining in detail what it's about. > > > > Tried to add the following pydoc: > > """ > > Runs tests under third_party/WebKit/PerformanceTests/Paint to measure > > paint performance. > > """ > > but it seems not to add any useful information. Every benchmark in > > benchmarks/blink_perf.py runs tests under a sub directory of > > third_party/WebKit/PerformanceTests to measure performance indicated by the > name > > of the sub directory. > > > > > Also we want to see justification for > > > adding the benchmark as outlined in > > > > > > https://docs.google.com/document/d/1bBKyYCW3VlUUPDpQE4xvrMFdA6tovQMZoqO9KCcmq... > > > > This CL just categorize two tests from blink_perf.layout to blink_perf.paint. > > > > 1. Demonstrate an example regression or improvement. > > CL: https://codereview.chromium.org/1549693002/ > > Try results: > > https://codereview.chromium.org/1542263004/ > > https://codereview.chromium.org/1543263003/ > > https://codereview.chromium.org/1543333002/ > > > > 2. Monitor the benchmark or metric until alerting is stable. > > The tests has been running under blink_perf.layout for some days. > > > > 3. Benchmarks should run for less than 5 minutes on linux perf trybots. > > The new benchmark run for a few seconds. > > > > 4. Whenever possible, avoid creating new benchmarks. > > The new benchmark includes some tests that previously under blink_perf.layout. > > This CL doesn't add new tests or new metrics. > > > > 5. Run benchmarks on perf trybots on all platforms before submitting. > > Done. > > lgtm > > Can you put 1. --> 5. in the description for archiving purpose? Done.
The CQ bit was checked by wangxianzhu@chromium.org
CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/patch-status/1550833002/1 View timeline at https://chromium-cq-status.appspot.com/patch-timeline/1550833002/1
Message was sent while issue was closed.
Description was changed from ========== PerformanceTests/Layout/large-table-background-change* to Paint The large-table-background-change* tests are actually testing performance of paint invalidation and paint. Justifications for the new benchmark (https://docs.google.com/document/d/1bBKyYCW3VlUUPDpQE4xvrMFdA6tovQMZoqO9KCcmq...): 1. Demonstrate an example regression or improvement. CL: https://codereview.chromium.org/1549693002/ Try results: https://codereview.chromium.org/1542263004/ https://codereview.chromium.org/1543263003/ https://codereview.chromium.org/1543333002/ 2. Monitor the benchmark or metric until alerting is stable. The tests has been running under blink_perf.layout for some days. 3. Benchmarks should run for less than 5 minutes on linux perf trybots. The new benchmark run for a few seconds. 4. Whenever possible, avoid creating new benchmarks. The new benchmark includes some tests that previously under blink_perf.layout. This CL doesn't add new tests or new metrics. 5. Run benchmarks on perf trybots on all platforms before submitting. Done. BUG=34736 CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_10_10_perf_bisect;tryserver.chromium.perf:win_perf_bisect ========== to ========== PerformanceTests/Layout/large-table-background-change* to Paint The large-table-background-change* tests are actually testing performance of paint invalidation and paint. Justifications for the new benchmark (https://docs.google.com/document/d/1bBKyYCW3VlUUPDpQE4xvrMFdA6tovQMZoqO9KCcmq...): 1. Demonstrate an example regression or improvement. CL: https://codereview.chromium.org/1549693002/ Try results: https://codereview.chromium.org/1542263004/ https://codereview.chromium.org/1543263003/ https://codereview.chromium.org/1543333002/ 2. Monitor the benchmark or metric until alerting is stable. The tests has been running under blink_perf.layout for some days. 3. Benchmarks should run for less than 5 minutes on linux perf trybots. The new benchmark run for a few seconds. 4. Whenever possible, avoid creating new benchmarks. The new benchmark includes some tests that previously under blink_perf.layout. This CL doesn't add new tests or new metrics. 5. Run benchmarks on perf trybots on all platforms before submitting. Done. BUG=34736 CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_10_10_perf_bisect;tryserver.chromium.perf:win_perf_bisect ==========
Message was sent while issue was closed.
Committed patchset #1 (id:1)
Message was sent while issue was closed.
Description was changed from ========== PerformanceTests/Layout/large-table-background-change* to Paint The large-table-background-change* tests are actually testing performance of paint invalidation and paint. Justifications for the new benchmark (https://docs.google.com/document/d/1bBKyYCW3VlUUPDpQE4xvrMFdA6tovQMZoqO9KCcmq...): 1. Demonstrate an example regression or improvement. CL: https://codereview.chromium.org/1549693002/ Try results: https://codereview.chromium.org/1542263004/ https://codereview.chromium.org/1543263003/ https://codereview.chromium.org/1543333002/ 2. Monitor the benchmark or metric until alerting is stable. The tests has been running under blink_perf.layout for some days. 3. Benchmarks should run for less than 5 minutes on linux perf trybots. The new benchmark run for a few seconds. 4. Whenever possible, avoid creating new benchmarks. The new benchmark includes some tests that previously under blink_perf.layout. This CL doesn't add new tests or new metrics. 5. Run benchmarks on perf trybots on all platforms before submitting. Done. BUG=34736 CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_10_10_perf_bisect;tryserver.chromium.perf:win_perf_bisect ========== to ========== PerformanceTests/Layout/large-table-background-change* to Paint The large-table-background-change* tests are actually testing performance of paint invalidation and paint. Justifications for the new benchmark (https://docs.google.com/document/d/1bBKyYCW3VlUUPDpQE4xvrMFdA6tovQMZoqO9KCcmq...): 1. Demonstrate an example regression or improvement. CL: https://codereview.chromium.org/1549693002/ Try results: https://codereview.chromium.org/1542263004/ https://codereview.chromium.org/1543263003/ https://codereview.chromium.org/1543333002/ 2. Monitor the benchmark or metric until alerting is stable. The tests has been running under blink_perf.layout for some days. 3. Benchmarks should run for less than 5 minutes on linux perf trybots. The new benchmark run for a few seconds. 4. Whenever possible, avoid creating new benchmarks. The new benchmark includes some tests that previously under blink_perf.layout. This CL doesn't add new tests or new metrics. 5. Run benchmarks on perf trybots on all platforms before submitting. Done. BUG=34736 CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_10_10_perf_bisect;tryserver.chromium.perf:win_perf_bisect Committed: https://crrev.com/cae008ceb3508fb4607a841bfd9d99492c304b05 Cr-Commit-Position: refs/heads/master@{#367344} ==========
Message was sent while issue was closed.
Patchset 1 (id:??) landed as https://crrev.com/cae008ceb3508fb4607a841bfd9d99492c304b05 Cr-Commit-Position: refs/heads/master@{#367344} |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
