| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
| Code-Review | +0 |
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
| Code-Review | +1 |
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
[base] Add feature to group MemoryDumpProviders which can run on same sequence.
Previously, MemoryDumpManager::CreateProcessDump ran this algorithm:
1. Group all MemoryDumpProviders by TaskRunner.
2. Get the next MemoryDumpProvider.
3. Is it bound to a TaskRunner which RunsTasksInCurrentSequence?
If yes: Go to step 4.
If no: Go to step 5.
4. Run the MemoryDumpProvider, remove it from the list,
and go back to step 2.
5. Post a task to the MemoryDumpProvider's TaskRunner,
and go back to step 2.
This algorithm did not attempt to group MemoryDumpProviders associated
with different TaskRunners which can run on the same sequence. This
could result in:
- More task posting than necessary
- If a MemoryDumpProvider is associated with a frozen TaskRunner,
it would not run, even if its TaskRunner indicates that it
could have run synchronously as part of the initial
MemoryDumpManager::CreateProcessDump call.
This CL solves these problem by introducing a feature which groups
MemoryDumpProvider which can run on the same sequence together (even if
that's through different TaskRunners).
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
CHECK(num_following_providers_.has_value());This CL would be safer to merge without this CHECK. Have a graceful failure or some way of indicating an error condition for `Memory.DumpProvider.FollowingProviders3.`
auto get_effective_task_runner = [&](const MeasuredMemoryDumpProviderInfo&
mdpinfo) {Are these changes absolutely required to gather metrics?
This sort of structural change dramatically increases the risk.
For a metrics FP++ merge, keep the changes as minimal as possible, excluding any proactive cleanup as well.
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
auto get_effective_task_runner = [&](const MeasuredMemoryDumpProviderInfo&
mdpinfo) {Are these changes absolutely required to gather metrics?
This sort of structural change dramatically increases the risk.
For a metrics FP++ merge, keep the changes as minimal as possible, excluding any proactive cleanup as well.
Ack, withdrew the FP++ request.
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |