Bug 202308 – Move service worker process termination delay disabling from process pool to website data store

• Component: New Bugs
• Status: RESOLVED FIXED

[Alex Christensen]

Move service worker process termination delay disabling from process pool to website data store

[Alex Christensen]

Created attachment 379719 [details]
Patch

[Chris Dumez]

Comment on attachment 379719 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=379719&action=review

> Source/WebCore/workers/service/server/SWServer.cpp:305
> +    , m_processTerminationDelayEnabled(processTerminationDelayEnabled)

This needs a "is" prefix since this is a Boolean variable.

> Source/WebCore/workers/service/server/SWServer.h:257
> +    bool m_processTerminationDelayEnabled { true };

The new name is not as good because it is missing a prefix, violating coding style.

> Source/WebKit/NetworkProcess/NetworkProcess.h:532
> +        bool processTerminationDelayEnabled { true };

Can you explain why we need this and we cannot simply get this information from the SWServer in m_swServers ? You already store this boolean there.

[Alex Christensen]

Comment on attachment 379719 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=379719&action=review

>> Source/WebCore/workers/service/server/SWServer.cpp:305
>> +    , m_processTerminationDelayEnabled(processTerminationDelayEnabled)
> 
> This needs a "is" prefix since this is a Boolean variable.

I'll update the name when committing.

>> Source/WebKit/NetworkProcess/NetworkProcess.h:532
>> +        bool processTerminationDelayEnabled { true };
> 
> Can you explain why we need this and we cannot simply get this information from the SWServer in m_swServers ? You already store this boolean there.

This information is used when initializing the SWServer because they are lazily initialized.  If we don't have this, then we'll need another HashMap.

[Chris Dumez]

(In reply to Alex Christensen from comment #3)
> Comment on attachment 379719 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=379719&action=review
> 
> >> Source/WebCore/workers/service/server/SWServer.cpp:305
> >> +    , m_processTerminationDelayEnabled(processTerminationDelayEnabled)
> > 
> > This needs a "is" prefix since this is a Boolean variable.
> 
> I'll update the name when committing.
> 
> >> Source/WebKit/NetworkProcess/NetworkProcess.h:532
> >> +        bool processTerminationDelayEnabled { true };
> > 
> > Can you explain why we need this and we cannot simply get this information from the SWServer in m_swServers ? You already store this boolean there.
> 
> This information is used when initializing the SWServer because they are
> lazily initialized.  If we don't have this, then we'll need another HashMap.

Ok, thanks.

[Alex Christensen]

http://trac.webkit.org/r250426

[Radar WebKit Bug Importer]

<rdar://problem/55783546>