Bug 104718 – Make LOG() work in WebProcess and NetworkProcess

• Component: WebKit2
• Status: RESOLVED FIXED

[Alexey Proskuryakov]

NetworkProcess doesn't initialize any logging at all. WebProcess initializes WebCore logging only, but not WebKit2 channels.

[Alexey Proskuryakov]

Created attachment 178872 [details]
proposed fix

[Alexey Proskuryakov]

Created attachment 178887 [details]
proposed fix

Thanks to thorton's vigilance, the patch is now a little larger.

[Alexey Proskuryakov]

Comment on attachment 178887 [details]
proposed fix

View in context: https://bugs.webkit.org/attachment.cgi?id=178887&action=review

> Source/WebKit2/NetworkProcess/NetworkProcess.cpp:104
> +    WebKit::initializeLogChannelsIfNecessary();

Should initialize WebCore logging here, too.

[Early Warning System Bot]

Comment on attachment 178887 [details]
proposed fix

Attachment 178887 [details] did not pass qt-ews (qt):
Output: http://queues.webkit.org/results/15282207

[Early Warning System Bot]

Comment on attachment 178887 [details]
proposed fix

Attachment 178887 [details] did not pass qt-wk2-ews (qt):
Output: http://queues.webkit.org/results/15277324

[Alexey Proskuryakov]

Created attachment 178891 [details]
with release build fix

[Build Bot]

Comment on attachment 178891 [details]
with release build fix

Attachment 178891 [details] did not pass win-ews (win):
Output: http://queues.webkit.org/results/15285037

[Alexey Proskuryakov]

Created attachment 178909 [details]
patch for landing with multiple port fixes

[WebKit Review Bot]

Comment on attachment 178909 [details]
patch for landing with multiple port fixes

Rejecting attachment 178909 [details] from commit-queue.

Failed to run "['/mnt/git/webkit-commit-queue/Tools/Scripts/webkit-patch', '--status-host=queues.webkit.org', '-..." exit_code: 2

Last 500 characters of output:
t/git/webkit-commit-queue/Source/WebKit/chromium/v8 --revision 13185 --non-interactive --force --accept theirs-conflict --ignore-externals' in '/mnt/git/webkit-commit-queue/Source/WebKit/chromium'
52>At revision 13185.

________ running '/usr/bin/python tools/clang/scripts/update.py --mac-only' in '/mnt/git/webkit-commit-queue/Source/WebKit/chromium'

________ running '/usr/bin/python gyp_webkit' in '/mnt/git/webkit-commit-queue/Source/WebKit/chromium'
Updating webkit projects from gyp files...

Full output: http://queues.webkit.org/results/15277387

[Build Bot]

Comment on attachment 178909 [details]
patch for landing with multiple port fixes

Attachment 178909 [details] did not pass mac-ews (mac):
Output: http://queues.webkit.org/results/15282464

[Alexey Proskuryakov]

Created attachment 179072 [details]
patch for landing

[WebKit Review Bot]

Comment on attachment 179072 [details]
patch for landing

Clearing flags on attachment: 179072

Committed r137486: <http://trac.webkit.org/changeset/137486>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.

[Eric Seidel (no email)]

I think this broke cr-linux:
../../Source/WebCore/platform/NotImplemented.cpp: In function ‘void logNotImplemented()’:
../../Source/WebCore/platform/NotImplemented.cpp:32:16: error: ‘printf’ was not declared in this scope

But should be an easy fix.

[Eric Seidel (no email)]

Actually, that whole file looks wrong.  Must have been a mistaken commit.  I'm not sure from this bug, since I saw ossy comment about this in another bug... /me goes to look.

[Alexey Proskuryakov]

That was from a different commit, which got reverted a moment ago anyway.

[Chris Dumez]

Comment on attachment 179072 [details]
patch for landing

View in context: https://bugs.webkit.org/attachment.cgi?id=179072&action=review

> Source/WebCore/platform/NotImplemented.h:51
> +            WTFLogVerbose(__FILE__, __LINE__, WTF_PRETTY_FUNCTION, notImplementedLoggingChannel(), "UNIMPLEMENTED: "); \

Why no WebCore:: namespace for notImplementedLoggingChannel() ? This causes a build break on EFL port. Is it mandatory for the caller to use "using namespace WebCore;" ?

[Tim Horton]

(In reply to comment #17)
> (From update of attachment 179072 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=179072&action=review
> 
> > Source/WebCore/platform/NotImplemented.h:51
> > +            WTFLogVerbose(__FILE__, __LINE__, WTF_PRETTY_FUNCTION, notImplementedLoggingChannel(), "UNIMPLEMENTED: "); \
> 
> Why no WebCore:: namespace for notImplementedLoggingChannel() ? This causes a build break on EFL port. Is it mandatory for the caller to use "using namespace WebCore;" ?

I think adding the WebCore:: here is more reasonable than silently forcing callers to import the namespace.

[Chris Dumez]

(In reply to comment #18)
> (In reply to comment #17)
> > (From update of attachment 179072 [details] [details])
> > View in context: https://bugs.webkit.org/attachment.cgi?id=179072&action=review
> > 
> > > Source/WebCore/platform/NotImplemented.h:51
> > > +            WTFLogVerbose(__FILE__, __LINE__, WTF_PRETTY_FUNCTION, notImplementedLoggingChannel(), "UNIMPLEMENTED: "); \
> > 
> > Why no WebCore:: namespace for notImplementedLoggingChannel() ? This causes a build break on EFL port. Is it mandatory for the caller to use "using namespace WebCore;" ?
> 
> I think adding the WebCore:: here is more reasonable than silently forcing callers to import the namespace.

Yes I agree, I'm fixing this in https://bugs.webkit.org/show_bug.cgi?id=104883