On 2015/08/27 13:05:52, dsinclair wrote:
On 2015/08/27 10:48:56, Yang wrote:
> On 2015/08/25 22:02:46, caseq wrote:
> >
https://codereview.chromium.org/988893003/diff/260001/include/v8-tracing.h
> > File include/v8-tracing.h (right):
> >
> >
>
https://codereview.chromium.org/988893003/diff/260001/include/v8-tracing.h#newcode49
> > include/v8-tracing.h:49: virtual EventTracer::Handle AddTraceEvent(
> > I'm very positive about the CL in general, but rather than introducing
> > additional class for embedders to implement, can we just add these to
> > v8::Platform? Besides simplifying the API slightly, there's another
small
> > advantage in that docs already stress well the need for
InitializePlatform
to
> be
> > called before v8 initialization, which I think is critical in case of
tracing
> > (due to caching of category pointers).
>
> I still feel very uncomfortable about dropping a copy of a huge file
from
> Chromium (trace-event-common.h) into V8 without any automated way to
know
> whether it needs to be updated. We have no precedence for this sort of
thing.
I
> would be a lot more comfortable if we had some way, maybe via gclient,
to
> synchronize.
The common.h file is there to be copied into other projects. We add to the
file
but we don't modify the current values of anything. So, while chrome may
get a
new macro that v8 doesn't have, until v8 needs that macro the header can
be
out
of sync. This is what is currently being done in blink, skia and nacl.
If v8 wants to add new macros they should be added into the chrome version
first, as mentioned in the .h file.
Yang, ping.
https://codereview.chromium.org/988893003/
--
--
v8-dev mailing list
[email protected]
http://groups.google.com/group/v8-dev
---
You received this message because you are subscribed to the Google Groups "v8-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
