Really? So after fork(), I'm getting all optimizations? What about debugger? After fork(), debuggers that talk to v8 via the debugger protocol seem really confused. If I don't fork(), then the debugger works great.
On Apr 27, 2012, at 10:37 AM, Jakob Kummerow wrote: > You don't need the ticker thread any more for optimizations. > > On Fri, Apr 27, 2012 at 19:27, Michael Schwartz <[email protected]> wrote: > This is on a freshly installed Ubuntu in a VM. There is no system wide > <v8.h> header. The -I./v8-read-only/include switch is having it read the v8 > from the include directory. > > Second, see: > https://groups.google.com/forum/?fromgroups#!topic/v8-users/MsrDzWXDxf4 > > I included the internal v8 header - to call AssureTickThreadStarted() (or > whatever that function is named) during testing. > > It does build with the internal include removed. It has built with that > header included for months. > > I'd appreciate a response to the link above. > > Regards, > > > On Apr 27, 2012, at 10:19 AM, Jakob Kummerow wrote: > >> a few points... >> You're #including v8-read-only/src/v8.h, which is an internal header file >> that's not supposed to be used by embedders. Try including >> v8-read-only/include/v8.h instead. >> In line 17 of process.cc, you include SilkJS.h, which has an "#include >> <v8.h>" statement, which pulls in a system-wide installed header file, which >> most likely doesn't match the version that's checked out into v8-read-only! >> Combined with the first point, I'm not at all surprised that you're getting >> all sorts of whackiness. >> At this point I believe that a vanilla V8 build would work just fine on your >> machine. Please try that next time before complaining about our build being >> broken. >> >> On Fri, Apr 27, 2012 at 19:02, mschwartz <[email protected]> wrote: >> Commenting this line in heap.h makes it compile (line 1426): >> >> // STATIC_CHECK(kempty_symbolRootIndex == Internals::kEmptySymbolRootIndex); >> >> >> On Friday, April 27, 2012 9:42:12 AM UTC-7, mschwartz wrote: >> This is the second time I've reported this kind of thing. >> >> Shouldn't v8 actually build before being put in trunk? >> >> g++ -fexceptions -fomit-frame-pointer -fdata-sections -ffunction-sections >> -fno-strict-aliasing -fvisibility=hidden -Wall -W -Wno-ignored-qualifiers >> -Wno-unused-function -Wno-unused-parameter -Wnon-virtual-dtor -m64 -O3 >> -fomit-frame-pointer -fdata-sections -ffunction-sections -ansi >> -fno-strict-aliasing -c -I./v8-read-only/include -I/usr/include/mysql -o >> process.o process.cpp >> In file included from v8-read-only/src/elements.h:32:0, >> from v8-read-only/src/objects-inl.h:38, >> from v8-read-only/src/v8.h:60, >> from process.cpp:18: >> v8-read-only/src/heap.h:1426:3: error: invalid application of ‘sizeof’ to >> incomplete type ‘StaticAssertion<false>’ >> v8-read-only/src/heap.h:1426:3: error: template argument 1 is invalid >> make[1]: *** [process.o] Error 1 >> make[1]: Leaving directory `/home/mschwartz/src/SilkJS/src' >> make: *** [all] Error 2 >> > > > -- > v8-users mailing list > [email protected] > http://groups.google.com/group/v8-users -- v8-users mailing list [email protected] http://groups.google.com/group/v8-users
