Short update: Yes, the "using namespace" is gone, and we also enabled the corresponding presubmit check: https://chromium-review.googlesource.com/c/v8/v8/+/647888
Kudos also go to Clemens for doing all of the legwork in the WebAssembly part of the code. On Mon, Sep 4, 2017 at 2:31 PM, Ross McIlroy <[email protected]> wrote: > Sounds like this is resolved, but +1 to avoiding "using namespace" rather > than renaming interpreter::Register if this fixes the issue. > > On 1 September 2017 at 11:10, Leszek Swirski <[email protected]> wrote: > >> Seems to build on my machine with current master and "use_jumbo_build". >> >> On Thu, Aug 31, 2017 at 11:46 PM, Jakob Kummerow <[email protected]> >> wrote: >> >>> Good point about "using namespace". Does https://chromium-review.g >>> ooglesource.com/c/v8/v8/+/646850 solve the issue then? (We should land >>> it either way.) >>> >>> On Thu, Aug 31, 2017 at 4:03 AM, Leszek Swirski <[email protected]> >>> wrote: >>> >>>> On Thu, Aug 31, 2017 at 11:54 AM, <[email protected]> wrote: >>>> >>>>> Here's an example of where I think jumbo builds could possibly >>>>> introduce ambiguity in type resolution, even if all code is inside >>>>> namespace blocks: >>>>> >>>>> Let's say a source file has some code in the v8::internal::intepreter >>>>> namespace which refers to "Register" which is resolved to >>>>> v8::internal::Register in non-jumbo builds because the code does not have >>>>> a >>>>> forward declaration or header include for v8::internal::interpreter: >>>>> :Register. >>>>> Then in jumbo builds, a header inclusion from a preceeding source file >>>>> makes v8::internal::interpreter::Register visible, and the compiler >>>>> (I assume) selects this for "Register" since it has the deepest matching >>>>> namespace. >>>>> >>>> >>>> Fair enough, that's a valid hypothetical situation, although unlikely >>>> to be the actual source of issues in this case since the interpreter >>>> doesn't care about machine registers afaik. >>>> >>>> Note that this isn't ambiguity, but rather incorrect resolution, and >>>> should show up as a different error. Ambiguity is only when there is no >>>> obvious best choice (https://godbolt.org/g/Kteb9b). >>>> >>> >>> -- >>> -- >>> 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. >>> >> >> > -- > -- > 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. > -- -- 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.
