Whoops, I meant ignition bytecode. As Turbofan is the optimizing compiler. On Thursday, April 26, 2018 at 4:03:10 PM UTC-4, [email protected] wrote: > > Hi all, > > So I have a decent understanding of v8 in most parts besides the > optimizing compiler. (Correct me if any of my understanding is wrong) > > My understanding of how v8 optimizes is it takes the turbofan bytecode and > compiles it to native assembly after some amount of calls to a function. > So starting from v8/src/compiler/pileline.cc::Pipeline::Impl it'll create > a sea of nodes graph from the original turbofan bytecode. Once the graph is > created > it is run through optimizing phases and finally generated as native code. > > My questions then are as follows: > 1) Is there any documentation on how v8's version of the sea of nodes is > implemented? > 2) How does one go about making sure the code for generating/optimizing > the graph is implemented correctly? > 3) What tools are available in v8 to assist in #2? > > Thanks, > Wes >
-- -- 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.
