On 2015/07/28 13:35:28, oth wrote:
On 2015/07/28 08:58:12, rmcilroy (OOO until 10th Aug) wrote: >
https://codereview.chromium.org/1257543003/diff/40001/src/interpreter/interpreter.cc#newcode57
> src/interpreter/interpreter.cc:57: void
> Interpreter::DoLoadSmi8(compiler::InterpreterAssembler* assembler) {
> On 2015/07/27 16:59:00, oth wrote:
> > On 2015/07/27 14:56:58, rmcilroy (OOO until 10th Aug) wrote:
> > > Is LoadLiteralSmi8 too wordy? (honest question).
> >
> > There are two issues here - how handlers are named and what the
disassembly
> > looks like. Personally, I'd lean towards compact and easy to scan disassembler> > output. Handlers could have the more descriptive names and the disassembly
a
> > compact and easy scan format by adding an extra argument to DECLARE_BYTECODE? > > No, I'd rather have the handlers named the same as the bytecode in disassembly. > This is fine I think.
The names are particularly important today, but LoadLiteralSmi8 will be insufficient now we're moving to registers plus accumulator assuming the interpreter supports loads to both. (LdaSmi8, LdrSmi8).
s/important/not important/ :-) https://codereview.chromium.org/1257543003/ -- -- 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.
