Talked to Kasper today, he brought up a good point of moving to Immediate() could allow us to eventually move to having registers be an enum, otherwise we have the problem of mov(reg, reg) and mov(reg, int) being ambiguous. I can look into assembler cleanup someday, but for now I think this patch makes sense.
Kasper? On Mon, Nov 10, 2008 at 7:13 AM, Dean McNamee <[EMAIL PROTECTED]> wrote: > I talked to Kasper, and I think the conclusion was to use Immediate(), > but I'm not positive. I'll see what he thinks. Thanks for the > review. > > -- dean > > On Mon, Nov 10, 2008 at 7:05 AM, <[EMAIL PROTECTED]> wrote: >> I like it. Even if it doesn't result in a shorter encoding, I think it >> reads better to get rid of Immediate in places where it isn't needed. >> >> >> http://codereview.chromium.org/10002/diff/401/604 >> File src/codegen-ia32.cc (right): >> >> http://codereview.chromium.org/10002/diff/401/604#newcode1044 >> Line 1044: __ mov(eax, Immediate(value)); >> Couldn't this just be __ mov(eax, value)? >> >> http://codereview.chromium.org/10002/diff/401/604#newcode4987 >> Line 4987: __ mov(eax, Immediate(reinterpret_cast<int32_t>(failure))); >> Similarly __ mov(eax, reinterpret_cast<int32_t>(failure))? >> >> http://codereview.chromium.org/10002/diff/401/604#newcode5005 >> Line 5005: __ mov(eax, Immediate(reinterpret_cast<int32_t>(failure))); >> Ditto. >> >> http://codereview.chromium.org/10002/diff/401/606 >> File src/macro-assembler-ia32.cc (right): >> >> http://codereview.chromium.org/10002/diff/401/606#newcode698 >> Line 698: mov(eax, Immediate(Factory::undefined_value())); >> Just mov(eax, Factory::undefined_value())? >> >> http://codereview.chromium.org/10002/diff/401/606#newcode790 >> Line 790: mov(edx, Immediate(code_constant)); >> Ditto. >> >> http://codereview.chromium.org/10002/diff/401/607 >> File src/stub-cache-ia32.cc (right): >> >> http://codereview.chromium.org/10002/diff/401/607#newcode398 >> Line 398: __ mov(ecx, Immediate(Handle<Map>(transition))); >> Ditto and below. >> >> http://codereview.chromium.org/10002 >> > --~--~---------~--~----~------------~-------~--~----~ v8-dev mailing list [email protected] http://groups.google.com/group/v8-dev -~----------~----~----~----~------~----~------~--~---
