X64: Move some methods in x64/ to their file and make codegen do int3.

X64: Move some methods in x64/ to their file and make codegen do int3.

[Lasse Reichstein, 2009-05-20 13:00:22]

Review please.

[Lasse Reichstein, 2009-05-20 13:03:27]

Apologies to laforge for miscompletion.

[Dean McNamee, 2009-05-20 13:09:42]

Ok with me

http://codereview.chromium.org/115567/diff/1/2
File src/x64/assembler-x64.cc (right):

http://codereview.chromium.org/115567/diff/1/2#newcode50
Line 50: *pc_++ = (x)
really needs to be 2 lines?

http://codereview.chromium.org/115567/diff/1/6
File src/x64/ic-x64.cc (right):

http://codereview.chromium.org/115567/diff/1/6#newcode174
Line 174: 
remote two extra new lines

[William Hesse, 2009-05-20 13:13:48]

It looks like pc(), a byte*, is being used to hold an int offset, which is being
manipulated.  See comment.

Otherwise, LGTM.

http://codereview.chromium.org/115567/diff/1/2
File src/x64/assembler-x64.cc (right):

http://codereview.chromium.org/115567/diff/1/2#newcode221
Line 221: int32_t* p = reinterpret_cast<int32_t*>(it.rinfo()->pc());
pc() is a byte*.  If you want to manipulate it as an int, cast it to intptr_t,
but I think you can do what you want directly:
it.rinfo()->set_pc( it.rinfo()->pc() - pc_delta );

[Lasse Reichstein, 2009-05-20 14:25:42]

http://codereview.chromium.org/115567/diff/1/2
File src/x64/assembler-x64.cc (right):

http://codereview.chromium.org/115567/diff/1/2#newcode221
Line 221: int32_t* p = reinterpret_cast<int32_t*>(it.rinfo()->pc());
Actually, it's not the pointer that is treated as an int32_t, but what it points
to. This should be safe, if relocation data are still integers.