Refactor word copying logic.

Refactor word copying logic.

Committed: http://code.google.com/p/v8/source/detail?r=4310

[antonm, 2010-03-29 14:28:55]

Vitaly,

may you have a look?

I'll need it for sort implementation.

[Vitaly Repeshko, 2010-03-29 17:02:04]

LGTM with a few nits.

http://codereview.chromium.org/1530005/diff/1/4
File src/utils.h (right):

http://codereview.chromium.org/1530005/diff/1/4#newcode600
src/utils.h:600: inline void CopyWords(Object** dst, Object** src, int
words_size) {
Maybe also implement CopyBytes and call it here? This way all clients of memcpy
get a faster alternative.

And "words_size" is a weird name. "num_words"?

[Vitaly Repeshko, 2010-03-29 17:04:22]

http://codereview.chromium.org/1530005/diff/1/4
File src/utils.h (right):

http://codereview.chromium.org/1530005/diff/1/4#newcode606
src/utils.h:606: memcpy(dst, src, words_size * kPointerSize);
Oh, and it should also be documented that dst and src must not overlap.

[antonm, 2010-03-29 17:23:23]

Thanks a lot for comments, Vitaly

http://codereview.chromium.org/1530005/diff/1/4
File src/utils.h (right):

http://codereview.chromium.org/1530005/diff/1/4#newcode600
src/utils.h:600: inline void CopyWords(Object** dst, Object** src, int
words_size) {
On 2010/03/29 17:02:04, Vitaly wrote:
> Maybe also implement CopyBytes and call it here? This way all clients of
memcpy
> get a faster alternative.
> 

I didn't see a lot of byte proper copying logic as we usually operate on words,
not bytes itself (maybe in string related code?), so if you don't mind, I won't
implement it right now.

> And "words_size" is a weird name. "num_words"?

Regarding words+size---I tried to follow byte_size in Heap::CopyBlock (but for
some reason made it in plural).

Given that which one you prefer: num_words, word_size, words_size?

http://codereview.chromium.org/1530005/diff/1/4#newcode606
src/utils.h:606: memcpy(dst, src, words_size * kPointerSize);
On 2010/03/29 17:04:22, Vitaly wrote:
> Oh, and it should also be documented that dst and src must not overlap.

Done.

[antonm, 2010-03-29 17:47:54]

assert added, param renamed

[Vitaly Repeshko, 2010-03-29 17:53:52]

Thanks, LGTM.


-- Vitaly