http://codereview.chromium.org/652119/diff/2001/3002
File src/objects.cc (right):

http://codereview.chromium.org/652119/diff/2001/3002#newcode3078
src/objects.cc:3078: if (code->type() == NORMAL) {
On 2010/02/25 13:26:24, Mads Ager wrote:
It is not completely trivial that MONOMORPHIC & NORMAL implies that
the code is
for accessing a global property.  Could you add a comment just stating
that this
is the case.  It is clear however, that MONOMORPHIC & NORMAL means
that there
can be an unbounded number of them and we therefore need a hash table.

Renamed GlobalAccess -> NormalType global_access -< normal_type, and
added a comment indicating that this is property cell access.

http://codereview.chromium.org/652119/diff/2001/3002#newcode3079
src/objects.cc:3079: // Make sure the a hash table is allocated got the
global access code cache.
On 2010/02/25 13:26:24, Mads Ager wrote:
the -> that
got -> for?

Done.

http://codereview.chromium.org/652119/diff/2001/3002#newcode3202
src/objects.cc:3202: // This is not used for global object IC's.
On 2010/02/25 13:26:24, Mads Ager wrote:
Should we just assert that code->type() != NORMAL in that case?

Done.

http://codereview.chromium.org/652119/diff/2001/3002#newcode3228
src/objects.cc:3228: // code object. The actual match is on the name and
the code flags. If this key
On 2010/02/25 13:26:24, Mads Ager wrote:
this -> a

Done.

http://codereview.chromium.org/652119/diff/2001/3002#newcode3229
src/objects.cc:3229: // is created using the flags in not a code object
it can only be used for
On 2010/02/25 13:26:24, Mads Ager wrote:
in -> and

Done.

http://codereview.chromium.org/652119/diff/2001/3003
File src/objects.h (right):

http://codereview.chromium.org/652119/diff/2001/3003#newcode3733
src/objects.h:3733: // Lookup Code object in the cache. Returns code
object if found.
On 2010/02/25 13:26:24, Mads Ager wrote:
Code -> code.

Document what it returns when not found?

Done.

http://codereview.chromium.org/652119/diff/2001/3003#newcode3737
src/objects.h:3737: // code object is not in that cache. This index can
be ised to later call
On 2010/02/25 13:26:24, Mads Ager wrote:
ised -> used

Done.

http://codereview.chromium.org/652119/diff/2001/3003#newcode3763
src/objects.h:3763: // Code cache layout. The code cache is a fixed
array where the first element
On 2010/02/25 13:26:24, Mads Ager wrote:
This comment looks wrong - left over from a previous version of this
change?

Sorry about that, fixed.

http://codereview.chromium.org/652119

--
v8-dev mailing list
[email protected]
http://groups.google.com/group/v8-dev

Reply via email to