Fix for issue 2233

Fix for issue 2233

- Fixed the bigint instance allocation path when full snapshots are read.

- JSRegExp objects are only serialized/deserialized in a message snapshot and
hence do not need that special path. There is an assertion in JSRegExp to
ensure that these objects are not part of a full snapshot.
Committed: https://code.google.com/p/dart/source/detail?r=5802

[cshapiro, 2012-03-24 01:48:07]

lgtm

https://chromiumcodereview.appspot.com/9835074/diff/5003/vm/bigint_operations.cc
File vm/bigint_operations.cc (right):

https://chromiumcodereview.appspot.com/9835074/diff/5003/vm/bigint_operations...
vm/bigint_operations.cc:126: intptr_t bigint_length = ((hex_length - 1) /
kHexCharsPerDigit) + 1;
Utils::RoundUp ?

https://chromiumcodereview.appspot.com/9835074/diff/5003/vm/bigint_operations.h
File vm/bigint_operations.h (right):

https://chromiumcodereview.appspot.com/9835074/diff/5003/vm/bigint_operations...
vm/bigint_operations.h:25: // Compute length of bigint instance created for the
specified hex string.
What is the unit for this length?

https://chromiumcodereview.appspot.com/9835074/diff/5003/vm/bigint_operations...
vm/bigint_operations.h:42: // The given string must be a null-terminated string
of decimal digits. It
just an fyi, nul (one 'l') is the name of the ascii character, but null (two
'l') is okay as well.  the original used one 'l'.

https://chromiumcodereview.appspot.com/9835074/diff/5003/vm/snapshot_test.cc
File vm/snapshot_test.cc (right):

https://chromiumcodereview.appspot.com/9835074/diff/5003/vm/snapshot_test.cc#...
vm/snapshot_test.cc:782: "  final int fld2_1 = 0xfffffffffff;\n"
maybe just name these bigintField or something just to make its purpose clear?

https://chromiumcodereview.appspot.com/9835074/diff/5003/vm/snapshot_test.cc#...
vm/snapshot_test.cc:784: "  static final int fld4 = 10;\n"
this could be smiField, etc.

[siva, 2012-03-24 02:25:03]

https://chromiumcodereview.appspot.com/9835074/diff/5003/vm/bigint_operations.cc
File vm/bigint_operations.cc (right):

https://chromiumcodereview.appspot.com/9835074/diff/5003/vm/bigint_operations...
vm/bigint_operations.cc:126: intptr_t bigint_length = ((hex_length - 1) /
kHexCharsPerDigit) + 1;
The current Roundup implementation doesn't quite work here as strings smaller
than 7 chars would return 0.

On 2012/03/24 01:48:07, cshapiro wrote:
> Utils::RoundUp ?

https://chromiumcodereview.appspot.com/9835074/diff/5003/vm/bigint_operations.h
File vm/bigint_operations.h (right):

https://chromiumcodereview.appspot.com/9835074/diff/5003/vm/bigint_operations...
vm/bigint_operations.h:25: // Compute length of bigint instance created for the
specified hex string.
Changed comment.

On 2012/03/24 01:48:07, cshapiro wrote:
> What is the unit for this length?

https://chromiumcodereview.appspot.com/9835074/diff/5003/vm/bigint_operations...
vm/bigint_operations.h:42: // The given string must be a null-terminated string
of decimal digits. It
Reverted to nul-terminated.

On 2012/03/24 01:48:07, cshapiro wrote:
> just an fyi, nul (one 'l') is the name of the ascii character, but null (two
> 'l') is okay as well.  the original used one 'l'.

https://chromiumcodereview.appspot.com/9835074/diff/5003/vm/snapshot_test.cc
File vm/snapshot_test.cc (right):

https://chromiumcodereview.appspot.com/9835074/diff/5003/vm/snapshot_test.cc#...
vm/snapshot_test.cc:782: "  final int fld2_1 = 0xfffffffffff;\n"
On 2012/03/24 01:48:07, cshapiro wrote:
> maybe just name these bigintField or something just to make its purpose clear?

Done.

https://chromiumcodereview.appspot.com/9835074/diff/5003/vm/snapshot_test.cc#...
vm/snapshot_test.cc:784: "  static final int fld4 = 10;\n"
On 2012/03/24 01:48:07, cshapiro wrote:
> this could be smiField, etc.

Done.