remove Sk64 from public API, and start to remove usage internally

remove Sk64 from public API, and start to remove usage internally

BUG=
R=robertphillips@google.com

Committed: https://code.google.com/p/skia/source/detail?r=12840

[reed1, 2013-12-20 16:18:17]

define SK_SUPPORT_LEGACY_SK64 for chrome (for now)

[robertphillips, 2013-12-20 16:40:21]

lgtm + questions

https://codereview.chromium.org/113823003/diff/1/include/core/SkBitmap.h
File include/core/SkBitmap.h (right):

https://codereview.chromium.org/113823003/diff/1/include/core/SkBitmap.h#newc...
include/core/SkBitmap.h:169: #ifdef SK_SUPPORT_LEGACY_SK64
use computeSize64?

https://codereview.chromium.org/113823003/diff/1/include/core/SkBitmap.h#newc...
include/core/SkBitmap.h:176: 
use computeSafeSize64?

https://codereview.chromium.org/113823003/diff/1/include/core/SkMath.h
File include/core/SkMath.h (right):

https://codereview.chromium.org/113823003/diff/1/include/core/SkMath.h#newcode48
include/core/SkMath.h:48: *  Return the 64bit argument as signed 32bits,
asserting in debug that the arg
performed?

https://codereview.chromium.org/113823003/diff/1/src/core/SkBitmap.cpp
File src/core/SkBitmap.cpp (right):

https://codereview.chromium.org/113823003/diff/1/src/core/SkBitmap.cpp#newcode47
src/core/SkBitmap.cpp:47: }
Don't we really want sk_64_asSizeT?

https://codereview.chromium.org/113823003/diff/1/src/core/SkBitmap.cpp#newcod...
src/core/SkBitmap.cpp:201: }
same here?

https://codereview.chromium.org/113823003/diff/1/src/core/SkBitmap.cpp#newcod...
src/core/SkBitmap.cpp:211: int64_t size = SkBitmap::ComputeSize64(c, width,
height);
and here?

https://codereview.chromium.org/113823003/diff/1/src/core/SkBitmap.cpp#newcod...
src/core/SkBitmap.cpp:232: int64_t safeSize = ComputeSafeSize64(config, width,
height, rowBytes);
Why not use sk_64_isS32 here?
sk_64_isSizeT?

https://codereview.chromium.org/113823003/diff/1/src/core/SkMallocPixelRef.cpp
File src/core/SkMallocPixelRef.cpp (right):

https://codereview.chromium.org/113823003/diff/1/src/core/SkMallocPixelRef.cp...
src/core/SkMallocPixelRef.cpp:67: int64_t bigSize = (int64_t)info.fHeight *
rowBytes;
isSizeT?

https://codereview.chromium.org/113823003/diff/1/src/core/SkMallocPixelRef.cp...
src/core/SkMallocPixelRef.cpp:71: 
asSizeT?

https://codereview.chromium.org/113823003/diff/1/src/core/SkMipMap.cpp
File src/core/SkMipMap.cpp (right):

https://codereview.chromium.org/113823003/diff/1/src/core/SkMipMap.cpp#newcod...
src/core/SkMipMap.cpp:116: int64_t size = sk_64_mul(levelCount + 1,
sizeof(Level)) + pixelSize;
same here?

https://codereview.chromium.org/113823003/diff/1/src/core/SkRegion_path.cpp
File src/core/SkRegion_path.cpp (right):

https://codereview.chromium.org/113823003/diff/1/src/core/SkRegion_path.cpp#n...
src/core/SkRegion_path.cpp:125: 
same here?

https://codereview.chromium.org/113823003/diff/1/src/core/SkRegion_path.cpp#n...
src/core/SkRegion_path.cpp:131: int64_t size = sk_64_mul(fStorageCount,
sizeof(SkRegion::RunType));
isSizeT?

https://codereview.chromium.org/113823003/diff/1/src/core/SkRegion_path.cpp#n...
src/core/SkRegion_path.cpp:135: 
asSizeT?

[bungeman-skia, 2013-12-20 17:00:30]

https://codereview.chromium.org/113823003/diff/1/src/core/SkBitmap.cpp
File src/core/SkBitmap.cpp (right):

https://codereview.chromium.org/113823003/diff/1/src/core/SkBitmap.cpp#newcode43
src/core/SkBitmap.cpp:43: int64_t size = (levelCount + 1) * sizeof(MipLevel);
also, mixing signed and unsigned (you're trying to determine a size_t here) is a
pain. Why not uint64_t? This calculation cannot possibly result in a negative
value.

https://codereview.chromium.org/113823003/diff/1/src/core/SkBitmap.cpp#newcode45
src/core/SkBitmap.cpp:45: if (!sk_64_isS32(size)) {
This is what I wrote SkTFitsIn for (it handles all such cases). Also, this
should be against size_t since that's what it's being used for.

if (!SkTFitsIn<size_t>(size)) {

[reed1, 2013-12-30 14:14:50]

Committed patchset #2 manually as r12840 (presubmit successful).

[reed1, 2013-12-30 14:24:17]

A revert of this CL has been created in
https://codereview.chromium.org/122283002/ by reed@google.com.

The reason for reverting is: need to update callsites in linux codecs.