Fix type punning error in (de)serializing audio frames.

Fix type punning error in (de)serializing audio frames.

*(reinterpret_cast<Foo*>(pointer_of_type_bar)) is not safe. Replace it with memcpy.

TBR=viettrungluu@chromium.org
BUG=none
TEST=CDM decrypt-and-decode audio is still working.

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=164555

[xhwang, 2012-10-26 18:53:50]

PTAL, Thanks!

[Ami GONE FROM CHROMIUM, 2012-10-26 21:48:44]

LGTM % nit

http://codereview.chromium.org/11304010/diff/1/webkit/media/crypto/ppapi/ffmp...
File webkit/media/crypto/ppapi/ffmpeg_cdm_audio_decoder.cc (right):

http://codereview.chromium.org/11304010/diff/1/webkit/media/crypto/ppapi/ffmp...
webkit/media/crypto/ppapi/ffmpeg_cdm_audio_decoder.cc:336: memcpy(data, &value,
sizeof(value));
You could alternatively extend the vector by sizeof(value) and memcpy directly
into the vector's data()+prev_size.

[xhwang, 2012-10-26 22:28:12]

http://codereview.chromium.org/11304010/diff/1/webkit/media/crypto/ppapi/ffmp...
File webkit/media/crypto/ppapi/ffmpeg_cdm_audio_decoder.cc (right):

http://codereview.chromium.org/11304010/diff/1/webkit/media/crypto/ppapi/ffmp...
webkit/media/crypto/ppapi/ffmpeg_cdm_audio_decoder.cc:336: memcpy(data, &value,
sizeof(value));
On 2012/10/26 21:48:44, Ami Fischman wrote:
> You could alternatively extend the vector by sizeof(value) and memcpy directly
> into the vector's data()+prev_size.

Done.

[commit-bot: I haz the power, 2012-10-26 22:30:53]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/xhwang@chromium.org/11304010/7001

[commit-bot: I haz the power, 2012-10-27 06:21:34]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/xhwang@chromium.org/11304010/7001

[commit-bot: I haz the power, 2012-10-27 06:31:17]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/xhwang@chromium.org/11304010/5003

[commit-bot: I haz the power, 2012-10-27 12:24:26]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/xhwang@chromium.org/11304010/5003

[commit-bot: I haz the power, 2012-10-27 12:25:47]

Failed to apply patch for webkit/media/crypto/ppapi/clear_key_cdm.cc:
While running patch -p1 --forward --force --no-backup-if-mismatch;
  patching file webkit/media/crypto/ppapi/clear_key_cdm.cc
  Hunk #1 FAILED at 494.
  1 out of 1 hunk FAILED -- saving rejects to file
webkit/media/crypto/ppapi/clear_key_cdm.cc.rej

Patch:       webkit/media/crypto/ppapi/clear_key_cdm.cc
Index: webkit/media/crypto/ppapi/clear_key_cdm.cc
diff --git a/webkit/media/crypto/ppapi/clear_key_cdm.cc
b/webkit/media/crypto/ppapi/clear_key_cdm.cc
index
83cf00c8f8e58e447d61a817f64f224cd75d8a64..4d46dacf14f1882fb7316f3fa43a32044f135675
100644
--- a/webkit/media/crypto/ppapi/clear_key_cdm.cc
+++ b/webkit/media/crypto/ppapi/clear_key_cdm.cc
@@ -494,9 +494,12 @@ void ClearKeyCdm::GenerateFakeAudioFrames(cdm::AudioFrames*
audio_frames) {
 
   const int kHeaderSize = sizeof(timestamp) + sizeof(frame_size);
   audio_frames->set_buffer(allocator_->Allocate(kHeaderSize + frame_size));
-  int64* data = reinterpret_cast<int64*>(audio_frames->buffer()->data());
-  *(data++) = timestamp;
-  *(data++) = frame_size;
+  uint8_t* data = audio_frames->buffer()->data();
+
+  memcpy(data, &timestamp, sizeof(timestamp));
+  data += sizeof(timestamp);
+  memcpy(data, &frame_size, sizeof(frame_size));
+  data += sizeof(frame_size);
   // You won't hear anything because we have all zeros here. But the video
   // should play just fine!
   memset(data, 0, frame_size);

[commit-bot: I haz the power, 2012-10-27 16:34:49]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/xhwang@chromium.org/11304010/2004

[commit-bot: I haz the power, 2012-10-27 18:35:12]

Change committed as 164555