roll to ffmpeg oct 10 version for chrome8... (issue3730003)

5 views
Skip to first unread message

fbar...@chromium.org

unread,
Oct 12, 2010, 1:12:49 PM10/12/10
to sche...@chromium.org, tomfi...@chromium.org, chromium...@chromium.org, v...@chromium.org, hc...@chromium.org, pam+...@chromium.org, ajw...@chromium.org, sche...@chromium.org
Reviewers: scherkus, tomf,

Message:
avformat/utils in chromium now requires id3v2 which was previously used
only in
chrome / chromeos
TBR=scherkus

Description:
roll to ffmpeg oct 10 version for chrome8
BUG=55399
TEST=layout tests should still pass

Please review this at http://codereview.chromium.org/3730003/show

SVN Base: svn://chrome-svn/chrome/trunk/src/

Affected files:
M DEPS


Index: DEPS
===================================================================
--- DEPS (revision 62287)
+++ DEPS (working copy)
@@ -11,7 +11,7 @@
"nacl_revision": "3365",
"libjingle_revision": "33",
"libvpx_revision": "60399",
- "ffmpeg_revision": "61805",
+ "ffmpeg_revision": "62291",
}

deps = {


tomfi...@chromium.org

unread,
Oct 13, 2010, 10:39:20 AM10/13/10
to fbar...@chromium.org, sche...@chromium.org, chromium...@chromium.org, v...@chromium.org, fbar...@chromium.org, hc...@chromium.org, pam+...@chromium.org, ajw...@chromium.org, sche...@chromium.org
On 2010/10/12 17:12:49, fbarchard wrote:
> avformat/utils in chromium now requires id3v2 which was previously used
> only
in
> chrome / chromeos
> TBR=scherkus

LGTM if the mac buildbot fail is unrelated (seems so...)

http://codereview.chromium.org/3730003/show

Frank Barchard

unread,
Oct 13, 2010, 12:51:26 PM10/13/10
to fbar...@chromium.org, sche...@chromium.org, tomfi...@chromium.org, chromium...@chromium.org, v...@chromium.org, hc...@chromium.org, pam+...@chromium.org, ajw...@chromium.org
ya, the roll went smoothly.  Which I think is a first!

Frank Barchard

unread,
Oct 13, 2010, 4:09:30 PM10/13/10
to fbar...@chromium.org, sche...@chromium.org, tomfi...@chromium.org, chromium...@chromium.org, v...@chromium.org, hc...@chromium.org, pam+...@chromium.org, ajw...@chromium.org
at least I think it went smoothly.  irc guy said he had an arm issue

15:26 <+kuan> fbarchard: r u there? ur r62300 broke the build for chromeos arm 
             with the following errors: 
15:28 <+kuan> third_party/ffmpeg/patched-ffmpeg-mt/libavcodec/arm/mpegvideo_arm.c:28: error: size of array 'x_Y_DC_SCALE' is negative
15:28 <+kuan> third_party/ffmpeg/patched-ffmpeg-mt/libavcodec/arm/mpegvideo_arm.c:29: error: size of array 'x_C_DC_SCALE' is negative
15:28 <+kuan> third_party/ffmpeg/patched-ffmpeg-mt/libavcodec/arm/mpegvideo_arm.c:30: error: size of array 'x_AC_PRED' is negative
15:28 <+kuan> third_party/ffmpeg/patched-ffmpeg-mt/libavcodec/arm/mpegvideo_arm.c:31: error: size of array 'x_BLOCK_LAST_INDEX' is negative
15:28 <+kuan> third_party/ffmpeg/patched-ffmpeg-mt/libavcodec/arm/mpegvideo_arm.c:32: error: size of array 'x_INTER_SCANTAB_RASTER_END' is negative
15:28 <+kuan> third_party/ffmpeg/patched-ffmpeg-mt/libavcodec/arm/mpegvideo_arm.c:33: error: size of array 'x_H263_AIC' is negative
15:28 <+kuan> make: * 
[c/Release/obj.target/ffmpegsumo/third_party/ffmpeg/patched-ffmpeg-mt/libavcodec/arm/mpegvideo_arm.o] Error 1

I just tested it and it built for chromium arm
  CC(target) out/Debug/obj.target/ffmpegsumo/third_party/ffmpeg/patched-ffmpeg-mt/libavutil/arm/cpu.o
  CC(target) out/Debug/obj.target/ffmpegsumo/third_party/ffmpeg/patched-ffmpeg-mt/libavcodec/arm/h264dsp_init_arm.o
  CC(target) out/Debug/obj.target/ffmpegsumo/third_party/ffmpeg/patched-ffmpeg-mt/libavcodec/arm/h264pred_init_arm.o
  CC(target) out/Debug/obj.target/ffmpegsumo/third_party/ffmpeg/patched-ffmpeg-mt/libavcodec/arm/mpegvideo_arm.o
  CC(target) out/Debug/obj.target/ffmpegsumo/third_party/ffmpeg/patched-ffmpeg-mt/libavcodec/arm/mpegvideo_armv5te.o
  CC(target) out/Debug/obj.target/ffmpegsumo/third_party/ffmpeg/patched-ffmpeg-mt/libavcodec/arm/mpegvideo_armv5te_s.o
  SOLINK(target) out/Debug/obj.target/third_party/ffmpeg/libffmpegsumo.so

he irc'ed back
(1:00:04 PM) kuan: fbarchard: pls ignore that, i was looking at the wrong waterfall, sorry abt that.

So I think its okay, but I cant fully test arm.
Reply all
Reply to author
Forward
0 new messages