|
|
Created:
8 years, 1 month ago by Paweł Hajdan Jr. Modified:
8 years, 1 month ago CC:
chromium-reviews Base URL:
svn://svn.chromium.org/chrome/trunk/src Visibility:
Public. |
DescriptionFix skia build when not on armv7.
BUG=none
TEST=none
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=165400
Patch Set 1 #
Total comments: 1
Patch Set 2 : cleanup #
Total comments: 1
Patch Set 3 : fixes #Patch Set 4 : fix build #Messages
Total messages: 14 (0 generated)
https://codereview.chromium.org/11342016/diff/1/skia/skia.gyp File skia/skia.gyp (right): https://codereview.chromium.org/11342016/diff/1/skia/skia.gyp#newcode754 skia/skia.gyp:754: '../third_party/skia/src/opts/opts_check_arm.cpp', Not your fault, but this thing is a growing mess. I would prefer that all of the armv7-specific files on line 725 were made conditional on armv7 (and the _none.cpp files were added in a new 'target_arch == "arm" and armv7 == 0 clause below it), rather than adding the _arm files there and then excluding them here. Could you take care of that?
PTAL
https://codereview.chromium.org/11342016/diff/3/skia/skia.gyp File skia/skia.gyp (right): https://codereview.chromium.org/11342016/diff/3/skia/skia.gyp#newcode750 skia/skia.gyp:750: '../third_party/skia/src/opts/opts_check_arm.cpp', No, this doesn't seem right for armv7. You're excluding the files here, but never adding them above. Let me try adding a patch to show you what I mean.
On 2012/10/29 22:16:35, Stephen White wrote: > https://codereview.chromium.org/11342016/diff/3/skia/skia.gyp > File skia/skia.gyp (right): > > https://codereview.chromium.org/11342016/diff/3/skia/skia.gyp#newcode750 > skia/skia.gyp:750: '../third_party/skia/src/opts/opts_check_arm.cpp', > No, this doesn't seem right for armv7. You're excluding the files here, but > never adding them above. > > Let me try adding a patch to show you what I mean. Here's a draft patch (untested). diff --git a/skia/skia.gyp b/skia/skia.gyp index 747427c..8e7dfac 100644 --- a/skia/skia.gyp +++ b/skia/skia.gyp @@ -721,6 +721,15 @@ 'cflags': [ '-fomit-frame-pointer', ], + }], + [ 'target_arch == arm' and 'armv7 == 0', { + 'sources': [ + '../third_party/skia/src/opts/SkBitmapProcState_opts_none.cpp', + '../third_party/skia/src/opts/SkBlitRow_opts_none.cpp', + '../third_party/skia/src/opts/SkBlitRow_opts_none.h', + ], + }], + [ 'armv7 == 1', { 'sources': [ '../third_party/skia/src/opts/SkBitmapProcState_opts_arm.cpp', '../third_party/skia/src/opts/SkBlitRow_opts_arm.cpp', @@ -744,14 +753,6 @@ '../third_party/skia/src/opts/SkBlitRow_opts_arm_neon.cpp', ], }], - [ 'target_arch == "arm" and armv7 != 1', { - 'sources': [ - '../third_party/skia/src/opts/SkBlitRow_opts_none.cpp', - ], - 'sources!': [ - '../third_party/skia/src/opts/SkBlitRow_opts_arm.cpp', - ], - }], ], }, # For the same lame reasons as what is done for skia_opts, we have to Note that I actually turned off SkBitmapProcState_opts_arm.cpp for non-armv7 too, just for symmetry. If in fact it does work, you can turn it back on.
On 2012/10/29 22:31:21, Stephen White wrote: > On 2012/10/29 22:16:35, Stephen White wrote: > > https://codereview.chromium.org/11342016/diff/3/skia/skia.gyp > > File skia/skia.gyp (right): > > > > https://codereview.chromium.org/11342016/diff/3/skia/skia.gyp#newcode750 > > skia/skia.gyp:750: '../third_party/skia/src/opts/opts_check_arm.cpp', > > No, this doesn't seem right for armv7. You're excluding the files here, but > > never adding them above. > > > > Let me try adding a patch to show you what I mean. > > Here's a draft patch (untested). > > diff --git a/skia/skia.gyp b/skia/skia.gyp > index 747427c..8e7dfac 100644 > --- a/skia/skia.gyp > +++ b/skia/skia.gyp > @@ -721,6 +721,15 @@ > 'cflags': [ > '-fomit-frame-pointer', > ], > + }], > + [ 'target_arch == arm' and 'armv7 == 0', { Whoops, should be: > + [ 'target_arch == "arm" and armv7 == 0', {
Is this more like what you want? Is there any trybot I could test this on?
On 2012/10/29 23:17:19, Paweł Hajdan Jr. wrote: > Is this more like what you want? Yes, LGTM thanks. > Is there any trybot I could test this on? I think the android bots should take care of armv7 and higher. I don't think we have any non-armv7 bots, but I could be wrong.
Does anyone else want to comment on this change? I'd like to commit within say a day if there is no reply.
CQ is trying da patch. Follow status at https://chromium-status.appspot.com/cq/phajdan.jr@chromium.org/11342016/10002
Step "update" is always a major failure. Look at the try server FAQ for more details.
Note that this change will conflict with https://codereview.chromium.org/11292003 ('[MIPS] Add build support in Skia for MIPS.'). I fear the incentives associated with such a warning, but: whoever commits last will have to manually merge the changes. |