Merge into development-branch : fix-and-enable-lto : Code : Mir

Reviewer	Review Type	Date Requested	Status
Mir CI Bot	continuous-integration	2016-01-11	Approve on 2016-03-01
Alan Griffiths			Approve on 2016-02-29
Cemil Azizoglu (community)			Approve on 2016-02-26
PS Jenkins bot (community)	continuous-integration	2016-01-11	Needs Fixing on 2016-02-24
Andreas Pokorny (community)		2016-01-11	Approve on 2016-01-21
Review via email: mp+282133@code.launchpad.net

Revision history for this message

PS Jenkins bot (ps-jenkins) wrote on 2015-12-22: Posted in a previous version of this proposal

#

Click here to trigger a rebuild:
http://s-jenkins.ubuntu-ci:8080/job/mir-ci/5916/rebuild

review: Needs Fixing (continuous-integration)

Revision history for this message

Chris Halse Rogers (raof) wrote on 2015-12-22: Posted in a previous version of this proposal

#

/usr/include/i386-linux-gnu/bits/stat.h:46:8: error: type 'struct stat' violates one definition rule [-Werror=odr]
struct stat
        ^
/usr/include/i386-linux-gnu/bits/stat.h:46:0: note: a different type is defined in another translation unit
struct stat
^
/usr/include/i386-linux-gnu/bits/stat.h:53:13: note: the first difference of corresponding definitions is field 'st_ino'
     __ino_t st_ino; /* File serial number. */
             ^
/usr/include/i386-linux-gnu/bits/stat.h:55:13: note: a field with different name is defined in another translation unit
     __ino_t __st_ino; /* 32bit file serial number. */

Interesting... this seems to be a genuine error; we've included stat.h with twice, with different compile options, really giving us two different definitions of struct stat.

Revision history for this message

Andreas Pokorny (andreas-pokorny) wrote on 2015-12-22: Posted in a previous version of this proposal

#

hmm will gcc-nm and gcc-ar work for cross compilation?

review: Needs Information

Revision history for this message

Chris Halse Rogers (raof) wrote on 2015-12-23: Posted in a previous version of this proposal

#

That's a fair question. Looks like it works, which somewhat surprised me :).

Of course, this also means that a cross-compile doesn't hit the issue that hit mediumtests-builder-vivid-armhf. Of course.

Revision history for this message

Andreas Pokorny (andreas-pokorny) wrote on 2015-12-23: Posted in a previous version of this proposal

#

ah the job of nm and ar is more elf-ish, tar-ish and less-ish so they might not need to interfere with the actual target format

review: Approve

Revision history for this message

Alan Griffiths (alan-griffiths) wrote on 2016-01-08: Posted in a previous version of this proposal

#

Not sure why LP has this a "Merged" - it clearly isn't.

Revision history for this message

PS Jenkins bot (ps-jenkins) wrote on 2016-01-08: Posted in a previous version of this proposal

#

Click here to trigger a rebuild:
http://s-jenkins.ubuntu-ci:8080/job/mir-ci/5961/rebuild

review: Needs Fixing (continuous-integration)

Revision history for this message

PS Jenkins bot (ps-jenkins) wrote on 2016-01-08: Posted in a previous version of this proposal

#

Click here to trigger a rebuild:
http://s-jenkins.ubuntu-ci:8080/job/mir-ci/5964/rebuild

review: Needs Fixing (continuous-integration)

Revision history for this message

Mir CI Bot (mir-ci-bot) wrote on 2016-01-08: Posted in a previous version of this proposal

#

PASSED: Continuous integration, rev:3236
https://mir-jenkins.ubuntu.com/job/mir-ci/15/
Executed test runs:
None: https://mir-jenkins.ubuntu.com/job/generic-update-mp/15/console

Click here to trigger a rebuild:
https://mir-jenkins.ubuntu.com/job/mir-ci/15/rebuild

review: Approve (continuous-integration)

Revision history for this message

PS Jenkins bot (ps-jenkins) wrote on 2016-01-08: Posted in a previous version of this proposal

#

Click here to trigger a rebuild:
http://s-jenkins.ubuntu-ci:8080/job/mir-ci/5967/rebuild

review: Needs Fixing (continuous-integration)

Revision history for this message

PS Jenkins bot (ps-jenkins) wrote on 2016-01-11:

#

FAILED: Continuous integration, rev:3238
http://jenkins.qa.ubuntu.com/job/mir-ci/5973/
Executed test runs:
    SUCCESS: http://jenkins.qa.ubuntu.com/job/mir-android-vivid-i386-build/5478
    SUCCESS: http://jenkins.qa.ubuntu.com/job/mir-clang-vivid-amd64-build/4385
    FAILURE: http://jenkins.qa.ubuntu.com/job/mir-mediumtests-vivid-touch/5434/console
    SUCCESS: http://jenkins.qa.ubuntu.com/job/mir-xenial-amd64-ci/297
        deb: http://jenkins.qa.ubuntu.com/job/mir-xenial-amd64-ci/297/artifact/work/output/*zip*/output.zip
    SUCCESS: http://jenkins.qa.ubuntu.com/job/mir-xenial-i386-ci/297
        deb: http://jenkins.qa.ubuntu.com/job/mir-xenial-i386-ci/297/artifact/work/output/*zip*/output.zip
    SUCCESS: http://jenkins.qa.ubuntu.com/job/mir-mediumtests-builder-vivid-armhf/5431
        deb: http://jenkins.qa.ubuntu.com/job/mir-mediumtests-builder-vivid-armhf/5431/artifact/work/output/*zip*/output.zip
    FAILURE: http://jenkins.qa.ubuntu.com/job/mir-mediumtests-runner-touch/7917/console
    SUCCESS: http://s-jenkins.ubuntu-ci:8080/job/touch-flash-device/26530

Click here to trigger a rebuild:
http://s-jenkins.ubuntu-ci:8080/job/mir-ci/5973/rebuild

review: Needs Fixing (continuous-integration)

Revision history for this message

Mir CI Bot (mir-ci-bot) wrote on 2016-01-11:

#

PASSED: Continuous integration, rev:3238
https://mir-jenkins.ubuntu.com/job/mir-ci/23/
Executed test runs:
None: https://mir-jenkins.ubuntu.com/job/generic-update-mp/23/console

Click here to trigger a rebuild:
https://mir-jenkins.ubuntu.com/job/mir-ci/23/rebuild

review: Approve (continuous-integration)

Revision history for this message

Andreas Pokorny (andreas-pokorny) wrote on 2016-01-21:

#

https://jenkins.qa.ubuntu.com/job/mir-mediumtests-runner-touch/7917/consoleFull says

[1452488342.218071] mirserver: Selected driver: android (version 0.19.0)
ERROR: Dynamic exception type: St12system_error
std::exception::what: Enable multithreading to use std::thread: Operation not permitted

Do we need to clutter cmake with more pthread flags?

Apart from that. I can live with that fancy solution.

review: Approve

Revision history for this message

Alan Griffiths (alan-griffiths) wrote on 2016-01-22:

#

Apart from the android failure looks good.

Should we also force use of gold with LTO?

review: Needs Information

Revision history for this message

PS Jenkins bot (ps-jenkins) wrote on 2016-02-17:

#

FAILED: Continuous integration, rev:3238
http://jenkins.qa.ubuntu.com/job/mir-ci/6311/
Executed test runs:
    FAILURE: http://jenkins.qa.ubuntu.com/job/mir-android-vivid-i386-build/5952/console
    None: http://jenkins.qa.ubuntu.com/job/mir-clang-vivid-amd64-build/4859/console
    None: http://jenkins.qa.ubuntu.com/job/mir-mediumtests-vivid-touch/5908/console
    FAILURE: http://jenkins.qa.ubuntu.com/job/mir-xenial-amd64-ci/635/console
    FAILURE: http://jenkins.qa.ubuntu.com/job/mir-xenial-i386-ci/635/console
    FAILURE: http://jenkins.qa.ubuntu.com/job/mir-mediumtests-builder-vivid-armhf/5905/console

Click here to trigger a rebuild:
http://s-jenkins.ubuntu-ci:8080/job/mir-ci/6311/rebuild

review: Needs Fixing (continuous-integration)

Revision history for this message

Andreas Pokorny (andreas-pokorny) wrote on 2016-02-17:

#

There is now a conflict, when resolved there is a chance that the problem on vivid/armhf is gone.

Revision history for this message

PS Jenkins bot (ps-jenkins) wrote on 2016-02-17:

#

FAILED: Continuous integration, rev:3238
http://jenkins.qa.ubuntu.com/job/mir-ci/6312/
Executed test runs:
    FAILURE: http://jenkins.qa.ubuntu.com/job/mir-android-vivid-i386-build/5953/console
    FAILURE: http://jenkins.qa.ubuntu.com/job/mir-clang-vivid-amd64-build/4860/console
    FAILURE: http://jenkins.qa.ubuntu.com/job/mir-mediumtests-vivid-touch/5909/console
    FAILURE: http://jenkins.qa.ubuntu.com/job/mir-xenial-amd64-ci/636/console
    FAILURE: http://jenkins.qa.ubuntu.com/job/mir-xenial-i386-ci/636/console
    FAILURE: http://jenkins.qa.ubuntu.com/job/mir-mediumtests-builder-vivid-armhf/5906/console

Click here to trigger a rebuild:
http://s-jenkins.ubuntu-ci:8080/job/mir-ci/6312/rebuild

review: Needs Fixing (continuous-integration)

Revision history for this message

Mir CI Bot (mir-ci-bot) wrote on 2016-02-24:

#

Click here to trigger a rebuild:
https://mir-jenkins.ubuntu.com/job/mir-ci/382/rebuild

review: Needs Fixing (continuous-integration)

Revision history for this message

Mir CI Bot (mir-ci-bot) wrote on 2016-02-24:

#

Click here to trigger a rebuild:
https://mir-jenkins.ubuntu.com/job/mir-ci/383/rebuild

review: Needs Fixing (continuous-integration)

Revision history for this message

PS Jenkins bot (ps-jenkins) wrote on 2016-02-24:

#

FAILED: Continuous integration, rev:3340
http://jenkins.qa.ubuntu.com/job/mir-ci/6376/
Executed test runs:
    SUCCESS: http://jenkins.qa.ubuntu.com/job/mir-android-vivid-i386-build/6034
    SUCCESS: http://jenkins.qa.ubuntu.com/job/mir-clang-vivid-amd64-build/4941
    FAILURE: http://jenkins.qa.ubuntu.com/job/mir-mediumtests-vivid-touch/5990/console
    FAILURE: http://jenkins.qa.ubuntu.com/job/mir-xenial-amd64-ci/700/console
    FAILURE: http://jenkins.qa.ubuntu.com/job/mir-xenial-i386-ci/700/console
    FAILURE: http://jenkins.qa.ubuntu.com/job/mir-mediumtests-builder-vivid-armhf/5987/console

Click here to trigger a rebuild:
http://s-jenkins.ubuntu-ci:8080/job/mir-ci/6376/rebuild

review: Needs Fixing (continuous-integration)

Revision history for this message

Mir CI Bot (mir-ci-bot) wrote on 2016-02-25:

#

Click here to trigger a rebuild:
https://mir-jenkins.ubuntu.com/job/mir-ci/398/rebuild

review: Approve (continuous-integration)

Revision history for this message

Alan Griffiths (alan-griffiths) wrote on 2016-02-25:

#

+static_assert(
+ std::is_pointer<EGLNativeWindowType>::value ||
+ sizeof(EGLNativeWindowType) == sizeof(void*),
+ "Oh, oh. EGLNativeWindowType isn't losslessly convertible to void*.\n"
+ "The ClientBufferStream requires that EGLNativeWindowType be convertible to void*");

It seems odd that this is needed for the LTO MP and that it is needed in two headers (out of the five that #include <EGL/eglplatform.h>).

What's the reasoning for this approach and should we have a single wrapper header for all uses?

review: Needs Information

Revision history for this message

Chris Halse Rogers (raof) wrote on 2016-02-26:

#

Moved the static_assert() check to a more relevant header, and gave it a more explicit comment.

It's here because the LTO turned out to allow a check which found that we were returning different types from different virtual overrides, because EGLNativeWindowType is a different type on different EGL platforms.

Revision history for this message

Mir CI Bot (mir-ci-bot) wrote on 2016-02-26:

#

Click here to trigger a rebuild:
https://mir-jenkins.ubuntu.com/job/mir-ci/405/rebuild

review: Needs Fixing (continuous-integration)

Revision history for this message

Chris Halse Rogers (raof) wrote on 2016-02-26:

#

Ah, there it is.

Different layouts for BufferStream because EGLNativeWindowType differs...

Revision history for this message

Mir CI Bot (mir-ci-bot) wrote on 2016-02-26:

#

Click here to trigger a rebuild:
https://mir-jenkins.ubuntu.com/job/mir-ci/408/rebuild

review: Needs Fixing (continuous-integration)

Revision history for this message

Chris Halse Rogers (raof) wrote on 2016-02-26:

#

I'm going to disable LTO on the clang build; it's too difficult to debug with clang being all not working in recent releases.

Revision history for this message

Mir CI Bot (mir-ci-bot) wrote on 2016-02-26:

#

Click here to trigger a rebuild:
https://mir-jenkins.ubuntu.com/job/mir-ci/409/rebuild

review: Approve (continuous-integration)

Revision history for this message

Alan Griffiths (alan-griffiths) wrote on 2016-02-26:

#

+ -DMIR_LINK_TIME_OPTIMIZATION=ON

While this is probably right, we ought to be explicit about which builds we want this optimization for (as it can slow things quite a bit).

I'd suggest not for the "normal" CI and developer builds where fast turnaround is desirable.

I think problems will be rare enough that enabling for autolanding would be fine as a final check. And for release.

Is that what happens?

(It also suggests the question of any compile-time optimizations we might want to enable.)

review: Needs Information

Revision history for this message

Cemil Azizoglu (cemil-azizoglu) wrote on 2016-02-26:

#

> I think problems will be rare enough that enabling for autolanding would be
> fine as a final check. And for release.

+1

review: Approve

Revision history for this message

Chris Halse Rogers (raof) wrote on 2016-02-28:

#

Ok. I've gated MIR_LINK_TIME_OPTIMIZATION on the presence or absence of “noopt” in the DEB_BUILD_OPTIONS environment variable. This is a standard interface for Debian package builds.

I've also started to hook this up to CI, so that we pass DEB_BUILD_OPTIONS=noopt for normal builds (so we'll get slightly faster -O0 builds for normal CI) and not pass it for autolanding.

Sadly CI is currently broken, having run out of disc space, and I don't think I have the ability to fix it without a shell on that machine.

Revision history for this message

Alan Griffiths (alan-griffiths) wrote on 2016-02-29:

#

LGTM

review: Approve

Revision history for this message

Mir CI Bot (mir-ci-bot) wrote on 2016-02-29:

#

Click here to trigger a rebuild:
https://mir-jenkins.ubuntu.com/job/mir-ci/432/rebuild

FAILED: Continuous integration, rev:3350
https://mir-jenkins.ubuntu.com/job/mir-ci/432/
Executed test runs:
    FAILURE: https://mir-jenkins.ubuntu.com/job/build-mir/242/console
    SUCCESS: https://mir-jenkins.ubuntu.com/job/build-0-fetch/266
    SUCCESS: https://mir-jenkins.ubuntu.com/job/build-1-sourcepkg/release=vivid+overlay/258
    SUCCESS: https://mir-jenkins.ubuntu.com/job/build-1-sourcepkg/release=xenial/258
    SUCCESS: https://mir-jenkins.ubuntu.com/job/build-2-binpkg-mir/arch=amd64,compiler=clang,platform=mesa,release=vivid+overlay/249
        deb: https://mir-jenkins.ubuntu.com/job/build-2-binpkg-mir/arch=amd64,compiler=clang,platform=mesa,release=vivid+overlay/249/artifact/output/*zip*/output.zip
    SUCCESS: https://mir-jenkins.ubuntu.com/job/build-2-binpkg-mir/arch=amd64,compiler=gcc,platform=mesa,release=xenial/249
        deb: https://mir-jenkins.ubuntu.com/job/build-2-binpkg-mir/arch=amd64,compiler=gcc,platform=mesa,release=xenial/249/artifact/output/*zip*/output.zip
    SUCCESS: https://mir-jenkins.ubuntu.com/job/build-2-binpkg-mir/arch=cross-armhf,compiler=gcc,platform=android,release=vivid+overlay/249
        deb: https://mir-jenkins.ubuntu.com/job/build-2-binpkg-mir/arch=cross-armhf,compiler=gcc,platform=android,release=vivid+overlay/249/artifact/output/*zip*/output.zip
    SUCCESS: https://mir-jenkins.ubuntu.com/job/build-2-binpkg-mir/arch=i386,compiler=gcc,platform=android,release=vivid+overlay/249
        deb: https://mir-jenkins.ubuntu.com/job/build-2-binpkg-mir/arch=i386,compiler=gcc,platform=android,release=vivid+overlay/249/artifact/output/*zip*/output.zip
    FAILURE: https://mir-jenkins.ubuntu.com/job/build-2-binpkg-mir/arch=i386,compiler=gcc,platform=mesa,release=xenial/249/console
        deb: https://mir-jenkins.ubuntu.com/job/build-2-binpkg-mir/arch=i386,compiler=gcc,platform=mesa,release=xenial/249/artifact/output/*zip*/output.zip

Click here to trigger a rebuild:
https://mir-jenkins.ubuntu.com/job/mir-ci/432/rebuild

review: Needs Fixing (continuous-integration)

Revision history for this message

Mir CI Bot (mir-ci-bot) wrote on 2016-03-01:

#

Click here to trigger a rebuild:
https://mir-jenkins.ubuntu.com/job/mir-ci/452/rebuild

review: Approve (continuous-integration)

Revision history for this message

Mir CI Bot (mir-ci-bot) on 2016-03-01:

#

review: Approve (continuous-integration)

Mir

Merge lp:~raof/mir/fix-and-enable-lto into lp:mir

Commit message

Description of the change

Preview Diff

Subscribers

1	=== modified file 'CMakeLists.txt'
2	--- CMakeLists.txt 2016-02-25 17:58:32 +0000
3	+++ CMakeLists.txt 2016-02-28 23:45:57 +0000
4	@@ -34,6 +34,8 @@
5	add_definitions(-DMIR_VERSION_MAJOR=${MIR_VERSION_MAJOR})
6	add_definitions(-DMIR_VERSION_MINOR=${MIR_VERSION_MINOR})
7	add_definitions(-DMIR_VERSION_MICRO=${MIR_VERSION_PATCH})
8	+add_definitions(-D_GNU_SOURCE)
9	+add_definitions(-D_FILE_OFFSET_BITS=64)
10
11	set(MIR_VERSION ${MIR_VERSION_MAJOR}.${MIR_VERSION_MINOR}.${MIR_VERSION_PATCH})
12
13	@@ -81,6 +83,11 @@
14	if(MIR_LINK_TIME_OPTIMIZATION)
15	set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -flto")
16	set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -flto")
17	+ if(${CMAKE_COMPILER_IS_GNUCXX})
18	+ set(CMAKE_NM "gcc-nm")
19	+ set(CMAKE_AR "gcc-ar")
20	+ set(CMAKE_RANLIB "gcc-ranlib")
21	+ endif()
22	endif()
23
24	string(TOLOWER "${CMAKE_BUILD_TYPE}" cmake_build_type_lower)
25
26	=== modified file 'benchmarks/frame-uniformity/CMakeLists.txt'
27	--- benchmarks/frame-uniformity/CMakeLists.txt 2016-01-29 08:18:22 +0000
28	+++ benchmarks/frame-uniformity/CMakeLists.txt 2016-02-28 23:45:57 +0000
29	@@ -15,6 +15,12 @@
30	${PROJECT_SOURCE_DIR}/tests/include/
31	)
32
33	+# We use mir-test-{doubles,framework}, which builds differently based on
34	+# the primary test platform.
35	+if (MIR_TEST_PLATFORM STREQUAL "android")
36	+ add_definitions(-DANDROID)
37	+endif()
38	+
39	mir_add_wrapped_executable(frame_uniformity_test_client NOINSTALL
40	touch_measuring_client.cpp