This was only used to mark executable as non-PIE. After the non-PIE Chromium
helper was removed, the last remaining use was for Valgrind. But Valgrind
support was also recently removed, so this can go away entirely.
change_mach_o_flags.py did explicitly set the MH_NO_HEAP_EXECUTION flag on
executables, but this is not necessary on x86_64 where the architecture default
is to not have an executable heap.
BUG=520680
Review URL: https://codereview.chromium.org/1885593003
Cr-Original-Commit-Position: refs/heads/master@{#387007}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: 806f385a1f9c2814de91bd1a98a50e74c4fa9a81
Reason for revert:
gn binary as been fixed and create_bundle now causes its data_deps to be built, reverting this temporary change.
Original issue's description:
> Work around a bug where create_bundle() is ignoring data_deps() in iOS GN build.
>
> It looks like create_bundle() is not checking for data_deps when it
> generates its ninja rules, and as a result we don't build iossim
> when we build base_unittests.app. By shifting the data_dep to the
> underlying executable target, this works around the issue, but we
> should probably fix the GN code generator for create_bundle() instead.
>
> R=sdefresne@chromium.org
> BUG=599203
>
> Committed: https://crrev.com/cd03fa72212077a789679a77992529c3aa6c08fe
> Cr-Commit-Position: refs/heads/master@{#384141}
TBR=dpranke@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=599203
Review URL: https://codereview.chromium.org/1883843002
Cr-Original-Commit-Position: refs/heads/master@{#386994}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: d76422e697ad4de2edac5b23f12da1ee5e9dcd56
Components such as cc/ and gpu/ that depend on ui::LatencyInfo
should not pull in all of ui/events.
This CL pulls ui::LatencyInfo into a separate component to
simplify this dependency. cc/ and gpu/ no longer depend on
ui/events.
BUG=601871
TBR=jam@chromium.org for top level BUILD.gn and build/all.gyp
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel
Committed: https://crrev.com/168cca4ea0abb22aa05575ec94c35d429dc08993
Cr-Commit-Position: refs/heads/master@{#386889}
Review URL: https://codereview.chromium.org/1871233002
Cr-Original-Commit-Position: refs/heads/master@{#386992}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: 7cc1dcccd28336397a468d1d5caa3be92675b456
Overview:
Browser process hangs are believed to be responsible for a large fraction of browser unclean shutdowns. This patch is to get visibility into the problem on windows using the Kasko reporter.
Note:
This functionality is hidden behind buildflags, and only triggers for users with stats collection consent on canary/unspecified channels. At this time, it is only considered for one-off releases.
Details:
- Use of the Kasko reporter is required for its ability to capture reports from other processes.
- Pulls out Kasko reporting code to kasko_utils
- Invoke Kasko on failed rendez-vous.
- Hidden behind build flags (kasko and kasko-rdv-capture).
BUG=478209
Review URL: https://codereview.chromium.org/1844023002
Cr-Original-Commit-Position: refs/heads/master@{#386983}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: f071e88974318cec29c02bd5b92fe61868da8569
MSVC 2015's linker regressed incremental linking with clang, so turn it off
for now.
Going forward, hopefully we can get that fixed, or we can switch to
linking with lld.
BUG=595702
TBR=dpranke
Review URL: https://codereview.chromium.org/1884973002 .
Cr-Original-Commit-Position: refs/heads/master@{#386978}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: 9f230704fde72e9ae4a7a023b6e6c3d8666e920f
Add support for copying public headers to the ios_framework_bundle
template and use this to implement the //ios/third_party/ochamcrest
target.
Add a set_defaults for ios_framework_bundle and mac_framework_bundle
to build/config/BUILDCONFIG.gn so that overriding "configs" works as
expected.
BUG=599322
Review URL: https://codereview.chromium.org/1879493002
Cr-Original-Commit-Position: refs/heads/master@{#386957}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: 8fab21c694073ca4db23495b715770385c597e1a
Fix the configuration of libjpeg_turbo to build on iOS by using the
same settings as Mac. This is not used by Chrome on iOS AFAIK, but
appears to be used by WebRTC.
BUG=602682
Review URL: https://codereview.chromium.org/1886443002
Cr-Original-Commit-Position: refs/heads/master@{#386941}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: cb3f4114b1b74302570e17e09b59b097d6ee406b
Reason for revert:
Highly suspected of breaking Win GN build. More details in http://crbug.com/601871#c2.
Today's sheriff
Original issue's description:
> Move LatencyInfo to ui/latency_info
>
> Components such as cc/ and gpu/ that depend on ui::LatencyInfo
> should not pull in all of ui/events.
>
> This CL pulls ui::LatencyInfo into a separate component to
> simplify this dependency. cc/ and gpu/ no longer depend on
> ui/events.
>
> BUG=601871
> TBR=jam@chromium.org for top level BUILD.gn and build/all.gyp
> CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel
>
> Committed: https://crrev.com/168cca4ea0abb22aa05575ec94c35d429dc08993
> Cr-Commit-Position: refs/heads/master@{#386889}
TBR=tdresser@chromium.org,sky@chromium.org,piman@chromium.org,dcheng@chromium.org,jam@chromium.org,fsamuel@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=601871
Review URL: https://codereview.chromium.org/1883763002
Cr-Original-Commit-Position: refs/heads/master@{#386930}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: ab57159df92cf956577149ce8d5c482e4c53a01b
Reason for revert:
Speculative revert: this breaks page_cycler.intl_hi_ru and page_cycler.intl_es_fr_pt-BR benchmarks
Original issue's description:
> Update blimp build bot args match build/args/blimp_engine.gn
>
> Reorder args so it's easier to spot differences.
>
> Remove gyp_defines because blimp is not designed to build with gyp.
>
> BUG=602696
>
> Committed: https://crrev.com/55bf089c901fb38204b642edc847b6259697ee9e
> Cr-Commit-Position: refs/heads/master@{#386760}
TBR=dpranke@chromium.org,maniscalco@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=602696
Review URL: https://codereview.chromium.org/1881343002
Cr-Original-Commit-Position: refs/heads/master@{#386912}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: 76edf5cc66829c9a2aed30dab102de2cc16da1c7
Components such as cc/ and gpu/ that depend on ui::LatencyInfo
should not pull in all of ui/events.
This CL pulls ui::LatencyInfo into a separate component to
simplify this dependency. cc/ and gpu/ no longer depend on
ui/events.
BUG=601871
TBR=jam@chromium.org for top level BUILD.gn and build/all.gyp
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel
Review URL: https://codereview.chromium.org/1871233002
Cr-Original-Commit-Position: refs/heads/master@{#386889}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: 168cca4ea0abb22aa05575ec94c35d429dc08993
Instead, pass the --plist to tweak on the command line. This way the
script can be used in both GYP and GN builds.
There is one behavior change: previously passing --keystone=1 would only
insert the Keystone keys if env['CONFIGURATION']=='Release' and after this
change, the keys will be added regardless of the build configuration if the
flag is passed.
BUG=297668
Review URL: https://codereview.chromium.org/1881053002
Cr-Original-Commit-Position: refs/heads/master@{#386854}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: adf34911dcb45fab4d7e13c4d7ef43e3dea840eb
This adds the final changes to cert_verify_proc_ios and various tests
in order to prepare for the switchover from NSS to BoringSSL on iOS.
BUG=591545
Review URL: https://codereview.chromium.org/1871043003
Cr-Original-Commit-Position: refs/heads/master@{#386819}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: b7f886b3d364497ff0422e9808d8ad04ffcb5c4d
Reason for revert:
This breaks win_clang
Original issue's description:
> build: Enable complex class copy ctor check by default.
>
> This patch ensures that our FindBadConstructs plugin starts detecting
> when a complex class requires an out of line copy constructor.
>
> R=thakis@chromium.org, dcheng@chromium.org
> BUG=436357
>
> Committed: https://crrev.com/319a17cb9761369818c8d8405da8d65c0a8bbc22
> Cr-Commit-Position: refs/heads/master@{#386787}
TBR=jochen@chromium.org,dcheng@chromium.org,thakis@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=436357
Review URL: https://codereview.chromium.org/1880243002
Cr-Original-Commit-Position: refs/heads/master@{#386794}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: 6a8849450d8ba833451921eee4adb2b1a1f4d84b
Reorder args so it's easier to spot differences.
Remove gyp_defines because blimp is not designed to build with gyp.
BUG=602696
Review URL: https://codereview.chromium.org/1878273003
Cr-Original-Commit-Position: refs/heads/master@{#386760}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: 55bf089c901fb38204b642edc847b6259697ee9e
Most output with annotations look like:
Class member attributes (count = 2):
- Deprecated attribute
- Runtime visible annotations attribute:
- Annotation [Ljava/lang/Deprecated;]:
- Method: <init>(Ljava/security/cert/X509Certificate;)V
But it was choking when it hit:
Class member attributes (count = 2):
- Runtime invisible annotations attribute:
- Annotation [Landroid/support/annotation/Nullable;]:
- Signature attribute:
- Utf8 [()Ljava/util/List<Ljava/lang/String;>;]
The code was treating - Utf8 as an annotation-related line.
BUG=525035
Review URL: https://codereview.chromium.org/1869083002
Cr-Original-Commit-Position: refs/heads/master@{#386706}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: 44a0b274c777c0a6aee6f5ba722c435e71d54273
This uses the new output_prefix_override flag to not prefix framework
shared libraries with "lib" and overrides the output_extension to be
empty.
BUG=297668
Review URL: https://codereview.chromium.org/1871143002
Cr-Original-Commit-Position: refs/heads/master@{#386692}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: c7905381e103462334ea24c20f1de670d89bd1ec
Extract commonalities to framework_bundle and rename mac_framework to
mac_framework_bundle for consistency & introduce ios_framework_bundle.
Rename app template from build/config/ios/rules.gni to ios_app_bundle
for consistency.
BUG=None
Review URL: https://codereview.chromium.org/1871413002
Cr-Original-Commit-Position: refs/heads/master@{#386673}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: 1a969442c07ee8642074b4f204cdfb4c9e630995
Ensure that ozone_platform_headless is the only ozone platform compiled
in the sample headless configuration. Other ozone platforms are included
by default and aren't necessarily desired.
BUG=none
Review URL: https://codereview.chromium.org/1876273002
Cr-Original-Commit-Position: refs/heads/master@{#386672}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: 7dfdf188b6c9bfc4e208949a1be8af557c0264b0
The gyp build uses -gdwarf-2 debug level so change GN to also use
that level instead of -g2 when targetting iOS.
BUG=None
Review URL: https://codereview.chromium.org/1873133002
Cr-Original-Commit-Position: refs/heads/master@{#386656}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: fa0530bfbfe97a969c23a00cb56d4f8d409c5eb9
VS 2015's warning about 32-bit shifts that are then assigned to a
64-bit target fires more frequently than in VS 2013. This type of
code triggers it:
int64_t size = 1 << shift_amount;
Because the '1' being shifted is a 32-bit int the result of the shift
will be a 32-bit result, so assigning it to a 64-bit variable is just
misleading and can lead to undefined behavior and lost bits.
During the port to VS 2015 this warning was globally suppressed. This
removes that global suppression.
There were ~nine C4334 warnings in Chromium with most of them being in
external repos such as pdfium, skia, webrtc, and angle. The external
repos have all been fixed. This fixes the Chromium repo warnings and
enables C4334 in gn builds.
In these cases the code that triggers it was assigning to a size_t
so it only showed up on 64-bit builds. In some cases there was already a
cast but it was after the shift, which is not as good as before the
shift. In one case the 32-bit constant was completely superfluous.
The Chromium specific warnings were:
net\base\mime_sniffer_perftest.cc(93)
third_party\webkit\source\platform\heap\heappage.cpp(738)
net\spdy\hpack\hpack_huffman_table.cc(172)
media\filters\vp8_parser.cc(157)
warning C4334: '<<': result of 32-bit shift implicitly converted to 64 bits
BUG=593448
Review URL: https://codereview.chromium.org/1858423002
Cr-Original-Commit-Position: refs/heads/master@{#386480}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: c8615da7ebb48668de21b3e23edec64c1685ffbb
Disable ozone auto platform selection for blimp engine builds. Ensure
blimp does not have a dependency on X11.
BUG=602312
Review URL: https://codereview.chromium.org/1881493003
Cr-Original-Commit-Position: refs/heads/master@{#386469}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: 83282ff073fe045d06df62468e9683db17f18e8c
This change will hopefully allow us to configure devil more based
on the build config. For example, it will let us use the same
android_sdk_tools to run tests that we use to build with.
BUG=
Review URL: https://codereview.chromium.org/1812383003
Cr-Original-Commit-Position: refs/heads/master@{#386461}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: c2aa4243a9e00a2d0e254d9f6f73b41c00cb644d
Reason for reland:
The original CL was reverted by crrev.com/1881523003 because it broke
clang builds. This CL contains the fix for clang (see diff between
patch-set 1 and 2).
The culprit of the fix is that LLVM libc++ uses _NOEXCEPT instead
of glibc's __THROW. The absence of it was causing errors like the
following:
../../base/allocator/allocator_shim_override_cpp_symbols.h:20:25: error: function previously declared with an explicit exception specification redeclared with an implicit exception specification [-Werror,-Wimplicit-exception-spec-mismatch]
SHIM_ALWAYS_EXPORT void operator delete(void* p) __THROW
Original issue's description:
> Enable allocator shim for Android
>
> This is a follow-up to crrev.com/1719433002, which introduced the
> shim for Android, and enables it by default by setting
> use_experimental_allocator_shim=true for Android.
>
> Build/Perf sheriffs heads up
> ----------------------------
> If you see any build error or crash related with __wrap_malloc,
> __wrap_free, __real_malloc, __real_free, etc this CL is to blame.
>
> Performance considerations
> ------------------------
> Binary size diff (GN, arm, static, official build): 24k
>
> I did a mixture of local and trybots run to estimate the perf impact
> of this change. Didn't get any conclusive data, everything I tried
> seems in the same ballpark, below noise levels. More in details:
>
> cc_perftests.PrepareTiles on a Nexus 4.
> Rationale of the choice: in a previous CL (crbug.com/593344), this
> benchmark revealed the presence of two mfences in the malloc path.
> Results: https://goo.gl/8VC3Jp in the same ballpark.
>
> page-cycler on Nexus 9 via trybots:
> Results: http://goo.gl/J3i50a seems to suggest that this CL improves
> both warm and cold times in most cases. I doubt it, more likely it's
> noise.
>
> All the other perf trybots failed. The perf waterfall seems to be in a
> bad state in these days.
>
> BUG=550886,598075
> TEST=base_unittests --gtest_filter=AllocatorShimTest.*
> TBR=thakis@chromium.org
>
> Committed: https://crrev.com/ebb95496c73dc0d5ce83968ac619921f154305f7
> Cr-Commit-Position: refs/heads/master@{#386386}
BUG=550886,598075
CC=dalecurtis@chromium.org
Review URL: https://codereview.chromium.org/1875173002
Cr-Original-Commit-Position: refs/heads/master@{#386444}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: 88bf797383cd290d9bd23db0045c5fb23e010beb
Icecc is confused by the two-argument "-target x" form, and decides to compile
everything locally. --target=x makes it happy.
Review URL: https://codereview.chromium.org/1871813003
Cr-Original-Commit-Position: refs/heads/master@{#386419}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: e4199397e4c6f32cf6a561e33581d59b5a7a0e9e
Reason for revert:
Looks like this breaks android clang builds.
Original issue's description:
> Enable allocator shim for Android
>
> This is a follow-up to crrev.com/1719433002, which introduced the
> shim for Android, and enables it by default by setting
> use_experimental_allocator_shim=true for Android.
>
> Build/Perf sheriffs heads up
> ----------------------------
> If you see any build error or crash related with __wrap_malloc,
> __wrap_free, __real_malloc, __real_free, etc this CL is to blame.
>
> Performance considerations
> ------------------------
> Binary size diff (GN, arm, static, official build): 24k
>
> I did a mixture of local and trybots run to estimate the perf impact
> of this change. Didn't get any conclusive data, everything I tried
> seems in the same ballpark, below noise levels. More in details:
>
> cc_perftests.PrepareTiles on a Nexus 4.
> Rationale of the choice: in a previous CL (crbug.com/593344), this
> benchmark revealed the presence of two mfences in the malloc path.
> Results: https://goo.gl/8VC3Jp in the same ballpark.
>
> page-cycler on Nexus 9 via trybots:
> Results: http://goo.gl/J3i50a seems to suggest that this CL improves
> both warm and cold times in most cases. I doubt it, more likely it's
> noise.
>
> All the other perf trybots failed. The perf waterfall seems to be in a
> bad state in these days.
>
> BUG=550886,598075
> TEST=base_unittests --gtest_filter=AllocatorShimTest.*
> TBR=thakis@chromium.org
>
> Committed: https://crrev.com/ebb95496c73dc0d5ce83968ac619921f154305f7
> Cr-Commit-Position: refs/heads/master@{#386386}
TBR=dalecurtis@chromium.org,primiano@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=550886,598075
Review URL: https://codereview.chromium.org/1881523003
Cr-Original-Commit-Position: refs/heads/master@{#386408}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: 6fdf2cfe0228e221c2cd738edabf3514459d6a89
This is a follow-up to crrev.com/1719433002, which introduced the
shim for Android, and enables it by default by setting
use_experimental_allocator_shim=true for Android.
Build/Perf sheriffs heads up
----------------------------
If you see any build error or crash related with __wrap_malloc,
__wrap_free, __real_malloc, __real_free, etc this CL is to blame.
Performance considerations
------------------------
Binary size diff (GN, arm, static, official build): 24k
I did a mixture of local and trybots run to estimate the perf impact
of this change. Didn't get any conclusive data, everything I tried
seems in the same ballpark, below noise levels. More in details:
cc_perftests.PrepareTiles on a Nexus 4.
Rationale of the choice: in a previous CL (crbug.com/593344), this
benchmark revealed the presence of two mfences in the malloc path.
Results: https://goo.gl/8VC3Jp in the same ballpark.
page-cycler on Nexus 9 via trybots:
Results: http://goo.gl/J3i50a seems to suggest that this CL improves
both warm and cold times in most cases. I doubt it, more likely it's
noise.
All the other perf trybots failed. The perf waterfall seems to be in a
bad state in these days.
BUG=550886,598075
TEST=base_unittests --gtest_filter=AllocatorShimTest.*
TBR=thakis@chromium.org
Review URL: https://codereview.chromium.org/1875043003
Cr-Original-Commit-Position: refs/heads/master@{#386386}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: ebb95496c73dc0d5ce83968ac619921f154305f7
This is a follow-up to the work of crrev.com/1675143004 and ports the
shim layer to Android.
Conversely to Linux, on Android symbol interposition is not possible.
This is because Android processes are fork-ed from the Android zygote,
which pre-loads libc.so, and later native code gets dlopen-ed.
Therefore a different approach is used on Android, that is:
- All linker units that depend on //base:allocator (via //base)
get the --Wl,-wrap,malloc (& friends) linker flag.
- Such linker flags causes all symbol references to malloc to be
rewritten as references to __wrap_malloc.
- The __wrap_malloc & friends symbols are defined by
allocator_shim_override_linker_wrapped_symbols.h. These symbols
route the malloc calls inside the shim layer.
- The shim layer ultimately dispatches the calls to the __real_malloc
symbol.
- The special __real_malloc symbols are resolved by the linker (as a
result of -wrap) against what would have been the original "malloc"
symbol.
In summary, this entire trickery is transparent to the dynamic loader,
which still sees undefined symbol references to malloc & friend symbols.
Those symbols will be resolved against libc.so (or whatever other weird
choice the Android OEM made) as usual.
An alternative approach would have been just redefining the malloc symbols
and use dlsym() to do the routing to bionic. This has the following
disadvantages:
- More risky, as Chrome for Android has insane layers of complexity around
loading (crazylinker pre M, relies on a special flavour of dlopen on M+).
- Won't be possible to support component builds, as all components
% libbase.so would resolve malloc against libc.so (unless violating ODR
and redefining malloc & friends in every component, which smells bad++).
All this trickery is a noop when use_experimental_allocator_shim=false.
This CL does NOT enable the shim by default on Android. This will happen
separately in crrev.com/1875043003
Motivations of this work
------------------------
- This allows the memory-infra heap-profiler to work on Android
(design doc: https://goo.gl/UPfbF4).
- This enables security checks (suicide on malloc()==null via
the std::new_handler) on Android.
- This fixes by accident crbug.com/598075 (Some Android device vendor
forgot to implement posix_memalign on J, as mandated by the NDK,
causing chrome to crash on startup).
Performance considerations
------------------------
Binary size diff (GN, arm, static, official build): 24k
I did a mixture of local and trybots run to estimate the perf impact
of this change. Didn't get any conclusive data, everything I tried
seems in the same ballpark, below noise levels. More in details:
cc_perftests.PrepareTiles on a Nexus 4.
Rationale of the choice: in a previous CL (crbug.com/593344), this
benchmark revealed the presence of two mfences in the malloc path.
Results: https://goo.gl/8VC3Jp in the same ballpark.
page-cycler on Nexus 9 via trybots:
Results: http://goo.gl/J3i50a seems to suggest that this CL improves
both warm and cold times in most cases. I doubt it, more likely it's
noise.
All the other perf trybots failed. The perf waterfall seems to be in a
bad state in these days.
BUG=550886,598075
TEST=base_unittests --gtest_filter=AllocatorShimTest.*
Review URL: https://codereview.chromium.org/1719433002
Cr-Original-Commit-Position: refs/heads/master@{#386382}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: 40b65c66b855b58cb2fda9228b7626f2904581a5
Stop bundling it in the installer (FILES.cfg). Remove the references
to the crash_service target from the perf builders' targets. (Issue
601762)
It's necessary to proceed with the removal of the start_crash_service
step on the bots, because putting it back now breaks the GPU bots
(Issue 601839).
The BUILD.gn removals are proactive, and not actually used yet.
BUG=601762, 601839
Review URL: https://codereview.chromium.org/1875613004
Cr-Original-Commit-Position: refs/heads/master@{#386239}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: 4cc1754d4ddb33f39dd9fe9cc9518f78bd840091
Lint's stderr filter actually was dropping all output of the tool
instead of just single 'Picking up _JAVA_OPTIONS' line. Fixing this
revealed a lot of internal lint errors.
Most of them were CRC errors from interface JARs, so I switched back to
use normal JARs.
Others were internal errors of InvalidPackageName check which was trying
to check Android SDK classes, because these classes were added to
classpath as a jar. It turns out that Lint treats SDK jar differently but
only if it is properly specified. The only way to tell Lint which SDK
jar to use is to create project.properties file (like in old Ant-based
build). So lint.py now accepts SDK version as a command-line argument
and creates dummy project.properties in temporary folder.
BUG=599052
Review URL: https://codereview.chromium.org/1870073002
Cr-Original-Commit-Position: refs/heads/master@{#386188}
Cr-Mirrored-From: https://chromium.googlesource.com/chromium/src
Cr-Mirrored-Commit: ef5e20f9fe84651ca03fb53ed15910121cedfc0a
svaldez
maniscalco
rsesek
sdefresne
fsamuel
manzagop
Nico Weber
dgn
mmoroz
vabr
ben
nednguyen
agrieve
bcf
vmpstr
forney
bpastene
kylechar
kjellander
pcc
brucedawson
mikecase
primiano
tsniatowski
kbr
cwallez
mostynb
mlopatkin
hans