- 30 Jul, 2016 2 commits
-
-
ben authored
(Mojo) Shell types have been renamed elsewhere in Chromium, specifically ServiceRegistry is split into two parts: - InterfaceRegistry, for registration of interface factories in code that implements the interfaces - InterfaceProvider, allowing binding of named interfaces in code that consumes them. This reflects this naming scheme into Blink, which uses only the Provider half. R=esprehn@chromium.org http://crbug.com/632874 Review-Url: https://codereview.chromium.org/2196843003 Cr-Commit-Position: refs/heads/master@{#408810}
-
fsamuel authored
This CL makes use of the newly introduced Array DataViews in cc StructTraits in order to improve serialization (and to a lesser degree deserialization) performance. This CL improves serialization performance by an average of 18% on an HP Z620. Deserialization performance improves between 7 to 10%. Prior to the ability to inspect Array and Struct DataViews in parent StructTraits, we had to do extra work at serialization time in order to capture state that spans individual structs within a larger struct. This CL addresses two of those cases: 1. Previously, QuadList maintained an extra array of materials. This array existed so that we could preallocate the appropriately typed DrawQuads during deserialization. StructTraits assume a preallocated type and populated fields in that type. Alternatively we could have deserialized DrawQuads into unique_ptrs, but cc required that we allocate DrawQuads inline within a buffer in order to minimize the cost of allocations (there can be hundreds of DrawQuads in a CompositorFrame). 2. Previously, SharedQuadStates were serialized independently of DrawQuads. However, in in-memory form, DrawQuads refer to SharedQuadStates via raw pointers. In order to allow DrawQuads and SharedQuadStates to deserialize independently, as was required previously by StructTraits, a third array was serialized "shared_quad_state_references" which was equal in size to the QuadList. Each slot corresponded to an index in the SharedQuadStateList. After the QuadList and SharedQuadStateList were serialized, shared_quad_state_references updated the raw pointers in the DrawQuads in the quad list to refer to the SharedQuadStates. This required revisiting all deserialized DrawQuads again. In both 1. and 2. additional state needed to be allocated on the heap during serialization. In pprof profiling, it became apparent that much of the cost of serialization was in SetupContext. This CL eliminates that cost. Performance measures are available here: https://docs.google.com/a/google.com/spreadsheets/d/13Y0I-BVDrQC4RHIzMSs0vl0QPe8xFAhYMUAXaxoEqgg/pubhtml BUG=624459 TBR=danakj@chromium.org CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel Review-Url: https://codereview.chromium.org/2174843003 Cr-Commit-Position: refs/heads/master@{#408809}
-
- 29 Jul, 2016 38 commits
-
-
japhet authored
This moves all of ResourceLoader's non-trivial response handling logic to ResourceFetcher, and changes the ordering of some steps to better match other callbacks. Old order: - Access control checks in ResourceLoader::didReceiveResponse() - Resource::responseReceived() - Access control checks in ResourceFetcher::didReceiveResponse() - FetchContext::dispatchDidReceiveResponse() New order: - Access control checks in ResourceLoader::didReceiveResponse() (now in ResourceFetcher) - Access control checks in ResourceFetcher::didReceiveResponse() - FetchContext::dispatchDidReceiveResponse() - Resource::responseReceived() BUG= Review-Url: https://codereview.chromium.org/1975373002 Cr-Commit-Position: refs/heads/master@{#408808}
-
pfeldman authored
TBR=dgozman Review-Url: https://codereview.chromium.org/2197823002 Cr-Commit-Position: refs/heads/master@{#408807}
-
caseq authored
Initially broken by https://codereview.chromium.org/2150783002 BUG=629530,630030 Review-Url: https://codereview.chromium.org/2193323002 Cr-Commit-Position: refs/heads/master@{#408806}
-
jyasskin authored
BUG=597178 Review-Url: https://codereview.chromium.org/2191323004 Cr-Commit-Position: refs/heads/master@{#408805}
-
kulshin authored
Previous change passed the sender to FontFallback init, but not DWriteFontCollectionProxy which is what really matters. This will ensure that both classes get the right sender. BUG=631254 Review-Url: https://codereview.chromium.org/2196493003 Cr-Commit-Position: refs/heads/master@{#408804}
-
estade authored
Side note: if we had just used .icon files, besides the massive proliferation of them that would have been necessary, we'd have worse rendering at fractional scales like 1.5x because we'd be filling fractional pixels. However, to actually get the better drawing at fractional scales, the network technology badges (LTE, G, etc.) will also have to be vectorized. BUG=617306 Review-Url: https://codereview.chromium.org/2201443002 Cr-Commit-Position: refs/heads/master@{#408803}
-
nyquist authored
The BlimpClientSession will be going away, so this CL takes part of it and splits it up into smaller parts and moves them to //blimp/client/core/session. They will be used by the BlimpClientContextImpl. BUG=611097 Review-Url: https://codereview.chromium.org/2187893002 Cr-Commit-Position: refs/heads/master@{#408802}
-
rsesek authored
R=jrummell@chromium.org Review-Url: https://codereview.chromium.org/2196453003 Cr-Commit-Position: refs/heads/master@{#408801}
-
vakh authored
BUG=543161, 624567 Review-Url: https://codereview.chromium.org/2196553002 Cr-Commit-Position: refs/heads/master@{#408800}
-
chrome://invalidationspnoland authored
invalidation_service_android never updated its logger instance's state, since the state is always enabled. This meant the logger only ever reported the state as its initial value, which is transient_error. This change causes invalidation_service_android to update the state to enabled upon construction. R=pavely@chromium.org BUG=630406 [sync] re-enable SyncTest#testRename Review-Url: https://codereview.chromium.org/2201503002 Cr-Commit-Position: refs/heads/master@{#408799}
-
agrieve authored
Just needed to make the aar's java_prebuilt targets depend on the resource targets so that would be deemed the resource owners within write_build_config.py BUG=632362 Review-Url: https://codereview.chromium.org/2190283002 Cr-Commit-Position: refs/heads/master@{#408798}
-
jshin authored
1. Update the IANA timezone DB to 2016f from 2016d. 2. Add support for Big Endian (should not affect Chrome builds) mainly for v8 embedders (mips/mips64 BE targets). 3. Make it possible to run foreign trybots. For more details, see https://chromium.googlesource.com/chromium/deps/icu.git/+log/b5ecbb29..ef5c7353 TBR=machenbach@chromium.org BUG=473288,v8:4828 TEST=Try the following in d8 or Chrome's JS console: > (new Date("07/23/2016 12:00Z")).toLocaleString("en", {timeZone: "Asia/Novosibirsk"}) 7/23/2016, 6:00:00 PM > (new Date("07/24/2016 12:00Z")).toLocaleString("en", {timeZone: "Asia/Novosibirsk"}) 7/24/2016, 7:00:00 PM Review-Url: https://codereview.chromium.org/2188333002 Cr-Commit-Position: refs/heads/master@{#408797}
-
mcasas authored
No new code, just shuffling files around -- commands: git mv content/browser/media/webrtc content/browser/ tools/git/mass-rename.py BUG=632020 Review-Url: https://codereview.chromium.org/2193343002 Cr-Commit-Position: refs/heads/master@{#408796}
-
wangxianzhu authored
The flakiness may be because sometimes the test was started before the image is loaded. Now start the test in onload. Remove wrong baselines that were created based on wrong results when the test failed on the bots. BUG=622898 Review-Url: https://codereview.chromium.org/2188383002 Cr-Commit-Position: refs/heads/master@{#408795}
-
fdoray authored
Currently, TaskRunner::PostTaskAndReply and WorkerPool::PostTaskAndReply can only be called when ThreadTaskRunnerHandle::IsSet(). The reply is posted back to the thread on which PostTaskAndReply was invoked. With this CL, these methods can also be called when SequencedTaskRunnerHandle::IsSet(). The reply is posted back to the sequence on which PostTaskReply was invoked. This change is a prerequisite to implement base::PostTaskAndReply which must work from a sequenced task https://docs.google.com/document/d/1S2AAeoo1xa_vsLbDYBsDHCqhrkfiMgoIPlyRi6kxa5k/edit#heading=h.gf6e3bwrxqqs BUG=553459 Review-Url: https://codereview.chromium.org/2180953002 Cr-Commit-Position: refs/heads/master@{#408794}
-
catapult-deps-roller authored
https://chromium.googlesource.com/external/github.com/catapult-project/catapult.git/+log/28aad26a05ee..395e5365f02a $ git log 28aad26a0..395e5365f --date=short --no-merges --format='%ad %ae %s' BUG=630300 TBR=catapult-sheriff@chromium.org Review-Url: https://codereview.chromium.org/2196483005 Cr-Commit-Position: refs/heads/master@{#408793}
-
kolos authored
Replace WebString with base::string16. AutofillAgent sends to PasswordAutofillAgent base::string16, but they are wrapped into WebString and and unwrapped to base::string16. That doesn't make sense. Review-Url: https://codereview.chromium.org/2187253002 Cr-Commit-Position: refs/heads/master@{#408792}
-
jbauman authored
The Microsoft MFT we're using doesn't pass through the bind flags correctly so sharing textures with ANGLE doesn't work. Review-Url: https://codereview.chromium.org/2188423002 Cr-Commit-Position: refs/heads/master@{#408791}
-
chrome-cron authored
Cr-Commit-Position: refs/heads/master@{#408790}
-
waffles authored
BUG=623804 Review-Url: https://codereview.chromium.org/2190083002 Cr-Commit-Position: refs/heads/master@{#408789}
-
scottmg authored
Use raw pointers instead of unique pointers to reduce size. See linked bug for size details. BUG=631416 Review-Url: https://codereview.chromium.org/2193693003 Cr-Commit-Position: refs/heads/master@{#408788}
-
enne authored
This bug started after r408172. In some situations when navigating, the SetNeedsBeginFrame messages that is sent from the CompositorExternalBeginFrameSource to the RenderWidgetHostView, but is received by the RenderWidgetHostImpl while the RenderViewHost is swapped out and before it has swapped back in. In these cases, if this message is dropped, the renderer will never tick new frames and the tab will freeze. This appears to be racy, as the renderer finds out that it is visible and sends this message before the view host is swapped in. To fix this race, always allow the SetNeedsBeginFrame message to be received while being swapped out so that the RenderWidgetHostView can be in the correct state when it is swapped in. R=sievers@chromium.org BUG=632276 Review-Url: https://codereview.chromium.org/2197783002 Cr-Commit-Position: refs/heads/master@{#408787}
-
qyearsley authored
Many test methods in unit tests have names much longer than 30 characters. I think this is fine for test method names, since they're meant to be descriptive, but not necessarily short and memorable. Review-Url: https://codereview.chromium.org/2192213002 Cr-Commit-Position: refs/heads/master@{#408786}
-
dewittj authored
This is an inefficient intermediate step to support multiple client IDs in a single API call. TODO: update native backend to support multiple Client ID search. BUG= Review-Url: https://codereview.chromium.org/2187033002 Cr-Commit-Position: refs/heads/master@{#408785}
-
brettw authored
Revert of Simplify landmines code a bit now that all branches of all platforms are on ninja (patchset #2 id:20001 of https://codereview.chromium.org/2071463002/ ) Reason for revert: I just synced on Windows and this is giving me: ________ running 'D:\depot_tools\python276_bin\python.exe src/build/landmines.py ' in 'D:\src\src1' Traceback (most recent call last): File "D:\src\src1\src\build\get_landmines.py", line 16, in <module> builder = landmine_utils.builder AttributeError: 'module' object has no attribute 'builder' Clobbering due to: --- old_landmines Tue Aug 12 16:34:51 2014 +++ new_landmines Fri Jul 29 15:16:56 2016 @@ -1,17 +0,0 @@ -Compile on cc_unittests fails due to symbols removed in r185063. -Improper dependency for create_nmf.py broke in r240802, fixed in r240860. -Switch to VS2015 Update 2 -Need to clobber everything due to an IDL change in r154579 (blink) -Need to clobber everything due to gen file moves in r175513 (Blink) -Clobber to get rid of obselete test plugin after r248358 -Clobber to rebuild GN files for V8 -Clobber to get rid of stale generated mojom.h files -Need to clobber everything due to build_nexe change in nacl r13424 -[chromium-dev] PSA: clobber build needed for IDR_INSPECTOR_* compil... -blink_resources.grd changed: crbug.com/400860 -ninja dependency cycle: crbug.com/408192 -Clobber to fix missing NaCl gyp dependencies (crbug.com/427427). -Another clobber for missing NaCl gyp deps (crbug.com/427427). -Clobber to fix GN not picking up increased ID range (crbug.com/444902) -Remove NaCl toolchains from the output dir (crbug.com/456902) -Clobber to delete stale generated files (crbug.com/510086) Original issue's description: > Simplify landmines code a bit now that all branches of all platforms are on ninja > > BUG=none > > Committed: https://crrev.com/7491d14275670fad399837445b6c1e76ff498a50 > Cr-Commit-Position: refs/heads/master@{#408739} TBR=iannucci@chromium.org,thakis@chromium.org # Skipping CQ checks because original CL landed less than 1 days ago. NOPRESUBMIT=true NOTREECHECKS=true NOTRY=true BUG=none Review-Url: https://codereview.chromium.org/2198583002 Cr-Commit-Position: refs/heads/master@{#408784}
-
wkorman authored
Review-Url: https://codereview.chromium.org/2193073004 Cr-Commit-Position: refs/heads/master@{#408783}
-
dpranke authored
TBR=brettw@chromium.org NOTRY=true BUG=623659 Review-Url: https://codereview.chromium.org/2192303002 Cr-Commit-Position: refs/heads/master@{#408782}
-
mfoltz authored
This converts presentation.mojom to use the new (std::) Mojo wrapper types: mojo::String => std::string mojo::Array => std::vector No behavior changes should ensue. Optional values are represented as base::Optional<Foo> with the new wrapper types requiring some additional changes. This work will simplify further changes for crbug.com/627655. BUG=627655 Review-Url: https://codereview.chromium.org/2181163003 Cr-Commit-Position: refs/heads/master@{#408781}
-
dpapad authored
BUG=632761 CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation Review-Url: https://codereview.chromium.org/2198443003 Cr-Commit-Position: refs/heads/master@{#408780}
-
jsbell authored
Better name and it matches the concept in the spec. Review-Url: https://codereview.chromium.org/2170043004 Cr-Commit-Position: refs/heads/master@{#408779}
-
ben authored
R=rockot@chromium.org BUG= Review-Url: https://codereview.chromium.org/2192263002 Cr-Commit-Position: refs/heads/master@{#408778}
-
rch authored
Review-Url: https://codereview.chromium.org/2193073003 Cr-Commit-Position: refs/heads/master@{#408777}
-
einbinder authored
BUG= Review-Url: https://codereview.chromium.org/2163393002 Cr-Commit-Position: refs/heads/master@{#408776}
-
kenobi authored
Unqualified page transitions are treated as link-clicks, which can get sent to android via the arc intent mechanism. This can result in infinite-looping if the URL that is being opened originated from android in the first place. Marking this as an API-generated page transition ensures that the arc intent code does not forward this URL back to android. BUG=632360 Review-Url: https://codereview.chromium.org/2189063002 Cr-Commit-Position: refs/heads/master@{#408775}
-
thomasanderson authored
The linux kernel stores and prints the data associated with /proc/vmstat as unsigned longs, so increase Chrome's storage of these fields accordingly. https://github.com/torvalds/linux/blob/e4fba88d00d21f50239eac7d653027bfaf844296/include/linux/vmstat.h#L25 Review-Url: https://codereview.chromium.org/2188043003 Cr-Commit-Position: refs/heads/master@{#408774}
-
rnephew authored
BUG=632809 CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.perf:android_s5_perf_cq;master.tryserver.chromium.perf:linux_perf_cq;master.tryserver.chromium.perf:mac_retina_perf_cq;master.tryserver.chromium.perf:winx64_10_perf_cq Review-Url: https://codereview.chromium.org/2198523003 Cr-Commit-Position: refs/heads/master@{#408773}
-
sievers authored
This removes some native references to ContentViewCore from the embedder, which were mostly unnecessary indirections and can use WebContents directly. BUG=626764 Review-Url: https://codereview.chromium.org/2179563003 Cr-Commit-Position: refs/heads/master@{#408772}
-
ssid authored
On Android the exit status is always marked as EXIT_CRASHED. So, the background traces are never uploaded. Since the check is useful only for preemptive mode, this CL removes this check for REACTIVE triggers on Android and adds extra check for preemptive trigger in that place. BUG=613198 Review-Url: https://codereview.chromium.org/2192093002 Cr-Commit-Position: refs/heads/master@{#408771}
-