Updates //docs/gpu/gpu_testing.md in a few ways:
1. Clarifies that vpython should be used instead of python
2. Explicitly lists all Telemetry-based test suites and the names they
can show up as on the bots. This list was generated by parsing the
//testing/buildbot JSON files.
3. Updates the instructions on how to obtain the command used to run
a test on swarming and how to download an isolate.
Bug: 1199154
Change-Id: I21bbbe7341d8788ef9e0715e5c090b8cc37469ec
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2828908
Auto-Submit: Brian Sheedy <bsheedy@chromium.org>
Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Cr-Commit-Position: refs/heads/master@{#873436}
I can't effectively review code in most of these directories anymore,
so removing myself. (I've left myself as an owner in the areas of code
I still remember, like the interstitials.)
Change-Id: Ie4b00e738509f83c5082a61b47cbd2cf9e3f25f9
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2830461
Reviewed-by: Emily Stark <estark@chromium.org>
Commit-Queue: Adrienne Porter Felt <felt@chromium.org>
Cr-Commit-Position: refs/heads/master@{#873359}
Use enum class instead of enum for View::EmphasisMetric.
enum class is more type safety.
and rename EmphasisMetric to Emphasis to keep usage sites
reading like they used to.
Bug: 940736
Change-Id: I84540655dcf1b009bb620e67829a798f5bdfd866
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2804936
Reviewed-by: Scott Violet <sky@chromium.org>
Reviewed-by: Peter Kasting <pkasting@chromium.org>
Commit-Queue: Hwanseung Lee <hs1217.lee@samsung.com>
Cr-Commit-Position: refs/heads/master@{#870759}
Most of the Chrome/Android developers frequently rely on custom
log messages during the development or bug-fixing phase.
Whether we add logging to validate the flow or inspect the data,
we frequently depend on certain patterns.
This change enables Developers to define environment variable,
`CHROMIUM_LOGCAT_HIGHLIGHT`, holding a regular expression that will
be applied to Tag and Message to verify whether a given log line is
relevant to the Developer. Identified relevant lines are highlighted.
Change-Id: Id4ae5a8698bb8d097151b863ea663b1b922e5925
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2815513
Commit-Queue: Tomasz Wiszkowski <ender@google.com>
Reviewed-by: Peter Wen <wnwen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#870718}
This CL adds use counters tracking how often a COOP
same-origin-allow-popups page has a document which is same-origin with a
document in a COOP unsafe-none page in the same browsing context group.
Bug: 1184718
Change-Id: I35023c9759eb4afdb30dc6c7bcb5b6beddb80790
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2735678
Commit-Queue: Camille Lamy <clamy@chromium.org>
Reviewed-by: Arthur Sonzogni <arthursonzogni@chromium.org>
Reviewed-by: Mike West <mkwst@chromium.org>
Cr-Commit-Position: refs/heads/master@{#870461}
After getting merge approval, it can be hard to find details of how,
exactly, one merges a change into a branch. Since merge requests
automatically link to docs/process/merge_request.md, have a link from
there to the doc with the technical details.
Bug: None
Change-Id: I429064fa06c6b0562d15584c14c04a8347cbafff
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2783154
Auto-Submit: Ian Barkley-Yeung <iby@chromium.org>
Reviewed-by: Alex Mineer <amineer@chromium.org>
Commit-Queue: Alex Mineer <amineer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#869642}
No change to logic, only docs. This updates the emulator docs to call
out a known issue with Android L-M emulator images. Unlike other OS
versions, all L-M emulator images are configured to expect the AOSP
WebView package.
Test: Upload to gerrit > open file > click "gitiles"
Change-Id: Ia48ab33dab246b89c98cfc512bd5e0007f760f85
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2806654
Auto-Submit: Nate Fischer <ntfschr@chromium.org>
Commit-Queue: Shimi Zhang <ctzsm@chromium.org>
Reviewed-by: Shimi Zhang <ctzsm@chromium.org>
Cr-Commit-Position: refs/heads/master@{#869554}
This makes a few updates to the documentation on using the rr debugger:
* clarify recommendation to build from source into one about using a
recent version (which might mean building from source), and explain
why
* mention both rr replay -p and -f, and explain the difference
* remove reference to a year-old bug that's now subsumed by the advice
on using a recent version
* add advice on workarounds for calling functions that use LOG() from
gdb
* link to relevant Chromium Chronicle episode
Change-Id: I13206a009103ae12f83a3890101a91e02fd78a0a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2794877
Reviewed-by: Christian Biesinger <cbiesinger@chromium.org>
Commit-Queue: David Baron <dbaron@chromium.org>
Cr-Commit-Position: refs/heads/master@{#867753}
The memfd_create(2) is necessary for creating a shared memory region in
the app zygote and letting it survive the fork(2). Surviving the
memfd-provided FD is only possible on Android R+ because of the new
behavior in FileDescriptorWhitelist.
All the functionality of memfd_create was introduced in kernel version
3.17. The minimum supported kernel version on R is 4.4. Hence no kernel
version check is performed in Chrome when running on R+.
The libandroid.so is providing a couple of functions to abstract away
creating and sealing the shared memory regions:
* ASharedMemory_create(),
* ASharedMemory_setProt().
There is a plan to remove ashmem support from the kernel releases, and
presumably switch to memfd. The switch to memfd has not happened in R,
hence the ModernLinker has to use the memfd explicitly.
For memfd-provided regions we implement these _create and _setProt
functions using memfd_create and fcntl(F_ADD_SEALS, ...). The difference
is that PROT_EXEC is not supported for memfd.
Unfortunately, the memfd regions have a different behavior with mmap(2).
While ashmem regions are always memory-mapped as MAP_SHARED, the
read-only memfd regions should use MAP_PRIVATE. Counter-intuitively,
those pages are still shared across processes. Even more: a read-only
memfd region can be mapped as a read-write with MAP_SHARED, but in this
case it would copy the physical pages (likely copy-on-write). This
required to add other functions that behave differently for ashmem and
memfd:
typedef bool (*MapReadOnlyFunction)(int fd, size_t size, void** out_address);
typedef bool (*MapReadOnlyFixedFunction)(int fd, size_t address, size_t size);
The "Fixed" variant performs a MAP_FIXED, which can only reliably work
on top of an address range that was previously reserved with mmap. This
will be needed later.
Creating RELRO FD from within the App Zygote is not enabled with this
change, it switches _all_ RELRO FD regions to memfd on R+ for
simplicity. Creating in zygote is coming soon. In order to reduce the
size of potential relands the switch is guarded by a compile time
constant, and not enabled yet. The native unittest, however, starts
covering the memfd support right now.
Bug: 1154224
Change-Id: I9cc63ae6db268fb14ee7b6197bc101b13257e93e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2791391
Reviewed-by: Benoit L <lizeb@chromium.org>
Reviewed-by: Gabriel Charette <gab@chromium.org>
Commit-Queue: Egor Pasko <pasko@chromium.org>
Cr-Commit-Position: refs/heads/master@{#867649}
The binary shouldn't depend on what system it was built on.
We already have a sysroot for chromium, so just use that.
Built at the same revision as previously (eb85e90350e).
No expected behavior change.
Bug: none
Change-Id: Iddced88221bd8864c5a33963ca7796b0660d0732
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2790525
Auto-Submit: Nico Weber <thakis@chromium.org>
Commit-Queue: Hans Wennborg <hans@chromium.org>
Reviewed-by: Hans Wennborg <hans@chromium.org>
Cr-Commit-Position: refs/heads/master@{#867197}
Normally we prefer having a dedicated binary each for intel and arm,
but clang-format:
1. is pulled via a .sha1 file and the binary will be next to the .sha1
file
2. the location of the binary is referenced from depot_tools
So adding a second binary would mean adding a second .sha1 file, and
then we'd have to teach depot_tools to look in both places. And
clang-format is reasonably small. So just make it a universal binary
instead.
I built clang-format at the same revision the intel-only binary
was built at (eb85e90350e), using the steps in
docs/updating_clang_format_binaries.md. The only change I made
was to also pass `'-DCMAKE_OSX_ARCHITECTURES=arm64;x86_64'` to
cmake (need to include the quotes, else the shell uses the `;` as
statement terminator -- luckily the `../llvm` part was after it
so cmake informed me of missing this at first).
Bug: 1190868
Change-Id: I092c4b1e9d37ddd7aeb4caef7d612a8df44092f2
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2786107
Reviewed-by: Hans Wennborg <hans@chromium.org>
Reviewed-by: Nico Weber <thakis@chromium.org>
Commit-Queue: Nico Weber <thakis@chromium.org>
Auto-Submit: Nico Weber <thakis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#866974}
This CL renames all the usages of LifecycleState with
LifecycleStateImpl.
This CL is the first part of the work to expose RenderFrameHost
lifecycle state to //content/public.
RenderFrameHostImpl::LifecycleStateImpl will be used exclusively
within //content, where it will represent all possible
lifecycle states. Future CLs will introduce
RenderFrameHost::LifecycleState, which will expose a high-level a
subset of these states publicly.
This is a no-op update.
BUG=1113357
Change-Id: I6e70090b76f7e52d252e229f371f861ad07cc0f3
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2774609
Commit-Queue: Sreeja Kamishetty <sreejakshetty@chromium.org>
Reviewed-by: Alexander Timin <altimin@chromium.org>
Reviewed-by: Alex Moshchuk <alexmos@chromium.org>
Cr-Commit-Position: refs/heads/master@{#866589}
This change removes calls to base::UTF8ToUTF16 with a single-line string
literal not containing escape sequences and replaces them with a u"..."
literal instead.
This is a mechanical change:
$ git grep -lw UTF8ToUTF16 | \
xargs sed -i 's/\(base::\)\?UTF8ToUTF16(\("[^\\"]*"\))/u\2/g'
$ git cl format
AX-Relnotes: n/a.
Bug: 1189439
Change-Id: I6df2e531f62926a12be3f610ce8627d94ec407f6
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2776021
Commit-Queue: Jan Wilken Dörrie <jdoerrie@chromium.org>
Commit-Queue: Daniel Cheng <dcheng@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Owners-Override: Daniel Cheng <dcheng@chromium.org>
Cr-Commit-Position: refs/heads/master@{#865390}
- Mention what CLs stand for.
- Mention "Find Owners" button in Gerrit.
- Provide links to sections, rather than referring to them as "above"
and "below".
- Use backticks to mark certain keywords more consistently.
- Change reference to the "blink" dir to third_party/blink.
Change-Id: Ied603ec93ea81256f2ef1fe0318a508989ce3c3c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2775960
Reviewed-by: Takuto Ikuta <tikuta@chromium.org>
Commit-Queue: Lei Zhang <thestig@chromium.org>
Cr-Commit-Position: refs/heads/master@{#865229}
Adds 2 additional include dirs:
- src/out/Default/gen, which allows CLion to understand generated files
(e.g. mojom files and debugging_buildflags.h)
- src/third_party/googletest/src/googletest/include, which allows CLion
to understand testing/gtest/include/gtest/gtest.h and its includes.
Change-Id: I1e93b4fae40da8d56e336d550d89dac5c9b8b8d7
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2775967
Commit-Queue: manuk hovanesian <manukh@chromium.org>
Reviewed-by: Orin Jaworski <orinj@chromium.org>
Cr-Commit-Position: refs/heads/master@{#864873}
