|
ed391dae
|
2022-03-25T16:36:39
|
|
Replace `extern thread_local` to avoid GCC < 9.1 bug
GCC generates broken code for mingw-w64 target when accessing `extern
thread_local` variables from another compilation unit.
See: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104862
See:
https://invent.kde.org/graphics/krita/-/blob/master/3rdparty/ext_googleangle/02-patches_krita/0017-Replace-extern-thread_local-to-avoid-GCC-mingw-w64-b.patch
Co-authored-by: Alvin Wong <alvinhochun@gmail.com>
Bug: None
Change-Id: Ief0036054e644d40b8b0e86ed990280e566da2a9
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4637587
Auto-Submit: L. E. Segovia <amy@amyspark.me>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
3545ae0c
|
2023-06-09T16:11:47
|
|
Add isContextMutexStateConsistent() ASSERT
Added `gl::Context::isContextMutexStateConsistent()` method. This method
is primarily used to check if "SharedContextMutex" activation worked
successfully. It is automatically called in the updated
`ScopedContextMutexLock` before unlocking. This is to catch possible
errors using ASSERT during normal ANGLE operation in applications and
tests.
The `ScopedContextMutexLock` is now also used instead of the
`std::lock_guard<egl::ContextMutex>` in the `SCOPED_SHARE_CONTEXT_LOCK`.
No performance regression observed.
Important note: `lockAndActivateSharedContextMutex()` is NOT 100% safe
regardless of the `kActivationDelayMicro` value, so `ASSERT` may still
fail. However, failure does not necessary mean that there will be an
undefined behavior, it means that UB might happen.
Bug: angleproject:6957
Bug: chromium:1336126
Change-Id: Iee7357fede0d37fa315fe2cc7d27a4e30a304194
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4610227
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Igor Nazarov <i.nazarov@samsung.com>
|
|
|
36c3e0f5
|
2023-01-17T17:42:59
|
|
Implement "Shared Context Mutex" functionality.
Existing implementation uses single `GlobalMutex` for
- EGL calls
- GL calls for Contexts with concurrent access.
This CL introduces abstract `egl::ContextMutex` with two
implementations:
- SingleContextMutex;
- SharedContextMutex<Mutex>;
Note:
`std::mutex` is used in this commit. It is very easy to change mutex
type either at compile-time or at run-time (single type per Display).
When Context:
- is not Shared;
- does not use `EGLImage`s;
- does not use EGL_DISPLAY_TEXTURE_SHARE_GROUP_ANGLE
- does not use EGL_DISPLAY_SEMAPHORE_SHARE_GROUP_ANGLE
then it will be using `SingleContextMutex` with minimal overhead.
Before such Context is used as `shareContext` or uses `EGLImage`
its mutex replaced by `SharedContextMutex<Mutex>`.
The `GlobalMutex` is only used for EGL calls, while `egl::ContextMutex`
implementations for GL calls. Because some EGL calls use Context,
explicit `egl::ContextMutex` lock is required. This is implemented by
generating "egl_context_mutex_autogen.h" header, and insertion of
`ANGLE_EGL_SCOPED_CONTEXT_LOCK()` macro before `ANGLE_EGL_VALIDATE()`
in each EGL entry point. Implementation in "egl_context_lock_impl.h"
returns lock for required APIs. Special cases of `egl::ContextMutex`
lock handled separately. `std::unique_lock<>` is not used for
performance reasons.
`egl::ContextMutex` explicitly locked when capturing EGL calls.
Fixes EGLImage problem:
https://chromium.googlesource.com/angle/angle/+/e18240d136d15e5cdfa4fa4a6355ca21c8d807b6
Mark contexts as shared when importing EGL images.
Details:
- EGLImage inherits Context's mutex when created.
Mutex is used when the EGLImage accessed or destroyed.
- When EGLImage is used in Context with other `egl::ContextMutex`,
two mutexes are merged into one.
- After the mutex merge, Context Groups will remain separate,
but will not be able to run in parallel.
Fixes race when checking `context->isShared()` in the
`SCOPED_SHARE_CONTEXT_LOCK()` macro. One Context may start executing GL
call while not "Shared", but become "Shared" inside the call. New
(second) "Shared" Context may immediately start using GL and potentially
corrupt some "Shared" state.
Possible performance benefit: allows parallel execution in some cases,
when single `GlobalMutex` would block.
Important note:
Process of replacing the `SingleContextMutex` by
`SharedContextMutex<Mutex>` is not 100% safe. This mean that
original Context may still be using `SingleContextMutex` after
activating `SharedContextMutex<Mutex>`. However, this was always
the case before introduction of this CL. Old `Context::mShared`
member update was not synchronized in any way at all. In other
words, this solution does not 100% fix the original problem.
For 100% safe solution `SingleContextMutex` should not be used
(always pass `SharedContextMutex<Mutex>` to the `gl::Context`
constructor). See `lockAndActivateSharedContextMutex()` for more
details.
CL adds new build option:
angle_enable_shared_context_mutex = true
Behavior with other build options:
- When:
`angle_enable_shared_context_mutex` is disabled or
`angle_enable_share_context_lock` is disabled or
`angle_force_context_check_every_call` is enabled,
Contexts will always have `SingleContextMutex`, however it will be
only used in special cases. `SCOPED_SHARE_CONTEXT_LOCK()` will use
`GlobalMutex` when applicable.
- Otherwise, `SCOPED_SHARE_CONTEXT_LOCK()` will use `egl::ContextMutex`.
Some GFXBench "1080p Driver Overhead 2 Offscreen" performance numbers.
Tested on S906B (Samsung Galaxy S22+) on old ANGLE base:
https://chromium.googlesource.com/angle/angle/+/807c94ea85e046c6f279d081d99f0fb1bcf1191a
Capture/Replay: Adjust tests do adhere to capture limits
Each test result is an average frame number from 6 runs.
SingleContextMutex 6579 ( +0.13%)
(old) GetContextLock() (mShared is false) 6570
Forced `mShared = true` or NOT using `SingleContextMutex`.
SharedContextMutex<std::mutex> FORCE 5061 (-22.97%)
(old) GetContextLock() FORCE 4766 (-27.46%)
Bug: angleproject:6957
Bug: chromium:1336126
Change-Id: Idcd919f9d4bf482b9ae489bd8b4415ec96048e32
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4374545
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
1acc6d40
|
2023-04-21T17:03:49
|
|
Vulkan: Use Android TLS for *valid* global context
... instead of just global context. This means that the majority of
calls which call GetValidGlobalContext() will be faster (because they
don't need to check for context loss).
Bug: angleproject:8135
Change-Id: Ia91f8fadb5799088542c58d417f31c3e5028b3e0
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4454769
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Igor Nazarov <i.nazarov@samsung.com>
Reviewed-by: Charlie Lao <cclao@google.com>
|
|
|
b22124bd
|
2023-01-17T17:44:06
|
|
Implement new "Global Mutex" functionality.
This CL improves performance compared to the existing code as well as
allows using non-std mutex implementations. Also acts as a base for
future changes.
CL adds new build option:
angle_enable_global_mutex_recursion = is_android && angle_enable_vulkan
"mutex_recursion" work same way as `std::recursive_mutex` before. It
will help in situations when Vulkan API may return back to the ANGLE.
For example: RenderDoc layer EGL deadlock.
Automatic loading of "libVkLayer_GLES_RenderDoc.so" layer causes
deadlock in EGL.
Recursion stack:
#01 pc 000000000029ea80 /vendor/lib64/egl/libGLESv2_angle.so (egl::GlobalMutexHelper::lock(int)+596)
#02 pc 000000000029c59c /vendor/lib64/egl/libGLESv2_angle.so (EGL_GetError+32)
#04 pc 0000000000062368 /system/lib64/libEGL.so (eglQueryString+20)
#05 pc 0000000000508fec /data/local/debug/vulkan/libVkLayer_GLES_RenderDoc.so
#20 pc 0000000000016690 /system/lib64/libvulkan.so (vulkan::api::EnumerateInstanceLayerProperties(unsigned int*, VkLayerProperties*)+40)
#21 pc 00000000005aa030 /vendor/lib64/egl/libGLESv2_angle.so (rx::RendererVk::initialize(rx::DisplayVk*, egl::Display*, char const*, char const*)+292)
#26 pc 000000000029c7e8 /vendor/lib64/egl/libGLESv2_angle.so (EGL_Initialize+192)
Additionally, recursive mutex will partially solve Android
SurfaceTexture deadlock (angleproject:4354).
Some performance numbers for 1000'000 `eglGetError()` calls.
Mutex Time (ms.)
Android S906B
egl::GetGlobalMutex()(std::recursive_mutex) 41.4
(Default) GlobalMutex (std::recursive_mutex) 39.1
(Recursive) GlobalMutex (std::mutex) 34.9
(Debug) GlobalMutex (std::mutex) 34.7
(Default) GlobalMutex (std::mutex) 34.4
Windows
egl::GetGlobalMutex()(std::recursive_mutex) 20.5
(Default) GlobalMutex (std::recursive_mutex) 20.0
(Recursive) GlobalMutex (std::mutex) 21.9
(Debug) GlobalMutex (std::mutex) 20.5
(Default) GlobalMutex (std::mutex) 19.9
Note: Recursive GlobalMutex enabled only for Android Vulkan by default.
Original fix:
https://chromium-review.googlesource.com/c/angle/angle/+/2029218
Bug: angleproject:8101
Bug: angleproject:4354
Test: angle_unittests --gtest_filter=GlobalMutexTest.*
Change-Id: I9e9d9b5c598ad1177ffa147ea690bd955946a712
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4401940
Commit-Queue: Igor Nazarov <i.nazarov@samsung.com>
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
e2baaff8
|
2023-04-06T00:11:23
|
|
Prevent recursive GlobalMutex locking by ANGLE itself.
Instead of adding `ANGLE_SCOPED_GLOBAL_LOCK()` for `glEGLImage*` GLES
APIs, this CL uses new `SCOPED_GLOBAL_AND_SHARE_CONTEXT_LOCK()` macro
in place of `SCOPED_SHARE_CONTEXT_LOCK()`.
This will remove dependency on recursive mutex for ANGLE itself.
Recursive mutex is still required for Android Vulkan or in specific
cases when using layer libraries.
Bug: chromium:1383195
Change-Id: I0c379c45c046b0f5e5dd3ea64a45d33b0ad3ee43
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4404777
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Charlie Lao <cclao@google.com>
Commit-Queue: Igor Nazarov <i.nazarov@samsung.com>
|
|
|
be3d7e6b
|
2023-03-30T14:35:18
|
|
Remove redundant g_SurfaceMutex.
This mutex become redundant after recent CL:
https://chromium.googlesource.com/angle/angle/+/23ad4fa2be35aed303cbb8c2d632a04714354ef0
Don't hold global surface lock during AcquireNextImage
Every place in the code where `ANGLE_SCOPED_GLOBAL_SURFACE_LOCK()`
is used also uses `ANGLE_SCOPED_GLOBAL_LOCK()`.
Bug: angleproject:6851
Change-Id: I464b6ca74743c9ee9fa23caad216f0e26c480655
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4385293
Commit-Queue: Igor Nazarov <i.nazarov@samsung.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Charlie Lao <cclao@google.com>
|
|
0f3aaebf
|
2022-12-14T01:00:42
|
|
[PATCH] Add riscv64 support
Bug: b/262585507
Change-Id: I7e081c5a1dacbcf1221d7580f33ce504442ebbbc
Signed-off-by: Zhang Ye <haocheng.zy@linux.alibaba.com>
Signed-off-by: Mao Han <han_mao@linux.alibaba.com>
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4104471
Commit-Queue: Charlie Lao <cclao@google.com>
Reviewed-by: Charlie Lao <cclao@google.com>
|
|
b9fd47b8
|
2022-08-12T10:55:49
|
|
Not use Android TLS slot for chromium
In the profile, GetGlobalContext() & GetValidGlobalContext() have
extra overhead for checking angle::gUseAndroidOpenGLTlsSlot.
Since angle::gUseAndroidOpenGLTlsSlot is always false for now,
so add a build flag to disable it for chromium build.
In the profile, with angle_use_android_tls_slot = false,
GetValidGlobalContext() is reduced from 0.153% to 0.044%.
Bug: chromium:1336126
Change-Id: Id5e1752b0a03289fb299e8de2ed6e962e35291c0
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3827955
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Auto-Submit: Peng Huang <penghuang@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
|
|
|
b78d471a
|
2022-07-21T09:53:37
|
|
Add angle_enable_share_context_lock build flag
This flags can be used to enable the share context lock.
Without this lock, the client need to use gl calls in
a threadsafe way. It is true by default.
Bug: chromium:1336126
Change-Id: I984f8cfb0379195f6ebe11b0997e401f2421affa
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3780582
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Peng Huang <penghuang@chromium.org>
|
|
91976352
|
2022-06-21T15:41:02
|
|
Use C++17 attributes instead of custom macros
Bug: angleproject:6747
Change-Id: Iad6c7cd8a18d028e01da49b647c5d01af11e0522
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3718999
Reviewed-by: Cody Northrop <cnorthrop@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
7a85d114
|
2022-03-25T15:01:17
|
|
Use [[nodiscard]] on RAII classes
Scoped* classes provide an RAII way of adding cleanup/restore state/etc
in a robust way. Unfortunatley, it's very easy to mistakenly leave the
variable name, leading to the destructor being called immediately
instead of at the end of the scope:
{
ScopedX(parameters); // instead of ScopedX x(parameters);
// Code here is run after destructor
}
The [[nodiscard]] attribute, if specified on the ScopedX class would
lead to a warning (turned to error with -Werror). This change does
that for classes named *Scoped* in ANGLE.
Bug: chromium:1103817
Change-Id: I65c9922c9b4eba1f9c033e093fe8fe534648ab62
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3552092
Reviewed-by: Lingfeng Yang <lfy@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
926b43e7
|
2022-01-06T13:31:54
|
|
Reland: Frontend: separate lock in swap prep
Swapchain-based backends like Vulkan might block a lot in
vkAcquireNextImageKHR, which is bad for overall fast progress if we also
hold the global EGL lock there.
This CL starts to split the global EGL lock. We release the EGL lock
when performing vkAcquireNextImageKHR, and only maintain a lock for
surfaces.
This is done via a new custom entry point, EGL_PrepareSwapBuffers, so
that we can control how the global lock is used throughout the entire
call.
Bug: angleproject:6851
Change-Id: I095cd8b3bdbb13c842cab0a46148e2122582cdfd
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3373426
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Lingfeng Yang <lfy@google.com>
|
|
9d668d6c
|
2022-01-07T18:18:23
|
|
Revert "Frontend: separate lock in swap prep"
This reverts commit 40c5cb255c0a07bdab574aa076ee603e7d791ab3.
Reason for revert: Regression in TSAN caused by this CL:
https://ci.chromium.org/ui/p/angle/builders/ci/linux-tsan-test/352/overview
Original change's description:
> Frontend: separate lock in swap prep
>
> Swapchain-based backends like Vulkan might block a lot in
> vkAcquireNextImageKHR, which is bad for overall fast progress if we also
> hold the global EGL lock there.
>
> This CL starts to split the global EGL lock. We release the EGL lock
> when performing vkAcquireNextImageKHR, and only maintain a lock for
> surfaces.
>
> Bug: angleproject:6851
> Change-Id: I329d5c4c579718a4980c4261590f77099ce1400e
> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3361249
> Reviewed-by: Charlie Lao <cclao@google.com>
> Reviewed-by: Geoff Lang <geofflang@chromium.org>
> Commit-Queue: Lingfeng Yang <lfy@google.com>
Bug: angleproject:6851
Change-Id: Ie03b784021f7b8b5c1ef95a911ef7da4029abd46
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3373165
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
40c5cb25
|
2022-01-06T13:31:54
|
|
Frontend: separate lock in swap prep
Swapchain-based backends like Vulkan might block a lot in
vkAcquireNextImageKHR, which is bad for overall fast progress if we also
hold the global EGL lock there.
This CL starts to split the global EGL lock. We release the EGL lock
when performing vkAcquireNextImageKHR, and only maintain a lock for
surfaces.
Bug: angleproject:6851
Change-Id: I329d5c4c579718a4980c4261590f77099ce1400e
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3361249
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Lingfeng Yang <lfy@google.com>
|
|
6c172e4b
|
2021-11-20T16:27:14
|
|
Add support for memory cleanup on process exit
This patch adds a callback to cleanup memory on process exit.
Bug: angleproject:6723
Test: Android CTS WrapperTest.testThreadCleanup
Change-Id: Ia517d4c6ae280ddc1f17a3b6f77d437aaaad0678
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3294581
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Mohan Maiya <m.maiya@samsung.com>
|
|
54d4bfe5
|
2021-09-28T17:27:57
|
|
Update ANGLE Metal to Webkit at Sept 29 2021
This commit merges changes from Webkit into ANGLE upstream.
The following commits were used:
Current:
https://git.webkit.org/?p=WebKit.git;a=commit;h=e01d0bda8f4b7dc2fd834b92802d15d8c15735f
Previous:
https://git.webkit.org/?p=WebKit.git;a=commit;h=492f078198748e8ff248eea0bb979cf79e5f5adfj
The following commits were merged in from the Webkit Repository:
(Hashes from git://git.webkit.org/WebKit-https.git)
03ea44c78ce5665d4ec9add271260121cbc7bc6c
Problems with drawElements in some conditions
https://bugs.webkit.org/show_bug.cgi?id=230107
c8dc8e0c4d1109d39a62eb197b45e95132380290
ANGLE Metal: single-component swizzles do not compile
https://bugs.webkit.org/show_bug.cgi?id=230472
7285dbaaf5af15877d6c332b30ef7a4d67225460
webgl-compressed-texture-s3tc-srgb.html fails on Intel+AMD Metal
https://bugs.webkit.org/show_bug.cgi?id=229941
4c72f92967ecd2a095666fef431384c4f5f60fb4
fragcolor-fragdata-invariant.html fails
https://bugs.webkit.org/show_bug.cgi?id=223317
cd943145467f54e5928793c0dd3dfa2313c007dd
ANGLE Metal index buffer restart range cache could be maintained..
https://bugs.webkit.org/show_bug.cgi?id=227451
f075ff77e592eabd54dd659a8e13617cc5faedc8
ANGLE Metal infinities and NaNs generated with incorrect syntax
https://bugs.webkit.org/show_bug.cgi?id=229439
5862073269122f4b2d43d96d3922757557755e86
[Metal ANGLE] Fix over-autorelease of
rx::DisplayMtl::getMetalDeviceMatchingAttribute()...
<https://webkit.org/b/229128>
85f797ad31db048cb82cbafd428ef77f0b839312 ANGLE Cocoa compiles....
https://bugs.webkit.org/show_bug.cgi?id=228987
a67918ba279ad4842b6ae84a79c3f1c0cdc35ace Avoid infinite recursion...
https://bugs.webkit.org/show_bug.cgi?id=228978
d341f67de0033adcf1ec6373ace6a54b06c4a031
Cherry-pick ANGLE: Revise WebGL's shaderSource validation
https://bugs.webkit.org/show_bug.cgi?id=228951
1e2714d981e97de8234ba055570dfdf56e8b6944
3.5 MB system-wide footprint impact due to thread-locals...
https://bugs.webkit.org/show_bug.cgi?id=228240
d32e5cca34081997d32504b0b56c18b9703ff3be
Build Default Metal library offline
https://bugs.webkit.org/show_bug.cgi?id=227333
33702279faccfd4c8d1c8a6d549925f9ca9a4e8f
WebGL2 demo doesn't work due to failing compilation....
https://bugs.webkit.org/show_bug.cgi?id=226865
0a075885d242db38c4e435a6597173dc3b082173
rAF driven WebGL submits excessive amount of GPU work...
https://bugs.webkit.org/show_bug.cgi?id=227059
f38a92b3e7c17efda269caa7066e7ffe2f828e72
WebGL shader link error in iOS 15 beta: "Internal error..."
https://bugs.webkit.org/show_bug.cgi?id=227723
98d48f011d561531470d97f26a022767b5452fb7
REGRESSION (r279466): [Big Sur] webgl/1.0.3/conformance &...
https://bugs.webkit.org/show_bug.cgi?id=227596
Bug: angleproject:6471
Change-Id: I07166d0dc4b5c3579d98353485b3245b81c7b882
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3194322
Reviewed-by: Jonah Ryan-Davis <jonahr@google.com>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Commit-Queue: Jonah Ryan-Davis <jonahr@google.com>
Commit-Queue: Kyle Piddington <kpiddington@apple.com>
|
|
|
a8e98a23
|
2021-07-10T15:25:20
|
|
Fix gl::Context::unMakeCurrent crashes
The crash is because Display::makeCurrent() may fail, in that case,
egl_stubs.cpp will not update the current context in global_state and
Thread accordingly. And then Context::refCount could be reach 0,
and be released, however egl_stubs.cpp still thinks the context is
current, and then user-after-free happens.
This CL fixes the problem by updating current context of Thread in
Display::makeCurrent, and reading the current context of Thread in
egl_stubs.cpp and setting it to global_state.
Bug: chromium:1171371
Change-Id: Ifc5fffb0e4902c9c72514839d03e5783d50fe283
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3017210
Reviewed-by: Jonah Ryan-Davis <jonahr@google.com>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Peng Huang <penghuang@chromium.org>
|
|
e7ca0aa2
|
2021-01-24T17:46:26
|
|
Fix crash in multi-context D3D11 backend.
The crash occured due to a lack of marking a context dirty on
eglMakeCurrent(), resulting in a situation where a thread B would
call eglMakeCurrent() but before issuing additional GL calls,
thread A would issue a draw command. Prior to this change, context
state would only be marked dirty on gl calls.
Test: dEQP-EGL.functional.sharing.gles2.multithread.*
dEQP-EGL.functional.multithread.*
dEQP-EGL.functional.render.multi_thread.*
dEQP-EGL.functional.color_clears.multi_thread.*
Bug: b/177602915
Change-Id: I765e0423002a373f94ea459b81b6e6f6942870d2
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2646609
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Doug Horn <doughorn@google.com>
|
|
192a0147
|
2020-10-08T16:47:35
|
|
Multithreading in D3D11 minimum viable product
Bug: b/168046573
Change-Id: I676a148333cbf5e9ca508768503e62cb14d8eeb0
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2629618
Commit-Queue: Doug Horn <doughorn@google.com>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Trevor David Black <vantablack@google.com>
|
|
d8529380
|
2020-10-20T13:55:23
|
|
Generate CONTEXT_LOST errors on every GL call.
The robustness spec states that all GL calls should generate context
lost errors when the context is lost. This behaviour was lost due to
some context lookup optimizations but can be re-added at no cost by
updating the autogenerated entry points.
Updated the ContextLostTest to cover the repeated error generation
behaviour.
Bug: chromium:1137241
Change-Id: I0b8d1cf4d2a37f13c466c8dd0e66edfe852f7e59
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2488180
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
|
|
|
cb286073
|
2020-09-20T21:59:24
|
|
Use c++ thread_local for current thread.
This solves a race condition using the global context.
Sadly we can't use an unprotected variable safely because
it can be written/read from multiple threads doing MakeCurrent.
Has about a 2 ns regression when tested in Linux/Release per
call. "null" benchmark test went from 27 -> 29 ns/iteration.
Fixes a TSAN warning that popped up in angle_end2end_tests.
Test: DrawCallPerfBenchmark.Run/vulkan_null
Test: EGLContextASANTest.DestroyContextInUse/ES3_Vulkan
Bug: b/168744561
Change-Id: Ic56f3faae81c1087b942a3cfc0e011b9ab439e0f
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2419641
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
f16e94cd
|
2020-06-17T10:43:19
|
|
Prefer retrieval of TLS values through ASM code on Android
On the Android platform prefer using the ASM code to retrieve
TLS context object even for single threaded apps. This helps
get rid of a branch in GetGlobalContext() and GetValidGlobalContext()
further improving the CPU perf of TLS operations.
Bug: angleproject:4717
Change-Id: I58d3d3b7061d613b24f945c07bed497c7c4be25c
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2250318
Commit-Queue: Mohan Maiya <m.maiya@samsung.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Courtney Goeltzenleuchter <courtneygo@google.com>
|
|
|
33c28e35
|
2020-06-12T08:17:47
|
|
Optimize thread specific storage and retrieval with native ASM
Android reserves specific TLS slots to store thread specific
values. Given that the Context object gets queried on every
call we leverage Android's ASM code to improve the speed of
this operation.
TLS_SLOT_OPENGL is an unused slot in Android and using that
in combination with the ASM code,rather than using the
pthread API, allows angle to store and retrieve thread specific
context object much more efficiently.
Bug: angleproject:4717
Change-Id: I27a117fe82e62407e01c8c372918b866aaea9ee5
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2231883
Commit-Queue: Mohan Maiya <m.maiya@samsung.com>
Reviewed-by: Courtney Goeltzenleuchter <courtneygo@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
0a6e118d
|
2020-01-27T13:37:29
|
|
Change g_Mutex from std::mutex to std::recursive_mutex
When running flatland on android-10.0.0_r21 (Pixel 3), libgui's ~EglImage calls
eglTerminate which grabs angle's EGL entry point mutex. The path continues
to libvulkan where eventually another egl call happens (eglDestroyImageKHR) and
it will attempt to take the mutex at the entry point again. So we try to get the
mutex multiple times from the same thread.
Change this mutex to a recursive_mutex to allow for this re-entry of EGL calls
Tests: android-10.0.0_r21/frameworks/native/cmds/flatland
Bug: angleproject:4354
Change-Id: If8a817df45e9f58d5f06884510350e17d7127fa9
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2029218
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Cody Northrop <cnorthrop@google.com>
|
|
9d737966
|
2019-08-14T12:25:12
|
|
Standardize copyright notices to project style
For all "ANGLE Project" copyrights, standardize to the format specified
by the style guide. Changes:
- "Copyright (c)" and "Copyright(c)" changed to just "Copyright".
- Removed the second half of date ranges ("Y1Y1-Y2Y2"->"Y1Y1").
- Fixed a small number of files that had no copyright date using the
initial commit year from the version control history.
- Fixed one instance of copyright being "The ANGLE Project" rather than
"The ANGLE Project Authors"
These changes are applied both to the copyright of source file, and
where applicable to copyright statements that are generated by
templates.
BUG=angleproject:3811
Change-Id: I973dd65e4ef9deeba232d5be74c768256a0eb2e5
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1754397
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
a2ec1344
|
2019-08-13T14:13:35
|
|
Add shared mutex to frontend
Add shared mutex to frontend API entrypoints that is only enabled when
ANGLE is being used with a shared context.
Bug: angleproject:2464
Change-Id: I0d918e37d9579dccd013dc88f563bed7de7ee55f
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1685712
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Courtney Goeltzenleuchter <courtneygo@google.com>
Commit-Queue: Courtney Goeltzenleuchter <courtneygo@google.com>
|
|
067687f4
|
2019-05-29T12:48:49
|
|
Removal global locks from GL entry points. Always lock in EGL.
The ANGLE Vulkan backend is now thread safe for non-share group contexts. This
means that a global GL lock only adds overhead for most use cases.
Remove the angle_force_thread_safety gn argument.
BUG=angleproject:2464
Change-Id: Ic6ba89e18b46e5dd72aa83d0f409097441fcca3a
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/1635749
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Tobin Ehlis <tobine@google.com>
|
|
|
b980c563
|
2018-11-27T11:34:27
|
|
Reformat all cpp and h files.
This applies git cl format --full to all ANGLE sources.
Bug: angleproject:2986
Change-Id: Ib504e618c1589332a37e97696cdc3515d739308f
Reviewed-on: https://chromium-review.googlesource.com/c/1351367
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
77db6f4a
|
2018-09-23T17:28:55
|
|
Use global Context lookup in single-threaded cases.
In single-threaded cases we don't need to use a TLS lookup for the
current Context.
Assume a global context has 3 states: unset, set, and multi-threaded:
* global context unset -> set the global context
* global context set, matches TLS context -> set the global context
* global context set, doesn't match TLS -> set multi-threaded
* global context multi-threaded -> leave it
This also inlines the global context/TLS check into global_state.h.
Increases performance of the binding perf overhead test by up to 25%
and the draw call perf test by up to 3%.
Bug: angleproject:2824
Change-Id: Ib95953ca918f2408fe72ca83061f4e2d1bd1adad
Reviewed-on: https://chromium-review.googlesource.com/1231993
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
|
|
|
b02fc662
|
2018-08-21T09:48:01
|
|
Lock around all EGL and GL calls with a global mutex.
BUG=angleproject:2464
Change-Id: I0231cc84777272f9cf26298c6a137f1ad3fd51d6
Reviewed-on: https://chromium-review.googlesource.com/1183441
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
|
|
|
75359664
|
2018-04-11T01:42:27
|
|
Implement EGL_KHR_debug.
BUG=angleproject:1618
Change-Id: I790944b49badc910b6c72266469fcb8e86ac4252
Reviewed-on: https://chromium-review.googlesource.com/1019387
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
e1a057e8
|
2018-06-07T15:09:00
|
|
Fully format some files.
Change-Id: Id6ea245849696d4c6d7eabc6860c0ac424dd8013
Reviewed-on: https://chromium-review.googlesource.com/1091309
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
|
|
|
7b8f3c9b
|
2016-11-10T10:42:48
|
|
Encapsulate the thread local storage into an egl::Thread class.
Instead of having separate GetGlobal* functions, interact with the global
objects through a single Thread object. This reduces the number of TLS
lookups in many EGL functions and allows the Thread object to be passed
down to other objects if needed.
BUG=angleproject:1618
Change-Id: I1f9a89e8899d637633f4e91fda0e38ac308dd020
Reviewed-on: https://chromium-review.googlesource.com/409637
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
|
|
|
e7c6e43c
|
2014-12-03T14:48:07
|
|
Implement all entry points in libGLES and have libEGL act as a shim.
This allows libANGLE to only be included in libGLESv2 and moves all TLS
data to libGLESv2.dll.
BUG=angle:733
Change-Id: I34f0b47987a5efbe906c290d3ca656142e69ea9a
Reviewed-on: https://chromium-review.googlesource.com/232962
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>
|