Commit graph

14434 commits

Author SHA1 Message Date
ReinUsesLisp
6d3a046caa query_cache: Address feedback 2020-02-14 17:38:27 -03:00
ReinUsesLisp
54a00ee4cf query_cache: Fix ambiguity in CacheAddr getter 2020-02-14 17:38:27 -03:00
ReinUsesLisp
cc0694559f query_cache: Add a recursive mutex for concurrent usage 2020-02-14 17:38:27 -03:00
ReinUsesLisp
bcd348f238 vk_query_cache: Implement generic query cache on Vulkan 2020-02-14 17:38:27 -03:00
ReinUsesLisp
c31382ced5 query_cache: Abstract OpenGL implementation
Abstract the current OpenGL implementation into the VideoCommon
namespace and reimplement it on top of that. Doing this avoids repeating
code and logic in the Vulkan implementation.
2020-02-14 17:38:27 -03:00
ReinUsesLisp
73d2d3342d gl_query_cache: Optimize query cache
Use a custom cache instead of relying on a ranged cache.
2020-02-14 17:38:27 -03:00
ReinUsesLisp
aae8c180cb gl_query_cache: Implement host queries using a deferred cache
Instead of waiting immediately for executed commands, defer the query
until the guest CPU reads it. This way we get closer to what the guest
program is doing.

To archive this we have to build a dependency queue, because host APIs
(like OpenGL and Vulkan) use ranged queries instead of counters like
NVN.

Waiting for queries implicitly uses fences and this requires a command
being queued, otherwise the driver will lock waiting until a timeout. To
fix this when there are no commands queued, we explicitly call glFlush.
2020-02-14 17:33:13 -03:00
ReinUsesLisp
ef9920e164 gl_rasterizer: Sort method declarations 2020-02-14 17:27:17 -03:00
ReinUsesLisp
fe1238be7a gl_rasterizer: Add queued commands counter
Keep track of the queued OpenGL commands that can signal a fence if
waited on. As a side effect, we avoid calls to glFlush when no commands
are queued.
2020-02-14 17:27:17 -03:00
ReinUsesLisp
2b58652f08 maxwell_3d: Slow implementation of passed samples (query 21)
Implements GL_SAMPLES_PASSED by waiting immediately for queries.
2020-02-14 17:27:17 -03:00
bunnei
f552d553ba
Merge pull request #3401 from FernandoS27/synchronization
Set of refactors for Kernel Synchronization and Hardware Constants
2020-02-14 14:40:20 -05:00
Morph
c3d0a0d627 Add 4:3 aspect ratio and address feedback 2020-02-14 14:39:04 -05:00
bunnei
63a59b9935
Merge pull request #3379 from ReinUsesLisp/cbuf-offset
shader/decode: Fix constant buffer offsets
2020-02-14 13:22:53 -05:00
Zach Hilman
4501bd8ca9
Merge pull request #3398 from brianclinkenbeard/fix-cmake-sdl2
Use config mode for finding SDL2 with CMake
2020-02-14 09:11:47 -05:00
Fernando Sahmkow
829d8c0d6b Core: Correct compilition in GCC 2020-02-14 05:53:30 -04:00
Morph
20dc2e3622 Address feedback 2020-02-14 00:06:26 -05:00
Morph
22f58cca5e Use enumeration instead of magic numbers 2020-02-13 23:13:23 -05:00
Morph
27e19f87c6 Add following aspect ratios: 16:9, 21:9, Stretch to Window
Available as a drop down within the configure graphics tab.
2020-02-13 22:17:28 -05:00
bunnei
74feed372c
Merge pull request #3400 from makigumo/patch-1
update hwopus DecodeInterleaved for FW 7.0.0+
2020-02-13 21:26:13 -05:00
ReinUsesLisp
3217400dd1 gl_resource_manager: Add managed query class 2020-02-13 22:25:55 -03:00
bunnei
3563af2364
Merge pull request #3395 from FernandoS27/queries
GPU: Refactor queries implementation and correct GPU Clock.
2020-02-13 20:18:26 -05:00
ReinUsesLisp
336a4f8e99 gl_rasterizer: Use the least generic OpenGL draw function possible
This may help some implementations.
2020-02-13 21:55:21 -03:00
ReinUsesLisp
cbea8c74de vk_shader_decompiler: Fix vertex id and instance id
Vulkan's VertexIndex and InstanceIndex don't match with hardware. This
is because Nvidia implements gl_VertexID and gl_InstanceID. The math
that relates these is:

gl_VertexIndex = gl_BaseVertex + gl_VertexID
gl_InstanceIndex = gl_InstanceIndex + gl_InstanceID

To emulate it using what Vulkan's SPIR-V offers (the *Index variants)
this commit substracts gl_Base* from gl_*Index to obtain the OpenGL and
hardware's equivalent.
2020-02-13 20:25:28 -03:00
Fernando Sahmkow
2bc949628d Core: Address Feedback 2020-02-13 19:10:33 -04:00
Fernando Sahmkow
d6ed31b9fa GPU: Address Feedback. 2020-02-13 18:16:07 -04:00
bunnei
8b9a56033a
Merge pull request #3405 from lioncash/thread
address_arbiter: Minor cleanup to list querying
2020-02-12 21:46:00 -05:00
Brian Clinkenbeard
0d85b6bfe1 Merge branch 'master' into fix-cmake-sdl2 2020-02-12 16:07:07 -08:00
Lioncash
be269e21a5 address_arbiter: Collapse loops in InsertThread() and RemoveThread()
Same behavior, but without the need to explicitly loop through
everything manually.
2020-02-12 15:34:07 -05:00
Lioncash
9f2c703137 address_arbiter: Simplify GetThreadsWaitingOnAddress()
Simplifies the overall function and also allows for it to become a
const-qualified member function.
2020-02-12 15:10:16 -05:00
bunnei
8f8dda2d5b
Merge pull request #3403 from lioncash/debug
bcat/backend: Prevent fmt exception in debug log within NullBackend::Clear()
2020-02-12 11:17:43 -05:00
bunnei
2506f7b3a1
Merge pull request #3402 from lioncash/sys-global
kernel/thread: Remove trivial usages of the global system accessor
2020-02-12 10:10:00 -05:00
Lioncash
f00a54f508 bcat/backend: Make formatting of passphrase consistent in NullBackend::SetPassphrase()
Aligns the '=' to be consistent with the rest of the logs within this
source file.
2020-02-12 01:18:29 -05:00
Lioncash
eefd97e80d bcat/backend: Prevent fmt exception in debug log within NullBackend::Clear()
A formatting specifier within Clear wasn't being used, which will cause
fmt to throw an exception. This fixes that.
2020-02-12 01:14:47 -05:00
Lioncash
b80c348b09 kernel/thread: Remove trivial usages of the global system accessor
We can just use the kernel member variable directly instead of going
through the system to obtain the same thing.
2020-02-12 01:00:41 -05:00
Fernando Sahmkow
1e6f8aba04 Core: Set all hardware emulation constants in a single file. 2020-02-11 20:19:11 -04:00
Fernando Sahmkow
d23d504d77 Kernel: Refactor synchronization to better match RE 2020-02-11 18:47:31 -04:00
makigumo
926ea5a16d
update hwopus DecodeInterleaved for FW 7.0.0+
trivial change,
see https://switchbrew.org/wiki/Audio_services#IHardwareOpusDecoder
2020-02-11 18:41:04 +01:00
Fernando Sahmkow
c5aefe42aa Kernel: Change WaitObject to Synchronization object. In order to better reflect RE. 2020-02-11 10:46:25 -04:00
bunnei
37f1cf8cbd
Merge pull request #3376 from ReinUsesLisp/point-sprite
gl_rasterizer: Implement GL_POINT_SPRITE
2020-02-11 08:26:07 -05:00
Brian Clinkenbeard
68043dd233 use config mode for finding SDL2 with CMake 2020-02-10 19:56:33 -08:00
Fernando Sahmkow
8e9a4944db GPU: Implement GPU Clock correctly. 2020-02-10 10:44:54 -04:00
Fernando Sahmkow
0cb3bcfbb7 Maxwell3D: Correct query reporting. 2020-02-10 10:41:43 -04:00
bunnei
84ea9c2b42 Merge pull request #3372 from ReinUsesLisp/fix-back-stencil
maxwell_3d: Fix stencil back mask
2020-02-09 22:29:28 -05:00
Zach Hilman
21c3f48279
Merge pull request #3391 from Morph1984/remove-unknown
Remove option "Show files with type 'Unknown'"
2020-02-09 12:08:01 -05:00
Morph
fcf3425b1b Remove option "Show files with type 'Unknown'" 2020-02-09 11:30:02 -05:00
bunnei
a952fbc5b3
Merge pull request #3388 from bunnei/service-shared-ptr
hle: services: Use std::shared_ptr instead of copy by value.
- This is a prerequisite to adding a mutex to `ServiceFramework`, which cannot be copied.
- This will be used for threaded services.
2020-02-08 21:35:30 -05:00
bunnei
e210835dd0
Merge pull request #3387 from bunnei/gpu-mpscqueue
gpu_thread: Use MPSCQueue for GPU commands.
2020-02-08 21:15:48 -05:00
bunnei
6536cc9741
Merge pull request #3386 from bunnei/gpu-mem-interface
video_core: memory_manager: Use GPU interface for cache functions.
2020-02-08 21:15:27 -05:00
bunnei
7b07e521ca hle: services: Use std::shared_ptr instead of copy by value. 2020-02-07 23:02:26 -05:00
bunnei
b5c13ee0eb gpu_thread: Use MPSCQueue for GPU commands.
- Necessary for multiple service threads.
2020-02-07 23:01:23 -05:00