Compare commits

..

182 Commits

Author SHA1 Message Date
Lioncash
cba78dc70a services/fsp_srv: Amend service function table
Adds new functions that have been given names to the table. Information
is based off what is provided on Switchbrew.
2018-10-02 21:34:33 -04:00
bunnei
bc679c9b8c Merge pull request #1330 from raven02/tlds
TLDS: Add 1D sampler
2018-09-30 21:53:38 -04:00
bunnei
7c61f0c322 Merge pull request #1420 from MerryMage/dynarmic
externals: Update dynarmic to 4e6848d
2018-09-30 21:25:35 -04:00
bunnei
8391048a83 Merge pull request #1322 from bunnei/tex-cubemap
gl_rasterizer_cache: Implement cubemap textures.
2018-09-30 21:19:00 -04:00
bunnei
decc319634 Merge pull request #1403 from DarkLordZach/install-sysnand
qt: Install System TitleTypes to System NAND
2018-09-30 21:08:44 -04:00
bunnei
8f2ad3a66d Merge pull request #1338 from raven02/service_vi
Implement ISystemDisplayService::GetDisplayMode
2018-09-30 15:39:54 -04:00
bunnei
5e2f23e2b1 Merge pull request #1417 from lioncash/context
svc: Implement svcGetThreadContext
2018-09-30 15:32:27 -04:00
bunnei
df3799a008 gl_rasterizer_cache: Fixes to how we do render to cubemap.
- Fixes issues with Splatoon 2.
2018-09-30 15:10:14 -04:00
MerryMage
3031aa9d27 externals: Update dynarmic to 4e6848d
4e6848d A32/ir_emitter: Bugfix: ExceptionRaised was producing incorrect PC
41ba9fd value: Move ImmediateToU64() to be a part of Value's interface
c6a6271 reg_alloc: Emit AVX instructions where able
aedd32a abi: Emit AVX instructions where able
f2d9337 a64_exclusive_monitor: Loosen memory ordering requirements
7ca709d travis: Make macOS builds use Xcode 10
14dd45e Fix VShift terminology
88554c4 emit_x64_vector: AVX512 implementation of EmitVectorLogicalVShiftS16
ab4e316 emit_x64_vector: AVX512 implementation of EmitVectorLogicalVShiftS64
0ea84f3 emit_x64_vector: AVX2 implementation of EmitVectorLogicalVShiftS32
c77a2c5 emit_x64_vector: AVX512 implementation of EmitVectorLogicalVShiftU16()
e9441fd emit_x64_vector: AVX2 implementation of EmitVectorLogicalVShiftU64()
0e9c33c emit_x64_vector: AVX2 implementation of EmitVectorLogicalVShiftU32()
8f85274 emit_x64_vector: SSSE3 variant of EmitVectorCountLeadingZeros8()
be05e75 Merge pull request #397 from VelocityRa/dec-shift-fix
bc328fc decoders: Cast to correctly-sized type before shifting
9c3d2d1 a64_emit_x64: Lowercase PAGE_SIZE
f538d29 emit_x64_vector_floating_point: SSE4.1 implementation of EmitFPVectorToFixed
1603a6e emit_x64_vector_floating_point: EmitFPVectorRoundInt: Use FCODE
2e1ccaf emit_x64_vector: AVX implementation for EmitVectorCountLeadingZeros8
555bfda emit_x64_vector: SSE implementation of EmitVectorCountLeadingZeros16
71c2589 externals: Update Xbyak to 5.73
1ec1b2f Squashed 'externals/xbyak/' changes from 1de435ed..42462ef9
2018-09-30 19:45:27 +01:00
bunnei
29782273ec gl_rasterizer_cache: Add check for array rendering to cubemap texture. 2018-09-30 14:31:58 -04:00
bunnei
f543b43fd0 gl_rasterizer_cache: Implement render to cubemap. 2018-09-30 14:31:58 -04:00
bunnei
15cc729ebd gl_shader_decompiler: TEXS: Implement TextureType::TextureCube. 2018-09-30 14:31:58 -04:00
bunnei
ed2e0e85c9 gl_rasterizer_cache: Add support for SurfaceTarget::TextureCubemap. 2018-09-30 14:31:58 -04:00
bunnei
871580dcd8 gl_rasterizer_cache: Implement LoadGLBuffer for Texture2DArray. 2018-09-30 14:31:57 -04:00
bunnei
a9aa1db552 gl_rasterizer_cache: Update BlitTextures to support non-Texture2D ColorTexture surfaces. 2018-09-30 14:31:57 -04:00
bunnei
2e1cdde994 gl_rasterizer_cache: Track texture target and depth in the cache. 2018-09-30 14:31:57 -04:00
bunnei
fefb003b23 gl_rasterizer_cache: Workaround for Texture2D -> Texture2DArray scenario. 2018-09-30 14:31:56 -04:00
bunnei
ce452049d3 gl_rasterizer_cache: Keep track of surface 2D size separately from total size. 2018-09-30 14:31:56 -04:00
raven02
b92b4bbeaf Fix trailing whitespace 2018-09-30 23:51:10 +08:00
Merry
f8e46d335f Merge pull request #1418 from FearlessTobi/port-4269
Port citra-emu/citra#4269: "OSX: Set MACOSX_DEPLOYMENT_TARGET to 10.13"
2018-09-30 13:51:21 +01:00
B3n30
56901912cb OSX: Set MACOSX_DEPLOYMENT_TARGET to 10.13 2018-09-30 13:34:49 +02:00
Lioncash
541c550753 kernel/svc: Implement svcGetThreadContext()
Now that we have all of the rearranging and proper structure sizes in
place, it's fairly trivial to implement svcGetThreadContext(). In the
64-bit case we can more or less just write out the context as is, minus
some minor value sanitizing. In the 32-bit case we'll need to clear out
the registers that wouldn't normally be accessible from a 32-bit
AArch32 exectuable (or process).
2018-09-30 05:29:40 -04:00
Lioncash
dccfe193a9 kernel/process: Add a data member to determine if a process is 64-bit or not.
This will be necessary for the implementation of svcGetThreadContext(),
as the kernel checks whether or not the process that owns the thread
that has it context being retrieved is a 64-bit or 32-bit process.

If the process is 32-bit, then the upper 15 general-purpose registers
and upper 16 vector registers are cleared to zero (as AArch32 only has
15 GPRs and 16 128-bit vector registers. not 31 general-purpose
registers and 32 128-bit vector registers like AArch64).
2018-09-30 05:29:40 -04:00
Lioncash
cf9d6c6f52 kernel/process: Make data member variables private
Makes the public interface consistent in terms of how accesses are done
on a process object. It also makes it slightly nicer to reason about the
logic of the process class, as we don't want to expose everything to
external code.
2018-09-30 02:30:01 -04:00
Lioncash
16145e2f21 arm_interface: Add missing fpsr/tpidr members to the ThreadContext struct
Internally within the kernel, it also includes a member variable for the
floating-point status register, and TPIDR, so we should do the same here to match
it.

While we're at it, also fix up the size of the struct and add a static
assertion to ensure it always stays the correct size.
2018-09-30 02:29:57 -04:00
raven02
4092907687 Implement ISystemDisplayService::GetDisplayMode 2018-09-30 10:04:03 +08:00
bunnei
334090fe6f Merge pull request #1414 from lioncash/ref
loader: Make the Load() function take a process as a regular reference, not a SharedPtr
2018-09-29 16:57:02 -04:00
Lioncash
a63e6f9dfd loader: Make the Load() function take a process as a regular reference, not a SharedPtr
A process should never require being reference counted in this
situation. If the handle to a process is freed before this function is
called, it's definitely a bug with our lifetime management, so we can
put the requirement in place for the API that the process must be a
valid instance.
2018-09-29 16:00:03 -04:00
bunnei
0921558a9f Merge pull request #1388 from FearlessTobi/port-4258
Port citra-emu/citra#4258: "Meta: Add gitattributes file"
2018-09-29 12:01:15 -04:00
bunnei
97c0ac3545 Merge pull request #1412 from lioncash/move
kernel/object: Remove unnecessary std::move from DynamicObjectCast()
2018-09-29 11:58:58 -04:00
bunnei
fe5962e073 Merge pull request #1411 from ReinUsesLisp/point-size
video_core: Implement point_size and add point state sync
2018-09-29 11:58:39 -04:00
bunnei
8d8366b602 Merge pull request #1406 from ReinUsesLisp/multibind-samplers
gl_state: Pack sampler bindings into a single ARB_multi_bind
2018-09-29 10:55:45 -04:00
bunnei
f7b69d61f2 Merge pull request #1395 from lioncash/vm
process/vm_manager: Initial modifications to load NPDM metadata
2018-09-29 10:54:39 -04:00
Lioncash
f4c24d0832 kernel/object: Remove unnecessary std::move from DynamicObjectCast()
boost::static_pointer_cast for boost::intrusive_ptr (what SharedPtr is),
takes its parameter by const reference. Given that, it means that this
std::move doesn't actually do anything other than obscure what the
function's actual behavior is, so we can remove this. To clarify, this
would only do something if the parameter was either taking its argument
by value, by non-const ref, or by rvalue-reference.
2018-09-28 02:17:57 -04:00
ReinUsesLisp
e3e51d3ddb video_core: Implement point_size and add point state sync 2018-09-28 02:13:29 -03:00
ReinUsesLisp
b8f1506aa5 gl_state: Pack sampler bindings into a single ARB_multi_bind 2018-09-28 02:04:22 -03:00
bunnei
f7da74d18e Merge pull request #1360 from FearlessTobi/port-3979
Port citra-emu/citra#3979 game_list: move SearchField to game_list_p.h and fix untranslated text
2018-09-27 17:09:11 -04:00
bunnei
fc2419e441 Merge pull request #1394 from lioncash/stream
stream: Preserve enum class type in GetState()
2018-09-27 17:05:03 -04:00
Mat M
29ff84ea99 Merge pull request #1389 from PhiBabin/valgrind
FPCR register was uninitialized at start up
2018-09-27 15:23:03 -04:00
bunnei
a75740b298 Merge pull request #1397 from spycrab/cmake_bin
CMake: Remove superfluous CMAKE_RUNTIME_OUTPUT_DIRECTORY assignment
2018-09-27 10:00:29 -04:00
bunnei
62048edc15 Merge pull request #1377 from FernandoS27/faster-swizzle
Improved Fast Swizzle and Legacy Swizzle
2018-09-27 10:00:04 -04:00
bunnei
8731ac87fa Merge pull request #1404 from lioncash/vfs
fsmitm_romfsbuild: Minor changes
2018-09-27 09:58:56 -04:00
Zach Hilman
70e86248fd qt: Install System TitleTypes to System NAND
Fixes an issue where installed system archive NCAs would be installed to user NAND and not recognized by games.
2018-09-27 09:20:06 -04:00
Lioncash
861580f6d2 fsmitm_romfsbuild: std::move std::vector instances in Build()
Avoids making copies of large std::vector instances where it's trivially
avoidable to do so.
2018-09-26 17:35:44 -04:00
Lioncash
7ecdaaf189 fsmitm_romfsbuild: Replace manual value aligning with Common::AlignUp()
Theres no need to do explicit bitwise arithmetic here, when we have a
function that does this with a more descriptive name.
2018-09-26 17:35:21 -04:00
bunnei
c0445006af Merge pull request #1399 from lioncash/sched
kernel/scheduler: Take ARM_Interface instances by reference
2018-09-26 16:17:18 -04:00
bunnei
efcb83fb41 Merge pull request #1400 from lioncash/header
service: Add missing headers inclusions where applicable
2018-09-26 16:11:19 -04:00
bunnei
cc866d1384 Merge pull request #1402 from ReinUsesLisp/asserts
video_core: Add asserts for CS, TFB and alpha testing
2018-09-26 16:10:55 -04:00
bunnei
92dd496fb9 Merge pull request #1401 from lioncash/vfs
vfs: Minor cleanup related changes to layered VFS code
2018-09-26 16:09:16 -04:00
bunnei
9d163c00cc Merge pull request #1398 from lioncash/macos
travis: Make macOS builds utilize Xcode 10
2018-09-26 16:08:39 -04:00
Lioncash
11104b4883 patch_manager: Invert conditionals within ApplyLayeredFS()
Avoids the need to nest code quite a bit by early-exiting in error
cases.
2018-09-25 20:09:23 -04:00
ReinUsesLisp
ab65fde9f4 video_core: Add asserts for CS, TFB and alpha testing
Add asserts for compute shader dispatching, transform feedback being
enabled and alpha testing. These have in common that they'll probably break
rendering without logging.
2018-09-25 21:07:00 -03:00
Lioncash
e3b2ef9170 vfs_vector: Amend initializer list order in VectorVfsFile's constructor initializer list
Orders the initializer list members to be in the same order that they
would be initialized in. Avoids compiler warnings.
2018-09-25 20:06:21 -04:00
Lioncash
4654f89618 fsmitm_romfsbuild: Avoid type truncation warnings
Cast where explicitly necessary and in other cases we can simply modify
the algorithm to accomodate larger data.
2018-09-25 20:06:21 -04:00
Lioncash
91b56c4928 fsmitm_romfsbuild: Remove unnecessary constructors and initializers for RomFSBuildFileContext and RomFSBuildDirectoryContext
There's no need to duplicate in-class initializers with a constructor
initializer list. std::strings also initialize to empty by default.
2018-09-25 20:06:21 -04:00
Lioncash
1f92cbc059 fsmitm_romfsbuild: Remove unnecessary loops in Build()
The std::vector instances are already initially allocated with all
entries having these values, there's no need to loop through and fill
them with it again when they aren't modified.
2018-09-25 20:06:21 -04:00
Lioncash
fa9e0f9c8b fsmitm_romfsbuild: Make auto variable into a std::size_t variable within Build()
auto x = 0;

auto-deduces x to be an int. This is undesirable when working with
unsigned values. It also causes sign conversion warnings. Instead, we
can make it a proper unsigned value with the correct width that the
following expressions operate on.
2018-09-25 20:06:21 -04:00
Lioncash
f646ca874d yuzu/main: Resolve precedence bug within CalculateRomFSEntrySize()
Ternary operators have a lower precedence than arithmetic operators, so
what was actually occurring here is "return (out + full) ? x : y" which most
definitely isn't intended, given we calculate out recursively above. We
were essentially doing a lot of work for nothing.
2018-09-25 20:06:21 -04:00
Lioncash
cbb146069a yuzu/main: Move functions stored into static std::function instances out of OnGameListDumpRomFS()
This can cause warnings about static constructors, and is also not ideal
performance-wise due to the indirection through std::function. This also
keeps the behavior itself separate from the surrounding code, which can
make it nicer to read, due to the size of the code.
2018-09-25 20:06:21 -04:00
Lioncash
57616f9758 vfs/etc: Append std:: to size_t usages
Given we just recently had a patch backport this from citra, let's try
and keep the convention uniform.
2018-09-25 20:06:21 -04:00
Lioncash
28bef31ea8 vfs_concat/vfs_layered: Remove friend declarations from ConcatenatedVfsFile
Given these are only added to the class to allow those functions to
access the private constructor, it's a better approach to just make them
static functions in the interface, to make the dependency explicit.
2018-09-25 20:06:01 -04:00
Lioncash
14e2df5610 vfs_static: Remove template byte parameter from StaticVfsFile
This converts it into a regular constructor parameter. There's no need
to make this a template parameter on the class when it functions
perfectly well as a constructor argument.

This also reduces the amount of code bloat produced by the compiler, as
it doesn't need to generate the same code for multiple different
instantiations of the same class type, but with a different fill value.
2018-09-25 17:40:53 -04:00
Lioncash
e42bb5e003 service: Add missing headers inclusions where applicable
Gets rid of a few indirect inclusions.
2018-09-25 17:14:38 -04:00
bunnei
7b81e1e525 Merge pull request #1365 from DarkLordZach/lfs
file_sys: Add support for LayeredFS mods
2018-09-25 16:59:44 -04:00
Lioncash
598e4d2f6c core_cpu: Make arm_interface instances a std::unique_ptr
This is only exposed by reference, so we can just make it a unique
pointer to get rid of the need to also use reference counting for the
pointer.
2018-09-25 16:04:56 -04:00
Lioncash
a58eefa7e4 kernel/scheduler: Take ARM_Interface instance by reference in the constructor
It doesn't make sense to allow a scheduler to be constructed around a
null pointer.
2018-09-25 16:00:17 -04:00
Lioncash
5873b14efd travis: Make macOS builds utilize Xcode 10
Keeps the used toolchain up to date and finally allows the use of
<optional> and <variant> standard library headers within the codebase.
2018-09-25 14:46:46 -04:00
spycrab
1008be9b67 CMake: Remove superfluous CMAKE_RUNTIME_OUTPUT_DIRECTORY assignment 2018-09-25 17:55:08 +02:00
bunnei
b67c1fdf38 Merge pull request #1393 from tech4me/svc
svc: Updated svc names
2018-09-25 10:47:12 -04:00
Lioncash
83377113bf memory: Dehardcode the use of fixed memory range constants
The locations of these can actually vary depending on the address space
layout, so we shouldn't be using these when determining where to map
memory or be using them as offsets for calculations. This keeps all the
memory ranges flexible and malleable based off of the virtual memory
manager instance state.
2018-09-24 22:16:03 -04:00
Lioncash
6c6f95d071 svc: Report correct memory-related values within some of the cases in svcGetInfo()
Previously, these were reporting hardcoded values, but given the regions
can change depending on the requested address spaces, these need to
report the values that the memory manager contains.
2018-09-24 22:16:03 -04:00
Lioncash
7fd598636e memory: Dehardcode the use of a 36-bit address space
Given games can also request a 32-bit or 39-bit address space, we
shouldn't be hardcoding the address space range as 36-bit.
2018-09-24 22:15:53 -04:00
Lioncash
75603b005b process/vm_manager: Amend API to allow reading parameters from NPDM metadata
Rather than hard-code the address range to be 36-bit, we can derive the
parameters from supplied NPDM metadata if the supplied exectuable
supports it. This is the bare minimum necessary for this to be possible.

The following commits will rework the memory code further to adjust to
this.
2018-09-24 17:24:50 -04:00
David
367c52ff0d Implemented fatal:u properly (#1347)
* Implemented fatal:u properly

fatal:u now is properly implemented with all the ipc cmds. Error reports/Crash reports are also now implemented for fatal:u. Crash reports save to yuzu/logs/crash_reports/
The register dump is currently known as sysmodules send all zeros. If there are any non zero values for the "registers" or the unknown values, let me know!

* Fatal:U fixups

* Made fatal:u execution break more clear

* Fatal fixups
2018-09-23 22:34:11 -04:00
David
2513e086ab Stubbed IRS (#1349)
* Stubbed IRS

Currently we have no ideal way of implementing IRS. For the time being we should have the functions stubbed until we come up with a way to emulate IRS properly.

* Added IRS to logging backend

* Forward declared shared memory for irs
2018-09-23 22:33:29 -04:00
bunnei
f2c1fd08f9 Merge pull request #1354 from ogniK5377/ssl-version
Corrected SSL::SetInterfaceVersion
2018-09-23 22:32:14 -04:00
Zach Hilman
b3c2ec362b fsmitm: Cleanup and modernize fsmitm port 2018-09-23 21:50:20 -04:00
Lioncash
2f6a611311 stream: Preserve enum class type in GetState()
Preserves the meaning/type-safetiness of the stream state instead of
making it an opaque u32. This makes it usable for other things outside
of the service HLE context.
2018-09-23 20:03:38 -04:00
tech4me
d42424ace0 svc: Updated svc names 2018-09-23 17:03:38 -07:00
David
9f3fc067bf Added glObjectLabels for renderdoc for textures and shader programs (#1384)
* Added glObjectLabels for renderdoc for textures and shader programs

* Changed hardcoded "Texture" name to reflect the texture type instead

* Removed string initialize
2018-09-23 17:55:41 -04:00
bunnei
6b05f71b67 Merge pull request #1387 from FearlessTobi/port-4245
Port citra-emu/citra#4245: "common/thread: remove YieldCPU()"
2018-09-23 17:51:53 -04:00
bunnei
5d9f001bb5 Merge pull request #1385 from FearlessTobi/port-4214
Port citra-emu/citra#4214: "Set citra-qt project as default StartUp Project in Visual Studio"
2018-09-23 17:49:27 -04:00
bunnei
d373a65d26 Merge pull request #1391 from ogniK5377/GetAudioRendererState
Added audren:u#GetAudioRendererState
2018-09-23 16:25:12 -04:00
bunnei
c6f5c7d291 Merge pull request #1392 from greggameplayer/correct-BC6H
Correction of the ComponentType of BC6H_UF16 & BC6H_SF16
2018-09-23 16:23:56 -04:00
greggameplayer
b91e2d55f3 correct BC6H 2018-09-23 19:17:22 +02:00
David Marcec
c461188f51 Added audren:u#GetAudioRendererState 2018-09-23 22:32:01 +10:00
Philippe Babin
fb6bc2c495 FPCR register was uninitialized at start up 2018-09-22 21:17:43 -04:00
bunnei
93fea4e179 Merge pull request #1378 from lioncash/thread
process: Make a few member variables private where applicable
2018-09-22 19:46:53 -04:00
James Rowe
05db48b3f8 Meta: Add gitattributes file
Github Linguist will read this file when calculating language stats for
the repository. We can use this to exclude any vendored dependencies in
externals and dist. Also makes all h files be considered cpp
2018-09-22 23:31:44 +02:00
Weiyi Wang
e0ce07aa7d common/thread: remove YieldCPU()
simply use the standard library yield()
2018-09-22 23:27:42 +02:00
James Rowe
5659d5d258 Merge pull request #1386 from jroweboy/oops
Build: Reintroduce Appveyor deploy
2018-09-22 13:15:20 -06:00
James Rowe
b23b30f9ef Build: Reintroduce Appveyor deploy
This was erroneously removed when we added travis mingw builds. We still
want to push MSVC builds to github for the time being
2018-09-22 13:09:32 -06:00
fearlessTobi
f90b75cc4c Port citra-emu/citra#4214: "Set citra-qt project as default StartUp Project in Visual Studio" 2018-09-22 12:24:02 +02:00
bunnei
fb9f273e90 Merge pull request #1380 from lioncash/const
shader_bytecode: Make operator== and operator!= of IpaMode const qualified
2018-09-22 01:37:36 -04:00
bunnei
73eea61614 Merge pull request #1382 from lioncash/inc
gl_state: Remove unused type alias
2018-09-22 01:36:21 -04:00
bunnei
0246db9ea2 Merge pull request #1376 from Subv/timestretch_trace
Logging: Change the TimeStretch::Process log from debug to trace level.
2018-09-22 01:30:28 -04:00
bunnei
2d5f51bb9b Merge pull request #1381 from valentinvanelslande/patch-1
yuzu_cmd: fix sd card directory set to nand directory
2018-09-22 01:29:28 -04:00
James Rowe
37bb2c45ee Merge pull request #1383 from DarkLordZach/game-list-interpolation
game_list: Add Qt SmoothTransformation to picture scaling
2018-09-21 20:55:49 -06:00
Zach Hilman
4f183123f5 game_list: Add Qt SmoothTransformation to picture scaling 2018-09-21 21:34:46 -04:00
Zach Hilman
ba0873d33c qt: Add UI elements for LayeredFS and related tools 2018-09-21 19:53:33 -04:00
Zach Hilman
050547b801 romfs: Implement CreateRomFS 2018-09-21 19:53:33 -04:00
Zach Hilman
6eada3c57d file_sys: Port Atmosphere-NX fs_mitm implementation 2018-09-21 19:53:33 -04:00
Zach Hilman
940a711caf filesystem: Add LayeredFS VFS directory getter 2018-09-21 19:53:33 -04:00
Zach Hilman
50a470eab8 bis_factory: Add mod directory VFS getter 2018-09-21 19:53:33 -04:00
Zach Hilman
16188acb50 patch_manager: Add LayeredFS mods support 2018-09-21 19:53:33 -04:00
Zach Hilman
44fdac334c vfs_concat: Rewrite and fix ConcatenatedVfsFile 2018-09-21 19:53:33 -04:00
Zach Hilman
3e5c3d0f16 vfs_layered: Add LayeredVfsDirectory
Reads multiple dirs through as if a waterfall.
2018-09-21 19:53:05 -04:00
Zach Hilman
b52343a428 vfs_vector: Add VectorVfsFile
Maps a vector into the VFS interface.
2018-09-21 19:53:05 -04:00
Zach Hilman
c65d4d119f vfs_static: Add StaticVfsFile
Always returns the template argument byte for all reads. Doesn't support writes.
2018-09-21 19:53:05 -04:00
Zach Hilman
f68e324672 vfs: Add and rewite VfsRawCopy functions 2018-09-21 19:53:05 -04:00
Zach Hilman
d6cbb3a3e0 vfs: Add GetEntries method
Maps name string to directory or file.
2018-09-21 19:53:05 -04:00
Zach Hilman
bd8db3f7f8 common_paths: Add Load and Dump dirs 2018-09-21 19:53:05 -04:00
Lioncash
90746c33c7 gl_state: Remove unused type alias
This isn't used anywhere within the header, so we can remove it, along
with the include that was previously necessary. This also uncovers an
indirect include in the cpp file for the assertion macros.
2018-09-21 18:22:43 -04:00
Valentin Vanelslande
19d9b0778a Update config.cpp 2018-09-21 16:16:51 -05:00
Lioncash
a8f5fd787f shader_bytecode: Lay out the Ipa-related enums better
This is more consistent with the surrounding enums.
2018-09-21 16:17:31 -04:00
Lioncash
272517cf7e shader_bytecode: Make operator== and operator!= of IpaMode const qualified
These don't affect the state of the struct and can be const member
functions.
2018-09-21 16:17:27 -04:00
bunnei
4f186de069 Merge pull request #1379 from lioncash/bitwise
gl_stream_buffer: Fix use of bitwise OR instead of logical OR in Map()
2018-09-21 14:02:00 -04:00
zhupengfei
44228ee3ed game_list: move SearchField to game_list_p.h and fix untranslated text
I have tested and made sure the text is translatable, but this would require a translation update to take effect.
2018-09-21 18:47:41 +02:00
FernandoS27
57b44200a2 Reverse stride align restriction on FastSwizzle due to lost performance 2018-09-21 12:09:59 -04:00
FernandoS27
d2dd1289bd Join both Swizzle methods within one interface function 2018-09-21 11:42:34 -04:00
FernandoS27
41c6c4593a Standarized Legacy Swizzle to look alike FastSwizzle and use a Swizzling Table instead 2018-09-21 11:34:54 -04:00
FernandoS27
f020319a45 Remove same output bpp restriction on FastSwizzle 2018-09-21 11:10:44 -04:00
FernandoS27
68aaa83836 Improved Legacy Swizzler to be better documented and work better 2018-09-21 10:57:12 -04:00
Lioncash
ba02dd9ebc gl_stream_buffer: Fix use of bitwise OR instead of logical OR in Map()
This was very likely intended to be a logical OR based off the
conditioning and testing of inversion in one case.

Even if this was intentional, this is the kind of non-obvious thing one
should be clarifying with a comment.
2018-09-21 07:59:03 -04:00
Lioncash
48b2eda492 svc: Move most process termination code to its own function within Process
Reduces the use of Process class members externally and keeps most code
related to tearing down a process with the rest of the process code.
2018-09-21 06:07:41 -04:00
Lioncash
acfc801d14 thread/process: Move TLS slot marking/freeing to the process class
Allows making several members of the process class private, it also
avoids going through Core::CurrentProcess() just to retrieve the owning
process.
2018-09-21 03:50:12 -04:00
David
1db1e013e0 Added support for uncompressed NSOs (#1374)
* Added support for uncompressed NSOs

* Moved compressed section check to NsoHeader
2018-09-21 00:39:30 -04:00
James Rowe
8606995515 Merge pull request #1225 from tech4me/travis-windows
travis: running mingw build on travis ci
2018-09-20 21:39:15 -06:00
bunnei
000da01c1d Merge pull request #1337 from DarkLordZach/create-fs-cmd
yuzu-cmd: Add filesystem service initializer call
2018-09-20 23:36:11 -04:00
bunnei
0285ddfbd4 Merge pull request #1372 from lioncash/thread
kernel/thread: Use owner_process when setting the page table in SetupMainThread()
2018-09-20 23:35:29 -04:00
bunnei
072053ab95 Merge pull request #1371 from lioncash/fwd-arm
arm_interface: Replace kernel vm_manager include with a forward declaration
2018-09-20 23:35:06 -04:00
bunnei
a373dddbd9 Merge pull request #1375 from Subv/gl_clear
RasterizerGL: Use the correct framebuffer when clearing via the CLEAR_BUFFERS register.
2018-09-20 23:34:39 -04:00
Subv
8ba21e28cf Logging: Change the TimeStretch::Process log from debug to trace level.
This function is called too many times and makes the debug logging basically unusable due to the spam.
2018-09-20 22:33:54 -05:00
Subv
9cd5c61fcf RasterizerGL: Use the correct framebuffer when clearing via the CLEAR_BUFFERS register.
Previously we were clearing the default backbuffer framebuffer.

Found thanks to a Piglit test :)
2018-09-20 22:31:53 -05:00
bunnei
c1c59617ad Merge pull request #1364 from lioncash/content
file-sys: Default heavy-weight class destructors in the cpp file
2018-09-20 23:31:31 -04:00
bunnei
cf97d00eb6 Merge pull request #1367 from lioncash/plural
game_list: Handle plurals within setFilterResult() better
2018-09-20 23:30:32 -04:00
bunnei
b0b57c21e6 Merge pull request #1368 from ogniK5377/nifm-fix
Added IRequest::Submit
2018-09-20 23:30:11 -04:00
bunnei
a0db7e2cf9 Merge pull request #1352 from lioncash/sharing
ring_buffer: Use std::hardware_destructive_interference_size to determine alignment size for avoiding false sharing
2018-09-20 23:29:10 -04:00
FernandoS27
bf2f2a715f Improved fast swizzle and removed restrictions to it 2018-09-20 23:06:53 -04:00
bunnei
9f92533cc2 Merge pull request #1373 from ogniK5377/revert-nifm
Revert GetRequestState
2018-09-20 22:19:58 -04:00
David Marcec
b918925bd5 Revert GetRequestState
Even though setting this value to 3 is more correct. We break more games than we fix due to missing implementations. We should keep this as 0 for the time being
2018-09-21 12:15:49 +10:00
Lioncash
05aa4aa01a kernel/thread: Use owner_process when setting the page table in SetupMainThread()
The owning process of a thread is required to exist before the thread,
so we can enforce this API-wise by using a reference. We can also avoid
the reliance on the system instance by using that parameter to access
the page table that needs to be set.
2018-09-20 21:10:00 -04:00
Lioncash
9b8fc2b689 arm_interface: Replace kernel vm_manager include with a forward declaration
Avoids an unnecessary inclusion and also uncovers three places where
indirect inclusions were relied upon, which allows us to also resolve
those.
2018-09-20 19:35:36 -04:00
Mat M
f237265187 Merge pull request #1370 from Hedges/GDBClean
Correct endianness of BRK
2018-09-20 17:25:56 -04:00
Mat M
1e35d8fa8f Merge pull request #1362 from MerryMage/dynarmic
externals: Update dynarmic to 171d116
2018-09-20 17:22:42 -04:00
Jarek Syrylak
74f966404d Correct endianness of BKPT 2018-09-20 22:03:57 +01:00
MerryMage
ef31487624 arm_dynarmic: Halt when BRK encountered 2018-09-20 19:12:42 +01:00
David Marcec
3f49725a51 Fixed submit 2018-09-21 00:51:13 +10:00
David Marcec
cfc9fe4460 Added IRequest::Submit
This fixes updated versions of SMO. Currently unable to test as I don't have an updated version
2018-09-21 00:47:30 +10:00
bunnei
8fe9572271 Merge pull request #1358 from DarkLordZach/temp-storage
savedata_factory: Add TemporaryStorage SaveDataType
2018-09-20 10:13:38 -04:00
bunnei
0931a711de Merge pull request #1363 from lioncash/control
control_metadata: Move language name array definition to the cpp file
2018-09-20 10:13:23 -04:00
bunnei
b02a1e38fa Merge pull request #1361 from lioncash/nax
xts_archive/nax: Minor interface changes
2018-09-20 10:13:10 -04:00
bunnei
e57aa62b11 Merge pull request #1366 from ogniK5377/splat-fix
Implemented NTC & IEnsureNetworkClockAvailabilityService
2018-09-20 10:12:35 -04:00
Lioncash
e980e90d6e game_list: Handle plurals within setFilterResult() better
Qt provides an overload of tr() that operates on quantities in relation
to pluralization. This also allows the translation to adapt based on the
target language rules better.

For example, the previous code would result in an incorrect translation
for the French language (which doesn't use the pluralized version of
"result" in the case of a total of zero. While in English it's
correct to use the pluralized version of "result", that is, "results"

---

For example:

English: "0 results"

French: "0 résultat" (uses the singular form)

In French, the noun being counted is singular if the quantity is 0 or 1.
In English, on the other hand, if the noun being counted has a quantity
of 0 or N > 1, then the noun is pluralized.

---

For another example in a language that has different counting methods
than the above, consider English and Irish. Irish has a special form of
of a grammatical number called a dual. Which alters how a word is
written when N of something is 2. This won't appear in this case with a
direct number "2", but it would change if we ever used "Two" to refer to
two of something. For example:

English: "Zero results"

Irish: "Toradh ar bith"

English: "One result"

Irish: "Toradh amháin"

English: "Two results"

Irish: "Dhá thorthaí" <- Dual case

Which is an important distinction to make between singular and plural,
because in other situations, "two" on its own would be written as "dó"
in Irish. There's also a few other cases where the order the words are
placed *and* whether or not the plural or singular variant of the word
is used *and* whether or not the word is placed after or between a set
of numbers can vary. Counting in Irish also differs depending on whether or not
you're counting things (like above) or counting people, in which case an
entirely different set of numbers are used.

It's not important for this case, but it's provided as an example as to why one
should never assume the placement of values in text will be like that of
English or other languages. Some languages have very different ways to
represent counting, and breaking up the translated string like this
isn't advisable because it makes it extremely difficult to get right
depending on what language a translator is translating text into due to
the ambiguity of the strings being presented for translation.

In this case a translator would see three fragmented strings on
Transifex (and not necessarily grouped beside one another, but even
then, it would still be annoying to decipher):

- "of"
- "result"
- "results"

There is no way a translator is going to know what those sets of words
are actually used for unless they look at the code to see what is being
done with them (which they shouldn't have to do).
2018-09-20 01:35:50 -04:00
David Marcec
583137709f Removed unneeded event clear 2018-09-20 15:16:08 +10:00
David Marcec
335e9d18ae Implemented NTC & IEnsureNetworkClockAvailabilityService
Needed because of the recent nim fixes
2018-09-20 15:14:07 +10:00
Lioncash
05ef9dfc10 file-sys: Default heavy-weight class destructors in the cpp file
Several classes have a lot of non-trivial members within them, or don't
but likely should have the destructor defaulted in the cpp file for
future-proofing/being more friendly to forward declarations.

Leaving the destructor unspecified allows the compiler to inline the
destruction code all over the place, which is generally undesirable from
a code bloat perspective.
2018-09-19 19:34:08 -04:00
Lioncash
3146502a12 control_metadata: Remove unnecessary else within GetLanguageEntry()
There's no need to indent the code here, given the if case contains a
return statement at the end of it.
2018-09-19 19:02:06 -04:00
Lioncash
05d49962b6 control_metadata: Move language name array definition to the cpp file
This was used in two different translation units
(deconstructed_rom_directory and patch_manager). This means we'd be
pointlessly duplicating the whole array twice due to it being defined
within the header.
2018-09-19 18:57:26 -04:00
MerryMage
3b4a158230 arm_dynarmic: Support BKPT instruction 2018-09-19 21:00:38 +01:00
MerryMage
321eb0b6b0 externals: Update dynarmic to 171d116
171d116 A64: Implement SCVTF, UCVTF (vector, fixed-point), scalar variant
f221bb0 emit_x64_floating_point: Reduce fallback LUT code in EmitFPToFixed
eb123e2 A64: Implement FCVTZS, FCVTZU, UCVTF, SCVTF (vector, fixed-point), vector variant
487d37a A64: Implement UQSHL's vector immediate and register variants
f698933 ir: Add opcodes for unsigned saturating left shifts
7148e66 A64/translate/impl: Make signatures consistent for unimplemented by-element SIMD variants
fdde4ca A64: Implement BRK
b1490db A64/imm: Add full range of comparison operators to Imm template
1ec40ef IR: Add fbits argument to FPVectorFrom{Signed,Unsigned}Fixed
d6d5e98 A64: Implement SCVTF, UCVTF (scalar, fixed-point)
6513595 opcodes.inc: Align columns to a tabstop of 4
6b0d2b5 IR: Add fbits argument to FixedToFP-related opcodes
c4b3831 A64: Implement SQSHL's vector immediate variant
e0d8d2d A64: Implement SQSHL's vector register variant
5327625 ir: Add opcodes for left signed saturated shifts
9705252 branch: Make variables const where applicable
650946e move_wide: Make variables const where applicable
62b3a6d load_store_register_unprivileged: Make variables const where applicable
3add1c7 load_store_register_immediate: Place conditional bodies on their own line
2fc4088 load_store_load_literal: Make variables const where applicable
b2c1462 data_processing_logical: Move datasize declarations after early-exit conditionals
028028f data_processing_conditional_select: Make variables const where applicable
c66042d data_processing_addsub: Move datasize declarations after early-exit conditionals
6bc546e data_processing_bitfield: Move datasize variables after early-exit conditionals
2aad5fa A64: Implement CLS's vector variant
6c877ff emit_x64_vector: Make EmitVectorUnsignedSaturatedAccumulateSigned() internally linked
4b5926d perf_map: Use std::string_view instead of std::string for PerfMapRegister()
7445947 A64: Implement SQRDMULH (vector), vector variant
03b80f2 A64: Implement SQDMULL (vector), vector variant
4a2c596 IR: Add VectorSignedSaturatedDoublingMultiplyLong
59dc33e emit_x64_vector: Changes to VectorSignedSaturatedDoublingMultiply
bbaebeb IR: Implement Vector{Signed,Unsigned}Multiply{16,32}
baac5a8 backend_x64/a64_interface: Re-enable the constant folding pass
e78ca19 emit_x64_vector_floating_point: Hardware FMA implementation for RSqrtStepFused
8a5ae9a emit_x64_vector_floating_point: Hardware FMA implementation of FPVectorRecipStepFused
39818f9 emit_x64_floating_point: Hardware FMA implementation of FPRSqrtStepFused
3d0a0b4 emit_x64_floating_point: Hardware FMA implementation of FPRecipStepFused{32,64}
2293dff emit_x64_vector: SSE implementation of VectorSignedSaturatedAccumulateUnsigned{8,16,32}
2047683 emit_x64_vector: Correct static asserts for < 64-bit type checks in saturated accumulate fallbacks
55e9e40 emit_x64_vector: EmitVectorSignedSaturatedAccumulateUnsigned64: SSE implementation
1076651 emit_x64_vector: Simplify fpsr_qc related code
4039030 A64: Implement CLZ's vector variant
0bb908f ir: Add opcodes for vector CLZ operations
3b13259 A64/translate: VectorZeroUpper for V(64) stores
1931d44 simd_two_register_misc: FNEG (vector) with Q == 0 had dirty upper
a0790f0 emit_x64_vector: Remove unnecessary [[maybe_unused]] attributes
b0e1eb5 A64: Implement USQADD's scalar and vector variants
28424c7 ir: Add opcodes form unsigned saturated accumulations of signed values
9923ea0 A64: Implement SUQADD's scalar and vector variants
4c0adbb ir: Add opcodes for signed saturated accumulations of unsigned values
799bfed A64: Implement SMLAL{2}, SMLSL{2}, UMLAL{2}, and UMLSL{2}'s vector by-element variants
94451ec A64: Implement UMULL{2}'s vector by-element variant
45867de A64: Implement SMULL{2}'s vector by-element variant
0235793 ir/value: Replace includes with forward declarations
450f721 ir/cond: Migrate to C++17 nested namespace specifiers
e649988 CMakeLists: Add missing cond.h header to file listing
d20e769 A64: Implement URSQRTE
4f3bde5 ir: Add opcodes for performing unsigned reciprocal square root estimates
cfeeaec A64: Implement URECPE
622b60e ir: Add opcodes for unsigned reciprocal estimate
d17599a Update Xbyak to 5.71
f7c26e9 Squashed 'externals/xbyak/' changes from 671fc805..1de435ed
8782b69 travis: Make macOS build with Xcode 9.4.1
b575b23 A64: Implement SQNEG's scalar and vector variant
06062a9 A64: Add opcodes for signed saturating negations
1c40579 emit_x64_vector: Simplify "position == 0" case for EmitVectorExtract()
e335050 emit_x64_vector: Simplify "position == 0" case for EmitVectorExtractLower()
8b13421 A64: Implement SQDMULH's by-element scalar variant
9122a6e A64: Implement SQDMULH's by-element vector variant
176e60e backend/x64: Do not clear fast_dispatch_table if not enabled
2018-09-19 20:31:39 +01:00
Lioncash
2e5f0e5024 xts_archive: Remove unused variables from CalculateHMAC256()
These variables aren't used, which still has an impact, as std::vector
cannot be optimized away by the compiler (it's constructor and
destructor are both non-trivial), so this was just wasting memory.
2018-09-19 14:23:13 -04:00
Lioncash
2752183883 xts_archive: Make AsNCA() return a std::unique_ptr instead of a std::shared_ptr
std::shared_ptr isn't strictly necessary here and is only ever used in
contexts where the object doesn't depend on being shared. This also
makes the interface more flexible, as it's possible to create a
std::shared_ptr from a std::unique_ptr (std::shared_ptr has a
constructor that accepts a std::unique_ptr), but not the other way
around.
2018-09-19 14:22:37 -04:00
Lioncash
45195a51a7 nax: Avoid re-parsing NAX data with GetFileType()
An instance of the NAX apploader already has an existing NAX instance in
memory. Calling directly into IdentifyType() directly would re-parse the
whole file again into yet another NAX instance, only to toss it away
again.

This gets rid of unnecessary/redundant file parsing and allocations.
2018-09-19 14:22:37 -04:00
Lioncash
c8c4105659 nax: Avoid unnecessary calls to AsNCA() in IdentifyType()
AsNCA() allocates an NCA instance every time it's called. In the current
manner it's used, it's quite inefficient as it's making a redundant
allocation.

We can just amend the order of the conditionals to make it easier to
just call it once.
2018-09-19 14:22:37 -04:00
Lioncash
f272261c21 xts_archive: Ensure NAX's type member is always initialized
Ensures that the member always has a deterministic value.
2018-09-19 14:22:37 -04:00
Lioncash
92a98a8b19 xts_archive: Amend initializer order of NAX's constructor
Orders the initializer list in the same order the members would be
initialized. Avoids compiler warnings.
2018-09-19 14:22:30 -04:00
tech4me
189cf0da7b Update MinGWCross.cmake to lowercase 2018-09-19 14:22:14 -04:00
tech4me
a8f54f96fc travis: running mingw build on travis ci
This commit also fixed a broken cmake dependency with unicorn
2018-09-19 14:22:14 -04:00
David
0432af5ad1 Reworked incorrect nifm stubs (#1355)
* Reworked incorrect nifm stubs

Need confirmation on `CreateTemporaryNetworkProfile`, unsure which game uses it but according to reversing. It should return a uuid which we currently don't do.

Any 0 client id is considered an invalid client id.

GetRequestState 0 is considered invalid.

* Fixups for nifm
2018-09-19 11:59:01 -04:00
bunnei
52fb86c65c Merge pull request #1356 from degasus/hotfix
gl_rasterizer: Fix StartAddress handling with indexed draw calls.
2018-09-19 11:58:22 -04:00
bunnei
8dff92c5f6 Merge pull request #1359 from ogniK5377/nes
Fixed GetAccountId stub, Added error code for OpenDirectory and added ActivateNpadWithRevision
2018-09-19 10:01:36 -04:00
bunnei
5fbc354261 Merge pull request #1353 from ogniK5377/remove-MakeBuilder
Removed the use of rp.MakeBuilder in the yuzu codebase
2018-09-19 10:00:23 -04:00
David Marcec
cbc7ad8f6d Fixed GetAccountId stub, Added error code for OpenDirectory and added ActivateNpadWithRevision
With these, `Nintendo Entertainment System - Nintendo Switch Online` loads
2018-09-19 23:25:00 +10:00
Zach Hilman
9bba103791 yuzu-cmd: Add call to CreateFactories
Ensures all relevant filesystem objects are initialized and eliminates a crash related to the RegisteredCache.
2018-09-19 09:09:11 -04:00
Zach Hilman
dfaea55be5 savedata_factory: Add TemporaryStorage SaveDataType
Seems to be used by NSO NES Emulator
2018-09-19 09:06:22 -04:00
raven02
c8f9bbbf85 Merge branch 'master' into tlds 2018-09-19 19:53:11 +08:00
Markus Wick
f465e4aaf2 gl_rasterizer: Fix StartAddress handling with indexed draw calls.
We uploaded the wrong data before. So the offset on the host GPU pointer may work for the first vertices, the last ones run out bounds.
Let's just offset the upload instead.
2018-09-19 09:22:30 +02:00
David Marcec
d06f4cfc63 Corrected SSL::SetInterfaceVersion
Should be a single u32
2018-09-19 16:46:11 +10:00
Lioncash
ab6dfa4fa5 ring_buffer: Use std::atomic_size_t in a static assert
Avoids the need to repeat "std::" twice
2018-09-18 23:36:04 -04:00
Lioncash
c51f8563a6 ring_buffer: Use std::hardware_destructive_interference_size to determine alignment size for avoiding false sharing
MSVC 19.11 (A.K.A. VS 15.3)'s C++ standard library implements P0154R1
(http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2016/p0154r1.html)
which defines two new constants within the <new> header, std::hardware_destructive_interference_size
and std::hardware_constructive_interference_size.

std::hardware_destructive_interference_size defines the minimum
recommended offset between two concurrently-accessed objects to avoid
performance degradation due to contention introduced by the
implementation (with the lower-bound being at least alignof(max_align_t)).
In other words, the minimum offset between objects necessary to avoid
false-sharing.

std::hardware_constructive_interference_size on the other hand defines
the maximum recommended size of contiguous memory occupied by two
objects accessed wth temporal locality by concurrent threads (also
defined to be at least alignof(max_align_t)). In other words the maximum
size to promote true-sharing.

So we can simply use this facility to determine the ideal alignment
size. Unfortunately, only MSVC supports this right now, so we need to
enclose it within an ifdef for the time being.
2018-09-18 23:35:57 -04:00
raven02
b91f7d5d67 Add 1D sampler for TLDS - TexelFetch (Mario Rabbids) 2018-09-17 23:25:18 +08:00
151 changed files with 3629 additions and 811 deletions

4
.gitattributes vendored Normal file
View File

@@ -0,0 +1,4 @@
dist/languages/* linguist-vendored
dist/qt_themes/* linguist-vendored
externals/* linguist-vendored
*.h linguist-language=cpp

View File

@@ -24,11 +24,24 @@ matrix:
- os: osx
env: NAME="macos build"
sudo: false
osx_image: xcode9.3
osx_image: xcode10
install: "./.travis/macos/deps.sh"
script: "./.travis/macos/build.sh"
after_success: "./.travis/macos/upload.sh"
cache: ccache
- os: linux
env: NAME="MinGW build"
sudo: required
dist: trusty
services: docker
addons:
apt:
packages:
- p7zip-full
install: "./.travis/linux-mingw/deps.sh"
script: "./.travis/linux-mingw/build.sh"
after_success: "./.travis/linux-mingw/upload.sh"
cache: ccache
deploy:
provider: releases

View File

@@ -11,6 +11,9 @@ if [ -z $TRAVIS_TAG ]; then
RELEASE_NAME=head
else
RELEASE_NAME=$(echo $TRAVIS_TAG | cut -d- -f1)
if [ "$NAME" = "MinGW build" ]; then
RELEASE_NAME="${RELEASE_NAME}-mingw"
fi
fi
mv "$REV_NAME" $RELEASE_NAME

3
.travis/linux-mingw/build.sh Executable file
View File

@@ -0,0 +1,3 @@
#!/bin/bash -ex
mkdir "$HOME/.ccache" || true
docker run --env-file .travis/common/travis-ci.env -v $(pwd):/yuzu -v "$HOME/.ccache":/root/.ccache ubuntu:18.04 /bin/bash -ex /yuzu/.travis/linux-mingw/docker.sh

3
.travis/linux-mingw/deps.sh Executable file
View File

@@ -0,0 +1,3 @@
#!/bin/sh -ex
docker pull ubuntu:18.04

59
.travis/linux-mingw/docker.sh Executable file
View File

@@ -0,0 +1,59 @@
#!/bin/bash -ex
cd /yuzu
MINGW_PACKAGES="sdl2-mingw-w64 qt5base-mingw-w64 qt5tools-mingw-w64 libsamplerate-mingw-w64 qt5multimedia-mingw-w64"
apt-get update
apt-get install -y gpg wget git python3-pip python ccache g++-mingw-w64-x86-64 gcc-mingw-w64-x86-64 mingw-w64-tools cmake
echo 'deb http://ppa.launchpad.net/tobydox/mingw-w64/ubuntu bionic main ' > /etc/apt/sources.list.d/extras.list
apt-key adv --keyserver keyserver.ubuntu.com --recv '72931B477E22FEFD47F8DECE02FE5F12ADDE29B2'
apt-get update
apt-get install -y ${MINGW_PACKAGES}
# fix a problem in current MinGW headers
wget -q https://raw.githubusercontent.com/Alexpux/mingw-w64/d0d7f784833bbb0b2d279310ddc6afb52fe47a46/mingw-w64-headers/crt/errno.h -O /usr/x86_64-w64-mingw32/include/errno.h
# override Travis CI unreasonable ccache size
echo 'max_size = 3.0G' > "$HOME/.ccache/ccache.conf"
# Dirty hack to trick unicorn makefile into believing we are in a MINGW system
mv /bin/uname /bin/uname1 && echo -e '#!/bin/sh\necho MINGW64' >> /bin/uname
chmod +x /bin/uname
# Dirty hack to trick unicorn makefile into believing we have cmd
echo '' >> /bin/cmd
chmod +x /bin/cmd
mkdir build && cd build
cmake .. -DCMAKE_TOOLCHAIN_FILE="$(pwd)/../CMakeModules/MinGWCross.cmake" -DUSE_CCACHE=ON -DYUZU_USE_BUNDLED_UNICORN=ON -DENABLE_COMPATIBILITY_LIST_DOWNLOAD=ON -DCMAKE_BUILD_TYPE=Release
make -j4
# Clean up the dirty hacks
rm /bin/uname && mv /bin/uname1 /bin/uname
rm /bin/cmd
ccache -s
echo "Tests skipped"
#ctest -VV -C Release
echo 'Prepare binaries...'
cd ..
mkdir package
QT_PLATFORM_DLL_PATH='/usr/x86_64-w64-mingw32/lib/qt5/plugins/platforms/'
find build/ -name "yuzu*.exe" -exec cp {} 'package' \;
# copy Qt plugins
mkdir package/platforms
cp "${QT_PLATFORM_DLL_PATH}/qwindows.dll" package/platforms/
cp -rv "${QT_PLATFORM_DLL_PATH}/../mediaservice/" package/
cp -rv "${QT_PLATFORM_DLL_PATH}/../imageformats/" package/
rm -f package/mediaservice/*d.dll
for i in package/*.exe; do
# we need to process pdb here, however, cv2pdb
# does not work here, so we just simply strip all the debug symbols
x86_64-w64-mingw32-strip "${i}"
done
pip3 install pefile
python3 .travis/linux-mingw/scan_dll.py package/*.exe "package/"

View File

@@ -0,0 +1,106 @@
import pefile
import sys
import re
import os
import queue
import shutil
# constant definitions
KNOWN_SYS_DLLS = ['WINMM.DLL', 'MSVCRT.DLL', 'VERSION.DLL', 'MPR.DLL',
'DWMAPI.DLL', 'UXTHEME.DLL', 'DNSAPI.DLL', 'IPHLPAPI.DLL']
# below is for Ubuntu 18.04 with specified PPA enabled, if you are using
# other distro or different repositories, change the following accordingly
DLL_PATH = [
'/usr/x86_64-w64-mingw32/bin/',
'/usr/x86_64-w64-mingw32/lib/',
'/usr/lib/gcc/x86_64-w64-mingw32/7.3-posix/'
]
missing = []
def parse_imports(file_name):
results = []
pe = pefile.PE(file_name, fast_load=True)
pe.parse_data_directories()
for entry in pe.DIRECTORY_ENTRY_IMPORT:
current = entry.dll.decode()
current_u = current.upper() # b/c Windows is often case insensitive
# here we filter out system dlls
# dll w/ names like *32.dll are likely to be system dlls
if current_u.upper() not in KNOWN_SYS_DLLS and not re.match(string=current_u, pattern=r'.*32\.DLL'):
results.append(current)
return results
def parse_imports_recursive(file_name, path_list=[]):
q = queue.Queue() # create a FIFO queue
# file_name can be a string or a list for the convience
if isinstance(file_name, str):
q.put(file_name)
elif isinstance(file_name, list):
for i in file_name:
q.put(i)
full_list = []
while q.qsize():
current = q.get_nowait()
print('> %s' % current)
deps = parse_imports(current)
# if this dll does not have any import, ignore it
if not deps:
continue
for dep in deps:
# the dependency already included in the list, skip
if dep in full_list:
continue
# find the requested dll in the provided paths
full_path = find_dll(dep)
if not full_path:
missing.append(dep)
continue
full_list.append(dep)
q.put(full_path)
path_list.append(full_path)
return full_list
def find_dll(name):
for path in DLL_PATH:
for root, _, files in os.walk(path):
for f in files:
if name.lower() == f.lower():
return os.path.join(root, f)
def deploy(name, dst, dry_run=False):
dlls_path = []
parse_imports_recursive(name, dlls_path)
for dll_entry in dlls_path:
if not dry_run:
shutil.copy(dll_entry, dst)
else:
print('[Dry-Run] Copy %s to %s' % (dll_entry, dst))
print('Deploy completed.')
return dlls_path
def main():
if len(sys.argv) < 3:
print('Usage: %s [files to examine ...] [target deploy directory]')
return 1
to_deploy = sys.argv[1:-1]
tgt_dir = sys.argv[-1]
if not os.path.isdir(tgt_dir):
print('%s is not a directory.' % tgt_dir)
return 1
print('Scanning dependencies...')
deploy(to_deploy, tgt_dir)
if missing:
print('Following DLLs are not found: %s' % ('\n'.join(missing)))
return 0
if __name__ == '__main__':
main()

13
.travis/linux-mingw/upload.sh Executable file
View File

@@ -0,0 +1,13 @@
#!/bin/bash -ex
. .travis/common/pre-upload.sh
REV_NAME="yuzu-windows-mingw-${GITDATE}-${GITREV}"
ARCHIVE_NAME="${REV_NAME}.tar.gz"
COMPRESSION_FLAGS="-czvf"
mkdir "$REV_NAME"
# get around the permission issues
cp -r package/* "$REV_NAME"
. .travis/common/post-upload.sh

View File

@@ -6,7 +6,9 @@ apt-get install --no-install-recommends -y build-essential git libqt5opengl5-dev
cd /yuzu
mkdir build && cd build
cmake .. -DYUZU_BUILD_UNICORN=ON -DCMAKE_BUILD_TYPE=Release -DCMAKE_C_COMPILER=/usr/lib/ccache/gcc -DCMAKE_CXX_COMPILER=/usr/lib/ccache/g++ -DENABLE_COMPATIBILITY_LIST_DOWNLOAD=ON -G Ninja
cmake .. -DYUZU_USE_BUNDLED_UNICORN=ON -DCMAKE_BUILD_TYPE=Release -DCMAKE_C_COMPILER=/usr/lib/ccache/gcc -DCMAKE_CXX_COMPILER=/usr/lib/ccache/g++ -DENABLE_COMPATIBILITY_LIST_DOWNLOAD=ON -G Ninja
ninja
ccache -s
ctest -VV -C Release

View File

@@ -2,14 +2,16 @@
set -o pipefail
export MACOSX_DEPLOYMENT_TARGET=10.12
export MACOSX_DEPLOYMENT_TARGET=10.13
export Qt5_DIR=$(brew --prefix)/opt/qt5
export UNICORNDIR=$(pwd)/externals/unicorn
export PATH="/usr/local/opt/ccache/libexec:$PATH"
mkdir build && cd build
cmake --version
cmake .. -DYUZU_BUILD_UNICORN=ON -DCMAKE_BUILD_TYPE=Release -DENABLE_COMPATIBILITY_LIST_DOWNLOAD=ON
cmake .. -DYUZU_USE_BUNDLED_UNICORN=ON -DCMAKE_BUILD_TYPE=Release -DENABLE_COMPATIBILITY_LIST_DOWNLOAD=ON
make -j4
ccache -s
ctest -VV -C Release

View File

@@ -123,8 +123,6 @@ else()
# Avoid windows.h from including some usually unused libs like winsocks.h, since this might cause some redefinition errors.
add_definitions(/DWIN32_LEAN_AND_MEAN)
# set up output paths for executable binaries (.exe-files, and .dll-files on DLL-capable platforms)
set(CMAKE_RUNTIME_OUTPUT_DIRECTORY ${CMAKE_BINARY_DIR}/bin)
set(CMAKE_CONFIGURATION_TYPES Debug Release CACHE STRING "" FORCE)
# Tweak optimization settings
@@ -269,10 +267,18 @@ if (YUZU_USE_BUNDLED_UNICORN)
find_package(PythonInterp 2.7 REQUIRED)
add_custom_command(OUTPUT ${LIBUNICORN_LIBRARY}
COMMAND ${CMAKE_COMMAND} -E env UNICORN_ARCHS="aarch64" PYTHON="${PYTHON_EXECUTABLE}" /bin/sh make.sh macos-universal-no
WORKING_DIRECTORY ${UNICORN_PREFIX}
)
if (MINGW)
add_custom_command(OUTPUT ${LIBUNICORN_LIBRARY}
COMMAND ${CMAKE_COMMAND} -E env UNICORN_ARCHS="aarch64" PYTHON="${PYTHON_EXECUTABLE}" /bin/sh make.sh cross-win64
WORKING_DIRECTORY ${UNICORN_PREFIX}
)
else()
add_custom_command(OUTPUT ${LIBUNICORN_LIBRARY}
COMMAND ${CMAKE_COMMAND} -E env UNICORN_ARCHS="aarch64" PYTHON="${PYTHON_EXECUTABLE}" /bin/sh make.sh macos-universal-no
WORKING_DIRECTORY ${UNICORN_PREFIX}
)
endif()
# ALL makes this custom target build every time
# but it won't actually build if LIBUNICORN_LIBRARY is up to date
add_custom_target(unicorn-build ALL
@@ -286,6 +292,7 @@ endif()
if (UNICORN_FOUND)
add_library(unicorn INTERFACE)
add_dependencies(unicorn unicorn-build)
target_link_libraries(unicorn INTERFACE "${LIBUNICORN_LIBRARY}")
target_include_directories(unicorn INTERFACE "${LIBUNICORN_INCLUDE_DIR}")
else()
@@ -431,8 +438,12 @@ enable_testing()
add_subdirectory(externals)
add_subdirectory(src)
# Set yuzu project as default StartUp Project in Visual Studio
set_property(DIRECTORY ${CMAKE_CURRENT_SOURCE_DIR} PROPERTY VS_STARTUP_PROJECT yuzu)
# Set yuzu project or yuzu-cmd project as default StartUp Project in Visual Studio depending on whether QT is enabled or not
if(ENABLE_QT)
set_property(DIRECTORY ${CMAKE_CURRENT_SOURCE_DIR} PROPERTY VS_STARTUP_PROJECT yuzu)
else()
set_property(DIRECTORY ${CMAKE_CURRENT_SOURCE_DIR} PROPERTY VS_STARTUP_PROJECT yuzu-cmd)
endif()
# Installation instructions

View File

@@ -0,0 +1,54 @@
set(MINGW_PREFIX /usr/x86_64-w64-mingw32/)
set(CMAKE_SYSTEM_NAME Windows)
set(CMAKE_SYSTEM_PROCESSOR x86_64)
# Actually a hack, w/o this will cause some strange errors
set(CMAKE_HOST_WIN32 TRUE)
set(CMAKE_FIND_ROOT_PATH ${MINGW_PREFIX})
set(SDL2_PATH ${MINGW_PREFIX})
set(MINGW_TOOL_PREFIX ${CMAKE_SYSTEM_PROCESSOR}-w64-mingw32-)
# Specify the cross compiler
set(CMAKE_C_COMPILER ${MINGW_TOOL_PREFIX}gcc-posix)
set(CMAKE_CXX_COMPILER ${MINGW_TOOL_PREFIX}g++-posix)
set(CMAKE_RC_COMPILER ${MINGW_TOOL_PREFIX}windres)
# Mingw tools
set(STRIP ${MINGW_TOOL_PREFIX}strip)
set(WINDRES ${MINGW_TOOL_PREFIX}windres)
set(ENV{PKG_CONFIG} ${MINGW_TOOL_PREFIX}pkg-config)
# ccache wrapper
option(USE_CCACHE "Use ccache for compilation" OFF)
if(USE_CCACHE)
find_program(CCACHE ccache)
if(CCACHE)
message(STATUS "Using ccache found in PATH")
set_property(GLOBAL PROPERTY RULE_LAUNCH_COMPILE ${CCACHE})
set_property(GLOBAL PROPERTY RULE_LAUNCH_LINK ${CCACHE})
else(CCACHE)
message(WARNING "USE_CCACHE enabled, but no ccache found")
endif(CCACHE)
endif(USE_CCACHE)
# Search for programs in the build host directories
set(CMAKE_FIND_ROOT_PATH_MODE_PROGRAM NEVER)
# Echo modified cmake vars to screen for debugging purposes
if(NOT DEFINED ENV{MINGW_DEBUG_INFO})
message("")
message("Custom cmake vars: (blank = system default)")
message("-----------------------------------------")
message("* CMAKE_C_COMPILER : ${CMAKE_C_COMPILER}")
message("* CMAKE_CXX_COMPILER : ${CMAKE_CXX_COMPILER}")
message("* CMAKE_RC_COMPILER : ${CMAKE_RC_COMPILER}")
message("* WINDRES : ${WINDRES}")
message("* ENV{PKG_CONFIG} : $ENV{PKG_CONFIG}")
message("* STRIP : ${STRIP}")
message("* USE_CCACHE : ${USE_CCACHE}")
message("")
# So that the debug info only appears once
set(ENV{MINGW_DEBUG_INFO} SHOWN)
endif()

View File

@@ -79,6 +79,10 @@ u32 AudioRenderer::GetMixBufferCount() const {
return worker_params.mix_buffer_count;
}
Stream::State AudioRenderer::GetStreamState() const {
return stream->GetState();
}
std::vector<u8> AudioRenderer::UpdateAudioRenderer(const std::vector<u8>& input_params) {
// Copy UpdateDataHeader struct
UpdateDataHeader config{};

View File

@@ -170,6 +170,7 @@ public:
u32 GetSampleRate() const;
u32 GetSampleCount() const;
u32 GetMixBufferCount() const;
Stream::State GetStreamState() const;
private:
class VoiceState;

View File

@@ -49,9 +49,14 @@ void Stream::Play() {
}
void Stream::Stop() {
state = State::Stopped;
ASSERT_MSG(false, "Unimplemented");
}
Stream::State Stream::GetState() const {
return state;
}
s64 Stream::GetBufferReleaseCycles(const Buffer& buffer) const {
const std::size_t num_samples{buffer.GetSamples().size() / GetNumChannels()};
return CoreTiming::usToCycles((static_cast<u64>(num_samples) * 1000000) / sample_rate);

View File

@@ -33,6 +33,12 @@ public:
Multi51Channel16,
};
/// Current state of the stream
enum class State {
Stopped,
Playing,
};
/// Callback function type, used to change guest state on a buffer being released
using ReleaseCallback = std::function<void()>;
@@ -72,13 +78,10 @@ public:
/// Gets the number of channels
u32 GetNumChannels() const;
private:
/// Current state of the stream
enum class State {
Stopped,
Playing,
};
/// Get the state
State GetState() const;
private:
/// Plays the next queued buffer in the audio stream, starting playback if necessary
void PlayNextBuffer();

View File

@@ -59,7 +59,7 @@ std::size_t TimeStretcher::Process(const s16* in, std::size_t num_in, s16* out,
m_stretch_ratio = std::max(m_stretch_ratio, 0.05);
m_sound_touch.setTempo(m_stretch_ratio);
LOG_DEBUG(Audio, "{:5}/{:5} ratio:{:0.6f} backlog:{:0.6f}", num_in, num_out, m_stretch_ratio,
LOG_TRACE(Audio, "{:5}/{:5} ratio:{:0.6f} backlog:{:0.6f}", num_in, num_out, m_stretch_ratio,
backlog_fullness);
m_sound_touch.putSamples(in, static_cast<u32>(num_in));

View File

@@ -33,6 +33,8 @@
#define NAND_DIR "nand"
#define SYSDATA_DIR "sysdata"
#define KEYS_DIR "keys"
#define LOAD_DIR "load"
#define DUMP_DIR "dump"
#define LOG_DIR "log"
// Filenames

View File

@@ -705,6 +705,8 @@ const std::string& GetUserPath(UserPath path, const std::string& new_path) {
#endif
paths.emplace(UserPath::SDMCDir, user_path + SDMC_DIR DIR_SEP);
paths.emplace(UserPath::NANDDir, user_path + NAND_DIR DIR_SEP);
paths.emplace(UserPath::LoadDir, user_path + LOAD_DIR DIR_SEP);
paths.emplace(UserPath::DumpDir, user_path + DUMP_DIR DIR_SEP);
paths.emplace(UserPath::SysDataDir, user_path + SYSDATA_DIR DIR_SEP);
paths.emplace(UserPath::KeysDir, user_path + KEYS_DIR DIR_SEP);
// TODO: Put the logs in a better location for each OS

View File

@@ -29,6 +29,8 @@ enum class UserPath {
NANDDir,
RootDir,
SDMCDir,
LoadDir,
DumpDir,
SysDataDir,
UserDir,
};

View File

@@ -183,6 +183,7 @@ void FileBackend::Write(const Entry& entry) {
SUB(Service, FS) \
SUB(Service, GRC) \
SUB(Service, HID) \
SUB(Service, IRS) \
SUB(Service, LBL) \
SUB(Service, LDN) \
SUB(Service, LDR) \

View File

@@ -70,6 +70,7 @@ enum class Class : ClassType {
Service_FS, ///< The FS (Filesystem) service
Service_GRC, ///< The game recording service
Service_HID, ///< The HID (Human interface device) service
Service_IRS, ///< The IRS service
Service_LBL, ///< The LBL (LCD backlight) service
Service_LDN, ///< The LDN (Local domain network) service
Service_LDR, ///< The loader service

View File

@@ -9,6 +9,7 @@
#include <atomic>
#include <cstddef>
#include <cstring>
#include <new>
#include <type_traits>
#include <vector>
#include "common/common_types.h"
@@ -29,7 +30,7 @@ class RingBuffer {
static_assert(capacity < std::numeric_limits<std::size_t>::max() / 2 / granularity);
static_assert((capacity & (capacity - 1)) == 0, "capacity must be a power of two");
// Ensure lock-free.
static_assert(std::atomic<std::size_t>::is_always_lock_free);
static_assert(std::atomic_size_t::is_always_lock_free);
public:
/// Pushes slots into the ring buffer
@@ -102,8 +103,15 @@ public:
private:
// It is important to align the below variables for performance reasons:
// Having them on the same cache-line would result in false-sharing between them.
alignas(128) std::atomic<std::size_t> m_read_index{0};
alignas(128) std::atomic<std::size_t> m_write_index{0};
// TODO: Remove this ifdef whenever clang and GCC support
// std::hardware_destructive_interference_size.
#if defined(_MSC_VER) && _MSC_VER >= 1911
alignas(std::hardware_destructive_interference_size) std::atomic_size_t m_read_index{0};
alignas(std::hardware_destructive_interference_size) std::atomic_size_t m_write_index{0};
#else
alignas(128) std::atomic_size_t m_read_index{0};
alignas(128) std::atomic_size_t m_write_index{0};
#endif
std::array<T, granularity * capacity> m_data;
};

View File

@@ -87,14 +87,6 @@ private:
void SleepCurrentThread(int ms);
void SwitchCurrentThread(); // On Linux, this is equal to sleep 1ms
// Use this function during a spin-wait to make the current thread
// relax while another thread is working. This may be more efficient
// than using events because event functions use kernel calls.
inline void YieldCPU() {
std::this_thread::yield();
}
void SetCurrentThreadName(const char* name);
} // namespace Common

View File

@@ -32,6 +32,8 @@ add_library(core STATIC
file_sys/control_metadata.h
file_sys/directory.h
file_sys/errors.h
file_sys/fsmitm_romfsbuild.cpp
file_sys/fsmitm_romfsbuild.h
file_sys/mode.h
file_sys/nca_metadata.cpp
file_sys/nca_metadata.h
@@ -59,10 +61,13 @@ add_library(core STATIC
file_sys/vfs.h
file_sys/vfs_concat.cpp
file_sys/vfs_concat.h
file_sys/vfs_layered.cpp
file_sys/vfs_layered.h
file_sys/vfs_offset.cpp
file_sys/vfs_offset.h
file_sys/vfs_real.cpp
file_sys/vfs_real.h
file_sys/vfs_static.h
file_sys/vfs_vector.cpp
file_sys/vfs_vector.h
file_sys/xts_archive.cpp

View File

@@ -6,7 +6,10 @@
#include <array>
#include "common/common_types.h"
#include "core/hle/kernel/vm_manager.h"
namespace Kernel {
enum class VMAPermission : u8;
}
namespace Core {
@@ -19,10 +22,16 @@ public:
std::array<u64, 31> cpu_registers;
u64 sp;
u64 pc;
u64 pstate;
u32 pstate;
std::array<u8, 4> padding;
std::array<u128, 32> vector_registers;
u64 fpcr;
u32 fpcr;
u32 fpsr;
u64 tpidr;
};
// Internally within the kernel, it expects the AArch64 version of the
// thread context to be 800 bytes in size.
static_assert(sizeof(ThreadContext) == 0x320);
/// Runs the CPU until an event happens
virtual void Run() = 0;

View File

@@ -12,8 +12,10 @@
#include "core/core.h"
#include "core/core_cpu.h"
#include "core/core_timing.h"
#include "core/gdbstub/gdbstub.h"
#include "core/hle/kernel/process.h"
#include "core/hle/kernel/svc.h"
#include "core/hle/kernel/vm_manager.h"
#include "core/memory.h"
namespace Core {
@@ -79,6 +81,17 @@ public:
case Dynarmic::A64::Exception::SendEventLocal:
case Dynarmic::A64::Exception::Yield:
return;
case Dynarmic::A64::Exception::Breakpoint:
if (GDBStub::IsServerEnabled()) {
parent.jit->HaltExecution();
parent.SetPC(pc);
Kernel::Thread* thread = Kernel::GetCurrentThread();
parent.SaveContext(thread->context);
GDBStub::Break();
GDBStub::SendTrap(thread, 5);
return;
}
[[fallthrough]];
default:
ASSERT_MSG(false, "ExceptionRaised(exception = {}, pc = {:X})",
static_cast<std::size_t>(exception), pc);
@@ -116,7 +129,8 @@ public:
};
std::unique_ptr<Dynarmic::A64::Jit> ARM_Dynarmic::MakeJit() const {
auto** const page_table = Core::CurrentProcess()->vm_manager.page_table.pointers.data();
auto& current_process = Core::CurrentProcess();
auto** const page_table = current_process->VMManager().page_table.pointers.data();
Dynarmic::A64::UserConfig config;
@@ -125,7 +139,7 @@ std::unique_ptr<Dynarmic::A64::Jit> ARM_Dynarmic::MakeJit() const {
// Memory
config.page_table = reinterpret_cast<void**>(page_table);
config.page_table_address_space_bits = Memory::ADDRESS_SPACE_BITS;
config.page_table_address_space_bits = current_process->VMManager().GetAddressSpaceWidth();
config.silently_mirror_page_table = false;
// Multi-process state
@@ -161,7 +175,7 @@ ARM_Dynarmic::ARM_Dynarmic(std::shared_ptr<ExclusiveMonitor> exclusive_monitor,
std::size_t core_index)
: cb(std::make_unique<ARM_Dynarmic_Callbacks>(*this)), core_index{core_index},
exclusive_monitor{std::dynamic_pointer_cast<DynarmicExclusiveMonitor>(exclusive_monitor)} {
ThreadContext ctx;
ThreadContext ctx{};
inner_unicorn.SaveContext(ctx);
PageTableChanged();
LoadContext(ctx);
@@ -233,15 +247,19 @@ void ARM_Dynarmic::SaveContext(ThreadContext& ctx) {
ctx.pstate = jit->GetPstate();
ctx.vector_registers = jit->GetVectors();
ctx.fpcr = jit->GetFpcr();
ctx.fpsr = jit->GetFpsr();
ctx.tpidr = cb->tpidr_el0;
}
void ARM_Dynarmic::LoadContext(const ThreadContext& ctx) {
jit->SetRegisters(ctx.cpu_registers);
jit->SetSP(ctx.sp);
jit->SetPC(ctx.pc);
jit->SetPstate(static_cast<u32>(ctx.pstate));
jit->SetPstate(ctx.pstate);
jit->SetVectors(ctx.vector_registers);
jit->SetFpcr(static_cast<u32>(ctx.fpcr));
jit->SetFpcr(ctx.fpcr);
jit->SetFpsr(ctx.fpsr);
SetTPIDR_EL0(ctx.tpidr);
}
void ARM_Dynarmic::PrepareReschedule() {

View File

@@ -12,6 +12,10 @@
#include "core/arm/exclusive_monitor.h"
#include "core/arm/unicorn/arm_unicorn.h"
namespace Memory {
struct PageTable;
}
namespace Core {
class ARM_Dynarmic_Callbacks;

View File

@@ -64,7 +64,7 @@ FileSys::VirtualFile GetGameFileFromPath(const FileSys::VirtualFilesystem& vfs,
if (concat.empty())
return nullptr;
return FileSys::ConcatenateFiles(concat, dir->GetName());
return FileSys::ConcatenatedVfsFile::MakeConcatenatedFile(concat, dir->GetName());
}
return vfs->OpenFile(path, FileSys::Mode::Read);
@@ -202,7 +202,7 @@ struct System::Impl {
return init_result;
}
const Loader::ResultStatus load_result{app_loader->Load(kernel.CurrentProcess())};
const Loader::ResultStatus load_result{app_loader->Load(*kernel.CurrentProcess())};
if (load_result != Loader::ResultStatus::Success) {
LOG_CRITICAL(Core, "Failed to load ROM (Error {})!", static_cast<int>(load_result));
Shutdown();

View File

@@ -55,16 +55,16 @@ Cpu::Cpu(std::shared_ptr<ExclusiveMonitor> exclusive_monitor,
if (Settings::values.use_cpu_jit) {
#ifdef ARCHITECTURE_x86_64
arm_interface = std::make_shared<ARM_Dynarmic>(exclusive_monitor, core_index);
arm_interface = std::make_unique<ARM_Dynarmic>(exclusive_monitor, core_index);
#else
arm_interface = std::make_shared<ARM_Unicorn>();
arm_interface = std::make_unique<ARM_Unicorn>();
LOG_WARNING(Core, "CPU JIT requested, but Dynarmic not available");
#endif
} else {
arm_interface = std::make_shared<ARM_Unicorn>();
arm_interface = std::make_unique<ARM_Unicorn>();
}
scheduler = std::make_shared<Kernel::Scheduler>(arm_interface.get());
scheduler = std::make_shared<Kernel::Scheduler>(*arm_interface);
}
Cpu::~Cpu() = default;

View File

@@ -76,7 +76,7 @@ public:
private:
void Reschedule();
std::shared_ptr<ARM_Interface> arm_interface;
std::unique_ptr<ARM_Interface> arm_interface;
std::shared_ptr<CpuBarrier> cpu_barrier;
std::shared_ptr<Kernel::Scheduler> scheduler;

View File

@@ -2,13 +2,14 @@
// Licensed under GPLv2 or any later version
// Refer to the license.txt file included.
#include <fmt/format.h>
#include "core/file_sys/bis_factory.h"
#include "core/file_sys/registered_cache.h"
namespace FileSys {
BISFactory::BISFactory(VirtualDir nand_root_)
: nand_root(std::move(nand_root_)),
BISFactory::BISFactory(VirtualDir nand_root_, VirtualDir load_root_)
: nand_root(std::move(nand_root_)), load_root(std::move(load_root_)),
sysnand_cache(std::make_shared<RegisteredCache>(
GetOrCreateDirectoryRelative(nand_root, "/system/Contents/registered"))),
usrnand_cache(std::make_shared<RegisteredCache>(
@@ -24,4 +25,11 @@ std::shared_ptr<RegisteredCache> BISFactory::GetUserNANDContents() const {
return usrnand_cache;
}
VirtualDir BISFactory::GetModificationLoadRoot(u64 title_id) const {
// LayeredFS doesn't work on updates and title id-less homebrew
if (title_id == 0 || (title_id & 0x800) > 0)
return nullptr;
return GetOrCreateDirectoryRelative(load_root, fmt::format("/{:016X}", title_id));
}
} // namespace FileSys

View File

@@ -17,14 +17,17 @@ class RegisteredCache;
/// registered caches.
class BISFactory {
public:
explicit BISFactory(VirtualDir nand_root);
explicit BISFactory(VirtualDir nand_root, VirtualDir load_root);
~BISFactory();
std::shared_ptr<RegisteredCache> GetSystemNANDContents() const;
std::shared_ptr<RegisteredCache> GetUserNANDContents() const;
VirtualDir GetModificationLoadRoot(u64 title_id) const;
private:
VirtualDir nand_root;
VirtualDir load_root;
std::shared_ptr<RegisteredCache> sysnand_cache;
std::shared_ptr<RegisteredCache> usrnand_cache;

View File

@@ -463,6 +463,8 @@ NCA::NCA(VirtualFile file_, VirtualFile bktr_base_romfs_, u64 bktr_base_ivfc_off
status = Loader::ResultStatus::Success;
}
NCA::~NCA() = default;
Loader::ResultStatus NCA::GetStatus() const {
return status;
}

View File

@@ -81,6 +81,8 @@ class NCA : public ReadOnlyVfsDirectory {
public:
explicit NCA(VirtualFile file, VirtualFile bktr_base_romfs = nullptr,
u64 bktr_base_ivfc_offset = 0);
~NCA() override;
Loader::ResultStatus GetStatus() const;
std::vector<std::shared_ptr<VfsFile>> GetFiles() const override;

View File

@@ -8,6 +8,14 @@
namespace FileSys {
const std::array<const char*, 15> LANGUAGE_NAMES = {
"AmericanEnglish", "BritishEnglish", "Japanese",
"French", "German", "LatinAmericanSpanish",
"Spanish", "Italian", "Dutch",
"CanadianFrench", "Portugese", "Russian",
"Korean", "Taiwanese", "Chinese",
};
std::string LanguageEntry::GetApplicationName() const {
return Common::StringFromFixedZeroTerminatedBuffer(application_name.data(), 0x200);
}
@@ -20,18 +28,20 @@ NACP::NACP(VirtualFile file) : raw(std::make_unique<RawNACP>()) {
file->ReadObject(raw.get());
}
NACP::~NACP() = default;
const LanguageEntry& NACP::GetLanguageEntry(Language language) const {
if (language != Language::Default) {
return raw->language_entries.at(static_cast<u8>(language));
} else {
for (const auto& language_entry : raw->language_entries) {
if (!language_entry.GetApplicationName().empty())
return language_entry;
}
// Fallback to English
return GetLanguageEntry(Language::AmericanEnglish);
}
for (const auto& language_entry : raw->language_entries) {
if (!language_entry.GetApplicationName().empty())
return language_entry;
}
// Fallback to English
return GetLanguageEntry(Language::AmericanEnglish);
}
std::string NACP::GetApplicationName(Language language) const {

View File

@@ -66,18 +66,15 @@ enum class Language : u8 {
Default = 255,
};
static constexpr std::array<const char*, 15> LANGUAGE_NAMES = {
"AmericanEnglish", "BritishEnglish", "Japanese",
"French", "German", "LatinAmericanSpanish",
"Spanish", "Italian", "Dutch",
"CanadianFrench", "Portugese", "Russian",
"Korean", "Taiwanese", "Chinese"};
extern const std::array<const char*, 15> LANGUAGE_NAMES;
// A class representing the format used by NX metadata files, typically named Control.nacp.
// These store application name, dev name, title id, and other miscellaneous data.
class NACP {
public:
explicit NACP(VirtualFile file);
~NACP();
const LanguageEntry& GetLanguageEntry(Language language = Language::Default) const;
std::string GetApplicationName(Language language = Language::Default) const;
std::string GetDeveloperName(Language language = Language::Default) const;

View File

@@ -0,0 +1,366 @@
/*
* Copyright (c) 2018 Atmosphère-NX
*
* This program is free software; you can redistribute it and/or modify it
* under the terms and conditions of the GNU General Public License,
* version 2, as published by the Free Software Foundation.
*
* This program is distributed in the hope it will be useful, but WITHOUT
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
* FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
* more details.
*
* You should have received a copy of the GNU General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
/*
* Adapted by DarkLordZach for use/interaction with yuzu
*
* Modifications Copyright 2018 yuzu emulator team
* Licensed under GPLv2 or any later version
* Refer to the license.txt file included.
*/
#include <cstring>
#include "common/alignment.h"
#include "common/assert.h"
#include "core/file_sys/fsmitm_romfsbuild.h"
#include "core/file_sys/vfs.h"
#include "core/file_sys/vfs_vector.h"
namespace FileSys {
constexpr u64 FS_MAX_PATH = 0x301;
constexpr u32 ROMFS_ENTRY_EMPTY = 0xFFFFFFFF;
constexpr u32 ROMFS_FILEPARTITION_OFS = 0x200;
// Types for building a RomFS.
struct RomFSHeader {
u64 header_size;
u64 dir_hash_table_ofs;
u64 dir_hash_table_size;
u64 dir_table_ofs;
u64 dir_table_size;
u64 file_hash_table_ofs;
u64 file_hash_table_size;
u64 file_table_ofs;
u64 file_table_size;
u64 file_partition_ofs;
};
static_assert(sizeof(RomFSHeader) == 0x50, "RomFSHeader has incorrect size.");
struct RomFSDirectoryEntry {
u32 parent;
u32 sibling;
u32 child;
u32 file;
u32 hash;
u32 name_size;
};
static_assert(sizeof(RomFSDirectoryEntry) == 0x18, "RomFSDirectoryEntry has incorrect size.");
struct RomFSFileEntry {
u32 parent;
u32 sibling;
u64 offset;
u64 size;
u32 hash;
u32 name_size;
};
static_assert(sizeof(RomFSFileEntry) == 0x20, "RomFSFileEntry has incorrect size.");
struct RomFSBuildFileContext;
struct RomFSBuildDirectoryContext {
std::string path;
u32 cur_path_ofs = 0;
u32 path_len = 0;
u32 entry_offset = 0;
std::shared_ptr<RomFSBuildDirectoryContext> parent;
std::shared_ptr<RomFSBuildDirectoryContext> child;
std::shared_ptr<RomFSBuildDirectoryContext> sibling;
std::shared_ptr<RomFSBuildFileContext> file;
};
struct RomFSBuildFileContext {
std::string path;
u32 cur_path_ofs = 0;
u32 path_len = 0;
u32 entry_offset = 0;
u64 offset = 0;
u64 size = 0;
std::shared_ptr<RomFSBuildDirectoryContext> parent;
std::shared_ptr<RomFSBuildFileContext> sibling;
VirtualFile source;
};
static u32 romfs_calc_path_hash(u32 parent, std::string path, u32 start, std::size_t path_len) {
u32 hash = parent ^ 123456789;
for (u32 i = 0; i < path_len; i++) {
hash = (hash >> 5) | (hash << 27);
hash ^= path[start + i];
}
return hash;
}
static u64 romfs_get_hash_table_count(u64 num_entries) {
if (num_entries < 3) {
return 3;
}
if (num_entries < 19) {
return num_entries | 1;
}
u64 count = num_entries;
while (count % 2 == 0 || count % 3 == 0 || count % 5 == 0 || count % 7 == 0 ||
count % 11 == 0 || count % 13 == 0 || count % 17 == 0) {
count++;
}
return count;
}
void RomFSBuildContext::VisitDirectory(VirtualDir root_romfs,
std::shared_ptr<RomFSBuildDirectoryContext> parent) {
std::vector<std::shared_ptr<RomFSBuildDirectoryContext>> child_dirs;
VirtualDir dir;
if (parent->path_len == 0)
dir = root_romfs;
else
dir = root_romfs->GetDirectoryRelative(parent->path);
const auto entries = dir->GetEntries();
for (const auto& kv : entries) {
if (kv.second == VfsEntryType::Directory) {
const auto child = std::make_shared<RomFSBuildDirectoryContext>();
// Set child's path.
child->cur_path_ofs = parent->path_len + 1;
child->path_len = child->cur_path_ofs + static_cast<u32>(kv.first.size());
child->path = parent->path + "/" + kv.first;
// Sanity check on path_len
ASSERT(child->path_len < FS_MAX_PATH);
if (AddDirectory(parent, child)) {
child_dirs.push_back(child);
}
} else {
const auto child = std::make_shared<RomFSBuildFileContext>();
// Set child's path.
child->cur_path_ofs = parent->path_len + 1;
child->path_len = child->cur_path_ofs + static_cast<u32>(kv.first.size());
child->path = parent->path + "/" + kv.first;
// Sanity check on path_len
ASSERT(child->path_len < FS_MAX_PATH);
child->source = root_romfs->GetFileRelative(child->path);
child->size = child->source->GetSize();
AddFile(parent, child);
}
}
for (auto& child : child_dirs) {
this->VisitDirectory(root_romfs, child);
}
}
bool RomFSBuildContext::AddDirectory(std::shared_ptr<RomFSBuildDirectoryContext> parent_dir_ctx,
std::shared_ptr<RomFSBuildDirectoryContext> dir_ctx) {
// Check whether it's already in the known directories.
const auto existing = directories.find(dir_ctx->path);
if (existing != directories.end())
return false;
// Add a new directory.
num_dirs++;
dir_table_size +=
sizeof(RomFSDirectoryEntry) + Common::AlignUp(dir_ctx->path_len - dir_ctx->cur_path_ofs, 4);
dir_ctx->parent = parent_dir_ctx;
directories.emplace(dir_ctx->path, dir_ctx);
return true;
}
bool RomFSBuildContext::AddFile(std::shared_ptr<RomFSBuildDirectoryContext> parent_dir_ctx,
std::shared_ptr<RomFSBuildFileContext> file_ctx) {
// Check whether it's already in the known files.
const auto existing = files.find(file_ctx->path);
if (existing != files.end()) {
return false;
}
// Add a new file.
num_files++;
file_table_size +=
sizeof(RomFSFileEntry) + Common::AlignUp(file_ctx->path_len - file_ctx->cur_path_ofs, 4);
file_ctx->parent = parent_dir_ctx;
files.emplace(file_ctx->path, file_ctx);
return true;
}
RomFSBuildContext::RomFSBuildContext(VirtualDir base_) : base(std::move(base_)) {
root = std::make_shared<RomFSBuildDirectoryContext>();
root->path = "\0";
directories.emplace(root->path, root);
num_dirs = 1;
dir_table_size = 0x18;
VisitDirectory(base, root);
}
RomFSBuildContext::~RomFSBuildContext() = default;
std::map<u64, VirtualFile> RomFSBuildContext::Build() {
const u64 dir_hash_table_entry_count = romfs_get_hash_table_count(num_dirs);
const u64 file_hash_table_entry_count = romfs_get_hash_table_count(num_files);
dir_hash_table_size = 4 * dir_hash_table_entry_count;
file_hash_table_size = 4 * file_hash_table_entry_count;
// Assign metadata pointers
RomFSHeader header{};
std::vector<u32> dir_hash_table(dir_hash_table_entry_count, ROMFS_ENTRY_EMPTY);
std::vector<u32> file_hash_table(file_hash_table_entry_count, ROMFS_ENTRY_EMPTY);
std::vector<u8> dir_table(dir_table_size);
std::vector<u8> file_table(file_table_size);
std::shared_ptr<RomFSBuildFileContext> cur_file;
// Determine file offsets.
u32 entry_offset = 0;
std::shared_ptr<RomFSBuildFileContext> prev_file = nullptr;
for (const auto& it : files) {
cur_file = it.second;
file_partition_size = Common::AlignUp(file_partition_size, 16);
cur_file->offset = file_partition_size;
file_partition_size += cur_file->size;
cur_file->entry_offset = entry_offset;
entry_offset += sizeof(RomFSFileEntry) +
Common::AlignUp(cur_file->path_len - cur_file->cur_path_ofs, 4);
prev_file = cur_file;
}
// Assign deferred parent/sibling ownership.
for (auto it = files.rbegin(); it != files.rend(); ++it) {
cur_file = it->second;
cur_file->sibling = cur_file->parent->file;
cur_file->parent->file = cur_file;
}
std::shared_ptr<RomFSBuildDirectoryContext> cur_dir;
// Determine directory offsets.
entry_offset = 0;
for (const auto& it : directories) {
cur_dir = it.second;
cur_dir->entry_offset = entry_offset;
entry_offset += sizeof(RomFSDirectoryEntry) +
Common::AlignUp(cur_dir->path_len - cur_dir->cur_path_ofs, 4);
}
// Assign deferred parent/sibling ownership.
for (auto it = directories.rbegin(); it->second != root; ++it) {
cur_dir = it->second;
cur_dir->sibling = cur_dir->parent->child;
cur_dir->parent->child = cur_dir;
}
std::map<u64, VirtualFile> out;
// Populate file tables.
for (const auto& it : files) {
cur_file = it.second;
RomFSFileEntry cur_entry{};
cur_entry.parent = cur_file->parent->entry_offset;
cur_entry.sibling =
cur_file->sibling == nullptr ? ROMFS_ENTRY_EMPTY : cur_file->sibling->entry_offset;
cur_entry.offset = cur_file->offset;
cur_entry.size = cur_file->size;
const auto name_size = cur_file->path_len - cur_file->cur_path_ofs;
const auto hash = romfs_calc_path_hash(cur_file->parent->entry_offset, cur_file->path,
cur_file->cur_path_ofs, name_size);
cur_entry.hash = file_hash_table[hash % file_hash_table_entry_count];
file_hash_table[hash % file_hash_table_entry_count] = cur_file->entry_offset;
cur_entry.name_size = name_size;
out.emplace(cur_file->offset + ROMFS_FILEPARTITION_OFS, cur_file->source);
std::memcpy(file_table.data() + cur_file->entry_offset, &cur_entry, sizeof(RomFSFileEntry));
std::memset(file_table.data() + cur_file->entry_offset + sizeof(RomFSFileEntry), 0,
Common::AlignUp(cur_entry.name_size, 4));
std::memcpy(file_table.data() + cur_file->entry_offset + sizeof(RomFSFileEntry),
cur_file->path.data() + cur_file->cur_path_ofs, name_size);
}
// Populate dir tables.
for (const auto& it : directories) {
cur_dir = it.second;
RomFSDirectoryEntry cur_entry{};
cur_entry.parent = cur_dir == root ? 0 : cur_dir->parent->entry_offset;
cur_entry.sibling =
cur_dir->sibling == nullptr ? ROMFS_ENTRY_EMPTY : cur_dir->sibling->entry_offset;
cur_entry.child =
cur_dir->child == nullptr ? ROMFS_ENTRY_EMPTY : cur_dir->child->entry_offset;
cur_entry.file = cur_dir->file == nullptr ? ROMFS_ENTRY_EMPTY : cur_dir->file->entry_offset;
const auto name_size = cur_dir->path_len - cur_dir->cur_path_ofs;
const auto hash = romfs_calc_path_hash(cur_dir == root ? 0 : cur_dir->parent->entry_offset,
cur_dir->path, cur_dir->cur_path_ofs, name_size);
cur_entry.hash = dir_hash_table[hash % dir_hash_table_entry_count];
dir_hash_table[hash % dir_hash_table_entry_count] = cur_dir->entry_offset;
cur_entry.name_size = name_size;
std::memcpy(dir_table.data() + cur_dir->entry_offset, &cur_entry,
sizeof(RomFSDirectoryEntry));
std::memset(dir_table.data() + cur_dir->entry_offset + sizeof(RomFSDirectoryEntry), 0,
Common::AlignUp(cur_entry.name_size, 4));
std::memcpy(dir_table.data() + cur_dir->entry_offset + sizeof(RomFSDirectoryEntry),
cur_dir->path.data() + cur_dir->cur_path_ofs, name_size);
}
// Set header fields.
header.header_size = sizeof(RomFSHeader);
header.file_hash_table_size = file_hash_table_size;
header.file_table_size = file_table_size;
header.dir_hash_table_size = dir_hash_table_size;
header.dir_table_size = dir_table_size;
header.file_partition_ofs = ROMFS_FILEPARTITION_OFS;
header.dir_hash_table_ofs = Common::AlignUp(header.file_partition_ofs + file_partition_size, 4);
header.dir_table_ofs = header.dir_hash_table_ofs + header.dir_hash_table_size;
header.file_hash_table_ofs = header.dir_table_ofs + header.dir_table_size;
header.file_table_ofs = header.file_hash_table_ofs + header.file_hash_table_size;
std::vector<u8> header_data(sizeof(RomFSHeader));
std::memcpy(header_data.data(), &header, header_data.size());
out.emplace(0, std::make_shared<VectorVfsFile>(std::move(header_data)));
std::vector<u8> metadata(file_hash_table_size + file_table_size + dir_hash_table_size +
dir_table_size);
std::size_t index = 0;
std::memcpy(metadata.data(), dir_hash_table.data(), dir_hash_table.size() * sizeof(u32));
index += dir_hash_table.size() * sizeof(u32);
std::memcpy(metadata.data() + index, dir_table.data(), dir_table.size());
index += dir_table.size();
std::memcpy(metadata.data() + index, file_hash_table.data(),
file_hash_table.size() * sizeof(u32));
index += file_hash_table.size() * sizeof(u32);
std::memcpy(metadata.data() + index, file_table.data(), file_table.size());
out.emplace(header.dir_hash_table_ofs, std::make_shared<VectorVfsFile>(std::move(metadata)));
return out;
}
} // namespace FileSys

View File

@@ -0,0 +1,70 @@
/*
* Copyright (c) 2018 Atmosphère-NX
*
* This program is free software; you can redistribute it and/or modify it
* under the terms and conditions of the GNU General Public License,
* version 2, as published by the Free Software Foundation.
*
* This program is distributed in the hope it will be useful, but WITHOUT
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
* FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
* more details.
*
* You should have received a copy of the GNU General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
/*
* Adapted by DarkLordZach for use/interaction with yuzu
*
* Modifications Copyright 2018 yuzu emulator team
* Licensed under GPLv2 or any later version
* Refer to the license.txt file included.
*/
#pragma once
#include <map>
#include <memory>
#include <string>
#include <boost/detail/container_fwd.hpp>
#include "common/common_types.h"
#include "core/file_sys/vfs.h"
namespace FileSys {
struct RomFSBuildDirectoryContext;
struct RomFSBuildFileContext;
struct RomFSDirectoryEntry;
struct RomFSFileEntry;
class RomFSBuildContext {
public:
explicit RomFSBuildContext(VirtualDir base);
~RomFSBuildContext();
// This finalizes the context.
std::map<u64, VirtualFile> Build();
private:
VirtualDir base;
std::shared_ptr<RomFSBuildDirectoryContext> root;
std::map<std::string, std::shared_ptr<RomFSBuildDirectoryContext>, std::less<>> directories;
std::map<std::string, std::shared_ptr<RomFSBuildFileContext>, std::less<>> files;
u64 num_dirs = 0;
u64 num_files = 0;
u64 dir_table_size = 0;
u64 file_table_size = 0;
u64 dir_hash_table_size = 0;
u64 file_hash_table_size = 0;
u64 file_partition_size = 0;
void VisitDirectory(VirtualDir filesys, std::shared_ptr<RomFSBuildDirectoryContext> parent);
bool AddDirectory(std::shared_ptr<RomFSBuildDirectoryContext> parent_dir_ctx,
std::shared_ptr<RomFSBuildDirectoryContext> dir_ctx);
bool AddFile(std::shared_ptr<RomFSBuildDirectoryContext> parent_dir_ctx,
std::shared_ptr<RomFSBuildFileContext> file_ctx);
};
} // namespace FileSys

View File

@@ -51,6 +51,8 @@ CNMT::CNMT(CNMTHeader header, OptionalHeader opt_header, std::vector<ContentReco
: header(std::move(header)), opt_header(std::move(opt_header)),
content_records(std::move(content_records)), meta_records(std::move(meta_records)) {}
CNMT::~CNMT() = default;
u64 CNMT::GetTitleID() const {
return header.title_id;
}

View File

@@ -87,6 +87,7 @@ public:
explicit CNMT(VirtualFile file);
CNMT(CNMTHeader header, OptionalHeader opt_header, std::vector<ContentRecord> content_records,
std::vector<MetaRecord> meta_records);
~CNMT();
u64 GetTitleID() const;
u32 GetTitleVersion() const;

View File

@@ -72,6 +72,8 @@ PartitionFilesystem::PartitionFilesystem(std::shared_ptr<VfsFile> file) {
status = Loader::ResultStatus::Success;
}
PartitionFilesystem::~PartitionFilesystem() = default;
Loader::ResultStatus PartitionFilesystem::GetStatus() const {
return status;
}

View File

@@ -25,6 +25,8 @@ namespace FileSys {
class PartitionFilesystem : public ReadOnlyVfsDirectory {
public:
explicit PartitionFilesystem(std::shared_ptr<VfsFile> file);
~PartitionFilesystem() override;
Loader::ResultStatus GetStatus() const;
std::vector<std::shared_ptr<VfsFile>> GetFiles() const override;

View File

@@ -11,6 +11,7 @@
#include "core/file_sys/patch_manager.h"
#include "core/file_sys/registered_cache.h"
#include "core/file_sys/romfs.h"
#include "core/file_sys/vfs_layered.h"
#include "core/hle/service/filesystem/filesystem.h"
#include "core/loader/loader.h"
@@ -31,8 +32,9 @@ std::string FormatTitleVersion(u32 version, TitleVersionFormat format) {
return fmt::format("v{}.{}.{}", bytes[3], bytes[2], bytes[1]);
}
constexpr std::array<const char*, 1> PATCH_TYPE_NAMES{
constexpr std::array<const char*, 2> PATCH_TYPE_NAMES{
"Update",
"LayeredFS",
};
std::string FormatPatchTypeName(PatchType type) {
@@ -41,6 +43,8 @@ std::string FormatPatchTypeName(PatchType type) {
PatchManager::PatchManager(u64 title_id) : title_id(title_id) {}
PatchManager::~PatchManager() = default;
VirtualDir PatchManager::PatchExeFS(VirtualDir exefs) const {
LOG_INFO(Loader, "Patching ExeFS for title_id={:016X}", title_id);
@@ -64,6 +68,44 @@ VirtualDir PatchManager::PatchExeFS(VirtualDir exefs) const {
return exefs;
}
static void ApplyLayeredFS(VirtualFile& romfs, u64 title_id, ContentRecordType type) {
const auto load_dir = Service::FileSystem::GetModificationLoadRoot(title_id);
if (type != ContentRecordType::Program || load_dir == nullptr || load_dir->GetSize() <= 0) {
return;
}
auto extracted = ExtractRomFS(romfs);
if (extracted == nullptr) {
return;
}
auto patch_dirs = load_dir->GetSubdirectories();
std::sort(patch_dirs.begin(), patch_dirs.end(),
[](const VirtualDir& l, const VirtualDir& r) { return l->GetName() < r->GetName(); });
std::vector<VirtualDir> layers;
layers.reserve(patch_dirs.size() + 1);
for (const auto& subdir : patch_dirs) {
auto romfs_dir = subdir->GetSubdirectory("romfs");
if (romfs_dir != nullptr)
layers.push_back(std::move(romfs_dir));
}
layers.push_back(std::move(extracted));
auto layered = LayeredVfsDirectory::MakeLayeredDirectory(std::move(layers));
if (layered == nullptr) {
return;
}
auto packed = CreateRomFS(std::move(layered));
if (packed == nullptr) {
return;
}
LOG_INFO(Loader, " RomFS: LayeredFS patches applied successfully");
romfs = std::move(packed);
}
VirtualFile PatchManager::PatchRomFS(VirtualFile romfs, u64 ivfc_offset,
ContentRecordType type) const {
LOG_INFO(Loader, "Patching RomFS for title_id={:016X}, type={:02X}", title_id,
@@ -87,6 +129,9 @@ VirtualFile PatchManager::PatchRomFS(VirtualFile romfs, u64 ivfc_offset,
}
}
// LayeredFS
ApplyLayeredFS(romfs, title_id, type);
return romfs;
}
@@ -112,6 +157,10 @@ std::map<PatchType, std::string> PatchManager::GetPatchVersionNames() const {
}
}
const auto lfs_dir = Service::FileSystem::GetModificationLoadRoot(title_id);
if (lfs_dir != nullptr && lfs_dir->GetSize() > 0)
out.insert_or_assign(PatchType::LayeredFS, "");
return out;
}

View File

@@ -26,6 +26,7 @@ std::string FormatTitleVersion(u32 version,
enum class PatchType {
Update,
LayeredFS,
};
std::string FormatPatchTypeName(PatchType type);
@@ -34,6 +35,7 @@ std::string FormatPatchTypeName(PatchType type);
class PatchManager {
public:
explicit PatchManager(u64 title_id);
~PatchManager();
// Currently tracked ExeFS patches:
// - Game Updates
@@ -41,6 +43,7 @@ public:
// Currently tracked RomFS patches:
// - Game Updates
// - LayeredFS
VirtualFile PatchRomFS(VirtualFile base, u64 ivfc_offset,
ContentRecordType type = ContentRecordType::Program) const;

View File

@@ -12,6 +12,10 @@
namespace FileSys {
ProgramMetadata::ProgramMetadata() = default;
ProgramMetadata::~ProgramMetadata() = default;
Loader::ResultStatus ProgramMetadata::Load(VirtualFile file) {
std::size_t total_size = static_cast<std::size_t>(file->GetSize());
if (total_size < sizeof(Header))
@@ -79,10 +83,12 @@ void ProgramMetadata::Print() const {
auto address_space = "Unknown";
switch (npdm_header.address_space_type) {
case ProgramAddressSpaceType::Is64Bit:
case ProgramAddressSpaceType::Is36Bit:
case ProgramAddressSpaceType::Is39Bit:
address_space = "64-bit";
break;
case ProgramAddressSpaceType::Is32Bit:
case ProgramAddressSpaceType::Is32BitNoMap:
address_space = "32-bit";
break;
}

View File

@@ -17,8 +17,10 @@ enum class ResultStatus : u16;
namespace FileSys {
enum class ProgramAddressSpaceType : u8 {
Is64Bit = 1,
Is32Bit = 2,
Is32Bit = 0,
Is36Bit = 1,
Is32BitNoMap = 2,
Is39Bit = 3,
};
enum class ProgramFilePermission : u64 {
@@ -36,6 +38,9 @@ enum class ProgramFilePermission : u64 {
*/
class ProgramMetadata {
public:
ProgramMetadata();
~ProgramMetadata();
Loader::ResultStatus Load(VirtualFile file);
bool Is64BitProgram() const;

View File

@@ -18,6 +18,10 @@
#include "core/loader/loader.h"
namespace FileSys {
// The size of blocks to use when vfs raw copying into nand.
constexpr size_t VFS_RC_LARGE_COPY_BLOCK = 0x400000;
std::string RegisteredCacheEntry::DebugInfo() const {
return fmt::format("title_id={:016X}, content_type={:02X}", title_id, static_cast<u8>(type));
}
@@ -121,7 +125,7 @@ VirtualFile RegisteredCache::OpenFileOrDirectoryConcat(const VirtualDir& dir,
if (concat.empty())
return nullptr;
file = FileSys::ConcatenateFiles(concat);
file = ConcatenatedVfsFile::MakeConcatenatedFile(concat, concat.front()->GetName());
}
return file;
@@ -480,7 +484,8 @@ InstallResult RegisteredCache::RawInstallNCA(std::shared_ptr<NCA> nca, const Vfs
auto out = dir->CreateFileRelative(path);
if (out == nullptr)
return InstallResult::ErrorCopyFailed;
return copy(in, out) ? InstallResult::Success : InstallResult::ErrorCopyFailed;
return copy(in, out, VFS_RC_LARGE_COPY_BLOCK) ? InstallResult::Success
: InstallResult::ErrorCopyFailed;
}
bool RegisteredCache::RawInstallYuzuMeta(const CNMT& cnmt) {

View File

@@ -27,7 +27,7 @@ struct ContentRecord;
using NcaID = std::array<u8, 0x10>;
using RegisteredCacheParsingFunction = std::function<VirtualFile(const VirtualFile&, const NcaID&)>;
using VfsCopyFunction = std::function<bool(VirtualFile, VirtualFile)>;
using VfsCopyFunction = std::function<bool(const VirtualFile&, const VirtualFile&, size_t)>;
enum class InstallResult {
Success,

View File

@@ -4,8 +4,10 @@
#include "common/common_types.h"
#include "common/swap.h"
#include "core/file_sys/fsmitm_romfsbuild.h"
#include "core/file_sys/romfs.h"
#include "core/file_sys/vfs.h"
#include "core/file_sys/vfs_concat.h"
#include "core/file_sys/vfs_offset.h"
#include "core/file_sys/vfs_vector.h"
@@ -98,7 +100,7 @@ void ProcessDirectory(VirtualFile file, std::size_t dir_offset, std::size_t file
}
}
VirtualDir ExtractRomFS(VirtualFile file) {
VirtualDir ExtractRomFS(VirtualFile file, RomFSExtractionType type) {
RomFSHeader header{};
if (file->ReadObject(&header) != sizeof(RomFSHeader))
return nullptr;
@@ -117,9 +119,22 @@ VirtualDir ExtractRomFS(VirtualFile file) {
VirtualDir out = std::move(root);
while (out->GetSubdirectory("") != nullptr)
out = out->GetSubdirectory("");
while (out->GetSubdirectories().size() == 1 && out->GetFiles().empty()) {
if (out->GetSubdirectories().front()->GetName() == "data" &&
type == RomFSExtractionType::Truncated)
break;
out = out->GetSubdirectories().front();
}
return out;
}
VirtualFile CreateRomFS(VirtualDir dir) {
if (dir == nullptr)
return nullptr;
RomFSBuildContext ctx{dir};
return ConcatenatedVfsFile::MakeConcatenatedFile(0, ctx.Build(), dir->GetName());
}
} // namespace FileSys

View File

@@ -5,6 +5,7 @@
#pragma once
#include <array>
#include <map>
#include "common/common_funcs.h"
#include "common/common_types.h"
#include "common/swap.h"
@@ -12,6 +13,8 @@
namespace FileSys {
struct RomFSHeader;
struct IVFCLevel {
u64_le offset;
u64_le size;
@@ -29,8 +32,18 @@ struct IVFCHeader {
};
static_assert(sizeof(IVFCHeader) == 0xE0, "IVFCHeader has incorrect size.");
enum class RomFSExtractionType {
Full, // Includes data directory
Truncated, // Traverses into data directory
};
// Converts a RomFS binary blob to VFS Filesystem
// Returns nullptr on failure
VirtualDir ExtractRomFS(VirtualFile file);
VirtualDir ExtractRomFS(VirtualFile file,
RomFSExtractionType type = RomFSExtractionType::Truncated);
// Converts a VFS filesystem into a RomFS binary
// Returns nullptr on failure
VirtualFile CreateRomFS(VirtualDir dir);
} // namespace FileSys

View File

@@ -28,11 +28,13 @@ RomFSFactory::RomFSFactory(Loader::AppLoader& app_loader) {
ivfc_offset = app_loader.ReadRomFSIVFCOffset();
}
RomFSFactory::~RomFSFactory() = default;
ResultVal<VirtualFile> RomFSFactory::OpenCurrentProcess() {
if (!updatable)
return MakeResult<VirtualFile>(file);
const PatchManager patch_manager(Core::CurrentProcess()->program_id);
const PatchManager patch_manager(Core::CurrentProcess()->GetTitleID());
return MakeResult<VirtualFile>(patch_manager.PatchRomFS(file, ivfc_offset));
}

View File

@@ -30,6 +30,7 @@ enum class StorageId : u8 {
class RomFSFactory {
public:
explicit RomFSFactory(Loader::AppLoader& app_loader);
~RomFSFactory();
ResultVal<VirtualFile> OpenCurrentProcess();
ResultVal<VirtualFile> Open(u64 title_id, StorageId storage, ContentRecordType type);

View File

@@ -20,6 +20,8 @@ std::string SaveDataDescriptor::DebugInfo() const {
SaveDataFactory::SaveDataFactory(VirtualDir save_directory) : dir(std::move(save_directory)) {}
SaveDataFactory::~SaveDataFactory() = default;
ResultVal<VirtualDir> SaveDataFactory::Open(SaveDataSpaceId space, SaveDataDescriptor meta) {
if (meta.type == SaveDataType::SystemSaveData || meta.type == SaveDataType::SaveData) {
if (meta.zero_1 != 0) {
@@ -79,16 +81,16 @@ std::string SaveDataFactory::GetFullPath(SaveDataSpaceId space, SaveDataType typ
// According to switchbrew, if a save is of type SaveData and the title id field is 0, it should
// be interpreted as the title id of the current process.
if (type == SaveDataType::SaveData && title_id == 0)
title_id = Core::CurrentProcess()->program_id;
title_id = Core::CurrentProcess()->GetTitleID();
std::string out;
switch (space) {
case SaveDataSpaceId::NandSystem:
out = "/system/save/";
out = "/system/";
break;
case SaveDataSpaceId::NandUser:
out = "/user/save/";
out = "/user/";
break;
default:
ASSERT_MSG(false, "Unrecognized SaveDataSpaceId: {:02X}", static_cast<u8>(space));
@@ -96,9 +98,12 @@ std::string SaveDataFactory::GetFullPath(SaveDataSpaceId space, SaveDataType typ
switch (type) {
case SaveDataType::SystemSaveData:
return fmt::format("{}{:016X}/{:016X}{:016X}", out, save_id, user_id[1], user_id[0]);
return fmt::format("{}save/{:016X}/{:016X}{:016X}", out, save_id, user_id[1], user_id[0]);
case SaveDataType::SaveData:
return fmt::format("{}{:016X}/{:016X}{:016X}/{:016X}", out, 0, user_id[1], user_id[0],
return fmt::format("{}save/{:016X}/{:016X}{:016X}/{:016X}", out, 0, user_id[1], user_id[0],
title_id);
case SaveDataType::TemporaryStorage:
return fmt::format("{}temp/{:016X}/{:016X}{:016X}/{:016X}", out, 0, user_id[1], user_id[0],
title_id);
default:
ASSERT_MSG(false, "Unrecognized SaveDataType: {:02X}", static_cast<u8>(type));

View File

@@ -48,6 +48,7 @@ static_assert(sizeof(SaveDataDescriptor) == 0x40, "SaveDataDescriptor has incorr
class SaveDataFactory {
public:
explicit SaveDataFactory(VirtualDir dir);
~SaveDataFactory();
ResultVal<VirtualDir> Open(SaveDataSpaceId space, SaveDataDescriptor meta);

View File

@@ -24,7 +24,7 @@ enum class ContentRecordType : u8;
class NSP : public ReadOnlyVfsDirectory {
public:
explicit NSP(VirtualFile file);
~NSP();
~NSP() override;
Loader::ResultStatus GetStatus() const;
Loader::ResultStatus GetProgramStatus(u64 title_id) const;

View File

@@ -399,6 +399,15 @@ bool VfsDirectory::Copy(std::string_view src, std::string_view dest) {
return f2->WriteBytes(f1->ReadAllBytes()) == f1->GetSize();
}
std::map<std::string, VfsEntryType, std::less<>> VfsDirectory::GetEntries() const {
std::map<std::string, VfsEntryType, std::less<>> out;
for (const auto& dir : GetSubdirectories())
out.emplace(dir->GetName(), VfsEntryType::Directory);
for (const auto& file : GetFiles())
out.emplace(file->GetName(), VfsEntryType::File);
return out;
}
std::string VfsDirectory::GetFullPath() const {
if (IsRoot())
return GetName();
@@ -454,13 +463,41 @@ bool DeepEquals(const VirtualFile& file1, const VirtualFile& file2, std::size_t
return true;
}
bool VfsRawCopy(VirtualFile src, VirtualFile dest) {
if (src == nullptr || dest == nullptr)
bool VfsRawCopy(const VirtualFile& src, const VirtualFile& dest, std::size_t block_size) {
if (src == nullptr || dest == nullptr || !src->IsReadable() || !dest->IsWritable())
return false;
if (!dest->Resize(src->GetSize()))
return false;
std::vector<u8> data = src->ReadAllBytes();
return dest->WriteBytes(data, 0) == data.size();
std::vector<u8> temp(std::min(block_size, src->GetSize()));
for (std::size_t i = 0; i < src->GetSize(); i += block_size) {
const auto read = std::min(block_size, src->GetSize() - i);
const auto block = src->Read(temp.data(), read, i);
if (dest->Write(temp.data(), read, i) != read)
return false;
}
return true;
}
bool VfsRawCopyD(const VirtualDir& src, const VirtualDir& dest, std::size_t block_size) {
if (src == nullptr || dest == nullptr || !src->IsReadable() || !dest->IsWritable())
return false;
for (const auto& file : src->GetFiles()) {
const auto out = dest->CreateFile(file->GetName());
if (!VfsRawCopy(file, out, block_size))
return false;
}
for (const auto& dir : src->GetSubdirectories()) {
const auto out = dest->CreateSubdirectory(dir->GetName());
if (!VfsRawCopyD(dir, out, block_size))
return false;
}
return true;
}
VirtualDir GetOrCreateDirectoryRelative(const VirtualDir& rel, std::string_view path) {

View File

@@ -4,6 +4,7 @@
#pragma once
#include <map>
#include <memory>
#include <string>
#include <string_view>
@@ -265,6 +266,10 @@ public:
// dest.
virtual bool Copy(std::string_view src, std::string_view dest);
// Gets all of the entries directly in the directory (files and dirs), returning a map between
// item name -> type.
virtual std::map<std::string, VfsEntryType, std::less<>> GetEntries() const;
// Interprets the file with name file instead as a directory of type directory.
// The directory must have a constructor that takes a single argument of type
// std::shared_ptr<VfsFile>. Allows to reinterpret container files (i.e NCA, zip, XCI, etc) as a
@@ -310,13 +315,19 @@ public:
bool Rename(std::string_view name) override;
};
// Compare the two files, byte-for-byte, in increments specificed by block_size
bool DeepEquals(const VirtualFile& file1, const VirtualFile& file2, std::size_t block_size = 0x200);
// Compare the two files, byte-for-byte, in increments specified by block_size
bool DeepEquals(const VirtualFile& file1, const VirtualFile& file2,
std::size_t block_size = 0x1000);
// A method that copies the raw data between two different implementations of VirtualFile. If you
// are using the same implementation, it is probably better to use the Copy method in the parent
// directory of src/dest.
bool VfsRawCopy(VirtualFile src, VirtualFile dest);
bool VfsRawCopy(const VirtualFile& src, const VirtualFile& dest, std::size_t block_size = 0x1000);
// A method that performs a similar function to VfsRawCopy above, but instead copies entire
// directories. It suffers the same performance penalties as above and an implementation-specific
// Copy should always be preferred.
bool VfsRawCopyD(const VirtualDir& src, const VirtualDir& dest, std::size_t block_size = 0x1000);
// Checks if the directory at path relative to rel exists. If it does, returns that. If it does not
// it attempts to create it and returns the new dir or nullptr on failure.

View File

@@ -5,17 +5,22 @@
#include <algorithm>
#include <utility>
#include "common/assert.h"
#include "core/file_sys/vfs_concat.h"
#include "core/file_sys/vfs_static.h"
namespace FileSys {
VirtualFile ConcatenateFiles(std::vector<VirtualFile> files, std::string name) {
if (files.empty())
return nullptr;
if (files.size() == 1)
return files[0];
static bool VerifyConcatenationMapContinuity(const std::map<u64, VirtualFile>& map) {
const auto last_valid = --map.end();
for (auto iter = map.begin(); iter != last_valid;) {
const auto old = iter++;
if (old->first + old->second->GetSize() != iter->first) {
return false;
}
}
return std::shared_ptr<VfsFile>(new ConcatenatedVfsFile(std::move(files), std::move(name)));
return map.begin()->first == 0;
}
ConcatenatedVfsFile::ConcatenatedVfsFile(std::vector<VirtualFile> files_, std::string name)
@@ -27,6 +32,48 @@ ConcatenatedVfsFile::ConcatenatedVfsFile(std::vector<VirtualFile> files_, std::s
}
}
ConcatenatedVfsFile::ConcatenatedVfsFile(std::map<u64, VirtualFile> files_, std::string name)
: files(std::move(files_)), name(std::move(name)) {
ASSERT(VerifyConcatenationMapContinuity(files));
}
ConcatenatedVfsFile::~ConcatenatedVfsFile() = default;
VirtualFile ConcatenatedVfsFile::MakeConcatenatedFile(std::vector<VirtualFile> files,
std::string name) {
if (files.empty())
return nullptr;
if (files.size() == 1)
return files[0];
return std::shared_ptr<VfsFile>(new ConcatenatedVfsFile(std::move(files), std::move(name)));
}
VirtualFile ConcatenatedVfsFile::MakeConcatenatedFile(u8 filler_byte,
std::map<u64, VirtualFile> files,
std::string name) {
if (files.empty())
return nullptr;
if (files.size() == 1)
return files.begin()->second;
const auto last_valid = --files.end();
for (auto iter = files.begin(); iter != last_valid;) {
const auto old = iter++;
if (old->first + old->second->GetSize() != iter->first) {
files.emplace(old->first + old->second->GetSize(),
std::make_shared<StaticVfsFile>(filler_byte, iter->first - old->first -
old->second->GetSize()));
}
}
// Ensure the map starts at offset 0 (start of file), otherwise pad to fill.
if (files.begin()->first != 0)
files.emplace(0, std::make_shared<StaticVfsFile>(filler_byte, files.begin()->first));
return std::shared_ptr<VfsFile>(new ConcatenatedVfsFile(std::move(files), std::move(name)));
}
std::string ConcatenatedVfsFile::GetName() const {
if (files.empty())
return "";
@@ -60,7 +107,7 @@ bool ConcatenatedVfsFile::IsReadable() const {
}
std::size_t ConcatenatedVfsFile::Read(u8* data, std::size_t length, std::size_t offset) const {
auto entry = files.end();
auto entry = --files.end();
for (auto iter = files.begin(); iter != files.end(); ++iter) {
if (iter->first > offset) {
entry = --iter;
@@ -68,20 +115,17 @@ std::size_t ConcatenatedVfsFile::Read(u8* data, std::size_t length, std::size_t
}
}
// Check if the entry should be the last one. The loop above will make it end().
if (entry == files.end() && offset < files.rbegin()->first + files.rbegin()->second->GetSize())
--entry;
if (entry == files.end())
if (entry->first + entry->second->GetSize() <= offset)
return 0;
const auto remaining = entry->second->GetSize() + offset - entry->first;
if (length > remaining) {
return entry->second->Read(data, remaining, offset - entry->first) +
Read(data + remaining, length - remaining, offset + remaining);
const auto read_in =
std::min<u64>(entry->first + entry->second->GetSize() - offset, entry->second->GetSize());
if (length > read_in) {
return entry->second->Read(data, read_in, offset - entry->first) +
Read(data + read_in, length - read_in, offset + read_in);
}
return entry->second->Read(data, length, offset - entry->first);
return entry->second->Read(data, std::min<u64>(read_in, length), offset - entry->first);
}
std::size_t ConcatenatedVfsFile::Write(const u8* data, std::size_t length, std::size_t offset) {
@@ -91,4 +135,5 @@ std::size_t ConcatenatedVfsFile::Write(const u8* data, std::size_t length, std::
bool ConcatenatedVfsFile::Rename(std::string_view name) {
return false;
}
} // namespace FileSys

View File

@@ -4,24 +4,30 @@
#pragma once
#include <map>
#include <memory>
#include <string_view>
#include <boost/container/flat_map.hpp>
#include "core/file_sys/vfs.h"
namespace FileSys {
// Wrapper function to allow for more efficient handling of files.size() == 0, 1 cases.
VirtualFile ConcatenateFiles(std::vector<VirtualFile> files, std::string name = "");
// Class that wraps multiple vfs files and concatenates them, making reads seamless. Currently
// read-only.
class ConcatenatedVfsFile : public VfsFile {
friend VirtualFile ConcatenateFiles(std::vector<VirtualFile> files, std::string name);
ConcatenatedVfsFile(std::vector<VirtualFile> files, std::string name);
ConcatenatedVfsFile(std::map<u64, VirtualFile> files, std::string name);
public:
~ConcatenatedVfsFile() override;
/// Wrapper function to allow for more efficient handling of files.size() == 0, 1 cases.
static VirtualFile MakeConcatenatedFile(std::vector<VirtualFile> files, std::string name);
/// Convenience function that turns a map of offsets to files into a concatenated file, filling
/// gaps with a given filler byte.
static VirtualFile MakeConcatenatedFile(u8 filler_byte, std::map<u64, VirtualFile> files,
std::string name);
std::string GetName() const override;
std::size_t GetSize() const override;
bool Resize(std::size_t new_size) override;
@@ -34,7 +40,7 @@ public:
private:
// Maps starting offset to file -- more efficient.
boost::container::flat_map<u64, VirtualFile> files;
std::map<u64, VirtualFile> files;
std::string name;
};

View File

@@ -0,0 +1,132 @@
// Copyright 2018 yuzu emulator team
// Licensed under GPLv2 or any later version
// Refer to the license.txt file included.
#include <algorithm>
#include <utility>
#include "core/file_sys/vfs_layered.h"
namespace FileSys {
LayeredVfsDirectory::LayeredVfsDirectory(std::vector<VirtualDir> dirs, std::string name)
: dirs(std::move(dirs)), name(std::move(name)) {}
LayeredVfsDirectory::~LayeredVfsDirectory() = default;
VirtualDir LayeredVfsDirectory::MakeLayeredDirectory(std::vector<VirtualDir> dirs,
std::string name) {
if (dirs.empty())
return nullptr;
if (dirs.size() == 1)
return dirs[0];
return std::shared_ptr<VfsDirectory>(new LayeredVfsDirectory(std::move(dirs), std::move(name)));
}
std::shared_ptr<VfsFile> LayeredVfsDirectory::GetFileRelative(std::string_view path) const {
for (const auto& layer : dirs) {
const auto file = layer->GetFileRelative(path);
if (file != nullptr)
return file;
}
return nullptr;
}
std::shared_ptr<VfsDirectory> LayeredVfsDirectory::GetDirectoryRelative(
std::string_view path) const {
std::vector<VirtualDir> out;
for (const auto& layer : dirs) {
auto dir = layer->GetDirectoryRelative(path);
if (dir != nullptr)
out.push_back(std::move(dir));
}
return MakeLayeredDirectory(std::move(out));
}
std::shared_ptr<VfsFile> LayeredVfsDirectory::GetFile(std::string_view name) const {
return GetFileRelative(name);
}
std::shared_ptr<VfsDirectory> LayeredVfsDirectory::GetSubdirectory(std::string_view name) const {
return GetDirectoryRelative(name);
}
std::string LayeredVfsDirectory::GetFullPath() const {
return dirs[0]->GetFullPath();
}
std::vector<std::shared_ptr<VfsFile>> LayeredVfsDirectory::GetFiles() const {
std::vector<VirtualFile> out;
for (const auto& layer : dirs) {
for (const auto& file : layer->GetFiles()) {
if (std::find_if(out.begin(), out.end(), [&file](const VirtualFile& comp) {
return comp->GetName() == file->GetName();
}) == out.end()) {
out.push_back(file);
}
}
}
return out;
}
std::vector<std::shared_ptr<VfsDirectory>> LayeredVfsDirectory::GetSubdirectories() const {
std::vector<std::string> names;
for (const auto& layer : dirs) {
for (const auto& sd : layer->GetSubdirectories()) {
if (std::find(names.begin(), names.end(), sd->GetName()) == names.end())
names.push_back(sd->GetName());
}
}
std::vector<VirtualDir> out;
out.reserve(names.size());
for (const auto& subdir : names)
out.push_back(GetSubdirectory(subdir));
return out;
}
bool LayeredVfsDirectory::IsWritable() const {
return false;
}
bool LayeredVfsDirectory::IsReadable() const {
return true;
}
std::string LayeredVfsDirectory::GetName() const {
return name.empty() ? dirs[0]->GetName() : name;
}
std::shared_ptr<VfsDirectory> LayeredVfsDirectory::GetParentDirectory() const {
return dirs[0]->GetParentDirectory();
}
std::shared_ptr<VfsDirectory> LayeredVfsDirectory::CreateSubdirectory(std::string_view name) {
return nullptr;
}
std::shared_ptr<VfsFile> LayeredVfsDirectory::CreateFile(std::string_view name) {
return nullptr;
}
bool LayeredVfsDirectory::DeleteSubdirectory(std::string_view name) {
return false;
}
bool LayeredVfsDirectory::DeleteFile(std::string_view name) {
return false;
}
bool LayeredVfsDirectory::Rename(std::string_view name_) {
name = name_;
return true;
}
bool LayeredVfsDirectory::ReplaceFileWithSubdirectory(VirtualFile file, VirtualDir dir) {
return false;
}
} // namespace FileSys

View File

@@ -0,0 +1,50 @@
// Copyright 2018 yuzu emulator team
// Licensed under GPLv2 or any later version
// Refer to the license.txt file included.
#pragma once
#include <memory>
#include "core/file_sys/vfs.h"
namespace FileSys {
// Class that stacks multiple VfsDirectories on top of each other, attempting to read from the first
// one and falling back to the one after. The highest priority directory (overwrites all others)
// should be element 0 in the dirs vector.
class LayeredVfsDirectory : public VfsDirectory {
LayeredVfsDirectory(std::vector<VirtualDir> dirs, std::string name);
public:
~LayeredVfsDirectory() override;
/// Wrapper function to allow for more efficient handling of dirs.size() == 0, 1 cases.
static VirtualDir MakeLayeredDirectory(std::vector<VirtualDir> dirs, std::string name = "");
std::shared_ptr<VfsFile> GetFileRelative(std::string_view path) const override;
std::shared_ptr<VfsDirectory> GetDirectoryRelative(std::string_view path) const override;
std::shared_ptr<VfsFile> GetFile(std::string_view name) const override;
std::shared_ptr<VfsDirectory> GetSubdirectory(std::string_view name) const override;
std::string GetFullPath() const override;
std::vector<std::shared_ptr<VfsFile>> GetFiles() const override;
std::vector<std::shared_ptr<VfsDirectory>> GetSubdirectories() const override;
bool IsWritable() const override;
bool IsReadable() const override;
std::string GetName() const override;
std::shared_ptr<VfsDirectory> GetParentDirectory() const override;
std::shared_ptr<VfsDirectory> CreateSubdirectory(std::string_view name) override;
std::shared_ptr<VfsFile> CreateFile(std::string_view name) override;
bool DeleteSubdirectory(std::string_view name) override;
bool DeleteFile(std::string_view name) override;
bool Rename(std::string_view name) override;
protected:
bool ReplaceFileWithSubdirectory(VirtualFile file, VirtualDir dir) override;
private:
std::vector<VirtualDir> dirs;
std::string name;
};
} // namespace FileSys

View File

@@ -14,6 +14,8 @@ OffsetVfsFile::OffsetVfsFile(std::shared_ptr<VfsFile> file_, std::size_t size_,
: file(file_), offset(offset_), size(size_), name(std::move(name_)),
parent(parent_ == nullptr ? file->GetContainingDirectory() : std::move(parent_)) {}
OffsetVfsFile::~OffsetVfsFile() = default;
std::string OffsetVfsFile::GetName() const {
return name.empty() ? file->GetName() : name;
}

View File

@@ -19,6 +19,7 @@ class OffsetVfsFile : public VfsFile {
public:
OffsetVfsFile(std::shared_ptr<VfsFile> file, std::size_t size, std::size_t offset = 0,
std::string new_name = "", VirtualDir new_parent = nullptr);
~OffsetVfsFile() override;
std::string GetName() const override;
std::size_t GetSize() const override;

View File

@@ -413,6 +413,23 @@ std::string RealVfsDirectory::GetFullPath() const {
return out;
}
std::map<std::string, VfsEntryType, std::less<>> RealVfsDirectory::GetEntries() const {
if (perms == Mode::Append)
return {};
std::map<std::string, VfsEntryType, std::less<>> out;
FileUtil::ForeachDirectoryEntry(
nullptr, path,
[&out](u64* entries_out, const std::string& directory, const std::string& filename) {
const std::string full_path = directory + DIR_SEP + filename;
out.emplace(filename, FileUtil::IsDirectory(full_path) ? VfsEntryType::Directory
: VfsEntryType::File);
return true;
});
return out;
}
bool RealVfsDirectory::ReplaceFileWithSubdirectory(VirtualFile file, VirtualDir dir) {
return false;
}

View File

@@ -98,6 +98,7 @@ public:
bool DeleteFile(std::string_view name) override;
bool Rename(std::string_view name) override;
std::string GetFullPath() const override;
std::map<std::string, VfsEntryType, std::less<>> GetEntries() const override;
protected:
bool ReplaceFileWithSubdirectory(VirtualFile file, VirtualDir dir) override;

View File

@@ -0,0 +1,79 @@
// Copyright 2018 yuzu emulator team
// Licensed under GPLv2 or any later version
// Refer to the license.txt file included.
#pragma once
#include <algorithm>
#include <memory>
#include <string_view>
#include "core/file_sys/vfs.h"
namespace FileSys {
class StaticVfsFile : public VfsFile {
public:
explicit StaticVfsFile(u8 value, std::size_t size = 0, std::string name = "",
VirtualDir parent = nullptr)
: value{value}, size{size}, name{std::move(name)}, parent{std::move(parent)} {}
std::string GetName() const override {
return name;
}
std::size_t GetSize() const override {
return size;
}
bool Resize(std::size_t new_size) override {
size = new_size;
return true;
}
std::shared_ptr<VfsDirectory> GetContainingDirectory() const override {
return parent;
}
bool IsWritable() const override {
return false;
}
bool IsReadable() const override {
return true;
}
std::size_t Read(u8* data, std::size_t length, std::size_t offset) const override {
const auto read = std::min(length, size - offset);
std::fill(data, data + read, value);
return read;
}
std::size_t Write(const u8* data, std::size_t length, std::size_t offset) override {
return 0;
}
boost::optional<u8> ReadByte(std::size_t offset) const override {
if (offset < size)
return value;
return boost::none;
}
std::vector<u8> ReadBytes(std::size_t length, std::size_t offset) const override {
const auto read = std::min(length, size - offset);
return std::vector<u8>(read, value);
}
bool Rename(std::string_view new_name) override {
name = new_name;
return true;
}
private:
u8 value;
std::size_t size;
std::string name;
VirtualDir parent;
};
} // namespace FileSys

View File

@@ -3,16 +3,72 @@
// Refer to the license.txt file included.
#include <algorithm>
#include <cstring>
#include <utility>
#include "core/file_sys/vfs_vector.h"
namespace FileSys {
VectorVfsFile::VectorVfsFile(std::vector<u8> initial_data, std::string name, VirtualDir parent)
: data(std::move(initial_data)), parent(std::move(parent)), name(std::move(name)) {}
VectorVfsFile::~VectorVfsFile() = default;
std::string VectorVfsFile::GetName() const {
return name;
}
size_t VectorVfsFile::GetSize() const {
return data.size();
}
bool VectorVfsFile::Resize(size_t new_size) {
data.resize(new_size);
return true;
}
std::shared_ptr<VfsDirectory> VectorVfsFile::GetContainingDirectory() const {
return parent;
}
bool VectorVfsFile::IsWritable() const {
return true;
}
bool VectorVfsFile::IsReadable() const {
return true;
}
std::size_t VectorVfsFile::Read(u8* data_, std::size_t length, std::size_t offset) const {
const auto read = std::min(length, data.size() - offset);
std::memcpy(data_, data.data() + offset, read);
return read;
}
std::size_t VectorVfsFile::Write(const u8* data_, std::size_t length, std::size_t offset) {
if (offset + length > data.size())
data.resize(offset + length);
const auto write = std::min(length, data.size() - offset);
std::memcpy(data.data(), data_, write);
return write;
}
bool VectorVfsFile::Rename(std::string_view name_) {
name = name_;
return true;
}
void VectorVfsFile::Assign(std::vector<u8> new_data) {
data = std::move(new_data);
}
VectorVfsDirectory::VectorVfsDirectory(std::vector<VirtualFile> files_,
std::vector<VirtualDir> dirs_, std::string name_,
VirtualDir parent_)
: files(std::move(files_)), dirs(std::move(dirs_)), parent(std::move(parent_)),
name(std::move(name_)) {}
VectorVfsDirectory::~VectorVfsDirectory() = default;
std::vector<std::shared_ptr<VfsFile>> VectorVfsDirectory::GetFiles() const {
return files;
}

View File

@@ -8,6 +8,31 @@
namespace FileSys {
// An implementation of VfsFile that is backed by a vector optionally supplied upon construction
class VectorVfsFile : public VfsFile {
public:
explicit VectorVfsFile(std::vector<u8> initial_data = {}, std::string name = "",
VirtualDir parent = nullptr);
~VectorVfsFile() override;
std::string GetName() const override;
std::size_t GetSize() const override;
bool Resize(std::size_t new_size) override;
std::shared_ptr<VfsDirectory> GetContainingDirectory() const override;
bool IsWritable() const override;
bool IsReadable() const override;
std::size_t Read(u8* data, std::size_t length, std::size_t offset) const override;
std::size_t Write(const u8* data, std::size_t length, std::size_t offset) override;
bool Rename(std::string_view name) override;
virtual void Assign(std::vector<u8> new_data);
private:
std::vector<u8> data;
VirtualDir parent;
std::string name;
};
// An implementation of VfsDirectory that maintains two vectors for subdirectories and files.
// Vector data is supplied upon construction.
class VectorVfsDirectory : public VfsDirectory {
@@ -15,6 +40,7 @@ public:
explicit VectorVfsDirectory(std::vector<VirtualFile> files = {},
std::vector<VirtualDir> dirs = {}, std::string name = "",
VirtualDir parent = nullptr);
~VectorVfsDirectory() override;
std::vector<std::shared_ptr<VfsFile>> GetFiles() const override;
std::vector<std::shared_ptr<VfsDirectory>> GetSubdirectories() const override;

View File

@@ -30,9 +30,6 @@ static bool CalculateHMAC256(Destination* out, const SourceKey* key, std::size_t
mbedtls_md_context_t context;
mbedtls_md_init(&context);
const auto key_f = reinterpret_cast<const u8*>(key);
const std::vector<u8> key_v(key_f, key_f + key_length);
if (mbedtls_md_setup(&context, mbedtls_md_info_from_type(MBEDTLS_MD_SHA256), 1) ||
mbedtls_md_hmac_starts(&context, reinterpret_cast<const u8*>(key), key_length) ||
mbedtls_md_hmac_update(&context, reinterpret_cast<const u8*>(data), data_length) ||
@@ -45,7 +42,7 @@ static bool CalculateHMAC256(Destination* out, const SourceKey* key, std::size_t
return true;
}
NAX::NAX(VirtualFile file_) : file(std::move(file_)), header(std::make_unique<NAXHeader>()) {
NAX::NAX(VirtualFile file_) : header(std::make_unique<NAXHeader>()), file(std::move(file_)) {
std::string path = FileUtil::SanitizePath(file->GetFullPath());
static const std::regex nax_path_regex("/registered/(000000[0-9A-F]{2})/([0-9A-F]{32})\\.nca",
std::regex_constants::ECMAScript |
@@ -65,13 +62,15 @@ NAX::NAX(VirtualFile file_) : file(std::move(file_)), header(std::make_unique<NA
}
NAX::NAX(VirtualFile file_, std::array<u8, 0x10> nca_id)
: file(std::move(file_)), header(std::make_unique<NAXHeader>()) {
: header(std::make_unique<NAXHeader>()), file(std::move(file_)) {
Core::Crypto::SHA256Hash hash{};
mbedtls_sha256(nca_id.data(), nca_id.size(), hash.data(), 0);
status = Parse(fmt::format("/registered/000000{:02X}/{}.nca", hash[0],
Common::HexArrayToString(nca_id, false)));
}
NAX::~NAX() = default;
Loader::ResultStatus NAX::Parse(std::string_view path) {
if (file->ReadObject(header.get()) != sizeof(NAXHeader))
return Loader::ResultStatus::ErrorBadNAXHeader;
@@ -138,9 +137,9 @@ VirtualFile NAX::GetDecrypted() const {
return dec_file;
}
std::shared_ptr<NCA> NAX::AsNCA() const {
std::unique_ptr<NCA> NAX::AsNCA() const {
if (type == NAXContentType::NCA)
return std::make_shared<NCA>(GetDecrypted());
return std::make_unique<NCA>(GetDecrypted());
return nullptr;
}

View File

@@ -33,12 +33,13 @@ class NAX : public ReadOnlyVfsDirectory {
public:
explicit NAX(VirtualFile file);
explicit NAX(VirtualFile file, std::array<u8, 0x10> nca_id);
~NAX() override;
Loader::ResultStatus GetStatus() const;
VirtualFile GetDecrypted() const;
std::shared_ptr<NCA> AsNCA() const;
std::unique_ptr<NCA> AsNCA() const;
NAXContentType GetContentType() const;
@@ -60,7 +61,7 @@ private:
VirtualFile file;
Loader::ResultStatus status;
NAXContentType type;
NAXContentType type{};
VirtualFile dec_file;

View File

@@ -37,7 +37,9 @@
#include "core/core.h"
#include "core/core_cpu.h"
#include "core/gdbstub/gdbstub.h"
#include "core/hle/kernel/process.h"
#include "core/hle/kernel/scheduler.h"
#include "core/hle/kernel/vm_manager.h"
#include "core/loader/loader.h"
#include "core/memory.h"
@@ -248,7 +250,7 @@ static void RegWrite(std::size_t id, u64 val, Kernel::Thread* thread = nullptr)
} else if (id == PC_REGISTER) {
thread->context.pc = val;
} else if (id == PSTATE_REGISTER) {
thread->context.pstate = val;
thread->context.pstate = static_cast<u32>(val);
} else if (id > PSTATE_REGISTER && id < FPCR_REGISTER) {
thread->context.vector_registers[id - (PSTATE_REGISTER + 1)][0] = val;
}
@@ -585,7 +587,8 @@ static void HandleQuery() {
strlen("Xfer:features:read:target.xml:")) == 0) {
SendReply(target_xml);
} else if (strncmp(query, "Offsets", strlen("Offsets")) == 0) {
std::string buffer = fmt::format("TextSeg={:0x}", Memory::PROCESS_IMAGE_VADDR);
const VAddr base_address = Core::CurrentProcess()->VMManager().GetCodeRegionBaseAddress();
std::string buffer = fmt::format("TextSeg={:0x}", base_address);
SendReply(buffer.c_str());
} else if (strncmp(query, "fThreadInfo", strlen("fThreadInfo")) == 0) {
std::string val = "m";
@@ -893,11 +896,11 @@ static void ReadMemory() {
static u8 reply[GDB_BUFFER_SIZE - 4];
auto start_offset = command_buffer + 1;
auto addr_pos = std::find(start_offset, command_buffer + command_length, ',');
VAddr addr = HexToLong(start_offset, static_cast<u64>(addr_pos - start_offset));
const auto addr_pos = std::find(start_offset, command_buffer + command_length, ',');
const VAddr addr = HexToLong(start_offset, static_cast<u64>(addr_pos - start_offset));
start_offset = addr_pos + 1;
u64 len =
const u64 len =
HexToLong(start_offset, static_cast<u64>((command_buffer + command_length) - start_offset));
LOG_DEBUG(Debug_GDBStub, "gdb: addr: {:016X} len: {:016X}", addr, len);
@@ -906,7 +909,9 @@ static void ReadMemory() {
SendReply("E01");
}
if (addr < Memory::PROCESS_IMAGE_VADDR || addr >= Memory::MAP_REGION_VADDR_END) {
const auto& vm_manager = Core::CurrentProcess()->VMManager();
if (addr < vm_manager.GetCodeRegionBaseAddress() ||
addr >= vm_manager.GetMapRegionEndAddress()) {
return SendReply("E00");
}
@@ -995,7 +1000,7 @@ static bool CommitBreakpoint(BreakpointType type, VAddr addr, u64 len) {
breakpoint.addr = addr;
breakpoint.len = len;
Memory::ReadBlock(addr, breakpoint.inst.data(), breakpoint.inst.size());
static constexpr std::array<u8, 4> btrap{{0xd4, 0x20, 0x7d, 0x0}};
static constexpr std::array<u8, 4> btrap{{0x00, 0x7d, 0x20, 0xd4}};
Memory::WriteBlock(addr, btrap.data(), btrap.size());
Core::System::GetInstance().InvalidateCpuInstructionCaches();
p.insert({addr, breakpoint});

View File

@@ -31,6 +31,7 @@ enum {
TooLarge = 119,
InvalidEnumValue = 120,
NoSuchEntry = 121,
AlreadyRegistered = 122,
InvalidState = 125,
ResourceLimitExceeded = 132,
};
@@ -58,6 +59,7 @@ constexpr ResultCode ERR_INVALID_MEMORY_PERMISSIONS(ErrorModule::Kernel,
constexpr ResultCode ERR_INVALID_HANDLE(ErrorModule::Kernel, ErrCodes::InvalidHandle);
constexpr ResultCode ERR_INVALID_PROCESSOR_ID(ErrorModule::Kernel, ErrCodes::InvalidProcessorId);
constexpr ResultCode ERR_INVALID_SIZE(ErrorModule::Kernel, ErrCodes::InvalidSize);
constexpr ResultCode ERR_ALREADY_REGISTERED(ErrorModule::Kernel, ErrCodes::AlreadyRegistered);
constexpr ResultCode ERR_INVALID_STATE(ErrorModule::Kernel, ErrCodes::InvalidState);
constexpr ResultCode ERR_INVALID_THREAD_PRIORITY(ErrorModule::Kernel,
ErrCodes::InvalidThreadPriority);

View File

@@ -16,6 +16,7 @@
#include "core/hle/kernel/object.h"
#include "core/hle/kernel/thread.h"
#include "core/hle/result.h"
#include "core/memory.h"
namespace Kernel {

View File

@@ -6,7 +6,6 @@
#include <atomic>
#include <string>
#include <utility>
#include <boost/smart_ptr/intrusive_ptr.hpp>
@@ -97,7 +96,7 @@ using SharedPtr = boost::intrusive_ptr<T>;
template <typename T>
inline SharedPtr<T> DynamicObjectCast(SharedPtr<Object> object) {
if (object != nullptr && object->GetHandleType() == T::HANDLE_TYPE) {
return boost::static_pointer_cast<T>(std::move(object));
return boost::static_pointer_cast<T>(object);
}
return nullptr;
}

View File

@@ -7,10 +7,13 @@
#include "common/assert.h"
#include "common/common_funcs.h"
#include "common/logging/log.h"
#include "core/core.h"
#include "core/file_sys/program_metadata.h"
#include "core/hle/kernel/errors.h"
#include "core/hle/kernel/kernel.h"
#include "core/hle/kernel/process.h"
#include "core/hle/kernel/resource_limit.h"
#include "core/hle/kernel/scheduler.h"
#include "core/hle/kernel/thread.h"
#include "core/hle/kernel/vm_manager.h"
#include "core/memory.h"
@@ -32,14 +35,22 @@ SharedPtr<Process> Process::Create(KernelCore& kernel, std::string&& name) {
process->name = std::move(name);
process->flags.raw = 0;
process->flags.memory_region.Assign(MemoryRegion::APPLICATION);
process->resource_limit = kernel.ResourceLimitForCategory(ResourceLimitCategory::APPLICATION);
process->status = ProcessStatus::Created;
process->program_id = 0;
process->process_id = kernel.CreateNewProcessID();
process->svc_access_mask.set();
kernel.AppendNewProcess(process);
return process;
}
void Process::LoadFromMetadata(const FileSys::ProgramMetadata& metadata) {
program_id = metadata.GetTitleID();
is_64bit_process = metadata.Is64BitProgram();
vm_manager.Reset(metadata.GetAddressSpaceType());
}
void Process::ParseKernelCaps(const u32* kernel_caps, std::size_t len) {
for (std::size_t i = 0; i < len; ++i) {
u32 descriptor = kernel_caps[i];
@@ -117,7 +128,7 @@ void Process::Run(VAddr entry_point, s32 main_thread_priority, u32 stack_size) {
// TODO(bunnei): This is heap area that should be allocated by the kernel and not mapped as part
// of the user address space.
vm_manager
.MapMemoryBlock(Memory::STACK_AREA_VADDR_END - stack_size,
.MapMemoryBlock(vm_manager.GetTLSIORegionEndAddress() - stack_size,
std::make_shared<std::vector<u8>>(stack_size, 0), 0, stack_size,
MemoryState::Mapped)
.Unwrap();
@@ -125,7 +136,92 @@ void Process::Run(VAddr entry_point, s32 main_thread_priority, u32 stack_size) {
vm_manager.LogLayout();
status = ProcessStatus::Running;
Kernel::SetupMainThread(kernel, entry_point, main_thread_priority, this);
Kernel::SetupMainThread(kernel, entry_point, main_thread_priority, *this);
}
void Process::PrepareForTermination() {
status = ProcessStatus::Exited;
const auto stop_threads = [this](const std::vector<SharedPtr<Thread>>& thread_list) {
for (auto& thread : thread_list) {
if (thread->owner_process != this)
continue;
if (thread == GetCurrentThread())
continue;
// TODO(Subv): When are the other running/ready threads terminated?
ASSERT_MSG(thread->status == ThreadStatus::WaitSynchAny ||
thread->status == ThreadStatus::WaitSynchAll,
"Exiting processes with non-waiting threads is currently unimplemented");
thread->Stop();
}
};
auto& system = Core::System::GetInstance();
stop_threads(system.Scheduler(0)->GetThreadList());
stop_threads(system.Scheduler(1)->GetThreadList());
stop_threads(system.Scheduler(2)->GetThreadList());
stop_threads(system.Scheduler(3)->GetThreadList());
}
/**
* Finds a free location for the TLS section of a thread.
* @param tls_slots The TLS page array of the thread's owner process.
* Returns a tuple of (page, slot, alloc_needed) where:
* page: The index of the first allocated TLS page that has free slots.
* slot: The index of the first free slot in the indicated page.
* alloc_needed: Whether there's a need to allocate a new TLS page (All pages are full).
*/
static std::tuple<std::size_t, std::size_t, bool> FindFreeThreadLocalSlot(
const std::vector<std::bitset<8>>& tls_slots) {
// Iterate over all the allocated pages, and try to find one where not all slots are used.
for (std::size_t page = 0; page < tls_slots.size(); ++page) {
const auto& page_tls_slots = tls_slots[page];
if (!page_tls_slots.all()) {
// We found a page with at least one free slot, find which slot it is
for (std::size_t slot = 0; slot < page_tls_slots.size(); ++slot) {
if (!page_tls_slots.test(slot)) {
return std::make_tuple(page, slot, false);
}
}
}
}
return std::make_tuple(0, 0, true);
}
VAddr Process::MarkNextAvailableTLSSlotAsUsed(Thread& thread) {
auto [available_page, available_slot, needs_allocation] = FindFreeThreadLocalSlot(tls_slots);
const VAddr tls_begin = vm_manager.GetTLSIORegionBaseAddress();
if (needs_allocation) {
tls_slots.emplace_back(0); // The page is completely available at the start
available_page = tls_slots.size() - 1;
available_slot = 0; // Use the first slot in the new page
// Allocate some memory from the end of the linear heap for this region.
auto& tls_memory = thread.GetTLSMemory();
tls_memory->insert(tls_memory->end(), Memory::PAGE_SIZE, 0);
vm_manager.RefreshMemoryBlockMappings(tls_memory.get());
vm_manager.MapMemoryBlock(tls_begin + available_page * Memory::PAGE_SIZE, tls_memory, 0,
Memory::PAGE_SIZE, MemoryState::ThreadLocal);
}
tls_slots[available_page].set(available_slot);
return tls_begin + available_page * Memory::PAGE_SIZE + available_slot * Memory::TLS_ENTRY_SIZE;
}
void Process::FreeTLSSlot(VAddr tls_address) {
const VAddr tls_base = tls_address - vm_manager.GetTLSIORegionBaseAddress();
const VAddr tls_page = tls_base / Memory::PAGE_SIZE;
const VAddr tls_slot = (tls_base % Memory::PAGE_SIZE) / Memory::TLS_ENTRY_SIZE;
tls_slots[tls_page].reset(tls_slot);
}
void Process::LoadModule(SharedPtr<CodeSet> module_, VAddr base_addr) {
@@ -145,8 +241,8 @@ void Process::LoadModule(SharedPtr<CodeSet> module_, VAddr base_addr) {
}
ResultVal<VAddr> Process::HeapAllocate(VAddr target, u64 size, VMAPermission perms) {
if (target < Memory::HEAP_VADDR || target + size > Memory::HEAP_VADDR_END ||
target + size < target) {
if (target < vm_manager.GetHeapRegionBaseAddress() ||
target + size > vm_manager.GetHeapRegionEndAddress() || target + size < target) {
return ERR_INVALID_ADDRESS;
}
@@ -181,8 +277,8 @@ ResultVal<VAddr> Process::HeapAllocate(VAddr target, u64 size, VMAPermission per
}
ResultCode Process::HeapFree(VAddr target, u32 size) {
if (target < Memory::HEAP_VADDR || target + size > Memory::HEAP_VADDR_END ||
target + size < target) {
if (target < vm_manager.GetHeapRegionBaseAddress() ||
target + size > vm_manager.GetHeapRegionEndAddress() || target + size < target) {
return ERR_INVALID_ADDRESS;
}

View File

@@ -17,6 +17,10 @@
#include "core/hle/kernel/thread.h"
#include "core/hle/kernel/vm_manager.h"
namespace FileSys {
class ProgramMetadata;
}
namespace Kernel {
class KernelCore;
@@ -131,6 +135,121 @@ public:
return HANDLE_TYPE;
}
/// Gets a reference to the process' memory manager.
Kernel::VMManager& VMManager() {
return vm_manager;
}
/// Gets a const reference to the process' memory manager.
const Kernel::VMManager& VMManager() const {
return vm_manager;
}
/// Gets the current status of the process
ProcessStatus GetStatus() const {
return status;
}
/// Gets the unique ID that identifies this particular process.
u32 GetProcessID() const {
return process_id;
}
/// Gets the title ID corresponding to this process.
u64 GetTitleID() const {
return program_id;
}
/// Gets the resource limit descriptor for this process
ResourceLimit& GetResourceLimit() {
return *resource_limit;
}
/// Gets the resource limit descriptor for this process
const ResourceLimit& GetResourceLimit() const {
return *resource_limit;
}
/// Gets the default CPU ID for this process
u8 GetDefaultProcessorID() const {
return ideal_processor;
}
/// Gets the bitmask of allowed CPUs that this process' threads can run on.
u32 GetAllowedProcessorMask() const {
return allowed_processor_mask;
}
/// Gets the bitmask of allowed thread priorities.
u32 GetAllowedThreadPriorityMask() const {
return allowed_thread_priority_mask;
}
u32 IsVirtualMemoryEnabled() const {
return is_virtual_address_memory_enabled;
}
/// Whether this process is an AArch64 or AArch32 process.
bool Is64BitProcess() const {
return is_64bit_process;
}
/**
* Loads process-specifics configuration info with metadata provided
* by an executable.
*
* @param metadata The provided metadata to load process specific info.
*/
void LoadFromMetadata(const FileSys::ProgramMetadata& metadata);
/**
* Parses a list of kernel capability descriptors (as found in the ExHeader) and applies them
* to this process.
*/
void ParseKernelCaps(const u32* kernel_caps, std::size_t len);
/**
* Applies address space changes and launches the process main thread.
*/
void Run(VAddr entry_point, s32 main_thread_priority, u32 stack_size);
/**
* Prepares a process for termination by stopping all of its threads
* and clearing any other resources.
*/
void PrepareForTermination();
void LoadModule(SharedPtr<CodeSet> module_, VAddr base_addr);
///////////////////////////////////////////////////////////////////////////////////////////////
// Memory Management
// Marks the next available region as used and returns the address of the slot.
VAddr MarkNextAvailableTLSSlotAsUsed(Thread& thread);
// Frees a used TLS slot identified by the given address
void FreeTLSSlot(VAddr tls_address);
ResultVal<VAddr> HeapAllocate(VAddr target, u64 size, VMAPermission perms);
ResultCode HeapFree(VAddr target, u32 size);
ResultCode MirrorMemory(VAddr dst_addr, VAddr src_addr, u64 size);
ResultCode UnmapMemory(VAddr dst_addr, VAddr src_addr, u64 size);
private:
explicit Process(KernelCore& kernel);
~Process() override;
/// Memory manager for this process.
Kernel::VMManager vm_manager;
/// Current status of the process
ProcessStatus status;
/// The ID of this process
u32 process_id = 0;
/// Title ID corresponding to the process
u64 program_id;
@@ -140,7 +259,7 @@ public:
/// The process may only call SVCs which have the corresponding bit set.
std::bitset<0x80> svc_access_mask;
/// Maximum size of the handle table for the process.
unsigned int handle_table_size = 0x200;
u32 handle_table_size = 0x200;
/// Special memory ranges mapped into this processes address space. This is used to give
/// processes access to specific I/O regions and device memory.
boost::container::static_vector<AddressMapping, 8> address_mappings;
@@ -154,29 +273,6 @@ public:
u32 allowed_processor_mask = THREADPROCESSORID_DEFAULT_MASK;
u32 allowed_thread_priority_mask = 0xFFFFFFFF;
u32 is_virtual_address_memory_enabled = 0;
/// Current status of the process
ProcessStatus status;
/// The ID of this process
u32 process_id = 0;
/**
* Parses a list of kernel capability descriptors (as found in the ExHeader) and applies them
* to this process.
*/
void ParseKernelCaps(const u32* kernel_caps, std::size_t len);
/**
* Applies address space changes and launches the process main thread.
*/
void Run(VAddr entry_point, s32 main_thread_priority, u32 stack_size);
void LoadModule(SharedPtr<CodeSet> module_, VAddr base_addr);
///////////////////////////////////////////////////////////////////////////////////////////////
// Memory Management
VMManager vm_manager;
// Memory used to back the allocations in the regular heap. A single vector is used to cover
// the entire virtual address space extents that bound the allocations, including any holes.
@@ -196,18 +292,12 @@ public:
/// This vector will grow as more pages are allocated for new threads.
std::vector<std::bitset<8>> tls_slots;
/// Whether or not this process is AArch64, or AArch32.
/// By default, we currently assume this is true, unless otherwise
/// specified by metadata provided to the process during loading.
bool is_64bit_process = true;
std::string name;
ResultVal<VAddr> HeapAllocate(VAddr target, u64 size, VMAPermission perms);
ResultCode HeapFree(VAddr target, u32 size);
ResultCode MirrorMemory(VAddr dst_addr, VAddr src_addr, u64 size);
ResultCode UnmapMemory(VAddr dst_addr, VAddr src_addr, u64 size);
private:
explicit Process(KernelCore& kernel);
~Process() override;
};
} // namespace Kernel

View File

@@ -17,7 +17,7 @@ namespace Kernel {
std::mutex Scheduler::scheduler_mutex;
Scheduler::Scheduler(Core::ARM_Interface* cpu_core) : cpu_core(cpu_core) {}
Scheduler::Scheduler(Core::ARM_Interface& cpu_core) : cpu_core(cpu_core) {}
Scheduler::~Scheduler() {
for (auto& thread : thread_list) {
@@ -59,9 +59,9 @@ void Scheduler::SwitchContext(Thread* new_thread) {
// Save context for previous thread
if (previous_thread) {
previous_thread->last_running_ticks = CoreTiming::GetTicks();
cpu_core->SaveContext(previous_thread->context);
cpu_core.SaveContext(previous_thread->context);
// Save the TPIDR_EL0 system register in case it was modified.
previous_thread->tpidr_el0 = cpu_core->GetTPIDR_EL0();
previous_thread->tpidr_el0 = cpu_core.GetTPIDR_EL0();
if (previous_thread->status == ThreadStatus::Running) {
// This is only the case when a reschedule is triggered without the current thread
@@ -88,13 +88,13 @@ void Scheduler::SwitchContext(Thread* new_thread) {
if (previous_process != current_thread->owner_process) {
Core::CurrentProcess() = current_thread->owner_process;
SetCurrentPageTable(&Core::CurrentProcess()->vm_manager.page_table);
SetCurrentPageTable(&Core::CurrentProcess()->VMManager().page_table);
}
cpu_core->LoadContext(new_thread->context);
cpu_core->SetTlsAddress(new_thread->GetTLSAddress());
cpu_core->SetTPIDR_EL0(new_thread->GetTPIDR_EL0());
cpu_core->ClearExclusiveState();
cpu_core.LoadContext(new_thread->context);
cpu_core.SetTlsAddress(new_thread->GetTLSAddress());
cpu_core.SetTPIDR_EL0(new_thread->GetTPIDR_EL0());
cpu_core.ClearExclusiveState();
} else {
current_thread = nullptr;
// Note: We do not reset the current process and current page table when idling because

View File

@@ -19,7 +19,7 @@ namespace Kernel {
class Scheduler final {
public:
explicit Scheduler(Core::ARM_Interface* cpu_core);
explicit Scheduler(Core::ARM_Interface& cpu_core);
~Scheduler();
/// Returns whether there are any threads that are ready to run.
@@ -72,7 +72,7 @@ private:
SharedPtr<Thread> current_thread = nullptr;
Core::ARM_Interface* cpu_core;
Core::ARM_Interface& cpu_core;
static std::mutex scheduler_mutex;
};

View File

@@ -8,6 +8,7 @@
#include "common/logging/log.h"
#include "core/core.h"
#include "core/hle/kernel/errors.h"
#include "core/hle/kernel/kernel.h"
#include "core/hle/kernel/shared_memory.h"
#include "core/memory.h"
@@ -34,11 +35,11 @@ SharedPtr<SharedMemory> SharedMemory::Create(KernelCore& kernel, SharedPtr<Proce
// Refresh the address mappings for the current process.
if (Core::CurrentProcess() != nullptr) {
Core::CurrentProcess()->vm_manager.RefreshMemoryBlockMappings(
Core::CurrentProcess()->VMManager().RefreshMemoryBlockMappings(
shared_memory->backing_block.get());
}
} else {
auto& vm_manager = shared_memory->owner_process->vm_manager;
auto& vm_manager = shared_memory->owner_process->VMManager();
// The memory is already available and mapped in the owner process.
auto vma = vm_manager.FindVMA(address);
@@ -71,7 +72,8 @@ SharedPtr<SharedMemory> SharedMemory::CreateForApplet(
shared_memory->other_permissions = other_permissions;
shared_memory->backing_block = std::move(heap_block);
shared_memory->backing_block_offset = offset;
shared_memory->base_address = Memory::HEAP_VADDR + offset;
shared_memory->base_address =
kernel.CurrentProcess()->VMManager().GetHeapRegionBaseAddress() + offset;
return shared_memory;
}
@@ -105,7 +107,7 @@ ResultCode SharedMemory::Map(Process* target_process, VAddr address, MemoryPermi
VAddr target_address = address;
// Map the memory block into the target process
auto result = target_process->vm_manager.MapMemoryBlock(
auto result = target_process->VMManager().MapMemoryBlock(
target_address, backing_block, backing_block_offset, size, MemoryState::Shared);
if (result.Failed()) {
LOG_ERROR(
@@ -115,14 +117,14 @@ ResultCode SharedMemory::Map(Process* target_process, VAddr address, MemoryPermi
return result.Code();
}
return target_process->vm_manager.ReprotectRange(target_address, size,
ConvertPermissions(permissions));
return target_process->VMManager().ReprotectRange(target_address, size,
ConvertPermissions(permissions));
}
ResultCode SharedMemory::Unmap(Process* target_process, VAddr address) {
// TODO(Subv): Verify what happens if the application tries to unmap an address that is not
// mapped to a SharedMemory.
return target_process->vm_manager.UnmapRange(address, size);
return target_process->VMManager().UnmapRange(address, size);
}
VMAPermission SharedMemory::ConvertPermissions(MemoryPermission permission) {

View File

@@ -51,8 +51,9 @@ static ResultCode SetHeapSize(VAddr* heap_addr, u64 heap_size) {
}
auto& process = *Core::CurrentProcess();
const VAddr heap_base = process.VMManager().GetHeapRegionBaseAddress();
CASCADE_RESULT(*heap_addr,
process.HeapAllocate(Memory::HEAP_VADDR, heap_size, VMAPermission::ReadWrite));
process.HeapAllocate(heap_base, heap_size, VMAPermission::ReadWrite));
return RESULT_SUCCESS;
}
@@ -169,7 +170,7 @@ static ResultCode GetProcessId(u32* process_id, Handle process_handle) {
return ERR_INVALID_HANDLE;
}
*process_id = process->process_id;
*process_id = process->GetProcessID();
return RESULT_SUCCESS;
}
@@ -325,26 +326,27 @@ static ResultCode GetInfo(u64* result, u64 info_id, u64 handle, u64 info_sub_id)
LOG_TRACE(Kernel_SVC, "called info_id=0x{:X}, info_sub_id=0x{:X}, handle=0x{:08X}", info_id,
info_sub_id, handle);
const auto& vm_manager = Core::CurrentProcess()->vm_manager;
const auto& current_process = Core::CurrentProcess();
const auto& vm_manager = current_process->VMManager();
switch (static_cast<GetInfoType>(info_id)) {
case GetInfoType::AllowedCpuIdBitmask:
*result = Core::CurrentProcess()->allowed_processor_mask;
*result = current_process->GetAllowedProcessorMask();
break;
case GetInfoType::AllowedThreadPrioBitmask:
*result = Core::CurrentProcess()->allowed_thread_priority_mask;
*result = current_process->GetAllowedThreadPriorityMask();
break;
case GetInfoType::MapRegionBaseAddr:
*result = Memory::MAP_REGION_VADDR;
*result = vm_manager.GetMapRegionBaseAddress();
break;
case GetInfoType::MapRegionSize:
*result = Memory::MAP_REGION_SIZE;
*result = vm_manager.GetMapRegionSize();
break;
case GetInfoType::HeapRegionBaseAddr:
*result = Memory::HEAP_VADDR;
*result = vm_manager.GetHeapRegionBaseAddress();
break;
case GetInfoType::HeapRegionSize:
*result = Memory::HEAP_SIZE;
*result = vm_manager.GetHeapRegionSize();
break;
case GetInfoType::TotalMemoryUsage:
*result = vm_manager.GetTotalMemoryUsage();
@@ -359,22 +361,35 @@ static ResultCode GetInfo(u64* result, u64 info_id, u64 handle, u64 info_sub_id)
*result = 0;
break;
case GetInfoType::AddressSpaceBaseAddr:
*result = vm_manager.GetAddressSpaceBaseAddr();
*result = vm_manager.GetCodeRegionBaseAddress();
break;
case GetInfoType::AddressSpaceSize:
*result = vm_manager.GetAddressSpaceSize();
case GetInfoType::AddressSpaceSize: {
const u64 width = vm_manager.GetAddressSpaceWidth();
switch (width) {
case 32:
*result = 0xFFE00000;
break;
case 36:
*result = 0xFF8000000;
break;
case 39:
*result = 0x7FF8000000;
break;
}
break;
}
case GetInfoType::NewMapRegionBaseAddr:
*result = Memory::NEW_MAP_REGION_VADDR;
*result = vm_manager.GetNewMapRegionBaseAddress();
break;
case GetInfoType::NewMapRegionSize:
*result = Memory::NEW_MAP_REGION_SIZE;
*result = vm_manager.GetNewMapRegionSize();
break;
case GetInfoType::IsVirtualAddressMemoryEnabled:
*result = Core::CurrentProcess()->is_virtual_address_memory_enabled;
*result = current_process->IsVirtualMemoryEnabled();
break;
case GetInfoType::TitleId:
*result = Core::CurrentProcess()->program_id;
*result = current_process->GetTitleID();
break;
case GetInfoType::PrivilegedProcessId:
LOG_WARNING(Kernel_SVC,
@@ -400,8 +415,36 @@ static ResultCode SetThreadActivity(Handle handle, u32 unknown) {
}
/// Gets the thread context
static ResultCode GetThreadContext(Handle handle, VAddr addr) {
LOG_WARNING(Kernel_SVC, "(STUBBED) called, handle=0x{:08X}, addr=0x{:X}", handle, addr);
static ResultCode GetThreadContext(VAddr thread_context, Handle handle) {
LOG_DEBUG(Kernel_SVC, "called, context=0x{:08X}, thread=0x{:X}", thread_context, handle);
auto& kernel = Core::System::GetInstance().Kernel();
const SharedPtr<Thread> thread = kernel.HandleTable().Get<Thread>(handle);
if (!thread) {
return ERR_INVALID_HANDLE;
}
const auto current_process = Core::CurrentProcess();
if (thread->owner_process != current_process) {
return ERR_INVALID_HANDLE;
}
if (thread == GetCurrentThread()) {
return ERR_ALREADY_REGISTERED;
}
Core::ARM_Interface::ThreadContext ctx = thread->context;
// Mask away mode bits, interrupt bits, IL bit, and other reserved bits.
ctx.pstate &= 0xFF0FFE20;
// If 64-bit, we can just write the context registers directly and we're good.
// However, if 32-bit, we have to ensure some registers are zeroed out.
if (!current_process->Is64BitProcess()) {
std::fill(ctx.cpu_registers.begin() + 15, ctx.cpu_registers.end(), 0);
std::fill(ctx.vector_registers.begin() + 16, ctx.vector_registers.end(), u128{});
}
Memory::WriteBlock(thread_context, &ctx, sizeof(ctx));
return RESULT_SUCCESS;
}
@@ -429,8 +472,8 @@ static ResultCode SetThreadPriority(Handle handle, u32 priority) {
// Note: The kernel uses the current process's resource limit instead of
// the one from the thread owner's resource limit.
SharedPtr<ResourceLimit>& resource_limit = Core::CurrentProcess()->resource_limit;
if (resource_limit->GetMaxResourceValue(ResourceType::Priority) > priority) {
const ResourceLimit& resource_limit = Core::CurrentProcess()->GetResourceLimit();
if (resource_limit.GetMaxResourceValue(ResourceType::Priority) > priority) {
return ERR_NOT_AUTHORIZED;
}
@@ -504,9 +547,9 @@ static ResultCode QueryProcessMemory(MemoryInfo* memory_info, PageInfo* /*page_i
if (!process) {
return ERR_INVALID_HANDLE;
}
auto vma = process->vm_manager.FindVMA(addr);
auto vma = process->VMManager().FindVMA(addr);
memory_info->attributes = 0;
if (vma == Core::CurrentProcess()->vm_manager.vma_map.end()) {
if (vma == Core::CurrentProcess()->VMManager().vma_map.end()) {
memory_info->base_address = 0;
memory_info->permission = static_cast<u32>(VMAPermission::None);
memory_info->size = 0;
@@ -530,35 +573,13 @@ static ResultCode QueryMemory(MemoryInfo* memory_info, PageInfo* page_info, VAdd
/// Exits the current process
static void ExitProcess() {
LOG_INFO(Kernel_SVC, "Process {} exiting", Core::CurrentProcess()->process_id);
auto& current_process = Core::CurrentProcess();
ASSERT_MSG(Core::CurrentProcess()->status == ProcessStatus::Running,
LOG_INFO(Kernel_SVC, "Process {} exiting", current_process->GetProcessID());
ASSERT_MSG(current_process->GetStatus() == ProcessStatus::Running,
"Process has already exited");
Core::CurrentProcess()->status = ProcessStatus::Exited;
auto stop_threads = [](const std::vector<SharedPtr<Thread>>& thread_list) {
for (auto& thread : thread_list) {
if (thread->owner_process != Core::CurrentProcess())
continue;
if (thread == GetCurrentThread())
continue;
// TODO(Subv): When are the other running/ready threads terminated?
ASSERT_MSG(thread->status == ThreadStatus::WaitSynchAny ||
thread->status == ThreadStatus::WaitSynchAll,
"Exiting processes with non-waiting threads is currently unimplemented");
thread->Stop();
}
};
auto& system = Core::System::GetInstance();
stop_threads(system.Scheduler(0)->GetThreadList());
stop_threads(system.Scheduler(1)->GetThreadList());
stop_threads(system.Scheduler(2)->GetThreadList());
stop_threads(system.Scheduler(3)->GetThreadList());
current_process->PrepareForTermination();
// Kill the current thread
GetCurrentThread()->Stop();
@@ -575,14 +596,14 @@ static ResultCode CreateThread(Handle* out_handle, VAddr entry_point, u64 arg, V
return ERR_INVALID_THREAD_PRIORITY;
}
SharedPtr<ResourceLimit>& resource_limit = Core::CurrentProcess()->resource_limit;
if (resource_limit->GetMaxResourceValue(ResourceType::Priority) > priority) {
const ResourceLimit& resource_limit = Core::CurrentProcess()->GetResourceLimit();
if (resource_limit.GetMaxResourceValue(ResourceType::Priority) > priority) {
return ERR_NOT_AUTHORIZED;
}
if (processor_id == THREADPROCESSORID_DEFAULT) {
// Set the target CPU to the one specified in the process' exheader.
processor_id = Core::CurrentProcess()->ideal_processor;
processor_id = Core::CurrentProcess()->GetDefaultProcessorID();
ASSERT(processor_id != THREADPROCESSORID_DEFAULT);
}
@@ -909,10 +930,10 @@ static ResultCode SetThreadCoreMask(Handle thread_handle, u32 core, u64 mask) {
}
if (core == static_cast<u32>(THREADPROCESSORID_DEFAULT)) {
ASSERT(thread->owner_process->ideal_processor !=
ASSERT(thread->owner_process->GetDefaultProcessorID() !=
static_cast<u8>(THREADPROCESSORID_DEFAULT));
// Set the target CPU to the one specified in the process' exheader.
core = thread->owner_process->ideal_processor;
core = thread->owner_process->GetDefaultProcessorID();
mask = 1ull << core;
}
@@ -1039,7 +1060,7 @@ static const FunctionDef SVC_Table[] = {
{0x2B, nullptr, "FlushDataCache"},
{0x2C, nullptr, "MapPhysicalMemory"},
{0x2D, nullptr, "UnmapPhysicalMemory"},
{0x2E, nullptr, "GetNextThreadInfo"},
{0x2E, nullptr, "GetFutureThreadInfo"},
{0x2F, nullptr, "GetLastThreadInfo"},
{0x30, nullptr, "GetResourceLimitLimitValue"},
{0x31, nullptr, "GetResourceLimitCurrentValue"},
@@ -1065,11 +1086,11 @@ static const FunctionDef SVC_Table[] = {
{0x45, nullptr, "CreateEvent"},
{0x46, nullptr, "Unknown"},
{0x47, nullptr, "Unknown"},
{0x48, nullptr, "AllocateUnsafeMemory"},
{0x49, nullptr, "FreeUnsafeMemory"},
{0x4A, nullptr, "SetUnsafeAllocationLimit"},
{0x4B, nullptr, "CreateJitMemory"},
{0x4C, nullptr, "MapJitMemory"},
{0x48, nullptr, "MapPhysicalMemoryUnsafe"},
{0x49, nullptr, "UnmapPhysicalMemoryUnsafe"},
{0x4A, nullptr, "SetUnsafeLimit"},
{0x4B, nullptr, "CreateCodeMemory"},
{0x4C, nullptr, "ControlCodeMemory"},
{0x4D, nullptr, "SleepSystem"},
{0x4E, nullptr, "ReadWriteRegister"},
{0x4F, nullptr, "SetProcessActivity"},
@@ -1104,7 +1125,7 @@ static const FunctionDef SVC_Table[] = {
{0x6C, nullptr, "SetHardwareBreakPoint"},
{0x6D, nullptr, "GetDebugThreadParam"},
{0x6E, nullptr, "Unknown"},
{0x6F, nullptr, "GetMemoryInfo"},
{0x6F, nullptr, "GetSystemInfo"},
{0x70, nullptr, "CreatePort"},
{0x71, nullptr, "ManageNamedPort"},
{0x72, nullptr, "ConnectToPort"},

View File

@@ -64,6 +64,11 @@ void SvcWrap() {
FuncReturn(func(Param(0), (s32)Param(1)).raw);
}
template <ResultCode func(u64, u32)>
void SvcWrap() {
FuncReturn(func(Param(0), static_cast<u32>(Param(1))).raw);
}
template <ResultCode func(u64*, u64)>
void SvcWrap() {
u64 param_1 = 0;

View File

@@ -65,10 +65,7 @@ void Thread::Stop() {
wait_objects.clear();
// Mark the TLS slot in the thread's page as free.
const u64 tls_page = (tls_address - Memory::TLS_AREA_VADDR) / Memory::PAGE_SIZE;
const u64 tls_slot =
((tls_address - Memory::TLS_AREA_VADDR) % Memory::PAGE_SIZE) / Memory::TLS_ENTRY_SIZE;
Core::CurrentProcess()->tls_slots[tls_page].reset(tls_slot);
owner_process->FreeTLSSlot(tls_address);
}
void WaitCurrentThread_Sleep() {
@@ -177,32 +174,6 @@ void Thread::ResumeFromWait() {
Core::System::GetInstance().CpuCore(processor_id).PrepareReschedule();
}
/**
* Finds a free location for the TLS section of a thread.
* @param tls_slots The TLS page array of the thread's owner process.
* Returns a tuple of (page, slot, alloc_needed) where:
* page: The index of the first allocated TLS page that has free slots.
* slot: The index of the first free slot in the indicated page.
* alloc_needed: Whether there's a need to allocate a new TLS page (All pages are full).
*/
static std::tuple<std::size_t, std::size_t, bool> GetFreeThreadLocalSlot(
const std::vector<std::bitset<8>>& tls_slots) {
// Iterate over all the allocated pages, and try to find one where not all slots are used.
for (std::size_t page = 0; page < tls_slots.size(); ++page) {
const auto& page_tls_slots = tls_slots[page];
if (!page_tls_slots.all()) {
// We found a page with at least one free slot, find which slot it is
for (std::size_t slot = 0; slot < page_tls_slots.size(); ++slot) {
if (!page_tls_slots.test(slot)) {
return std::make_tuple(page, slot, false);
}
}
}
}
return std::make_tuple(0, 0, true);
}
/**
* Resets a thread context, making it ready to be scheduled and run by the CPU
* @param context Thread context to reset
@@ -264,32 +235,7 @@ ResultVal<SharedPtr<Thread>> Thread::Create(KernelCore& kernel, std::string name
thread->owner_process = owner_process;
thread->scheduler = Core::System::GetInstance().Scheduler(processor_id);
thread->scheduler->AddThread(thread, priority);
// Find the next available TLS index, and mark it as used
auto& tls_slots = owner_process->tls_slots;
auto [available_page, available_slot, needs_allocation] = GetFreeThreadLocalSlot(tls_slots);
if (needs_allocation) {
tls_slots.emplace_back(0); // The page is completely available at the start
available_page = tls_slots.size() - 1;
available_slot = 0; // Use the first slot in the new page
// Allocate some memory from the end of the linear heap for this region.
const std::size_t offset = thread->tls_memory->size();
thread->tls_memory->insert(thread->tls_memory->end(), Memory::PAGE_SIZE, 0);
auto& vm_manager = owner_process->vm_manager;
vm_manager.RefreshMemoryBlockMappings(thread->tls_memory.get());
vm_manager.MapMemoryBlock(Memory::TLS_AREA_VADDR + available_page * Memory::PAGE_SIZE,
thread->tls_memory, 0, Memory::PAGE_SIZE,
MemoryState::ThreadLocal);
}
// Mark the slot as used
tls_slots[available_page].set(available_slot);
thread->tls_address = Memory::TLS_AREA_VADDR + available_page * Memory::PAGE_SIZE +
available_slot * Memory::TLS_ENTRY_SIZE;
thread->tls_address = thread->owner_process->MarkNextAvailableTLSSlotAsUsed(*thread);
// TODO(peachum): move to ScheduleThread() when scheduler is added so selected core is used
// to initialize the context
@@ -311,13 +257,14 @@ void Thread::BoostPriority(u32 priority) {
}
SharedPtr<Thread> SetupMainThread(KernelCore& kernel, VAddr entry_point, u32 priority,
SharedPtr<Process> owner_process) {
Process& owner_process) {
// Setup page table so we can write to memory
SetCurrentPageTable(&Core::CurrentProcess()->vm_manager.page_table);
SetCurrentPageTable(&owner_process.VMManager().page_table);
// Initialize new "main" thread
const VAddr stack_top = owner_process.VMManager().GetTLSIORegionEndAddress();
auto thread_res = Thread::Create(kernel, "main", entry_point, priority, 0, THREADPROCESSORID_0,
Memory::STACK_AREA_VADDR_END, std::move(owner_process));
stack_top, &owner_process);
SharedPtr<Thread> thread = std::move(thread_res).Unwrap();

View File

@@ -62,6 +62,9 @@ enum class ThreadWakeupReason {
class Thread final : public WaitObject {
public:
using TLSMemory = std::vector<u8>;
using TLSMemoryPtr = std::shared_ptr<TLSMemory>;
/**
* Creates and returns a new thread. The new thread is immediately scheduled
* @param kernel The kernel instance this thread will be created under.
@@ -134,6 +137,14 @@ public:
return thread_id;
}
TLSMemoryPtr& GetTLSMemory() {
return tls_memory;
}
const TLSMemoryPtr& GetTLSMemory() const {
return tls_memory;
}
/**
* Resumes a thread from waiting
*/
@@ -269,7 +280,7 @@ private:
explicit Thread(KernelCore& kernel);
~Thread() override;
std::shared_ptr<std::vector<u8>> tls_memory = std::make_shared<std::vector<u8>>();
TLSMemoryPtr tls_memory = std::make_shared<TLSMemory>();
};
/**
@@ -281,7 +292,7 @@ private:
* @return A shared pointer to the main thread
*/
SharedPtr<Thread> SetupMainThread(KernelCore& kernel, VAddr entry_point, u32 priority,
SharedPtr<Process> owner_process);
Process& owner_process);
/**
* Gets the current thread

View File

@@ -9,6 +9,7 @@
#include "common/logging/log.h"
#include "core/arm/arm_interface.h"
#include "core/core.h"
#include "core/file_sys/program_metadata.h"
#include "core/hle/kernel/errors.h"
#include "core/hle/kernel/vm_manager.h"
#include "core/memory.h"
@@ -54,30 +55,32 @@ bool VirtualMemoryArea::CanBeMergedWith(const VirtualMemoryArea& next) const {
}
VMManager::VMManager() {
Reset();
// Default to assuming a 39-bit address space. This way we have a sane
// starting point with executables that don't provide metadata.
Reset(FileSys::ProgramAddressSpaceType::Is39Bit);
}
VMManager::~VMManager() {
Reset();
Reset(FileSys::ProgramAddressSpaceType::Is39Bit);
}
void VMManager::Reset() {
vma_map.clear();
void VMManager::Reset(FileSys::ProgramAddressSpaceType type) {
Clear();
InitializeMemoryRegionRanges(type);
page_table.Resize(address_space_width);
// Initialize the map with a single free region covering the entire managed space.
VirtualMemoryArea initial_vma;
initial_vma.size = MAX_ADDRESS;
initial_vma.size = address_space_end;
vma_map.emplace(initial_vma.base, initial_vma);
page_table.pointers.fill(nullptr);
page_table.special_regions.clear();
page_table.attributes.fill(Memory::PageType::Unmapped);
UpdatePageTableForVMA(initial_vma);
}
VMManager::VMAHandle VMManager::FindVMA(VAddr target) const {
if (target >= MAX_ADDRESS) {
if (target >= address_space_end) {
return vma_map.end();
} else {
return std::prev(vma_map.upper_bound(target));
@@ -291,7 +294,7 @@ ResultVal<VMManager::VMAIter> VMManager::CarveVMARange(VAddr target, u64 size) {
const VAddr target_end = target + size;
ASSERT(target_end >= target);
ASSERT(target_end <= MAX_ADDRESS);
ASSERT(target_end <= address_space_end);
ASSERT(size > 0);
VMAIter begin_vma = StripIterConstness(FindVMA(target));
@@ -382,6 +385,85 @@ void VMManager::UpdatePageTableForVMA(const VirtualMemoryArea& vma) {
}
}
void VMManager::InitializeMemoryRegionRanges(FileSys::ProgramAddressSpaceType type) {
u64 map_region_size = 0;
u64 heap_region_size = 0;
u64 new_map_region_size = 0;
u64 tls_io_region_size = 0;
switch (type) {
case FileSys::ProgramAddressSpaceType::Is32Bit:
address_space_width = 32;
code_region_base = 0x200000;
code_region_end = code_region_base + 0x3FE00000;
map_region_size = 0x40000000;
heap_region_size = 0x40000000;
break;
case FileSys::ProgramAddressSpaceType::Is36Bit:
address_space_width = 36;
code_region_base = 0x8000000;
code_region_end = code_region_base + 0x78000000;
map_region_size = 0x180000000;
heap_region_size = 0x180000000;
break;
case FileSys::ProgramAddressSpaceType::Is32BitNoMap:
address_space_width = 32;
code_region_base = 0x200000;
code_region_end = code_region_base + 0x3FE00000;
map_region_size = 0;
heap_region_size = 0x80000000;
break;
case FileSys::ProgramAddressSpaceType::Is39Bit:
address_space_width = 39;
code_region_base = 0x8000000;
code_region_end = code_region_base + 0x80000000;
map_region_size = 0x1000000000;
heap_region_size = 0x180000000;
new_map_region_size = 0x80000000;
tls_io_region_size = 0x1000000000;
break;
default:
UNREACHABLE_MSG("Invalid address space type specified: {}", static_cast<u32>(type));
return;
}
address_space_base = 0;
address_space_end = 1ULL << address_space_width;
map_region_base = code_region_end;
map_region_end = map_region_base + map_region_size;
heap_region_base = map_region_end;
heap_region_end = heap_region_base + heap_region_size;
new_map_region_base = heap_region_end;
new_map_region_end = new_map_region_base + new_map_region_size;
tls_io_region_base = new_map_region_end;
tls_io_region_end = tls_io_region_base + tls_io_region_size;
if (new_map_region_size == 0) {
new_map_region_base = address_space_base;
new_map_region_end = address_space_end;
}
}
void VMManager::Clear() {
ClearVMAMap();
ClearPageTable();
}
void VMManager::ClearVMAMap() {
vma_map.clear();
}
void VMManager::ClearPageTable() {
std::fill(page_table.pointers.begin(), page_table.pointers.end(), nullptr);
page_table.special_regions.clear();
std::fill(page_table.attributes.begin(), page_table.attributes.end(),
Memory::PageType::Unmapped);
}
u64 VMManager::GetTotalMemoryUsage() const {
LOG_WARNING(Kernel, "(STUBBED) called");
return 0xF8000000;
@@ -392,14 +474,80 @@ u64 VMManager::GetTotalHeapUsage() const {
return 0x0;
}
VAddr VMManager::GetAddressSpaceBaseAddr() const {
LOG_WARNING(Kernel, "(STUBBED) called");
return 0x8000000;
VAddr VMManager::GetAddressSpaceBaseAddress() const {
return address_space_base;
}
VAddr VMManager::GetAddressSpaceEndAddress() const {
return address_space_end;
}
u64 VMManager::GetAddressSpaceSize() const {
LOG_WARNING(Kernel, "(STUBBED) called");
return MAX_ADDRESS;
return address_space_end - address_space_base;
}
u64 VMManager::GetAddressSpaceWidth() const {
return address_space_width;
}
VAddr VMManager::GetCodeRegionBaseAddress() const {
return code_region_base;
}
VAddr VMManager::GetCodeRegionEndAddress() const {
return code_region_end;
}
u64 VMManager::GetCodeRegionSize() const {
return code_region_end - code_region_base;
}
VAddr VMManager::GetHeapRegionBaseAddress() const {
return heap_region_base;
}
VAddr VMManager::GetHeapRegionEndAddress() const {
return heap_region_end;
}
u64 VMManager::GetHeapRegionSize() const {
return heap_region_end - heap_region_base;
}
VAddr VMManager::GetMapRegionBaseAddress() const {
return map_region_base;
}
VAddr VMManager::GetMapRegionEndAddress() const {
return map_region_end;
}
u64 VMManager::GetMapRegionSize() const {
return map_region_end - map_region_base;
}
VAddr VMManager::GetNewMapRegionBaseAddress() const {
return new_map_region_base;
}
VAddr VMManager::GetNewMapRegionEndAddress() const {
return new_map_region_end;
}
u64 VMManager::GetNewMapRegionSize() const {
return new_map_region_end - new_map_region_base;
}
VAddr VMManager::GetTLSIORegionBaseAddress() const {
return tls_io_region_base;
}
VAddr VMManager::GetTLSIORegionEndAddress() const {
return tls_io_region_end;
}
u64 VMManager::GetTLSIORegionSize() const {
return tls_io_region_end - tls_io_region_base;
}
} // namespace Kernel

View File

@@ -12,6 +12,10 @@
#include "core/memory.h"
#include "core/memory_hook.h"
namespace FileSys {
enum class ProgramAddressSpaceType : u8;
}
namespace Kernel {
enum class VMAType : u8 {
@@ -110,12 +114,6 @@ struct VirtualMemoryArea {
*/
class VMManager final {
public:
/**
* The maximum amount of address space managed by the kernel.
* @todo This was selected arbitrarily, and should be verified for Switch OS.
*/
static constexpr VAddr MAX_ADDRESS{0x1000000000ULL};
/**
* A map covering the entirety of the managed address space, keyed by the `base` field of each
* VMA. It must always be modified by splitting or merging VMAs, so that the invariant
@@ -130,7 +128,7 @@ public:
~VMManager();
/// Clears the address space map, re-initializing with a single free area.
void Reset();
void Reset(FileSys::ProgramAddressSpaceType type);
/// Finds the VMA in which the given address is included in, or `vma_map.end()`.
VMAHandle FindVMA(VAddr target) const;
@@ -195,12 +193,63 @@ public:
/// Gets the total heap usage, used by svcGetInfo
u64 GetTotalHeapUsage() const;
/// Gets the total address space base address, used by svcGetInfo
VAddr GetAddressSpaceBaseAddr() const;
/// Gets the address space base address
VAddr GetAddressSpaceBaseAddress() const;
/// Gets the total address space address size, used by svcGetInfo
/// Gets the address space end address
VAddr GetAddressSpaceEndAddress() const;
/// Gets the total address space address size in bytes
u64 GetAddressSpaceSize() const;
/// Gets the address space width in bits.
u64 GetAddressSpaceWidth() const;
/// Gets the base address of the code region.
VAddr GetCodeRegionBaseAddress() const;
/// Gets the end address of the code region.
VAddr GetCodeRegionEndAddress() const;
/// Gets the total size of the code region in bytes.
u64 GetCodeRegionSize() const;
/// Gets the base address of the heap region.
VAddr GetHeapRegionBaseAddress() const;
/// Gets the end address of the heap region;
VAddr GetHeapRegionEndAddress() const;
/// Gets the total size of the heap region in bytes.
u64 GetHeapRegionSize() const;
/// Gets the base address of the map region.
VAddr GetMapRegionBaseAddress() const;
/// Gets the end address of the map region.
VAddr GetMapRegionEndAddress() const;
/// Gets the total size of the map region in bytes.
u64 GetMapRegionSize() const;
/// Gets the base address of the new map region.
VAddr GetNewMapRegionBaseAddress() const;
/// Gets the end address of the new map region.
VAddr GetNewMapRegionEndAddress() const;
/// Gets the total size of the new map region in bytes.
u64 GetNewMapRegionSize() const;
/// Gets the base address of the TLS IO region.
VAddr GetTLSIORegionBaseAddress() const;
/// Gets the end address of the TLS IO region.
VAddr GetTLSIORegionEndAddress() const;
/// Gets the total size of the TLS IO region in bytes.
u64 GetTLSIORegionSize() const;
/// Each VMManager has its own page table, which is set as the main one when the owning process
/// is scheduled.
Memory::PageTable page_table;
@@ -240,5 +289,36 @@ private:
/// Updates the pages corresponding to this VMA so they match the VMA's attributes.
void UpdatePageTableForVMA(const VirtualMemoryArea& vma);
/// Initializes memory region ranges to adhere to a given address space type.
void InitializeMemoryRegionRanges(FileSys::ProgramAddressSpaceType type);
/// Clears the underlying map and page table.
void Clear();
/// Clears out the VMA map, unmapping any previously mapped ranges.
void ClearVMAMap();
/// Clears out the page table
void ClearPageTable();
u32 address_space_width = 0;
VAddr address_space_base = 0;
VAddr address_space_end = 0;
VAddr code_region_base = 0;
VAddr code_region_end = 0;
VAddr heap_region_base = 0;
VAddr heap_region_end = 0;
VAddr map_region_base = 0;
VAddr map_region_end = 0;
VAddr new_map_region_base = 0;
VAddr new_map_region_end = 0;
VAddr tls_io_region_base = 0;
VAddr tls_io_region_end = 0;
};
} // namespace Kernel

View File

@@ -69,7 +69,7 @@ private:
template <>
inline SharedPtr<WaitObject> DynamicObjectCast<WaitObject>(SharedPtr<Object> object) {
if (object != nullptr && object->IsWaitable()) {
return boost::static_pointer_cast<WaitObject>(std::move(object));
return boost::static_pointer_cast<WaitObject>(object);
}
return nullptr;
}

View File

@@ -130,11 +130,10 @@ private:
void GetAccountId(Kernel::HLERequestContext& ctx) {
LOG_WARNING(Service_ACC, "(STUBBED) called");
// TODO(Subv): Find out what this actually does and implement it. Stub it as an error for
// now since we do not implement NNID. Returning a bogus id here will cause games to send
// invalid IPC requests after ListOpenUsers is called.
IPC::ResponseBuilder rb{ctx, 2};
rb.Push(ResultCode(-1));
// Should return a nintendo account ID
IPC::ResponseBuilder rb{ctx, 4};
rb.Push(RESULT_SUCCESS);
rb.PushRaw<u64>(1);
}
};

View File

@@ -10,6 +10,7 @@
#include "common/alignment.h"
#include "common/common_funcs.h"
#include "common/logging/log.h"
#include "core/core.h"
#include "core/hle/ipc_helpers.h"
#include "core/hle/kernel/event.h"
#include "core/hle/kernel/hle_ipc.h"
@@ -25,7 +26,7 @@ public:
{0, &IAudioRenderer::GetAudioRendererSampleRate, "GetAudioRendererSampleRate"},
{1, &IAudioRenderer::GetAudioRendererSampleCount, "GetAudioRendererSampleCount"},
{2, &IAudioRenderer::GetAudioRendererMixBufferCount, "GetAudioRendererMixBufferCount"},
{3, nullptr, "GetAudioRendererState"},
{3, &IAudioRenderer::GetAudioRendererState, "GetAudioRendererState"},
{4, &IAudioRenderer::RequestUpdateAudioRenderer, "RequestUpdateAudioRenderer"},
{5, &IAudioRenderer::StartAudioRenderer, "StartAudioRenderer"},
{6, &IAudioRenderer::StopAudioRenderer, "StopAudioRenderer"},
@@ -62,6 +63,13 @@ private:
LOG_DEBUG(Service_Audio, "called");
}
void GetAudioRendererState(Kernel::HLERequestContext& ctx) {
IPC::ResponseBuilder rb{ctx, 3};
rb.Push(RESULT_SUCCESS);
rb.Push<u32>(static_cast<u32>(renderer->GetStreamState()));
LOG_DEBUG(Service_Audio, "called");
}
void GetAudioRendererMixBufferCount(Kernel::HLERequestContext& ctx) {
IPC::ResponseBuilder rb{ctx, 3};
rb.Push(RESULT_SUCCESS);

View File

@@ -2,8 +2,17 @@
// Licensed under GPLv2 or any later version
// Refer to the license.txt file included.
#include <array>
#include <cstring>
#include <ctime>
#include <fmt/time.h>
#include "common/file_util.h"
#include "common/logging/log.h"
#include "common/scm_rev.h"
#include "common/swap.h"
#include "core/core.h"
#include "core/hle/ipc_helpers.h"
#include "core/hle/kernel/process.h"
#include "core/hle/service/fatal/fatal.h"
#include "core/hle/service/fatal/fatal_p.h"
#include "core/hle/service/fatal/fatal_u.h"
@@ -15,16 +24,142 @@ Module::Interface::Interface(std::shared_ptr<Module> module, const char* name)
Module::Interface::~Interface() = default;
struct FatalInfo {
std::array<u64_le, 31> registers{}; // TODO(ogniK): See if this actually is registers or
// not(find a game which has non zero valeus)
u64_le unk0{};
u64_le unk1{};
u64_le unk2{};
u64_le unk3{};
u64_le unk4{};
u64_le unk5{};
u64_le unk6{};
std::array<u64_le, 32> backtrace{};
u64_le unk7{};
u64_le unk8{};
u32_le backtrace_size{};
u32_le unk9{};
u32_le unk10{}; // TODO(ogniK): Is this even used or is it just padding?
};
static_assert(sizeof(FatalInfo) == 0x250, "FatalInfo is an invalid size");
enum class FatalType : u32 {
ErrorReportAndScreen = 0,
ErrorReport = 1,
ErrorScreen = 2,
};
static void GenerateErrorReport(ResultCode error_code, const FatalInfo& info) {
const auto title_id = Core::CurrentProcess()->GetTitleID();
std::string crash_report =
fmt::format("Yuzu {}-{} crash report\n"
"Title ID: {:016x}\n"
"Result: 0x{:X} ({:04}-{:04d})\n"
"\n",
Common::g_scm_branch, Common::g_scm_desc, title_id, error_code.raw,
2000 + static_cast<u32>(error_code.module.Value()),
static_cast<u32>(error_code.description.Value()), info.unk8, info.unk7);
if (info.backtrace_size != 0x0) {
crash_report += "Registers:\n";
// TODO(ogniK): This is just a guess, find a game which actually has non zero values
for (size_t i = 0; i < info.registers.size(); i++) {
crash_report +=
fmt::format(" X[{:02d}]: {:016x}\n", i, info.registers[i]);
}
crash_report += fmt::format(" Unknown 0: {:016x}\n", info.unk0);
crash_report += fmt::format(" Unknown 1: {:016x}\n", info.unk1);
crash_report += fmt::format(" Unknown 2: {:016x}\n", info.unk2);
crash_report += fmt::format(" Unknown 3: {:016x}\n", info.unk3);
crash_report += fmt::format(" Unknown 4: {:016x}\n", info.unk4);
crash_report += fmt::format(" Unknown 5: {:016x}\n", info.unk5);
crash_report += fmt::format(" Unknown 6: {:016x}\n", info.unk6);
crash_report += "\nBacktrace:\n";
for (size_t i = 0; i < info.backtrace_size; i++) {
crash_report +=
fmt::format(" Backtrace[{:02d}]: {:016x}\n", i, info.backtrace[i]);
}
crash_report += fmt::format("\nUnknown 7: 0x{:016x}\n", info.unk7);
crash_report += fmt::format("Unknown 8: 0x{:016x}\n", info.unk8);
crash_report += fmt::format("Unknown 9: 0x{:016x}\n", info.unk9);
crash_report += fmt::format("Unknown 10: 0x{:016x}\n", info.unk10);
}
LOG_ERROR(Service_Fatal, "{}", crash_report);
const std::string crashreport_dir =
FileUtil::GetUserPath(FileUtil::UserPath::LogDir) + "crash_logs";
if (!FileUtil::CreateFullPath(crashreport_dir)) {
LOG_ERROR(
Service_Fatal,
"Unable to create crash report directory. Possible log directory permissions issue.");
return;
}
const std::time_t t = std::time(nullptr);
const std::string crashreport_filename =
fmt::format("{}/{:016x}-{:%F-%H%M%S}.log", crashreport_dir, title_id, *std::localtime(&t));
auto file = FileUtil::IOFile(crashreport_filename, "wb");
if (file.IsOpen()) {
file.WriteString(crash_report);
LOG_ERROR(Service_Fatal, "Saving error report to {}", crashreport_filename);
} else {
LOG_ERROR(Service_Fatal, "Failed to save error report to {}", crashreport_filename);
}
}
static void ThrowFatalError(ResultCode error_code, FatalType fatal_type, const FatalInfo& info) {
LOG_ERROR(Service_Fatal, "Threw fatal error type {}", static_cast<u32>(fatal_type));
switch (fatal_type) {
case FatalType::ErrorReportAndScreen:
GenerateErrorReport(error_code, info);
[[fallthrough]];
case FatalType::ErrorScreen:
// Since we have no fatal:u error screen. We should just kill execution instead
ASSERT(false);
break;
// Should not throw a fatal screen but should generate an error report
case FatalType::ErrorReport:
GenerateErrorReport(error_code, info);
break;
};
}
void Module::Interface::ThrowFatal(Kernel::HLERequestContext& ctx) {
LOG_ERROR(Service_Fatal, "called");
IPC::RequestParser rp{ctx};
auto error_code = rp.Pop<ResultCode>();
ThrowFatalError(error_code, FatalType::ErrorScreen, {});
IPC::ResponseBuilder rb{ctx, 2};
rb.Push(RESULT_SUCCESS);
}
void Module::Interface::ThrowFatalWithPolicy(Kernel::HLERequestContext& ctx) {
LOG_ERROR(Service_Fatal, "called");
IPC::RequestParser rp(ctx);
u32 error_code = rp.Pop<u32>();
LOG_WARNING(Service_Fatal, "(STUBBED) called, error_code=0x{:X}", error_code);
auto error_code = rp.Pop<ResultCode>();
auto fatal_type = rp.PopEnum<FatalType>();
ThrowFatalError(error_code, fatal_type, {}); // No info is passed with ThrowFatalWithPolicy
IPC::ResponseBuilder rb{ctx, 2};
rb.Push(RESULT_SUCCESS);
}
void Module::Interface::ThrowFatalWithCpuContext(Kernel::HLERequestContext& ctx) {
LOG_WARNING(Service_Fatal, "(STUBBED) called");
LOG_ERROR(Service_Fatal, "called");
IPC::RequestParser rp(ctx);
auto error_code = rp.Pop<ResultCode>();
auto fatal_type = rp.PopEnum<FatalType>();
auto fatal_info = ctx.ReadBuffer();
FatalInfo info{};
ASSERT_MSG(fatal_info.size() == sizeof(FatalInfo), "Invalid fatal info buffer size!");
std::memcpy(&info, fatal_info.data(), sizeof(FatalInfo));
ThrowFatalError(error_code, fatal_type, info);
IPC::ResponseBuilder rb{ctx, 2};
rb.Push(RESULT_SUCCESS);
}

View File

@@ -15,6 +15,7 @@ public:
explicit Interface(std::shared_ptr<Module> module, const char* name);
~Interface() override;
void ThrowFatal(Kernel::HLERequestContext& ctx);
void ThrowFatalWithPolicy(Kernel::HLERequestContext& ctx);
void ThrowFatalWithCpuContext(Kernel::HLERequestContext& ctx);

View File

@@ -8,7 +8,7 @@ namespace Service::Fatal {
Fatal_U::Fatal_U(std::shared_ptr<Module> module) : Module::Interface(std::move(module), "fatal:u") {
static const FunctionInfo functions[] = {
{0, nullptr, "ThrowFatal"},
{0, &Fatal_U::ThrowFatal, "ThrowFatal"},
{1, &Fatal_U::ThrowFatalWithPolicy, "ThrowFatalWithPolicy"},
{2, &Fatal_U::ThrowFatalWithCpuContext, "ThrowFatalWithCpuContext"},
};

View File

@@ -197,7 +197,7 @@ ResultVal<FileSys::VirtualDir> VfsDirectoryServiceWrapper::OpenDirectory(const s
auto dir = GetDirectoryRelativeWrapped(backing, path);
if (dir == nullptr) {
// TODO(DarkLordZach): Find a better error code for this
return ResultCode(-1);
return FileSys::ERROR_PATH_NOT_FOUND;
}
return MakeResult(dir);
}
@@ -343,6 +343,15 @@ std::shared_ptr<FileSys::RegisteredCache> GetSDMCContents() {
return sdmc_factory->GetSDMCContents();
}
FileSys::VirtualDir GetModificationLoadRoot(u64 title_id) {
LOG_TRACE(Service_FS, "Opening mod load root for tid={:016X}", title_id);
if (bis_factory == nullptr)
return nullptr;
return bis_factory->GetModificationLoadRoot(title_id);
}
void CreateFactories(const FileSys::VirtualFilesystem& vfs, bool overwrite) {
if (overwrite) {
bis_factory = nullptr;
@@ -354,9 +363,11 @@ void CreateFactories(const FileSys::VirtualFilesystem& vfs, bool overwrite) {
FileSys::Mode::ReadWrite);
auto sd_directory = vfs->OpenDirectory(FileUtil::GetUserPath(FileUtil::UserPath::SDMCDir),
FileSys::Mode::ReadWrite);
auto load_directory = vfs->OpenDirectory(FileUtil::GetUserPath(FileUtil::UserPath::LoadDir),
FileSys::Mode::ReadWrite);
if (bis_factory == nullptr)
bis_factory = std::make_unique<FileSys::BISFactory>(nand_directory);
bis_factory = std::make_unique<FileSys::BISFactory>(nand_directory, load_directory);
if (save_data_factory == nullptr)
save_data_factory = std::make_unique<FileSys::SaveDataFactory>(std::move(nand_directory));
if (sdmc_factory == nullptr)

View File

@@ -52,6 +52,8 @@ std::shared_ptr<FileSys::RegisteredCache> GetSystemNANDContents();
std::shared_ptr<FileSys::RegisteredCache> GetUserNANDContents();
std::shared_ptr<FileSys::RegisteredCache> GetSDMCContents();
FileSys::VirtualDir GetModificationLoadRoot(u64 title_id);
// Creates the SaveData, SDMC, and BIS Factories. Should be called once and before any function
// above is called.
void CreateFactories(const FileSys::VirtualFilesystem& vfs, bool overwrite = true);

View File

@@ -468,6 +468,7 @@ FSP_SRV::FSP_SRV() : ServiceFramework("fsp-srv") {
{80, nullptr, "OpenSaveDataMetaFile"},
{81, nullptr, "OpenSaveDataTransferManager"},
{82, nullptr, "OpenSaveDataTransferManagerVersion2"},
{83, nullptr, "OpenSaveDataTransferProhibiterForCloudBackUp"},
{100, nullptr, "OpenImageDirectoryFileSystem"},
{110, nullptr, "OpenContentStorageFileSystem"},
{200, &FSP_SRV::OpenDataStorageByCurrentProcess, "OpenDataStorageByCurrentProcess"},
@@ -495,6 +496,7 @@ FSP_SRV::FSP_SRV() : ServiceFramework("fsp-srv") {
{613, nullptr, "VerifySaveDataFileSystemBySaveDataSpaceId"},
{614, nullptr, "CorruptSaveDataFileSystemBySaveDataSpaceId"},
{615, nullptr, "QuerySaveDataInternalStorageTotalSize"},
{616, nullptr, "GetSaveDataCommitId"},
{620, nullptr, "SetSdCardEncryptionSeed"},
{630, nullptr, "SetSdCardAccessibility"},
{631, nullptr, "IsSdCardAccessible"},

View File

@@ -338,7 +338,7 @@ public:
{106, &Hid::AcquireNpadStyleSetUpdateEventHandle, "AcquireNpadStyleSetUpdateEventHandle"},
{107, &Hid::DisconnectNpad, "DisconnectNpad"},
{108, &Hid::GetPlayerLedPattern, "GetPlayerLedPattern"},
{109, nullptr, "ActivateNpadWithRevision"},
{109, &Hid::ActivateNpadWithRevision, "ActivateNpadWithRevision"},
{120, &Hid::SetNpadJoyHoldType, "SetNpadJoyHoldType"},
{121, &Hid::GetNpadJoyHoldType, "GetNpadJoyHoldType"},
{122, &Hid::SetNpadJoyAssignmentModeSingleByDefault, "SetNpadJoyAssignmentModeSingleByDefault"},
@@ -603,6 +603,12 @@ private:
rb.Push(RESULT_SUCCESS);
LOG_WARNING(Service_HID, "(STUBBED) called");
}
void ActivateNpadWithRevision(Kernel::HLERequestContext& ctx) {
IPC::ResponseBuilder rb{ctx, 2};
rb.Push(RESULT_SUCCESS);
LOG_WARNING(Service_HID, "(STUBBED) called");
}
};
class HidDbg final : public ServiceFramework<HidDbg> {

View File

@@ -2,6 +2,11 @@
// Licensed under GPLv2 or any later version
// Refer to the license.txt file included.
#include "common/swap.h"
#include "core/core.h"
#include "core/core_timing.h"
#include "core/hle/ipc_helpers.h"
#include "core/hle/kernel/shared_memory.h"
#include "core/hle/service/hid/irs.h"
namespace Service::HID {
@@ -9,28 +14,145 @@ namespace Service::HID {
IRS::IRS() : ServiceFramework{"irs"} {
// clang-format off
static const FunctionInfo functions[] = {
{302, nullptr, "ActivateIrsensor"},
{303, nullptr, "DeactivateIrsensor"},
{304, nullptr, "GetIrsensorSharedMemoryHandle"},
{305, nullptr, "StopImageProcessor"},
{306, nullptr, "RunMomentProcessor"},
{307, nullptr, "RunClusteringProcessor"},
{308, nullptr, "RunImageTransferProcessor"},
{309, nullptr, "GetImageTransferProcessorState"},
{310, nullptr, "RunTeraPluginProcessor"},
{311, nullptr, "GetNpadIrCameraHandle"},
{312, nullptr, "RunPointingProcessor"},
{313, nullptr, "SuspendImageProcessor"},
{314, nullptr, "CheckFirmwareVersion"},
{315, nullptr, "SetFunctionLevel"},
{316, nullptr, "RunImageTransferExProcessor"},
{317, nullptr, "RunIrLedProcessor"},
{318, nullptr, "StopImageProcessorAsync"},
{319, nullptr, "ActivateIrsensorWithFunctionLevel"},
{302, &IRS::ActivateIrsensor, "ActivateIrsensor"},
{303, &IRS::DeactivateIrsensor, "DeactivateIrsensor"},
{304, &IRS::GetIrsensorSharedMemoryHandle, "GetIrsensorSharedMemoryHandle"},
{305, &IRS::StopImageProcessor, "StopImageProcessor"},
{306, &IRS::RunMomentProcessor, "RunMomentProcessor"},
{307, &IRS::RunClusteringProcessor, "RunClusteringProcessor"},
{308, &IRS::RunImageTransferProcessor, "RunImageTransferProcessor"},
{309, &IRS::GetImageTransferProcessorState, "GetImageTransferProcessorState"},
{310, &IRS::RunTeraPluginProcessor, "RunTeraPluginProcessor"},
{311, &IRS::GetNpadIrCameraHandle, "GetNpadIrCameraHandle"},
{312, &IRS::RunPointingProcessor, "RunPointingProcessor"},
{313, &IRS::SuspendImageProcessor, "SuspendImageProcessor"},
{314, &IRS::CheckFirmwareVersion, "CheckFirmwareVersion"},
{315, &IRS::SetFunctionLevel, "SetFunctionLevel"},
{316, &IRS::RunImageTransferExProcessor, "RunImageTransferExProcessor"},
{317, &IRS::RunIrLedProcessor, "RunIrLedProcessor"},
{318, &IRS::StopImageProcessorAsync, "StopImageProcessorAsync"},
{319, &IRS::ActivateIrsensorWithFunctionLevel, "ActivateIrsensorWithFunctionLevel"},
};
// clang-format on
RegisterHandlers(functions);
auto& kernel = Core::System::GetInstance().Kernel();
shared_mem = Kernel::SharedMemory::Create(
kernel, nullptr, 0x8000, Kernel::MemoryPermission::ReadWrite,
Kernel::MemoryPermission::Read, 0, Kernel::MemoryRegion::BASE, "IRS:SharedMemory");
}
void IRS::ActivateIrsensor(Kernel::HLERequestContext& ctx) {
IPC::ResponseBuilder rb{ctx, 2};
rb.Push(RESULT_SUCCESS);
LOG_WARNING(Service_IRS, "(STUBBED) called");
}
void IRS::DeactivateIrsensor(Kernel::HLERequestContext& ctx) {
IPC::ResponseBuilder rb{ctx, 2};
rb.Push(RESULT_SUCCESS);
LOG_WARNING(Service_IRS, "(STUBBED) called");
}
void IRS::GetIrsensorSharedMemoryHandle(Kernel::HLERequestContext& ctx) {
IPC::ResponseBuilder rb{ctx, 2, 1};
rb.Push(RESULT_SUCCESS);
rb.PushCopyObjects(shared_mem);
LOG_DEBUG(Service_IRS, "called");
}
void IRS::StopImageProcessor(Kernel::HLERequestContext& ctx) {
IPC::ResponseBuilder rb{ctx, 2};
rb.Push(RESULT_SUCCESS);
LOG_WARNING(Service_IRS, "(STUBBED) called");
}
void IRS::RunMomentProcessor(Kernel::HLERequestContext& ctx) {
IPC::ResponseBuilder rb{ctx, 2};
rb.Push(RESULT_SUCCESS);
LOG_WARNING(Service_IRS, "(STUBBED) called");
}
void IRS::RunClusteringProcessor(Kernel::HLERequestContext& ctx) {
IPC::ResponseBuilder rb{ctx, 2};
rb.Push(RESULT_SUCCESS);
LOG_WARNING(Service_IRS, "(STUBBED) called");
}
void IRS::RunImageTransferProcessor(Kernel::HLERequestContext& ctx) {
IPC::ResponseBuilder rb{ctx, 2};
rb.Push(RESULT_SUCCESS);
LOG_WARNING(Service_IRS, "(STUBBED) called");
}
void IRS::GetImageTransferProcessorState(Kernel::HLERequestContext& ctx) {
IPC::ResponseBuilder rb{ctx, 5};
rb.Push(RESULT_SUCCESS);
rb.PushRaw<u64>(CoreTiming::GetTicks());
rb.PushRaw<u32>(0);
LOG_WARNING(Service_IRS, "(STUBBED) called");
}
void IRS::RunTeraPluginProcessor(Kernel::HLERequestContext& ctx) {
IPC::ResponseBuilder rb{ctx, 2};
rb.Push(RESULT_SUCCESS);
LOG_WARNING(Service_IRS, "(STUBBED) called");
}
void IRS::GetNpadIrCameraHandle(Kernel::HLERequestContext& ctx) {
IPC::ResponseBuilder rb{ctx, 3};
rb.Push(RESULT_SUCCESS);
rb.PushRaw<u32>(device_handle);
LOG_WARNING(Service_IRS, "(STUBBED) called");
}
void IRS::RunPointingProcessor(Kernel::HLERequestContext& ctx) {
IPC::ResponseBuilder rb{ctx, 2};
rb.Push(RESULT_SUCCESS);
LOG_WARNING(Service_IRS, "(STUBBED) called");
}
void IRS::SuspendImageProcessor(Kernel::HLERequestContext& ctx) {
IPC::ResponseBuilder rb{ctx, 2};
rb.Push(RESULT_SUCCESS);
LOG_WARNING(Service_IRS, "(STUBBED) called");
}
void IRS::CheckFirmwareVersion(Kernel::HLERequestContext& ctx) {
IPC::ResponseBuilder rb{ctx, 2};
rb.Push(RESULT_SUCCESS);
LOG_WARNING(Service_IRS, "(STUBBED) called");
}
void IRS::SetFunctionLevel(Kernel::HLERequestContext& ctx) {
IPC::ResponseBuilder rb{ctx, 2};
rb.Push(RESULT_SUCCESS);
LOG_WARNING(Service_IRS, "(STUBBED) called");
}
void IRS::RunImageTransferExProcessor(Kernel::HLERequestContext& ctx) {
IPC::ResponseBuilder rb{ctx, 2};
rb.Push(RESULT_SUCCESS);
LOG_WARNING(Service_IRS, "(STUBBED) called");
}
void IRS::RunIrLedProcessor(Kernel::HLERequestContext& ctx) {
IPC::ResponseBuilder rb{ctx, 2};
rb.Push(RESULT_SUCCESS);
LOG_WARNING(Service_IRS, "(STUBBED) called");
}
void IRS::StopImageProcessorAsync(Kernel::HLERequestContext& ctx) {
IPC::ResponseBuilder rb{ctx, 2};
rb.Push(RESULT_SUCCESS);
LOG_WARNING(Service_IRS, "(STUBBED) called");
}
void IRS::ActivateIrsensorWithFunctionLevel(Kernel::HLERequestContext& ctx) {
IPC::ResponseBuilder rb{ctx, 2};
rb.Push(RESULT_SUCCESS);
LOG_WARNING(Service_IRS, "(STUBBED) called");
}
IRS::~IRS() = default;

View File

@@ -4,14 +4,41 @@
#pragma once
#include "core/hle/kernel/object.h"
#include "core/hle/service/service.h"
namespace Kernel {
class SharedMemory;
}
namespace Service::HID {
class IRS final : public ServiceFramework<IRS> {
public:
explicit IRS();
~IRS() override;
private:
void ActivateIrsensor(Kernel::HLERequestContext& ctx);
void DeactivateIrsensor(Kernel::HLERequestContext& ctx);
void GetIrsensorSharedMemoryHandle(Kernel::HLERequestContext& ctx);
void StopImageProcessor(Kernel::HLERequestContext& ctx);
void RunMomentProcessor(Kernel::HLERequestContext& ctx);
void RunClusteringProcessor(Kernel::HLERequestContext& ctx);
void RunImageTransferProcessor(Kernel::HLERequestContext& ctx);
void GetImageTransferProcessorState(Kernel::HLERequestContext& ctx);
void RunTeraPluginProcessor(Kernel::HLERequestContext& ctx);
void GetNpadIrCameraHandle(Kernel::HLERequestContext& ctx);
void RunPointingProcessor(Kernel::HLERequestContext& ctx);
void SuspendImageProcessor(Kernel::HLERequestContext& ctx);
void CheckFirmwareVersion(Kernel::HLERequestContext& ctx);
void SetFunctionLevel(Kernel::HLERequestContext& ctx);
void RunImageTransferExProcessor(Kernel::HLERequestContext& ctx);
void RunIrLedProcessor(Kernel::HLERequestContext& ctx);
void StopImageProcessorAsync(Kernel::HLERequestContext& ctx);
void ActivateIrsensorWithFunctionLevel(Kernel::HLERequestContext& ctx);
Kernel::SharedPtr<Kernel::SharedMemory> shared_mem;
const u32 device_handle{0xABCD};
};
class IRS_SYS final : public ServiceFramework<IRS_SYS> {

View File

@@ -3,6 +3,7 @@
// Refer to the license.txt file included.
#include "common/logging/log.h"
#include "core/core.h"
#include "core/hle/ipc_helpers.h"
#include "core/hle/kernel/event.h"
#include "core/hle/service/hid/hid.h"

Some files were not shown because too many files have changed in this diff Show More