Commit graph

4474 commits

Author SHA1 Message Date
Lioncash
73e1e929a2 XCI: Add function for checking the existence of the program NCA
The only reason the getter existed was to check whether or not the
program NCA was null. Instead, we can just provide a function to query
for the existence of it, instead of exposing it entirely.
2018-10-16 11:36:58 -04:00
bunnei
78f2a6a9e1
Merge pull request #1443 from DarkLordZach/lower-loader-logs-1
content_archive/patch_manager: Lower log levels to eliminate some unnecessary logs
2018-10-16 11:26:54 -04:00
David
92d8ad3770 Implement VI ConvertScalingMode (#1475)
* Implement VI ConvertScalingMode

* Fixed push enum

* Scale mode now uses Nintendo scale mode as an enum as well
2018-10-16 11:25:42 -04:00
bunnei
88b8383da2
Merge pull request #1502 from lioncash/unique
core: Convert shared_ptr instances into unique_ptr instances where applicable for System and Cpu
2018-10-16 11:21:42 -04:00
Zach Hilman
9d4e6176eb savedata_factory: Add TemporaryStorage SaveDataSpaceId
Required for TemporaryStorage saves (in addition to SaveDataType)
2018-10-16 10:20:04 -04:00
Zach Hilman
74890cf2da savedata_factory: Add support for DeviceSaveData
Uses the same path as SaveData except with UID 0. Adds a warning if UID is not 0.
2018-10-16 10:19:21 -04:00
Lioncash
39ae73b356 file_sys/registered_cache: Use unique_ptr and regular pointers instead of shared_ptrs where applicable
The data retrieved in these cases are ultimately chiefly owned by either
the RegisteredCache instance itself, or the filesystem factories. Both
these should live throughout the use of their contained data. If they
don't, it should be considered an interface/design issue, and using
shared_ptr instances here would mask that, as the data would always be
prolonged after the main owner's lifetime ended.

This makes the lifetime of the data explicit and makes it harder to
accidentally create cyclic references. It also makes the interface
slightly more flexible than the previous API, as a shared_ptr can be
created from a unique_ptr, but not the other way around, so this allows
for that use-case if it ever becomes necessary in some form.
2018-10-16 09:38:52 -04:00
bunnei
548958bcaf
Merge pull request #1473 from lioncash/cmake
web_service: Make linkage of web_service-related externals and the library private
2018-10-15 21:33:32 -04:00
Lioncash
76fc8b59b2 file_sys/control_metadata: Get rid of magic constants
These are just the size of the data being passed in, so we can specify
that via the size() member function.
2018-10-15 20:11:44 -04:00
bunnei
9b21fbd1eb
Merge pull request #1494 from DarkLordZach/aoc-signature-fixes
aoc: Fix various bugs in current AOC implementation
2018-10-15 18:34:02 -04:00
bunnei
50e6205c21
Merge pull request #1499 from lioncash/nro
nro/nso: Minor error handling changes
2018-10-15 17:48:36 -04:00
Lioncash
bed872ed38 nso: Return an optional address from LoadModule
If a malformed NSO is attempted to be loaded, we shouldn't continue
onwards. We should be reporting an error and bailing out.
2018-10-15 17:02:11 -04:00
Lioncash
5484742fda core_cpu: Make Cpu scheduler instances unique_ptrs instead of shared_ptrs 2018-10-15 14:15:56 -04:00
Lioncash
59f872a8e0 core: Make the live Cpu instances unique_ptrs instead of shared_ptrs
There's no need for shared ownership here, as the only owning class
instance of those Cpu instances is the System class itself. We can also
make the thread_to_cpu map use regular pointers instead of shared_ptrs,
given that the Cpu instances will always outlive the cases where they're
used with that map.
2018-10-15 14:15:56 -04:00
Lioncash
aeadbfa790 core: Make the exclusive monitor a unique_ptr instead of a shared_ptr
Like the barrier, this is owned entirely by the System and will always
outlive the encompassing state, so shared ownership semantics aren't
necessary here.
2018-10-15 14:15:50 -04:00
Lioncash
c34efbbd60 core: Make CPUBarrier a unique_ptr instead of a shared_ptr
This will always outlive the Cpu instances, since it's destroyed after
we destroy the Cpu instances on shutdown, so there's no need for shared
ownership semantics here.
2018-10-15 09:11:47 -04:00
Zach Hilman
720d36ca71 crypto: Various crypto fixes for quickstart guide 2018-10-14 21:57:52 -04:00
Lioncash
bb9cf8a127 nso: Make LoadModule take a VfsFile by const reference 2018-10-14 20:38:19 -04:00
Lioncash
0732786ddc nro: Make LoadNro take a VfsFile by const reference
This function doesn't need to care about ownership semantics, so we can
just pass it a reference to the file itself, rather than a
std::shared_ptr alias.
2018-10-14 20:24:18 -04:00
Lioncash
90f8474fc1 svc: Clarify enum values for AddressSpaceBaseAddr and AddressSpaceSize in svcGetInfo()
So, one thing that's puzzled me is why the kernel seemed to *not* use
the direct code address ranges in some cases for some service functions.
For example, in svcMapMemory, the full address space width is compared
against for validity, but for svcMapSharedMemory, it compares against
0xFFE00000, 0xFF8000000, and 0x7FF8000000 as upper bounds, and uses
either 0x200000 or 0x8000000 as the lower-bounds as the beginning of the
compared range. Coincidentally, these exact same values are also used in
svcGetInfo, and also when initializing the user address space, so this
is actually retrieving the ASLR extents, not the extents of the address
space in general.
2018-10-14 20:11:16 -04:00
Zach Hilman
5737441374 aoc: Read DLC base title ID from RegisteredCache
Falls back to title ID + 0x1000, which is what HOS does.
2018-10-14 18:58:14 -04:00
bunnei
b3cca34f50
Merge pull request #1486 from lioncash/file
key_manager/partition_data_manager: Minor changes
2018-10-14 14:46:47 -04:00
bunnei
0d2ba0a320
Merge pull request #1491 from lioncash/reference
filesystem: Make CreateFactories() and InstallInterface() take a VfsFilesystem by reference
2018-10-14 14:42:57 -04:00
bunnei
2f8ca32020
Merge pull request #1492 from lioncash/proc
svc: Implement svcGetProcessInfo
2018-10-14 14:37:58 -04:00
David Marcec
92fae7e1ab Stop all threads on svcBreak
This should help diagnose crashes easier and prevent many users thinking that a game is still running when in fact it's just an audio thread still running(this is typically not killed when svcBreak is hit since the game expects us to do this)
2018-10-14 18:14:51 +11:00
Zach Hilman
7e2096db8a aoc: Return size in ListAddOnContent 2018-10-13 22:52:54 -04:00
Lioncash
1c7a7ed79b svc: Implement svcGetProcessInfo
A fairly basic service function, which only appears to currently support
retrieving the process state. This also alters the ProcessStatus enum to
contain all of the values that a kernel process seems to be able of
reporting with regards to state.
2018-10-13 17:00:43 -04:00
Lioncash
0149162dba filesystem: Make CreateFactories() and InstallInterface() take a VfsFilesystem instance by reference
Neither of these functions alter the ownership of the provided pointer,
so we can simply make the parameters a reference rather than a direct
shared pointer alias. This way we also disallow passing incorrect memory values like
nullptr.
2018-10-13 11:36:35 -04:00
Lioncash
6467b01de2 partition_data_manager: Reserve and insert data within output vector in DecryptPackage2()
We can just reserve the memory then perform successive insertions
instead of needing to use memcpy. This also avoids the need to zero out
the output vector's memory before performing the insertions.

We can also std::move the output std::vector into the destination so
that we don't need to make a completely new copy of the vector, getting
rid of an unnecessary allocation.

Additionally, we can use iterators to determine the beginning and end
ranges of the std::vector instances that comprise the output vector, as
the end of one range just becomes the beginning for the next successive
range, and since std::vector's iterator constructor copies data within
the range [begin, end), this is more straightforward and gets rid of the
need to have an offset variable that keeps getting incremented to
determine where to do the next std::memcpy.
2018-10-13 09:50:08 -04:00
Lioncash
781fd7983c partition_data_manager: Remove unused std::map instance within DecryptPackage2()
Aside from emplacing elements into the map, the map itself is never
actually queried for contained data.
2018-10-13 09:27:12 -04:00
Lioncash
e0c76226ad partition_data_manager: Take package2_keys by const reference
These are only ever read from, so we don't need to make a copy of all
the keys here.
2018-10-13 09:24:41 -04:00
Lioncash
3d9df49619 partition_data_manager: Move IV data to where it's needed in DecryptPackage2()
Given it's only used in one spot and has a fairly generic name, we can
just specify it directly in the function call. This also the benefit of
automatically moving it.
2018-10-13 09:20:21 -04:00
Lioncash
bc2196bb09 partition_data_manager: Remove commented out code
Commented out code shouldn't be left in without a reason indicating why
in a comment.
2018-10-13 09:17:02 -04:00
Lioncash
6da2ed4232 key_manager/partition_data_manager: Silence truncation compiler warnings 2018-10-13 09:13:19 -04:00
Lioncash
f56a8da46a partition_data_manager: Dehardcode array bounds
Instead, we can make it part of the type and make named variables for
them, so they only require one definition (and if they ever change for
whatever reason, they only need to be changed in one spot).
2018-10-13 08:52:37 -04:00
Lioncash
d257a3b56c partition_data_manager: Take VirtualFile by const reference in constructor
Given the VirtualFile instance isn't stored into the class as a data
member, or written to, this can just be turned into a const reference,
as the constructor doesn't need to make a copy of it.
2018-10-13 08:39:05 -04:00
Lioncash
e96d69c328 partition_data_manager: Amend constructor initializer list order
Orders the members in the exact order they would be initialized. This
also prevents compiler warnings about this sort of thing.
2018-10-13 08:36:26 -04:00
Lioncash
aaca7543f0 partition_data_manager: Remove unused includes
Gets unused includes out of the headers and moves them into the cpp file
if they're used there instead.
2018-10-13 08:33:49 -04:00
Lioncash
06898263f6 key_manager: Use std::vector's insert() instead of std::copy with a back_inserter
If the data is unconditionally being appended to the back of a
std::vector, we can just directly insert it there without the need to
insert all of the elements one-by-one with a std::back_inserter.
2018-10-13 08:29:35 -04:00
Lioncash
e70c08b543 key_manager: Brace long conditional body
If a conditional (or it's body) travels more than one line, it should be
braced.
2018-10-13 08:24:21 -04:00
Lioncash
ef5639bfbb key_manager: Don't assume file seeks and reads will always succeed
Given the filesystem should always be assumed to be volatile, we should
check and bail out if a seek operation isn't successful. This'll prevent
potentially writing/returning garbage data from the function in rare
cases.

This also allows removing a check to see if an offset is within the
bounds of a file before perfoming a seek operation. If a seek is
attempted beyond the end of a file, it will fail, so this essentially
combines two checks into one in one place.
2018-10-13 08:24:18 -04:00
Lioncash
82ea1cf35a key_manager: Remove unnecessary seek in DeriveSDSeed()
Given the file is opened a few lines above and no operations are done,
other than check if the file is in a valid state, the read/write pointer
will always be at the beginning of the file.
2018-10-13 08:08:44 -04:00
Zach Hilman
f61379f8d2 patch_manager: Move non-Program RomFS patch log to Debug
Normal Program-type patches will still be logged to aid in debugging, but for others (mainly Control), it was moved to Debug.
2018-10-12 23:27:19 -04:00
Zach Hilman
90c07e0d33 content_archive: Move get key log to Trace level
Avoids printing live keys in the general log.
2018-10-12 23:25:59 -04:00
bunnei
1584fb6b38
Merge pull request #1409 from DarkLordZach/key-derivation
crypto: Add support for full key derivation
2018-10-12 22:55:49 -04:00
bunnei
c2aa4293ec
Merge pull request #1483 from lioncash/codeset
kernel/process: Make CodeSet a regular non-inherited object
2018-10-12 22:52:12 -04:00
bunnei
ffcda6c08e
Merge pull request #1481 from lioncash/typo
svc: Fix typos in sanitizing checks for MapMemory/UnmapMemory
2018-10-12 20:46:55 -04:00
bunnei
2946d4bdbe
Merge pull request #1467 from ogniK5377/svcbreak-type-fix
Fixed incorrect types for svcBreak
2018-10-12 12:08:08 -04:00
Lioncash
1abed2f4c4 kernel/process: Make CodeSet a regular non-inherited object
These only exist to ferry data into a Process instance and end up going
out of scope quite early. Because of this, we can just make it a plain
struct for holding things and just std::move it into the relevant
function. There's no need to make this inherit from the kernel's Object
type.
2018-10-12 12:07:32 -04:00
bunnei
0f7ab3e21a
Merge pull request #1478 from ogniK5377/remap-invalidhandle-remap
Passing an invalid nmap handle to Remap should throw an error
2018-10-12 12:07:14 -04:00
bunnei
f9d03b1d41
Merge pull request #1482 from lioncash/init
thread: Remove unnecessary memset from ResetThreadContext()
2018-10-12 12:06:51 -04:00
bunnei
dc328440c8
Merge pull request #1479 from ogniK5377/nmap-revamped
Added error codes for nvmap
2018-10-12 12:06:22 -04:00
Lioncash
b492d43e63 thread: Remove unnecessary memset from ResetThreadContext()
Regular value initialization is adequate here for zeroing out data. It
also has the benefit of not invoking undefined behavior if a non-trivial
type is ever added to the struct for whatever reason.
2018-10-12 10:57:31 -04:00
David Marcec
4d2de6564f Returned an error before processing other remaps 2018-10-12 17:10:41 +11:00
David Marcec
c55b5de0fb Made the minimum alignment more clear 2018-10-12 17:06:46 +11:00
Lioncash
4ccf30dfaa svc: Fix typos in sanitizing checks for MapMemory/UnmapMemory 2018-10-12 01:48:26 -04:00
David Marcec
98b760c645 Wip 2018-10-12 16:28:00 +11:00
bunnei
9bf409f275
Merge pull request #1474 from ogniK5377/hwopus-decodeinterleavedwithperformance
HwOpus, Implemented DecodeInterleavedWithPerformance
2018-10-11 16:52:13 -04:00
bunnei
3fd26b7147
Merge pull request #1472 from lioncash/san
svc: Add missing address range sanitizing checks to MapMemory/UnmapMemory
2018-10-11 16:51:41 -04:00
David Marcec
85b0d9a7be Dynamically decide handheld variant based on supported npad id priority
Kirby input still doesn't work, should fix a lot of other games
2018-10-12 02:56:49 +11:00
David Marcec
c7763603ef Added error codes for nvmap 2018-10-11 23:06:34 +11:00
David Marcec
5dd538cace Passing an invalid nmap handle to Remap should throw an error
Added error for invalid nmap handles
2018-10-11 20:32:21 +11:00
bunnei
bf795edac4 nvhost_as_gpu: Flush CPU VAddr on UnmapBuffer. 2018-10-11 00:19:36 -04:00
Lioncash
28ec921d0d core/CMakeLists: Make all web_service-related libraries private
Now that all external dependencies are hidden, we can remove
json-headers from the publically linked libraries, as the use of this
library is now completely hidden from external users of the web_service
library. We can also make the web_services library private as well,
considering it's not a requirement. If a library needs to link in
web_service, it should be done explicitly -- not via indirect linking.
2018-10-10 22:29:39 -04:00
David Marcec
fa10905e1e HwOpus, Implemented DecodeInterleavedWithPerformance
Used by sonic ages
2018-10-11 13:06:56 +11:00
Lioncash
6e6ce2ce39 core/CMakeLists: Use target_compile_definitions instead of add_definitions for specifying ENABLE_WEB_SERVICE
Avoids introducing the definition to the whole directory space and
localizes it to being added to the library that needs it.
2018-10-10 20:54:02 -04:00
Lioncash
72e9cb523e svc: Add missing address range sanitizing checks to MapMemory/UnmapMemory
This adds the missing address range checking that the service functions
do before attempting to map or unmap memory. Given that both service
functions perform the same set of checks in the same order, we can wrap
these into a function and just call it from both functions, which
deduplicates a little bit of code.
2018-10-10 20:30:49 -04:00
David Marcec
9e924f2ef2 Added BeginPermitVibrationSession and EndPermitVibrationSession
Used by Mario Party
2018-10-11 00:58:47 +11:00
David Marcec
3d75c9cd7a Added GetLedPattern and HandheldVariant
HandheldVariant is for specific games which expect handheld controllers to be at position 8(kirby), however this doesn't fix all games as some games require handhelds to be at position 0(snipperclips)
2018-10-10 21:38:43 +11:00
Lioncash
5c0408596f kernel/thread: Use a regular pointer for the owner/current process
There's no real need to use a shared pointer in these cases, and only
makes object management more fragile in terms of how easy it would be to
introduce cycles. Instead, just do the simple thing of using a regular
pointer. Much of this is just a hold-over from citra anyways.

It also doesn't make sense from a behavioral point of view for a
process' thread to prolong the lifetime of the process itself (the
process is supposed to own the thread, not the other way around).
2018-10-10 02:04:55 -04:00
David Marcec
46cdeb4549 Kirby expects handheld controllers to be at position 8 2018-10-10 14:21:56 +11:00
bunnei
5461b21c7a
Merge pull request #1461 from lioncash/warn
ips_layer: Silence truncation and conversion warnings
2018-10-09 22:30:01 -04:00
bunnei
3ac874c32e
Merge pull request #1464 from lioncash/unique
patch_manager: Return a std::unique_ptr from ParseControlNCA() and GetControlMetadata() instead of a std::shared_ptr
2018-10-09 22:29:39 -04:00
David Marcec
f43815af5d Added the ability to "disconnect" individual npads
Fixes arms
2018-10-10 13:15:39 +11:00
David Marcec
b79c294c02 Removed unneeded forward declarations 2018-10-10 13:15:37 +11:00
David Marcec
5857aea94e Addressed changes for better hid 2018-10-10 13:15:37 +11:00
David Marcec
56f35ab262 "Better Hid" rework part 1 2018-10-10 13:15:35 +11:00
David Marcec
2db37ddea9 Changed all casts in svc_wrap.h to be static_cast instead 2018-10-10 12:49:08 +11:00
David Marcec
09b6dda8f0 Use a better name than "dont_kill_application"
signal_debugger seems like a more fitting name
2018-10-10 12:27:44 +11:00
David Marcec
a4412c8e22 Fixed incorrect types for svcBreak
svcBreak reason should be a u32, not a u64.
2018-10-10 12:23:50 +11:00
bunnei
fe16905de1
Merge pull request #1459 from ogniK5377/break
svcBreak, Signalling to the debugger should not kill execution
2018-10-09 16:57:37 -04:00
bunnei
89939be9e6
Merge pull request #1465 from lioncash/telemetry
telemetry_session: Minor miscellaneous changes
2018-10-09 16:56:56 -04:00
Lioncash
e3b4d31f4e telemetry_session: Remove doxygen comment for a non-existent parameter
There's no "func" parameter, so this can just be removed.
2018-10-09 14:52:10 -04:00
Lioncash
8aa4889e76 telemetry_session: Add missing includes
Prevents potential compilation issues in the future by including missing
headers for certain functions and types.
2018-10-09 14:51:39 -04:00
Lioncash
1964f4bbb3 telemetry_session: Remove unimplemented FinalizeAsyncJob prototype
This isn't implemented anywhere, so it can just be removed.
2018-10-09 14:46:31 -04:00
Lioncash
8723cc8798 telemetry_session: Use a std::array in GenerateTelemetryId()
We don't need to potentially heap-allocate a std::string instance here,
given the data is known ahead of time. We can just place it within an
array and pass this to the mbedtls functions.
2018-10-09 14:46:26 -04:00
Lioncash
6636f3ff47 patch_manager: Return a std::unique_ptr from ParseControlNCA() and GetControlMetadata() instead of a std::shared_ptr
Neither of these functions require the use of shared ownership of the
returned pointer. This makes it more difficult to create reference
cycles with, and makes the interface more generic, as std::shared_ptr
instances can be created from a std::unique_ptr, but the vice-versa
isn't possible. This also alters relevant functions to take NCA
arguments by const reference rather than a const reference to a
std::shared_ptr. These functions don't alter the ownership of the memory
used by the NCA instance, so we can make the interface more generic by
not assuming anything about the type of smart pointer the NCA is
contained within and make it the caller's responsibility to ensure the
supplied NCA is valid.
2018-10-09 14:38:03 -04:00
Lioncash
465175cdf5 ips_layer: Avoid constructing std::vector instances where not necessary
We can just compare the existing std::vector instance with a constexpr
std::array containing the desired match. This is lighter resource-wise,
as we don't need to allocate on the heap.
2018-10-09 14:10:22 -04:00
Lioncash
9ff743bc0a ips_layer: Remove unnecessary explicit std::pair constructor in std::array
Makes the layout of the array consistent, by making all elements match,
instead of special-casing the first one.
2018-10-09 14:10:22 -04:00
Lioncash
f7d2889fb4 ips_layer: Add missing includes
Adds missing includes to prevent potential compilation issues in the
future. Also moves the definition of a struct into the cpp file, so that
some includes don't need to be introduced within the header.
2018-10-09 14:10:13 -04:00
Lioncash
93ac8d0fea ips_layer: std::move data within PatchIPS() and Apply()
We don't need to make a copy of the read data, so we can std::move it
into the make_shared call here.
2018-10-09 14:06:44 -04:00
Lioncash
567e818440 ips_layer: Silence truncation and conversion warnings
Makes type conversions explicit to avoid compiler warnings.
2018-10-09 13:18:23 -04:00
David Marcec
f5631e78d1 Added bitfield instead of manually checking if the bit is set 2018-10-09 12:11:14 +11:00
David Marcec
af3ba94b2a Actual kill execution when the bit isn't set, not the other way around 2018-10-09 11:14:48 +11:00
David Marcec
c50f66a8eb svcBreak, Signalling to the debugger should not kill execution
When loading NROs, svcBreak is called to signal to the debugger that a new "module" is loaded. As no debugger is technically attached we shouldn't be killing the programs execution.
2018-10-09 11:10:30 +11:00
bunnei
561d79e034
Merge pull request #1423 from DarkLordZach/romfs-file-exts
fsmitm_romfsbuild: Add support for stubbing and IPS patches in LFS
2018-10-08 12:31:27 -04:00
bunnei
6b48ba5271
Merge pull request #1424 from DarkLordZach/ips-witch
ips_layer: Add support for IPSwitch executable patches
2018-10-08 12:30:33 -04:00
bunnei
fd891ee9c0
Merge pull request #1456 from ogniK5377/aoc-u-fixups
Fixed assertion due to CountAddOnContent & Casting warnings
2018-10-08 01:21:05 -04:00
bunnei
3f1f82a8c4
Merge pull request #1457 from ogniK5377/unmap-buffer
Unmapping an unmapped buffer should succeed
2018-10-08 01:20:18 -04:00
David Marcec
c5c184246d Unmapping an unmapped buffer should succeed
Hardware tests show that trying to unmap an unmapped buffer already should always succeed. Hardware test was tested up to 32 iterations of attempting to unmap
2018-10-08 13:26:48 +11:00