Optimize for speed and memory.
JIRA issue: CORE-13950
CDirectoryList class exists just for remembering which file item is a directory or
not, in order to notify the filesystem item changes. This information can
become a tree data structure.
- Add CFSPathIterator and CFSNode helper classes.
- CFSNode is a class for tree nodes.
- Re-implement CDirectoryList class by using tree nodes.
- Delete CDirectoryItem class.
Fix playing wave header multiple times in case the caller requests to do it.
In Windows, it is supported by WHDR_BEGINLOOP and WHDR_ENDLOOP flags (specified together) and dwLoops member of WAVEHDR structure (ususally set to 0xFFFFFFFF (INFINITE constant)).
- Check whenther WHDR_BEGINLOOP | WHDR_ENDLOOP flags are set by the caller.
- If they are, get the amount of times to play the header from WAVEHDR.dwLoops.
- Perform wave header competion only when the loop count is equal to zero. Otherwise, don't do it.
- When the header is entirely committed, in case completion is not needed, reset committed bytes count to the starting value to allow the header play again.
- Decrement loop count in case it isn't set to INFINITE, to mark loop as played correctly. When this count becomes zero, then the playback is finished.
- Get rid from SOUND_OVERLAPPED.PerformCompletion member. Use SOUND_DEVICE_INSTANCE.LoopsRemaining == 0 condition instead.
- Do this only for WaveOut devices, since MSDN states it works only with output buffers. Hence, there is nothing changed for WaveIn.
- Update an appropriate statement about unimplemented functionality from mmebuddy notes.
TODO: handle the case when multiple headers are requested to be looped (WHDR_BEGINLOOP and WHDR_ENDLOOP are set separatedly: 1st flag - in the 1st header, and 2nd in the last header). Currently, only looping a single wave header is supported.
This fixes the playback in the apps those request looped wave playback of some audio data (e. g., BRD Demo app, which did play its sound only first 500 ms before, now it plays endlessly until closing it manually).
CORE-10118
Make the testcase more stable and faster. Reduce code.
JIRA issue: CORE-13950
- Use WM_COPYDATA message to communicate with shell32_apitest_sub.exe.
- Use concept of "stage" and "step".
- Do multiple checks in one step.
- Flush notification by using (SHCNF_FLUSH | SHCNF_FLUSHNOWAIT) flags.
- Don't wait 1 second for each processor
- Use AFFINITY_MASK
- Use Prcb->Number, which is more future proof than KeGetCurrentProcessorNumber
- Improve for loop
Implementing missing features...
JIRA issue: CORE-13950
- Add WM_DEVICECHANGE message handler in the shell window to detect
DBT_DEVICEARRIVAL and DBT_DEVICEREMOVECOMPLETE.
- Use GetLogicalDrives function to detect drives.
- Use SHChangeNotify to send SHCNE_DRIVEADD and
SHCNE_DRIVEREMOVED notifications.
- Modify CDefView::OnChangeNotify.
ShellExecute on a .zip file fails because the command in the registry
uses rundll32 that passes in an ANSI string, which is the default for
exported functions without a W suffix.
- This should also fix RAPPS not being able to open downloaded .zip files
- I made it use the full path to Explorer for good measure
Fixes regression introduced by 0.4.15-dev-6343-g bf2cec186c (#5411)
CORE-19506
And deprecate corresponding boot types "Drive" and "Partition".
These are replaced by the more general "BootSector" boot type.
Finish the unification of the code, started in commit ff85aa0c3,
that loads and boots disk MBR, partition VBR or boot sector in file.
A "WarnDeprecated()" helper is added to warn the user about the
deprecated features, and to inform them to adjust their FREELDR.INI
file in accordance.
In addition, bump FreeLoader version to 3.2 (at last!): a lot of
features have been added or deprecated since its last release.
Calculate size of the additinal data based on the format tag specified by the caller.
For PCM wave format, it should be zero, and for all others,
it should be taken from cbSize member of WAVEFORMATEX structure.
- Initialize data size for each format appropriately.
- Append it to the total pin connect allocation size, to allocate enough memory for it.
- Use it as additional data size for requested wave format.
- Append it to the total size of data format.
- Add SAL2 annotations for MMixerInitializeDataFormat function.
Fixes two crashes in winmm:capture and winmm:wave winetests, which have been introduced by 12b3272d9e.
So this is an addendum to that commit.
CORE-19514
[FREELDR] Add "WindowsVista" boot type
[FREELDR] Set GDT correctly for Vista
[FREELDR] Map first page of memory, this is an observed behavior, and
also increases stability boot Checked windows 2003 SP2 ntoskrnl with
freeloader.
[SDK] Don't assert on big keys in bootloader
Co-authored-by: Justin Miller <justin.miller@reactos.org>
CORE-19191
Reproduce the same behaviour of the calculator for Windows XP when an error condition is met.
When the result is NaN and it cannot be printed, the error condition is cleared by pressing the CLEAR button.
The code itself isn't wrong but we're entirely missing the logic needed
to make it work. This leads to spontaneous crashes with video drivers and
various unity games.
After a back and fourth investigation I've just decided to disable pdb generation for this ROSAPP for now.
We tried taking the code apart a bit and only building parts but alas it didnt yield any sucess.
disabling the AV also didn't help.
JIRA issue: CORE-19516
- Add ClampSplitBarX helper function to delete
duplicated code.
- Use a signed value in WM_LBUTTONUP handling.
- Use ClampSplitBarX in ResizeWnd function,
WM_LBUTTONDOWN, WM_MOUSEMOVE, and
WM_LBUTTONUP handling.
- UiMessageBox(): Enlarge the default buffer used to printf msgbox strings.
- TuiUpdateDateTime(): When displaying the time, don't pad too much
with spaces on the left.
- TuiDrawShadow():
* Pre-calculate whether we need to show the right/bottom shadows,
and whether the right-shadow has a double width.
* Cap the right and bottom upper-bound coordinates before looping,
in order to avoid buffer overflows if the given coordinates go
beyond the screen.
- TuiDrawMsgBoxCommon(): Improve how the message box centering
calculations are done:
* When the "full-UI" is used (and center-menu is used), try to
center the message box in the space between the header and
the status bar.
* Ensure the top-left box corner is inside the screen, in case
the calculated message box borders go off-screen.
- TuiCalcMenuBoxSize():
* Uniformize the way the menu box coordinates are calculated.
* Reduce the space between menu box margins and the longest item.
* Ensure the top-left menu corner is inside the screen, in case
the calculated menu box borders go off-screen.
A NULL pointer (not necessarily the terminating one) is a valid entry
in the Argv argument vector, according to the ARC specification
(Section 4.4 "Loaded-Program Conventions").
Thus, such pointer needs to be ignored when searching over the
argument vector.
Addendum to commits d05be0da3 and bd451f240.
This is required for POSIX compliance, which the ARC specification obeys.
See Section 2.1.2.2.1 (ANSI X3.159-1989) or 5.1.2.2.1 (ISO/IEC 9899:x)
"Program startup":
```
If they are declared, the parameters to the main function shall obey
the following constraints:
— The value of argc shall be nonnegative.
— argv[argc] shall be a null pointer.
[...]
```
See also https://pubs.opengroup.org/onlinepubs/9699919799/functions/exec.html
This saves a few bytes in the binary, e.g.
comdlg32.dll master 0.4.15-dev-7887-g64a59a1 RosBEWin2.2.2 GCC8.4.0 dbg x86 shrinks
from 1.011.712 bytes to 1.009.664 bytes.
Furthermore it syncs the translations at this line, because pt-PT.rc already applied such a tweak.
Co-authored-by: Katayama Hirofumi MZ <katayama.hirofumi.mz@gmail.com>
fixes CORE-18807 'Our explorer.exe ATL-asserts when executed on XPSP3'
which most likely got unhidden when we enabled the ATL-assertions in
0.4.15-dev-5023-g 55343f04b0
The registers that are saved/restored in the trap / exception frame need to be consistent between all entry/exit points as well as the functions that convert between trap/exception frame and context. The trap frame contains only the non-volatile registers and rbp, the rest is saved in the exception frame.
The previous code didn't save rbp in the syscall handler, which led to it being clobbered when exiting though KiServiceExit2 rather than returning back to the syscall exit path. Also KeContextToTrapFrame would use rbx, rsi and rdi from the trap frame, which wouldn't be saved there by the syscall handler.
