[WS2_32_APITESTS] Add tests for SIO_GET_INTERFACE_LIST. Brought to you by Andreas Maier. Thanks.

Minor modifications by me on comparison and expected flags.


    • -0
    • +232
[ACPPAGE] Remove unneeded initialization. Should make the x64 bot a tiny bit happier. Spotted by Mark Jansen.
[PSDK] Add missing RpcBindingServerFromClient.
    • -0
    • +2
[URLMON] Stub and export HlinkGoForward(). CORE-12196
[RPCRT4_WINETEST] Addendum to r73006.
    • -0
    • +40
[RPCRT4] Import Wine commit 01290cd by Colin and Christoph: Implement RpcBindingServerFromClient and populate NetworkAddr for each transport.
    • -2
    • +32
    • -3
    • +17

Import Wine commit:

- d9c9d9b10f58844a780d8bfc9ed1b6713e37d487, Implement connected resources enumeration.


    • -85
    • +71

Build with "ninja -k 0", so that Ninja continues with other source files after building one of them failed. The Buildslave log will then show all source files with compilation errors in a single run.

Suggested by Hermes

[WINETESTS] Rename generated_i386.c to generated.c, adjust the CMakeLists.txt accordingly.
    • -7183
    • +0
    • -3925
    • +0
    • -2669
    • +0
    • -13560
    • +0
    • -1866
    • +0
  1. … 12 more files in changeset.
[SDK] Fix our Delayload notify hook for MSVC x64
    • -0
    • +7
[APPHELP_APITEST] This should not be here.
    • -2
    • +0
[KS]: Fix annotations, etc.
    • -16
    • +12

- add missing definition

    • -0
    • +41

- implement support for KSPROPSETID_General

    • -3
    • +3
[CMD]: Documentation improvement: we actually support two types of filename completions, windows-cmd like (the default one), and "bash-style" one (that is available through a compile switch). I prefer calling it "bash-style" over "unix-style" because this style of completion looks like the one from the GNU ReadLine library, that is used by Bash, which is not exclusive to unix per se...
  1. … 7 more files in changeset.
[KERNEL32]: Mark Get/SetEnvironmentVariableA/W as hotpatchable since it's needed for Clink 0.4.8 (<rant>btw, if GCC was able to set exported stdcall functions as hotpatchable at a global level, as MSVC does, those "DECLSPEC_HOTPATCH" hacks would not be necessary! </rant>)

- Localization of output based on a patch by Lee Schroeder, but where I haven't modified yet the output localization of errors.

- Dynamically underline the headers of the device status reports (by me).


    • -0
    • +47

- Rearrange ResizeTextConsole so that it correctly computes the Top & Bottom parts of the window size rectangle;

- Fix its return type;

- Display an error if we failed resizing the console;

- Clear the screen before resizing the console, as the real MODE.COM does.

- Fix a parsing error in SetConsoleStateOld and in SetConsoleCPState.

[MODE]: Code style changes + remove unused parameter for ResizeTextConsole.

- Call TermGetLargestConsoleWindowSize to obtain the largest console window size allowed on the system, and use it for GetConsoleScreenBufferInfo and for SetConsoleWindowInfo too, where it is used to check if the given user window size is not too large.

- Improve GuiGetLargestConsoleWindowSize for multi-monitor situations.

- Remove the redundant definition of GetScreenBufferSizeUnits in guiterm.c (it already exists in conwnd.c).

[CONSRV]: Code style changes only.

Replace error codes that don't even exist in MS headers (but they are referenced on MSDN?!)

Replace their usage in code


- Let the zero page thread decide itself whether it is active or not.

- Raise the low memory limit for our good old balancer

- Allow the balancer thread to wait for a page to be freed, with a lower limit than for "regular" page faults

- Let ARM3 notify RosMm when a page gets freed

CORE-12047 #comment Whole patch got in in r72988, 72989 and 72990

    • -0
    • +2
    • -2
    • +10
    • -21
    • +46

Miscellaneous fixes for legacy Mm section implementation

- Always allocate a private page for IMAGE_SCN_CNT_UNINITIALIZED_DATA

- Make sure a shared page is present before writing on a COW mapping and making a private copy.

- Fix a race condition : when paging out a file section, old Mm lists all of the process maps, removing them one after the other and lowering the page reference count in the process. Sometimes a page fault occur in the process, the mapping is added, but the page refcount is not bumped because it requires locking the corresponding segment. Manage page refcount under segment lock.


    • -106
    • +102

- Do not allocate zeroed pagefile entries for legacy Mm



- Acquire Paging IO Resource shared when performing paged writes

Greenlighted by Pierre and tests.

CORE-12184 #resolve


Make DECLARE_CONST_UNICODE_STRING() work with something else than MSVC

    • -0
    • +10
[KERNEL32_APITEST]: SetConsoleWindowInfo test:

- Improve the checks done for the tests;

- Add a new test (Test7) that tests the allowed console window sizes wrt. the csbi.dwMaximumWindowSize size.


- highlight excessive machine reboots as requested by Thomas


[CONSRV]: Fixes for SetConsoleWindowInfo, where one notices that again the MSDN documentation on SetConsoleWindowInfo is partially wrong.

This makes all the kernel32_apitest:SetConsoleWindowInfo tests pass now.

Also, notify the console window about the size change.