[PATCH 0/1] MR7445: ntdll/tests: Skip FileFsFullSizeInformationEx test on older windows.

[Bernhard Übelacker (＠bernhardu)]

[Testbot shows failures](https://test.winehq.org/data/patterns-tb-win.html#ntdll:file) since a few days in ntdll:file on older windows versions.

These failures seem to be related to 37c5fa3d and f15bd29c (CC: @cmccarthy).

This patch just adds `STATUS_INVALID_INFO_CLASS` and moves the `ok` statement below the `win_skip`.

[PATCH 1/1] ntdll/tests: Skip FileFsFullSizeInformationEx test on older windows.

[Bernhard Übelacker]

From: Bernhard Übelacker

---
 dlls/ntdll/tests/file.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/dlls/ntdll/tests/file.c b/dlls/ntdll/tests/file.c
index cc0afdbdb64..34962db9426 100644
--- a/dlls/ntdll/tests/file.c
+++ b/dlls/ntdll/tests/file.c
@@ -1388,14 +1388,13 @@ static void test_file_full_size_information(void)

/* FileFsFullSizeInformationEx is supported on Windows 10 build 1809 and later */
     res = pNtQueryVolumeInformationFile(h, &io, &ffsie, sizeof ffsie, FileFsFullSizeInformationEx);
-    ok(res == STATUS_INVALID_PARAMETER || res == STATUS_SUCCESS, "cannot get attributes, res %lx\n", res);
-
-    if (res == STATUS_NOT_IMPLEMENTED || res == STATUS_INVALID_PARAMETER)
+    if (res == STATUS_NOT_IMPLEMENTED || res == STATUS_INVALID_PARAMETER || res == STATUS_INVALID_INFO_CLASS)
     {
         win_skip( "FileFsFullSizeInformationEx not supported.\n" );
         CloseHandle( h );
         return;
     }
+    ok(res == STATUS_SUCCESS, "cannot get attributes, res %lx\n", res);

expected = ffsie.ActualAvailableAllocationUnits + ffsie.ActualPoolUnavailableAllocationUnits
             + ffsie.UsedAllocationUnits + ffsie.TotalReservedAllocationUnits;

[Jinoh Kang (＠iamahuman)]

Jinoh Kang (@iamahuman) commented about dlls/ntdll/tests/file.c:
>  /* FileFsFullSizeInformationEx is supported on Windows 10 build 1809 and later */
>  res = pNtQueryVolumeInformationFile(h, &io, &ffsie, sizeof ffsie, FileFsFullSizeInformationEx);
> - ok(res == STATUS_INVALID_PARAMETER || res == STATUS_SUCCESS, "cannot get attributes, res %lx\n", res);
> -
> - if (res == STATUS_NOT_IMPLEMENTED || res == STATUS_INVALID_PARAMETER)
> - if (res == STATUS_NOT_IMPLEMENTED || res == STATUS_INVALID_PARAMETER || res == STATUS_INVALID_INFO_CLASS)

Doesn't `STATUS_INVALID_PARAMETER` still belong to the ok() statement?

[Bernhard Übelacker (＠bernhardu)]

On Fri Feb 28 16:13:14 2025 +0000, Jinoh Kang wrote:
> Doesn't `STATUS_INVALID_PARAMETER` still belong to the ok() statement?

As there is a return after the skip I removed it from the ok statement.

Is there a downside to moving the ok below the skip?

[Jinoh Kang (＠iamahuman)]

On Fri Feb 28 16:20:29 2025 +0000, Bernhard Übelacker wrote:
> As there is a return after the skip I removed it from the ok statement.
> Is there a downside to moving the ok below the skip?

I'm fine with moving the ok. I'm asking for the ok condition to include `res == STATUS_INVALID_PARAMETER`, just like it has been before, instead of in the skip condition.

[Bernhard Übelacker (＠bernhardu)]

On Fri Feb 28 16:25:10 2025 +0000, Jinoh Kang wrote:
> I'm fine with moving the ok. I'm asking for the ok condition to include
> `res == STATUS_INVALID_PARAMETER`, just like it has been before, instead
> of in the skip condition.

I am not sure if I understand you right. Before this patch `res == STATUS_INVALID_PARAMETER` was part of the ok and the if condition around the skip. After my patch the ok should never get reached with `res == STATUS_INVALID_PARAMETER`, so isn't checking for it superfluous?

I am happy with either way, should I add it again?