August 2016 - Wine-devel - list.winehq.org

Re: [PATCH] wined3d: Fix ATI1 buffer size, try 2
by Henri Verbeet 24 Aug '16

24 Aug '16

On 22 August 2016 at 18:53, Patrick Rudolph <siro(a)das-labor.org> wrote: > Apitrace crashes on WINE when trying to playback a trace containing ATI1 textures. > > Add a tests to show that "unknown" formats like ATI1 are backed by a buffer > of size pitch * height. Without the attaced fix the tests will crash. > > Calculate correct buffer size in wined3d_format_calculate_pitch for formats that > have WINED3DFMT_FLAG_BROKEN_PITCH flag set. > Remove no longer needed check in texture_resource_sub_resource_map. > > Signed-off-by: Patrick Rudolph <siro(a)das-labor.org> > --- > dlls/d3d9/tests/device.c | 8 ++++++++ > dlls/wined3d/texture.c | 10 +--------- > dlls/wined3d/utils.c | 9 ++++++++- > 3 files changed, 17 insertions(+), 10 deletions(-) > This breaks the tests. For example: ../../../tools/runtest -q -P wine -T ../../.. -M d3d9.dll -p d3d9_test.exe.so visual && touch visual.ok visual.c:9512: Tests skipped: Card has unconditional NP2 support, skipping conditional NP2 tests visual.c:13651: Tests skipped: No NVDB (depth bounds test) support, skipping tests. visual.c:17111: Tests skipped: DXT5 volume textures are not supported, skipping test. visual.c:17966: Test failed: Got unexpected color 0x008080ff, case 2. visual.c:17966: Test failed: Got unexpected color 0x008080ff, case 3. visual.c:19801: Tests skipped: Volume ATI2N textures are not supported, skipping some tests. Makefile:276: recipe for target 'visual.ok' failed make: *** [visual.ok] Error 2

1 0

Re: [PATCH 2/7] d3d8/tests: Prefer higher resolutions in test_reset().
by Henri Verbeet 24 Aug '16

24 Aug '16

On 22 August 2016 at 14:53, Józef Kucia <jkucia(a)codeweavers.com> wrote: > @@ -21,6 +21,8 @@ > * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA > */ > > +#include <stdlib.h> > + > #define WINVER 0x0602 /* for CURSOR_SUPPRESSED */ This causes warnings from make crosstest for me: device.c:26:0: warning: "WINVER" redefined #define WINVER 0x0602 /* for CURSOR_SUPPRESSED */ ^ In file included from /usr/share/mingw-w64/include/crtdefs.h:10:0, from /usr/share/mingw-w64/include/stdlib.h:9, from device.c:24: /usr/share/mingw-w64/include/_mingw.h:225:0: note: this is the location of the previous definition #define WINVER 0x0502 ^

1 0

Re: ntoskrnl.exe: Implement ZwLoadDriver and ZwUnloadDriver based on services API.
by Sebastian Lackner 23 Aug '16

23 Aug '16

On 22.08.2016 16:12, Sebastian Lackner wrote: > Signed-off-by: Sebastian Lackner <sebastian(a)fds-team.de> > --- > > This hides the complicated services API behind some easy to use functions :) > The timeout are a bit arbitrary and can be adjusted if they do not work well > enough in practice. Based on Arics feedback this solution is confirmed to work > for his purpose. > > dlls/ntoskrnl.exe/Makefile.in | 1 > dlls/ntoskrnl.exe/ntoskrnl.c | 164 ++++++++++++++++++++++++++++++++++++ > dlls/ntoskrnl.exe/ntoskrnl.exe.spec | 4 > 3 files changed, 167 insertions(+), 2 deletions(-) Aric already signed off, but please ignore this patch. I am planning to send a new version shortly. The functions should also strip off the registry prefix.

1 0

Re: Adding binfmt configuration to official Wine packages
by Alex Henrie 23 Aug '16

23 Aug '16

2016-08-23 4:45 GMT-06:00 Rosanne DiMesio <dimesio(a)earthlink.net>: > On Mon, 22 Aug 2016 20:26:51 -0600 > Alex Henrie <alexhenrie24(a)gmail.com> wrote: > >> In fact, it's easy to >> forget that you have to explicitly give execution permissions to a >> downloaded file before Linux will let you execute it. If the file is >> on a USB drive, unless the drive has a Linux filesystem, the binary >> must be moved off of the drive before it can receive execution >> permissions. >> > > I have portable Firefox on a FAT32 formatted flash drive and Wine is able to run it. Wine will run any EXE as long as you start it (or your desktop environment starts it) with `wine foobar.exe`. If you start it with `./foobar.exe` instead, the file must be executable and binfmt must be set up correctly. Interestingly, it looks like Arch Linux and Ubuntu have both started to mount FAT32 filesystems with the "showexec" option. This makes EXE files on USB drives executable automatically. (It doesn't look very smart, it just turns on executability if the filename ends in ".exe".) So I stand corrected; you're right that it's no longer necessary to change the mount options or copy the EXE to another filesystem. Still, no desktop environment is going to execute files off of a USB drive without asking. -Alex

1 0

Adding binfmt configuration to official Wine packages
by Bruno Jesus 23 Aug '16

23 Aug '16

Hi, I would like to please call for attention on bug https://bugs.winehq.org/show_bug.cgi?id=39884 As far as I understand there are only benefits to users if we allow .exe files to run through binfmt, would it pose any kind of problem to add this configuration to our packages? Best wishes, Bruno

6 19

Re: [PATCH] Get the position from the driver on every GetCursorPos run
by Alexandre Julliard 23 Aug '16

23 Aug '16

Max Qian <public(a)maxqia.com> writes: > This reverts the GetCursorPos changes in commits 6576703 and f1a3480 to fix bug 24951. This is going to reintroduce the performance issues that the changes were supposed to fix. Doing a server round-trip on every GetCursorPos gives bad results in many apps. You'll need to find a fix that doesn't require this. -- Alexandre Julliard julliard(a)winehq.org

2 2

Re: [resend PATCH] ntdll: Tolerate null handle in DeregisterWait
by Bruno Jesus 23 Aug '16

23 Aug '16

On Fri, Aug 19, 2016 at 6:24 PM, Keno Fischer <keno(a)juliacomputing.com> wrote: Hi, is the subject supposed to be Tolerate NULL handle in UnregisterWait? > diff --git a/dlls/kernel32/tests/thread.c b/dlls/kernel32/tests/thread.c > index a9a04df..44c6278 100644 > --- a/dlls/kernel32/tests/thread.c > +++ b/dlls/kernel32/tests/thread.c > @@ -1245,6 +1245,11 @@ static void test_RegisterWaitForSingleObject(void) > > ret = pUnregisterWait(wait_handle); > ok(ret, "UnregisterWait failed with error %d\n", GetLastError()); > + > + /* Test UnregisterWait null argument case */ > + ret = pUnregisterWait(0); > + ok(!ret && GetLastError() == ERROR_INVALID_HANDLE, > + "UnregisterWait failed with unexpected error %d\n", GetLastError()); > } Please test ret and GetLastError separately as the rest of the code does most times. If you see other tests that test GetLastError you will notice that they call SetLastError before calling the tested function to ensure the error is not from a previous test. So add a SetLastError(0xdeadbeef) before the call to pUnregisterWait to ensure the ERROR_INVALID_HANDLE returned is from UnregisterWait. Also for consistency why not pUnregisterWait(NULL) instead of (0)?

3 2

Re: [PATCH] dinput: 0 is a valid fd (Coverity)
by Bruno Jesus 23 Aug '16

23 Aug '16

On Mon, Aug 22, 2016 at 6:19 PM, Marcus Meissner <marcus(a)jet.franken.de> wrote: > 1371727 Resource leak > (will not happen in Wine, but it is valid) > > Signed-off-by: Marcus Meissner <marcus(a)jet.franken.de> This patch is superseeded by http://source.winehq.org/patches/data/125859

1 0

Re: dinput: Check the return of open() properly (Coverity)
by Bruno Jesus 22 Aug '16

22 Aug '16

On Mon, Aug 22, 2016 at 6:07 PM, Bruno Jesus <00cpxxx(a)gmail.com> wrote: > Signed-off-by: Bruno Jesus <00cpxxx(a)gmail.com> This patch is superseeded by http://source.winehq.org/patches/data/125859

1 0

Re: [PATCH] wined3d: Fix ATI1 buffer size, try 2
by Marvin 22 Aug '16

22 Aug '16

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=25322 Your paranoid android. === w864 (64 bit device) === device.c:4392: Test failed: cw is 0xf60, expected 0x7f. device.c:4401: Test failed: Callback cw is 0xf60, expected 0x7f. === w1064 (64 bit device) === device.c:4392: Test failed: cw is 0xf60, expected 0x7f. device.c:4401: Test failed: Callback cw is 0xf60, expected 0x7f.

1 0