July 2010 - Wine-devel - list.winehq.org

nightly kernel testbot up
by Austin English 10 Jul '10

10 Jul '10

Howdy, I wrote a quick script to test the latest wine and linux kernel, to hopefully prevent more regressions like the WoW login bug. The test machine is in a virtualbox vm, so of course graphics tests won't be useful, but the rest should be... There are already a few results up (in the future, it'll be one per day, but I had to debug). tag is 'ae-lucid-nighkernel': http://test.winehq.org/data/1cd170fd2e0e7df5083e7c66f8422783c57ed256/wine_a… there are a few test failures that look interesting: kernel32:debugger kernel32:process kernel32:thread kernel32:toolhelp kernel32:virtual ntdll:exception ntdll:info psapi:psapi_main rpcrt4:rpc rpcrt4:server For the curious, script is at http://winezeug.googlecode.com/svn/trunk/build_script/auto-update-winetest.… -- -Austin

1 1

Re: git rebase to _insert_ commits?
by Mike Kaplinskiy 10 Jul '10

10 Jul '10

You can move commits around in an interactive rebase. Just create an empty commit, move it into place with rebase and edit it. Sorry for top post, mobile gmail doesn't let you bottom post. On Jul 9, 2010 8:56 PM, "Misha Koshelev" <misha680(a)gmail.com> wrote: Dear All: I am still learning git and it seems that my solution does not quite work. Specifically, my idea was to use: git rebase -i upstream/master Mark the commit _after_ which I'd like to insert a commit with edit. Then, when I am at that commit, I do: git commit --amend -a save commit Then, _before_ doing git rebase --continue, I edit files and again do: git commit -a -n Finally I do: git rebase --continue The problem is that this seems to squish further commits, and leads to problems which I am now trying to untangle. However, as I still cannot insert a commit, I am not sure how to proceed, except by creating a whole new repository. Any ideas/hints? Thank you Misha

2 2

re: [PATCH] Fix EndDialog behaviour for OrCad.
by Dan Kegel 09 Jul '10

09 Jul '10

Thanks for sending that patch. It seems your editor stripped whitespace from the ends of lines, causing three whitespace-only hunks on the end of your patch. Please don't do that; Wine deveopers frown on unrelated whitespace changes. You might have better luck getting your change in if you include a test. Even then, we're in code freeze, your fix might have to wait until 1.2 is released.

3 5

Can Relay miss some calls in initialization stage?
by Hayan Lee 09 Jul '10

09 Jul '10

Hi, all I'm working on relay layer in WINE. When I turn on WINEDEBUG=+relay, relay looks to miss some calls. When is the relay layer loaded? Before dll loaded or after? If WINE does something in its initialization step, relay can catch those or not? Thanks in advance. Hayan

3 3

Portuguese (Brazil) translation of Wine at 100%
by Paul Vriens 09 Jul '10

09 Jul '10

Hi, Many thanks to Gustavo who brought the Portuguese (Brazil) translation to a 100%. The list of 100% translations is slowly growing, we now have 7 (not counting the ones that are mostly inherited from the neutral ones)!! -- Cheers, Paul.

4 4

WINED3DUSAGE_MASK?
by Misha Koshelev 09 Jul '10

09 Jul '10

Dear All: Sorry to bother - I just ran into something quite curious in terms of a Wine-specific feature that is certainly not matched in Windows XP and was wondering what its purpose was... Also, should I use todo_wine to test for this, or is there a better way since this seems something that obviously we're not trying to "make conform" in Wine but rather something that seems to be an intended feature (please correct me if I am incorrect). Specifically, this is a usage flag for index buffer creation that seems to have been introduced here: commit 3529560cd7a4f752596e80d4084e9527b348eb22 Author: Stefan Dösinger <stefandoesinger(a)gmx.at> Date: Thu Oct 5 17:56:54 2006 +0200 wined3d: Add WINED3DUSAGE_OVERLAY. diff --git a/dlls/d3d9/indexbuffer.c b/dlls/d3d9/indexbuffer.c index fc85357..269552a 100644 --- a/dlls/d3d9/indexbuffer.c +++ b/dlls/d3d9/indexbuffer.c @@ -175,7 +175,7 @@ HRESULT WINAPI IDirect3DDevice9Impl_CreateIndexBuffer(LPDIRECT3DDEVICE9 iface, object->lpVtbl = &Direct3DIndexBuffer9_Vtbl; object->ref = 1; TRACE("Calling wined3d create index buffer\n"); - hrc = IWineD3DDevice_CreateIndexBuffer(This->WineD3DDevice, Length, Usage, Format, (WINED3DPOOL) Pool, &object->wineD3DIndexBuffer, pSharedHandle, (IUnknown *)object); + hrc = IWineD3DDevice_CreateIndexBuffer(This->WineD3DDevice, Length, Usage & WINED3DUSAGE_MASK, Format, (WINED3DPOOL) Pool, &object->wineD3DIndexBuffer, pSharedHandle, (IUnknown *)object); if (hrc != D3D_OK) { /* free up object */ Thank you Misha

2 2

Re: [1/8] msi/tests: Run the automation tests in silent mode to avoid a dialog.
by testbot＠testbot.winehq.org 09 Jul '10

09 Jul '10

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 http://testbot.winehq.org/JobDetails.pl?Key=3251 Your paranoid android. === W7PROX64 (64 bit automation) === automation.c:2487: Test failed: Expected imaname, got automation.c:2504: Test failed: Expected imaname, got

2 1

Re: kernel32: Loosen trailing \ requirement on GetVolumeInformationAW.
by Alexandre Julliard 09 Jul '10

09 Jul '10

<Joerg-Cyril.Hoehle(a)t-systems.com> writes: > The tests prove that not all calls without a trailing slash must fail. > So until we know more precisely the exact conditions, I feel it's > better to revert to the old behaviour that allowed the copy > protection scheme to work. That's only moving the bug around. We have tests for the current behavior, any further improvement must keep the existing tests working. -- Alexandre Julliard julliard(a)winehq.org

2 4

TestBot: new VMs and components
by Greg Geldorp 08 Jul '10

08 Jul '10

Upon request, I've added a couple of VMs for your testing pleasure: WXPPROARSP3 - Windows XP Pro Arabic SP3 WXPPROJASP3 - Windows XP Pro Japanese SP3 WXPPRORUSP3 - Windows XP Pro Russian SP3 WXPPROZHSP3 - Windows XP Pro Chinese simplified SP3 I must say, installing these VMs with their non-Western scripts was quite a challenge :-). I'm also in the process of adding some components to all applicable VMs: Microsoft Visual C++ 2005 Runtime Microsoft Visual C++ 2008 Runtime DirectX End-User Runtimes (June 2010) It will take some time before these components are available everywhere. If you'd like to see more components added please let me know. Ge.

1 0

Re: shell32: Add stub for SHGetPropertyStoreFromParsingName
by Hans Leidekker 08 Jul '10

08 Jul '10

On Thu, 2010-07-08 at 21:19 +0200, André Hentschel wrote: > +@ stdcall SHGetPropertyStoreFromParsingName(str ptr long ptr ptr) The first parameter should be wstr.

2 1