June 2008 - Wine-devel - list.winehq.org

Re: ddraw: Reset the viewport data before each test
by Paul Vriens 15 Jun '08

15 Jun '08

James Hawkins wrote: > Hi, > > Fixes bug 13643 and the ddraw:d3d tests. > > Changelog: > * Reset the viewport data before each test. > > dlls/ddraw/tests/d3d.c | 31 +++++++++++++++++++++------------ > 1 files changed, 20 insertions(+), 11 deletions(-) > > > > ------------------------------------------------------------------------ > > Can you explain why we have to do this? The fact that only Paul Millar's winetest seems to suffer from this is a bit odd, isn't it? I'm just wondering if there are more issues like this that introduces failures in the tests or even worse lets tests succeed. -- Cheers, Paul.

3 2

d3dx9_28: add version resource
by David Adam 15 Jun '08

15 Jun '08

Hello, maybe you should do this for every d3dx9_xx dll, not only the d3dx9_28 one. David

1 0

Unhandled page fault on prefix creation on osx 10.5
by Cesar Izurieta 15 Jun '08

15 Jun '08

I just deleted my .wine directory and run wineboot. The .wine directory is recreated but I get: wine: created the configuration directory '/Users/cesar/.wine' wine: Unhandled page fault on read access to 0x008cc65f at address 0xffff0eb8 (thread 000b), starting debugger... couldn't load main module (2) Unhandled exception: page fault on read access to 0x008cc65f in 32-bit code (0xffff0eb8). And a lot of dumps and backtraces. I did a regression testing and found out that 58913b439b797462a70794774b87e63283109617 commit is what is causing this problem. Reverting this commit everything works perfectly. More information about the commit (from git bisect): 58913b439b797462a70794774b87e63283109617 is first bad commit commit 58913b439b797462a70794774b87e63283109617 Author: Alexandre Julliard <julliard(a)winehq.org> Date: Fri Feb 15 12:07:56 2008 +0100 kernel32: On Mac OS, recompose the Unicode strings we get from the OS. :040000 040000 f3d03f060068424b166b0ce68169c5a3eddab96e f1a567176edfd42947239dbacff3cc8ad469a497 M dlls I'm using osx 10.5.3 in spanish on a macbook pro. Does anybody know about this?

2 3

new winetricks: new verbs comctl32.ocx, autohotkey
by Dan Kegel 14 Jun '08

14 Jun '08

Can somebody who needs comctl32.ocx or mscomctl.ocx test this? I added this at the suggestion of a comment on http://mike.kronenberg.org, but don't have an app handy to test with. Changes: Use "MS" uniformly in front of choices that are Microsoft downloads. Add comctl32.ocx Add comctl32 as alias for cc580 Added autohotkey Cosmetic changes to menu: make 1st column wider, make gui window wide enough Restore winversion on dotnet20 (Austin) As always, the latest winetricks is online at http://kegel.com/wine/winetricks and the change history is online at http://winezeug.googlecode.com/ - Dan

1 0

Re: winedbg: Add support for printing c++ bool values
by Eric Pouech 14 Jun '08

14 Jun '08

Jon Griffiths a écrit : > Hi, > > Subject says it all. > > + dbg_printf("%s", val_int ? "true" : "false"); what about the simpler: + dbg_printf(val_int ? "true" : "false"); A+ -- Eric Pouech "The problem with designing something completely foolproof is to underestimate the ingenuity of a complete idiot." (Douglas Adams)

2 1

re: Bugs in Wine research
by Dan Kegel 14 Jun '08

14 Jun '08

tomecek wrote: > we are students from Department of computer science at Aalborg > university. During this semester we were working on project on static > analysis using the Coccinelle tool > (http://www.emn.fr/x-info/coccinelle/) and Flawfinder > (http://en.wikipedia.org/wiki/Flawfinder) We decided to search for > bugs in Wine source code. Great, thanks! We didn't have a wiki page about static analysis and Wine yet, so I created one, http://wiki.winehq.org/StaticAnalysis At least a couple of the memory leaks you found seem like they might be worth fixing. If you could put your full results up somewhere, and link to them from that wiki page, we'll take a look.

3 4

Re: Bugs in Wine research
by tomecek＠cs.aau.dk 14 Jun '08

14 Jun '08

> Dan Kegel wrote: >> James McKenzie <jjmckenzie51(a)sprintpcs.com> wrote: >>>>> we are students from Department of computer science at Aalborg >>>>> university. During this semester we were working on project on static >>>>> analysis using the Coccinelle tool >>>> http://wiki.winehq.org/StaticAnalysis >>> I was wondering if results from Wine 0.9.57 would continue to be valid? >>> Maybe the effort could shift to RC1? >> >> Sure, lots of the unused vars bugs in particular are probably >> already fixed. But the other issues have a pretty good >> chance of still being valid. >> > > Are there any volunteers to do this? > > My experience with Wine C code is fairly minimal, but it shouldn't be > too hard to look at one of their changes and see if it applies to > current git, and if it does just forward-port it. > > Thanks, > Scott Ritchie > > > > Hi, the complete results are available at http://www.cs.aau.dk/~dechouxb/results/complete/, the scripts used at http://www.cs.aau.dk/~dechouxb/cocci_scripts/done/. It is valid for Wine 0.9.57. Sorry for the delay, we were very busy. Unfortunately we cannot rerun the tests because we are leaving Aalborg University in 3 weeks. If you are interested I should be able to rerun these tests (with some improvements) at my home university during late August and September. Jaroslav Tomecek ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program.

3 2

test.winehq.org doesn't always show the download link for the executable
by Paul Vriens 14 Jun '08

14 Jun '08

Hi, I was just reading a Dutch site were somebody mentioned to help Wine and referred to: http://test.winehq.com/data/062d61a5a4ba2d7972a0011387ceda64c79dd4e3/ (with a remark about possible reboots of your system). The current page however doesn't have a link to the winetest executable. This is because the last report that is picked up by the 'gather' script fills in this executable. If the last report came from somebody running it's own winetest on Wine (instead of the Paul Millars one) there will be 'N/A' (I also saw '-') instead of a proper executable. How could we improve this? - if the last report mentions 'N/A' use a link to http://www.astro.gla.ac.uk/users/paulm/WRT/CrossBuilt/winetest-latest.exe? The problem is however that this doesn't have to be the same winetest 'version'. - if any of the reports contains a link to Paul Millars executable, use that. - if none of the reports refer to a proper executable use that winetest-latest again. Any ideas, suggestions? -- Cheers, Paul.

1 0

ddraw: Fix Setmaterial with a test
by David Adam 14 Jun '08

14 Jun '08

>David Adam wrote: >>* The test succeded in my windows XP box. *>>* *>You have only ddraw test and no d3d test. > >Vitaliy. Giving a NULL value to SetMaterial in d3d9 or d3d8 makes Windows crashes. It is a forbidden value (see bug 13646). So, no test can be performed for d3d8 and d3d9. I sent a patch to removed these test earlier. David

3 6

Re: user32: Allow a NULL foreground window in the tests
by Dmitry Timoshkov 14 Jun '08

14 Jun '08

"James Hawkins" <jhawkins(a)codeweavers.com> wrote: > According to msdn, "The foreground window can be NULL in certain > circumstances, such as when a window is losing activation. " which is > exactly when these failures occur. > > Changelog: > * Allow a NULL foreground window in the tests. NULL foreground window is already allowed, you just need to fix the callers of check_wnd_state() to pass 0 for foreground. -- Dmitry.

2 11