November 2001 - Wine-Devel - list.winehq.org

Re: GetDlgItem16 - is id signed or not?
by Gerard Patel Nov. 12, 2001

Nov. 12, 2001

At 07:07 PM 11/11/2001 -0500, you wrote: > >It hasn't gotten itself fixed in the ordinary way of progress, so I have >finally gotten around to a regression in juno-1.49. It crashes just >after getting back 0 from GetDlgItem16, so I added a little trace inside >GetDlgItem. Now I know why it crashes, but I don't know - is id really >unsigned, or is GetControl16 incorrectly failing to extend the sign? >I guess I could try it each way and see which one does least damage, but >probably several of you know which way it is meant to go. I'd cast id to unsigned in GetDlgItem16. Gerard

2 1

Re: Regression errors
by Eric Pouech Nov. 12, 2001

Nov. 12, 2001

> At 11:06 AM 10/11/2001 -0500, you wrote: > > > >Thanks Gerard. I will try it Tuesday. Going out of town till then. > > Well, I have advanced a bit this weekend. I noticed that one of my > test app does indeed show this regression after the patch we are talking > about. > > As the case is a lot more simple than the one I was trying to debug > before for this problem, I have been able to understand better what > is happening. The problem shows itself in the following case : > > - invalidating a window > - hiding it > - deleting it IIRC, before the big window overhaul (when paint count was attached to each window), when a window was hidden, its paint count wasn't set to 0 (or when the window was actually shown, its paint count wasn't set to 1). This lead to some painting errors... I don't know if this is fully related (and I didn't follow closely the new window stuff), but it may be related... in case this helps... A+ --------------- Eric Pouech (http://perso.wanadoo.fr/eric.pouech/) The future will be better tomorrow, Vice President Dan Quayle ____________________________________________________________ Faites un voeu et puis Voila ! www.voila.fr Avec Voila Mail, consultez vos e-mails sur votre mobile Wap.

1 0

Re: More overlapped I/O issues (important)
by Mike McCormack Nov. 12, 2001

Nov. 12, 2001

Hi Martin, You're right, i never considered multiple overlapped requests at the same time... To fix the problem correctly, you will need to move (some of) the overlapped code back to the wineserver, so that requests from different processes can be handled correctly. (i moved it out of the wineserver about a year ago for efficiency reasons.) Have a look at rev 1.7 of server/async.c to see how this worked. http://cvs.winehq.com/cvsweb/wine/server/async.c?rev=1.7&content-type= text/x-cvsweb-markup The current overlapped code (with the above exception) is similar to what win 9x supports, so it does not support overlapped on normal files, and file offsets in OVERLAPPED structures. If that could be implemented too, it would be good. i guess there is choice now... implement overlapped sockets with the existing structure, or make a rather large change to the wine architecture, which might take a while. Maybe a good first step is to back out my change which put overlapped operations into the client side code... Alexandre thought this could be done using POLLIO instead... Mike Original message from: Martin Wilck <Martin.Wilck(a)fujitsu-siemens.com> > > >Hi Mike, all, > >I see a severe problem with the current overlapped I/O implementation, >especially wrt extending it to Winsock2 overlapped I/O. > >The problem I see is with buffer ordering. > >For normal files, the file position from which to read is on the >OVERLAPPED struct. This was not in the Wine implementation, the patch in >my last email fixes it. > >For sockets and other FIFOs, this makes no sense. In this case Windows >specifies that buffers are read / written **in the same order they were >submitted**. This matters only if users spawn several asynchronous I/O >requests at the same time, but this is allowed and actually beneficial >for application performance (at least on Windows). > >Unless I oversee something essential, the current wine implementation >does not guarantee such behaviour. Actually, since new requests are >inserted at the head of the NtCurrentTeb() list, they will be considered >first in subsequent asynchronous reads - i.e. the order will be reversed. >If I/O requests don't complete in a single read() or write() call, the >ordering will be basically random. > >IMO this implies that for sockets, asynchronous reads and writes must >be scheduled such that only one request for a given socket is actually >reading at a time, and all others are blocked until this request is >finished. Again, this holds only for FIFO-type files. > >With the current structure, it seems to be very complex to sort this out. >One could recheck the fd list in check_async and remove duplicate fds. >One could also try to "lock" the fd somehow. Still it would be necessary >to ascertain that the correct request gets the lock first. > >With this min mind, it seems more natural for me to organize the >asynchronous requests in the "file" data structure itself. One would >have a reader's and a writers's list there, and make sure that for FIFOS >only the first list element can actually issue a read(), write(), send(), >or recv() system call. For normal files all elements in the lists could be >doing IO at the same time, question is if that makes sense or not. > >As I see it, this would mean both the readers and writers lists must >be stored in the file data structure in the wine server. >This would of course strongly affect the current implementation. > >Note that with Winsock2 overlapped I/O in place, it must >be possible to use also ReadFile(), ReadFileEx() etc. on >overlapped sockets, and get correct behaviour. >Thus the ReadFile() etc. code must be able to deal with both sockets >and ordinary files, and it makes no sense to intriduce a different >approach for sockets only. > >Please give me your opinion on these issues ASAP. > >Regards >Martin > >-- >Martin Wilck Phone: +49 5251 8 15113 >Fujitsu Siemens Computers Fax: +49 5251 8 20409 >Heinz-Nixdorf-Ring 1 mailto:Martin.Wilck(a)Fujitsu-Siemens.com >D-33106 Paderborn http://www.fujitsu-siemens.com/primergy ------------------------------------------ mailto:Mike_McCormack(a)start.com.au ph +82 16 430 0425 __________________________________________________________________ Get your free Australian email account at http://www.Looksmart.com.au

2 1

Loop (and other) problem in audio.c 1.45 located
by Simon Britnell Nov. 12, 2001

Nov. 12, 2001

The problem is that I advance to the next wavehdr automatically in FeedDSP, which skips the first wavehdr sent to the wodPlayer. Doh! A patch will follow to fix this (which will be built on top of the race condition fix patch). __________________________________________________ Do You Yahoo!? Find a job, post your resume. http://careers.yahoo.com

1 0

Re: Winsock2 & overlapped IO
by Mike McCormack Nov. 12, 2001

Nov. 12, 2001

Hi Martin, Original message from: Martin Wilck <Martin.Wilck(a)fujitsu-siemens.com> >I have taken a closer look at this code. I believe a very similar approach >is possible for Winsock. However, there are a few problems: Good to see you are a man of action, not just words! >typedef struct async_private .. >What do you think about this? > >It would be cleaner to define a "general" async structure >containing either a type-specific union or a pointer to >type-specific data, and using void* for pointers. However, that would >require even more changes in the async file I/O code, which I'd like to >avoid at the moment. i agree with you it would be better to use a union... i'll have a go at making a patch to do this. i'll probably understand the mess i've created in files/file.c more easily ;-) >Another difference is that the overlapped Winsock routines cause >asynchronous success notification (i.e. calling the completion routine or >signalling the lpOverlapped->hEvent) even if they are immediately >successful, which seems not to be the case for asynchronous file I/O. > >That would require a minor change in the check_async_list() routine. Cool, go for it. >Please tell me your opinion! > >Regards, >Martin i wrote a longer reply with a union definition in it, but accidently resized my Netscape window an lost it when the page reloaded :-( Could be about time to give up on this web-based email thing. Mike btw. in the patch in your other mail, you have: -typedef struct _OVERLAPPED * LPWSAOVERLAPPED; typedef HANDLE WSAEVENT; + +typedef struct _WSAOVERLAPPED { + DWORD Internal; + DWORD InternalHigh; + DWORD Offset; + DWORD OffsetHigh; + WSAEVENT hEvent; +} WSAOVERLAPPED, *LPWSAOVERLAPPED; + can you just use the following? typedef OVERLAPPED WSAOVERLAPPED; typedef LPOVERLAPPED LPWSAOVERLAPPED; typedef HANDLE WSAEVENT; or even #define... ------------------------------------------ mailto:Mike_McCormack(a)start.com.au ph +82 16 430 0425 __________________________________________________________________ Get your free Australian email account at http://www.Looksmart.com.au

3 3

GetDlgItem16 - is id signed or not?
by lawson_whitney＠juno.com Nov. 12, 2001

Nov. 12, 2001

Good day! It hasn't gotten itself fixed in the ordinary way of progress, so I have finally gotten around to a regression in juno-1.49. It crashes just after getting back 0 from GetDlgItem16, so I added a little trace inside GetDlgItem. Now I know why it crashes, but I don't know - is id really unsigned, or is GetControl16 incorrectly failing to extend the sign? I guess I could try it each way and see which one does least damage, but probably several of you know which way it is meant to go. trace:dialog:DIALOG_CreateControls BEGIN trace:dialog:DIALOG_GetControl16 STATIC '' 1164, 12, 4, 232, 45, 50020000, 00000000 trace:dialog:DIALOG_GetControl16 BUTTON '&Get and Send Mail' 1162, 10, 55, 56, 28, 5003000b, 00000000 trace:dialog:DIALOG_GetControl16 BUTTON '&Just Get Mail' 1163, 70, 55, 56, 28, 5001000b, 00000000 trace:dialog:DIALOG_GetControl16 BUTTON 'Cancel' 1161, 130, 55, 56, 28, 5001000b, 00000000 trace:dialog:DIALOG_GetControl16 BUTTON '&Help' 57669, 190, 55, 56, 28, 5001000b, 00000000 trace:dialog:DIALOG_CreateControls END trace:dialog:GetDlgItem GWL_ID 48c id 48a trace:dialog:GetDlgItem GWL_ID 48a id 48a trace:dialog:GetDlgItem GWL_ID 48c id 48b trace:dialog:GetDlgItem GWL_ID 48a id 48b trace:dialog:GetDlgItem GWL_ID 48b id 48b trace:dialog:GetDlgItem GWL_ID 48c id 489 trace:dialog:GetDlgItem GWL_ID 48a id 489 trace:dialog:GetDlgItem GWL_ID 48b id 489 trace:dialog:GetDlgItem GWL_ID 489 id 489 trace:dialog:GetDlgItem GWL_ID 48c id ffffe145 trace:dialog:GetDlgItem GWL_ID 48a id ffffe145 trace:dialog:GetDlgItem GWL_ID 48b id ffffe145 trace:dialog:GetDlgItem GWL_ID 489 id ffffe145 trace:dialog:GetDlgItem GWL_ID e145 id ffffe145 wine: Unhandled exception, starting debugger... Regards, Lawson The Devil finds work for idle minds.

1 0

Latest CVS Wine winmm
by Ori Pessach Nov. 10, 2001

Nov. 10, 2001

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hello, I tried running Pajama Sam today (a Humongous Entertainment game), which used to run, with some sound problems, a while ago. Now, I get a dialog box on startup, saying that "No sound card was detected". 'wine -debugmsg +wave,+msys' shows multiple instances of the message: trace:mmsys:DriverCallback Null ! - From a quick look at mmsystem.c, it's not immediately clear whether this is a problem or not. At some point later on, I get: err:msacm:wodCallback Bad data But, this only happens after I close the dialog box, so that can't be the cause of the failure. Again, this used to work before. I first tried it with Simon's recent patch, and then checked out current the CVS version of mmsystem.c, and I don't see a difference. Any ideas? I have a complete dump the trace messages, but I'd rather not post it here, since it's fairly big. Ori Pessach -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.6 (GNU/Linux) Comment: http://www.oripessach.com iD8DBQE77KsXzA/uWWyw6uQRApmPAKDOJ5LmmUzwWczXJ/So263BrFR26ACfeLkr SbhwMyhft7A5pmgIO1azzfA= =VyBD -----END PGP SIGNATURE----- _________________________________________________________ Do You Yahoo!? Get your free @yahoo.com address at http://mail.yahoo.com

3 2

Re: Regression errors
by Gerard Patel Nov. 10, 2001

Nov. 10, 2001

At 12:38 AM 10/11/2001 -0500, you wrote: >1. SetTimer call is not always generating WM_TIMER messages as it should. I >see: > >0806bb20:Call user32.SetTimer(00020057,00001003,00000000,6003a04c) >ret=6001e9e4 >0806bb20:Ret user32.SetTimer() retval=00001003 ret=6001e9e4 > >in the trace, but never is the callback routine invoked. Without the above >patch, the callback is invoked and does the painting. I am not sure if it's related, but I have seen something like that. I have seen it since much longer than this patch, though. It's about the paint count in the server. When there is a paint count > 0 in the server and the number of windows found to need a repaint (at the client level) is less than the paint count, the client can never get a timer event again. Try this patch to dlls/user/message.c to see if your problem is related : --- message.c.orig Sat Nov 10 07:03:29 2001 +++ message.c Sat Nov 10 07:08:27 2001 @@ -1861,6 +1861,7 @@ USER_Driver.pMsgWaitForMultipleObjectsEx( 0, NULL, 0, 0, 0 ); hwnd = WIN_GetFullHandle( hwnd ); +yetanothergoto: locks = WIN_SuspendWndsLock(); if (!MSG_peek_message( &msg, hwnd, first, last, @@ -1884,8 +1885,19 @@ /* need to fill the window handle for WM_PAINT message */ if (msg.message == WM_PAINT) { - if (!(msg.hwnd = WIN_FindWinToRepaint( hwnd ))) return FALSE; - + if (!(msg.hwnd = WIN_FindWinToRepaint( hwnd ))) + { + TRACE("FIXME : server paint count > 0 but no window needs painting\n"); + SERVER_START_REQ( inc_queue_paint_count ) + { + req->id = (void *)GetWindowThreadProcessId( hwnd, NULL ); + if (!req->id) req->id = (void *)GetCurrentThreadId(); + req->incr = -100000; /* reset paint count */ + SERVER_CALL(); + } + SERVER_END_REQ; + goto yetanothergoto; + } if (IsIconic( msg.hwnd ) && GetClassLongA( msg.hwnd, GCL_HICON )) { msg.message = WM_PAINTICON; I have no idea about your second problem. Gerard

2 1

Help with Revelant demo
by Paul Millar Nov. 10, 2001

Nov. 10, 2001

Hi, [I'm prefacing this by saying know next to nothing about Windows] I've had a quick go at getting the (newer) demo of Eidos' Revenant game to run. The install works ok, but when I try to run the demo, I get a msgbox saying: "FATAL ERROR", "Unable to initialize Network" and the following appearing on stderr: > fixme:ole:CoCreateInstance no classfactory created for > {2fe8f810-b2a5-11d0-a787-0000f803abfc}, hres is 0x80040154 Rerunning with the "-debugmsg +ole" option, I get some (not terribly useful) extra info: > trace:ole:OLE32_DllEntryPoint 0x4087b000 0x1 0x1 > trace:ole:CoInitializeEx ((nil), 2) > trace:ole:CoInitializeEx () - Initializing the COM libraries > trace:ole:RunningObjectTableImpl_Initialize () > trace:ole:WINE_StringFromCLSID > 0x5a27f8->{2FE8F810-B2A5-11D0-A787-0000F803ABFC} > trace:ole:CoGetClassObject > CLSID: {2fe8f810-b2a5-11d0-a787-0000f803abfc}, > IID: {00000001-0000-0000-c000-000000000046} > fixme:ole:CoCreateInstance no classfactory created for > {2fe8f810-b2a5-11d0-a787-0000f803abfc}, hres is 0x80040154 > trace:ole:OLE32_DllEntryPoint 0x4087b000 0x0 0x1 I did a Google search and "2fe8f810-b2a5-11d0-a787-0000f803abfc" seems to be the CLSID of DirectPlayLobby. On a seperate note, I can't figure out which bit (or lack) of WINE is causing the "network error". From what I can see, there doesn't seem to be any network activity as rerunning with "-debugmsg +wnet,+winsock" doesn't produce any extra output. FWIW, there is an .INI file with the following: > [Network] > ConnectType=none > SessionName=none > Provider=IPX > ConnectTime=5 So my questions are: 1. How easy would it be to fix/add CLSID_DirectPlayLobby? (where do I start)? 2. Any suggestions how to go about debugging the apparently fatal "No Network" msgbox? Cheers, Paul. ------------------------------------------------------------------------------ Paul Millar yo-yo, n. : Particle Physics Theory Group Something that is occasionally Department of Physics and Astronomy up but normally down. University of Glasgow, (see also Computer) Glasgow G12 8QQ, paulm(a)physics.gla.ac.uk Scotland +44 (0)141 330 4717 ------------------------------------------------------------------------------

1 0

Regression errors
by Guy L. Albertelli Nov. 10, 2001

Nov. 10, 2001

I have noticed two other regression issues with the patch to "Store in the server all the window info... " (01/10/22 15:08:34). Both of these occur in Outlook Express 4. 1. SetTimer call is not always generating WM_TIMER messages as it should. I see: 0806bb20:Call user32.SetTimer(00020057,00001003,00000000,6003a04c) ret=6001e9e4 0806bb20:Ret user32.SetTimer() retval=00001003 ret=6001e9e4 in the trace, but never is the callback routine invoked. Without the above patch, the callback is invoked and does the painting. 2. A WM_KEYUP (wparam is VK_CNTL) message is generated and processed even though no key is touched on the keyboard. If "-debugmsg +key" is specified the message never appears, nor does any trace:key: entry. Has anybody else seen or trapped these bugs? Guy Albertelli <<galberte(a)neo.lrun.com>>

1 0