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

Re: RFC: client side implementation of async I/O
by Alexandre Julliard March 6, 2001

March 6, 2001

Michael McCormack <mccormac(a)aals27.alcatel.com.au> writes: > This patch moves responsibility for asynchronous I/O to the client process. > i'm posting it to wine-devel because i want to know what other developers here > think about the technique i've used to make this work... Your approach is not going to work with the latest changes I made to the server. The good news is that the changes I'm making are in part to allow making server calls in signal handlers, so when this works you should be able to use SIGIO to do async IO. > i think this implementation is more efficient, as it makes fewer server calls > and duplicates fewer file descriptors, while maintaining correctness. It's not clear to me that it is more efficient; it seems it will do a lot more work on every single server call if an async IO is pending. Of course the only way to tell is to run benchmarks. -- Alexandre Julliard julliard(a)winehq.com

3 2

TCP/IP networking and registry keys
by Alan Chandler March 6, 2001

March 6, 2001

I spent most of the weekend trying to get Grand Prix Legends GPL to work. I achieved my first goal of getting it to run in server mode. One of the things I needed to get the game to do, was to recognise that there where some interfaces it could talk TCP/IP out of. It took me some time to realise that under windows it was doing this by looking for the key [System\\CurrentControlSet\\Services\\Class\\NetTrans] and was then enumerating the keys under it to establish an IP address ie the following [System\\CurrentControlSet\\Services\\Class\\NetTrans\\0000] "IPAddress"="10.0.10.100" The thought occurred to me that maybe the tool the builds the registry during wine install could actually create these keys - since it is essentially a key part of the system (and other similar keys are also built in the same way). I don't think I have the skills to do it myself yet. Alan alan(a)chandlerfamily.org.uk http://www.chandler.u-net.com

3 2

Re: Speeding up wineserver syncronization objects with shared memory
by David Howells March 5, 2001

March 5, 2001

Gavriel State <gav(a)transgaming.com> wrote: > As it stands, your approach won't be useful for general Wine usage until > you've got *everything* done. True. But I think there are valid reasons for doing it this way. One major problem is handles. Either the kernel must allocate all handles, or userspace must allocate all handles. Take something like the implementation of DuplicateHandle() - dead easy really in kernel space, including duplication to another process. If userspace tells the kernel to use handle X for a mutex, say, and handle Y for a semaphore, then userspace still has to go to atomic management code to (a) allocate a handle, and (b) access a handle. And to implement cross-process duplication, you either need some sort of interrupt & IPC mechanism, or an external process (the wineserver). This being the case, you lose any gains by putting the stuff in-kernel. Alternatively, if userspace is allowed to implement arbitrary handles that are allocated by the kernel, then the kernel waiting mechanism can get a little tricky. However, that said, for non-waitable objects, this might not be so bad. In fact, I'm planning on implementing the registry access handles this way and may well do some of the process and thread control this way too. Furthermore, the kernel would have to be given object handles not object pointers, otherwise you have a gaping security hole. And another major problem is context switches... they are horribly expensive really. They currently make Wine with at least 20 times the system call latency that the kernel method is capable of. > And then there will be boatloads of debugging to do. Perhaps not as much as you think... I'm doing as much debugging as I can as I go along, just making sure the kernel objects work as I'd expect (not necessarily the same thing, true, as having "compatible" behaviour). > If there's any way that you can implement your kernel module more within the > context of the existing server architecture - replacing objects on a piece > by piece fashion rather than all at once - that might make it easier to > adopt. See above for why the piece-by-piece method is difficult. One alternative would be to invent a new network protocol (say AF_WINE), but that again requires a complete implementation before it is really useful. > For example, you might try implementing the core object/handle management > and waiting code in the kernel module, and have the wine server rely on the > kernel module for that low-level functionality. By this, do you mean actually having the wineserver process talk to the kernel module on behalf of the Wine application? > Waiting for objects would be implemented on the client side through a call > to the kernel module. That's what I'm currently doing, though it's not fully implemented yet. > When something needs to be done with an object, we would call either the > wineserver or the kernel module, depending on how that object is > implemented. > For example, you could do mutexes entirely within the kernel module, but > leave file objects on the wine-server side initially. My main gripe is the slow speed of access to files... Every Read/WriteFile goes to the wineserver to convert the handle into a file descriptor and to check for locking. The FD is then passed back over a UNIX domain socket, used once and then closed. I suspect it can't really be done otherwise, particularly if ZwCloseObject (or whatever it is called) is implemented, since this allows handles in another process to be closed. Actually, I've done a fair amount of the file object stuff... Most of it involves mapping down to a "struct file *", which is how the kernel views files, and then invoking appropriate kernel method. > In my experience Alexandre far prefers incremental change to kind of > approach you're taking. Using that kind of approach will improve the > chances that your code will make it into Wine at some point. Hmmm... It's difficult to determine how to do it incrementally without making for even more work, but I think I know what you mean. > One thing I've been wondering that you might be able to answer is this: > exactly why is the current Wine Server architecture so slow? Is it just the > context switching? Context switching is the main element of it. Going to the wineserver and back again just for a ReadFile() call or a Wait*() function incurs a fairly serious penalty (particularly on an X86, I think). Plus there's no requirement for the kernel to pass the remains of your timeslice to the wineserver and back again. Also, you have to bundle lots of data through AF_UNIX network packets and/or copy lots of data into and out of _shared_ memory without killing other threads. One of the problems with the context switch is that you have to flush all the CPU caches, muck around with the MMU and execute scheduling algorithms. > Is it that the kernel isn't giving the wineserver a high enough priority > once the client blocks after having written to the socket? I don't think priority has anything much to do with it. A more convenient scheduling algorithm might help a little, though. > Is it other socket overhead (routing, perhaps)? Simply speeding up the > communications path between the clients and the server would remove the need > for most of the kernel level services. I don't think so... To communicate with the wineserver you have to use some sort of waitable UNIX object (ie: a socket, a pipe, or a SYSV semaphore or message), do busy waiting (& kill your CPU), or send signals (context switch _and_ signal overhead). To do it without a wineserver (ie: using shared memory) is also tricky... You have to be able to, for instance, recover from processes going away without releasing mutexes. > Also, just FYI, the PE image mapping work is nice, but isn't likely to affect > speed all that much, since Wine can just mmap in most PE images created with > recent compilers. WordPerfect, for example, launches in around 10-12 seconds > on my machine (down from 45-60 seconds before the mmaping). I can't imagine > that doing the fixups when paging instead would do too much better. Look at the VM size of something like MS Word, and think of not having to allocate buffers to store all those DLLs, thus eating a massive chunk out of your machines total VM. Plus, it should be even quicker launching because fixups don't have to be done unless they're necessary. You certainly don't have to go through an fixup a few tens of megs of DLL. Cheers, David

1 0

Wine Press
by Eric Pouech March 5, 2001

March 5, 2001

http://linux.cnet.com/linux/0-2136864-7-4961586.html worth the reading, even if Wine gets the lowest ranking my reading of it is that Wine is not yet a product (compared to the two others). I definitively think we should focus more on a release target. It of course means that we have to make choices (do not include some functionnalities), spend pretty much more time on testing... but that's the way to go if we want to have Wine 1.0 around on of these days A+ --------------- Eric Pouech (http://perso.wanadoo.fr/eric.pouech/) "The future will be better tomorrow", Vice President Dan Quayle

3 2

Re: RESUBMIT: afm
by Gavriel State March 5, 2001

March 5, 2001

Back in Febuary, Marcus Meissner wrote: > > Changelog: > fixed one NULL ptr problem > added loading of .afm files from several common locations > (ghostscript, a2ps, enscript, (teTeX), X11) > > Index: afm.c > =================================================================== > RCS file: /home/wine/wine/dlls/wineps/afm.c,v > retrieving revision 1.4 > diff -u -r1.4 afm.c > --- afm.c 2001/01/02 20:30:16 1.4 > +++ afm.c 2001/02/03 18:50:42 > BOOL PSDRV_GetFontMetrics(void) > { > int idx = 0; > char key[256]; > char value[256]; > + char *path; > + DIR *dir; > + > + /* some packages with afm files in that directory */ > + PSDRV_ReadAFMDir("/usr/share/ghostscript/fonts/"); > + PSDRV_ReadAFMDir("/usr/share/a2ps/afm/"); > + PSDRV_ReadAFMDir("/usr/share/enscript/"); > + PSDRV_ReadAFMDir("/usr/X11R6/lib/X11/fonts/Type1/"); We just did an integration with WineHQ, and this code has started to grate on me (mostly because of the fact that the AFM parser spits out a bunch of error messages while trying to parse my AFM files. Rather than hardcoding the paths into the code, could we do this instead: while (PROFILE_EnumWineIniString( "afmdirs", idx++, key, sizeof(key), value, sizeof(value))) PSDRV_ReadAFMDir(value); This has the added benefit of allowing the user to specify only those AFM directories that their rasterizer knows about too. With the above, I get a whole bunch of printer fonts listed as available in wine that I can't actually print if Ghostscript isn't configured to find the font files. -Gav -- Gavriel State, CEO TransGaming Technologies Inc. http://www.transgaming.com gav(a)transgaming.com

1 0

[RESEND] Re: faster 8bpp to 24 bit depth conversion in SetImageBits_8
by Vedran Rodic March 4, 2001

March 4, 2001

On Sun, Mar 04, 2001 at 01:51:16PM +0100, Ove Kaaven wrote: > > > Since this is a 32-bit move (movl), you should use %%eax, not %%ax. The > assembler doesn't complain about this now, but technically it should, and > might someday. > > Since you're converting to 32bpp, not 16bpp, you should multiply "left" by > 4, not by 2, or strange things may happen in some apps. Thanks. Fixed patch attached. I did change left multiplicator from 2 to 4 in my earlier patches, but I forgot to include it in my final patch being 3 AM and so... I'll try to be more careful next time. Vedran

1 1

Re: for loop fixes
by Francois Gouget March 4, 2001

March 4, 2001

On Sat, 3 Mar 2001, Andreas Mohr wrote: [...] > - changed all terminating ";" of isolated for loops to " ;" to indicate the > loop's isolation. IMHO, for such loops the following would be much clearer: <init_expression>; while (<test_condition>) { <update_expression>; } (ok, you can also put '{' on the next line) > Please do the same for these for loops in the future. ??? > Somebody might want to write a Wine coding guidelines article... $ wg . -E 'for *$' | grep -E '$ *; *$'` | wc -l 96 (wg is my recursive source grep script) Maybe we should ask people to put the for and the instruction it applies to on separate lines. Combined with the suggestion above this would make it easier to check for this kind of bug. > P.S.: I just *knew* that at least one other for loop in Wine would be wrong, > too ;-) Yep, I should have done that myself (if you find a generic bug, always check for other similar bugs). -- Francois Gouget fgouget(a)free.fr http://fgouget.free.fr/ In a world without fences who needs Gates?

4 4

RFC: client side implementation of async I/O
by Michael McCormack March 4, 2001

March 4, 2001

Hi all, This patch moves responsibility for asynchronous I/O to the client process. i'm posting it to wine-devel because i want to know what other developers here think about the technique i've used to make this work... i've relaxed some of the assumptions wineserver makes about client processes. i think this implementation is more efficient, as it makes fewer server calls and duplicates fewer file descriptors, while maintaining correctness. Mike -- __________________________________________________________________ mailto:Michael.McCormack(a)alcatel.com.au phone: (+8232) 667 0425 (h) (+8211) 430 0425 (m)

1 0

Re: Add explicit WM_CLOSE processing in the edit control
by Alexandre Julliard March 3, 2001

March 3, 2001

Dmitry Timoshkov <dmitry(a)sloboda.ru> writes: > Since handler of WM_CLOSE in DefWindowProc calls DestroyWindow, > which eventually sends WM_DESTROY, EDITSTATE gets destroyed before > call to EDIT_UnlockBuffer in the end of processing of WM_CLOSE, > therefore EDIT_UnlockBuffer gets confused. This is not specific to WM_CLOSE, it can happen with other messages too; any time we transfer control back to the application the window can be destroyed. I think it would be better to either unlock the buffer before calling DefWindowProc, or make EDIT_UnlockBuffer cope with a destroyed window. -- Alexandre Julliard julliard(a)winehq.com

1 0

Re: erase bkgnd on page scroller control
by Alexandre Julliard March 3, 2001

March 3, 2001

Susan Farley <susan(a)codeweavers.com> writes: > @@ -720,7 +721,8 @@ > PAGER_INFO *infoPtr; > DWORD dwStyle = GetWindowLongA (hwnd, GWL_STYLE); > > - dwStyle |= WS_CLIPCHILDREN; > + /* background color of the child should be the same as the pager */ > + dwStyle &= ~WS_CLIPCHILDREN; Shouldn't we simply leave WS_CLIPCHILDREN alone and use whatever the app specified when creating the window, instead of forcing it one way or another? -- Alexandre Julliard julliard(a)winehq.com

1 0