May 2010 - Wine-Devel - list.winehq.org

Re: Wine looses a bit gaming and wins mobility?

by Julius Schwartzenberg

Edward Savage wrote: > It is a shame that it is just a rumor but really that's all it's been > for the last half decade. Plus, Phoronix refuses to release this > binary their meant to have found which I tend to think means they > don't have it. It is just a rumor and the binaries there were available were for the dedicated servers of some games. I made a copy here in case you're really interested: http://haar.student.utwente.nl/~julius/steam_linux/

May 6, 2010

Re: Stack corruption in dsound.c

by Henri Verbeet

On 6 May 2010 11:18, <wylda(a)volny.cz> wrote: > * The first 0022: and 0018: are thread numbers? Yes > * Can be the problem in 0018: be caused by different thread? Yes > * What is that 0/19? That means the first of 19 tail bytes was overwritten. The message as a whole means that there's a block of memory at 0x19543568, 5 bytes large (which is pretty small for something allocated on the heap, actually), and something wrote past the end of that block. Finding out where that block was allocated may help with identifying where it was corrupted. > * Can i make assumption, that it was overwritten by gdi32.wglGetCurrentContext()? No, not necessarily.

May 6, 2010

Re: Stack corruption in dsound.c

by wylda＠volny.cz

> > Allocates me 3E0 (rounded 3d4) bytes filled with > > zeros, begining at adr1 and this: > > a) adr1 should be "relatively close" to the 0x00110000 > > b) adr1 should be same as adr2 > > > You can't really make those assumptions, though I guess > they may > sometimes be true in practice. > OK, i lets start from the beginning. I did +relay,+heap and found: trace:heap:RtlAllocateHeap (0x4df4000,70000062,00000413): returning 0x4edded0 0018:Ret ntdll.RtlAllocateHeap() retval=04edded0 ret=78134d83 0023:Ret KERNEL32.GetCurrentThreadId() retval=00000023 ret=016633ec 0023:Call KERNEL32.GetTickCount() ret=016669e8 0023:Ret KERNEL32.GetTickCount() retval=00051289 ret=016669e8 0023:Call KERNEL32.GetCurrentThreadId() ret=0166342f 0023:Ret KERNEL32.GetCurrentThreadId() retval=00000023 ret=0166342f 0023:Call KERNEL32.InterlockedExchange(020876a8,00000001) ret=0166343b 0023:Ret KERNEL32.InterlockedExchange() retval=00000000 ret=0166343b 0023:Call KERNEL32.GetCurrentThreadId() ret=01663472 0023:Ret KERNEL32.GetCurrentThreadId() retval=00000023 ret=01663472 0023:Call KERNEL32.Sleep(00000032) ret=016632f9 trace:heap:RtlAllocateHeap (0x4df4000,70000062,00000047): returning 0x4e93558 0018:Ret ntdll.RtlAllocateHeap() retval=04e93558 ret=78134d83 0018:Call ntdll.RtlAllocateHeap(04df4000,00000000,00000113) ret=78134d83 trace:heap:RtlAllocateHeap (0x4df4000,70000062,00000113): returning 0x4e744f8 0018:Ret ntdll.RtlAllocateHeap() retval=04e744f8 ret=78134d83 0018:Call ntdll.RtlAllocateHeap(04df4000,00000000,00000047) ret=78134d83 trace:heap:RtlFreeHeap (0x110000,70000062,0x164740): returning TRUE 0022:Ret ntdll.RtlFreeHeap() retval=00000001 ret=7e1d68e8 0022:Call user32.WindowFromDC(00000534) ret=7e0f9ee7 0022:Ret user32.WindowFromDC() retval=00040056 ret=7e0f9ee7 0022:Call KERNEL32.TlsGetValue(00000008) ret=7e0fa801 0022:Ret KERNEL32.TlsGetValue() retval=19dfcfe0 ret=7e0fa801 0022:Call gdi32.wglGetCurrentContext() ret=7e0faeea 0022:Ret gdi32.wglGetCurrentContext() retval=19a66a00 ret=7e0faeea 0022:Call KERNEL32.TlsGetValue(00000008) ret=7e0fa801 0022:Ret KERNEL32.TlsGetValue() retval=19dfcfe0 ret=7e0fa801 trace:heap:RtlAllocateHeap (0x4df4000,70000062,00000047): returning 0x4e997d0 0018:Ret ntdll.RtlAllocateHeap() retval=04e997d0 ret=78134d83 0018:Call ntdll.RtlAllocateHeap(04df4000,00000000,00000213) ret=78134d83 err:heap:HEAP_ValidateInUseArena Heap 0x110000: block 0x19543568 tail overwritten at 0x1954356d (byte 0/19 == 0xcf) trace:heap:RtlAllocateHeap (0x110000,70000062,00000100): returning 0x19008c58 err:heap:HEAP_ValidateInUseArena Heap 0x110000: block 0x19543568 tail overwritten at 0x1954356d (byte 0/19 == 0xcf) * The first 0022: and 0018: are thread numbers? * Can be the problem in 0018: be caused by different thread? * What is that 0/19? * Can i make assumption, that it was overwritten by gdi32.wglGetCurrentContext()?

May 6, 2010

Re: Stack corruption in dsound.c

by Charles Davis

On 5/6/10 1:57 AM, wylda(a)volny.cz wrote: > BTW: why it takes so long (cca 6 hours) a post to appear, when i send > mail to wine-devel? Some kind of mail graylist? If you're not subscribed to wine-devel, your messages are subject to moderation. Chip

May 6, 2010

Re: Stack corruption in dsound.c

by Henri Verbeet

2010/5/6 <wylda(a)volny.cz>: >> What exactly is wrong with that address? > > Hi Henri, to my understending, when i take real values: > > adr1 = HeapAlloc(0x00110000, 0x00000008, 0x000003d4); > HeapFree (0x00110000, 0, adr1); > > adr2 = HeapAlloc(0x00110000, 0x00000008, 0x000003d4); > HeapFree (0x00110000, 0, adr2); > > Allocates me 3E0 (rounded 3d4) bytes filled with zeros, begining at adr1 > and this: > a) adr1 should be "relatively close" to the 0x00110000 > b) adr1 should be same as adr2 > You can't really make those assumptions, though I guess they may sometimes be true in practice. >> Stefan Dösinger: >> There's HeapValidate which may be better than Alloc+Free in spotting >> > bugs. > > Hi, this kind of help i'm looking for... Furhter googling lead me to > nice articles about memory damage. > WINEDEBUG="warn+heap" also does some basic validation. > BTW: why it takes so long (cca 6 hours) a post to appear, when i send > mail to wine-devel? Some kind of mail graylist? > If you're not subscribed to wine-devel your mails will go through moderation, which means they won't come through until a moderator looks at them. If you *are* subscribed mail should generally arrive within a couple of minutes, if it doesn't something is wrong somewhere.

May 6, 2010

Re: LinuxTag Berlin - call for projects

by Jan Gerrit Möltgen

Hello Kai, I've discard the stand already, but if someone would go to Berlin we could be reaccepted for LinuxTag. However, if we want, we schould act as soon as possible. A part of Mail from projects(a)linuxtag.org (06.05.'10): Falls doch noch die Möglichkeit besteht, jemand anderes aus eurem Projekt zu mobilisieren, gib uns bitte baldmöglichst Bescheid. Wir finden dann eine Möglichkeit. Greetings, Jan Gerrit Am Samstag, 1. Mai 2010, um 08:56:57 schrieb Kai Blin: > > Until when do you need to know? > > Cheers, > Kai -- Dipl.-Wirt.-Inf. Jan Gerrit Möltgen . . . . . .gerrit.moeltgen(a)itomig.de ITOMIG GmbH . . . . . . . . . . . . . . . . . . Tel.: +49 7031 389 25 06 Otto-Lilienthal-Str. 36 . . . . . . . . . . . . . . http://www.itomig.de D-71034 Böblingen . . . . . . . . . . . . . . . http://www.itomig.de/fr Geschäftsführer: David Gümbel . . . . . HRB 728519 Amtsgericht Stuttgart Sitz der Gesellschaft: Böblingen . . . . . . . . . . . . . . . . . . . .

May 6, 2010

Stack corruption in dsound.c

by wylda＠volny.cz

> > #274 before wined3d_mutex_unlock(), 0x0020a3f0=HeapAlloc(), > > #274 after wined3d_mutex_unlock(), 0x18fa7048=HeapAlloc() > > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > > > What exactly is wrong with that address? Hi Henri, to my understending, when i take real values: adr1 = HeapAlloc(0x00110000, 0x00000008, 0x000003d4); HeapFree (0x00110000, 0, adr1); adr2 = HeapAlloc(0x00110000, 0x00000008, 0x000003d4); HeapFree (0x00110000, 0, adr2); Allocates me 3E0 (rounded 3d4) bytes filled with zeros, begining at adr1 and this: a) adr1 should be "relatively close" to the 0x00110000 b) adr1 should be same as adr2 Both these a) and b) perfectly works 273 times, when injected in funtion wined3d_mutex_unlock(). So that makes me believe, that there is memory on 274 entrance when calling LeaveCriticalSection(&wined3d_cs) inside wined3d_mutex_unlock(). > Stefan Dösinger: > There's HeapValidate which may be better than Alloc+Free in spotting > > bugs. Hi, this kind of help i'm looking for... Furhter googling lead me to nice articles about memory damage. BTW: why it takes so long (cca 6 hours) a post to appear, when i send mail to wine-devel? Some kind of mail graylist?

May 6, 2010

Re: Wine looses a bit gaming and wins mobility?

by Edward Savage

It is a shame that it is just a rumor but really that's all it's been for the last half decade. Plus, Phoronix refuses to release this binary their meant to have found which I tend to think means they don't have it. Steam for Linux would really be a huge win for native gaming but from Valves point of view it might be too risky? A better solution which you'd hope they'd approach Codeweavers or the likes of Ubuntu about would be integrating Wine into a native Steam to run Win32 binaries of non-native games. On top of native Source games that is. This would be pretty easy, and using the Steam first-time-run script system we could actually make a native version support Win32 binaries via Wine ourselves (if it agreed to download and install them!). On the Steam topic I should note that Steam is now at the point where you can run it in the background 24/7. There is a patch that allows you to buy things via the Steam store and using the latest Beta Client update from the opt-in menu removes the last couple of CPU intensive problems with running it (this is distinct from the new UI update). I've had it open for the last week just like a native program and it's great to catch up with friends playing various games instead of them having to tell you to login and join!

May 6, 2010

Re: Stack corruption in dsound.c

by Henri Verbeet

On 5 May 2010 01:39, <wylda(a)volny.cz> wrote: > #274 before wined3d_mutex_unlock(), 0x0020a3f0=HeapAlloc(), &wined3d_cs=0x7e22fda0 > #274 after wined3d_mutex_unlock(), 0x18fa7048=HeapAlloc(), &wined3d_cs=0x7e22fda0 > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > Any suggestion? > What exactly is wrong with that address?

May 6, 2010

Re: Wine looses a bit gaming and wins mobility?

by Jerome Leclanche

Knowing that they are developing an opengl version of their engines (for mac os), I'm guessing porting it to linux will be trivial. J. Leclanche On Thu, May 6, 2010 at 3:07 AM, Ben Klein <shacklein(a)gmail.com> wrote: > On 6 May 2010 10:01, Evil Jay <wine(a)eternaldusk.com> wrote: >> On 05/05/2010 02:34 PM, André Hentschel wrote: >>> Hi Folks, >>> Steam seems to add a Linux client in the near future, so maybe Wine will not be needed anymore for that. or did i get something wrong? >>> That might reduce our "market share" a bit as i guess that many Wineusers play steam games. >>> On the other hand Intel presented its Z600, which is basically x86 but without IDE, SATA or a BIOS. Thats the reason that it cant run Windows. >>> So Linux runs and so should Wine i guess. So we might see some mobil advantures in the near future? >>> >> >> I would not hold my breath waiting for that Steam client, there's still >> been no official announcement and it could just as well be something >> they just play around with as a side project for the next half decade. >> >> Phoronix has been claiming Steam for Linux was imminent for 2 years... >> but for how many years did they (and even the developers) say that UT3 >> was right around the corner? A few 32-bit god-awfully incomplete >> binaries that can't even launch the actual GUI is not near enough to >> start me waiting in a virtual line. >> >> To take the Devil's advocacy a step farther: For all we know, Steam for >> Linux is not even being actively developed; They may have started it >> with the Mac port and abandoned it as too resource intensive for the >> expected payoff. The few updates that we've seen (before the Linux >> build was pulled) may well be the result of of the automated package >> builder recompiling the Linux files due to changes in some shared >> upstream code (being changed for the benefit of the Mac or Windows client). >> >> Now, personally I am hoping like crazy that Steam for Linux does >> materialize in the next year or two. But, if it does, it will hardly >> put a dent in Wine's use for gaming. Steam's not a game, so naturally >> we must assume that Valve will port the Source engine too, so that they >> will have some games to sell. That's great, but how many games in the >> AppDB are using the Source engine? How many games sold through Steam >> even use it, and will the non-Valve developers take the time to build >> Linux packages? Wine has plenty of gaming left in its future. > > The other solution is they pull a Picasa/Google Earth manoeuvre and > bundle Wine with Steam (or rely on Wine to play the games). > >> -J >> >> >> > > >

May 6, 2010