January 2014 - Wine-Devel - list.winehq.org

Re: [PATCH] Add support for GNU/Hurd

by Svante Signell

On Mon, 2014-01-06 at 15:30 +0100, Svante Signell wrote: > On Fri, 2014-01-03 at 16:49 +0100, Svante Signell wrote: > > Hi, > > > > On Fri, 2014-01-03 at 15:35 +0100, André Hentschel wrote: > > > Am 03.01.2014 12:32, schrieb Svante Signell: > > > > Hi, > > > > > > > > Currently wine-1.6.1 FTBFS on Debian GNU/Hurd, see > ... > > > > > > This winterbreak i looked into this again, too. I was quite suprised > > > that i had no problems with socket credentials, but it might still be a > > > problem in other hurd distros. > > > > The only credentials needed by wine is SCM_RIGHTS, and that is already > > supported. > > I found the reason why wine runs now, it was solved by fixing the > SCM_CREDS bug in: SCM_RIGHTS!! > #724681: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=724681 > resulting in an error: EKERN_INVALID_ADDRESS by returning code 1 from > the sendmsg when file descriptor flags are non-zero (=1). > > FYI for #733605: The upstream wine patches are at: > https://source.winehq.org/patches/data/10143 > and > http://wiki.winehq.org/Hurd >

Jan. 6, 2014

Re: strmbase: Store BasePinFuncTable inside Base*PinFuncTable structs.

by Jacek Caban

On 01/03/14 19:10, Aric Stewart wrote: > Also just so you are aware, all the strmbase stuff is based off actual microsoft strmbase stuff for directshow > > http://msdn.microsoft.com/en-us/library/windows/desktop/dd318897%28v=vs.85%… > > > Not that this patch specifically breaks that, I have not dug that much, but just so that you are aware of the base for the implementation. Sure, the patch that I sent just cleans up C representation of Base*Pin classes a bit. Thanks, Jacek

Jan. 6, 2014

Re: [1/7] gdi32: Relax vertical face requirements.

by Akihiro Sagawa

On Sat, 4 Jan 2014 20:02:35 +0900, Dmitry Timoshkov wrote: > > + const DWORD DBCS_MASK = 0x3e0000; > dlls/gdi32/font.c,GetFontLanguageInfo() defines DBCS_MASK as 0x3f0000, > I'd guess that one of these places needs to be fixed. Good point. I think Wine's GetFontLanaguageInfo needs to update. There are two reasons. Firstly, GetFontLanguageInfo()'s DBCS_MASK is not suitable for DBCS and vertical writing. Because 0x3f0000 sets bit 16 (0x10000), that is Thai charset. Secondly, I tested native GetFontLanguageInfo(). The result[1] shows even DBCS font, like MS Gothic, doesn't return GCP_MBCS flag. I've not figure out what condition is needed for GCP_DBCS yet. If you have a good idea, could you let me know. Thanks in advance. [1] http://newtestbot.winehq.org/JobDetails.pl?Key=4002&log_206=1#k206 font.c:5993 is an additional test for GCP_xxx flags Regards, Akihiro

Jan. 6, 2014

Re: Build fixes for wine-1.6.1 (FreeType2)

by Vincent Povirk

Scott Ritchie has already requested this on the bug (http://bugs.winehq.org/show_bug.cgi?id=35010) by adding the target milestone of 1.6.x. The patch will most likely be in 1.6.2, if and when that release happens. Patches aren't added to the stable branch except when making new releases.

Jan. 6, 2014

WineHQ Downtime

by Jeremy Newman

Starting tomorrow at 10:00 CST we will be performing updates and maintenance to the WineHQ server. The service window should be around 2 hours at most. Various services will experience outages during this time frame. This will affect www.winehq.org, appdb.winehq.org, bugs.winehq.org, forum.winehq.org, source.winehq.org, mailing lists, and GIT. wiki.winehq.org will not be affected.

Jan. 6, 2014

Re: [PATCH] Add support for GNU/Hurd

by Svante Signell

On Fri, 2014-01-03 at 16:49 +0100, Svante Signell wrote: > Hi, > > On Fri, 2014-01-03 at 15:35 +0100, André Hentschel wrote: > > Am 03.01.2014 12:32, schrieb Svante Signell: > > > Hi, > > > > > > Currently wine-1.6.1 FTBFS on Debian GNU/Hurd, see ... > > > > This winterbreak i looked into this again, too. I was quite suprised > > that i had no problems with socket credentials, but it might still be a > > problem in other hurd distros. > > The only credentials needed by wine is SCM_RIGHTS, and that is already > supported. I found the reason why wine runs now, it was solved by fixing the SCM_CREDS bug in: #724681: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=724681 resulting in an error: EKERN_INVALID_ADDRESS by returning code 1 from the sendmsg when file descriptor flags are non-zero (=1). FYI for #733605: The upstream wine patches are at: https://source.winehq.org/patches/data/10143 and http://wiki.winehq.org/Hurd

Jan. 6, 2014

Re: msvcp: Return if module is not loaded (Coverity)

by Piotr Caban

On 01/04/14 19:24, André Hentschel wrote: > 1147949 > --- > dlls/msvcp90/msvcp_main.c | 9 ++++++++- > 1 file changed, 8 insertions(+), 1 deletion(-) > > diff --git a/dlls/msvcp90/msvcp_main.c b/dlls/msvcp90/msvcp_main.c > index 489aacb..d946f3b 100644 > --- a/dlls/msvcp90/msvcp_main.c > +++ b/dlls/msvcp90/msvcp_main.c > @@ -67,7 +67,14 @@ static void init_cxx_funcs(void) > { > HMODULE hmod = GetModuleHandleA( MSVCRT_NAME(_MSVCP_VER) ); > > - if (!hmod) FIXME( "%s not loaded\n", MSVCRT_NAME(_MSVCP_VER) ); > + if (!hmod) > + { > + FIXME( "%s not loaded\n", MSVCRT_NAME(_MSVCP_VER) ); > + MSVCRT_operator_new = NULL; > + MSVCRT_operator_delete = NULL; > + MSVCRT_set_new_handler = NULL; > + return; > + } If we want to handle GetModuleHandle failure we should probably fail to load msvcpXX.dll (I'm not sure if it may ever happen unless we forget to import proper msvcrXX dll in makefile).

Jan. 6, 2014

Re: ws2_32: Fix FreeBSD compiling issue on ipx address structure

by Francois Gouget

On Sun, 5 Jan 2014, Bruno Jesus wrote: > Manually tested on PC-BSD which also shows the issue. I think it would even work to compare sipx_network and sipx_node in one go. There is already code that copies them that way in ws_sockaddr_ws2u(): memcpy(&uipx->sipx_network,wsipx->sa_netnum,sizeof(uipx->sipx_network)+sizeof(uipx->sipx_node)); So we could do (compiles but not tested): diff --git a/dlls/ws2_32/socket.c b/dlls/ws2_32/socket.c index 0391554..a92c5c3 100644 --- a/dlls/ws2_32/socket.c +++ b/dlls/ws2_32/socket.c @@ -1473,7 +1473,10 @@ static BOOL is_sockaddr_bound(const struct sockaddr *uaddr, int uaddrlen) { static const struct sockaddr_ipx emptyAddr; struct sockaddr_ipx *ipx = (struct sockaddr_ipx*) uaddr; - return ipx->sipx_port || ipx->sipx_network || memcmp(&ipx->sipx_node, &emptyAddr.sipx_node, sizeof(emptyAddr.sipx_node)); + + /* compare sipx_network and sipx_node in one go to ease portability */ + return ipx->sipx_port || + memcmp(&ipx->sipx_network, &emptyAddr.sipx_network, sizeof(emptyAddr.sipx_network) + sizeof(emptyAddr.sipx_node)); } #endif case AF_INET6: But your patch works too. -- Francois Gouget <fgouget(a)free.fr> http://fgouget.free.fr/ Research is the transformation of money to knowledge. Innovation is the transformation of knowledge to money. -- Dr. Hans Meixner

Jan. 6, 2014

Re: [PATCH] msvcp90: mark _invalid_parameter as noreturn (Coverity)

by Piotr Caban

On 01/04/14 18:25, Marcus Meissner wrote: > 1147947 Dereference after null check > > after not seeing that _invalid_parameter will not return, > so just mark the function as such, perhaps coverity then notices it. The function may return if application sets custom invalid parameter handler. I guess it shouldn't be marked as DECLSPEC_NORETURN. Probably all the places where it's used should be changed to return an error after calling _invalid_parameter.

Jan. 6, 2014

Re: Build fixes for wine-1.6.1 (FreeType2)

by Scott Ritchie

The relevant bug is http://bugs.winehq.org/show_bug.cgi?id=35010, and I've already tagged it for 1.6.x milestone, which is our way of nominating bug fixes for the stable branch :) Additionally, a couple weeks ago Alexandre hinted to me in IRC that a 1.6.2 would in fact be coming. On Thu, Jan 2, 2014 at 10:38 PM, David Naylor <dbn(a)freebsd.org> wrote: > Dear Wine-developers, > > On behalf of Gerald Pfeifer and the FreeBSD/Wine team I would like to report > that wine-1.6.1 does not build against recent versions of FreeType2. This has > been fixed in master in commit 67f9b6e3c5d2397446d39ab454f691c1ee46d2c0 by > Alexandre Julliard. > > We would like to request this commit be back-ported to the stable branch. > > Regards, > > David > > P.S. I'm not on the mailing list, please CC me in related correspondence, > thanks. > >

Jan. 6, 2014