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

Re: [0/9] wined3d: Rework card pci id selection mechanism

by Henri Verbeet

On 8 July 2014 21:38, Andrei Slăvoiu <andrei.slavoiu(a)gmail.com> wrote: > I resent parts 4 and 5 that had comments and also parts 6,7 and 9 that did not > apply cleanly after the changes to 5. > I don't mean to be rude about it, but it looks like you missed some of my comments on the patches.

July 9, 2014

Re: kernel32: Partially implement GetGeoInfo() (try2)

by Alexandre Julliard

Nikolay Sivov <nsivov(a)codeweavers.com> writes: > + ptr = &geoinfodata[geoid]; > + if (!*ptr->iso2W) { > + SetLastError(ERROR_INVALID_PARAMETER); > + return 0; > + } > + > + len = get_geoinfo_len(geoid, geotype); > + if (!data || !data_len) > + return len; You could check this at the end before copying the string instead of having a duplicate function for the length. > +INT WINAPI GetGeoInfoA(GEOID geoid, GEOTYPE geotype, LPSTR data, int data_len, LANGID lang) > +{ > + WCHAR buffW[12]; > + INT len = 0; > + > + TRACE("%d %d %p %d %d\n", geoid, geotype, data, data_len, lang); > + > + if (!data || !data_len) > + return GetGeoInfoW(geoid, geotype, NULL, 0, lang); You can't just return the W length. -- Alexandre Julliard julliard(a)winehq.org

July 9, 2014

Re: shell32: Return NULL-terminated list of arguments in CommandLineToArgvW.

by Marvin

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=7904 Your paranoid android. === w2000pro (32 bit shlexec) === shlexec.c:2447: Test failed: ShellExecuteEx(mask="0x500", file="C:\Documents and Settings\winetest\Local Settings\Temp\wt1.tmp\test file.sde") failed: rc=33 err=1400 shlexec.c:2447: Test failed: ShellExecuteEx(mask="0x500", file="C:\Documents and Settings\winetest\Local Settings\Temp\wt1.tmp\test file.sde") failed: rc=33 err=1400 shlexec.c:2447: Test failed: ShellExecuteEx(mask="0x500", file="C:\Documents and Settings\winetest\Local Settings\Temp\wt1.tmp\test file.sde") failed: rc=33 err=1400 shlexec.c:2447: Test failed: ShellExecuteEx(mask="0x500", file="C:\Documents and Settings\winetest\Local Settings\Temp\wt1.tmp\test file.sde") failed: rc=33 err=1400 shlexec.c:2447: Test failed: ShellExecuteEx(mask="0x500", file="C:\Documents and Settings\winetest\Local Settings\Temp\wt1.tmp\test file.sde") failed: rc=33 err=1400 shlexec.c:2447: Test failed: ShellExecuteEx(mask="0x500", file="C:\Documents and Settings\winetest\Local Settings\Temp\wt1.tmp\test file.sde") failed: rc=33 err=1400 === wxppro (32 bit shlexec) === shlexec.c:1222: Test failed: expected 1 args, got 4 === w8 (32 bit shlexec) === shlexec.c:139: Test failed: ShellExecute(verb="", file="C:\Users\winetest\AppData\Local\Temp\wtD88E.tmp\test file.noassoc") WaitForSingleObject returned 258 shlexec.c:139: Test failed: ShellExecute(file="C:\Users\winetest\AppData\Local\Temp\wtD88E.tmp\test file.noassoc") WaitForSingleObject returned 258 shlexec.c:1545: Test failed: ShellExecute(file="C:\Users\winetest\AppData\Local\Temp\wtD88E.tmp\test file.noassoc") failed: rc=33 err=2 shlexec.c:1577: Test failed: argvA4 expected 'C:\Users\winetest\AppData\Local\Temp\wtD88E.tmp\test file.noassoc', but key not found or empty shlexec.c:139: Test failed: ShellExecute(verb="", file="C:\Users\winetest\AppData\Local\Temp\wtD88E.tmp\test file.shlexec.noassoc") WaitForSingleObject returned 258 shlexec.c:139: Test failed: ShellExecute(file="C:\Users\winetest\AppData\Local\Temp\wtD88E.tmp\test file.shlexec.noassoc") WaitForSingleObject returned 258 shlexec.c:1545: Test failed: ShellExecute(file="C:\Users\winetest\AppData\Local\Temp\wtD88E.tmp\test file.shlexec.noassoc") failed: rc=33 err=2 shlexec.c:1577: Test failed: argvA4 expected 'C:\Users\winetest\AppData\Local\Temp\wtD88E.tmp\test file.shlexec.noassoc', but key not found or empty shlexec.c:139: Test failed: ShellExecute(verb="", file="C:\Users\winetest\AppData\Local\Temp\wtD88E.tmp\test file.noassoc", params="shlexec "C:\Users\winetest\AppData\Local\Temp\wtD88E.tmp\wtD8AF.tmp" Exec") WaitForSingleObject returned 258 shlexec.c:139: Test failed: ShellExecute(file="C:\Users\winetest\AppData\Local\Temp\wtD88E.tmp\test file.noassoc", params="shlexec "C:\Users\winetest\AppData\Local\Temp\wtD88E.tmp\wtD8AF.tmp" Exec") WaitForSingleObject returned 258 shlexec.c:2115: Test failed: ShellExecute(file="C:\Users\winetest\AppData\Local\Temp\wtD88E.tmp\test file.noassoc", params="shlexec "C:\Users\winetest\AppData\Local\Temp\wtD88E.tmp\wtD8AF.tmp" Exec") succeeded: rc=42

July 8, 2014

Re: (resend)[3/3] usp10: Implement N0: bracketing pairs

by Alexandre Julliard

Aric Stewart <aric(a)codeweavers.com> writes: > +{ > + WCHAR open_stack[iso_run->length]; > + int stack_index[iso_run->length]; That's not portable. -- Alexandre Julliard julliard(a)winehq.org

July 8, 2014

Re: [0/9] wined3d: Rework card pci id selection mechanism

by Andrei Slăvoiu

În ziua de Vin 04 Iul 2014, la 01:40:04, Andrei Slăvoiu a scris: > This series of patches will change how the PCI id of the current card is > selected for both known cards as well as the fall-back for unrecognised > ones. > > Intended changes in behaviour: > - known cards will be selected by render string only for binary drivers > - for unknown Nvidia cards running the binary driver the PCI id will be > selected by the fall-back mechanism instead of returning the hard-coded > CARD_NVIDIA_RIVA_TNT > - the fall-back mechanism will now select D3D9 SM3 capable cards when > appropriate for AMD and Intel as well. Also, D3D11 cards are now selected > for all vendors > - for Intel CARD_INTEL_845G will be returned instead of CARD_INTEL_915G > for D3D8 or less capable cards > > Question: is it intended to return an AMD card as fall-back for VMware? I resent parts 4 and 5 that had comments and also parts 6,7 and 9 that did not apply cleanly after the changes to 5.

July 8, 2014

Re: [2/5] server: A new function "set_sd_defaults_from_token" that sets the security descriptor along with a token that will be used to gather defaults, instead of always using the primary token.

by Joris van der Wel

Alright, I will resubmit without that check. I thought "current" being NULL is a valid condition because when I was experiment with default security descriptors I noticed that current is sometimes NULL when objects are created. I will address that in another way when I get around to that. On Tue, Jul 8, 2014 at 7:08 PM, Alexandre Julliard <julliard(a)winehq.org> wrote: > Joris van der Wel <joris(a)jorisvanderwel.com> writes: > >> +/** Set the security descriptor using the current primary token for defaults. */ >> +int default_set_sd( struct object *obj, const struct security_descriptor *sd, >> + unsigned int set_info ) >> +{ >> + return set_sd_defaults_from_token( obj, sd, set_info, current ? current->process->token : NULL ); > > There shouldn't be any need to test the current pointer for null. If > it's needed, that probably indicates some other problem. > > -- > Alexandre Julliard > julliard(a)winehq.org

July 8, 2014

Re: [2/5] server: A new function "set_sd_defaults_from_token" that sets the security descriptor along with a token that will be used to gather defaults, instead of always using the primary token.

by Alexandre Julliard

Joris van der Wel <joris(a)jorisvanderwel.com> writes: > +/** Set the security descriptor using the current primary token for defaults. */ > +int default_set_sd( struct object *obj, const struct security_descriptor *sd, > + unsigned int set_info ) > +{ > + return set_sd_defaults_from_token( obj, sd, set_info, current ? current->process->token : NULL ); There shouldn't be any need to test the current pointer for null. If it's needed, that probably indicates some other problem. -- Alexandre Julliard julliard(a)winehq.org

July 8, 2014

Re: FOSDEM and Wineconf

by Michael Stefaniuc

On 07/08/2014 05:41 PM, Jeremy White wrote: > The FOSDEM call for rooms is out for next year: > https://fosdem.org/2015/news/2014-07-01-call-for-participation/ > > I left FOSDEM this year feeling that it was a great two years, but not > something that made sense for 2015, at least not in the same form (a > dedicated dev room). To keep in line with the tradition next years WineConf should happen in the US. I have heard that Minneapolis is beautiful in January/February ;) > I had the sense that a table like the distros had, maybe with a few > talks in the general tracks, made more sense. A table needs 3+ people for each day. I would be willing to do a talk at FOSDEM; either an updated - "Is Wine Still Relevant" talk that I held at LinuxTag or a - "Getting your bad code past Alexandre" talk that I promised Alexandre to hold at the next WineConf. > But the bottom line is that I'm not going to go in 2015, and as such, it > doesn't make sense for me to do the coordination of it. > > Is anyone else willing to pick this up? Maybe, I need to think a little bit more about it. > Anyone else have ideas for if and/or how we should participate in FOSDEM > 2015? bye michael

July 8, 2014

FOSDEM and Wineconf

by Jeremy White

Hey Folks, The FOSDEM call for rooms is out for next year: https://fosdem.org/2015/news/2014-07-01-call-for-participation/ I left FOSDEM this year feeling that it was a great two years, but not something that made sense for 2015, at least not in the same form (a dedicated dev room). I had the sense that a table like the distros had, maybe with a few talks in the general tracks, made more sense. But the bottom line is that I'm not going to go in 2015, and as such, it doesn't make sense for me to do the coordination of it. Is anyone else willing to pick this up? Anyone else have ideas for if and/or how we should participate in FOSDEM 2015? Cheers, Jeremy

July 8, 2014

Re: Wine development questions

by Vincent Povirk

> Implementing the correct behaviour is not the best choice if the rest of the > structure suffers functional regressions due to it by not being prepared. I > know the 1.7.x versions are development releases but if something is not > done its go into the stable release. My understanding is that Ken's work also fixed real applications, and therefore reverting it would also create regressions. At least when we take the correct approach, the bugs can be fixed without introducing new ones. We shouldn't be making stable releases when a feature like WGL is in the middle of major work. > This bug http://bugs.winehq.org/show_bug.cgi?id=33125 and the latter fix. If it's the patch adding DECLSPEC_HOTPATCH to CreateSemaphoreExW, that's already in wine and has been since 1.5.28. "Patch which disables optimization for CreateEventExW" would be rejected in seconds. It looks like something that only works by coincidence. It seems Anastasius Focht figured out what was really going on and wrote this comment: http://bugs.winehq.org/show_bug.cgi?id=33125#c110 . So it just needs someone to write the patch and maybe add a test.

July 8, 2014