- wine-devel - winehq.org

Re: [v2] comctl32/taskdialog: Implement nDefaultButton and add tests
by Fabian Maurer 06 Nov '17

06 Nov '17

I reworked my patch according to your suggestions. >>> We already iterate over custom buttons and check every common button >>> flag before this point, so it shouldn't be necessary to have another loop. >> >> I'm open to suggestion, but how would you do that? >> >> We could add three checks each time before taskdialog_init_button is called, >> or alternatively one check in taskdialog_init_button. Like giving it a pointer >> to default_button, and checking if the id equal the taskdialog default button. >> But I find my current solution to be cleaner to be honest, what's your >> suggestion? >Yes, something like that. taskdialog_init_button() has all the info to > do that. In the beginning I wanted to have default_button be -1 if no button was found yet, but as the tests show this is impossible since windows accepts a negative ID. Now we have to pass 2 pointers around.. Your thoughts on that? I find that the second loop makes things a lot cleaner. Regards, Fabian Maurer

2 1

Re: [PATCH] msvcp100: add time_get::get function with format
by Piotr Caban 06 Nov '17

06 Nov '17

Hi, On 11/05/17 13:26, Vijay Kiran Kamuju wrote: > diff --git a/dlls/msvcp110/msvcp110.spec b/dlls/msvcp110/msvcp110.spec > index bf54ac5f53..efb13361ff 100644 > --- a/dlls/msvcp110/msvcp110.spec > +++ b/dlls/msvcp110/msvcp110.spec > @@ -2599,8 +2599,8 @@ > @ thiscall -arch=i386 ?get@?$time_get@DV?$istreambuf_iterator@DU?$char_traits@D@std@@@std@@@std@@QBE?AV?$istreambuf_iterator@DU?$char_traits@D@std@@@2@V32@0AAVios_base@2@AAHPAUtm@@DD@Z(ptr ptr ptr long ptr long ptr ptr ptr long long) time_get_char_get > @ cdecl -arch=win64 ?get@?$time_get@DV?$istreambuf_iterator@DU?$char_traits@D@std@@@std@@@std@@QEBA?AV?$istreambuf_iterator@DU?$char_traits@D@std@@@2@V32@0AEAVios_base@2@AEAHPEAUtm@@DD@Z(ptr ptr ptr long ptr long ptr ptr ptr long long) time_get_char_get > @ stub -arch=arm ?get@?$time_get@DV?$istreambuf_iterator@DU?$char_traits@D@std@@@std@@@std@@QBA?AV?$istreambuf_iterator@DU?$char_traits@D@std@@@2@V32@0AAVios_base@2@AAHPAUtm@@PBD4@Z > -@ stub -arch=i386 ?get@?$time_get@DV?$istreambuf_iterator@DU?$char_traits@D@std@@@std@@@std@@QBE?AV?$istreambuf_iterator@DU?$char_traits@D@std@@@2@V32@0AAVios_base@2@AAHPAUtm@@PBD4@Z > -@ stub -arch=win64 ?get@?$time_get@DV?$istreambuf_iterator@DU?$char_traits@D@std@@@std@@@std@@QEBA?AV?$istreambuf_iterator@DU?$char_traits@D@std@@@2@V32@0AEAVios_base@2@AEAHPEAUtm@@PEBD4@Z > +@ thiscall -arch=i386 ?get@?$time_get@DV?$istreambuf_iterator@DU?$char_traits@D@std@@@std@@@std@@QBE?AV?$istreambuf_iterator@DU?$char_traits@D@std@@@2@V32@0AAVios_base@2@AAHPAUtm@@PBD4@Z(ptr ptr ptr long ptr long ptr ptr ptr ptr ptr) time_get_char_get_fmt > +@ cdecl -arch=win64 ?get@?$time_get@DV?$istreambuf_iterator@DU?$char_traits@D@std@@@std@@@std@@QEBA?AV?$istreambuf_iterator@DU?$char_traits@D@std@@@2@V32@0AEAVios_base@2@AEAHPEAUtm@@PEBD4@Z(ptr ptr ptr long ptr long ptr ptr ptr ptr ptr) time_get_char_get_fmt Please also add arm export. Also win64 arguments needs fixing (structures are passed as pointers in this case). > +istreambuf_iterator_char* __thiscall time_get_char_get_fmt(const time_get_char *this, > + istreambuf_iterator_char *ret, istreambuf_iterator_char s, istreambuf_iterator_char e, > + ios_base *base, int *err, struct tm *t, char *fmtstart, char *fmtend) The fmtstart and fmtend arguments should be const. Thanks, Piotr

1 0

Re: [v2 PATCH] cmd: Implement builtin mklink command
by Alexandre Julliard 06 Nov '17

06 Nov '17

Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> writes: > @@ -310,6 +310,15 @@ CHOICE is mainly used to build a menu selection in a batch file.\n" > "EXIT terminates the current command session and returns to the operating\n\ > system or shell from which you invoked cmd.\n" > > + WCMD_MKLINK, > +"Creates a symbolic link.\n\ > +MKLINK [[/D] | [/H] | [/J]] Link Target\n\ > +/D Creates a directory symbolic link. Default is a file symbolic link.\n\ > +/H Creates a hard link instead of a symbolic link.\n\ > +/J Creates a Directory Junction.\n\ > +Link specifies the new symbolic link name.\n\ > +Target specifies the path (relative or absolute) that the new link refers to.\n" This is a straight copy from the Microsoft version. Not only is this not allowed, it also doesn't fit in with the description style of the other commands. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [PATCH] user32: check GetClipboardFormatName length
by Marvin 06 Nov '17

06 Nov '17

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=33944 Your paranoid android. === w8 (32 bit clipboard) === clipboard.c:402: Test failed: GlobalGetAtomNameA should fail clipboard.c:403: Test failed: err -559038737

2 1

Re: [v3] cmd: Added support for "&&" and "||" condition checking.
by Marvin 06 Nov '17

06 Nov '17

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=33950 Your paranoid android. === build (build) === Patch failed to apply

1 0

Re: [PATCH] ole32: handle length=0 in save datacache
by Dmitry Timoshkov 05 Nov '17

05 Nov '17

Marcus Meissner <marcus(a)jet.franken.de> wrote: > The ole2 test crashes for me, reason is that length is 0 > and we use length-1 for accessing whats returned. ... > + if (!length) > + return DV_E_CLIPFORMAT; The test intentionally creates the storage streams with length field set to 0, and as the tests show this is not an error. So the fix should be probably different. -- Dmitry.

2 1

Re: [PATCH] comctl32/taskdialog: Implement nDefaultButton and add tests
by Nikolay Sivov 05 Nov '17

05 Nov '17

On 04.11.2017 18:22, Fabian Maurer wrote: > Signed-off-by: Fabian Maurer <dark.shadow4(a)web.de> > --- > dlls/comctl32/taskdialog.c | 21 +++++-- > dlls/comctl32/tests/taskdialog.c | 115 +++++++++++++++++++++++++++++++++++++++ > 2 files changed, 130 insertions(+), 6 deletions(-) > > diff --git a/dlls/comctl32/taskdialog.c b/dlls/comctl32/taskdialog.c > index 5c4af2ab03..4f80311a61 100644 > --- a/dlls/comctl32/taskdialog.c > +++ b/dlls/comctl32/taskdialog.c > @@ -151,7 +151,7 @@ static void taskdialog_get_text_extent(const struct taskdialog_template_desc *de > } > > static unsigned int taskdialog_add_control(struct taskdialog_template_desc *desc, WORD id, const WCHAR *class, > - HINSTANCE hInstance, const WCHAR *text, short x, short y, short cx, short cy) > + HINSTANCE hInstance, const WCHAR *text, DWORD style, short x, short y, short cx, short cy) > { > struct taskdialog_control *control = Alloc(sizeof(*control)); > unsigned int size, class_size, text_size; > @@ -178,7 +178,7 @@ static unsigned int taskdialog_add_control(struct taskdialog_template_desc *desc > control->template = template = Alloc(size); > control->template_size = size; > > - template->style = WS_VISIBLE; > + template->style = WS_VISIBLE | style; > template->dwExtendedStyle = 0; > template->x = x; > template->y = y; > @@ -206,7 +206,7 @@ static unsigned int taskdialog_add_static_label(struct taskdialog_template_desc > taskdialog_get_text_extent(desc, str, TRUE, &sz); > > desc->dialog_height += DIALOG_SPACING; > - size = taskdialog_add_control(desc, id, WC_STATICW, desc->taskconfig->hInstance, str, DIALOG_SPACING, > + size = taskdialog_add_control(desc, id, WC_STATICW, desc->taskconfig->hInstance, str, 0, DIALOG_SPACING, > desc->dialog_height, sz.cx, sz.cy); > desc->dialog_height += sz.cy + DIALOG_SPACING; > return size; > @@ -268,6 +268,7 @@ static unsigned int taskdialog_add_buttons(struct taskdialog_template_desc *desc > unsigned int location_x, *line_widths, alignment = ~0u; > const TASKDIALOGCONFIG *taskconfig = desc->taskconfig; > struct taskdialog_button_desc *buttons; > + int default_button; > > /* Allocate enough memory for the custom and the default buttons. Maximum 6 default buttons possible. */ > buttons_size = 6; > @@ -340,18 +341,26 @@ static unsigned int taskdialog_add_buttons(struct taskdialog_template_desc *desc > alignment = new_alignment; > } > > + /* Check if the default button is actually valid, if no default to the first created button */ > + default_button = buttons[0].id; > + for (i = 0; i < count; i++) > + if (buttons[i].id == taskconfig->nDefaultButton) > + default_button = taskconfig->nDefaultButton; > + We already iterate over custom buttons and check every common button flag before this point, so it shouldn't be necessary to have another loop. > @@ -583,7 +592,7 @@ HRESULT WINAPI TaskDialogIndirect(const TASKDIALOGCONFIG *taskconfig, int *butto > Free(template); > > if (button) *button = ret; > - if (radio_button) *radio_button = taskconfig->nDefaultButton; > + if (radio_button) *radio_button = FALSE; > if (verification_flag_checked) *verification_flag_checked = TRUE; > Same as with verification flag, we don't implement radio buttons currently so it's better to leave it as is to avoid possible regressions until we add support for them. Also you're supposed to return button id, not binary flags. > > +static TASKDIALOG_BUTTON* buttons_make(void) > +{ > + static const WCHAR str_format[] = {'%','0','2','d',0}; > + static TASKDIALOG_BUTTON buttons[TEST_NUM_BUTTONS]; > + static WCHAR titles[TEST_NUM_BUTTONS * 3]; /* Each button has two digits as title, plus null-terminator */ > + int i; > + > + for (i = 0; i < TEST_NUM_BUTTONS; i++) > + { > + WCHAR *text = &titles[i * 3]; > + wsprintfW(text, str_format, i); > + > + buttons[i].pszButtonText = text; > + buttons[i].nButtonID = ID_START_BUTTON + i; > + } > + return buttons; > +} > + Please merge this to test function.

2 2

Unhandled exception
by Ran Shalit 05 Nov '17

05 Nov '17

Hello, I have an issue trying to run application with wine, in which I get the below exception. It first (before getting this exception) said that mfc100u.dll is required, so I had to download and put it in the same folder as .exe, but then trying to run the application, I got this exception. I use wine 1.8. Thank you for any suggestion. user@user-VirtualBox:~/2.1.2$ wine record_channels.exe system.reg is not a valid registry file userdef.reg is not a valid registry file user.reg is not a valid registry file err:ole:marshal_object couldn't get IPSFactory buffer for interface {00000131-0000-0000-c000-000000000046} err:ole:marshal_object couldn't get IPSFactory buffer for interface {6d5140c1-7436-11ce-8034-00aa006009fa} err:ole:StdMarshalImpl_MarshalInterface Failed to create ifstub, hres=0x80004002 err:ole:CoMarshalInterface Failed to marshal the interface {6d5140c1-7436-11ce-8034-00aa006009fa}, 80004002 err:ole:get_local_server_stream Failed: 80004002 fixme:thread:GetThreadPreferredUILanguages 52, 0x32f9d8, 0x32fc54 0x32f9e0 fixme:heap:RtlSetHeapInformation (nil) 1 (nil) 0 stub fixme:thread:GetThreadPreferredUILanguages 52, 0x32f680, 0x32f8fc 0x32f688 wine: Unhandled exception 0xc06d007e in thread 12 at address 0x7b83c3ae (thread 0012), starting debugger... Unhandled exception: 0xc06d007e in 32-bit code (0x7b83c3ae). Register dump: CS:0073 SS:007b DS:007b ES:007b FS:0033 GS:003b EIP:7b83c3ae ESP:0086dfc4 EBP:0086e038 EFLAGS:00000287( - -- I S - -P-C) EAX:7b827a69 EBX:7b8c0000 ECX:0086e060 EDX:0086dfe4 ESI:c06d007e EDI:00000000 Stack dump: 0x0086dfc4: 0086e0b0 00000004 7b85ac31 c06d007e 0x0086dfd4: 00000000 00000000 7b83c3ae 00000001 0x0086dfe4: 0086e060 00000000 00000000 4d430003 0x0086dff4: 7b85abfc 0086e020 7b8c0000 0043db3c 0x0086e004: 00000000 0086e038 7b85ac9d 00000000 0x0086e014: 00000000 0086e038 7b85ac9d 0042d7b0 Backtrace: =>0 0x7b83c3ae in kernel32 (+0x2c3ae) (0x0086e038) 1 0x00411033 in record_channels (+0x11032) (0x0086e0a4) 2 0x00410e0e in record_channels (+0x10e0d) (0x0086e600) 3 0x00427e10 in record_channels (+0x27e0f) (0x0086e7a0) 4 0x004285f6 in record_channels (+0x285f5) (0x0086e9cc) 5 0x00428e09 in record_channels (+0x28e08) (0x0086e9d8) 6 0x7bc84470 call_thread_func_wrapper+0xb() in ntdll (0x0086e9e8) 7 0x7bc8763d call_thread_func+0x7c() in ntdll (0x0086eab8) 8 0x7bc8444e RtlRaiseException+0x21() in ntdll (0x0086ead8) 9 0x7bc8fe69 in ntdll (+0x7fe68) (0x0086f358) 10 0xb74fff70 start_thread+0xcf() in libpthread.so.0 (0x0086f428) 11 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 12 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 13 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 14 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 15 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 16 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 17 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 18 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 19 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 20 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 21 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 22 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 23 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 24 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 25 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 26 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 27 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 28 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 29 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 30 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 31 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 32 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 33 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 34 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 35 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 36 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 37 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 38 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 39 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 40 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 41 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 42 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 43 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 44 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 45 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 46 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 47 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 48 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 49 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 50 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 51 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 52 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 53 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 54 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 55 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 56 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 57 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 58 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 59 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 60 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 61 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 62 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 63 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 64 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 65 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 66 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 67 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 68 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 69 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 70 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 71 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 72 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 73 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 74 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 75 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 76 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 77 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 78 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 79 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 80 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 81 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 82 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 83 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 84 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 85 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 86 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 87 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 88 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 89 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 90 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 91 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 92 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 93 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 94 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 95 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 96 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 97 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 98 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 99 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 100 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 101 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 102 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 103 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 104 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 105 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 106 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 107 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 108 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 109 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 110 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 111 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 112 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 113 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 114 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 115 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 116 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 117 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 118 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 119 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 120 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 121 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 122 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 123 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 124 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 125 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 126 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 127 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 128 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 129 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 130 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 131 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 132 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 133 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 134 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 135 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 136 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 137 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 138 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 139 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 140 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 141 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 142 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 143 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 144 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 145 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 146 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 147 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 148 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 149 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 150 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 151 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 152 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 153 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 154 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 155 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 156 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 157 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 158 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 159 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 160 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 161 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 162 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 163 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 164 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 165 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 166 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 167 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 168 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 169 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 170 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 171 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 172 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 173 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 174 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 175 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 176 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 177 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 178 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 179 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 180 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 181 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 182 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 183 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 184 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 185 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 186 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 187 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 188 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 189 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 190 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 191 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 192 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 193 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 194 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 195 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 196 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 197 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 198 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 199 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 200 0xb743050e __clone+0x5d() in libc.so.6 (0x00000000) 0x7b83c3ae: subl $4,%esp Modules: Module Address Debug info Name (79 modules) PE 400000- 44d000 Export record_channels PE 10000000-1000a000 Deferred recordings_channels_dll PE 785f0000-78a22000 Deferred mfc100u ELF 7b800000-7ba6a000 Dwarf kernel32<elf> \-PE 7b810000-7ba6a000 \ kernel32 ELF 7bc00000-7bcf3000 Dwarf ntdll<elf> \-PE 7bc10000-7bcf3000 \ ntdll ELF 7bf00000-7bf04000 Deferred <wine-loader> ELF 7dacc000-7db10000 Deferred usp10<elf> \-PE 7dad0000-7db10000 \ usp10 ELF 7db9a000-7dbd3000 Deferred uxtheme<elf> \-PE 7dba0000-7dbd3000 \ uxtheme ELF 7dbd3000-7dbd9000 Deferred libxfixes.so.3 ELF 7dbd9000-7dbe4000 Deferred libxcursor.so.1 ELF 7dbe4000-7dbf4000 Deferred libxi.so.6 ELF 7dbf4000-7dbf8000 Deferred libxcomposite.so.1 ELF 7dbf8000-7dc03000 Deferred libxrandr.so.2 ELF 7dc03000-7dc0e000 Deferred libxrender.so.1 ELF 7dc0e000-7dc14000 Deferred libxxf86vm.so.1 ELF 7dc14000-7dc18000 Deferred libxinerama.so.1 ELF 7dc18000-7dc1f000 Deferred libxdmcp.so.6 ELF 7dc1f000-7dc23000 Deferred libxau.so.6 ELF 7dc23000-7dc45000 Deferred libxcb.so.1 ELF 7dc45000-7dd79000 Deferred libx11.so.6 ELF 7dd79000-7dd8c000 Deferred libxext.so.6 ELF 7dd90000-7dda6000 Deferred dwmapi<elf> \-PE 7dda0000-7dda6000 \ dwmapi ELF 7dda8000-7de3c000 Deferred winex11<elf> \-PE 7ddb0000-7de3c000 \ winex11 ELF 7de3c000-7de61000 Deferred imm32<elf> \-PE 7de40000-7de61000 \ imm32 ELF 7dffa000-7e023000 Deferred libexpat.so.1 ELF 7e023000-7e05e000 Deferred libfontconfig.so.1 ELF 7e05e000-7e086000 Deferred libpng12.so.0 ELF 7e086000-7e0a0000 Deferred libz.so.1 ELF 7e0a0000-7e140000 Deferred libfreetype.so.6 ELF 7e140000-7e162000 Deferred libtinfo.so.5 ELF 7e162000-7e187000 Deferred libncurses.so.5 ELF 7e1a3000-7e1df000 Deferred ws2_32<elf> \-PE 7e1b0000-7e1df000 \ ws2_32 ELF 7e201000-7e310000 Deferred msvcp100<elf> \-PE 7e240000-7e310000 \ msvcp100 ELF 7e310000-7e55e000 Deferred shell32<elf> \-PE 7e320000-7e55e000 \ shell32 ELF 7e55e000-7e572000 Deferred msimg32<elf> \-PE 7e560000-7e572000 \ msimg32 ELF 7e572000-7e67e000 Deferred comctl32<elf> \-PE 7e580000-7e67e000 \ comctl32 ELF 7e67e000-7e6f9000 Deferred shlwapi<elf> \-PE 7e690000-7e6f9000 \ shlwapi ELF 7e6f9000-7e7c2000 Deferred msvcr100<elf> \-PE 7e710000-7e7c2000 \ msvcr100 ELF 7e7c2000-7e7ed000 Deferred msacm32<elf> \-PE 7e7d0000-7e7ed000 \ msacm32 ELF 7e7ed000-7e872000 Deferred rpcrt4<elf> \-PE 7e800000-7e872000 \ rpcrt4 ELF 7e872000-7e9b7000 Deferred ole32<elf> \-PE 7e890000-7e9b7000 \ ole32 ELF 7e9b7000-7ea33000 Deferred advapi32<elf> \-PE 7e9c0000-7ea33000 \ advapi32 ELF 7ea33000-7eb54000 Deferred gdi32<elf> \-PE 7ea40000-7eb54000 \ gdi32 ELF 7eb54000-7ecb1000 Deferred user32<elf> \-PE 7eb70000-7ecb1000 \ user32 ELF 7ecb1000-7ed6c000 Deferred winmm<elf> \-PE 7ecc0000-7ed6c000 \ winmm ELF 7ef6c000-7ef79000 Deferred libnss_files.so.2 ELF 7ef79000-7ef85000 Deferred libnss_nis.so.2 ELF 7ef85000-7ef9e000 Deferred libnsl.so.1 ELF 7ef9e000-7efe4000 Deferred libm.so.6 ELF 7efe6000-7f000000 Deferred version<elf> \-PE 7eff0000-7f000000 \ version ELF b7344000-b74f3000 Dwarf libc.so.6 ELF b74f3000-b74f8000 Deferred libdl.so.2 ELF b74f9000-b7515000 Dwarf libpthread.so.0 ELF b7517000-b7520000 Deferred libnss_compat.so.2 ELF b7531000-b76e7000 Dwarf libwine.so.1 ELF b76e9000-b770b000 Deferred ld-linux.so.2 ELF b770b000-b770c000 Deferred [vdso].so Threads: process tid prio (all id:s are in hex) 00000008 (D) Z:\home\user\2.1.2\record_channels.exe 00000013 0 <== 00000012 0 00000009 0 0000000c services.exe 0000000f 0 0000000e 0 0000000d 0 00000010 explorer.exe 00000011 0 System information: Wine build: wine-1.8 Platform: i386 Version: Windows XP Host system: Linux Host version: 3.16.0-30-generic user@user-VirtualBox:~/2.1.2$ Best Regards, Ran

1 1

Re: xmllite: Avoid warning in test_read_state related to -1 not being in XmlReadState.
by Nikolay Sivov 05 Nov '17

05 Nov '17

On 11/03/2017 01:26 AM, Gerald Pfeifer wrote: > Hi Nikolay and others, > > the patch below avoids a warning of the kind > > reader.c:276:20: warning: comparison of constant -1 with expression of type > 'XmlReadState' (aka 'enum XmlReadState') is always false > [-Wtautological-constant-out-of-range-compare] > if (exp_broken == -1) > ~~~~~~~~~~ ^ ~~ > > which clang issues, so I hope it can be included? > > Gerald > > > Signed-off-by: Gerald Pfeifer <gerald(a)pfeifer.com> > --- > dlls/xmllite/tests/reader.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/dlls/xmllite/tests/reader.c b/dlls/xmllite/tests/reader.c > index 0f03c3b42e..2a93a76ddb 100644 > --- a/dlls/xmllite/tests/reader.c > +++ b/dlls/xmllite/tests/reader.c > @@ -273,7 +273,7 @@ static void test_read_state(IXmlReader *reader, XmlReadState expected, > state = -1; /* invalid state value */ > IXmlReader_GetProperty(reader, XmlReaderProperty_ReadState, &state); > > - if (exp_broken == -1) > + if (exp_broken == (XmlReadState)-1) > broken_state = FALSE; > else > broken_state = broken(exp_broken == state); Hi, Gerald. Is this warning enabled by default in some release clang version? I'm not necessarily opposed in silencing it like that, but I imagine you'll see many more instances like that, and if it's not default set of warnings I don't know if we should bother.

2 3

Re: [PATCH] oleaut32: fixed 64bit crash on tmarshal test
by Donna Whisnant 04 Nov '17

04 Nov '17

> On 04.11.2017 17:28, Marcus Meissner wrote: > >> > >> Any idea why is this specific to 64 bit? > > > > Well, the call fails in because we have no 64bit typelib threaded marshaller. > > > > ... > > > > We could exit PSFacBuf_CreateProxy earlier on x86_64 with a different error. > > > > (or implement the typelib marshaller ;) > > Right, there is a fix for that linked from bug > https://bugs.winehq.org/show_bug.cgi?id=26768. It would be nice to have > that for 3.0. > And while you guys are working on it, it would also be nice to have typelib marshaller support for ARM in 3.0 too. ;-)

1 0

Re: [PATCH] oleaut32: fixed 64bit crash on tmarshal test
by Nikolay Sivov 04 Nov '17

04 Nov '17

On 03.11.2017 14:15, Marcus Meissner wrote: > Test will fail now, but no longer crash. > > Signed-off-by: Marcus Meissner <marcus(a)jet.franken.de> > --- > dlls/oleaut32/tests/tmarshal.c | 3 ++- > 1 file changed, 2 insertions(+), 1 deletion(-) > > diff --git a/dlls/oleaut32/tests/tmarshal.c b/dlls/oleaut32/tests/tmarshal.c > index d2b8dc92d6..f19cbf451b 100644 > --- a/dlls/oleaut32/tests/tmarshal.c > +++ b/dlls/oleaut32/tests/tmarshal.c > @@ -2010,7 +2010,8 @@ static void test_external_connection(void) > todo_wine > ok(external_connections == 2, "external_connections = %d\n", external_connections); > > - ITestSecondDisp_Release(second); > + if (hres == S_OK) > + ITestSecondDisp_Release(second); > todo_wine > ok(external_connections == 2, "external_connections = %d\n", external_connections); > > Any idea why is this specific to 64 bit?

2 2

Re: ws2_32/tests: Add some tests for sockets bound to a completion port.
by Alexandre Julliard 04 Nov '17

04 Nov '17

Dmitry Timoshkov <dmitry(a)baikal.ru> writes: > + if (pNtClose) > + { > + pNtClose((HANDLE)src); > + break; > + } > + win_skip("NtClose is not available on this platform"); > + how_to_close = 1; The test and skip are not necessary, we no longer support win9x in tests. > + ok(ovl_iocp->Internal == STATUS_CONNECTION_ABORTED || STATUS_LOCAL_DISCONNECT /* XP */, "got %#lx\n", ovl_iocp->Internal); The second part of the || is always true so this is not testing anything. > + ok(ovl_iocp->Internal == STATUS_CONNECTION_ABORTED || STATUS_LOCAL_DISCONNECT /* XP */, "got %#lx\n", ovl_iocp->Internal); Same here. -- Alexandre Julliard julliard(a)winehq.org

2 1

Setting a source directory in winedbg
by Mihai Todor 03 Nov '17

03 Nov '17

Hi Everyone, I noticed that I can set the source directory via the `dir` command in winedbg, but, for some reason, it rejects absolute paths, like Z:\bla\blah, claiming that ":" is a syntax error. I tried to enclose the path in quotes / double quotes, but then the quote / double quote is rejected as a syntax error. I don't have sufficient knowledge of bison to grok the code in programs/winedbg/dbg.y, so is there a way to escape the colon symbol? As a workaround, I mapped my build directory under Wine, but that is really annoying. Thank you, Mihai Todor

1 1

Re: [PATCH v2] kernel32/tests: Detect and handle a redirected stdin.
by Zebediah Figura 03 Nov '17

03 Nov '17

On 11/03/2017 05:43 AM, Alex Henrie wrote: > Signed-off-by: Alex Henrie <alexhenrie24(a)gmail.com> > --- > v2: Simply skip the tests if the input is redirected. > --- > dlls/kernel32/tests/console.c | 5 +++++ > 1 file changed, 5 insertions(+) > > diff --git a/dlls/kernel32/tests/console.c b/dlls/kernel32/tests/console.c > index a14bc45034..926c052686 100644 > --- a/dlls/kernel32/tests/console.c > +++ b/dlls/kernel32/tests/console.c > @@ -2582,6 +2582,11 @@ static void test_ReadConsole(void) > > SetLastError(0xdeadbeef); > ret = GetFileSize(std_input, NULL); > + if (GetLastError() == 0xdeadbeef) > + { > + skip("stdin is redirected\n"); > + return; > + } > ok(ret == INVALID_FILE_SIZE, "expected INVALID_FILE_SIZE, got %#x\n", ret); > ok(GetLastError() == ERROR_INVALID_HANDLE || > GetLastError() == ERROR_INVALID_FUNCTION, /* Win 8, 10 */ > Couldn't we just skip those two tests? The others don't seem to fail under that circumstance.

2 1

Re: Józef Kucia : wined3d: Implement indirect draws.
by Ken Thomases 03 Nov '17

03 Nov '17

Hi Józef, On Nov 2, 2017, at 5:28 PM, Alexandre Julliard <julliard(a)winehq.org> wrote: > > Module: wine > Branch: master > Commit: f4aa2911a689f95783fe2061a516440dc3d3f39e > URL: http://source.winehq.org/git/wine.git/?a=commit;h=f4aa2911a689f95783fe2061a… > > Author: Józef Kucia <jkucia(a)codeweavers.com> > Date: Wed Nov 1 22:08:42 2017 +0100 > > wined3d: Implement indirect draws. If I'm reading this commit correctly, the new indexed indirect draw code does not make use of state->index_offset (via idx_data, for example). Is that how it should be? Or should DrawIndexedInstancedIndirect() respect the index buffer offset? Thanks, Ken

2 1

Re: [PATCH] ieframe: avoid crash on missing IID_IWebBrowser2 interface
by Jacek Caban 03 Nov '17

03 Nov '17

On 03.11.2017 12:21, Marcus Meissner wrote: > diff --git a/dlls/ieframe/tests/ie.c b/dlls/ieframe/tests/ie.c > index ac055e248d..459854cb64 100644 > --- a/dlls/ieframe/tests/ie.c > +++ b/dlls/ieframe/tests/ie.c > @@ -264,6 +264,8 @@ static void test_InternetExplorer(void) > > hres = IUnknown_QueryInterface(unk, &IID_IWebBrowser2, (void**)&wb); > ok(hres == S_OK, "Could not get IWebBrowser2 interface: %08x\n", hres); > + if (hres != S_OK) > + return; You leak unk and don't unadvise connection point here. I think that moving QueryInterface() call before advise_cp() call and properly releasing unk in error case would do the trick. Thanks, Jacek

1 0

Learning C and Wine
by Luke Short 03 Nov '17

03 Nov '17

Greetings everyone, I am looking to learn how to program in C and, more specifically, want to focus on how to do so with Wine development. I've already been contributing to this project via testing applications and updating/submitting bug reports and I hope to contribute even more. As a developer, I've been professionally making Python programs for years now and I figure it's about time I really learn C. Ideally I will eventually contribute more of my time and code back into the Wine project. Beyond just Wine, knowing C should also help give me an edge on creating high-performance functions using Python's C API. My question is: are there any recommended resources for learning C from the ground up and also using that knowledge to help with Wine? What are some materials that you have used to get yourself started with C and/or Wine beyond the Wine Developer's Guide? I understand that learning C will be more of a gradual thing over the course of many years. And, like most things, mastery is an ongoing process. I am looking to get my feet wet so that I can eventually dive into the Wine development pool. Thank you guys for any advice you can give to a newcomer! Sincerely, Luke Short

2 1

RFC winevulkan design
by Roderick Colenbrander 03 Nov '17

03 Nov '17

Hi all, Based on conversations on IRC, I reconsidered the design of the ICD. The idea is to have a thin winevulkan proxy library. It hosts the thunks and loads the appropriate ICD based the HWND similar how opengl32 operates. In general the driver (e.g. winex11) is responsible for loading the required function pointers used by the thunks. These function pointers are stored in the Vulkan objects (Vulkan operates exactly the same way). The driver can decide by itself for which APIs it needs to provide a custom implementation. Usually this are at least vkCreateInstance, vkCreateDevice and others. I wrote some pseudo code into this RFC patch to illustrate the design. Here and there I added some detailed comments and questions. The code obviously does not compile and variable/function names are just for illustration purposes. Let me know your thoughts. Thanks, Roderick --- dlls/winevulkan/vulkan.c | 100 ++++++++++++++++++++++++++++++++++++++ dlls/winevulkan/vulkan_thunks.c | 80 ++++++++++++++++++++++++++++++ dlls/winex11.drv/vulkan.c | 77 +++++++++++++++++++++++++++++ dlls/winex11.drv/winex11.drv.spec | 4 ++ include/wine/vulkan_driver.h | 78 +++++++++++++++++++++++++++++ 5 files changed, 339 insertions(+) create mode 100644 dlls/winevulkan/vulkan.c create mode 100644 dlls/winevulkan/vulkan_thunks.c create mode 100644 dlls/winex11.drv/vulkan.c create mode 100644 include/wine/vulkan_driver.h diff --git a/dlls/winevulkan/vulkan.c b/dlls/winevulkan/vulkan.c new file mode 100644 index 0000000000..fd73acaf10 --- /dev/null +++ b/dlls/winevulkan/vulkan.c @@ -0,0 +1,100 @@ +#include "wine/vulkan.h" +#include "wine/vulkan_driver.h" + +/* Vulkan global functions. Stateless functions of which there very few, + * but which we need to call directly outside of dispatch table to the driver. + */ + +MAKE_FUNCPTR(vkCreateInstance) +MAKE_FUNCPTR(vkEnumerateInstanceExtensionProperties) +MAKE_FUNCPTR(vkGetInstanceProcAddr) + +const struct vulkan_func vk_global_dispatch_table[] = { + {"vkCreateInstance", NULL, &wine_vkCreateInstance}, + {"vkEnumerateInstanceExtensionProperties", NULL, &wine_vkEnumerateInstanceExtensionProperties}, + {"vkGetInstanceProcAddr", NULL, &wine_vkGetInstanceProcAddr}, +}; +const int vk_global_dispatch_table_size = 3; + +void* wine_vk_get_global_proc_addr(pName) +{ + //search global dispatch table +} + + +static BOOL vulkan_initialized = FALSE; +static struct vulkan_icd_funcs *icd_funcs = NULL; + +static BOOL wine_vk_init(void) +{ + HDC hdc = GetDC(0) + vulkan_funcs = __wine_get_vk_driver(hdc); + if (!icd_funcs) + { + ReleaseDC(hdc); + return FALSE; + } + +#define LOAD_FUNCPTR(x) icd_funcs->vk_icdGetInstanceProcAddr(x) + LOAD_FUNCPTR(vkCreateInstance) + LOAD_FUNCPTR(vkEnumerateInstanceExtensionProperties) + LOAD_FUNCPTR(vkGetInstanceProcAddr) + + ReleaseDC(hdc) + return TRUE; +} + + +/* Implementation of global Vulkan functions, which are state-less. + * There are only a handful of these. */ + +static PFN_vkVoidFunction WINE_VKAPI wine_vkGetInstanceProcAddr(VkInstance instance, const char* pName) +{ + return wine_vk_icdGetInstanceProcAddr(instance, pName) +} + +static VkResult WINE_VKAPI wine_vkCreateInstance(..) +{ + return icd_funcs->vkCreateInstance(..) +} + +static VkResult WINE_VKAPI wine_vkEnumerateInstanceExtensionProperties(const char *pLayerName, uint32_t *pPropertyCount, + VkExtensionProperties* pProperties) +{ + return icd_funcs->vkEnumerateInstanceExtensionProperties(..) +} + + +/* ICD entry points, only 2 exist these days. */ + +void* WINAPI vk_icdGetInstanceProcAddr(VkInstance instance, const char* pName) +{ + if (!wine_vk_init()) + return NULL; + + if (!instance) + { + return wine_vk_get_global_proc_addr(pName); + } + + /* Now find any instance function. */ + func = wine_vk_get_instance_proc_addr(pName); + if (func) + return func; + + func = wine_vk_get_device_proc_addr(pName); + if (func) + return func; + + TRACE("Function %s not found\n", pName); + return NULL; +} + +VkResult WINAPI vk_icdNegotiateLoaderICDInterfaceVersion(uint32_t* pSupportedVersion) +{ + if (!wine_vk_init()) + return VK_ERROR_INCOMPATIBLE_DRIVER; + + *pSupportedVersion = min(*pSupportedVersion, WINE_VULKAN_ICD_VERSION); + return VK_SUCCESS; +} diff --git a/dlls/winevulkan/vulkan_thunks.c b/dlls/winevulkan/vulkan_thunks.c new file mode 100644 index 0000000000..0dfb5ac7fd --- /dev/null +++ b/dlls/winevulkan/vulkan_thunks.c @@ -0,0 +1,80 @@ +/* Automatically generated from Vulkan vk.xml; DO NOT EDIT! */ + +#include "wine/debug.h" +#include "wine/vulkan.h" +#include "vulkan_private.h" + +WINE_DEFAULT_DEBUG_CHANNEL(vulkan); + +static VkResult WINAPI wine_vkAllocateCommandBuffers(VkDevice device, const VkCommandBufferAllocateInfo * pAllocateInfo, VkCommandBuffer * pCommandBuffers) +{ + TRACE(" %p, %p, %p\n", device, pAllocateInfo, pCommandBuffers); + /* Vulkan excepts dispatch tables to stored in 'dispatchable' objects. + * Each object will have a dispatch table to either the 'instance' or 'device' table. + * This helps us as the graphics driver winex11 will fill this function table. + * Another nice benefit is that the native libvulkan-1.so as used by winex11 may + * load multiple ICDs behind our back in a multi-gpu system, but we don't have to bother + * with any of the details as the function pointers will always be right as they are per object. + */ + return device->funcs->vkAllocateCommandBuffers(device, pAllocateInfo, pCommandBuffers); +} +.. +.. +static void WINAPI X11DRV_vkGetPhysicalDeviceFeatures(VkPhysicalDevice physicalDevice, VkPhysicalDeviceFeatures * pFeatures) +{ + TRACE("%p, %p\n", physicalDevice, pFeatures); + physicalDevice->funcs.p_vkGetPhysicalDeviceFeatures(physicalDevice->phys_dev, pFeatures); +} + + +const struct vulkan_func vk_device_dispatch_table[] = { + {"vkAllocateCommandBuffers", NULL, &wine_vkAllocateCommandBuffers}, + .. + {"vkWaitForFences", NULL, &wine_vkWaitForFences}, +}; +const int vk_device_dispatch_table_size = 120; + +const struct vulkan_func vk_instance_dispatch_table[] = { + {"vkCreateDevice", NULL, &wine_vkCreateDevice}, + {"vkDestroyInstance", NULL, &wine_vkDestroyInstance}, + {"vkEnumerateDeviceExtensionProperties", NULL, &wine_vkEnumerateDeviceExtensionProperties}, + {"vkEnumerateDeviceLayerProperties", NULL, &wine_vkEnumerateDeviceLayerProperties}, + {"vkEnumeratePhysicalDevices", NULL, &wine_vkEnumeratePhysicalDevices}, + {"vkGetDeviceProcAddr", NULL, &wine_vkGetDeviceProcAddr}, + {"vkGetPhysicalDeviceFeatures", NULL, &wine_vkGetPhysicalDeviceFeatures}, + {"vkGetPhysicalDeviceFormatProperties", NULL, &wine_vkGetPhysicalDeviceFormatProperties}, + {"vkGetPhysicalDeviceImageFormatProperties", NULL, &wine_vkGetPhysicalDeviceImageFormatProperties}, + {"vkGetPhysicalDeviceMemoryProperties", NULL, &wine_vkGetPhysicalDeviceMemoryProperties}, + {"vkGetPhysicalDeviceProperties", NULL, &wine_vkGetPhysicalDeviceProperties}, + {"vkGetPhysicalDeviceQueueFamilyProperties", NULL, &wine_vkGetPhysicalDeviceQueueFamilyProperties}, + {"vkGetPhysicalDeviceSparseImageFormatProperties", NULL, &wine_vkGetPhysicalDeviceSparseImageFormatProperties}, +}; +const int vk_instance_dispatch_table_size = 13; + +void* wine_vk_get_device_proc_addr(const char* name) +{ + int i; + for (i = 0; i < vk_device_dispatch_table_size; i++) + { + if (strcmp(name, vk_device_dispatch_table[i].name) == 0) + { + TRACE("Found pName=%s in device table\n", name); + return vk_device_dispatch_table[i].func; + } + } + return NULL; +} + +void* wine_vk_get_instance_proc_addr(const char* name) +{ + int i; + for (i = 0; i < vk_instance_dispatch_table_size; i++) + { + if (strcmp(name, vk_instance_dispatch_table[i].name) == 0) + { + TRACE("Found pName=%s in instance table\n", name); + return vk_instance_dispatch_table[i].func; + } + } + return NULL; +} diff --git a/dlls/winex11.drv/vulkan.c b/dlls/winex11.drv/vulkan.c new file mode 100644 index 0000000000..5b83d43360 --- /dev/null +++ b/dlls/winex11.drv/vulkan.c @@ -0,0 +1,77 @@ +struct vk_global_functions = +{ + {"vkCreateInstance", X11DRV_vkCreateInstance}, + .. +} + +/* Functions creating dispatchable objects and a few others (e.g. vkGetDeviceQueue) + * need special implementations. */ +struct vk_overrides = +{ + {"vkCreateDevice", X11DRV_vkCreateDevice} + .. +} + +struct winex11_vkinstance +{ + struct VkInstance_T; + + /* Native instance */ + VkInstance *instance; +}; + +static VkResult WINE_VKAPI X11DRV_vkCreateInstance(..) +{ + instance = wine_vk_alloc_dispatchable_object(..); + + pvkCreateInstance(,, &instance->instance); + + /* Load all instance functions we are aware of. Note the loader takes care + * of any filtering for extensions which were not requested, but which the + * ICD may support. + */ +#define USE_VK_FUNC(name) \ + instance->funcs.p_##name = (void*)pvkGetInstanceProcAddr(instance->instance, #name); + ALL_VK_INSTANCE_FUNCS() +#undef USE_VK_FUNC + + return VK_SUCCESS; +} + + +static VkResult WINE_VKAPI X11DRV_vkCreateDevice(..) +{ + device = wine_vk_alloc_dispatchable_object(..); + + pvkCreateDevice(,, &device->device); + + /* Just load all function pointers we are aware off. The loader takes care of filtering. */ +#define USE_VK_FUNC(name) \ + device->funcs.p_##name = (void*)pvkGetDeviceProcAddr(device->device, #name); \ + if (device->funcs.p_##name == NULL) \ + TRACE("Not found %s\n", #name); + ALL_VK_DEVICE_FUNCS() +#undef USE_VK_FUNC + + return VK_SUCCESS; +} + +void* X11DRV_vk_icdGetInstanceProcAddr(VkInstance instance, const char* pName) +{ + if (!instance) + { + // search vk_global_functions + } + + // search table of overrides + + /* Ask the native ICD library. Since VkInstance is a dispatchable object + * retrieve the native instance and pass it down. + */ + return native_vkGetInstanceProcAddr(instance->native_instance, pName); +} + +VkResult X11DRV_vk_icdNegotiateLoaderICDInterfaceVersion(uint32_t* pSupportedVersion) +{ + return VK_INCOMPLETE; +} diff --git a/dlls/winex11.drv/winex11.drv.spec b/dlls/winex11.drv/winex11.drv.spec index 614f0b90f4..d701488b31 100644 --- a/dlls/winex11.drv/winex11.drv.spec +++ b/dlls/winex11.drv/winex11.drv.spec @@ -78,3 +78,7 @@ @ stdcall ImeProcessKey(long long long ptr) @ stdcall ImeGetRegisterWordStyle(long ptr) @ stdcall ImeGetImeMenuItems(long long long ptr ptr long) + +# Vulkan +@ cdecl vk_icdGetInstanceProcAddr(ptr str) X11DRV_vk_icdGetInstanceProcAddr +@ cdecl vk_icdNegotiateLoaderICDInterfaceVersion(ptr) X11DRV_vk_icdNegotiateLoaderICDInterfaceVersion diff --git a/include/wine/vulkan_driver.h b/include/wine/vulkan_driver.h new file mode 100644 index 0000000000..eb3159f098 --- /dev/null +++ b/include/wine/vulkan_driver.h @@ -0,0 +1,78 @@ +/* I'm thinking of using 'WINE_VKAPI' as the calling convention for all calls implemented + * by winex11/android. This convention should be the same as the calling convention used + * by the native Vulkan library. + */ +#define WINE_VKAPI + +/* Struct returned by gdi32 back to winevulkan. */ +struct vulkan_icd_funcs +{ + /* To use WINE_VKAPI calling convention? */ + Function Pointer to vk_icdGetInstanceProcAddr + /* Technically this is an ICD export, but we may not need, will see. */ + Function Pointer to vk_icdNegotiateLoaderICDInterfaceVersion +}; + +struct vulkan_device_funcs +{ + vkAllocateCommandBuffers; + .. + vkWaitForFences; +}; + +struct vulkan_instance_funcs +{ + vkCreateDevice; + .. + vkGetPhysicalDeviceSparseImageFormatProperties; +}; + +#define ALL_VK_DEVICE_FUNCS() \ + USE_VK_FUNC(vkAllocateCommandBuffers) \ + USE_VK_FUNC(vkAllocateDescriptorSets) \ + USE_VK_FUNC(vkAllocateMemory) \ + .. + + +#define ALL_VK_INSTANCE_FUNCS() \ + USE_VK_FUNC(vkCreateDevice)\ + USE_VK_FUNC(vkDestroyInstance)\ + USE_VK_FUNC(vkEnumerateDeviceExtensionProperties)\ + .. + + +/* Magic value defined by Vulkan ICD / Loader spec */ +#define VULKAN_ICD_MAGIC_VALUE 0x01CDC0DE + +/* Base 'class' for our Vulkan dispatchable objects such as VkDevice and VkInstance. + * This structure MUST be the first element of a dispatchable object as the ICD + * loader depends on it. For now only contains loader_magic, but over time more common + * functionality is expected. + */ +struct wine_vk_base +{ + /* Special section in each dispatchable object for use by the ICD loader for + * storing dispatch tables. The start contains a magical value '0x01CDC0DE'. + */ + uintptr_t loader_magic; +}; + +/* Vulkan defines handles for objects through VK_DEFINE_HANDLE. As part of this + * it defines forward declarations for structs '_T', which the implementor (us) + * can use. + */ +struct VkDevice_T +{ + struct wine_vk_base base; + struct vulkan_device_funcs funcs; + + /* driver private data */ +}; + +struct VkInstance_T +{ + struct wine_vk_base base; + struct vulkan_instance_funcs funcs; + + /* driver private data */ +}; -- 2.13.6

1 0

Re: [PATCH] kernel32/tests: Test console input instead of stdin in test_ReadConsole.
by Marvin 02 Nov '17

02 Nov '17

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=33874 Your paranoid android. === wxppro (32 bit console) === console.c:2582: Test failed: expected INVALID_FILE_SIZE, got 0 console.c:2583: Test failed: expected ERROR_INVALID_HANDLE, got -559038737 === wvistau64 (32 bit console) === console.c:2582: Test failed: expected INVALID_FILE_SIZE, got 0x6000 console.c:2583: Test failed: expected ERROR_INVALID_HANDLE, got -559038737 === w2008s64 (32 bit console) === console.c:2582: Test failed: expected INVALID_FILE_SIZE, got 0x4000 console.c:2583: Test failed: expected ERROR_INVALID_HANDLE, got -559038737 === w7pro64 (64 bit console) === console.c:2582: Test failed: expected INVALID_FILE_SIZE, got 0x1000 console.c:2583: Test failed: expected ERROR_INVALID_HANDLE, got -559038737

1 0

Re: [PATCH] kernel32/tests: Detect and handle a redirected stdin.
by Alexandre Julliard 02 Nov '17

02 Nov '17

Alex Henrie <alexhenrie24(a)gmail.com> writes: > @@ -2572,7 +2572,7 @@ static void test_ReadConsoleOutputAttribute(HANDLE output_handle) > ok(count == 1, "Expected count to be 1, got %u\n", count); > } > > -static void test_ReadConsole(void) > +static void test_ReadConsole(HANDLE input_handle) > { > HANDLE std_input; > DWORD ret, bytes; > @@ -2582,6 +2582,13 @@ static void test_ReadConsole(void) > > SetLastError(0xdeadbeef); > ret = GetFileSize(std_input, NULL); > + if (GetLastError() == 0xdeadbeef) > + { > + /* print a warning and cheat a little so that the tests can continue */ > + skip("stdin appears to be redirected; if it's not then this is an error"); > + std_input = input_handle; > + ret = GetFileSize(std_input, NULL); > + } I'm not sure why we'd need to use stdin at all then, we might as well always use the proper console handle. -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: [PATCH] iphlpapi: Remove check for negative value.
by Alexandre Julliard 02 Nov '17

02 Nov '17

Andrey Gusev <andrey.goosev(a)gmail.com> writes: > +++ b/dlls/iphlpapi/ipstats.c > @@ -2056,7 +2056,7 @@ static unsigned int find_owning_pid( struct pid_map *map, unsigned int num_entri > for (i = 0; i < num_entries; i++) > { > fd_len = proc_pidinfo( map[i].unix_pid, PROC_PIDLISTFDS, 0, NULL, 0 ); > - if (fd_len <= 0) continue; > + if (fd_len == 0) continue; The function returns int, probably the variable type should be changed instead. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [v3] oleaut32: Add ARM support to DispCallFunc().
by Alexandre Julliard 02 Nov '17

02 Nov '17

Donna Whisnant <dewhisna(a)dewtronics.com> writes: > Adds ARM ABI support to DispCallFunc() to allow IDispatch invoke calls > to succeed on ARM platforms. This change specifically targets only > 32-bit little-endian (ARMEL) platform CPUs. It's believed to likely be > compatible with big-endian (ARMEB) platforms, but testing for that > platform should be completed before enabling. > > Tested on Raspbian Stretch 2017-09-07 RPi image using a Qemu systemd > container to compile and run it on an x86-64 Host.. When running on > Raspberry Pi hardware, a 3G/1G split kernel needs to be used for Wine > to function. It breaks the Android build: arm-linux-androideabi-gcc -c -o typelib.o ../../../wine/dlls/oleaut32/typelib.c -I. -I../../../wine/dlls/oleaut32 \ -I../../include -I../../../wine/include -D__WINESRC__ -D_OLEAUT32_ -D_REENTRANT -fPIC -Wall -pipe \ -fno-strict-aliasing -Wdeclaration-after-statement -Wempty-body -Wignored-qualifiers \ -Wstrict-prototypes -Wtype-limits -Wunused-but-set-parameter -Wvla -Wwrite-strings -Wpointer-arith \ -Wlogical-op -gdwarf-2 -gstrict-dwarf -g -O2 -fno-diagnostics-show-caret -D__ANDROID_API__=26 -marm {standard input}: Assembler messages: {standard input}:32: Error: selected processor does not support ARM mode `vldm r3!,{s0-s15}' Makefile:516: recipe for target 'typelib.o' failed make: *** [typelib.o] Error 1 -- Alexandre Julliard julliard(a)winehq.org

3 5

Re: [PATCH] dsound: Don't use lrintf if it's not available.
by Matteo Bruni 02 Nov '17

02 Nov '17

2017-10-29 21:59 GMT+01:00 Alex Henrie <alexhenrie24(a)gmail.com>: > Signed-off-by: Alex Henrie <alexhenrie24(a)gmail.com> > --- > Fixes https://bugs.winehq.org/show_bug.cgi?id=43000 > > lrintf is not part of C89, and we were already using HAVE_LRINTF in > msvcrt. > --- > dlls/dsound/dsound_convert.c | 16 ++++++++++++++++ > 1 file changed, 16 insertions(+) > > diff --git a/dlls/dsound/dsound_convert.c b/dlls/dsound/dsound_convert.c > index 6887bae558..ad1aba3737 100644 > --- a/dlls/dsound/dsound_convert.c > +++ b/dlls/dsound/dsound_convert.c > @@ -120,7 +120,11 @@ static inline unsigned char f_to_8(float value) > return 0; > if(value >= 1.f * 0x7f / 0x80) > return 0xFF; > +#ifdef HAVE_LRINTF > return lrintf((value + 1.f) * 0x80); > +#else > + return (value + 1) * 0x80; > +#endif > } The fallback should probably add 0.5f before truncating. Not entirely equivalent but probably closer to the original code. BTW, we also use lrint() in d3dx9, so we might want to fix it there too. Actually, wouldn't it be better to add generic lrint() / lrintf() fallbacks in libs/port/ instead?

2 1

Re: [PATCH 00/10] winex11: Add Vulkan ICD.
by Alexandre Julliard 02 Nov '17

02 Nov '17

Roderick Colenbrander <thunderbird2k(a)gmail.com> writes: > The amount of code sharing between ICDs, I expect to be mostly in > generated code and maybe a few helper functions. For now > I think it is best to duplicate code and longer-term if > a little bit of sharing makes sense put this in a staticly > linked helper library. I want to avoid implementing our own > 'winevulkan' glue ICD library in between loader and display > driver as I fear this will cause many issues, because the object > design didn't lend for this (assumes 1 ICD = 1 driver). Besides > some applications load the ICD directly. > > This first wave of patches gets some initial ICD infrastructure > in place. Enough to create a minimal Vulkan Instance and get > some function loading going. It will take another 6-7 patches > or so before 'vulkaninfo' the Vulkan equivalent of glxinfo > will work to some degree. > > For now the easiest way to load the ICD is to add a registry key: > [HKEY_LOCAL_MACHINE\SOFTWARE\Khronos\Vulkan\Drivers\] > "C:\Windows\System32\winex11.json"=dword:00000000 > > winex11.json: > { > "file_format_version": "1.0.0", > "ICD": { > "library_path": "c:\\windows\\system32\\winex11.drv", > "api_version": "1.0.51" > } > } How would this be able to support multiple drivers with the same install? It seems to me that an intermediate library that calls the currently active driver would be better. What sort of issues do you think it would cause? -- Alexandre Julliard julliard(a)winehq.org

3 7

Re: [PATCH 02/10] winex11: Add stub Vulkan ICD
by Józef Kucia 01 Nov '17

01 Nov '17

On Tue, Oct 31, 2017 at 6:24 PM, Roderick Colenbrander <thunderbird2k(a)gmail.com> wrote: > Signed-off-by: Roderick Colenbrander <thunderbird2k(a)gmail.com> > --- > configure.ac | 9 +++++++ > dlls/winex11.drv/Makefile.in | 1 + > dlls/winex11.drv/vulkan.c | 54 +++++++++++++++++++++++++++++++++++++++ > dlls/winex11.drv/winex11.drv.spec | 4 +++ > include/config.h.in | 3 +++ > 5 files changed, 71 insertions(+) > create mode 100644 dlls/winex11.drv/vulkan.c > > diff --git a/configure.ac b/configure.ac > index c02253bb86..4766c0e6c0 100644 > --- a/configure.ac > +++ b/configure.ac > @@ -78,6 +78,7 @@ AC_ARG_WITH(sane, AS_HELP_STRING([--without-sane],[do not use SANE (scanner > AC_ARG_WITH(tiff, AS_HELP_STRING([--without-tiff],[do not use TIFF])) > AC_ARG_WITH(udev, AS_HELP_STRING([--without-udev],[do not use udev (plug and play support)])) > AC_ARG_WITH(v4l, AS_HELP_STRING([--without-v4l],[do not use v4l1 (v4l support)])) > +AC_ARG_WITH(vulkan, AS_HELP_STRING([--without-vulkan],[do not use vulkan])) > AC_ARG_WITH(xcomposite,AS_HELP_STRING([--without-xcomposite],[do not use the Xcomposite extension]), > [if test "x$withval" = "xno"; then ac_cv_header_X11_extensions_Xcomposite_h=no; fi]) > AC_ARG_WITH(xcursor, AS_HELP_STRING([--without-xcursor],[do not use the Xcursor extension]), > @@ -1786,6 +1787,14 @@ then > WINE_WARNING([No sound system was found. Windows applications will be silent.]) > fi > > +dnl *** Check for vulkan *** > +if test "x$with_vulkan" != "xno" > +then > + WINE_CHECK_SONAME(vulkan, vkGetInstanceProcAddr) > +fi > +WINE_NOTICE_WITH(vulkan,[test "x$ac_cv_lib_soname_vulkan" = "x"], > + [libvulkan ${notice_platform}development files not found, vulkan won't be supported.]) > + > dnl **** Check for gcc specific options **** > > AC_SUBST(EXTRACFLAGS,"") > diff --git a/dlls/winex11.drv/Makefile.in b/dlls/winex11.drv/Makefile.in > index 463eefdcfb..747f509b44 100644 > --- a/dlls/winex11.drv/Makefile.in > +++ b/dlls/winex11.drv/Makefile.in > @@ -20,6 +20,7 @@ C_SRCS = \ > pen.c \ > settings.c \ > systray.c \ > + vulkan.c \ > window.c \ > wintab.c \ > x11drv_main.c \ > diff --git a/dlls/winex11.drv/vulkan.c b/dlls/winex11.drv/vulkan.c > new file mode 100644 > index 0000000000..936e5620d2 > --- /dev/null > +++ b/dlls/winex11.drv/vulkan.c > @@ -0,0 +1,54 @@ > +/* X11DRV Vulkan ICD implementation > + * > + * Copyright 2017 Roderick Colenbrander > + * > + * This library is free software; you can redistribute it and/or > + * modify it under the terms of the GNU Lesser General Public > + * License as published by the Free Software Foundation; either > + * version 2.1 of the License, or (at your option) any later version. > + * > + * This library is distributed in the hope that it will be useful, > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU > + * Lesser General Public License for more details. > + * > + * You should have received a copy of the GNU Lesser General Public > + * License along with this library; if not, write to the Free Software > + * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA > + */ > + > +#include "config.h" > +#include "wine/port.h" > + > +#include "wine/debug.h" > +#include "wine/vulkan.h" > + > +#ifdef SONAME_LIBVULKAN > + > +WINE_DEFAULT_DEBUG_CHANNEL(vulkan); > + > +void* WINAPI X11DRV_vk_icdGetInstanceProcAddr(VkInstance instance, const char* pName) Please use consistent style for pointers. I think that "void * WINAPI" and "const char *pName" would be preferred. > +{ > + FIXME("stub: %p, %s\n", instance, pName ? pName : "NULL"); Wouldn't it be better to use debugstr_a() or something similar? Please also note that some strings in Vulkan API can be UTF-8 according to the spec. > + return NULL; > +} > + > +VkResult WINAPI X11DRV_vk_icdNegotiateLoaderICDInterfaceVersion(uint32_t* pSupportedVersion) > +{ > + FIXME("stub: %p\n", pSupportedVersion); > + return VK_INCOMPLETE; > +} > + > +#else /* No vulkan */ > + > +void* WINAPI X11DRV_vk_icdGetInstanceProcAddr(VkInstance instance, const char* pName) > +{ > + return NULL; > +} > + > +VkResult WINAPI X11DRV_vk_icdNegotiateLoaderICDInterfaceVersion(uint32_t* pSupportedVersion) > +{ > + return VK_INCOMPLETE; > +} > + > +#endif /* defined SONAME_LIBVULKAN */ > diff --git a/dlls/winex11.drv/winex11.drv.spec b/dlls/winex11.drv/winex11.drv.spec > index 614f0b90f4..0932a2c88e 100644 > --- a/dlls/winex11.drv/winex11.drv.spec > +++ b/dlls/winex11.drv/winex11.drv.spec > @@ -78,3 +78,7 @@ > @ stdcall ImeProcessKey(long long long ptr) > @ stdcall ImeGetRegisterWordStyle(long ptr) > @ stdcall ImeGetImeMenuItems(long long long ptr ptr long) > + > +# Vulkan > +@ stdcall vk_icdGetInstanceProcAddr(ptr str) X11DRV_vk_icdGetInstanceProcAddr > +@ stdcall vk_icdNegotiateLoaderICDInterfaceVersion(ptr) X11DRV_vk_icdNegotiateLoaderICDInterfaceVersion > diff --git a/include/config.h.in b/include/config.h.in > index a6ec41b5fa..444f4595d4 100644 > --- a/include/config.h.in > +++ b/include/config.h.in > @@ -1541,6 +1541,9 @@ > /* Define to the soname of the libXxf86vm library. */ > #undef SONAME_LIBXXF86VM > > +/* Define to the soname of the libvulkan library. */ > +#undef SONAME_LIBVULKAN > + > /* Define to 1 if the `S_IS*' macros in <sys/stat.h> do not work properly. */ > #undef STAT_MACROS_BROKEN > > -- > 2.13.6 > > >

2 1

Re: [PATCH 1/4] gdiplus/metafile: Support hatch brushes in playback
by Vincent Povirk 01 Nov '17

01 Nov '17

+ status = GdipCreateHatchBrush(data->BrushData.hatch.HatchStyle, *(ARGB *)&data->BrushData.hatch.ForeColor, + *(ARGB *)&data->BrushData.hatch.BackColor, (GpHatch **)brush); I guess this is OK for this patch, but maybe we should modify the definition of EmfPlusARGB to avoid needing things like this?

2 4

Re: [PATCH 01/10] vulkan: Add Wine vulkan header.
by Józef Kucia 01 Nov '17

01 Nov '17

On Tue, Oct 31, 2017 at 6:24 PM, Roderick Colenbrander <thunderbird2k(a)gmail.com> wrote: > Signed-off-by: Roderick Colenbrander <thunderbird2k(a)gmail.com> > --- > include/wine/vulkan.h | 2889 +++++++++++++++++++++++++++++++++++++++++++++++++ > 1 file changed, 2889 insertions(+) > create mode 100644 include/wine/vulkan.h > > diff --git a/include/wine/vulkan.h b/include/wine/vulkan.h > new file mode 100644 > index 0000000000..b86702e43f > --- /dev/null > +++ b/include/wine/vulkan.h > @@ -0,0 +1,2889 @@ > +/* Automatically generated from Vulkan vk.xml; DO NOT EDIT! */ This file is generated by a custom code generator, right? When do you plan to submit the code generator? In a way it might be easier to review the code generator than the generated code. > + > +#ifndef __WINE_VULKAN_H > +#define __WINE_VULKAN_H

2 1

Re: [PATCH v2] vbscript/tests: Fix test for WeekDayName(foo, bar, 0).
by Alexandre Julliard 01 Nov '17

01 Nov '17

Jacek Caban <jacek(a)codeweavers.com> writes: > Hi Alex, > > On 30.10.2017 18:17, Alex Henrie wrote: >> -static BOOL is_lang_english(void) >> +static void detect_locale(void) >> { >> - static HMODULE hkernel32 = NULL; >> - static LANGID (WINAPI *pGetThreadUILanguage)(void) = NULL; >> - static LANGID (WINAPI *pGetUserDefaultUILanguage)(void) = NULL; >> + is_english = (PRIMARYLANGID(GetThreadUILanguage()) == LANG_ENGLISH && >> + PRIMARYLANGID(GetUserDefaultUILanguage()) == LANG_ENGLISH && >> + PRIMARYLANGID(GetUserDefaultLangID()) == LANG_ENGLISH); > > > GetThreadUILanguage is available since Vista, so we still need to > check it in runtime. BTW, I'm not sure what's up with TestBot. There > are failures (as expected), but it has OK status on patch status page. Probably that's because there are no failure messages, only timeouts; the testbot doesn't seem to handle that case correctly. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [PATCH 1/4] dx8vb: Add stubs for D3DX8 interface
by Henri Verbeet 01 Nov '17

01 Nov '17

Perhaps this is sufficiently far removed from core Direct3D that nobody cares, but for what it's worth: On 31 October 2017 at 03:01, Fabian Maurer <dark.shadow4(a)web.de> wrote: > +typedef struct > +{ > + ID3DX8 ID3DX8_iface; > + LONG ref; > +} d3dx8; Please don't typedef structures just to save keystrokes. > +static HRESULT WINAPI d3dx8_QueryInterface(ID3DX8 *iface, REFIID riid, void **ppv) > +{ > + d3dx8 *This = impl_from_ID3DX8(iface); Please use lowercase variable names. > + > + TRACE("(%p/%p)->(%s,%p)\n", iface, This, debugstr_guid(riid), ppv); > + > + *ppv = NULL; > + > + if (IsEqualGUID(riid, &IID_IUnknown)) > + *ppv = &This->ID3DX8_iface; > + else if(IsEqualGUID(riid, &IID_ID3DX8)) > + *ppv = &This->ID3DX8_iface; > + else if(IsEqualGUID(riid, &IID_IPersistPropertyBag)) > + FIXME("No interface for IID_IPersistPropertyBag\n"); > + else if(IsEqualGUID(riid, &IID_IPersistStreamInit)) > + FIXME("No interface for IID_IPersistStreamInit\n"); > + else > + FIXME("No interface for %s\n", debugstr_guid(riid)); Not implementing an arbitrary interface is a WARN at best. > + > + if (*ppv) > + { > + IUnknown_AddRef((IUnknown *)(*ppv)); > + return S_OK; > + } > + > + return E_NOINTERFACE; > +} I think Nikolay already hinted at this, but this is fairly different from the canonical QueryInterface() implementation, especially for an object that only implements a single interface. > +static HRESULT WINAPI d3dx8_CreateFont(ID3DX8 *iface, Direct3DDevice8 *device, > + LONG hFont, D3DXFont **retFont) Please use lowercase parameter names, and avoid the stray 'h'. > +static HRESULT WINAPI d3dx8_AssembleShaderFromFile(ID3DX8 *iface, BSTR file, LONG flags, BSTR *log, > + D3DXBuffer **constants, D3DXBuffer **ppVertexShader) > +{ > + FIXME("(%s, %i, %p, %p, %p): stub!\n", debugstr_w(file), flags, log, constants, ppVertexShader); I certainly have my preferences when it comes to trace formatting, but I think "%i" is questionable for flags regardless of preference. > +static const ID3DX8Vtbl ID3DX8_Vtbl = "d3dx8_vtbl" > +HRESULT d3dx8_create(IUnknown *outer_unk, void **ppv) > +{ > + d3dx8 *object; > + > + TRACE("(%p,%p)\n", outer_unk, ppv); > + > + object = HeapAlloc(GetProcessHeap(), HEAP_ZERO_MEMORY, sizeof(d3dx8)); "sizeof(*object)"

4 6

Re: [PATCH] regedit: Always start the GUI as 64-bit when on a 64-bit prefix.
by Austin English 01 Nov '17

01 Nov '17

On Tue, Oct 31, 2017 at 6:27 PM, Zebediah Figura <z.figura12(a)gmail.com> wrote: > Same as we do with winecfg. This should help to eliminate some confusion > with users who may want to edit keys with WOW64 redirection (e.g. > Software\\Wine\\Drives). > > Signed-off-by: Zebediah Figura <z.figura12(a)gmail.com> > --- > programs/regedit/main.c | 28 ++++++++++++++++++++++++++++ > 1 file changed, 28 insertions(+) > > diff --git a/programs/regedit/main.c b/programs/regedit/main.c > index 8f82580..460a951 100644 > --- a/programs/regedit/main.c > +++ b/programs/regedit/main.c > @@ -24,10 +24,13 @@ > #include <stdlib.h> > #include <stdio.h> > #include <fcntl.h> > +#include "wine/debug.h" > > #define REGEDIT_DECLARE_FUNCTIONS > #include "main.h" > > +WINE_DEFAULT_DEBUG_CHANNEL(regedit); > + > WCHAR g_pszDefaultValueName[64]; > > BOOL ProcessCmdLine(WCHAR *cmdline); > @@ -138,11 +141,36 @@ int WINAPI wWinMain(HINSTANCE hInstance, HINSTANCE hPrevInstance, LPWSTR lpCmdLi > { > MSG msg; > HACCEL hAccel; > + BOOL is_wow64; > > if (ProcessCmdLine(GetCommandLineW())) { > return 0; > } > > + if (IsWow64Process( GetCurrentProcess(), &is_wow64 ) && is_wow64) > + { > + STARTUPINFOW si; > + PROCESS_INFORMATION pi; > + WCHAR filename[MAX_PATH]; > + void *redir; > + DWORD exit_code; > + > + memset( &si, 0, sizeof(si) ); > + si.cb = sizeof(si); > + GetModuleFileNameW( 0, filename, MAX_PATH ); > + > + Wow64DisableWow64FsRedirection( &redir ); > + if (CreateProcessW( filename, GetCommandLineW(), NULL, NULL, FALSE, 0, NULL, NULL, &si, &pi )) > + { > + WINE_TRACE( "restarting %s\n", wine_dbgstr_w(filename) ); > + WaitForSingleObject( pi.hProcess, INFINITE ); > + GetExitCodeProcess( pi.hProcess, &exit_code ); > + ExitProcess( exit_code ); > + } > + else WINE_ERR( "failed to restart 64-bit %s, err %d\n", wine_dbgstr_w(filename), GetLastError() ); > + Wow64RevertWow64FsRedirection( redir ); > + } > + > /* Initialize global strings */ > LoadStringW(hInstance, IDS_APP_TITLE, szTitle, COUNT_OF(szTitle)); > LoadStringW(hInstance, IDS_REGISTRY_DEFAULT_VALUE, g_pszDefaultValueName, COUNT_OF(g_pszDefaultValueName)); > -- > 2.7.4 How should we expect users to edit the 32-bit entries then? Through Wow6432node or some other way? -- -Austin GPG: 14FB D7EA A041 937B

2 2

Re: [PATCH 07/10] winex11: add logging functions for vkCreateInstance
by Józef Kucia 01 Nov '17

01 Nov '17

On Tue, Oct 31, 2017 at 6:24 PM, Roderick Colenbrander <thunderbird2k(a)gmail.com> wrote: > Signed-off-by: Roderick Colenbrander <thunderbird2k(a)gmail.com> > --- > dlls/winex11.drv/vulkan.c | 44 ++++++++++++++++++++++++++++++++++++++++++++ > 1 file changed, 44 insertions(+) > > diff --git a/dlls/winex11.drv/vulkan.c b/dlls/winex11.drv/vulkan.c > index 4171ee0fd6..8632c46010 100644 > --- a/dlls/winex11.drv/vulkan.c > +++ b/dlls/winex11.drv/vulkan.c > @@ -39,6 +39,13 @@ WINE_DEFAULT_DEBUG_CHANNEL(vulkan); > > void* WINAPI X11DRV_vk_icdGetInstanceProcAddr(VkInstance instance, const char* pName); > > +/* All Vulkan structures use this structure for the first elements. */ > +struct wine_vk_structure_header > +{ > + VkStructureType sType; > + const void *pNext; > +}; > + > > static BOOL wine_vk_init(void) > { > @@ -67,10 +74,47 @@ static void *wine_vk_get_global_proc_addr(const char* name) > return NULL; > } > > +static void wine_vk_application_dump_info(const VkApplicationInfo *info) > +{ > + TRACE("pApplicationName \"%s\", applicationVersion %d\n", info->pApplicationName, info->applicationVersion); > + TRACE("pEngineName \"%s\", engineVersion %d\n", info->pEngineName, info->engineVersion); > + TRACE("apiVersion %#x\n", info->apiVersion); > +} I guess that "%s" doesn't work very well for NULL. We should probably use debugstr_a(). > + > +static void wine_vk_instance_dump_create_info(const VkInstanceCreateInfo *info) > +{ > + int i, num_extensions, num_layers; > + const struct wine_vk_structure_header *header; > + > + wine_vk_application_dump_info(info->pApplicationInfo); pApplicationInfo can be NULL. > + > + /* Should be 0, loader should have filtered this out. */ > + num_layers = info->enabledLayerCount; > + TRACE("Enabled layers: %d\n", num_layers); > + for (i = 0; i < num_layers; i++) > + { > + TRACE("Layer %d: %s\n", i, info->ppEnabledLayerNames[i]); > + } > + > + num_extensions = info->enabledExtensionCount; > + TRACE("Enabled extensions: %d\n", num_extensions); > + for (i = 0; i < num_extensions; i++) > + { > + TRACE("Extension %d: %s\n", i, info->ppEnabledExtensionNames[i]); > + } > + > + for (header = info->pNext; header; header = header->pNext) > + { > + TRACE("pNext: %p sType:%d\n", header, header->sType); > + } > +} > + > VkResult WINAPI X11DRV_vkCreateInstance(const VkInstanceCreateInfo *pCreateInfo, const VkAllocationCallbacks *pAllocator, > VkInstance *pInstance) > { > FIXME("stub: %p, %p, %p\n", pCreateInfo, pAllocator, pInstance); > + > + wine_vk_instance_dump_create_info(pCreateInfo); > return VK_ERROR_INCOMPATIBLE_DRIVER; > } > > -- > 2.13.6 > > >

1 0

Re: [PATCH 04/10] winex11: Add stubs for global Vulkan functions.
by Józef Kucia 01 Nov '17

01 Nov '17

On Tue, Oct 31, 2017 at 6:24 PM, Roderick Colenbrander <thunderbird2k(a)gmail.com> wrote: > Signed-off-by: Roderick Colenbrander <thunderbird2k(a)gmail.com> > --- > dlls/winex11.drv/Makefile.in | 1 + > dlls/winex11.drv/vulkan.c | 65 ++++++++++++++++++++++++++++++++++++++- > dlls/winex11.drv/vulkan_private.h | 45 +++++++++++++++++++++++++++ > dlls/winex11.drv/vulkan_thunks.c | 18 +++++++++++ > 4 files changed, 128 insertions(+), 1 deletion(-) > create mode 100644 dlls/winex11.drv/vulkan_private.h > create mode 100644 dlls/winex11.drv/vulkan_thunks.c > > diff --git a/dlls/winex11.drv/Makefile.in b/dlls/winex11.drv/Makefile.in > index 747f509b44..fc6c288d66 100644 > --- a/dlls/winex11.drv/Makefile.in > +++ b/dlls/winex11.drv/Makefile.in > @@ -21,6 +21,7 @@ C_SRCS = \ > settings.c \ > systray.c \ > vulkan.c \ > + vulkan_thunks.c \ > window.c \ > wintab.c \ > x11drv_main.c \ > diff --git a/dlls/winex11.drv/vulkan.c b/dlls/winex11.drv/vulkan.c > index f58bad9c5a..cff474190a 100644 > --- a/dlls/winex11.drv/vulkan.c > +++ b/dlls/winex11.drv/vulkan.c > @@ -25,6 +25,8 @@ > > #ifdef SONAME_LIBVULKAN > > +#include "vulkan_private.h" > + > WINE_DEFAULT_DEBUG_CHANNEL(vulkan); > > /* For now default to 3 as it felt like a reasonable version feature wise to support. > @@ -34,10 +36,71 @@ WINE_DEFAULT_DEBUG_CHANNEL(vulkan); > */ > #define WINE_VULKAN_ICD_VERSION 3 > > +void* WINAPI X11DRV_vk_icdGetInstanceProcAddr(VkInstance instance, const char* pName); > + > + > +static void *wine_vk_get_global_proc_addr(const char* name) > +{ > + int i; I think that "unsigned int" is preferred in this case. > diff --git a/dlls/winex11.drv/vulkan_private.h b/dlls/winex11.drv/vulkan_private.h > new file mode 100644 > index 0000000000..bb722ddb06 > --- /dev/null > +++ b/dlls/winex11.drv/vulkan_private.h > @@ -0,0 +1,45 @@ > +/* X11DRV Vulkan ICD implementation > + * > + * Copyright 2017 Roderick Colenbrander > + * > + * This library is free software; you can redistribute it and/or > + * modify it under the terms of the GNU Lesser General Public > + * License as published by the Free Software Foundation; either > + * version 2.1 of the License, or (at your option) any later version. > + * > + * This library is distributed in the hope that it will be useful, > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU > + * Lesser General Public License for more details. > + * > + * You should have received a copy of the GNU Lesser General Public > + * License along with this library; if not, write to the Free Software > + * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA > + */ > + > +#ifndef __WINE_VULKAN_PRIVATE_H /* __WINE_VULKAN_PRIVATE_H */ I guess that /* __WINE_VULKAN_PRIVATE_H */ is unintentional. > +#define __WINE_VULKAN_PRIVATE_H > + > +#ifndef __WINE_CONFIG_H > +# error You must include config.h to use this header > +#endif > + > +#ifdef SONAME_LIBVULKAN > + > +struct vulkan_func > +{ > + const char *name; > + const char *extension; > + void *func; > +}; > + > +/* For use by vk_icdGetInstanceProcAddr / vkGetInstanceProcAddr */ > +extern const struct vulkan_func vk_global_dispatch_table[] DECLSPEC_HIDDEN; > +extern const int vk_global_dispatch_table_size DECLSPEC_HIDDEN; Could we make the dispatch tables static and local to vulkan_thunks.c? It would be nicer if only wine_vk_get_global_proc_addr() would be accessible to other files. > + > +VkResult WINAPI X11DRV_vkCreateInstance(const VkInstanceCreateInfo*, const VkAllocationCallbacks*, VkInstance*) DECLSPEC_HIDDEN; > +VkResult WINAPI X11DRV_vkEnumerateInstanceExtensionProperties(const char*, uint32_t*, VkExtensionProperties*) DECLSPEC_HIDDEN; > +PFN_vkVoidFunction WINAPI X11DRV_vkGetInstanceProcAddr(VkInstance, const char*) DECLSPEC_HIDDEN; > + > +#endif /* SONAME_LIBVULKAN */ > +#endif /* __WINE_VULKAN_PRIVATE_H */ > diff --git a/dlls/winex11.drv/vulkan_thunks.c b/dlls/winex11.drv/vulkan_thunks.c > new file mode 100644 > index 0000000000..fc3c65762a > --- /dev/null > +++ b/dlls/winex11.drv/vulkan_thunks.c > @@ -0,0 +1,18 @@ > +/* Automatically generated from Vulkan vk.xml; DO NOT EDIT! */ When do you plan to submit the code generator? > + > +#include "config.h" > +#include "wine/port.h" > + > +#ifdef SONAME_LIBVULKAN > + > +#include "wine/vulkan.h" > +#include "vulkan_private.h" > + > +const struct vulkan_func vk_global_dispatch_table[] = { > + {"vkCreateInstance", NULL, &X11DRV_vkCreateInstance}, > + {"vkEnumerateInstanceExtensionProperties", NULL, &X11DRV_vkEnumerateInstanceExtensionProperties}, > + {"vkGetInstanceProcAddr", NULL, &X11DRV_vkGetInstanceProcAddr}, > +}; > +const int vk_global_dispatch_table_size = 3; > + > +#endif /* SONAME_LIBVULKAN */ > -- > 2.13.6 > > >

1 0

Re: [PATCH 06/10] winex11: Load vulkan library.
by Józef Kucia 01 Nov '17

01 Nov '17

On Tue, Oct 31, 2017 at 6:24 PM, Roderick Colenbrander <thunderbird2k(a)gmail.com> wrote: > Signed-off-by: Roderick Colenbrander <thunderbird2k(a)gmail.com> > --- > dlls/winex11.drv/vulkan.c | 21 +++++++++++++++++++++ > 1 file changed, 21 insertions(+) > > diff --git a/dlls/winex11.drv/vulkan.c b/dlls/winex11.drv/vulkan.c > index e0e45d47ce..4171ee0fd6 100644 > --- a/dlls/winex11.drv/vulkan.c > +++ b/dlls/winex11.drv/vulkan.c > @@ -21,6 +21,7 @@ > #include "wine/port.h" > > #include "wine/debug.h" > +#include "wine/library.h" > #include "wine/vulkan.h" > > #ifdef SONAME_LIBVULKAN > @@ -39,6 +40,19 @@ WINE_DEFAULT_DEBUG_CHANNEL(vulkan); > void* WINAPI X11DRV_vk_icdGetInstanceProcAddr(VkInstance instance, const char* pName); > > > +static BOOL wine_vk_init(void) > +{ > + static BOOL init_done = FALSE; > + static void *vulkan_handle; > + > + if (init_done) return (vulkan_handle != NULL); > + init_done = TRUE; It's probably better to use InitOnceExecuteOnce().

1 0

Re: [PATCH 09/10] winex11: Add stubs for core Vulkan instance functions.
by Józef Kucia 01 Nov '17

01 Nov '17

On Tue, Oct 31, 2017 at 6:24 PM, Roderick Colenbrander <thunderbird2k(a)gmail.com> wrote: > Signed-off-by: Roderick Colenbrander <thunderbird2k(a)gmail.com> > --- > dlls/winex11.drv/vulkan.c | 24 +++++++++- > dlls/winex11.drv/vulkan_private.h | 2 + > dlls/winex11.drv/vulkan_thunks.c | 98 +++++++++++++++++++++++++++++++++++++++ > 3 files changed, 122 insertions(+), 2 deletions(-) > > diff --git a/dlls/winex11.drv/vulkan.c b/dlls/winex11.drv/vulkan.c > index f1c41908d8..70adbe5a92 100644 > --- a/dlls/winex11.drv/vulkan.c > +++ b/dlls/winex11.drv/vulkan.c > @@ -98,6 +98,20 @@ static void *wine_vk_get_global_proc_addr(const char* name) > return NULL; > } > > +static void *wine_vk_get_instance_proc_addr(const char* name) It would be nice to have asterisks placed consistently after a space. A consistent style would also be nice in vulkan_thunks.c, if possible.

1 0

Re: [PATCH 00/10] winex11: Add Vulkan ICD.
by Józef Kucia 01 Nov '17

01 Nov '17

Thanks for working on this! On Tue, Oct 31, 2017 at 6:24 PM, Roderick Colenbrander <thunderbird2k(a)gmail.com> wrote: > Hi all, > > This patch series starts the implementation of a Vulkan ICD > into winex11.drv (and later in other drivers). > > For people not familiar with Vulkan, let provide a brief overview > on why we need an ICD and why it should be in the display drivers > themselves. > > Application developers typically interface with Vulkan through the > Vulkan loader library (vulkan-1.dll on Windows). The loader is > open source (more on this later) and its upto the developer to > link it dynamically or statically (yuck). Some developers skip > the loader and directly load the ICD in the hope to get that > magical 0.1% fps boost. > > The loader hides loading of the ICDs which are provided by the > hardware vendors. In addition the loader provides a layering > mechanism, which allows layer libraries to get inserted in > between loader and ICD. Some reference layers can help with > debugging, but e.g. Steam uses layers to inject overlays. > > A Vulkan implementation in Wine must be in the form of an ICD > as demonstrated above. The implemention in Wine staging just > replaces the loader and would have the issues mentioned above. Yes, a Vulkan implementation in the form of an ICD is preferred to make using layers possible. > > Vulkan was retrofitted into existing operating systems, so > integration with Win32 is fairly minimal. For example there > is no path like with gdi32 by which Vulkan can reach the > graphics drivers, so Vulkan loads the display driver directly. > Registry settings and json files provide Vulkan this detail. > > I decided to place the ICD directly in the display driver > as ultimately we need some tight coupling for window > rendering and also child window rendering (this is possible, > though I haven't seen applications use it). > > The amount of code sharing between ICDs, I expect to be mostly in > generated code and maybe a few helper functions. For now > I think it is best to duplicate code and longer-term if > a little bit of sharing makes sense put this in a staticly > linked helper library. I want to avoid implementing our own > 'winevulkan' glue ICD library in between loader and display > driver as I fear this will cause many issues, because the object > design didn't lend for this (assumes 1 ICD = 1 driver). Besides > some applications load the ICD directly. > > This first wave of patches gets some initial ICD infrastructure > in place. Enough to create a minimal Vulkan Instance and get > some function loading going. It will take another 6-7 patches > or so before 'vulkaninfo' the Vulkan equivalent of glxinfo > will work to some degree. > > For now the easiest way to load the ICD is to add a registry key: > [HKEY_LOCAL_MACHINE\SOFTWARE\Khronos\Vulkan\Drivers\] > "C:\Windows\System32\winex11.json"=dword:00000000 > > winex11.json: > { > "file_format_version": "1.0.0", > "ICD": { > "library_path": "c:\\windows\\system32\\winex11.drv", > "api_version": "1.0.51" > } > } > > One other comment in regards to licensing. For now I'm generating > code from vk.xml version 1.0.51. This is NOT the latest version, > but is the last BSD licensed version. Newer versions are Apache > licensed and can't be used by us as we are a derived work and > would be Apache. I have reached out to Khronos a few weeks ago > and they want to fix this. They thought Apache made things easy, > but didn't think about generated code and the license implications. > They are actively working on a solution either dual licensing or > additing some kind of code generation exception. This is a slow > process as it involves 100+ companies, but the Neil Trevett of > Khronos wants to get this done. > > Thanks, > Roderick > > Roderick Colenbrander (10): > vulkan: Add Wine vulkan header. > winex11: Add stub Vulkan ICD > winex11: Implement vk_icdNegotiateLoaderICDInterfaceVersion > winex11: Add stubs for global Vulkan functions. > winex11: Add initial vkEnumerateInstanceExtensionProperties > implementation. > winex11: Load vulkan library. > winex11: add logging functions for vkCreateInstance > winex11: Add initial vkCreateInstance implementation. > winex11: Add stubs for core Vulkan instance functions. > winex11: Implement vkDestroyInstance. > > configure.ac | 9 + > dlls/winex11.drv/Makefile.in | 2 + > dlls/winex11.drv/vulkan.c | 332 +++++ > dlls/winex11.drv/vulkan_private.h | 75 + > dlls/winex11.drv/vulkan_thunks.c | 110 ++ > dlls/winex11.drv/winex11.drv.spec | 4 + > include/config.h.in | 3 + > include/wine/vulkan.h | 2889 +++++++++++++++++++++++++++++++++++++ > 8 files changed, 3424 insertions(+) > create mode 100644 dlls/winex11.drv/vulkan.c > create mode 100644 dlls/winex11.drv/vulkan_private.h > create mode 100644 dlls/winex11.drv/vulkan_thunks.c > create mode 100644 include/wine/vulkan.h > > -- > 2.13.6 > > >

1 0

Re: [PATCH] wmvcore/tests: Access IWMDRMReader(|2|3) interfaces on Windows.
by Nikolay Sivov 31 Oct '17

31 Oct '17

On 24.10.2017 8:36, Jefferson Carpenter wrote: > +#if defined(__MINGW32__) || defined(__MINGW64__) > +#ifdef _WIN64 > + flagPtrAddr = (unsigned long long)(reader) - 24; > + flagMem[556] = 1; > +#elif _WIN32 > + flagPtrAddr = (unsigned int)(reader) - 12; > + flagMem[288] = 1; > +#endif > + *(unsigned char **)flagPtrAddr = flagMem; > +#endif Why would you need something like that to match potential Wine's implementation? How do applications access this thing?

3 10

Re: [PATCH 4/4] dx8vb: Implement Direct3D8 methods
by Stefan Dösinger 31 Oct '17

31 Oct '17

Hi Fabian, Why do you need a proxy IDirect3D8 implementation in dx8vb? Is there any difference between dx8vb and d3d8? Stefan > Am 31.10.2017 um 00:31 schrieb Fabian Maurer <dark.shadow4(a)web.de>: > > Signed-off-by: Fabian Maurer <dark.shadow4(a)web.de> > --- > dlls/dx8vb/Makefile.in | 2 +- > dlls/dx8vb/directx8.c | 112 ++++++++++++++++++++++++++++++++++++++----------- > 2 files changed, 89 insertions(+), 25 deletions(-) > > diff --git a/dlls/dx8vb/Makefile.in b/dlls/dx8vb/Makefile.in > index 8438d022f4..33fe651503 100644 > --- a/dlls/dx8vb/Makefile.in > +++ b/dlls/dx8vb/Makefile.in > @@ -1,5 +1,5 @@ > MODULE = dx8vb.dll > -IMPORTS = uuid ole32 > +IMPORTS = uuid ole32 d3d8 > > C_SRCS = \ > d3dx8.c \ > diff --git a/dlls/dx8vb/directx8.c b/dlls/dx8vb/directx8.c > index 338862a4ff..dd50c86008 100644 > --- a/dlls/dx8vb/directx8.c > +++ b/dlls/dx8vb/directx8.c > @@ -21,6 +21,9 @@ > #include "config.h" > > #include "dx8vb_private.h" > +#define __WINE_D3D8TYPES_H > +#define __WINE_D3D8CAPS_H > +#include "d3d8.h" > #include "ocidl.h" > > #include "wine/debug.h" > @@ -36,6 +39,7 @@ typedef struct > typedef struct > { > Direct3D8 Direct3D8_iface; > + IDirect3D8 *direct3d8; > LONG ref; > } direct3d8; > > @@ -95,7 +99,10 @@ static ULONG WINAPI direct3d8_Release(Direct3D8 *iface) > TRACE("(%p/%p)->(): new ref %d\n", iface, This, ref); > > if (!ref) > + { > + IDirect3D8_Release(This->direct3d8); > HeapFree(GetProcessHeap(), 0, This); > + } > > return ref; > } > @@ -104,91 +111,145 @@ static ULONG WINAPI direct3d8_Release(Direct3D8 *iface) > > static HRESULT WINAPI direct3d8_RegisterSoftwareDevice(Direct3D8 *iface, void *initializeFunction) > { > - FIXME("(%p): stub!\n", initializeFunction); > + direct3d8 *This = impl_from_Direct3D8(iface); > > - return E_NOTIMPL; > + HRESULT ret = IDirect3D8_RegisterSoftwareDevice(This->direct3d8, initializeFunction); > + > + FIXME("(%p) -> (%x)\n", initializeFunction, ret); > + > + return ret; > } > > static int WINAPI direct3d8_GetAdapterCount(Direct3D8 *iface) > { > - FIXME("(): stub!\n"); > + direct3d8 *This = impl_from_Direct3D8(iface); > + > + int ret = IDirect3D8_GetAdapterCount(This->direct3d8); > + > + FIXME("() -> (%i)\n", ret); > > - return 0; > + return ret; > } > > static HRESULT WINAPI direct3d8_GetAdapterIdentifier(Direct3D8 *iface, int adapter, > LONG flags, D3DADAPTER_IDENTIFIER8 *identifier) > { > - FIXME("(%i, %i, %p): stub!\n", adapter, flags, identifier); > + direct3d8 *This = impl_from_Direct3D8(iface); > > - return E_NOTIMPL; > + HRESULT ret = IDirect3D8_GetAdapterIdentifier(This->direct3d8, adapter, flags, identifier); > + > + FIXME("(%i, %i, %p) -> (%x)\n", adapter, flags, identifier, ret); > + > + return ret; > } > > static int WINAPI direct3d8_GetAdapterModeCount(Direct3D8 *iface, int adapter) > { > - FIXME("(%i): stub!\n", adapter); > + direct3d8 *This = impl_from_Direct3D8(iface); > > - return 0; > + int ret = IDirect3D8_GetAdapterModeCount(This->direct3d8, adapter); > + > + FIXME("(%i) -> (%i)\n", adapter, ret); > + > + return ret; > } > > static HRESULT WINAPI direct3d8_EnumAdapterModes(Direct3D8 *iface, int adapter, int mode, D3DDISPLAYMODE *displayMode) > { > - FIXME("(%i, %i, %p): stub!\n", adapter, mode, displayMode); > + direct3d8 *This = impl_from_Direct3D8(iface); > > - return E_NOTIMPL; > + HRESULT ret = IDirect3D8_EnumAdapterModes(This->direct3d8, adapter, mode, displayMode); > + > + FIXME("(%i, %i, %p) -> (%x)\n", adapter, mode, displayMode, ret); > + > + return ret; > } > > static HRESULT WINAPI direct3d8_GetAdapterDisplayMode(Direct3D8 *iface, int adapter, D3DDISPLAYMODE *displayMode) > { > - FIXME("(%i, %p): stub!\n", adapter, displayMode); > + direct3d8 *This = impl_from_Direct3D8(iface); > + HRESULT ret; > > - return E_NOTIMPL; > + ret = IDirect3D8_GetAdapterDisplayMode(This->direct3d8, adapter, displayMode); > + > + TRACE("(%i, %p) -> (%x)\n", adapter, displayMode, ret); > + > + return ret; > } > > static LONG WINAPI direct3d8_CheckDeviceType(Direct3D8 *iface, int adapter, D3DDEVTYPE checkType, > D3DFORMAT displayFormat, D3DFORMAT backBufferFormat, LONG bWindowed) > { > - FIXME("(%i, %u, %u, %u, %i): stub!\n", adapter, checkType, displayFormat, backBufferFormat, bWindowed); > + direct3d8 *This = impl_from_Direct3D8(iface); > > - return 0; > + LONG ret = IDirect3D8_CheckDeviceType(This->direct3d8, adapter, checkType, > + displayFormat, backBufferFormat, bWindowed); > + > + FIXME("(%i, %u, %u, %u, %i) -> (%i)\n", adapter, checkType, displayFormat, backBufferFormat, bWindowed, ret); > + > + return ret; > } > > static LONG WINAPI direct3d8_CheckDeviceFormat(Direct3D8 *iface, int adapter, D3DDEVTYPE deviceType, > D3DFORMAT adapterFormat, LONG usage, D3DRESOURCETYPE rType, D3DFORMAT checkFormat) > { > - FIXME("(%i, %u, %u, %i, %u, %u): stub!\n", adapter, deviceType, adapterFormat, usage, rType, checkFormat); > + direct3d8 *This = impl_from_Direct3D8(iface); > > - return 0; > + LONG ret = IDirect3D8_CheckDeviceFormat(This->direct3d8, adapter, deviceType, > + adapterFormat, usage, rType, checkFormat); > + > + FIXME("(%i, %u, %u, %i, %u, %u) -> (%i)\n", adapter, deviceType, adapterFormat, usage, rType, checkFormat, ret); > + > + return ret; > } > > static LONG WINAPI direct3d8_CheckDeviceMultiSampleType(Direct3D8 *iface, int adapter, D3DDEVTYPE deviceType, > D3DFORMAT renderTargetFormat, LONG windowed, D3DMULTISAMPLE_TYPE multiSampleType) > { > - FIXME("(%i, %u, %u, %i, %u): stub!\n", adapter, deviceType, renderTargetFormat, windowed, multiSampleType); > + direct3d8 *This = impl_from_Direct3D8(iface); > + > + LONG ret = IDirect3D8_CheckDeviceMultiSampleType(This->direct3d8, adapter, deviceType, > + renderTargetFormat, windowed, multiSampleType); > + > + FIXME("(%i, %u, %u, %i, %u) -> (%i)\n", adapter, deviceType, renderTargetFormat, windowed, multiSampleType, ret); > > - return 0; > + return ret; > } > > static LONG WINAPI direct3d8_CheckDepthStencilMatch(Direct3D8 *iface, int adapter, D3DDEVTYPE deviceType, > D3DFORMAT adapterFormat, D3DFORMAT renderTargetFormat, D3DFORMAT depthStencilFormat) > { > - FIXME("(%i, %u, %u, %u, %u): stub!\n", adapter, deviceType, adapterFormat, renderTargetFormat, depthStencilFormat); > + direct3d8 *This = impl_from_Direct3D8(iface); > + > + LONG ret = IDirect3D8_CheckDepthStencilMatch(This->direct3d8, adapter, deviceType, adapterFormat, > + renderTargetFormat, depthStencilFormat); > + > + TRACE("(%i, %u, %u, %u, %u) ->(%i)\n", adapter, deviceType, adapterFormat, > + renderTargetFormat, depthStencilFormat, ret); > > - return 0; > + return ret; > } > > static HRESULT WINAPI direct3d8_GetDeviceCaps(Direct3D8 *iface, int adapter, D3DDEVTYPE deviceType, D3DCAPS8 *caps) > { > - FIXME("(%i, %u, %p): stub!\n", adapter, deviceType, caps); > + direct3d8 *This = impl_from_Direct3D8(iface); > > - return E_NOTIMPL; > + HRESULT ret = IDirect3D8_GetDeviceCaps(This->direct3d8, adapter, deviceType, caps); > + > + TRACE("(%i, %u, %p) -> (%x)\n", adapter, deviceType, caps, ret); > + > + return ret; > } > > static LONG WINAPI direct3d8_GetAdapterMonitor(Direct3D8 *iface, int adapter) > { > - FIXME("(%i): stub!\n", adapter); > + direct3d8 *This = impl_from_Direct3D8(iface); > + > + LONG ret = (LONG)IDirect3D8_GetAdapterMonitor(This->direct3d8, adapter); > + > + TRACE("(%i) -> (%i)!\n", adapter, ret); > > - return 0; > + return ret; > } > > static HRESULT WINAPI direct3d8_CreateDevice(Direct3D8 *iface, int adapter, D3DDEVTYPE deviceType, HWND hFocusWindow, > @@ -235,6 +296,9 @@ HRESULT direct3d8_create(Direct3D8 **ppv) > > object->Direct3D8_iface.lpVtbl = &Direct3D8_Vtbl; > object->ref = 1; > + object->direct3d8 = Direct3DCreate8(D3D_SDK_VERSION); > + if(!object->direct3d8) > + ERR("Can't create D3D8 object\n"); > > *ppv = &object->Direct3D8_iface; > > -- > 2.15.0 > > >

2 3

Re: [PATCH 1/4] dx8vb: Add stubs for D3DX8 interface
by Nikolay Sivov 31 Oct '17

31 Oct '17

On Tue, Oct 31, 2017 at 2:31 AM, Fabian Maurer <dark.shadow4(a)web.de> wrote: > > main.c > > IDL_SRCS = dx8vb.idl > diff --git a/dlls/dx8vb/d3dx8.c b/dlls/dx8vb/d3dx8.c > new file mode 100644 > index 0000000000..2959849547 > --- /dev/null > +++ b/dlls/dx8vb/d3dx8.c > @@ -0,0 +1,861 @@ > + /* > + * Copyright 2017 Fabian Maurer > + * > + * This library is free software; you can redistribute it and/or > + * modify it under the terms of the GNU Lesser General Public > + * License as published by the Free Software Foundation; either > + * version 2.1 of the License, or (at your option) any later version. > + * > + * This library is distributed in the hope that it will be useful, > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU > + * Lesser General Public License for more details. > + * > + * You should have received a copy of the GNU Lesser General Public > + * License along with this library; if not, write to the Free Software > + * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, > USA > + */ > + > +#define COBJMACROS > + > +#include "config.h" > + > +#include "dx8vb_private.h" > +#include "ocidl.h" > + > +#include "wine/debug.h" > + > +WINE_DEFAULT_DEBUG_CHANNEL(dx8vb); > + > +typedef struct > +{ > + ID3DX8 ID3DX8_iface; > + LONG ref; > +} d3dx8; > + > +static inline d3dx8 *impl_from_ID3DX8(ID3DX8 *iface) > +{ > + return CONTAINING_RECORD(iface, d3dx8, ID3DX8_iface); > +} > + > +/*** d3dx8 - IUnknown methods ***/ > + > Comment seems a bit useless, you're not going to use this file for anything else. > +static HRESULT WINAPI d3dx8_QueryInterface(ID3DX8 *iface, REFIID riid, > void **ppv) > +{ > + d3dx8 *This = impl_from_ID3DX8(iface); > + > + TRACE("(%p/%p)->(%s,%p)\n", iface, This, debugstr_guid(riid), ppv); > + > + *ppv = NULL; > + > + if (IsEqualGUID(riid, &IID_IUnknown)) > + *ppv = &This->ID3DX8_iface; > + else if(IsEqualGUID(riid, &IID_ID3DX8)) > + *ppv = &This->ID3DX8_iface; > + else if(IsEqualGUID(riid, &IID_IPersistPropertyBag)) > + FIXME("No interface for IID_IPersistPropertyBag\n"); > + else if(IsEqualGUID(riid, &IID_IPersistStreamInit)) > + FIXME("No interface for IID_IPersistStreamInit\n"); > + else > Where does this come from?

2 1

Re: [PATCH 3/4] dx8vb: Add Direct3D8 interface with stubs
by Nikolay Sivov 31 Oct '17

31 Oct '17

> > > static HRESULT WINAPI directx8_QueryInterface(IDirectX8 *iface, REFIID > riid, void **ppv) > @@ -237,9 +440,11 @@ static HRESULT WINAPI directx8_ > DirectPlayAddressCreate(IDirectX8 *iface, DirectP > > static HRESULT WINAPI directx8_Direct3DCreate(IDirectX8 *iface, > Direct3D8 **ret) > { > - FIXME("(%p): stub!\n", ret); > + HRESULT res = direct3d8_create(ret); > > - return E_NOTIMPL; > + TRACE("(%p) -> (%x)\n", ret, res); > + > + return res; > } > That's an unusual way to arrange things. Why don't you return as "return direct3d8_create();" ?

2 1

Wineconference keynote 2017
by Marcus Meissner 30 Oct '17

30 Oct '17

Hi folks, I have uploaded Alexandres Wineconference 2017 Keynote to youtube: https://www.youtube.com/watch?v=AokFgDSLMWU&t=772s Ciao, Marcus

2 1

WineConf agenda
by Jacek Caban 29 Oct '17

29 Oct '17

Hi all, It's getting close to WineConf and I think it's time to think about agenda. Does anyone have anything he'd like to talk about? Any ideas about conversations we should have? Feel free to add them to https://wiki.winehq.org/WineConf2017 Cheers, Jacek

4 4

Re: [v2] oleaut32: Add ARM support to DispCallFunc().
by Donna Whisnant 28 Oct '17

28 Oct '17

V3 has been posted. Other comments are inlined below... > > Hi, > > thanks, but it seems I missed minor things last time: > > > + args = heap_alloc(sizeof(VARIANT) * cActuals + sizeof(DWORD) * 4 ); > > Try to be consistent with the spaces for each function call. > Sorry. That one came from a cut-and-paste of the i386 code and I missed fixing it. > > + case VT_R4: /* these must be 4-byte aligned, and put in 's' regs or stack, as they are single-floats */ > > + if (scount < 16) > > + regs.sd.s[scount++] = V_R4(arg); > > + else > > + args[argspos++] = V_UI4(arg); > > + break; > ... > > + case VT_VARIANT: > > + /* 8-byte align 'r' and/or stack: */ > > + if (rcount < 3) > > + { > > + rcount += (rcount % 2); > > + } > > + else > > + { > > + rcount = 4; > > + argspos += (argspos % 2); > > + } > ... > > + case VT_BOOL: /* VT_BOOL is 16-bit but BOOL is 32-bit, needs to be extended */ > > + if (rcount < 4) > > + regs.r[rcount++] = V_BOOL(arg); > > + else > > + args[argspos++] = V_BOOL(arg); > > + break; > > + default: > > + if (rcount < 4) > > + regs.r[rcount++] = V_UI4(arg); > > + else > > + args[argspos++] = V_UI4(arg); > > + break; > > + } > > Maybe one of those is not really conistent wrt curly brackets > That was actually intentional, as I made the 'if' in the VT_VARIANT case consistent with itself. All of the others cited needed no curly braces on either half of the 'if'. However, that one had to have curly braces on the 'else', so I made it consistent on the other side too and used curly braces throughout the 'if'. My usage of curly braces with respect to 'if' statements was to use them on both halves or neither half... But since you want them removed, I have removed the extra curly braces on that one in the v3 post. > > + case VT_CY: > > + V_UI8(pvargResult) = call_method(func, argspos, args, (DWORD*)&regs ); > > + break; > > + default: > > + V_UI4(pvargResult) = call_method( func, argspos, args, (DWORD*)&regs ); > > + break; > > space issue again, maybe somewhere else too? > Oops, I missed that one space. Fixed. I didn't spot any others. The rest appear correct or are at least consistent throughout.

1 0

Re: [v2] oleaut32: Add ARM support to DispCallFunc().
by André Hentschel 28 Oct '17

28 Oct '17

Am 26.10.2017 um 06:39 schrieb Donna Whisnant: > Adds ARM ABI support to DispCallFunc() to allow IDispatch invoke calls > to succeed on ARM platforms. This change specifically targets only > 32-bit little-endian (ARMEL) platform CPUs. It's believed to likely be > compatible with big-endian (ARMEB) platforms, but testing for that > platform should be completed before enabling. > > Tested on Raspbian Stretch 2017-09-07 RPi image using a Qemu systemd > container to compile and run it on an x86-64 Host.. When running on > Raspberry Pi hardware, a 3G/1G split kernel needs to be used for Wine > to function. Hi, thanks, but it seems I missed minor things last time: > + args = heap_alloc(sizeof(VARIANT) * cActuals + sizeof(DWORD) * 4 ); Try to be consistent with the spaces for each function call. > + case VT_R4: /* these must be 4-byte aligned, and put in 's' regs or stack, as they are single-floats */ > + if (scount < 16) > + regs.sd.s[scount++] = V_R4(arg); > + else > + args[argspos++] = V_UI4(arg); > + break; ... > + case VT_VARIANT: > + /* 8-byte align 'r' and/or stack: */ > + if (rcount < 3) > + { > + rcount += (rcount % 2); > + } > + else > + { > + rcount = 4; > + argspos += (argspos % 2); > + } ... > + case VT_BOOL: /* VT_BOOL is 16-bit but BOOL is 32-bit, needs to be extended */ > + if (rcount < 4) > + regs.r[rcount++] = V_BOOL(arg); > + else > + args[argspos++] = V_BOOL(arg); > + break; > + default: > + if (rcount < 4) > + regs.r[rcount++] = V_UI4(arg); > + else > + args[argspos++] = V_UI4(arg); > + break; > + } Maybe one of those is not really conistent wrt curly brackets > + case VT_CY: > + V_UI8(pvargResult) = call_method(func, argspos, args, (DWORD*)&regs ); > + break; > + default: > + V_UI4(pvargResult) = call_method( func, argspos, args, (DWORD*)&regs ); > + break; space issue again, maybe somewhere else too? Otherwise I'm quite tempted to sign it off

1 0

Re: Fwd: Re: oleaut32: Add ARM support to DispCallFunc().
by Donna Whisnant 26 Oct '17

26 Oct '17

Thanks. I did miss it, as I had subscribed to the patches list, but forgot to subscribe the devel list. And thanks for your feedback. One additional comment regarding the tests, as I didn't see mention of that in this thread. My code passes the tests for this DispCallFunc() function (I did verify that) except for the specialized case that tests for win64 functionality in DispCallFunc(). That test was designed to verify stdcall vs. cdecl on 32-bit x86 platforms. However, like the 64-bit x86, ARM (both 32 and 64) doesn't use stdcall in the traditional sense of stack cleaning. Thus, that test needs to be #ifdef'd out on ARM as it isn't valid. But I didn't like how that looked and it wasn't consistent with the existing 64-bit version. I thought a better solution would be to change it to only be there for "__i386__" instead, as I think that's the only mode that will ever need it. So if you are wondering why my patch set didn't touch the test code and why that particular test technically fails, that's why. I figured you guys can decide which way you want to tweak that... My other replies are inlined below... On Tue, Oct 24, 2017, at 13:37, André Hentschel wrote: > hi, just in case you missed it > > > -------- Weitergeleitete Nachricht -------- > Betreff: Re: oleaut32: Add ARM support to DispCallFunc(). > Datum: Mon, 23 Oct 2017 23:07:43 +0200 > Von: André Hentschel <nerv(a)dawncrow.de> > An: wine-devel(a)winehq.org > > Hi Donna and Welcome to the Wine Project! > > Whats your motivation for this patch? > You did a very good job for a firsttimer! > Thanks! While new to this project, I've been programming for over 37 years. My motivation on this patch was porting some ActiveX components over to ARM to run on a Raspberry Pi Zero W for a data logging device I'm designing for the CAN bus. Those components use IDispatch for invocation rather than the vtable, as they were designed to be extensible without requiring users to recompile their code, and this was the last piece of the puzzle to make them fully work on ARM. > Some more notes below: > > Am 22.10.2017 um 03:04 schrieb Donna Whisnant: > > Adds ARM ABI support to DispCallFunc() to allow IDispatch invoke calls > > to succeed on ARM platforms. This change specifically targets only > > 32-bit little-endian (ARMEL) platform CPUs. It's believed to likely be > > compatible with 64-bit and/or big-endian (ARMEB) platforms, but testing > > for those other platforms should be completed before enabling. > > > > Tested on Raspbian Stretch 2017-09-07 RPi image using a Qemu systemd > > container to compile and run it on an x86-64 Host.. When running on > > Raspberry Pi hardware, a 3G/1G split kernel needs to be used for Wine > > to function. > > > > Signed-off-by: Donna Whisnant <dewhisna(a)dewtronics.com> > > --- > > dlls/oleaut32/typelib.c | 221 +++++++++++++++++++++++++++++++++++++++++++++++- > > 1 file changed, 219 insertions(+), 2 deletions(-) > > > > diff --git a/dlls/oleaut32/typelib.c b/dlls/oleaut32/typelib.c > > index ebf6d85..d7ef212 100644 > > --- a/dlls/oleaut32/typelib.c > > +++ b/dlls/oleaut32/typelib.c > > @@ -6302,8 +6302,48 @@ static HRESULT WINAPI ITypeInfo_fnGetIDsOfNames( ITypeInfo2 *iface, > > return DISP_E_UNKNOWNNAME; > > } > > > > +#if defined(__arm__) && defined(__ARM_32BIT_STATE) && !defined(__ARMEB__) > > +/* Note: this may work OK on 64-bit and/or ARMEB, but has currently only been tested for 32-bit ARMEL */ > > While i doubt there is something risky here regarding endianess you can > keep the ARMEB check, but get rid of the 32bit check, it might be about > Thunk(16bit) vs. ARM(32bit) but I'm not sure. > Anyway, this won't work on 64bit, because the instruction set is totally > different, but it inspires me to do the 64bit version as soon this lands > upstream ;) > Interesting. Between the ARMEB/ARMEL modes and the 32/64 bit check, I would have left the 32-bit check and tossed the !ARMEB check, which is the opposite of your instructions. I'm fairly certain there isn't anything endian-specific. But as you pointed out, the instruction set on the 64-bit is different. So I'm confused why you say to get rid of the 32-bit check. Are you talking about removing the 'define' check for 32-bit or just the comment? Can you please clarify? > > > > -#ifdef __i386__ > > +extern LONGLONG call_method( void *func, int nb_stk_args, const DWORD *stk_args, const DWORD *reg_args ); > > +__ASM_GLOBAL_FUNC( call_method, > > + /* r0 = *func > > + * r1 = nb_stk_args > > + * r2 = *stk_args (pointer to 'nb_stk_args' DWORD values to push on stack) > > + * r3 = *reg_args (pointer to 8, 64-bit d0-d7 (double) values OR as 16, 32-bit s0-s15 (float) values, followed by 4, 32-bit (DWORD) r0-r3 values) > > + */ > > + > > + "push {fp, lr}\n\t" /* Save frame pointer and return address (stack still aligned to 8 bytes) */ > > + __ASM_CFI(".cfi_adjust_cfa_offset 4\n\t") > > + __ASM_CFI(".cfi_rel_offset %fp,0\n\t") > > + "add fp, sp, #4\n\t" /* Setup frame pointer */ > > Is this adjusted FP used somewhere? (I forgot a lot of things regarding > arm32...) > In my first version of the code, I was using r4 and r5 (which had to be saved/restored) and originally had slightly different arguments coming into the function (more than was fitting in the registers) and so I was setting up the frame for easier debugging. But then I remembered I have the 'ip' register to my disposal, and that was sufficient and allowed me to toss the saving of those other registers. And, my original function signature was passing both the floating point registers and the core registers in separate structures using separate pointers to them, but then I realized that could be a single structure with both, so I tossed the extra argument and everything would then fit into the registers alone for the call, meaning the stack frame itself isn't really needed anymore. However, it's still necessary to set 'fp', though, as it's needed for cleaning the stack on the return from the callee, but I could toss the +4 adjustment of it or use "#0" for the adjustment like the compiler probably would (as I think it maps to the same instruction binary). If I do that, the ASM_CFI macro usage could be dropped too as there would be nothing for them to report. > > + __ASM_CFI(".cfi_def_cfa_register %fp\n\t") > > + > > + "lsls r1, r1, #2\n\t" /* r1 = nb_stk_args * sizeof(DWORD) */ > > + "beq 1f\n\t" /* Skip allocation if no stack args */ > > + "add r2, r2, r1\n" /* Calculate ending address of incoming stack data */ > > + "2:\tldr ip, [r2, #-4]!\n\t" /* Get next value */ > > + "str ip, [sp, #-4]!\n\t" /* Push it on the stack */ > > + "subs r1, r1, #4\n\t" /* Decrement count */ > > + "bgt 2b\n\t" /* Loop till done */ > > + > > + "1:\tvldm r3!, {s0-s15}\n\t" /* Load the s0-s15/d0-d7 arguments */ > > + "mov ip, r0\n\t" /* Save the function call address to ip before we nuke r0 with arguments to pass */ > > + "ldm r3, {r0-r3}\n\t" /* Load the r0-r3 arguments */ > > + > > + "blx ip\n\t" /* Call the target function */ > > + > > + "sub sp, fp, #4\n\t" /* Clean the stack using 'fp' to allow for either stdcall or cdecl calls */ > > + __ASM_CFI(".cfi_def_cfa %sp,4\n\t") > > + "pop {fp, pc}\n\t" /* Restore FP and return */ > > + __ASM_CFI(".cfi_same_value %fp\n\t") /* Here for completeness and symmetry with other cpus, but won't be reached */ > > + ) > > + > > +/* same function but returning single/double floating point */ > > +static float (* const call_float_method)(void *, int, const DWORD *, const DWORD *) = (void *)call_method; > > +static double (* const call_double_method)(void *, int, const DWORD *, const DWORD *) = (void *)call_method; > > + > > +#elif defined(__i386__) > > > > extern LONGLONG call_method( void *func, int nb_args, const DWORD *args, int *stack_offset ); > > __ASM_GLOBAL_FUNC( call_method, > > @@ -6637,7 +6677,184 @@ DispCallFunc( > > void* pvInstance, ULONG_PTR oVft, CALLCONV cc, VARTYPE vtReturn, UINT cActuals, > > VARTYPE* prgvt, VARIANTARG** prgpvarg, VARIANT* pvargResult) > > { > > -#ifdef __i386__ > > +#if defined(__arm__) && defined(__ARM_32BIT_STATE) && !defined(__ARMEB__) > > +/* Note: this may work OK on 64-bit and/or ARMEB, but has currently only been tested for 32-bit ARMEL */ > > Same as above > > > + int argspos; > > + void *func; > > + UINT i; > > + DWORD *args; > > + struct { > > + union { > > + float s[16]; > > + double d[8]; > > + } sd; > > + DWORD r[4]; > > + } regs; > > + int rcount; /* 32-bit register index count */ > > + int scount; /* single-precision float register index count (will be incremented twice for doubles, plus alignment) */ > > + > > + TRACE("(%p, %ld, %d, %d, %d, %p, %p, %p (vt=%d))\n", > > + pvInstance, oVft, cc, vtReturn, cActuals, prgvt, prgpvarg, > > + pvargResult, V_VT(pvargResult)); > > You can go up to 100 chars per line, so be free to rearrange this > I could, but the trace line was a cut-and-paste from the i386 path that was already there and I figured consistency in the code was good. > > + > > + if (cc != CC_STDCALL && cc != CC_CDECL) > > + { > > + FIXME("unsupported calling convention %d\n",cc); > > + return E_INVALIDARG; > > + } > > + > > + argspos = 0; > > + rcount = 0; > > + scount = 0; > > + > > + /* Determine if we need to pass a pointer for the return value as arg 0. If so, do that */ > > + /* first as it will need to be in the 'r' registers: */ > > + switch (vtReturn) > > + { > > + case VT_DECIMAL: > > + case VT_VARIANT: > > + regs.r[rcount++] = (DWORD)pvargResult; /* arg 0 is a pointer to the result */ > > + break; > > + case VT_HRESULT: > > + WARN("invalid return type %u\n", vtReturn); > > + return E_INVALIDARG; > > + default: /* And all others are in 'r', 's', or 'd' registers or have no return value */ > > + break; > > + } > > + > > + if (pvInstance) > > + { > > + const FARPROC *vtable = *(FARPROC **)pvInstance; > > + func = vtable[oVft/sizeof(void *)]; > > + regs.r[rcount++] = (DWORD)pvInstance; /* the This pointer is always the first parameter */ > > + } > > + else func = (void *)oVft; > > + > > + /* maximum size for an argument is sizeof(VARIANT). Also allow for return pointer and stack alignment. */ > > + args = heap_alloc(sizeof(VARIANT) * cActuals + sizeof(DWORD) * 4 ); > > + > > + for (i = 0; i < cActuals; i++) > > + { > > + VARIANT *arg = prgpvarg[i]; > > + DWORD *pdwarg = (DWORD *)(arg); /* a reinterpret_cast of the variant, used for copying structures when they are split between registers and stack */ > > + int ntemp; /* Used for counting words split between registers and stack */ > > + > > + switch (prgvt[i]) > > + { > > + case VT_EMPTY: > > + break; > > + case VT_R4: /* these must be 4-byte aligned, and put in 's' regs or stack, as they are single-floats */ > > + if (scount < 16) > > + { > > + regs.sd.s[scount++] = V_R4(arg); > > + } else { > > + args[argspos++] = V_UI4(arg); > > + } > > I think no curly brackets are necessary here at all, but we usually don't > open them in the same line as the if/else statement for new code > Oh sorry. I thought I was being consistent with the rest of the code-base. I have the curly brace on the next line after the 'if', but didn't think about the 'else'. I will fix that and resubmit it. I personally like the curly braces even for single-line statements, though, as it keeps from either misreading it or accidentally adding something for the conditional and forgetting to add them. But I will fix the formatting. > > + break; > > + case VT_R8: /* these must be 8-byte aligned, and put in 'd' regs or stack, as they are double-floats */ > > + case VT_DATE: > > + if (scount < 15) > > + { > > + scount += (scount % 2); /* align scount to next whole double */ > > + regs.sd.d[scount/2] = V_R8(arg); > > + scount += 2; > > + } else { > > + scount = 16; /* Make sure we flag that all 's' regs are full */ > > + argspos += (argspos % 2); /* align argspos to 8-bytes */ > > + memcpy( &args[argspos], &V_R8(arg), sizeof(V_R8(arg)) ); > > + argspos += sizeof(V_R8(arg)) / sizeof(DWORD); > > + } > > + break; > > + case VT_I8: /* these must be 8-byte aligned, and put in 'r' regs or stack, as they are long-longs */ > > + case VT_UI8: > > + case VT_CY: > > + if (rcount < 3) > > + { > > + rcount += (rcount % 2); /* align rcount to 8-byte register pair */ > > + memcpy( &regs.r[rcount], &V_UI8(arg), sizeof(V_UI8(arg)) ); > > + rcount += sizeof(V_UI8(arg)) / sizeof(DWORD); > > + } else { > > + rcount = 4; /* Make sure we flag that all 'r' regs are full */ > > + argspos += (argspos % 2); /* align argspos to 8-bytes */ > > + memcpy( &args[argspos], &V_UI8(arg), sizeof(V_UI8(arg)) ); > > + argspos += sizeof(V_UI8(arg)) / sizeof(DWORD); > > + } > > + break; > > + case VT_DECIMAL: /* these structures are 8-byte aligned, and put in 'r' regs or stack, can be split between the two */ > > + case VT_VARIANT: > > + /* 8-byte align 'r' and/or stack: */ > > + if (rcount < 3) > > + { > > + rcount += (rcount % 2); > > + } else { > > + rcount = 4; > > + argspos += (argspos % 2); > > + } > > + ntemp = sizeof(*arg) / sizeof(DWORD); > > + while (ntemp > 0) > > + { > > + if (rcount < 4) > > + { > > + regs.r[rcount++] = *pdwarg++; > > + } else { > > + args[argspos++] = *pdwarg++; > > + } > > + --ntemp; > > + } > > + break; > > + case VT_BOOL: /* VT_BOOL is 16-bit but BOOL is 32-bit, needs to be extended */ > > + if (rcount < 4) > > + { > > + regs.r[rcount++] = V_BOOL(arg); > > + } else { > > + args[argspos++] = V_BOOL(arg); > > + } > > + break; > > + default: > > + if (rcount < 4) > > + { > > + regs.r[rcount++] = V_UI4(arg); > > + } else { > > + args[argspos++] = V_UI4(arg); > > + } > > + break; > > + } > > + TRACE("arg %u: type %s %s\n", i, debugstr_vt(prgvt[i]), debugstr_variant(arg)); > > + } > > + > > + argspos += (argspos % 2); /* Make sure stack function alignment is 8-byte */ > > + > > + TRACE("rcount: %d, scount: %d, argspos: %d\n", rcount, scount, argspos); > > + > > + switch (vtReturn) > > + { > > + case VT_EMPTY: /* EMPTY = no return value */ > > + case VT_DECIMAL: /* DECIMAL and VARIANT already have a pointer argument passed (see above) */ > > + case VT_VARIANT: > > + call_method( func, argspos, args, (DWORD*)&regs ); > > + break; > > + case VT_R4: > > + V_R4(pvargResult) = call_float_method( func, argspos, args, (DWORD*)&regs ); > > + break; > > + case VT_R8: > > + case VT_DATE: > > + V_R8(pvargResult) = call_double_method( func, argspos, args, (DWORD*)&regs ); > > + break; > > + case VT_I8: > > + case VT_UI8: > > + case VT_CY: > > + V_UI8(pvargResult) = call_method(func, argspos, args, (DWORD*)&regs ); > > + break; > > + default: > > + V_UI4(pvargResult) = call_method( func, argspos, args, (DWORD*)&regs ); > > + break; > > + } > > + heap_free( args ); > > + if (vtReturn != VT_VARIANT) V_VT(pvargResult) = vtReturn; > > + TRACE("retval: %s\n", debugstr_variant(pvargResult)); > > + return S_OK; > > + > > +#elif defined(__i386__) > > int argspos, stack_offset; > > void *func; > > UINT i; > > > > >

2 2

Re: [PATCH 4/4] d3d9/tests: Windows 10 15.07 doesn't detect external mode changes.
by Henri Verbeet 25 Oct '17

25 Oct '17

On 19 October 2017 at 10:09, Stefan Dösinger <stefan(a)codeweavers.com> wrote: > This fixes this test failure: > http://test.winehq.org/data/9787af34b1c23deb67902fc7dbf2c8421b987048/win10_… > > What I have tried to debug or try to work around this issue: > *) Checked that the resolution is actually changed. It is changed from > 640x480 to 800x600. > *) IDirect3DSwapchain9Ex::GetDisplayModeEx sees the new mode. Calling > this method does not make Present see it though. > *) Replace Present with PresentEx. > *) Call flush_events before calling Present. > > I'm happy to try more ideas, but I think this is just broken in this > Windows build. It is working OK in my 17.03 VM, and I believe it was > working on 16.07 too, but I don't have 16.07 around anymore because it > auto updated to 17.03... Should we update the Windows 10 VMs at some point? Do we want to keep the old versions around?

3 2

Re: [PATCH 3/4] mshtml/tests: Skip postMessage test on IE7.
by Marvin 24 Oct '17

24 Oct '17

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=33746 Your paranoid android. === w7pro64 (32 bit htmldoc) === htmldoc.c:3118: Test failed: Incorrect error code: -2146697211 htmldoc.c:3123: Test failed: Page address: L"http://test.winehq.org/tests/winehq_snapshot/" htmldoc.c:5895: Test failed: expected OnChanged_1012 htmldoc.c:5896: Test failed: expected Exec_HTTPEQUIV htmldoc.c:5898: Test failed: expected Exec_SETTITLE htmldoc.c:5939: Test failed: expected FireNavigateComplete2

1 0

Re: [PATCH 2/3] gdiplus: Add internal string format properties used by EMF+ records
by Vincent Povirk 24 Oct '17

24 Oct '17

I think the margins in default_drawstring_format should be 1/6. It's not currently used in GdipDrawString or GdipMeasureString, but that's something we may change in the future. Also, they're copied over to generic_default_format which would've had 1/6 margins previously.

1 0

Re: [01/10] secur32: Implement AcquireCredentialsHandle for Kerberos.
by Alexandre Julliard 24 Oct '17

24 Oct '17

Hans Leidekker <hans(a)codeweavers.com> writes: > Signed-off-by: Hans Leidekker <hans(a)codeweavers.com> > --- > configure.ac | 18 +++ > dlls/secur32/Makefile.in | 2 +- > dlls/secur32/kerberos.c | 276 +++++++++++++++++++++++++++++++++++++++++--- > dlls/secur32/secur32.c | 1 + > dlls/secur32/secur32_priv.h | 1 + > 5 files changed, 282 insertions(+), 16 deletions(-) I liked the idea of putting this in kerberos.dll. Have you given up on this? -- Alexandre Julliard julliard(a)winehq.org

3 8

Re: [PATCH] cmd: Implement builtin mklink command
by Alexandre Julliard 24 Oct '17

24 Oct '17

Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> writes: > + if(hard && !CreateHardLinkW(file1, file2, NULL)) > + WCMD_output_stderr(WCMD_LoadMessage(WCMD_READFAIL), file1); > + else if(!junction && !CreateSymbolicLinkW(file1, file2, isdir)) > + WCMD_output_stderr(WCMD_LoadMessage(WCMD_READFAIL), file1); > + else if(junction) > + WINE_TRACE("Juction links currently not supported.\n"); Surely the options have to be exclusive. It makes no sense to try to create multiple links for the same file. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [PATCH] msxml/tests: Add test for \u escape sequences in regular expressions.
by Nikolay Sivov 24 Oct '17

24 Oct '17

On 12.10.2017 22:37, Zebediah Figura wrote: > For bug #43581. > > Signed-off-by: Zebediah Figura <zfigura(a)codeweavers.com> > --- > dlls/msxml3/tests/schema.c | 26 ++++++++++++++++++++++++++ > 1 file changed, 26 insertions(+) > > diff --git a/dlls/msxml3/tests/schema.c b/dlls/msxml3/tests/schema.c > index ee84b92158..dc65811adb 100644 > --- a/dlls/msxml3/tests/schema.c > +++ b/dlls/msxml3/tests/schema.c > @@ -407,6 +407,15 @@ static const CHAR szOpenSeqXML2[] = "<test><x/><x/><y/><z/><z/><w/></test>"; > static const CHAR szOpenSeqXML3[] = "<test><w/><x/><x/><y/><z/><z/></test>"; > static const CHAR szOpenSeqXML4[] = "<test><x/><x/><y/><z/><z/><v/></test>"; > > +static const CHAR szUnicodeEscapeXDR[] = > +"<schema xmlns='http://www.w3.org/2001/XMLSchema'>" > +" <simpleType name='type1'>" > +" <restriction base='string'>" > +" <pattern value='\\u0020'/>" > +" </restriction>" > +" </simpleType>" > +"</schema>"; > + > #define check_ref_expr(expr, n) { \ > LONG ref = expr; \ > ok(ref == n, "expected %i refs, got %i\n", n, ref); \ > @@ -1154,6 +1163,23 @@ static void test_XDR_schemas(void) Is it XDR "feature" only, that they didn't add to XSDs? > IXMLDOMDocument2_Release(schema); > IXMLDOMSchemaCollection_Release(cache); > > + /* test for \u escapes */ > + schema = create_document_version(60, &IID_IXMLDOMDocument2); > + cache = create_cache_version(60, &IID_IXMLDOMSchemaCollection); > + if (!schema || !cache) > + { > + if (schema) IXMLDOMDocument2_Release(schema); > + if (cache) IXMLDOMSchemaCollection_Release(cache); > + return; > + } Is this really specific to msxml6? It's better to test on all of them to get a full picture, but probably all of them support this extension if 6th does. If you want to skip on unsupported class, just return after first failure, no need to check both pointers. > + > + ole_check(IXMLDOMDocument2_loadXML(schema, _bstr_(szUnicodeEscapeXDR), &b)); > + ok(b == VARIANT_TRUE, "failed to load XML string\n"); > + todo_wine ole_check(IXMLDOMSchemaCollection_add(cache, _bstr_(""), _variantdoc_(schema))); I think it's better to get rid of ole_check(), but it's up to you. > + > + IXMLDOMDocument2_Release(schema); > + IXMLDOMSchemaCollection_Release(cache); > + > free_bstrs(); > } > >

2 1

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

wine-devel