- wine-devel - winehq.org

Re: [v9 4/6] d3dx9: Add basic opcodes to preshader.
by Matteo Bruni 07 Apr '16

07 Apr '16

2016-04-01 13:21 GMT+02:00 Paul Gofman <gofmanp(a)gmail.com>: > +static double pres_rsq(double *args, int ncomp) > +{ > +double v; Indentation. You could also probably rename ncomp to just n. Also I get this on 32 bit: effect.c:3838: Test succeeded inside todo block: Operation rsq, component 2, expected 0x3f2c985c, got 0x3f2c985c (0.6742). effect.c:3838: Test succeeded inside todo block: Operation rsq, component 3, expected 0x1f800001, got 0x1f800001 (5.42101e-20).

2 2

Re: [v9 3/6] d3dx9: Implement fxlc constants (expressions) in effect.
by Matteo Bruni 07 Apr '16

07 Apr '16

2016-04-01 13:21 GMT+02:00 Paul Gofman <gofmanp(a)gmail.com>: > +static double regstore_get_double(struct d3dx_regstore *rs, unsigned int table, unsigned int offset) > +{ > + BYTE *p; > + > + p = (BYTE *)rs->tables[table] + table_info[table].component_size * offset; > + switch (table_info[table].type) > + { > + case PRES_VT_FLOAT : return *(float *)p; > + case PRES_VT_DOUBLE: return *(double *)p; > + case PRES_VT_INT : return *(int *)p; > + case PRES_VT_BOOL : return *(BOOL *)p ? 1.0f : 0.0f; The 'f' suffix is now wrong since you're returning double. Something like "return !!*(BOOL *)p;" would be probably preferable, although I don't think you're ever supposed to read from PRES_VT_INT or PRES_VT_BOOL tables so maybe replacing the last two cases with a FIXME or similar is even better. > +static void regstore_set_double(struct d3dx_regstore *rs, unsigned int table, unsigned int offset, double v) > +{ > + BYTE *p; > + unsigned int reg_idx; > + > + p = (BYTE *)rs->tables[table] + table_info[table].component_size * offset; > + switch (table_info[table].type) > + { > + case PRES_VT_FLOAT : *(float *)p = v; break; > + case PRES_VT_DOUBLE: *(double *)p = v; break; > + case PRES_VT_INT : *(int *)p = roundf(v); break; Same here. Also it's not clear that rounding away from 0 is correct, this would need tests (although IIRC preshaders generated by the MS compiler explicitly compute integral values for integer constants, that would make this hard to test and mostly irrelevant in practice). So please try to test for rounding behavior with 0.5 / -0.5 / 1.5, if nothing else to confirm that my memory serves me right. BTW unless testing proves otherwise lrint() is probably the best option here since it generates much nicer code (at least when building with gcc for 32 bits). > + case PRES_VT_BOOL : *(BOOL *)p = !!v; break; The !! here might also be unnecessary strictly speaking but this isn't quite as significant and it's okay to me either way. > + transpose = (desc.Class == D3DXPC_MATRIX_COLUMNS && param->class == D3DXPC_MATRIX_ROWS) || > + (param->class == D3DXPC_MATRIX_COLUMNS && desc.Class == D3DXPC_MATRIX_ROWS); It looks like you forgot to move the operator at the start of the next line here (while you did that e.g. in patch 6/6).

2 2

[GSoC] Further work on the Visual C++ iostream library
by Piotr Caban 07 Apr '16

07 Apr '16

Hi Iván, could you please provide us with timeline for your project? It doesn't have to be very specific. Thanks, Piotr

2 1

Re: [PATCH] msvcp90: Add codecvt<wchar_t>::do_encoding that returns 0
by Piotr Caban 07 Apr '16

07 Apr '16

On 04/06/16 18:51, Daniel Lehman wrote: > seems to be only msvcp90 that returns 1 > > all others return 0 (which seems more correct anyway) Msvcp100 behaves the same as msvcp90. Please also update codecvt_short virtual functions table. Thanks, Piotr

1 0

Wine is "Community Choice" SourceForge Project of the Month
by Marcus Meissner 07 Apr '16

07 Apr '16

Hi, We are the "Community Choice" Project of the Month April 2016 at SourceForge. https://sourceforge.net/blog/april-2016-community-choice-project-of-the-mon… Ciao, Marcus

2 2

Re: ntdll: Do not update iosb.Status after FSCTL_PIPE_LISTEN call. (v2)
by Marvin 07 Apr '16

07 Apr '16

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=21960 Your paranoid android. === build (build) === Patch failed to apply

1 0

Re: ntdll: Do not update iosb.Status after FSCTL_PIPE_LISTEN call. (v2)
by Marvin 07 Apr '16

07 Apr '16

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=21959 Your paranoid android. === build (build) === Patch failed to apply

1 0

Re: [4/4] ntdll: Do not update iosb.Status after FSCTL_PIPE_LISTEN call.
by Alexandre Julliard 07 Apr '16

07 Apr '16

Sebastian Lackner <sebastian(a)fds-team.de> writes: > Signed-off-by: Sebastian Lackner <sebastian(a)fds-team.de> > --- > > The change only affects FSCTL_PIPE_LISTEN because we probably want > more tests before changing the rest. Indeed we do, so you should probably move this out of the default case. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [4/4] ntdll: Do not update iosb.Status after FSCTL_PIPE_LISTEN call.
by Marvin 07 Apr '16

07 Apr '16

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=21958 Your paranoid android. === w864 (32 bit pipe) === pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: c0000000, GetLastError: 231 === w1064 (32 bit pipe) === pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: c0000000, GetLastError: 231 === w864 (64 bit pipe) === pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: c0000000, GetLastError: 231 === w1064 (64 bit pipe) === pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: c0000000, GetLastError: 231

1 0

Re: [2/4] ntdll/tests: Add tests for iosb.Status value after pending FSCTL_PIPE_LISTEN call.
by Marvin 07 Apr '16

07 Apr '16

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=21955 Your paranoid android. === w864 (32 bit pipe) === pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: c0000000, GetLastError: 231 === w1064 (32 bit pipe) === pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: c0000000, GetLastError: 231 === w864 (64 bit pipe) === pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: c0000000, GetLastError: 231 === w1064 (64 bit pipe) === pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: c0000000, GetLastError: 231

2 1

Re: [1/4] ntdll/tests: Fix some test failures on Win 8+.
by Marvin 07 Apr '16

07 Apr '16

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=21954 Your paranoid android. === w864 (32 bit pipe) === pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: c0000000, GetLastError: 231 === w1064 (32 bit pipe) === pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: c0000000, GetLastError: 231 === w864 (64 bit pipe) === pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: c0000000, GetLastError: 231 === w1064 (64 bit pipe) === pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 1, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 2, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 0, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 80000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: 40000000, GetLastError: 231 pipe.c:240: Test failed: CreateFile failed for sharing 3, access: c0000000, GetLastError: 231

1 0

Re: [PATCH 1/2] d3drm/tests: Add tests for IDirect3DRM*::CreateObject.
by Stefan Dösinger 06 Apr '16

06 Apr '16

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 Hi, Some curiosities: Am 2016-04-03 um 20:04 schrieb Aaryaman Vasishta: > + for (i = 0; i < sizeof(tests) / sizeof(*tests); ++i) > + { > + hr = IDirect3DRM_CreateObject(d3drm1, tests[i].clsid, NULL, NULL, NULL); > + todo_wine ok(hr == D3DRMERR_BADVALUE, "Test %u: expected hr == D3DRMERR_BADVALUE, got %#x.\n", i, hr); What does clsid != NULL, refiid == NULL, output object != NULL do? I guess that clsid is the object you create, and iid is the interface to that object that is returned. So I guess in theory you can pass clsid = IDirect3DRMTexture3, iid = IDirect3DRMTexture and you'd get the same result as creating a d3drmtexture3 object and then calling QI(IDirect3DRMTexture). It might be worth adding explicit tests for this in future patches. (And if you use aggregation you'll want to pass iid == IUnknown, otherwise there's no way you get the inner IUnknown and will never be able to release the object) > + ref2 = get_refcount((IUnknown *)d3drm1); > + ok(ref2 == ref1, "Test %u: expected ref2 == ref1, got ref1 = %u, ref2 = %u.\n", i, ref1, ref2); > + IUnknown_Release(unknown); > + ref2 = get_refcount((IUnknown *)d3drm1); > + ok(ref2 == ref1, "Test %u: expected ref2 == ref1, got ref1 = %u, ref2 = %u.\n", i, ref1, ref2); So what will increment the refcount? InitFrom*? > + hr = IDirect3DRM2_CreateObject(d3drm2, tests[i].clsid, NULL, tests[i].iid, (void **)&unknown); > + ok(SUCCEEDED(hr), "Test %u: expected hr == D3DRM_OK, got %#x.\n", i, hr); > + ref2 = get_refcount((IUnknown *)d3drm1); > + ok(ref2 == ref1, "Test %u: expected ref2 == ref1, got ref1 = %u, ref2 = %u.\n", i, ref1, ref2); > + ref3 = get_refcount((IUnknown *)d3drm2); > + ok(ref3 == ref1, "Test %u: expected ref3 == ref1, got ref1 = %u, ref3 = %u.\n", i, ref1, ref3); > ... > + > + hr = IDirect3DRM3_CreateObject(d3drm3, tests[i].clsid, NULL, tests[i].iid, (void **)&unknown); > + ok(SUCCEEDED(hr), "Test %u: expected hr == D3DRM_OK, got %#x.\n", i, hr); > + ref2 = get_refcount((IUnknown *)d3drm1); > + ok(ref2 == ref1, "Test %u: expected ref2 == ref1, got ref1 = %u, ref2 = %u.\n", i, ref1, ref2); > + ref3 = get_refcount((IUnknown *)d3drm3); > + ok(ref3 == ref1, "Test %u: expected ref3 == ref1, got ref1 = %u, ref3 = %u.\n", i, ref1, ref3); Since you're testing the refcounts of 1,2 and 1,3 you might as well test the refcounts of all interfaces in all 3 versions. Cheers, Stefan -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBCAAGBQJXAXWIAAoJEN0/YqbEcdMwbkIQAI9ItseZE8bef+z24Cisul9g UYxXhve3Ll9Pz5vVB+jG0H0F5iPk4vdmoHIwsJgd018xBnwXc1C0+QkTknsTHcfI LOb1l4PbjherhDNDn+4t95Hy3fdHKrutFkMhYp73m1+0V5GwY7MANCmIxu4b/z+b nW545EvcpPGluZXPcPyWiwGi9qB33VkGRvDkUpii8qQtwC+vsrDuerWcT2V7fbXj 2R0O9QSuvlbd83p16i3nVx2R8ZQRweYQoSKKdLBca07cIt3/xW6lBJOz34yUmupE 5PdKK0wOPWMKXVCuAplWuDGrvgCOtyMcIB1FmsUfxPfLjonfX3UeXoWfTwMmrMBo XkzgZj8S6JTPK2qOegzg8nGz8FZXvsvWkUZWzCtypuuAQ3rGn5p940JwIu3pW+Cq 4Z3K+r7Ka9lYxSJfiliNyhqPIlrGytHbrpN0aTmWJGn79Jqs7Ffd01+ujOQ2NlOf POWffmglWRHoYDZjSOS2JaL597hhna7ktM0xlhTWHT7HFgRB4oCV2DJAHmSeZUME egudzSfSbRgB9u8bQQpfrYr/kV6MJ0OE2d3VoTn8RRJ3RJL5UFp08N7m+YXkQW9T LAzfiyaAjNILLTCj4tpqjEikH2ftGE/hBH+tInXTKlwnf1NyB6iz5/1+dAm+7JYk MKdVO/nQCpGffRqmYNtG =H6sj -----END PGP SIGNATURE-----

2 1

Re: [PATCH] wined3d: Add Nvidia GeForce 820M
by Henri Verbeet 06 Apr '16

06 Apr '16

On 6 April 2016 at 16:06, Caron Wills <caron(a)codeweavers.com> wrote: > for bug 40348 > This ends up looking like this on the patches site: https://source.winehq.org/patches/data/121067. That's actually a valid patch, but it's not exactly ideal. Could you please use git send-email to send patches in the future?

1 0

Re: [PATCH] msvcr120: Add remainder (try3)
by Piotr Caban 06 Apr '16

06 Apr '16

Hi Daniel, On 04/06/16 01:00, Daniel Lehman wrote: > Try2: handle y==0 > Try3: add tests and error handling for both x and y, but I have a question: > > Based on the tests, remainder on 32-bit Windows differs from 64-bit Windows in errno handling. Other math functions, like acosf, behave the same - if tests for it were added, the current Wine implementation would differ from 32-bit Windows > > The #ifdef is admittedly ugly but makes it match both Windows 32-bit and 64-bit > > is it preferable to match Windows exactly? or could I drop the #ifdef and use the 64-bit errno handling, like acosf? could I skip the tests that differ on 32-bit? It depends on the function. In this case I don't think anything bad will happen if errno is set on invalid input even so it's not on Windows. I think it's OK to ignore the difference in this case and run some tests only in 32-bit or 64-bit environment. If you decide to be not fully compatible with native - please leave a comment in code. Currently the tests are not working on wine (both i386 and x86_64). Thanks, Piotr

2 2

Re: [PATCH 5/5] wined3d: Alpha test emulation for core profile contexts.
by Henri Verbeet 06 Apr '16

06 Apr '16

On 31 March 2016 at 19:51, Matteo Bruni <mbruni(a)codeweavers.com> wrote: > + shader_addline(buffer, "if (alpha_test)\n"); > + if (alpha_func != WINED3D_CMP_NEVER) > + shader_addline(buffer, " if (!(%s[0].a %s alpha_ref))\n", > + get_fragment_output(gl_info), comparison_operator[alpha_func - WINED3D_CMP_NEVER]); > + shader_addline(buffer, " discard;\n"); Is there really an advantage to having the "alpha_test" uniform instead of just mapping disabled alpha test to WINED3D_CMP_ALWAYS? > @@ -973,6 +981,7 @@ struct ps_compile_args { > WORD texcoords_initialized; /* MAX_TEXTURES, 8 */ > BOOL pointsprite; > BOOL flatshading; > + enum wined3d_cmp_func alpha_func; > }; > > enum fog_src_type { > @@ -2022,6 +2031,7 @@ struct ffp_frag_settings > unsigned char pointsprite : 1; > unsigned char flatshading : 1; > unsigned char padding : 5; > + enum wined3d_cmp_func alpha_func; "alpha_func" only needs 4 bits (3 if you just mask the 4th bit), so would still fit in the padding. That applies somewhat to struct ps_compile_args as well, although we may care less there.

2 3

Re: [v9 6/6] d3dx9: Implement setting named shader constants in effect.
by Matteo Bruni 06 Apr '16

06 Apr '16

2016-04-01 13:21 GMT+02:00 Paul Gofman <gofmanp(a)gmail.com>: A few nitpicks: > +HRESULT d3dx_set_shader_const_fxlc(struct ID3DXEffectImpl *effect, struct d3dx_pass *pass, > + struct d3dx_parameter *param, BOOL vs) > +{ > + IDirect3DDevice9 *device = effect->device; > + HRESULT hr, ret; > + struct d3dx_parameter **params; > + D3DXCONSTANT_DESC *cdesc; > + unsigned int parameters_count; > + unsigned int i, j; > + > + if (!param->param_eval) > + { > + FIXME("Preshader structure is null.\n"); > + return D3DERR_INVALIDCALL; > + } The name of the function and the FIXME above seem somewhat mixed up WRT the naming change we had earlier. > + if (FAILED(hr = d3dx_param_eval_set_shader_constants(device, param->param_eval))) > + return hr; > + params = param->param_eval->shader_inputs.inputs_param; > + cdesc = param->param_eval->shader_inputs.inputs; > + parameters_count = param->param_eval->shader_inputs.input_count; > + ret = D3D_OK; > + for (i = 0; i < parameters_count; i++) > + { > + if (params[i] && params[i]->class == D3DXPC_OBJECT && (params[i]->type == D3DXPT_SAMPLER > + || params[i]->type == D3DXPT_SAMPLER1D || params[i]->type == D3DXPT_SAMPLER2D > + || params[i]->type == D3DXPT_SAMPLER3D || params[i]->type == D3DXPT_SAMPLERCUBE)) Line continuations should have 8 spaces indentation. You can also replace i++ with ++i like in the other patches. > + TRACE("setting %u consts at %u.\n", n, start); Capitalize "setting", spell out "constants".

1 0

Re: [v9 5/6] d3dx9: Implement array selectors in effect.
by Matteo Bruni 06 Apr '16

06 Apr '16

2016-04-01 13:21 GMT+02:00 Paul Gofman <gofmanp(a)gmail.com>: > case ST_ARRAY_SELECTOR: > - FIXME("Array selector.\n"); > - break; > + { > + struct d3dx_parameter array_idx_par; I'd call that array_idx_param, for consistency. Also you can probably declare it static const and initialize it just once. > + if (FAILED(hr = d3dx9_get_param_value_ptr(effect, pass, state, &param_value, &param))) > + /* Native d3dx returns OK from BeginPass or Commit involving out of bound array > + access, and does not touch affected state. */ Nitpicks: please spell D3D_OK in full, boundS, drop the comma. Usually we put a '*' at the start of following comment lines for multiline comments, aligned with the '*' from the first line.

1 0

Re: [v9 1/6] d3dx9: Implement preshader parsing.
by Matteo Bruni 06 Apr '16

06 Apr '16

2016-04-01 13:21 GMT+02:00 Paul Gofman <gofmanp(a)gmail.com>: > Signed-off-by: Paul Gofman <gofmanp(a)gmail.com> Looks good! Just a few tiny comments: > +static void dump_bytecode(void *data, unsigned int size) > +{ > + unsigned int *bytecode = (unsigned int *)data; The cast is unnecessary. > +static unsigned int *find_bytecode_comment(unsigned int *ptr, unsigned int count, > + unsigned int fourcc, unsigned int *size) > +{ > + /* Provide at least one value in comment section on non-NULL return. */ > + while (count > 2 && (*ptr & 0xffff) == 0xfffe) > + { > + unsigned int section_size; > + > + section_size = (*ptr >> 16); > + if (!section_size || section_size + 1 > count) > + break; I guess a WARN() would be nice there. Also probably an "explicit" return NULL is a bit nicer. > +static HRESULT parse_preshader(struct d3dx_preshader *pres, unsigned int *ptr, unsigned int count, struct d3dx9_base_effect *base) Try to stay below 120 characters per line (targeting ~100). > + p = find_bytecode_comment(ptr + 1, count - 1, FOURCC_CLIT, &section_size); > + if (p) > + { > + const_count = *p++; > + if (const_count > (section_size - 1) / (sizeof(double) / sizeof(unsigned int))) > + { > + WARN("Byte code buffer ends unexpectedly.\n"); > + return D3DXERR_INVALIDDATA; > + } > + dconst = (double *)p; > + } > + else > + { > + const_count = 0; > + dconst = NULL; > + } > + TRACE("%u double constants.\n", const_count); > + > + p = find_bytecode_comment(ptr + 1, count - 1, FOURCC_FXLC, &section_size); You could decrement "count" by the amount of section_size before looking for the next comment. It probably makes sense to set *size to 0 at the beginning of find_bytecode_comment() so that you can unconditionally decrement here. > + count = byte_code_size / sizeof(unsigned int); > + if (!byte_code || !count) > + { > + *peval_out = NULL; > + return; > + } It just occurred to me, you could also check for byte_code_size % sizeof(unsigned int) while you're at it.

1 0

Re: [tools] winemono.php: Set default version to 4.6.2.
by Alexandre Julliard 06 Apr '16

06 Apr '16

Vincent Povirk <madewokherd(a)gmail.com> writes: > @@ -4,7 +4,7 @@ > */ > > // Default version if none given > -$sVersion = '4.6.0'; > +$sVersion = '4.6.2'; Do we need this? Is there anything that actually uses the default version? -- Alexandre Julliard julliard(a)winehq.org

4 3

Re: [4/4] webservices: Always skip comment and empty text nodes in WsReadType.
by Sebastian Lackner 06 Apr '16

06 Apr '16

On 06.04.2016 14:09, Hans Leidekker wrote: > + if ((hr = read_node( reader ) != S_OK)) return hr; Only moved by your patch, but this looks like a typo.

1 0

Re: [3/4] webservices: Reset the read position if an opening bracket is not followed by a valid name character.
by Sebastian Lackner 06 Apr '16

06 Apr '16

On 06.04.2016 14:09, Hans Leidekker wrote: > +static inline void read_rewind( struct reader *reader, unsigned int count ) > +{ > + while (reader->read_pos && count) > + { > + reader->read_pos--; > + count--; > + } > +} I know its also done similar in the function above, but it still looks weird to me. Is it really necessary to unroll simple math functions like addition and subtraction?

1 0

Re: user32: Use top level window as dialog parent for modal dialogs.
by Dmitry Timoshkov 06 Apr '16

06 Apr '16

Hi Jacek, Jacek Caban <jacek(a)codeweavers.com> wrote: > This is part of the fix for bug 40282. Could you please add actual tests that show the handling of disabled window state? Neither current tests nor the new ones actually check that at all, having such tests would prevent new regressions in this area. Feel free to use my tests from the staging repository as a base if desired, or write your own ones that use the same approach. -- Dmitry.

4 23

Re: krnl386.exe16: Do not reassign default handles after they got closed.
by Alexandre Julliard 06 Apr '16

06 Apr '16

Sebastian Lackner <sebastian(a)fds-team.de> writes: > @@ -363,7 +363,7 @@ HFILE16 WINAPI _lclose16( HFILE16 hFile ) > } > TRACE("%d (handle32=%p)\n", hFile, dos_handles[hFile] ); > CloseHandle( dos_handles[hFile] ); > - dos_handles[hFile] = 0; > + dos_handles[hFile] = (hFile < 5) ? INVALID_HANDLE_VALUE : 0; This will cause functions that take default handles to not return the correct errors. Is that really how it's supposed to work? What bug is this for? -- Alexandre Julliard julliard(a)winehq.org

2 2

Re: user32: Use top level window as dialog parent for modal dialogs.
by Marvin 05 Apr '16

05 Apr '16

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=21903 Your paranoid android. === w864 (64 bit win) === win.c:8544: Test failed: msg.message = 320

1 0

Re: [PATCH 2/3] mshtml: Added IHTMLWindow6::postMessage semi-stub implementation.
by Marvin 05 Apr '16

05 Apr '16

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=21874 Your paranoid android. === wxppro (32 bit) === Found no trace of the test summary line or of a crash === w2003std (32 bit) === Found no trace of the test summary line or of a crash === wvistau64 (32 bit) === Found no trace of the test summary line or of a crash === w2008s64 (32 bit) === Found no trace of the test summary line or of a crash === w7u (32 bit) === Found no trace of the test summary line or of a crash === w7pro64 (32 bit) === Found no trace of the test summary line or of a crash === w8 (32 bit) === Found no trace of the test summary line or of a crash === w864 (32 bit) === Found no trace of the test summary line or of a crash === w1064 (32 bit) === Found no trace of the test summary line or of a crash === wvistau64 (64 bit) === Found no trace of the test summary line or of a crash === w2008s64 (64 bit) === Found no trace of the test summary line or of a crash === w7pro64 (64 bit) === Found no trace of the test summary line or of a crash === w864 (64 bit) === Found no trace of the test summary line or of a crash === w1064 (64 bit) === Found no trace of the test summary line or of a crash

1 0

FW: [PATCH 1/2] ucrtbase: Add _get_initial_wide_environment
by Daniel Lehman 05 Apr '16

05 Apr '16

ignore - didn't sign. Will resend later

1 0

Wine Download Server
by Nathan 04 Apr '16

04 Apr '16

I am having difficulty downloading the three files: https://dl.winehq.org/wine/wine-appdb-*.tar.gz https://dl.winehq.org/wine/wine-bugzilla-*.tar.gz https://dl.winehq.org/wine/wine-wiki-*.xml.gz * not part of the actual file names.

6 10

Re: [PATCH 1/2] ntoskrnl.exe: add PsRemoveLoadImageNotifyRoutine stub
by Sebastian Lackner 04 Apr '16

04 Apr '16

On 04.04.2016 02:59, Austin English wrote: > /*********************************************************************** > + * PsRemoveLoadImageNotifyRoutine (NTOSKRNL.EXE.@) > + */ > + NTSTATUS WINAPI PsRemoveLoadImageNotifyRoutine(PLOAD_IMAGE_NOTIFY_ROUTINE NotifyRoutine) > + { > + FIXME( "stub: %p\n", NotifyRoutine ); > + return STATUS_SUCCESS; > + } Its too late to fix it now, but please check the formatting a bit more careful the next time. Some lines contain an unnecessary space at the beginning.

1 0

Re: kernel32: add GetVolumeInformationByHandleW stub
by Nikolay Sivov 04 Apr '16

04 Apr '16

> +@ stdcall GetVolumeInformationByHandleW(ptr wstr long ptr ptr ptr wstr long) All string buffers are out parameters. > + FIXME("%p %s %d %p %p %p %s %d\n", handle, debugstr_w(volnamebuf), volnamesize, volserial, maxlength, flags, debugstr_w(fsnamebuf), fsnamesize); Should be fixed too.

1 0

Re: [PATCH] msvcr120: Add remainder (try 2)
by Piotr Caban 04 Apr '16

04 Apr '16

On 04/04/16 17:05, Daniel Lehman wrote: > Try2: handle y == 0 You should probably also check if x and y is finite (I haven't tested it myself). Thanks, Piotr

1 0

Re: [PATCH] ucrtbase: Add some wide functions called on startup
by Piotr Caban 04 Apr '16

04 Apr '16

The functions looks like stubs. Is this is a correct implementation? If not please change TRACE to FIXME. Thanks, Piotr

2 1

Re: [PATCH 2/2] msvcr120: Add remainder
by Piotr Caban 04 Apr '16

04 Apr '16

On 04/02/16 02:09, Daniel Lehman wrote: > +/********************************************************************* > + * remainder (MSVCR120.@) > + */ > +double CDECL MSVCR120_remainder(double x, double y) > +{ > +#ifdef HAVE_REMAINDER > + return remainder(x, y); > +#else > + FIXME( "not implemented\n" ); > + return 0.0; > +#endif Please handle y==0 case. Thanks, Piotr

1 0

Re: [PATCH] wininet: Dont fail on invalid Certificates when masking errors.
by Jacek Caban 04 Apr '16

04 Apr '16

Hi Alistair, On 04/04/16 11:48, Alistair Leslie-Hughes wrote: > Signed-off-by: Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> > --- > dlls/wininet/netconnection.c | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/dlls/wininet/netconnection.c b/dlls/wininet/netconnection.c > index 7305860..723705a 100644 > --- a/dlls/wininet/netconnection.c > +++ b/dlls/wininet/netconnection.c > @@ -200,7 +200,10 @@ static DWORD netconn_verify_cert(netconn_t *conn, PCCERT_CONTEXT cert, HCERTSTOR > conn->server->cert_chain = NULL; > } > if(conn->mask_errors) > + { > conn->server->security_flags |= conn->security_flags & _SECURITY_ERROR_FLAGS_MASK; > + err = ERROR_SUCCESS; > + } I don't think we should do that. mask_errors means that we should combine errors, not ignore them. I don't know what you're trying to fix, but I'd suggest to concentrate on error source (maybe there shouldn't be an error in the first place or it should be ignored due to security_flags). If you think that the error is right, maybe we don't handle failure correctly. Cheers, Jacek

2 1

Re: [PATCH] comctrl32/tooltips: AddTool doesn't fails on empty TTTOOLINFO structure
by Marvin 04 Apr '16

04 Apr '16

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=21855 Your paranoid android. === wxppro (32 bit tooltips) === tooltips.c:382: Test failed: Adding the tool to the tooltip failed === w2003std (32 bit tooltips) === tooltips.c:382: Test failed: Adding the tool to the tooltip failed === wvistau64 (32 bit tooltips) === tooltips.c:382: Test failed: Adding the tool to the tooltip failed === w2008s64 (32 bit tooltips) === tooltips.c:382: Test failed: Adding the tool to the tooltip failed === w7u (32 bit tooltips) === tooltips.c:382: Test failed: Adding the tool to the tooltip failed === w7pro64 (32 bit tooltips) === tooltips.c:382: Test failed: Adding the tool to the tooltip failed === w8 (32 bit tooltips) === tooltips.c:382: Test failed: Adding the tool to the tooltip failed === w864 (32 bit tooltips) === tooltips.c:382: Test failed: Adding the tool to the tooltip failed === w1064 (32 bit tooltips) === tooltips.c:382: Test failed: Adding the tool to the tooltip failed === wvistau64 (64 bit tooltips) === tooltips.c:382: Test failed: Adding the tool to the tooltip failed === w2008s64 (64 bit tooltips) === tooltips.c:382: Test failed: Adding the tool to the tooltip failed === w7pro64 (64 bit tooltips) === tooltips.c:382: Test failed: Adding the tool to the tooltip failed === w864 (64 bit tooltips) === tooltips.c:382: Test failed: Adding the tool to the tooltip failed === w1064 (64 bit tooltips) === tooltips.c:382: Test failed: Adding the tool to the tooltip failed

1 0

Re: sfc_os: add SRSetRestorePointA/W stubs
by Alistair Leslie-Hughes 04 Apr '16

04 Apr '16

Hi Austin, On 04/04/16 10:24, Austin English wrote: > For https://bugs.winehq.org/show_bug.cgi?id=40397 / > https://bugs.winehq.org/show_bug.cgi?id=40410 > > > +BOOL WINAPI SRSetRestorePointA(RESTOREPOINTINFOA *restorepoint, STATEMGRSTATUS *status) >... > + return 0; > +} > It would be better to return TRUE or FALSE for the BOOL type. Best Regards Alistair

1 0

wine-1.8.2-rc2 Available
by Michael Stefaniuc 03 Apr '16

03 Apr '16

Hello, I have pushed now wine-1.8.2-rc2 (Warning: forced update). Many thanks to Sebastian for reviewing 1.8.2-rc1. As I've missed 3 dependency patches for the fix for bug #13683 I have rebased the stable branch and added a fix for a fix and two test patches. The 1.8.2-rc1 tag is still available. Missed dependencies: 85faafc user32: Ensure that button responds only to changes in WM_CAPTURECHANGED handler. 0bea2f7 comctl32: Ensure that trackbar respond only to changes in WM_CAPTURECHANGED handler. ff063d9 comctl32: Ensure that toolbar respond only to changes in WM_CAPTURECHANGED handler. Fix: 97c70c8 wined3d: Fix driver version check. Tests 1bebee9 kernel32/tests: Add test to show that CompareStringW aborts on first non-matching character. 32f7298 kernel32/tests: Add tests for RtlInterlockedPushListSList[Ex]. The changes excluding tests are: comctl32/toolbar.c | 1 + comctl32/trackbar.c | 1 + user32/button.c | 1 + wined3d/directx.c | 3 +-- 4 files changed, 4 insertions(+), 2 deletions(-) bye michael

1 0

Re: [PATCH] loader: Fix wld_prctl prototype.
by Qian Hong 03 Apr '16

03 Apr '16

On Mon, Apr 4, 2016 at 1:42 AM, Qian Hong <qhong(a)codeweavers.com> wrote: > reserve debugging Typo, should be "Reverse Debugging". One can debug with "backward step in" and "backward next". Maybe useful for Wine developers. -- Regards, Qian Hong - http://www.codeweavers.com

1 0

Re: [PATCH 1/2] t2embed: Implement TTIsEmbeddingEnabledForFacename()
by Dmitry Timoshkov 02 Apr '16

02 Apr '16

Nikolay Sivov <nsivov(a)codeweavers.com> wrote: > + if (!strcasecmp(name, facename)) strcasecmp() is not suitable for strings in windows encodings. -- Dmitry.

1 0

VRAM size autodetection on OS X
by Theodore Dubois 02 Apr '16

02 Apr '16

Sometimes, when I'm playing games on Wine, the game crashes, and then I look in the log and see "Out of adapter memory". Then I run `winetricks videomemorysize=1024`, and it runs fine. My graphics card has 1536 MB of total VRAM. It seems to me that Wine isn't correctly autodetecting the amount of VRAM I have. (DISCLAIMER: The following analysis is not based on any log output. It's based on my reading of the code, which may well be incorrect. If so, please correct me.) I just had a look at the code. In dlls/wined3d/directx.c, query_gpu_description uses wglQueryCurrentRendererIntegerWINE to obtain the GPU vendor, device, and available VRAM. wglQueryCurrentRendererIntegerWINE is part of the WGL_WINE_QUERY_RENDERER extension, which...is not implemented in winemac.drv. It's only implemented in winex11.drv. I don't use X11 if I can help it. So, with absolutely no information about what the graphics card is, Wine defaults to an Nvidia Riva 128 with WINE_DEFAULT_VIDMEM (64MB) of VRAM. This is not close. I would like to write a patch to fix this problem. I'm thinking of adding an implementation of WGL_WINE_QUERY_RENDERER to winemac.drv. There are 12 different things that can be queried about the renderer, but only WGL_RENDERER_VENDOR_ID_WINE, WGL_RENDERER_DEVICE_ID_WINE, and WGL_RENDERER_VIDEO_MEMORY_WINE are used to my knowledge, so I’d only implement those. Does this sound correct? ~Theodore

3 2

Re: ntoskrnl.exe: Make IoAllocateIrp not crash on negative values. (try 2)
by Sebastian Lackner 02 Apr '16

02 Apr '16

On 29.03.2016 22:13, Bernhard Übelacker wrote: > https://bugs.winehq.org/show_bug.cgi?id=39734 > > This patch should avoid crash in acedrv11.sys. > IoAllocateIrp is called with a stack_size of -128. > Therefore ExAllocatePool gets a negative size value. > > Tested against Windows XP. > (See the test based on wine-staging "driver testing framework" attached to the bug.) > ( https://newtestbot.winehq.org/JobDetails.pl?Key=21722 testrun by Sebastian Lackner.) > > Try 1: https://www.winehq.org/pipermail/wine-patches/2016-March/148587.html > Review 1: https://www.winehq.org/pipermail/wine-devel/2016-March/112476.html > > Changes since try 1: > - Fix usage of wrong variable. > - Use a better name for variable. > - Simplify if statement. > > Signed-off-by: Bernhard Übelacker <bernhardu(a)vr-web.de> > --- > dlls/ntoskrnl.exe/ntoskrnl.c | 9 +++++++-- > 1 file changed, 7 insertions(+), 2 deletions(-) Thanks for the updated version. Could you maybe also add a couple of additional tests with charge_quota = TRUE? Since you are changing both versions, I'm fearing a bit that your changes could break other situations. If there are no unexpected test failures it looks good to me.

2 1

Re: [PATCH] user32: Add stub for IsProcessDPIAware
by Alexandre Julliard 01 Apr '16

01 Apr '16

Louis Lenders <xerox.xerox2000x(a)gmail.com> writes: > Signed off: Louis Lenders <xerox.xerox2000x(a)gmail.com> This is still not a valid Signed-off-by header. Could you please use git to generate it, so that it matches the standard format? -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [v8 1/6] d3dx9: Implement preshader parsing.
by Matteo Bruni 31 Mar '16

31 Mar '16

2016-03-29 17:22 GMT+02:00 Paul Gofman <gofmanp(a)gmail.com>: > Signed-off-by: Paul Gofman <gofmanp(a)gmail.com> > +enum pres_reg_tables > +{ > + PRES_REGTAB_IMMED, /* immediate double constants from CLIT */ I don't quite recall, I might be contradicting my own earlier comments here but at this point this comment doesn't seem to add much. General (and late) note, we usually tend to be more verbose with variable or type names and avoid too cryptic abbreviations. I don't think it's worth changing everything now though. > +struct d3dx_preshader > +{ > + struct d3dx_regstore regs; > + > + unsigned int ins_count; > + struct d3dx_pres_ins *ins; > + > + struct d3dx_const_tab inputs; > +}; > + > + > +struct d3dx_param_eval That double empty line seems unnecessary. > -HRESULT d3dx_create_param_eval(struct d3dx9_base_effect *base_effect, void *byte_code, unsigned int byte_code_size, > +enum pres_ops > +{ > + PRESHADER_OP_NOP, > + PRESHADER_OP_MOV, > + PRESHADER_OP_MAX_ENUM > +}; Maybe rename _MAX_ENUM to _UNKNOWN? At least for this patch series you don't really need that enum value though (see below). > +static HRESULT regstore_alloc_table(struct d3dx_regstore *rs, unsigned int table) > +{ > + unsigned int size; > + > + size = rs->table_sizes[table] * table_info[table].reg_component_count * table_info[table].component_size; > + if (size) > + { > + rs->tables[table] = HeapAlloc(GetProcessHeap(), HEAP_ZERO_MEMORY, size); > + rs->table_value_set[table] = HeapAlloc(GetProcessHeap(), HEAP_ZERO_MEMORY, > + sizeof(*rs->table_value_set[table]) * > + ((rs->table_sizes[table] + PRES_BITMASK_BLOCK_SIZE - 1) / PRES_BITMASK_BLOCK_SIZE)); Indentation. > +static void regstore_set_values(struct d3dx_regstore *rs, unsigned int table, void *data, > + unsigned int start_offset, unsigned int count) > +{ > + unsigned int block_idx, start, end, start_block, end_block; > + > + if (!count) > + return; > + > + memcpy((BYTE *)rs->tables[table] + start_offset * table_info[table].component_size, > + data, count * table_info[table].component_size); Indentation. > + for (i = 0; i < PRESHADER_OP_MAX_ENUM; ++i) > + if (ins_code == pres_op_info[i].opcode) > + break; > + if (i == PRESHADER_OP_MAX_ENUM) > + { > + FIXME("Unknown opcode %#x, raw %#x.\n", ins_code, ins_raw); > + return NULL; > + } I'd use ARRAY_SIZE(pres_op_info) instead of PRESHADER_OP_MAX_ENUM. > + cdesc = HeapAlloc(GetProcessHeap(), HEAP_ZERO_MEMORY, sizeof(*cdesc) * desc.Constants); > + inputs_param = HeapAlloc(GetProcessHeap(), HEAP_ZERO_MEMORY, sizeof(*inputs_param) * desc.Constants); AFAICS neither really needs HEAP_ZERO_MEMORY. You should also check those allocations and bail out if either one fails. > +static void update_table_sizes_consts(unsigned int *table_sizes, struct d3dx_const_tab *ctab) > +{ > + unsigned int i, table, max_register; > + > + for (i = 0; i < ctab->input_count; ++i) > + { > + max_register = ctab->inputs[i].RegisterIndex + ctab->inputs[i].RegisterCount; > + table = ctab->regset2table[ctab->inputs[i].RegisterSet]; > + if (table < PRES_REGTAB_COUNT) > + table_sizes[table] = max(table_sizes[table], max_register); > + } > +} > + > +static void update_table_size(unsigned int *table_sizes, unsigned int table, unsigned int offset) > +{ > + table_sizes[table] = max(table_sizes[table], get_reg_offset(table, offset) + 1); > +} Why not use it also in update_table_sizes_consts()? The "table < PRES_REGTAB_COUNT" check should then go into the update_table_size() helper. > + if (const_count > UINT_MAX / sizeof(double)) > + { > + WARN("Invalid immediate constants count %u.\n", const_count); > + return NULL; > + } > + if (p - ptr + const_count * (sizeof(double) / sizeof(unsigned int)) >= count) > + { > + WARN("Byte code buffer ends unexpectedly.\n"); > + return NULL; > + } The first check seems unnecessary, if the idea was to avoid integer overflows in the second check you can rewrite that one instead (e.g. by dividing by sizeof(unsigned int) both sides of the inequality). That second check can probably be improved by verifying that the constants fit into the CLIT comment instead (see below). BTW, the function return type is HRESULT, those returns generate warnings for me. > + pres->ins = HeapAlloc(GetProcessHeap(), HEAP_ZERO_MEMORY, sizeof(*pres->ins) * pres->ins_count); You should probably check this allocation too. > + for (i = 0; i < pres->ins_count; ++i) > + { > + unsigned int *ptr_next; > + > + ptr_next = parse_pres_ins(p, count, &pres->ins[i]); > + if (!ptr_next) > + return D3DXERR_INVALIDDATA; > + count -= ptr_next - p; > + p = ptr_next; > + } > + > + if (!count) > + { > + WARN("Byte code buffer ends unexpectedly.\n"); > + return D3DXERR_INVALIDDATA; > + } These kinds of size checks probably work but they are somewhat hard to follow. You could instead check the size of each comment section (making sure that it doesn't overflow the whole bytecode buffer) and then check that you don't go outside of each comment while parsing it. find_bytecode_comment() could do the first check before returning and, for example, also return the comment size so that the caller can use it to validate the relevant accesses. Hopefully what I mean is clear enough, otherwise as usual feel free to ask. > +err_out: > + FIXME("Error creating parameter evaluator.\n"); > + d3dx_free_param_eval(peval); > *peval_out = NULL; > - return E_NOTIMPL; > + return; > +} That return is now unnecessary.

1 0

Re: [PATCH 2/2] d3dx10: Add D3DX10UnsetAllDeviceObjects stub
by Matteo Bruni 31 Mar '16

31 Mar '16

2016-03-29 9:07 GMT+02:00 Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com>: > Reference: https://bugs.winehq.org/show_bug.cgi?id=27035 > > Signed-off-by: Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> Applying: d3dx10: Add D3DX10UnsetAllDeviceObjects stub .git/rebase-apply/patch:35: new blank line at EOF. + .git/rebase-apply/patch:52: new blank line at EOF. + warning: 2 lines add whitespace errors. Looks okay otherwise.

1 0

Re: [PATCH v2 1/5] ddraw/tests: Add a test for color varyings clamping.
by Marvin 31 Mar '16

31 Mar '16

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=21816 Your paranoid android. === w1064 (32 bit ddraw7) === ddraw7.c:2275: Test failed: Expected {0, 0, 512, 384}, got {0, 0, 1024, 768}. === w1064 (64 bit ddraw7) === ddraw7.c:2275: Test failed: Expected {0, 0, 512, 384}, got {0, 0, 1024, 768}. === w1064 (32 bit ddraw4) === ddraw4.c:2545: Test failed: Expected {0, 0, 512, 384}, got {0, 0, 1024, 768}. === w1064 (64 bit ddraw4) === ddraw4.c:2545: Test failed: Expected {0, 0, 512, 384}, got {0, 0, 1024, 768}.

1 0

wine-1.8.2-rc1 Available
by Michael Stefaniuc 31 Mar '16

31 Mar '16

Hello guys, sorry for the delay in preparing the next Stable release. Too much stuff going on and I couldn't focus on Wine. I have pushed today an informal wine-1.8.2-rc1 to https://github.com/mstefani/wine-stable https://static-winehq.rhcloud.com/stable/ 1.8.2-rc1 is fairly complete(*) from my point of view. The plan is to give the interested developers about a week time to look at the proposed 1.8.2. On Wednesday next week I plan to hand over the "final" to Alexandre, with a probable release on Friday next week. Depending on the feedback / outcry I might do another release candidate before Wednesday. (*) I wanted to include: Bug 39092 - Regression introduced with xaudio2_8 too. It is a beast to cherry-pick due to the dependencies and dependencies of dependencies... 1.8.2-rc1 is already big enough: - I have included this time interesting commits I have found interesting, mostly based on the additional information in the patch submission. - 32 bugs fixed (of which 6 are regressions). - 86 patches (of which 7 are tests only). - With test file changes filtered out: 95 files changed, 1873 insertions(+), 770 deletions(-) - The biggest "offender" for those changes is Bug 25086 - Multiple games and applications need kernel32.GetLogicalProcessorInformationEx implementation - I have attached the filtered diffstat While I have added a (lightweight) tag to it, I didn't bother with the rest of the release stuff. I have included that info in this email though. ---------------------------------------------------------------- Bugs fixed in 1.8.2-rc1 (total 32) [12652] Tom Clancy's Rainbow Six: Vegas 2 fails to launch (number of quality levels for D3DMULTISAMPLE_NONMASKABLE multiple-sample type must not exceed 8) [13683] Ultima IX: Mouse pointer missing when reading books, under inventory [14617] Multiple Visual Basic 5/6 applications using ActiveX components depend on Standard OLE Types Typelib which is missing from 'olepro32.dll' [15473] Adobe After Effects 4 crashes on startup [25086] Multiple games and applications need kernel32.GetLogicalProcessorInformationEx implementation (CPU-Z 1.72, Splinter Cell: Blacklist, Adobe CC 2015, Kontakt 5) [30687] Powerpoint 2003/2007: embedded video play/pause on click no longer works [32827] Legrand XlPro 3 (MFC app) crashes on startup (failure to load registry-free COM servers via app hook, 'OleCreate' must ignore HRESULT of 'IOleObject::GetMiscStatus') [33787] IE8 crashes on new tab [33988] Star Wars: The Force Unleashed II - Crashes with a page fault [34174] MinGW installer 0.5-beta-20120426-1 fails to process log files (sharing violation on temp file used for piped output) [34318] Multiple games and applications need ntdll.dll.EtwRegisterTraceGuidsW (Process Hacker 2.x, Far Cry 4) [35187] Jade Empire Special Edition configuration tool always resetting video settings [36013] bleem! 1.4 (PlayStation emulator) fails on startup, complaining with 'Not enough memory to load program!' [37115] Calling ReadFile from WINBASEAPI (WinBase.h) with zero read length causes hang [37288] Not implemented feature System.Drawing.Pen.ResetTransform() [37536] Native MSVCP120.dll override requires implementation of msvcr120.dll._W_Gettnames (Desura client) [37556] CompareStringW should abort on the first nonmatching character to avoid invalid memory access. [38314] Mouse Wheel doesn't work for Jagged Alliance 2 version 1.13 (it uses MOUSEHOOKSTRUCTEX) [39591] Crimzon Clover: Page fault on launch [39838] Wreckfest Server: Crashes at startup, unimplemented function KERNEL32.dll.GetConsoleScreenBufferInfoEx & SetConsoleScreenBufferInfoEx [39866] XIIZeal (steam) immediately exits on start [39875] Starcraft can't see Direct Cable Connection games hosted on another computer [39878] Diablo II does not start on 64bit system with more than 2GB VRAM cards [39908] SuperTux 0.3.5a crashes on startup, needs msvcr120.dll.strtof [40021] IDEA v1.7 fails to start [40053] Romance of the Three Kingdoms 13 (三國志13) fails to start [40083] Heroes of the Storm Freezes/Crashes after its recent updates [40129] MSI Reading and Writing Behaves differently than Windows [40147] Assertion 'm->state == STATE_PASSIVE' failed at pulse/mainloop.c:827 on starting Skype 7.18.0.112 [40201] EccoPro only works on Wine 1.6.2 [40204] Steam client fails to start with UmePlus font installed, when running in dwrite mode [40222] IMalloc(Spy)::Free nullptr ---------------------------------------------------------------- Changes since 1.8.1: Alex Henrie (1): ntdll: Finish serial reads of size 0 immediately. Alexandre Julliard (3): make_unicode: Fix typo in Indic syllable table generation. server: Implement serial port read interval timeouts. advapi32: Fix null termination in RegQueryInfoKeyA. Alistair Leslie-Hughes (7): user32: Don't crash in DrawTextEx when tab length is zero. olepro32: Add typelib resource. include: Add GdipGetPenTransform, GdipSetPenTransform declarations. gdiplus/tests: Add Pen Transform tests. gdiplus: Implement GdipGetPenTransform. gdiplus: Implement GdipSetPenTransform. gdiplus: Implement GdipResetPenTransform. Andrew Eikum (5): oleaut32: Don't clear BSTR size. mciqtz: Implement MCI_RESUME. xaudio2: Handle openal source allocation failure. winepulse.drv: Check that the server connection actually succeeded. ntdll: Implement SystemLogicalProcessorInformationEx. André Hentschel (2): winemaker: Markup fixes in German translation. ntdll: Set TPIDRURW also on ARMv8-A. Anton Baskanov (1): quartz: Set the result to 1.0 in FilterGraph::GetRate. Aric Stewart (1): winejoystick.drv: Initialize unhandled dwPOV to JOY_POVCENTERED. Austin English (1): kernel32: Add GetConsoleScreenBufferInfoEx/SetConsoleScreenBufferInfoEx stubs. Bernhard Übelacker (2): include: Implement todo_wine_if for usage in wine tests. msvcr120: Implement strtof and _strtof_l. Charles Davis (3): winex11.drv: Actually query the swap method attribute. dwrite: Fix typos, and fix definition of Mac LangCodes for Welsh and up. configure: Check for __builtin_ms_va_list on all x86_64 platforms. Christopher Thielen (1): user32: A window may be notified with WM_CAPTURECHANGED about itself if SetCapture() is called twice. Dmitry Timoshkov (1): kernel32: CompareStringW should abort on the first nonmatching character to avoid an invalid memory access. Frédéric Delanoy (1): winemaker: Markup fixes in French translation. Hans Leidekker (2): msiexec: Fix manpage markup for repair and logging modifiers. msi: Fix the CHAR column type. Henri Verbeet (1): ntdll: Add special handling for int $0x2d exceptions. Hugh McMaster (1): reg: Remove stub output from the 'add' and 'delete' functions. Huw D. M. Davies (1): ole32: The type of the size of IMalloc'ed blocks is SIZE_T. Jacek Caban (4): ntdll: Pass SYNCHRONIZE flag in NtOpenFile and NtCreateFile calls. ntdll: Don't call NtOpenKeyEx directly from NtOpenKey. kernel32: Don't set unconditionally SECTION_QUERY flag in OpenFileMapping. ntdll: Moved RegisterTraceGuids[AW] implementation to ntdll. Jens Reyer (2): loader: Fix macro-not-defined error from man. winemaker: Fix macro-not-defined errors from man. Józef Kucia (3): wined3d: Validate render state in wined3d_device_set_render_state(). wined3d: Limit amount of video memory for older Windows versions. wined3d: Allow to query if surface format supports gamma correction. Kira Backes (1): user32: Add MOUSEHOOKSTRUCTEX for mouse wheel support. Marcus Meissner (1): manpages: Markup fixes. Matteo Bruni (2): wined3d: Limit the number of WINED3D_MULTISAMPLE_NON_MASKABLE quality levels reported. wined3d: Add WINED3DFMT_FLAG_RENDERTARGET to WINED3DFMT_B5G5R5X1_UNORM. Michael Müller (6): msvideo.dll16: Translate 16 bit address in lparam for MCIWNDM_SETTIMEFORMATA message. include: Add definitions for Win8 pseudo handles. server: Add support for Win8 pseudo handles. advapi32/tests: Add tests for Win8 pseudo handles. kernel32: Allow to pass NULL as old protection in VirtualProtect for Win9x compatibility. user32: Preserve beginning of extra data for MDI windows. Miklós Máté (2): winex11: Add 1280x960 resolution. wined3d: Warn only once about unsupported wrap state. Nikolay Sivov (12): ole2disp: Implement SafeArrayGetDim(). ole2disp: Implement SafeArrayGetElemsize(). ole2disp: Implement SafeArrayLock()/SafeArrayUnlock(). ole2disp: Implement SafeArrayAccessData()/SafeArrayUnaccessData(). include: Fix IsBad* prototypes. ctl3d32: Claim success in Ctl3dRegister(). ole32: Ignore GetMiscStatus() returned value in OleCreate(). dwrite: Handle font data initialization failure when building collections. ntdll/tests: Some tests for NtQuerySystemInformationEx(). ntdll: Add NtQuerySystemInformationEx() stub. kernel32: Forward GetLogicalProcessorInformationEx() to ntdll. ole32: Don't call spy on IMalloc::Free(NULL). Paul Gofman (1): d3dx9: Fix D3DXIntersectTri function. Piotr Caban (5): msxml3: Update xmlNode reference count when nodes are merged in node_insert_before. msxml3: Store information about ignorrable whitespace nodes in xmlNode._private variable. msxml3: Reimplement node_get_text helper so it uses information about ignorred white space nodes. msxml3: Add support for CDATA nodes in node_get_text. msvcr110: Add _W_Gettnames implementation. Sebastian Lackner (13): vcomp/tests: Reenable architecture dependent tests. vcomp: Implement 64-bit atomic instructions. ntdll: Implement RtlInterlockedPushListSListEx and fix calling convention of RtlInterlockedPushListSList. kernel32/tests: Clean up interlocked SList tests. kernel32: Forward InterlockedPushListSList[Ex] to ntdll. cmd: Fix sharing permissions of piped output files. vmm.vxd: Fix protection flags passed to VirtualAlloc. ntdll: Mimic Windows behaviour when calling debug service. ntdll: Implement handling of int $0x2d for x86_64. ntdll/tests: Add more tests for calling int $0x2d on x86 and x86_64. ntdll/tests: Add additional tests for instruction pointer in debug service tests. ntdll/tests: Add separate tests for exception handling of int $0x3. ntdll: Fix up instruction pointer in context for EXCEPTION_BREAKPOINT on x86_64. bye michael

1 0

Re: [PATCH 1/2] d3dx10: Add D3D10CreateEffectPoolFromMemory stub
by Józef Kucia 31 Mar '16

31 Mar '16

On Tue, Mar 29, 2016 at 9:07 AM, Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> wrote: > dlls/d3d10/d3d10.spec | 2 +- > dlls/d3d10/d3d10_main.c | 9 +++++++++ > dlls/d3d10_1/d3d10_1.spec | 2 +- > include/d3d10effect.h | 2 ++ The subject prefix should be "d3d10:". > +HRESULT WINAPI D3D10CreateEffectPoolFromMemory(void *data, SIZE_T data_size, UINT fx_flags, > + ID3D10Device *device, ID3D10EffectPool **effectpool) > +{ Please use "effect_pool" instead of "effectpool", mainly for consistency (snake_case). > + FIXME("data %p, data_size %lu, fx_flags %#x, device %p, effectpool %p.\n", > + data, data_size, fx_flags, device, effectpool); The FIXME() message should end with "stub!" or "stub.". > + return E_NOTIMPL; > +} > + I think you can remove this additional empty line. > diff --git a/include/d3d10effect.h b/include/d3d10effect.h > index 1494894..5dafdae 100644 > --- a/include/d3d10effect.h > +++ b/include/d3d10effect.h > @@ -832,6 +832,8 @@ HRESULT WINAPI D3D10CompileEffectFromMemory(void *data, SIZE_T data_size, const > ID3D10Blob **effect, ID3D10Blob **errors); > HRESULT WINAPI D3D10CreateEffectFromMemory(void *data, SIZE_T data_size, UINT flags, > ID3D10Device *device, ID3D10EffectPool *effect_pool, ID3D10Effect **effect); > +HRESULT WINAPI D3D10CreateEffectPoolFromMemory(void *data, SIZE_T data_size, UINT fx_flags, > + ID3D10Device *device, ID3D10EffectPool **effectpool); It should be "effect_pool" instead of "effectpool".

1 0

Re: [PATCH (try2)] setupapi: Fix spec entries for handle parameters
by Sebastian Lackner 31 Mar '16

31 Mar '16

On 30.03.2016 17:00, Andrew Eikum wrote: > Signed-off-by: Andrew Eikum <aeikum(a)codeweavers.com> > --- > > try2: Turns out there's a whole bunch more of these. I think I got > most/all of them. > > dlls/setupapi/setupapi.spec | 118 ++++++++++++++++++++++---------------------- > 1 file changed, 59 insertions(+), 59 deletions(-) Is there any specific reason why you want to change that? Using "long" for handles is usually perfectly fine and it depends on the exact implementation details if its a pointer or just an index into a handle table.

3 2

Re: [PATCH] msvcr120: Add scalbn
by Piotr Caban 31 Mar '16

31 Mar '16

On 03/31/16 08:45, Daniel Lehman wrote: @ cdecl scalbn(double long) MSVCRT__scalb It will not compile in case of i386 wine. MSVCRT__scalb is defined inside ifdef. Thanks, Piotr

1 0

Need some advice on shell32
by Mathy Vanvoorden 31 Mar '16

31 Mar '16

Hello, In an effort to get myself acquainted with the code I searched for some bugs I might be able to solve in Bugzilla. I found this one: https://bugs.winehq.org/show_bug.cgi?id=32079 Basically it says that an open file dialog in for example notepad or notepad++ doesn't show hidden files when the "Show hidden files" option is set in winecfg. I have traced this back to the fact that the ShowDotFiles option is only checked in the NT functions in ntdll, but notepad (and notepad++) use the comdlg32 open dialog which does not use those NT functions but the shell32 ones. It seems to me that the comdlg32 open dialog should check if ShowDotFiles is active and if so, to pass the SHCONTF_INCLUDEHIDDEN flag to the shell32 objects. But I might (probably?) be wrong. Can someone verify that this is the way to go before I start writing a patch? -- Met vriendelijke groeten, Best regards, Mathy Vanvoorden

3 7

Re: [PATCH 4/4] mshtml: Added IHTMLElement::outerText tests.
by Marvin 30 Mar '16

30 Mar '16

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=21790 Your paranoid android. === wxppro (32 bit dom) === dom.c:6968: Test failed: put_outerText failed: 00000000 dom.c:6971: Test failed: expected len = 2, got 1 === wvistau64 (32 bit dom) === dom.c:6968: Test failed: put_outerText failed: 00000000 dom.c:6971: Test failed: expected len = 2, got 1 === w7u (32 bit dom) === dom.c:6968: Test failed: put_outerText failed: 00000000 dom.c:6971: Test failed: expected len = 2, got 1 === w7pro64 (32 bit dom) === dom.c:6968: Test failed: put_outerText failed: 00000000 dom.c:6971: Test failed: expected len = 2, got 1 === w8 (32 bit dom) === dom.c:6968: Test failed: put_outerText failed: 00000000 dom.c:6971: Test failed: expected len = 2, got 1 === w864 (32 bit dom) === dom.c:6968: Test failed: put_outerText failed: 00000000 dom.c:6971: Test failed: expected len = 2, got 1 === w1064 (32 bit dom) === dom.c:6968: Test failed: put_outerText failed: 00000000 dom.c:6971: Test failed: expected len = 2, got 1 === wvistau64 (64 bit dom) === dom.c:6968: Test failed: put_outerText failed: 00000000 dom.c:6971: Test failed: expected len = 2, got 1 === w7pro64 (64 bit dom) === dom.c:6968: Test failed: put_outerText failed: 00000000 dom.c:6971: Test failed: expected len = 2, got 1 === w864 (64 bit dom) === dom.c:6968: Test failed: put_outerText failed: 00000000 dom.c:6971: Test failed: expected len = 2, got 1 === w1064 (64 bit dom) === dom.c:6968: Test failed: put_outerText failed: 00000000 dom.c:6971: Test failed: expected len = 2, got 1

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