June 2017 - Wine-Devel - list.winehq.org

Re: [PATCH 2/4] winemp3.acm: Disallow operations encoding to MPEG.
by Marvin June 9, 2017

June 9, 2017

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=31546 Your paranoid android. === wxppro (32 bit msacm) === msacm.c:1162: Test failed: expected 0, got 41 === w2003std (32 bit msacm) === msacm.c:1162: Test failed: expected 0, got 41 === w2008s64 (32 bit msacm) === msacm.c:1159: Test failed: failed with error 0x200 msacm.c:1160: Test failed: expected 30, got 0 msacm.c:1167: Test failed: failed with error 0x6 msacm.c:1168: Test failed: expected 1, got 0 msacm.c:1169: Test failed: expected 44100, got 0 msacm.c:1170: Test failed: expected 16, got 0 msacm.c:1171: Test failed: expected 2, got 0 msacm.c:1172: Test failed: expected 88200, got 0 === w2008s64 (64 bit msacm) === msacm.c:1159: Test failed: failed with error 0x200 msacm.c:1160: Test failed: expected 30, got 0 msacm.c:1167: Test failed: failed with error 0x6 msacm.c:1168: Test failed: expected 1, got 0 msacm.c:1169: Test failed: expected 44100, got 0 msacm.c:1170: Test failed: expected 16, got 0 msacm.c:1171: Test failed: expected 2, got 0 msacm.c:1172: Test failed: expected 88200, got 0

1 0

Re: [PATCH 3/3] d3drm/tests: Some tests for mesh builder and faces
by Henri Verbeet June 9, 2017

June 9, 2017

On 9 June 2017 at 01:36, Nikolay Sivov <nsivov(a)codeweavers.com> wrote: > +#define test_vector_eq(a, b, c, d) test_vector_eq_(__LINE__, a, b, c, d) > +static void test_vector_eq_(unsigned int line, D3DVECTOR *v, float x, float y, float z) "v" should probably be const. More generally, with the scheme we've been using for the other D3D tests, this would look like the following: static void check_vector_(unsigned int line, const D3DVECTOR *v, float x, float y, float x, unsigned int ulps) and print the ok_() as something like "Got unexpected vector {%.8e, %.8e, %.8e}, expected {%.8e, %.8e, %.8e}.\n". Note that the %.8e is because %f is lossy. See e.g. compare_vec4() in the ddraw and d3d11 tests. > + memset(v, 0, sizeof(v)); > + hr = IDirect3DRMFace_GetVertex(face1, 0, v, n); > + ok(SUCCEEDED(hr), "Failed to get vertex data, hr %#x.\n", hr); > + ok(U1(v[0]).x == 1.0f, "Wrong component v[0].x = %f (expected %f)\n", U1(v[0]).x, 1.0f); > + ok(U2(v[0]).y == 2.0f, "Wrong component v[0].y = %f (expected %f)\n", U2(v[0]).y, 2.0f); > + ok(U3(v[0]).z == 3.0f, "Wrong component v[0].z = %f (expected %f)\n", U3(v[0]).z, 3.0f); > + That should probably use check_vector(), and I think there are more places in the d3drm tests that would benefit from using check_vector(). > + color = IDirect3DRMFace_GetColor(face1); > +todo_wine > + ok(color == 0xff00ff00, "Got wrong color, %#x.\n", color); %#x is a little awkward for colours, 0x%08x would be more appropriate. I'd argue for "unexpected" over "wrong". > + hr = IDirect3DRMMeshBuilder3_GetFace(mesh_builder3, 0, &face2); > + ok(SUCCEEDED(hr), "Failed to delete a face, %#x.\n", hr); s/delete/retrieve/ I think that should be "the" instead of "a", but I think simply omitting the article would also be fine. > -#define IDirect3DRMFace_GetColor(p,a) (p)->lpVtbl->GetColor(p,a) > +#define IDirect3DRMFace_GetColor(p) (p)->lpVtbl->GetColor(p) IDirect3DRMFace2 seems to have the same issue.

1 0

Re: [PATCH v2] comctl32/taskdialog: Implement buttons.
by Nikolay Sivov June 9, 2017

June 9, 2017

On 01.06.2017 20:02, Fabian Maurer wrote: > Signed-off-by: Fabian Maurer <dark.shadow4(a)web.de> > --- > dlls/comctl32/taskdialog.c | 243 ++++++++++++++++++++++++++++++++++----------- > 1 file changed, 185 insertions(+), 58 deletions(-) > It does not apply to current git head, most likely because of my latest patch to this file. Could you please rebase, I'd like to test new wrapping logic.

1 0

Re: [2/2] ws2_32/tests: Test IP address conversion in getaddrinfo.
by Marvin June 9, 2017

June 9, 2017

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=31521 Your paranoid android. === w7u (32 bit sock) === sock.c:5979: Test failed: Expected event sequence [FD_READ(0)], got [] sock.c:5988: Test failed: Expected event sequence [FD_READ(0)], got [FD_READ(0) FD_READ(0)]

2 1

Re: Alexandre Julliard : wineandroid: Implement ToUnicodeEx.
by Dmitry Timoshkov June 9, 2017

June 9, 2017

Alexandre Julliard <julliard(a)winehq.org> wrote: > +INT CDECL ANDROID_ToUnicodeEx( UINT virt, UINT scan, const BYTE *state, > + LPWSTR buf, int size, UINT flags, HKL hkl ) > +{ > + WCHAR buffer[2]; > + BOOL shift = (state[VK_SHIFT] & 0x80) || (state[VK_CAPITAL] & 0x01); Treating Shift and CapsLock states same way leads to not correct results for the not letter keys. > + BOOL ctrl = state[VK_CONTROL] & 0x80; > + BOOL numlock = state[VK_NUMLOCK] & 0x01; > + > + buffer[0] = buffer[1] = 0; > + > + if (scan & 0x8000) return 0; /* key up */ > + > + /* FIXME: hardcoded layout */ > + > + if (!ctrl) > + { > + switch (virt) > + { > + case VK_BACK: buffer[0] = '\b'; break; > + case VK_OEM_1: buffer[0] = shift ? ':' : ';'; break; > + case VK_OEM_2: buffer[0] = shift ? '?' : '/'; break; > + case VK_OEM_3: buffer[0] = shift ? '~' : '`'; break; > + case VK_OEM_4: buffer[0] = shift ? '{' : '['; break; > + case VK_OEM_5: buffer[0] = shift ? '|' : '\\'; break; > + case VK_OEM_6: buffer[0] = shift ? '}' : ']'; break; > + case VK_OEM_7: buffer[0] = shift ? '"' : '\''; break; > + case VK_OEM_COMMA: buffer[0] = shift ? '<' : ','; break; > + case VK_OEM_MINUS: buffer[0] = shift ? '_' : '-'; break; > + case VK_OEM_PERIOD: buffer[0] = shift ? '>' : '.'; break; > + case VK_OEM_PLUS: buffer[0] = shift ? '+' : '='; break; > + case VK_RETURN: buffer[0] = '\r'; break; > + case VK_SPACE: buffer[0] = ' '; break; > + case VK_TAB: buffer[0] = '\t'; break; > + case VK_MULTIPLY: buffer[0] = '*'; break; > + case VK_ADD: buffer[0] = '+'; break; > + case VK_SUBTRACT: buffer[0] = '-'; break; > + case VK_DIVIDE: buffer[0] = '/'; break; > + default: > + if (virt >= '0' && virt <= '9') > + { > + buffer[0] = shift ? ")!@#$%^&*("[virt - '0'] : virt; While this compiles even with MSVC there are certainly more readable ways to do the same thing. -- Dmitry.

1 0

Re: [PATCH v2] shell32: Implement FCIDM_SHVIEW_CREATELINK
by Nikolay Sivov June 8, 2017

June 8, 2017

On 05.06.2017 18:42, Aric Stewart wrote: > > v2: suggestions from Nikolay Sivov > > Presently the wine file explorer has a create shortcut entry that > does nothing. This implements the FCIDM_SHVIEW_CREATELINK command. > > Signed-off-by: Aric Stewart <aric(a)codeweavers.com> > --- > dlls/shell32/shlview_cmenu.c | 49 ++++++++++++++++++++++++++++++++++++++++++++ > 1 file changed, 49 insertions(+) > > I tested this a bit more, and it's more complicated than that: - created file name format is "<target file name> - Shortcut.lnk", where "Shortcut" is localized. If such file exists, it tries to create unique name using extended format "<target file name> - Shortcut (%u).lnk", where %u starts with 2; - for special folders link is placed on the desktop; - for fs folders without write access user is prompted with message box, asking if shortcut file should be placed on the desktop or not. I think we should address at least some of those, like generating unique name. It's possible this needs extension for ISFHelper. Regarding patch itself, there's still some room for improvement: - QI check is not required, IPersistFile is always supported; - still !SUCCEEDED(); - path concatenation should use some API imho; - no allocation failure check; - MAX_PATH assumptions.

1 0

WineConf 2017
by Jacek Caban June 8, 2017

June 8, 2017

Hi all, We'd like to share with everyone current plans for WineConf. It will most likely be on 28th and 29th October 2017 in Wrocław, Sofitel hotel. It's still not final yet, but we feel that it's close enough to be worth sharing. We will send an announcement with more details once we have everything sorted out. Looking forward to see you all in Wrocław, Piotr and Jacek PS: I don't seem to have enough permissions to create a page on Wiki. Could someone please do that?

1 1

Re: [PATCH 1/2] imaadp32.acm: Fix block size calculations.
by Andrew Eikum June 8, 2017

June 8, 2017

On Wed, Jun 07, 2017 at 10:29:42AM -0500, Zebediah Figura wrote: > Signed-off-by: Zebediah Figura <z.figura12(a)gmail.com> > --- > dlls/imaadp32.acm/imaadp32.c | 11 ++--------- > 1 file changed, 2 insertions(+), 9 deletions(-) > > diff --git a/dlls/imaadp32.acm/imaadp32.c b/dlls/imaadp32.acm/imaadp32.c > index 78a0ac4c6c..ecc111c071 100644 > --- a/dlls/imaadp32.acm/imaadp32.c > +++ b/dlls/imaadp32.acm/imaadp32.c > @@ -154,7 +154,7 @@ static void init_wfx_ima_adpcm(IMAADPCMWAVEFORMAT* awfx/*, DWORD nba*/) > } > pwfx->cbSize = sizeof(WORD); > > - awfx->wSamplesPerBlock = (pwfx->nBlockAlign - (4 * pwfx->nChannels) * 2) / pwfx->nChannels + 1; > + awfx->wSamplesPerBlock = pwfx->nBlockAlign * (2 / pwfx->nChannels) - 7; If this reference is accurate: https://wiki.multimedia.cx/index.php/Microsoft_IMA_ADPCM Then I think the old version is actually almost correct. I think they just distributed the 2 factor incorrectly. Can you try this and double check my math? + awfx->wSamplesPerBlock = ((pwfx->nBlockAlign - (4 * pwfx->nChannels)) * 2) / pwfx->nChannels + 1; Andrew

2 1

Re: [PATCH v2 2/4] msacm32: Add invalid parameter checks for acmFormatDetails().
by Andrew Eikum June 7, 2017

June 7, 2017

This one adds test failures on my machine: msacm.c:542: Test failed: acmFormatDetailsA(): rc = 00000200, should be 0000000b msacm.c:542: Test failed: acmFormatDetailsA(): rc = 00000200, should be 0000000b msacm.c:542: Test failed: acmFormatDetailsA(): rc = 00000200, should be 0000000b msacm.c:542: Test failed: acmFormatDetailsA(): rc = 00000200, should be 0000000b msacm.c:542: Test failed: acmFormatDetailsA(): rc = 00000200, should be 0000000b msacm.c:542: Test failed: acmFormatDetailsA(): rc = 00000200, should be 0000000b Andrew On Wed, Jun 07, 2017 at 10:28:40AM -0500, Zebediah Figura wrote: > v2: restore zero struct, check for error from acmFormatTagDetailsW() > Signed-off-by: Zebediah Figura <z.figura12(a)gmail.com> > --- > dlls/msacm32/format.c | 34 ++++++++++-- > dlls/msacm32/tests/msacm.c | 132 +++++++++++++++++++++++++++++++++++++++++++++ > 2 files changed, 163 insertions(+), 3 deletions(-) > > diff --git a/dlls/msacm32/format.c b/dlls/msacm32/format.c > index 902807cb72..359fbe8366 100644 > --- a/dlls/msacm32/format.c > +++ b/dlls/msacm32/format.c > @@ -68,7 +68,7 @@ static BOOL CALLBACK MSACM_FillFormatTagsCB(HACMDRIVERID hadid, > HACMDRIVER had; > > if (acmDriverOpen(&had, hadid, 0) == MMSYSERR_NOERROR) { > - ACMFORMATDETAILSW afd; > + ACMFORMATDETAILSW afd = {0}; > unsigned int i, len; > MMRESULT mmr; > WCHAR buffer[ACMFORMATDETAILS_FORMAT_CHARS+16]; > @@ -87,6 +87,7 @@ static BOOL CALLBACK MSACM_FillFormatTagsCB(HACMDRIVERID hadid, > int j; > > afd.dwFormatIndex = i; > + afd.fdwSupport = 0; > mmr = acmFormatDetailsW(had, &afd, ACM_FORMATDETAILSF_INDEX); > if (mmr == MMSYSERR_NOERROR) { > lstrcpynW(buffer, afd.szFormat, ACMFORMATTAGDETAILS_FORMATTAG_CHARS + 1); > @@ -112,7 +113,7 @@ static BOOL CALLBACK MSACM_FillFormatTagsCB(HACMDRIVERID hadid, > HACMDRIVER had; > > if (acmDriverOpen(&had, hadid, 0) == MMSYSERR_NOERROR) { > - ACMFORMATDETAILSW afd; > + ACMFORMATDETAILSW afd = {0}; > > afd.cbStruct = sizeof(afd); > afd.dwFormatTag = paftd->dwFormatTag; > @@ -376,10 +377,14 @@ MMRESULT WINAPI acmFormatDetailsA(HACMDRIVER had, PACMFORMATDETAILSA pafd, > ACMFORMATDETAILSW afdw; > MMRESULT mmr; > > + if (!pafd) > + return MMSYSERR_INVALPARAM; > + > memset(&afdw, 0, sizeof(afdw)); > afdw.cbStruct = sizeof(afdw); > afdw.dwFormatIndex = pafd->dwFormatIndex; > afdw.dwFormatTag = pafd->dwFormatTag; > + afdw.fdwSupport = pafd->fdwSupport; > afdw.pwfx = pafd->pwfx; > afdw.cbwfx = pafd->cbwfx; > > @@ -401,11 +406,24 @@ MMRESULT WINAPI acmFormatDetailsW(HACMDRIVER had, PACMFORMATDETAILSW pafd, DWORD > MMRESULT mmr; > static const WCHAR fmt1[] = {'%','d',' ','H','z',0}; > static const WCHAR fmt2[] = {';',' ','%','d',' ','b','i','t','s',0}; > + ACMFORMATTAGDETAILSW aftd = {0}; > > TRACE("(%p, %p, %d)\n", had, pafd, fdwDetails); > > + if (!pafd) > + return MMSYSERR_INVALPARAM; > + > if (pafd->cbStruct < sizeof(*pafd)) return MMSYSERR_INVALPARAM; > > + if (!pafd->pwfx) > + return MMSYSERR_INVALPARAM; > + > + if (pafd->cbwfx < sizeof(PCMWAVEFORMAT)) > + return MMSYSERR_INVALPARAM; > + > + if (pafd->fdwSupport) > + return MMSYSERR_INVALPARAM; > + > switch (fdwDetails) { > case ACM_FORMATDETAILSF_FORMAT: > if (pafd->dwFormatTag != pafd->pwfx->wFormatTag) { > @@ -430,7 +448,16 @@ MMRESULT WINAPI acmFormatDetailsW(HACMDRIVER had, PACMFORMATDETAILSW pafd, DWORD > } > break; > case ACM_FORMATDETAILSF_INDEX: > - /* should check pafd->dwFormatIndex < aftd->cStandardFormats */ > + aftd.cbStruct = sizeof(aftd); > + aftd.dwFormatTag = pafd->dwFormatTag; > + mmr = acmFormatTagDetailsW(had, &aftd, ACM_FORMATTAGDETAILSF_FORMATTAG); > + if (mmr != MMSYSERR_NOERROR) > + break; > + if (pafd->dwFormatIndex >= aftd.cStandardFormats) > + { > + mmr = MMSYSERR_INVALPARAM; > + break; > + } > mmr = MSACM_Message(had, ACMDM_FORMAT_DETAILS, (LPARAM)pafd, fdwDetails); > break; > default: > @@ -581,6 +608,7 @@ static BOOL MSACM_FormatEnumHelper(PWINE_ACMDRIVERID padid, HACMDRIVER had, > for (j = 0; j < aftd.cStandardFormats; j++) { > pafd->dwFormatIndex = j; > pafd->dwFormatTag = aftd.dwFormatTag; > + pafd->fdwSupport = 0; > if (acmFormatDetailsW(had, pafd, ACM_FORMATDETAILSF_INDEX) != MMSYSERR_NOERROR) > continue; > > diff --git a/dlls/msacm32/tests/msacm.c b/dlls/msacm32/tests/msacm.c > index f5ab168290..1b47fe2087 100644 > --- a/dlls/msacm32/tests/msacm.c > +++ b/dlls/msacm32/tests/msacm.c > @@ -38,9 +38,110 @@ static BOOL CALLBACK FormatTagEnumProc(HACMDRIVERID hadid, > DWORD_PTR dwInstance, > DWORD fdwSupport) > { > + MMRESULT rc; > + HACMDRIVER had; > + > if (winetest_interactive) > trace(" Format 0x%04x: %s\n", paftd->dwFormatTag, paftd->szFormatTag); > > + rc = acmDriverOpen(&had, hadid, 0); > + ok(rc == MMSYSERR_NOERROR || rc == MMSYSERR_NODRIVER, > + "acmDriverOpen(): rc = %08x, should be %08x\n", > + rc, MMSYSERR_NOERROR); > + > + if (rc == MMSYSERR_NOERROR) > + { > + ACMFORMATDETAILSA fd = {0}; > + WAVEFORMATEX *pwfx; > + DWORD i; > + > + fd.cbStruct = sizeof(fd); > + if (paftd->cbFormatSize < sizeof(WAVEFORMATEX)) > + pwfx = HeapAlloc(GetProcessHeap(), HEAP_ZERO_MEMORY, sizeof(WAVEFORMATEX)); > + else > + pwfx = HeapAlloc(GetProcessHeap(), HEAP_ZERO_MEMORY, paftd->cbFormatSize); > + fd.pwfx = pwfx; > + fd.cbwfx = paftd->cbFormatSize; > + fd.dwFormatTag = paftd->dwFormatTag; > + > + /* try bad pwfx */ > + fd.pwfx = NULL; > + rc = acmFormatDetailsA(had, &fd, ACM_FORMATDETAILSF_FORMAT); > + ok(rc == MMSYSERR_INVALPARAM, > + "acmFormatDetailsA(): rc = %08x, should be %08x\n", > + rc, MMSYSERR_INVALPARAM); > + fd.pwfx = pwfx; > + > + /* try bad wFormatTag */ > + fd.pwfx->wFormatTag = WAVE_FORMAT_UNKNOWN; > + rc = acmFormatDetailsA(had, &fd, ACM_FORMATDETAILSF_FORMAT); > + ok(rc == MMSYSERR_INVALPARAM, > + "acmFormatDetailsA(): rc = %08x, should be %08x\n", > + rc, MMSYSERR_INVALPARAM); > + fd.pwfx->wFormatTag = paftd->dwFormatTag; > + > + /* try bad fdwSupport */ > + fd.fdwSupport = 0xdeadbeef; > + rc = acmFormatDetailsA(had, &fd, ACM_FORMATDETAILSF_FORMAT); > + ok(rc == MMSYSERR_INVALPARAM, > + "acmFormatDetailsA(): rc = %08x, should be %08x\n", > + rc, MMSYSERR_INVALPARAM); > + fd.fdwSupport = 0; > + > + /* try bad pwfx structure size */ > + fd.cbwfx = sizeof(PCMWAVEFORMAT)-1; > + rc = acmFormatDetailsA(had, &fd, ACM_FORMATDETAILSF_FORMAT); > + ok(rc == MMSYSERR_INVALPARAM, > + "acmFormatDetailsA(): rc = %08x, should be %08x\n", > + rc, MMSYSERR_INVALPARAM); > + fd.cbwfx = paftd->cbFormatSize; > + > + /* test bad parameters (all zero) */ > + rc = acmFormatDetailsA(had, &fd, ACM_FORMATDETAILSF_FORMAT); > + ok(rc == ACMERR_NOTPOSSIBLE, > + "acmFormatDetailsA(): rc = %08x, should be %08x\n", > + rc, ACMERR_NOTPOSSIBLE); > + > + /* test index */ > + for (i = 0; i < paftd->cStandardFormats; i++) > + { > + fd.dwFormatIndex = i; > + > + fd.fdwSupport = 0; > + fd.cbwfx = paftd->cbFormatSize; > + fd.pwfx->cbSize = 0xbeef; > + rc = acmFormatDetailsA(had, &fd, ACM_FORMATDETAILSF_INDEX); > + ok(rc == MMSYSERR_NOERROR, > + "acmFormatDetailsA(): rc = %08x, should be %08x\n", > + rc, MMSYSERR_NOERROR); > + > + /* Windows will write cbSize (and other data) even if the > + * given cbwfx is not large enough */ > + fd.fdwSupport = 0; > + fd.cbwfx = sizeof(PCMWAVEFORMAT); > + fd.pwfx->cbSize = 0xbeef; > + rc = acmFormatDetailsA(had, &fd, ACM_FORMATDETAILSF_INDEX); > + todo_wine_if(rc != MMSYSERR_NOERROR) /* remove when fixed */ > + ok(rc == MMSYSERR_NOERROR, > + "acmFormatDetailsA(): rc = %08x, should be %08x\n", > + rc, MMSYSERR_NOERROR); > + if (paftd->dwFormatTag != WAVE_FORMAT_PCM) > + todo_wine_if(fd.pwfx->cbSize != paftd->cbFormatSize - sizeof(WAVEFORMATEX)) /* remove when fixed */ > + ok(fd.pwfx->cbSize == paftd->cbFormatSize - sizeof(WAVEFORMATEX), > + "expected cbSize %d, got %d\n", > + paftd->cbFormatSize - sizeof(WAVEFORMATEX), fd.pwfx->cbSize); > + } > + > + /* one more */ > + fd.dwFormatIndex = paftd->cStandardFormats; > + fd.fdwSupport = 0; > + rc = acmFormatDetailsA(had, &fd, ACM_FORMATDETAILSF_INDEX); > + ok(rc == MMSYSERR_INVALPARAM, > + "acmFormatDetailsA(): rc = %08x, should be %08x\n", > + rc, MMSYSERR_INVALPARAM); > + > + HeapFree(GetProcessHeap(), 0, pwfx); > + } > return TRUE; > } > > @@ -411,6 +512,37 @@ static BOOL CALLBACK DriverEnumProc(HACMDRIVERID hadid, > "acmFormatTagEnumA(): rc = %08x, should be %08x\n", > rc, MMSYSERR_NOERROR); > > + /* try bad pointer */ > + rc = acmFormatDetailsA(had, NULL, ACM_FORMATDETAILSF_INDEX); > + ok(rc == MMSYSERR_INVALPARAM, > + "acmFormatDetailsA(): rc = %08x, should be %08x\n", > + rc, MMSYSERR_INVALPARAM); > + > + /* try bad structure size */ > + ZeroMemory(&fd, sizeof(fd)); > + rc = acmFormatDetailsA(had, &fd, ACM_FORMATDETAILSF_INDEX); > + ok(rc == MMSYSERR_INVALPARAM, > + "acmFormatDetailsA(): rc = %08x, should be %08x\n", > + rc, MMSYSERR_INVALPARAM); > + > + fd.cbStruct = sizeof(fd) - 1; > + rc = acmFormatDetailsA(had, &fd, ACM_FORMATDETAILSF_INDEX); > + ok(rc == MMSYSERR_INVALPARAM, > + "acmFormatDetailsA(): rc = %08x, should be %08x\n", > + rc, MMSYSERR_INVALPARAM); > + > + fd.cbStruct = sizeof(fd); > + fd.pwfx = pwfx; > + ZeroMemory(fd.pwfx, dwSize); > + fd.cbwfx = dwSize; > + fd.dwFormatTag = WAVE_FORMAT_UNKNOWN; > + > + /* try WAVE_FORMAT_UNKNOWN */ > + rc = acmFormatDetailsA(had, &fd, ACM_FORMATDETAILSF_INDEX); > + ok(rc == MMSYSERR_INVALPARAM, > + "acmFormatDetailsA(): rc = %08x, should be %08x\n", > + rc, MMSYSERR_INVALPARAM); > + > HeapFree(GetProcessHeap(), 0, pwfx); > > /* try invalid handle */ > -- > 2.13.0 > > >

1 0

Re: [PATCH] kernel32: Don't limit physical memory size even if not large address aware.
by Sebastian Lackner June 7, 2017

June 7, 2017

On 07.06.2017 19:39, Zebediah Figura wrote: > From testing on Windows 7, it appears Windows doesn't do this either. Also, > the virtual size is already limited in ntdll, so that check is unnecessary > as well. > > This fixes a bug with sndrec32.exe giving OOM errors when recording. > > Signed-off-by: Zebediah Figura <z.figura12(a)gmail.com> > --- > dlls/kernel32/heap.c | 10 ---------- > 1 file changed, 10 deletions(-) Are you sure you tested it correctly (with a binary that does not have IMAGE_FILE_LARGE_ADDRESS_AWARE set)? And what about Adobe Illustrator 8 which depends on this, according to the comment in the source code? Regards, Sebastian

2 1