February 2017 - Wine-Bugs - list.winehq.org

[Bug 42546] New: DSOUND_PrimaryOpen() incorrect set buf size
by wine-bugs＠winehq.org Dec. 21, 2018

Dec. 21, 2018

https://bugs.winehq.org/show_bug.cgi?id=42546 Bug ID: 42546 Summary: DSOUND_PrimaryOpen() incorrect set buf size Product: Wine Version: 2.2 Hardware: x86 OS: FreeBSD Status: UNCONFIRMED Severity: major Priority: P2 Component: directx-dsound Assignee: wine-bugs(a)winehq.org Reporter: rozhuk.im(a)gmail.com Wine (1.x, 2.x) crashes on FreeBSD 11 release with wineoss.drv With winealsa.drv on FreeBSD 11 - OK. With winoss.drv on FreeBSD 10 - OK. I made some investigation and make patch to fix problem. I test it and it work. But looks like dsound needs review. DSOUND_ReopenDevice() call DSOUND_PrimaryOpen() with forcewave = FALSE DSOUND_PrimaryOpen() calculate: new_buflen = device->buflen; new_buflen -= new_buflen % wfx->nBlockAlign; (wrong alig code, but newer mind) Then calculated DWORD alloc_len = frames * sizeof(float); and allocated buffer if (device->buffer) newbuf = HeapReAlloc(GetProcessHeap(), 0, device->buffer, alloc_len); else newbuf = HeapAlloc(GetProcessHeap(), 0, alloc_len); final: save new buffer pointer and size: device->buffer = newbuf; device->buflen = new_buflen; !!! We allocate 6144 (0x00001800) bytes but set buf size to 65536. DSOUND_MixToPrimary() and norm16() in DSOUND_mixthread use device->buflen and corrupt heap after first call. First I make more proper alignment: new_buflen = (device->buflen + wfx->nBlockAlign); new_buflen -= (new_buflen % wfx->nBlockAlign); and change calculation to: DWORD alloc_len = ((new_buflen / wfx->nBlockAlign) * sizeof(float)); It works. Next I replace alloc_len->new_buflen and return original calc code: new_buflen = (frames * sizeof(float)); This work but sound with a bit noise and game crash after some time. Finnaly: new_buflen = ((frames + 1) * wfx->nBlockAlign * sizeof(float)); This work OK. This not looks like proper buf size. Also in DSOUND_ReopenDevice() after DSOUND_PrimaryOpen(): device->fraglen = frag_frames * wfx->nBlockAlign; device->aclen = aclen_frames * wfx->nBlockAlign; device->buflen should somehow be synced with this. 606723.622:0034:0035:trace:heap:RtlAllocateHeap (0x110000,70000062,00000030): returning 0x11d438 606723.622:0034:0035:trace:oss:AudioClient_Start (0x1ecb38) now playing... 606723.622:0034:0035:trace:oss:AudioClient_GetStreamLatency (0x1ecb38)->(0x335f590) 606723.622:0034:0035:trace:oss:AudioClient_GetBufferSize (0x1ecb38)->(0x335f58c) 606723.622:0034:0035:trace:oss:AudioClient_GetBufferSize buffer size: 3840 606723.622:0034:0035:trace:dsound:DSOUND_ReopenDevice period 11 ms fraglen 2048 buflen 6144 606723.622:0034:0035:trace:dsound:DSOUND_PrimaryOpen (0x11e4d8) 606723.622:0034:0035:trace:heap:RtlAllocateHeap (0x110000,70000062,00001800): returning 0x1fc1e8 606723.622:0034:0035:trace:dsound:DSOUND_PrimaryOpen buflen: 65536, fraglen: 0 606723.622:0034:0035:trace:oss:AudioClient_IsFormatSupported (0x1ecb38)->(0, 0x335f650, 0x335f64c) ... 606723.688:0034:003e:trace:dsound:DSOUND_mixthread (0x11e4d8) 606723.688:0034:003e:trace:dsound:DSOUND_PerformMix (0x11e4d8) 606723.688:0034:003e:trace:oss:AudioClient_GetCurrentPadding (0x1ecb38)->(0x1448f6b8) 606723.688:0034:003e:trace:oss:AudioClient_GetCurrentPadding pad: 0 606723.688:0034:003e:warn:dsound:DSOUND_PerformMix Probable buffer underrun 606723.688:0034:003e:trace:oss:AudioRenderClient_GetBuffer (0x1ecb38)->(1536, 0x1448f6b0) 606723.690:0034:003e:trace:heap:RtlAllocateHeap (0x110000,70000062,00001810): returning 0x141908b0 606723.690:0034:003e:trace:dsound:DSOUND_MixToPrimary (0,6144) 606723.690:0034:003e:trace:dsound:norm16 0x1fc1e8 - 0x141908b0 6144 606723.690:0034:003e:trace:oss:AudioRenderClient_ReleaseBuffer (0x1ecb38)->(1536, 0) 606723.690:0034:003e:trace:oss:AudioRenderClient_ReleaseBuffer writen: 6144 606723.690:0034:003d:err:heap:HEAP_ValidateInUseArena Heap 0x110000: block 0x1fc1e8 tail overwritten at 0x1fd9e8 (byte 0/8 == 0x00) Heap: 0x110000 ... Sub-heap 0x110014: base=0x110000 size=00110000 committed=00110000 Block Arena Stat Size Id ... 0x1f0118 00bedead pend 00000430 0x1f0550 00bedead pend 00002c18 0x1f3170 00bedead pend 00000430 0x1f35a8 00bedead pend 00008c30 0x1fc1e0 00455355 used 00001808 0x1fd9f0 00000000 pend 00000000 0x1fd9f8 00000000 pend 00000000 0x1fda00 00000000 pend 00000000 0x1fda08 00000000 pend 00000000 -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

1 22

[Bug 36354] New: valgrind shows a leak in ddraw/tests/dsurface.c
by wine-bugs＠winehq.org Dec. 21, 2018

Dec. 21, 2018

https://bugs.winehq.org/show_bug.cgi?id=36354 Bug ID: 36354 Summary: valgrind shows a leak in ddraw/tests/dsurface.c Product: Wine Version: 1.7.18 Hardware: x86 OS: Linux Status: NEW Keywords: download, source, testcase Severity: normal Priority: P2 Component: directx-d3d Assignee: wine-bugs(a)winehq.org Reporter: austinenglish(a)gmail.com ==8690== 24 bytes in 1 blocks are definitely lost in loss record 250 of 943 ==8690== at 0x7BC50A9B: RtlAllocateHeap (heap.c:255) ==8690== by 0x4A868D9: ddraw7_CreatePalette (ddraw.c:3385) ==8690== by 0x4A86D09: ddraw1_CreatePalette (ddraw.c:3458) ==8690== by 0x4DB8A62: PaletteTest (dsurface.c:2247) ==8690== by 0x4DC8FA4: func_dsurface (dsurface.c:3962) ==8690== by 0x4DDD6DC: run_test (test.h:584) ==8690== by 0x4DDDACB: main (test.h:654) ==8690== -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

1 7

[Bug 28766] New: DIB engine multiple invalid memory accesses
by wine-bugs＠winehq.org Dec. 21, 2018

Dec. 21, 2018

http://bugs.winehq.org/show_bug.cgi?id=28766 Bug #: 28766 Summary: DIB engine multiple invalid memory accesses Product: Wine Version: 1.3.30 Platform: x86 OS/Version: Linux Status: NEW Severity: normal Priority: P2 Component: gdi32 AssignedTo: wine-bugs(a)winehq.org ReportedBy: dank(a)kegel.com Classification: Unclassified In gdi32/tests, running "make bitmap.ok dib.ok font.ok pen.ok" makes Valgrind go bonkers. Huw asked for a bug to track these, and said one big bug would suffice. I'll attach the initial log, and will do more valgrind runs on request. -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 11

[Bug 26042] New: advapi32/crypt tests show two leaks under valgrind
by wine-bugs＠winehq.org Dec. 21, 2018

Dec. 21, 2018

http://bugs.winehq.org/show_bug.cgi?id=26042 Summary: advapi32/crypt tests show two leaks under valgrind Product: Wine Version: 1.3.13 Platform: x86 OS/Version: Linux Status: NEW Keywords: download, source, testcase Severity: minor Priority: P2 Component: advapi32 AssignedTo: wine-bugs(a)winehq.org ReportedBy: austinenglish(a)gmail.com 12 bytes in 1 blocks are definitely lost at notify_alloc (heap.c:254) by RtlAllocateHeap (heap.c:1701) by HeapAlloc (heap.c:267) by GlobalAlloc (heap.c:360) by LocalAlloc (heap.c:960) by CryptGenKey (crypt.c:1433) by test_incorrect_api_usage (crypt.c:241) by func_crypt (crypt.c:1130) by run_test (test.h:556) by main (test.h:624) ... 996 bytes in 1 blocks are definitely lost at notify_alloc (heap.c:254) by RtlAllocateHeap (heap.c:1701) by ??? by ??? by ??? by CryptGenKey (crypt.c:1441) by test_incorrect_api_usage (crypt.c:241) by func_crypt (crypt.c:1130) by run_test (test.h:556) by main (test.h:624) -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 8

[Bug 36655] New: valgrind shows an unitialized variable in mountmgr.sys/device.c (d3d9/tests/d3d9ex.c)
by wine-bugs＠winehq.org Dec. 21, 2018

Dec. 21, 2018

https://bugs.winehq.org/show_bug.cgi?id=36655 Bug ID: 36655 Summary: valgrind shows an unitialized variable in mountmgr.sys/device.c (d3d9/tests/d3d9ex.c) Product: Wine Version: 1.7.19 Hardware: x86 OS: Linux Status: NEW Keywords: download, source, testcase, valgrind Severity: normal Priority: P2 Component: mountmgr.sys Assignee: wine-bugs(a)winehq.org Reporter: austinenglish(a)gmail.com ==6715== Thread 4: ==6715== Syscall param writev(vector[...]) points to uninitialised byte(s) ==6715== at 0x431A82C: writev (writev.c:54) ==6715== by 0x7BC7FE0B: send_request (server.c:213) ==6715== by 0x7BC7FF9D: wine_server_call (server.c:294) ==6715== by 0x5B66A38: put_message_in_queue (message.c:3141) ==6715== by 0x5B66DE9: send_inter_thread_message (message.c:3209) ==6715== by 0x5B670CA: send_message (message.c:3277) ==6715== by 0x5B67511: SendMessageTimeoutW (message.c:3414) ==6715== by 0x5B692FD: bcast_childwindow (message.c:4265) ==6715== by 0x5B9B07A: EnumDesktopWindows (win.c:3336) ==6715== by 0x5B6967E: bcast_desktop (message.c:4315) ==6715== by 0x5BA9587: EnumDesktopsW (winstation.c:454) ==6715== by 0x5B697DB: bcast_winsta (message.c:4329) ==6715== by 0x5BA8CFB: EnumWindowStationsW (winstation.c:238) ==6715== by 0x5B69B06: BroadcastSystemMessageExW (message.c:4398) ==6715== by 0x5B698F9: BroadcastSystemMessageW (message.c:4350) ==6715== by 0x5710683: send_notify (device.c:186) ==6715== by 0x571214A: add_dos_device (device.c:812) ==6715== by 0x570F70C: udisks2_add_device (dbus.c:479) ==6715== by 0x570F90D: udisks2_add_devices (dbus.c:517) ==6715== by 0x570FEC6: dbus_thread (dbus.c:717) ==6715== Address 0x599e5ce is on thread 1's stack ==6715== Uninitialised value was created by a stack allocation ==6715== at 0x571062B: send_notify (device.c:178) ==6715== -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

1 3

[Bug 36162] New: valgrind shows several leaks in programs/cmd/tests/batch.c
by wine-bugs＠winehq.org Dec. 21, 2018

Dec. 21, 2018

https://bugs.winehq.org/show_bug.cgi?id=36162 Bug ID: 36162 Summary: valgrind shows several leaks in programs/cmd/tests/batch.c Product: Wine Version: 1.7.17 Hardware: x86 OS: Linux Status: NEW Keywords: download, source, testcase Severity: normal Priority: P2 Component: cmd Assignee: wine-bugs(a)winehq.org Reporter: austinenglish(a)gmail.com ==19001== 24 bytes in 2 blocks are definitely lost in loss record 77 of 302 ==19001== at 0x7BC4C735: notify_alloc (heap.c:255) ==19001== by 0x7BC50F79: RtlAllocateHeap (heap.c:1716) ==19001== by 0x4B2544E: heap_alloc (wcmdmain.c:444) ==19001== by 0x4B1E203: WCMD_pushoperator (builtins.c:3615) ==19001== by 0x4B1F18F: WCMD_handleExpression (builtins.c:3932) ==19001== by 0x4B1F84D: WCMD_setshow_env (builtins.c:4098) ==19001== by 0x4B282D6: WCMD_execute (wcmdmain.c:1553) ==19001== by 0x4B29F36: WCMD_process_commands (wcmdmain.c:2294) ==19001== by 0x4B130E0: WCMD_batch (batch.c:95) ==19001== by 0x4B27185: WCMD_run_program (wcmdmain.c:1187) ==19001== by 0x4B28467: WCMD_execute (wcmdmain.c:1618) ==19001== by 0x4B29F36: WCMD_process_commands (wcmdmain.c:2294) ==19001== by 0x4B2ADC4: wmain (wcmdmain.c:2590) ==19001== by 0x4B2B328: __wine_spec_exe_wentry (exe_wentry.c:36) ==19001== by 0x7B864247: ??? (process.c:1045) ==19001== by 0x7B864394: start_process (process.c:1097) ==19001== by 0x7BC87087: ??? (signal_i386.c:2571) ==19001== by 0x7BC870D0: call_thread_func (signal_i386.c:2630) ==19001== by 0x7BC87065: ??? (signal_i386.c:2571) ==19001== by 0x7BC5A1C0: start_process (loader.c:2840) ==19001== -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

1 4

[Bug 26070] New: user32/class tests show some valgrind warnings
by wine-bugs＠winehq.org Dec. 21, 2018

Dec. 21, 2018

http://bugs.winehq.org/show_bug.cgi?id=26070 Summary: user32/class tests show some valgrind warnings Product: Wine Version: 1.3.13 Platform: x86 OS/Version: Linux Status: NEW Keywords: download, source, testcase Severity: minor Priority: P2 Component: user32 AssignedTo: wine-bugs(a)winehq.org ReportedBy: austinenglish(a)gmail.com Invalid read of size 2 at RtlImageNtHeader (loader.c:2508) by RtlImageDirectoryEntryToData (loader.c:2684) by find_entry (resource.c:189) by LdrFindResource_U (resource.c:303) by find_resourceW (resource.c:139) by FindResourceExW (resource.c:183) by FindResourceW (resource.c:192) by LoadMenuW (menu.c:4359) by WIN_CreateWindowEx (win.c:1372) by CreateWindowExA (win.c:1550) by check_instance_ (class.c:266) by test_instances (class.c:441) by func_class (class.c:995) by run_test (test.h:556) by main (test.h:624) Address 0x12345678 is not stack'd, malloc'd or (recently) free'd ... 28 bytes in 1 blocks are definitely lost at notify_alloc (heap.c:254) by RtlAllocateHeap (heap.c:1701) by DIALOG_get_info (defdlg.c:335) by DIALOG_CreateIndirect (dialog.c:669) by CreateDialogIndirectParamAorW (dialog.c:755) by CreateDialogIndirectParamA (dialog.c:764) by CreateDialogParamA (dialog.c:729) by CreateDialogParamTest (class.c:878) by func_class (class.c:990) by run_test (test.h:556) by main (test.h:624) ... 42,420 bytes in 21 blocks are possibly lost at notify_alloc (heap.c:254) by RtlAllocateHeap (heap.c:1701) by get_tls_data (test.h:240) by winetest_set_location (test.h:275) by check_instance_ (class.c:263) by thread_proc (class.c:292) by ??? (signal_i386.c:2473) by call_thread_entry_point (signal_i386.c:2499) by start_thread (thread.c:404) by start_thread (in /lib/libpthread-2.11.2.so) by clone (in /lib/libc-2.11.2.so) -- Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email Do not reply to this email, post in Bugzilla using the above URL to reply. ------- You are receiving this mail because: ------- You are watching all bug changes.

1 6

[Bug 36328] New: valgrind shows a leak in rsaenh/tests/rsaenh.c
by wine-bugs＠winehq.org Dec. 21, 2018

Dec. 21, 2018

https://bugs.winehq.org/show_bug.cgi?id=36328 Bug ID: 36328 Summary: valgrind shows a leak in rsaenh/tests/rsaenh.c Product: Wine Version: 1.7.18 Hardware: x86 OS: Linux Status: NEW Keywords: download, source, testcase Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: austinenglish(a)gmail.com ==16347== 88,188 (416 direct, 87,772 indirect) bytes in 7 blocks are definitely lost in loss record 2,195 of 2,200 ==16347== at 0x4006B11: malloc (in /usr/lib/valgrind/vgpreload_memcheck-x86-linux.so) ==16347== by 0x4ED03DAB: ??? (in /usr/lib/libfreetype.so.6.10.2) ==16347== by 0x4ED08C5A: ft_mem_qalloc (in /usr/lib/libfreetype.so.6.10.2) ==16347== by 0x4ED08CCE: ft_mem_alloc (in /usr/lib/libfreetype.so.6.10.2) ==16347== by 0x4ED0DF8E: FT_Add_Module (in /usr/lib/libfreetype.so.6.10.2) ==16347== by 0x4ED040FE: FT_Add_Default_Modules (in /usr/lib/libfreetype.so.6.10.2) ==16347== by 0x4ED04169: FT_Init_FreeType (in /usr/lib/libfreetype.so.6.10.2) ==16347== by 0x6296169: ??? ==16347== by 0x6298F85: ??? ==16347== by 0x62B4A52: ??? ==16347== by 0x7BC53B24: ??? (in /home/austin/wine-valgrind-known/dlls/ntdll/ntdll.dll.so) ==16347== by 0x7BC5644E: MODULE_InitDLL (loader.c:1068) ==16347== by 0x7BC56AE7: process_attach (loader.c:1157) ==16347== by 0x7BC56AB9: process_attach (loader.c:1149) ==16347== by 0x7BC56AB9: process_attach (loader.c:1149) ==16347== by 0x7BC56AB9: process_attach (loader.c:1149) ==16347== by 0x7BC596E6: LdrLoadDll (loader.c:2129) ==16347== by 0x7B8581FE: load_library (module.c:933) ==16347== by 0x7B8582C9: LoadLibraryExW (module.c:990) ==16347== by 0x7B8584AC: LoadLibraryW (module.c:1032) ==16347== -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

1 3

[Bug 36356] New: valgrind shows a possible leak in dinput/tests/device.c
by wine-bugs＠winehq.org Dec. 21, 2018

Dec. 21, 2018

https://bugs.winehq.org/show_bug.cgi?id=36356 Bug ID: 36356 Summary: valgrind shows a possible leak in dinput/tests/device.c Product: Wine Version: 1.7.18 Hardware: x86 OS: Linux Status: NEW Keywords: download, source, testcase Severity: normal Priority: P2 Component: directx-dinput Assignee: wine-bugs(a)winehq.org Reporter: austinenglish(a)gmail.com ==9098== 28 bytes in 1 blocks are possibly lost in loss record 230 of 825 ==9098== at 0x7BC50A9B: RtlAllocateHeap (heap.c:255) ==9098== by 0x5981BB8: apartment_getclassobject (compobj.c:526) ==9098== by 0x5984FC6: get_inproc_class_object (compobj.c:2894) ==9098== by 0x5987D2F: CoGetClassObject (compobj.c:3032) ==9098== by 0x59884B7: CoCreateInstance (compobj.c:3197) ==9098== by 0x4A56A80: func_device (device.c:216) ==9098== by 0x4A5E18C: run_test (test.h:584) ==9098== by 0x4A5E57B: main (test.h:654) ==9098== -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

1 4

[Bug 36290] New: valgrind shows a leak in mscms/tests/profile.c
by wine-bugs＠winehq.org Dec. 21, 2018

Dec. 21, 2018

https://bugs.winehq.org/show_bug.cgi?id=36290 Bug ID: 36290 Summary: valgrind shows a leak in mscms/tests/profile.c Product: Wine Version: 1.7.18 Hardware: x86 OS: Linux Status: NEW Keywords: download, source, testcase Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: austinenglish(a)gmail.com ==17956== 19,636 (192 direct, 19,444 indirect) bytes in 1 blocks are definitely lost in loss record 225 of 225 ==17956== at 0x4006B11: malloc (in /usr/lib/valgrind/vgpreload_memcheck-x86-linux.so) ==17956== by 0x4ED03DAB: ??? (in /usr/lib/libfreetype.so.6.10.2) ==17956== by 0x4ED08C5A: ft_mem_qalloc (in /usr/lib/libfreetype.so.6.10.2) ==17956== by 0x4ED08CCE: ft_mem_alloc (in /usr/lib/libfreetype.so.6.10.2) ==17956== by 0x4ED09CCB: FT_New_Library (in /usr/lib/libfreetype.so.6.10.2) ==17956== by 0x4ED04142: FT_Init_FreeType (in /usr/lib/libfreetype.so.6.10.2) ==17956== by 0x550C7F0: ??? ==17956== by 0x550D39B: ??? ==17956== by 0x551B635: ??? ==17956== by 0x553A646: ??? ==17956== by 0x7BC52FC0: ??? (loader.c:138) ==17956== by 0x7BC555FE: MODULE_InitDLL (loader.c:1068) ==17956== by 0x7BC5597C: process_attach (loader.c:1157) ==17956== by 0x7BC5592C: process_attach (loader.c:1149) ==17956== by 0x7BC5592C: process_attach (loader.c:1149) ==17956== by 0x7BC5592C: process_attach (loader.c:1149) ==17956== by 0x7BC585E2: LdrLoadDll (loader.c:2129) ==17956== by 0x7B85A79A: load_library (module.c:933) ==17956== by 0x7B85A908: LoadLibraryExW (module.c:990) ==17956== by 0x7B85A869: LoadLibraryExA (module.c:970) ==17956== -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

1 5