Re: kernel32: Implement K32GetProcessImageFileNameA().
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 http://testbot.winehq.org/JobDetails.pl?Key=13315 Your paranoid android. === WVISTAADM (32 bit psapi_main) === psapi_main.c:110: Test failed: cbNeeded=100 === W2K8SE (32 bit psapi_main) === psapi_main.c:110: Test failed: cbNeeded=100 === W7PRO (32 bit psapi_main) === psapi_main.c:110: Test failed: cbNeeded=96 === W7PROX64 (32 bit psapi_main) === psapi_main.c:110: Test failed: cbNeeded=108
On Fri, 5 Aug 2011, Marvin wrote: [...]
=== WVISTAADM (32 bit psapi_main) === psapi_main.c:110: Test failed: cbNeeded=100
=== W2K8SE (32 bit psapi_main) === psapi_main.c:110: Test failed: cbNeeded=100
=== W7PRO (32 bit psapi_main) === psapi_main.c:110: Test failed: cbNeeded=96
=== W7PROX64 (32 bit psapi_main) === psapi_main.c:110: Test failed: cbNeeded=108
These are unrelated to the parts I changed. (EnumProcessModules() instead of GetProcessImageFileNameA()) -- Francois Gouget <fgouget(a)codeweavers.com>
On Fri, 5 Aug 2011, Marvin wrote: [...]
Full results can be found at http://testbot.winehq.org/JobDetails.pl?Key=13315 [...] === WVISTAADM (32 bit psapi_main) === psapi_main.c:110: Test failed: cbNeeded=100
This got me looking into this test. It's a test of EnumProcessModules(). It's supposed to only return the 4 modules that psapi_crosstest is linked with: psapi_test.exe, ntdll.dll, kernel32.dll psapi.dll. On Windows there's two extra modules: msvcrt.dll and kernelbase.dll (probably brought in by kernel32.dll). So far so good. But on the TestBot VMs there's a whole raft of extra dlls: psapi_main.c:118: Test failed: cbNeeded=108 psapi_main.c:125: i=0 hmod=00400000 path=[C:\winetest\psapi_crosstest.exe] psapi_main.c:125: i=1 hmod=77DD0000 path=[C:\Windows\SysWOW64\ntdll.dll] psapi_main.c:125: i=2 hmod=77850000 path=[C:\Windows\syswow64\kernel32.dll] psapi_main.c:125: i=3 hmod=76E00000 path=[C:\Windows\syswow64\KERNELBASE.dll] psapi_main.c:125: i=4 hmod=777A0000 path=[C:\Windows\syswow64\msvcrt.dll] psapi_main.c:125: i=5 hmod=75340000 path=[C:\Windows\system32\apphelp.dll] psapi_main.c:125: i=6 hmod=752B0000 path=[C:\Windows\AppPatch\AcLayers.DLL] psapi_main.c:125: i=7 hmod=75940000 path=[C:\Windows\syswow64\SspiCli.dll] psapi_main.c:125: i=8 hmod=77010000 path=[C:\Windows\syswow64\RPCRT4.dll] psapi_main.c:125: i=9 hmod=75930000 path=[C:\Windows\syswow64\CRYPTBASE.dll] psapi_main.c:125: i=10 hmod=76E50000 path=[C:\Windows\SysWOW64\sechost.dll] psapi_main.c:125: i=11 hmod=76A10000 path=[C:\Windows\syswow64\USER32.dll] psapi_main.c:125: i=12 hmod=772F0000 path=[C:\Windows\syswow64\GDI32.dll] psapi_main.c:125: i=13 hmod=76B10000 path=[C:\Windows\syswow64\LPK.dll] psapi_main.c:125: i=14 hmod=76970000 path=[C:\Windows\syswow64\USP10.dll] psapi_main.c:125: i=15 hmod=76610000 path=[C:\Windows\syswow64\ADVAPI32.dll] psapi_main.c:125: i=16 hmod=759A0000 path=[C:\Windows\syswow64\SHELL32.dll] psapi_main.c:125: i=17 hmod=767B0000 path=[C:\Windows\syswow64\SHLWAPI.dll] psapi_main.c:125: i=18 hmod=77640000 path=[C:\Windows\syswow64\ole32.dll] psapi_main.c:125: i=19 hmod=76B70000 path=[C:\Windows\syswow64\OLEAUT32.dll] psapi_main.c:125: i=20 hmod=756E0000 path=[C:\Windows\system32\USERENV.dll] psapi_main.c:125: i=21 hmod=756D0000 path=[C:\Windows\system32\profapi.dll] psapi_main.c:125: i=22 hmod=753F0000 path=[C:\Windows\system32\WINSPOOL.DRV] psapi_main.c:125: i=23 hmod=75290000 path=[C:\Windows\system32\MPR.dll] psapi_main.c:125: i=24 hmod=76FB0000 path=[C:\Windows\system32\IMM32.DLL] psapi_main.c:125: i=25 hmod=766B0000 path=[C:\Windows\syswow64\MSCTF.dll] psapi_main.c:125: i=26 hmod=77DA0000 path=[C:\Windows\syswow64\psapi.dll] I suspect they are brought in by AcLayers.dll which appears to be an 'Application Compatibility Layer' dll that got added in Vista+ (isn't it ironic that it breaks compatibility for our tests?). However, although it's present on disk, it does not get loaded on my Windows 7 VM. Can anyone shed more light on this? And what should we do? Increase the upper bound check (e.g. to 30 modules) or remove it altogether? I'm attaching a patch for your testing pleasure. -- Francois Gouget <fgouget(a)free.fr> http://fgouget.free.fr/ Demander si un ordinateur peut penser revient à demander si un sous-marin peut nager.
participants (3)
-
Francois Gouget -
Francois Gouget -
Marvin