Ros-dev August 2010

ros-dev@reactos.org

18 participants
18 discussions

by Olaf Siejka

Allow me to present a small sub-project Gabriel and I are working on. We did a small scale consulting, but its the time to present it on a more wider scale. As a comment, specifically to Timo's post: 2nd class tags should not be conflicting with Component field when possible. In case of Kernel and win32k, a submodule should be used instead of module name: for example in case of kernel bugs, Components should be set to Kernel, whereas 2nd class tag should be pointing to kernel submodule, like iomgr:, ob:, mm: etc. All opinions are most welcome Best regards

14 years, 3 months

Bugzilla interface

by Timo Kreuzer

Hi I'd like to propose an overhaul of our bugzilla interface. The reason is that the current interface is ugly and confusing and I think we can make filing and handling bugs less unenjoyable ;-) First I think it would be very useful, if we could edit the description field of the bugs. This way we don't need to browse through dozens of comments to find all neccessary info, while the description only contains useless stuff. This should be restricted to the original reporter and testers / developers. Then when filing a new bug or looking at a bug that is already filed, the layout is horrible. I'm not a webdesigner, so it's hard for me to say how it should look like, but definately not the way it is. This might be appropriate for projects whose website look like http://www.gnu.org/software/binutils/ but I'm sure, we can do better. Then when filing a bug there are the following fields: - reporter - I know who I am, so why show it to me? Unneccessary - Component: I don't think this field is really useful as it is. First: Patches are definately not a component. Then it's often hard to tell where the bug is. for example, if rapps doesn't download anything, is that related to network or is it a kernel bug or win32 or rapps or is only the server down? You often don't know it when filing a bug. Also win32 covers a lot from win32k to shell32, but are these more closely related than drivers and networking? - Severity: while I think this field is useful and important, it should only be editable by testers and developers and should not appear when a bug is filed. - Platform: should be removed - OS: should be removed - AssignTo, Cc: rarely used on first filing, should IMO only be editable by testers / developers - URL: While we use this field from time to time, I think the URL could as well, if neccessary be put into the description field (provided, it was editable). This versatile field should imo change it's purpose from URL to TAG. So we can add different tags, like REGRESSION, PATCH, HACK, that are currently put into the summery field. It could also contain the regression range or guilty version - Alias: we don't use this, or rather currently only misuse this for the guilty version, which is problematic, as soon as 2 bugs have the same guilty version, IMO unneccessary - Summery: should be as wide as the description field - Description: To get better bugreports, I suggest dividing this field into "Steps to reproduce:" "Experienced behaviour:" "Remarks" These fields can very well be automatically merged into one field, It's just to show people that they must provide steps to reproduce, instead of writing dozens of lines about their hardware configuration and how they feel about the bug. Regards, Timo

14 years, 3 months

Regarding 48632

by Ros Arm

Hello, Many thanks for fixing this idiotic oversight of mine. I apologies for the long hours of debugging I must have caused. This fix is somewhat incomplete/at the wrong place; could this be marked as such in the code, perhaps with a FIXME under my name, so that I may fix this correctly upon my return to the US? Much obliged! -r Author: mjmartin Date: Sat Aug 28 00:26:02 2010 New Revision: 48632 URL: http://svn.reactos.org/svn/reactos?rev=48632&view=rev Log: [ntoskrnl/ps] - When deleting a Process remove the Process from the MmProcessList. Fixes random NonPaged Pool corruptions. Thanks aicom for assistance. Modified: trunk/reactos/ntoskrnl/ps/kill.c Modified: trunk/reactos/ntoskrnl/ps/kill.c URL: http://svn.reactos.org/svn/reactos/trunk/reactos/ntoskrnl/ps/kill.c?rev=486… ============================================================================== --- trunk/reactos/ntoskrnl/ps/kill.c [iso-8859-1] (original) +++ trunk/reactos/ntoskrnl/ps/kill.c [iso-8859-1] Sat Aug 28 00:26:02 2010 @@ -300,6 +300,8 @@ /* Detach */ KeUnstackDetachProcess(&ApcState); + + RemoveEntryList(&Process->MmProcessLinks); /* Completely delete the Address Space */ MmDeleteProcessAddressSpace(Process);

14 years, 3 months

Re: [ros-dev] [ros-diffs] [cgutman] 48593: [DHCPCSVC] - Write the DNS servers in a REG_MULTI_SZ value [IPHLPAPI] - Rewrite the registry reading code - Use HeapFree to free memory from the allocated from heap

by Timo Kreuzer

It looks like this change has caused some regressions in sysreg http://www.reactos.org/testman/compare.php?ids=3934,3935 cgutman(a)svn.reactos.org wrote: > Author: cgutman > Date: Sun Aug 22 22:22:27 2010 > New Revision: 48593 > > URL: http://svn.reactos.org/svn/reactos?rev=48593&view=rev > Log: > [DHCPCSVC] > - Write the DNS servers in a REG_MULTI_SZ value > [IPHLPAPI] > - Rewrite the registry reading code > - Use HeapFree to free memory from the allocated from heap > > Modified: > trunk/reactos/dll/win32/dhcpcsvc/dhcp/dhclient.c > trunk/reactos/dll/win32/iphlpapi/iphlpapi_private.h > trunk/reactos/dll/win32/iphlpapi/registry.c > trunk/reactos/dll/win32/iphlpapi/resinfo_reactos.c > > Modified: trunk/reactos/dll/win32/dhcpcsvc/dhcp/dhclient.c > URL: http://svn.reactos.org/svn/reactos/trunk/reactos/dll/win32/dhcpcsvc/dhcp/dh… > ============================================================================== > --- trunk/reactos/dll/win32/dhcpcsvc/dhcp/dhclient.c [iso-8859-1] (original) > +++ trunk/reactos/dll/win32/dhcpcsvc/dhcp/dhclient.c [iso-8859-1] Sun Aug 22 22:22:27 2010 > @@ -507,24 +507,25 @@ > char *nsbuf; > int i, addrs = > new_lease->options[DHO_DOMAIN_NAME_SERVERS].len / sizeof(ULONG); > - > - nsbuf = malloc( addrs * sizeof(IP_ADDRESS_STRING) ); > + int len = 0; > + > + nsbuf = malloc( addrs * sizeof(IP_ADDRESS_STRING) + 1 ); > > if( nsbuf) { > - nsbuf[0] = 0; > + memset(nsbuf, 0, addrs * sizeof(IP_ADDRESS_STRING) + 1); > for( i = 0; i < addrs; i++ ) { > nameserver.len = sizeof(ULONG); > memcpy( nameserver.iabuf, > new_lease->options[DHO_DOMAIN_NAME_SERVERS].data + > (i * sizeof(ULONG)), sizeof(ULONG) ); > strcat( nsbuf, piaddr(nameserver) ); > - if( i != addrs-1 ) strcat( nsbuf, "," ); > + len += strlen(nsbuf) + 1; > } > > DH_DbgPrint(MID_TRACE,("Setting DhcpNameserver: %s\n", nsbuf)); > > - RegSetValueExA( RegKey, "DhcpNameServer", 0, REG_SZ, > - (LPBYTE)nsbuf, strlen(nsbuf) + 1 ); > + RegSetValueExA( RegKey, "DhcpNameServer", 0, REG_MULTI_SZ, > + (LPBYTE)nsbuf, len + 1 ); > free( nsbuf ); > } > > > Modified: trunk/reactos/dll/win32/iphlpapi/iphlpapi_private.h > URL: http://svn.reactos.org/svn/reactos/trunk/reactos/dll/win32/iphlpapi/iphlpap… > ============================================================================== > --- trunk/reactos/dll/win32/iphlpapi/iphlpapi_private.h [iso-8859-1] (original) > +++ trunk/reactos/dll/win32/iphlpapi/iphlpapi_private.h [iso-8859-1] Sun Aug 22 22:22:27 2010 > @@ -139,6 +139,7 @@ > PWCHAR GetNthChildKeyName( HANDLE RegHandle, DWORD n ); > void ConsumeChildKeyName( PWCHAR Name ); > PWCHAR QueryRegistryValueString( HANDLE RegHandle, PWCHAR ValueName ); > +PWCHAR *QueryRegistryValueStringMulti( HANDLE RegHandle, PWCHAR ValueName ); > void ConsumeRegValueString( PWCHAR NameServer ); > BOOL isInterface( TDIEntityID *if_maybe ); > BOOL hasArp( HANDLE tcpFile, TDIEntityID *arp_maybe ); > > Modified: trunk/reactos/dll/win32/iphlpapi/registry.c > URL: http://svn.reactos.org/svn/reactos/trunk/reactos/dll/win32/iphlpapi/registr… > ============================================================================== > --- trunk/reactos/dll/win32/iphlpapi/registry.c [iso-8859-1] (original) > +++ trunk/reactos/dll/win32/iphlpapi/registry.c [iso-8859-1] Sun Aug 22 22:22:27 2010 > @@ -66,25 +66,123 @@ > if (Name) HeapFree( GetProcessHeap(), 0, Name ); > } > > -PWCHAR QueryRegistryValueString( HANDLE RegHandle, PWCHAR ValueName ) { > - PWCHAR Name; > - DWORD ReturnedSize = 0; > +PVOID QueryRegistryValue(HANDLE RegHandle, PWCHAR ValueName, LPDWORD RegistryType, LPDWORD Length) > +{ > + PVOID ReadValue = NULL; > + DWORD Error; > > - if (RegQueryValueExW( RegHandle, ValueName, NULL, NULL, NULL, > - &ReturnedSize ) != 0) { > - return 0; > - } else { > - Name = malloc( ReturnedSize); > - RegQueryValueExW( RegHandle, ValueName, NULL, NULL, (PVOID)Name, > - &ReturnedSize ); > - return Name; > - } > + *Length = 0; > + *RegistryType = REG_NONE; > + > + while (TRUE) > + { > + Error = RegQueryValueExW(RegHandle, ValueName, NULL, RegistryType, ReadValue, Length); > + if (Error == ERROR_SUCCESS) > + { > + if (ReadValue) break; > + } > + else if (Error == ERROR_MORE_DATA) > + { > + HeapFree(GetProcessHeap(), 0, ReadValue); > + } > + else break; > + > + ReadValue = HeapAlloc(GetProcessHeap(), 0, *Length); > + if (!ReadValue) return NULL; > + } > + > + if (Error != ERROR_SUCCESS) > + { > + if (ReadValue) HeapFree(GetProcessHeap(), 0, ReadValue); > + > + *Length = 0; > + *RegistryType = REG_NONE; > + ReadValue = NULL; > + } > + > + return ReadValue; > +} > + > +PWCHAR TerminateReadString(PWCHAR String, DWORD Length) > +{ > + PWCHAR TerminatedString; > + > + TerminatedString = HeapAlloc(GetProcessHeap(), 0, Length + sizeof(WCHAR)); > + if (TerminatedString == NULL) > + return NULL; > + > + memcpy(TerminatedString, String, Length); > + > + TerminatedString[Length / sizeof(WCHAR)] = UNICODE_NULL; > + > + return TerminatedString; > +} > + > +PWCHAR QueryRegistryValueString( HANDLE RegHandle, PWCHAR ValueName ) > +{ > + PWCHAR String, TerminatedString; > + DWORD Type, Length; > + > + String = QueryRegistryValue(RegHandle, ValueName, &Type, &Length); > + if (!String) return NULL; > + if (Type != REG_SZ) > + { > + DbgPrint("Type mismatch for %S (%d != %d)\n", ValueName, Type, REG_SZ); > + //HeapFree(GetProcessHeap(), 0, String); > + //return NULL; > + } > + > + TerminatedString = TerminateReadString(String, Length); > + HeapFree(GetProcessHeap(), 0, String); > + if (!TerminatedString) return NULL; > + > + return TerminatedString; > } > > void ConsumeRegValueString( PWCHAR Value ) { > - if (Value) free(Value); > + if (Value) HeapFree(GetProcessHeap(), 0, Value); > } > > PWCHAR *QueryRegistryValueStringMulti( HANDLE RegHandle, PWCHAR ValueName ) { > - return 0; /* FIXME if needed */ > + PWCHAR String, TerminatedString, Tmp; > + PWCHAR *Table; > + DWORD Type, Length, i, j; > + > + String = QueryRegistryValue(RegHandle, ValueName, &Type, &Length); > + if (!String) return NULL; > + if (Type != REG_MULTI_SZ) > + { > + DbgPrint("Type mismatch for %S (%d != %d)\n", ValueName, Type, REG_MULTI_SZ); > + //HeapFree(GetProcessHeap(), 0, String); > + //return NULL; > + } > + > + TerminatedString = TerminateReadString(String, Length); > + HeapFree(GetProcessHeap(), 0, String); > + if (!TerminatedString) return NULL; > + > + for (Tmp = TerminatedString, i = 0; *Tmp; Tmp++, i++) while (*Tmp) Tmp++; > + > + Table = HeapAlloc(GetProcessHeap(), 0, (i + 1) * sizeof(PWCHAR)); > + if (!Table) > + { > + HeapFree(GetProcessHeap(), 0, TerminatedString); > + return NULL; > + } > + > + for (Tmp = TerminatedString, j = 0; *Tmp; Tmp++, j++) > + { > + PWCHAR Orig = Tmp; > + > + for (i = 0; *Tmp; i++, Tmp++); > + > + Table[j] = HeapAlloc(GetProcessHeap(), 0, i * sizeof(WCHAR)); > + memcpy(Table[j], Orig, i * sizeof(WCHAR)); > + } > + > + Table[j] = NULL; > + > + HeapFree(GetProcessHeap(), 0, TerminatedString); > + > + return Table; > } > > Modified: trunk/reactos/dll/win32/iphlpapi/resinfo_reactos.c > URL: http://svn.reactos.org/svn/reactos/trunk/reactos/dll/win32/iphlpapi/resinfo… > ============================================================================== > --- trunk/reactos/dll/win32/iphlpapi/resinfo_reactos.c [iso-8859-1] (original) > +++ trunk/reactos/dll/win32/iphlpapi/resinfo_reactos.c [iso-8859-1] Sun Aug 22 22:22:27 2010 > @@ -122,47 +122,24 @@ > > void EnumNameServers( HANDLE RegHandle, PWCHAR Interface, > PVOID Data, EnumNameServersFunc cb ) { > - PWCHAR NameServerString = > - QueryRegistryValueString(RegHandle, L"DhcpNameServer"); > + PWCHAR *NameServerString = > + QueryRegistryValueStringMulti(RegHandle, L"DhcpNameServer"); > + DWORD i; > > if (!NameServerString) > - NameServerString = QueryRegistryValueString(RegHandle, L"NameServer"); > - > - if (NameServerString) { > - /* Now, count the non-empty comma separated */ > - DWORD ch; > - DWORD LastNameStart = 0; > - for (ch = 0; NameServerString[ch]; ch++) { > - if (NameServerString[ch] == ',') { > - if (ch - LastNameStart > 0) { /* Skip empty entries */ > - PWCHAR NameServer = > - malloc(((ch - LastNameStart) + 1) * sizeof(WCHAR)); > - if (NameServer) { > - memcpy(NameServer,NameServerString + LastNameStart, > - (ch - LastNameStart) * sizeof(WCHAR)); > - NameServer[ch - LastNameStart] = 0; > - cb( Interface, NameServer, Data ); > - free(NameServer); > - LastNameStart = ch +1; > - } > - } > - LastNameStart = ch + 1; /* The first one after the comma */ > - } > - } > - if (ch - LastNameStart > 0) { /* A last name? */ > - PWCHAR NameServer = malloc(((ch - LastNameStart) + 1) * sizeof(WCHAR)); > - if (NameServer) { > - memcpy(NameServer,NameServerString + LastNameStart, > - (ch - LastNameStart) * sizeof(WCHAR)); > - NameServer[ch - LastNameStart] = 0; > - cb( Interface, NameServer, Data ); > - free(NameServer); > - } > - } > - ConsumeRegValueString(NameServerString); > - } > -} > - > + NameServerString = QueryRegistryValueStringMulti(RegHandle, L"NameServer"); > + > + if (!NameServerString) return; > + > + for (i = 0; NameServerString[i]; i++) > + { > + cb(Interface, NameServerString[i], Data); > + > + HeapFree(GetProcessHeap(), 0, NameServerString[i]); > + } > + > + HeapFree(GetProcessHeap(), 0, NameServerString); > +} > static void CreateNameServerListEnumNamesFuncCount( PWCHAR Interface, > PWCHAR Server, > PVOID _Data ) { > > > >

14 years, 4 months

Fedora 13 Build Error

by James Tabor

I ran into this on Fedora 13, [HOST-CC] tools/cabman/cabinet.cxx tools/cabman/cabinet.cxx: In member function ‘bool CCabinet::CreateSimpleCabinet()’: tools/cabman/cabinet.cxx:2177: error: no matching function for call to ‘stat::stat(char [260], stat*)’ /usr/include/bits/stat.h:40: note: candidates are: stat::stat() /usr/include/bits/stat.h:40: note: stat::stat(const stat&) from http://code.google.com/p/libproxy/issues/detail?id=122 the patch is based on this; http://code.google.com/p/libproxy/issues/attachmentText?id=122&aid=20023750… Index: tools/cabman/cabinet.cxx =================================================================== --- tools/cabman/cabinet.cxx (revision 48297) +++ tools/cabman/cabinet.cxx (working copy) @@ -21,7 +21,8 @@ #if !defined(WIN32) # include <dirent.h> #endif -#if defined(__FreeBSD__) || defined(__APPLE__) +#if !defined(WIN32) || defined(__FreeBSD__) || defined(__APPLE__) +# include <sys/types.h> # include <sys/stat.h> #endif // __FreeBSD__ #include "cabinet.h" Not sure this is right, so~ posted here for fast review and fix, James

14 years, 4 months

Re: [ros-dev] [ros-diffs] [akhaldi] 48535: [CMAKE] - Add calc to build.

by Colin Finck

akhaldi(a)svn.reactos.org wrote: > +++ branches/cmake-bringup/base/applications/calc/CMakeLists.txt > [...] > > +file(GLOB_RECURSE SOURCE *.c) > +list(REMOVE_ITEM SOURCE > + ${CMAKE_CURRENT_SOURCE_DIR}/fun_mpfr.c > + ${CMAKE_CURRENT_SOURCE_DIR}/rpn_mpfr.c > + ${CMAKE_CURRENT_SOURCE_DIR}/utl_mpfr.c) Should GLOB_RECURSE and later removing unwanted source files really be the way to go? The only advantage I see are smaller CMakeLists files. On the contrary, it is quite an uncommon way and just adding a single unrelated file into an application's directory could render it unbuildable. Additionally, please take a look at http://public.kitware.com/pipermail/cmake/2010-July/037833.html. According to this user, CMake cannot automatically detect added files when using GLOB_RECURSE. In my opinion, we should list every source file to be included to prevent unexpected side effects. Cheers, Colin

14 years, 4 months

Truncating ros-diffs subject lines

by Colin Finck

Hi everybody, Looking at http://www.reactos.org/pipermail/ros-diffs/2010-August/038197.html and the same message in Thunderbird, I now see a need to automatically truncate all subject lines in ros-diffs' mails to 200 characters. I don't think that anybody reads the full commit messages through subject lines, so 200 characters should be enough to identify a commit easily. If there are no objections, I'll try to change this in the next few days. Cheers, Colin

14 years, 4 months

Re: [ros-dev] [ros-diffs] [tkreuzer] 48576: [WIN32K] - co_IntTranslateMouseMessage: properly initialize *HitTest and only send WM_NCHITTEST when the message is going to be removed - co_IntPeekMessage: Prevent possible use of uninitialized HitTest by ProcessMou

by Monkey Libre

2010/8/20, tkreuzer(a)svn.reactos.org <tkreuzer(a)svn.reactos.org>: > Author: tkreuzer > Date: Fri Aug 20 19:24:48 2010 > New Revision: 48576 > > URL: http://svn.reactos.org/svn/reactos?rev=48576&view=rev > Log: > [WIN32K] > - co_IntTranslateMouseMessage: properly initialize *HitTest and only send > WM_NCHITTEST when the message is going to be removed > - co_IntPeekMessage: Prevent possible use of uninitialized HitTest by > ProcessMouseMessage() > - Patch by Jan Roeloffzen [jroeloffzen at hotmail dot com] > - Fixes bug 2139 > > Modified: > trunk/reactos/subsystems/win32/win32k/ntuser/message.c > > Modified: trunk/reactos/subsystems/win32/win32k/ntuser/message.c > URL: > http://svn.reactos.org/svn/reactos/trunk/reactos/subsystems/win32/win32k/nt… > ============================================================================== > --- trunk/reactos/subsystems/win32/win32k/ntuser/message.c [iso-8859-1] > (original) > +++ trunk/reactos/subsystems/win32/win32k/ntuser/message.c [iso-8859-1] Fri > Aug 20 19:24:48 2010 > @@ -579,14 +579,20 @@ > return TRUE; > } > > + *HitTest = HTCLIENT; > + > UserRefObjectCo(Window, &Ref); > > if ( ThreadQueue == Window->pti->MessageQueue && > ThreadQueue->CaptureWindow != Window->hSelf) > { > /* only send WM_NCHITTEST messages if we're not capturing the window! > */ > - *HitTest = co_IntSendMessage(Window->hSelf, WM_NCHITTEST, 0, > - MAKELONG(Msg->pt.x, Msg->pt.y)); > + if (Remove ) > + { > + *HitTest = co_IntSendMessage(Window->hSelf, WM_NCHITTEST, 0, > + MAKELONG(Msg->pt.x, Msg->pt.y)); > + } > + /* else we are going to see this message again, but then with Remove > == TRUE */ > > if (*HitTest == (USHORT)HTTRANSPARENT) > { > @@ -625,10 +631,6 @@ > UserDerefObjectCo(DesktopWindow); > } > } > - } > - else > - { > - *HitTest = HTCLIENT; > } > > if ( gspv.bMouseClickLock && > @@ -800,6 +802,8 @@ > * WM_TIMER messages > */ > CheckMessages: > + > + HitTest = HTNOWHERE; > > Present = FALSE; > > > > -- My translator and tester profile: https://launchpad.net/~monkey-libre -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) iEYEARECAAYFAkq9LEMACgkQY9l6fYUypMPuvACeIZFzQSH3yxMy4wHpQpjK5y9t l34An2b5eHl/hA/AaB8GKbJczM2v0UcT =4iFN -----END PGP SIGNATURE-----

14 years, 4 months

MyReactOS: A Proof of Concept

by victor martinez

Hello everyone! In the latest 6 months, Niski and I have been working on a testing ground/proof of concept of a ReactOS community site, currently called the ReactOS Portal. The idea of a ReactOS community site has been floating around for years due to the fact that our followers (Users and Non-Users) keep expecting more info and more features. For this purpose, and having in mind that a ReactOS community site needs to focus on the followers' needs, we've created a light, eyecandy, easy to maintain Drupal-based site, to serve as a testing ground, and also as a proof of concept for the ReactOS Community idea. Despite the fact that the community site is mainly targeted towards the project followers, we didn't forget our developers and official testers needs. We've already included some (hopefully) useful features, and we welcome ideas and suggestions. The community site is made for everyone, by everyone. To avoid quite a lengthy email, we'll enumerate the Top Features we've added. We've also prepared a PDF and a Video, that you can download from this URL: http://download.myreactos.com/?dir=Vicmarcal ) ************ALL+IN+1*********** All the info in ONE click (latest blog posts, forum replies, commits, ML posts and wiki activity) ******Multiblogging System***** Several blogs, managed from just one Blogging System. Easy to post and maintain. ***********IRC-CHAT************ An integrated IRC client for the ReactOS communiy. *********Contact Page********** Redesigned, full of info. **************FAQ************* A much more user oriented FAQ with new Tester and Translator sections. ************RosKarma*********** A new forum module : "More Karma an User has, more reliable he is" and a new Autobanning System : "Bye,bye Trolls". *********Answered Post********* Another new forum module : Answered topics turn Green(easy to spot) and get locked ("No more Necromancy"). ***********Rosquests*********** Another new Forum module : "Use Forum Users skills for ReactOS project". How? Read the PDF or watch the video for more details. **********MultivideoBox******** "An image is worth than thousand words". Videotutorials, ReactOS presentations... and much more. *******Social Networking******* "Facebook, Vimeo, Youtube, and Tweeter integration". And after getting some feedback we've added more features : ***Unlimited/Free FTP account*** Now Official Developers/Testers can have FTP accounts to easily and conveniently share ReactOS ISOs (or any other files). We're also considering integrating the Software Compatibility Database into the Portal. We hope you like it, your feedback is very important, so that we can create a comfortable, highly efficient Community Site. More info in the Video+PDF: http://download.myreactos.com/?dir=Vicmarcal Thanks to all the ReactOS members that are joining, posting and translating. Thanks to everyone involved ;) Victor Martinez Calvo Maciej Bialas www.myreactos.com

14 years, 4 months

Re: [ros-dev] [ros-diffs] [cgutman] 48522: [MSAFD] - Update lpcbBytesReturned when FIONREAD is called - Fix the value inside lpErrno when XxxSocketInformation() fails - Pass the unrecognized IOCTLs to the winsock helper (no behavior change for this yet because wsh

by Gabriel ilardi

Was this meant for trunk instead of a branch ? After this commit DNS resolution doesn't work. Thanks. Gabriel. > Date: Thu, 12 Aug 2010 13:13:21 +0000 > To: ros-diffs(a)reactos.org > From: cgutman(a)svn.reactos.org > Subject: [ros-diffs] [cgutman] 48522: [MSAFD] - Update lpcbBytesReturned when FIONREAD is called - Fix the value inside lpErrno when XxxSocketInformation() fails - Pass the unrecognized IOCTLs to the winsock helper (no behavior change for this yet because wshtcpip's WSHIoctl is unimplemented) > > Author: cgutman > Date: Thu Aug 12 13:13:19 2010 > New Revision: 48522 > > URL: http://svn.reactos.org/svn/reactos?rev=48522&view=rev > Log: > [MSAFD] > - Update lpcbBytesReturned when FIONREAD is called > - Fix the value inside lpErrno when XxxSocketInformation() fails > - Pass the unrecognized IOCTLs to the winsock helper (no behavior change for this yet because wshtcpip's WSHIoctl is unimplemented) > > Modified: > trunk/reactos/dll/win32/msafd/misc/dllmain.c > > Modified: trunk/reactos/dll/win32/msafd/misc/dllmain.c > URL: http://svn.reactos.org/svn/reactos/trunk/reactos/dll/win32/msafd/misc/dllma… > ============================================================================== > --- trunk/reactos/dll/win32/msafd/misc/dllmain.c [iso-8859-1] (original) > +++ trunk/reactos/dll/win32/msafd/misc/dllmain.c [iso-8859-1] Thu Aug 12 13:13:19 2010 > @@ -1910,6 +1910,7 @@ > OUT LPINT lpErrno) > { > PSOCKET_INFORMATION Socket = NULL; > + BOOLEAN NeedsCompletion; > > /* Get the Socket Structure associate to this Socket*/ > Socket = GetSocketStructure(Handle); > @@ -1918,6 +1919,8 @@ > *lpErrno = WSAENOTSOCK; > return SOCKET_ERROR; > } > + > + *lpcbBytesReturned = 0; > > switch( dwIoControlCode ) > { > @@ -1928,17 +1931,44 @@ > return SOCKET_ERROR; > } > Socket->SharedData.NonBlocking = *((PULONG)lpvInBuffer) ? 1 : 0; > - return SetSocketInformation(Socket, AFD_INFO_BLOCKING_MODE, (PULONG)lpvInBuffer, NULL); > + *lpErrno = SetSocketInformation(Socket, AFD_INFO_BLOCKING_MODE, (PULONG)lpvInBuffer, NULL); > + if (*lpErrno != NO_ERROR) > + return SOCKET_ERROR; > + else > + return NO_ERROR; > case FIONREAD: > if( cbOutBuffer < sizeof(INT) || IS_INTRESOURCE(lpvOutBuffer) ) > { > *lpErrno = WSAEFAULT; > return SOCKET_ERROR; > } > - return GetSocketInformation(Socket, AFD_INFO_RECEIVE_CONTENT_SIZE, (PULONG)lpvOutBuffer, NULL); > + *lpErrno = GetSocketInformation(Socket, AFD_INFO_RECEIVE_CONTENT_SIZE, (PULONG)lpvOutBuffer, NULL); > + if (*lpErrno != NO_ERROR) > + { > + *lpcbBytesReturned = sizeof(ULONG); > + return SOCKET_ERROR; > + } > + else > + return NO_ERROR; > default: > - *lpErrno = WSAEINVAL; > - return SOCKET_ERROR; > + *lpErrno = Socket->HelperData->WSHIoctl(Socket->HelperContext, > + Handle, > + Socket->TdiAddressHandle, > + Socket->TdiConnectionHandle, > + dwIoControlCode, > + lpvInBuffer, > + cbInBuffer, > + lpvOutBuffer, > + cbOutBuffer, > + lpcbBytesReturned, > + lpOverlapped, > + lpCompletionRoutine, > + (LPBOOL)&NeedsCompletion); > + > + if (*lpErrno != NO_ERROR) > + return SOCKET_ERROR; > + else > + return NO_ERROR; > } > } > > >

14 years, 4 months

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Ros-dev August 2010