Ros-dev March 2005

ros-dev@reactos.org

64 participants
200 discussions

Assertion in /objects/dc.c
by dolphin_＠gmx.net 08 Mar '05

08 Mar '05

Hello! I have just updates to the latest SVN version and now I have the problem, that I cannot install ReactOS anymore. At the first reboot after installation, I get the following Assertion (cut from debug output): (devinst.c:1008:SetupDiInstallClassW) Assertion 'NewDC->DMW.dmBitsPerPel == BitsPerFormat(SurfObj->iBitmapFormat)' fai led at objects/dc.c line 808 Entered debugger on embedded INT3 at 0x0008:0x80006243. When I ignore this Assertion (by continuing), I can do so serveral times, after that my PC reboots. The error occurs on both QEMU and real HW. I guess this error has something to do with the latest changes. Perhaps someone can verify/fix this (GvG ?). I discoverd one or two other problems too (that have nothing to do with the latest SVN build), but I have to recheck them to ensure they still exist. This is what I've done: I have clicked wildly with the right mouse button on the Desktop background to open the context menu and randomly used the context menu of the Explorer to move the Explorer windows arround in between. When I do this for a couple of seconds, ReactOS crashes. (I posted a BSOD yesterday in IRC, but I guess nobody noticed it, here is the link: http://www.rafb.net/paste/results/kxur3L28.html As I said, I will recheck this in the evening today. Perhaps someone else will have a look on this too. Best regards, Dolphin

2 2

new build system: informal performance test
by Royce Mitchell III 07 Mar '05

07 Mar '05

each test was run from a "mingw32-make clean" or "make clean" tree with KDBG=0 DBG=0 old build system build command: mingw32-make mingw32-make bootcd 21 minutes new build system: make bootcd 11 minutes new build system: make -j 5 bootcd 13 minutes I built the old build system with two separate commands because last time I tried, you could do a "mingw32-make bootcd" from a clean tree. So, besides the other benefits outlined when we started this project, the new build system is cutting build time in half. Now to just fix the remaining bug(s) ;)

7 9

Proposed change to usermode exception reporting
by art yerkes 07 Mar '05

07 Mar '05

I want to change the usermode exception reporter to make stack traces in this form: (KERNEL32:except/except.c:184) 6dc00000+cb991 C:\dl\libgtk-0.dll rather than this from: (KERNEL32:except/except.c:184) 6dccb991 C:\dl\libgtk-0.dll My form is less compact but gives more information (namely, the offset in the target module). It's a simple change but I wanted to see if anyone objected strongly. -- Here's a simple experiment. Stand on a train track between two locomotives which are pushing on you with equal force in opposite directions. You will exhibit no net motion. None the less, you may soon begin to notice that something important is happening. -- Robert Stirniman

2 1

0.2.6 branch
by Steven Edwards 06 Mar '05

06 Mar '05

Hi, Robert if you are out there I think we want to try to branch for 0.2.6 sometime in the middle of the coming week. Alex is going to merge his changes in to the trunk and then we should be mostly ready. If Robert is not free and no one else wants to do it then I will plan on branching on Wednesday. I expect we can let the 0.2.6 branch run for about a week so lets plan on shipping 0.2.6-final on 2005/03/16. Thanks Steven __________________________________ Celebrate Yahoo!'s 10th Birthday! Yahoo! Netrospective: 100 Moments of the Web http://birthday.yahoo.com/netrospective/

1 0

Re: [ros-diffs] [ekohl] 13825: Fix calls to ReadFile and WriteFile for asynchronous I/O.
by Hartmut Birr 06 Mar '05

06 Mar '05

Hi, you should send this mail to the Ros-Dev list because some people with a deal-up connection dosn't read the Ros-Diffs list. - Hartmut Gunnar Dalsnes schrieb: > Eric: Ping? > > Gunnar Dalsnes wrote: > >> From MSDN OVERLAPPED: "Functions such as WriteFile set the event to >> the nonsignaled state before they begin an I/O operation" >> >> So the ResetEvent additions are meaningless... Why did you add them? >> >> Gunnar >> >> ekohl(a)svn.reactos.com wrote: >> >>> Fix calls to ReadFile and WriteFile for asynchronous I/O. >>> >>> Modified: trunk/reactos/lib/rpcrt4/rpc_message.c >>> >>> ------------------------------------------------------------------------ >>> >>> *Modified: trunk/reactos/lib/rpcrt4/rpc_message.c* >>> >>> --- trunk/reactos/lib/rpcrt4/rpc_message.c 2005-03-05 11:27:15 >>> UTC (rev 13824) >>> +++ trunk/reactos/lib/rpcrt4/rpc_message.c 2005-03-05 11:38:48 >>> UTC (rev 13825) >>> @@ -265,10 +265,15 @@ >>> >>> } >>> >>> /* transmit packet header */ >>> >>> - if (!WriteFile(Connection->conn, Header, hdr_size, &count, >>> NULL)) { >>> >>> + ResetEvent(Connection->ovl.hEvent); >>> + if (!WriteFile(Connection->conn, Header, hdr_size, &count, >>> &Connection->ovl)) { >>> >>> WARN("WriteFile failed with error %ld\n", GetLastError()); >>> return GetLastError(); >>> } >>> >>> + if (!GetOverlappedResult(Connection->conn, &Connection->ovl, >>> &count, TRUE)) { >>> + WARN("GetOverlappedResult failed with error %ld\n", >>> GetLastError()); >>> + return GetLastError(); >>> + } >>> >>> >>> /* fragment consisted of header only and is the last one */ >>> if (hdr_size == Header->common.frag_len && >>> @@ -277,10 +282,15 @@ >>> >>> } >>> >>> /* send the fragment data */ >>> >>> - if (!WriteFile(Connection->conn, buffer_pos, >>> Header->common.frag_len - hdr_size, &count, NULL)) { >>> >>> + ResetEvent(Connection->ovl.hEvent); >>> + if (!WriteFile(Connection->conn, buffer_pos, >>> Header->common.frag_len - hdr_size, &count, &Connection->ovl)) { >>> >>> WARN("WriteFile failed with error %ld\n", GetLastError()); >>> return GetLastError(); >>> } >>> >>> + if (!GetOverlappedResult(Connection->conn, &Connection->ovl, >>> &count, TRUE)) { >>> + WARN("GetOverlappedResult failed with error %ld\n", >>> GetLastError()); >>> + return GetLastError(); >>> + } >>> >>> >>> Header->common.flags &= ~RPC_FLG_FIRST; >>> } >>> @@ -309,9 +319,15 @@ >>> >>> TRACE("(%p, %p, %p)\n", Connection, Header, pMsg); >>> >>> /* read packet common header */ >>> >>> - if (!ReadFile(Connection->conn, &common_hdr, sizeof(common_hdr), >>> &dwRead, NULL)) { >>> >>> + ResetEvent(Connection->ovl.hEvent); >>> + if (!ReadFile(Connection->conn, &common_hdr, sizeof(common_hdr), >>> &dwRead, &Connection->ovl)) { >>> + WARN("ReadFile failed with error %ld\n", GetLastError()); >>> + status = RPC_S_PROTOCOL_ERROR; >>> + goto fail; >>> + } >>> + if (!GetOverlappedResult(Connection->conn, &Connection->ovl, >>> &dwRead, TRUE)) { >>> >>> if (GetLastError() != ERROR_MORE_DATA) { >>> >>> - WARN("ReadFile failed with error %ld\n", GetLastError()); >>> >>> + WARN("GetOverlappedResult failed with error %ld\n", >>> GetLastError()); >>> >>> status = RPC_S_PROTOCOL_ERROR; >>> goto fail; >>> } >>> @@ -339,10 +355,16 @@ >>> >>> memcpy(*Header, &common_hdr, sizeof(common_hdr)); >>> >>> /* read the rest of packet header */ >>> >>> + ResetEvent(Connection->ovl.hEvent); >>> >>> if (!ReadFile(Connection->conn, &(*Header)->common + 1, >>> >>> - hdr_length - sizeof(common_hdr), &dwRead, NULL)) { >>> >>> + hdr_length - sizeof(common_hdr), &dwRead, >>> &Connection->ovl)) { >>> + WARN("ReadFile failed with error %ld\n", GetLastError()); >>> + status = RPC_S_PROTOCOL_ERROR; >>> + goto fail; >>> + } >>> + if (!GetOverlappedResult(Connection->conn, &Connection->ovl, >>> &dwRead, TRUE)) { >>> >>> if (GetLastError() != ERROR_MORE_DATA) { >>> >>> - WARN("ReadFile failed with error %ld\n", GetLastError()); >>> >>> + WARN("GetOverlappedResult failed with error %ld\n", >>> GetLastError()); >>> >>> status = RPC_S_PROTOCOL_ERROR; >>> goto fail; >>> } >>> @@ -352,6 +374,7 @@ >>> >>> goto fail; >>> } >>> >>> >>> + >>> >>> /* read packet body */ >>> switch (common_hdr.ptype) { >>> case PKT_RESPONSE: >>> @@ -379,13 +402,20 @@ >>> >>> goto fail; >>> } >>> >>> >>> - if (data_length == 0) dwRead = 0; else >>> - if (!ReadFile(Connection->conn, buffer_ptr, data_length, >>> &dwRead, NULL)) { >>> - if (GetLastError() != ERROR_MORE_DATA) { >>> >>> + if (data_length == 0) dwRead = 0; else { >>> + ResetEvent(Connection->ovl.hEvent); >>> + if (!ReadFile(Connection->conn, buffer_ptr, data_length, >>> &dwRead, &Connection->ovl)) { >>> >>> WARN("ReadFile failed with error %ld\n", GetLastError()); >>> status = RPC_S_PROTOCOL_ERROR; >>> goto fail; >>> } >>> >>> + if (!GetOverlappedResult(Connection->conn, &Connection->ovl, >>> &dwRead, TRUE)) { >>> + if (GetLastError() != ERROR_MORE_DATA) { >>> + WARN("GetOverlappedResult failed with error %ld\n", >>> GetLastError()); >>> + status = RPC_S_PROTOCOL_ERROR; >>> + goto fail; >>> + } >>> + } >>> >>> } >>> if (dwRead != data_length) { >>> status = RPC_S_PROTOCOL_ERROR; >>> @@ -403,10 +433,16 @@ >>> >>> TRACE("next header\n"); >>> >>> /* read the header of next packet */ >>> >>> - if (!ReadFile(Connection->conn, *Header, hdr_length, &dwRead, >>> NULL)) { >>> >>> + ResetEvent(Connection->ovl.hEvent); >>> + if (!ReadFile(Connection->conn, *Header, hdr_length, &dwRead, >>> &Connection->ovl)) { >>> + WARN("ReadFile failed with error %ld\n", GetLastError()); >>> + status = GetLastError(); >>> + goto fail; >>> + } >>> + if (!GetOverlappedResult(Connection->conn, &Connection->ovl, >>> &dwRead, TRUE)) { >>> >>> if (GetLastError() != ERROR_MORE_DATA) { >>> >>> - WARN("ReadFile failed with error %ld\n", GetLastError()); >>> - status = GetLastError(); >>> >>> + WARN("GetOverlappedResult failed with error %ld\n", >>> GetLastError()); >>> + status = RPC_S_PROTOCOL_ERROR; >>> >>> goto fail; >>> } >>> } >>> >>> >>> ------------------------------------------------------------------------ >>> >>>

2 1

Re: [ros-svn] [ion] 13847: Merge with 13846. Fixes QEMU+KDBG problem.
by James Tabor 06 Mar '05

06 Mar '05

Hi! ion(a)svn.reactos.com wrote: > Merge with 13846. Fixes QEMU+KDBG problem. > > > Added files: > branches/alex_devel_branch/reactos/media/drivers/etc/KDBinit > > Updated files: > branches/alex_devel_branch/reactos/Makefile > branches/alex_devel_branch/reactos/ntoskrnl/dbg/kdb.h > branches/alex_devel_branch/reactos/ntoskrnl/dbg/kdb_cli.c > branches/alex_devel_branch/reactos/ntoskrnl/dbg/kdb_keyboard.c > branches/alex_devel_branch/reactos/ntoskrnl/dbg/kdb_serial.c > > Deleted files: > branches/alex_devel_branch/reactos/media/drivers/etc/KDB.init > I think is is caused by blight patch. Gcc 3.4.2 bin utils 2.15.94. ntoskrnl: [CC] dbg/kdb_cli.c dbg/kdb_cli.c: In function `KdbpPrint': dbg/kdb_cli.c:1721: warning: implicit declaration of function `isalpha' make[1]: *** [dbg/kdb_cli.o] Error 1 make: *** [ntoskrnl] Error 2 Thanks, James

2 3

Build ros on ros is broken
by Hartmut Birr 06 Mar '05

06 Mar '05

Hi, since my last svn update, I'm not able to compile ros on ros. It seems that something is wrong with the pipes. - Hartmut f:\Sandbox\Ros\ReactOS>_make ReactOS Build Number Generator Current date : 2005-03-06 ROS Version : 0.3-CVS (build 20050306) ar: creating zlib.host.a ntoskrnl: [DLLTOOL] libntoskrnl.a hal: [CC] hal.c gcc: pipe: Invalid argument (EINVAL) _make[1]: *** [hal.o] Error 1 _make: *** [hallib] Error 2

2 5

Re: Re: Re: [ros-dev] added a proposed patch to bugzilla
by Roman Hoegg 06 Mar '05

06 Mar '05

1 0

Re: [ros-svn] [blight] 13841: Little KDB update ; -) If you have any problems and/or questions let me know. I hope it was tested enough and works well enough for everybody.
by Hartmut Birr 06 Mar '05

06 Mar '05

Hi, compiling fails if KDBG = 0 and DBG = 1. objects/ke.o: In function `InitializeListHead': M:/Sandbox/ros_work/reactos/ntoskrnl/../include/ntos/rtl.h:96: undefined reference to `KdbInit' objects/ps.o: In function `PsUnlockProcess': M:/Sandbox/ros_work/reactos/ntoskrnl/ps/process.c:2973: undefined reference to `KdbDeleteProcessHook' collect2: ld returned 1 exit status _make[1]: *** [ntoskrnl.nostrip.exe] Error 1 _make: *** [ntoskrnl] Error 2 - Hartmut blight(a)svn.reactos.com schrieb: >Little KDB update ;-) If you have any problems and/or questions let me know. I hope it was tested enough and works well enough for everybody. > > >Added files: >trunk/reactos/media/drivers/etc/KDB.init >trunk/reactos/ntoskrnl/dbg/i386/longjmp.S >trunk/reactos/ntoskrnl/dbg/i386/setjmp.S >trunk/reactos/ntoskrnl/dbg/kdb_cli.c >trunk/reactos/ntoskrnl/dbg/kdb_expr.c >trunk/reactos/ntoskrnl/dbg/kdb_string.c > >Updated files: >trunk/reactos/Makefile >trunk/reactos/drivers/input/keyboard/keyboard.c >trunk/reactos/ntoskrnl/Makefile >trunk/reactos/ntoskrnl/dbg/i386/i386-dis.c >trunk/reactos/ntoskrnl/dbg/i386/kdb_help.S >trunk/reactos/ntoskrnl/dbg/kdb.c >trunk/reactos/ntoskrnl/dbg/kdb.h >trunk/reactos/ntoskrnl/dbg/kdb_keyboard.c >trunk/reactos/ntoskrnl/dbg/kdb_serial.c >trunk/reactos/ntoskrnl/dbg/kdb_symbols.c >trunk/reactos/ntoskrnl/include/internal/i386/ke.h >trunk/reactos/ntoskrnl/include/internal/kd.h >trunk/reactos/ntoskrnl/ke/catch.c >trunk/reactos/ntoskrnl/ke/i386/trap.s >trunk/reactos/ntoskrnl/ke/i386/tskswitch.S >trunk/reactos/ntoskrnl/ke/kthread.c >trunk/reactos/ntoskrnl/ke/main.c >trunk/reactos/ntoskrnl/ps/w32call.c > >_______________________________________________ >Ros-svn mailing list >Ros-svn(a)reactos.com >http://reactos.com:8080/mailman/listinfo/ros-svn > > > >

1 0

ATTN CASPER: case-conflicts in svn
by Royce Mitchell III 06 Mar '05

06 Mar '05

Casper: could we please have this hook installed in the svn server?: http://svn.collab.net/viewcvs/svn/trunk/contrib/hook-scripts/check-case-ins…

1 0

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Ros-dev March 2005