Ros-dev

ros-dev@reactos.org

5789 discussions

Re: [ros-svn] [hbirr] 14964: Do always set the UserIosb of an irp in IoSecondStageCompletion.

by Alex Ionescu

hbirr(a)svn.reactos.com wrote: >Do always set the UserIosb of an irp in IoSecondStageCompletion. > > > >Updated files: >trunk/reactos/ntoskrnl/io/irp.c > > This is incorrect. 1) The IOSB should not always be set. Create a driver and fail an operation that you send to yourself by an IRP. Make that IRP not SYNCH_API, or better yet, make sure you don't have a File Object. You will notice that the Status Block is not touched. 2) The IOSB is not checked if it exists, it should ALWAYS be there. IRPs without a IOSB are invalid. To verify this, set the IOSB of your IRP to 0 and run Windows with a Debugger. You will see that it will break in many places, because Windows has simply placed SEH to make sure that the write is valid. So the correct thing to do is wrap the write in SEH, which protects both against invalid pointers and zero ones, but that still doesn't mean they are"valid" and should be checked that way. Best regards, Alex Ionescu

20 years, 3 months

Both CMD Consoles Go Puff!

by James Tabor

Hi All! Wow! Ros on ros build and I walk away come back and both consoles are gone. Thank god for taskmgr! It shows both cmds', ctm, and three make.exe, not running but still in memory. Build 14598 (KERNEL32:mem/global.c:412) Memory Load: 23 (KERNEL32:mem/global.c:412) Memory Load: 23 (MSVCRT:io/open.c:103) not valid fd 0, g_fdend 5, fdinfo 7802d2f0, bucket 7802d2 f0, fdflags 0 (MSVCRT:io/open.c:700) _setmode: inval fd (0) (NTDLL:ldr/utils.c:2097) Relocating (77ed0000 -> 780000) C:\ReactOS\System32\gdi 32.dll (NTDLL:ldr/utils.c:2097) Relocating (77ed0000 -> 850000) C:\ReactOS\system32\gdi 32.dll (NTDLL:rtl/critical.c:446) Deadlock: 8517f8 Unhandled exception Address: 7c90a05c C:\ReactOS\system32\ntdll.dll CS:EIP 1b:7c90a3d2 DS 23 ES 23 FS 3b GS 23 EAX: 00000000 EBX: 00000000 ECX: 00000000 EDX: 00000000 EBP: 019afdd8 ESI: 00000000 ESP: 019afde0 EDI: 00000000 EFLAGS: 00000212 Frames: 7c900000+a05c C:\ReactOS\system32\ntdll.dll 5ffb0000+8381 C:\ReactOS\system32\win32csr.dll 5ffb0000+8eac C:\ReactOS\system32\win32csr.dll 220000+12a8f C:\ReactOS\system32\user32.dll 220000+12ced C:\ReactOS\system32\user32.dll 5ffb0000+91c4 C:\ReactOS\system32\win32csr.dll 7c800000+26554 C:\ReactOS\system32\kernel32.dll (NTDLL:rtl/critical.c:446) Deadlock: 8517f8 (ke/catch.c:191) Unhandled UserMode exception, terminating thread Explorer is running btw, James

20 years, 3 months

ping

by Hartmut Birr

20 years, 3 months

RE: [ros-dev] gcc problem or not

by Waldo Alvarez Cañizares

Hi: I doubt this could be a bug. A lot of code would break with it. Maybe it does violates the short circuit in the small number of cases where that can be done. As in this case, it doesn't matters. Regards Waldo ________________________________ From: ros-dev-bounces(a)reactos.com on behalf of Hartmut Birr Sent: Tue 5/3/2005 2:43 PM To: ReactOS Development List Subject: [ros-dev] gcc problem or not Hi, I've looked on a map file from an optimized build. I found the following code: /* * If the directory containing the file to open doesn't exist then * fail immediately */ if (Status == STATUS_OBJECT_PATH_NOT_FOUND || 12015: 83 c4 0c add $0xc,%esp 12018: 89 c3 mov %eax,%ebx 1201a: 3d 3a 00 00 c0 cmp $0xc000003a,%eax 1201f: 0f 94 c2 sete %dl 12022: 3d 0d 00 00 c0 cmp $0xc000000d,%eax 12027: 0f 94 c0 sete %al 1202a: 09 d0 or %edx,%eax 1202c: a8 01 test $0x1,%al 1202e: 0f 85 42 01 00 00 jne 12176 <_VfatCreateFile+0x2e9> 12034: 81 fb 56 00 00 c0 cmp $0xc0000056,%ebx 1203a: 0f 84 36 01 00 00 je 12176 <_VfatCreateFile+0x2e9> Status == STATUS_INVALID_PARAMETER || Status == STATUS_DELETE_PENDING) { if (ParentFcb) { vfatReleaseFCB (DeviceExt, ParentFcb); } return(Status); } I was always the opinion that the examination of a test condition stops if the result can not change again. A test condition like this: if (pointer == NULL || pointer->member == 0) should never access pointer->member if pointer is zero. Compared with the code above, it is possible that gcc build the result from the right side of the OR statement. This may hit a page fault. Is this a bug in gcc? - Hartmut _______________________________________________ Ros-dev mailing list Ros-dev(a)reactos.com http://reactos.com:8080/mailman/listinfo/ros-dev

20 years, 3 months

gcc problem or not

by Hartmut Birr

Hi, I've looked on a map file from an optimized build. I found the following code: /* * If the directory containing the file to open doesn't exist then * fail immediately */ if (Status == STATUS_OBJECT_PATH_NOT_FOUND || 12015: 83 c4 0c add $0xc,%esp 12018: 89 c3 mov %eax,%ebx 1201a: 3d 3a 00 00 c0 cmp $0xc000003a,%eax 1201f: 0f 94 c2 sete %dl 12022: 3d 0d 00 00 c0 cmp $0xc000000d,%eax 12027: 0f 94 c0 sete %al 1202a: 09 d0 or %edx,%eax 1202c: a8 01 test $0x1,%al 1202e: 0f 85 42 01 00 00 jne 12176 <_VfatCreateFile+0x2e9> 12034: 81 fb 56 00 00 c0 cmp $0xc0000056,%ebx 1203a: 0f 84 36 01 00 00 je 12176 <_VfatCreateFile+0x2e9> Status == STATUS_INVALID_PARAMETER || Status == STATUS_DELETE_PENDING) { if (ParentFcb) { vfatReleaseFCB (DeviceExt, ParentFcb); } return(Status); } I was always the opinion that the examination of a test condition stops if the result can not change again. A test condition like this: if (pointer == NULL || pointer->member == 0) should never access pointer->member if pointer is zero. Compared with the code above, it is possible that gcc build the result from the right side of the OR statement. This may hit a page fault. Is this a bug in gcc? - Hartmut

20 years, 3 months

RE: [ros-svn] [ion] 14799: KD System Rewrite:

by Ge van Geldorp

> From: ion(a)svn.reactos.com > > KD System Rewrite: This utterly broke the GDB stub, to the point that I can't even use DbgPrint anymore to try and figure out what's wrong. May I suggest you go and read http://www.joelonsoftware.com/articles/fog0000000069.html first, revert this patch (so I can get on with my work), and then, if you still feel this overwhelming urge to rewrite, test all the functionality you're replacing (you obviously didn't even consider a "/DEBUGPORT=GDB" and boot test necessary) before recommiting. If I sound a little bit pissed-off, that's because it's exactly how I feel. This is not the first time I've been bitten by a rewrite. Back in January I spent a lot of time improving symbol handling and profiling, only to see you remove the profiling code a few weeks later, for a "new and improved" rewrite. Only problem is, the "new and improved" profiling system doesn't produce any output and is therefore useless. You said you would fix that in a weeks time, but now, months later, still nothing. I'll be damned if I'm going to let the same happen to the GDB stub, which I use on a daily basis. We all break stuff sometimes (God knows I do) and I can live with that. I can even live with rewrites when they're necessary for binary compatibility (after all, that's what this project is all about) but I'm highly suspicious of rewrites because of "well, uhmm, I think my way is cleaner". Gé van Geldorp.

20 years, 3 months

keyboard problem in qemu

by Hartmut Birr

Hi, since the keyboard driver has changed, I'm not able to install ros within qemu if i want change the selection of some components. The cursor keys doesn't work. They has worked with the old driver and they do work with freeldr or other OS. I've enabled the debug output from keyboard driver. There is a little difference between qemu and real hardware. The debug output is from hitting the cursor up key. - Hartmut QEMU (keyboard.c:107) PortStatus: 1d (keyboard.c:120) Got: 48 (keyboard.c:145) Irq is keyboard input (keyboard.c:74) Irq completes key (keyboard.c:231) Send a key (kbdclass.c:81) ServiceCallback called (kbdclass.c:64) Send scancode: 48 (kbdclass.c:68) Success! (kbdclass.c:164) DeviceObject 8086b618 (kbdclass.c:165) Irp 808c94a8 (kbdclass.c:167) Dispatch: stk->MajorFunction 3 (kbdclass.c:168) AlreadyOpened 1 (kbdclass.c:189) Queueing packet (kbdclass.c:115) KeyboardStartIo(DeviceObject 8086b618 Irp 808c94a8) (kbdclass.c:62) Pending... (NrRead 0, NrToRead 1 (keyboard.c:107) PortStatus: 1d (keyboard.c:120) Got: c8 (keyboard.c:145) Irq is keyboard input (keyboard.c:74) Irq completes key (keyboard.c:231) Send a key (kbdclass.c:81) ServiceCallback called (kbdclass.c:64) Send scancode: 48 (kbdclass.c:68) Success! (kbdclass.c:164) DeviceObject 8086b618 (kbdclass.c:165) Irp 808c94a8 (kbdclass.c:167) Dispatch: stk->MajorFunction 3 (kbdclass.c:168) AlreadyOpened 1 (kbdclass.c:189) Queueing packet (kbdclass.c:115) KeyboardStartIo(DeviceObject 8086b618 Irp 808c94a8) (kbdclass.c:62) Pending... (NrRead 0, NrToRead 1 Real Hardware (keyboard.c:107) PortStatus: 11 (keyboard.c:120) Got: e0 (keyboard.c:145) Irq is keyboard input (keyboard.c:107) PortStatus: 11 (keyboard.c:120) Got: 48 (keyboard.c:145) Irq is keyboard input (keyboard.c:74) Irq completes key (keyboard.c:231) Send a key (kbdclass.c:81) ServiceCallback called (kbdclass.c:64) Send scancode: 48 (kbdclass.c:68) Success! (kbdclass.c:164) DeviceObject 80873a50 (kbdclass.c:165) Irp 808d2ed8 (kbdclass.c:167) Dispatch: stk->MajorFunction 3 (kbdclass.c(keyboard.c:107) PortStatus: 11 (keyboard.c:120) Got: e0 (keyboard.c:145) Irq is keyboard input (keyboard.c:107) PortStatus: 11 (keyboard.c:120) Got: c8 (keyboard.c:145) Irq is keyboard input (keyboard.c:74) Irq completes key (keyboard.c:231) Send a key (kbdclass.c:81) ServiceCallback called :168) AlreadyOpened 1 (kbdclass.c:189) Queueing packet (kbdclass.c:115) KeyboardStartIo(DeviceObject 80873a50 Irp 808d2ed8) (kbdclass.c:64) Send scancode: 48 (kbdclass.c:68) Success! (kbdclass.c:164) DeviceObject 80873a50 (kbdclass.c:165) Irp 808d2ed8 (kbdclass.c:167) Dispatch: stk->MajorFunction 3 (kbdclass.c:168) AlreadyOpened 1 (kbdclass.c:189) Queueing packet (kbdclass.c:115) KeyboardStartIo(DeviceObject 80873a50 Irp 808d2ed8) (kbdclass.c:62) Pending... (NrRead 0, NrToRead 1

20 years, 3 months

Using BOCHS and external gdb

by Mark Junker

Hi, I just want to mention that COM port redirection is possible in Bochs CVS HEAD. Currently it only supports "null", "file", "raw", and "mouse". I extended it in my local copy so that you can use "pipe" too. I attached my patch and my own version of pipetunnel. This patch is already on the Bochs "Patches" page. I hope that it'll be added ASAP to the main source trunk. BTW: I've seen a patch to allow COM <-> TCP/IP redirection on the "Patches" page. When there is interest, I can combine the two patches and publish them here ... Regards, Mark Index: config.cc =================================================================== RCS file: /cvsroot/bochs/bochs/config.cc,v retrieving revision 1.34 diff -u -r1.34 config.cc --- config.cc 3 Apr 2005 15:00:44 -0000 1.34 +++ config.cc 3 May 2005 00:39:58 -0000 @@ -863,6 +863,7 @@ static char *serial_mode_list[] = { "null", "file", + "pipe", "term", "raw", "mouse", Index: iodev/serial.cc =================================================================== RCS file: /cvsroot/bochs/bochs/iodev/serial.cc,v retrieving revision 1.64 diff -u -r1.64 serial.cc --- iodev/serial.cc 19 Jan 2005 18:21:37 -0000 1.64 +++ iodev/serial.cc 3 May 2005 00:54:44 -0000 @@ -45,6 +45,148 @@ #define LOG_THIS theSerialDevice-> +#ifdef WIN32 +class TPipe { +private: + HANDLE hPipe; + char *pszPipeName; + +public: + TPipe(const char *pszPipeName_arg=NULL) { + pszPipeName = NULL; + hPipe = INVALID_HANDLE_VALUE; + Open(pszPipeName_arg); + } + ~TPipe(void) { + Close(); + if (pszPipeName!=NULL) { + free(pszPipeName); + } + } + + bool Open(const char *pszPipeName_arg=NULL) { + Close(); + if (pszPipeName_arg!=NULL) { + if (pszPipeName!=NULL) { + free(pszPipeName); + } + pszPipeName = strdup(pszPipeName_arg); + } + if (pszPipeName==NULL) + return false; + hPipe = + CreateFileA(pszPipeName, + GENERIC_READ | GENERIC_WRITE, + 0, + NULL, + OPEN_EXISTING, + 0, + NULL); + if (hPipe!=INVALID_HANDLE_VALUE) { + DWORD dwMode = PIPE_READMODE_BYTE | PIPE_NOWAIT; + if (!SetNamedPipeHandleState(hPipe, + &dwMode, + NULL, + NULL)) + { + Close(); + } + } + } + void Close(void) { + if (hPipe!=INVALID_HANDLE_VALUE) { + CloseHandle(hPipe); + hPipe = INVALID_HANDLE_VALUE; + } + } + + bool Ok(void) const { + return hPipe!=INVALID_HANDLE_VALUE; + } + + size_t Write(const void *pcv,size_t size) { + assert(hPipe!=INVALID_HANDLE_VALUE); + DWORD dwWritten = 0; + if (!WriteFile(hPipe, + pcv, + (DWORD) size, + &dwWritten, + NULL)) + return 0; + return (size_t) dwWritten; + } + + size_t Read(void *pv,size_t size) { + assert(hPipe!=INVALID_HANDLE_VALUE); + DWORD dwRead = 0; + if (!ReadFile(hPipe, + pv, + (DWORD) size, + &dwRead, + NULL)) + { + return 0; + } + return (size_t) dwRead; + } + + bool Flush(void) { + assert(hPipe!=INVALID_HANDLE_VALUE); + return FlushFileBuffers(hPipe)!=0; + } + + bool Terminated(void) { + assert(hPipe!=INVALID_HANDLE_VALUE); + DWORD dwBytesAvail = 0; + if (!PeekNamedPipe(hPipe, + NULL, + 0, + NULL, + &dwBytesAvail, + NULL)) + return true; + return false; + } + + HANDLE GetHandle(void) const { + return hPipe; + } +}; +#else +class TPipe +{ +private: + int iHandle; + +public: + TPipe(const char *pszFileName) { + iHandle = open(pszFileName, O_RDWR | O_NONBLOCK, 600); + } + ~TPipe(void) { + if (iHandle >=0 ) + close(iHandle); + } + bool Ok(void) const { + return iHandle >= 0; + } + size_t Write(const void *pv, size_t size) + { + return write(iHandle, pv, size); + } + size_t Read(void *pv, size_t size) + { + return read(iHandle, pv, size); + } + bool Flush(void) { + return commit(iHandle)==0; + } + + int GetHandle(void) const { + return iHandle; + } +}; +#endif + bx_serial_c *theSerialDevice = NULL; int @@ -82,6 +224,11 @@ if (BX_SER_THIS s[i].output != NULL) fclose(BX_SER_THIS s[i].output); break; + case BX_SER_MODE_PIPE: + // Close the handler + // It must not happen that io_handler is NULL at this point + delete (TPipe*) BX_SER_THIS s[i].io_handler; + break; case BX_SER_MODE_TERM: #if defined(SERIAL_ENABLE) if (s[i].tty_id >= 0) { @@ -227,6 +374,23 @@ if (BX_SER_THIS s[i].output) BX_SER_THIS s[i].io_mode = BX_SER_MODE_FILE; } + } else if (!strcmp(mode, "pipe")) { + if (strlen(bx_options.com[i].Odev->getptr ()) > 0) { + TPipe *pIo = new TPipe(bx_options.com[i].Odev->getptr ()); + if (!pIo->Ok()) { + // It's not an error if the connection to the pipe failed + // because the listening process might be available only + // when a developer wants to debug something without changing + // the configuration file + BX_INFO(("open of com%d (%s) failed\n", i+1, bx_options.com[i].Odev->getptr ())); + // Delete TPipe class to avoid a memory leak + delete pIo; + } else { + BX_DEBUG(("com%d pipe: %s", i+1, bx_options.com[i].Odev->getptr ())); + BX_SER_THIS s[i].io_mode = BX_SER_MODE_PIPE; + BX_SER_THIS s[i].io_handler = pIo; + } + } } else if (!strcmp(mode, "term")) { #if defined(SERIAL_ENABLE) if (strlen(bx_options.com[i].Odev->getptr ()) > 0) { @@ -1031,6 +1195,17 @@ fputc(BX_SER_THIS s[port].tsrbuffer, BX_SER_THIS s[port].output); fflush(BX_SER_THIS s[port].output); break; + case BX_SER_MODE_PIPE: + { + // Get the redirection target object. + TPipe *pIo = (TPipe*) BX_SER_THIS s[port].io_handler; + BX_DEBUG(("com%d: write: '%c'", port+1, BX_SER_THIS s[port].tsrbuffer)); + // The status of this object must be Ok()==true so there's no need + // to test for this status explicitly. + pIo->Write((bx_ptr_t) & BX_SER_THIS s[port].tsrbuffer, 1); + pIo->Flush(); + } + break; case BX_SER_MODE_TERM: #if defined(SERIAL_ENABLE) BX_DEBUG(("com%d: write: '%c'", port+1, BX_SER_THIS s[port].tsrbuffer)); @@ -1178,6 +1353,18 @@ } #endif break; + case BX_SER_MODE_PIPE: + { + // Get the redirection target object. + TPipe *pIo = (TPipe*) BX_SER_THIS s[port].io_handler; + // The status of this object must be Ok()==true so there's no need + // to test for this status explicitly. + if (pIo->Read(&chbuf, 1)==1) { + BX_DEBUG(("com%d: read: '%c'", port+1, chbuf)); + data_ready = 1; + } + } + break; case BX_SER_MODE_MOUSE: if (BX_SER_THIS mouse_internal_buffer.num_elements > 0) { chbuf = BX_SER_THIS mouse_internal_buffer.buffer[BX_SER_THIS mouse_internal_buffer.head]; Index: iodev/serial.h =================================================================== RCS file: /cvsroot/bochs/bochs/iodev/serial.h,v retrieving revision 1.25 diff -u -r1.25 serial.h --- iodev/serial.h 5 Dec 2004 20:23:39 -0000 1.25 +++ iodev/serial.h 3 May 2005 00:41:34 -0000 @@ -67,6 +67,7 @@ #define BX_SER_MODE_TERM 2 #define BX_SER_MODE_RAW 3 #define BX_SER_MODE_MOUSE 4 +#define BX_SER_MODE_PIPE 5 enum { BX_SER_INT_IER, @@ -110,6 +111,7 @@ int io_mode; int tty_id; FILE *output; + void *io_handler; #if USE_RAW_SERIAL serial_raw* raw; // // pipetunnel.cpp // // Martin Fuchs, 30.11.2003 // // // Invoke as: "pipetunnel [pipe_name]", // for example: "pipetunnel com_2" // // Then start up RectOS in VMWare, wait for the serial connect. // After that you can connect GDB using the command "target remote :9999". // #define _WIN32_WINNT 0x0501 #define WIN32_LEAN_AND_MEAN #include <windows.h> #include <winsock.h> #ifdef _MSC_VER #pragma comment(lib, "wsock32") #endif #include <stdio.h> #include <stddef.h> #include <string.h> #include <assert.h> #include <errno.h> #include <queue> #include <vector> // This definition currently missing in MinGW. #ifndef FILE_FLAG_FIRST_PIPE_INSTANCE #define FILE_FLAG_FIRST_PIPE_INSTANCE 0x00080000 #endif void dumphex(const char *buf, int len, int pos) { int i, j; for(j = 0; j < len; j += 16) { for(i = 0; i < 16; i++) { if(i + j < len) printf("%02x%c", (unsigned char)buf[i + j], j + i + 1 == pos ? '*' : ' '); else printf(" "); } for(i = 0; i < 16; i++) { if(i + j < len) printf("%c", buf[i + j] >= ' ' ? buf[i + j] : '.'); else printf(" "); } printf("\n"); } } class WinSockInit { private: WSADATA wsa_data; bool bIsOK; public: WinSockInit(void) { bIsOK = WSAStartup(MAKEWORD(2,2), &wsa_data)==0; } ~WinSockInit(void) { if (bIsOK) WSACleanup(); } bool Ok(void) const { return bIsOK; } }; class TSocketServer { private: SOCKET sckServer; SOCKADDR_IN addr; SOCKET sckClient; public: TSocketServer(void) { memset(&addr, 0, sizeof(addr)); addr.sin_family = AF_INET; addr.sin_addr.s_addr = htonl(INADDR_ANY); addr.sin_port = htons(9999); sckClient = INVALID_SOCKET; sckServer = socket(PF_INET, SOCK_STREAM, 0); if (sckServer==INVALID_SOCKET) return; u_long nbio_value = 1; ioctlsocket(sckServer, FIONBIO, &nbio_value); if (bind(sckServer, (struct sockaddr*) &addr, sizeof(addr))!=0) { sckServer = INVALID_SOCKET; return; } if (listen(sckServer, 4)!=0) { sckServer = INVALID_SOCKET; return; } } bool Ok(void) const { return (sckServer!=INVALID_SOCKET); } bool IsConnected(void) { if (sckClient==INVALID_SOCKET) { sckClient = accept(sckServer, NULL, NULL); if (sckClient!=INVALID_SOCKET) { BOOL bValue = TRUE; setsockopt(sckClient, IPPROTO_TCP, TCP_NODELAY, (const char *) &bValue, sizeof(BOOL)); } } return sckClient!=INVALID_SOCKET; } size_t Write(const void *pcv,size_t size) { assert(sckClient!=INVALID_SOCKET); int count = send(sckClient, (const char *) pcv, (int) size, 0); if (count==SOCKET_ERROR) { sckClient = INVALID_SOCKET; return 0; } return (size_t) count; } size_t Read(void *pv,size_t size) { assert(sckClient!=INVALID_SOCKET); int count = recv(sckClient, (char*) pv, (int) size, 0); if (count==SOCKET_ERROR) { int iErrCode = WSAGetLastError(); switch (iErrCode) { case WSAEWOULDBLOCK: break; case WSAECONNRESET: sckClient = INVALID_SOCKET; break; } return 0; } else if (count==0) { sckClient = INVALID_SOCKET; } return (size_t) count; } }; class TPipeServer { private: HANDLE hPipe; char *pszPipeName; private: bool Listen(void) { assert(hPipe!=INVALID_HANDLE_VALUE); if (ConnectNamedPipe(hPipe, NULL)!=0) return true; if (GetLastError()==ERROR_PIPE_LISTENING) return true; return false; } public: TPipeServer(const char *pszPipeName_arg=NULL) { pszPipeName = NULL; hPipe = INVALID_HANDLE_VALUE; Open(pszPipeName_arg); } ~TPipeServer(void) { Close(); if (pszPipeName!=NULL) { free(pszPipeName); } } bool Open(const char *pszPipeName_arg=NULL) { if (hPipe!=INVALID_HANDLE_VALUE) { Close(); } if (pszPipeName_arg!=NULL) { if (pszPipeName!=NULL) { free(pszPipeName); } pszPipeName = strdup(pszPipeName_arg); } if (pszPipeName==NULL) return false; hPipe = CreateNamedPipeA(pszPipeName, PIPE_ACCESS_DUPLEX | FILE_FLAG_FIRST_PIPE_INSTANCE | FILE_FLAG_WRITE_THROUGH, PIPE_TYPE_BYTE | PIPE_READMODE_BYTE | PIPE_NOWAIT, 1, 4096, 4096, 0, NULL); if (hPipe!=INVALID_HANDLE_VALUE) { if (!Listen()) { Close(); } } return true; } void Close(void) { if (hPipe!=INVALID_HANDLE_VALUE) { Flush(); DisconnectNamedPipe(hPipe); CloseHandle(hPipe); hPipe = INVALID_HANDLE_VALUE; } } bool Ok(void) const { return hPipe!=INVALID_HANDLE_VALUE; } bool IsConnected(void) { assert(hPipe!=INVALID_HANDLE_VALUE); if (ConnectNamedPipe(hPipe, NULL)!=0) return false; DWORD dwErrCode = GetLastError(); if (dwErrCode==ERROR_PIPE_CONNECTED) return true; if (dwErrCode==ERROR_NO_DATA) { // reopen Open(); } return false; } size_t Write(const void *pcv,size_t size) { assert(hPipe!=INVALID_HANDLE_VALUE); DWORD dwWritten = 0; if (!WriteFile(hPipe, pcv, (DWORD) size, &dwWritten, NULL)) return 0; return (size_t) dwWritten; } size_t Read(void *pv,size_t size) { assert(hPipe!=INVALID_HANDLE_VALUE); DWORD dwRead = 0; if (!ReadFile(hPipe, pv, (DWORD) size, &dwRead, NULL)) return 0; return (size_t) dwRead; } bool Flush(void) { assert(hPipe!=INVALID_HANDLE_VALUE); return FlushFileBuffers(hPipe)!=0; } }; typedef std::vector<unsigned char> TByteBuffer; typedef std::queue<TByteBuffer> TBufferQueue; int main(int argc, char** argv) { // initialize winsock WinSockInit SocketInit; if (!SocketInit.Ok()) { fprintf(stderr, "Failed to initialize sockets\n"); return 1; } // Get name of named pipe char path[MAX_PATH]; const char* pipe_name; if (argc > 1) { pipe_name = *++argv; } else { pipe_name = "com_2"; } sprintf(path, "\\\\.\\pipe\\%s", pipe_name); // increment priority to be faster than the cpu eating VMWare process SetPriorityClass(GetCurrentProcess(), HIGH_PRIORITY_CLASS); SetThreadPriority(GetCurrentThread(), THREAD_PRIORITY_HIGHEST); TPipeServer PipeServer(path); if (!PipeServer.Ok()) { fprintf(stderr, "Failed to create pipe %s\n", path); return 1; } TSocketServer SocketServer; if (!SocketServer.Ok()) { fprintf(stderr, "Failed to create socket\n"); return 1; } bool bSocketWasOpen = false; bool bPipeWasOpen = false; TBufferQueue PipeReadQueue; TBufferQueue SocketReadQueue; TByteBuffer ReadBuffer(4096); for (;;) { if (PipeServer.IsConnected()) { if (!bPipeWasOpen) { printf("Pipe connected\n"); bPipeWasOpen = true; } size_t BufSize = PipeServer.Read(&*ReadBuffer.begin(), ReadBuffer.size()); if (BufSize!=0) { // send to socket TByteBuffer TempBuffer(BufSize); memcpy(&*TempBuffer.begin(), &*ReadBuffer.begin(), BufSize); printf("PIPE >>\n"); dumphex((const char *) &*TempBuffer.begin(), TempBuffer.size(), 0); PipeReadQueue.push(TempBuffer); } if (SocketReadQueue.size()!=0) { const TByteBuffer &TempBuffer(SocketReadQueue.front()); printf("PIPE <<\n"); dumphex((const char *) &*TempBuffer.begin(), TempBuffer.size(), 0); PipeServer.Write(&*TempBuffer.begin(), TempBuffer.size()); SocketReadQueue.pop(); } } else { if (bPipeWasOpen) { printf("Pipe disconnected\n"); // reopen pipe PipeServer.Open(); bPipeWasOpen = false; PipeReadQueue = TBufferQueue(); } } if (SocketServer.IsConnected()) { if (!bSocketWasOpen) { printf("TCP connection established\n"); bSocketWasOpen = true; } size_t BufSize = SocketServer.Read(&*ReadBuffer.begin(), ReadBuffer.size()); if (BufSize!=0) { // send to socket TByteBuffer TempBuffer(BufSize); memcpy(&*TempBuffer.begin(), &*ReadBuffer.begin(), BufSize); printf("TCP >>\n"); dumphex((const char *) &*TempBuffer.begin(), TempBuffer.size(), 0); SocketReadQueue.push(TempBuffer); } if (PipeReadQueue.size()!=0) { const TByteBuffer &TempBuffer(PipeReadQueue.front()); printf("TCP <<\n"); dumphex((const char *) &*TempBuffer.begin(), TempBuffer.size(), 0); SocketServer.Write(&*TempBuffer.begin(), TempBuffer.size()); PipeReadQueue.pop(); } } else { if (bSocketWasOpen) { printf("TCP connection closed\n"); bSocketWasOpen = false; SocketReadQueue = TBufferQueue(); } } if (!bPipeWasOpen && !bSocketWasOpen) { Sleep(50); } else { Sleep(1); } } return 0; }

20 years, 3 months

memory leakage

by Hartmut Birr

Hi, we have somewhere a leakage which doesn't free unicode strings in kernel mode. After a short time of compiling ros on ros, I see many blocks with the tag USTR: ... Tag 52545355 (USTR) Blocks 714270 Total Size 53114920 Average Size 74 ... TotalBlocks 758557 TotalSize 66314608 AverageSize 87 Freeblocks 21427 TotalFreeSize 323088 AverageFreeSize 15 - Hartmut

20 years, 3 months

Re: [ros-svn] [Phreak] 14886: Renamed another makefile in the xmlbuildsystem branch that had the wrong case

by Steven Edwards

Hi, --- Phreak(a)svn.reactos.com wrote: > Renamed another makefile in the xmlbuildsystem branch that had the wrong case I don't really care which case we use but if its going to be upper case Makefile can you add it somewhere on the Wiki? Also maybe note about MixEd CaSeFileNames in SoUrce FIles. Thanks Steven __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

20 years, 3 months

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