Re: src/winsup/cygwin ChangeLog dcrt0.cc
On 27.01.2012 10:46, Václav Zeman wrote: +#define is_dos_path(s) (isdrive(s) \ + || ((s)[0] == '\\' \ +(s)[1] == '\\' \ +isalpha ((s)[2]) \ +strchr ((s) + 3, '\\'))) Is it safe to access 2nd, 3rd and 4th characters of the word without checking the length first? Yes, if s is a NUL-terminated string. If s[0] != '\0' (because, e.g., s[0] == '\\') then s[1] is valid; if s[1] != '\0' then s[2] is valid, and so on. Chris -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: [patch/rebase] Improve peflags
- fprintf (f, Copyright (c) 2009 Charles Wilson, Dave Korn, Jason Tishler\n); + fprintf (f, Copyright (c) 2009, 2011 Charles Wilson, Dave Korn, Jason Tishler\n); May I suggest if you are updating the copyright notice you also add et al. to the list of names (since you are now a contributor too). Regards, Chris
Re: procps aborts on Cygwin 1.7.5
Hello Bryan, On Wed, 14 Apr 2010 10:13:30 -0500, Thrall, Bryan wrote: thr...@pc1163-8413-xp ~/visual $ procps Signal 6 (ABRT) caught by ps (procps version 3.2.7). Please send bug reports to feedback at lists.sf.net or albert at users.sf.net thr...@pc1163-8413-xp ~/visual I believe this was working as of Cygwin 1.7.3. I am ignoring the request to send bug reports to sf.net because this appears to be due to the Cygwin change; if that's not a good enough reason, let me know and I'll forward this report to albert. Please try the suggestions you've already received and if the problem persists send the gzip'ed output of: strace procps Regards, Chris January (procps maintainer) -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Incorrect permissions on mount point when user does not have permission to access mounted directory
Hello, I have drive I: mounted at /backup but the ACLs deny access to anyone not in the Administrators group. I am in the Administrators group, but running Windows 7 so I can only access the drive if I am running with elevated privileges. If I run Cygwin without elevated privileges and type ls -l /backup, Cygwin shows the following: $ ls -l /backup -rw-r--r-- 1 Chris None 0 2006-12-01 00:00 /backup i.e. Cygwin incorrectly reports the mount point is a regular file and not a directory. $ mount C:/cygwin/bin on /usr/bin type ntfs (binary,auto) C:/cygwin/lib on /usr/lib type ntfs (binary,auto) C:/cygwin on / type ntfs (binary,auto) I: on /backup type unknown (binary,noacl) C: on /cygdrive/c type ntfs (binary,posix=0,user,noumount,auto) $ uname -a CYGWIN_NT-6.1 bonsai 1.7.0(0.218/5/3) 2009-12-04 17:08 i686 Cygwin Chris -- http://www.atomice.com -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Did I missing the official announcement that 1.7 went official?
On Mon, 21 Dec 2009 22:42:39 -0500, Larry Hall (Cygwin) reply-to-list-only...@cygwin.com wrote: On 12/21/2009 10:41 PM, Paul McFerrin wrote: Doing my bi-weekly update of my 1.7 installation, I noticed that it download my whole installation again and took at least an hour to perform at RR speeds. I was in the hospital for a while so I was out of it for a while when I deleted a bunch of old mail messageg but I looked first. My cygwin1.dll is dated: -rwxr-xr-x 1 paul root 2477884 2009-12-04 11:10 /bin/cygwin1.dll and the sub-sersion is 68. Did I catch it at the wrong time? 1.7 isn't released yet. cygwin.com confirms that. What about Netcraft? Chris -- http://www.atomice.com -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: CYGWIN_NT-5.1 1.7.0(0.214/5/3) ps (cygwin) 1.11 - how do I display the arguments to the processes?
On Wed, 18 Nov 2009 04:00:21 -0800 (PST), Larry W. Virden wrote: A developer has asked me to see if there is a way in Cygwin to get output more similar to Solaris or other Unix system ps commands. He specifically needs to get information about a process, its pid, and the arguments passed to the process. When I try the various flags for ps, I don't see any which displays the arguments. Install the procps package and then use the procps command. Chris -- http://www.atomice.com -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Avoiding the final setup.exe page
On Tue, 22 Sep 2009 01:45:07 -0400, ABCD en.a...@gmail.com wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Christopher Faylor wrote: Also, as far as I can tell, there is no remembering of anything going on now. The buttons are off by default. Is that right or am I missing something in the complicated setup code? - From what I've seen, the buttons are *on* by default, and only default to *off* if the shortcut already exists. If the shortcut is removed, then they turn back on (to my annoyance). Yes, that's the behaviour I see. Just verified with setup-1.7.exe form the Cygwin website. Setup checks the entries for any icons that don't already exist. So if you uncheck the Create icon on Desktop the first time you run setup, then the next time it will be checked again because the icon doesn't exist. If you don't want a desktop icon you must uncheck it every time you run setup. Regards, Chris -- http://www.atomice.com
Re: [Preliminary Patch] setup.exe size/position restore on startup
Hello Jonathon, On Wed, 13 May 2009 21:59:38 -0400, Jonathon Merz jonathon.m...@gmail.com wrote: Hello, Per Dave Korn's suggestion in: http://cygwin.com/ml/cygwin/2009-05/msg00208.html - If setup.exe exits while in a maximized state, it will be maximized on next startup, and recall it's last non-maximized size and position for restoration to normal state. You really ought to be using Get/SetWindowPlacement rather than GetWindowPos, IsZoomed etc. GetWindowPlacement works in workspace coordinates, whereas GetWindowPos works with screen coordinates so your code won't work properly if the user moves the taskbar or changes the monitor layout. Also GetWindowPlacement will get the 'restored' geometry of the window even when its maximised saving you a lot of code. Cheers, Chris
Re: graphics programming - beginner
Hello, On Wed, Feb 25, 2009 at 1:39 AM, H Le wrote: Hi, The previous message I asked was the path to graphics.h library but it seems to get more confusion. After searching for awhile, it seems that the graphics.h library is obsoleted. I try to compile some simple graphics program but unsure what libraries to #include and what compile parameters are needed. If someone shows me or has a link to someplace where it has some simple program, and step by step to compile it to run under cygwin, I appreciate it. I think this is what you're looking for: http://www.cs.colorado.edu/~main/bgi/install.html It's designed for use with MinGW, however, and not Cygwin. If you don't need Cygwin's Linux API emulation layer for your program I suggest you use MinGW itself, following the instructions on that website. You may be able to compile it from source to work on Cygwin. The source code is here: http://www.cs.colorado.edu/~main/bgi/source/ The makefile will need some changes to work. Regards, Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: dd in cygwin vs dd in linux - number of records is always different
On Mon, Feb 9, 2009 at 5:18 PM, Corinna Vinschen wrote: I guess I'll ask on one of the Microsoft newsgroups if this is a known effect and especially what I'm doing wrong or what I have to do to get the desired behaviour. Btw., if anybody has an idea what's going wrong, please speak up :} Maybe this excerpt from the CreateFile documentation applies? Note To read or write to the last few sectors of the volume, you must call DeviceIoControl and specify FSCTL_ALLOW_EXTENDED_DASD_IO. This signals the file system driver not to perform any I/O boundary checks on partition read or write calls. Instead, boundary checks are performed by the device driver. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Signal handling in Win32 GUI programs
On Tue, Jan 13, 2009 at 7:29 AM, Andy Koppe wrote: Actually I still can't quite get signal handling in MinTTY to work right. SIGINT is fine, but SIGTERM, SIGHUP, and SIGKILL don't seem to get to sigwait(), instead still invoking the default handler, i.e. terminating MinTTY without SIGHUP being sent to the command inside it. Anything obvious missing? Shouldn't pthread_sigmask be called for signal_thread as well? From the Linux manual page: In a multithreaded program, the signal should be blocked in all threads to prevent the signal being delivered to a thread other than the one calling sigwaitinfo() or sigtimedwait()). But maybe that doesn't apply to Cygwin. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: [Patch] Allow access to /proc/registry/HKEY_PERFORMANCE_DATA
On Fri, Dec 19, 2008 at 1:58 PM, Christian Franke wrote: (fhandler_registry::fill_filebuf): Use larger buffer to speed up access to HKEY_PERFORMANCE_DATA values. Remove check for possible subkey. Add RegCloseKey (). + /* RegQueryValueEx () opens HKEY_PERFORMANCE_DATA. */ + RegCloseKey (handle); I'm slightly puzzled by this change. handle is usually closed in fhandler_register::close. If you close it here then won't CloseHandle be called with an invalid handle in that method? Chris
Re: Kill cygwin process form task manager
On Nov 14, 2008 10:20am, Michael Wiedmann wrote: I need hints how to act if someone kills my Cygwin (parent) process using Windows task manager and I want to kill all forked child processes. I already keep a list of all childs pids and can kill them successfully in an SIGINT handler (if the programm is started in foreground and is interrupted e.g. by Ctrl-C). Untested on Cygwin, but works on Linux, etc.: 1. Create an anonymous pipe using the pipe function. 2. Fork the child. 3. Close the read fd in the parent and the write fd in the child. 4. Have child select on the read fd. 5. When the parent dies the select will return. 6. Try to read from the pipe. If read returns 0, i.e. end-of-file the parent is dead. Alternatively, do it the other way round, with the child occasionally writing to the pipe and the parent draining it. The child will get SIGPIPE when the parent dies. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Rsync and NTFS - FAT32 problem
On Sat, Sep 13, 2008 at 8:01 PM, Richard Ivarson wrote: Hello, I use Rsync successully to sync, for example, a NTFS formatted WinXP PC (where Cygwin's Rsync runs) with a Linux PC, and to sync the Win-PC with a n GByte USB memory stick. When the memory stick is formatted to NTFS all works fine (exact command see below please). But since I formatted the memory stick to FAT32 because it's more usuable on Linux PCs, suddenly the very same Rsync command shows different results! It copies many files from the Win-PC to the USB stick which already are up to date... I couldn't figure out what these files have in common... or why Rsync would get wrong information from the FAT32 filesystem... In the end I tried the Rsync parameter --checksum which solved the problem, but slows down the whole thing a lot. Then I found the --update parameter which does the trick, too. OK, so it prints many directories which --checksum did not print, but ... it's rather optical. The command is : * rsync --delete --update --relative --recursive --times --progress Sourcefolder/ /cygdrive/u/Destinationfolder Using parameters like --owner or -permission didn't help. Does anybody know what is the problem with FAT32 (and Rsync using it) ? Thanks. FAT32 only has 2 second file modification time granularity. You need to pass the --modify-window option to rsync. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: /proc/loadavg
On 27/02/2008, Jeff Fulmer wrote: I installed procps in the hopes of getting 1,5,15 minute load averages. I have them, but the values to meet my expectations. No matter how much load the system is under, the values are always 0.0 0.0 0.0 Are load averages supported? Is there something I need to do in order to set them up? From the source: /* * not really supported - Windows doesn't keep track of these values * Windows 95/98/me does have the KERNEL/CPUUsage performance counter * which is similar. */ To my knowledge there are no kernel counters on Windows NT/2000/XP/Vista from which we could get these values. Regards, Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: ps executable does not appear to match source
On 23/01/2008, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: That said, however, the other way of dealing with this is to modify procps to deal with Windows pids. Then we wouldn't need the cygwin ps. If you want to provide a patch to do that, then it's likely that the procps maintainer would accept it -- assuming that it isn't so intrusive as to cause an ongoing maintenance problem. I would rather see a patch that added Windows pids to /proc than only to procps. Then the functionality would be available to other programs, like top. If procps can be made to do all of the things that ps now does then there would be no reason to keep ps around. I am interested. However, I would want to ensure from the beginning the it is possible to achieve. Would Cygwin accept a ps that did not produce identically formatted output for each option of the historically older version? What about all those people who have crafted their shell or Perl or Python code to interpret the output of the historically older version? To support scripts that rely on the format and options of the old Cygwin ps we could add a new 'Cygwin' personality to procps. Cheers, Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: ps executable does not appear to match source
On 23/01/2008, Corinna Vinschen [EMAIL PROTECTED] wrote: On Jan 23 09:34, Chris January wrote: I would rather see a patch that added Windows pids to /proc than only to procps. Then the functionality would be available to other programs, like top. Care to implement it? You basically wrote the /proc stuff in Cygwin anyway. Possibly. I'll look into it in a couple of weeks time. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Installation with root privilage: can't download Intel Fortran F90 compiler
It won't work because Cygwin is not a Linux emulator. There are no recent Linux emulators/translation layers for Windows and even if you did use one the compiler would not generate Windows binaries. You need a Windows version of the Intel Fortran compiler or you could use a virtual machine (e.g. VirtualBox) / Co-operative Linux to run a full Linux session under Windows. How to do any of this is off-topic for this mailing list. On 18/01/07, K. Basu [EMAIL PROTECTED] wrote: The problem with running filename.sh was fixed with d2u - thanks! But there's another problem now! I am trying to download the Intel Fortran F90 compiler (which is originally intended for Linux) to work on windows with Cygwin. It requires installing as 'root'. I tried to install it from the windows Administrator account, but it didn't work. Would you know how to fix this? Thanks, KB On 1/17/07, Dave Korn [EMAIL PROTECTED] wrote: On 17 January 2007 20:06, Kasturi Basu wrote: Hi Dave, Hi KB, http://cygwin.com/acronyms#PPIOSPE please! Haven't used d2u before. A google search revealed several versions: http://linux.maruhn.com/sec/d2u.html Would you recommend any particular one? The one that comes as part of cygwin! If you don't already have it, use setup.exe to install the cygutils package. cheers, DaveK -- Can't think of a witty .sigline today -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
New procps-3.2.7 for upload
Hello, I have finally packaged a new version of procps that fixes the incorrect page size bug. http://www.avocado.plus.com/procps-3.2.7/procps-3.2.7-1-bin.tar.bz2 http://www.avocado.plus.com/procps-3.2.7/procps-3.2.7-1-src.tar.bz2 Cheers, Chris
Re: Error in 'cat /proc/uptime' output
Could you include full cygcheck information please. On 21/11/06, Rajesh Tiwari [EMAIL PROTECTED] wrote: Hi All, I'm using cygwin 1.5.9(0.112/4/2). I'm facing this issue: 'cat /proc/uptime' gives output as- $ cat /proc/uptime 380224.81 0.00 and this output leads to 100% CPU usage. I searched for any patches for this issue but couldn't succeed. Please advise if there are any patches for the above issue. Thanks and Regards! Rajesh -**Nihilent*** *** All information contained in this communication is confidential, proprietary, privileged and is intended for the addressees only. If you have received this E-mail in error please notify mail administrator by telephone on +91-20-39846100 or E-mail the sender by replying to this message, and then delete this E-mail and other copies of it from your computer system. Any unauthorized dissemination, publication, transfer or use of the contents of this communication, with or without modifications is punishable under the relevant law. Nihilent has scanned this mail with current virus checking technologies. However, Nihilent makes no representations or warranties to the effect that this communication is virus-free. Nihilent reserves the right to monitor all E-mail communications through its Corporate Network. *** *- -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Reboot vs. Restart Windows
On 30/10/06, Andrew DeFaria [EMAIL PROTECTED] wrote: Mike Maxwell wrote: That's not what I said, go back and re-read. Wait, I'll save you the trouble: I said that 99% of the words we know--not 99% of the people who know words--are our definitions that we infer from usage, rather than from looking them up. The second thing that shows me that you can't read, is that I also did not suggest changing terminology. I suggested changing a message. And the change is away from a non-standard usage (in the Windows world) to a standard usage (restart Windows). As for my guesstimate, I am a linguist, and it is standard knowledge among linguists that most of the vocabulary we use (in our first language--second language learning is often different) is not from looking definitions up in dictionaries. It's still a number you've pulled from your ask thus it stinks. Listen dude - I am often in the Windows world and standard usage in the Windows world is reboot - trust me! From http://support.microsoft.com/default.aspx?scid=%2Fsupport%2Fglossary%2FR.asp: reboot vb. To restart a computer by reloading the operating system. See also boot2, cold boot, warm boot. :) Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Top reports wrong memoryuse
It would be possible to change /proc/*/stat{m} to use the page size of 64K, but this would make the values in those files imprecise. It might be better just to change top. I'll see if I can patch top to get these right. IIRC there is a constant defined somewhere with the correct page size. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: chown chown on /proc/registry/HKEY_PERFORMANCE_DATA
On 01/06/06, Lester Ingber [EMAIL PROTECTED] wrote: I was looking over my /proc directory @lester:/% ls -ld proc dr-xr-xr-x 11 ingber None 0 May 31 2006 proc/ and found @lester:/proc/registry% ls -la total 0 dr-xr-xr-x9 ingber None 0 May 31 22:30 ./ dr-xr-xr-x 11 ingber None 0 May 31 22:30 ../ dr-xr-x--- 8418 Administrators SYSTEM 0 May 18 10:05 HKEY_CLASSES_ROOT/ dr-xr-x---4 Administrators SYSTEM 0 Sep 11 2004 HKEY_CURRENT_CONFIG/ dr-xr-x--- 22 Administrators SYSTEM 0 May 31 07:19 HKEY_CURRENT_USER/ dr-xr-xr-x1 ingber None 0 May 31 22:30 HKEY_DYN_DATA/ dr-xr-xr--7 Administrators SYSTEM 0 May 31 07:14 HKEY_LOCAL_MACHINE/ d-2 0 Dec 31 1969 HKEY_PERFORMANCE_DATA/ dr-xr-xr-- 10 Administrators SYSTEM 0 May 31 18:10 HKEY_USERS/ I do not see how to change the combination of owner and permissions (and date?) to access the info under HKEY_PERFORMANCE_DATA. HKEY_PERFORMANCE_DATA is not a normal registry key and needs to be queried in a special way but Cygwin doesn't support this at present. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Windows 95 support ?
On 20/04/06, Christopher Faylor why-does-gmail-insist-on-quoting-addresses wrote: On Thu, Apr 20, 2006 at 10:42:14PM +0200, Samuel Thibault wrote: I tried to install cygwin on win95 this afternoon, but setup.exe can't start, because it is linked with MSVCRT.DLL, which is not available under windows 95. MSVCRT.DLL or its variant has been available since Windows 3.1, AFAIK. I'm sure you can find it somewhere. MSVCRT.DLL only comes as part of the base operating system install with Windows 95 OSR 2 or higher. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Windows 95 support ?
On 21/04/06, Samuel Thibault no-no-no wrote: Chris January, le Fri 21 Apr 2006 08:51:07 +0100, a écrit : On 20/04/06, Christopher Faylor why-does-gmail-insist-on-quoting-addresses wrote: On Thu, Apr 20, 2006 at 10:42:14PM +0200, Samuel Thibault wrote: I tried to install cygwin on win95 this afternoon, but setup.exe can't start, because it is linked with MSVCRT.DLL, which is not available under windows 95. MSVCRT.DLL or its variant has been available since Windows 3.1, AFAIK. I'm sure you can find it somewhere. MSVCRT.DLL only comes as part of the base operating system install with Windows 95 OSR 2 or higher. Mmm, even a fresh install of OSR2 didn't bring me it. My mistake. It's actually a part of Windows 95 OSR2.5 which includes IE4. http://support.microsoft.com/kb/q175430/ Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
New procps 3.2.6 release
Hello, I have packaged a new version of procps that fixes the bug that Corinna found in top and also tracks the latest upstream version. I have tweaked setup.hint slightly to have a shorter 'long' description. http://www.avocado.plus.com/procps-3.2.6/ Cheers, Chris
Re: New procps 3.2.6 release
On Apr 4 13:56, Chris January wrote: Hello, I have packaged a new version of procps that fixes the bug that Corinna found in top and also tracks the latest upstream version. I have tweaked setup.hint slightly to have a shorter 'long' description. http://www.avocado.plus.com/procps-3.2.6/ Can you please mail the full paths to the files to upload? http://www.avocado.plus.com/procps-3.2.6/procps-3.2.6-1.tar.bz2 http://www.avocado.plus.com/procps-3.2.6/procps-3.2.6-1-src.tar.bz2 http://www.avocado.plus.com/procps-3.2.6/setup.hint Cheers, Chris
Re: nice under tcsh does not register under top? [Attn top maintainer]
According to Lester Ingber on 4/4/2006 7:36 AM: I have a tcsh script that runs some processes, e.g., `nice +19 gmake run`. I monitor the NI and %CPU columns under `top`. This works fine under FreeBSD and Solaris/SPARC, but under Cygwin the NI column always reads 0? Is it not possible to affect Windows priorities via Cygwin? It's possible, since cygwin 1.5.13 or so (for example, '/bin/nice /bin/nice' outputs 10, since the first nice defaults the second to +10, and the second displays its current nice value with no argument). In tcsh, nice is a shell builtin, which defaults to +4 instead of +10, but my testing shows that it works. It looks like top is not displaying nice values properly. This isn't a problem with top but a missing feature of the /proc filesystem. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: 1.5.19(0.150/4/2): missing device in /proc/partition after removing USB disk
On 08/03/06, Loh, Joe wrote: Then we proceed to removing the first USB disk, i.e. /dev/sdb, using the the Safe Removal method from Windows. Once that operation completes we cat the /proc/partitions again. This time, the only disk shown is /dev/sda, which is the system disk. The second USB disk was never removed. The second dd on /dev/sdc proves that the disk is still accessible eventhough /proc/partitions never shows it. The sequence of event is captured in the following text. We are also attaching the cygcheck output as well. This is a bug. The format_proc_partitions assumes that drives are numbered consecutively - which is not the case if you hot remove a drive. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: 1st summary (was Re: [HEADSUP] ALL Maintainers, please reply.)
procps I'm still maintaining this. (I've been having problems sending to the list.) Chris
Re: suggestions for cygwin developers
Alex Goldman wrote: When Cygwin gets set up, it would be more user-friendly if it placed two icons on the desktop: one should start maximized Rxvt; another should start X with a couple of xterms or whatever. First-time users might think that the MS-DOS terminal is as good as it gets, and this is not good for Cygwin. Others still have to figure out how to start Rxvt automatically and how to configure it to look pretty. Also, it would be neat to be able to keep Cygwin up-to-date automatically. How about a prompt when a new Cygwin DLL version is available: A new version of Cygwin is available. Run setup to install the latest version. This could be included as part of the /etc/profile or something. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Blind people using setup.exe?
Brian Dessent wrote: Igor Pechtchanski wrote: Oh, that reminds me, --no-md5 is broken, I just noticed the other day - gotta look at that too. It's not broken, it doesn't exist any more. Hehe. While we're at it, would you like to make setup print out a list of its command-line options, either in a message box or by doing an AllocConsole()? I know the latter approach will pop up a console box if setup is started from a pty or via a shortcut (which will go away once the output is done), but it's better than nothing, and should work just fine if setup is started from a console shell. I think having a spurious console appear and then disappear is not very friendly. I have considered having it do a messagebox with the --help output, which I guess is better than what we have now. You can use AttachConsole (ATTACH_PARENT_PROCESS) to attach to the parent process console. If the parent process does not have a console the call will fail and you can popup a message box instead. IMHO this is the most user-friendly way of doing things. Chris
Re: Blind people using setup.exe?
Dave Korn wrote: Original Message From: Chris January Sent: 28 June 2005 12:06 You can use AttachConsole (ATTACH_PARENT_PROCESS) to attach to the parent process console. If the parent process does not have a console the call will fail and you can popup a message box instead. IMHO this is the most user-friendly way of doing things. IMHO you should have finished reading the thread before sending this reply I agree. Sorry about that! Chris
Re: Possible Bug in /proc/partitions ??
Bengt-Arne Fjellner wrote: either /proc/prtitions has something wrong or i have. This is how it looks. $ cat /proc/partitions major minor #blocks name 8 0 19535040 sdaOK 816 78124095 sdbOK 817 56196 sdb1 OK 818 61978770 sdb2 OK 819514080 sdb3 OK 820 15575017 sdb4 extended no name?? 821 14546826 sdb5 found as sdb4 822 1020127 sdb6 extended no name?? 825 1020096 sdb9 found as sdb5 832 120624052 sdcOK 833 120624021 sdc1 OK 848 58613152 sddOK 849 58613121 sdd1 OK 865253984 sde1 extended no name?? 869 2062305 sde5 found as sde1 I dont think that the extended partitions should have a device?? and i find the other partitions as i have stated. Thanks for the bug report. I will look into this when I get the time. Incidentally, do you know what Linux's behaviour is in this regard, i.e. if you run the same command on Linux (if you have it installed), what is the result? Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Unicode in filenames support?
Jaeho Shin wrote: I'm having problem with accessing files that have Unicode in their filenames. 1. I use Windows XP Korean version (so the codepage must be 949?). 2. I use iTunes to listen to my music. 3. Files in iTunes Library have filenames in the following format: {Artist}/{Album}/{Track#} {Title}.mp3 where names inside braces are values from its ID3-tag. 4. Some of my mp3s have Japanese or Latin characters, e.g. (Latin small letter e with acute). In ID3-tags, those characters seem to be in UCS-2 encoding or so, but not in CP949 or EUC-KR. 5. I want to rsync those files to my other Linux machine. 6. But rsync complains some files (whose name contains such special/Unicode characters perhaps?) have vanished! :'( With Windows Explorer, I can copy them to a Samba share (with utf-8 encoding) without any problem. However, from the Cygwin environment, it seems that there is no way I can access those files. I tried the mount -o managed option which escapes capitals and other non-ascii characters in filenames. It wasn't a solution for me since iTunes (not Cygwin) mainly manages the files. Since I really want to use rsync, I hope Cygwin to be able to access Unicode filenames. It would be great if I could mount a filesystem with a charset or encoding specified. Is there any nice way already I can solve this problem? Some time ago I wrote a patch for Cygwin that converted Unicode files to UTF-8 and back. Maybe you can dig that up and see if you can get it working with the latest Cygwin code. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Problem with £ sign at Cygwin prompt.
[EMAIL PROTECTED] wrote: Hi all, I've noticed a minor issue while typing at the bash prompt in cmd.exe. It seems that when I try to type a £ sign, a hash character and newline is entered instead. I'm aware of the issues surrounding these two characters, but as far as I can tell all of my language settings in Windows are configured for the UK. Anyone else noticed this? Yes - me too now you come to mention it... That means either this is a new bug or I've never typed a £ before at a Cygwin bash prompt. I wonder which it is? Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
FW: new setup for testing
Oops - wrong list. Brian Dessent wrote: Gerrit P. Haase wrote: The column headers disappear partially when switching back from full screen view to normal view, see attached screenshot. It works at first, breaks after using the radio buttons one time, e.g. switch from 'curr' to 'exp' then toggle back from full screen view to default. Thanks for the report. I think this is related to another problem I'm aware of and need to write code to fix: when you resize the package chooser, the scroll info is not recalculated. It's only recalculated when you e.g. switch the view, or toggle the hide obsolete checkbox. Example: Maximize, then cycle the view to full. Now restore to original size without touching the view, and scroll all the way down. You can't get to the very bottom of the package list because the scroll info has not been recalculated. Likewise if you switch to full then maximize you can scroll down past the last line, for the same reason. I don't think the radio buttons should really be radio buttons at all because when you select one it changes the state of another widget: namely the package selection widget. It also collapses the tree if you have it expanded, which is another no-no. The radio buttons are actually commands and therefore should be command buttons (i.e. push buttons). If you select Keep, for example, the action for every package changes to Keep. I can, however, then change the action for specific packages by clicking the cycle icon beside the package name. The wizard page is now no longer in the Keep state so the radio buttons no longer reflect the current state of the page. Changing the current view, on the other hand, is not a command and therefore shouldn't be a command button but instead a cycle widget/drop-down list. So a revised set of controls might look like this: View: [Not Installed]v Set all packages to: |Current| |Previous| |Update| |Unstable| Where Current, Previous, Update and Unstable are command buttons are Not Installed is an item in a drop down list. The View option has moved to the left hand side since it is, IMHO, the most frequently used control and should therefore be as close to the top left corner for left-to-right languages as possible. I've used Current here to refer to the currently installed package and Update to refer to the latest stable version of a package available. This usage is consistent with the usage of Current in the package selection listview below, whereas the current usage is inconsistent. Also selecting any control (including Hide Obsolete Packages) resets the listview. This should not happen as the user may have spent some considerable time expanding the tree and arranging the columns to find the items they want to change only to have the tree collapsed again and the columns resized when they click another control. Regards, Chris
Re: new setup for testing
Brian Dessent wrote: Gerrit P. Haase wrote: The column headers disappear partially when switching back from full screen view to normal view, see attached screenshot. It works at first, breaks after using the radio buttons one time, e.g. switch from 'curr' to 'exp' then toggle back from full screen view to default. Thanks for the report. I think this is related to another problem I'm aware of and need to write code to fix: when you resize the package chooser, the scroll info is not recalculated. It's only recalculated when you e.g. switch the view, or toggle the hide obsolete checkbox. Example: Maximize, then cycle the view to full. Now restore to original size without touching the view, and scroll all the way down. You can't get to the very bottom of the package list because the scroll info has not been recalculated. Likewise if you switch to full then maximize you can scroll down past the last line, for the same reason. I don't think the radio buttons should really be radio buttons at all because when you select one it changes the state of another widget: namely the package selection widget. It also collapses the tree if you have it expanded, which is another no-no. The radio buttons are actually commands and therefore should be command buttons (i.e. push buttons). If you select Keep, for example, the action for every package changes to Keep. I can, however, then change the action for specific packages by clicking the cycle icon beside the package name. The wizard page is now no longer in the Keep state so the radio buttons no longer reflect the current state of the page. Changing the current view, on the other hand, is not a command and therefore shouldn't be a command button but instead a cycle widget/drop-down list. So a revised set of controls might look like this: View: [Not Installed]v Set all packages to: |Current| |Previous| |Update| |Unstable| Where Current, Previous, Update and Unstable are command buttons are Not Installed is an item in a drop down list. The View option has moved to the left hand side since it is, IMHO, the most frequently used control and should therefore be as close to the top left corner for left-to-right languages as possible. I've used Current here to refer to the currently installed package and Update to refer to the latest stable version of a package available. This usage is consistent with the usage of Current in the package selection listview below, whereas the current usage is inconsistent. Also selecting any control (including Hide Obsolete Packages) resets the listview. This should not happen as the user may have spent some considerable time expanding the tree and arranging the columns to find the items they want to change only to have the tree collapsed again and the columns resized when they click another control. Regards, Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: [ANNOUNCEMENT] New package: stow
[EMAIL PROTECTED] wrote: The stow package is now available in the Cygwin distribution. Stow is an installation manager for local software packages. It creates sets of symlinks from the installed location (e.g. /usr/local) to a stow directory (e.g. /usr/local/stow/emacs) where the real files live. This allows you to keep packages separate, while making them appear to be installed in the same place. Note that because stow uses symlinks to install files, it will probably only be effective for software that is used only in the Cygwin environment and doesn't install any DLLs. The reason is that Cygwin symlinks are implemented in the Windows file system as shortcuts (.lnk files), but Windows shortcuts are fundamentally broken: Windows will not, as a rule (the only exception being for GUI operations in Windows Explorer), interpret a shortcut as a pointer to a file. Cygwin corrects this and emulates the Unix behavior, but Windows programs won't, and PATH searches for DLLs won't follow shortcuts either. Does stow have support for hard links at all? If not is that an easy thing to add in? Such an option would make stow more useful on Cygwin, IMHO. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Hippo icon!
Ashwin N wrote: Hi, I noticed that the icon for Cygwin on the website has changed from the green-n-black C to a roaring hippo. http://www.cygwin.com/hippo.jpg It's to shake off the impression that Cygwin is bloated and slow. No, wait Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: /proc/self/exename - exe ? (Re: Retrieving name of executable)
Corinna Vinschen wrote: On Mar 28 23:36, Anthony Heading wrote: On Fri, Jan 14, 2005 at 02:35:28PM -0500, Christopher Faylor wrote: On Fri, Jan 14, 2005 at 10:39:23AM -0800, Earl Chew wrote: I think the name of the current executable is stored in myself-progname within cygwin1.dll. The more easily accessible __progname returns the basename of the executable. Is there a way for an application to obtain myself-program, other than resorting to raw win32 call to GetModuleFileName()? argv[0] or __argv[0] or /proc/self/exename Has this link been renamed? It appears to have changed to /proc/self/exe now, which has broken a lot of my programs... The old exename was a file containing the path, the new exe is a symlink pointing to the binary. This is how it's done on Linux. Maybe exename should be kept for backwards compatibility? Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: 1.5.13-1 : can't cd anywhere with non-admin account
François-David Collin wrote: Corinna, thanks you very much! the notraverse keyword did the trick. I don't really understand what do you mean by it checks the permissions of the parent directory... As long as I can traverse with windows acls in the windows world (and it is presently the case) I should be able to do the same with cygwin semantics, right? Windows has traverse checking off by default. This means a user can access their home directory, for example, without being given access to to the parent directory or the root of the drive. It's a security 'feature'. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: New procps release
On Feb 21 21:21, Chris January wrote: I've finally found time to update the procps package. The new version is based off procps 3.2.5 from procps.sourceforge.net. Download links: http://www.atomice.com/downloads/procps-3.2.5-1-bin.tar.bz2 http://www.atomice.com/downloads/procps-3.2.5-1-src.tar.bz2 The only Cygwin specific change in this version is greater tolerance for different clocks on SMP Windows. Slabinfo is included but does not work (no /proc/slabinfo file exists at present). Pmap requires cygserver. Sysctl is useless. Cygwin 1.5.13 is required (use current CVS for now). If Cygwin 1.5.13 is required, this should be marked as test release. Sure. Are you implying I need to send a setup.hint or is it good to be uploaded as-is? Chris
New procps release
I've finally found time to update the procps package. The new version is based off procps 3.2.5 from procps.sourceforge.net. Download links: http://www.atomice.com/downloads/procps-3.2.5-1-bin.tar.bz2 http://www.atomice.com/downloads/procps-3.2.5-1-src.tar.bz2 The only Cygwin specific change in this version is greater tolerance for different clocks on SMP Windows. Slabinfo is included but does not work (no /proc/slabinfo file exists at present). Pmap requires cygserver. Sysctl is useless. Cygwin 1.5.13 is required (use current CVS for now). Chris -- http://www.atomice.com
RE: [ANNOUNCEMENT] Updated (in test): coreutils-5.3.0-2
Dates can be entered via integer counts of seconds since 1970 when prefixed by `@'. For example, [EMAIL PROTECTED]' represents 1970-01-01 00:05:21 UTC. I think that example might be wrong... :) Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Multiple installations and 3PPs
Well, it was proposed once already, but nobody did the work: Get rid of the registry entries completly. Cygwin could use something similar to /etc/fstab to manage its mount modes. And how do you know where / is mounted? Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: EFS encrypted files ssh
Is it normal that during an SSH connection EFS-encrypted files are not accessible? Is it for the way the SSH token autentication is made? Yes, it probably is. I belive the user's private EFS is encrypted using their password hash. If the SSH token was generated without using a password (e.g. because you are using RSA authenitcation) then the EFS key can't be decrypted and used. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
PATCH: Replace spaces with tabs in /proc/pid/status.
2004-12-22 Chris January [EMAIL PROTECTED] * fhandler_process.cpp (format_process_status): Use tabs in formatting instead of spaces. Chris -- http://www.atomice.com fhandler_process_status_tabs.patch Description: Binary data
RE: why is -L/usr/local/lib necessary?
why doesn't gcc -lfoo (ld) find /usr/local/lib/foo.dll? what do I do to avoid this? Edit /usr/lib/gcc-lib/i686-pc-cygwin/3.3.3/specs Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: [Bug Cygwin Applications/575] Unknown HZ value! message from procps commands
Volker Zell writes: Chris January writes: In both cases what is the actual value of HZ you are seeing? 01:02 AM [555] w Unknown HZ value! (483) Assume 100. 08:34:05 up 11:33, 3 users, load average: 0.00, 0.00, 0.00 USER TTY FROM LOGIN@ IDLE JCPU PCPU WHAT vzelltty0 127.0.0.1:0.023:020.00s 0.00s ? - vzelltty1 127.0.0.1:0.023:020.00s 0.00s ? - vzelltty2 127.0.0.1:0.023:020.00s 0.00s ? - [EMAIL PROTECTED] /usr/src 08:34 AM [556] procps Unknown HZ value! (484) Assume 100. PID TTY TIME CMD 1864 ?00:00:08 bash 2036 ?00:00:01 procps [EMAIL PROTECTED] /usr/src 08:34 AM [557] vmstat.exe Unknown HZ value! (484) Assume 100. procs memoryswap io system cpu r b w swpd free buff cache si sobi bo incs us sy id 0 29 0 18112 615880 0 0 136 134 0 83 100 8 5 88 [EMAIL PROTECTED] /usr/src 08:35 AM [558] uptime Unknown HZ value! (484) Assume 100. 08:35:05 up 11:34, 3 users, load average: 0.00, 0.00, 0.00 The values seem to change after each reboot: 09:20 AM [498] vmstat Unknown HZ value! (438) Assume 100. procs memoryswap io system cpu r b w swpd free buff cache si sobibo in cs us sy id 0 36 0 19640 493428 0 0 10 132 02 66 4 4 92 [EMAIL PROTECTED] /tmp 08:15 AM [499] uptime Unknown HZ value! (438) Assume 100. 08:15:40 up 22:59, 3 users, load average: 0.00, 0.00, 0.00 [EMAIL PROTECTED] /tmp 08:15 AM [500] w Unknown HZ value! (438) Assume 100. 08:15:42 up 22:59, 3 users, load average: 0.00, 0.00, 0.00 USER TTY FROM LOGIN@ IDLE JCPU PCPU WHAT vzelltty0 127.0.0.1:0.0Mon090.00s 0.00s ? - vzelltty1 127.0.0.1:0.0Mon090.00s 0.00s ? - vzelltty2 127.0.0.1:0.0Mon090.00s 0.00s ? - I'm in the process of preparing a new procps release. Please try again with that when it's announced. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: ioctls.h not found
When I am compiling ZSNES in cygwin. asm/ioctls.h: No such file or directory Error. Please Help Me Did you sacrifice the goat? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: ls taking too long
If a directory contains a large number of files (I have 4) where most of them are named such that the first character is a 1 and you do an ls x* where only one of the files begin with x, the ls takes an inordinate amount of time, but going to a plain dos window and dir e* is really fast. The cmd/dos window's filename filtering is done in the kernel, whereas Cygwin's is done in user mode so every file is processed. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: [Bug Cygwin Applications/575] Unknown HZ value! message from procps commands
[please, let's keep the discussion on the list] dr dot volker dot zell at oracle dot com schrieb: --- Additional Comments From dr dot volker dot zell at oracle dot com 2004-12-04 16:40 --- (In reply to comment #3) This is known problem with multiple processors. Workaround: Disabeling hyperthreading or - in case you're on a real multiprocessor system - using the /NUMPROC=1 switch for booting up windows might have the desired effect. See the thread at http://cygwin.com/ml/cygwin/2004-03/msg01110.html I get this too, on a single processor machine with W2k SP4 Interesting. For which cmdline exactly? cygcheck -s -v -r cygcheck.out also please, since this is the first time I heard it on a SMP. In both cases what is the actual value of HZ you are seeing? Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: FDISK support
According to this post: http://sources.redhat.com/ml/cygwin/2002-11/msg01300.html , you can now use fdisk. Can someone please tell me how to get fdisk onto cygwin? I couldn't find the package with the cygwin setup file. Thanks everyone. There is no package available under setup because I didn't want to deal with people misusing fdisk and wiping their hard disk. You can compile fdisk yourself from sources (a few minor changes required). Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: ls /dev/*
* Christopher Faylor [EMAIL PROTECTED] [2004-11-02 15:01:13 -0500]: On Tue, Nov 02, 2004 at 02:55:39PM -0500, Sam Steingold wrote: why isn't /dev a more usual directory? cd /dev, ls /dev all fail, while cat /dev/clipboard works. No one has implemented the special handling required for /dev which would enable things like opendir/readdir or cd to work. Thanks, I guessed that much. I also know about PTC. (fhandler_proc.cc is too long, I guess fhandler_dev.cc would be just as long, and I suspect that fhandler_dev.cc is not the only this missing). Is this on anyone's TODO list? Actually it's not that difficult. I've already implemented it once. See this patch: http://www.cygwin.com/ml/cygwin-patches/2002-q2/msg00191.html It was due to be merged sometime around 1.3.12 but I think I and the maintainers forgot about it. Feel free to update the patch to latest CVS and re-submit it. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: ls /dev/*
* Chris January [EMAIL PROTECTED] [2004-11-03 09:08:44 +]: * Christopher Faylor [EMAIL PROTECTED] [2004-11-02 15:01:13 -0500]: On Tue, Nov 02, 2004 at 02:55:39PM -0500, Sam Steingold wrote: why isn't /dev a more usual directory? cd /dev, ls /dev all fail, while cat /dev/clipboard works. No one has implemented the special handling required for /dev which would enable things like opendir/readdir or cd to work. Thanks, I guessed that much. I also know about PTC. (fhandler_proc.cc is too long, I guess fhandler_dev.cc would be just as long, and I suspect that fhandler_dev.cc is not the only this missing). Is this on anyone's TODO list? Actually it's not that difficult. I've already implemented it once. See this patch: http://www.cygwin.com/ml/cygwin-patches/2002-q2/msg00191.html It was due to be merged sometime around 1.3.12 but I think I and the maintainers forgot about it. Feel free to update the patch to latest CVS and re-submit it. I am sure it would be much easier for you to update your own patch. Could you please do it? Oh of course - if you are willing to compensate me for my time... I don't need this feature and am not inclined to work on it voluntarily at the moment. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: ls /dev/*
On Wed, Nov 03, 2004 at 09:08:44AM -, Chris January wrote: * Christopher Faylor [EMAIL PROTECTED] [2004-11-02 15:01:13 -0500]: On Tue, Nov 02, 2004 at 02:55:39PM -0500, Sam Steingold wrote: why isn't /dev a more usual directory? cd /dev, ls /dev all fail, while cat /dev/clipboard works. No one has implemented the special handling required for /dev which would enable things like opendir/readdir or cd to work. Thanks, I guessed that much. I also know about PTC. (fhandler_proc.cc is too long, I guess fhandler_dev.cc would be just as long, and I suspect that fhandler_dev.cc is not the only this missing). Is this on anyone's TODO list? Actually it's not that difficult. I've already implemented it once. See this patch: http://www.cygwin.com/ml/cygwin-patches/2002-q2/msg00191.html It was due to be merged sometime around 1.3.12 but I think I and the maintainers forgot about it. Feel free to update the patch to latest CVS and re-submit it. Actually, please don't. I think you misinterpret the discussion in cygwin-developers. Now that you've reacquainted me with the discussion, I remember why it wasn't applied as-is. My plan was for /dev to go away as a special mount. Now that mknod works, this is more doable than it was in 2002. Ah yes - I remember now why it wasn't committed in the first place. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: cygwin 1.5.10: ImageMagick 6.0.3 binaries fail
Harold L Hunt II wrote: Okay, it is fixed now. There is a 6.0.4 version that is posted. The release notes explain the source of the problem (read: me). In addition, I installed jasper, lcms, and libfpx so support for these was compiled in. ImageMagick doesn't seem to do anything with libwmf... is that correct? Hmm... oh, I see: libwmf support is not built in unless --with-modules=yes is used. We have not been enabling module support in the past and I didn't enable it in this release so libwmf was ignored. Comments on whether we should be enabling modules? Would it break anything? If not then I'd say go for it. Chris
RE: cygwin 1.5.10: ImageMagick 6.0.3 binaries fail
I just updated my cygwin and related apps to the latest versions using the setup.exe tool. Things seemed to be working fine before the update, but now I have trouble running any ImageMagick tools. For example, the following command and resulting error: $ convert a.jpg b.png assertion list_info != (LinkedListInfo *) NULL failed: file /home/harold/ports/ImageMagick/ImageMagick-6.0.3/magick/hashmap.c, line 1033 I've received this error with all of the ImageMagick binaries I've tried to run. Just thought I'd let people know and see if others out there get the same error. I've attached the output from cygcheck -s -v -r in case anyone needs to look at it. Can the ImageMagick maintainer (Harold) please take note of this as there has been a slew of messages in this vein? Maybe it's a packaging problem as it seems all the programs are failing this assertion. Chris J
[PATCH] Fix AMD flags in /proc/cpuinfo
This patch extends Tomas Ukkonen's earlier AMD fix by removing Intel-specific flags from /proc/cpuinfo on AMD processors. It also adds support for a few more AMD-specific flags. Output for the flags field on /proc/cpuinfo on my AMD Athlon XP now matches Linux. I changed a few of the names for Intel extended features to match Linux. Chris -- http://www.atomice.com proc_amd.patch Description: Binary data proc_amd.ChangeLog Description: Binary data
Wanted: Cygwin TWiki maintainer
A while ago I set up a Cygwin wiki here: www.doc.ic.ac.uk/~ccj00/cygwin Unfortunately I've now graduated form Imperial and my account will be closed at the end of this month. I'm looking for someone willing to host and maintain the wiki. Regards, Chris January -- http://www.atomice.com -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: cat /proc/registry/HKEY_PERFOMANCE_DATA/@ hangs
Yes, I'm still seeing the segfault in the latest snapshot, but only when run under gdb or strace. Here are some sample tests: $ cat /proc/registry/HKEY_PERFORMANCE_DATA/\@ e.out cat: /proc/registry/HKEY_PERFORMANCE_DATA/@: No such file or directory $ # no segfault $ strace -o cat_HKPD.strace cat /proc/registry/HKEY_PERFORMANCE_DATA/\@ e.out 2262669 [main] cat 2400 handle_exceptions: Exception: STATUS_ACCESS_VIOLATION 2264445 [main] cat 2400 open_stackdumpfile: Dumping stack trace to cat.exe.stackdump $ I can't reproduce this with CVS. Can you? Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: cat /proc/registry/HKEY_PERFOMANCE_DATA/@ hangs
Yes, I'm still seeing the segfault in the latest snapshot, but only when run under gdb or strace. Here are some sample tests: $ cat /proc/registry/HKEY_PERFORMANCE_DATA/\@ e.out cat: /proc/registry/HKEY_PERFORMANCE_DATA/@: No such file or directory $ # no segfault $ strace -o cat_HKPD.strace cat /proc/registry/HKEY_PERFORMANCE_DATA/\@ e.out 2262669 [main] cat 2400 handle_exceptions: Exception: STATUS_ACCESS_VIOLATION 2264445 [main] cat 2400 open_stackdumpfile: Dumping stack trace to cat.exe.stackdump $ I can't reproduce this with CVS. Can you? Chris Yes, I can. I just compiled ([1]) the latest CVS cygwin0.dll (as of today, 14:31 UTC), then tested with the new DLL ([2]). I get the same exact errors (including the stackdump). Interestingly enough, I had other unexplained exceptions with this test approach, so I'd appreciate if anyone points out what's wrong with it. Could you send me a copy of the stackdump and the strace output by private mail please? Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: cat /proc/registry/HKEY_PERFOMANCE_DATA/@ hangs
However, the fix is not as simple as inserting a size = bufalloc; just before the RegQueryValueEx. When I do that, I get a SIGSEGV in the guts of iasperf.dll, which I have yet to track down. This happens on the second iteration, FWIW, with buffer increment of 1000. I'm going to investigate some more, but I'd say that with the above bug, this key was never tested, so I have no idea what's going on. Hopefully Chris (January) can use this to help him track down the problem. I'm back from my honeymoon (!) and I've just been catching up on this thread. Are you still seeing the segfault Igor? If so I'll try to track it down if I have any spare time. As you can probably tell I never tested the HKEY_PERFORMANCE_DATA key. Increasing the buffer size in increments is of course boilerplate code but I managed to cod it up regardless. Sigh. Chris January -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Bug fix to /proc/cpuinfo implementation
Hi I hope I email to correct mailing list... It seems that '/proc/cpuinfo' doesn't report 3dnow and 3dnowext support correctly. Because I had been using it for recognizing processor features I looked briefly into fhandler_proc.cc (taken freshly from cvs) and I think I fixed the problem. The changed file compiles and *should* correctly detect presence of 3dnow instruction support (by using AMD's 0x8000..1 extended function). I haven't have time to test it (= figure out how build+install process actually works) because I decided to write my own cpuid detection routines from scratch instead. Actually nearly all of the processor feature flags are wrong for AMD processors. This is a known bug. Your patch looks to be going about the right way to fix it. I'll try to add the rest of the AMD flags in. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Cygwin's chmod +X
I noticed while creating a bash script to backup my parents outlook mydocuments folers, that WindowsXP does not recognize a superuser as being allowed access to a users folders! You can only access a folder if you have been given permission (in the ACL) unless you open it in backup mode. You can only change the ACL if you are the owner of the folder. If you are an Administrator you can take ownership of the folder and change the permissions. What you actually want to do is change the owner of these folders to the Administrators group and then: Before backing up: cacls * /t /e /g yourusername:F After backing up: cacls * /t /e /r yourusername or something like that... In Local Security Policy you can change the default owner of objects created by members of the Administrators group to Administrators by changing the System Object: Default owner for objects created by members of the administrator group option in Local Policies/Security Options. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: A new method of storing package data base information, proposing packages, and announcing updates
Anyway, is this crackrock? Good stuff? Suggest any tweaks? This looks great. It would be good if the site could subscribe to the cygwin-apps mailing list and parse replies to the initial announcement so they also got included on the website. Chris
RE: co-linux
On Wed, Apr 14, 2004 at 02:50:42PM +0200, Pinhas Krengel wrote: I have just read about beta release of co-linux (linux on windows). Is this product going to kill cygwin. No. What will be the benefits of using cygwin in that case. Having a nicely integrated UNIX environment. Colinux just allows you to run a separate linux subsystem under windows. There isn't any real communication between windows and linux other than via networking. I spent some time working on porting Linux to Windows myself (with some success, but I shelved the project after CoLinux came out) and it may be possible to use CoLinux to service Linux system calls in a standard Windows process giving you the best of both worlds. AFAICT the only thing required to do this is to introduce the notion of 'foreign' processes and allow the Linux kernel to read/write to/from a Windows process' address space. That way you could link against both Windows DLLs and Linux shared objects (via the Linux loader as ported in LINE). CoLinux could run as a Windows service. Not that this would actually gain you much mind, since I can't think of anyone who releases binary only code for Linux but not for Windows. It would be really nice to be able to run Valgrind on Windows as well, but that's another story. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Quick hack to implement gethostbyname_r() through gethostbyname()+mutex lock
-Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Dave Korn Sent: 15 April 2004 14:03 To: [EMAIL PROTECTED] Subject: RE: Quick hack to implement gethostbyname_r() through gethostbyname()+mutex lock -Original Message- From: cygwin-owner On Behalf Of Enzo Michelangeli Sent: 15 April 2004 13:49 Another self-followup :-) - Original Message - From: Enzo Michelangeli [EMAIL PROTECTED] To: [EMAIL PROTECTED] Cc: Brian Ford [EMAIL PROTECTED] Sent: Thursday, April 15, 2004 12:03 PM Subject: Re: 1.5.9-1: socket() appears NOT to be thread-safe P.S. By the way, Corinna: couldn't I just put my gethostbyname_r() in the public domain, rather than going through the bureaucratic chore of the copyright assignment? Also because I feel that implementing it through mutex-protection of gethostbyname(), as I did, is just a quick hack, as it unnecessarily blocks other threads that could access the name server in parallel (with separate network I/O and properly re-entrant code). It may help other implementors to solve an urgent problem, but I don't think it should be released as part of the Cygwin code. Well, OK, here is the code, hereby placed in the public domain. Everybody can do with it whatever s/he likes; attribution will be appreciated. Of course, no guarantees etc. Ah, but it's not a matter of it having no copyright, but of the copyright existing and belonging to the FSF so that the GPL can be enforced on the file. If you submit a completely PD bit of source to a GPL project, other people can take that code, modify it and release it as binaries without being obliged by the GPL to provide sources, because they can claim they're working on your PD version rather than any version distributed under GPL. IOW, making code PD makes it impossible to apply and enforce the GPL to it. IIUIC. The inclusion of PD code in a GPL project makes a derivative work that is also licensed under the GPL. Someone can independently take the PD code and do whatever they want with it, but they must abide by the terms of the GPL if they wish to modify the derivative work. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: ANNOUNCE: cygbug - generic bug reporting tool (like Debian reportbug)
* Tue 2004-03-23 Christopher Faylor cgf-no-personal-reply-please-rDBXBDvO6BXQT0dZR+AlfA AT public.gmane.org * | On Tue, Mar 23, 2004 at 05:10:03PM +0200, Jari Aalto+mail.linux wrote: | | It can be used to deliver messages to upstream developers concerning | cygwin related bugs. It can also be used to send patches to other | Cugwin developer (I just sent one to ssmtp maintainer). | | The ssmtp maintainer? So in other words, you're sending private email | to people whom you deem to be the maintainer? That's really counter to | policy. The maintainer information is read from installed *README files. If there is some better way, let me know. I'm not happy with something that e-mails the maintainers directly lest this be mistaken for policy. Chris J
RE: procps returns \Unknown HZ value\
Redirecting to the cygwin mailing list. You probably need to upgrade your version of the Cygwin DLL. The /proc filesystem doesn't exist? Does something need to be installed using cygrunsrv? -richard -Original Message- From: Chris January [mailto:[EMAIL PROTECTED] Sent: Wed 3/24/2004 10:45 AM To: Duran, Richard Cc: Subject: RE: procps returns \Unknown HZ value\ Is there a parameter I can pass to procs.exe to avoid getting \Unknown HZ value (168) Assume 100.\? I don't think so. What's the output of /proc/cpuinfo on your system? Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Is the command top in cygwin?
I got two packages. Which one shall I use? What does 010801-1 mean? procps/procps-010801-1 Utilities for monitoring your system and processes on your system. procps/procps-010801-2 Utilities for monitoring your system and processes on your system. year month day 01 0801 1 = first revision 2 = second revision Chris - Original Message - From: Igor Pechtchanski [EMAIL PROTECTED] To: Peng Yu [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Sent: Wednesday, March 03, 2004 8:21 AM Subject: Re: Is the command top in cygwin? On Wed, 3 Mar 2004, Peng Yu wrote: Hi, Is the command top in cygwin? Which package I have to download, if I want to monitor the CPU resources? Peng, If you're looking for a specific program, first try searching the Cygwin packages at http://cygwin.com/packages/. For programs with short names (like top) it's usually a good idea to prepend bin/, and limit the suffixes, e.g., search for bin/top(.exe| |$) (the search page supports arbitrary Perl regular expressions). The next step is Google: http://www.google.com/search?q=cygwin+top+program. BTW, how to run a program with a lower priority from command line? man nice. Not Cygwin-specific. Igor -- http://cs.nyu.edu/~pechtcha/ |\ _,,,---,,_ [EMAIL PROTECTED] ZZZzz /,`.-'`'-. ;-;;,_ [EMAIL PROTECTED] |,4- ) )-,_. ,\ ( `'-' Igor Pechtchanski, Ph.D. '---''(_/--' `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-. Meow! I have since come to realize that being between your mentor and his route to the bathroom is a major career booster. -- Patrick Naughton -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: cygwin rebooting computer
I am a big fan of cygwin and has just learned the TAB completion thing. Now, my harddisk is partitioned in 2 so sometimes (not always) when doing cd /cygdrive/d TAB, teh computer goes black = softboot = reset Am I doing anything wrong or what Try looking in the System event log ('Control Panel'-'Administrative Tools'-'Event Viewer' then click System) to see if that sheds any light on the crashes. Assuming you are using Windows XP: Try opening 'Control Panel'-'System'. Then 'Advanced' tab-'Settings'. Uncheck 'Automatically restart'. Next time you should see a blue screen instead of the computer rebooting. See if this helps you solve the problem. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Output of 'top' lacking load average output
I am wondering why the output of top does not produce any output for the load averages, I only get this... 13:44:32 up 2 days, 3:03, 2 users, load average: 0.00, 0.00, 0.00 Windows doesn't record load averages. In the absence of a process to collect load average data these values are not available. (Note: this isn't strictly true as there are similar performance counters on Windows 95 - feel free to write some code to read them and add them to /proc/loadavg.) Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: MS offers Services For Unix free of charge
On Wed, Jan 14, 2004 at 04:26:03PM -0500, Robb, Sam wrote: But beyond curiosity, there's not many reasons to install and use both, at least concurrently. Cygwin and SFU both address the same needs and Cygwin covers a wider range of tools. We'll see what happens though. One thing that Cygwin does lack, and SFU has, is an NFS client :-/ I know that alone will probably entice me into taking a look at SFU. It would be rather interesting to add nfs to cygwin. We could develop filesystem plug-ins which could be generalized for stuff like NFS, EXTFS, etc. Didn't someone say they had a free month? Perfect project. :-) Isn't the SFU NFS client an installable file system, i.e. you can use it anywhere in Windows, not just with the SFU stuff? Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Unable to replicate Win2000 accounts
I have recently installed Cygwin on a Win2000 Pro system. While trying to replicate the accounts from windows to Cygwin, I used the commands: mkgroup -l /etc/group mkpasswd -l /etc/passwd the output of my group and passwd files are, respectively: $ /etc/group try cat /etc/group $ /etc/passwd try cat /etc/passwd Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Any current plans for an SDL package?
Frédéric L. W. Meunier wrote: On Sun, 23 Nov 2003 [EMAIL PROTECTED] wrote: The only thing I found missing really are the libraries' header files. One such library is libjpeg, which has no development package, and i believe there was some compression lib also. The headers are in jpeg-6b-11.tar.bz2. As I don't see libjpeg or any compression library linked in my libSDL and libSDL_mixer libraries on Linux (I never tried on Windows), I'm assuming all missing headers are to compile SDL-image. It'd be nice if you could post the errors or any links to e-mails. Remember that I am talking about SDL, SDL-mixer, and SDL-image, as most programs that use one use all 3 (but it would not be difficult to package them as 3 seperate packages) As i now have the headers i cannot post the errors. At this point i have decided that I should try to make sdl linked with cygwin1.dll, this should be possible without too much work, but using mingw sounds like it is at least slightly more common. If i can make the lib like that i will have to see if i can find anybody who can maintain the packages. I would love to have a Cygwin linked SDL and am willing to maintain it if you are prepared to package it in the first place (along with SDL-mixer and SDL-image). Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: system() refuses to work!!! HELP
It did not work. Will it be that I found a bug??? How much luck I have.. :-) It follows the source. that program does not do anything of important! It is alone a test. what we most can do? I changed the command for DIR who list the directories of the windows just to facilitate the comprehension and so that it to do not are necessary to do an upload of sound.exe. Try, with that program execute some other command! Here, already tried of everything, and anything (not) worked. * #include stdlib.h #include math.h #include stdio.h main() { int a,b,c,d,e,f,g,h,j; FILE *stream; printf(Programa Gerador de Resultados para Teste:); printf(\n\nEntre com as Faixas); printf(\nNúmero a ser comparado: );scanf(%d,c); printf(\nDe: );scanf(%d,a); printf(\n.A: );scanf(%d,b); d = c; f = a; g = b; j = 1; while(g = c){ e = f * g; h = d - e; if((stream = fopen(dtr1.txt, at)) != NULL){ fprintf(stream,\n %d * %d = %d -r = %d,f,g,e,h); } fflush(stdout); fclose(stream); if(j = c){ f = a; j = 0; g++; printf(\nF: %d\tG %d:,f,g); fflush(stdout); } f++; j++; } if((system(dir)) == 0){ printf(\n\nComando executado com sucesso); } else{ printf(\n\ncomando falhou); } } This program will not work. There is no program called 'dir' in Windows. You probably want something like 'cmd /c dir'. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Bash extremely slow when started via task scheduler
Why not use cron instead? Jason Thank you for suggestion but it's an application server managed by non-UNIX 'mouse users' which may manually start the task, so it should be simple for them. It seems that not having an output window slows down the script dramatically on each 'echo'. I assume you are using Windows 2000? Could it be that Cygwin is waiting for the caption of the console window to change, but it never does because it's on a hidden desktop? Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: #! not a recognized internal or external command
Did you check the line endings? -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Neil Messmer Sent: 02 October 2003 23:30 To: [EMAIL PROTECTED] Subject: Re: #! not a recognized internal or external command I should also mention that the script was executed within a cygwin shell window when this error occurred. Neil Messmer [EMAIL PROTECTED] wrote in message news:[EMAIL PROTECTED] I would have thought the paths set in your particuliar environment would have enabled to the find the correct shell for proper execution. I am still convinced it is a setup issue as it works on another machine here. OS versions and hardware are identical. Corinna Vinschen [EMAIL PROTECTED] wrote in message news:[EMAIL PROTECTED] On Thu, Oct 02, 2003 at 12:08:25PM -0700, Neil Messmer wrote: I have just installed the latest version of cygwin and get the following error message when running my scripts while running cygwin under Win XP. It does not matter what shell I specify for the script. My paths on the win pc is set to /usr/local/bin; /usr/bin; /bin; /usr/x11r6/bin. The simple test script contains one line: #! /bin/tsch s/tsch/tcsh but the *real* problem is that you're trying to start a shell script under cmd.exe. That won't work. The error message is generated by cmd.exe because it correctly doesn't recognize #! as a command. The #! syntax requires support by the starting application, in your case, by Cygwin. If the starting application is not a Cygwin shell, you must start the script as a parameter to the right shell: C:\foo tcsh script-name Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Developer mailto:[EMAIL PROTECTED] Red Hat, Inc. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/ -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Setup uninstall order
AFAICT uninstall of existing packages in Setup is done in an arbitrary (or possible alphabetical) order. Uninstall needs to be done in the reverse order of package dependancies (if that makes sense) if it is to succeed. i.e. if you build a dependancy graph/tree of all the packages, then the leaves should be uninstalled, then their parents and so on. The problem is that, for example, if the cygwin package is uninstalled, and then another package that has an uninstall script, the script can't run because the cygwin DLL has already been uninstalled. Chris -- http://www.atomice.com
RE: GIFs in the tcm binary package
On Mon, Sep 29, 2003 at 01:31:23PM +0200, Benjamin Riefenstahl wrote: Hi Ronald, Ronald Landheer-Cieslak [EMAIL PROTECTED] writes: I just downloaded the binary package and found a lot of GIF files in them. I was wondering whether they were LZW-compressed: `file' doesn't tell met and `mozilla' shows them properly, but I was worried anyway (patents in Europe...) AFAIK, Unisys only wants license fees for software that creates GIF files that use LZW. Use and display of GIF files should not be a problem. But of course I am not layer and I may be missing something ... I thought (but am by no means certain) that the algorithm for decompressing LZW is also covered by the patent, in which case you'd need a license for any software that decompresses it as well. Again, IANAL - I was just worried that someone might get into trouble.. Where does this patent apply? The relevant patent expired in the US and software patents are not (yet) valid in member countries in the EU. Chris
RE: Skipping the /proc filesystem
Randall R Schulz wrote: At 18:15 2003-07-22, David A. Cobb wrote: I would wish to tell find not to get involved with the /proc filesystem at all. Can that easily be done? Very easily: % find / -path '/proc' -prune -o -print Would it make sense to identify the inodes under /proc/registry as not regular files (type f), but, say, devices (or other such special files)? No. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: problem with history
Hi, my cygwin bash does not write anything in .bash_history . I think it was working fine before and some changes done by me has caused it to stop updating this file. rxvt does not have this problem. any pointers where could be the problem. Close bash by typing exit rather than clicking the Close button. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: start_time patch for fhandler_process.cc
On Tue, Jul 08, 2003 at 07:09:12PM +0100, Chris January wrote: Try this Chris and see if it solves the start time problem. Chris 2003-07-28 Chris January [EMAIL PROTECTED]@atomice.net * fhandler_process.cc (format_process_stat): Changed the calculation for start_time. Sorry, no. Unknown HZ value! (250) Assume 100. USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND cgf 3452 0.0 1.0 2544 2680 ?RAug08 0:00 procps auwx Now that I've read the description of what the field is supposed to contain, I'm wondering if the culprit is the Unknown HZ value! (250) Assume 100. Maybe sysconf (_SC_NPROCESSORS_CONF) is reporting the wrong amount if the problem is indeed you are running on an SMP machine. _SC_NPROCESSORS_CONF returns two, as it should. _SC_NPROCESSORS_ONLN returned three, which was wrong, but I just checked in a fix for that. No change after that, though. I guess I'll build procps and see what's up. There are a couple of problems. 1) procps is not allowing a valid 500MHZ setting for my system (patch enclosed for procps). 2) /proc/stat is not reporting times for all cpus (patch enclosed and applied). With these two patches, procps reports accurate times. This requires a new procps release, though, unfortunately. cgf A new procps release is due anyway so I shall make one sometime in the next couple of weeks. The /proc/stat patch looks good. Chris
RE: Read access to all keys in /proc/registry
find /proc/registry/HKEY_LOCAL_MACHINE/ does not go down the /proc/registry/HKEY_LOCAL_MACHINE/SOFTWARE directory(key) because there is no read access for people outside of the SYSTEM group: $ ls -la /proc/registry/HKEY_LOCAL_MACHINE/ total 0 dr-xr-xr--5 Administ SYSTEM 0 Jul 10 08:50 . dr-xr-xr-x9 00 0 Jul 10 13:03 .. dr-xr-xr--4 Administ SYSTEM 0 Jul 10 08:50 HARDWARE dr-xr-xr--1 Administ SYSTEM 0 Mar 15 2001 SAM dr-xr-xr-x1 00 0 Jul 10 13:03 SECURITY dr-xr-x--- 65 Administ SYSTEM 0 Mar 15 2001 SOFTWARE dr-xr-x---6 Administ SYSTEM 0 Mar 15 2001 SYSTEM The SYSTEM group is not a normal Windows group, so how can I add myself to the SYSTEM group (Cygwin doesn't provide the usual Linux group commands e.g. usermod, newgrp). Otherwise, how can I modify the permissions for read access to others so that the find command goes down the SOFTWARE directory? 1. Can you access the key using ls? 2. Can you access the key in regedit? 3. Are you running as an Administrator? Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: start_time patch for fhandler_process.cc
On Tue, Jul 08, 2003 at 07:09:12PM +0100, Chris January wrote: Try this Chris and see if it solves the start time problem. Chris 2003-07-28 Chris January [EMAIL PROTECTED]@atomice.net * fhandler_process.cc (format_process_stat): Changed the calculation for start_time. Sorry, no. Unknown HZ value! (250) Assume 100. USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND cgf 3452 0.0 1.0 2544 2680 ?RAug08 0:00 procps auwx Now that I've read the description of what the field is supposed to contain, I'm wondering if the culprit is the Unknown HZ value! (250) Assume 100. Could that be it? Almost certainly. That would give you dates 250/100 = 2.5 times into the future I should think. Could the problem be you have an SMP machine? procps calibrates HZ using uptime and total cpu jiffies. The /proc implementation asks NT how many ticks it spent in kernel, user mode, etc. However maybe NT counts ticks more than once if you have more than one processor? Chris
RE: start_time patch for fhandler_process.cc
On Tue, Jul 08, 2003 at 07:09:12PM +0100, Chris January wrote: Try this Chris and see if it solves the start time problem. Chris 2003-07-28 Chris January [EMAIL PROTECTED]@atomice.net * fhandler_process.cc (format_process_stat): Changed the calculation for start_time. Sorry, no. Unknown HZ value! (250) Assume 100. USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND cgf 3452 0.0 1.0 2544 2680 ?RAug08 0:00 procps auwx Now that I've read the description of what the field is supposed to contain, I'm wondering if the culprit is the Unknown HZ value! (250) Assume 100. Maybe sysconf (_SC_NPROCESSORS_CONF) is reporting the wrong amount if the problem is indeed you are running on an SMP machine. Chris
RE: mmap() and gcc precompiled headers
On Thu, Jul 03, 2003 at 10:19:42AM -0400, Christopher Faylor wrote: On Thu, Jul 03, 2003 at 11:47:28AM +0200, Corinna Vinschen wrote: Do you mean something like this: If addr is given, check if it's 64K aligned. If not, align and raise the memory requirement accordingly. Call MapViewOfFileEx with the aligned address. If it works, return the addr given as parameter, otherwise return MapViewOfFileEx(NULL). How about, instead, just use the address and if it fails and is not MAP_FIXED, use MapViewOfFileEx without the address? Yep, that's the simple approach. I dropped this suggestion from my original reply since it requires addr to be on a 64k boundary. Unfortunately I have no idea if the chance to succeed might be better or worse than using the more complex approach. Just out of curiosity would a patch to support mmap on 4k boundaries be useful/accepted? Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: killall utility
On Wed, 2 Jul 2003, Christopher Faylor wrote: On Wed, Jul 02, 2003 at 11:47:04AM +0100, Elfyn McBratney wrote: I have written a killall utility based on the code already in utils/kill.cc . Would this make a worthy addition to Cygwin? If so, there's a bit of code duplication, so maybe moving the generic code into a file called `sigutil.cc' or something would be sufficient, having kill{,all}.exe depending on `sigutil.o'. Any ideas bofore I submit a patch? Can't you do something like this with the kill in procps? I did look and from the usage info, it doesn't look that way. However you can compile the source for the /proc-based killall and it should work as-is. Chris
RE: Weird top bug?
I've stumbled across some peculiar behaviour for top. I've got the following script (vping) that I use to keep a VPN connection alive: #!/bin/sh while true; do ping -n 1 remoteMachine /dev/null sleep 60 done I typically run this as a background task (vping ), then telnet to remoteMachine. Now when I quit telnet, then exit the shell (vping was run from), the shell stays around. My script has a stdout handle, I suppose. That's OK, I can close the window with the mouse, and then vping dies. Seems normal. But if I leave the shell open after typing exit, then run top in another shell, it clears the screen, shows exactly one line of output (in this specific case): 15:23:03 up 8:00, 2 users, load average: 0.00, 0.00, 0.00 and locks up. Control C does not regain shell control. If I open another shell and use ps to find the process number for top, I can kill it (kill pid, no explicit signal type). It doesn't seem to me that my specific script should have anything to do with how top is behaving, but I supose it is possible. Seems more like top is having trouble because the parent process of my script is no longer valid. Maybe the parent process is gone, but top uses a windows thing to enumerate processes, and the open shell still has an entry in that list? I can't reproduce this. Can you try something like strace top somefile 21? Then take the steps you outlined above to get top to lock up and send me the last few thousand lines of the file? (to personal e-mail) Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Weird top bug?
I've stumbled across some peculiar behaviour for top. I've got the following script (vping) that I use to keep a VPN connection alive: #!/bin/sh while true; do ping -n 1 remoteMachine /dev/null sleep 60 done I typically run this as a background task (vping ), then telnet to remoteMachine. Now when I quit telnet, then exit the shell (vping was run from), the shell stays around. My script has a stdout handle, I suppose. That's OK, I can close the window with the mouse, and then vping dies. Seems normal. But if I leave the shell open after typing exit, then run top in another shell, it clears the screen, shows exactly one line of output (in this specific case): 15:23:03 up 8:00, 2 users, load average: 0.00, 0.00, 0.00 and locks up. Control C does not regain shell control. If I open another shell and use ps to find the process number for top, I can kill it (kill pid, no explicit signal type). It doesn't seem to me that my specific script should have anything to do with how top is behaving, but I supose it is possible. Seems more like top is having trouble because the parent process of my script is no longer valid. Maybe the parent process is gone, but top uses a windows thing to enumerate processes, and the open shell still has an entry in that list? Please post the output of cygcheck. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: win32api_NtQuerySystemInformation compiling problem
Hi, I've just try to use native w32api function NtQuerySystemInformation, but linker fail with following. Is this something trivial I am missing? You forgot -lntdll Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Mozilla 1.3 built on cygwin?
If you have such great insight into this type of thing, it won't take you any time at all to duplicate. You've been complaining about this and other cygwin performance issues for months. Why don't *you* do something? I figured fork/exec/signals out from scratch. Certainly the brighter bulbs than I who would be finding the problem would have no problems generating a new and better implementation. It's a wonder why no one has done so yet. There are several people on the list with more skills as I have and I was hoping that somebody of this gurus could fix this I'm wondering too. It seems really it is on me to buy this Nebett book and to see, what I can do. It will need some time to get familiar with this stuff and I have currently two libtool relating outstanding tasks, which has to be finished first You could buy the Nebett book. Or instead download the sample source code (including the fork example) from http://www.newriders.com/content/images/1578701996/downloads/1578701996.zip and http://www.newriders.com/content/images/1578701996/downloads/ntdll.zip. You can find most of the API that's documented in Nebbet's book at http://undocumented.ntinternals.net/. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Mozilla 1.3 built on cygwin?
On Sat, Mar 29, 2003 at 12:04:01AM -, Chris January wrote: On Thu, Mar 27, 2003 at 11:58:50PM +0100, Ralf Habacker wrote: I can't prove a fact, that forking is the most anonying problem and there were some initial work from some people (I remember Chris Faylor, Chris January and other) to identify the problems and to implement a new copy-on-write semantic, which will be much faster, You misremember. I did hobble together a copy-on-write implementation and found that it was actually slower. The generic win32 implementation of copy-on-write isn't powerful enough to completely implement fork anyway. Noone has explained, however, *why* the copy-on-write implementation was slower. Perhaps we have just been using the wrong tests. Does copy-on-write actually perform slower in real world tests? I don't know, because I only used the skeleton example found in Nebbit's book. I implemented it with both the win32 api and with the skeleton example. Neither was a speed daemon. I can't think of a better test than doing a bunch of forks and measuring the results. Who knows why it is slower? Maybe ReadProcessMemory is doing copy-on-write already or something. For the record my own tests involved a single parent process forking, then sleeping for a set period of time and exiting. The child process wrote all over the heap while the parent was sleeping, thus forcing all of the shared pages to be copied. This was faster with Cygwin's fork than with the copy-on-write fork, even with Cygwin's extra process launching overhead, but I could not explain why. My copy-on-write fork code doesn't work on XP SP1 so I can't retest right now. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: RPM-4.1 port to cygwin available
Peter Ring wrote: There's substantial evidence that RPM based distribution of Cygwin is feasible: http://www.holonlinux.com/product/xonwin/index.html Just in case you don't read Japanese, go directly to the FTP site: ftp://xow.holonlinux.com/pub/XonWindows/ PETER! (In case anyone was wondering, Peter was one of those hardy souls working on porting rpm 'back in the day' -- IIRC Peter was working on early 4.0.x versions...) Yes, an RPM-based cygwin is feasible -- but the last time I looked, most of the competitors said something like: First do (X) to install a basic cygwin system, and then use this tarball of rpm.exe, run rpm --initdb, then use rpm to install and/or update other parts of your system Where (X) is unpack a tarball or piggyback off setup.exe and only install these three packages or somesuch. While *feasible,* that's not really *practical* as a complete distribution. Further, none of the schemes out there were capable of updating the cygwin dll itself -- because rpm.exe uses it. Nor could they update any other in-use files. However, things may have changed over the years. I dunno, and I'm too lazy to check now. :-) Personally, I'd welcome an official setup-installable package providing rpm. Here's why: 1) we'd probably see a number of folks -- those who don't want to permanently maintain a package, but want to provide it for people to use -- who'd choose to pack their contribution as rpms. (Preferably, these ad-hoc rpms would go somewhere like /usr/local or /opt/ or ANYWHERE except /usr and /usr/X11R6/ ). 2) as these numbers grow, folks might begin wondering how to (and provding code for) help setup.exe and rpm coexist -- updating each other's databases, maybe even linking setup.exe against librpm, etc etc. Of course, this requires that someone really really smart figure out the best way to create a native port of librpm -- that can still figure out where /var/cache/rpm and /etc and suchlike are really located... This goes back to that other thread of figuring out where / is from a non-Cygwin application. There would still have to be some auxillary program to set / in the first place. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Mozilla 1.3 built on cygwin?
On Thu, Mar 27, 2003 at 11:58:50PM +0100, Ralf Habacker wrote: I can't prove a fact, that forking is the most anonying problem and there were some initial work from some people (I remember Chris Faylor, Chris January and other) to identify the problems and to implement a new copy-on-write semantic, which will be much faster, You misremember. I did hobble together a copy-on-write implementation and found that it was actually slower. The generic win32 implementation of copy-on-write isn't powerful enough to completely implement fork anyway. Noone has explained, however, *why* the copy-on-write implementation was slower. Perhaps we have just been using the wrong tests. Does copy-on-write actually perform slower in real world tests? I don't know, because I only used the skeleton example found in Nebbit's book. Unfortunately I can't work on this anymore as I have seen the fork () code in WinNT POSIX. That code is the kind of thing it would be nice to have in Cygwin. I can't compare perfomance, however, as WinNT POSIX has significantly different overheads to Cygwin. I am trying to persuade Andrew to release his code under another license since non-GPL compatible open source programs can't currently be linked against it. If he does choose to do this and the new license is GPL compatible I will look at this again. I am also willing to talk anyone else through the process of writing a copy-on-write fork () implementation. The problem in KDE is that you can't just optimise away the fork/exec pairs using vfork/spawn, because the fork is abstracted in a class and you don't know what the calling code's intentions are. I also am working on a native exec () implementation that doesn't spawn a new process. However I don't think this will give any significant speed-up over CreateProcess and I doubt very much you will ever see it in Cygwin. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: tar and gzip
Hi all! I wrote a small script in Python, but it requires two programs to run correctly: tar.exe and gzip.exe. Both are in CygWin package. And that's my question: can I bundle both programs and cygwin1.dll with my script? Script is free, but the program that the script comes with is not. -- Krzysiek 'Nelchael' Pawlik | C/C++, PHP, OpenGL, WinAPI [EMAIL PROTECTED] | Network Administrator - BAFH http://www.ps.nq.pl/pcfaq/ | http://www.ps.nq.pl/nelchael/ These are just my thoughts and I'm not a lawyer. It doesn't sound like your proprietary program is derived from or based on any Cygwin source code. Does it execute the Python script which executes tar.exe? If it does, I don't think even that would put it under the GPL. The GPL states that the act of running the Program is not restricted. Your program can execute Cygwin binaries without it becoming GPL software. If you link to Cygwin source code, then your program would be a derivative work under the GPL. However, I believe you could also link to another proprietary third party library without providing it's source code. For instance, you could link to a Microsoft library without being required to provide Microsoft source code. This is not true. It is ok to link with certain Microsoft DLLs because the GPL makes the following exception: However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable. However this exception does not apply to other DLLs, only those considered part of the operating system. Going one step further, you could put your proprietary code into a standalone DLL built using Microsoft tools. You could market the DLL as a separate product. The DLL would have no dependencies on any Cygwin source or binary. Your Cygwin based application could us it just like any other third party library without providing source code for the DLL. I don't see GPL language that would prevent this. From the GPL FAQ (http://www.gnu.org/licenses/gpl-faq.html#LinkingWithGPL): You have a GPL'ed program that I'd like to link with my code to build a proprietary program. Does the fact that I link with your program mean I have to GPL my program? Yes. and: What is the difference between mere aggregation and combining two modules into one program? Mere aggregation of two programs means putting them side by side on the same CD-ROM or hard disk. We use this term in the case where they are separate programs, not parts of a single program. In this case, if one of the programs is covered by the GPL, it has no effect on the other program. Combining two modules means connecting them together so that they form a single larger program. If either part is covered by the GPL, the whole combination must also be released under the GPL--if you can't, or won't, do that, you may not combine them. What constitutes combining two parts into one program? This is a legal question, which ultimately judges will decide. We believe that a proper criterion depends both on the mechanism of communication (exec, pipes, rpc, function calls within a shared address space, etc.) and the semantics of the communication (what kinds of information are interchanged). If the modules are included in the same executable file, they are definitely combined in one program. *** - If modules are designed to run linked together in a shared address space, that almost surely means combining them into one program. - *** By contrast, pipes, sockets and command-line arguments are communication mechanisms normally used between two separate programs. So when they are used for communication, the modules normally are separate programs. But if the semantics of the communication are intimate enough, exchanging complex internal data structures, that too could be a basis to consider the two parts as combined into a larger program. and: I'd like to incorporate GPL-covered software in my proprietary system. Can I do this? You cannot incorporate GPL-covered software in a proprietary system. The goal of the GPL is to grant everyone the freedom to copy, redistribute, understand, and modify a program. If you could incorporate GPL-covered software into a non-free system, it would have the effect of making the GPL-covered software non-free too. A system incorporating a GPL-covered program is an extended version of that program. The GPL says that any extended version of the program must be released under the GPL if it is released at all. This is for two reasons: to make sure that users who get the software get the freedom they should have, and to encourage people to give back improvements that they make. However, in many cases you can distribute the
RE: [PATCH] performance patch for /proc/registry -- version 2
How common are ACLs 4096 bytes? Could you try calling RegKeyGetSecurity twice? First with a length of 0. Then RegKeyGetSecurity will set length to the required buffer size which you can allocate dynamically using new. Chris Here is a second version of the patch, with the code indented properly -- no other changes. I just ran indent -nut on it after looking around for Cygwin coding standards info. 2003-03-25 Joe Buehler [EMAIL PROTECTED] * autoload.cc: added RegGetKeySecurity() * security.cc (get_nt_object_attribute): use RegGetKeySecurity() for performance. Index: autoload.cc === RCS file: /cvs/src/src/winsup/cygwin/autoload.cc,v retrieving revision 1.65 diff -u -r1.65 autoload.cc --- autoload.cc 13 Mar 2003 22:53:15 - 1.65 +++ autoload.cc 25 Mar 2003 19:28:24 - @@ -375,6 +373,7 @@ LoadDLLfunc (SetSecurityDescriptorGroup, 12, advapi32) LoadDLLfunc (SetSecurityDescriptorOwner, 12, advapi32) LoadDLLfunc (SetTokenInformation, 16, advapi32) +LoadDLLfunc (RegGetKeySecurity, 16, advapi32) LoadDLLfunc (NetApiBufferFree, 4, netapi32) LoadDLLfuncEx (NetGetDCName, 12, netapi32, 1) Index: security.cc === RCS file: /cvs/src/src/winsup/cygwin/security.cc,v retrieving revision 1.141 diff -u -r1.141 security.cc --- security.cc 19 Mar 2003 21:34:38 - 1.141 +++ security.cc 26 Mar 2003 14:08:30 - @@ -1443,19 +1444,73 @@ PSECURITY_DESCRIPTOR psd = NULL; cygpsid owner_sid; cygpsid group_sid; - PACL acl; + PACL acl = NULL; - if (ERROR_SUCCESS != GetSecurityInfo (handle, object_type, - DACL_SECURITY_INFORMATION | - GROUP_SECURITY_INFORMATION | - OWNER_SECURITY_INFORMATION, - (PSID *) owner_sid, - (PSID *) group_sid, - acl, NULL, psd)) + if (object_type == SE_REGISTRY_KEY) { - __seterrno (); - debug_printf (GetSecurityInfo %E); - return -1; + // use different code for registry handles, for performance reasons + char sd_buf[4096]; + PSECURITY_DESCRIPTOR psd2 = (PSECURITY_DESCRIPTOR) sd_buf[0]; + DWORD len = sizeof (sd_buf); + if (ERROR_SUCCESS != RegGetKeySecurity ((HKEY) handle, + DACL_SECURITY_INFORMATION | + GROUP_SECURITY_INFORMATION | + OWNER_SECURITY_INFORMATION, + psd2, len)) +{ + __seterrno (); + debug_printf (RegGetKeySecurity %E); + return -1; +} + + BOOL bDaclPresent; + BOOL bDaclDefaulted; + if (!GetSecurityDescriptorDacl (psd2, + bDaclPresent, acl, bDaclDefaulted)) +{ + __seterrno (); + debug_printf (GetSecurityDescriptorDacl %E); + return -1; +} + if (!bDaclPresent) +{ + acl = NULL; +} + + BOOL bGroupDefaulted; + if (!GetSecurityDescriptorGroup (psd2, + (PSID *) group_sid, + bGroupDefaulted)) +{ + __seterrno (); + debug_printf (GetSecurityDescriptorGroup %E); + return -1; +} + + BOOL bOwnerDefaulted; + if (!GetSecurityDescriptorOwner (psd2, + (PSID *) owner_sid, + bOwnerDefaulted)) +{ + __seterrno (); + debug_printf (GetSecurityDescriptorOwner %E); + return -1; +} +} + else +{ + if (ERROR_SUCCESS != GetSecurityInfo (handle, object_type, +DACL_SECURITY_INFORMATION | +GROUP_SECURITY_INFORMATION | +OWNER_SECURITY_INFORMATION, +(PSID *) owner_sid, +(PSID *) group_sid, +acl, NULL, psd)) +{ + __seterrno (); + debug_printf (GetSecurityInfo %E); + return -1; +} } __uid32_t uid; -- Joe Buehler
RE: procps and top output
Randall R Schulz wrote: What options to procps are you using to get that output format? I cannot reproduce it. It's the output of top, and yes, I see the problem too. The size column is always around 400 (+/-) MB, however large or small the process. The RSS size is correct - it matches the resident set size in the Windows task manager process display. It's only the size (== VM size in task manager?) that seems to be off. I noticed this when I was porting procps but I never had time to investigate exactly why this is so. It might be because VM size also includes DLLs and file mappings. Chris -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/