May 2005 - Darwin-dev - lists.apple.com

Re: Filesystem View
by Hamish Allan 19 May '05

19 May '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hi Mike, $ ls /.spotlightfs/kMDItemFSName==\'test.txt\'/ would give you the same result as: $ mdfind "kMDItemFSName=='test.txt'" Thanks, Hamish On 19 May 2005, at 20:19, Mike Smith wrote: From: Hamish Allan <hamish.allan(a)gold.ac.uk> Subject: Filesystem View Hi, I'm looking for a way to create a filesystem that is actually a view onto another filesystem. Something akin to what is described here: http://bryanmills.net:8086/archives/2005/05/meta-filesystem/ Bryan's implementation, however, uses FUSE, which as far as I can tell is Linux-specific, as is LUFS. Is there anything similar for Darwin / OS X? If not, how do I best go about my task? I tried looking for information about the VFS, but all I found were mails on this list saying that it was now finally stable in Tiger. Anyway I'm not sure that VFS would be quite what I'm looking for (please excuse my ignorance, I'm usually an application developer). = Mike _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… Many thanks for your reply. What I really want to do is create a filesystem in which Spotlight queries map directly to paths, e.g.: This shares some common ground with volfs, in that a readdir in /.spotlightfs/ would yield only . and .. (as enumerating all the possible entries would make no sense). Also, like volfs it should report identical inode numbers for the files it references (i.e., they should be akin to hard links). However, as you say I would want it to behave like WebDAVfs in caching its results (using kernel notification or "live updates" in Spotlight terminology). Having thought about this some more I realise, given that MDQuery apparently requires a run loop to yield results (although the documentation says it can run synchronously), I would probably need a daemon process to take queries from feed results to my filesystem (via Mach messages?). It strikes me that the process "mds" is already that daemon (feeding results to the MDQuery), and it would be a shame to duplicate the functionality, but I don't suppose you know how MDQuery and the mds daemon actually communicate? On May 18, 2005, at 12:13 PM, darwin-dev-request(a)lists.apple.com wrote: Please ignore the other suggestions you've been abused with to date; including the reference to volfs which isn't very helpful for a variety of reasons. If you are looking to make a straight pathname->pathname transformation, i.e. you're able to use lookup by pathname on the base filesystem, you should take a look at WebDAVfs. Essentially, it presents a filesystem hierarchy as represented on the server, but backs it with cache files in a directory on a local filesystem; this is the same disconnect between namespace geometries that I think you're looking for. If you're able to post a few more details on the namespace transform that you're looking to achieve, it might be possible to offer some more concrete suggestions, however I think for most cases the WebDAVfs model will work fine. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Building CF-Lite on Linux
by Garth Cummings 19 May '05

19 May '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hi Nicola, On May 19, 2005, at 10:45 AM, Nicola Vitacolonna wrote: I have downloaded CF-299.35 from http://www.opensource.apple.com/darwinsource/Current/ instead of CF-368.1 from <snip...> Hope this helps, --gc __________________________________________________________________ Garth Cummings Apple Developer Technical Support gcummings(a)apple.com http://developer.apple.com/technicalsupport Spotlight on Innovation Worldwide Developers Conference 2005 June 6-10 in San Francisco http://developer.apple.com/wwdc _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com http://www.opensource.apple.com/darwinsource/10.4.1/ Should I always get projects from the Current directory? Current should always point to the latest release, 10.4.1 in this case. When you downloaded CF, Current was pointing to 10.4 instead. That has been fixed. smime.p7s

1 0

Re: Filesystem View
by Mike Smith 19 May '05

19 May '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On May 18, 2005, at 12:13 PM, darwin-dev-request(a)lists.apple.com wrote: From: Hamish Allan <hamish.allan(a)gold.ac.uk> Subject: Filesystem View Hi, I'm looking for a way to create a filesystem that is actually a view onto another filesystem. Something akin to what is described here: http://bryanmills.net:8086/archives/2005/05/meta-filesystem/ Bryan's implementation, however, uses FUSE, which as far as I can tell is Linux-specific, as is LUFS. Is there anything similar for Darwin / OS X? If not, how do I best go about my task? I tried looking for information about the VFS, but all I found were mails on this list saying that it was now finally stable in Tiger. Anyway I'm not sure that VFS would be quite what I'm looking for (please excuse my ignorance, I'm usually an application developer). If you are looking to make a straight pathname->pathname transformation, i.e. you're able to use lookup by pathname on the base filesystem, you should take a look at WebDAVfs. If you're able to post a few more details on the namespace transform that you're looking to achieve, it might be possible to offer some more concrete suggestions, however I think for most cases the WebDAVfs model will work fine. = Mike _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… Please ignore the other suggestions you've been abused with to date; including the reference to volfs which isn't very helpful for a variety of reasons. Essentially, it presents a filesystem hierarchy as represented on the server, but backs it with cache files in a directory on a local filesystem; this is the same disconnect between namespace geometries that I think you're looking for. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Building CF-Lite on Linux
by Nicola Vitacolonna 19 May '05

19 May '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com I have downloaded CF-299.35 from http://www.opensource.apple.com/darwinsource/Current/ instead of CF-368.1 from http://www.opensource.apple.com/darwinsource/10.4.1/ But this might not be CF's fault. Nicola Il giorno 19/mag/05, alle ore 16:59, Nicola Vitacolonna ha scritto: - Messaggio controllato dal sistema antivirus del Dipartimento di Matematica e Informatica dell'Universita' degli Studi di Udine. Email scanned by the antivirus system of the Department of Mathematics and Informatics of the University of Udine (Italy). _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… Following the instructions at http://developer.apple.com/darwin/ cflite.html (in particular, I had to apply the patch file CF-299.linux.diff as explained) I was able to build the static library on Debian Linux. I succeeded with gcc 3.3.5, but not with gcc 2.96. I have not yet been able to compile the example, though, because I get the following: tmp/CoreFoundation.dst/usr/local/lib/libCoreFoundation.a (CFUniChar.opt.o)(.text+0xcf): In function `__CFUniCharLoadFile': StringEncodings.subproj/CFUniChar.c:66: undefined reference to `strlcpy' /tmp/CoreFoundation.dst/usr/local/lib/libCoreFoundation.a (CFUniChar.opt.o)(.text+0xe7):StringEncodings.subproj/CFUniChar.c:67: undefined reference to `strlcat' /tmp/CoreFoundation.dst/usr/local/lib/libCoreFoundation.a (CFUniChar.opt.o)(.text+0xfe):StringEncodings.subproj/CFUniChar.c: 112: undefined reference to `strlcat' Are projects in 10.4.1 directory supposed to compile only with Darwin? Should I always get projects from the Current directory? Maybe, a bit more information within projects would help newbies like me... Hi, I'm trying to build CF-Lite (CF-368) on Linux. According to the article at http://developer.apple.com/darwin/cflite.html this should be unproblematic. In fact, when I try "make install" I get the following error message: /bin/cp: cannot stat `OSXCompatibilityHeaders/*.h': No such file or directory - Messaggio controllato dal sistema antivirus del Dipartimento di Matematica e Informatica dell'Universita' degli Studi di Udine. Email scanned by the antivirus system of the Department of Mathematics and Informatics of the University of Udine (Italy). This email sent to site_archiver(a)lists.apple.com

1 0

Re: poll() Bug in 10.4 / Darwin 8.0 ?
by Dave Zarzycki 19 May '05

19 May '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com davez On May 19, 2005, at 8:09 AM, Steven Palm wrote: I end up here by tracing with gdb: The hang is here: It appears to hang on the read call. info locals shows: descriptors = {{ fd = 0, events = 1, revents = 32 }} ptr = 0xbffff424 "" i = 1023 j = -1073744832 count = 1 state = 0 The author of wy60 responded with: if (!(descriptors->revents & POLLIN)) i = 0; else before the line that reads if (i == 0) This email sent to n9yty(a)n9yty.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/zarzycki%40apple.com This email sent to zarzycki(a)apple.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… Bug number 3710161 at Apple is tracking this. In short, kqueue and poll on anything in /dev does not behave correctly. I was trying to use a program I've used for a long time in 10.2 and 10.3, a terminal emulator "shim" called "wy60". I haven't tried to wrap this in any sort of simple test case, nor have I filed this as a bug. I thought I'd make sure it was worthy of being filed and not a programming error, although the code does work on various unixes as-is. Also, if this is terribly off-base for this group, I apologize, but it's decidedly at a Darwin level, no GUI (except possibly curses and terminal stuff) involved here. Here are some notes (and workaround) I had posted to the wy60 group, with the author's response: In trying wy60 on MacOS X 10.4 (Tiger), I'm having an odd blocking issue that does not happen with 10.2 or 10.3. #0 0x90013784 in read () #1 0x00007d44 in readResponse (timeout=500, query=0x15a74 "\033 [0c", buffer=0xbffff424 "", firstChar=27, lastChar=99, discard=0, maxLength=80) at wy60.c:3127 #2 0x00010458 in initTerminal (pty=-1) at wy60.c:4919 #3 0x000134e0 in main (argc=1, argv=0xbffff628) at wy60.c:6123 do { switch (poll(descriptors, 1, 0)) { case -1: case 0: i = 0; break; default: i = sizeof(extraData) - extraDataLength; if (i == 0) { *buffer = '\000'; return(buffer); } else if ((count = read(0, extraData + extraDataLength, i)) > 0) { extraDataLength+= count; } else i = 0; } } while (i); Now, why this is an issue in 10.4 and was not in any earlier version, I do not know. I first attempted to run the binary from an earlier system without luck, and then compiled a fresh copy on the 10.4 machine and it has the same problem. Well, the fix seems easy enough, but the underlying answer not so easy.... I simply edited my config.h to remove HAVE_POLL_H so it wouldn't use the system poll call, and it now works using select and/or poll emulation. My thinking was that poll() was telling the system it had data to read when it really didn't have anything there. I see the author of the tun/tap kext for MacOS X also claims the poll() call is broken for character devices in Darwin8. For some reason, it seems to think that stdin is not a valid file descriptor. Not quite sure why that would happen. But in any case, can you try adding This should make the code stop trying to read from stdin, if poll returned with anything other than POLLIN. Hopefully, if I understand your problem description correctly, this will fix the bug that you are seeing. Although, you might now run the risk of occasionally seeing garbage characters when you press function keys. If that happens, we'll need to take a closer look at what the system calls are actually returning in all of these cases. _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-userlevel mailing list (Darwin-userlevel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-userlevel/n9yty% 40n9yty.com This email sent to site_archiver(a)lists.apple.com

1 0

poll() Bug in 10.4 / Darwin 8.0 ?
by Steven Palm 19 May '05

19 May '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com I end up here by tracing with gdb: The hang is here: It appears to hang on the read call. info locals shows: descriptors = {{ fd = 0, events = 1, revents = 32 }} ptr = 0xbffff424 "" i = 1023 j = -1073744832 count = 1 state = 0 The author of wy60 responded with: if (!(descriptors->revents & POLLIN)) i = 0; else before the line that reads if (i == 0) This email sent to n9yty(a)n9yty.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… I was trying to use a program I've used for a long time in 10.2 and 10.3, a terminal emulator "shim" called "wy60". I haven't tried to wrap this in any sort of simple test case, nor have I filed this as a bug. I thought I'd make sure it was worthy of being filed and not a programming error, although the code does work on various unixes as-is. Also, if this is terribly off-base for this group, I apologize, but it's decidedly at a Darwin level, no GUI (except possibly curses and terminal stuff) involved here. Here are some notes (and workaround) I had posted to the wy60 group, with the author's response: In trying wy60 on MacOS X 10.4 (Tiger), I'm having an odd blocking issue that does not happen with 10.2 or 10.3. #0 0x90013784 in read () #1 0x00007d44 in readResponse (timeout=500, query=0x15a74 "\033[0c", buffer=0xbffff424 "", firstChar=27, lastChar=99, discard=0, maxLength=80) at wy60.c:3127 #2 0x00010458 in initTerminal (pty=-1) at wy60.c:4919 #3 0x000134e0 in main (argc=1, argv=0xbffff628) at wy60.c:6123 do { switch (poll(descriptors, 1, 0)) { case -1: case 0: i = 0; break; default: i = sizeof(extraData) - extraDataLength; if (i == 0) { *buffer = '\000'; return(buffer); } else if ((count = read(0, extraData + extraDataLength, i)) > 0) { extraDataLength+= count; } else i = 0; } } while (i); Now, why this is an issue in 10.4 and was not in any earlier version, I do not know. I first attempted to run the binary from an earlier system without luck, and then compiled a fresh copy on the 10.4 machine and it has the same problem. Well, the fix seems easy enough, but the underlying answer not so easy.... I simply edited my config.h to remove HAVE_POLL_H so it wouldn't use the system poll call, and it now works using select and/or poll emulation. My thinking was that poll() was telling the system it had data to read when it really didn't have anything there. I see the author of the tun/tap kext for MacOS X also claims the poll () call is broken for character devices in Darwin8. For some reason, it seems to think that stdin is not a valid file descriptor. Not quite sure why that would happen. But in any case, can you try adding This should make the code stop trying to read from stdin, if poll returned with anything other than POLLIN. Hopefully, if I understand your problem description correctly, this will fix the bug that you are seeing. Although, you might now run the risk of occasionally seeing garbage characters when you press function keys. If that happens, we'll need to take a closer look at what the system calls are actually returning in all of these cases. _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-userlevel mailing list (Darwin-userlevel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-userlevel/n9yty% 40n9yty.com This email sent to site_archiver(a)lists.apple.com

1 0

Re: Float and Double precision representations
by Justin Walker 19 May '05

19 May '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On May 19, 2005, at 3:48, Warwick Hall wrote: Hi, Regards, Justin PS: don't cross-post. -- Justin C. Walker, Curmudgeon at Large Institute for General Semantics ----------- My wife 'n kids 'n dog are gone, I can't get Jesus on the phone, But Ol' Milwaukee's Best is my best friend. ----------- _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… I am developing an open source C library data/ network protocol that will be cross platform for MacOSX, Win32, Win64, Linux, and Solaris. [snip] Does anyone know if and how the above mentioned platforms differ in native floating point number representations? You have a tough job ahead of you if you don't understand that none of the systems you mention have "native floating point number representations". In all cases, what determines the representation is the hardware, not the OS. Sun's RPC/XDR protocol deals with floats, if memory serves. It sounds like you want to use it. This email sent to site_archiver(a)lists.apple.com

1 0

Building CF-Lite on Linux
by Nicola Vitacolonna 19 May '05

19 May '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Best regards Nicola Hi, I'm trying to build CF-Lite (CF-368) on Linux. According to the article at http://developer.apple.com/darwin/cflite.html this should be unproblematic. In fact, when I try "make install" I get the following error message: /bin/cp: cannot stat `OSXCompatibilityHeaders/*.h': No such file or directory A comment in the Makefile explicitly mentions that the "compatibility headers" are needed on all platforms other than Darwin. But, what are they? I have not been able to find any information about them. Can anybody help me? PS: I have copied AssertMacros.h, AvailabilityMacros.h and TargetConditionals.h from /usr/include in Mac OS X 10.4 into /usr/ include of my Linux system. _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

Float and Double precision representations
by Warwick Hall 19 May '05

19 May '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hi, Regards, Warwick Hall _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… I am developing an open source C library data/ network protocol that will be cross platform for MacOSX, Win32, Win64, Linux, and Solaris. The aim is to transmit floating point numbers in the same format used by Java virtual machines in the Class File Format (see Java Virtual Machine Specification 2nd Edition Sections 4.4.4 and 4.4.5). The problem arises when a server converts a native (platform specific) float or double into the Java format, transmits it, and the client converts it back to the client's native float or double format that there can be a loss of information such as precision. Does anyone know if and how the above mentioned platforms differ in native floating point number representations? Are there any cross platform libraries that deal with converting between platforms? This email sent to site_archiver(a)lists.apple.com

1 0

Re: Bsd sockets and GCC 4.0
by plumber 19 May '05

19 May '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Good samples code and docs for beginner can be found here http://www.alhem.net/ Best Regards -----BEGIN PLUMBER CODE BLOCK----- Version: 1.0 ¬ | | > Regards Plumber || GNU Darwin. ¬ \\\ plumber.gnu-darwin.org ¬ ¬ | | > OOP --\///\ (0)=(0) Darwin/Power Mac G4. ¬ | | > ( __ò ó__ ) The best for Fede ¬ \\\ federicafontana.it ¬ ------END PLUMBER CODE BLOCK------ -----BEGIN GEEK CODE BLOCK----- Version: 3.1 GMU/S d+@ s: a- K+++++ PGP++++ K---- dpu s--:-- C++++ C-- B L X U++++ P+++ C-- P! L--- E+++ W+++ P! L--- N+++ o-- K+++++ ------END GEEK CODE BLOCK------ Plain, Loving, Useful, Mischievous, Brutal, Explosive, Responsible _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0