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

Re: /AppleInternal
by Peter Bierman 16 May '05

16 May '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Heh. No. I think the name "AppleInternal" is explanation enough. Why do you think you need it? -pmb _______________________________________________ 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… At 5:49 PM -0400 5/16/05, Dale Walsh wrote: Can someone assist me by providing a complete contents tar of /AppleInternal please? This email sent to site_archiver(a)lists.apple.com

1 0

/AppleInternal
by Dale Walsh 16 May '05

16 May '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Mta-interface: amavisd-new-2.3.1-rc3-E (2005-05-06) at daleenterprise.com -- Dale _______________________________________________ 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… Can someone assist me by providing a complete contents tar of / AppleInternal please? This email sent to site_archiver(a)lists.apple.com

1 0

Re: Bsd sockets and GCC 4.0
by Jonas Maebe 16 May '05

16 May '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On 16 May 2005, at 23:39, Ryan Lum wrote: if(bind(s, (struct sockaddr *)&sa, sizeof(sa) < 0)) { Your code has been wrong for 10 years, it should be if(bind(s, (struct sockaddr *)&sa, sizeof(sa)) < 0) { Jonas _______________________________________________ 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… With gcc 3.3 no problems, but it is returning -1 with 4.0. I am no socket expert, and any debugging advice would help. You were passing 0 as namelen parameter (since sizeof(sa) > 0). Not sure why this would work with gcc 3 and not with 4 though. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Bsd sockets and GCC 4.0
by Josh Graessley 16 May '05

16 May '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On May 16, 2005, at 2:39 PM, Ryan Lum wrote: First, though I was getting strange compiler warnings. 1789: getsockname(s, (struct sockaddr *)&isa, &i); 1804: t = accept(s, (struct sockaddr *)&isa, &i); 1863: rc = getpeername(desc, (struct sockaddr *) &sock, &size); I fixed the warnings by making (i) an unsigned int. The bind code I am having a problem with is: if(bind(s, (struct sockaddr *)&sa, sizeof(sa) < 0)) { When you get -1 as the return value, what is errno set to? -josh _______________________________________________ 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 I have some code that is over 10 years old. With the recent change to GCC 4.0 my socket code won't compile anymore. The only way I can compile my code is if I use gcc_select. The problem is with the bind() call. comm.c: In function 'co1900_new_connection': comm.c:1798: warning: pointer targets in passing argument 3 of 'getsockname' differ in signedness comm.c:1804: warning: pointer targets in passing argument 3 of 'accept' differ in signedness comm.c: In function 'co2000_new_descriptor': comm.c:1863: warning: pointer targets in passing argument 3 of 'getpeername' differ in signedness With gcc 3.3 no problems, but it is returning -1 with 4.0. I am no socket expert, and any debugging advice would help. smime.p7s

1 0

Re: Bsd sockets and GCC 4.0
by Dave Zarzycki 16 May '05

16 May '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com You have: if(bind(s, (struct sockaddr *)&sa, sizeof(sa) < 0)) { You want: if(bind(s, (struct sockaddr *)&sa, sizeof(sa)) < 0) { davez On May 16, 2005, at 2:39 PM, Ryan Lum wrote: First, though I was getting strange compiler warnings. 1789: getsockname(s, (struct sockaddr *)&isa, &i); 1804: t = accept(s, (struct sockaddr *)&isa, &i); 1863: rc = getpeername(desc, (struct sockaddr *) &sock, &size); I fixed the warnings by making (i) an unsigned int. The bind code I am having a problem with is: if(bind(s, (struct sockaddr *)&sa, sizeof(sa) < 0)) { _______________________________________________ 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… I have some code that is over 10 years old. With the recent change to GCC 4.0 my socket code won't compile anymore. The only way I can compile my code is if I use gcc_select. The problem is with the bind() call. comm.c: In function 'co1900_new_connection': comm.c:1798: warning: pointer targets in passing argument 3 of 'getsockname' differ in signedness comm.c:1804: warning: pointer targets in passing argument 3 of 'accept' differ in signedness comm.c: In function 'co2000_new_descriptor': comm.c:1863: warning: pointer targets in passing argument 3 of 'getpeername' differ in signedness With gcc 3.3 no problems, but it is returning -1 with 4.0. I am no socket expert, and any debugging advice would help. This email sent to site_archiver(a)lists.apple.com

1 0

Bsd sockets and GCC 4.0
by Ryan Lum 16 May '05

16 May '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com User-agent: Microsoft-Entourage/11.1.0.040913 I have some code that is over 10 years old. With the recent change to GCC 4.0 my socket code won't compile anymore. The only way I can compile my code is if I use gcc_select. The problem is with the bind() call. First, though I was getting strange compiler warnings. comm.c: In function 'co1900_new_connection': comm.c:1798: warning: pointer targets in passing argument 3 of 'getsockname' differ in signedness comm.c:1804: warning: pointer targets in passing argument 3 of 'accept' differ in signedness comm.c: In function 'co2000_new_descriptor': comm.c:1863: warning: pointer targets in passing argument 3 of 'getpeername' differ in signedness 1789: getsockname(s, (struct sockaddr *)&isa, &i); 1804: t = accept(s, (struct sockaddr *)&isa, &i); 1863: rc = getpeername(desc, (struct sockaddr *) &sock, &size); I fixed the warnings by making (i) an unsigned int. The bind code I am having a problem with is: if(bind(s, (struct sockaddr *)&sa, sizeof(sa) < 0)) { With gcc 3.3 no problems, but it is returning -1 with 4.0. I am no socket expert, and any debugging advice would help. _______________________________________________ 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

Re: specifying the kernel version
by Dean Reece 16 May '05

16 May '05

site_archiver(a)lists.apple.com Delivered-To: Darwin-dev(a)lists.apple.com Cheers, - Dean Dear all, is there a way using #ifdef to determine the kernel version built during compilation? config/MasterVersion suggests but libkern/version.h is not built on my machine. Best regards Michael _______________________________________________ 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 file is new to Tiger. I don't believe there are any #defines available on releases (or SDKs) prior to Tiger. On May 16, 2005, at 2:13 PM, Michael Tuexen wrote: # The correct way to make use of the kernel version within kernel code or a # kext is to include libkern/verison.h. version.h contains defines that can # be used for build-time version logic and prototypes for variables that can # be used for run-time version logic. This email sent to site_archiver(a)lists.apple.com

1 0

specifying the kernel version
by Michael Tuexen 16 May '05

16 May '05

site_archiver(a)lists.apple.com Delivered-To: Darwin-dev(a)lists.apple.com Dear all, is there a way using #ifdef to determine the kernel version built during compilation? config/MasterVersion suggests but libkern/version.h is not built on my machine. Best regards Michael _______________________________________________ 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… # The correct way to make use of the kernel version within kernel code or a # kext is to include libkern/verison.h. version.h contains defines that can # be used for build-time version logic and prototypes for variables that can # be used for run-time version logic. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Darwin-dev Digest, Vol 2, Issue 112
by Justin Walker 16 May '05

16 May '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On May 16, 2005, at 10:09, Alexey Proskuryakov wrote: On 16.05.2005 20:49, "Justin Walker" <justin(a)mac.com> wrote: It seems like a stretch to say that, because your favorite bug isn't fixed, Apple isn't interested. I've always found them to be interested, but often there are higher priority issues to deal with. Never be afraid to imagine :-} Seriously, I thought that I implied the same by saying "isn't very interested", rather than "isn't interested". That's way too subtle, especially on a mailing list. Cheers, Justin -- 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… FWIW, last August I reported how a file can be deleted via SMB, and still be seen as existing on the same (!) local system (rdar://3749257). It's still not fixed, so looks like Apple is not very interested in such issues. Well, perhaps I'm too afraid to imagine a file system issue worse than this in a shipping system :). This email sent to site_archiver(a)lists.apple.com

1 0

Re: DiskArbitration API
by Jesper Papmehl 16 May '05

16 May '05

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hi! Thank you! /Jesper 2 maj 2005 kl. 23.11 skrev Dan Markarian: Hi Jesper, Dan On 2 May 2005, at 5:52 AM, Jesper Papmehl wrote: Hello! Thanks for any help! /Jesper Papmehl Propellerheads _______________________________________________ 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… Thanks, you are exactly right. I wasn't doing anything with the port that I got from DiskArbStart(). I now add it the the current run loop, and everything works like it did before. Disk Arbitration SPIs do continue to work on Tiger. What do you do with the port you are handed back in DiskArbStart()? Do you add it do your run loop? You must add it to your run loop in order to service it and have your notifications fire. You can also use DiskArbMsgLoop() to service it as an alternative. You may ask how some applications do work on Panther even with the oversight. The port from the SPI is a global port, for historical reasons. You may get lucky in case someone else in your application does do the right thing by adding it to the main run loop. It was probably File Manager. File Manager has since moved onto the new APIs. You must service the port you are handed back to be sure to receive notifications. The same is true of the new APIs, which are also present on Panther, and most other notification APIs, as a general rule. You can take a look at the disktool.c sample code, in the DiskArbitration project, to see how to service that port with a run loop on the old SPI. You are welcome to use the new API as well. You may sign up with DARegisterDiskPeekCallback() and claim your disk in that callback with DADiskClaim(). We have an app that reads (removable) disks that uses a file system that Mac OS X doesn't recognize. We have been using the DiskArbitration API to make the OS ignore the disks as they are inserted, to prevent the "disk cannot be read" dialog from appearing. This has been working fine, until Tiger, where it stopped working because the DiskArbitration API has been completely redesigned. (Not that I'm complaining, I know that we have been using a private, unsupported API.) I have not been able to find any documentation for the new API (apart from the brief comments in the headers), and I'm having a bit of trouble getting started. What I want to do is to get notified whenever a disk is about to be mounted, and then tell the OS to ignore the disk. Does anyone know how to do this using the Tiger version of DiskArbitration? This email sent to site_archiver(a)lists.apple.com

1 0