April 2002 - Darwin-kernel - lists.apple.com

Re: also: Darwin 6.0 compile problems
by Jean-Edouard BABIN 16 Apr '02

16 Apr '02

i also think that in osfmk/kern/task.c ligne 576 if (task->bsd_info) return(KERN_FAILURE); should be changed to #ifdef MACH_BSD if (task->bsd_info) return(KERN_FAILURE); #endif because in osfmk/kern/task.h thereis: #ifdef MACH_BSD void *bsd_info; #endif in struct task there is the same probleme in: osfmk/kern/thread.c:732 osfmk/kern/thread.c:877 osfmk/kern/thread_act.c:949 i also have a question, how can i chose the name of my kernel at boot ? thereis anything to change in the firmware ? Jean-Edouard BABIN icrit: > I have the same problem, I have deleted bsd/net/ndrv_var.h then do a cvs > update > > I think that > > struct ndrv_multiaddr > { > struct ndrv_multiaddr* next; > struct sockaddr addr; > }; > > Must be added to this .h > > I also think that struct ndrv_multiaddr* nd_multiaddrs; must be > added in struct ndrv_cb... > > After that work > > # cat bsd/net/ndrv_var.h > /* [SNIP] */ > #ifndef _NET_NDRV_VAR_H > #define _NET_NDRV_VAR_H > #ifdef KERNEL > /* [SNIP] */ > struct ndrv_cb > { > struct ndrv_cb *nd_next; /* Doubly-linked list - Obsolete */ > struct ndrv_cb *nd_prev; > struct socket *nd_socket; /* Back to the socket */ > unsigned int nd_signature; /* Just double-checking */ > struct sockaddr_ndrv *nd_faddr; > struct sockaddr_ndrv *nd_laddr; > struct sockproto nd_proto; /* proto family, protocol */ > int nd_descrcnt; /* # elements in nd_dlist - Obsolete > */ > TAILQ_HEAD(dlist, dlil_demux_desc) nd_dlist; /* Descr. list */ > struct ifnet *nd_if; > u_long nd_send_tag; > u_long nd_tag; > u_long nd_family; > short nd_unit; > }; > > #define sotondrvcb(so) ((struct ndrv_cb *)(so)->so_pcb) > #define NDRV_SIGNATURE 0x4e445256 /* "NDRV" */ > > /* Nominal allocated space for NDRV sockets */ > #define NDRVSNDQ 8192 > #define NDRVRCVQ 8192 > > extern struct ndrv_cb ndrvl; /* Head of controlblock list */ > #endif /* KERNEL */ > #endif /* _NET_NDRV_VAR_H */ > > > On 16/04/02 8:17, Josh Graessley (jgraessley(a)apple.com) wrote: > > > ndrv_multiaddr should be declared in bsd/net/ndrv_var.h. > > > > I wonder...if you're only doing cvs updates and not a checkout, perhaps you > > didn't get any files added to the net directory. ndrv_var.h was added a > > while ago. > > > > -josh > > > > On 4/15/02 11:08 PM, "Nathan T. Hjelm" <hjelmn(a)cs.unm.edu> wrote: > > > >> I am having some problems with compiling the latest HEAD cvs source. The > >> problem come up in bsd/net/ndrv.c where I get compile errors because > >> struct ndrv_multiaddr is not defined in any of the header files. Am I > >> missing somthing or is this a problem with the current cvs source? > > -- > Jean-Edouard BABIN > Mail : Jeb(a)jeb.com.fr > Web : www.Jeb.com.fr > Icq : 27305084 > Irc : Jeb @#Mac-fr , @#So .. UnderNet.org > _______________________________________________ > darwin-kernel mailing list | darwin-kernel(a)lists.apple.com > Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel > Do not post admin requests to the list. They will be ignored. -- Jean-Edouard BABIN Jeb-Soft: http://www.jeb.com.fr _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

Re: also: Darwin 6.0 compile problems
by Jean-Edouard BABIN 16 Apr '02

16 Apr '02

I have the same problem, I have deleted bsd/net/ndrv_var.h then do a cvs update I think that struct ndrv_multiaddr { struct ndrv_multiaddr* next; struct sockaddr addr; }; Must be added to this .h I also think that struct ndrv_multiaddr* nd_multiaddrs; must be added in struct ndrv_cb... After that work # cat bsd/net/ndrv_var.h /* [SNIP] */ #ifndef _NET_NDRV_VAR_H #define _NET_NDRV_VAR_H #ifdef KERNEL /* [SNIP] */ struct ndrv_cb { struct ndrv_cb *nd_next; /* Doubly-linked list - Obsolete */ struct ndrv_cb *nd_prev; struct socket *nd_socket; /* Back to the socket */ unsigned int nd_signature; /* Just double-checking */ struct sockaddr_ndrv *nd_faddr; struct sockaddr_ndrv *nd_laddr; struct sockproto nd_proto; /* proto family, protocol */ int nd_descrcnt; /* # elements in nd_dlist - Obsolete */ TAILQ_HEAD(dlist, dlil_demux_desc) nd_dlist; /* Descr. list */ struct ifnet *nd_if; u_long nd_send_tag; u_long nd_tag; u_long nd_family; short nd_unit; }; #define sotondrvcb(so) ((struct ndrv_cb *)(so)->so_pcb) #define NDRV_SIGNATURE 0x4e445256 /* "NDRV" */ /* Nominal allocated space for NDRV sockets */ #define NDRVSNDQ 8192 #define NDRVRCVQ 8192 extern struct ndrv_cb ndrvl; /* Head of controlblock list */ #endif /* KERNEL */ #endif /* _NET_NDRV_VAR_H */ On 16/04/02 8:17, Josh Graessley (jgraessley(a)apple.com) wrote: > ndrv_multiaddr should be declared in bsd/net/ndrv_var.h. > > I wonder...if you're only doing cvs updates and not a checkout, perhaps you > didn't get any files added to the net directory. ndrv_var.h was added a > while ago. > > -josh > > On 4/15/02 11:08 PM, "Nathan T. Hjelm" <hjelmn(a)cs.unm.edu> wrote: > >> I am having some problems with compiling the latest HEAD cvs source. The >> problem come up in bsd/net/ndrv.c where I get compile errors because >> struct ndrv_multiaddr is not defined in any of the header files. Am I >> missing somthing or is this a problem with the current cvs source? -- Jean-Edouard BABIN Mail : Jeb(a)jeb.com.fr Web : www.Jeb.com.fr Icq : 27305084 Irc : Jeb @#Mac-fr , @#So .. UnderNet.org _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

Re: also: Darwin 6.0 compile problems
by Justin C. Walker 16 Apr '02

16 Apr '02

On Monday, April 15, 2002, at 11:08 PM, Nathan T. Hjelm wrote: I am having some problems with compiling the latest HEAD cvs source. The problem come up in bsd/net/ndrv.c where I get compile errors because struct ndrv_multiaddr is not defined in any of the header files. Am I missing somthing or is this a problem with the current cvs source? You probably shouldn't ever mess with HEAD cvs source for Darwin. It can only lead to problems. Check the archives (including darwin-development) for further discussions. Regards, Justin -- Justin C. Walker, Curmudgeon-At-Large * Institute for General Semantics | Men are from Earth. | Women are from Earth. | Deal with it. *--------------------------------------*-------------------------------* _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

Re: also: Darwin 6.0 compile problems
by Josh Graessley 16 Apr '02

16 Apr '02

ndrv_multiaddr should be declared in bsd/net/ndrv_var.h. I wonder...if you're only doing cvs updates and not a checkout, perhaps you didn't get any files added to the net directory. ndrv_var.h was added a while ago. -josh On 4/15/02 11:08 PM, "Nathan T. Hjelm" <hjelmn(a)cs.unm.edu> wrote: > I am having some problems with compiling the latest HEAD cvs source. The > problem come up in bsd/net/ndrv.c where I get compile errors because > struct ndrv_multiaddr is not defined in any of the header files. Am I > missing somthing or is this a problem with the current cvs source? > _______________________________________________ > darwin-kernel mailing list | darwin-kernel(a)lists.apple.com > Help/Unsubscribe/Archives: > http://www.lists.apple.com/mailman/listinfo/darwin-kernel > Do not post admin requests to the list. They will be ignored. _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

Re: installing a new kernel broke carbon
by Justin C. Walker 16 Apr '02

16 Apr '02

On Monday, April 15, 2002, at 11:05 PM, Nathan T. Hjelm wrote: Hi, I have seen this problem internally when the Carbon bridge libraries, which bridge CFM binaries to macho's, get out of whack. However, I have never seen xnu affect anything like this, so I'm curious to know if any other changes were made. That's the issue. Nathan's report indicates all he changed was 'xnu', which makes the rest of the problem sound bizarre. Exactly, thats what makes this so weird. As far as I can tell with all the files in /System, none of them have a modification date that corresponds with the xnu build. I tries running the FixupResourceForks command but it seemed to do no good either. What I can gather from the error is there is some problem accessing a function inside the CarbonLib but nm does no good in helping me determine if CFMPriv_System is defined becuase it says that it isn't an object. But I have never encountered a problem like this before. You haven't mentioned what kernel you have built and tried to run. It's conceivable that the version you chose to build is sufficiently out of sync with the rest of the system as installed on your disk that you have hosed some of your files. We won't know for sure until we know what kernel you ran (and even then, we may not know; for sure, without that info, there's no way to be certain). Regards, Justin -- Justin C. Walker, Curmudgeon-At-Large * Institute for General Semantics | Men are from Earth. | Women are from Earth. | Deal with it. *--------------------------------------*-------------------------------* _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

also: Darwin 6.0 compile problems
by Nathan T. Hjelm 16 Apr '02

16 Apr '02

I am having some problems with compiling the latest HEAD cvs source. The problem come up in bsd/net/ndrv.c where I get compile errors because struct ndrv_multiaddr is not defined in any of the header files. Am I missing somthing or is this a problem with the current cvs source? _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

Re: installing a new kernel broke carbon
by Nathan T. Hjelm 16 Apr '02

16 Apr '02

> > Hi, > > > > I have seen this problem internally when the Carbon bridge > > libraries, which bridge CFM binaries to macho's, get out of whack. > > > > However, I have never seen xnu affect anything like this, so I'm > > curious to know if any other changes were made. > > That's the issue. Nathan's report indicates all he changed was 'xnu', > which makes the rest of the problem sound bizarre. Exactly, thats what makes this so weird. As far as I can tell with all the files in /System, none of them have a modification date that corresponds with the xnu build. I tries running the FixupResourceForks command but it seemed to do no good either. What I can gather from the error is there is some problem accessing a function inside the CarbonLib but nm does no good in helping me determine if CFMPriv_System is defined becuase it says that it isn't an object. But I have never encountered a problem like this before. > > It sounds like Nathan is running Mac OS X, he may want to try > > running /System/Library/CoreServices/FixupResourceForks. > > Maybe he'll eventually let us know what's going on, and whether the > above helps in any way. I sent what have done several times but found out it wasnt working cause mail has the wrong default email address. opps. _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

Re: installing a new kernel broke carbon
by Justin C. Walker 16 Apr '02

16 Apr '02

On Monday, April 15, 2002, at 08:57 PM, Dallas De Atley wrote: Hi, I have seen this problem internally when the Carbon bridge libraries, which bridge CFM binaries to macho's, get out of whack. However, I have never seen xnu affect anything like this, so I'm curious to know if any other changes were made. That's the issue. Nathan's report indicates all he changed was 'xnu', which makes the rest of the problem sound bizarre. It sounds like Nathan is running Mac OS X, he may want to try running /System/Library/CoreServices/FixupResourceForks. Maybe he'll eventually let us know what's going on, and whether the above helps in any way. Regards, Justin -- Justin C. Walker, Curmudgeon-At-Large * Institute for General Semantics | Men are from Earth. | Women are from Earth. | Deal with it. *--------------------------------------*-------------------------------* _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

Re: installing a new kernel broke carbon
by Dallas De Atley 16 Apr '02

16 Apr '02

Hi, I have seen this problem internally when the Carbon bridge libraries, which bridge CFM binaries to macho's, get out of whack. However, I have never seen xnu affect anything like this, so I'm curious to know if any other changes were made. It sounds like Nathan is running Mac OS X, he may want to try running /System/Library/CoreServices/FixupResourceForks. - Dallas On Monday, April 15, 2002, at 08:47 PM, Justin C. Walker wrote: On Monday, April 15, 2002, at 07:03 PM, Louis Gerbarg wrote: Well, what I was thinking was that if he followed some of the more common build instructions he would have not just replaced the kernel, but also the System, IOKit, and Kernel frameworks. If he then reverted by simply copying the old kernel he might have a kernel/library mismatch. On Monday, April 15, 2002, at 08:07 PM, Jim Magee wrote: It sounds to me like he installed the "less functional" Darwin version of CoreFoundation, and Mac OS X wants, no needs, the full version. I didn't get either interpretation. Nathan said the following: I built and installed the darwin kernel on my OS 10.1.3 machine and now i get messages like this one when I try to run any carbon based application: The application "Mozilla" could not be launched because of a shared library error: "&<Mozilla><CarbonLib><CFMPriv_System><>" The problem remained when I put the old kernel back. Is there a way to correct this problem without reinstalling macos 10.1 from cd? which to me indicates that the kernel is the only thing that changed. Nathan, can you clarify: - did you install anything other than the kernel (and if so, what)? - what version of the kernel did you build and install? In the meanwhile, can anyone comment on what the error message means? Regards, Justin -- Justin C. Walker, Curmudgeon-At-Large * Institute for General Semantics | If you're not confused, | You're not paying attention *--------------------------------------*-------------------------------* _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored. _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0

Re: installing a new kernel broke carbon
by Justin C. Walker 16 Apr '02

16 Apr '02

On Monday, April 15, 2002, at 07:03 PM, Louis Gerbarg wrote: Well, what I was thinking was that if he followed some of the more common build instructions he would have not just replaced the kernel, but also the System, IOKit, and Kernel frameworks. If he then reverted by simply copying the old kernel he might have a kernel/library mismatch. On Monday, April 15, 2002, at 08:07 PM, Jim Magee wrote: It sounds to me like he installed the "less functional" Darwin version of CoreFoundation, and Mac OS X wants, no needs, the full version. I didn't get either interpretation. Nathan said the following: I built and installed the darwin kernel on my OS 10.1.3 machine and now i get messages like this one when I try to run any carbon based application: The application "Mozilla" could not be launched because of a shared library error: "&<Mozilla><CarbonLib><CFMPriv_System><>" The problem remained when I put the old kernel back. Is there a way to correct this problem without reinstalling macos 10.1 from cd? which to me indicates that the kernel is the only thing that changed. Nathan, can you clarify: - did you install anything other than the kernel (and if so, what)? - what version of the kernel did you build and install? In the meanwhile, can anyone comment on what the error message means? Regards, Justin -- Justin C. Walker, Curmudgeon-At-Large * Institute for General Semantics | If you're not confused, | You're not paying attention *--------------------------------------*-------------------------------* _______________________________________________ darwin-kernel mailing list | darwin-kernel(a)lists.apple.com Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-kernel Do not post admin requests to the list. They will be ignored.

1 0