July 2005 - Darwin-kernel - lists.apple.com

ctl_enqueuedata ENOBUFS error
by Allan Hoeltje 11 Jul '05

11 Jul '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com User-agent: Microsoft-Entourage/10.1.4.030702.0 I am using ctl_enqueuedata() in my KEXT to send data to a client application. The client is reading the socket with recv( s, buff, BUFFER_LENGTH, MSG_WAITALL ). Everything seems to work fine but the ctl_enqueuedata() call in my KEXT eventually returns "ENOBUFS - The queue is full or there are no free mbufs." So, it appears that recv() does not remove data from the queue. What do I need to do so that ctl_enqueuedata() does not fill the queue? I do not see any other function calls in sys/kern_control.h that look appropriate. -Allan _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com

1 0

Re: copyout across 256MB boundary bug in 10.4.1
by Andrew Gallatin 11 Jul '05

11 Jul '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com I apologize for the delayed reply.. I was on vacation all last week. William Kucharski writes: > On Jun 30, 2005, at 4:55 PM, Andrew Gallatin wrote: > > > It seems rock solid for small buffers, but once I get over a few > > hundred megs, I see some "interesting" corruption. The corruption > > starts at user address 0x10000000 and continues through the page > > starting at 0x1006f000. Another "bad spot" appears at 0x20000000 and > > continues through the page starting at 0x2006f000. Moreover, the > > contents which should appear at 0x20000000..0x2006ffff appear at > > 0x10000000..0x1006ffff, while I see stale data at > > 0x20000000..0x2006ffff. This makes it seem like large copyouts which > > cross a 256MB boundary may wrap around, and write to exactly 256MB > > below the desired address. > > 256MB boundaries are "magic" as they denote the PPC "segment" to be > referenced, and segment registers are part of the PPC VM translation > mechanism. Simply put, each 256MB-aligned chunk is translated using > a different segment register value. (For 32-bit PPCs, the segment ID > is the virtual address & 0xf0000000). > > The copyin/copyout mechanism works in an interesting manner, mapping > 512MB chunks of the user memory space into the kernel address range > 0xe0000000 - 0xffffffff. > > So for example, accesses to virtual address 0x1000 would be mapped via > the kernel to kernel address 0xe0001000. > > For what you're seeing to make sense, what must be happening is that > since you're wrapping around a 256MB boundary, the segment register > that SHOULD be programmed for the next 256MB "chunk" is actually being > programmed to point to the original 256MB chunk. Excellent. I'm glad this makes sense in some way ;) > Do you see this same failure on 32-bit PPCs, or just on G5s? I know > you mentioned your test machine was a G5, but I'm just curious if we > can rule out the 32-bit VM code here. I have not tested yet on G4s. My only G4 is running 10.3, and I've only been able to reproduce the problem under 10.4. Additionally, I'm not 100% certain that a G4 will accept the PCI-X card whose driver is exercising this failure. If, after a week, it would still be helpful to you to know if this happens on G4s, I will try to install a card in a G4 and update it to 10.4. Thanks, Drew _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com

1 0

Re: File level action protocol
by Thomas Tempelmann 09 Jul '05

09 Jul '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Mike, >How much work do you plan to do on each notification? Do you need to >know about each and every operation as it happens, or is it enough >for you know that the file has been modified at some point prior to >"now"? The latter is enough. >If the delayed notification is acceptable, how long a delay can you >tolerate? Several seconds are no problem. Goal is to be able to tell the _user_ eventually what files have changed on his computer without having to scan the entire directory tree and compare it to a previously saved tree. >What about your other users? Is it important that it only be files >that a specific user can see? I like to know about all changed files that are present on a disk, including system files. >lookup passes a parent object and a name fragment to a >filesystem and gets back a child object or an error. You can also look up by a file ID instead of parID+name, right? So, all I need is to know a file's ID for my project. >Lookups performed by fileid may not have a meaningful parent >identifier at all On HFS+, they do, and that's all I care about. As long as I can learn the File ID while I do the recording, I should be fine. >The file handle is not relevant at the layer you're interested in; >it's a per-process association between a descriptor number and a vnode. So, what I need then is a (later) association between a vnode and a File ID. Isn't the Vnode identified by a unique number that is, like the dir and file ID, remaining constant as long as its file/dir exists? Can I record this vnode number when I get a "file change" notification? _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com

1 0

Re: Connection to KEXT with kern_control
by Allan Hoeltje 08 Jul '05

08 Jul '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com User-agent: Microsoft-Entourage/10.1.4.030702.0 Geesh!, I found _my_ problem. I misread the kern_ctl_reg description and had set the ctl_flags field to CTL_FLAG_REG_SOCK_STREAM. It should be 0 for data gram sockets. Have a nice weekend! -Allan "Allan Hoeltje" <ahoeltje(a)lmi.net> wrote on 7/7/05 11:13 AM: > My socket woes continue. > > I am attempting to open a socket from my client application to my loaded > KEXT. My client is using the sample code copied nearly verbatim from page > 20 in the "Network Kernel Extensions Programming Guide". > > The code successfully executes the call to socket( PF_SYSTEM, SOCK_DGRAM, > SYSPROTO_CONTROL ) and the call to ioctl() successfully obtains the dynamic > ctl_id of my loaded KEXT. However, when I make the call to connect() I get > a -1 return and errno contains: "41 Protocol wrong type for socket". > > What is wrong? My search for clues turned up nothing pertaining to KEXTs. > > -Allan > > > _______________________________________________ > Do not post admin requests to the list. They will be ignored. > Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) > Help/Unsubscribe/Update your Subscription: > http://lists.apple.com/mailman/options/darwin-kernel/ahoeltje%40lmi.net > > This email sent to ahoeltje(a)lmi.net > _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com

1 0

Re: File level action protocol
by Mike Smith 08 Jul '05

08 Jul '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com

1 0

Including my Filesystem in DiskUtility Format Options
by Thane Heninger 08 Jul '05

08 Jul '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Thanks, Thane Heninger Iomega _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… I placed our filesystem in /System/Library/Filesystems where it seems to function fine for probing and mounting disks. I made reference to an /sbin/newfs_ style tool in the same manner as the /sbin/mount_ with nearly identical FSPersonalities dictionary as is used by the Info.plist of hfs.fs. I thought that would be all that would be required to list our filesystem as an option in the Erase tab of DiskUtility, but it isn't showing up. I can use the filesystem as an option to /usr/sbin/diskutil eraseDisk or eraseVolume and it seems to work correctly. Also, if I supply a dummy name for the format parameter to diskutil, the error message lists my filesystem's name as a "Valid Filesystem". Is there something else I must do to allow my filesystem to be in DiskUtility's "Volume Format" menu? This email sent to site_archiver(a)lists.apple.com

1 0

Connection to KEXT with kern_control
by Allan Hoeltje 07 Jul '05

07 Jul '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com User-agent: Microsoft-Entourage/10.1.4.030702.0 My socket woes continue. I am attempting to open a socket from my client application to my loaded KEXT. My client is using the sample code copied nearly verbatim from page 20 in the "Network Kernel Extensions Programming Guide". The code successfully executes the call to socket( PF_SYSTEM, SOCK_DGRAM, SYSPROTO_CONTROL ) and the call to ioctl() successfully obtains the dynamic ctl_id of my loaded KEXT. However, when I make the call to connect() I get a -1 return and errno contains: "41 Protocol wrong type for socket". What is wrong? My search for clues turned up nothing pertaining to KEXTs. -Allan _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com

1 0

Re: connect or sock_connect?
by Josh Graessley 07 Jul '05

07 Jul '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com -josh On Jul 6, 2005, at 6:09 PM, Justin C. Walker wrote: On Jul 6, 2005, at 17:52 , Allan Hoeltje wrote: I would assume it needs sys/socket.h? User code should use <sys/socket.h> Regards, Justin -- Justin C. Walker, Curmudgeon-At-Large Institute for General Semantics -------- "Weaseling out of things is what separates us from the animals. Well, except the weasel." - Homer J Simpson -------- This email sent to jgraessley(a)apple.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com Justin is correct. Another hint is that sys/kpi_socket.h only appears in the Kernel framework. If you look at sys/socket.h in the kernel framework, you'll find that functions like socket, accept, listen, bind, connect, etc are not declared. When working in the kernel, you use the headers from the Kernel Framework. When working in user space, you use headers from a variety of places but not the kernel framework. The Network Kernel Extensions Programming Guide, section KEXT Controls and Notifications, describes how an application client can communicate with a KEXT via sockets. There is a bare bones code snippet that uses socket, ioctl, connect, and setsockopt calls. It does not say which framework nor which header file to use. However, there is a header sys/kpi_socket.h that defines a socket api and all funtions are prefixed with "sock_". The parameters are slightly different from the functions defined in sys/socket.h. What are these sock_ calls for? Is the NKE doc in error? Or are the sock_ calls for sockets used only within KEXTs and not a client app talking to a KEXT? Anything in the kpi world is for the use of kexts, not for user- mode code. _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/jgraessley% 40apple.com smime.p7s

1 0

Re: connect or sock_connect?
by Justin C. Walker 07 Jul '05

07 Jul '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com On Jul 6, 2005, at 17:52 , Allan Hoeltje wrote: User code should use <sys/socket.h> Regards, Justin -- Justin C. Walker, Curmudgeon-At-Large Institute for General Semantics -------- "Weaseling out of things is what separates us from the animals. Well, except the weasel." - Homer J Simpson -------- _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… The Network Kernel Extensions Programming Guide, section KEXT Controls and Notifications, describes how an application client can communicate with a KEXT via sockets. There is a bare bones code snippet that uses socket, ioctl, connect, and setsockopt calls. It does not say which framework nor which header file to use. I would assume it needs sys/socket.h? However, there is a header sys/kpi_socket.h that defines a socket api and all funtions are prefixed with "sock_". The parameters are slightly different from the functions defined in sys/socket.h. What are these sock_ calls for? Is the NKE doc in error? Or are the sock_ calls for sockets used only within KEXTs and not a client app talking to a KEXT? Anything in the kpi world is for the use of kexts, not for user-mode code. This email sent to site_archiver(a)lists.apple.com

1 0

connect or sock_connect?
by Allan Hoeltje 07 Jul '05

07 Jul '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com User-agent: Microsoft-Entourage/10.1.4.030702.0 The Network Kernel Extensions Programming Guide, section KEXT Controls and Notifications, describes how an application client can communicate with a KEXT via sockets. There is a bare bones code snippet that uses socket, ioctl, connect, and setsockopt calls. It does not say which framework nor which header file to use. I would assume it needs sys/socket.h? However, there is a header sys/kpi_socket.h that defines a socket api and all funtions are prefixed with "sock_". The parameters are slightly different from the functions defined in sys/socket.h. What are these sock_ calls for? Is the NKE doc in error? Or are the sock_ calls for sockets used only within KEXTs and not a client app talking to a KEXT? -Allan _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-kernel mailing list (Darwin-kernel(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-kernel/site_archiver%40lists.… This email sent to site_archiver(a)lists.apple.com

1 0