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

IOStorage synchronous write routine crashing
by Yogesh P 05 Jul '05

05 Jul '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Hi Folks, I am using a synchronous write call in filter scheme driver. I have created a IOMemoryDescriptor buffer by calling withAddress function and pass this buffer to the sync write call, but i am getting crash when this write routine gets called. My code snippet for the same is as follows: void com_My_driver::write(IOService* provider, UInt64 byteStart, IOMemoryDescriptor* buffer, IOStorageCompletion completion) { // Allocate a buffer to write the data on disk. UInt32 WriteBufSize = 512; char * bufWrite = new char[512]; strcpy(bufWrite, "DRIVER ON MACOS X"); // Create memory descriptor for the buffer IOMemoryDescriptor *WriteBuffer = NULL; WriteBuffer = IOMemoryDescriptor::withAddress((void*) bufWrite, (IOByteCount) WriteBufSize, kIODirectionOut); if(WriteBuffer) { filteredMedia->write(this, 512, WriteBuffer); /* crashing just here and getting memory access exception*/ } else { WriteBuffer->release(); WriteBuffer = 0; } Is it i am missing something while using the sync write call? I am able to make the code work if I do memory allocations in User Client and successfully do the write() on disk. But not able to do the same if I don’t use user client and do allocation in filter scheme driver. I think I am not using memory management functions correctly. Is there any documentation available,how to use IOMemoryDescriptor in kernel level drivers? I have read KernelProgramming.pdf, WritingDeviceDrivers.pdf but I didn’t find it useful in my case. Look forward to pointers so that I can solve this issue. Thanks in advance, Yogesh Pahilwan _______________________________________________ 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

File Close Event on Mac OS X
by Imran Kazi 05 Jul '05

05 Jul '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Thread-index: AcWBbJ+2I+gOjjVXRSOr3o/0ciu44w== _______________________________________________ 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 read/write routine in Driver Code
by William Kucharski 05 Jul '05

05 Jul '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com On Jul 5, 2005, at 3:12 AM, santosh katvate wrote: Hello everybody how to do file read within driver code(i.e SampleFilterScheme.kext). As i want to read some config data from that file. You really, really don't want to read a file from within the kernel. Instead, have a user level program read and parse the data and pass the resulting data into the kernel. William Kucharski kucharsk(a)mac.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

file read/write routine in Driver Code
by santosh katvate 05 Jul '05

05 Jul '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:reply-to:to:subject:mime-version:content-type:content-transfer-encoding:content-disposition; b=H4Dct70D146sJ7hpor1TthvB88/i+TLlihzZhMdtC6emX0T1wNoiKJwMSb/DeotDGL6gmXQH9tXFlsxRmjDqI8o7hTMwpRHFHUd5NCGIeSdpNet+kiX6Fn9X2xrxo8B/Yb/MhJVX8UczlFL7pzoRidB70nnFivkPLkWICmCA5LA= Hello everybody how to do file read within driver code(i.e SampleFilterScheme.kext). As i want to read some config data from that file. Thanks in Advance santosh k _______________________________________________ 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: enetlognke and KEXT questions
by Allan Hoeltje 04 Jul '05

04 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 Raj, Thank you for that information! The only reason why I thought the pkthdr_header had to be set was the comment in enetlognke that mentions "system panic". Maybe it is only a concern with Interface filters. -Allan "Raj Doshi" <admin(a)tacosw.com> wrote on 7/3/05 6:30 PM: > Hi, > > I implemented an IP filter with injection like you are describing > without ever modifying mbuf_pkthdr_header. Even when I made a copy > of an mbuf and injected the copy instead of the original, I did not > set the mbuf_pkthdr_header (although there are other fields that need > to be set like checksum requested/performed in this case). So, I > think the answer is no, you do not need to worry about > mbuf_pkthdr_header in an IP filter. You should be able to inject a > filtered mbuf, as is. > > Raj Doshi > > On Jul 1, 2005, at 8:27 PM, Allan Hoeltje wrote: > >> First off, thank you Quinn and Josh for the pointers you have given >> me so >> far. >> >> Second, I need some more pointers! >> >> I have been testing the enetlognke interface filter sample and >> mundged it >> into what I hope is the beginning of an IP filter. >> >> The problem I am running into is that the mbuf_t data passed to my >> ipf_input_func (and ipf_output_func) always has a NULL >> mbuf_pkthdr_header. >> >> For enetlognke this is OK since the interface filter >> enet_input_func is >> passed a reference to a frame header. There is even a comment that >> says: >> "Before we swallow the packet, we need to make sure that the very >> first mbuf >> has the pkthdr_header field set, else the system will panic when we >> inject >> the packet. The input routine is passed the frame_header pointer, >> so we can >> use this value to set as the pkthdr_header." >> >> Is there some api I can use to get the frame header in my >> ipf_input_func? >> Does any of this matter for an IP filter? >> >> Sorry if this is a dumb question but there is very little >> documentation and >> even less sample code for mbufs and kernel extensions. >> >> Help! >> >> -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 William Kucharski 04 Jul '05

04 Jul '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com 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. 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. William Kucharski kucharsk(a)mac.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

Re: enetlognke and KEXT questions
by Raj Doshi 04 Jul '05

04 Jul '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Hi, Raj Doshi On Jul 1, 2005, at 8:27 PM, Allan Hoeltje wrote: Second, I need some more pointers! Help! This email sent to admin(a)tacosw.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.… I implemented an IP filter with injection like you are describing without ever modifying mbuf_pkthdr_header. Even when I made a copy of an mbuf and injected the copy instead of the original, I did not set the mbuf_pkthdr_header (although there are other fields that need to be set like checksum requested/performed in this case). So, I think the answer is no, you do not need to worry about mbuf_pkthdr_header in an IP filter. You should be able to inject a filtered mbuf, as is. First off, thank you Quinn and Josh for the pointers you have given me so far. I have been testing the enetlognke interface filter sample and mundged it into what I hope is the beginning of an IP filter. The problem I am running into is that the mbuf_t data passed to my ipf_input_func (and ipf_output_func) always has a NULL mbuf_pkthdr_header. For enetlognke this is OK since the interface filter enet_input_func is passed a reference to a frame header. There is even a comment that says: "Before we swallow the packet, we need to make sure that the very first mbuf has the pkthdr_header field set, else the system will panic when we inject the packet. The input routine is passed the frame_header pointer, so we can use this value to set as the pkthdr_header." Is there some api I can use to get the frame header in my ipf_input_func? Does any of this matter for an IP filter? Sorry if this is a dumb question but there is very little documentation and even less sample code for mbufs and kernel extensions. -Allan (Everybody starts out as a newbie and it is only through trial, error, and good documentation do they become experts. - ALH) _______________________________________________ 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/admin% 40tacosw.com This email sent to site_archiver(a)lists.apple.com

1 0

enetlognke and KEXT questions
by Allan Hoeltje 02 Jul '05

02 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 First off, thank you Quinn and Josh for the pointers you have given me so far. Second, I need some more pointers! I have been testing the enetlognke interface filter sample and mundged it into what I hope is the beginning of an IP filter. The problem I am running into is that the mbuf_t data passed to my ipf_input_func (and ipf_output_func) always has a NULL mbuf_pkthdr_header. For enetlognke this is OK since the interface filter enet_input_func is passed a reference to a frame header. There is even a comment that says: "Before we swallow the packet, we need to make sure that the very first mbuf has the pkthdr_header field set, else the system will panic when we inject the packet. The input routine is passed the frame_header pointer, so we can use this value to set as the pkthdr_header." Is there some api I can use to get the frame header in my ipf_input_func? Does any of this matter for an IP filter? Sorry if this is a dumb question but there is very little documentation and even less sample code for mbufs and kernel extensions. Help! -Allan (Everybody starts out as a newbie and it is only through trial, error, and good documentation do they become experts. - ALH) _______________________________________________ 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: Problem on Mac 10.4 Tiger with the drivers porting
by Karunakar Reddy G 01 Jul '05

01 Jul '05

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:reply-to:to:subject:cc:in-reply-to:mime-version:content-type:references; b=tUmBacdS5dKCN0QdU3BNSuJSmte+dlSxZVWwf/c/b+ECagUCSI6aFKLgNTBEH/pMWM+ZcA23wH0p0zZg+bzZQtsEkxnMylUqpnKGchUPsIqHk5j/GP+3s/2BiCJTzqfpMkWlWHTPWOKBNr/NNlFQkUt7vv4oM39rp28v6kcrpvY= _______________________________________________ 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