June 2006 - Darwin-kernel - lists.apple.com

Boundary communication using IOMemory* and IOMemoryDescriptor*. Comments awaited.
by rohit dhamija 12 Jun '06

12 Jun '06

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:to:subject:mime-version:content-type; b=S5d8B+OY2Km5bzIGSStaSxrHXeyEmN20TJthr0xqriODncVk0A0Ta4Dy3v6hbzrLaLMa4CJl6LLy5HotX+MwZjUNDPNFY12qH7YRp65kcctlGcNC2IvBoREI3tmDvFfWVgJhqF/th9p0gaTJLLpsfJMpGlnw5PUY7z+jRAeMXd4= _______________________________________________ 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:mbuf data access (repost)
by jy.cases＠mac.com 12 Jun '06

12 Jun '06

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com On 7 juin 06, at 19:26, Josh Graessley wrote: Is there a way to find out if those mbufs are shared ? Jean-Yves _______________________________________________ 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.… mbuf_len like mbuf_data works whether or not MBUF_EXT is set. MBUF_EXT just indicates that the actual implementation of the mbuf is storing the data in an external cluster and not in the mbuf data structure itself. The one reason this would be important to a client of the mbuf is the case where you may modify data. In some cases, the reference to the external cluster may be shared. If this is the case, you will need to duplicate the data before modifying it. Duplicate data means duplicate mbuf or the whole chain, or only the data returned with mbuf_datastart ? What method should i use : mbuf_dup, mbuf_copym or mbuf_copydata ? In my code (in sf_data_in and sf_data_out) , i check out if mbuf was tagged, in this case filter simply return, otherwise i set a tag for the mbuf and push it in a queue for re- injection. When timer is fired mbuf is retrieve from queue and the mbuf is processed. So, do i need to duplicate mbuf before tag ? before push it in queue ? or before re-injection ? This email sent to site_archiver(a)lists.apple.com

1 0

Re: manual network configuration problem
by Michael Smith 11 Jun '06

11 Jun '06

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com = Mike _______________________________________________ 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.… On Jun 11, 2006, at 4:54 PM, proclus(a)gnu-darwin.org wrote: Any clues for the helpless? Yes. Don't spam irrelevant lists. Don't crosspost between lists, period. This email sent to site_archiver(a)lists.apple.com

1 0

manual network configuration problem
by proclus＠gnu-darwin.org 11 Jun '06

11 Jun '06

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com This might be a bug, so I am sending it to a few relevant lists. I have an x86 machine running Darwin-8.0.3. I can get the ethernet interface configured, but the resolver isn't working. Although the 'host' command works, no other apps can get IP numbers, and they all give dns lookup errors. Things work fine in DHCP mode, but I need it to work in MANUAL mode as well. Any clues for the helpless? Netstat and ifconfig utilities give proper indications that the interface is up and configured correctly. I can ssh to and from the machine. The /etc/hosts file is read by ssh, but it cannot find outside internet addresses. I have a correct resolv.conf file, which works fine on my Darwin-7.2.1 machine, but not on the 8.0.3 machine. I also tested the 8.0.1 kernel to no avail. I can ping and connect by IP address, but not by name. I was thinking that this DNS issue could be related to the lookup order for lookupd. Could someone please suggest a lookup order where NetInfo and DNS both work, which can be set manually. I know how to set this with nicl I think, if you will just give me then lookupd names in proper order. Thank you. Regards, proclus http://www.gnu-darwin.org/ Regards, proclus http://www.gnu-darwin.org/ -- Visit proclus realm! http://proclus.tripod.com/ -----BEGIN GEEK CODE BLOCK----- Version: 3.1 GMU/S d+@ s: a+ C++++ UBULI++++$ P+ L+++(++++) E--- W++ N- !o K- w--- !O M++@ V-- PS+++ PE Y+ PGP-- t+++(+) 5+++ X+ R tv-(--)@ b !DI D- G e++++ h--- r+++ y++++ ------END GEEK CODE BLOCK------ _______________________________________________ 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 pgps02Ld0F3wn.pgp

1 0

Re: kext loading issues on Mac OS 10.3.9, but works fine on Mac OS X 10.4 ppc and intel. Why?
by Mike Smith 11 Jun '06

11 Jun '06

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com On Jun 11, 2006, at 10:33 AM, rohit dhamija wrote: You should not be using copyin in an I/O Kit kext. = Mike _______________________________________________ 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 am developing kernel extension for my USB bulk and control devices for "Mac OS X 10.4 powerpc and Intel Mac. and Mac OS X 10.3.9 powerpc" 1. I had used following key to use "copyin" function in my kext code. Assuming this is the same exercise you were discussing previously, please re-examine the example I gave, which clearly demonstrates how to use an IOMemoryDescriptor to do I/O against user-owned memory. I note in one of your previous replies, you refer to dynamically allocating the command structure; this is probably also a pointless waste of time, since you can for about the same cost just construct the user command inside a fixed-size buffer already associated with the command. This email sent to site_archiver(a)lists.apple.com

1 0

kext loading issues on Mac OS 10.3.9, but works fine on Mac OS X 10.4 ppc and intel. Why?
by rohit dhamija 11 Jun '06

11 Jun '06

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:to:subject:mime-version:content-type:content-transfer-encoding:content-disposition; b=XYYdlw9x/ylQbQgT+xCr/fFcR/xboISe/fFEniNcrx/6aQCW5tyoYtxZF5cldDt6XCk+fuJR6L9yQc3pW/NsdTIC8PG9N37AxqAQ++Zx+4/zJFlG5lgP0wo3jTNeOekPNXtePF37KSq2qYpwdqLlaorR2N2xvGKBMQc6QIgIm48= Dear All, I am developing kernel extension for my USB bulk and control devices for "Mac OS X 10.4 powerpc and Intel Mac. and Mac OS X 10.3.9 powerpc" 1. I had used following key to use "copyin" function in my kext code. copyin() is exposed in kpi which is available for Mac OS X 10.4 and above. <key>com.apple.kpi.libkern</key> <string>8.0.0</string> But I require to run my kext on Mac OS X 10.3.9 too. So is there any alternative to copyin function ?? Or alternative to com.apple.kpi.libkern. I know that kpi's are introduced for Mac OS 10.4 and above Please suggest. 2. The kext works fine on 10.4 but throws following error on Mac OS 10.3.9. Please note that I have already commented com.apple.kpi.libkern and the corresponding copyin function, but still my kext throws following error whiel loading on 10.3.9:: kextload: extension myDriver.kext appears to be valid kextload: notice: extension myDriver.kext has debug properties set 18myDriverUserClient is not compatible with its superclass, 12IOUserClient superclass changed? kextload: a link/load error occured for kernel extension myDriver.kext load failed for extension myDriver.kext (run kextload with -t for diagnostic output) Why is myDriverUserClient : IOUserClient not compatible on Mac OS X 10.3.9, although the same works great on Mac OS X 10.4 ? Did any body encountered similiar situation ?? Please help. Thanks and Regards, Rohit Dhamija _______________________________________________ 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

launch agent, non-Admin user, CoreMIDI/bootstrap namespace errors
by Alex Sheh 10 Jun '06

10 Jun '06

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Thread-index: AcaMMCRhXUdzoq2OS6Gx9ku+NkRmgQ== Thread-topic: launch agent, non-Admin user, CoreMIDI/bootstrap namespace errors _______________________________________________ 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: [OT] WOA URLs [was: Re: dlil_output_list and interface filters]
by Anton Kuzmin 10 Jun '06

10 Jun '06

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com User-agent: Thunderbird 1.5.0.4 (Windows/20060516) Anton Cheers [apologies for being preachy], _______________________________________________ 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.… Yes, I see it now in my sent folder. It looks like a bug in Thunderbird - I just copied the ids from the bugreporter in Firefox into Thunderbird and somehow it managed to include their urls, even though the message was plain text and in email composer it only displayed ids to me without links. Graham J Lee wrote: On 10/6/06 00:45, Anton Kuzmin wrote: <https://bugreport.apple.com/cgi-bin/WebObjects/RadarWeb.woa/9/wo/XMN4cdKQjp…> This isn't a personal jab, I see this in emails quite often and just want to spread the word. WebObjects URLs which contain a "/wo/" like the above are not transferrable to other users, as they depend on your current session (the SID is contained in the URL). So you shouldn't paste them into mail or whatever, as no-one else will be able to access them anyway. In the case of Radar, the bug ID is sufficient for those people who have access to it. Graham. This email sent to site_archiver(a)lists.apple.com

1 0

Re: dlil_output_list and interface filters
by Anton Kuzmin 09 Jun '06

09 Jun '06

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com User-agent: Thunderbird 1.5.0.4 (Windows/20060516) Anton Hi Anton, You should file bugs for all of the issues you came across. -josh On Jun 9, 2006, at 2:59 PM, Anton Kuzmin wrote: Anton Thanks, -josh On Jun 9, 2006, at 12:20 PM, Anton Kuzmin wrote: _______________________________________________ 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.… Josh, thanks for looking into this, I always file reports when I come across kernel bugs, it just that sometimes I want to confirm first that something is a bug and not a "feature" that mere mortals cannot comprehend :-) Anyway, just FYI the bug ids are 4581001 <https://bugreport.apple.com/cgi-bin/WebObjects/RadarWeb.woa/9/wo/XMN4cdKQjp…> and 4581051. <https://bugreport.apple.com/cgi-bin/WebObjects/RadarWeb.woa/9/wo/XMN4cdKQjp…> Josh Graessley wrote: I could respond and say that, from looking at the code, it looks like the re-factoring fixes those bugs too but I don't know when it will make it out. That's all speculation on my part with no official documentation to track it. Bug reports act as that documentation ensuring that stuff does get taken care of. I can't comment on whether or not a specific fix will make it in to a software update. There are things that make it more likely that a fix will make it in to a software update. There must be a bug, preferably filed by someone outside of Apple. Contacting DTS to escalate the bug may also help. Thanks, I'll file a report later today, and what about the second question, about the way it's designed to free the whole list if a single packet fails or gets filtered? Or is that also a design flaw and going to be changed in the next release? Also, is there any chance of fixing this in 10.4.7 update? We've got some pretty unhappy customers due to this and I couldn't find any straightforward workaround. Josh Graessley wrote: That is definitely a bug. Please file a bug report. It does look like this is fixed in the sources for the next major release (part of some re-factoring), but it doesn't look like there were any changes made for a software update fix. I've been trying to nail down a mysterious kernel panic that occurs when our KPI interface filter is enabled and stumbled across a possible reason for it: in dlil_output_list after a call to filt_output, a check for EJUSTRETURN leads to what seems to be an erroneous "continue", which skips to the next interface filter instead of bailing out. So basically any packet consumed by an interface filter still gets passed to if_output and freed, which leads to a nasty surprise for an unsuspecting filter still referencing the freed mbuf. Could someone do a sanity check on this and if it's indeed a bug, I'll file a radar. Also, I would be grateful if someone could explain why a failure to process a single packet from a list should cause freeing of the whole list. This email sent to site_archiver(a)lists.apple.com

1 0

Re: dlil_output_list and interface filters
by Josh Graessley 09 Jun '06

09 Jun '06

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Hi Anton, You should file bugs for all of the issues you came across. -josh On Jun 9, 2006, at 2:59 PM, Anton Kuzmin wrote: Anton Thanks, -josh On Jun 9, 2006, at 12:20 PM, Anton Kuzmin wrote: _______________________________________________ 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 I could respond and say that, from looking at the code, it looks like the re-factoring fixes those bugs too but I don't know when it will make it out. That's all speculation on my part with no official documentation to track it. Bug reports act as that documentation ensuring that stuff does get taken care of. I can't comment on whether or not a specific fix will make it in to a software update. There are things that make it more likely that a fix will make it in to a software update. There must be a bug, preferably filed by someone outside of Apple. Contacting DTS to escalate the bug may also help. Thanks, I'll file a report later today, and what about the second question, about the way it's designed to free the whole list if a single packet fails or gets filtered? Or is that also a design flaw and going to be changed in the next release? Also, is there any chance of fixing this in 10.4.7 update? We've got some pretty unhappy customers due to this and I couldn't find any straightforward workaround. Josh Graessley wrote: That is definitely a bug. Please file a bug report. It does look like this is fixed in the sources for the next major release (part of some re-factoring), but it doesn't look like there were any changes made for a software update fix. I've been trying to nail down a mysterious kernel panic that occurs when our KPI interface filter is enabled and stumbled across a possible reason for it: in dlil_output_list after a call to filt_output, a check for EJUSTRETURN leads to what seems to be an erroneous "continue", which skips to the next interface filter instead of bailing out. So basically any packet consumed by an interface filter still gets passed to if_output and freed, which leads to a nasty surprise for an unsuspecting filter still referencing the freed mbuf. Could someone do a sanity check on this and if it's indeed a bug, I'll file a radar. Also, I would be grateful if someone could explain why a failure to process a single packet from a list should cause freeing of the whole list. smime.p7s

1 0