November 2008 - Darwin-kernel

Where is asl_private.h?
by James C 18 Nov '08

18 Nov '08

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:date:from:to :subject:mime-version:content-type:content-transfer-encoding :content-disposition; bh=zyTzbcIbJk4JONrNgcH7eXhsHqtU6fh+p7ajHHZFQP0=; b=ny2SOGyL4HCl/geKcj7fB2zDSyGgRs5vcUss3e8+b9wLcmp0qlGLzdZWcAeicXxyRk fjrFGHlhmOCuBEhaHyfODZtZqYCA93mk5qrvZYvqcvSdmmj028jNxCY0aPkpopXrkmdd 4BufUJW1ZaZO2PFZghQn3Oj2Oj8TxpAsCKmCA= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:mime-version:content-type :content-transfer-encoding:content-disposition; b=gd/vRWt0f+dlNK/UgpTGYFjj150b5pm6xVui5FpWV68jIPYnxiVuaFHla6vQ6kedRp TPnxZAfRhL3ByclJY7hv8ZVbPlqQ3GSs1wvllZkDwZl9hjkV48fs86HiKSE+gGXaOQbu OtP93gHz/cL7FQ9mLg2AwFD2Wgd/7/UDTe2fQ= Hi All, In this list of Darwin kernel modules: http://www.opensource.apple.com/darwinsource/10.5.5/ the module: syslog-64 depends on a file asl_private.h, but does not contain it. How do I find the module that contains this file? Regards, James. ps I'm trying to build syslogd with debug, to diagnose spinning while accessing asl.db, after shortages of cpu-time, which is suspect is triggered by a race condition. So far, Engineering have classified it as a fault in the most cpu-intensive application that I run, so I think I'll have to help myself. _______________________________________________ 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: Interrupt handling in multi-processor environment
by Rob Cas 17 Nov '08

17 Nov '08

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:date:from:to :subject:in-reply-to:mime-version:content-type :content-transfer-encoding:content-disposition:references; bh=ySFesvJiFBEN5MX3PXO3oB9ZiFz+c7TMKzkJtFVnQqc=; b=u5mqHciP4dIqFt1y/aCaHyJ3ZzP2tY8W+j5uYhE/gAF4cCpggRgdivdiCW1LR/OL7Q Ihzr26QTz1p0PC1epDjFjaVIIRlIg19wGuYaE5lHct/Z51FE4wLpf7rpS5LiAKC+YdXp at2iXqVMjKwQRQ66gcpkvi9I74HvCr8+2Re7k= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:in-reply-to:mime-version :content-type:content-transfer-encoding:content-disposition :references; b=BHFqSt6KiQA2PnZZ+oqZczGE3gIZoEaCQ+FgKUaR7xNNxXy7Yw/bUhxsq917as2Zco NK/za2rJ6cW4eg873gP2Y0NUQzR8PAy1rXAP+3PhK+WyQqHai6E2/RHoloMV/LgSfZkm 2TOwTim69S3V2diGwKRiv/osxmg7VLaiq5BH0= Thanks Terry, One further question: When the sti/cli instruction is executed by IOSimpleLockLockDisableInterrupts(), does it somehow have the effect of disabling interrupts across all CPUs (or specifically the 'interrupt handling CPU'), or just on the CPU executing the code? ie. If we're using IOSimpleLockLockDisableInterrupts(), do we need to somehow make sure we call it while we're running on the 'interrupt handling CPU' to ensure it has the desired effect? Regards, Rob. On Tue, Nov 18, 2008 at 5:17 AM, Terry Lambert <tlambert(a)apple.com> wrote: > Interrupts are all routed to a single CPU. Disable them sparingly, and if > possible, restructure your code so you don't need to disable them at all. > > -- Terry _______________________________________________ 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: Interrupt handling in multi-processor environment
by Terry Lambert 17 Nov '08

17 Nov '08

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com -- Terry On Nov 16, 2008, at 10:28 PM, Rob Cas <sacbor(a)gmail.com> wrote: Hi, I'm trying to understand how IOSimpleLockLockDisableInterrupt works to prevent interrupt preemption in a multi-processor environment. Perhaps somebody could help clarify my (mis)understandings. IOSimpleLockLockDisableInterrupt() appears to disable interrupts (via ml_set_interrupts_enabled()) using the sti/cli instructions which act only on a single CPU. Does this mean that if such a lock is held by a task on the same workloop as an IOFilterInterruptEventSource, it is still possible for an interrupt filter function to be called via an interrupt occurring on a different CPU? Looking at how interrupts are delivered/controlled inside IOKit via IOInterruptControllers/IOSharedInterruptControllers, IOFilterInterruptEventSource, etc it looks like there is just a single 'master' interrupt controller (AppleACPICPUInterruptController) attached to just one of the CPUs that (ultimately) causes interrupts to be dispatched to the filter functions of IOFilterInterruptEventSources. Are interrupts from all CPUs somehow routed through to this single master interrupt controller? (The context of this query is trying to serialise access to registers (including register reads) on a PCI device. An alternative approach that occurs to me (if IOSimpleLockLockDisableInterrupt is not suitable) is to disable the IOFilterInterruptEventSource temporarily whilst the registers are accessed outside of the interrupt filter function, which has the side-effect of calling disableInterrupt() on the shared interrupt controller, thus preventing preemption by the interrupt filter function. I'm not sure whether disabling a shared interrupt controller for very brief times is a good idea though - will it have an adverse impact on other parts of the system sharing those interrupts. ) Thanks in advance, Rob. _______________________________________________ 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.… Interrupts are all routed to a single CPU. Disable them sparingly, and if possible, restructure your code so you don't need to disable them at all. This email sent to site_archiver(a)lists.apple.com

1 0

Re: Interrupt handling in multi-processor environment
by mm w 17 Nov '08

17 Nov '08

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:date:from:to :subject:cc:in-reply-to:mime-version:content-type :content-transfer-encoding:content-disposition:references; bh=ojhDQCOa+sgKmkAqNkqbJM3IobujKE8vKxQACkIgpIY=; b=CoqCttkukhLBGfRiK2p+AqND9Z9bzV4GaAP739d+Ahw7TR1zuuuCgYW7L+b6ZNCk92 dxvtT9QFv7p2EX1SMJztuRwAyN5VSxem7e576lZHa4GGcTiWbngf7jSa8GHvmiCv37Ut qYKRYHz14dF8roqStO9bexsY7RxgCpMLlpmCM= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:cc:in-reply-to:mime-version :content-type:content-transfer-encoding:content-disposition :references; b=p/3oCw7YNrgHIHbKhXpLQBZJRB5mXdzElAms8u62ResSF9dHIQ3hiDkWKX3HEJb5tU DrViKuiPoOlpxTIa6ldD4ZnrvO4AD2OAyq+gGJ+i+BZrSHMhuIqjA1jCe36w+UmNlatH +tyAFw07LIEKWWcdo4JcUpxY0tPkztyxXJ92k= hi Rob, I answered you on this part (it sounds like you are porting existing code, or try to apply foreign concepts to macos) > (The context of this query is trying to serialise access to registers > (including register reads) on a PCI device. i will recommend to read this document carefully http://developer.apple.com/documentation/DeviceDrivers/Conceptual/WritingDe… Cheers! On Sun, Nov 16, 2008 at 10:28 PM, Rob Cas <sacbor(a)gmail.com> wrote: > Hi, > > I'm trying to understand how IOSimpleLockLockDisableInterrupt works to > prevent interrupt preemption in a multi-processor environment. Perhaps > somebody could help clarify my (mis)understandings. > > IOSimpleLockLockDisableInterrupt() appears to disable interrupts (via > ml_set_interrupts_enabled()) using the sti/cli instructions which act > only on a single CPU. > > Does this mean that if such a lock is held by a task on the same > workloop as an IOFilterInterruptEventSource, it is still possible for > an interrupt filter function to be called via an interrupt occurring > on a different CPU? > > Looking at how interrupts are delivered/controlled inside IOKit via > IOInterruptControllers/IOSharedInterruptControllers, > IOFilterInterruptEventSource, etc it looks like there is just a single > 'master' interrupt controller (AppleACPICPUInterruptController) > attached to just one of the CPUs that (ultimately) causes interrupts > to be dispatched to the filter functions of > IOFilterInterruptEventSources. Are interrupts from all CPUs somehow > routed through to this single master interrupt controller? > > (The context of this query is trying to serialise access to registers > (including register reads) on a PCI device. > > An alternative approach that occurs to me (if > IOSimpleLockLockDisableInterrupt is not suitable) is to disable the > IOFilterInterruptEventSource temporarily whilst the registers are > accessed outside of the interrupt filter function, which has the > side-effect of calling disableInterrupt() on the shared interrupt > controller, thus preventing preemption by the interrupt filter > function. I'm not sure whether disabling a shared interrupt controller > for very brief times is a good idea though - will it have an adverse > impact on other parts of the system sharing those interrupts. > ) > > Thanks in advance, > Rob. > _______________________________________________ > 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/openspecies%40gmail.com > > This email sent to openspecies(a)gmail.com > -- -mmw _______________________________________________ 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

Interrupt handling in multi-processor environment
by Rob Cas 17 Nov '08

17 Nov '08

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:date:from:to :subject:mime-version:content-type:content-transfer-encoding :content-disposition; bh=BnGISGZTTh2/ROYuuUWzgUdUp4rSHR9GfHAkLg/HiSM=; b=X0g7ax/RRl6/hwwWuwl+gzSLgsSizZ2joUJS8mv404w5Zvm1joXf8XWY3askOPdAWO CxIZ+/PfDigh1Xl/38V1Baf9SCD54WcLjmHW8KVCvH4JDbsSPzdq80o1wdxr3LA2oPxK P5juKNOsN5wnlaZ8OA8VuJrl9gx6PhYixB2eU= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:mime-version:content-type :content-transfer-encoding:content-disposition; b=OFf++/bEEDRhqs8vCu/ZlNXuMpwC7UBz1ehi8DOD0m3b2rPqpMyNGygP4DLrhsLLY4 EjqrCIT/1w1oyWhUTu/n2jMsHDuMFYxETgAUlnH4753DsxnfxtK4mR87nUNV7a6gb+9C 6610vyk6cP376QPPRMp+j2oxlDiET+SAVXa0s= Hi, I'm trying to understand how IOSimpleLockLockDisableInterrupt works to prevent interrupt preemption in a multi-processor environment. Perhaps somebody could help clarify my (mis)understandings. IOSimpleLockLockDisableInterrupt() appears to disable interrupts (via ml_set_interrupts_enabled()) using the sti/cli instructions which act only on a single CPU. Does this mean that if such a lock is held by a task on the same workloop as an IOFilterInterruptEventSource, it is still possible for an interrupt filter function to be called via an interrupt occurring on a different CPU? Looking at how interrupts are delivered/controlled inside IOKit via IOInterruptControllers/IOSharedInterruptControllers, IOFilterInterruptEventSource, etc it looks like there is just a single 'master' interrupt controller (AppleACPICPUInterruptController) attached to just one of the CPUs that (ultimately) causes interrupts to be dispatched to the filter functions of IOFilterInterruptEventSources. Are interrupts from all CPUs somehow routed through to this single master interrupt controller? (The context of this query is trying to serialise access to registers (including register reads) on a PCI device. An alternative approach that occurs to me (if IOSimpleLockLockDisableInterrupt is not suitable) is to disable the IOFilterInterruptEventSource temporarily whilst the registers are accessed outside of the interrupt filter function, which has the side-effect of calling disableInterrupt() on the shared interrupt controller, thus preventing preemption by the interrupt filter function. I'm not sure whether disabling a shared interrupt controller for very brief times is a good idea though - will it have an adverse impact on other parts of the system sharing those interrupts. ) Thanks in advance, Rob. _______________________________________________ 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: ACL handling for NFSv4
by Rick Macklem 14 Nov '08

14 Nov '08

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com On Fri, 14 Nov 2008, Quinn wrote: What are the guid_t owner and group fields of a kauth_filesec and is there a way a client can map those to/from uid/gid or do they have to be stored/retrieved from the file system? Yes, that does look like it'll be useful. Thanks a lot. (I now see that it checks a cache and calls a userland daemon upon a miss. I also see that dsmemberutil(1) knows how to figure these out based upon a user or group.) As an aside, NFSv4 ACLs look just like NT4 ACLs, which appear to look just like Leopard ACLs, so the only trick seems to be mapping the "who" (called ace_applicable on Leopard). It looks like the above, plus kauth_wellknown_guid() will do that. (I haven't found an inverse function to kauth_wellknown_guid() that creates one, but I can easily make one up.) In NFSv4, ACLs are checked/enforced by the server, so I don't see that as an issue for the client. So, the Get ACL case looks ok, but that still leaves the Set ACL case as a bit of an issue. It looks to me that it is always done via vnode_set_filesec() which ends up doing a VNOP_SETXATTR() for the filesec. Which finally gets me to the questions: Can I get away with ignoring the fsec_owner and fsec_group fields, since I have no way of storing them on the server? Does setting these imply a change in file ownership? Thanks again, rick _______________________________________________ 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.… At 15:38 -0500 13/11/08, Rick Macklem wrote: <sys/kauth.h> has KPIs to map from GUIDs to UID/GID (kauth_cred_guid2uid, kauth_cred_guid2gid) and vice versa (kauth_cred_uid2guid, kauth_cred_gid2guid). While I'm not an NFSv4 expert, so I really have no idea whether this is the right thing to do, I suspect you might find them useful (-: If anyone happens to have answers to these, thanks in advance for them, but in any case, with Quinn's help, it looks like I can at least Get ACLs ok. (Great, I like it:-) This email sent to site_archiver(a)lists.apple.com

1 0

Re: ACL handling for NFSv4
by Quinn 14 Nov '08

14 Nov '08

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com What are the guid_t owner and group fields of a kauth_filesec and is there a way a client can map those to/from uid/gid or do they have to be stored/retrieved from the file system? S+E -- Quinn "The Eskimo!" <http://www.apple.com/developer/> Apple Developer Relations, Developer Technical Support, Core OS/Hardware _______________________________________________ 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.… At 15:38 -0500 13/11/08, Rick Macklem wrote: <sys/kauth.h> has KPIs to map from GUIDs to UID/GID (kauth_cred_guid2uid, kauth_cred_guid2gid) and vice versa (kauth_cred_uid2guid, kauth_cred_gid2guid). While I'm not an NFSv4 expert, so I really have no idea whether this is the right thing to do, I suspect you might find them useful (-: This email sent to site_archiver(a)lists.apple.com

1 0

ACL handling for NFSv4
by Rick Macklem 13 Nov '08

13 Nov '08

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Looking at xnu-1228 sources, it seems that getting an ACL from a file system is done via vnode_getattr(), but is set via vn_setxattr() setting KAUTH_FILESEC_XATTR. Is that correct? (Or, does the vfs always set the kauth_filesec at the same time it sets an ACL?) Part of my problem is that I don't have extended attribute support in the server, so I don't have a way to store/retrieve the guid_t owner and group fields in kauth_filesec on the server, which leads me to... What are the guid_t owner and group fields of a kauth_filesec and is there a way a client can map those to/from uid/gid or do they have to be stored/retrieved from the file system? (If I have to store/retrieve them, I suspect I'm snookered.) Thanks in advance for any information on this, rick _______________________________________________ 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 finally have an NFSv4 server that includes support for NFSv4 ACLs, so I was hoping to tweak my NFSv4 Leopard client (http://code.google.com/p/macnfsv4) to make use of them. (Note this is an experimental open source project using a Kext that already goes beyond the KPIs, so I realize that the internal interfaces may change at any time...) This email sent to site_archiver(a)lists.apple.com

1 0

Re: choosing kernel on intel?
by Andrew Gallatin 04 Nov '08

04 Nov '08

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=simple/simple; d=cs.duke.edu; s=mail; t=1225833328; bh=4QAsUdhZstPL1vk9hg3YYOoLj1cD08kotSIud6ZH0Yg=; h=Message-ID:Date:From:MIME-Version:To:CC:Subject:References: In-Reply-To:Content-Type:Content-Transfer-Encoding; b=UJOV1F8PO/jr hI4FnXI37waoWYPvvvzbtHZRFAXq/oceH0J4+3rBlZ0szlb5yB8IIBUgICAC0HyB7uu RzNYGyqFyPJUDZQzeHVtaodVExjaYORKVP/b9J/CWeHEnZhq3z5IaR/2iFtFjslszku Poq6ZIYiPAaoLjlT35SoniYKQ= User-agent: Thunderbird 2.0.0.16 (X11/20080724) BTW, do you know if there are any EFI drivers for USB->serial dongles? I hate having to have a KVM around just to choose boot devices, and I'd love to control a MacPro via a serial console the same way I do all my generic x86 test boxes. 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.… Shantonu Sen wrote: You can get an EFI shell from a few places, for instance rEFIt (<http://refit.sourceforge.net>) or tianocore.org. I wouldn't recommend it, but if you go this approach, you can do something like: fs2:/> \System\Library\CoreServices\boot.efi Kernel=mach_kernel.test This email sent to site_archiver(a)lists.apple.com

1 0

Re: choosing kernel on intel?
by Andrew Gallatin 04 Nov '08

04 Nov '08

site_archiver(a)lists.apple.com Delivered-To: darwin-kernel(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=simple/simple; d=cs.duke.edu; s=mail; t=1225832939; bh=1yyVeq7my8NJxxfGxdaNE8B6GyoNuD17ZbsIG70jAgU=; h=Message-ID:Date:From:MIME-Version:To:CC:Subject:References: In-Reply-To:Content-Type:Content-Transfer-Encoding; b=IPgUWoIsiX1S f613bjCL9Az+mq9fe0MwtM4nJpk3KmAe3mcbEo6mo6LSH4jSCMsNY1lj2fLN+wvP1QI aHX3mC1Zx9/mT/R7UD7jJZJEstx4pKHdRqvlY3D+aS4xJSToT1pfzEy7X6RzRtuXLg9 JVVzfkq/L/KyvIxwcA73ze8Ac= User-agent: Thunderbird 2.0.0.16 (X11/20080724) Thanks.. that seemed to work. 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.… Shantonu Sen wrote: You can get an EFI shell from a few places, for instance rEFIt (<http://refit.sourceforge.net>) or tianocore.org. I wouldn't recommend it, but if you go this approach, you can do something like: fs2:/> \System\Library\CoreServices\boot.efi Kernel=mach_kernel.test This email sent to site_archiver(a)lists.apple.com

1 0