March 2003 - Darwin-kernel - lists.apple.com

Re: Non-executable stack
by Jim Magee 31 Mar '03

31 Mar '03

On Sunday, March 30, 2003, at 7:48 PM, Brian Tabone wrote: I read the MMU overview for the PowerPC and it seems that the MMU supports fairly granular execute privileges (IE per page versus per segment), which should make implementing the non-execute stack (and perhaps non-execute data segment) simpler. I think you need to read that again. The effect of the no-execute bit is discussed on a page basis, but the bit itself is only settable at the segment level. So, you have to give up 256 MB section(s) of your address space to non-execute status, and you have to assure all stacks are in that range. That was deemed too restrictive at the time. --Jim _______________________________________________ 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

Non-executable stack
by Brian Tabone 31 Mar '03

31 Mar '03

List, I searched the archives and could not find any reference to a non-executable stack discussion. Has their been any consideration given to setting the process level stacks to non-executable to help circumvent or at least reduce the problem of buffer overflow exploits. I know that OpenBSD has some work in this area, and was curious if Darwin was looking in this direction as well. Are their any gotcha's in either the GNU compiler or dynamic library loader mechanism that would make a non-executable stack difficult to do in Darwin? I read the MMU overview for the PowerPC and it seems that the MMU supports fairly granular execute privileges (IE per page versus per segment), which should make implementing the non-execute stack (and perhaps non-execute data segment) simpler. -Brian Tabone _______________________________________________ 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: libkld.a error
by Quinn 30 Mar '03

30 Mar '03

At 20:14 +1000 30/3/03, Wade Tregaskis wrote: As a pre-emptive to the inevitable response: Thus, this response must not be inevitable (-: Why don't you just not upgrade to a new system update until its corresponding Darwin sources show up? All of my systems are still running 10.2.3. S+E -- Quinn "The Eskimo!" <http://www.apple.com/developer/> Apple Developer Technical Support * Networking, Communications, Hardware _______________________________________________ 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: libkld.a error
by Wade Tregaskis 30 Mar '03

30 Mar '03

[ updating libsa.o RELEASE ] /usr/bin/ld -static -r -arch ppc -o libsa.o bootstrap.cpo catalogue.cpo c++rem3.o kld_patch.o kmod.cpo mach.o misc.o mkext.o vers_rsrc.o bsearch.o malloc.o sort.o strrchr.o strstr.o setjmp.o vers.o /usr/local/lib/libkld.a /usr/bin/ld: table of contents for archive: /usr/local/lib/libkld.a is out of date; rerun ranlib(1) (can't load from it) make[6]: *** [libsa.o] Error 1 make[5]: *** [build_all] Error 2 make[4]: *** [do_all] Error 2 make[3]: *** [build_all] Error 2 make[2]: *** [build_all] Error 2 make[1]: *** [build_all] Error 2 make: *** [all] Error 2 I am trying to compile the latest version of xnu kernel, and I get this error. What is wrong? >From memory this error is self-explanatory - the library libkld.a, in /usr/local/lib, is too old (or possibly too new, if you're probably building a 10.2.x kernel on 10.2.4). Apply ranlib to it ("ranlib /usr/local/lib/libkld.a" might do it). At least, that's what I remember doing to overcome that problem. I'd be interested, if you are building a 10.2.x kernel on 10.2.4, to hear how well it goes. All my attempts have indicated you can't do any kernel development with MacOS X, because the Apple CVS sources are always at least one minor version behind the current release. As a pre-emptive to the inevitable response: I'd love to install Darwin on another partition so as to do kernel stuff, but apart from the fact my hard drive is full of things I need [and would lose in a repartition & format], the stuff I want to do relates directly to userland OS X apps. :( Wade Tregaskis _______________________________________________ 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

libkld.a error
by Luke Stempien 30 Mar '03

30 Mar '03

[ updating libsa.o RELEASE ] /usr/bin/ld -static -r -arch ppc -o libsa.o bootstrap.cpo catalogue.cpo c++rem3.o kld_patch.o kmod.cpo mach.o misc.o mkext.o vers_rsrc.o bsearch.o malloc.o sort.o strrchr.o strstr.o setjmp.o vers.o /usr/local/lib/libkld.a /usr/bin/ld: table of contents for archive: /usr/local/lib/libkld.a is out of date; rerun ranlib(1) (can't load from it) make[6]: *** [libsa.o] Error 1 make[5]: *** [build_all] Error 2 make[4]: *** [do_all] Error 2 make[3]: *** [build_all] Error 2 make[2]: *** [build_all] Error 2 make[1]: *** [build_all] Error 2 make: *** [all] Error 2 I am trying to compile the latest version of xnu kernel, and I get this error. What is wrong? Thanks :) _______________________________________________ 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: How many number of cpus can Darwin suppot?
by Justin Walker 26 Mar '03

26 Mar '03

On Tuesday, Mar 25, 2003, at 19:13 US/Pacific, Wu Qingbo wrote: Hi, all, I am new to this mail list. I do not know how many cpus Darwin can support for SMP system? If someone knows, please give me a reply. Thanks in advance! Nominally, Darwin will support the number of CPUs present in your system. There are some reality checks to consider, though: - On PowerPC, I don't believe it's been run with more than two CPUs - On *86, I don't believe it's been run with more than one CPU Corrections welcome. Regards, Justin -- /~\ The ASCII Justin C. Walker, Curmudgeon-at-Large \ / Ribbon Campaign X Help cure HTML Email / \ _______________________________________________ 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

How many number of cpus can Darwin suppot?
by Wu Qingbo 26 Mar '03

26 Mar '03

Hi, all, I am new to this mail list. I do not know how many cpus Darwin can support for SMP system? If someone knows, please give me a reply. Thanks in advance! Barry _______________________________________________ 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: Disk Scheduling
by Justin Walker 25 Mar '03

25 Mar '03

On Tuesday, Mar 25, 2003, at 14:16 US/Pacific, Shawn Erickson wrote: On Tuesday, March 25, 2003, at 12:06 PM, Joey Echeverria wrote: I'm talking about the low level scheduling of writing logical blocks out to disk. For example, FIFO, SSTF, LOOK, C-LOOK, etc. The reason that I'm interested is that we are looking into power efficiency in operating systems. One thing we wanted to investigate is if there is any effect to the energy usage when using different algorithms. One idea that we had was to be able to queue requests for longer in an operating system buffer and only write to disk when the queue becomes a certain size. We were thinking that under such a scheme, you could keep the disk stopped for longer periods of time to save the energy from keeping it spinning. To my knowledge Mac OS X does no buffering of writes to disk, I think that the kernel actually does buffer writes to disk, in the "unified buffer cache", using variations on 'classical' Unix cache-management schemes. Depending on what the data is, and how it's being written, it could be a while before the data ends up on the disk. There are, however, no scheduling mechanisms of the sort referred to above. I believe the reasoning is that, these days, the OS does not have a good idea of the actual disk geometry and other performance characteristics, and so we leave that level of scheduling up to the controllers in the disk drive itself. I think, for example, that recent IBM drives have some variant of OS/2 running in the controller. Since the OS knows nothing about heads, tracks, cylinders for modern commodity disks, it's futile to try to schedule I/O for them. it simply uses a write through caching scheme (file system structure updates may be cached for a short bit I think). I also believe it does no scheduling or coalescing of IO requests, just simply FIFO (or whatever the bus/disk driver is implemented to do). It leaves the scheduling up to the disk / array controller. Coalescing is, in fact, done by the kernel, as part of the UBC operation. As for the rest, the controllers tend to be sufficiently intelligent that the OS can only confuse the issue by trying to apply strategy. Corrections welcome. Regards, Justin -- Justin C. Walker, Curmudgeon-At-Large * Institute for General Semantics | When LuteFisk is outlawed | Only outlaws will have | LuteFisk *--------------------------------------*-------------------------------* _______________________________________________ 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: Disk Scheduling
by Shawn Erickson 25 Mar '03

25 Mar '03

On Tuesday, March 25, 2003, at 12:06 PM, Joey Echeverria wrote: I'm talking about the low level scheduling of writing logical blocks out to disk. For example, FIFO, SSTF, LOOK, C-LOOK, etc. The reason that I'm interested is that we are looking into power efficiency in operating systems. One thing we wanted to investigate is if there is any effect to the energy usage when using different algorithms. One idea that we had was to be able to queue requests for longer in an operating system buffer and only write to disk when the queue becomes a certain size. We were thinking that under such a scheme, you could keep the disk stopped for longer periods of time to save the energy from keeping it spinning. To my knowledge Mac OS X does no buffering of writes to disk, it simply uses a write through caching scheme (file system structure updates may be cached for a short bit I think). I also believe it does no scheduling or coalescing of IO requests, just simply FIFO (or whatever the bus/disk driver is implemented to do). It leaves the scheduling up to the disk / array controller. Mac OS X does host a broad set of applications that have a wide set of IO profiles that makes implementing an in OS disk scheduling system problematic. Such a scheme can also compete or in the least be redundant to logic housed in the disk / array controllers logic. I would agree that caching could allow one to reduce the amount of disk spin up for sparse in time writes. I think write caching would have far more of a potential impact on power consumption then choosing a particular block scheduling algorithm (especially given current drives having rather intelligent embedded head schedulers). To do things correctly... the write caching logic would have to be adaptive so it can learn the current behavior of the IO profile to correctly defer writes to maximize disk down time and that weighted against the power draw related to spinning things back up (as I am sure you know, it take less power to keep a disk spinning then spinning it back up to speed). -Shawn _______________________________________________ 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: Disk Scheduling
by Joey Echeverria 25 Mar '03

25 Mar '03

I'm talking about the low level scheduling of writing logical blocks out to disk. For example, FIFO, SSTF, LOOK, C-LOOK, etc. The reason that I'm interested is that we are looking into power efficiency in operating systems. One thing we wanted to investigate is if there is any effect to the energy usage when using different algorithms. One idea that we had was to be able to queue requests for longer in an operating system buffer and only write to disk when the queue becomes a certain size. We were thinking that under such a scheme, you could keep the disk stopped for longer periods of time to save the energy from keeping it spinning. Thanks, Joey Echeverria Department of Electrical and Computer Engineering Carnegie Mellon University On Tuesday, March 25, 2003, at 01:25 PM, Shawn Erickson wrote: On Monday, March 17, 2003, at 06:35 PM, Joey Echeverria wrote: I'm working on a project where we are using the Darwin kernel. I need to look into disk scheduling but was having a hard time traversing the source. I was hoping that you may be able to point me in the right direction. What do you mean by "disk scheduling" in this context? -Shawn _______________________________________________ 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