May 2004 - Darwin-kernel - lists.apple.com

Re: Kernel build problem
by Felix Kronlage 28 May '04

28 May '04

On Fri, May 28, 2004 at 09:18:04AM -0500, Steven Bytnar wrote: > You should try applying parts of this diff that make sense: > http://www.opendarwin.org/~fkr/xnu/make-xnu-517.3.7-compile.diff > In particular, check out this part: > diff -urN xnu-517.3.7/osfmk/ppc/chud/chud_spr.h xnu-517.3.7-patched/osfmk/ppc/chud/chud_spr.h > --- xnu-517.3.7/osfmk/ppc/chud/chud_spr.h Fri Dec 19 23:38:23 2003 > +++ xnu-517.3.7-patched/osfmk/ppc/chud/chud_spr.h Tue Dec 30 22:01:26 2003 > > Things like this have been a recurring problem for many of the XNU releases > for 10.3.x. It's very annoying. It makes it a pain in the butt to rebuild > anything from the Apple provided source snapshots without applying some > amount of patches. > > Instead of building the Apple source snapshots directly, you will have better > luck with the opendarwin CVS version (which is currently the 10.3.2 kernel > though *cough* *cough*): > http://cvs.opendarwin.org/index.cgi/src/xnu/ > where this problem is fixed: > http://cvs.opendarwin.org/index.cgi/src/xnu/osfmk/ppc/chud/chud_spr.h yes, I have not done any more imports since I still want to finish a release with what we got. Also importing (and doing all the merging including removing cruft) takes _a_lot_ of time, especially considering there are not a lot of active helpers out there. Likely I'll will not import the full source of Darwin in the future, mainly to not waste valuable time that much rather is spent hacking on useful stuff (like wireless support, xnu, etc. etc.) and making somewhat useful releases from the tarballs that Apple releases (plus the stuff we develope). Likely (especially because people asked me) I'll keep importing xnu and of course other third party sources that we work on, things like ncutil, yum, rpm. With the tarballs from Apple and the DarwinPorts Portfiles in src/Ports we can easily build the sources and with adding Patchfiles even patch things without the hassle of importing/merging, so this will hopefully allow me to concentrate more on the real nifty things ;) I agree with people that it's much nicer to see code changes in cvs, but too few people (at least it seems like that) make use of our cvs to actually see such changes that it is not worth the effort. > If you want the 10.3.4 kernel, you'll need to fix the source yourself. > If you want a wrapped up kernel that's likely to build without problems, > use the one in opendarwin CVS. I'll take a look tonight and see about providing a tarball. -fkr -- gpg-fingerprint: 076E 1E87 3E05 1C7F B1A0 8A48 0D31 9BD3 D9AC 74D0 Felix Kronlage | FKR-RIPE | http://www.hazardous.org/fkr/ fkr(a){grummel.net|opendarwin.org} | http://www.opendarwin.org/~fkr/ _______________________________________________ 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: Kernel build problem
by Kevin Van Vechten 28 May '04

28 May '04

xnu-517.7.7 on <http://www.opensource.apple.com/darwinsource> has been updated to correct this problem. It had already been corrected in previous releases as far as I know (please let me know if this isn't correct.) The newer xnu-517.7.7 contains the same fixes as Felix's patch mentioned below. - Kevin On May 28, 2004, at 7:18 AM, Steven Bytnar wrote: You should try applying parts of this diff that make sense: http://www.opendarwin.org/~fkr/xnu/make-xnu-517.3.7-compile.diff In particular, check out this part: diff -urN xnu-517.3.7/osfmk/ppc/chud/chud_spr.h xnu-517.3.7-patched/osfmk/ppc/chud/chud_spr.h --- xnu-517.3.7/osfmk/ppc/chud/chud_spr.h Fri Dec 19 23:38:23 2003 +++ xnu-517.3.7-patched/osfmk/ppc/chud/chud_spr.h Tue Dec 30 22:01:26 2003 Things like this have been a recurring problem for many of the XNU releases for 10.3.x. It's very annoying. It makes it a pain in the butt to rebuild anything from the Apple provided source snapshots without applying some amount of patches. Instead of building the Apple source snapshots directly, you will have better luck with the opendarwin CVS version (which is currently the 10.3.2 kernel though *cough* *cough*): http://cvs.opendarwin.org/index.cgi/src/xnu/ where this problem is fixed: http://cvs.opendarwin.org/index.cgi/src/xnu/osfmk/ppc/chud/chud_spr.h If you want the 10.3.4 kernel, you'll need to fix the source yourself. If you want a wrapped up kernel that's likely to build without problems, use the one in opendarwin CVS. --Steve On Fri, May 28, 2004 at 10:33:03PM +1000, Alastair Rankine wrote: Hi all, I've got a problem building the darwin kernel, following the instructions here: http://www.opendarwin.org/doc/en/articles/building_xnu/ building_xnu.php . Google tells me that one other person has seen the same (or similar) problem, but I can't see a fix. Any help appreciated. /usr/bin/cc -c -static -g -nostdinc -nostdlib -no-cpp-precomp -fno-builtin -finline -fno-keep-inline-functions -msoft-float -fsigned-bitfields -Wpointer-arith -arch ppc -Dppc -DPPC -D__PPC__ -DPAGE_SIZE_FIXED -O2 -mcpu=750 -mmultiple -fschedule-insns -DAPPLE -DNeXT -DKERNEL_PRIVATE -D__MACHO__=1 -Dvolatile=__volatile -DMACH -DMACH_KERNEL -DMACH_LOAD -DMACH_BSD -DIOKIT -DMACH_PE -DPOWERMAC -DMACH_KERNEL_PRIVATE -MD -I. -I/Users/alastair/src/xnu-517.7.7/osfmk -I/Users/alastair/src/xnu-517.7.7/BUILD/obj/EXPORT_HDRS/bsd -I/Users/alastair/src/xnu-517.7.7/BUILD/obj/EXPORT_HDRS/iokit -I/Users/alastair/src/xnu-517.7.7/BUILD/obj/EXPORT_HDRS/pexpert -I/Users/alastair/src/xnu-517.7.7/BUILD/obj/EXPORT_HDRS/libkern -I/Users/alastair/src/xnu-517.7.7/BUILD/obj/EXPORT_HDRS/libsa -I/Users/alastair/src/xnu-517.7.7/BUILD/obj/EXTERN_HDRS -I/Users/alastair/src/xnu-517.7.7/EXTERNAL_HEADERS -I/Users/alastair/src/xnu-517.7.7/EXTERNAL_HEADERS/bsd -I/Users/alastair/src/xnu-517.7.7/osfmk/libsa /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c: In function `chudxnu_set_shadowed_spr': /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:224: error: `chud_7400_msscr1' undeclared (first use in this function) /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:224: error: (Each undeclared identifier is reported only once /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:224: error: for each function it appears in.) /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c: In function `chudxnu_set_shadowed_spr64': /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:260: error: `chud_970_hid0' undeclared (first use in this function) /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:272: error: `chud_970_hid1' undeclared (first use in this function) /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:284: error: `chud_970_hid4' undeclared (first use in this function) /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:296: error: `chud_970_hid5' undeclared (first use in this function) /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c: In function `chudxnu_clear_cpu_rupt_counters': /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:444: warning: passing arg 1 of `bzero' from incompatible pointer type make[6]: *** [chud_cpu.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 _______________________________________________ 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. _______________________________________________ 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. _______________________________________________ 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

Darwin 7.4 kernel compilation fails
by piechurski, sebastien 28 May '04

28 May '04

Hi, I am trying to compile the Darwin xnu kernel version 517.7.7 . I have come across the problem described earlier by Alastair Rankine (undeclared chud_* symbols), but I am now getting this error : /usr/bin/cc -static -D__ASSEMBLER__ -force_cpusubtype_ALL -arch ppc -Dppc -DPPC -D__PPC__ -DPAGE_SIZE_FIXED -DAPPLE -DNeXT -DKERNEL_PRIVATE -D__MACHO__=1 -Dvolatile=__volatile -DMACH -DMACH_KERNEL -DMACH_LOAD -DMACH_BSD -DIOKIT -DMACH_PE -DPOWERMAC -c chud_cpu_asm.s; /bin/rm -f chud_cpu_asm.pp chud_cpu_asm.s chud_cpu_asm.s:282:Parameter error: expression must be absolute (parameter 2) chud_cpu_asm.s:553:Parameter error: expression must be absolute (parameter 1) make[6]: *** [chud_cpu_asm.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 which corresponds to : osfmk/ppc/chud/chud_cpu_asm.s:282 : .globl EXT(chudxnu_mfscomd_64) EXT(chudxnu_mfscomd_64): I don't understand the error... I can't find the difference between this line and the ones above which compile correctly... Please help. 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: Kernel build problem
by Steven Bytnar 28 May '04

28 May '04

You should try applying parts of this diff that make sense: http://www.opendarwin.org/~fkr/xnu/make-xnu-517.3.7-compile.diff In particular, check out this part: diff -urN xnu-517.3.7/osfmk/ppc/chud/chud_spr.h xnu-517.3.7-patched/osfmk/ppc/chud/chud_spr.h --- xnu-517.3.7/osfmk/ppc/chud/chud_spr.h Fri Dec 19 23:38:23 2003 +++ xnu-517.3.7-patched/osfmk/ppc/chud/chud_spr.h Tue Dec 30 22:01:26 2003 Things like this have been a recurring problem for many of the XNU releases for 10.3.x. It's very annoying. It makes it a pain in the butt to rebuild anything from the Apple provided source snapshots without applying some amount of patches. Instead of building the Apple source snapshots directly, you will have better luck with the opendarwin CVS version (which is currently the 10.3.2 kernel though *cough* *cough*): http://cvs.opendarwin.org/index.cgi/src/xnu/ where this problem is fixed: http://cvs.opendarwin.org/index.cgi/src/xnu/osfmk/ppc/chud/chud_spr.h If you want the 10.3.4 kernel, you'll need to fix the source yourself. If you want a wrapped up kernel that's likely to build without problems, use the one in opendarwin CVS. --Steve On Fri, May 28, 2004 at 10:33:03PM +1000, Alastair Rankine wrote: > Hi all, > > I've got a problem building the darwin kernel, following the > instructions here: > http://www.opendarwin.org/doc/en/articles/building_xnu/building_xnu.php > . Google tells me that one other person has seen the same (or similar) > problem, but I can't see a fix. > > Any help appreciated. > > > /usr/bin/cc -c -static -g -nostdinc -nostdlib -no-cpp-precomp > -fno-builtin -finline -fno-keep-inline-functions -msoft-float > -fsigned-bitfields -Wpointer-arith -arch ppc -Dppc -DPPC -D__PPC__ > -DPAGE_SIZE_FIXED -O2 -mcpu=750 -mmultiple -fschedule-insns -DAPPLE > -DNeXT -DKERNEL_PRIVATE -D__MACHO__=1 -Dvolatile=__volatile -DMACH > -DMACH_KERNEL -DMACH_LOAD -DMACH_BSD -DIOKIT -DMACH_PE -DPOWERMAC > -DMACH_KERNEL_PRIVATE -MD -I. -I/Users/alastair/src/xnu-517.7.7/osfmk > -I/Users/alastair/src/xnu-517.7.7/BUILD/obj/EXPORT_HDRS/bsd > -I/Users/alastair/src/xnu-517.7.7/BUILD/obj/EXPORT_HDRS/iokit > -I/Users/alastair/src/xnu-517.7.7/BUILD/obj/EXPORT_HDRS/pexpert > -I/Users/alastair/src/xnu-517.7.7/BUILD/obj/EXPORT_HDRS/libkern > -I/Users/alastair/src/xnu-517.7.7/BUILD/obj/EXPORT_HDRS/libsa > -I/Users/alastair/src/xnu-517.7.7/BUILD/obj/EXTERN_HDRS > -I/Users/alastair/src/xnu-517.7.7/EXTERNAL_HEADERS > -I/Users/alastair/src/xnu-517.7.7/EXTERNAL_HEADERS/bsd > -I/Users/alastair/src/xnu-517.7.7/osfmk/libsa > /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c > /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c: In function > `chudxnu_set_shadowed_spr': > /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:224: error: > `chud_7400_msscr1' undeclared (first use in this function) > /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:224: error: > (Each undeclared identifier is reported only once > /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:224: error: > for each function it appears in.) > /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c: In function > `chudxnu_set_shadowed_spr64': > /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:260: error: > `chud_970_hid0' undeclared (first use in this function) > /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:272: error: > `chud_970_hid1' undeclared (first use in this function) > /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:284: error: > `chud_970_hid4' undeclared (first use in this function) > /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:296: error: > `chud_970_hid5' undeclared (first use in this function) > /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c: In function > `chudxnu_clear_cpu_rupt_counters': > /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:444: warning: > passing arg 1 of `bzero' from incompatible pointer type > make[6]: *** [chud_cpu.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 > _______________________________________________ > 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. _______________________________________________ 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

Kernel build problem
by Alastair Rankine 28 May '04

28 May '04

Hi all, I've got a problem building the darwin kernel, following the instructions here: http://www.opendarwin.org/doc/en/articles/building_xnu/building_xnu.php . Google tells me that one other person has seen the same (or similar) problem, but I can't see a fix. Any help appreciated. /usr/bin/cc -c -static -g -nostdinc -nostdlib -no-cpp-precomp -fno-builtin -finline -fno-keep-inline-functions -msoft-float -fsigned-bitfields -Wpointer-arith -arch ppc -Dppc -DPPC -D__PPC__ -DPAGE_SIZE_FIXED -O2 -mcpu=750 -mmultiple -fschedule-insns -DAPPLE -DNeXT -DKERNEL_PRIVATE -D__MACHO__=1 -Dvolatile=__volatile -DMACH -DMACH_KERNEL -DMACH_LOAD -DMACH_BSD -DIOKIT -DMACH_PE -DPOWERMAC -DMACH_KERNEL_PRIVATE -MD -I. -I/Users/alastair/src/xnu-517.7.7/osfmk -I/Users/alastair/src/xnu-517.7.7/BUILD/obj/EXPORT_HDRS/bsd -I/Users/alastair/src/xnu-517.7.7/BUILD/obj/EXPORT_HDRS/iokit -I/Users/alastair/src/xnu-517.7.7/BUILD/obj/EXPORT_HDRS/pexpert -I/Users/alastair/src/xnu-517.7.7/BUILD/obj/EXPORT_HDRS/libkern -I/Users/alastair/src/xnu-517.7.7/BUILD/obj/EXPORT_HDRS/libsa -I/Users/alastair/src/xnu-517.7.7/BUILD/obj/EXTERN_HDRS -I/Users/alastair/src/xnu-517.7.7/EXTERNAL_HEADERS -I/Users/alastair/src/xnu-517.7.7/EXTERNAL_HEADERS/bsd -I/Users/alastair/src/xnu-517.7.7/osfmk/libsa /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c: In function `chudxnu_set_shadowed_spr': /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:224: error: `chud_7400_msscr1' undeclared (first use in this function) /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:224: error: (Each undeclared identifier is reported only once /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:224: error: for each function it appears in.) /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c: In function `chudxnu_set_shadowed_spr64': /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:260: error: `chud_970_hid0' undeclared (first use in this function) /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:272: error: `chud_970_hid1' undeclared (first use in this function) /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:284: error: `chud_970_hid4' undeclared (first use in this function) /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:296: error: `chud_970_hid5' undeclared (first use in this function) /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c: In function `chudxnu_clear_cpu_rupt_counters': /Users/alastair/src/xnu-517.7.7/osfmk/ppc/chud/chud_cpu.c:444: warning: passing arg 1 of `bzero' from incompatible pointer type make[6]: *** [chud_cpu.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 _______________________________________________ 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 to debug the early startup stages of OS X?
by Adhamh Findlay 28 May '04

28 May '04

Corey, About all you could do is boot into OF by holding down command option O F. Then set the verbose boot-args flag. setenv boot-args -v then do: mac-boot IF you can get anymore info that would do it. Adhamh Corey O'Connor wrote: A little background: I installed the 10.3.4 updater to my G4 last night. Sadly, the computer would then refuse to go past the grey apple on light-grey screen. No spinning disk or anything, just the grey screen. The comp has heavily tweaked HW & SW so I'm not bitching, just downgraded to 10.3.3 and everything is fine. Correct me if I'm wrong but the usual boot sequence is: 1. OF loads the first stage bootloader 2. First stage bootloader initializes the computer then loads the kernel 3. The kernel goes off and does it's thing. Now, I'd imagine the problem is occurring either early in stage 3 or late in stage 2. How do I go about debugging these steps? Is there any pointers to some way I could get OF to spit out more info or anything along that lines? Corey O'Connor DogHeadBone LLC http://www.dogheadbone.com _______________________________________________ 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. _______________________________________________ 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 to debug the early startup stages of OS X?
by Josh de Cesare 28 May '04

28 May '04

Corey, Start by adding '-s' to the boot-args. This will force a single user boot. If you are getting to the kernel, it should become obvious when kernel start verbose logging. If all you see is grey screen with the apple, then it probably did not make it out of BootX. In that case, stop at Open Firmware (hold CMD-OPT-O-F at boot). Then type 'boot' at the prompt then immediately hold down CMD-S. This will force single user boot, but it will also enable logging to the screen from BootX. You should see messages about BootX loading files: mach_kernel, and Extensions.mkext or all the many drivers. Josh At 2:50 PM -0700 5/27/04, Corey O'Connor wrote: A little background: I installed the 10.3.4 updater to my G4 last night. Sadly, the computer would then refuse to go past the grey apple on light-grey screen. No spinning disk or anything, just the grey screen. The comp has heavily tweaked HW & SW so I'm not bitching, just downgraded to 10.3.3 and everything is fine. Correct me if I'm wrong but the usual boot sequence is: 1. OF loads the first stage bootloader 2. First stage bootloader initializes the computer then loads the kernel 3. The kernel goes off and does it's thing. Now, I'd imagine the problem is occurring either early in stage 3 or late in stage 2. How do I go about debugging these steps? Is there any pointers to some way I could get OF to spit out more info or anything along that lines? Corey O'Connor DogHeadBone LLC http://www.dogheadbone.com _______________________________________________ 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. _______________________________________________ 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 to debug the early startup stages of OS X?
by Corey O'Connor 27 May '04

27 May '04

A little background: I installed the 10.3.4 updater to my G4 last night. Sadly, the computer would then refuse to go past the grey apple on light-grey screen. No spinning disk or anything, just the grey screen. The comp has heavily tweaked HW & SW so I'm not bitching, just downgraded to 10.3.3 and everything is fine. Correct me if I'm wrong but the usual boot sequence is: 1. OF loads the first stage bootloader 2. First stage bootloader initializes the computer then loads the kernel 3. The kernel goes off and does it's thing. Now, I'd imagine the problem is occurring either early in stage 3 or late in stage 2. How do I go about debugging these steps? Is there any pointers to some way I could get OF to spit out more info or anything along that lines? Corey O'Connor DogHeadBone LLC http://www.dogheadbone.com _______________________________________________ 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: cpu_to_le16 and le16_to_cpu
by Michael Cashwell 27 May '04

27 May '04

On May 27, 2004, at 4:08 AM, Dan Christiansen wrote: Lalit Chandivade wrote: What is the equivalent of Linux apis cpu_to_le16 and le16_to_cpu on Mac OS X? I believe they are in <architecture/byte_order.h> and are called NXSwapHostShortToLittle() and NXSwapLittleShortToHost(). (Sorry for the repeat Dan. I meant to send this to the list too so here goes...) The Swap functions are useful when the data are already in host memory. But if I recall the behavior of the cpu_to and to_cpu functions on Linux, a closer match would be functions like OSWriteLittleInt32(). These boil down to the PPC instructions to do the swap as part of the reads and writes and offer other useful things like byte offsets via an argument. They are in <libkern/OSByteOrder.h> but I don't explicitly include that. You probably can, but I get it due to some higher-level include from IOKit apparently though I don't know which one. Perhaps <IOKit/system.h>? -Mike _______________________________________________ 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: cpu_to_le16 and le16_to_cpu
by Brian Bergstrand 27 May '04

27 May '04

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I wrote some compat macros while porting Ext2. They work in userland as well as kernel. <http://cvs.sourceforge.net/viewcvs.py/ext2fsx/src/ext2_bigendian.h? rev=1.2&view=auto> Note: Mail adds spaces to long URL's, so if the above doesn't work when you click on it, look for any %20's in the URL and remove them. HTH. On May 27, 2004, at 1:24 AM, Lalit Chandivade wrote: > Hi, > > > What is the equivalent of Linux apis cpu_to_le16 and le16_to_cpu on > Mac OS > X? > > Brian Bergstrand <http://www.bergstrand.org/brian/>, AIM: triryche206 PGP Key: <http://www.bergstrand.org/brian/misc/public_key.txt> Parkinson's Law: Work expands to fill the time allotted it. -----BEGIN PGP SIGNATURE----- Version: PGP 8.0.3 iQA/AwUBQLYDR3nR2Fu2x7aiEQJbtgCg9yhRp3b3PexzkULjFzg5roHafdMAnifX ft2E/vg6WYdKSfJ5lJKPRgJx =2whl -----END PGP SIGNATURE----- _______________________________________________ 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