How to PANIC a NeXT (mach kernel bug )

Thread View: comp.sys.next.programmer

3 messages

[T] Back to Threads [V] Tree View [G] Group Articles

3 total messages Started by jolly Tue, 01 Sep 1992 21:11

#985

Author: jolly
Date: Tue, 01 Sep 1992 21:11

22 lines
1367 bytes

Hi folks !

I think it is very useful to know, that the NeXT-Mach Kernel
does not support more than 2oo threads !
Running a program under the gdb doesn t change anything ! To solve
this little problem is to use not more than 2oo threads. I think
that the most applications won t use all threads at the same time -
some will die after a while - so that the thread_set_limt() makes
them only a little bit slowlier.
I don t know if 3.o solves that kernel-bug. ( using 2.1 )

				Jolly B)

Post Scriptum : Don t try to test the bug, cause it may damage some
		files ( 2 files were destructed on my host due to the
		fsck !!)

Post Post Scriptum : I think that NeXT should not publish files, which
		     will panic the system after inserting only one
		     character. ( The only thing you have to do is to
		     insert a "2" in front of the "15" in the NeXT 
		     Operating System thread-example ).

Re: How to PANIC a NeXT (mach kernel bug )

#1013

Author: Christopher D Audley
Date: Wed, 02 Sep 1992 13:10

31 lines
2079 bytes

In article <1992Sep1.211133.21384@news.lrz-muenchen.de> jolly@next2.cis.uni-muenchen.de () writes:
>I think it is very useful to know, that the NeXT-Mach Kernel
>does not support more than 2oo threads !
[..munched..]
>I don t know if 3.o solves that kernel-bug. ( using 2.1 )

This is not a bug.  All operating system kernels have resource limits. Congrats,
You've discovered one of mach's.  There's more than one way to kill a kernel by
exceeding its limits, many I'm sure will remember that oldie but goodie
"while(1) fork() ;".  You have to program around these limits, you can't just
claim its a bug and push it onto someone elses plate.  I don't remember
which header file contains a macro indicating the max number of threads per
task, but I do know it exists ( anyone care to jump in with that info? ).

>Post Post Scriptum : I think that NeXT should not publish files, which
>		     will panic the system after inserting only one
>		     character. ( The only thing you have to do is to
>		     insert a "2" in front of the "15" in the NeXT 
>		     Operating System thread-example ).

And files that will hang the system if 17 characters are inserted!! ( See
above )  Is it to late to have NeXT wipe the Examples directory clean in
3.0!!!  And we must get rid of that nasty compiler, it could be twisted
into a tool of system destruction! :-) :-) :-)

-=Chris

-- 
Christopher D. Audley                 Internet: audley@jhunix.hcf.jhu.edu
Elec & Comp Engineering                      -= NeXTmail Welcome =-
The Johns Hopkins University          Phone: (410)889-3639

Re: How to PANIC a NeXT (mach kernel bug )

#840

Author: Matthew Dillon
Date: Tue, 08 Sep 1992 02:52

41 lines
2297 bytes

In article <1992Sep2.131055.25827@jhunix.hcf.jhu.edu> audley@jhunix.hcf.jhu.edu (Christopher D Audley) writes:
>In article <1992Sep1.211133.21384@news.lrz-muenchen.de> jolly@next2.cis.uni-muenchen.de () writes:
>>I think it is very useful to know, that the NeXT-Mach Kernel
>>does not support more than 2oo threads !
>[..munched..]
>>I don t know if 3.o solves that kernel-bug. ( using 2.1 )
>
>This is not a bug.  All operating system kernels have resource limits. Congrats,
>You've discovered one of mach's.  There's more than one way to kill a kernel by
>exceeding its limits, many I'm sure will remember that oldie but goodie
>"while(1) fork() ;".  You have to program around these limits, you can't just
>claim its a bug and push it onto someone elses plate.	I don't remember
>which header file contains a macro indicating the max number of threads per
>task, but I do know it exists ( anyone care to jump in with that info? ).

    The fact that it crashes the kernel *is* a bug, however.

    I've noticed that the NeXT, 3.0 included, appears to ignore
    many of the resource limits (limit command from CSH).. it
    looks at the coredumpsize and stacksize but completely ignores
    datasize and memoryuse.

    This is exceedingly dangerous NeXT!  It means that *one* process with a
    memory leak can crash the machine.	I'd rather just the one process go
    away instead, thank you!

				-Matt

>-=Chris
>
>--
>Christopher D. Audley		       Internet: audley@jhunix.hcf.jhu.edu
>Elec & Comp Engineering		      -= NeXTmail Welcome =-
>The Johns Hopkins University	       Phone: (410)889-3639

--

    Matthew Dillon	    dillon@Overload.Berkeley.CA.US
    891 Regal Rd.	    uunet.uu.net!overload!dillon
    Berkeley, Ca. 94708     ham: KC6LVW (no mail drop)
    USA 		    Sandel-Avery Engineering (702)831-8000

Thread Navigation

This is a paginated view of messages in the thread with full content displayed inline.

Messages are displayed in chronological order, with the original post highlighted in green.

Use pagination controls to navigate through all messages in large threads.

Back to All Threads