Jeff:

Nice idea but inapplicable in this case.  This hang is more like the 
compiler losing track of the stack depth and returning from a function 
to never-never land.  By the time anything breaks the CPU is so far away 
from the cause that backtracking is not possible.  And many times 
nothing breaks -- it just keeps executing in the pasture and can't be 
interrupted -- or it kills the debugger stub, or some such.

-- Dave

Jeff Goldszer wrote:

>Here is a "tunnel view" approach troubleshooting this SMP issue and possibly get more 
>information on this problem.
>
>I call this a tunnel view approach because it is assuming the hang is occurring 
>because of Deadlock on a SMP Lock.
>
>Modify the code LiS Locking code to busy loop (trylock) instead of calling the lock 
>macro. The busy loop will loop for some period of time after which it will assume a 
>hang and printout the thread trying to catch the lock and the thread that owns the 
>lock to the console.
>
>The approach make a lot of assumptions but in some circumstances can be useful.
>
>
>
>
>Jeff Goldszer
>Senior Software Engineer
>AlebraTechnologies
>65C Commodore Lane
>West Babylon NY, 11704
>Phone: 631-321-5118
>FAX: 631-321-5119
>
> -----Original Message-----
>From:  Dave Grothe [mailto:[EMAIL PROTECTED]] 
>Sent:  Monday, May 20, 2002 12:06 PM
>To:    Linux STREAMS
>Subject:       [Linux-streams] LiS vs gcc 2.96
>
>I have observed that gcc 2.96 contains a code optimization error that 
>causes mysterious system hangs when running LiS and certain driver 
>combinations that include interactions with the ldl driver.  The hangs 
>are so bad that a 2 CPU SMP system with NMI watchdog timer enabled 
>cannot break into the loop.
>
>Due to the severity of the hang I was never able to get close enough to 
>the problem to pinpoint it.  Therefore it cannot be usefully reported to 
>the gcc authors.
>
>The solution is to use -O2 instead of -O3 to build LiS and its included 
>drivers.  The next published version of LiS will be configured to use 
>-O2.  In the meantime if anyone is experiencing these sorts of problems, 
>try editing the LiS file config.mk directly and changing the 
>optimization level.
>
>I am uncertain whether it is necesary to build your STREAMS drivers with 
>-O2 rather than -O3.
>
>gcc 2.96 is distributed with both Red Hat 7.2 and 7.3.  Both versions 
>have the problem.
>
>-- Dave
>
>
>_______________________________________________
>Linux-streams mailing list
>[EMAIL PROTECTED]
>http://gsyc.escet.urjc.es/mailman/listinfo/linux-streams
>



_______________________________________________
Linux-streams mailing list
[EMAIL PROTECTED]
http://gsyc.escet.urjc.es/mailman/listinfo/linux-streams

Reply via email to