atom feed7 messages in org.freebsd.freebsd-smpRe: Call to arms..-SMP (fwd)
FromSent OnAttachments
Julian ElischerJun 25, 1999 12:07 am 
Julian ElischerJun 25, 1999 12:08 am 
Julian ElischerJun 25, 1999 12:08 am 
Julian ElischerJun 25, 1999 12:08 am 
Mike SmithJun 25, 1999 12:13 am 
Terry LambertJun 25, 1999 9:51 am 
Julian ElischerJun 25, 1999 11:01 am 
Subject:Re: Call to arms..-SMP (fwd)
From:Terry Lambert (tlam@primenet.com)
Date:Jun 25, 1999 9:51:41 am
List:org.freebsd.freebsd-smp

Not much assembly is need. All we need to do is to make SYSCALL_LOCK a nop, and handle the locking in syscall() the C function.

Don't even do it here; put the lock acquisitions/releases in the syscall implementation functions themselves.

Last time I looked, portions of the trap code and the signal trampoline were not process reentrant. This should not be done until they are, since a user space call conversion scheduler or use of async system calls (e.g. aioread) that occurred simultaneously with a subsequent call could result in a kernel stack corruption of the already pending async call.

Terry Lambert ter@lambert.org

--- Any opinions in this posting are my own and not those of my present or previous employers.

To Unsubscribe: send mail to majo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message