On Sun, 5 May 2002, Bruce Evans wrote: > On Sat, 4 May 2002, Jake Burkholder wrote: > > > Apparently, On Sun, May 05, 2002 at 10:44:44AM +1000, > > Bruce Evans said words to the effect of; > > > > > > I have seen signs of a generic pipe bug in vi: vi's i/o buffer for > > > pipes is sometimes invalid (kern/sys_pipe.c:pipe_build_write_buffer() > > > gets an error faulting it in). This doesn't usually cause signals; > > > it just confuses vi. > > > > Can you try backing out rev 1.104 of kern/sys_pipe.c? > > I first noticed vi getting confused in the same way (but not the faultin > failure) long before rev.1.104 (in late Jan or early Feb. this year).
Anyway, the failure is in the vm_fault_quick() call which is one line before the pmap_*extract() line that was changed in rev.1.104. (I have some debugging code that traps if the i386 fubyte() fails, but fubyte() fails so rarely that I had forgotten about it. It didn't trap back in Jan/Feb, but I may have been running a plain current kernel without the debugging code.) Bruce To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
