|
From: Casper H. <ch...@us...> - 2002-05-19 14:24:25
|
s=F8n, 2002-05-19 kl. 17:10 skrev David Welch: > On Sun, May 19, 2002 at 03:46:05PM +0200, Casper Hornstrup wrote: > > Fixed, please try again. > >=20 > I do not think this fix is correct; we can certainly try to lock an=20 > fast mutex when the current thread doesn't already own it but if the > current thread does already own it then we will deadlock since fast=20 > mutexes aren't recursive. The real problem is locking the address space > when we already own it so I had have reverted your changes and modified=20 > section.c not to do this. Yes, you are right, I wasn't thinking there. |