whack-a-mole: no need to set_fs(USER_DS) in {start,flush}_thread()
authorAl Viro <viro@zeniv.linux.org.uk>
Mon, 23 Feb 2015 08:21:31 +0000 (03:21 -0500)
committerAl Viro <viro@zeniv.linux.org.uk>
Sun, 12 Apr 2015 02:24:31 +0000 (22:24 -0400)
flush_old_exec() has already done that.  Back on 2011 a bunch of
instances like that had been kicked out, but that hadn't taken
care of then-out-of-tree architectures, obviously, and they served
as reinfection vector...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
arch/arc/kernel/process.c
arch/c6x/kernel/process.c
arch/hexagon/kernel/process.c
arch/metag/include/asm/processor.h
arch/nios2/kernel/process.c
arch/openrisc/kernel/process.c

index 98c00a2..f46efd1 100644 (file)
@@ -155,8 +155,6 @@ int copy_thread(unsigned long clone_flags,
  */
 void start_thread(struct pt_regs * regs, unsigned long pc, unsigned long usp)
 {
-       set_fs(USER_DS); /* user space */
-
        regs->sp = usp;
        regs->ret = pc;
 
index 57d2ea8..3ae9f5a 100644 (file)
@@ -101,7 +101,6 @@ void start_thread(struct pt_regs *regs, unsigned int pc, unsigned long usp)
         */
        usp -= 8;
 
-       set_fs(USER_DS);
        regs->pc  = pc;
        regs->sp  = usp;
        regs->tsr |= 0x40; /* set user mode */
Simple merge
Simple merge
Simple merge
Simple merge