powerpc, hw_breakpoints: Fix racy access to ptrace breakpoints
authorFrederic Weisbecker <fweisbec@gmail.com>
Fri, 8 Apr 2011 15:29:36 +0000 (17:29 +0200)
committerFrederic Weisbecker <fweisbec@gmail.com>
Mon, 25 Apr 2011 15:33:54 +0000 (17:33 +0200)
While the tracer accesses ptrace breakpoints, the child task may
concurrently exit due to a SIGKILL and thus release its breakpoints
at the same time. We can then dereference some freed pointers.

To fix this, hold a reference on the child breakpoints before
manipulating them.

Reported-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Acked-by: Prasad <prasad@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: v2.6.33.. <stable@kernel.org>
Link: http://lkml.kernel.org/r/1302284067-7860-4-git-send-email-fweisbec@gmail.com
arch/powerpc/kernel/ptrace.c

index 55613e33e263ac7abb3b1d478f4a27bb9912b86f..4edeeb325429dd6b8e10c44ccf51057659ea5d5b 100644 (file)
@@ -1591,7 +1591,10 @@ long arch_ptrace(struct task_struct *child, long request,
        }
 
        case PTRACE_SET_DEBUGREG:
+               if (ptrace_get_breakpoints(child) < 0)
+                       return -ESRCH;
                ret = ptrace_set_debugreg(child, addr, data);
+               ptrace_put_breakpoints(child);
                break;
 
 #ifdef CONFIG_PPC64