From: Oleg Nesterov Date: Mon, 8 Jul 2013 23:01:06 +0000 (-0700) Subject: ptrace/x86: flush_ptrace_hw_breakpoint() shoule clear the virtual debug registers X-Git-Url: https://git.stricted.de/?a=commitdiff_plain;h=f7da04c9e363e479258135ac825734d78aecd2b0;p=GitHub%2Fmoto-9609%2Fandroid_kernel_motorola_exynos9610.git ptrace/x86: flush_ptrace_hw_breakpoint() shoule clear the virtual debug registers flush_ptrace_hw_breakpoint() destroys the counters set by ptrace, but "leaks" ->debugreg6 and ->ptrace_dr7. The problem is minor, but still it doesn't look right and flush_thread() did this until commit 66cb59172959 ("hw-breakpoints: use the new wrapper routines to access debug registers in process/thread code"). Now that PTRACE_DETACH does flush_ too this makes even more sense. Signed-off-by: Oleg Nesterov Cc: Benjamin Herrenschmidt Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jan Kratochvil Cc: Michael Neuling Cc: Paul Mackerras Cc: Paul Mundt Cc: Will Deacon Cc: Prasad Cc: Russell King Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- diff --git a/arch/x86/kernel/hw_breakpoint.c b/arch/x86/kernel/hw_breakpoint.c index 02f07634d265..f66ff162dce8 100644 --- a/arch/x86/kernel/hw_breakpoint.c +++ b/arch/x86/kernel/hw_breakpoint.c @@ -393,6 +393,9 @@ void flush_ptrace_hw_breakpoint(struct task_struct *tsk) unregister_hw_breakpoint(t->ptrace_bps[i]); t->ptrace_bps[i] = NULL; } + + t->debugreg6 = 0; + t->ptrace_dr7 = 0; } void hw_breakpoint_restore(void)