3.16.72-rc1 review patch. If anyone has any objections, please let me know.
------------------
From: Thomas Gleixner tglx@linutronix.de
commit 2f5fb19341883bb6e37da351bc3700489d8506a7 upstream.
Mikhail reported a lockdep splat related to the AMD specific ssb_state lock:
CPU0 CPU1 lock(&st->lock); local_irq_disable(); lock(&(&sighand->siglock)->rlock); lock(&st->lock); <Interrupt> lock(&(&sighand->siglock)->rlock);
*** DEADLOCK ***
The connection between sighand->siglock and st->lock comes through seccomp, which takes st->lock while holding sighand->siglock.
Make sure interrupts are disabled when __speculation_ctrl_update() is invoked via prctl() -> speculation_ctrl_update(). Add a lockdep assert to catch future offenders.
Fixes: 1f50ddb4f418 ("x86/speculation: Handle HT correctly on AMD") Reported-by: Mikhail Gavrilov mikhail.v.gavrilov@gmail.com Signed-off-by: Thomas Gleixner tglx@linutronix.de Tested-by: Mikhail Gavrilov mikhail.v.gavrilov@gmail.com Cc: Thomas Lendacky thomas.lendacky@amd.com Link: https://lkml.kernel.org/r/alpine.DEB.2.21.1904141948200.4917@nanos.tec.linut... Signed-off-by: Ben Hutchings ben@decadent.org.uk --- arch/x86/kernel/process.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
--- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -351,6 +351,8 @@ static __always_inline void __speculatio u64 msr = x86_spec_ctrl_base; bool updmsr = false;
+ lockdep_assert_irqs_disabled(); + /* * If TIF_SSBD is different, select the proper mitigation * method. Note that if SSBD mitigation is disabled or permanentely @@ -402,10 +404,12 @@ static unsigned long speculation_ctrl_up
void speculation_ctrl_update(unsigned long tif) { + unsigned long flags; + /* Forced update. Make sure all relevant TIF flags are different */ - preempt_disable(); + local_irq_save(flags); __speculation_ctrl_update(~tif, tif); - preempt_enable(); + local_irq_restore(flags); }
/* Called from seccomp/prctl update */