[PATCH AUTOSEL 4.9 14/36] powerpc/xmon: Fix invocation inside lock region

[PATCH AUTOSEL 4.9 14/36] powerpc/xmon: Fix invocation inside lock region

[Sasha Levin]

From: Breno Leitao <leitao@debian.org>

[ Upstream commit 8d4a862276a9c30a269d368d324fb56529e6d5fd ]

Currently xmon needs to get devtree_lock (through rtas_token()) during its
invocation (at crash time). If there is a crash while devtree_lock is being
held, then xmon tries to get the lock but spins forever and never get into
the interactive debugger, as in the following case:

	int *ptr = NULL;
	raw_spin_lock_irqsave(&devtree_lock, flags);
	*ptr = 0xdeadbeef;

This patch avoids calling rtas_token(), thus trying to get the same lock,
at crash time. This new mechanism proposes getting the token at
initialization time (xmon_init()) and just consuming it at crash time.

This would allow xmon to be possible invoked independent of devtree_lock
being held or not.

Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Thiago Jung Bauermann <bauerman@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 arch/powerpc/xmon/xmon.c | 18 ++++++++++++++----
 1 file changed, 14 insertions(+), 4 deletions(-)

diff --git a/arch/powerpc/xmon/xmon.c b/arch/powerpc/xmon/xmon.c
index 760545519a0b..687e8b8bf5c6 100644
--- a/arch/powerpc/xmon/xmon.c
+++ b/arch/powerpc/xmon/xmon.c
@@ -73,6 +73,9 @@ static int xmon_gate;
 #define xmon_owner 0
 #endif /* CONFIG_SMP */
 
+#ifdef CONFIG_PPC_PSERIES
+static int set_indicator_token = RTAS_UNKNOWN_SERVICE;
+#endif
 static unsigned long in_xmon __read_mostly = 0;
 
 static unsigned long adrs;
@@ -340,7 +343,6 @@ static inline void disable_surveillance(void)
 #ifdef CONFIG_PPC_PSERIES
 	/* Since this can't be a module, args should end up below 4GB. */
 	static struct rtas_args args;
-	int token;
 
 	/*
 	 * At this point we have got all the cpus we can into
@@ -349,11 +351,11 @@ static inline void disable_surveillance(void)
 	 * If we did try to take rtas.lock there would be a
 	 * real possibility of deadlock.
 	 */
-	token = rtas_token("set-indicator");
-	if (token == RTAS_UNKNOWN_SERVICE)
+	if (set_indicator_token == RTAS_UNKNOWN_SERVICE)
 		return;
 
-	rtas_call_unlocked(&args, token, 3, 1, NULL, SURVEILLANCE_TOKEN, 0, 0);
+	rtas_call_unlocked(&args, set_indicator_token, 3, 1, NULL,
+			   SURVEILLANCE_TOKEN, 0, 0);
 
 #endif /* CONFIG_PPC_PSERIES */
 }
@@ -3227,6 +3229,14 @@ static void xmon_init(int enable)
 		__debugger_iabr_match = xmon_iabr_match;
 		__debugger_break_match = xmon_break_match;
 		__debugger_fault_handler = xmon_fault_handler;
+
+#ifdef CONFIG_PPC_PSERIES
+		/*
+		 * Get the token here to avoid trying to get a lock
+		 * during the crash, causing a deadlock.
+		 */
+		set_indicator_token = rtas_token("set-indicator");
+#endif
 	} else {
 		__debugger = NULL;
 		__debugger_ipi = NULL;
-- 
2.19.1

[PATCH AUTOSEL 4.9 15/36] powerpc/pseries/cpuidle: Fix preempt warning

[Sasha Levin]

From: Breno Leitao <leitao@debian.org>

[ Upstream commit 2b038cbc5fcf12a7ee1cc9bfd5da1e46dacdee87 ]

When booting a pseries kernel with PREEMPT enabled, it dumps the
following warning:

   BUG: using smp_processor_id() in preemptible [00000000] code: swapper/0/1
   caller is pseries_processor_idle_init+0x5c/0x22c
   CPU: 13 PID: 1 Comm: swapper/0 Not tainted 4.20.0-rc3-00090-g12201a0128bc-dirty #828
   Call Trace:
   [c000000429437ab0] [c0000000009c8878] dump_stack+0xec/0x164 (unreliable)
   [c000000429437b00] [c0000000005f2f24] check_preemption_disabled+0x154/0x160
   [c000000429437b90] [c000000000cab8e8] pseries_processor_idle_init+0x5c/0x22c
   [c000000429437c10] [c000000000010ed4] do_one_initcall+0x64/0x300
   [c000000429437ce0] [c000000000c54500] kernel_init_freeable+0x3f0/0x500
   [c000000429437db0] [c0000000000112dc] kernel_init+0x2c/0x160
   [c000000429437e20] [c00000000000c1d0] ret_from_kernel_thread+0x5c/0x6c

This happens because the code calls get_lppaca() which calls
get_paca() and it checks if preemption is disabled through
check_preemption_disabled().

Preemption should be disabled because the per CPU variable may make no
sense if there is a preemption (and a CPU switch) after it reads the
per CPU data and when it is used.

In this device driver specifically, it is not a problem, because this
code just needs to have access to one lppaca struct, and it does not
matter if it is the current per CPU lppaca struct or not (i.e. when
there is a preemption and a CPU migration).

That said, the most appropriate fix seems to be related to avoiding
the debug_smp_processor_id() call at get_paca(), instead of calling
preempt_disable() before get_paca().

Signed-off-by: Breno Leitao <leitao@debian.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/cpuidle/cpuidle-pseries.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/cpuidle/cpuidle-pseries.c b/drivers/cpuidle/cpuidle-pseries.c
index 166ccd711ec9..83203c59bf59 100644
--- a/drivers/cpuidle/cpuidle-pseries.c
+++ b/drivers/cpuidle/cpuidle-pseries.c
@@ -230,7 +230,13 @@ static int pseries_idle_probe(void)
 		return -ENODEV;
 
 	if (firmware_has_feature(FW_FEATURE_SPLPAR)) {
-		if (lppaca_shared_proc(get_lppaca())) {
+		/*
+		 * Use local_paca instead of get_lppaca() since
+		 * preemption is not disabled, and it is not required in
+		 * fact, since lppaca_ptr does not need to be the value
+		 * associated to the current CPU, it can be from any CPU.
+		 */
+		if (lppaca_shared_proc(local_paca->lppaca_ptr)) {
 			cpuidle_state_table = shared_states;
 			max_idle_state = ARRAY_SIZE(shared_states);
 		} else {
-- 
2.19.1