Skip to contentSkip to conversation

powerpc/crash: Use NMI context for printk after crashing other CPUs

PATCH In review Leonardo Bras 2 participants 3 messages Apr 6, 2020 – Apr 8, 2020 PowerPC

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 powerpc/crash: Use NMI context for printk after crashing other CPUs

Currently, if printk lock (logbuf_lock) is held by other thread during crash, there is a chance of deadlocking the crash on next printk, and blocking a possibly desired kdump.

After sending IPI to all other CPUs, make printk enter in NMI context, as it will use per-cpu buffers to store the message, and avoid locking logbuf_lock.

---
 arch/powerpc/kexec/crash.c | 1 +
 1 file changed, 1 insertion(+)
arch/powerpc/kexec/crash.c +1 -0
index d488311efab1..9b73e3991bf4 100644
@@ -115,6 +115,7 @@ static void crash_kexec_prepare_cpus(int cpu)
 
 	crash_send_ipi(crash_ipi_callback);
 	smp_wmb();
+	printk_nmi_enter();
 
 again:
 	/*
Signature · 1 line
2.25.1
#1 lore raw

arch/powerpc/kexec/crash.c static void crash_kexec_prepare_cpus(int cpu)

Quoted 18 lines · Leonardo Bras

Currently, if printk lock (logbuf_lock) is held by other thread during crash, there is a chance of deadlocking the crash on next printk, and blocking a possibly desired kdump.

After sending IPI to all other CPUs, make printk enter in NMI context, as it will use per-cpu buffers to store the message, and avoid locking logbuf_lock.

---
 arch/powerpc/kexec/crash.c | 1 +
 1 file changed, 1 insertion(+)
arch/powerpc/kexec/crash.c +0 -0
index d488311efab1..9b73e3991bf4 100644
@@ -115,6 +115,7 @@ static void crash_kexec_prepare_cpus(int cpu)

Added context:

	printk(KERN_EMERG "Sending IPI to other CPUs\n");

	if (crash_wake_offline)
		ncpus = num_present_cpus() - 1;

arch/powerpc/kexec/crash.c static void crash_kexec_prepare_cpus(int cpu) · line 118

Quoted 4 lines
arch/powerpc/kexec/crash.c +1 -0
 	crash_send_ipi(crash_ipi_callback);
 	smp_wmb();
+	printk_nmi_enter();

Why did you decide to put it there, rather than at the start of default_machine_crash_shutdown() like I did?

The printk() above could have already deadlocked if another CPU is stuck with the logbuf lock held.

cheers

#2 lore raw

On Wed, 2020-04-08 at 22:13 +1000, Michael Ellerman wrote:
[...]

Quoted 15 lines

Added context:

	printk(KERN_EMERG "Sending IPI to other CPUs\n");

	if (crash_wake_offline)
		ncpus = num_present_cpus() - 1;

arch/powerpc/kexec/crash.c static void crash_kexec_prepare_cpus(int cpu) · line 118

Quoted 4 lines
arch/powerpc/kexec/crash.c +1 -0
 	crash_send_ipi(crash_ipi_callback);
 	smp_wmb();
+	printk_nmi_enter();

Why did you decide to put it there, rather than at the start of default_machine_crash_shutdown() like I did?

Quoted 2 lines

The printk() above could have already deadlocked if another CPU is stuck with the logbuf lock held.

Oh, I thought the CPUs would start crashing after crash_send_ipi(), so only printk() after that would possibly deadlock.

I was not able to see how the printk() above would deadlock, but I see no problem adding that at the start of the function.

Best regards,
Leonardo Bras