arm64: Demote boot and shutdown messages to pr_debug
PATCH 1/1 arm64: Demote boot and shutdown messages to pr_debug
Similar to commits c68b0274fb3cf ("ARM: reduce "Booted secondary processor" message to debug level") and 035e787543de7 ("ARM: 8644/1: Reduce "CPU:
shutdown" message to debug level"), demote the secondary_start_kernel() and __cpu_die() messages from info, respectively notice to debug. While we are at it, also do this for cpu_psci_cpu_kill() which is redundant with __cpu_die().
This helps improve the amount of possible hotplug cycles by around +50% on ARCH_BRCMSTB.
--- arch/arm64/kernel/psci.c | 2 +- arch/arm64/kernel/smp.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-)
arch/arm64/kernel/psci.c +1 -1
@@ -96,7 +96,7 @@ static int cpu_psci_cpu_kill(unsigned int cpu)for (i =0 ; i <10 ; i++) { err = psci_ops.affinity_info (cpu_logical_map (cpu),0 );if (err ==PSCI_0_2_AFFINITY_LEVEL_OFF ) { -pr_info ("CPU%d killed.\n" , cpu); +pr_debug ("CPU%d killed.\n" , cpu);return 0 ; }
arch/arm64/kernel/smp.c +2 -2
@@ -259,7 +259,7 @@ asmlinkage notrace void secondary_start_kernel(void)* the CPU migration code to notice that the CPU is online * before we continue. */ -pr_info ("CPU%u: Booted secondary processor 0x%010lx [0x%08x]\n" , +pr_debug ("CPU%u: Booted secondary processor 0x%010lx [0x%08x]\n" , cpu, (unsigned long )mpidr,read_cpuid_id ());update_cpu_boot_status (CPU_BOOT_SUCCESS );
@@ -348,7 +348,7 @@ void __cpu_die(unsigned int cpu)pr_crit ("CPU%u: cpu didn't die\n" , cpu);return ; } -pr_notice ("CPU%u: shutdown\n" , cpu); +pr_debug ("CPU%u: shutdown\n" , cpu);/* * Now that the dying CPU is beyond the point of no return w.r.t.
Signature · 1 line
2.17.1
Mark Rutland
reviewerQuoted 9 lines · Florian Fainelli
Similar to commits
c68b0274fb3cf("ARM: reduce "Booted secondary processor" message to debug level") and035e787543de7("ARM: 8644/1: Reduce "CPU:
shutdown" message to debug level"), demote thesecondary_start_kernel()and__cpu_die()messages from info, respectively notice to debug. While we are at it, also do this forcpu_psci_cpu_kill()which is redundant with__cpu_die().This helps improve the amount of possible hotplug cycles by around +50% on
ARCH_BRCMSTB.
Could you elaborate on why that matters?
e.g. is this just for testing, or does this matter in some shutdown or hibernate scenario?
Quoted 24 lines
Signed-off-by Florian Fainelli <[email protected]>--- arch/arm64/kernel/psci.c | 2 +- arch/arm64/kernel/smp.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-)arch/arm64/kernel/psci.c +1
-1@@ -96,7 +96,7 @@ static int cpu_psci_cpu_kill(unsigned int cpu)for (i =0 ; i <10 ; i++) { err = psci_ops.affinity_info (cpu_logical_map (cpu),0 );if (err ==PSCI_0_2_AFFINITY_LEVEL_OFF ) { -pr_info ("CPU%d killed.\n" , cpu); +pr_debug ("CPU%d killed.\n" , cpu);return 0 ; }arch/arm64/kernel/smp.c +0
-0
arch/arm64/kernel/smp.c asmlinkage notrace void secondary_start_kernel(void) · line 264
Quoted 7 lines
arch/arm64/kernel/smp.c +1
-1@@ -259,7 +259,7 @@ asmlinkage notrace void secondary_start_kernel(void)* the CPU migration code to notice that the CPU is online * before we continue. */ -pr_info ("CPU%u: Booted secondary processor 0x%010lx [0x%08x]\n" , +pr_debug ("CPU%u: Booted secondary processor 0x%010lx [0x%08x]\n" , cpu, (unsigned long )mpidr,read_cpuid_id ());
I generally agree that we don't need to be verbose, and demoting these to debug is fine, but it's a shame that these won't be accessible in defconfig.
I wonder if we should enable DYNAMIC_DEBUG so that we can turn these on from the kernel command line, or if we should have something like a verbose_hotplug option specifically for these messages.
Thanks,
Mark.
Quoted 13 lines
arch/arm64/kernel/smp.c +1
-1update_cpu_boot_status (CPU_BOOT_SUCCESS );@@ -348,7 +348,7 @@ void __cpu_die(unsigned int cpu)pr_crit ("CPU%u: cpu didn't die\n" , cpu);return ; } -pr_notice ("CPU%u: shutdown\n" , cpu); +pr_debug ("CPU%u: shutdown\n" , cpu);/* * Now that the dying CPU is beyond the point of no return w.r.t. Signature · 1 line
2.17.1
Quoted 52 lines · Mark Rutland
Quoted 9 lines · Florian Fainelli
Similar to commits
c68b0274fb3cf("ARM: reduce "Booted secondary processor" message to debug level") and035e787543de7("ARM: 8644/1: Reduce "CPU:
shutdown" message to debug level"), demote thesecondary_start_kernel()and__cpu_die()messages from info, respectively notice to debug. While we are at it, also do this forcpu_psci_cpu_kill()which is redundant with__cpu_die().This helps improve the amount of possible hotplug cycles by around +50% on
ARCH_BRCMSTB.Could you elaborate on why that matters?
e.g. is this just for testing, or does this matter in some shutdown or hibernate scenario?
Quoted 24 lines
Signed-off-by Florian Fainelli <[email protected]>--- arch/arm64/kernel/psci.c | 2 +- arch/arm64/kernel/smp.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-)arch/arm64/kernel/psci.c +1
-1@@ -96,7 +96,7 @@ static int cpu_psci_cpu_kill(unsigned int cpu)for (i =0 ; i <10 ; i++) { err = psci_ops.affinity_info (cpu_logical_map (cpu),0 );if (err ==PSCI_0_2_AFFINITY_LEVEL_OFF ) { -pr_info ("CPU%d killed.\n" , cpu); +pr_debug ("CPU%d killed.\n" , cpu);return 0 ; }arch/arm64/kernel/smp.c +0
-0arch/arm64/kernel/smp.c asmlinkage notrace void secondary_start_kernel(void) · line 264
Quoted 7 lines
arch/arm64/kernel/smp.c +1
-1@@ -259,7 +259,7 @@ asmlinkage notrace void secondary_start_kernel(void)* the CPU migration code to notice that the CPU is online * before we continue. */ -pr_info ("CPU%u: Booted secondary processor 0x%010lx [0x%08x]\n" , +pr_debug ("CPU%u: Booted secondary processor 0x%010lx [0x%08x]\n" , cpu, (unsigned long )mpidr,read_cpuid_id ());I generally agree that we don't need to be verbose, and demoting these to debug is fine, but it's a shame that these won't be accessible in defconfig.
Quoted 3 lines
I wonder if we should enable
DYNAMIC_DEBUGso that we can turn these on from the kernel command line, or if we should have something like averbose_hotplugoption specifically for these messages.
I really wonder what kind of workload is hotplug-happy enough that these two messages become a bottleneck. Why isn't turning down the loglevel enough?
Thanks,
M.
Signature · 1 line
Jazz is not dead. It just smells funny...
Quoted 8 lines · Mark Rutland
Quoted 7 lines · Florian Fainelli
Similar to commits
c68b0274fb3cf("ARM: reduce "Booted secondary processor" message to debug level") and035e787543de7("ARM: 8644/1: Reduce "CPU:
shutdown" message to debug level"), demote thesecondary_start_kernel()and__cpu_die()messages from info, respectively notice to debug. While we are at it, also do this forcpu_psci_cpu_kill()which is redundant with__cpu_die().
Quoted 4 lines
Quoted 2 lines
This helps improve the amount of possible hotplug cycles by around +50% on
ARCH_BRCMSTB.Could you elaborate on why that matters?
Yeah, in general if you have a slow serial console then removing all the prints from the kernel makes lots of things much faster, but that's not necessarily a good argument for doing so. If that's a problem that really concerns users then I'd have to ask why they aren't using a stricter loglevel or a different console to begin with.
Quoted 39 lines
e.g. is this just for testing, or does this matter in some shutdown or hibernate scenario?
Quoted 24 lines
Signed-off-by Florian Fainelli <[email protected]>--- arch/arm64/kernel/psci.c | 2 +- arch/arm64/kernel/smp.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-)arch/arm64/kernel/psci.c +1
-1@@ -96,7 +96,7 @@ static int cpu_psci_cpu_kill(unsigned int cpu)for (i =0 ; i <10 ; i++) { err = psci_ops.affinity_info (cpu_logical_map (cpu),0 );if (err ==PSCI_0_2_AFFINITY_LEVEL_OFF ) { -pr_info ("CPU%d killed.\n" , cpu); +pr_debug ("CPU%d killed.\n" , cpu);return 0 ; }arch/arm64/kernel/smp.c +0
-0arch/arm64/kernel/smp.c asmlinkage notrace void secondary_start_kernel(void) · line 264
Quoted 7 lines
arch/arm64/kernel/smp.c +1
-1@@ -259,7 +259,7 @@ asmlinkage notrace void secondary_start_kernel(void)* the CPU migration code to notice that the CPU is online * before we continue. */ -pr_info ("CPU%u: Booted secondary processor 0x%010lx [0x%08x]\n" , +pr_debug ("CPU%u: Booted secondary processor 0x%010lx [0x%08x]\n" , cpu, (unsigned long )mpidr,read_cpuid_id ());I generally agree that we don't need to be verbose, and demoting these to debug is fine, but it's a shame that these won't be accessible in defconfig.
Quoted 3 lines
I wonder if we should enable
DYNAMIC_DEBUGso that we can turn these on from the kernel command line, or if we should have something like averbose_hotplugoption specifically for these messages.
We've had DYNAMIC_DEBUG=y in defconfig for a while already :/
Robin.
Quoted 12 lines
Thanks,
Mark.Quoted 8 lines
arch/arm64/kernel/smp.c +1
-1update_cpu_boot_status (CPU_BOOT_SUCCESS );@@ -348,7 +348,7 @@ void __cpu_die(unsigned int cpu)pr_crit ("CPU%u: cpu didn't die\n" , cpu);return ; } -pr_notice ("CPU%u: shutdown\n" , cpu); +pr_debug ("CPU%u: shutdown\n" , cpu);
Quoted 11 lines
Quoted 5 lines
arch/arm64/kernel/smp.c +0
-0/* * Now that the dying CPU is beyond the point of no return w.r.t. Signature · 1 line
2.17.1_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Quoted 13 lines · Robin Murphy
Quoted 12 lines · Mark Rutland
Quoted 10 lines · Florian Fainelli
Similar to commits
c68b0274fb3cf("ARM: reduce "Booted secondary processor" message to debug level") and035e787543de7("ARM: 8644/1:
Reduce "CPU:
shutdown" message to debug level"), demote thesecondary_start_kernel()and__cpu_die()messages from info, respectively notice to debug. While we are at it, also do this forcpu_psci_cpu_kill()which is redundant with__cpu_die().This helps improve the amount of possible hotplug cycles by around +50% on
ARCH_BRCMSTB.
Quoted 7 lines
Quoted 1 line
Could you elaborate on why that matters?
Yeah, in general if you have a slow serial console then removing all the prints from the kernel makes lots of things much faster, but that's not necessarily a good argument for doing so. If that's a problem that really concerns users then I'd have to ask why they aren't using a stricter loglevel or a different console to begin with.
See my response to Mark for specific use cases. Teaching users about changing their default print levels is certainly an option, although they will likely start wondering why other messages are now gone as well. There is not unfortunately a choice of a faster console on those platforms.
The print levels are not necessarily consistent (info vs. notice) and we have plenty of messages for when the CPU does not come online so telling us when it does is completely superfluous and does not bring much value and just gets in the way of being able to do that more often.
Quoted 69 lines
Quoted 39 lines
e.g. is this just for testing, or does this matter in some shutdown or hibernate scenario?
Quoted 23 lines
Signed-off-by Florian Fainelli <[email protected]>---
arch/arm64/kernel/psci.c| 2 +-
arch/arm64/kernel/smp.c | 4 ++--2 files changed, 3 insertions(+), 3 deletions(-)arch/arm64/kernel/psci.c +1
-1@@ -96,7 +96,7 @@ static int cpu_psci_cpu_kill(unsigned int cpu) for (i = 0; i < 10; i++) { err = psci_ops.affinity_info(cpu_logical_map(cpu), 0); if (err == PSCI_0_2_AFFINITY_LEVEL_OFF) { - pr_info("CPU%d killed.\n", cpu); + pr_debug("CPU%d killed.\n", cpu); return 0; }diff --git
a/arch/arm64/kernel/smp.cb/arch/arm64/kernel/smp.cindex824de7038967..71fd2b5a3f0e100644arch/arm64/kernel/smp.c asmlinkage notrace void secondary_start_kernel(void) · line 264
Quoted 7 lines
arch/arm64/kernel/smp.c +1
-1@@ -259,7 +259,7 @@ asmlinkage notrace void secondary_start_kernel(void)* the CPU migration code to notice that the CPU is online * before we continue. */ - pr_info("CPU%u: Booted secondary processor 0x%010lx [0x%08x]\n", + pr_debug("CPU%u: Booted secondary processor 0x%010lx [0x%08x]\n", cpu, (unsigned long)mpidr, read_cpuid_id());I generally agree that we don't need to be verbose, and demoting these to debug is fine, but it's a shame that these won't be accessible in defconfig.
Quoted 3 lines
I wonder if we should enable
DYNAMIC_DEBUGso that we can turn these on from the kernel command line, or if we should have something like averbose_hotplugoption specifically for these messages.We've had DYNAMIC_DEBUG=y in defconfig for a while already :/
Robin.
Quoted 22 lines
Thanks,
Mark.Quoted 9 lines
arch/arm64/kernel/smp.c +1
-1update_cpu_boot_status(CPU_BOOT_SUCCESS);@@ -348,7 +348,7 @@ void __cpu_die(unsigned int cpu) pr_crit("CPU%u: cpu didn't die\n", cpu); return; } - pr_notice("CPU%u: shutdown\n", cpu); + pr_debug("CPU%u: shutdown\n", cpu); /** Now that the dying CPU is beyond the point of no return w.r.t. Quoted 3 lines
--
2.17.1_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Signature · 1 line
Florian
Quoted 8 lines · Mark Rutland
Quoted 7 lines · Florian Fainelli
Similar to commits
c68b0274fb3cf("ARM: reduce "Booted secondary processor" message to debug level") and035e787543de7("ARM: 8644/1: Reduce "CPU:
shutdown" message to debug level"), demote thesecondary_start_kernel()and__cpu_die()messages from info, respectively notice to debug. While we are at it, also do this forcpu_psci_cpu_kill()which is redundant with__cpu_die().
Quoted 7 lines
Quoted 2 lines
This helps improve the amount of possible hotplug cycles by around +50% on
ARCH_BRCMSTB.Could you elaborate on why that matters?
e.g. is this just for testing, or does this matter in some shutdown or hibernate scenario?
It matters to both testing, e.g.: how many hotplug cycles you can go through within a given time frame to help assess system stsability, and during opportunistic suspend that e.g.: Android does.
Quoted 37 lines
Quoted 25 lines
Signed-off-by Florian Fainelli <[email protected]>--- arch/arm64/kernel/psci.c | 2 +- arch/arm64/kernel/smp.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-)arch/arm64/kernel/psci.c +1
-1@@ -96,7 +96,7 @@ static int cpu_psci_cpu_kill(unsigned int cpu)for (i =0 ; i <10 ; i++) { err = psci_ops.affinity_info (cpu_logical_map (cpu),0 );if (err ==PSCI_0_2_AFFINITY_LEVEL_OFF ) { -pr_info ("CPU%d killed.\n" , cpu); +pr_debug ("CPU%d killed.\n" , cpu);return 0 ; }arch/arm64/kernel/smp.c +0
-0arch/arm64/kernel/smp.c asmlinkage notrace void secondary_start_kernel(void) · line 264
Quoted 7 lines
arch/arm64/kernel/smp.c +1
-1@@ -259,7 +259,7 @@ asmlinkage notrace void secondary_start_kernel(void)* the CPU migration code to notice that the CPU is online * before we continue. */ -pr_info ("CPU%u: Booted secondary processor 0x%010lx [0x%08x]\n" , +pr_debug ("CPU%u: Booted secondary processor 0x%010lx [0x%08x]\n" , cpu, (unsigned long )mpidr,read_cpuid_id ());I generally agree that we don't need to be verbose, and demoting these to debug is fine, but it's a shame that these won't be accessible in defconfig.
Quoted 3 lines
I wonder if we should enable
DYNAMIC_DEBUGso that we can turn these on from the kernel command line, or if we should have something like averbose_hotplugoption specifically for these messages.
Based on Robin's answer (CONFIG_DYNAMIC_DEBUG already enabled in the default configuration), would you prefer this to become a dynamic debug message instead?
Signature · 1 line
Florian