patch "x86: MOVE PCI IO ECS code to x86/pci" breaks CPU hotplug
Quoted 9 lines
Converting
__cpuinitfunctions called out ofinit_amd()(and similar others) to__init(and making them subject ofxxx_initcall()handling isn't valid, as they would no longer be called for hot plugged CPUs.Further, since it's likely that in virtualized environments the MSR write would at best be ignored, I'd also recommend using the fault-safe accessors here *and* check that the bit actually got set before setting
PCI_HAS_IO_ECS(one would obviously have toBUG()when hot-plugged CPUs fail to set the bit when those available at boot successfully did so).
Even worse - this would even try to access the MSR on non-AMD CPUs (currently probably prevented just by the fact that only AMD ones use family values of 0x10 or higher).
Jan
Quoted 3 lines · Jan Beulich
Even worse - this would even try to access the MSR on non-AMD CPUs (currently probably prevented just by the fact that only AMD ones use family values of
0x10or higher).
Ups, really worse. Code was AMD specific before and became generic. Will fix this.
-Robert
Signature · 3 lines
Advanced Micro Devices, Inc. Operating System Research Center email: [email protected]
Quoted 8 lines · Robert Richter
Quoted 3 lines · Jan Beulich
Even worse - this would even try to access the MSR on non-AMD CPUs (currently probably prevented just by the fact that only AMD ones use family values of
0x10or higher).Ups, really worse. Code was AMD specific before and became generic. Will fix this.
it seems you didn't address the concern about hotplug as you promised
http://lkml.org/lkml/2008/6/13/151
YH
Quoted 6 lines · Yinghai Lu
Quoted 5 lines · Robert Richter
Quoted 3 lines · Jan Beulich
Even worse - this would even try to access the MSR on non-AMD CPUs (currently probably prevented just by the fact that only AMD ones use family values of
0x10or higher).
Quoted 7 lines
Quoted 3 lines
Ups, really worse. Code was AMD specific before and became generic. Will fix this.
it seems you didn't address the concern about hotplug as you promised
Yinghai, I know. You made a comment on this in reply to the patch and this is still an open issue. Really, will fix this. :-)
-Robert
Signature · 3 lines
Advanced Micro Devices, Inc. Operating System Research Center email: [email protected]
PATCH 1/1 x86: fix: do not run code in amd_bus.c on non-AMD CPUs
In reply to Jan BeulichQuoted 3 lines · Jan Beulich
Even worse - this would even try to access the MSR on non-AMD CPUs (currently probably prevented just by the fact that only AMD ones use family values of
0x10or higher).
This patch adds cpu vendor check to the postcore_initcalls.
--- arch/x86/pci/amd_bus.c | 19 ++++++++++++++++--- 1 files changed, 16 insertions(+), 3 deletions(-)
arch/x86/pci/amd_bus.c +16 -3
@@ -555,9 +555,11 @@ static int __init early_fill_mp_bus_info(void) return 0; } -postcore_initcall (early_fill_mp_bus_info); +#else /* !CONFIG_X86_64 */ -#endif +static int __init early_fill_mp_bus_info (void ) {return 0 ; } + +#endif /* !CONFIG_X86_64 */ /* common 32/64 bit code */
@@ -583,4 +585,15 @@ static int __init enable_pci_io_ecs(void)return 0 ; } -postcore_initcall (enable_pci_io_ecs); +static int __init amd_postcore_init (void ) +{ +if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +return 0 ; + +early_fill_mp_bus_info (); +enable_pci_io_ecs (); + +return 0 ; +} + +postcore_initcall (amd_postcore_init);
Signature · 1 line
1.5.6.4
Re: [PATCH] x86: fix: do not run code in amd_bus.c on non-AMD CPUs
Quoted 39 lines · Robert Richter
Quoted 3 lines · Jan Beulich
Even worse - this would even try to access the MSR on non-AMD CPUs (currently probably prevented just by the fact that only AMD ones use family values of
0x10or higher).This patch adds cpu vendor check to the
postcore_initcalls.Cc Jan Beulich <[email protected]>Cc Yinghai Lu <[email protected]>Signed-off-by Robert Richter <[email protected]>--- arch/x86/pci/amd_bus.c | 19 ++++++++++++++++--- 1 files changed, 16 insertions(+), 3 deletions(-)arch/x86/pci/amd_bus.c +6
-3@@ -555,9 +555,11 @@ static int __init early_fill_mp_bus_info(void) return 0; } -postcore_initcall (early_fill_mp_bus_info); +#else /* !CONFIG_X86_64 */ -#endif +static int __init early_fill_mp_bus_info (void ) {return 0 ; } + +#endif /* !CONFIG_X86_64 */ /* common 32/64 bit code */ @@ -583,4 +585,15 @@ static int __init enable_pci_io_ecs(void)return 0 ; } -postcore_initcall (enable_pci_io_ecs); +static int __init amd_postcore_init (void ) +{
arch/x86/pci/amd_bus.c static int __init enable_pci_io_ecs(void) · line 599
Quoted 10 lines
arch/x86/pci/amd_bus.c +10
-0+if (boot_cpu_data.x86_vendor !=X86_VENDOR_AMD ) +return 0 ; + +early_fill_mp_bus_info (); +enable_pci_io_ecs (); + +return 0 ; +} + +postcore_initcall (amd_postcore_init);
1. early_fill_mp_bus_info has checking with vendor/deviceid in itself.
2. you didn't address: when cpu is hot plug in later, enable_pci_to_ecs_per_cpu will not be called. ( maxcpus and addtional_cpus)
YH
Re: [PATCH] x86: fix: do not run code in amd_bus.c on non-AMD CPUs
Quoted 1 line · Yinghai Lu
1.
early_fill_mp_bus_infohas checking with vendor/deviceid in itself.
Right, but scanning the pci bus could be skipped for non-AMD cpus. If you don't like this check for your function I will resubmit a new patch.
Quoted 3 lines
2. you didn't address: when cpu is hot plug in later,
enable_pci_to_ecs_per_cpuwill not be called. ( maxcpus andaddtional_cpus)
Next patch?
-Robert
Signature · 3 lines
Advanced Micro Devices, Inc. Operating System Research Center email: [email protected]
Re: [PATCH] x86: fix: do not run code in amd_bus.c on non-AMD CPUs
Quoted 6 lines · Robert Richter
Quoted 1 line · Yinghai Lu
1.
early_fill_mp_bus_infohas checking with vendor/deviceid in itself.Right, but scanning the pci bus could be skipped for non-AMD cpus. If you don't like this check for your function I will resubmit a new patch.
just fine, if you check vendor before that, we should remove check vendor again in later function.
YH
Re: [PATCH] x86: fix: do not run code in amd_bus.c on non-AMD CPUs
Quoted 10 lines · Yinghai Lu
Quoted 6 lines · Robert Richter
Quoted 1 line · Yinghai Lu
1.
early_fill_mp_bus_infohas checking with vendor/deviceid in itself.Right, but scanning the pci bus could be skipped for non-AMD cpus. If you don't like this check for your function I will resubmit a new patch.
just fine, if you check vendor before that, we should remove check vendor again in later function.
Right. There are no more checks in, only the PCI vendor check that can not be removed.
-Robert
Quoted 3 lines
YH
Signature · 3 lines
Advanced Micro Devices, Inc. Operating System Research Center email: [email protected]
PATCH 1/1 x86: fix: make PCI ECS for AMD CPUs hotplug capable
In reply to Jan BeulichUntil now, PCI ECS setup was performed at boot time only and for cpus that are enabled then. This patch fixes this and adds cpu hotplug.
Tests sequence (check if ECS bit is set when bringing cpu online again):
# ( perl -e 'sysseek(STDIN, 0xC001001F, 0)'; hexdump -n 8 -e '2/4 "%08x " "\n"' ) < /dev/cpu/1/msr
00000008 00404010
# ( perl -e 'sysseek(STDOUT, 0xC001001F, 0); print pack "l*", 8, 0x00400010' ) > /dev/cpu/1/msr
# ( perl -e 'sysseek(STDIN, 0xC001001F, 0)'; hexdump -n 8 -e '2/4 "%08x " "\n"' ) < /dev/cpu/1/msr
00000008 00400010
# echo 0 > /sys/devices/system/cpu/cpu1/online # echo 1 > /sys/devices/system/cpu/cpu1/online # ( perl -e 'sysseek(STDIN, 0xC001001F, 0)'; hexdump -n 8 -e '2/4 "%08x " "\n"' ) < /dev/cpu/1/msr 00000008 00404010
--- arch/x86/pci/amd_bus.c | 35 +++++++++++++++++++++++++++++++---- 1 files changed, 31 insertions(+), 4 deletions(-)
arch/x86/pci/amd_bus.c +31 -4
@@ -1,6 +1,7 @@#include <linux/init.h> #include <linux/pci.h> #include <linux/topology.h> +#include <linux/cpu.h> #include "pci.h" #ifdef CONFIG_X86_64
@@ -565,7 +566,7 @@ static int __init early_fill_mp_bus_info(void) { return 0; }#define ENABLE_CF8_EXT_CFG (1ULL << 46) -static void enable_pci_io_ecs_per_cpu (void *unused) +static void enable_pci_io_ecs (void *unused) {u64 reg;rdmsrl (MSR_AMD64_NB_CFG , reg);
@@ -575,13 +576,39 @@ static void enable_pci_io_ecs_per_cpu(void *unused) } } -static int __init enable_pci_io_ecs (void ) +static int __cpuinit amd_cpu_notify (struct notifier_block *self, +unsigned long action,void *hcpu) { +int cpu = (long )hcpu; +switch (action) { +case CPU_ONLINE : +case CPU_ONLINE_FROZEN : +smp_call_function_single (cpu, enable_pci_io_ecs,NULL ,0 ); +break ; +default : +break ; + } +return NOTIFY_OK ; +} + +static struct notifier_block __cpuinitdata amd_cpu_notifier = { + .notifier_call = amd_cpu_notify, +}; + +static int __init pci_io_ecs_init (void ) +{ +int cpu; +/* assume all cpus from fam10h have IO ECS */ if (boot_cpu_data.x86 <0x10 )return 0 ; -on_each_cpu (enable_pci_io_ecs_per_cpu,NULL ,1 ); + +register_cpu_notifier (&amd_cpu_notifier); +for_each_online_cpu (cpu) +amd_cpu_notify (&amd_cpu_notifier, (unsigned long )CPU_ONLINE , + (void *)(long )cpu); pci_probe |=PCI_HAS_IO_ECS ; +return 0 ; }
@@ -591,7 +618,7 @@ static int __init amd_postcore_init(void)return 0 ;early_fill_mp_bus_info (); -enable_pci_io_ecs (); +pci_io_ecs_init ();return 0 ; }
Signature · 1 line
1.5.6.4
Re: [PATCH] x86: fix: make PCI ECS for AMD CPUs hotplug capable
Quoted 70 lines · Robert Richter
Until now, PCI ECS setup was performed at boot time only and for cpus that are enabled then. This patch fixes this and adds cpu hotplug.
Tests sequence (check if ECS bit is set when bringing cpu online again):
# ( perl -e '
sysseek(STDIN, 0xC001001F, 0)'; hexdump -n 8 -e '2/4 "%08x " "\n"' ) </dev/cpu/1/msr
00000008 00404010
# ( perl -e 'sysseek(STDOUT, 0xC001001F, 0); print pack "l*", 8,0x00400010' ) >/dev/cpu/1/msr
# ( perl -e 'sysseek(STDIN, 0xC001001F, 0)'; hexdump -n 8 -e '2/4 "%08x " "\n"' ) </dev/cpu/1/msr
00000008 00400010# echo 0 > /sys/devices/system/cpu/cpu1/online # echo 1 > /sys/devices/system/cpu/cpu1/online # ( perl -e 'sysseek(STDIN, 0xC001001F, 0)'; hexdump -n 8 -e '2/4 "%08x " "\n"' ) < /dev/cpu/1/msr 00000008 00404010Cc Jan Beulich <[email protected]>Cc Yinghai Lu <[email protected]>Signed-off-by Robert Richter <[email protected]>--- arch/x86/pci/amd_bus.c | 35 +++++++++++++++++++++++++++++++---- 1 files changed, 31 insertions(+), 4 deletions(-)arch/x86/pci/amd_bus.c +23
-2@@ -1,6 +1,7 @@#include <linux/init.h> #include <linux/pci.h> #include <linux/topology.h> +#include <linux/cpu.h> #include "pci.h" #ifdef CONFIG_X86_64 @@ -565,7 +566,7 @@ static int __init early_fill_mp_bus_info(void) { return 0; }#define ENABLE_CF8_EXT_CFG (1ULL << 46) -static void enable_pci_io_ecs_per_cpu (void *unused) +static void enable_pci_io_ecs (void *unused) {u64 reg;rdmsrl (MSR_AMD64_NB_CFG , reg);@@ -575,13 +576,39 @@ static void enable_pci_io_ecs_per_cpu(void *unused) } } -static int __init enable_pci_io_ecs (void ) +static int __cpuinit amd_cpu_notify (struct notifier_block *self, +unsigned long action,void *hcpu) { +int cpu = (long )hcpu; +switch (action) { +case CPU_ONLINE : +case CPU_ONLINE_FROZEN : +smp_call_function_single (cpu, enable_pci_io_ecs,NULL ,0 ); +break ; +default : +break ; + } +return NOTIFY_OK ; +} + +static struct notifier_block __cpuinitdata amd_cpu_notifier = { + .notifier_call = amd_cpu_notify, +}; + +static int __init pci_io_ecs_init (void ) +{ +int cpu;
arch/x86/pci/amd_bus.c static void enable_pci_io_ecs_per_cpu(void *unused) · line 609
Quoted 10 lines
arch/x86/pci/amd_bus.c +6
-1+/* assume all cpus from fam10h have IO ECS */ if (boot_cpu_data.x86 <0x10 )return 0 ; -on_each_cpu (enable_pci_io_ecs_per_cpu,NULL ,1 ); + +register_cpu_notifier (&amd_cpu_notifier); +for_each_online_cpu (cpu) +amd_cpu_notify (&amd_cpu_notifier, (unsigned long )CPU_ONLINE , + (void *)(long )cpu);
wonder if those two lines should be reversed.
Quoted 19 lines
arch/x86/pci/amd_bus.c +2
-2pci_probe |=PCI_HAS_IO_ECS ; +return 0 ; }@@ -591,7 +618,7 @@ static int __init amd_postcore_init(void)return 0 ;early_fill_mp_bus_info (); -enable_pci_io_ecs (); +pci_io_ecs_init ();return 0 ; } --1.5.6.4
Re: [PATCH] x86: fix: make PCI ECS for AMD CPUs hotplug capable
On 22.08.08 11:56:08, Yinghai Lu wrote:
[...]
Quoted 7 lines
Quoted 7 lines
arch/x86/pci/amd_bus.c +4
-0+static int __init pci_io_ecs_init (void ) +{ +int cpu; +/* assume all cpus from fam10h have IO ECS */ if (boot_cpu_data.x86 <0x10 )return 0 ;
Quoted 8 lines
arch/x86/pci/amd_bus.c static void enable_pci_io_ecs_per_cpu(void *unused) · line 609
Quoted 6 lines
arch/x86/pci/amd_bus.c +5
-1-on_each_cpu (enable_pci_io_ecs_per_cpu,NULL ,1 ); + +register_cpu_notifier (&amd_cpu_notifier); +for_each_online_cpu (cpu) +amd_cpu_notify (&amd_cpu_notifier, (unsigned long )CPU_ONLINE , + (void *)(long )cpu);wonder if those two lines should be reversed.
Do you mean setting PCI_HAS_IO_ECS before for_each_online_cpu(...)? PCI_HAS_IO_ECS will be used only in pci_direct_init(). PCI is initialized in a later init stage, so it doesn't matter. My intention was to set the bit after the setup of all online cpus is finished.
-Robert
Quoted 21 lines
Quoted 20 lines
pci_probe |= PCI_HAS_IO_ECS;arch/x86/pci/amd_bus.c +2
-2+return 0 ; }@@ -591,7 +618,7 @@ static int __init amd_postcore_init(void)return 0 ;early_fill_mp_bus_info (); -enable_pci_io_ecs (); +pci_io_ecs_init ();return 0 ; } --1.5.6.4
Signature · 3 lines
Advanced Micro Devices, Inc. Operating System Research Center email: [email protected]
Re: [PATCH] x86: fix: make PCI ECS for AMD CPUs hotplug capable
Quoted 20 lines · Robert Richter
On 22.08.08 11:56:08, Yinghai Lu wrote:
[...]
Quoted 7 lines
Quoted 7 lines
arch/x86/pci/amd_bus.c +4
-0+static int __init pci_io_ecs_init (void ) +{ +int cpu; +/* assume all cpus from fam10h have IO ECS */ if (boot_cpu_data.x86 <0x10 )return 0 ;Quoted 8 lines
arch/x86/pci/amd_bus.c static void enable_pci_io_ecs_per_cpu(void *unused) · line 609
Quoted 6 lines
arch/x86/pci/amd_bus.c +5
-1-on_each_cpu (enable_pci_io_ecs_per_cpu,NULL ,1 ); + +register_cpu_notifier (&amd_cpu_notifier); +for_each_online_cpu (cpu) +amd_cpu_notify (&amd_cpu_notifier, (unsigned long )CPU_ONLINE , + (void *)(long )cpu);wonder if those two lines should be reversed.
Quoted 4 lines
Do you mean setting
PCI_HAS_IO_ECSbeforefor_each_online_cpu(...)?PCI_HAS_IO_ECSwill be used only inpci_direct_init(). PCI is initialized in a later init stage, so it doesn't matter. My intention was to set the bit after the setup of all online cpus is finished.
==>
+ for_each_online_cpu(cpu)
+ amd_cpu_notify(&amd_cpu_notifier, (unsigned long)CPU_ONLINE,
+ (void *)(long)cpu);
+ register_cpu_notifier(&amd_cpu_notifier);
YH
Re: [PATCH] x86: fix: make PCI ECS for AMD CPUs hotplug capable
Quoted 28 lines · Yinghai Lu
Quoted 20 lines · Robert Richter
On 22.08.08 11:56:08, Yinghai Lu wrote:
[...]
Quoted 7 lines
Quoted 7 lines
arch/x86/pci/amd_bus.c +4
-0+static int __init pci_io_ecs_init (void ) +{ +int cpu; +/* assume all cpus from fam10h have IO ECS */ if (boot_cpu_data.x86 <0x10 )return 0 ;Quoted 8 lines
arch/x86/pci/amd_bus.c static void enable_pci_io_ecs_per_cpu(void *unused) · line 609
Quoted 6 lines
arch/x86/pci/amd_bus.c +5
-1-on_each_cpu (enable_pci_io_ecs_per_cpu,NULL ,1 ); + +register_cpu_notifier (&amd_cpu_notifier); +for_each_online_cpu (cpu) +amd_cpu_notify (&amd_cpu_notifier, (unsigned long )CPU_ONLINE , + (void *)(long )cpu);wonder if those two lines should be reversed.
Quoted 4 lines
Do you mean setting
PCI_HAS_IO_ECSbeforefor_each_online_cpu(...)?PCI_HAS_IO_ECSwill be used only inpci_direct_init(). PCI is initialized in a later init stage, so it doesn't matter. My intention was to set the bit after the setup of all online cpus is finished.==>
arch/x86/pci/amd_bus.c static void enable_pci_io_ecs_per_cpu(void *unused)
Quoted 4 lines
arch/x86/pci/amd_bus.c +4
-0+for_each_online_cpu (cpu) +amd_cpu_notify (&amd_cpu_notifier, (unsigned long )CPU_ONLINE , + (void *)(long )cpu); +register_cpu_notifier (&amd_cpu_notifier);
Hmm, most code in the kernel registers first. Theoretical, with this code, if a cpu would go online in between, it wouldn't be initialized.
-Robert
Quoted 3 lines
YH
Signature · 3 lines
Advanced Micro Devices, Inc. Operating System Research Center email: [email protected]
Re: [PATCH] x86: fix: make PCI ECS for AMD CPUs hotplug capable
In reply to Robert RichterQuoted 2 lines · Robert Richter
Until now, PCI ECS setup was performed at boot time only and for cpus that are enabled then. This patch fixes this and adds cpu hotplug.
this patch does not apply cleanly to -git.
Ingo
Re: [PATCH] x86: fix: make PCI ECS for AMD CPUs hotplug capable
Quoted 7 lines · Ingo Molnar
Quoted 3 lines · Robert Richter
Until now, PCI ECS setup was performed at boot time only and for cpus that are enabled then. This patch fixes this and adds cpu hotplug.
this patch does not apply cleanly to -git.
i see, it depends on the other patch. Applied both to tip/x86/urgent.
Ingo
PATCH 1/1 x86: fix: make PCI ECS for AMD CPUs hotplug capable (-v2)
In reply to Ingo MolnarUntil now, PCI ECS setup was performed at boot time only and for cpus that are enabled then. This patch fixes this and adds cpu hotplug.
Tests sequence (check if ECS bit is set when bringing cpu online again):
# ( perl -e 'sysseek(STDIN, 0xC001001F, 0)'; hexdump -n 8 -e '2/4 "%08x " "\n"' ) < /dev/cpu/1/msr
00000008 00404010
# ( perl -e 'sysseek(STDOUT, 0xC001001F, 0); print pack "l*", 8, 0x00400010' ) > /dev/cpu/1/msr
# ( perl -e 'sysseek(STDIN, 0xC001001F, 0)'; hexdump -n 8 -e '2/4 "%08x " "\n"' ) < /dev/cpu/1/msr
00000008 00400010
# echo 0 > /sys/devices/system/cpu/cpu1/online # echo 1 > /sys/devices/system/cpu/cpu1/online # ( perl -e 'sysseek(STDIN, 0xC001001F, 0)'; hexdump -n 8 -e '2/4 "%08x " "\n"' ) < /dev/cpu/1/msr 00000008 00404010
--- arch/x86/pci/amd_bus.c | 35 +++++++++++++++++++++++++++++++---- 1 files changed, 31 insertions(+), 4 deletions(-)
arch/x86/pci/amd_bus.c +31 -4
@@ -1,6 +1,7 @@#include <linux/init.h> #include <linux/pci.h> #include <linux/topology.h> +#include <linux/cpu.h> #include "pci.h" #ifdef CONFIG_X86_64
@@ -565,7 +566,7 @@ static int __init early_fill_mp_bus_info(void) { return 0; }#define ENABLE_CF8_EXT_CFG (1ULL << 46) -static void enable_pci_io_ecs_per_cpu (void *unused) +static void enable_pci_io_ecs (void *unused) {u64 reg;rdmsrl (MSR_AMD64_NB_CFG , reg);
@@ -575,13 +576,39 @@ static void enable_pci_io_ecs_per_cpu(void *unused) } } -static int __init enable_pci_io_ecs (void ) +static int __cpuinit amd_cpu_notify (struct notifier_block *self, +unsigned long action,void *hcpu) { +int cpu = (long )hcpu; +switch (action) { +case CPU_ONLINE : +case CPU_ONLINE_FROZEN : +smp_call_function_single (cpu, enable_pci_io_ecs,NULL ,0 ); +break ; +default : +break ; + } +return NOTIFY_OK ; +} + +static struct notifier_block __cpuinitdata amd_cpu_notifier = { + .notifier_call = amd_cpu_notify, +}; + +static int __init pci_io_ecs_init (void ) +{ +int cpu; +/* assume all cpus from fam10h have IO ECS */ if (boot_cpu_data.x86 <0x10 )return 0 ; -on_each_cpu (enable_pci_io_ecs_per_cpu,NULL ,1 ); + +register_cpu_notifier (&amd_cpu_notifier); +for_each_online_cpu (cpu) +amd_cpu_notify (&amd_cpu_notifier, (unsigned long )CPU_ONLINE , + (void *)(long )cpu); pci_probe |=PCI_HAS_IO_ECS ; +return 0 ; }
@@ -591,7 +618,7 @@ static int __init amd_postcore_init(void)return 0 ;early_fill_mp_bus_info (); -enable_pci_io_ecs (); +pci_io_ecs_init ();return 0 ; }
Signature · 1 line
1.5.6.4
Re: [PATCH] x86: fix: make PCI ECS for AMD CPUs hotplug capable (-v2)
Quoted 2 lines · Robert Richter
Until now, PCI ECS setup was performed at boot time only and for cpus that are enabled then. This patch fixes this and adds cpu hotplug.
i already had v1 applied to tip/x86/urgent, so i took the -v2 delta and queued it up in tip/x86/cleanups - see the commit below.
Ingo
--------------->
Quoted 1 line
From
ed21763e7b0b3fb50e4efd9d4bc17ef5b035d304Mon Sep 17 00:00:00 2001
From: Robert Richter <[email protected]>
Date: Fri, 22 Aug 2008 20:23:38 +0200
Subject: [PATCH] x86: cleanup in amd_cpu_notify()
small coding style fix.
--- arch/x86/pci/amd_bus.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
arch/x86/pci/amd_bus.c +1 -1
@@ -580,7 +580,7 @@ static int __cpuinit amd_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { int cpu = (long)hcpu; -switch (action) { +switch (action) {case CPU_ONLINE :case CPU_ONLINE_FROZEN :smp_call_function_single (cpu, enable_pci_io_ecs,NULL ,0 );