PCI: pciehp: hide ENTRY messages behind ctrl_dbg()
PATCH 1/1 PCI: pciehp: hide ENTRY messages behind ctrl_dbg()
In each suspend and resume cycle my laptop prints these messages at KERN_INFO level:
pciehp 0000:00:1c.1:pcie04: pciehp_suspend ENTRY
pciehp 0000:00:1c.0:pcie04: pciehp_suspend ENTRYand
pciehp 0000:00:1c.0:pcie04: pciehp_resume ENTRY
pciehp 0000:00:1c.1:pcie04: pciehp_resume ENTRYMessages like these are probably only useful while debugging pciehp, so let's hide them behind the ctrl_dbg() macro.
---
0) Compile tested only!
1) Perhaps a better solution is to drop these messages entirely.
drivers/pci/hotplug/pciehp_core.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/pci/hotplug/pciehp_core.c +3 -2
@@ -293,7 +293,8 @@ static void pciehp_remove(struct pcie_device *dev)#ifdef CONFIG_PM static int pciehp_suspend (struct pcie_device *dev) { -dev_info (&dev->device,"%s ENTRY\n" , __func__); +struct controller *ctrl =get_service_data (dev); +ctrl_dbg (ctrl,"%s ENTRY\n" , __func__);return 0 ; }
@@ -303,8 +304,8 @@ static int pciehp_resume (struct pcie_device *dev)struct slot *slot;u8 status; -dev_info (&dev->device,"%s ENTRY\n" , __func__); ctrl =get_service_data (dev); +ctrl_dbg (ctrl,"%s ENTRY\n" , __func__);/* reinitialize the chipset's event detection logic */ pcie_enable_notification (ctrl);
Signature · 1 line
1.7.11.7
Bjorn Helgaas
maintainer[+cc Rafael]
Quoted 26 lines · Paul Bolle
In each suspend and resume cycle my laptop prints these messages at
KERN_INFOlevel:pciehp 0000:00:1c.1:pcie04: pciehp_suspend ENTRY pciehp 0000:00:1c.0:pcie04: pciehp_suspend ENTRYand
pciehp 0000:00:1c.0:pcie04: pciehp_resume ENTRY pciehp 0000:00:1c.1:pcie04: pciehp_resume ENTRYMessages like these are probably only useful while debugging pciehp, so let's hide them behind the
ctrl_dbg()macro.Signed-off-by Paul Bolle <[email protected]>---
0) Compile tested only!1) Perhaps a better solution is to drop these messages entirely.
drivers/pci/hotplug/pciehp_core.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)drivers/pci/hotplug/pciehp_core.c +0
-0
drivers/pci/hotplug/pciehp_core.c static void pciehp_remove(struct pcie_device *dev) · line 298
Quoted 7 lines
drivers/pci/hotplug/pciehp_core.c +2
-1@@ -293,7 +293,8 @@ static void pciehp_remove(struct pcie_device *dev)#ifdef CONFIG_PM static int pciehp_suspend (struct pcie_device *dev) { -dev_info (&dev->device,"%s ENTRY\n" , __func__); +struct controller *ctrl =get_service_data (dev); +ctrl_dbg (ctrl,"%s ENTRY\n" , __func__);return 0 ;
I'd propose to just drop these messages completely. Rafael, would you have any objection to that?
Quoted 15 lines
drivers/pci/hotplug/pciehp_core.c +1
-2}@@ -303,8 +304,8 @@ static int pciehp_resume (struct pcie_device *dev)struct slot *slot;u8 status; -dev_info (&dev->device,"%s ENTRY\n" , __func__); ctrl =get_service_data (dev); +ctrl_dbg (ctrl,"%s ENTRY\n" , __func__);/* reinitialize the chipset's event detection logic */ pcie_enable_notification (ctrl); --1.7.11.7
Quoted 24 lines · Bjorn Helgaas
[+cc Rafael]
Quoted 21 lines · Paul Bolle
In each suspend and resume cycle my laptop prints these messages at
KERN_INFOlevel:pciehp 0000:00:1c.1:pcie04: pciehp_suspend ENTRY pciehp 0000:00:1c.0:pcie04: pciehp_suspend ENTRYand
pciehp 0000:00:1c.0:pcie04: pciehp_resume ENTRY pciehp 0000:00:1c.1:pcie04: pciehp_resume ENTRYMessages like these are probably only useful while debugging pciehp, so let's hide them behind the
ctrl_dbg()macro.Signed-off-by Paul Bolle <[email protected]>---
0) Compile tested only!1) Perhaps a better solution is to drop these messages entirely.
drivers/pci/hotplug/pciehp_core.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
Quoted 15 lines
drivers/pci/hotplug/pciehp_core.c static void pciehp_remove(struct pcie_device *dev) · line 298
Quoted 12 lines
drivers/pci/hotplug/pciehp_core.c +2
-1@@ -293,7 +293,8 @@ static void pciehp_remove(struct pcie_device *dev)#ifdef CONFIG_PM static int pciehp_suspend (struct pcie_device *dev) { -dev_info (&dev->device,"%s ENTRY\n" , __func__); +struct controller *ctrl =get_service_data (dev); +ctrl_dbg (ctrl,"%s ENTRY\n" , __func__);return 0 ;I'd propose to just drop these messages completely. Rafael, would you have any objection to that?
No, I wouldn't.
Quoted 7 lines
Quoted 7 lines
drivers/pci/hotplug/pciehp_core.c +0
-0}@@ -303,8 +304,8 @@ static int pciehp_resume (struct pcie_device *dev)struct slot *slot;u8 status;
Quoted 7 lines
drivers/pci/hotplug/pciehp_core.c static int pciehp_resume (struct pcie_device *dev)
Quoted 7 lines
drivers/pci/hotplug/pciehp_core.c +1
-2-dev_info (&dev->device,"%s ENTRY\n" , __func__); ctrl =get_service_data (dev); +ctrl_dbg (ctrl,"%s ENTRY\n" , __func__);/* reinitialize the chipset's event detection logic */ pcie_enable_notification (ctrl); --
Thanks,
Rafael
Signature · 2 lines
I speak only for myself. Rafael J. Wysocki, Intel Open Source Technology Center.
PATCH v2 1/1 PCI: pciehp: drop ENTRY messages
In each suspend and resume cycle my laptop prints these messages at KERN_INFO level:
pciehp 0000:00:1c.1:pcie04: pciehp_suspend ENTRY
pciehp 0000:00:1c.0:pcie04: pciehp_suspend ENTRYand
pciehp 0000:00:1c.0:pcie04: pciehp_resume ENTRY
pciehp 0000:00:1c.1:pcie04: pciehp_resume ENTRYDrop these messages, that were probably used to debug the suspend and resume code, but now serve no purpose.
---
0) v1 was called "PCI: pciehp: hide ENTRY messages behind ctrl_dbg()". But Bjorn and Rafael prefer to drop these messages completely.
1) Not even compile tested!
drivers/pci/hotplug/pciehp_core.c | 2 -- 1 file changed, 2 deletions(-)
drivers/pci/hotplug/pciehp_core.c +0 -2
@@ -293,7 +293,6 @@ static void pciehp_remove(struct pcie_device *dev)#ifdef CONFIG_PM static int pciehp_suspend (struct pcie_device *dev) { -dev_info (&dev->device,"%s ENTRY\n" , __func__);return 0 ; }
@@ -303,7 +302,6 @@ static int pciehp_resume (struct pcie_device *dev)struct slot *slot;u8 status; -dev_info (&dev->device,"%s ENTRY\n" , __func__); ctrl =get_service_data (dev);/* reinitialize the chipset's event detection logic */
Signature · 1 line
1.7.11.7
Re: [PATCH v2] PCI: pciehp: drop ENTRY messages
Quoted 5 lines · Paul Bolle
In each suspend and resume cycle my laptop prints these messages at
KERN_INFOlevel:pciehp 0000:00:1c.1:pcie04: pciehp_suspend ENTRY pciehp 0000:00:1c.0:pcie04: pciehp_suspend ENTRY
Quoted 8 lines
and
pciehp 0000:00:1c.0:pcie04: pciehp_resume ENTRY pciehp 0000:00:1c.1:pcie04: pciehp_resume ENTRYDrop these messages, that were probably used to debug the suspend and resume code, but now serve no purpose.
Signed-off-by Paul Bolle <[email protected]>
I applied this to pci/misc for v3.9. Thanks!
Bjorn
Quoted 32 lines
---
0) v1 was called "PCI: pciehp: hide ENTRY messages behindctrl_dbg()". But Bjorn and Rafael prefer to drop these messages completely.1) Not even compile tested!
drivers/pci/hotplug/pciehp_core.c | 2 -- 1 file changed, 2 deletions(-)drivers/pci/hotplug/pciehp_core.c +0
-3@@ -293,7 +293,6 @@ static void pciehp_remove(struct pcie_device *dev)#ifdef CONFIG_PM static int pciehp_suspend (struct pcie_device *dev) { -dev_info (&dev->device,"%s ENTRY\n" , __func__);return 0 ; }@@ -303,7 +302,6 @@ static int pciehp_resume (struct pcie_device *dev)struct slot *slot;u8 status; -dev_info (&dev->device,"%s ENTRY\n" , __func__); ctrl =get_service_data (dev);/* reinitialize the chipset's event detection logic */ --1.7.11.7