Skip to contentSkip to conversation

drivers: pci: This patch fix the following checkpatch warning.

PATCH In review Mohan Kumar 2 participants 2 messages Apr 16, 2019 – Apr 17, 2019 PCI

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 drivers: pci: This patch fix the following checkpatch warning.

Use pr_err instead of printk

WARNING: Prefer [subsystem eg: netdev]_err([subsystem]dev, ... then dev_err(dev, ... then pr_err(... to printk(KERN_ERR ...

---
 drivers/pci/bus.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/pci/bus.c +1 -1
index 5cb40b2..2179a8b 100644
@@ -23,7 +23,7 @@ void pci_add_resource_offset(struct list_head *resources, struct resource *res,
 
 	entry = resource_list_create_entry(res, 0);
 	if (!entry) {
-		printk(KERN_ERR "PCI: can't add host bridge window %pR\n", res);
+		pr_err("PCI: can't add host bridge window %pR\n", res);
 		return;
 	}
 
Signature · 1 line
2.7.4
#1 lore raw

Hi Mohan,

Quoted 1 line · Mohan Kumar

Use pr_err instead of printk

I don't mind taking patches like these, but they do tend to clutter the revision history if we're not careful. So here are some hints:

- Make the subject line useful. The one here really doesn't tell me anything. Run "git log --oneline drivers/pci/bus.c" and follow the conventions there. Something like:

PCI: Convert printk(KERN_*) to pr_*()

- Do all the similar changes at once. I notice similar things in these files:

drivers/pci/pci-acpi.c
drivers/pci/pci-stub.c
drivers/pci/pci-sysfs.c
drivers/pci/pci.c
drivers/pci/pcie/aspm.c
drivers/pci/quirks.c
drivers/pci/setup-bus.c
drivers/pci/slot.c

plus several in drivers/pci/hotplug/. These could all be done in a single patch.

- If there are several in a file that use the same prefix, add a "#define pr_fmt()". This should be a separate patch so each patch does only one type of change, which makes them easier to review.

- There are several uses of "printk(KERN_DEBUG)", "dev_printk(KERN_DEBUG)", and "pci_printk(KERN_DEBUG)". The obvious thing would be to convert those to pr_debug(), dev_dbg(), and pci_dbg(), but I don't like that approach because pr_debug() and friends do different things depending on how the kernel is built. The uses in PCI provide information that I want to appear in the dmesg log *always*. So they could be (a) left alone or (b) converted from KERN_DEBUG to KERN_INFO (i.e., converted to pr_info(), dev_info(), etc).

Quoted 24 lines

WARNING: Prefer [subsystem eg: netdev]_err([subsystem]dev, ... then dev_err(dev, ... then pr_err(... to printk(KERN_ERR ...

---
 drivers/pci/bus.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/pci/bus.c +1 -1
index 5cb40b2..2179a8b 100644
@@ -23,7 +23,7 @@ void pci_add_resource_offset(struct list_head *resources, struct resource *res,
 
 	entry = resource_list_create_entry(res, 0);
 	if (!entry) {
-		printk(KERN_ERR "PCI: can't add host bridge window %pR\n", res);
+		pr_err("PCI: can't add host bridge window %pR\n", res);
 		return;
 	}
 
Signature · 1 line
2.7.4