Skip to contentSkip to conversation

PCI: make pci_name use dev_name

PATCH In review Greg Kroah-Hartman 6 participants 19 messages Jul 2, 2008 – Aug 7, 2008

Subsystems (5)
Network drivers USB PCI Memory technology devices (MTD) +1 more subsystems
lore
Share & tools
lore

Greg Kroah-Hartman

author
#0 lore raw

PATCH 1/1 PCI: make pci_name use dev_name

From: Greg Kroah-Hartman <[email protected]>

Also fixes up the sparc PCI code that was assuming this is not a constant.

This is done in anticipation of removing the bus_id field from struct driver.

Signed-off-by Kay Sievers <[email protected]>
Signed-off-by Greg Kroah-Hartman <[email protected]>
---
 arch/sparc64/kernel/pci.c |    2 +-
 include/linux/pci.h       |    4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)
arch/sparc64/kernel/pci.c +1 -1
@@ -408,7 +408,7 @@ struct pci_dev *of_create_pci_dev(struct
 	dev->class = class >> 8;
 	dev->revision = class & 0xff;
 
-	sprintf(pci_name(dev), "%04x:%02x:%02x.%d", pci_domain_nr(bus),
+	sprintf(dev->dev.bus_id, "%04x:%02x:%02x.%d", pci_domain_nr(bus),
 		dev->bus->number, PCI_SLOT(devfn), PCI_FUNC(devfn));
 
 	if (ofpci_verbose)
include/linux/pci.h +2 -2
@@ -977,9 +977,9 @@ static inline void pci_set_drvdata(struc
 /* If you want to know what to call your pci_dev, ask this function.
  * Again, it's a wrapper around the generic device.
  */
-static inline char *pci_name(struct pci_dev *pdev)
+static inline const char *pci_name(struct pci_dev *pdev)
 {
-	return pdev->dev.bus_id;
+	return dev_name(&pdev->dev);
 }
author
#1 lore raw
Quoted 30 lines · Greg Kroah-Hartman

From: Greg Kroah-Hartman <[email protected]>

Also fixes up the sparc PCI code that was assuming this is not a constant.

This is done in anticipation of removing the bus_id field from struct driver.

Signed-off-by Kay Sievers <[email protected]>
Signed-off-by Greg Kroah-Hartman <[email protected]>
---
 arch/sparc64/kernel/pci.c |    2 +-
 include/linux/pci.h       |    4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)
arch/sparc64/kernel/pci.c +1 -1
@@ -408,7 +408,7 @@ struct pci_dev *of_create_pci_dev(struct
 	dev->class = class >> 8;
 	dev->revision = class & 0xff;
 
-	sprintf(pci_name(dev), "%04x:%02x:%02x.%d", pci_domain_nr(bus),
+	sprintf(dev->dev.bus_id, "%04x:%02x:%02x.%d", pci_domain_nr(bus),
 		dev->bus->number, PCI_SLOT(devfn), PCI_FUNC(devfn));
 
 	if (ofpci_verbose)

include/linux/pci.h static inline void pci_set_drvdata(struc · line 980

Quoted 5 lines
include/linux/pci.h +1 -1
@@ -977,9 +977,9 @@ static inline void pci_set_drvdata(struc
 /* If you want to know what to call your pci_dev, ask this function.
  * Again, it's a wrapper around the generic device.
  */
-static inline char *pci_name(struct pci_dev *pdev)
+static inline const char *pci_name(struct pci_dev *pdev)

Andrew just pointed out to me that this causes a bunch of compiler warnings to be generated. I'll go audit the tree to fix them as well and send you a follow-on patch if that's ok.

thanks,

greg k-h

#2 lore raw
Quoted 27 lines · Greg Kroah-Hartman
Quoted 26 lines · Greg Kroah-Hartman

From: Greg Kroah-Hartman <[email protected]>

Also fixes up the sparc PCI code that was assuming this is not a constant.

This is done in anticipation of removing the bus_id field from struct driver.

Signed-off-by Kay Sievers <[email protected]>
Signed-off-by Greg Kroah-Hartman <[email protected]>
---
 arch/sparc64/kernel/pci.c |    2 +-
 include/linux/pci.h       |    4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)
arch/sparc64/kernel/pci.c +1 -1
@@ -408,7 +408,7 @@ struct pci_dev *of_create_pci_dev(struct
 	dev->class = class >> 8;
 	dev->revision = class & 0xff;
 
-	sprintf(pci_name(dev), "%04x:%02x:%02x.%d", pci_domain_nr(bus),
+	sprintf(dev->dev.bus_id, "%04x:%02x:%02x.%d", pci_domain_nr(bus),
 		dev->bus->number, PCI_SLOT(devfn), PCI_FUNC(devfn));
Quoted 13 lines

include/linux/pci.h static inline void pci_set_drvdata(struc · line 980

Quoted 9 lines
arch/sparc64/kernel/pci.c +0 -0
 	if (ofpci_verbose)
include/linux/pci.h +1 -1
@@ -977,9 +977,9 @@ static inline void pci_set_drvdata(struc
 /* If you want to know what to call your pci_dev, ask this function.
  * Again, it's a wrapper around the generic device.
  */
-static inline char *pci_name(struct pci_dev *pdev)
+static inline const char *pci_name(struct pci_dev *pdev)

Andrew just pointed out to me that this causes a bunch of compiler warnings to be generated. I'll go audit the tree to fix them as well and send you a follow-on patch if that's ok.

Yeah, works for me. I've got at least one other patch along these lines already. Assuming Dave doesn't want to handle these SPARC PCI patches anyway...

Thanks,
Jesse

author
#3 lore raw
Quoted 42 lines · Jesse Barnes
Quoted 27 lines · Greg Kroah-Hartman
Quoted 26 lines · Greg Kroah-Hartman

From: Greg Kroah-Hartman <[email protected]>

Also fixes up the sparc PCI code that was assuming this is not a constant.

This is done in anticipation of removing the bus_id field from struct driver.

Signed-off-by Kay Sievers <[email protected]>
Signed-off-by Greg Kroah-Hartman <[email protected]>
---
 arch/sparc64/kernel/pci.c |    2 +-
 include/linux/pci.h       |    4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)
arch/sparc64/kernel/pci.c +1 -1
@@ -408,7 +408,7 @@ struct pci_dev *of_create_pci_dev(struct
 	dev->class = class >> 8;
 	dev->revision = class & 0xff;
 
-	sprintf(pci_name(dev), "%04x:%02x:%02x.%d", pci_domain_nr(bus),
+	sprintf(dev->dev.bus_id, "%04x:%02x:%02x.%d", pci_domain_nr(bus),
 		dev->bus->number, PCI_SLOT(devfn), PCI_FUNC(devfn));
Quoted 13 lines

include/linux/pci.h static inline void pci_set_drvdata(struc · line 980

Quoted 9 lines
arch/sparc64/kernel/pci.c +0 -0
 	if (ofpci_verbose)
include/linux/pci.h +1 -1
@@ -977,9 +977,9 @@ static inline void pci_set_drvdata(struc
 /* If you want to know what to call your pci_dev, ask this function.
  * Again, it's a wrapper around the generic device.
  */
-static inline char *pci_name(struct pci_dev *pdev)
+static inline const char *pci_name(struct pci_dev *pdev)

Andrew just pointed out to me that this causes a bunch of compiler warnings to be generated. I'll go audit the tree to fix them as well and send you a follow-on patch if that's ok.

Quoted 3 lines

Yeah, works for me. I've got at least one other patch along these lines already. Assuming Dave doesn't want to handle these SPARC PCI patches anyway...

Well, it's only used in 837 different places, so it might take me a few hours to ensure I catch all the usages :(

thanks,

greg k-h

author
#4 lore raw
Quoted 43 lines · Greg KH
Quoted 42 lines · Jesse Barnes
Quoted 40 lines · Greg Kroah-Hartman
Quoted 30 lines · Greg Kroah-Hartman

From: Greg Kroah-Hartman <[email protected]>

Also fixes up the sparc PCI code that was assuming this is not a constant.

This is done in anticipation of removing the bus_id field from struct driver.

Signed-off-by Kay Sievers <[email protected]>
Signed-off-by Greg Kroah-Hartman <[email protected]>
---
 arch/sparc64/kernel/pci.c |    2 +-
 include/linux/pci.h       |    4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)
arch/sparc64/kernel/pci.c +1 -1
@@ -408,7 +408,7 @@ struct pci_dev *of_create_pci_dev(struct
 	dev->class = class >> 8;
 	dev->revision = class & 0xff;
 
-	sprintf(pci_name(dev), "%04x:%02x:%02x.%d", pci_domain_nr(bus),
+	sprintf(dev->dev.bus_id, "%04x:%02x:%02x.%d", pci_domain_nr(bus),
 		dev->bus->number, PCI_SLOT(devfn), PCI_FUNC(devfn));
 
 	if (ofpci_verbose)

include/linux/pci.h static inline void pci_set_drvdata(struc · line 980

Quoted 5 lines
include/linux/pci.h +1 -1
@@ -977,9 +977,9 @@ static inline void pci_set_drvdata(struc
 /* If you want to know what to call your pci_dev, ask this function.
  * Again, it's a wrapper around the generic device.
  */
-static inline char *pci_name(struct pci_dev *pdev)
+static inline const char *pci_name(struct pci_dev *pdev)

Andrew just pointed out to me that this causes a bunch of compiler warnings to be generated. I'll go audit the tree to fix them as well and send you a follow-on patch if that's ok.

Quoted 6 lines
Quoted 3 lines

Yeah, works for me. I've got at least one other patch along these lines already. Assuming Dave doesn't want to handle these SPARC PCI patches anyway...

Well, it's only used in 837 different places, so it might take me a few hours to ensure I catch all the usages :(

Ok, here are 4 patches that fix the few places that this pci change caused problems.

Jesse, do you want to add these to your tree? I can take the single USB one if you want, or you can take all 4, it's fine with me.

thanks,

greg k-h

author
#5 lore raw

PATCH 1/4 MTD: handle pci_name() being const

This changes the MTD core to handle pci_name() now returning a constant string.

Cc David Woodhouse <[email protected]>
Signed-off-by Greg Kroah-Hartman <[email protected]>
---
 drivers/mtd/devices/block2mtd.c |    8 +++++---
 include/linux/mtd/map.h         |    2 +-
 include/linux/mtd/mtd.h         |    2 +-
 3 files changed, 7 insertions(+), 5 deletions(-)
drivers/mtd/devices/block2mtd.c +5 -3
@@ -241,6 +241,7 @@ static struct block2mtd_dev *add_device(
 {
 	struct block_device *bdev;
 	struct block2mtd_dev *dev;
+	char *name;
 
 	if (!devname)
 		return NULL;

@@ -279,12 +280,13 @@ static struct block2mtd_dev *add_device(
 
 	/* Setup the MTD structure */
 	/* make the name contain the block device in */
-	dev->mtd.name = kmalloc(sizeof("block2mtd: ") + strlen(devname),
+	name = kmalloc(sizeof("block2mtd: ") + strlen(devname) + 1,
 			GFP_KERNEL);
-	if (!dev->mtd.name)
+	if (!name)
 		goto devinit_err;
 
-	sprintf(dev->mtd.name, "block2mtd: %s", devname);
+	sprintf(name, "block2mtd: %s", devname);
+	dev->mtd.name = name;
 
 	dev->mtd.size = dev->blkdev->bd_inode->i_size & PAGE_MASK;
 	dev->mtd.erasesize = erase_size;
include/linux/mtd/map.h +1 -1
@@ -189,7 +189,7 @@ typedef union {
 */
 
 struct map_info {
-	char *name;
+	const char *name;
 	unsigned long size;
 	resource_size_t phys;
 #define NO_XIP (-1UL)
include/linux/mtd/mtd.h +1 -1
@@ -121,7 +121,7 @@ struct mtd_info {
 	u_int32_t oobavail;  // Available OOB bytes per block
 
 	// Kernel-only stuff starts here.
-	char *name;
+	const char *name;
 	int index;
 
 	/* ecc layout structure pointer - read only ! */
author
#6 lore raw

PATCH 2/4 PCI: handle pci_name() being const

In reply to Greg KH

This changes pci_setup_device to handle pci_name() now returning a constant string.

Cc Jesse Barns <[email protected]>
Signed-off-by Greg Kroah-Hartman <[email protected]>
---
 drivers/pci/probe.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/pci/probe.c +3 -2
@@ -711,8 +711,9 @@ static int pci_setup_device(struct pci_d
 {
 	u32 class;
 
-	sprintf(pci_name(dev), "%04x:%02x:%02x.%d", pci_domain_nr(dev->bus),
-		dev->bus->number, PCI_SLOT(dev->devfn), PCI_FUNC(dev->devfn));
+	dev_set_name(&dev->dev, "%04x:%02x:%02x.%d", pci_domain_nr(dev->bus),
+		     dev->bus->number, PCI_SLOT(dev->devfn),
+		     PCI_FUNC(dev->devfn));
 
 	pci_read_config_dword(dev, PCI_CLASS_REVISION, &class);
 	dev->revision = class & 0xff;
#7 lore raw

Re: [patch 02/04] PCI: handle pci_name() being const

Quoted 14 lines · Greg Kroah-Hartman

This changes pci_setup_device to handle pci_name() now returning a constant string.

Cc Jesse Barns <[email protected]>
Signed-off-by Greg Kroah-Hartman <[email protected]>
---
 drivers/pci/probe.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/pci/probe.c +0 -0
@@ -711,8 +711,9 @@ static int pci_setup_device(struct pci_d
 {

drivers/pci/probe.c static int pci_setup_device(struct pci_d · line 719

Quoted 10 lines
drivers/pci/probe.c +3 -2
 	u32 class;
 
-	sprintf(pci_name(dev), "%04x:%02x:%02x.%d", pci_domain_nr(dev->bus),
-		dev->bus->number, PCI_SLOT(dev->devfn), PCI_FUNC(dev->devfn));
+	dev_set_name(&dev->dev, "%04x:%02x:%02x.%d", pci_domain_nr(dev->bus),
+		     dev->bus->number, PCI_SLOT(dev->devfn),
+		     PCI_FUNC(dev->devfn));
 
 	pci_read_config_dword(dev, PCI_CLASS_REVISION, &class);
 	dev->revision = class & 0xff;

Just applied this to my linux-next branch (fortunately I already had the dev_set_name commit). Doing a quick build & test before I push it out.

Thanks,
Jesse

author
#8 lore raw

PATCH 3/4 3c59x: handle pci_name() being const

In reply to Greg KH

This changes vortex_prob1() to handle pci_name() now returning a constant string.

Cc Steffen Klassert <[email protected]>
Signed-off-by Greg Kroah-Hartman <[email protected]>
---
 drivers/net/3c59x.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/3c59x.c +1 -1
@@ -1010,7 +1010,7 @@ static int __devinit vortex_probe1(struc
 	static int printed_version;
 	int retval, print_info;
 	struct vortex_chip_info * const vci = &vortex_info_tbl[chip_idx];
-	char *print_name = "3c59x";
+	const char *print_name = "3c59x";
 	struct pci_dev *pdev = NULL;
 	struct eisa_device *edev = NULL;
 	DECLARE_MAC_BUF(mac);
#9 lore raw

Re: [patch 03/04] 3c59x: handle pci_name() being const

Quoted 14 lines · On Wed, 2 Jul 2008 13:24:53 -0700 Greg Kroah-Hartman <[email protected]> wrote:

This changes vortex_prob1() to handle pci_name() now returning a constant string.

Cc Steffen Klassert <[email protected]>
Signed-off-by Greg Kroah-Hartman <[email protected]>
---
 drivers/net/3c59x.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/net/3c59x.c static int __devinit vortex_probe1(struc · line 1016

Quoted 8 lines
drivers/net/3c59x.c +1 -1
@@ -1010,7 +1010,7 @@ static int __devinit vortex_probe1(struc
 	static int printed_version;
 	int retval, print_info;
 	struct vortex_chip_info * const vci = &vortex_info_tbl[chip_idx];
-	char *print_name = "3c59x";
+	const char *print_name = "3c59x";
 	struct pci_dev *pdev = NULL;
 	struct eisa_device *edev = NULL;
 	DECLARE_MAC_BUF(mac);

Well this one applied. The others didn't.

Please prefer to not raise patches against a 30MB-old kernel ;)

#10 lore raw

Re: [patch 03/04] 3c59x: handle pci_name() being const

In reply to Greg Kroah-Hartman
Quoted 3 lines · Greg Kroah-Hartman

This changes vortex_prob1() to handle pci_name() now returning a constant string.

I got the same patch from David Howells already,
that one is acked and send to Jeff Garzik.

Steffen

author
#11 lore raw

Re: [patch 03/04] 3c59x: handle pci_name() being const

Quoted 6 lines · Steffen Klassert
Quoted 3 lines · Greg Kroah-Hartman

This changes vortex_prob1() to handle pci_name() now returning a constant string.

I got the same patch from David Howells already,
that one is acked and send to Jeff Garzik.

Great, thanks for letting us know.

greg k-h

#12 lore raw

Re: [patch 03/04] 3c59x: handle pci_name() being const

In reply to Steffen Klassert
Quoted 6 lines · Steffen Klassert
Quoted 3 lines · Greg Kroah-Hartman

This changes vortex_prob1() to handle pci_name() now returning a constant string.

I got the same patch from David Howells already,
that one is acked and send to Jeff Garzik.

could you resend, I only have the patch in quoted email...

#13 lore raw

Re: [patch 03/04] 3c59x: handle pci_name() being const

Quoted 9 lines · Jeff Garzik
Quoted 6 lines · Steffen Klassert
Quoted 3 lines · Greg Kroah-Hartman

This changes vortex_prob1() to handle pci_name() now returning a constant string.

I got the same patch from David Howells already,
that one is acked and send to Jeff Garzik.

could you resend, I only have the patch in quoted email...

The "3c59x: handle pci_name() being const" patch is already mainline, so there is no need to resend.

Steffen

author
#14 lore raw

PATCH 4/4 USB: handle pci_name() being const

In reply to Greg KH

This changes usb_create_hcd() to be able to handle the fact that pci_name() has changed to a constant string.

Signed-off-by Greg Kroah-Hartman <[email protected]>
---
 drivers/usb/core/hcd.c |    2 +-
 drivers/usb/core/hcd.h |    2 +-
 include/linux/usb.h    |    2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)
drivers/usb/core/hcd.c +1 -1
@@ -1755,7 +1755,7 @@ EXPORT_SYMBOL_GPL (usb_hc_died);
  * If memory is unavailable, returns NULL.
  */
 struct usb_hcd *usb_create_hcd (const struct hc_driver *driver,
-		struct device *dev, char *bus_name)
+		struct device *dev, const char *bus_name)
 {
 	struct usb_hcd *hcd;
drivers/usb/core/hcd.h +1 -1
@@ -233,7 +233,7 @@ extern void usb_hcd_disable_endpoint(str
 extern int usb_hcd_get_frame_number(struct usb_device *udev);
 
 extern struct usb_hcd *usb_create_hcd(const struct hc_driver *driver,
-		struct device *dev, char *bus_name);
+		struct device *dev, const char *bus_name);
 extern struct usb_hcd *usb_get_hcd(struct usb_hcd *hcd);
 extern void usb_put_hcd(struct usb_hcd *hcd);
 extern int usb_add_hcd(struct usb_hcd *hcd,
include/linux/usb.h +1 -1
@@ -293,7 +293,7 @@ struct usb_devmap {
 struct usb_bus {
 	struct device *controller;	/* host/master side hardware */
 	int busnum;			/* Bus number (in order of reg) */
-	char *bus_name;			/* stable id (PCI slot_name etc) */
+	const char *bus_name;		/* stable id (PCI slot_name etc) */
 	u8 uses_dma;			/* Does the host controller use DMA? */
 	u8 otg_port;			/* 0, or number of OTG/HNP port */
 	unsigned is_b_host:1;		/* true during some HNP roleswitches */
#15 lore raw
In reply to Greg KH
Quoted 9 lines · Greg KH
Quoted 2 lines · Greg KH

Well, it's only used in 837 different places, so it might take me a few hours to ensure I catch all the usages :(

Ok, here are 4 patches that fix the few places that this pci change caused problems.

Jesse, do you want to add these to your tree? I can take the single USB one if you want, or you can take all 4, it's fine with me.

I'd rather just take the PCI patch (which I'm doing now). The other stuff can come in through the appropriate tree.

Thanks,
Jesse

#16 lore raw
Quoted 5 lines · Jesse Barnes
Quoted 4 lines · Greg KH
Quoted 2 lines · Greg KH

Well, it's only used in 837 different places, so it might take me a few hours to ensure I catch all the usages :(

Quoted 8 lines
Quoted 5 lines

Ok, here are 4 patches that fix the few places that this pci change caused problems.

Jesse, do you want to add these to your tree? I can take the single USB one if you want, or you can take all 4, it's fine with me.

I'd rather just take the PCI patch (which I'm doing now). The other stuff can come in through the appropriate tree.

Fair enough, I'll make sure they don't get dropped.

thanks,

greg k-h

#17 lore raw
Quoted 9 lines · Greg KH
Quoted 8 lines · Jesse Barnes
Quoted 7 lines · Greg KH
Quoted 2 lines · Greg KH

Well, it's only used in 837 different places, so it might take me a few hours to ensure I catch all the usages :(

Ok, here are 4 patches that fix the few places that this pci change caused problems.

Quoted 7 lines
Quoted 5 lines
Quoted 2 lines

Jesse, do you want to add these to your tree? I can take the single USB one if you want, or you can take all 4, it's fine with me.

I'd rather just take the PCI patch (which I'm doing now). The other stuff can come in through the appropriate tree.

Fair enough, I'll make sure they don't get dropped.

Cool thanks. I assume you'll only push the pci_name change once all or most of the compiler warnings have been fixed up?

Thanks,
Jesse

#18 lore raw
Quoted 13 lines · Jesse Barnes
Quoted 12 lines · Greg KH
Quoted 11 lines · Jesse Barnes
Quoted 9 lines · Greg KH
Quoted 2 lines · Greg KH

Well, it's only used in 837 different places, so it might take me a few hours to ensure I catch all the usages :(

Ok, here are 4 patches that fix the few places that this pci change caused problems.

Jesse, do you want to add these to your tree? I can take the single USB one if you want, or you can take all 4, it's fine with me.

Quoted 7 lines
Quoted 4 lines
Quoted 2 lines

I'd rather just take the PCI patch (which I'm doing now). The other stuff can come in through the appropriate tree.

Fair enough, I'll make sure they don't get dropped.

Cool thanks. I assume you'll only push the pci_name change once all or most of the compiler warnings have been fixed up?

As there were only 4 patches for the warnings, yes, I can handle that :)

thanks,

greg k-h