[PATCH 00/12] Fix various section mismatches and build errors.

[PATCH 00/12] Fix various section mismatches and build errors.

[Ralf Baechle]

I'm getting screen and screens full of section mismatches from my test
builds of the current kernel to the point where it's sometimes more
meaningful messages get hidden by the bulk of mismatches.  This is the
first round of fixes with more to come.

  Ralf

 drivers/gpu/drm/radeon/radeon_clocks.c |    4 ++--
 drivers/leds/leds-lp5521.c             |    4 ++--
 drivers/leds/leds-lp5523.c             |    4 ++--
 drivers/misc/ioc4.c                    |    2 +-
 drivers/net/3c509.c                    |    2 +-
 drivers/net/3c59x.c                    |    2 +-
 drivers/net/depca.c                    |   25 +++++++++++++------------
 drivers/net/hp100.c                    |    2 +-
 drivers/net/ne3210.c                   |   12 +++++++-----
 drivers/net/tulip/de4x5.c              |    2 +-
 drivers/scsi/sim710.c                  |    2 +-
 drivers/tty/serial/Kconfig             |    2 +-
 12 files changed, 33 insertions(+), 30 deletions(-)

[PATCH 02/12] NET: 3c509: Fix section mismatch

[Ralf Baechle]

WARNING: drivers/net/3c509.o(.data+0x190): Section mismatch in reference from the variable el3_eisa_driver to the function .init.text:el3_eisa_probe()
The variable el3_eisa_driver references
the function __init el3_eisa_probe()
If the reference is valid then annotate the
variable with __init* or __refdata (see linux/init.h) or name the variable:
*_template, *_timer, *_sht, *_ops, *_probe, *_probe_one, *_console

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
To: "David S. Miller" <davem@davemloft.net>
Cc: netdev@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: linux-mips@linux-mips.org
---
 drivers/net/3c509.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/net/3c509.c b/drivers/net/3c509.c
index 44b28b2..dc4bf07 100644
--- a/drivers/net/3c509.c
+++ b/drivers/net/3c509.c
@@ -671,7 +671,7 @@ static int __init el3_mca_probe(struct device *device)
 #endif /* CONFIG_MCA */
 
 #ifdef CONFIG_EISA
-static int __init el3_eisa_probe (struct device *device)
+static int __devinit el3_eisa_probe(struct device *device)
 {
 	short i;
 	int ioaddr, irq, if_port;
-- 
1.7.4.4

[PATCH 03/12] NET: 3c59x: Fix section mismatch.

[Ralf Baechle]

WARNING: drivers/net/3c59x.o(.data+0x40): Section mismatch in reference from the variable vortex_eisa_driver to the function .init.text:vortex_eisa_probe()
The variable vortex_eisa_driver references
the function __init vortex_eisa_probe()
If the reference is valid then annotate the
variable with __init* or __refdata (see linux/init.h) or name the variable:
*_template, *_timer, *_sht, *_ops, *_probe, *_probe_one, *_console

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
To: "David S. Miller" <davem@davemloft.net>
Cc: Steffen Klassert <klassert@mathematik.tu-chemnitz.de>
Cc: netdev@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: linux-mips@linux-mips.org
---
 drivers/net/3c59x.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/net/3c59x.c b/drivers/net/3c59x.c
index 8cc2256..f33fecc 100644
--- a/drivers/net/3c59x.c
+++ b/drivers/net/3c59x.c
@@ -908,7 +908,7 @@ static struct eisa_device_id vortex_eisa_ids[] = {
 };
 MODULE_DEVICE_TABLE(eisa, vortex_eisa_ids);
 
-static int __init vortex_eisa_probe(struct device *device)
+static int __devinit vortex_eisa_probe(struct device *device)
 {
 	void __iomem *ioaddr;
 	struct eisa_device *edev;
-- 
1.7.4.4

[PATCH 04/12] NET: depca: Fix bucketload full of section mismatches.

[Ralf Baechle]

WARNING: drivers/net/depca.o(.data+0x34): Section mismatch in reference from the variable depca_eisa_driver to the function .init.text:depca_eisa_probe()
The variable depca_eisa_driver references
the function __init depca_eisa_probe()
If the reference is valid then annotate the
variable with __init* or __refdata (see linux/init.h) or name the variable:
*_template, *_timer, *_sht, *_ops, *_probe, *_probe_one, *_console

WARNING: drivers/net/depca.o(.devinit.text+0x2c): Section mismatch in reference from the function depca_isa_probe() to the function .init.text:depca_common_init()
The function __devinit depca_isa_probe() references
a function __init depca_common_init().
If depca_common_init is only used by depca_isa_probe then
annotate depca_common_init with a matching annotation.

WARNING: drivers/net/depca.o(.devinit.text+0x44): Section mismatch in reference from the function depca_isa_probe() to the function .init.text:depca_shmem_probe()
The function __devinit depca_isa_probe() references
a function __init depca_shmem_probe().
If depca_shmem_probe is only used by depca_isa_probe then
annotate depca_shmem_probe with a matching annotation.

WARNING: drivers/net/depca.o(.devinit.text+0x8c): Section mismatch in reference from the function depca_isa_probe() to the function .init.text:depca_hw_init()
The function __devinit depca_isa_probe() references
a function __init depca_hw_init().
If depca_hw_init is only used by depca_isa_probe then
annotate depca_hw_init with a matching annotation.

Fixing these in turn triggers yet more mismatches, fix those as well.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
To: "David S. Miller" <davem@davemloft.net>
Cc: netdev@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: linux-mips@linux-mips.org
---
 drivers/net/depca.c |   25 +++++++++++++------------
 1 files changed, 13 insertions(+), 12 deletions(-)

diff --git a/drivers/net/depca.c b/drivers/net/depca.c
index 8b0084d..d40536a 100644
--- a/drivers/net/depca.c
+++ b/drivers/net/depca.c
@@ -331,7 +331,7 @@ static struct {
                          "DE422",\
                          ""}
 
-static char* __initdata depca_signature[] = DEPCA_SIGNATURE;
+static char* __devinitdata depca_signature[] = DEPCA_SIGNATURE;
 
 enum depca_type {
 	DEPCA, de100, de101, de200, de201, de202, de210, de212, de422, unknown
@@ -541,9 +541,9 @@ static void SetMulticastFilter(struct net_device *dev);
 static int load_packet(struct net_device *dev, struct sk_buff *skb);
 static void depca_dbg_open(struct net_device *dev);
 
-static u_char de1xx_irq[] __initdata = { 2, 3, 4, 5, 7, 9, 0 };
-static u_char de2xx_irq[] __initdata = { 5, 9, 10, 11, 15, 0 };
-static u_char de422_irq[] __initdata = { 5, 9, 10, 11, 0 };
+static u_char de1xx_irq[] __devinitdata = { 2, 3, 4, 5, 7, 9, 0 };
+static u_char de2xx_irq[] __devinitdata = { 5, 9, 10, 11, 15, 0 };
+static u_char de422_irq[] __devinitdata = { 5, 9, 10, 11, 0 };
 static u_char *depca_irq;
 
 static int irq;
@@ -580,7 +580,8 @@ static const struct net_device_ops depca_netdev_ops = {
 	.ndo_validate_addr	= eth_validate_addr,
 };
 
-static int __init depca_hw_init (struct net_device *dev, struct device *device)
+static int __devinit depca_hw_init (struct net_device *dev,
+	struct device *device)
 {
 	struct depca_private *lp;
 	int i, j, offset, netRAM, mem_len, status = 0;
@@ -1302,7 +1303,7 @@ static void SetMulticastFilter(struct net_device *dev)
 	}
 }
 
-static int __init depca_common_init (u_long ioaddr, struct net_device **devp)
+static int __devinit depca_common_init (u_long ioaddr, struct net_device **devp)
 {
 	int status = 0;
 
@@ -1333,7 +1334,7 @@ static int __init depca_common_init (u_long ioaddr, struct net_device **devp)
 /*
 ** Microchannel bus I/O device probe
 */
-static int __init depca_mca_probe(struct device *device)
+static int __devinit depca_mca_probe(struct device *device)
 {
 	unsigned char pos[2];
 	unsigned char where;
@@ -1497,7 +1498,7 @@ static void __init depca_platform_probe (void)
 	}
 }
 
-static enum depca_type __init depca_shmem_probe (ulong *mem_start)
+static enum depca_type __devinit depca_shmem_probe (ulong *mem_start)
 {
 	u_long mem_base[] = DEPCA_RAM_BASE_ADDRESSES;
 	enum depca_type adapter = unknown;
@@ -1558,7 +1559,7 @@ static int __devinit depca_isa_probe (struct platform_device *device)
 */
 
 #ifdef CONFIG_EISA
-static int __init depca_eisa_probe (struct device *device)
+static int __devinit depca_eisa_probe (struct device *device)
 {
 	enum depca_type adapter = unknown;
 	struct eisa_device *edev;
@@ -1629,7 +1630,7 @@ static int __devexit depca_device_remove (struct device *device)
 ** and Boot (readb) ROM. This will also give us a clue to the network RAM
 ** base address.
 */
-static int __init DepcaSignature(char *name, u_long base_addr)
+static int __devinit DepcaSignature(char *name, u_long base_addr)
 {
 	u_int i, j, k;
 	void __iomem *ptr;
@@ -1698,7 +1699,7 @@ static int __init DepcaSignature(char *name, u_long base_addr)
 ** PROM address counter is correctly positioned at the start of the
 ** ethernet address for later read out.
 */
-static int __init DevicePresent(u_long ioaddr)
+static int __devinit DevicePresent(u_long ioaddr)
 {
 	union {
 		struct {
@@ -1751,7 +1752,7 @@ static int __init DevicePresent(u_long ioaddr)
 ** reason: access the upper half of the PROM with x=0; access the lower half
 ** with x=1.
 */
-static int __init get_hw_addr(struct net_device *dev)
+static int __devinit get_hw_addr(struct net_device *dev)
 {
 	u_long ioaddr = dev->base_addr;
 	struct depca_private *lp = netdev_priv(dev);
-- 
1.7.4.4

[PATCH 05/12] NET: hp100: Fix section mismatches

[Ralf Baechle]

WARNING: drivers/net/hp100.o(.data+0x70): Section mismatch in reference from the variable hp100_eisa_driver to the function .init.text:hp100_eisa_probe()
The variable hp100_eisa_driver references
the function __init hp100_eisa_probe()
If the reference is valid then annotate the
variable with __init* or __refdata (see linux/init.h) or name the variable:
*_template, *_timer, *_sht, *_ops, *_probe, *_probe_one, *_console

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
To: "David S. Miller" <davem@davemloft.net>
Cc: Jaroslav Kysela <perex@perex.cz>
Cc: netdev@vger.kernel.org
Cc: linux-mips@linux-mips.org
---
 drivers/net/hp100.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/net/hp100.c b/drivers/net/hp100.c
index c3ecb11..9cd575f 100644
--- a/drivers/net/hp100.c
+++ b/drivers/net/hp100.c
@@ -2843,7 +2843,7 @@ static void cleanup_dev(struct net_device *d)
 }
 
 #ifdef CONFIG_EISA
-static int __init hp100_eisa_probe (struct device *gendev)
+static int __devinit hp100_eisa_probe(struct device *gendev)
 {
 	struct net_device *dev = alloc_etherdev(sizeof(struct hp100_private));
 	struct eisa_device *edev = to_eisa_device(gendev);
-- 
1.7.4.4

[PATCH 06/12] NET: ne3210: Fix bucketload full of section mismatches.

[Ralf Baechle]

WARNING: drivers/net/ne3210.o(.data+0x40): Section mismatch in reference from the variable ne3210_eisa_driver to the function .init.text:ne3210_eisa_probe()
The variable ne3210_eisa_driver references
the function __init ne3210_eisa_probe()
If the reference is valid then annotate the
variable with __init* or __refdata (see linux/init.h) or name the variable:
*_template, *_timer, *_sht, *_ops, *_probe, *_probe_one, *_console

Fixing this mismatch triggers yet more mismatches, fix those as well.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
To: "David S. Miller" <davem@davemloft.net>
Cc: netdev@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: linux-mips@linux-mips.org
---
 drivers/net/ne3210.c |   12 +++++++-----
 1 files changed, 7 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ne3210.c b/drivers/net/ne3210.c
index 243ed2a..e30b8ff 100644
--- a/drivers/net/ne3210.c
+++ b/drivers/net/ne3210.c
@@ -80,17 +80,19 @@ static void ne3210_block_output(struct net_device *dev, int count, const unsigne
 
 #define NE3210_DEBUG	0x0
 
-static unsigned char irq_map[] __initdata = {15, 12, 11, 10, 9, 7, 5, 3};
-static unsigned int shmem_map[] __initdata = {0xff0, 0xfe0, 0xfff0, 0xd8, 0xffe0, 0xffc0, 0xd0, 0x0};
-static const char *ifmap[] __initdata = {"UTP", "?", "BNC", "AUI"};
-static int ifmap_val[] __initdata = {
+static unsigned char irq_map[] __devinitdata = {15, 12, 11, 10, 9, 7, 5, 3};
+static unsigned int shmem_map[] __devinitdata = {
+	0xff0, 0xfe0, 0xfff0, 0xd8, 0xffe0, 0xffc0, 0xd0, 0x0
+};
+static const char *ifmap[] __devinitdata = {"UTP", "?", "BNC", "AUI"};
+static int ifmap_val[] __devinitdata = {
 		IF_PORT_10BASET,
 		IF_PORT_UNKNOWN,
 		IF_PORT_10BASE2,
 		IF_PORT_AUI,
 };
 
-static int __init ne3210_eisa_probe (struct device *device)
+static int __devinit ne3210_eisa_probe (struct device *device)
 {
 	unsigned long ioaddr, phys_mem;
 	int i, retval, port_index;
-- 
1.7.4.4

[PATCH 07/12] NET: de4x5: Fix section mismatch

[Ralf Baechle]

WARNING: drivers/net/tulip/de4x5.o(.data+0x34): Section mismatch in reference from the variable de4x5_eisa_driver to the function .init.text:de4x5_eisa_probe()
The variable de4x5_eisa_driver references
the function __init de4x5_eisa_probe()
If the reference is valid then annotate the
variable with __init* or __refdata (see linux/init.h) or name the variable:
*_template, *_timer, *_sht, *_ops, *_probe, *_probe_one, *_console

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
To: "David S. Miller" <davem@davemloft.net>
Cc: Grant Grundler <grundler@parisc-linux.org>
Cc: netdev@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: linux-mips@linux-mips.org
---
 drivers/net/tulip/de4x5.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/net/tulip/de4x5.c b/drivers/net/tulip/de4x5.c
index efaa1d6..ea473b6 100644
--- a/drivers/net/tulip/de4x5.c
+++ b/drivers/net/tulip/de4x5.c
@@ -1995,7 +1995,7 @@ SetMulticastFilter(struct net_device *dev)
 
 static u_char de4x5_irq[] = EISA_ALLOWED_IRQ_LIST;
 
-static int __init de4x5_eisa_probe (struct device *gendev)
+static int __devinit de4x5_eisa_probe (struct device *gendev)
 {
 	struct eisa_device *edev;
 	u_long iobase;
-- 
1.7.4.4

Re: [PATCH 00/12] Fix various section mismatches and build errors.

[David Miller]

See commit:

commit 948252cb9e01d65a89ecadf67be5018351eee15e
Author: David S. Miller <davem@davemloft.net>
Date:   Tue May 31 19:27:48 2011 -0700

    Revert "net: fix section mismatches"
    
    This reverts commit e5cb966c0838e4da43a3b0751bdcac7fe719f7b4.
    
    It causes new build regressions with gcc-4.2 which is
    pretty common on non-x86 platforms.
    
    Reported-by: James Bottomley <James.Bottomley@HansenPartnership.com>
    Signed-off-by: David S. Miller <davem@davemloft.net>

and postings that led to this revert including:

http://marc.info/?l=linux-netdev&m=130653748205263&w=2

Re: [PATCH 00/12] Fix various section mismatches and build errors.

[Ralf Baechle]

On Mon, Jun 27, 2011 at 10:12:57PM -0700, David Miller wrote:

> commit 948252cb9e01d65a89ecadf67be5018351eee15e
> Author: David S. Miller <davem@davemloft.net>
> Date:   Tue May 31 19:27:48 2011 -0700
> 
>     Revert "net: fix section mismatches"
>     
>     This reverts commit e5cb966c0838e4da43a3b0751bdcac7fe719f7b4.
>     
>     It causes new build regressions with gcc-4.2 which is
>     pretty common on non-x86 platforms.
>     
>     Reported-by: James Bottomley <James.Bottomley@HansenPartnership.com>
>     Signed-off-by: David S. Miller <davem@davemloft.net>
> 
> and postings that led to this revert including:
> 
> http://marc.info/?l=linux-netdev&m=130653748205263&w=2

Thanks for the pointers; I looked into it a bit deeper and found that the
construct which hppa64-linux-gcc 4.2.4 doesn't like is the combination of
const and __devinitconst __devinitdata.

My patches are minimalistic and don't do any constification and seem to
work fine for PA-RISC.

A possible alternative to allow the use of Michał's reverted patch would
be to conditionalize the definition of __devinitconst.  There is no
user of __devexitconst so I left that unchanged.

  Ralf

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

 include/linux/init.h |    8 ++++++++
 1 files changed, 8 insertions(+), 0 deletions(-)

diff --git a/include/linux/init.h b/include/linux/init.h
index 577671c..e12fd85 100644
--- a/include/linux/init.h
+++ b/include/linux/init.h
@@ -84,7 +84,15 @@
 /* Used for HOTPLUG */
 #define __devinit        __section(.devinit.text) __cold
 #define __devinitdata    __section(.devinit.data)
+#if defined __GNUC__ && (__GNUC__ == 4) && (__GNUC_MINOR__ == 2)
+/*
+ * GCC 4.2 will sometimes throw an error if the combination of const and
+ * __devinitconst is being used.  As a workaround make __devinitconst a noop
+ */
+#define __devinitconst
+#else
 #define __devinitconst   __section(.devinit.rodata)
+#endif
 #define __devexit        __section(.devexit.text) __exitused __cold
 #define __devexitdata    __section(.devexit.data)
 #define __devexitconst   __section(.devexit.rodata)

Re: [PATCH 00/12] Fix various section mismatches and build errors.

[James Bottomley]

On Wed, 2011-06-29 at 14:07 +0100, Ralf Baechle wrote:
> On Mon, Jun 27, 2011 at 10:12:57PM -0700, David Miller wrote:
> 
> > commit 948252cb9e01d65a89ecadf67be5018351eee15e
> > Author: David S. Miller <davem@davemloft.net>
> > Date:   Tue May 31 19:27:48 2011 -0700
> > 
> >     Revert "net: fix section mismatches"
> >     
> >     This reverts commit e5cb966c0838e4da43a3b0751bdcac7fe719f7b4.
> >     
> >     It causes new build regressions with gcc-4.2 which is
> >     pretty common on non-x86 platforms.
> >     
> >     Reported-by: James Bottomley <James.Bottomley@HansenPartnership.com>
> >     Signed-off-by: David S. Miller <davem@davemloft.net>
> > 
> > and postings that led to this revert including:
> > 
> > http://marc.info/?l=linux-netdev&m=130653748205263&w=2
> 
> Thanks for the pointers; I looked into it a bit deeper and found that the
> construct which hppa64-linux-gcc 4.2.4 doesn't like is the combination of
> const and __devinitconst __devinitdata.
> 
> My patches are minimalistic and don't do any constification and seem to
> work fine for PA-RISC.
> 
> A possible alternative to allow the use of Michał's reverted patch would
> be to conditionalize the definition of __devinitconst.  There is no
> user of __devexitconst so I left that unchanged.

To be honest, my own take on this is that, apart from the compiler
cockups trying to do read only annotations, which affect various
versions of gcc not just the parisc ones, the _devX annotations are
pretty pointless.  They only really do something in the non-hotplug
case, so since 95% of the world seems to use hotplug now and the other
5% doesn't care that much about the odd page of memory (which you rarely
get, since modules sections are accumulated per module, not aggregate),
I'd just favour stripping __init and __exit where there's a problem.

I think we should simply concentrate on __init and __exit; that's where
most of the discard value lies and stop expending huge efforts on the
__devX stuff which adds huge complexity for no real gain.

James

Re: [PATCH 00/12] Fix various section mismatches and build errors.

[Greg KH]

On Wed, Jun 29, 2011 at 08:58:19AM -0500, James Bottomley wrote:
> I think we should simply concentrate on __init and __exit; that's where
> most of the discard value lies and stop expending huge efforts on the
> __devX stuff which adds huge complexity for no real gain.

I have long felt that those __devX markings should just go away as they
cause nothing but problems and have no real gain as you point out.

greg k-h

Re: [PATCH 00/12] Fix various section mismatches and build errors.

[Ralf Baechle]

On Wed, Jun 29, 2011 at 08:14:24AM -0700, Greg KH wrote:

> 
> On Wed, Jun 29, 2011 at 08:58:19AM -0500, James Bottomley wrote:
> > I think we should simply concentrate on __init and __exit; that's where
> > most of the discard value lies and stop expending huge efforts on the
> > __devX stuff which adds huge complexity for no real gain.
> 
> I have long felt that those __devX markings should just go away as they
> cause nothing but problems and have no real gain as you point out.

The suggestion to do that has been floated around before but seems to
have missed sufficient thrust.  I'm all for it; the manual tagging with
__devX has not been very efficient on developer time.  I just want to see
meaningful warnings again over all that noise the current mechanisn may
produce.

  Ralf

Re: [PATCH 00/12] Fix various section mismatches and build errors.

[James Bottomley]

On Wed, 2011-06-29 at 17:19 +0100, Ralf Baechle wrote:
> On Wed, Jun 29, 2011 at 08:14:24AM -0700, Greg KH wrote:
> 
> > 
> > On Wed, Jun 29, 2011 at 08:58:19AM -0500, James Bottomley wrote:
> > > I think we should simply concentrate on __init and __exit; that's where
> > > most of the discard value lies and stop expending huge efforts on the
> > > __devX stuff which adds huge complexity for no real gain.
> > 
> > I have long felt that those __devX markings should just go away as they
> > cause nothing but problems and have no real gain as you point out.
> 
> The suggestion to do that has been floated around before but seems to
> have missed sufficient thrust.  I'm all for it; the manual tagging with
> __devX has not been very efficient on developer time.  I just want to see
> meaningful warnings again over all that noise the current mechanisn may
> produce.

For me, just go ahead and fix the actual problems: so _init sections and
_exit sections that are used from the main body, just strip the
annotations, don't try to change them for _devX ones.

Thanks,

James