Thread (1 message) 1 message, 1 author, 2012-07-13

Re: [PATCH 00/11] make struct of_device_id.data const

From: Uwe Kleine-König <hidden>
Date: 2012-07-13 17:49:35
Also in: linux-arm-kernel

Hello,

On Fri, Jul 13, 2012 at 04:24:26PM +0000, Arnd Bergmann wrote:
quoted hunk
On Friday 13 July 2012, Uwe Kleine-König wrote:
quoted
On Fri, Jul 13, 2012 at 07:41:02AM -0500, Rob Herring wrote:
quoted
On 07/13/2012 07:32 AM, y@pengutronix.de wrote:
quoted
From: Uwe Kleine-König <redacted>

this series' goal is to make struct of_device_id.data const.
Conceptually a driver must not modify the data contained there so making
it const is the right thing.

This change might introduce warnings because drivers don't use const to
access the data pointed to by of_device_id.data. In most cases the fix
is to add some consts.

This series (apart from adding this const in patch 10) fixes all drivers
covered by all arm defconfigs (based on v3.5-rc6).
What about other arches?
I didn't looked at them, mainly because I don't have the scripts for
arches other than arm to do such testing. Also note that it introduces
only some warnings, but no breakage. So adding the const in the merge
window might be ok?!
I think it makes sense to change the drivers that are easy to find now:

git grep -A6 of_device_id | grep '\<data\>.*='  | brain

arch/c6x/platforms/plldata.c-   { .compatible = "ti,c6455-pll", .data = c6455_setup_clocks },
arch/mips/lantiq/irq.c- { .compatible = "lantiq,icu", .data = icu_of_init },
arch/powerpc/platforms/83xx/suspend.c-          .data = &pmc_types[0],
arch/powerpc/platforms/cell/celleb_pci.c-               .data = &celleb_fake_pci_spec,
arch/powerpc/platforms/cell/celleb_pci.c-               .data = &celleb_epci_spec,
arch/powerpc/sysdev/fsl_msi.c-          .data = (void *)&mpic_msi_feature,
drivers/atm/fore200e.c-         .data = (void *) &fore200e_bus[1],
drivers/char/xilinx_hwicap/xilinx_hwicap.c-     { .compatible = "xlnx,opb-hwicap-1.00.b", .data = &buffer_icap_config},
drivers/i2c/busses/i2c-mpc.c-   {.compatible = "mpc5200-i2c", .data = &mpc_i2c_data_52xx, },
drivers/macintosh/mediabay.c-   .data           = &keylargo_mb_ops,
drivers/mfd/da9052-i2c.c-       { .compatible = "dlg,da9053-aa", .data = &da9052_i2c_id[1] },
drivers/net/can/mscan/mpc5xxx_can.c-    { .compatible = "fsl,mpc5200-mscan", .data = &mpc5200_can_data, },
drivers/scsi/qlogicpti.c-               .data = &qpti_template,
drivers/usb/host/fsl-mph-dr-of.c-       { .compatible = "fsl-usb2-mph", .data = &fsl_usb2_mpc8xxx_pd, },
drivers/watchdog/mpc8xxx_wdt.c-         .data = &(struct mpc8xxx_wdt_type) {

I'm optimistic that this list covers the vast majority of the non-ARM
drivers, and some of them don't even need any changes. I've done
patches for the others below.

Signed-off-by: Arnd Bergmann <redacted>
---
 arch/powerpc/platforms/83xx/suspend.c    |    2 +-
 arch/powerpc/platforms/cell/celleb_pci.c |    2 +-
 arch/powerpc/sysdev/fsl_msi.c            |    8 ++++----
 drivers/i2c/busses/i2c-mpc.c             |   12 ++++++------
 drivers/macintosh/mediabay.c             |    8 ++++----
 drivers/mfd/da9052-i2c.c                 |    4 ++--
 drivers/net/can/mscan/mpc5xxx_can.c      |    6 +++---
 drivers/scsi/qlogicpti.c                 |    2 +-
 drivers/watchdog/mpc8xxx_wdt.c           |    2 +-
 10 files changed, 23 insertions(+), 26 deletions(-)
diff --git a/arch/powerpc/platforms/83xx/suspend.c b/arch/powerpc/platforms/83xx/suspend.c
index 1a04671..1d769a2 100644
--- a/arch/powerpc/platforms/83xx/suspend.c
+++ b/arch/powerpc/platforms/83xx/suspend.c
@@ -326,7 +326,7 @@ static int pmc_probe(struct platform_device *ofdev)
 	const struct of_device_id *match;
 	struct device_node *np = ofdev->dev.of_node;
 	struct resource res;
-	struct pmc_type *type;
+	const struct pmc_type *type;
 	int ret = 0;
 
 	match = of_match_device(pmc_match, &ofdev->dev);
diff --git a/arch/powerpc/platforms/cell/celleb_pci.c b/arch/powerpc/platforms/cell/celleb_pci.c
index 5822141..abc8af4 100644
--- a/arch/powerpc/platforms/cell/celleb_pci.c
+++ b/arch/powerpc/platforms/cell/celleb_pci.c
@@ -472,7 +472,7 @@ int __init celleb_setup_phb(struct pci_controller *phb)
 {
 	struct device_node *dev = phb->dn;
 	const struct of_device_id *match;
-	struct celleb_phb_spec *phb_spec;
+	const struct celleb_phb_spec *phb_spec;
 	int rc;
 
 	match = of_match_node(celleb_phb_match, dev);
diff --git a/arch/powerpc/sysdev/fsl_msi.c b/arch/powerpc/sysdev/fsl_msi.c
index 6e097de..51ffafa 100644
--- a/arch/powerpc/sysdev/fsl_msi.c
+++ b/arch/powerpc/sysdev/fsl_msi.c
@@ -368,7 +368,7 @@ static int __devinit fsl_of_msi_probe(struct platform_device *dev)
 	int err, i, j, irq_index, count;
 	int rc;
 	const u32 *p;
-	struct fsl_msi_feature *features;
+	const struct fsl_msi_feature *features;
 	int len;
 	u32 offset;
 	static const u32 all_avail[] = { 0, NR_MSI_IRQS };
@@ -502,15 +502,15 @@ static const struct fsl_msi_feature vmpic_msi_feature = {
 static const struct of_device_id fsl_of_msi_ids[] = {
 	{
 		.compatible = "fsl,mpic-msi",
-		.data = (void *)&mpic_msi_feature,
+		.data = &mpic_msi_feature,
This looks unrelated.
quoted hunk
 	},
 	{
 		.compatible = "fsl,ipic-msi",
-		.data = (void *)&ipic_msi_feature,
+		.data = &ipic_msi_feature,
 	},
 	{
 		.compatible = "fsl,vmpic-msi",
-		.data = (void *)&vmpic_msi_feature,
+		.data = &vmpic_msi_feature,
 	},
 	{}
 };
diff --git a/drivers/i2c/busses/i2c-mpc.c b/drivers/i2c/busses/i2c-mpc.c
index b76731e..57f7703 100644
--- a/drivers/i2c/busses/i2c-mpc.c
+++ b/drivers/i2c/busses/i2c-mpc.c
@@ -647,7 +647,7 @@ static int __devinit fsl_i2c_probe(struct platform_device *op)
 	}
 
 	if (match->data) {
-		struct mpc_i2c_data *data = match->data;
+		const struct mpc_i2c_data *data = match->data;
 		data->setup(op->dev.of_node, i2c, clock, data->prescaler);
 	} else {
 		/* Backwards compatibility */
@@ -730,24 +730,24 @@ static int mpc_i2c_resume(struct device *dev)
 SIMPLE_DEV_PM_OPS(mpc_i2c_pm_ops, mpc_i2c_suspend, mpc_i2c_resume);
 #endif
 
-static struct mpc_i2c_data mpc_i2c_data_512x __devinitdata = {
+static const struct mpc_i2c_data mpc_i2c_data_512x __devinitdata = {
I guess this results in a warning until after patch 10 is applied? (A
few more down below.)

In the meantime I expanded my defconfig build script to handle powerpc,
too. I'll check these. I think I'll have to let the build run over
night...

Thanks Arnd for your effort.

Best regards
Uwe

-- 
Pengutronix e.K.                           | Uwe Kleine-König            |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help