i2c: versatile: Allow compile test build
PATCH 1/2 i2c: versatile: Allow compile test build
There is no build dependency for this driver, so enable COMPILE_TEST to get better build coverage.
--- drivers/i2c/busses/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/i2c/busses/Kconfig +1 -1
@@ -924,7 +924,7 @@ config I2C_UNIPHIER_F config I2C_VERSATILE tristate "ARM Versatile/Realview I2C bus support" - depends on ARCH_VERSATILE || ARCH_REALVIEW || ARCH_VEXPRESS + depends on ARCH_VERSATILE || ARCH_REALVIEW || ARCH_VEXPRESS || COMPILE_TEST select I2C_ALGOBIT help Say yes if you want to support the I2C serial bus on ARMs Versatile
Signature · 1 line
2.5.0
PATCH 2/2 i2c: versatile: Convert to use resource managed devm_* APIs
Use devm_* APIs to simplify the code a bit.
This patch also fixes the memory leak when unload the module.
---
Hi,
I don't have this h/w, I'd appreciate if someone can test this patch.
drivers/i2c/busses/i2c-versatile.c | 46 +++++++++++--------------------------- 1 file changed, 13 insertions(+), 33 deletions(-)
drivers/i2c/busses/i2c-versatile.c +13 -33
@@ -70,28 +70,14 @@ static int i2c_versatile_probe(struct platform_device *dev)struct resource *r;int ret; + i2c =devm_kzalloc (&dev->dev,sizeof (struct i2c_versatile ),GFP_KERNEL ); +if (!i2c) +return -ENOMEM ; + r =platform_get_resource (dev,IORESOURCE_MEM ,0 ); -if (!r) { - ret = -EINVAL ; - goto err_out; - } - -if (!request_mem_region (r->start,resource_size (r),"versatile-i2c" )) { - ret = -EBUSY ; - goto err_out; - } - - i2c =kzalloc (sizeof (struct i2c_versatile ),GFP_KERNEL ); -if (!i2c) { - ret = -ENOMEM ; - goto err_release; - } - - i2c->base =ioremap (r->start,resource_size (r)); -if (!i2c->base) { - ret = -ENOMEM ; - goto err_free; - } + i2c->base =devm_ioremap_resource (&dev->dev, r); +if (IS_ERR (i2c->base)) +return PTR_ERR (i2c->base);writel (SCL |SDA , i2c->base +I2C_CONTROLS );
@@ -105,18 +91,12 @@ static int i2c_versatile_probe(struct platform_device *dev) i2c->adap.nr = dev->id; ret =i2c_bit_add_numbered_bus (&i2c->adap); -if (ret >=0 ) { -platform_set_drvdata (dev, i2c); -return 0 ; - } - -iounmap (i2c->base); - err_free: -kfree (i2c); - err_release: -release_mem_region (r->start,resource_size (r)); - err_out: -return ret; +if (ret <0 ) +return ret; + +platform_set_drvdata (dev, i2c); + +return 0 ; }static int i2c_versatile_remove (struct platform_device *dev)
Signature · 1 line
2.5.0
Re: [PATCH 2/2] i2c: versatile: Convert to use resource managed devm_* APIs
2016-04-14 22:31 GMT+08:00 Axel Lin <[email protected]>:
Quoted 7 lines
Use devm_* APIs to simplify the code a bit.
This patch also fixes the memory leak when unload the module.Signed-off-by Axel Lin <[email protected]>---
Hi,
I don't have this h/w, I'd appreciate if someone can test this patch.
Obviously the module unload path has memory leak. Just wondering if anyone can review/test this patch.
PATCH 2/2 i2c: versatile: Convert to use resource managed devm_* APIs
Dnia 2016-06-25, Sat o godzinie 09:41 +0800, Axel Lin pisze:
Quoted 4 lines
2016-04-14 22:31 GMT+08:00 Axel Lin <[email protected]>:
Quoted 3 lines
Use devm_* APIs to simplify the code a bit.
This patch also fixes the memory leak when unload the module.
Quoted 8 lines
Quoted 5 lines
Signed-off-by Axel Lin <[email protected]>---
Hi,
I don't have this h/w, I'd appreciate if someone can test this patch.Obviously the module unload path has memory leak. Just wondering if anyone can review/test this patch.
Gentlemen? (referring to the VE maintainers here)
Original thread: thread.gmane.org/gmane.linux.drivers.i2c/27277
Thanks!
Pawel
PATCH 2/2 i2c: versatile: Convert to use resource managed devm_* APIs
Quoted 11 lines · Pawel Moll
Dnia 2016-06-25, Sat o godzinie 09:41 +0800, Axel Lin pisze:
Quoted 10 lines
2016-04-14 22:31 GMT+08:00 Axel Lin <[email protected]>:
Quoted 8 lines
Use devm_* APIs to simplify the code a bit.
This patch also fixes the memory leak when unload the module.Signed-off-by Axel Lin <[email protected]>---
Hi,
I don't have this h/w, I'd appreciate if someone can test this patch.
Quoted 6 lines
Quoted 2 lines
Obviously the module unload path has memory leak. Just wondering if anyone can review/test this patch.
Gentlemen? (referring to the VE maintainers here)
Original thread: thread.gmane.org/gmane.linux.drivers.i2c/27277
We weren't CC'ed, thanks for the heads-up anyway we will have a look and test shortly (@Axel next time please CC us if you want us to test your patches, thanks).
Lorenzo