Re: [PATCH v3] i2c: busses: i2c-pxa.c: Fix for possible null pointer dereferenc

From: Rickard Strandqvist
Date: Sat Jul 05 2014 - 08:34:42 EST


2014-07-04 19:24 GMT+02:00 Emil Goode <emilgoode@xxxxxxxxx>:
> Hello,
>
> I noticed one more thing.
>
> On Fri, Jul 04, 2014 at 07:07:48PM +0200, Rickard Strandqvist wrote:
>> 2014-07-04 11:10 GMT+02:00 Emil Goode <emilgoode@xxxxxxxxx>:
>> > Hello Rickard,
>> >
>> > Since this is a probe function there is also no need to release the devm_*
>> > resources in the i2c_pxa_remove function, this leads to double free.
>> >
>> > Also I have a few nit-pick comments below.
>> >
>> > On Thu, Jul 03, 2014 at 10:19:16PM +0200, Rickard Strandqvist wrote:
>> >> Fix for possible null pointer dereferenc, and there is a risk for memory leak if something unexpected happens and the function returns.
>> >> It now use Managed Device Resource instead.
>> >>
>> >> Signed-off-by: Rickard Strandqvist <rickard_strandqvist@xxxxxxxxxxxxxxxxxx>
>> >> ---
>> >> drivers/i2c/busses/i2c-pxa.c | 37 ++++++++++++++++---------------------
>> >> 1 file changed, 16 insertions(+), 21 deletions(-)
>> >>
>> >> diff --git a/drivers/i2c/busses/i2c-pxa.c b/drivers/i2c/busses/i2c-pxa.c
>> >> index be671f7..2edb633 100644
>> >> --- a/drivers/i2c/busses/i2c-pxa.c
>> >> +++ b/drivers/i2c/busses/i2c-pxa.c
>> >> @@ -1141,10 +1141,10 @@ static int i2c_pxa_probe(struct platform_device *dev)
>> >> struct resource *res = NULL;
>> >> int ret, irq;
>> >>
>> >> - i2c = kzalloc(sizeof(struct pxa_i2c), GFP_KERNEL);
>> >> + i2c = devm_kzalloc(&dev->dev, sizeof(struct pxa_i2c), GFP_KERNEL);
>> >> if (!i2c) {
>> >> ret = -ENOMEM;
>> >> - goto emalloc;
>> >> + goto err_nothing_to_release;
>> >
>> > Perhaps its a good idea to return directly here, then the label
>> > err_nothing_to_release can be removed.
>> >
>> >> }
>> >>
>> >> /* Default adapter num to device id; i2c_pxa_probe_dt can override. */
>> >> @@ -1154,18 +1154,19 @@ static int i2c_pxa_probe(struct platform_device *dev)
>> >> if (ret > 0)
>> >> ret = i2c_pxa_probe_pdata(dev, i2c, &i2c_type);
>> >> if (ret < 0)
>> >> - goto eclk;
>> >> + goto err_nothing_to_release;
>> >
>> > Same here.
>> >
>> >>
>> >> res = platform_get_resource(dev, IORESOURCE_MEM, 0);
>> >> irq = platform_get_irq(dev, 0);
>> >> if (res == NULL || irq < 0) {
>> >> ret = -ENODEV;
>> >> - goto eclk;
>> >> + goto err_nothing_to_release;
>> >
>> > Same here.
>> >
>> >> }
>> >>
>> >> - if (!request_mem_region(res->start, resource_size(res), res->name)) {
>> >> + if (!devm_request_mem_region(&dev->dev, res->start,
>> >> + resource_size(res), res->name)) {
>> >> ret = -ENOMEM;
>> >> - goto eclk;
>> >> + goto emalloc;
>> >
>> > We could also return directly here since the release_mem_region call should
>> > be removed, as mentioned in Jingoos reply.
>> >
>> >> }
>> >>
>> >> i2c->adap.owner = THIS_MODULE;
>> >> @@ -1176,16 +1177,16 @@ static int i2c_pxa_probe(struct platform_device *dev)
>> >>
>> >> strlcpy(i2c->adap.name, "pxa_i2c-i2c", sizeof(i2c->adap.name));
>> >>
>> >> - i2c->clk = clk_get(&dev->dev, NULL);
>> >> + i2c->clk = devm_clk_get(&dev->dev, NULL);
>> >> if (IS_ERR(i2c->clk)) {
>> >> ret = PTR_ERR(i2c->clk);
>> >> - goto eclk;
>> >> + goto emalloc;
>> >
>> > Same here.
>> >
>> >> }
>> >>
>> >> - i2c->reg_base = ioremap(res->start, resource_size(res));
>> >> - if (!i2c->reg_base) {
>> >> + i2c->reg_base = devm_ioremap_resource(&adev->dev, res));
>> >> + if (IS_ERR(i2c->reg_base)) {
>> >> ret = -EIO;
>> >> - goto eremap;
>> >> + goto emalloc;
>> >
>> > Same here.
>> >
>> >> }
>> >>
>> >> i2c->reg_ibmr = i2c->reg_base + pxa_reg_layout[i2c_type].ibmr;
>> >> @@ -1227,10 +1228,10 @@ static int i2c_pxa_probe(struct platform_device *dev)
>> >> i2c->adap.algo = &i2c_pxa_pio_algorithm;
>> >> } else {
>> >> i2c->adap.algo = &i2c_pxa_algorithm;
>> >> - ret = request_irq(irq, i2c_pxa_handler, IRQF_SHARED,
>> >> - dev_name(&dev->dev), i2c);
>> >> + ret = devm_request_irq(&dev->dev, irq, i2c_pxa_handler,
>> >> + IRQF_SHARED, dev_name(&dev->dev), i2c);
>> >> if (ret)
>> >> - goto ereqirq;
>> >> + goto emalloc;
>> >
>> > Same here.
>> >
>> >> }
>> >>
>> >> i2c_pxa_reset(i2c);
>> >> @@ -1261,15 +1262,9 @@ static int i2c_pxa_probe(struct platform_device *dev)
>> >> eadapt:
>> >> if (!i2c->use_pio)
>> >> free_irq(irq, i2c);
>> >
>> > The free_irq() call should also be removed since devm_request_irq() was used.
>> >
>> >> -ereqirq:
>> >> - clk_disable_unprepare(i2c->clk);
>
> I think we should keep the clk_disable_unprepare() call.
>
>> >> - iounmap(i2c->reg_base);
>> >> -eremap:
>> >> - clk_put(i2c->clk);
>> >> -eclk:
>> >> - kfree(i2c);
>> >> emalloc:
>> >> release_mem_region(res->start, resource_size(res));
>> >> +err_nothing_to_release:
>> >> return ret;
>> >> }
>> >>
>> >
>> > Best regards,
>> >
>> > Emil Goode
>>
>> Hi All!
>>
>> Ok, so there is literally nothing to do release anymore.
>> A good system Devres, I understand you want everyone to use it.
>>
>> I also think it is also better to return directly then.
>> But who decides, Wolfram?
>
> Yes, Wolfram is the maintainer.
>
> Personally I would do the devm_* convertion in a second patch.

Hi

Ok then, Wolfram!

Keep only:

ereqirq:
clk_disable_unprepare(i2c->clk);

And let everyone else just return?


And Emil, as I said before, it would in that case the first patch just be:
emalloc:
- release_mem_region(res->start, resource_size(res));
+ if(res)
+ release_mem_region(res->start, resource_size(res));

It is rather pointless to add something in a patch, to remove it in the next.



Kind regards
Rickard Strandqvist
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/