This series allows regmaps without a device (e.g. syscon) to be registered
in debugfs.
David Lechner (2):
regmap: fix NULL pointer dereference in regmap_name_read_file()
regmap: use debugfs even when no device
drivers/base/regmap/regmap-debugfs.c | 6 +++++-
drivers/base/regmap/regmap.c | 2 ++
2 files changed, 7 insertions(+), 1 deletion(-)
--
2.7.4
This registers regmaps with debugfs even when they do not have an
associated device. For example, this is common for syscon regmaps.
Signed-off-by: David Lechner <[email protected]>
---
drivers/base/regmap/regmap.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c
index 8d516a9..ebad970 100644
--- a/drivers/base/regmap/regmap.c
+++ b/drivers/base/regmap/regmap.c
@@ -1106,6 +1106,8 @@ struct regmap *__regmap_init(struct device *dev,
ret = regmap_attach_dev(dev, map, config);
if (ret != 0)
goto err_regcache;
+ } else {
+ regmap_debugfs_init(map, config->name);
}
return map;
--
2.7.4
This fixes a possible NULL pointer dereference oops in
regmap_name_read_file() when the regmap does not have a device
associated with it.
Signed-off-by: David Lechner <[email protected]>
---
drivers/base/regmap/regmap-debugfs.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/drivers/base/regmap/regmap-debugfs.c b/drivers/base/regmap/regmap-debugfs.c
index 36ce351..0df7379 100644
--- a/drivers/base/regmap/regmap-debugfs.c
+++ b/drivers/base/regmap/regmap-debugfs.c
@@ -40,6 +40,7 @@ static ssize_t regmap_name_read_file(struct file *file,
loff_t *ppos)
{
struct regmap *map = file->private_data;
+ const char *name = NULL;
int ret;
char *buf;
@@ -47,7 +48,10 @@ static ssize_t regmap_name_read_file(struct file *file,
if (!buf)
return -ENOMEM;
- ret = snprintf(buf, PAGE_SIZE, "%s\n", map->dev->driver->name);
+ if (map->dev && map->dev->driver)
+ name = map->dev->driver->name;
+
+ ret = snprintf(buf, PAGE_SIZE, "%s\n", name);
if (ret < 0) {
kfree(buf);
return ret;
--
2.7.4
On Tue, Jan 23, 2018 at 04:53:42PM -0600, David Lechner wrote:
> This fixes a possible NULL pointer dereference oops in
> regmap_name_read_file() when the regmap does not have a device
> associated with it.
How are you managing to get a regmap without a device associated with
it?
On 01/24/2018 05:30 AM, Mark Brown wrote:
> On Tue, Jan 23, 2018 at 04:53:42PM -0600, David Lechner wrote:
>
>> This fixes a possible NULL pointer dereference oops in
>> regmap_name_read_file() when the regmap does not have a device
>> associated with it.
>
> How are you managing to get a regmap without a device associated with
> it?
>
I'm using syscon_regmap_lookup_by_compatible(), which calls:
regmap_init_mmio(NULL, base, &syscon_config)
NULL being the struct device *dev parameter.
There are several clock drivers that do that as well. I'm writing
another clock driver and will need to do the same thing for some non-DT
boards. I tried using a device, but clocks are initialized fairly early
in the boot process and kobject is not initialized yet, so you can't
register a platform device at that point.
On 01/23/2018 11:53 PM, David Lechner wrote:
> This fixes a possible NULL pointer dereference oops in
> regmap_name_read_file() when the regmap does not have a device
> associated with it.
>
> Signed-off-by: David Lechner <[email protected]>
> ---
> drivers/base/regmap/regmap-debugfs.c | 6 +++++-
> 1 file changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/base/regmap/regmap-debugfs.c b/drivers/base/regmap/regmap-debugfs.c
> index 36ce351..0df7379 100644
> --- a/drivers/base/regmap/regmap-debugfs.c
> +++ b/drivers/base/regmap/regmap-debugfs.c
> @@ -40,6 +40,7 @@ static ssize_t regmap_name_read_file(struct file *file,
> loff_t *ppos)
> {
> struct regmap *map = file->private_data;
> + const char *name = NULL;
> int ret;
> char *buf;
>
> @@ -47,7 +48,10 @@ static ssize_t regmap_name_read_file(struct file *file,
> if (!buf)
> return -ENOMEM;
>
> - ret = snprintf(buf, PAGE_SIZE, "%s\n", map->dev->driver->name);
> + if (map->dev && map->dev->driver)
> + name = map->dev->driver->name;
> +
> + ret = snprintf(buf, PAGE_SIZE, "%s\n", name);
Won't this print "(null)" now? Not sure if that is the best approach.
On Wed, Jan 24, 2018 at 05:14:50PM +0100, Lars-Peter Clausen wrote:
> On 01/23/2018 11:53 PM, David Lechner wrote:
> > This fixes a possible NULL pointer dereference oops in
> > regmap_name_read_file() when the regmap does not have a device
> > associated with it.
> > - ret = snprintf(buf, PAGE_SIZE, "%s\n", map->dev->driver->name);
> > + if (map->dev && map->dev->driver)
> > + name = map->dev->driver->name;
> > +
> > + ret = snprintf(buf, PAGE_SIZE, "%s\n", name);
> Won't this print "(null)" now? Not sure if that is the best approach.
Indeed. It is an improvement but not exactly helpful.
On 01/24/2018 10:14 AM, Lars-Peter Clausen wrote:
> On 01/23/2018 11:53 PM, David Lechner wrote:
>> This fixes a possible NULL pointer dereference oops in
>> regmap_name_read_file() when the regmap does not have a device
>> associated with it.
>>
>> Signed-off-by: David Lechner <[email protected]>
>> ---
>> drivers/base/regmap/regmap-debugfs.c | 6 +++++-
>> 1 file changed, 5 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/base/regmap/regmap-debugfs.c b/drivers/base/regmap/regmap-debugfs.c
>> index 36ce351..0df7379 100644
>> --- a/drivers/base/regmap/regmap-debugfs.c
>> +++ b/drivers/base/regmap/regmap-debugfs.c
>> @@ -40,6 +40,7 @@ static ssize_t regmap_name_read_file(struct file *file,
>> loff_t *ppos)
>> {
>> struct regmap *map = file->private_data;
>> + const char *name = NULL;
>> int ret;
>> char *buf;
>>
>> @@ -47,7 +48,10 @@ static ssize_t regmap_name_read_file(struct file *file,
>> if (!buf)
>> return -ENOMEM;
>>
>> - ret = snprintf(buf, PAGE_SIZE, "%s\n", map->dev->driver->name);
>> + if (map->dev && map->dev->driver)
>> + name = map->dev->driver->name;
>> +
>> + ret = snprintf(buf, PAGE_SIZE, "%s\n", name);
>
> Won't this print "(null)" now? Not sure if that is the best approach.
>
The commit adding regmap_name_read_file() just says:
regmap: Expose the driver name in debugfs
Add a file called 'name' containing the name of the driver.
So, if a regmap doesn't have a driver, then "(null)" seems OK to me.
This is just for debugging anyway.
On 01/24/2018 10:19 AM, Mark Brown wrote:
> On Wed, Jan 24, 2018 at 05:14:50PM +0100, Lars-Peter Clausen wrote:
>> On 01/23/2018 11:53 PM, David Lechner wrote:
>>> This fixes a possible NULL pointer dereference oops in
>>> regmap_name_read_file() when the regmap does not have a device
>>> associated with it.
>
>>> - ret = snprintf(buf, PAGE_SIZE, "%s\n", map->dev->driver->name);
>>> + if (map->dev && map->dev->driver)
>>> + name = map->dev->driver->name;
>>> +
>>> + ret = snprintf(buf, PAGE_SIZE, "%s\n", name);
>
>> Won't this print "(null)" now? Not sure if that is the best approach.
>
> Indeed. It is an improvement but not exactly helpful.
>
Would it be better if it said "this regmap does not have driver" instead
of "(null)"?
On Wed, Jan 24, 2018 at 10:27:22AM -0600, David Lechner wrote:
> On 01/24/2018 10:19 AM, Mark Brown wrote:
> > Indeed. It is an improvement but not exactly helpful.
> Would it be better if it said "this regmap does not have driver" instead
> of "(null)"?
That's both not idiomatically good and isn't even true... ideally we'd
be able to get some sort of string out of the user, but otherwise
"nodev" or perhaps printing the pointer to the regmap (to avoid
collisions if there's more than one of these in a system).
The patch
regmap: use debugfs even when no device
has been applied to the regmap tree at
https://git.kernel.org/pub/scm/linux/kernel/git/broonie/regmap.git
All being well this means that it will be integrated into the linux-next
tree (usually sometime in the next 24 hours) and sent to Linus during
the next merge window (or sooner if it is a bug fix), however if
problems are discovered then the patch may be dropped or reverted.
You may get further e-mails resulting from automated or manual testing
and review of the tree, please engage with people reporting problems and
send followup patches addressing any issues that are reported if needed.
If any updates are required or you are submitting further changes they
should be sent as incremental updates against current git, existing
patches will not be replaced.
Please add any relevant lists and maintainers to the CCs when replying
to this mail.
Thanks,
Mark
From 9b947a13e7f6017f18470f665992dbae267852b3 Mon Sep 17 00:00:00 2001
From: David Lechner <[email protected]>
Date: Mon, 19 Feb 2018 15:43:02 -0600
Subject: [PATCH] regmap: use debugfs even when no device
This registers regmaps with debugfs even when they do not have an
associated device. For example, this is common for syscon regmaps.
Signed-off-by: David Lechner <[email protected]>
Signed-off-by: Mark Brown <[email protected]>
---
drivers/base/regmap/regmap.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c
index ee302ccdfbc8..f5fa1ddc65e9 100644
--- a/drivers/base/regmap/regmap.c
+++ b/drivers/base/regmap/regmap.c
@@ -1116,6 +1116,8 @@ struct regmap *__regmap_init(struct device *dev,
ret = regmap_attach_dev(dev, map, config);
if (ret != 0)
goto err_regcache;
+ } else {
+ regmap_debugfs_init(map, config->name);
}
return map;
--
2.16.1