Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1666638 > unrolled thread

[PATCH] drm: Reduce scope of 'state' variable

Started byDawid Kurek <dawikur@gmail.com>
First post2017-06-15 12:50 +0200
Last post2017-06-15 19:00 +0200
Articles 5 — 2 participants

Back to article view | Back to linux.kernel

Contents

  [PATCH] drm: Reduce scope of 'state' variable Dawid Kurek <dawikur@gmail.com> - 2017-06-15 12:50 +0200
    Re: [PATCH] drm: Reduce scope of 'state' variable Jani Nikula <jani.nikula@linux.intel.com> - 2017-06-15 13:20 +0200
      Re: [PATCH] drm: Reduce scope of 'state' variable Dawid Kurek <dawikur@gmail.com> - 2017-06-15 16:30 +0200
        Re: [PATCH] drm: Reduce scope of 'state' variable Jani Nikula <jani.nikula@linux.intel.com> - 2017-06-15 17:20 +0200
          Re: [PATCH] drm: Reduce scope of 'state' variable Dawid Kurek <dawikur@gmail.com> - 2017-06-15 19:00 +0200

#1666638 — [PATCH] drm: Reduce scope of 'state' variable

FromDawid Kurek <dawikur@gmail.com>
Date2017-06-15 12:50 +0200
Subject[PATCH] drm: Reduce scope of 'state' variable
Message-ID<tSAvE-1ng-13@gated-at.bofh.it>
Smaller scope reduces visibility of variable and makes usage of
uninitialized variable less possible.
---
 drivers/gpu/drm/drm_atomic.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c
index f32506a..ea5a9a7 100644
--- a/drivers/gpu/drm/drm_atomic.c
+++ b/drivers/gpu/drm/drm_atomic.c
@@ -108,10 +108,11 @@ struct drm_atomic_state *
 drm_atomic_state_alloc(struct drm_device *dev)
 {
 	struct drm_mode_config *config = &dev->mode_config;
-	struct drm_atomic_state *state;
 
 	if (!config->funcs->atomic_state_alloc) {
-		state = kzalloc(sizeof(*state), GFP_KERNEL);
+		struct drm_atomic_state *state
+			= kzalloc(sizeof(*state), GFP_KERNEL);
+
 		if (!state)
 			return NULL;
 		if (drm_atomic_state_init(dev, state) < 0) {
-- 
2.10.0


---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

[toc] | [next] | [standalone]

#1666664

FromJani Nikula <jani.nikula@linux.intel.com>
Date2017-06-15 13:20 +0200
Message-ID<tSAYF-1Mg-9@gated-at.bofh.it>
In reply to#1666638
On Thu, 15 Jun 2017, Dawid Kurek <dawikur@gmail.com> wrote:
> Smaller scope reduces visibility of variable and makes usage of
> uninitialized variable less possible.
> ---
>  drivers/gpu/drm/drm_atomic.c | 5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c
> index f32506a..ea5a9a7 100644
> --- a/drivers/gpu/drm/drm_atomic.c
> +++ b/drivers/gpu/drm/drm_atomic.c
> @@ -108,10 +108,11 @@ struct drm_atomic_state *
>  drm_atomic_state_alloc(struct drm_device *dev)
>  {
>  	struct drm_mode_config *config = &dev->mode_config;
> -	struct drm_atomic_state *state;
>  
>  	if (!config->funcs->atomic_state_alloc) {
> -		state = kzalloc(sizeof(*state), GFP_KERNEL);
> +		struct drm_atomic_state *state
> +			= kzalloc(sizeof(*state), GFP_KERNEL);

Separate declaration and initialization would lead to a cleaner patch
and result.

BR,
Jani.

> +
>  		if (!state)
>  			return NULL;
>  		if (drm_atomic_state_init(dev, state) < 0) {

-- 
Jani Nikula, Intel Open Source Technology Center

[toc] | [prev] | [next] | [standalone]

#1666757

FromDawid Kurek <dawikur@gmail.com>
Date2017-06-15 16:30 +0200
Message-ID<tSDWy-3yH-13@gated-at.bofh.it>
In reply to#1666664
On 15/06/17, Jani Nikula wrote:
> On Thu, 15 Jun 2017, Dawid Kurek <dawikur@gmail.com> wrote:
> > Smaller scope reduces visibility of variable and makes usage of
> > uninitialized variable less possible.
> > ---
> >  drivers/gpu/drm/drm_atomic.c | 5 +++--
> >  1 file changed, 3 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c
> > index f32506a..ea5a9a7 100644
> > --- a/drivers/gpu/drm/drm_atomic.c
> > +++ b/drivers/gpu/drm/drm_atomic.c
> > @@ -108,10 +108,11 @@ struct drm_atomic_state *
> >  drm_atomic_state_alloc(struct drm_device *dev)
> >  {
> >  	struct drm_mode_config *config = &dev->mode_config;
> > -	struct drm_atomic_state *state;
> >  
> >  	if (!config->funcs->atomic_state_alloc) {
> > -		state = kzalloc(sizeof(*state), GFP_KERNEL);
> > +		struct drm_atomic_state *state
> > +			= kzalloc(sizeof(*state), GFP_KERNEL);
> 
> Separate declaration and initialization would lead to a cleaner patch
> and result.

I saw combining declaration and initialization is quite common, i.e. in
drm_atomic file. Personally, I also prefer those in one statement. But yes, it
looks cleaner here, in two lines.

v2 sent :)

Thanks,
Dawid

> 
> BR,
> Jani.
> 
> > +
> >  		if (!state)
> >  			return NULL;
> >  		if (drm_atomic_state_init(dev, state) < 0) {
> 
> -- 
> Jani Nikula, Intel Open Source Technology Center

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

[toc] | [prev] | [next] | [standalone]

#1666796

FromJani Nikula <jani.nikula@linux.intel.com>
Date2017-06-15 17:20 +0200
Message-ID<tSEIW-46h-19@gated-at.bofh.it>
In reply to#1666757
On Thu, 15 Jun 2017, Dawid Kurek <dawikur@gmail.com> wrote:
> On 15/06/17, Jani Nikula wrote:
>> Separate declaration and initialization would lead to a cleaner patch
>> and result.
>
> I saw combining declaration and initialization is quite common, i.e. in
> drm_atomic file. Personally, I also prefer those in one statement. But yes, it
> looks cleaner here, in two lines.

I'd say the rule of thumb is that combined declaration and
initialization is fine if the initialization is trivial, in particular
can never fail. If you need to check the return value, like in this
case, I'd prefer separate initialization.

BR,
Jani.


-- 
Jani Nikula, Intel Open Source Technology Center

[toc] | [prev] | [next] | [standalone]

#1666901

FromDawid Kurek <dawikur@gmail.com>
Date2017-06-15 19:00 +0200
Message-ID<tSGhH-4U2-7@gated-at.bofh.it>
In reply to#1666796
On 15/06/17, Jani Nikula wrote:
> On Thu, 15 Jun 2017, Dawid Kurek <dawikur@gmail.com> wrote:
> > On 15/06/17, Jani Nikula wrote:
> >> Separate declaration and initialization would lead to a cleaner patch
> >> and result.
> >
> > I saw combining declaration and initialization is quite common, i.e. in
> > drm_atomic file. Personally, I also prefer those in one statement. But yes, it
> > looks cleaner here, in two lines.
> 
> I'd say the rule of thumb is that combined declaration and
> initialization is fine if the initialization is trivial, in particular
> can never fail. If you need to check the return value, like in this
> case, I'd prefer separate initialization.
> 
Yeah, makes sense. If you need to check then it exceeds simple initialization,
and then it is not declare&initialize. Yes, now I see it.

Thanks a lot,
Dawid

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web