[PATCH] video/smscufx: fix line counting in fb_write

[PATCH] video/smscufx: fix line counting in fb_write

[Alexander Holler]

Line 0 and 1 were both written to line 0 (on the display) and all subsequent
lines had an offset of -1. The result was that the last line on the display
was never overwritten by writes to /dev/fbN.

The origin of this bug seems to have been udlfb.

Signed-off-by: Alexander Holler <holler@ahsoftware.de>
---
 drivers/video/smscufx.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/video/smscufx.c b/drivers/video/smscufx.c
index ccbfef5..1e1e2d2 100644
--- a/drivers/video/smscufx.c
+++ b/drivers/video/smscufx.c
@@ -904,7 +904,7 @@ static ssize_t ufx_ops_write(struct fb_info *info, const char __user *buf,
 	result = fb_sys_write(info, buf, count, ppos);
 
 	if (result > 0) {
-		int start = max((int)(offset / info->fix.line_length) - 1, 0);
+		int start = max((int)(offset / info->fix.line_length), 0);
 		int lines = min((u32)((result / info->fix.line_length) + 1),
 				(u32)info->var.yres);
 
-- 
1.7.6.5

[PATCH] video/smscufx: fix line counting in fb_write

[Alexander Holler]

Line 0 and 1 were both written to line 0 (on the display) and all subsequent
lines had an offset of -1. The result was that the last line on the display
was never overwritten by writes to /dev/fbN.

The origin of this bug seems to have been udlfb.

Signed-off-by: Alexander Holler <holler@ahsoftware.de>
---
 drivers/video/smscufx.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/video/smscufx.c b/drivers/video/smscufx.c
index ccbfef5..1e1e2d2 100644
--- a/drivers/video/smscufx.c
+++ b/drivers/video/smscufx.c
@@ -904,7 +904,7 @@ static ssize_t ufx_ops_write(struct fb_info *info, const char __user *buf,
 	result = fb_sys_write(info, buf, count, ppos);
 
 	if (result > 0) {
-		int start = max((int)(offset / info->fix.line_length) - 1, 0);
+		int start = max((int)(offset / info->fix.line_length), 0);
 		int lines = min((u32)((result / info->fix.line_length) + 1),
 				(u32)info->var.yres);
 
-- 
1.7.6.5

Re: [PATCH] video/smscufx: fix line counting in fb_write

[Alexander Holler]

Hello,

as for the patch for udlfb, I forgot to mention that this is a candidate 
for all stable trees 3.2 and above.

Btw., the address of the maintainer doesn't seem to be valid anymore.

Regards,

Alexander

Am 21.04.2012 00:11, schrieb Alexander Holler:
> Line 0 and 1 were both written to line 0 (on the display) and all subsequent
> lines had an offset of -1. The result was that the last line on the display
> was never overwritten by writes to /dev/fbN.
>
> The origin of this bug seems to have been udlfb.
>
> Signed-off-by: Alexander Holler<holler@ahsoftware.de>

Cc: stable@vger.kernel.org

> ---
>   drivers/video/smscufx.c |    2 +-
>   1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/drivers/video/smscufx.c b/drivers/video/smscufx.c
> index ccbfef5..1e1e2d2 100644
> --- a/drivers/video/smscufx.c
> +++ b/drivers/video/smscufx.c
> @@ -904,7 +904,7 @@ static ssize_t ufx_ops_write(struct fb_info *info, const char __user *buf,
>   	result = fb_sys_write(info, buf, count, ppos);
>
>   	if (result>  0) {
> -		int start = max((int)(offset / info->fix.line_length) - 1, 0);
> +		int start = max((int)(offset / info->fix.line_length), 0);
>   		int lines = min((u32)((result / info->fix.line_length) + 1),
>   				(u32)info->var.yres);
>

Re: [PATCH] video/smscufx: fix line counting in fb_write

[Alexander Holler]

Hello,

as for the patch for udlfb, I forgot to mention that this is a candidate 
for all stable trees 3.2 and above.

Btw., the address of the maintainer doesn't seem to be valid anymore.

Regards,

Alexander

Am 21.04.2012 00:11, schrieb Alexander Holler:
> Line 0 and 1 were both written to line 0 (on the display) and all subsequent
> lines had an offset of -1. The result was that the last line on the display
> was never overwritten by writes to /dev/fbN.
>
> The origin of this bug seems to have been udlfb.
>
> Signed-off-by: Alexander Holler<holler@ahsoftware.de>

Cc: stable@vger.kernel.org

> ---
>   drivers/video/smscufx.c |    2 +-
>   1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/drivers/video/smscufx.c b/drivers/video/smscufx.c
> index ccbfef5..1e1e2d2 100644
> --- a/drivers/video/smscufx.c
> +++ b/drivers/video/smscufx.c
> @@ -904,7 +904,7 @@ static ssize_t ufx_ops_write(struct fb_info *info, const char __user *buf,
>   	result = fb_sys_write(info, buf, count, ppos);
>
>   	if (result>  0) {
> -		int start = max((int)(offset / info->fix.line_length) - 1, 0);
> +		int start = max((int)(offset / info->fix.line_length), 0);
>   		int lines = min((u32)((result / info->fix.line_length) + 1),
>   				(u32)info->var.yres);
>

Re: [PATCH] video/smscufx: fix line counting in fb_write

[Florian Tobias Schandinat]

Hi Alexander,

On 04/21/2012 11:26 AM, Alexander Holler wrote:
> Hello,
> 
> as for the patch for udlfb, I forgot to mention that this is a candidate
> for all stable trees 3.2 and above.
> 
> Btw., the address of the maintainer doesn't seem to be valid anymore.

it is better to cc me on patches to the framebuffer subsystem for such
cases. I don't have much free time so it's rare that I come around to
dig in the mailing list.

> 
> Regards,
> 
> Alexander
> 
> Am 21.04.2012 00:11, schrieb Alexander Holler:
>> Line 0 and 1 were both written to line 0 (on the display) and all
>> subsequent
>> lines had an offset of -1. The result was that the last line on the
>> display
>> was never overwritten by writes to /dev/fbN.
>>
>> The origin of this bug seems to have been udlfb.
>>
>> Signed-off-by: Alexander Holler<holler@ahsoftware.de>
> 
> Cc: stable@vger.kernel.org

Patch looks good to me but can be made simpler.

> 
>> ---
>>   drivers/video/smscufx.c |    2 +-
>>   1 files changed, 1 insertions(+), 1 deletions(-)
>>
>> diff --git a/drivers/video/smscufx.c b/drivers/video/smscufx.c
>> index ccbfef5..1e1e2d2 100644
>> --- a/drivers/video/smscufx.c
>> +++ b/drivers/video/smscufx.c
>> @@ -904,7 +904,7 @@ static ssize_t ufx_ops_write(struct fb_info *info,
>> const char __user *buf,
>>       result = fb_sys_write(info, buf, count, ppos);
>>
>>       if (result>  0) {
>> -        int start = max((int)(offset / info->fix.line_length) - 1, 0);
>> +        int start = max((int)(offset / info->fix.line_length), 0);

the cast to int as well as the max is superfluous without the -1 as the
value can no longer be negative.

>>           int lines = min((u32)((result / info->fix.line_length) + 1),
>>                   (u32)info->var.yres);
>>
> 

Best regards,

Florian Tobias Schandinat

Re: [PATCH] video/smscufx: fix line counting in fb_write

[Alexander Holler]

Hi Florian,

sorry for the late answer.

Am 13.05.2012 14:47, schrieb Florian Tobias Schandinat:
> Hi Alexander,
>
> On 04/21/2012 11:26 AM, Alexander Holler wrote:
>> Hello,
>>
>> as for the patch for udlfb, I forgot to mention that this is a candidate
>> for all stable trees 3.2 and above.
>>
>> Btw., the address of the maintainer doesn't seem to be valid anymore.
>
> it is better to cc me on patches to the framebuffer subsystem for such
> cases. I don't have much free time so it's rare that I come around to
> dig in the mailing list.
>
>>
>> Regards,
>>
>> Alexander
>>
>> Am 21.04.2012 00:11, schrieb Alexander Holler:
>>> Line 0 and 1 were both written to line 0 (on the display) and all
>>> subsequent
>>> lines had an offset of -1. The result was that the last line on the
>>> display
>>> was never overwritten by writes to /dev/fbN.
>>>
>>> The origin of this bug seems to have been udlfb.
>>>
>>> Signed-off-by: Alexander Holler<holler@ahsoftware.de>
>>
>> Cc: stable@vger.kernel.org
>
> Patch looks good to me but can be made simpler.
>
>>
>>> ---
>>>    drivers/video/smscufx.c |    2 +-
>>>    1 files changed, 1 insertions(+), 1 deletions(-)
>>>
>>> diff --git a/drivers/video/smscufx.c b/drivers/video/smscufx.c
>>> index ccbfef5..1e1e2d2 100644
>>> --- a/drivers/video/smscufx.c
>>> +++ b/drivers/video/smscufx.c
>>> @@ -904,7 +904,7 @@ static ssize_t ufx_ops_write(struct fb_info *info,
>>> const char __user *buf,
>>>        result = fb_sys_write(info, buf, count, ppos);
>>>
>>>        if (result>  0) {
>>> -        int start = max((int)(offset / info->fix.line_length) - 1, 0);
>>> +        int start = max((int)(offset / info->fix.line_length), 0);
>
> the cast to int as well as the max is superfluous without the -1 as the
> value can no longer be negative.

I had that impression too, but I wanted to change as less as possible, 
so I didn't had the need to check types and (their) sizes for possible 
overflows or such. I was lazy and just wanted to fix that one bug. ;)

>
>>>            int lines = min((u32)((result / info->fix.line_length) + 1),
>>>                    (u32)info->var.yres);
>>>

Regards,

Alexander

Re: [PATCH] video/smscufx: fix line counting in fb_write

[Alexander Holler]

Hi Florian,

sorry for the late answer.

Am 13.05.2012 14:47, schrieb Florian Tobias Schandinat:
> Hi Alexander,
>
> On 04/21/2012 11:26 AM, Alexander Holler wrote:
>> Hello,
>>
>> as for the patch for udlfb, I forgot to mention that this is a candidate
>> for all stable trees 3.2 and above.
>>
>> Btw., the address of the maintainer doesn't seem to be valid anymore.
>
> it is better to cc me on patches to the framebuffer subsystem for such
> cases. I don't have much free time so it's rare that I come around to
> dig in the mailing list.
>
>>
>> Regards,
>>
>> Alexander
>>
>> Am 21.04.2012 00:11, schrieb Alexander Holler:
>>> Line 0 and 1 were both written to line 0 (on the display) and all
>>> subsequent
>>> lines had an offset of -1. The result was that the last line on the
>>> display
>>> was never overwritten by writes to /dev/fbN.
>>>
>>> The origin of this bug seems to have been udlfb.
>>>
>>> Signed-off-by: Alexander Holler<holler@ahsoftware.de>
>>
>> Cc: stable@vger.kernel.org
>
> Patch looks good to me but can be made simpler.
>
>>
>>> ---
>>>    drivers/video/smscufx.c |    2 +-
>>>    1 files changed, 1 insertions(+), 1 deletions(-)
>>>
>>> diff --git a/drivers/video/smscufx.c b/drivers/video/smscufx.c
>>> index ccbfef5..1e1e2d2 100644
>>> --- a/drivers/video/smscufx.c
>>> +++ b/drivers/video/smscufx.c
>>> @@ -904,7 +904,7 @@ static ssize_t ufx_ops_write(struct fb_info *info,
>>> const char __user *buf,
>>>        result = fb_sys_write(info, buf, count, ppos);
>>>
>>>        if (result>  0) {
>>> -        int start = max((int)(offset / info->fix.line_length) - 1, 0);
>>> +        int start = max((int)(offset / info->fix.line_length), 0);
>
> the cast to int as well as the max is superfluous without the -1 as the
> value can no longer be negative.

I had that impression too, but I wanted to change as less as possible, 
so I didn't had the need to check types and (their) sizes for possible 
overflows or such. I was lazy and just wanted to fix that one bug. ;)

>
>>>            int lines = min((u32)((result / info->fix.line_length) + 1),
>>>                    (u32)info->var.yres);
>>>

Regards,

Alexander

Re: [PATCH] video/smscufx: fix line counting in fb_write

[Florian Tobias Schandinat]

On 07/02/2012 09:25 PM, Alexander Holler wrote:
> Hi Florian,
> 
> sorry for the late answer.
> 
> Am 13.05.2012 14:47, schrieb Florian Tobias Schandinat:
>> Hi Alexander,
>>
>> On 04/21/2012 11:26 AM, Alexander Holler wrote:
>>> Hello,
>>>
>>> as for the patch for udlfb, I forgot to mention that this is a candidate
>>> for all stable trees 3.2 and above.
>>>
>>> Btw., the address of the maintainer doesn't seem to be valid anymore.
>>
>> it is better to cc me on patches to the framebuffer subsystem for such
>> cases. I don't have much free time so it's rare that I come around to
>> dig in the mailing list.
>>
>>>
>>> Regards,
>>>
>>> Alexander
>>>
>>> Am 21.04.2012 00:11, schrieb Alexander Holler:
>>>> Line 0 and 1 were both written to line 0 (on the display) and all
>>>> subsequent
>>>> lines had an offset of -1. The result was that the last line on the
>>>> display
>>>> was never overwritten by writes to /dev/fbN.
>>>>
>>>> The origin of this bug seems to have been udlfb.
>>>>
>>>> Signed-off-by: Alexander Holler<holler@ahsoftware.de>
>>>
>>> Cc: stable@vger.kernel.org
>>
>> Patch looks good to me but can be made simpler.
>>
>>>
>>>> ---
>>>>    drivers/video/smscufx.c |    2 +-
>>>>    1 files changed, 1 insertions(+), 1 deletions(-)
>>>>
>>>> diff --git a/drivers/video/smscufx.c b/drivers/video/smscufx.c
>>>> index ccbfef5..1e1e2d2 100644
>>>> --- a/drivers/video/smscufx.c
>>>> +++ b/drivers/video/smscufx.c
>>>> @@ -904,7 +904,7 @@ static ssize_t ufx_ops_write(struct fb_info *info,
>>>> const char __user *buf,
>>>>        result = fb_sys_write(info, buf, count, ppos);
>>>>
>>>>        if (result>  0) {
>>>> -        int start = max((int)(offset / info->fix.line_length) - 1, 0);
>>>> +        int start = max((int)(offset / info->fix.line_length), 0);
>>
>> the cast to int as well as the max is superfluous without the -1 as the
>> value can no longer be negative.
> 
> I had that impression too, but I wanted to change as less as possible,
> so I didn't had the need to check types and (their) sizes for possible
> overflows or such. I was lazy and just wanted to fix that one bug. ;)

Well, as this patch fixes a bug I applied it as is.

> 
>>
>>>>            int lines = min((u32)((result / info->fix.line_length) + 1),
>>>>                    (u32)info->var.yres);
>>>>


Best regards,

Florian Tobias Schandinat

Re: [PATCH] video/smscufx: fix line counting in fb_write

[Alexander Holler]

Hello,

Am 26.07.2012 19:26, schrieb Florian Tobias Schandinat:

> Well, as this patch fixes a bug I applied it as is.

Thanks a lot, actually I was more interested to get that one-line-patch 
for udlfb, because I've discovered that bug using one of those Mimo-LCDs. ;)

Are you responsible for that too?

Regards,

Alexander

Re: [PATCH] video/smscufx: fix line counting in fb_write

[Alexander Holler]

Hello,

Am 26.07.2012 19:26, schrieb Florian Tobias Schandinat:

> Well, as this patch fixes a bug I applied it as is.

Thanks a lot, actually I was more interested to get that one-line-patch 
for udlfb, because I've discovered that bug using one of those Mimo-LCDs. ;)

Are you responsible for that too?

Regards,

Alexander