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

Groups > linux.kernel > #1157827 > unrolled thread

Re: [PATCH 2/2] ts2020: Provide DVBv5 API signal strength

Started byDavid Howells <dhowells@redhat.com>
First post2015-06-03 18:40 +0200
Last post2015-06-03 18:50 +0200
Articles 5 — 3 participants

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: [PATCH 2/2] ts2020: Provide DVBv5 API signal strength David Howells <dhowells@redhat.com> - 2015-06-03 18:40 +0200
    Re: [PATCH 2/2] ts2020: Provide DVBv5 API signal strength Antti Palosaari <crope@iki.fi> - 2015-06-03 18:50 +0200
      Re: [PATCH 2/2] ts2020: Provide DVBv5 API signal strength Malcolm Priestley <tvboxspy@gmail.com> - 2015-06-03 19:10 +0200
      Re: [PATCH 2/2] ts2020: Provide DVBv5 API signal strength David Howells <dhowells@redhat.com> - 2015-06-03 19:20 +0200
    Re: [PATCH 2/2] ts2020: Provide DVBv5 API signal strength Malcolm Priestley <tvboxspy@gmail.com> - 2015-06-03 18:50 +0200

#1157827 — Re: [PATCH 2/2] ts2020: Provide DVBv5 API signal strength

FromDavid Howells <dhowells@redhat.com>
Date2015-06-03 18:40 +0200
SubjectRe: [PATCH 2/2] ts2020: Provide DVBv5 API signal strength
Message-ID<pxjRV-6Dp-9@gated-at.bofh.it>
Malcolm Priestley <tvboxspy@gmail.com> wrote:

> >> Yes, also, the workqueue appears not to be initialized when using the dvb
> >> attached method.
> >
> > I'm not sure what you're referring to.  It's initialised in ts2020_probe()
> > just after the ts2020_priv struct is allocated - the only place it is
> > allocated.
> >
> ts2020_probe() isn't touched by devices not converted to I2C binding.

Hmmm...  Doesn't that expose a larger problem?  The only place the ts2020_priv
struct is allocated is in ts2020_probe() within ts2020.c and the struct
definition is private to that file and so it can't be allocated from outside.
So if you don't pass through ts2020_probe(), fe->tuner_priv will remain NULL
and the driver will crash.

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

[toc] | [next] | [standalone]

#1157833

FromAntti Palosaari <crope@iki.fi>
Date2015-06-03 18:50 +0200
Message-ID<pxk1A-6Pj-5@gated-at.bofh.it>
In reply to#1157827
On 06/03/2015 07:37 PM, David Howells wrote:
> Malcolm Priestley <tvboxspy@gmail.com> wrote:
>
>>>> Yes, also, the workqueue appears not to be initialized when using the dvb
>>>> attached method.
>>>
>>> I'm not sure what you're referring to.  It's initialised in ts2020_probe()
>>> just after the ts2020_priv struct is allocated - the only place it is
>>> allocated.
>>>
>> ts2020_probe() isn't touched by devices not converted to I2C binding.
>
> Hmmm...  Doesn't that expose a larger problem?  The only place the ts2020_priv
> struct is allocated is in ts2020_probe() within ts2020.c and the struct
> definition is private to that file and so it can't be allocated from outside.
> So if you don't pass through ts2020_probe(), fe->tuner_priv will remain NULL
> and the driver will crash.

Malcolm misses some pending patches where attach() is wrapped to I2C 
model probe().
http://git.linuxtv.org/cgit.cgi/anttip/media_tree.git/log/?h=ts2020

regards
Antti

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

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

#1157869

FromMalcolm Priestley <tvboxspy@gmail.com>
Date2015-06-03 19:10 +0200
Message-ID<pxkkX-7sr-37@gated-at.bofh.it>
In reply to#1157833

On 03/06/15 17:43, Antti Palosaari wrote:
> On 06/03/2015 07:37 PM, David Howells wrote:
>> Malcolm Priestley <tvboxspy@gmail.com> wrote:
>>
>>>>> Yes, also, the workqueue appears not to be initialized when using
>>>>> the dvb
>>>>> attached method.
>>>>
>>>> I'm not sure what you're referring to.  It's initialised in
>>>> ts2020_probe()
>>>> just after the ts2020_priv struct is allocated - the only place it is
>>>> allocated.
>>>>
>>> ts2020_probe() isn't touched by devices not converted to I2C binding.
>>
>> Hmmm...  Doesn't that expose a larger problem?  The only place the
>> ts2020_priv
>> struct is allocated is in ts2020_probe() within ts2020.c and the struct
>> definition is private to that file and so it can't be allocated from
>> outside.
>> So if you don't pass through ts2020_probe(), fe->tuner_priv will
>> remain NULL
>> and the driver will crash.
>
> Malcolm misses some pending patches where attach() is wrapped to I2C
> model probe().
> http://git.linuxtv.org/cgit.cgi/anttip/media_tree.git/log/?h=ts2020
>

Hmmm... Yes, I am indeed missing those patches.

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

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

#1157913

FromDavid Howells <dhowells@redhat.com>
Date2015-06-03 19:20 +0200
Message-ID<pxkuE-7Ey-59@gated-at.bofh.it>
In reply to#1157833
Antti Palosaari <crope@iki.fi> wrote:

> Malcolm misses some pending patches where attach() is wrapped to I2C model
> probe().
> http://git.linuxtv.org/cgit.cgi/anttip/media_tree.git/log/?h=ts2020

Aha!  That explains it.

	ts2020: register I2C driver from legacy media attach

removes the allocation from attach() in the branch I'm working on top of.

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

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

#1157839

FromMalcolm Priestley <tvboxspy@gmail.com>
Date2015-06-03 18:50 +0200
Message-ID<pxk1B-6Pj-27@gated-at.bofh.it>
In reply to#1157827

On 03/06/15 17:37, David Howells wrote:
> Malcolm Priestley <tvboxspy@gmail.com> wrote:
>
>>>> Yes, also, the workqueue appears not to be initialized when using the dvb
>>>> attached method.
>>>
>>> I'm not sure what you're referring to.  It's initialised in ts2020_probe()
>>> just after the ts2020_priv struct is allocated - the only place it is
>>> allocated.
>>>
>> ts2020_probe() isn't touched by devices not converted to I2C binding.
>
> Hmmm...  Doesn't that expose a larger problem?  The only place the ts2020_priv
> struct is allocated is in ts2020_probe() within ts2020.c and the struct
> definition is private to that file and so it can't be allocated from outside.
> So if you don't pass through ts2020_probe(), fe->tuner_priv will remain NULL
> and the driver will crash.
>

No, it is also allocated in ts2020_attach.

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

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web