From: Shuah Khan <shuahkh@osg.samsung.com>
To: Johannes Stezenbach <js@linuxtv.org>
Cc: Shuah Khan <shuah.kh@samsung.com>,
linux-media@vger.kernel.org,
Mauro Carvalho Chehab <m.chehab@samsung.com>,
Shuah Khan <shuahkh@osg.samsung.com>
Subject: Re: em28xx breaks after hibernate
Date: Thu, 25 Sep 2014 11:40:45 -0600 [thread overview]
Message-ID: <5424539D.8090503@osg.samsung.com> (raw)
In-Reply-To: <20140925160134.GA6207@linuxtv.org>
On 09/25/2014 10:01 AM, Johannes Stezenbach wrote:
> Hi Shuah,
>
> On Thu, Sep 25, 2014 at 07:45:37AM -0600, Shuah Khan wrote:
>> On 09/25/2014 06:53 AM, Johannes Stezenbach wrote:
>>> ever since your patchset which implements suspend/resume
>>> for em28xx, hibernating the system breaks the Hauppauge WinTV HVR 930C driver.
>>> In v3.15.y and v3.16.y it throws a request_firmware warning
>>> during hibernate + resume, and the /dev/dvb/ device nodes disappears after
>>> resume. In current git v3.17-rc6-247-g005f800, it hangs
>>> after resume. I bisected the hang in qemu to
>>> b89193e0b06f "media: em28xx - remove reset_resume interface",
>>> the hang is fixed if I revert this commit on top of v3.17-rc6-247-g005f800.
>>>
>>> Regarding the request_firmware issue. I think a possible
>>> fix would be:
>
> I think you should take a closer look at the code you snipped.
> Maybe this fixes the root of the issue you worked around
> with the DVB_FE_DEVICE_RESUME thing, namely calling
> fe->ops.tuner_ops.init from wrong context.
Sorry for aggressive snipping. :)
Right. I introduced DVB_FE_DEVICE_RESUME code to resume
problems in drx39xxj driver. Because I had to make it not
toggle power on the fe for resume. In other words, for it
to differentiate between disconnect and resume conditions.
dvb_frontend_resume() is used by dvb_usbv2 dvb_usb_core -
dvb_usbv2_resume_common()
Calling dvb_frontend_reinitialise() from dvb_frontend_resume()
could break dvb_usbv2 drivers because it has handling for
reset_resume in its core in dvb_usbv2_reset_resume()
reverting media: em28xx - remove reset_resume interface
might be a short-term solution. I think the longterm
solution is adding a dvb_frontend_reset_resume() that
does dvb_frontend_reinitialise() just like you suggested.
In addition, em28xx will call dvb_frontend_reset_resume()
from its reset_resume
What do you think?
thanks,
-- Shuah
--
Shuah Khan
Sr. Linux Kernel Developer
Samsung Research America (Silicon Valley)
shuahkh@osg.samsung.com | (970) 217-8978
next prev parent reply other threads:[~2014-09-25 17:40 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-09-25 12:53 em28xx breaks after hibernate Johannes Stezenbach
2014-09-25 13:45 ` Shuah Khan
2014-09-25 14:10 ` Mauro Carvalho Chehab
2014-09-25 15:07 ` Shuah Khan
2014-09-25 16:01 ` Johannes Stezenbach
2014-09-25 17:36 ` Johannes Stezenbach
2014-09-25 18:01 ` Shuah Khan
2014-09-25 17:40 ` Shuah Khan [this message]
2014-09-25 18:17 ` Johannes Stezenbach
2014-09-25 18:45 ` Shuah Khan
2014-09-25 19:03 ` Shuah Khan
2014-09-26 7:41 ` Johannes Stezenbach
2014-09-26 8:00 ` Johannes Stezenbach
2014-09-26 8:08 ` Johannes Stezenbach
2014-09-26 10:14 ` Mauro Carvalho Chehab
2014-09-26 11:07 ` Johannes Stezenbach
2014-09-26 11:42 ` Mauro Carvalho Chehab
2014-09-26 12:02 ` Johannes Stezenbach
2014-09-26 12:11 ` Mauro Carvalho Chehab
2014-09-26 12:03 ` Mauro Carvalho Chehab
2014-09-26 12:27 ` Johannes Stezenbach
2014-09-26 13:12 ` Mauro Carvalho Chehab
2014-09-26 13:25 ` Johannes Stezenbach
2014-09-26 14:25 ` Shuah Khan
2014-09-26 14:25 ` Johannes Stezenbach
2014-09-26 14:30 ` Shuah Khan
2014-09-26 15:06 ` Johannes Stezenbach
2014-09-26 15:22 ` Johannes Stezenbach
2014-09-26 15:43 ` Mauro Carvalho Chehab
2014-09-28 10:55 ` Johannes Stezenbach
2014-09-28 11:12 ` Mauro Carvalho Chehab
2014-09-28 11:54 ` Johannes Stezenbach
2014-09-28 12:35 ` Mauro Carvalho Chehab
2014-09-28 12:57 ` Mauro Carvalho Chehab
2014-09-29 2:38 ` Mauro Carvalho Chehab
2014-09-26 15:22 ` Shuah Khan
2014-09-26 15:35 ` Mauro Carvalho Chehab
2014-09-26 15:40 ` Mauro Carvalho Chehab
2014-09-26 15:42 ` Johannes Stezenbach
2014-09-26 15:48 ` Mauro Carvalho Chehab
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=5424539D.8090503@osg.samsung.com \
--to=shuahkh@osg.samsung.com \
--cc=js@linuxtv.org \
--cc=linux-media@vger.kernel.org \
--cc=m.chehab@samsung.com \
--cc=shuah.kh@samsung.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.