Created attachment 299 [details]
em8300 make log

Created attachment 300 [details]
em8300 driver uninstall details

It's related to 2.6.38 kernel changes regarding V4L.
Maybe you can give a try to https://github.com/austriancoder/v4l2-em8300 ?

CC: (none) => balcaen.john

Hi John, thanx for your reply and sorry to be back so late in this tread... But I'm not a developper and I don't know how to deal with source code. And the old PC I try to keep alive with Mageia (a good old K6-II 300; in the list of recent and easy-to-use distros, only Mageia can run on it!) is owned by an only-end-user.

Today I did an update of my Mageia and I retry to install em8300 as I did before but with same results.

So I hope that someone can rebuild the em8300 driver....... And I'm sorry to be unable to do so..

Johan

What is the status of this bug ? still present ? (and sorry for the biiig delay :) )

Keywords: (none) => NEEDINFO
Component: New RPM package request => RPM Packages

Hi there,
yes it is, I am having a look at it. em8300 indeed uses v4l1 which is no longer present in current kernels.

A solution is indeed to import the new v4l2 module mentioned above, but it does not support all the old stuff apparently.

Another way would be to still use the old v4l1 api, we still have it under libv4l1-devel. I am exploring this.

Cheers,
Chris.

CC: (none) => dirteat

Thomas, what do you think about this?

CC: (none) => tmb

(In reply to comment #6)
> Hi there,
> yes it is, I am having a look at it. em8300 indeed uses v4l1 which is no longer
> present in current kernels.
> 
> A solution is indeed to import the new v4l2 module mentioned above, but it does
> not support all the old stuff apparently.
> 
> Another way would be to still use the old v4l1 api, we still have it under
> libv4l1-devel. I am exploring this.
> 
> Cheers,
> Chris.


@ Chris
Did your exploring result in anything useful?


(In reply to comment #7)
> Thomas, what do you think about this?


@ tmb

Did you see Oliver's question?

Hi,
yes, unfortunately, I think it is dead.

I fixed wmwebcam to use libv4l1, but, tmb should confirm, as far as I understand now, a kernel module (by definition) cannot be linked to any lib outside the kernel. So, using libv4l1 to make em8300 working is a dead-end.

The only option looks indeed an import of the new package

https://github.com/austriancoder/v4l2-em8300

and we should drop em8300?

Chris.

(In reply to comment #9)
> Hi,
> yes, unfortunately, I think it is dead.
> 
> I fixed wmwebcam to use libv4l1, but, tmb should confirm, as far as I
> understand now, a kernel module (by definition) cannot be linked to any lib
> outside the kernel. So, using libv4l1 to make em8300 working is a dead-end.
> 
> The only option looks indeed an import of the new package
> 
> https://github.com/austriancoder/v4l2-em8300
> 
> and we should drop em8300?
> 
> Chris.

assigning to em8300 maintainer

Assignee: bugsquad => anssi.hannula

Hi,

This bug was filed against cauldron, but we do not have cauldron at the moment.

Please report whether this bug is still valid for Mageia 2.

Thanks :)

Cheers,
marja

Keywords: (none) => NEEDINFO

Hi,

Within the next few days, I will try to install my old Hollywood+ PCI board with the em8300 driver + kernel module that are currently available in the Mageia 2 repositories (em8300 v0.18.0 rev. 4.mga2).

I will answer back soon if the driver works on my PC or if it is still broken.

Regards,
Johan

(In reply to comment #12)
> Hi,
> 
> Within the next few days, I will try to install my old Hollywood+ PCI board
> with the em8300 driver + kernel module that are currently available in the
> Mageia 2 repositories (em8300 v0.18.0 rev. 4.mga2).
> 
> I will answer back soon if the driver works on my PC or if it is still broken.
> 
> Regards,
> Johan

Hi Johan,

Maybe you installed and forgot to give the feedback here?

Cheers,
Marja

Created attachment 2453 [details]
Contains dmesg, boot.log, em8300.bin, hardware config and em8300 debug install

Hi everyone,

(I'm still sorry 1. to be late to answer, and 2. for my bad english....)

I installed the em8300 driver with the kernel module and with the binary firmware (em8300.bin) and ... nothing works!

I tried many folders for the binary firmware, such as /lib/firmware/, /lib/firmware/em8300/, /etc/firmware/, /etc/firmware/em8300/ and still nothing.

I didn't found where else to place the firmware...

Modprobe em8300 returns no module found.

When booting up in text mode (by pressing ESC at the graphical boot up screen), I've got a message about installing em8300 module, but it fails!

In the Mageia Control Center -> Hardware -> Detected hardware, my Hollywood+ is detected but there is no driver loaded: the message is "Module: unknown".

As attachment #2453 [details], I added an archive <Mageia_2_em8300_retry_2012-06-14.zip> that contains:
- the last boot.log
- the last dmesg
- Hardware_Config.log that contains hardware informations about my config
- em8300_Debug_Install.txt: the last debug install try of the em8300 + dkms-em8300 packages (after em8300xx previous uninstall)
- The binary em8300.bin firmware that I currently use


In the boot.log, an "install" message starts at line 78 and ends later at lines 89-90. Here it is:
"
em8300 (0.18.0-4.mga2): Installing module.
..........................(bad exit status: 10)
  Build failed.  Installation skipped.
"


So, I don't know what to do else, I don't know where to find any debug information and I don't know where I can find any install documentation...

My Mageia 2 is up to date, I only use two channels: Core Release and Nonfree Release.

Regards,
Johan

Thx for all the feedback, Johan.

@ Anssi

If you agree this bug was assigned correctly, please set it to ASSIGNED or put OK on the whiteboard.

Thx

Keywords: NEEDINFO => (none)
Source RPM: em8300-0.18.0-3.mga1 => em8300-0.18.0-4.mga2
Whiteboard: (none) => MGA2TOO

It is assigned correctly, unfortunately it will be some time before I'll have time to look at this properly.

Status: NEW => ASSIGNED

Hi guys,
I think we should drop this package as I won't see it working in a reasonable future no?

cheers.

This bug report hasn't seen any action since before Mga4 release. Is it still valid in current cauldron (or, for traditional installer: with the 6dev1 snapshot, or for live isos with Mga5)?

Keywords: (none) => NEEDINFO

Hi,
as mentionned above, this bug cannot be fixed without a serious rewritting of this driver. I am closing at WONTFIX.

The only way to use it is to use a former kernel, in the 2.6.* branch, which also makes sense as the Hollywood card is almost 20 years old now!

Status: ASSIGNED => RESOLVED
Resolution: (none) => WONTFIX