Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 31 Oct 2006 18:35:02 +0200
From:      "Konstantin Dimitrov" <kosio.dimitrov@gmail.com>
To:        pyunyh@gmail.com
Cc:        freebsd-multimedia@freebsd.org
Subject:   Re: Status of VIA Envy24 audio controller
Message-ID:  <8103ad500610310835y414824eek7a6a355168d4246b@mail.gmail.com>
In-Reply-To: <20061031083219.GE55910@cdnetworks.co.kr>
References:  <200606072039.13422.shoesoft@gmx.net> <787520ee0607180732r6bcb1ae4h1591230d2db366a@mail.gmail.com> <200610111331.39605.lrntct@gmail.com> <8103ad500610111048v756778c6pcac4006746ec3f0a@mail.gmail.com> <ef10de9a0610280043x570c7b81tfd93a456038154fc@mail.gmail.com> <20061028082147.GB44390@cdnetworks.co.kr> <8103ad500610291101l291c5e04xfc6874e4abda6f83@mail.gmail.com> <20061031005233.GC55910@cdnetworks.co.kr> <8103ad500610302209y4f76f8c3pdc3370097f587f94@mail.gmail.com> <20061031083219.GE55910@cdnetworks.co.kr>

next in thread | previous in thread | raw e-mail | index | archive | help
On 10/31/06, Pyun YongHyeon <pyunyh@gmail.com> wrote:
>
> On Tue, Oct 31, 2006 at 08:09:41AM +0200, Konstantin Dimitrov wrote:
> > On 10/31/06, Pyun YongHyeon <pyunyh@gmail.com> wrote:
> > >
> > >On Sun, Oct 29, 2006 at 09:01:30PM +0200, Konstantin Dimitrov wrote:
> > >> sorry for my late answer, i'm too busy, but today i had some free
> time
> > >to
> > >> look at the problems. the two described problems are different and
> not
> > >> related. here is the new working code:
> > >>
> > >> for Envy24 (fix mutex problems on current and panic on verbose boot)
> :
> > >> http://envy24.svobodno.com/driver/envy24/current/envy24.c
> > >>
> > >> for Envy24HT(fix mutex problems on current and panic on verbose boot,
> > >enable
> > >> vchans):
> > >> http://envy24.svobodno.com/driver/envy24ht/current/envy24ht.c
> > >> http://envy24.svobodno.com/driver/envy24ht/current/envy24ht.h
> > >>
> > >> vchans support is now enabled in the Envy24HT code and hope it's
> working
> > >ok.
> > >>
> > >
> > >Thanks. It now loads ok on CURRENT but I can't hear anything.
> > >I can't sure but I have to make a cfg_table for
> > >"ONKYO SE-150PCI PLUS". ATM it seems that it uses generic
> > >configuration for the hardware. I'll let you if I mananged to
> > >make it work.
> >
> >
> > you can't hear anything, because DAC on your ONKYO SE-150PCI PLUS is not
> > initialized.
> >
> > Laurent Courty (lrntct at gmail.com) has ONKYO SE-90PCI and I have been
> > trying to get it running for more than 2 weeks now, but without success
> and
> > filling the cfg_table is not so easy at all, because it involves knowing
> the
> > hardware specific design of the board, so the process on adding support
> for
> > new unsupported card can be summarized in the following steps:
> >
> > 0) examine the board to see what ICs they use for DACs, ADCs, DITs, etc.
> > 1) make EEPROM dump (i wrote small kernel module that do the job:
> > http://www.nbritton.org/uploads/envy24/envy24_dump_v01.tar)
> > 2) find the hardware specific design of the board (wiring between Envy24
> > chip and other important chips on the board like DACs)
> > 3) fill the cfg_table and if needed write the necessary card specific
> code
> > (codec init code, etc.)
> >
> > on the table here: http://envy24.svobodno.com , you can see all the
> above
> > info for supported boards till now - ICs, EEPROM dumps, wiring info,
> etc.
> >
> > so do the steps 0) and 1) and mail me that info, then i will try to help
> you
> > with 2) and 3).
> >
>
> Thanks a lot. I've checked the hardware but I can't sure what ICs
> are used for DAC/ADC etc. The information I can get from the vendor's
> site and checking the hardware indicates the following.
> 2ch only DAC : WM8716SEDS
> CODEC : WM8776S
> 6ch DAC : WM8766G
> See http://onkyo.jp/wavio/150pci/03_03.htm for their description
> for the hardware.


ONKYO SE-90PCI also uses WM8716, so maybe they use the same wiring for both
cards and WM8716 is the DAC used for stereo output on both cards. do you
think that ONKYO will provide the wiring information for both cards if we
ask them for ? because the alternative is to use digital ohmmeter to reverse
engineering the wiring and it's very hard, that is why i have problems with
ONKYO SE-90PCI ...

I've attached EEPROM dump and patch for the dump program. Personally
> I think it should be included in envy24ht(4) and users can extract
> that information with sysctl(8) interface. You can add that information
> with SYSCTL_ADD_PROC.

BTW, if we need EEPROM information to initialize the hardware why
> hardcode cfg_table for each hardware model?


most of the Envy24HT-based cards have not valid EEPROM data or even empty
EEPROM (that contains only PCI SubVendor and SubDevice IDs), good examples
for such cards, are  all cards made by Terratec and Audiotrak, so their init
values are hard coded in their windows drivers and in this case all EEPROM
related routines in the driver are worthless. however if  the card has valid
eeprom data that can help to find the right wiring and also can be used to
initialize the card.

> >On 10/28/06, Pyun YongHyeon <pyunyh@gmail.com> wrote:
> > >> >
> > >> >On Sat, Oct 28, 2006 at 02:43:53AM -0500, Nikolas Britton wrote:
> > >> >> Hey Konstantin, I think I found a bug in the code. If I boot
> FreeBSD
> > >> >> with verbose logging the envy24 driver will panic the system
> during
> > >> >> the boot process. fatal trap 12, page fault, and something about
> > >> >> "envy24(play): setmap; xxxxx", I'd have to build a debug kernel
> etc.
> > >> >> if you want the full message and backtrace but this should be easy
> to
> > >> >> reproduce. I'm using envy24 sources from -CURRENT that I manually
> > >> >> MFC'd to FreeBSD 6.2-PRERELEASE. Oh... I'm using an M Audio
> > >Audiophile
> > >> >> 2496.
> > >> >>
> > >> >
> > >> >I saw the panic on envy24ht(4) too and sent a mail to Konstantin but
> > >> >got no reply yet. With WITNESS enabled kernel it shows several LORs
> > >> >and panicked due to use of recursed mutex(channel lock can't be
> > >> >acquired recursively). I guess correct fix could be done by moving
> > >> >buffer resizing code to channel_setblocksize method instead of
> > >> >channel_setformat method. Unfortunately I haven't have free time
> > >> >to investiage further due to msk(4) work and still have little time.
> > >:-(
> > >> >
> > >> >--
> > >> >Regards,
> > >> >Pyun YongHyeon
> > >> >
> > >
> > >--
> > >Regards,
> > >Pyun YongHyeon
> > >
>
> --
> Regards,
> Pyun YongHyeon
>
>
>



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?8103ad500610310835y414824eek7a6a355168d4246b>