Hi,
it looks like select() is broken for the cs4281 sound driver
in the 2.4 kernel. This breaks pretty much all GUI ham radio
applications that use the sound card ;(
I've done some tracing on various ham radio applications, but
none of the ones that use select() get any data from the cs4281
driver.
The applications tried include glfer and hamfax.
kind regards,
Rik
--
Bravely reimplemented by the knights who say "NIH".
http://www.surriel.com/ http://distro.conectiva.com/
Spamtraps of the month: [email protected] [email protected]
which 2.4 version are you using? 2.4.19? There was a ham-radio
select(2) fix in the cs4281 driver back in 2.4.17 era, think that
it got into the 2.4.18 kernel tree. I just checked the 2.4.19
kernel and the fix does seem to be included. i'll place a tarbz2
file out on an FTP server, and if you can try this driver, that
would help me out. if the driver does not build under your tree,
let me know, it's circa 2.4.17 and i don't know the 2.4.19 mods
concerning drivers.
server: ftp1.cirrus.com
Username: ftppclink
Password: cSPxQMd
i'll put cs4281-src-20011214-01n-tar.bz2 into \cs4281 directory.
thanks
Tom
-----Original Message-----
From: Rik van Riel [mailto:[email protected]]
Sent: Thursday, September 05, 2002 7:20 PM
To: [email protected]
Cc: [email protected]; [email protected]
Subject: cs4281 & select in 2.4
Hi,
it looks like select() is broken for the cs4281 sound driver
in the 2.4 kernel. This breaks pretty much all GUI ham radio
applications that use the sound card ;(
I've done some tracing on various ham radio applications, but
none of the ones that use select() get any data from the cs4281
driver.
The applications tried include glfer and hamfax.
kind regards,
Rik
--
Bravely reimplemented by the knights who say "NIH".
http://www.surriel.com/ http://distro.conectiva.com/
Spamtraps of the month: [email protected] [email protected]
On Fri, 6 Sep 2002, Rik van Riel wrote:
> On Fri, 6 Sep 2002, Woller, Thomas wrote:
>
> > which 2.4 version are you using? 2.4.19? There was a ham-radio
> > select(2) fix in the cs4281 driver back in 2.4.17 era, think that
> > i'll put cs4281-src-20011214-01n-tar.bz2 into \cs4281 directory.
>
> Thanks. I've just grabbed the tarball and will let you know
> if this fixes things.
You're right, all updates were already included in the 2.4.19
kernel. The only thing extra in your tarball is the 2.2 compat
code and two copy_to_user() return code fixes in cs4281_ioctl().
I wonder if it's some strange bug with the CS4281 hardware in
this Toshiba laptop:
00:08.0 Multimedia audio controller: Cirrus Logic Crystal CS4281 PCI Audio (rev 01)
Subsystem: Toshiba America Info Systems: Unknown device ff00
Flags: bus master, medium devsel, latency 64, IRQ 5
Memory at fc010000 (32-bit, non-prefetchable) [size=4K]
Memory at fc000000 (32-bit, non-prefetchable) [size=64K]
Does this ring a bell (pun intended) ?
regards,
Rik
--
Bravely reimplemented by the knights who say "NIH".
http://www.surriel.com/ http://distro.conectiva.com/
Spamtraps of the month: [email protected] [email protected]
On Fri, 6 Sep 2002, Woller, Thomas wrote:
> which 2.4 version are you using? 2.4.19? There was a ham-radio
> select(2) fix in the cs4281 driver back in 2.4.17 era, think that
> it got into the 2.4.18 kernel tree. I just checked the 2.4.19
> kernel and the fix does seem to be included.
I'm using 2.4.19, but select() doesn't work just as it
didn't in 2.4.16.
> i'll place a tarbz2 file out on an FTP server, and if you can try this
> driver, that would help me out. if the driver does not build under your
> tree, let me know, it's circa 2.4.17 and i don't know the 2.4.19 mods
> concerning drivers.
> i'll put cs4281-src-20011214-01n-tar.bz2 into \cs4281 directory.
Thanks. I've just grabbed the tarball and will let you know
if this fixes things.
kind regards,
Rik
--
Bravely reimplemented by the knights who say "NIH".
http://www.surriel.com/ http://distro.conectiva.com/
Spamtraps of the month: [email protected] [email protected]
Toshiba refused to lend us any laptops after repeated requests,
so all the toshiba work has been without a unit to test with.
There is at least one (that i remember) specific fix for toshiba
systems in the driver.
also does recording work at all? i mostly tested with grecord.
so, look at cs4281_setup_record_src(), and maybe commenting out
the toshiba specific code concerning the recording sources might
make a difference. or if the unit is NOT recognized, then forcing
the code to setup as your unit as a 1640CDT and see if that
helps.
i'll think about it some more. let me know
tom
On Fri, 6 Sep 2002, Rik van Riel wrote:
> On Fri, 6 Sep 2002, Woller, Thomas wrote:
>
> > which 2.4 version are you using? 2.4.19? There was a
ham-radio
> > select(2) fix in the cs4281 driver back in 2.4.17 era, think
that
> > i'll put cs4281-src-20011214-01n-tar.bz2 into \cs4281
directory.
>
> Thanks. I've just grabbed the tarball and will let you know
> if this fixes things.
You're right, all updates were already included in the 2.4.19
kernel. The only thing extra in your tarball is the 2.2 compat
code and two copy_to_user() return code fixes in cs4281_ioctl().
I wonder if it's some strange bug with the CS4281 hardware in
this Toshiba laptop:
00:08.0 Multimedia audio controller: Cirrus Logic Crystal CS4281
PCI Audio (rev 01)
Subsystem: Toshiba America Info Systems: Unknown device
ff00
Flags: bus master, medium devsel, latency 64, IRQ 5
Memory at fc010000 (32-bit, non-prefetchable) [size=4K]
Memory at fc000000 (32-bit, non-prefetchable) [size=64K]
Does this ring a bell (pun intended) ?
regards,
Rik
--
Bravely reimplemented by the knights who say "NIH".
http://www.surriel.com/ http://distro.conectiva.com/
Spamtraps of the month: [email protected] [email protected]
On Fri, 6 Sep 2002, Woller, Thomas wrote:
> Toshiba refused to lend us any laptops after repeated requests,
> so all the toshiba work has been without a unit to test with.
> There is at least one (that i remember) specific fix for toshiba
> systems in the driver.
> also does recording work at all?
Works fine with kpsk and other programs that just read.
It seems to break only with select()
> so, look at cs4281_setup_record_src(), and maybe commenting out
> the toshiba specific code concerning the recording sources might
> make a difference. or if the unit is NOT recognized, then forcing
> the code to setup as your unit as a 1640CDT and see if that
> helps.
I'll switch the debugging on and I'll see if anything
happens.
regards,
Rik
--
Bravely reimplemented by the knights who say "NIH".
http://www.surriel.com/ http://distro.conectiva.com/
Spamtraps of the month: [email protected] [email protected]