Re: pcm status

2001-01-20 Thread John Baldwin


On 20-Jan-01 Donald J . Maddox wrote:
> This is directly traceable to entropy harvesting by /dev/*random.
> I know it's not an option if you need to use ssh, but not loading
> random.ko will fix the sound problems when moving the mouse or
> typing.  It doesn't fix the 'hwptr went backwards' messages, though.

The entropy stuff only kicks in on the keyboard and mouse.  It can't really
help the hwptr backwards stuff, as that is due to the additional overhead in
the current interrupt path as well as asynchronous handling of interrupts.

-- 

John Baldwin <[EMAIL PROTECTED]> -- http://www.FreeBSD.org/~jhb/
PGP Key: http://www.baldwin.cx/~john/pgpkey.asc
"Power Users Use the Power to Serve!"  -  http://www.FreeBSD.org/


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message



Re: pcm status

2001-01-20 Thread Mark Murray

OK, please try this, then; change the HARVEST_RING_SIZE in
sys/dev/random/yarrow.h to a smaller power-of-2 (256 or 128 would
be good), and let me know how that helps-or-not.

Thanks!

M

> If it helps at all, it's marginal.  I still get skips.
> 
> On Sat, Jan 20, 2001 at 05:36:08PM +0200, Mark Murray wrote:
> > > This is directly traceable to entropy harvesting by /dev/*random.
> > > I know it's not an option if you need to use ssh, but not loading
> > > random.ko will fix the sound problems when moving the mouse or
> > > typing.  It doesn't fix the 'hwptr went backwards' messages, though.
> > 
> > Please tell me if
> > 
> > sysctl -w kern.random.yarrow.bins=2
> > makes a difference for you.
> > 
> > Thanks!
> > 
> > M
> > -- 
> > Mark Murray
> > Warning: this .sig is umop ap!sdn
> > 
> > 
> > To Unsubscribe: send mail to [EMAIL PROTECTED]
> > with "unsubscribe freebsd-current" in the body of the message
> 
-- 
Mark Murray
Warning: this .sig is umop ap!sdn


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message



Re: pcm status

2001-01-20 Thread Donald J . Maddox

If it helps at all, it's marginal.  I still get skips.

On Sat, Jan 20, 2001 at 05:36:08PM +0200, Mark Murray wrote:
> > This is directly traceable to entropy harvesting by /dev/*random.
> > I know it's not an option if you need to use ssh, but not loading
> > random.ko will fix the sound problems when moving the mouse or
> > typing.  It doesn't fix the 'hwptr went backwards' messages, though.
> 
> Please tell me if
> 
> sysctl -w kern.random.yarrow.bins=2
> makes a difference for you.
> 
> Thanks!
> 
> M
> -- 
> Mark Murray
> Warning: this .sig is umop ap!sdn
> 
> 
> To Unsubscribe: send mail to [EMAIL PROTECTED]
> with "unsubscribe freebsd-current" in the body of the message


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message



Re: pcm status

2001-01-20 Thread Mark Murray

> This is directly traceable to entropy harvesting by /dev/*random.
> I know it's not an option if you need to use ssh, but not loading
> random.ko will fix the sound problems when moving the mouse or
> typing.  It doesn't fix the 'hwptr went backwards' messages, though.

Please tell me if

sysctl -w kern.random.yarrow.bins=2
makes a difference for you.

Thanks!

M
-- 
Mark Murray
Warning: this .sig is umop ap!sdn


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message



Re: pcm status

2001-01-20 Thread Donald J . Maddox

This is directly traceable to entropy harvesting by /dev/*random.
I know it's not an option if you need to use ssh, but not loading
random.ko will fix the sound problems when moving the mouse or
typing.  It doesn't fix the 'hwptr went backwards' messages, though.

On Fri, Jan 19, 2001 at 11:42:32PM -0700, Warner Losh wrote:
> In message <[EMAIL PROTECTED]> Sean Kelly writes:
> : I did not have this behavior in -STABLE, but now I do since I upgraded
> : to -CURRENT.
> 
> -current's interrupt latency is horrible these days, which seems to
> manifest itself in problem with the sound driver.  I have a -current
> laptop and get all kinds of artifacts on it.  I have a laptop that is
> running -stable and has 1/3 the power of the -current laptop and plays
> w/o a hitch.
> 
> Warner
> 
> 
> To Unsubscribe: send mail to [EMAIL PROTECTED]
> with "unsubscribe freebsd-current" in the body of the message


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message



Re: pcm status

2001-01-19 Thread Warner Losh

In message <[EMAIL PROTECTED]> Sean Kelly writes:
: I did not have this behavior in -STABLE, but now I do since I upgraded
: to -CURRENT.

-current's interrupt latency is horrible these days, which seems to
manifest itself in problem with the sound driver.  I have a -current
laptop and get all kinds of artifacts on it.  I have a laptop that is
running -stable and has 1/3 the power of the -current laptop and plays
w/o a hitch.

Warner


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message



pcm status

2001-01-19 Thread Sean Kelly

What is the current status of the pcm driver?  I ask this because I
am able to hear sticking and slowdowns in sound when I move my PS/2
mouse.

pcm0:  at io 0xe800 irq 11 (4p/3r channels duplex)

I did not have this behavior in -STABLE, but now I do since I upgraded
to -CURRENT.

FreeBSD edgemaster 5.0-CURRENT FreeBSD 5.0-CURRENT #0: Mon Jan 15
  02:38:13 CST 2001  root@edgemaster:/usr/obj/usr/src/sys/EDGEMASTER  i386

I also get many of these:
pcm0: hwptr went backwards 2816 -> 2720
pcm0: hwptr went backwards 1760 -> 1568
pcm0: hwptr went backwards 1952 -> 1728
pcm0: hwptr went backwards 0 -> 3872


Other than that, -CURRENT has been very stable (except the time Ioaded vn.ko
and the unloaded it and the kernel exploded, leaving me no useful crashdumps
no less).

-- 
Sean Kelly | PGP KeyID: 77042C7B
[EMAIL PROTECTED] | http://www.zombie.org

For PGP key, send e-mail with subject "send pgp key"


To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message