Re: iBook clamshell audio

Hello.

I enjoyed the test and it worked fine on my iBook.

First I tried with the internal speaker playing WAV file with
audioplay. But there was no sound at all.
Then I set the volume with mixerctl up to 250,250 (default is 80,80).
I could listen to the sound putting my ear on the speaker.
I think it is no good because 250,250 is the maximum level.

Next plugging stereo headphones, it worked with no problems.
Setting the volume to 80,80 again, it was enough to use with my
headphones, and 250,250 was to much.

I could see the message on the shell.
" i2s0: powering down
  dac3550a0: entering low power mode"

I'm looking forward the next time to test.
Thank you.

Masafumi Hotta

monitor.output=speaker  { speaker headphones line }
monitor.master=80,80 volume
record.record=0,0 volume
monitor.treble=128,0 
monitor.bass=128,0 

Re: iBook clamshell audio

Hi,

Thank you very much for your report; the information is very useful to
improve the driver.

> First I tried with the internal speaker playing WAV file with
> audioplay. But there was no sound at all.
> Then I set the volume with mixerctl up to 250,250 (default is 80,80).
> I could listen to the sound putting my ear on the speaker.
> I think it is no good because 250,250 is the maximum level.

I think I know what's the problem with the built-in speaker, it needs
special configuration due to it being a mono device connected to
stereo outputs.

> Next plugging stereo headphones, it worked with no problems.



> Setting the volume to 80,80 again, it was enough to use with my
> headphones, and 250,250 was to much.

Yes, 250 is expected to be a high volume. It corresponds to +18 dB.
The 0 dB point corresponds to value 176, so this should be the highest
possible value that can be used without distortion. The value of 80
corresponds to -36 dB.

Thinking about this, it would be nice if mixerctl(1) could use dB
values instead of the 0-255 scale; this would be good for consistency
between different audio devices.

Re: iBook clamshell audio

Hi,

On 12/1/07, Marco Trillo <marcotrillo <at> gmail.com> wrote:
> > Setting the volume to 80,80 again, it was enough to use with my
> > headphones, and 250,250 was to much.
>
> Yes, 250 is expected to be a high volume. It corresponds to +18 dB.
> The 0 dB point corresponds to value 176, so this should be the highest
> possible value that can be used without distortion.

Oops, that of the 176 value and the 'distortion' should be ignored;
please excuse.

Another thing I forgot to mention is that I have to improve the
interaction between the 'i2s' driver, the codec driver -- in this case
'dac3550a' -- and the mixer layer, so the codec driver can control its
own mixer items instead of having a set of common items for 'i2s'.

For example, the 'bass' and 'treble' items are not relevant for the
dac3350a, since it does not support bass and treble.

Greetings,
Marco

Re: Firefox on recent NetBSDs?

>>>Is the problem with the disappearing menus and randomly
>>>resizing/disappearing elements fixed, then?

>>Looks sane now.

>>>Firefox on macppc is a disaster.

>>The rendering bugs disappeared when Andy replaced the thread library,  so 
>>in 4.0 they're likely still present.

>The workaround for this I've used with NetBSD 3.x is to disable all
>compiler optimizations when building Firefox.  This makes Firefox
>slower, but the glitches go away.  (I had understood the glitches were
>due to compiler bugs in the GCC 3.x series.  I've been meaning to
>upgrade to a 4.0 RC but haven't done so yet, so I can't comment on if
>the newer GCC release fixes any of these troubles.)

This may not be a revelation to anyone at this point, but I've been
using Firefox built with/running on the 4.0 release candidates and I'm
no longer seeing most of the rendering glitches and menu annoyances that
were present when built under the NetBSD 3.x series with the older GCC.
I say "most" because occasionally I do see a weirdness where I have to
click to open a menu more than once before it responds, and the odd time
it makes strange choices about fonts but other than that, it seems fine.

I didn't have to tweak anything for it to behave.  Has anyone else's
experience been different?

Regards,

ki2c.c problem? kernel GENERIC couldn't build

Hi, I can't compile GENERIC kernel of macppc.
src/xsrc keeping newest sources.May be ki2c.c 's bug?
please help me.

Command : $ time ./build -j3 -m macppc tools kernel=GENERIC

Build machine:
/* Athlon64 X2 4800+(2.5GHz; L2=512KB) - 2GB(1+1) DDR RAM(Dual Channel)
  - Asrock AM2NF3-VSTA M/B - 250GB Seagate SATAII 7200rpm
  - Nvidia Riva TNT AGP(4x; 2MB VRAM) - 3.5inch FDD
  - USB2.0 x 6 - UltraATA133 x 2 - SATA x 2
  - Intel 82557 NIC PCI - SymBios 53c870 SCSI /w Boot Rom
  - Texas FireWire TSB12LV26 IEEE 1394 Host Controller
  - Serial(9pin; male) - Parallel(25pin; female) - PS/2 Keyboard/Mouse
  */

--- ki2c.o ---
#   compile  GENERIC/ki2c.o
/home/amink/NetBSD-build/src/../tools/macppc/bin/powerpc--netbsd-gcc 
-mno-strict-align -msoft-float -Wa,-maltivec -ffreestanding 
-fno-zero-initialized-in-bss -O2 -Wreturn-type -Werror -Wall -Wno-main 
-Wno-format-zero-length -Wpointer-arith -Wmissing-prototypes 
-Wstrict-prototypes -Wswitch -Wshadow -Wcast-qual -Wwrite-strings 
-Wno-sign-compare -Wno-pointer-sign -Wno-attributes -fno-strict-aliasing 
-Dmacppc=macppc -I. 
-I/home/amink/NetBSD-build/src/sys/contrib/dev/ath/netbsd 
-I/home/amink/NetBSD-build/src/sys/../common/include 
-I/home/amink/NetBSD-build/src/sys/arch 
-I/home/amink/NetBSD-build/src/sys -nostdinc -DLKM -DZS_CONSOLE_ABORT 
-DFORCE_FUNCTION_KEYS -DRADEONFB_MMAP_BARS -DMAXUSERS=32 -D_KERNEL 

Re: ki2c.c problem? kernel GENERIC couldn't build

Hello,

On Dec 7, 2007, at 08:26, Ryo Chiba wrote:

> Hi, I can't compile GENERIC kernel of macppc.
> src/xsrc keeping newest sources.May be ki2c.c 's bug?
> please help me.

Not a bug, just a pasto. I'll fix that.

have fun
Michael

Re: ki2c.c problem? kernel GENERIC couldn't build

Hello,

On Dec 7, 2007, at 10:37, Michael Lorenz wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hello,
>
> On Dec 7, 2007, at 08:26, Ryo Chiba wrote:
>
>> Hi, I can't compile GENERIC kernel of macppc.
>> src/xsrc keeping newest sources.May be ki2c.c 's bug?
>> please help me.
>
> Not a bug, just a pasto. I'll fix that.

Done. You need to update macppc/dev/ki2cvar.h

have fun
Michael

Re: ki2c.c problem? kernel GENERIC couldn't build

 >>>Done. You need to update macppc/dev/ki2cvar.h

Thanks a lot for fixing the code!

R.C.

XFree86 problem in -current

Temporarily abandoning X.org, I'm trying to use XFree86 from xsrc along
with recent NetBSD/macppc-current (compiled with jemalloc).  However, X
doesn't start, giving this error message:

Fatal server error:
R_PPC_REL24 PLT offset 105748d4 too large

(full log attached)

Has anyone else seen this?  Would it be worth to try recompiling libc
without jemalloc?

Magnus

XFree86 Version 4.5.0
Release Date: 16 March 2005
X Protocol Version 11, Revision 0
Build Operating System:NetBSD/macppc 4.99.40 - The NetBSD Foundation, Inc.
Current Operating System: NetBSD zemdatav 4.99.40 NetBSD 4.99.40 (IBOOK) #0: Fri Dec  7 23:58:02 CET 2007 
magnus <at> zemdatav:/media/e/netbsd/obj/sys/arch/macppc/compile/IBOOK macppc
Build Date: 7 December 2007
	Before reporting problems, check http://www.XFree86.Org/
	to make sure that you have the latest version.
Module Loader present
Command line: X :0
Markers: (--) probed, (**) from config file, (==) default setting,
	(++) from command line, (!!) notice, (EE) error, (WW) warning,

Re: XFree86 problem in -current

Hello,

On Dec 10, 2007, at 12:31, Magnus Henoch wrote:

> Temporarily abandoning X.org, I'm trying to use XFree86 from xsrc  
> along
> with recent NetBSD/macppc-current (compiled with jemalloc).   
> However, X
> doesn't start, giving this error message:
>
> Fatal server error:
> R_PPC_REL24 PLT offset 105748d4 too large
>
> (full log attached)
>
> Has anyone else seen this?

Yes.

> Would it be worth to try recompiling libc without jemalloc?

Yes, works here at least. Someone with more ELF clue than me needs to  
fix the module loader, apparently it relies on some peculiarities of  
the old malloc.

have fun
Michael