PS3 2.6.21 Kernel ebuild / Controller via Bluetooth

andreyua · n00b Joined: 27 May 2007 Posts: 6

I have tried boot with this kernel configuration
http://homepage.ntlworld.com/garlicbread/Projects/PS3Linux/ps3-kernel-build-2.6.21-r1.tar.bz2
but sound is not work.
Any reason ?

mplayer -ao oss shrek.avi
MPlayer dev-SVN-rUNKNOWN-4.1.1 (C) 2000-2007 MPlayer Team
AltiVec found
CPU: PowerPC

Playing shrek.avi.
AVI file format detected.
VIDEO: [DX50] 640x352 24bpp 25.000 fps 971.0 kbps (118.5 kbyte/s)
Clip info:
Software: Nandub v0.26.1
==========================================================================
Opening video decoder: [ffmpeg] FFmpeg's libavcodec codec family
Selected video codec: [ffodivx] vfm: ffmpeg (FFmpeg MPEG-4)
==========================================================================
==========================================================================
Opening audio decoder: [mp3lib] MPEG layer-2, layer-3
AUDIO: 48000 Hz, 2 ch, s16be, 128.0 kbit/8.33% (ratio: 16000->192000)
Selected audio codec: [mp3] afm: mp3lib (mp3lib MPEG layer-2, layer-3)
==========================================================================
[AO OSS] audio_setup: Can't open audio device /dev/dsp: No such file or directory
Could not open/initialize audio device -> no sound.
Audio: no sound
Starting playback...
VDec: vo config request - 640 x 352 (preferred colorspace: Planar YV12)
Movie-Aspect is 1.82:1 - prescaling to correct movie aspect.
[swscaler @ 0x107338ec]ALTIVEC: Color Space ARGB
[swscaler @ 0x107338ec]SwScaler: using unscaled yuv420p -> rgb32 special converter
VO: [fbdev] 640x352 => 640x352 ARGB
V: 12.1 303/303 7% 3% 0.0% 0 0

MPlayer interrupted by signal 2 in module: sleep_timer
localhost Cartoons #

localhost Cartoons # mplayer -ao alsa shrek.avi
MPlayer dev-SVN-rUNKNOWN-4.1.1 (C) 2000-2007 MPlayer Team
AltiVec found
CPU: PowerPC

Playing shrek.avi.
AVI file format detected.
VIDEO: [DX50] 640x352 24bpp 25.000 fps 971.0 kbps (118.5 kbyte/s)
Clip info:
Software: Nandub v0.26.1
==========================================================================
Opening video decoder: [ffmpeg] FFmpeg's libavcodec codec family
Selected video codec: [ffodivx] vfm: ffmpeg (FFmpeg MPEG-4)
==========================================================================
==========================================================================
Opening audio decoder: [mp3lib] MPEG layer-2, layer-3
AUDIO: 48000 Hz, 2 ch, s16be, 128.0 kbit/8.33% (ratio: 16000->192000)
Selected audio codec: [mp3] afm: mp3lib (mp3lib MPEG layer-2, layer-3)
==========================================================================
[AO_ALSA] alsa-lib: confmisc.c:848:(snd_func_card_driver) cannot find card '0'
[AO_ALSA] alsa-lib: conf.c:3500:(_snd_config_evaluate) function snd_func_card_driver returned error: No such device
[AO_ALSA] alsa-lib: confmisc.c:397:(snd_func_concat) error evaluating strings
[AO_ALSA] alsa-lib: conf.c:3500:(_snd_config_evaluate) function snd_func_concat returned error: No such device
[AO_ALSA] alsa-lib: confmisc.c:1248:(snd_func_refer) error evaluating name
[AO_ALSA] alsa-lib: conf.c:3500:(_snd_config_evaluate) function snd_func_refer returned error: No such device
[AO_ALSA] alsa-lib: conf.c:3972:(snd_config_expand) Evaluate error: No such device
[AO_ALSA] alsa-lib: pcm.c:2145:(snd_pcm_open_noupdate) Unknown PCM default
[AO_ALSA] Playback open error: No such device
Could not open/initialize audio device -> no sound.
Audio: no sound
Starting playback...
VDec: vo config request - 640 x 352 (preferred colorspace: Planar YV12)
Movie-Aspect is 1.82:1 - prescaling to correct movie aspect.
[swscaler @ 0x107338ec]ALTIVEC: Color Space ARGB
[swscaler @ 0x107338ec]SwScaler: using unscaled yuv420p -> rgb32 special converter
VO: [fbdev] 640x352 => 640x352 ARGB
V: 1.9 49/ 49 5% 4% 0.0% 0 0

MPlayer interrupted by signal 2 in module: sleep_timer
localhost Cartoons #

garlicbread · Apprentice Joined: 06 Mar 2004 Posts: 182

it sounds like your trying to play sound using oss but haven't enabled oss in the kernel config
Instead of using oss use alsa
i.e.

andreyua · n00b Joined: 27 May 2007 Posts: 6

garlicbread · Apprentice Joined: 06 Mar 2004 Posts: 182

if you've compiled alsa as a module
on 2.6.21 and above there should be a file
/lib/modules/<kernel version>/kernel/sound/ppc/snd_ps3.ko
make sure this exists

if you've compiled directly into the kernel then I'm not sure but

Shadow Skill · Veteran Joined: 04 Dec 2004 Posts: 1023

Does anyone here know if it is possible to get a bluetooth keyboard to automatically connect using the internal bluetooth controller of the PS3 when the machine starts up so that one may type at the kboot prompt using said keyboard? I can't tell you how annoying it is to have to steal my desktop keyboard to type at the kboot prompt if I want to switch to the gameos.
_________________
Ware wa mutekinari.
Wa ga kage waza ni kanau mono nashi.
Wa ga ichigeki wa mutekinari.

"First there was nothing, so the lord gave us light. There was still nothing, but at least you could see it."

garlicbread · Apprentice Joined: 06 Mar 2004 Posts: 182

since the kboot image only contains the basics to boot it's unlikley
you'd probably have to custom graft your own kboot image with bluetooth support to get it to work
and I don't know much about grafting kboot images

in theory the kernel during normal operation should support switching to the game os
but it looks like this is another problem with the current version, so hopefully this will be fixed sometime in the future

in the mean time I've got one of those mini usb keyboards

http://www.maplin.co.uk/Module.aspx?ModuleNo=47313&criteria=USB%20keyboard&doy=28m5

Crymson · Posted: Tue May 29, 2007 5:02 am Post subject:

Ok, I more or less got the kernel upgraded.

I used the instructions above for using the latest git clone, rather than git --bare.

I then did the steps garlicbread outlined earlier - make modules && make modules_install.

I also tried to do genkernel initrd, but it failed when compiling busybox. Not sure why, but after changing the lines in /etc/kboot.conf and /etc/fstab I was able to boot in to the new kernel.

I'm now (slowly) on my way to building gnome.

I did try to emerge snes9x, and it's telling me it's masked because of a missing keyword? What keyword am I missing, does anyone know? I won't be able to do anything with it until the rest of this stuff finishes compiling, but I'd like to know ahead of time.

Also - to get the new kernel (2.6.22-rc2 is what I used) compiled, I had to use make ps3_defconfig, then manually edit the .config file to include the joystick support. Then it compiled with no issues.

Sorry for the scatterbrained post. I'll update as I get further along.
_________________
Knowledge is Power // Power Corrupts // Study Hard // Be Evil

RaZoR1394 · Guru Joined: 09 Jan 2005 Posts: 356

I'm now running 2.6.22-rc1 from garlicbread's layman overlay. I used genkernel and it worked this time. I noticed one error about a kernel(?) file when booting:

garlicbread · Apprentice Joined: 06 Mar 2004 Posts: 182

I think alsa was disabled in the stage 4 by default as 2.6.16 had originally issues with alsa and you had to use oss instead originally

for snes9x the default ebuilds in the gentoo tree lack the ppc64 keyword
in the overlay I added in the ppc64 keyword and used the more up-to-date version of snes9x that was 64bit compatible (see bugzilla for the original ebuild
https://bugs.gentoo.org/show_bug.cgi?id=139347
it looks like 1.51 is out, I've not tried this one yet so I'm not sure of the differences

the initird doesn't appear to be needed at this point unless you've selected anything critical to be compiled as a module
I think with 2.6.16 it may have been the case that at least one thing to do with storage had to be a module
which was the original reason I think as to why the initrd was needed

genkernel initrd
probably isn't working since it uses loopback fs
and since they've changed across to a dynamic type setup, you'll need to create /dev/loop0 manually to get it to work (see post above)

Crymson · Posted: Tue May 29, 2007 4:58 pm Post subject:

Thanks garlicbread. Once I get gnome installed, I'll get back to using the overlay to get snes9x installed.

Also, one additional change I had to do. In /etc/fstab, I had to change /dev/sda2 to /dev/ps3da2 for swap to work. I also saw /dev/ps3flash, but I'm not sure which device that maps to. Again, more as I play with it.
_________________
Knowledge is Power // Power Corrupts // Study Hard // Be Evil

garlicbread · Apprentice Joined: 06 Mar 2004 Posts: 182

to be honest I'd hold off doing anything related to the flash (such as boot-game-os) until we have a more stable kernel
it's safer just to run it from the kboot prompt instead

I haven't got time tonight but I think it should be possible to setup an ebuild that uses the git eclass
to automate the git-clone etc via emerge
I might have a go with dtc / and the PS3 git tree when I've got time

Crymson · Posted: Wed May 30, 2007 1:57 am Post subject:

I don't mean the flash for booting in to the game OS, I figured this was for the flash drives. Are /dev/sdc and /dev/sdd still valid for those (the compact flash reader, etc)? Since the hard drive was /dev/ps3da1, and swap became /dev/ps3da2 I was just thinking the rest of the devices might have changed too.

I dropped off the ~ppc64 branch, since a lot of those ebuilds are broken. I'm recompiling a bunch of stuff on just regular ppc64, and it seems to be going OK.

I can't wait to get Gnome done, so I can start playing with wireless, bluetooth, and the emus.
_________________
Knowledge is Power // Power Corrupts // Study Hard // Be Evil

garlicbread · Apprentice Joined: 06 Mar 2004 Posts: 182

from what I've observed the flash readers (CF / SD etc) seem have stayed using the sdx devices
it's just that originally the HD was sda but now sda will be one of the flash drives instead

only the PS3 hard disk has got the new name from the looks of things
I think something like "dmesg | grep sd" may give some clues as to which sd devices relate to which flash ports
another idea maybe try using "udevinfo -a -p /sys/block/sdx" etc to identify which drive is which

andreyua · n00b Joined: 27 May 2007 Posts: 6

I have a trouble, when compile net-wireless/bluez-hcidump-1.30-r1

>>> Unpacking source...
>>> Unpacking bluez-hcidump-1.30.tar.gz to /var/tmp/portage/net-wireless/bluez-hcidump-1.30-r1/work
>>> Source unpacked.
>>> Compiling source in /var/tmp/portage/net-wireless/bluez-hcidump-1.30-r1/work/bluez-hcidump-1.30 ...
./configure --prefix=/usr --host=powerpc64-unknown-linux-gnu --mandir=/usr/share/man --infodir=/usr/share/info --datadir=/usr/share --sysconfdir=/etc --localstatedir=/var/lib --libdir=/usr/lib64 --build=powerpc64-unknown-linux-gnu
checking for a BSD-compatible install... /usr/bin/install -c
checking whether build environment is sane... yes
checking for gawk... gawk
checking whether make sets $(MAKE)... yes
checking whether to enable maintainer-specific portions of Makefiles... no
checking for powerpc64-unknown-linux-gnu-gcc... powerpc64-unknown-linux-gnu-gcc
checking for C compiler default output file name... a.out
checking whether the C compiler works... yes
checking whether we are cross compiling... no
checking for suffix of executables...
checking for suffix of object files... o
checking whether we are using the GNU C compiler... yes
checking whether powerpc64-unknown-linux-gnu-gcc accepts -g... yes
checking for powerpc64-unknown-linux-gnu-gcc option to accept ANSI C... none needed
checking for style of include used by make... GNU
checking dependency style of powerpc64-unknown-linux-gnu-gcc... gcc3
checking whether powerpc64-unknown-linux-gnu-gcc accepts -fPIE... yes
checking for a BSD-compatible install... /usr/bin/install -c
checking for powerpc64-unknown-linux-gnu-ranlib... powerpc64-unknown-linux-gnu-ranlib
checking how to run the C preprocessor... powerpc64-unknown-linux-gnu-gcc -E
checking for egrep... grep -E
checking for ANSI C header files... yes
checking for sys/types.h... yes
checking for sys/stat.h... yes
checking for stdlib.h... yes
checking for string.h... yes
checking for memory.h... yes
checking for strings.h... yes
checking for inttypes.h... yes
checking for stdint.h... yes
checking for unistd.h... yes
checking bluetooth/bluetooth.h usability... yes
checking bluetooth/bluetooth.h presence... yes
checking for bluetooth/bluetooth.h... yes
checking for hci_open_dev in -lbluetooth... yes
checking for sdp_connect in -lbluetooth... yes
configure: creating ./config.status
config.status: creating Makefile
config.status: creating src/Makefile
config.status: creating parser/Makefile
config.status: creating config.h
config.status: executing depfiles commands
make all-recursive
make[1]: Entering directory `/var/tmp/portage/net-wireless/bluez-hcidump-1.30-r1/work/bluez-hcidump-1.30'
Making all in parser
make[2]: Entering directory `/var/tmp/portage/net-wireless/bluez-hcidump-1.30-r1/work/bluez-hcidump-1.30/parser'
if powerpc64-unknown-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I. -I.. -I/usr/include -O2 -pipe -mcpu=970 -mtune=970 -mabi=altivec -D_FORTIFY_SOURCE=2 -MT parser.o -MD -MP -MF ".deps/parser.Tpo" \
-c -o parser.o `test -f 'parser.c' || echo './'`parser.c; \
then mv -f ".deps/parser.Tpo" ".deps/parser.Po"; \
else rm -f ".deps/parser.Tpo"; exit 1; \
fi
if powerpc64-unknown-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I. -I.. -I/usr/include -O2 -pipe -mcpu=970 -mtune=970 -mabi=altivec -D_FORTIFY_SOURCE=2 -MT lmp.o -MD -MP -MF ".deps/lmp.Tpo" \
-c -o lmp.o `test -f 'lmp.c' || echo './'`lmp.c; \
then mv -f ".deps/lmp.Tpo" ".deps/lmp.Po"; \
else rm -f ".deps/lmp.Tpo"; exit 1; \
fi
if powerpc64-unknown-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I. -I.. -I/usr/include -O2 -pipe -mcpu=970 -mtune=970 -mabi=altivec -D_FORTIFY_SOURCE=2 -MT hci.o -MD -MP -MF ".deps/hci.Tpo" \
-c -o hci.o `test -f 'hci.c' || echo './'`hci.c; \
then mv -f ".deps/hci.Tpo" ".deps/hci.Po"; \
else rm -f ".deps/hci.Tpo"; exit 1; \
fi
hci.c: In function ×Ђ˜sniff_subrate_dump×Ђ™:
hci.c:748: error: ×Ђ˜sniff_subrate_cp×Ђ™ has no member named ×Ђ˜max_remote_latency×Ђ™
hci.c:748: error: ×Ђ˜sniff_subrate_cp×Ђ™ has no member named ×Ђ˜max_local_latency×Ђ™
hci.c: In function ×Ђ˜write_link_supervision_timeout_dump×Ђ™:
hci.c:965: error: ×Ђ˜write_link_supervision_timeout_cp×Ђ™ has no member named ×Ђ˜link_sup_to×Ђ™
hci.c: In function ×Ђ˜read_link_supervision_timeout_dump×Ђ™:
hci.c:1490: error: ×Ђ˜read_link_supervision_timeout_rp×Ђ™ has no member named ×Ђ˜link_sup_to×Ђ™
hci.c: In function ×Ђ˜sniff_subrate_event_dump×Ђ™:
hci.c:2384: error: ×Ђ˜evt_sniff_subrate×Ђ™ has no member named ×Ђ˜max_remote_latency×Ђ™
hci.c:2384: error: ×Ђ˜evt_sniff_subrate×Ђ™ has no member named ×Ђ˜max_local_latency×Ђ™
hci.c: In function ×Ђ˜sco_dump×Ђ™:
hci.c:2618: warning: ignoring return value of ×Ђ˜write×Ђ™, declared with attribute warn_unused_result
make[2]: *** [hci.o] Error 1
make[2]: Leaving directory `/var/tmp/portage/net-wireless/bluez-hcidump-1.30-r1/work/bluez-hcidump-1.30/parser'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/var/tmp/portage/net-wireless/bluez-hcidump-1.30-r1/work/bluez-hcidump-1.30'
make: *** [all] Error 2

!!! ERROR: net-wireless/bluez-hcidump-1.30-r1 failed.
Call stack:
ebuild.sh, line 1615: Called dyn_compile
ebuild.sh, line 972: Called qa_call 'src_compile'
ebuild.sh, line 44: Called src_compile
bluez-hcidump-1.30-r1.ebuild, line 19: Called die

!!! emake failed
!!! If you need support, post the topmost build error, and the call stack if relevant.
!!! A complete build log is located at '/var/tmp/portage/net-wireless/bluez-hcidump-1.30-r1/temp/build.log'.

garlicbread · Apprentice Joined: 06 Mar 2004 Posts: 182

you don't really need hcidump to get bluetooth working, it can just be used as a test
I've noticed "powerpc64-unknown-linux-gnu-gcc" above
are you using a 32bit or 64bit system? I've said before I've only tried this on 64bit so far

andreyua · n00b Joined: 27 May 2007 Posts: 6

I'm using 64 bit system and kernel 2.6.21-ps3-r2 version

How I can test bluetoth work on sisxaxis?

And, several question.
1. What better ? Alsa in modules or Alsa compiled in kernel ?
2. When I try emerge mplayer:

Crymson · Posted: Fri Jun 01, 2007 4:10 am Post subject:

I'm having some issues compiling snes9x. I'm using the 1.50 ebuild in garlicbreads overlay, which won't compile on its own. I had to create another patch that would fix the extern problems (some things were externed "C", and duplicated in C++ so I had to fix that in global.cpp)

In any case, now my hard drive is just spinning a lot, and it's stuck on compiling the /unix/X11 files.

I also tried to compile the source for 1.51 taken from the official snes9x site, but it hung on the very FIRST file.

Is there something in my USE flags that I should have set? Since garlicbread didn't mention needing another patch for 1.50, I think something isn't right.

Any insight in to this? Thanks!
_________________
Knowledge is Power // Power Corrupts // Study Hard // Be Evil

Crymson · Posted: Fri Jun 01, 2007 7:20 pm Post subject:

Hey all, sorry to be such a burden - now I'm having issues with wireless.

My router is set up to use WPA-PSK/TKIP, and I was looking at the config for /etc/conf.d/net and saw the commands for WPA, but when I try to restart eth0, it says that it isn't recognized. So, I tried entering the WPA password as a WEP key, and then it says it connects, but it really doesn't.

I get the following:

Crymson · Posted: Fri Jun 01, 2007 11:05 pm Post subject:

More on wireless.

I tried to use wpa_supplicant, and I think it will work, except the light for wifi never turns on. If I use iwconfig, it'll turn on, but then I'm limited to WEP, and as I said earlier, it won't go anywhere.

Has anyone gotten wpa_supplicant to work? I'm trying to use the -D wext driver, but that doesn't seem to be correct. madwifi doesn't work either.

This is so frustrating.

I'm using the 2.6.22-rc1 kernel from the git tree. Everything else is working great.

Oh, side note - you *can* use the controller when it's plugged in via USB - just press the PS button. When running jstest, I was getting data from the button pushes and whatnot.

So, any help with my snes9x issue, and the wifi would be very helpful.

Thanks again.
_________________
Knowledge is Power // Power Corrupts // Study Hard // Be Evil

Crymson · Posted: Sat Jun 02, 2007 6:34 am Post subject:

Wireless:

wpa_supplicant does not have the driver (ps3_gelic) to work properly. So, that means you either have to be running WEP, or no security at all. WPA will not work at this time.

I ended up just unsecuring my network, because WEP acts strangely with my other computers. I wish I could have gotten WPA to work, but alas.

Here's my /etc/conf.d/net file if someone else wants to use it.

garlicbread · Apprentice Joined: 06 Mar 2004 Posts: 182

that's an odd error regarding snes9x
I didn't actually patch the code for this one, all I did was add a small change to the ebuild
1. add the ppc64 keyword so we can unmask it
2. only make "offsets" if it's not the ppc64 arch
this bit

garlicbread · Apprentice Joined: 06 Mar 2004 Posts: 182

just spotted the problem you were talking about for snes9x-1.50 about globals.cpp
I believe the solution is to turn off distcc in /etc/make.conf under features

for 1.50 this has an issue with distcc being enabled
for 1.51 this has no problem with distcc being enabled (although I'm not sure if it uses it)

for 1.51 it appears to work but the sounds a bit crappy
I think it has something to do with someone setting the default output to 32Khz or so
it uses /dev/dsp by default which I believe is the old oss interface
so unless there's an option I've missed somewhere to use alsa instead or change the sampling rate
I'll have to keep fiddling with the settings

Pizza67 · n00b Joined: 19 Mar 2007 Posts: 6

Hi all!

I'm trying to make the BT remote works with Freevo or Mythv on my PS3.
I'm currently using 2.6.21 garlicbread overlay (thanks man!) and I successfully paired the remote with the PS3. Unfortunately no key events are found in evtest.

Some news about this problem?

Anybody have tried this?

http://apebox.org/modules/code/cakemote.pie (Rename in .py)

It returns an error:

garlicbread · Apprentice Joined: 06 Mar 2004 Posts: 182

I've already put another post further up on this one on the 2nd page, but for info
you won't get anything through the evtest untill there's been an update with the kernel or bluez-utils

at the moment when you hit the sync button on the remote
a connection is established remote -> PS3 to the running hidd service (if you've switched it on for the controller)
but hidd can't handle the odd data packet from the remote so nothing shows up through evtest

the existing python scripts I've seen do it the other way round instead (don't use the sync button on the remote, or the event device)
connection from the PS3 -> Remote control
this way the script controls the pairing and the type of connection made (I think L2CAPP or some kind of raw data connection) and the script interprets the meaning of the data itself
One thing they tend not to mention is that when your trying to connect from a device such as the PS3 to the remote (instead of the remote to the PS3) you need to hold down the Start / Enter buttons on the remote to allow the script to find / pair with the device, otherwise the remote is hidden from an outside point of view

also in the script I think you have to manually enter in the BT address instead of auto discover (if it's the same one I've used before)

it's not an ideal way of doing it but it does work for now
a more ideal way would be if the running hidd service could make sense of the incomming data
that way everything could be seen through the /dev/input/event entry
and the only thing you'd need to do would be to hit the sync button on the remote instead of messing around with a script
but we'll need to wait for ether the kernel or bluez-utils to be updated before that happens