gc-linux-devel — Development-related discussion.

[Gc-linux-devel] How was the GameCube hacked history?

[Louise H. <lou...@he...>]

Dear readers

I would very much like to read how the GameCube was hacked, but I
haven't been able to find anything detailed with Google.

Wikipedia had a short description, but I was wondering, if there excist
a history like the one Bunny wrote for the XBox? =)

Lots of love,
Louise

Re: [Gc-linux-devel] GC Cluster

[Adam T. <ad...@io...>]

On Dec 9, 2005, at 12:55 PM, richard klecker wrote:

> Does anyone have any info about the possiblity of doing a GC  
> Cluster?  I am wondering because this has not been done yet as far  
> as I can tell.  Thanx

I've chatted about it; there seems to be no technical reason you  
can't do it, but the difficulty of booting into Linux, and the  
limitation of a single network interface, would make it more trouble  
than I think it would be worth.

Adam

[Gc-linux-devel] GC Cluster

[richard k. <hyp...@ya...>]

  Does anyone have any info about the possiblity of doing a GC Cluster?  I am wondering because this has not been done yet as far as I can tell.  Thanx


			
---------------------------------
Yahoo! Shopping
 Find Great Deals on Holiday Gifts at Yahoo! Shopping 

Re: [Gc-linux-devel] Alive?

[Damian G. <dg...@af...>]

Thus spake Shaun Jackman (sja...@gm...) [05/11/05 13:25]:
: 2005/11/4, Damian Gerow <dg...@af...>:
: > Thus spake Shaun Jackman (sja...@gm...) [04/11/05 09:58]:
: > : > Hrm..  I've never heard of that.  I'll give it a shot, though I doubt it
: > : > will help me out much.
: > :
: > : Well, it will prevent the out-of-memory killer from slaying your
: > : processes. Whether the performance is sufficient for your needs is
: > : another matter, but I suspect you'll be pleased.
: >
: > Uh...  How will setting the priority on my only swap device change anything
: > from the current configuration?  I'm already using it as swap; I don't
: > understand how setting its priority will help me out.  If I'm out of memory,
: > I'm out of memory.
: 
: Reread my reply; it was in reference to using a NBD swap device, not
: setting the swap priority.

I did.

I said I'd never heard of setting a swap priority.  You said it will prevent
the out of memory killer.  Hence my confusion.

I'll try the NBD, as that will likely help me out.

Re: [Gc-linux-devel] Alive?

[Shaun J. <sja...@gm...>]

2005/11/4, Damian Gerow <dg...@af...>:
> Thus spake Shaun Jackman (sja...@gm...) [04/11/05 09:58]:
> : > Hrm..  I've never heard of that.  I'll give it a shot, though I doubt=
 it
> : > will help me out much.
> :
> : Well, it will prevent the out-of-memory killer from slaying your
> : processes. Whether the performance is sufficient for your needs is
> : another matter, but I suspect you'll be pleased.
>
> Uh...  How will setting the priority on my only swap device change anythi=
ng
> from the current configuration?  I'm already using it as swap; I don't
> understand how setting its priority will help me out.  If I'm out of memo=
ry,
> I'm out of memory.

Reread my reply; it was in reference to using a NBD swap device, not
setting the swap priority.

> So I presume the image provided does *not* use uclibc.  I have to get a
> Linux system going to build stuff on first, then I'll work on getting a
> uclibc-based image.
>
> Thanks for the pointers!

No problem! Good luck and cheers,
Shaun

Re: [Gc-linux-devel] Alive?

[daniel t. <da...@re...>]

Quoting Damian Gerow <dg...@af...>:
> Thus spake Shaun Jackman (sja...@gm...) [04/11/05 09:58]:
> : > Hrm..  I've never heard of that.  I'll give it a shot, though I doubt it
> : > will help me out much.
> : 
> : Well, it will prevent the out-of-memory killer from slaying your
> : processes. Whether the performance is sufficient for your needs is
> : another matter, but I suspect you'll be pleased.
> 
> Uh...  How will setting the priority on my only swap device change anything
> from the current configuration?  I'm already using it as swap; I don't
> understand how setting its priority will help me out.  If I'm out of memory,
> I'm out of memory.

Setting the priority of your only swap device will not alter anything. However 
if you configure swap to a network block device it will help preserve 
performance by prefering to store pages in ARAM rather than via the network.

I use both on my GC and the only time I really see NBD being used is either 
during compiling (which takes for ever - much better to cross compile!) or 
using apt-get.

I added a few notes to the wiki regarding NBD but have not finished it. It 
tells you how to set up the GC but not how to set up the NBD server. Since I 
haven't quite figured how the wiki is currently laid out I created a new 
node 'Documentation' and linked to my new material from there. This made a lot 
more sense to me than the current 'How to develop/run' node which contains 
links to pretty much all the actual (software) documentation (but I only 
realised this after adding my material.

-- 
Daniel Thompson (Merlin) <da...@re...>

If at first you don't succeed then sky diving is probably not for you.

Re: [Gc-linux-devel] Alive?

[Damian G. <dg...@af...>]

Thus spake Shaun Jackman (sja...@gm...) [04/11/05 09:58]:
: > Hrm..  I've never heard of that.  I'll give it a shot, though I doubt it
: > will help me out much.
: 
: Well, it will prevent the out-of-memory killer from slaying your
: processes. Whether the performance is sufficient for your needs is
: another matter, but I suspect you'll be pleased.

Uh...  How will setting the priority on my only swap device change anything
from the current configuration?  I'm already using it as swap; I don't
understand how setting its priority will help me out.  If I'm out of memory,
I'm out of memory.

: I ran a Debian distro on my GameCube, which uses glibc. I don't know
: of a uClibc [1] distro for the GameCube, but it would be a good fit
: along with Busybox [2]. You could trying building your own uClibc
: rootfs using Buildroot [3]. uClibc's goal is to be a drop-in
: replacement for glibc. It may not meet this goal 100%, but its
: coverage is very good.

So I presume the image provided does *not* use uclibc.  I have to get a
Linux system going to build stuff on first, then I'll work on getting a
uclibc-based image.

Thanks for the pointers!

  - Damian

Re: [Gc-linux-devel] Alive?

[Shaun J. <sja...@gm...>]

Hello Damian,

2005/11/4, Damian Gerow <dg...@af...>:
>> The other technique is to swap to a network block device (NBD) but I
>> haven't any canned instructions for that so you'll have to google to
>> figure it out.
>
> Hrm..  I've never heard of that.  I'll give it a shot, though I doubt it
> will help me out much.

Well, it will prevent the out-of-memory killer from slaying your
processes. Whether the performance is sufficient for your needs is
another matter, but I suspect you'll be pleased.

> That being said... do you know if the system is using uclibc or standard
> glibc?  Using uclibc would be a great way to minimize memory footprint of
> all libc-linked applications.  And is, IIRC, functionally a 1:1 replaceme=
nt
> for glibc (though feel free to correct me if I'm wrong).

I ran a Debian distro on my GameCube, which uses glibc. I don't know
of a uClibc [1] distro for the GameCube, but it would be a good fit
along with Busybox [2]. You could trying building your own uClibc
rootfs using Buildroot [3]. uClibc's goal is to be a drop-in
replacement for glibc. It may not meet this goal 100%, but its
coverage is very good.

Cheers,
Shaun

[1] http://uclibc.org
[2] http://busybox.net
[2] http://buildroot.uclibc.org

Re: [Gc-linux-devel] Alive?

[Damian G. <dg...@af...>]

Thus spake Daniel Thompson (da...@re...) [04/11/05 06:57]:
: Two good techniques to increase available memory are to add swap
: devices.

... Especially if said swap devices are RAM.  :)

: The first one permits swap to 'audio RAM'. In addition to its 24M of
: system RAM the GameCube contains 16M that is normally used to hold audio
: data buffers. The CPU cannot address the audio RAM directly so this
: memory cannot be used like normal system memory but it is possible to
: use the audio RAM as a swap device.

Already done.  When I hit my errors, I've maxed my memory, and run out of
swap.

Note that this process is outlined in the wiki.

: /dev/aram       swap            swap    pri=32767       0       0
: 
: Note: pri=32767 ensures that the ARAM is the kernel's preferred swap
: device, that is pages will be stored in ARAM in preference to other swap
: devices such as via ethernet. At present there is no other hardware for
: a GameCube that offers more performant swap device so giving it maximal
: priority really does make sense.

That, however, is not -- I'll throw it in, though it won't make a difference
in my case.

: The other technique is to swap to a network block device (NBD) but I
: haven't any canned instructions for that so you'll have to google to
: figure it out.

Hrm..  I've never heard of that.  I'll give it a shot, though I doubt it
will help me out much.

That being said... do you know if the system is using uclibc or standard
glibc?  Using uclibc would be a great way to minimize memory footprint of
all libc-linked applications.  And is, IIRC, functionally a 1:1 replacement
for glibc (though feel free to correct me if I'm wrong).

  - Damian

Re: [Gc-linux-devel] Alive?

[Daniel T. <da...@re...>]

On Fri, 2005-11-04 at 06:29 -0500, Damian Gerow wrote:
> I realize this is more of a dev list, but I've got a question: I recently
> got Linux booting on my GC, but I can't do much with the system, as it
> continuously runs out of memory, and starts killing processes.  Which, as=
 it
> happens, are generally the processes I'm trying to run.

Two good techniques to increase available memory are to add swap
devices.

The first one permits swap to 'audio RAM'. In addition to its 24M of
system RAM the GameCube contains 16M that is normally used to hold audio
data buffers. The CPU cannot address the audio RAM directly so this
memory cannot be used like normal system memory but it is possible to
use the audio RAM as a swap device.

Create a device node:

mknod /dev/aram b 37 0

Add the following line to /etc/fstab:

/dev/aram       swap            swap    pri=3D32767       0       0

Note: pri=3D32767 ensures that the ARAM is the kernel's preferred swap
device, that is pages will be stored in ARAM in preference to other swap
devices such as via ethernet. At present there is no other hardware for
a GameCube that offers more performant swap device so giving it maximal
priority really does make sense.

Finally after each reboot (this could be integrated into the boot
scripts):

mkswap /dev/aram
swapon -a

The other technique is to swap to a network block device (NBD) but I
haven't any canned instructions for that so you'll have to google to
figure it out.

--=20
Daniel Thompson (Merlin) <da...@re...>
signature.asc? http://www.redfelineninja.dsl.pipex.com/signature.html

Did Sigmund's wife wear Freudian slips?

[Gc-linux-devel] Alive?

[Damian G. <dg...@af...>]

Anyone alive?  I've not seen much activity here since I signed up (all of
two-three days ago).

I realize this is more of a dev list, but I've got a question: I recently
got Linux booting on my GC, but I can't do much with the system, as it
continuously runs out of memory, and starts killing processes.  Which, as it
happens, are generally the processes I'm trying to run.

It doesn't *look* like it, so I'm hoping someone here can confirm: does the
root image provided for download on gc-linux.org use the uclibc?  It looks
like it's using the standard libc, but it's been some time since I've used
Debian.

(If not, I'd like to see about swapping the image over to uclibc, as that
will likely help with the limited available memory.)

  - Damian

[Gc-linux-devel] [PATCH 2/2] ppc32: Add cputable entry for 750CXe DD2.4 ("Gekko")

[<a.o...@bl...>]

Add a table entry for 750CXe DD2.4 ("Gekko") as found in the GameCube
from Nintendo:

  http://www-306.ibm.com/chips/techlib/techlib.nsf/techdocs/291C8D0EF3EAEC1687256B72005C745C#C1

Signed-off-by: Arthur Othieno <a.o...@bl...>


 cputable.c |   14 ++++++++++++++
 1 files changed, 14 insertions(+)

diff -uprN a/arch/ppc/kernel/cputable.c b/arch/ppc/kernel/cputable.c
--- a/arch/ppc/kernel/cputable.c	2005-08-30 18:18:28.000000000 -0400
+++ b/arch/ppc/kernel/cputable.c	2005-08-30 19:21:15.000000000 -0400
@@ -240,6 +240,20 @@ struct cpu_spec	cpu_specs[] = {
 		.num_pmcs		= 4,
 		.cpu_setup		= __setup_cpu_750cx
 	},
+	{	/* 750CXe "Gekko" (83214) */
+		.pvr_mask		= 0xffffffff,
+		.pvr_value		= 0x00083214,
+		.cpu_name		= "750CXe",
+		.cpu_features		= CPU_FTR_COMMON |
+			CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_MAYBE_CAN_DOZE |
+			CPU_FTR_USE_TB | CPU_FTR_L2CR | CPU_FTR_TAU |
+			CPU_FTR_HPTE_TABLE | CPU_FTR_MAYBE_CAN_NAP,
+		.cpu_user_features	= COMMON_PPC,
+		.icache_bsize		= 32,
+		.dcache_bsize		= 32,
+		.num_pmcs		= 4,
+		.cpu_setup		= __setup_cpu_750cx
+	},
 	{	/* 745/755 */
 		.pvr_mask		= 0xfffff000,
 		.pvr_value		= 0x00083000,

[Gc-linux-devel] [PATCH 1/2] ppc32: Re-order cputable for 750CXe DD2.4 entry

[<a.o...@bl...>]

"745/755" (pvr_value:0x00083000) is a catch-all entry.
Since arch/ppc/kernel/misc.S:identify_cpu() returns on first match,
move this lower in the table so 750CXe DD2.4 (pvr_value:0x00083214)
may be correctly enumerated.

Signed-off-by: Arthur Othieno <a.o...@bl...>


 cputable.c |   28 ++++++++++++++--------------
 1 files changed, 14 insertions(+), 14 deletions(-)

diff -uprN a/arch/ppc/kernel/cputable.c b/arch/ppc/kernel/cputable.c
--- a/arch/ppc/kernel/cputable.c	2005-08-30 15:29:31.000000000 -0400
+++ b/arch/ppc/kernel/cputable.c	2005-08-30 18:18:28.000000000 -0400
@@ -198,20 +198,6 @@ struct cpu_spec	cpu_specs[] = {
 		.num_pmcs		= 4,
 		.cpu_setup		= __setup_cpu_750
 	},
-	{	/* 745/755 */
-		.pvr_mask		= 0xfffff000,
-		.pvr_value		= 0x00083000,
-		.cpu_name		= "745/755",
-		.cpu_features		= CPU_FTR_COMMON |
-			CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_MAYBE_CAN_DOZE |
-			CPU_FTR_USE_TB | CPU_FTR_L2CR | CPU_FTR_TAU |
-			CPU_FTR_HPTE_TABLE | CPU_FTR_MAYBE_CAN_NAP,
-		.cpu_user_features	= COMMON_PPC,
-		.icache_bsize		= 32,
-		.dcache_bsize		= 32,
-		.num_pmcs		= 4,
-		.cpu_setup		= __setup_cpu_750
-	},
 	{	/* 750CX (80100 and 8010x?) */
 		.pvr_mask		= 0xfffffff0,
 		.pvr_value		= 0x00080100,
@@ -254,6 +240,20 @@ struct cpu_spec	cpu_specs[] = {
 		.num_pmcs		= 4,
 		.cpu_setup		= __setup_cpu_750cx
 	},
+	{	/* 745/755 */
+		.pvr_mask		= 0xfffff000,
+		.pvr_value		= 0x00083000,
+		.cpu_name		= "745/755",
+		.cpu_features		= CPU_FTR_COMMON |
+			CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_MAYBE_CAN_DOZE |
+			CPU_FTR_USE_TB | CPU_FTR_L2CR | CPU_FTR_TAU |
+			CPU_FTR_HPTE_TABLE | CPU_FTR_MAYBE_CAN_NAP,
+		.cpu_user_features	= COMMON_PPC,
+		.icache_bsize		= 32,
+		.dcache_bsize		= 32,
+		.num_pmcs		= 4,
+		.cpu_setup		= __setup_cpu_750
+	},
 	{	/* 750FX rev 1.x */
 		.pvr_mask		= 0xffffff00,
 		.pvr_value		= 0x70000100,

[Gc-linux-devel] [PATCH 0/2] ppc32: Add cputable entry for 750CXe DD2.4

[<a.o...@bl...>]

Greetings,

This adds a cputable entry for 750CXe DD2.4, as found in the
GameCube from Nintendo. To be correctly enumerated, however,
the table has to be slightly re-ordered first.

Details within. Thanks.

	Arthur

[Gc-linux-devel] Re: bit-banging memory card reader

[Masken <ma...@em...>]

Ali Campbell wrote:
> Hello there,
> 
> This is slightly off topic, because it does not relate directly to Linux 
> development, but I am attempting to build an external bit-banging device 
> to read and write gamecube memory cards, using an old 6502-based 
> microcomputer equipped with a very useful 6522 VIA.
> 
> It would be helpful for the purposes of testing the device to know what 
> would be an appropriate response to the 0x85 "read ID" command. I notice 
> from the Macronix 4004/05 datasheets that the Macronix manufacturer code 
> (the first byte of the two-byte response to an 0x85) is 0xC2 - does this 
> hold for the memory cards as well ?
> 
> Any other information that might be useful would be appreciated. I'm 
> doing most of my work with a nintendo original 59 block card. I'm aware 
> the cards are locked by default, but I think I may have found a way 
> round that.
> 
> many thanks.
> 
> Ali

Hello,

There's unlocking code in libogc, but since you bypass the exi bus, it 
might take some doing to make it work for you.

I've tried to find a list of card IDs, but have come up empty. If you 
find any, I'd appreciate if you let me know.

//Masken

[Gc-linux-devel] bit-banging memory card reader

[Ali C. <gam...@al...>]

Hello there,

This is slightly off topic, because it does not relate directly to Linux 
development, but I am attempting to build an external bit-banging device 
to read and write gamecube memory cards, using an old 6502-based 
microcomputer equipped with a very useful 6522 VIA.

It would be helpful for the purposes of testing the device to know what 
would be an appropriate response to the 0x85 "read ID" command. I notice 
from the Macronix 4004/05 datasheets that the Macronix manufacturer code 
(the first byte of the two-byte response to an 0x85) is 0xC2 - does this 
hold for the memory cards as well ?

Any other information that might be useful would be appreciated. I'm 
doing most of my work with a nintendo original 59 block card. I'm aware 
the cards are locked by default, but I think I may have found a way 
round that.

many thanks.

Ali

Re: [Gc-linux-devel] binutils-2.16.1

[<a.o...@bl...>]

On Mon, Jul 18, 2005 at 09:05:54PM -0400, Arthur Othieno wrote:
> On Fri, Jul 15, 2005 at 04:57:51PM -0400, Arthur Othieno wrote:
> > GNU diff-style patches will be made available on the File Release System
> > for those who don't want to pull a hefty ~95MB just to build a toolchain ;)
> > These will also be bundled with an updated version of the uClibc
> > toolchain (buildroot) which should be available soon.
> 
> Patches uploaded here:
> https://sourceforge.net/project/showfiles.php?group_id=98581&package_id=158168&release_id=342875
 
Just a quick note; these may also be grabbed using cvs rdiff. So, for
example, to grab the equivalent of binutils-2.16.1-20041107.patch, one
would do:

  $ cvs -d:pserver:ano...@cv...:/cvsroot/gc-linux \
  > rdiff -u -r binutils_2_16_1_1 -r binutils_2_16_1_20041107 \
  > binutils >binutils-2.16.1-20041107.patch

Or, to diff against HEAD, simply s/-r binutls_2_16_1_20041107//

	Arthur

[Gc-linux-devel] Trouble with SSH

[Adam E. <a9...@ya...>]

I am trying to boot the Gamecube to Linux 2.6.8 release found on the web (and I am fairly new to running linux on the cube).  I am able to get up to the login prompt on the cube and according to the kernal messages on the screen, it appears as if SSH is started.  However when I try: ssh -v -v -v root@192.168.0.254 I get the following:
 
[root@birdo log]# ssh -v root@192.168.0.254
OpenSSH_3.6.1p2, SSH protocols 1.5/2.0, OpenSSL 0x0090701f
debug1: Reading configuration data /etc/ssh/ssh_config
debug1: Applying options for *
debug1: Rhosts Authentication disabled, originating port will not be trusted.
debug1: Connecting to 192.168.0.254 [192.168.0.254] port 22.
debug1: Connection established.
debug1: identity file /root/.ssh/identity type -1
debug1: identity file /root/.ssh/id_rsa type -1
debug1: identity file /root/.ssh/id_dsa type -1
ssh_exchange_identification: Connection closed by remote host
debug1: Calling cleanup 0x8062c30(0x0)
 
The /var/log/auth.log on the cube has the following message in it:
 
Jul 29 10:13:13 cube sshd[152]: error: Bind to port 22 on 192.168.0.4 failed: Cannot assign requested address.
Jul 29 10:13:13 cube sshd[152]: fatal: Cannot bind any address.
 
I have the Listenaddress in sshd_config set to 192.168.0.4 (Linux box serving the file system), but I have also tried 192.168.0.254 (cube address).  With Listenaddress set to the latter, the message in auth.log on the cube says that the address is already in use.
 
During these ssh failures I do see some new messages in the /var/log/wcstatus.log of the Linux box serving the file system:
 
3063229360 : device : (Fri 29 Jul 2005 10:13:35 AM CDT) : adapterStateHasChanged: eth1 has changed link state from "active" to "inactive"
3063229360 : device : (Fri 29 Jul 2005 10:13:37 AM CDT) : adapterStateHasChanged: eth1 adapter has just become active
3063229360 : device : (Fri 29 Jul 2005 10:13:37 AM CDT) : wg_dhcp_renew_function: The adapter does not have corresponding /var/run/dhclient-eth1.pid file
 
In /etc/init.d/ssh on the cube, these are the lines to start the ssh daemon:
 
echo -n "Starting OpenBSD Secure Shell server: sshd"
start-stop-daemon --start --quiet --pidfile /var/run/sshd.pid --exec /usr/sbin/sshd
echo "."
 
Is there anything I am missing in the setup that would cause ssh to fail in this way? 
 
Thanks

		
---------------------------------
 Start your day with Yahoo! - make it your home page 
		
---------------------------------
 Start your day with Yahoo! - make it your home page 

Re: [Gc-linux-devel] [ANN] buildroot uClibc toolchain: powerpc-linux-uclibc-10901

[<a.o...@bl...>]

On Sat, Jul 23, 2005 at 07:26:06PM -0400, Arthur Othieno wrote:
> Greetings,
> 
> This took a little longer than I anticipated. I got distracted squishing
> a few bugs in other places along the way ;-)
> 
> Release notes here:
> https://sourceforge.net/project/shownotes.php?release_id=344329
> 
> Download here:
> https://sourceforge.net/project/showfiles.php?group_id=98581&package_id=120824&release_id=344329
> 
> Enjoy.
> 
> 	Arthur
 
Ok, I've shuffled things around a little bit since. You'll find these
under package "buildroot", previously "buildroot uClibc Toolchain"
(previously "uClibc Toolchain" ;)

Links to the release notes and actual download are still valid.

	Arthur

Re: [Gc-linux-devel] binutils-2.16.1

[Groepaz <gr...@gm...>]

On Sunday 17 July 2005 03:18, Arthur Othieno wrote:

> Also, additional patent numbers for these, FWIW:
>
> U.S. Pat. 6,468,160; _6,712,704_ (Security system for video game system
> with hard disk drive and internet access capability)
>
> U.S. Pat. 6,681,296; _6,859,862_ (Method and apparatus for software
> management of on-chip cache)

thanks

--
http://www.hitmen-console.org
http://www.gc-linux.org/docs/yagcd.html
http://www.pokefinder.org

[Gc-linux-devel] [ANN] buildroot uClibc toolchain: powerpc-linux-uclibc-10901

[<a.o...@bl...>]

Greetings,

This took a little longer than I anticipated. I got distracted squishing
a few bugs in other places along the way ;-)

Release notes here:
https://sourceforge.net/project/shownotes.php?release_id=344329

Download here:
https://sourceforge.net/project/showfiles.php?group_id=98581&package_id=120824&release_id=344329

Enjoy.

	Arthur

Re: [Gc-linux-devel] GC Linux Wiki?

[Daniel T. <da...@re...>]

On Wed, 2005-07-20 at 14:58 +0200, Michael Steil wrote:
> Hi!
>=20
> Isobel, Apgo and me discussed whether it would make sense to migrate =20
> to a wiki system on the http://www.gc-linux.org website, so that it =20
> <snip>
> So what do the members of the mailing list think about it?

I think this is a very good idea.

I've been starting to write some documentation but its a little too
'bits 'n pieces' to be worth releasing. I could easily copy the good
bits into a wiki though.

--=20
Daniel Thompson (Merlin) <da...@re...>
signature.asc? http://www.redfelineninja.dsl.pipex.com/signature.html

Did Sigmund's wife wear Freudian slips?

[Gc-linux-devel] GC Linux Wiki?

[Michael S. <st...@in...>]

Hi!

Isobel, Apgo and me discussed whether it would make sense to migrate  
to a wiki system on the http://www.gc-linux.org website, so that it  
is easier to write, fix and extend documentation, and so that  
externals can contribute more easily.

http://www.xbox-linux.org has been a wiki for over a year now. We  
have had some problems with the MediaWiki system lately, but they had  
been exclusively because of the server setup at SourceForge, and now  
that more and more projects on SourceForge use MediaWiki, their  
current version 1.4.7 works very well on their servers. Try xbox- 
linux.org...

So what do the members of the mailing list think about it?

   Michael

Re: [Gc-linux-devel] Debian and SD cards

[Rob R. <ro...@re...>]

Personally I think the best way is to use kernel/drivers/mmc/mmc.c. 
Write/adapt a host interface for that and perhaps patch it to accept SD 
as well. SD just needs a little different initialisation, code is in the 
existing sd card driver. Unfortunately, I won't have the time to do 
that, working about 24/7 on this: http://www.solarteamtwente.nl/ . This 
is a project which has a deadline end of august, and we need to get a 
lot done to get everything working on time :(

Regards Rob


GoatZilla wrote:

>Hello again.
>
>I just started looking at the driver when the thought occurred to me...
>
>Would this driver speed things up?  It was written to stick an MMC
>card on one of those Linksys routers.  Yank out the bit twiddling of
>the GPIO port, stuff the EXI calls in their place?
>
>http://kiel.kool.dk:27/mmc.c
>
>
>
>
>
>On 7/10/05, Rob Reilink <ro...@re...> wrote:
>  
>
>>GoatZilla wrote:
>>
>>    
>>
>>>Hi, I had a few questions about the GC Linux happenings...
>>>
>>>
>>>1.  How fast is the SD card access on the GC?
>>>
>>>
>>>
>>>      
>>>
>>Max speed of the SD card in 1 bit mode (which must be used on the GC) is
>>25Mbit/s. Unfortunately, the GC allows to choose for 16Mbit or 27Mbit.
>>So with a little overclocking (which is really just a software setting)
>>27MBit/s is possible
>>
>>    
>>
>>>2.  Are there any plans underway to just enable running Debian on the
>>>GC (non-NFS)?  It would be nice to be able to simply boot the PowerPC
>>>netinst image with a minor amount of hacking, and just do a netinst
>>>onto and run off of SD, since flash is so cheap nowadays.  Hence my
>>>question #1...
>>>
>>>
>>>
>>>      
>>>
>>Plans have been there to enable a GC to run just off an SD card w/o NFS
>>(or without network at all if you boot another way than PSO). It should
>>be no problem, except:
>>-afaik the SD driver has been disabled since the EXI framework has been
>>changed
>>-I (writer of the original SD driver) have no time to fix it or to get
>>playing with the SD-only boot method
>>
>>Rob
>>
>>    
>>
>>>Thanks!
>>>
>>>
>>>-------------------------------------------------------
>>>This SF.Net email is sponsored by the 'Do More With Dual!' webinar happening
>>>July 14 at 8am PDT/11am EDT. We invite you to explore the latest in dual
>>>core and dual graphics technology at this free one hour event hosted by HP,
>>>AMD, and NVIDIA.  To register visit http://www.hp.com/go/dualwebinar
>>>_______________________________________________
>>>Gc-linux-devel mailing list
>>>Gc-...@li...
>>>https://lists.sourceforge.net/lists/listinfo/gc-linux-devel
>>>
>>>
>>>      
>>>
>>
>>-------------------------------------------------------
>>This SF.Net email is sponsored by the 'Do More With Dual!' webinar happening
>>July 14 at 8am PDT/11am EDT. We invite you to explore the latest in dual
>>core and dual graphics technology at this free one hour event hosted by HP,
>>AMD, and NVIDIA.  To register visit http://www.hp.com/go/dualwebinar
>>_______________________________________________
>>Gc-linux-devel mailing list
>>Gc-...@li...
>>https://lists.sourceforge.net/lists/listinfo/gc-linux-devel
>>
>>    
>>
>
>
>-------------------------------------------------------
>SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
>from IBM. Find simple to follow Roadmaps, straightforward articles,
>informative Webcasts and more! Get everything you need to get up to
>speed, fast. http://ads.osdn.com/?ad_idt77&alloc_id492&op=click
>_______________________________________________
>Gc-linux-devel mailing list
>Gc-...@li...
>https://lists.sourceforge.net/lists/listinfo/gc-linux-devel
>  
>

Re: [Gc-linux-devel] Debian and SD cards

[GoatZilla <goa...@gm...>]

Hello again.

I just started looking at the driver when the thought occurred to me...

Would this driver speed things up?  It was written to stick an MMC
card on one of those Linksys routers.  Yank out the bit twiddling of
the GPIO port, stuff the EXI calls in their place?

http://kiel.kool.dk:27/mmc.c





On 7/10/05, Rob Reilink <ro...@re...> wrote:
> GoatZilla wrote:
>=20
> >Hi, I had a few questions about the GC Linux happenings...
> >
> >
> >1.  How fast is the SD card access on the GC?
> >
> >
> >
> Max speed of the SD card in 1 bit mode (which must be used on the GC) is
> 25Mbit/s. Unfortunately, the GC allows to choose for 16Mbit or 27Mbit.
> So with a little overclocking (which is really just a software setting)
> 27MBit/s is possible
>=20
> >2.  Are there any plans underway to just enable running Debian on the
> >GC (non-NFS)?  It would be nice to be able to simply boot the PowerPC
> >netinst image with a minor amount of hacking, and just do a netinst
> >onto and run off of SD, since flash is so cheap nowadays.  Hence my
> >question #1...
> >
> >
> >
> Plans have been there to enable a GC to run just off an SD card w/o NFS
> (or without network at all if you boot another way than PSO). It should
> be no problem, except:
> -afaik the SD driver has been disabled since the EXI framework has been
> changed
> -I (writer of the original SD driver) have no time to fix it or to get
> playing with the SD-only boot method
>=20
> Rob
>=20
> >Thanks!
> >
> >
> >-------------------------------------------------------
> >This SF.Net email is sponsored by the 'Do More With Dual!' webinar happe=
ning
> >July 14 at 8am PDT/11am EDT. We invite you to explore the latest in dual
> >core and dual graphics technology at this free one hour event hosted by =
HP,
> >AMD, and NVIDIA.  To register visit http://www.hp.com/go/dualwebinar
> >_______________________________________________
> >Gc-linux-devel mailing list
> >Gc-...@li...
> >https://lists.sourceforge.net/lists/listinfo/gc-linux-devel
> >
> >
>=20
>=20
>=20
> -------------------------------------------------------
> This SF.Net email is sponsored by the 'Do More With Dual!' webinar happen=
ing
> July 14 at 8am PDT/11am EDT. We invite you to explore the latest in dual
> core and dual graphics technology at this free one hour event hosted by H=
P,
> AMD, and NVIDIA.  To register visit http://www.hp.com/go/dualwebinar
> _______________________________________________
> Gc-linux-devel mailing list
> Gc-...@li...
> https://lists.sourceforge.net/lists/listinfo/gc-linux-devel
>