Re: [9fans] IEEE 1394 support?

Re: [9fans] IEEE 1394 support?

[jmk]

On Fri Aug  3 15:16:22 EDT 2001, cross@math.psu.edu wrote:
> In article <20010803173049.658A4199D5@mail.cse.psu.edu> you write:
> >The Shuttle FV24 motherboard would be a good place to start, although I have
> >no direct experience with it myself:
> 
> This looks really neat, except for one thing:
> 
> >	RealTek RTL8139C 10/100 Ethernet
> 
> This is that lame RealTek ethernet interface, right?  Hrmph; why
> can't someone create a dinky little motherboard exactly like this
> one, but with an Intel 82557 or something on it?  *sigh*
> 
> 	- Dan C.
> 

The 8255x chips have their faults too. For anything other than an
extremely specialised application the differences between the two
are not worth bothering about.

Re: [9fans] IEEE 1394 support?

[Dan Cross]

In article <20010803192416.7945D199D5@mail.cse.psu.edu> you write:
>The 8255x chips have their faults too. For anything other than an
>extremely specialised application the differences between the two
>are not worth bothering about.

Really?  For some reason, I got the impression from the BSD people
that there were perceptible problems with the Realtek chips.  Maybe
not.  Still, it sort of sticks out like a sore thumb on an otherwise
very nice piece of gear.

	- Dan C.

Re: [9fans] IEEE 1394 support?

[Dan Cross]

In article <20010804020251.2BDA0199D7@mail.cse.psu.edu> you write:
>As I've mentioned before, I started doing a driver for the Realtek chip
>and gave up when I tried to supplement the meagre datasheet by looking at
>the Linux and one of the *BSD drivers. Both drivers seemed to have fixes
>for seemingly awful chip problems, but not the same ones.

Exactly my point....

>If there was an 8255x on the Shuttle motherboard it might be the only Intel
>chip on it.

Hmm, perhaps a 3COM 3c905x, or a DEC 21040 (who makes the Tulip now,
anyway?), or even an NS83815, which is even mostly compatible with the
SiS900.  Surely there's an alternative?  I suppose one could use the
only PCI slot for a better ethernet card, but I'd rather use it for
something more interesting.

Some of the comments from the BSD driver are kind of funny:

/*
 * The RealTek 8139 PCI NIC redefines the meaning of 'low end.' This is
 * probably the worst PCI ethernet controller ever made, with the possible
 * exception of the FEAST chip made by SMC. The 8139 supports bus-master
 * DMA, but it has a terrible interface that nullifies any performance
 * gains that bus-master DMA usually offers.
 *
 * For transmission, the chip offers a series of four TX descriptor
 * registers. Each transmit frame must be in a contiguous buffer, aligned
 * on a longword (32-bit) boundary. This means we almost always have to
 * do mbuf copies in order to transmit a frame, except in the unlikely
 * case where a) the packet fits into a single mbuf, and b) the packet
 * is 32-bit aligned within the mbuf's data area. The presence of only
 * four descriptor registers means that we can never have more than four
 * packets queued for transmission at any one time.
 *
 * Reception is not much better. The driver has to allocate a single large
 * buffer area (up to 64K in size) into which the chip will DMA received
 * frames. Because we don't know where within this region received packets
 * will begin or end, we have no choice but to copy data from the buffer
 * area into mbufs in order to pass the packets up to the higher protocol
 * levels.
 *
 * It's impossible given this rotten design to really achieve decent
 * performance at 100Mbps, unless you happen to have a 400Mhz PII or
 * some equally overmuscled CPU to drive it.
 *
 * On the bright side, the 8139 does have a built-in PHY, although
 * rather than using an MDIO serial interface like most other NICs, the
 * PHY registers are directly accessible through the 8139's register
 * space. The 8139 supports autonegotiation, as well as a 64-bit multicast
 * filter.
 *
 * The 8129 chip is an older version of the 8139 that uses an external PHY
 * chip. The 8129 has a serial MDIO interface for accessing the MII where
 * the 8139 lets you directly access the on-board PHY registers. We need
 * to select which interface to use depending on the chip type.
 */

Granted most CPU's are now faster than a 400MHz Pentium (wow; just
consider that for a moment), but still....

	- Dan C.

Re: [9fans] IEEE 1394 support?

[jmk]

On Fri Aug  3 21:46:22 EDT 2001, cross@math.psu.edu wrote:
> In article <20010803192416.7945D199D5@mail.cse.psu.edu> you write:
> >The 8255x chips have their faults too. For anything other than an
> >extremely specialised application the differences between the two
> >are not worth bothering about.
> 
> Really?  For some reason, I got the impression from the BSD people
> that there were perceptible problems with the Realtek chips.  Maybe
> not.  Still, it sort of sticks out like a sore thumb on an otherwise
> very nice piece of gear.
> 
> 	- Dan C.

As I've mentioned before, I started doing a driver for the Realtek chip
and gave up when I tried to supplement the meagre datasheet by looking at
the Linux and one of the *BSD drivers. Both drivers seemed to have fixes
for seemingly awful chip problems, but not the same ones.

If there was an 8255x on the Shuttle motherboard it might be the only Intel
chip on it.

--jim

Re: [9fans] IEEE 1394 support?

[Dan Cross]

In article <20010803173049.658A4199D5@mail.cse.psu.edu> you write:
>The Shuttle FV24 motherboard would be a good place to start, although I have
>no direct experience with it myself:

This looks really neat, except for one thing:

>	RealTek RTL8139C 10/100 Ethernet

This is that lame RealTek ethernet interface, right?  Hrmph; why
can't someone create a dinky little motherboard exactly like this
one, but with an Intel 82557 or something on it?  *sigh*

	- Dan C.

Re: [9fans] IEEE 1394 support?

[anothy]

// I'd just as soon see 9P2000 and this new file
// server before either, though.

yeah, okay, i'd agree with that.
-α.

Re: [9fans] IEEE 1394 support?

[jmk]

On Fri Aug  3 12:28:21 EDT 2001, anothy@cosym.net wrote:
> what equipment had you chosen?
> 
> of cource, maybe we shouldn't distract you guys so we
> can get USB support out here...
> -α.

We tried to kill 2 birds with one stone, the person interested in doing
1394 also needed a laptop so we looked for one that had 1394 and that we
felt we could get working with Plan 9 easily. Unfortunately the one we
chose was rejected by purchasing as coming from a manufacturer with reliability
problems. By then we were distracted and beleaguered.

The Shuttle FV24 motherboard would be a good place to start, although I have
no direct experience with it myself:

	socket 370, VIA chipset with integrated VGA
	1x32 bit PCI slot
	VIA AC97 audio codec
	RealTek RTL8139C 10/100 Ethernet
	dual 1394 controller (Lucent FW323)
	optional TV-out
	ata/100
	1 serial
	1 parallel
	ps/2
	4 USB ports

It's a MicroATX motherboard (190x175mm) and should cost somewhere between
$100 and $140. With a 'fast enough' processor, might be able to make a system
without a fan in a little box.

Re: [9fans] IEEE 1394 support?

[Dan Cross]

In article <20010803162735.06560199D7@mail.cse.psu.edu> you write:
>what equipment had you chosen?
>
>of cource, maybe we shouldn't distract you guys so we
>can get USB support out here...

Interestingly, the last time I looked at a firewire device, I was
intrigued to find out that the UHCI/OHCI interface on the host for it
was very close if not the same as that used with USB host interfaces.

But...I concur.  USB support is probably more pressing than firewire.
I'd just as soon see 9P2000 and this new file server before either,
though.  :-)

	- Dan C.

Re: [9fans] IEEE 1394 support?

[anothy]

what equipment had you chosen?

of cource, maybe we shouldn't distract you guys so we
can get USB support out here...
-α.

Re: [9fans] IEEE 1394 support?

[jmk]

On Fri Aug  3 05:14:23 EDT 2001, jas@corpus-callosum.com wrote:
> 
> Has there been any work on getting IEEE 1394 working for Plan 9?
> 
> As more and more devices are becoming available, it would be nice to be able
> to utilize some of them (video, audio, storage ...) with Plan 9 in order
> to potentially share the these connected devices with other machines on the
> network.
> 
> jas

We were interested in doing this a few months ago, when Lucent was merely a
struggling telecommunications manufacturer. Unfortunately, by the time we'd
chosen some equipment we had become beleaguered and no longer had the money
to purchase it.

--jim

[9fans] IEEE 1394 support?

[Jeff Sickel]

Has there been any work on getting IEEE 1394 working for Plan 9?

As more and more devices are becoming available, it would be nice to be able
to utilize some of them (video, audio, storage ...) with Plan 9 in order
to potentially share the these connected devices with other machines on the
network.

jas